Citizenship, Exile, and Natural Rights in Medieval Roman Law, ea. 
1200-1400 
Cornelius Maximilian Riethdorf 
Peterhouse 
June 2016 
This dissertation is submitted for the degree of Doctor of Philosophy. 
CAMBRIDGE 
UNIVERSITY . 
LIBRARY 
Abstract 
Citizenship, Exile, and Natural Rights in Medieval Roman Law, ea. 1200-1400 
Cornelius M Riethdorf 
This dissertation lies at the intersection of the history of political thought and legal 
history. It concerns the relationship between the political community (civitas) and the 
individual, as citizen (civis) and as human being, in the medieval Roman law 
tradition from ea. 1200 to 1400. My primary sources are the writings of the most 
famous north-Italian jurists (including Accursius, Bartolus of Sassoferrato, and 
Baldus de Ubaldis ), who were at the centre of the late-medieval reception of classical 
Roman law. I analyse three types of exile found in the legal theory of the Roman 
lawyers: deportation, imperial banishment, and communal banishment. This analysis 
is framed contextually with reference to the social and political conditions of 
northern Italy, where exile served not only as a political measure, but also formed 
part of the ordinary administration of justice. The legal procedure of exiling is 
informative about what we would today call due process of law and states of 
emergency. Furthermore, since exile caused loss of citizenship, by evaluating the 
legal qualities lost and retained by exiles, it is possible to establish the meaning of 
civic belonging, as well as the meaning of not belonging to any civitas - of being 
merely a human person with natural rights (statelessness, in modem terminology). 
My findings allow me to revise existing narratives about citizenship and natural 
rights in medieval Roman law. I argue that citizenship was not fundamentally 
contractual, and that it was primarily (if not exclusively) associated with private-law 
rights rather than political rights and public duties. I reject the current historical 
consensus that sees the origins of natural rights language only in medieval canon 
law, philosophy, and theology, by demonstrating that the Roman lawyers also 
articulated a theory of natural rights in terms of subjective ius naturale. 
Preface 
Declaration 
This dissertation is the result of my own work and includes nothing which is the 
outcome of work done in collaboration except as declared in the preface and 
specified in the text. 
It is not substantially the same as any that I have submitted, or, is being concurrently 
submitted for a degree or diploma or other qualification at the University of 
Cambridge or any other university or similar institution except as declared in the 
preface and specified in the text. I further state that no substantial part of my 
dissertation has already been submitted, or, is being concurrently submitted for any 
such degree, diploma or other qualification at the University of Cambridge or any 
other university or similar institution except as declared in the preface and specified 
in the text. 
It does not exceed the prescribed word limit for the relevant Degree Committee. The 
length of this dissertation is 79,889 words. 
Relationship to M.Phil. Thesis 
This dissertation has developed from work undertaken for my 2012 M.Phil. thesis, 
'Fourteenth-Century Roman Lawyers on Banishment and Exile'. While the topic -
exile in medieval Roman law - is the same, I have considerably expanded the 
primary-source base, which now includes sources from the thirteenth century as well 
as many from the fourteenth century that I did not previously cover. Naturally, the 
primary material discussed in my M.Phil. thesis can also be found at various points 
in this dissertation. I have, however, completely revised and reworked this material 
in conjunction with the new sources at my disposal. Moreover, this dissertation 
engages with a far broader range of secondary literature, which has influenced my 
narrative and altered its focus in important respects. As a result, although certain 
aspects of my interpretation of (some of) the sources are inevitably the same or 
similar to the interpretation given in my M.Phil. thesis, my arguments have not only 
substantially changed and developed, but also cover much more ground and have a 
much wider scope. At no point have I merely inserted or paraphrased entire segments 
of my M.Phil. thesis. 
A Note on Transcription, Translation, Gender, and References 
While all my transcriptions of the early modem printed editions keep the original 
standard spelling and use of capitals, I have modernised punctuation and provided 
numerical references to the Roman and canon law. All translations are my own 
unless otherwise specified. 
The medieval civilians and their legal authorities refer to the person as such as a 'he', 
and indeed primarily think of that person as a 'he', except when they very 
specifically speak of women. I normally follow this usage in my prose in order to 
reflect the gendered nature of the discourse and to maintain grammatical consistency 
with the sources. 
Unless otherwise stated, all citations of the Corpus iuris civilis follow the Berlin 
edition of Krueger, Mommsen, and Scholl, and all citations of the Corpus iuris 
canonici follow the Leipzig edition of Richter and Friedberg. All citations of 
Accursius' Glossa ordinaria refer to the version printed with the 1558-60 Lyon 
edition of the Corpus iuris civilis. 
All references in the footnotes to primary and secondary sources are in abbreviated 
form. Full references are provided in the bibliography attached to the end of this 
dissertation. 
II 
Abbreviations 
BIMA Bibliotheca iuridica medii aevi 
cons. cons ilium/cons ilia 
CQT Consilia, quaestiones, et tractatus 
DBGI Dizionario biografico dei giuristi italiani 
DBI Dizionario biografico degli italiani 
MGH Monumenta Germaniae Historica 
OIR Opera iuridica rariora 
Corpus iuris civilis 
Inst. Institutes 
D. Digest 
C. Code 
Nov. Novels 
Corpus iuris canonici 
X. 
Sext. 
Clem. 
Ext. com. 
Liber extra 
Liber sextus 
Clementines 
Extravagantes communes 
111 
Contents 
Introduction ....... ... ..... ........ ....... ................................... ............................. .... ... ..... ... 1-6 
Chapter I - Citizenship, Exile, and Natural Rights: A Review .... ................... .. ... 7-32 
Chapter 2 - Deportation ....................... .... .................. .... ......... ..... .. .................... . 33-92 
Chapter 3 - The Imperial Ban ...... ........ ............. ... .. ...... ..... .............. .................. 93-174 
Chapter 4-The Communal Ban .. ... ............... ......... .. ...... ................. ......... ...... 175-268 
Bibliography ... ....... ....................... ........ ............. ....................... ........... ............ 269-282 
V 
Introduction 
The reception of classical Roman law in late-medieval Europe has long been 
regarded as fundamental in the development of Western political thought. Much has 
been written about the contribution of the Roman law tradition to the concept of the 
state during the medieval and early-modem period. Pride of place is usually granted 
to the medieval jurists of northern Italy, whose political thought was shaped a great 
deal by the competing claims of the Holy Roman Empire, the papacy, and the cities 
in that region. 1 The basic political unit of their discourse was the civitas, the city, 
understood not as a physical collection of buildings, but in the sense of res publica -
the political community.2 
This thesis explores the relationship between the civitas and the individual in 
medieval Roman law. The subject at hand is the jurists' theory of citizens who had 
lost membership of the civitas and had become exiles and outlaws. It is a study that 
necessarily lies at the intersection of legal history and the history of political thought. 
While the source material will consist of texts that are primarily legal, the ideas 
pursued therein will be political. By analysing the position of exiles - by evaluating 
the qualities they lost and those they retained - we can establish the meaning of 
citizenship on the one hand, and the meaning of being without citizenship on the 
other. 3 Since certain categories of exiles were deprived of all civic forms of 
1 Important Anglophone works include: Woolf, Barto/us of Sassoferrato ; Carlyle and 
Carlyle, Mediaeval Political Theory VI; Kantorowicz, 'Kingship '; Ullmann, 'De Barto li 
sententia'; Tierney, Religion, Law, and the Growth of Constitutional Thought; Canning, 
'Ideas of the State', Baldus de Ubaldis, 'Law, Sovereignty, and Corporation Theory', ' Ital ian 
Juristic Thought'; Hopfl, 'Fundamental Law'; Stein, 'Roman Law' ; Pennington, 'Law, 
Legislative Authority, and Theories of Government', The Prince and the Law; Ryan, 
'Bartolus of Sassoferrato', 'Rulers and Justice', 'Freedom, Law, and the Medieval State', 
' Political Thought'. 
2 The civitas is the entity that would be identified with the state towards the end of the early 
modem period: in the words of Hobbes, Leviathan, p. 9, 'that great Leviathan called a 
Common-Wealth, or State, (in latine Civitas) .' Cf. Brett, Changes of State, pp. 1-3, who 
provides an excellent introductory explanation of the term civitas in early modern political 
thought. 
3 Some recent studies concerning the north-Italian cities in the period from the twelfth to the 
fourteenth century have applied the method of examining citizenship through forms of 
exclusion, albeit predominantly from a practical rather than theoretical Roman-law 
perspective. See: Milani, L 'esclusione dal comune, esp. his opening remarks on pp. 24-5; 
belonging, we can also establish the meaning of existing simply as a human being -
statelessness, in modern terminology. According to the thought of the jurists, this left 
the exile under the provisions of natural law, leading them to develop an important 
discourse on natural rights. A recurring theme in this thesis will be the clash between 
the demands of these rights and the demands of the city upon its former citizens.4 
Moreover, exile was often the result of political conflict and occurred at moments of 
great instability. The judicial process of becoming an exile is therefore revealing 
about the tension between what we would today call due process of law and states of 
emergency. 
The purpose of this study is to provide an exegesis of legal texts. Certain exceptions 
aside, we will not consider the practical implications of juristic theory. The aim of 
bringing in social and political contexts, and of discussing contemporary legal 
practice, will be to elucidate the jurists' ideas. This exegesis will focus exclusively 
on the Roman lawyers - the civilians - and secular forms of exile. Excommunication 
by the church, whether in Roman law or in its contemporary sister discipline of 
canon law, will only feature if it intersects with our topic. Although the revival of 
Roman law began with the rediscovery of Justinian's Corpus iuris civilis in the 
second half of the eleventh century, we will confine ourselves to the period between 
ea. 1200 and 1400, covering civilian thought from the Ordinary Gloss of Accursius 
( d. 1260/3) to the works of the most famous jurists of the medieval era: Bartol us of 
Sassoferrato ( d. 1357) and his pupil Baldus de Ubaldis ( d. 1400). Concentrating our 
efforts on this period makes sense not only because it oversaw the pinnacle of 
medieval Roman legal science. There is little incentive to include the twelfth century 
since it was only around the middle of the thirteenth century that the jurists began to 
consider medieval exile in detail; the glossators, up to the time of Accursius, mostly 
restricted themselves to analysing the ancient forms of exile in the Corpus iuris. 
There is equally little incentive to include the fifteenth century, which has been 
traditionally viewed as an age of little originality, when juristic thought was largely 
derivative of the achievements of previous generations. Indeed, what little work has 
been done on the civilian theory of exile in the fifteenth century confirms that 
and Vallerani, 'Logica della documentazione', 'Diritti di cittadinanza', and 'Evasione 
fiscale'. 
4 This aspect of the thesis is similar in conception to Brett's Changes of State. In essence, 
just as her book, which pursues this theme in the political thought of the sixteenth and 
seventeenth centuries, it concerns 'the conflicted relationship between nature and the city -
the fraught intersection of the political and the natural world' (p. 3). 
2 
assessment.5 In geographical terms, our investigation will be limited to those jurists 
who were based in the northern half o.f the Italian peninsula: in the regnum Italicum, 
a part of the Holy Roman Empire, and in the papal lands; the French and Neapolitan 
schools of Roman law have been omitted. Northern Italy offers itself as an object of 
study not only because it produced the leading juristic minds of the medieval era 
(such as Accursius, Bartolus, and Baldus), but also because exile, as a legal 
institution, occupied a position of importance in the region's life unrivalled anywhere 
else in Europe. For the north-Italian lawyers exile was a pressing contemporary legal 
problem. Civic exclusion was a central feature of the ordinary administration of 
criminal justice as well as a political phenomenon. 
Most historical attention has been devoted to studying exile as part of the political 
conflicts experienced by northern Italy from the early eleventh century until the end 
of our period and beyond.6 By the thirteenth century, the region's cities had emerged 
as semi-independent and, in some cases, fully independent political actors, governed 
under a range of republican and seigniorial regimes. Their political life was marked 
5 Ghisalberti, 'La condanna', and Cavalca, II bando, consider some of the material on the 
communal ban from the fifteenth century (and beyond). Cavalca (pp. 99-100) concludes: 
' Dopo le ampie ed approfondite ricerche dei giuristi Postaccursiani, ulteriormente ampliate 
da Bartolo, Baldo e Alberico da Rosciate, la materia de! bando continuo a suscitare 
l 'interesse di altri Commentatori, come Bartolomeo da Salecito, Paolo di Castro, Raffaele 
Fulgosio e Giason de! Maino. Questi autori non raggiungono pero l'originalita dei 
Postglossatori e non offrono spunti di particolare interesse. Parimenti prive di originalita 
sono le opere dei Pratici, dopo i primi Tractatus de maleficiis ... Importanza a parte riveste 
invece !'opera di Nello da S. Gimignano ... per ampiezza, organicita e citazione di fonti ... ' 
Mooney, 'The Legal Ban ', includes a relatively detailed study ofNello's (d. 1430) Tractatus 
de bannitis (pub. 1423). Mooney claims (esp. at pp. 72-5, 117-19, and 194-223) that in 
certain important respects Nello's thought represents substantial progress, particularly in 
relation to Bartolus' work. He is led to this conclusion by an incomplete and, at times, 
incorrect understanding of the fourteenth-century civilian tradition on banishment - which is 
excusable, given his focus on Nello's treatise and the statutory law of Florence. Careful 
readers of this thesis (and, indeed, of Cavalca's book) will notice that those ideas portrayed 
by Mooney as innovative were in fact already an established part of the discourse; minor 
differences aside, Mooney's contrasts between Bartolus ' and Nello's thought are 
unwarranted. As Cavalca argues, the importance of Nello 's treatise lies in its 
comprehensiveness: fundamentally, it is a majestic synthesis of Bartolist communis opinio 
on the communal ban. 
6 Cf. Bowsky, Henry VII in Italy; Kent, 'I Medici in esilio'; Heers, Parties and Political Life, 
and L 'esilio; Laurent, 'The Exiles and the Signory'; Stam, Contrary Commonwealth; Heers 
and Bee, Exil et Civilisation; Lansing, The Florentine Magnates; Baxendale, ' Exiles in 
Practice'; Ganz, 'Paying the Price'; Zaremska, Les bannis au MoyenAge, eh. 3, pp. 65-110; 
Shaw, Politics of Exile; Brown, 'Insiders and Outsiders' ; Milani , L 'esclusione dal comune; 
Ricciardelli, Politics of Exclusion; Blanshei, Politics and Justice . 
3 
by chronic violent strife between the factions of the citizen body: between noble 
magnates and popolo, between pro-papal Guelphs and pro-imperial Ghibellines. 
Internal conflicts were much exacerbated by continual imperial and papal 
interventions, intended to make good traditional claims of jurisdiction over the area. 
While citizens fought each other, emperors, such as Frederick II (d. 1250), fought 
popes as well as the cities when they crossed the Alps in their attempts to assert a 
level of control. Exile was a weapon employed by all parties in the struggle against 
their opponents. It was common practice of both emperors and cities to place 
individuals and whole groups under the ban ( bannum ), a type of outlawry that 
permitted anybody to injure the banished (bannitus, -i) in their person and property 
with impunity. In the cities, the period of greatest volatility occurred between the 
middle of the thirteenth century and the 1330s. During that time in particular, one 
can observe the zero-sum nature of the competition between city factions, which 
meant that the accession to power of one ordinarily resulted in the expulsion of the 
other. The defeated left their cities on such a large scale that they could form 
rivalling counter-communes in exile. The intractability of the situation is nicely 
illustrated by an ambassadorial report to Emperor Henry VII in April 1313, 
informing him that Modenese exiles 'hold many castles in the district of Modena and 
are waging war against the town'. 7 The most famous exile of this period is Dante, 
who was expelled from Florence with the White Guelphs in 1302, and subsequently 
wrote his Divine Comedy while wandering Italy. 
The centrality of the cities' ban in the day-to-day life of northern Italy explains why 
the overwhelming majority of the civilian discourse on exile concerns this institution, 
which is usually called the communal ban, and which will be the subject of chapter 
4. The jurists also considered other kinds of exile, albeit to a much smaller extent. 
Chapter 3 is devoted to their thought on the above-mentioned ban of the emperors. 
While the material in these chapters refers to actual medieval practices, chapter 2 
concerns deportation, a criminal penalty of the classical Roman empire that, by our 
period, only existed in the Corpus iuris and in the jurists' exposition of its texts. In 
Justinian's compilation, deportation was the most prominent form of exile involving 
the loss of citizenship. Hence, it served as a continual and crucial reference point for 
the jurists when they analysed medieval banishment. This is the reason why we will 
discuss deportation first. It is logical for that discussion to be followed by an 
examination of imperial banishment, since this, just as deportation, caused the loss of 
universal Roman citizenship, before we move on to the communal ban, which only 
7 Doenniges (ed.), Acta Henrici VII, p. 178, quoted and translated in Stam, Contrary Commonwealth, p. 58. 
4 
resulted in the loss of the local citizenship of an Italian city. Chapter 1 serves as a 
historiographical review of a number of themes and contexts that need to be 
considered if we are to gain an accurate understanding of the civilian theory of exile, 
and a full appreciation of its contribution to medieval political thought. 
5 
·----------------·· 
6 
Chapter 1 
Citizenship, Exile, and Natural Rights: A Review 
1. Citizenship in the Corpus iuris civilis 
The thought of the medieval Roman lawyers was profoundly shaped by the Corpus 
iuris, which, in their eyes, was the ultimate authority for legal argument. A study of 
the jurists' thought should therefore begin with an assessment of Justinian's 
compilation itself, whose texts, and the ideas contained therein, constituted the raw 
materials of medieval civilian theory. Since a central part of this thesis concerns 
citizenship and the loss thereof, it will be helpful to raise a number of key points 
about the type of citizenship (or rather citizenships) in the Corpus iuris, and to 
situate these in relation to the broader developments surrounding citizenship in the 
later Roman empire. 
Roman citizenship after Emperor Caracalla's Constitutio Antoniniana (AD 212), far 
from being the exclusive reserve of a privileged minority that it was during the 
Republic and Principate, became the property of nearly all free persons under Roman 
rule. 1 The universalism of later imperial ideology meant that post-classical Roman 
law rarely considered the category of Roman citizen (civis Romanus) and its inverse 
status of foreigner (peregrinus). The aim was to unite the whole world (orbis 
terrarum) under Rome's rule and under her laws. The Roman (or civil) law - the ius 
1 For this reason, the vast majority of modem scholarship on Roman citizenship has focused 
on developments prior to AD 212, after which time this status, given its universality, has 
generally been deemed to be of little importance. Cf. Sherwin-White, The Roman Citizenship 
(the classic Anglophone study); Balsdon, Romans and Aliens; Donati Giacomini and Poma, 
Cittadini e non cittadini; Gardner, Being a Roman Citizen; Noy, Foreigners at Rome. 
Relatively recently, a small number of articles have in one way or another questioned this 
trend, exploring, though often only briefly, some of the implications of Roman citizenship in 
the late-antique empire through a variety of legal and non-legal sources: Goria, 'Romani, 
Cittadinanza'; Liebeschuetz, 'Citizen Status'; Garnsey, 'Roman Citizenship'; Mathisen, 
'Peregrini, Barbari, and Cives Romani', and 'Concepts of Citizenship' (a condensed version 
of the former article). 
7 
civile - and Roman citizenship were to apply to all. 2 If we use the early sixth century 
as a terminus, the Corpus iuris represents the culmination of these trends. References 
to Roman citizenship are largely absent; the terms civis and civitas normally refer to 
one of the cities within the empire.3 It is with these entities that the law on citizenship 
at D.50.1 and C.10.39-40 is concerned. While Roman citizenship was taken for 
granted as an attribute of all free persons subject to the emperor, and Rome was thus 
assumed to be the 'common homeland' (communis patria), 4 an individual was 
distinguished by membership of one or several of the provincial cities. 5 Each person 
therefore had a type of dual citizenship: a limited local one as well as an overarching 
universal (Roman) one.6 
Since the provincial city essentially became 'a medium for the collection of revenue' 
in the late-imperial bureaucracy,7 much of the material in the Corpus iuris pertaining 
to the cities is part of an elaborate tax code. 8 Local citizenship ( civitas )9 is one of 
2 Mathisen, 'Peregrini, Barbari, and Cives Romani', esp. p. 1037: 'By making their law 
available to all, the Romans manifested their claim to rule all that mattered in the whole 
world and establish the closest thing ever known to a "citizenship of the world".' Cf. Goria, 
'Romani, Cittadinanza', pp. 340-2. 
· 
3 Goria, 'Romani, cittadinanza' , p. 285. Goria's work is summarised in English by Quaglioni, 
'The Legal Definition', pp. 158-9. 
4 D.50.1.33 : 'Roma communis nostra patria est.' 
. 
5 The possibility of possessing multiple citizenships of cities within the empire is raised, for 
example, in the context of manumission at D.50.1.27: 'Eius, qui manumisit, municeps est 
manumissus, non domicilium eius, sed patriam secutus. et si patronum habeat duarum 
civitatium municipem, per manumissionem earundem civitatium erit municeps.' 
6 The fact of overlapping local and universal citizenships in the late-antique period is 
mentioned by Coli, 'Civitas', p. 341. His views are also summarised in English by 
Quaglioni, 'The Legal Definition', p. 157. Norr, 'Origo', denies that these overlapping local 
and universal spheres of belonging constituted a real double citizenship ('echtes 
Doppelbiirgerrecht'). It is a fair point insofar as he is right that the modern (presumably 
German) legal concept of Doppelbiirgerrecht is not a useful one with which to approach 
post-classical Roman law. 
7 Abbott and Johnson, Municipal Administration, p. 194. 
8 The following is heavily indebted to: Liebenam, Stiidteverwaltung; Reid, Municipalities; 
Abbot and Johnson, Municipal Administration; Norr, ' Origo'; Jones, The Later Roman 
Empire, vol. 2, eh. 19 ('The Cities'), pp. 712-66; Coli, 'Civitas'. A good overview of the 
concept of origo in the Corpus iuris is also provided by Kirshner, 'Mulier alibi nupta', pp. 
149-51, with a focus on how this relates to women and marriage. 
9 Civitas in the sense of 'citizenship' - i.e. denoting the quality of belonging to a civitas -
rather than in its primary sense of 'city', although well attested when it comes to Roman 
citizenship (civitas Romana), is rare in the municipal context. I can only find two examples 
(D.50.1.1 ,1 and D.50.1.6,1), and even these are somewhat ambiguous (see ns 10 and 20 
below). The meaning of civitas is further discussed in eh. 2, pp. 37-43. 
8 
three categories that define the place where each person is liable to render his due; 
the other two are origin (origo) and domicile (domicilium). If born within a lawful 
marriage, an individual's origin is his father's city of origin. Since origin causes a 
person to be a citizen (civis) of that place, citizenship is mainly derived through 
patrilineal descent. 10 Although citizenship is brought about by origin, there is no 
complete overlap between these two concepts. Origin is the result of one's birth and 
thus a natural fact of life that can never be lawfully altered or lost. 11 Citizenship, 
however, may also be acquired by other means, e.g. adoption, 12 with the result that 
an individual can possess one or more citizenships independently of his origin. 13 
Domicile is distinct from both origin and citizenship, and does not lead to the 
acquisition of the latter; it merely denotes the city where one is ordinarily resident. 14 
The law is emphatic that origin, citizenship, and domicile all create full liability to 
certain impositions - honores and munera (syn. onera) - demanded by the 
authorities, such that one person can be liable in multiple cities. 15 Honores are higher 
municipal offices, while a wide range of taxes and personal services constitute 
munera. 16 The distinction between the two loses its importance in the sense that the 
law regularly considers them together as a pair of expensive, troublesome obligations 
'
0 D.50.1 .1: 'Municipem aut nativitas facit aut manum1ss10 aut adoptio. [ ... ] (2) Qui ex 
duobus igitur Campanis parentibus natus est, Campanus est. sed si ex patre Campano, matre 
Puteolana, aeque municeps Campanus est...' D.50.1.6, 1: 'Filius civitatem, ex qua pater eius 
naturalem originem ducit, non domicilium sequitur.' D.50.1.9: 'Eius, qui iustum patrem non 
habet, prima origo a matre eoque die, quo ex ea editus est, numerari debet.' C.10.39(38).3: 
'Filios apud originem patris, non in materna civitate, etsi ibi nati sunt, si modo non 
domiciliis retineantur, ad honores seu munera posse compelli explorati iuris est. ' 
C.10.40(39).7: 'Cives quidem origo manumissio adlectio adoptio, incolas vero, sicut et divus 
Hadrianus edicto suo manifestissime declaravit, domicilium facit.' 
11 D.50.1.6: 'Adsumptio originis, quae non est, veritatem naturae non peremit: errore enim 
veritas originis non amittitur nee mendacio dicentis se esse, unde non sit, deponitur: neque 
recusando quis patriam, ex qua oriundus est, neque mentiendo de ea, quam non habet, 
veritatem mutare potest.' Cf. C.10.39(38).4. 
12 . D.50.1.1; C.10.40(39).7. Seen. 10 above. 
13 D.50.1.15,3: 'Ius originis in honoribus obeundis ac muneribus suscipiendis adoptione non 
mutatur: sed novis quoque muneribus filius per adoptivum patrem adstringitur.' D.50.1.16: 
'Sed si emancipatur ab adoptivo patre, non tantum filius, sed etiam civis eius civitatis, cuius 
per adoptionem fuerat factus, esse desinit.' Cf. D.50.1.17,9. 
14 C.10.40(39). 7 (seen. 10 above). Cf. C. l 0.40(39).7,1; D.50.1.17, 13. 
15 D.50.1.29: 'Incola et his magistratibus parere debet, apud quos incola est, et illis, apud 
quos civis est: nee tantum municipali iurisdictioni in utroque municipio subiectus est, verum 
etiam omnibus publicis muneribus fungi debet.' C. l 0.39(38).1: 'Cum te byblium origine, 
incolam autem apud Berytios esse proponis, merito apud utrasque civitates muneribus fungi 
compelleris.' Cf. C. l 0.39(38).3/5; C. l 0.40(39). I /2/4/5/6. 
16 Cf. D.50.4 (De muneribus et honoribus). 
9 
to which one is either compelled 17 or from which one is immune. 18 Fundamentally, 
in the later empire, municipal office-holding became just another kind of service 
because the main function of municipal officers (decurions) was to collect taxes for 
the imperial treasury and to make good any shortfalls in expected revenue from their 
private estates. 19 In sum, local citizenship in the Corpus iuris is onerous and fiscal. It 
is fitting that the Digest gives Ulpian 's definition of the municipal citizen as 
someone subject to munera. In fact, according to Ulpian, municeps - another term 
for the municipal civis - is derived from the phrase 'muneris participes' (i.e. 'those 
who participate in munera').20 
2. Municipal Citizenship in Medieval Roman law 
The medieval civilians considered the Roman empire a living entity and its ancient 
law normative for their own times; hence, they viewed the north-Italian city as the 
modem equivalent of the provincial city within Justinian' s compendium. The way 
the jurists did, and did not, negotiate the inevitable conceptual gap between their 
legal authorities and their medieval surroundings goes to the very heart of their 
enterprise. There is now a small but growing historical literature devoted to medi~val 
civilian thought concerning municipal citizenship. From this it is clear that the jurists 
17 E.g. D.50.4.3 ,15: 'Praeses provinciae provideat munera et honores in civitatibus aequaliter 
per vices secundum aetates et dignitates, ut gradus munerum honorumque qui antiquitus 
statuti sunt, iniungi, ne sine discrimine et frequenter isdem oppressis simul viris et viribus res 
publicae destituantur.' Cf. C. I 0.39(38).3 (see n. 10 above). 
18 E.g. D.50.5.10,4: 'Defensores rei publicae ab honoribus et muneribus eodem tempore 
vacant.' A vast section of the law describes the highly complex system of immunity from 
honores and munera enjoyed by various privileged groups such as the imperial aristocracy. 
D.50.5-6 is exclusively devoted to this topic. 
19 See, for instance: D.50.2.1; D.50.4.18,26; C. I 0.32(31 ).2/18; C. I 0.59(57).1; C.11 .59(58).1 ; 
Nov. 38 pr. Hence, the position of decurion came to be seen as highly detrimental, an 
attitude that can be observed just by browsing through D.50.1-6 and C. I 0.32(31)-59(57). 
Although this is the dominant narrative in the Corpus iuris, some of the classical material 
contained in the Digest points to an older conception of municipal government, from the 
second century AD and before, when local offices were still prized possessions - e.g. 
D.50.2.3,1/2,2/7,2/9/9,1; D.50.4.7; D.50.12.11. 
20 D.50.1. 1, I: 'Et proprie quidem municipes appellantur muneris participes, recepti in 
civitatem, ut munera nobiscum facerent: sed nunc abusive municipes dicimus suae cuiusque 
civitatis cives, ut puta Campanos, Puteolanos.' 
10 
took from their sources the conceptual categories civis, civitas, origo, and 
domicilium, and developed them in interesting ways.21 
While the Roman lawyers continued to view citizenship as a status ordinarily 
acquired through birth (origo), studies by Julius Kirshner and Joseph Canning 
centred on Bartolus and Baldus have demonstrated that the commentators also 
created sophisticated theories on the assumption of citizenship by the grant of a city 
(i.e. naturalisation). Both jurists argued that this citizenship, which was brought into 
being on the basis of a city's statutes and thus by civil law, could be fully equal to 
citizenship acquired by birth. For Bartolus, the reason was that all citizens were such 
by civil law, which conferred this status upon the satisfaction of certain conditions; it 
did not matter whether these related to birth or statutory grant. While Baldus did not 
deny the full validity of citizenship created by statute, he did fundamentally 
differentiate the latter from original citizenship, which he saw as a product of nature 
rather than civil law. Canning argues that this stemmed from the Aristotelian 
influence on Baldus' thought: man was by nature a political animal and hence a 
citizen - natural law was an operative principle. 22 We should note that in this 
context, the jurists regularly used the term civilitas rather than civitas to refer to 
citizenship; this, however, did not indicate a different concept. Since civitas could 
mean both 'city' and 'citizenship' (as we shall discuss further in chapter 2),23 the 
term civilitas seems to have arisen as a clearer way of referring to the membership of 
a city. In the context of exile, the jurists normally continued to use the older term 
civitas. 
21 Bizzarri, 'Ricerche sul diritto di cittadinanza' , p. 51 ff. ; Cortese, 'Cittadinanza', pp. 138-9; 
Bowsky, 'A New Consilium'; Rummer, 'A Fourteenth-Century Legal Opinion '; Ullmann, 
'The Rebirth of the Citizen'; Barni, "'Civis" e "Civitas"'; Riesenberg, 'Civism and Roman 
Law', 'Citizenship and Equality', 'Citizenship at Law', all then incorporated into his 
Citizenship in the Western Tradition; Kirshner, 'Paolo di Castro', 'Civitas sibi facial civem' , 
'A Consilium of Baldus on Naturalisation ' , 'A consilium of Rosello dei Roselli', 'Between 
Nature and Culture', 'Wives' Claims', ' Mulier alibi nupta', 'Donne maritate altrove ', 
'Cittadinanza come genere '; Canning, 'A Fourteenth-Century Contribution', later subsumed 
into his Baldus de Ubaldis, eh. 4, pp. 159-84; Quaglioni, 'Le radici theoriche', subsequently 
republished in English as 'The Legal Definition'; Piergiovanni, 'Alcuni consigli' ; Meier, 
Mensch und Burger, part 4, pp. 127-211; Costa, Civitas, eh. 1, pp. 1-50; Gilli , 'Comment 
cesser d'etre etranger'; Menzinger, 'Fisco, giurisdizione e cittadinanza', 'La donna 
medievale', 'Verso la costruzione' (which includes the previous two articles), 'Diritti di 
cittadinanza'. 
22 Kirshner, 'Paolo di Castro', 'Civitas sibi facial civem', 'A Consilium of Baldus on 
Naturalisation', 'Between Nature and Culture'; Canning, Baldus de Ubaldis, eh. 4, pp. 159-
84. 
23 h C . 2, pp. 37-43 . 
11 
Recent studies by Sara Menzinger have characterised municipal citizenship in 
civilian theory as a status that predominantly comprised duties.24 She provides a 
wealth of evidence that the jurists, from the glossator Rogerius ( d. ea. 11 70) to the 
commentator Albericus de Rosate ( d. 1360), discussed the citizen as a bearer of 
munera, understood primarily as fiscal obligations but also as public services.25 For 
Menzinger, this was an expression of citizenship in legal practice, which tied the 
individual to the systems of taxation that the cities were developing. 26 It is indeed 
extremely likely that the glossators' had this view of municipal citizenship not least 
because their enterprise consisted in a largely self-contained exegesis of the Corpus 
iuris, which, as we have seen, portrays local citizenship as the obligation to pay taxes 
and render services to the city. It also seems beyond doubt that the commentators, 
including Bartolus and Baldus, considered munera to be one essential part of 
citizenship.27 This was a reflection not only of the material in their authorities, but 
24 Menzinger, 'Fisco, giurisdizione e cittadinanza', 'Verso la costruzione', and 'Diritti di 
cittadinanza'. 
25 E.g. Menzinger, 'Diritti di cittadinanza', p. 3: 'Dando uno sguardo di insieme al vasto 
materiale questionante duecentesco e inizio-trecentesco sulla cittadinanza, cio che balza in 
primo luogo agli occhi e l'implicita ed errata associazione che compiamo traduce~do 
automaticamente ii termine cittadinanza come insieme di diritti, di privilegi . Dalle 
quaestiones emerge bene quanta ii significato prevalente del termine civis, almeno per i 
primi secoli della storia comunale, rimandi a un insieme di doveri molto piu che di diritti, 
che sono doveri certo militari, ma soprattutto fiscali.' Her characterisation of citizenship is 
shared by Vallerani, 'Diritti di cittadinanza', p. 2: ' .. .la cittadinanza - intesa come insieme 
piu o meno definito di obblighi verso la citta .. .' 
26 Menzinger, 'Verso la costruzione', p. ccvii: 'Le politiche di cittadinanza estensive messe 
in atto dal primo Comune ... ben si riflettono nel pensiero politico-giuridico della stessa fase, 
che vede la cittadinanza come un insieme di doveri prima che di diritti .. .'; 'Diritti di 
cittadinanza', pp. 4-5: 'Le opinioni di questi autori ben testimoniano le origini delle teorie di 
cittadinanza, che ... sono animate da finalita e cercano risposte ad esigenze assai concrete ... 
Stabilire chi e civis e chi non comincia ad interessare in conseguenza dell ' introduzione di 
forme di imposizione su base reale e degli estimi, per capire in sostanza chi debba pagare 
cosa e dove.' 
27 In his consilium 1.62 on naturalisation (edited by Kirshner, 'Civitas sibifaciat civem', p. 
713), Bartolus writes: 'et vere et proprie civis est, quicunque recipitur, ut munera subeat.' 
One of Baldus' consilia on a Venetian grant of citizenship, edited by Kirshner, 'Between 
Nature and Culture', pp. 204-6, demonstrates how participation in munera could be used as a 
test to determine citizen status: 'Et videtur recipi in civem, si longo tempore est tractatus ut 
civis, solvendo munera personalia vel mixta, ut collectas ... ' (p. 206). As Canning argues, 
Baldus de Ubaldis, p. 178: 'Jurists agreed that a citizen was identified by the munera to 
which he was liable ... It was the essence of the citizen's status that he was subditus (subject) 
to the city's authority and the obligations it imposed.' Note, however, that while munera 
were an ordinary part of citizenship, the jurists made certain exceptions: clerics were 
generally exempt from all munera (Canning, Baldus de Ubaldis, p. 134 ff.); women were 
exempt from the liabilities of citizenship in their city of origin if they lived with their 
12 
also of the important position that a citizen's fiscal responsibilities clearly held in 
civic practice,28 which became a proper object of legal study from the second half of 
the thirteenth century. The main problem with Menzinger's narrative is that it 
focuses too much on the duty side of citizenship. In practice, although citizenship did 
involve certain duties, it was, on balance, a desirable status of privilege, which, 
legally speaking, made the citizen a superior person compared to the foreigner by 
bestowing an array of valuable rights29 (references to which are conspicuously absent 
from Menzinger's work): for example, the ability to hold political offices, to buy 
property in city territory, to exercise certain professions, the enjoyment of customs 
exemptions, and a greater degree of protection in the justice system. 30 This is the 
reason why so many foreigners applied to acquire citizenship via statutory grant in 
our period. 31 It would be remarkable, therefore, if civilian theory, after the practical 
tum of the mid-thirteenth century, did not account for the privilege side of 
citizenship in a fundamental way. 
Menzinger' s argument inadvertently links to the now rather dated work of Peter 
Riesenberg,32 who, although aware of the personal social and economic benefits that 
citizenship of an Italian city conferred, pressed Roman law into the service of a 
broader narrative ( concerning both medieval thought and practice) that portrays this 
citizenship essentially as the rebirth of an ancient, active, participatory model, in 
which public duties - fiscal and political - outweighed rights in the private sphere.33 
husbands elsewhere (Kirshner, 'Mulier alibi nupta'). Menzinger, 'Diritti di cittadinanza', pp. 
6-8, also shows that the jurists allowed citizenship to be conferred upon foreigners along 
with a special privilege freeing them from munera. 
28 The significance of citizens' fiscal duties in the legal practice of the cities was already 
recognised by Bizzarri, 'Ricerche sul diritto di cittadinanza' , pp. 22-3, and has recently been 
re-emphasised by Vallerani, 'Evasione fiscale' and 'Logica della documentazione'. Cf. 
Mueller, lmmigrazione e cittadinanza, pp. 42-5, 102, and 110-11. 
29 A point made by Bizzarri, 'Ricerche sul diritto di cittadinanza', p. 16, and more recently 
by Piergiovanni, 'Alcuni consigli', and Kirshner, 'Mulier alibi nupta' (seen. 48 below). 
30 Good summaries of the rights (and duties) of citizenship are provided by Bizzarri, 
'Ricerche sul diritto di cittadinanza', pp. 13-17, and Costa, Civitas, pp. 15-17 (a section 
derived from Bizzarri's article). 
31 A fact that is brought out nicely by Kirshner's work on naturalisation: 'Civitas sibi facial 
civem', pp. 694-6; 'A consilium of Baldus' , pp. 290-4; and 'Between Nature and Culture', 
pp. 185-6. 
32 Riesenberg, 'Civism and Roman Law', 'Citizenship and Equality', 'Citizenship at Law', 
and Citizenship in the Western Tradition . · 
33 Riesenberg, Citizenship in the Western Tradition, p. 118: 'Politically the new man is the 
citizen .. . who responded with some enthusiasm when called upon to expend not only his 
time and property in the service of the community, but also quite possibly his life. This 
13 
While Riesenberg provides some limited evidence from the Roman law tradition that 
seems to support such an interpretation, his work is problematic in that he undertakes 
little real, in-depth analysis of the civilian material.34 His narrative is closely related 
to Walter Ullmann's classic thesis that the revival of Roman law was part of a 
broader proto-Renaissance, humanist moment that oversaw the retreat of the passive 
medieval 'subject' (subditus) and the re-mergence of the ' citizen' (civis) of antiquity, 
understood as active political man.35 
The centrality of a citizen's political capacity has been emphasised more recently by 
Ulrich Meier, whose argument in this respect owes much to Canning's discussion of 
Baldus' thought on citizenship created by statutory grant. 36 He underlines one of 
Baldus' comments in which the latter states that citizens take part in both munera 
and honores, but that those 'who do not participate in public honours are not 
properly called citizens, because they are not treated as citizens in that which is the 
supreme and greatest proof of citizenship' .37 (We should note that the commentators 
no longer understood honores as the vexing, expensive offices described in the 
Corpus iuris, but as the exclusive, desirable, and influential positions of power that 
they were in the government of the Italian cities.38) Meier argues that, in the jurists ' 
view, the citizenry 's political participation was graded: some possessed unrestricted 
political rights and thus possessed the fullest form of citizenship, being, in Baldus ' 
words, citizens ' actively and passively' ; others, while still citizens, were such merely 
in a qualified sense because they participated to a lesser extent or were politically 
passive.39 This is considered to reflect Bartolus' statement in De regimine civitatis 
willingness was based upon an uneasy but growing acceptance of the ancient commitment to 
the purposeful active life.' 
34 E.g. Riesenberg, Citizenship in the Western Tradition, pp. 128-9, 131-2, 136, 144, 183-4. 
Perhaps his most striking examples used to illustrate the idea of citizenship as duty are two 
comments by Baldus, who, at C.4.43 .1, compares citizenship to slavery (p. 129), and, at 
C.8.52.2, describes the citizen's re lationship to his city as analogous to the vassal's 
relationship to his lord (p. 136). Cf. Gilli, 'Comment cesser d 'etre etranger', p. 62. 
35 Ullmann, 'The Rebirth of the Citizen' , p. 25 : 'The law was fetched down from the heavens 
and put into the hands of the citizens .. . ' 
36 Meier, Mensch und Burger, pp. 176-203 (esp. pp. 19 1-3); Canning, Baldus de Ubaldis, eh. 
4, pp. 159-84 (esp. pp. 177-8 1). While Meier's account is indebted to Canning's, the latter 
does not stress the theme of political participation to the same extent. 
37 Baldus to C.6.23.9, quoted and translated by Canning, Baldus de Ubaldis, p. 178. 
38 Kirshner's and Canning's studies on Bartolus' and Baldus ' theories of citizenship acquired 
by grant reveal that nativist trends to exclude newly created citizens from office was a large 
part of the background to their thought. See p. 11, n. 22 above for references. 
39 Meier, Mensch und Burger, p. 193: 'Der "aktive" Biirger, derjenige, der auch Amter 
bekleiden kann, ist der Biirger im eigentlichen Sinne.' Cf. Baldus, cons. 1.460, fol. 147v: 
14 
that in a well-ordered city 'honores and munera are distributed according to rank'.40 
A key part of Meier' s argument is that there were fundamental parallels between 
juristic thought and that of contemporary neo-Aristotelian philosophers, who 
developed Aristotle's conception of the civis as someone who holds political and 
judicial offices in the city: to take medieval conditions into account, they contended 
that a citizen who participates fully is a citizen properly speaking (simpliciter), but 
that a more passive one is only a citizen broadly speaking (secundum quid).41 
It is common amongst historians to characterise the civilian idea of municipal 
citizenship as contractual.42 Kirshner has shown that Bartolus applied the idea of 
contract to citizenship acquired by statutory grant, which the latter called civilitas 
contracta.43 He convincingly argues that, according to Bartolus, if a prospective 
citizen has fulfilled certain statutory requirements for the acquisition of citizenship, 
the city is contractually obligated to confer it and cannot capriciously refuse to do 
so.
44 Kirshner also argues that Bartolus saw citizenship itself - the thing conferred -
' ... quis receptus ad plenissimam civitatem .. . iste consequitur omnes favores civium, et active 
et passive habetur pro cive .. . ' (Canning, Baldus de Ubaldis, p. I 80, n. 62). Canning, too, 
brings out the internal differentiation of the citizenry in Baldus ' thought. Hierarchy and 
different modes of belonging also constitute the central theme of Costa's general account of 
medieval citizenship ( Civitas , eh. I , pp. 1-50); we will consider his argument in eh . 2, pp. 
88-90, and again in eh. 4, pp. 265-6. 
40 Quoted in Meier, Mensch und Burger, p. 195 : ' ... si honores et munera secundum gradus 
debitos distribuuntur, bonum est regimen .. .' 
41 The classic account is Grignaschi, 'La definition du 'civis' dans la scholastique'. Meier, 
Mensch und Burger, part 3, pp. 63-126, contains a more comprehensive treatment. 
42 Riesenberg, Citizenship in the Western Tradition , p. I 52 ('Most simply, it was a contract 
for reciprocal benefits.'); Canning, Baldus de Ubaldis, p. 178 (' the contract of citizenship ' ); 
Menzinger, 'Diritti di cittadinanza' , p. 7 ('Un rapporto contrattuale quasi privatistico'); 
Vallerani, 'Diritti di cittadinanza', p. 8 ('La cittadinanza ... e leggibile anche come un 
rapporto di debito con la respublica' , and hence there is a 'sottofondo contrattuale', a 
' sistema contrattato'). 
43 Kirshner, ' Paolo di Castro', pp. 257-61, 'Civitas sibi facial civem' , pp. 707-9, and 
'Between Nature and Culture', p. 190. On the contractual acquisition of citizenship in legal 
practice, see Bizzarri , 'Ricerche sul diritto di cittadinanza', pp. 7-21. 
44 Kirshner, 'Civitas sibi facial civem', p. 707 (n. 54), cites Bartolus to D.1.1.9, where he 
writes (my transcription), fol. 1 lra: ' Statuto cavetur quod qui venit ad habitandum in tali 
castro ha beat irnmunitatem perpetuo. quidam venerunt. nunc civitas vult revocare statutum ... 
Certe in praeiudicium eorum qui iam venerunt non potest revocari . Secus in his qui nondum 
venerunt. Nam dictum statutum transivit in contractum vel quasi contractum: do ut facias, 
vel facio ut facias, idest, concedo tibi immunitatem ut venias. si aliqui venerunt, ex utraque 
parte perfectus est contractus .. .' Although the case under discussion does not refer to 
citizenship directly, Kirshner reasonably views it as 'an almost identical situation ' given 
fourteenth-century practice. 
15 
While Riesenberg provides some limited evidence from the Roman law tradition that 
seems to support such an interpretation, his work is problematic in that he undertakes 
little real, in-depth analysis of the civilian material.34 His narrative is closely related 
to Walter Ullmann' s classic thesis that the revival of Roman law was part of a 
broader proto-Renaissance, humanist moment that oversaw the retreat of the passive 
medieval 'subject' (subditus) and the re-mergence of the 'citizen' (civis) of antiquity, 
understood as active political man. 35 
The centrality of a citizen' s political capacity has been emphasised more recently by 
Ulrich Meier, whose argument in this respect owes much to Canning's discussion of 
Baldus' thought on citizenship created by statutory grant. 36 He underlines one of 
Baldus' comments in which the latter states that citizens take part in both munera 
and honores, but that those 'who do not participate in public honours are not 
properly called citizens, because they are not treated as citizens in that which is the 
supreme and greatest proof of citizenship ' .37 (We should note that the commentators 
no longer understood honores as the vexing, expensive offices described in the 
Corpus iuris, but as the exclusive, desirable, and influential positions of power that 
they were in the government of the Italian cities.38) Meier argues that, in the jurists ' 
view, the citizenry ' s political participation was graded: some possessed unrestricted 
political rights and thus possessed the fullest form of citizenship, being, in Baldus ' 
words, citizens 'actively and passively' ; others, while still citizens, were such merely 
in a qualified sense because they participated to a lesser extent or were politically 
passive.39 This is considered to reflect Bartolus ' statement in De regimine civitatis 
willingness was based upon an uneasy but growing acceptance of the ancient commitment to 
the purposeful active life.' 
34 E.g. Riesenberg, Citizenship in the Western Tradition, pp. 128-9, 131-2, 136, 144, 183-4. 
Perhaps his most striking examples used to illustrate the idea of citizenship as duty are two 
comments by Baldus, who, at C.4.43 .1, compares citizenship to slavery (p. 129), and, at 
C.8.52.2, describes the citizen's relationship to his city as analogous to the vassal's 
relationship to his lord (p. 136). Cf. Gill i, 'Comment cesser d'etre etranger', p. 62. 35 Ullmann, 'The Rebirth of the Citizen ', p. 25: 'The law was fetched down from the heavens 
and put into the hands of the citizens .. .' 
36 Meier, Mensch und Burger, pp. 176-203 (esp. pp. 191-3); Canning, Baldus de Ubaldis, eh. 
4, pp. 159-84 (esp. pp. 177-81 ). While Meier's account is indebted to Canning's, the latter 
does not stress the theme of political participation to the same extent. 
37 Baldus to C.6.23.9, quoted and translated by Canning, Baldus de Ubaldis, p. 178. 38 Kirshner's and Canning's studies on Bartolus' and Baldus' theories of citizenship acquired 
by grant reveal that nativist trends to exclude newly created citizens from office was a large 
part of the background to their thought. See p. 11, n. 22 above for references. 
39 Meier, Mensch und Burger, p. 193: 'Der "aktive" Burger, derjenige, der auch Amter 
bekleiden kann, ist der Burger im eigentlichen Sinn e.' Cf. Baldus, cons. I.460, fol. 14 7v: 
14 
that in a well-ordered city 'honores and munera are distributed according to rank' .40 
A key part of Meier' s argument is that there were fundamental parallels between 
juristic thought and that of contemporary neo-Aristotelian philosophers, who 
developed Aristotle ' s conception of the civis as someone who holds political and 
judicial offices in the city: to take medieval conditions into account, they contended 
that a citizen who participates fully is a citizen properly speaking (simpliciter), but 
that a more passive one is only a citizen broadly speaking (secundum quid).41 
It is common amongst historians to characterise the civilian idea of municipal 
citizenship as contractual.42 Kirshner has shown that Bartolus applied the idea of 
contract to citizenship acquired by statutory grant, which the latter called civilitas 
contracta.43 He convincingly argues that, according to Bartolus, if a prospective 
citizen has fulfilled certain statutory requirements for the acquisition of citizenship, 
the city is contractually obligated to confer it and cannot capriciously refuse to do 
so.
44 Kirshner also argues that Bartolus saw citizenship itself - the thing conferred -
' .. . quis receptus ad plenissimam civitatem ... iste consequitur omnes favores civium, et active 
et passive habetur pro cive .. .' (Canning, Baldus de Ubaldis , p. 180, n. 62). Canning, too, 
brings out the internal differentiation of the citizenry in Baldus' thought. Hierarchy and 
different modes of belonging also constitute the central theme of Costa 's general account of 
medieval citizenship (Civitas, eh. 1, pp. 1-50); we will consider his argument in eh. 2, pp. 
88-90, and again in eh. 4, pp. 265-6. 
40 Quoted in Meier, Mensch und Burger, p. 195: ' ... si honores et munera secundum gradus 
debitos distribuuntur, bonum est regimen ... ' 
41 The classic account is Grignaschi, ' La definition du 'civis' dans la scholastique'. Meier, 
Mensch und Burger, part 3, pp. 63-126, contains a more comprehensive treatment. 42 Riesenberg, Citizenship in the Western Tradition, p. 152 ('Most simply, it was a contract 
for reciprocal benefits.'); Canning, Baldus de Ubaldis, p. 178 ('the contract of citizenship'); 
Menzinger, ' Diritti di cittadinanza', p. 7 ('Un rapporto contrattuale quasi privatistico'); 
Vallerani, 'Diritti di cittadinanza', p. 8 ('La cittadinanza... e leggibile anche come un 
rapporto di debito con la respublica', and hence there is a ' sottofondo contrattuale', a 
'sistema contrattato'). 
43 Kirshner, ' Paolo di Castro', pp. 257-61, 'Civitas sibi facial civem', pp. 707-9, and 
'Between Nature and Culture', p. 190. On the contractual acquisition of citizenship in legal 
practice, see Bizzarri, 'Ricerche sul diritto di cittadinanza', pp. 7-21. 
44 Kirshner, 'Civitas sibi facial civem', p. 707 (n. 54), cites Bartol us to D.1 .1.9, where he 
writes (my transcription), fol. l lra: 'Statuto cavetur quod qui venit ad habitandum in tali 
castro habeat immunitatem perpetuo. quidam venerunt. nunc civitas vult revocare statutum ... 
Certe in praeiudicium eorum qui iam. venerunt non po test revocari. Secus in his qui nondum 
venerunt. Nam dictum statutum transivit in contractum vel quasi c.ontractum: do ut facias, 
vel facio ut facias, idest, concedo tibi immunitatem ut venias. si aliqui venerunt, ex utraque 
parte perfectus est contractus .. .' Although the case under discussion does not refer to 
citizenship directly, Kirshner reasonably views it as 'an almost identical situation' given 
fourteenth-century practice. 
15 
as a contract, but does not provide compelling evidence to prove this. 45 In other 
words, only the mode of acquiring citizenship seems to be contractual in Bartolus' 
thought: citizenship is the object of contract, not the contract itself. It is true that in 
civilian theory citizenship is generally reciprocal in that it normally confers rights as 
well as duties. One can even find a handful of instances in which the jurists depict 
citizenship as an exchange, and the requirement to fulfil obligations ( especially 
munera) as a condition for the enjoyment of the advantages of citizenship.46 That is 
why historians like to speak of contract. But the jurists themselves very rarely used 
the technical vocabulary of contract to describe the relationship between citizen and 
community, that is, between the rights and duties of citizenship. One noteworthy 
exception is a comment by Baldus: 
I argue that citizenship entails obligations both ways .. . for a mutual bond is contracted 
on both sides; for just as they are to be protected as duty requires, so also they are 
bound to obey and submit to the bond of our citizenship ... Therefore it is not possible 
for its advantages to be accepted and its disadvantages rejected .. .. 47 
Baldus clearly wanted to stress the reciprocal element of citizenship and reached for 
the language of contract to do so. Nevertheless, it appears either that he is speaking 
loosely in this passage and did not actually view citizenship as contractual in a 
45 Kirshner, 'Civitas sibi f aciat civem' , p. 707: 'When the new citizen has performed the 
services demanded of him by the enactment granting him citizenship, his relationship with 
the civitas also takes the form of a contract.' He quotes the following of Bartolus ' comments 
in this regard on p. 708, ns 55-8: 'Secundo no. quod aliqui possunt recipi in civitate ut 
munera nobiscum faciant, et sic receptione contrahitur civilitas '; 'ut receptione propria ex 
contractu habito cum aliquo civitas sibi fac iat civem' (both to D.50. 1.1); 'civilitas acquisita 
per adoptionem vel manumissionem transit ad fil ios ... civilitas quaesita per conventionem 
praedictis aequiparatur, quia utrumque fi t per conventionem, ergo, etcetera' ; 'Sed civilitas 
contracta vel nativitate vel aliis modis ... non potest mutari libera voluntate' (both to 
D.50.1.6); ' quod civilitas contracta ex conventione transeat ad filios, quia quis est obligatus 
civitati, nee potest libera voluntate recedere' (to D.50.1.17). 
46 Two good examples are offered by Menzinger, 'Diritti di cittadinanza', pp. 4 and 6-7. Cf. 
Riesenberg, Citizenship in the Western Tradition , p. 128. 
47 Baldus to C.6.23 .9, quoted and translated by Canning, Baldus de Ubaldis, p. 178, which 
leads Canning to argue for a 'contract of citizenship' . The only other clear example is in 
Riesenberg, 'Citizenship and Equality', p. 432, where he quotes Johannes de Legnano's De 
hello, de represaliis, et de duello: 'quia in civibus ex nativitate perficitur obligatio inter 
ipsum [sic] et civitatem, quae non potest mutari' . According to Riesenberg, the obligation 
under discussion refers to the citizen's duty to undertake munera and the city's duty in return 
to aid the citizen by granting reprisals. Depending on the precise nature of the analogy, 
Baldus' comparison of the citizen to the vassal (see p. 14, n. 34 above) might also be 
adduced as an example in this regard, given that lord and vassal were mutually bound by the 
feudal contract. 
16 
technical sense, or that the idea of contract did not permeate his ( or any other 
civilian's) thought on citizenship to a significant extent. For, as we shall see, 
banishment resulted in a loss of citizenship, yet breach of contract did not at all 
feature in the discourse on the communal ban. 
This discourse also allows us to improve our understanding of the concept of 
municipal citizenship in other ways. While the existing literature describes 
citizenship as a status conferring both rights and duties, it tends either to portray the 
latter as much more significant than the former (citizenship as obligation), or, if it 
does emphasise citizens' prerogatives, to locate their primary significance in the 
exercise of political office. 48 If one reads previous contributions carefully, it is 
possible to observe that the commentators also discussed various rights that civic 
statutes granted citizens beyond the political sphere, but such rights have not 
received the attention they deserve. 49 The theory of banishment reveals that in 
civilian thought, municipal citizenship bestowed rights more so than duties, and that 
rights in private law were at the very heart of the concept. Naturally, this argument 
only applies to the jurists writing after the middle of the thirteenth century, since it 
was only with the practical tum of these years that the ban became the object of 
sustained legal analysis. 
48 Of course, there are exceptions. Kirshner, 'Mulier alibi nupta', highlights that the legal 
discourse recognised certain rights of citizenship unrelated to office-holding (see n. 49 
below), and while he accepts the importance of munera (p. 154), argues that citizenship 
constituted 'sia in teoria che in pratica ... un privilegio' (p. 171). Elsewhere, Kirshner 
criticises the _tendency of historians to reduce citizenship in this context 'alla detenzione di 
un ufficio pubblico e all'esecuzione di pubbliche funzioni' ('Cittadinanza come genere', p. 
3 ). Piergiovanni, 'Alcuni consigli ', shows that the cons ilia of the Genovese jurist 
Bartolomeo Bosco (d. ea. 1437) portray Genovese citizenship as a privileged status by 
placing citizens in a superior position to foreigners in civil litigation and inheritance law; this 
was a reflection of citizen status in Genoa's statutes. 
49 E.g. the right to request reprisals from the city (Riesenberg, 'Citizenship and Equality', pp. 
431-2; Kirshner, 'Civitas sibi faciat civem', pp. 708-10; Canning, Baldus de Ubaldis , pp. 
183-4); better protection in criminal law than foreigners (Riesenberg, 'Civism and Roman 
Law, p. 243; Kirshner, 'Mulier alibi nupta', p. 162; Menzinger, 'Diritti di cittadinanza', p. 
4); advantages in civil litigation and inheritance law (Piergiovanni, 'Alcuni consigli', pp. 6-
10); the ability to buy real-estate in city territory (Kirshner, 'Mulier alibi nupta', p. 162; 
Gilli, 'Comment cesser d'etre etranger', p. 71); exemption from certain tariffs (Kirshner, 
'Between Nature and Culture'). 
17 
3. The Imperial and the Communal Ban in Legal Practice 
Much of the older ( especially German) historiography considered banishment, in the 
guise of outlawry (Friedlosigkeit), to be an ancient, pan-'Germanic' institution 
present throughout the early Middle Ages. 50 In the late 1950s, Hinrich Siuts 
disproved that narrative and demonstrated that in Germany a royal ban, at least of the 
type common in the central and later Middle Ages, did not exist before the tenth 
century. 51 Parts of his conclusion rest on Julius Ficker's meticulous 1868 
Forschungen zur Reichs- und Rechtsgeschichte ltaliens, which still remains the most 
comprehensive compendium of Italian legal documents on this topic. 52 According to 
Ficker, the royal ban migrated with the emperors from Germany to Italy, where there 
is no evidence for this or similar forms of outlawry before the end of the eleventh 
century. 53 The predominant catalyst for the ban's subsequent advance seems to have 
been the growing conflict between the Hohenstaufen emperors and the Italian 
cities. 54 While the cities, for their part, already exiled individuals in the twelfth 
century, concrete evidence in the statute books for the communal ban as a legal 
mechanism involving outlawry first appears in the early thirteenth century. 55 In 
Ficker's view, this chronology suggests that the cities copied imperial practice in this 
regard. 56 We should note that in the Italian sources before the end of the eleventh 
century, the term bannum, rather than denoting outlawry as it increasingly. did 
thereafter, referred to the order of a public authority, the threat of a fine, or the fine 
itself. 57 The term retained these senses in later centuries. 58 
5
° Cf. Pertile, Storial de! diritto italiano V, pp. 311-43; Calisse, Storia de/ diritto penale 
italiano, pp. I 05-121 and 260-66; Eichman, Acht und Bann; Poetsch, Die Reichsacht; 
Planitz, Deutsche Rechtsgeschichte. 
51 Siuts, Bann und Acht. 
52 Milani, L 'esclusione dal comune, pp. 35-9, 115-17, 425-9, also has useful, if quite brief, 
sections on imperial banishment. 
53 Ficker, Forschungen I, pp. 73-92. 
54 Milani, L 'esclusione dal comune, p. 35 ff. 
55 Ficker, Forschungen I, pp. 92-6 and 133-4. This chronological development can also be 
observed in Milani's more recent presentation of the evidence: L 'esclusione dal comune, chs 
2-4, pp. 27-144. In his subsequent 'Giuristi, giudici e fuoriusciti', pp. 598-9, Milani suggests 
that the communal ban entailed loss of legal protection for person and property already in the 
twelfth century. His sources from that period, however, only speak of expulsion from the 
city, not of outlawry. 
56 Ficker, Forschungen I, pp. 224-5. 
57 Ibid., pp. 62-73 . In Lombard law, the term bannum could also refer to the temporary 
seizure of a contumacious defendant ' s estate for one year, after which it was confiscated if 
he remained contumacious (pp. 32-5). 
18 
Generally speaking, during our period the imperial ban in Italy can be divided into 
two types: the temporary or coercive procedural ban, and the permanent punitive 
ban. 59 Both were sanctions employed by the emperor ( or his vicars) against 
individuals, groups, and even whole cities, that were not under the power of the 
imperial authorities and remained contumacious, that is, persistently disobedient to 
an official order, such as a summons to appear at the imperial court to stand trial for 
an alleged crime. Indeed, the emperors were very careful in their sentences of 
banishment to underline that a lawful summons had been issued; it seems that 
defendants had to be given at least the chance to present a defence before banishment 
proceedings could be concluded in their absence. The only exceptions come from 
Henry VII's reign. In those sentences passed after his imperial coronation, one can 
observe the conviction that a summons was not necessary in cases of notorious 
crimes, justified by the Roman-law concept that the emperor was legibus solutus 
(above human law).60 Once under the ban, the bannitus was subject to a range of 
penalties, prominent amongst which was the liability to be harmed in person and 
property with legal impunity by anyone. The purpose of the temporary ban was to 
penalise disobedience and to force the contumacious party to submit to the emperor's 
demands. The bannitus remained in a position to effect the cancellation of his ban by 
returning to obedience, which usually involved the payment of a monetary fine. The 
permanent ban, on the other hand, offered no such possibility since it principally 
acted as a punishment for a crime; hence it could only be lifted by the emperor's 
judicial pardon. And whereas the temporary ban was used in relation to a variety of 
civil and criminal offences, the permanent ban was always associated with acts of 
treason and rebellion against the emperor; it was a highly political measure.61 
We will now tum to the ban as it can be found in the statutory law and legal practice 
of the Italian cities during the thirteenth and fourteenth centuries. 62 The term bannum 
58 Ibid., pp. 94-5 and 164. 
59 Ficker, ibid., eh. 7, pp. 147-70 ('Losbarer Reichsbann') and eh. 8, pp. 170-218 
('Bestandiger Reichsbann') covers these in detail. The following summary is derived from 
Picker's extremely lucid treatment. 
6
° Ficker, ibid., p. 190, recalls Henry's banishment of the city of Padua as an instance of a 
sentence in which a summons had been completely omitted. In eh. 3, p. 128 ff., we will 
discuss in depth the controversy caused by Henry's banishment of Robert of Naples. 
61 Note that the conceptual division between these two types of ban. was not always clear in 
practice. For instance, a coercive procedural ban could merge into a permanent punitive one 
if continued disobedience was deemed a serious enough infraction of the emperor's majesty . 
62 The following is a synthesis of the studies devoted to this topic. No historian has yet 
surpassed the comprehensiveness of Ficker, Forschungen I, pp. 92-147. The most detailed 
subsequent treatments are: Mooney, 'The Legal Ban' ; Cavalca, II bando (which focuses both 
19 
possessed a variety of meanings, and could refer to the order of a public authority or 
a monetary fine. It was used in relation to a number of stages in the procedure of 
summoning and sentencing a defendant in civil as well as criminal matters. In that 
context, bannum regularly denoted a sanction imposed on a defendant, the severity of 
which could differ depending on the nature of his offence (or alleged offence). We 
will focus here exclusively on the criminal ban, the bannum pro maleficio, in its 
harshest form, which found application in cases involving major crimes such as 
murder or treason. As we shall see in chapter 4, this type of ban overwhelmingly 
constituted the subject of the civilian discourse on banishment. Communal banniti 
subject to it were in a similar position to imperial banniti in the sense that they were 
free to be injured in person and property with legal impunity (in the territory of the 
city from which they were banished). These banniti also suffered further penalties, 
including, for instance, the confiscation of the property they owned at the time of 
their banishment. In contrast, many defendants under bans for minor crimes or civil 
offences could not be killed with impunity. 
The communal criminal ban (again, always understood in its most severe form) was 
typically employed in ways that resembled the functions of the imperial ban. 
Generally speaking, it can be divided into two versions: one that is temporary, 
coercive, and procedural, and another that is permanent and punitive. In chapter 4, 
we will see that civilian theory labelled the former bannum pro contumacia and the 
latter bannum pro delicto. 63 Both were used against a contumacious defendant, 
meaning a defendant who had been summoned to court to answer for a crime but 
persistently refused to appear and face justice. 64 After a certain deadline had passed, 
the defendant was taken to have confessed (pro confesso) by virtue of his contumacy, 
sentenced for the alleged crime, and placed under the ban. If he was subject to the 
bannum pro contumacia, the ban acted as a penalty in lieu of the ordinary one that 
the bannitus would have suffered had he been present and found guilty. Its aim was 
not only to ensure that the bannitus was adequately punished (for his disobedience as 
on legal practice as well as civilian theory); Pazzaglini, The Criminal Ban; Milani, 
L 'esclusione dal comune. Also useful for the criminal ban are: Montorsi , 'Plebescitae 
Bononiae'; Langeli and Corbucci, 'I "libri dei banditi"' ; Blanshei, 'Crime and Law 
Enforcement', and Politics and Justice; Vallerani, II sistema giudizario, esp. pp. 28-9, and 
'L'amministrazione della giustizia'; Milani, ' Prime note '; Mazzoni, 'Note sulla confisca'. 
For the debt ban, see: Planitz, 'Der Schuldbann'; Gaulin, 'Les registres'; Mehu, 'Structure et 
utilisation'. 
63 Cf. eh. 4, pp. 213-14 and 241-3. 
64 According to Ficker, Forschungen I, pp. 97, 104, 107-8, 110, and 113, some statutes 
indicate that the ban could also be imposed on a present defendant (who was not 
contumacious). 
20 
well as the alleged crime), but also to coerce the latter into submitting once more to 
the city's jurisdiction, and to ensure that he remained beyond city territory if he 
chose not to do so. The bannitus remained in a position to take himself out of the 
ban, ordinarily by paying a monetary fine to the city and by officially reconciling 
with the injured party (or their family). If an unrepentant bannitus was captured, he 
was unable to contest his conviction and immediately subject to punishment, which 
not uncommonly entailed the implementation of the death sentence. The bannum pro 
delicto was different in that it constituted the ordinary punishment for the alleged 
crime and could not be cancelled except by judicial pardon. It resulted from crimes 
that were considered to be particularly detrimental to the city and that were usually 
connected to treason and rebellion of some sort. Hence, banniti under the permanent 
ban were frequently called rebels (rebelles), enemies (inimici), and traitors 
(proditores) of their city. Those who were expelled due to the conflict between civic 
factions were normally cast into the role of rebels and therefore suffered this type of 
banishment. Given, however, that the coercive procedural ban was the routine way in 
which the cities proceeded against contumacious defendants in a range of ordinary, if 
serious, criminal matters, and given also the high rates of contumacy in our period, it 
seems probable that the average bannitus pro maleficio was not a political exile -
save perhaps during the years that immediately followed one of the great expulsions, 
especially those of the later thirteenth and early fourteenth centuries.65 It will become 
clear in chapter 4 that further evidence for this view is provided by the way the 
jurists discussed the criminal ban.66 
As we saw above, a lawful summons was a necessary part of the procedure resulting 
in a sentence of imperial banishment. The city statutes, too, make it clear that a 
summons was required before absent defendants could be legitimately banished; in 
65 As Pazzaglini, The Criminal Ban, p. 99, has stressed: 'The typical bannitus pro maleficio 
was not an .aristocratic political refugee, Guelf or Ghibelline, but a criminal fugitive, 
probably without a set political ideology, especially so when under a ban for rape, sodomy, 
gambling, adultery, abortion, forgery, blasphemy, and theft at night.' This point has also 
been made by Mooney, 'The Legal Ban', pp. 84-8, who provides some valuable statistics to 
illustrate the high rate of contumacy in proceedings for crimes that were not of a political 
nature. His evidence is conviction rates in Florence for crimes against persons and property (ea. 40% of total crime) during two three-year periods (1352-5 and 1380-3). Since these 
crimes could result in the criminal ban if the defendant remained contumacious, the 
proportion of cases that ended in convictions in contumacy is indicative of the number of 
criminal bans issued: 'The rate of contumacy is consistently high, often as high as 75% and 
rarely lower than 50%. Murder has a remarkable rate of conviction in contumacy - 86% in 
the period 1352-5, and 87% in the period 1380-1383' (p. 87). 
66 Ch. 4, esp. pp. 213-14. 
21 
other words, the latter had to be granted at least the opportunity to make their case, 
even when abbreviated or summary procedure was used. Interesting questions arise 
over exceptional statutory provisions specifying that the perpetrator of certain crimes 
would be placed under the ban as a punishment upon the very commission of the 
criminal act, i.e. ipso iure or ipso facto. As we would expect from crimes leading to 
the bannum pro delicto, these were deemed to constitute a revolt against the city's 
political order or a fundamental threat to civic peace. 67 It is not unreasonable to infer 
that it was the intent of the law to withdraw legal protection from the perpetrator 
immediately without any legal proceedings (and thus without a summons);68 indeed, 
one Florentine statute states this very explicitly.69 The broader question is to what 
extent those subject to the political ban, whether or not it was issued ipso iure, were 
granted the minimal procedural right to a summons and a defence. According to 
Desiderio Cavalca, the expulsions that resulted from the conflict between civic 
factions did not follow legal norms in general, though he provides very little concrete 
evidence in this regard.70 In contrast, Sarah Blanshei's recent detailed study of the 
Bolognese sources demonstrates that city magistrates were careful to summon 
defendants, and that they did this even in politically sensitive cases during the later 
thirteenth and early fourteenth centuries. Of course there were exceptions, and 
sometimes the brevity of the interval in which the summoned person had to appear 
67 Mooney, 'The Legal Ban ', pp. 122-30, contains a discussion of ipso iure banishment ih the 
statutes of Florence. According to Mooney, there were at least four crimes in Florence that 
led to the ipso iure acquisition of bannitus status: the proposal of an official to submit the 
city to a lord; raising a flag of insurrection in the city; limiting the freedom of a Florentine 
citizen on the basis of an authority other than that of the city itself; killing someone in the 
proximity of the central public buildings or squares. 
68 That is how Mooney, ibid., pp. 122-3, and Cavalca, 11 bando, pp. 119-20, interpret the 
relevant statutes of Florence and Bologna respectively. 
69 This statute can be found in the Statuta Populi et Communis Florentiae of 1415. It is 
quoted and translated by Mooney, 'The Legal Ban', p. 128: ' .. . anyone who will commit a 
homicide, or procures it against a person who could not be killed with impunity according to 
the statutes, or who was prepared to commit [such a crime], or had it done for him in any of 
the aforementioned places [administrative center and markets] can with impunity be 
offended after the homicide has been committed and before any condemnation.' 
7° Cavalca, // bando, pp. 103, 107-8, and 172. On p. 172 he cites a quaestio in Nello da San 
Gimignano's De bannitis (1.3, q. 4, fol. 11 6va-b) to support the argument that the formalities 
surrounding summonses were not extended to political enemies. Cavalca points to the pro 
section of the quaestio, which puts forward the argument, with reference to one of Bai1olus' 
consilia, that a summons is not required in notorious cases. In his solutio, however, Nello 
concludes that a notorious criminal has to be summoned at least once before a judge can 
banish him. As we shall see in eh. 4 (pp. 213-221 ), all the jurists, including Bartolus, agreed 
that summonses were indispensable in communal banishment proceedings (although there 
was a degree of ambiguity regarding ipso iure sentencing). 
22 
turned the right to a summons into a rather hollow formal ity.7 1 As we shall see in 
chapter 4, the civilian discourse on the procedure ?f communal banishment supports 
what we can learn from Blanshei's work. It not only indicates that summonses were 
conventional in political cases, but also that the legal requirement for a summons was 
not normally circumvented with ipso iure provisions. 72 
4. Rights, Natural Rights, Natural Law 
In historical jurisprudence it is usual to distinguish between subjective and objective 
right. Subjective right refers to what is normally called a right in English, i.e. the 
legal entitlement of an individual (such as a claim, power, or liberty), which is 
opposed to a duty (an obligation or liability). Objective right refers to that which is 
right - a situation corresponding to the demands of justice - and hence also law, 
which describes that which is right. During the past seventy years there developed a 
fairly extensive and complex debate amongst historians to what extent, and precisely 
in which ways, the Latin term ius (pl. iura) in classical and medieval Roman law, as 
well as in medieval philosophy, theology, and canon law, possessed the senses just 
described, most often in connection to arguments about the origins and progress of 
the theory of natural rights. The size and complexity of the existing literature 
precludes a full review here. 73 I merely wish to outline a number of broader points 
that can be drawn from the current state of the field. 
Michel Villey's argument, that ius in Roman law - especially in classical Roman law 
- did not denote subjective right, 74 gained no ground amongst legal historians. 75 
7 1 Blanshei , Politics and Justice, eh. 5, pp. 313-497. 
72 Ch. 4, pp. 213-21. 
73 In the words of Oakley, in Natural Law, Laws of Nature, and Natural Rights, a relatively 
recent historiographical survey, at p. 88: ' ... the smoke of battle still hangs low over what has 
become of recent years a rather crowded and confusing field on which competing hypothesis 
still maneuver warily for advantage ... In this conflicted realm, then, complexity reigns now 
as king. ' 
74 Of particular importance are Villey ' s 'L'idee du droit subjectif, '.Du sens de !'expression 
jus in re en droit romain classique', 'Le "jus in re" du droit romain classique au droit 
moderne ' , ' Suum jus cuique tribuens ', 'Les origines de la notion du droit subjectif' , 'La 
genese du droit subjectif' , La formation de la pensee juridique moderne. Probably the best 
review in English of Villey 's work is by Tierney, The Idea of Natural Rights, eh. 1, pp. 13-
42. 
23 
While most Romanists never bothered to engage with it, there are subsequent studies 
which convincingly demonstrate that ius in classical Roman law had a range of 
subjective and objective senses. Although subjective ius was nowhere near as 
prominent as it is in modem legal discourse, 76 it was nonetheless common for ius to 
refer to what we mean by a right today.77 It is also clear, as one would expect, that 
medieval Roman law, along with the closely related contemporary discipline of 
canon law, took on board all the ancient senses of ius. 78 One noteworthy 
development was the increased willingness of medieval jurists to speak in subjective 
terms, to describe legal situations with reference to a person's ius or iura when their 
classical counterparts did not. 79 
75 For a brief discussion of Villey's impact, or rather lack thereof, on Roman-law 
scholarship, see Garnsey, Thinking about Property, pp. 184-5, where Gamsey's quote of an 
'anonymous distinguished German Romanist' gives us a good idea of why Villey's claims 
were often ignored in this field: 'In my opinion his [Villey's] thesis is so absurd that it hardly 
seems worth taking issue with it.' 
76 Coing, 'Zur Geschichte des Begriffs "subjektives Recht"', pp. 9-12. 
77 The classic critique is by Pugliese, '"Res corporales", "res incorporales"' . Pugliese's 
argument does not so much centre on the meaning of the term ius as on a number of 
technical criticisms of Villey's ideas and on the claim that, even if ius did not denote 
subjective right, the Romans certainly had the concept, although expressed in other words 
and evident in legal institutions such as the actio or dominium. See also his Actio e diritto 
subiettivo. More recent, important critiques based on careful analysis of the term ius jn the 
legal sources are Kaser, 'Zurn "Ius"-Begriff der Romer' , and Donahue, '!us in the Subjective 
Sense'. According to Donahue, subjective ius can be defined using the scheme of Hohfeld 
(Fundamental Legal Conceptions). Kaser argues (p. 63): 'Die Romer... gebrauchen ius bald 
in objektiver Bedeutung, um das "Recht" als Rechtsordnung und ihre Elemente, die 
Einrichtungen und Normen, zu bezeichnen, bald in subjektiver, mit der sie die 
"Berechtigung" des Einzelnen erfassen ... ' Cf. Garnsey, Thinking about Property, pp. 179-95. 
78 This is largely uncontroversial. For a good summary of the state of research concerning 
medieval Roman law (by 1992), see Norr, 'Zur Frage des subjektiven Rechts' . The only 
scholar in recent years to have cast some doubt on the appropriateness of speaking of 
subjective right in medieval juristic thought, is Kriechbaum, Actio, !us, und Dominium. In 
my mind, her claims are ill-founded and have been successfully rebutted by Tierney's review 
article of her book, 'Medieval Rights and Powers'. Even Villey, in 'Le "jus in re" du droit 
romain classique au droit modeme', which represents a less stringent, earlier version of his 
thesis, acknowledges many instances of ius being used in a subjective sense by the 
glossators, although, in his mind, most of these iura were not rights in the proper modem 
sense. For medieval canon law, see: Tierney, Rights, Laws, and Infallibility, and The Idea of 
Natural Rights, esp. eh. 2 in both works; Reid, 'The Canonistic Contribution to the Western 
Rights Tradition', and 'Thirteenth-Century Canon Law and Rights'. 
79 Coing, 'Zur Geschichte des Begriffs "subjektives Recht" ', pp. 13-14. For medieval 
developments see also: Tuck, Natural Rights Theories, pp. 13-17; Brett, Liberty, Right, and 
Nature, pp. 20-2; Garnsey, Thinking about Property, p. 195-203 . 
24 
In Roman law, classical and medieval, ius is a difficult word to translate. While its 
meanings certainly include both (subjective) right and (objective) law, it is not 
uncommon for the sense to be ambiguous. Different senses may shade into each 
other, and sometimes ius can even be subjective and objective at the same time, 
referring to a legal provision, perhaps, while also implying the rights an individual 
derives from that legal provision. 80 According to Charles Donahue, the real value of 
Villey's work lay in emphasising the meaning of ius as 'an objective situation that 
was right' - id quod iustum/iuris est - after its neglect in modem Romanist 
scholarship.81 In Donahue's view, this is how we should understand certain rare 
instances of ius in classical Roman law, when it is the attribute of a person but refers 
to a liability or obligation: he argues that ius should be interpreted as that person's 
objective due, not his right. For example, the ius liberti, the ius of the freedman, does 
not refer to the freedman's entitlement but to his obligation towards his patron.82 We 
should note that in these cases, ius is an immediate, not a mediate duty. In other 
words, it is not a right in the modem sense that could also be interpreted as a duty ( a 
duty-right), such as, for example, the right of a public magistrate to exercise the 
lawful power granted to him, which in some cases, by moral and legal evaluation, he 
is duty-bound to exercise. The ius we are considering simply is a duty, not a legal 
entitlement that provides the means for the performance of a duty. 83 Donahue is of 
course correct when he argues that ius as duty in this sense does not fit the modem 
definition of subjective right. On the other hand, in the language of the sources, these 
iura are described as inhering in individuals just as much as iura that are 
entitlements. I therefore suggest that rather than treating them as instances of 
objective usage - id quod iustum/iuris est - as Donahue does, we should categorise 
them as subjective rights and, furthermore, translate them as 'rights' in English. 
Although this is very counterintuitive, it is a much better reflection of the subjectivity 
of these iura in the sources, and should not cause confusion as long as the modem 
reader is made aware that in legal Latin subjective ius can both refer to an 
individual's entitlement as well as his immediate obligation. We will discuss this 
point further with reference to the medieval material in chapters 2 and 3. 84 
80 A point raised by Kaser in respect of the classical material: 'Zurn "lus"-Begriff der 
Romer', p. 69. 
81 Donahue, '!us in the Subjective Sense', pp. 507-8. 
82 Ibid., p. 520. Other examples can be found on pp. 512, 518, and 527-8. 
83 For duty-rights, although not with reference to the civilian tradition, see: Tierney, The Idea 
of Natural Rights, esp. eh. 3, pp. 78-89; Brett, Liberty, Right, and Nature, p. 2; J. Kilkullen, 
'Medieval and Modem Concepts of Rights: How do they differ?', in Makinen (ed.), The 
Nature of Rights, pp. 31-62, at p. 51. 
84 Ch. 2, pp. 46-7; eh. 3, pp. 165-71. 
25 
When it comes to natural rights, their origins and development, generally speaking, 
one can find two narratives in the modem historiography about the contribution of 
Roman law. In the first, dominant narrative, the language of subjective natural right 
in works of canon law dating to the second half of the twelfth century is viewed as 
the ultimate point of origin. It was Brian Tierney's achievement to highlight the 
existence of subjective natural right in canonistic doctrine from this era as well as the 
thirteenth century.85 Regarding the thirteenth century and beyond, however, there is a 
permanent shift of focus in the historical literature to the scholastic philosophers and 
theologians, whose writings constitute the subject of the vast bulk of natural rights 
scholarship on the medieval period; pride of place is often granted to the Franciscan 
poverty dispute, and in that context, to the thought of William of Ockham ( d. 134 7). 
While it is acknowledged that the ideas of the canonists were precursors to those of 
the theologians and philosophers, and influenced the latter in some important 
respects, there is a common tendency to view the canonistic contribution as a 
relatively minor part of a tradition whose true heart is the theological-philosophical 
discourse. According to this narrative, Roman law, classical and medieval, had no 
direct part to play in the creation of natural rights theory: the civilian tradition, it is 
claimed, knew of subjective right, but not of subjective natural right. 86 The strongest 
statements to this effect have been made by Peter Garnsey, who describes the idea of 
natural rights as 'post-Roman and un-Roman', and furthermore writes: 'In the 
medieval period Glossators and Commentators were already betraying in their work 
a degree of exposure to canon law, but their treatment of property and rights was 
little affected. Yet it was from the canonistic tradition that the idea of natural rights 
emerged ... ' 87 These are by no means unreasonable conclusions to reach given the 
current state of the field. 
85 Tierney's two seminal articles, 'Origins of Natural Rights Language', and '!us and 
Metonymy in Rufinus', were both reprinted in the collection of his essays, Rights, Laws and 
Jrifallibility, as chs 2 and 3 respectively, and then combined and slightly developed in The 
Idea of Natural Rights, as eh. 2, pp. 43-77. Note that Tierney relies to a considerable extent 
on the canonistic material concerning natural right unearthed by Weigand, Die 
Naturrechtslehre, part 2 (Kanonisches Recht), pp. 121-443. 
86 Cf. Tuck, Natural Rights Theories, eh. I, pp. 5-31; Tierney, The Idea of Natural Rights, 
chs 3-8, pp. 78-203; Brett, Liberty, Right and Nature, chs 1-3, pp. 10-122; Korkman and 
Makinen (eds), Transformations; Garnsey, Thinking about Property, chs 7-8, pp. 177-232; 
Makinen (ed.), The Nature of Rights. For historiographical syntheses and surveys (in 
chronological order), see: Coleman, 'Medieval Discussions of Human Rights'; Pennington, 
'The History of Rights'; Oakley, Natural Law, Laws of Nature, Natural Rights, eh. 4, pp. 87-
109; Makinen, 'The Evolution of Natural Rights', and 'Justice, Law, Power, and Agency: 
Defining the Nature of Right(s)', in Makinen (ed.), The Nature of Rights, pp. 7-28; Kilcullen, 
'Medieval Theories of Natural Rights' . 
87 Garnsey, Thinking about Property, pp. 182 and 203 respectively. 
26 
The second narrative about Roman law and natural rights seems to be diametrically 
opposed to the first. In some parts of the literature one can find the claim that 
medieval canon as well as civil lawyers discussed a broad range of natural rights, 
that they defined, for example, 'rights of property, self-defence, non-Christians, 
marriage and procedure as being rooted in natural and inalienable, not positive 
law'. 88 Kenneth Pennington, whose book The Prince and the Law plays an important 
role in this narrative, feels comfortable speaking of rights and natural rights even 
though, as he openly states, 'the jurists did not call these rights natural' - nor rights, 
one should add, at least in the sources he presents. Nevertheless, Pennington argues, 
the jurists, including the civilians, 'developed doctrines that supported a theory of 
subjective, immutable, individual rights' . 89 It appears that the basis of the second 
narrative is the medieval Romanist discourse on natural law, which constitutes the 
focus of what is, for the most part, an older Italian and German historiography; in 
many ways, Pennington's book should be viewed as a more recent part of, and 
contribution to, the latter.90 With the exception of Pennington, however, historians 
who worked in this field were not specifically concerned with (subjective) natural 
rights theory and did not normally present their findings in terms of natural rights. 
This is a reflection of their sources, which speak of ius naturale only in an objective 
sense. 
The civilian natural law discourse can be summarised as follows. 91 Certain 
transcendent legal norms existed independently of, and beyond, positive civil law 
(ius civile). At the highest level was a divinised natural law, commonly associated 
with the precepts of natural reason and equity, but also with God's law as revealed in 
Scripture. In a subsidiary sense, natural law could be understood as the law of 
nations (ius gentium), which, in the words of Justinian's Institutes, 'natural reason 
has established among all mankind and which is equally observed among all 
88 Makinen, 'The Evolution of Natural Rights', p. 107. On one occasion, in 'Origins of 
Natural Rights Language', pp. 638-9, Tierney makes a similarly broad statement regarding 
the civil and canon lawyers. 
89 Pennington, The Prince and the Law, p. 124. 
90 Rota, 'II diritto naturale' ; Nicolini, La proprieta; Cortese, La norma giuridica; Weigand, 
Die Naturrechtslehre. 
91 My summary is largely based on: Nicolini, La proprieta; Cortese, La norma giuridica, 
esp. vol. 1, chs 1-3, pp. 1-141; Weigand, Die Naturrechtslehre, part 1 (Romisches Recht), pp. 
8-121; Pennington, The Prince and the Law, esp. chs 1-4, pp. 1-1 '64. Woolf, Barto/us of 
Sassoferrato, pp. 45-53, and Canning, Baldus de Ubaldis, pp. 71-92 and 154-8, the classic 
studies of Bartolus' and Baldus ' political thought, also contain good discussions of natural 
law and its relationship to civil law. 
27 
peoples'. 92 The jurists generally thought of the ius gentium as the source of property, 
contracts, slavery, and political authority. 93 Beneath the law of nations was the 
Roman civil law, the universal ius commune, continually valid, at least de iure, in the 
lands of the eternal Roman empire. While the jurists considered this empire to be 
coterminous with western Christendom, their theory allowed for the many political 
entities within it, such as the Italian cities, that de facto were no longer ruled by the 
emperor. 94 Since higher natural norms were fundamentally immutable, the Roman 
civil law and its supreme authority, the emperor (or prince, i.e. princeps), could not 
abrogate them, although they could with good cause (iusta causa) derogate from 
them. Pennington has brought out the casuistry in civilian thought concerning when 
and to what extent higher norms could be suspended.95 If it was reasonable, just, 
directed towards public utility, while not violating notions of decency and morality, 
cause had its place. For example, the lawyers commonly allowed the prince to 
confiscate property ex causa. On the other hand, as Cinus de Pistoia (d. 1336) made 
clear, he could not ever, even with cause, 'make a rescript that I should marry my 
mother'. 96 
As should have become increasingly clear by now, the difference between the two 
narratives about Roman law and natural rights stems from a difference in approach. 
The first narrative is the product of historians whose arguments are based on the 
language of subjective and objective ius in the sources, whereas other historians, 
such as Pennington, speak of rights even when subjective ius is not used by the 
jurists themselves. These historians feel comfortable speaking of rights because the 
civilians discussed certain juridical situations in the natural law discourse that result 
in what we would call a right today; rights are seen as being implicit. For example, it 
was argued that the prince could not lawfully deprive an individual of ownership 
(dominium) of a thing without just cause since dominium is derived from the law of 
92 Inst. 1.2, 1: ' ... quod vero natural is ratio inter omnes homines constituit, id apud omnes 
populos peraeque custoditur vocaturque ius gentium .. .' The translation is from Scott, The 
Civil Law. 
93 They took their lead from D.1. 1.5: 'Ex hoe iure gentium introducta bella, discretae gentes, 
regna condita, dominia distincta, agris termini positi, aedificia collocata, commercium, 
emptiones venditiones, locationes conductiones, obligationes institutae: exceptis quibusdam 
quae iure civili introductae sunt.' As this law indicates, although some contracts were 
introduced by the law of nations, others were introduced by the civil law of Rome. 
94 Cf. Woolf, Barto/us of Sasso/errata; Canning, Baldus de Ubaldis. 
95 Pennington, The Prince and the Law, pp. 130-2. 
96 Cinus de Pistoia to C.1.19(22). 7 ( quoted in Pennington, The Prince and the Law, p. 131 ): 
' ... aut rescribit super his que non possunt procedere causa subsistente, ut rescribit quod ego 
contraham cum matre mea. I stud non valet...' 
28 
nations or natural law. It would not be wholly unreasonable, if somewhat 
anachronistic, for us to speak of a natural right to property in this context. Once 
again, however, the civilians themselves, in the natural law discourse subject to 
previous historical analysis, only refer to ius naturale and ius gentium objectively, 
and consider certain aspects of the legal system established by higher norms ( e.g. 
property, contracts, and slavery) as legal institutions, not as rights. One could muster 
further support for narrative two by pointing to the fact that the jurists, in different 
contexts, where they are not discussing natural law, describe some natural legal 
institutions using a language of subjective ius. Bartolus' definition of dominium, as a 
person' s ius of disposing over a corporeal object, is well known. 97 But what 
historians have not demonstrated thus far is that all the various legal institutions 
derived from natural law could be framed in terms of an individual's ius; nor have 
they shown that any one of these institutions was explicitly associated with an 
individual's ius naturale/ius gentium. At best, therefore, the evidence presented by 
the existing historiography suggests that the medieval civilians were on the cusp of 
an articulated theory of natural rights, but that the final steps were missing. 
Since both natural law and subjective ius were established features of the civilian 
tradition, classical and medieval, it would have been far from unreasonable for a 
jurist (or legislator) to use the phrase ius naturale subjectively in order to describe a 
right derived from a higher norm. It is probably this intuition that led Donahue to re-
examine all the instances of ius naturale in Justinian's Digest to check for a 
subjective usage.98 His returns are meagre. According to Donahue, out of twenty-five 
occurrences, one is straightforwardly subjective, 99 while four others are more 
97 Bartolus to D.41.2.1 7,1 (quoted, for instance, by Nicolini, La proprieta, p. 47, and Brett, 
Liberty, Right, and Nature, p. 22): 'dominium est ius de re corporali perfecte disponendi nisi 
lege prohibeatur'. Accursius' gloss to Inst.4.6 pr. ( quoted by Coing, 'Zur Geschichte des 
Begriffs "subJektives Recht'", p. 13 ), where he describes actions and contractual claims as 
subjective iura, is also well known: 'Nota quod actio est ius, quo persequimur, sed obligatio 
est ius, propter quod persequimur.' As mentioned above, according to the civilians, certain 
contracts were derived from the ius gentium, and as we shall see in eh. 2 (pp. 72-5 and 80-6), 
they also viewed certain actions as rooted in natural law. 98 Donahue, '!us in the Subjective Sense', pp. 520-1 and 526-8. 
99 D.38.6.4: 'Liberi et capite minuti per edictum praetoris ad bonorum possesstonem 
vocantur parentium, nisi si adoptivi fuerint: hi enirn et liberorum nomen amittunt post 
emancipationem. sed si naturales emancipati et adoptati iterum emancipati sint, habent ius 
naturale liberorum.' Cf. Watson (ed.), The Digest of Justinian, vol. 3, p. 345: 'Children who 
have suffered a change of civil status are also called to bonorum possessio of their parents' 
property by the praetor's Edict, unless they have been adopted; for these last also lose the 
title of children after emancipation. But if natural children have been emancipated and 
29 
ambiguous but could be interpreted in a subjective manner. '00 He argues 
convincingly that in two of the latter four cases, a subjective interpretation makes 
better sense than an objective one. As we shall see in chapter 2, one of these two 
passages - D.4.5 .8 (1. Eas) - would become important in the medieval civilian 
discourse on deportation. 101 D.4.5.8 states: 
It is obvious that those obligations which are understood to hold good in natural law do 
not perish with a change of civil status [capitis deminutione], because a rule of civil 
law cannot destroy natural rights [civilis ratio naturalia iura corrumpere non potest] . 
Therefore, the action concerning the dowry, because it is framed with reference to what 
is right and just, continues to exist even after change of civil status ... 102 
According to this passage, obligations that are natural (rather than civil) in character 
remain unaffected by a change of civil condition ( capitis deminutio ), which could be 
brought about, for example, by parental emancipation or by loss of citizenship upon 
deportation. The obligation under consideration is that of the husband to return to his 
wife the dowry for which she is suing. The passage seems to indicate that the wife's 
claim to her dowry is her natural right, one of her iura naturalia, and as such 
enforceable with the action for dowry even after capitis deminutio. 103 It is true, 
however, that a degree of ambiguity remains concerning the meaning of ius in this 
text and in most of the others considered by Donahue. Given also the very small 
number of relevant texts, there are only really hints of a language of subjective 
natural right in classical Roman law. Nevertheless, we can conclude, with all due 
reserve, that this language does not appear to be completely alien to the ancient 
civilian tradition. 
[subsequently] adopted and have [then] been emancipated a second time, they keep the 
natural right of children.' 
JOO D.4.5.8; D.9.2.50; D.37.4.8,7; D.45 .1.83,5. 
101 Cf. eh. 2, pp. 72-5. The other passage is D.37.4.8,7: 'Si quis emancipatum filium 
exheredaverit eumque postea adrogaverit, Papinianus libro duodecimo quaestionum ait iura 
naturalia in eo praevalere : idcirco exheredationem nocere.' Watson (ed. ), The Digest of 
Justinian, vol. 3, p. 281 , also renders ius subjectively: 'If a man has disinherited his 
emancipated son and subsequently adopted him by adrogatio, Papinian, in the twelfth book 
of his Questions, says that his natural rights are paramount, and for that reason disherison 
prejudices his position.' 
102 
'Eas obligationes, quae naturalem praestationem habere intelleguntur, palam est capitis 
deminutione non perire, quia civilis ratio naturalia iura corrumpere non potest. Itaque de dote 
actio, quia in bonum et aequum concepta est, nihilo minus durat etiam post capitis 
deminutionem.' The translation is from Watson (ed.), The Digest of Justinian, vol. 1, p. 139. 
103 Cf. Donahue, '!us in the Subjective Sense', pp. 526-7. 
30 
One central argument of chapters 2 and 3 of this thesis is that the medieval civilians, 
in the context of exile, developed a theory of natural rights expressed in terms of 
subjective ius naturale, and thereby significantly expanded a language that is at best 
marginal in the Corpus iuris. 104 This appears to be a reflection of a broader trend: as 
mentioned above, subjective ius in general took on a greater role in medieval Roman 
law than it had in the jurists ' authorities. The person of the exile became a focal point 
for those elements of the legal system that the civilians considered to be natural, 
some of which will be familiar to us from the existing historiography, some of which 
will not. Rather than speaking of law or legal institutions, however, the jurists spoke 
quite explicitly of the exile ' s natural rights, his iura naturalia. This is a feature of the 
discourse right from the days of Accursius' Ordinary Gloss, composed mostly in the 
first half of the thirteenth century, to the fourteenth-century commentaries of 
Bartolus and Baldus. Contrary, therefore, to the dominant historical narrative about 
Roman law and natural rights, the medieval civilian tradition produced a natural 
rights theory, articulated in terms of subjective ius, that rivalled those of 
contemporary canonists, philosophers, and theologians; any origins narrative must 
include Roman law. Moreover, an awareness of the combined canonistic and civilian 
contribution allows us to raise the profile of the medieval legal tradition in this 
context, and correct the undue emphasis that historians often place on scholastic 
theologians and philosophers. 
104 Donahue, in '!us in the Subjective Sense', p. 530, is therefore right to suspect 'that some 
of the developments that Tierney sees in the canonists may also be seen in the civilians.' 
Initially, in his article, Donahue had wanted to cover the medieval jurists ' response to those 
texts in the Digest in which ius naturale can be interpreted subjectively. This he did not do, 
lamenting that 'the medieval story must await another day' . Already Tierney, aware of the 
provisional nature of his own work, pointed out that 'there is room for further research in 
several areas', including the thought of 'the medieval glossators of Roman law' and of ' the 
late-medieval jurists, both civilians and canonists' (The Idea of Natural Rights, p. 9). 
31 
32 
Chapter 2 
Deportation 
1. Introduction 
When the medieval jurists analysed contemporary forms of banishment, they looked 
to the ancient texts of the Roman law for examples and analogies. In the Corpus iuris 
civilis, deportation (deportare; deportatus, -i) was the most important prototype of 
exile that resulted in the loss of citizenship. This occurred when the Roman emperor, 
or other high imperial officials, confined a person to an island or an oasis within the 
Roman empire. As part of the sentence, that individual would lose his Roman 
citizenship and associated civil rights, but retain the rights derived from the law of 
nations and the law of nature. The central purpose of this chapter is therefore 
twofold. Since deported persons were excluded from the Roman civitas and existed 
purely as human beings under natural law, by examining what these individuals lost 
and what they retained, it is possible to give some tangible content to the concept of 
citizenship and civil rights on the one hand, and to the concept of natural rights on 
the other. As I made clear in the previous chapter, the medieval civilians not only 
possessed a substantive concept of natural rights, but also employed a language of 
natural rights: in other words, they articulated their theory in terms of subjective ius.1 
Furthermore, because we will examine the natural rights of a particular person in a 
particular context - when these are put under pressure by the power of the public 
authorities - it will be possible to draw a number of conclusions about how we 
should conceptualise these rights. 
While a substantial part of this chapter is devoted to the civilians' theory of natural 
rights, their discourse on deportation is equally fruitful in respect of Roman 
citizenship. There exists a small literature on medieval civilian theories about the 
local citizenship of cities within the Roman empire, as we saw in chapter 1,2 whereas 
very little indeed has been written about the concept of universal Roman citizenship, 
that is, citizenship of the empire. Historians have merely not~d several legal dicta 
1 Cf. eh. 1. pp. 23-31 . 
2 Ibid., pp. l 0-17. 
33 
concerning Roman citizenship that the medieval civilians created on the basis of their 
authorities without analysing the full implications of this status. It is therefore 
possible to find the idea that Roman citizenship is 'full freedom' ,3 that Roman 
citizens are 'all those subject to the empire',4 and that Rome is the communis patria. 5 
Not unjustifiably one could receive the impression that for the medieval jurists 
Roman citizenship was a rather empty category. Our study of deportation, however, 
will reveal a Roman citizenship that was a far richer concept than that. This concept 
contrasts with the way historians have portrayed the civilian idea of citizenship in the 
municipal context. Roman citizenship was not considered to be a contract between 
the individual and the community, it did not involve the performance of public 
duties, nor was it primarily defined by the exercise of political office. For the most 
part, Roman citizenship comprised a set of entitlements in the private sphere, 
especially in inheritance and family law. 
Reflecting historical interest in the Italian cities and their citizenship, some in-depth 
work has already been done concerning the jurists' thought on those banished from 
these cities. 6 On the other hand, there presently exists no devoted study of 
deportation in medieval Roman law. This is presumably one of the reasons why the 
concept of Roman citizenship has received such little attention. The best existing 
guide to deportation still remains Theodor Mommsen's 1899 encyclopaedic account 
of Roman criminal law, Romisches Strafrecht, which summarises the most impQrtant 
facets of deportation in the classical and postclassical periods. 7 The few authors who 
have more recently engaged with medieval theories of exile - e.g. Ghisalberti, 
Cavalca, Mooney - if they mention deportation at all, either base their own 
treatments of the deportatus on Mommsen's authoritative account or at least do not 
surpass it; nor do they analyse in detail what the medieval jurists made of the 
3 Quagl ioni, 'Le radici teoriche', p. 129, quoting the Dictionarium of Albericus de Rosate: 
'Civitas Romana, idest plena libertas.' 
4 Post, 'Review', p. 389, quoting Accursius to D.1.6.3: 'Cives Romanos dico omnes 
subiectos Imperii.' 
5 Bizzarri, 'Ricerche sul diritto di cittadinanza' , p. 56; Menzinger, ' Pisco, giurisdizione e 
ci ttadinanza', p. 41. 
6 Cf. eh. 4, pp. 176-7. 
7 Mommsen, Romisches Strafrecht, p. 956 ff. and p. 964 ff. For a very recent, detailed 
treatment of the phenomenon of deportation in the later Roman empire, using a variety of 
legal and non-legal sources, see Washburn, Banishment. Washburn's discussion of the law, 
however, does not add anything of substance to Mommsen's. Part of the explanation must be 
to do with the sparseness of the later legal sources themselves; they can only give so much. I 
would suggest that it is not until the medieval glosses and commentaries that interesting 
developments can be detected. 
34 
relevant legal texts. 8 The following chapter therefore constitutes the first 
comprehensive analysis of deportation.in the medieval civilian tradition. 
2. Further Considerations on Roman Citizenship in the Corpus iuris civilis 
Most allusions to Roman citizenship in the Corpus iuris come in the context of the 
manumission of slaves. Since, as we saw in chapter I, 9 all free persons were 
considered Roman citizens, to be manumitted or enslaved resulted in the automatic 
gain or loss of Roman citizenship. 10 Given that the status of freedom and the status of 
citizenship were conferred or lost together and at the same time, a study of slavery 
does not lend itself to singling out the features of Roman citizenship. The same can 
be said of postliminium, the right of Roman citizens, who had been enslaved by the 
enemy in a time of war, to regain their freedom and citizenship upon return from 
captivity. 11 The only other significant area where Roman citizenship is mentioned is 
that of capitis deminutio media, the legal mechanism by which citizenship was 
forfeited by virtue of a criminal penalty while freedom was retained.12 Given that the 
prime example of capitis deminutio media is deportation, a study of this penalty 
presents itself as a fruitful way of gleaning the significance of Roman citizenship, 
since it reveals the legal qualities that differentiate a citizen from a free foreigner. 
A result of Justinian' s universalism is that the term peregrinus in the Corpus iuris 
normally refers to an outsider of a particular city or province within the empire and 
only very rarely to a free non-Roman. 13 In fact, some instances of the unusual, latter 
usage are in the context of deportation, when a former civis has been punitively 
reduced to peregrinus status. For this reason any other mentions of a non-Roman 
8 Deportation is briefly discussed by: Ghisalberti, 'La condanna' , pp. 7-8 and 14-15; 
Cavalca, II bando, pp. 22-8 and 79-82; Mooney, 'The Legal Ban', pp. 161 -6. 
9 Ch. l,pp. 7-8. 
10 
E.g. Inst. l.5,3 (text on p. 38, n. 22 below); D. 38, 16, 3, l ('Libertum accipere debemus 
eum, quern quis ex servitute ad civitatem Romanam perduxit... '); C. 7 .6.1,2 (' ... servi ad 
civitatem producantur Romanam ... '). 
11 
For introductions to postliminium, see Berger, Encyclopedic Dictionary, p. 639, and 
Buckland and Stein, A Text-Book of Roman Law, pp. 67-8. The ·relevant Digest title 1s 
D.49.15 (De captivis et de postliminio et redemptis ab hostibus). 
12 Goria, 'Romani, cittadinanza', pp. 285-91. 
13 Cf. Goria, 'Romani, cittadinanza', pp. 287-91; Liebeschuetz, 'Citizen Status', p. 135; 
Mathisen, 'Peregrini, Barbari, and Cives Romani', pp. 120-1; Washburn, Banishment, p. 25. 
35 
peregrinus, when the term refers to an original foreigner rather than an ex-citizen, 
should be of great interest to us. A comparison of these two types of peregrini could 
potentially inform our conclusions about citizenship and natural rights in the context 
of exile. Unfortunately, although one can learn a thing or two from a handful of 
scattered references, Justinian does not leave his reader in a position to conduct an 
in-depth analysis of original peregrini. 14 
Justinian's compilation contains a great deal of law from many periods of Roman 
history, including much classical material, particularly in the Institutes and the 
Digest. It is therefore not unreasonable to suppose that vestiges or traces of the older 
Roman citizenship, that prominent and exclusive status before AD 212, could be 
found and were found by medieval jurists when they synthesised the scattered texts 
in order to bring some definition to the elusive category of civis Romanus. 15 If the 
current scholarly consensus is correct about citizenship under the Republic and the 
Principate, they would have found elements of a civic status that brought with it 
certain legal entitlements and obligations unavailable to the foreigner. On balance, 
the side of entitlement and privilege prevailed over that of obligation and duty. It was 
' an elite legal status' ,16 ' a valuable and much desired commodity'. 17 'To be a Roman 
citizen was to be a superior person'. 18 For this reason, the emperors used to bestow 
this status upon loyal provincials to reward them for their services to the empire. It is 
thus already likely, although perhaps not necessary, that the medieval glossators and 
commentators, simply by virtue of their sources, considered the civis Romanus in a 
broadly similar way. We will see how their treatment of deportation bears out this 
susp1c1on. 
At this point we should note an important tension that the ancient material brought to 
the medieval civilian tradition. On the one hand, Roman citizenship was described by 
the medieval lawyers, following their authorities, as the status of all free persons 
subject to the empire - perhaps not an unfair characterisation given its universality. 
On the other hand, as mentioned above, it will become very clear that deported 
persons lost their citizenship while retaining their freedom; they were free, subject to 
14 In the words of Wash burn, Banishment, p. 25: 'The legal literature from the later empire 
does little to explain their legal standing. Their existence seems plain enough; how they lived 
does not.' 
15 Goria, 'Romani, cittadinanza' , p. 287, suggests that traces of the older Roman citizenship 
can be found in the classical material of the Institutes and Digest. 
16 Mathisen, 'Peregrini, Barbari, and Cives Romani' , p. 1013 . 
17 Liebeschuetz, 'Citizen Status', p. 134. 
18 Ibid., p. 131. 
36 
the empire, but they were not Romans. This points to the anomalous nature of 
deported persons as well as the lack of real engagement with the category of 
peregrinus in Justinian' s law. This also points to a number of conceptual and 
methodological problems concerning the interpretation of terms such as ius civile, 
civitas, and peregrinus in the medieval material to which we shall now tum. 
3. Problems of Interpretation 
i . Civitas, ius civile, ea quae sunt iuris civilis 
Civitas, like ius, is a difficult word. Civitas primarily means ' city', which in itself 
contains a range of senses, from an assembly of physical buildings (the modem 
meaning of the word city) to a political community. For this reason, modem 
translations, especially older ones, often, but anachronistically, resort to the word 
' state '. Civitas as political community is associated with the term respublica, which 
of course does not merely denote 'republic' , but should be translated more neutrally 
as 'commonwealth '. Hence, as Accursius explains, the city of Rome (civitas 
Romana) is the Roman commonwealth (respublica Romana) as well as the Roman 
empire (imperium Romanum). In the same way, the terms civitas and respublica can 
describe a political community, a city, within the empire, such as the city of Florence 
or Perugia. 19 One of the reasons why we should not translate civitas as ' state ' is that 
it does not ever just refer to the institutional apparatus of government. It 
encompasses this apparatus, but more accurately refers to the entire community, the 
country, its people, its institutions, and its government as a totality. This is why I will 
join those historians of political thought who nowadays prefer using the very literal 
'city' in the _broader political sense just described.20 Accordingly, the term 'civil' in 
Roman law does not refer to civil society in the modem usage, which commonly 
19 Accursius to Coll.1.1 , § occupatis (=Nov. l.pr.) , col. 1: 'Reipubl icae. id est, totius imperii. 
[ ... ] Et nota quod tribus modis respublica dicitur. Primo, Romanorum, ut hie. Item pro 
civitate Romana tan tum, et tune proprie ... Item pro qualibet civitate, et tune improprie ... ' A 
succinct later example of the equation between civitas Romana and imperium Romanum can 
be found in Bartolus, Quaestio I (Lucanae civitatis), CQT (Basel) - henceforth, Bartolus, 
Lucanae civitatis - p. 205b: ' iura vero gentium et iura communia civitatis Romanae, seu 
Romani imperii , non perdit.' 
20 A good example is Brett, Changes of State. 
37 
opposes civil society (charities, trade unions, corporations etc.) to the state. Nor do 
civil rights (iura civilia) have the restrictive association with the civil liberties and 
the civic equality of the citizenry. Rather, 'civil' refers to the entire sphere of the 
civitas, again in that broad political sense. 
While the primary meaning of civitas is city, a secondary meaning is also usually 
assigned to it. In Latin, both ancient and medieval, this term was regularly used in a 
way that has induced modem translators to reach for the term 'citizenship'. One 
could, however, legitimately question the extent to which the medieval jurists had a 
meaningful concept of citizenship if they did not have a separate word for it. On the 
other hand, the texts of the Roman law at times deploy civitas in such a way that it is 
difficult to translate it as anything other than 'citizenship' if one is to avoid an undue 
literalism. And it is possible to have a concept without a separate word for it, in the 
same way that ius could signify both subjective right and objective law. In such 
passages, the sense of civitas seems to be figurative and denote the quality of 
belonging. For example, someone can be said to 'retain' or be 'without' civitas.
21 
Elsewhere civitas Romana is 'given' or 'granted' to, or adversely, 'taken away' from 
a person.22 As we would expect, the medieval jurists replicated the linguistic usage 
of their sources, and, in the context of deportation, speak of cives Romani who 
' retain ' or 'lose' their civitas and thus continue or cease to be cives.23 If we go by the 
terminology alone, a range of theoretical possibilities therefore exists. Citizenship 
could be a relatively strong idea, clearly implied or understood under the term 
civitas, although superficially obscured for want of a separate expression. Yet it 
might also be an underdeveloped idea and not emancipated in a significant way from 
the overpowering concept of city as a totality. In other words, an important question 
must be to what extent the medieval jurists had an idea of membership of the city 
which approaches something like the notion of a distinct, subjective status with 
21 D.48.19.8,8: ' ... si vero ad tempus damnantur, retinent civitatem.' D.48.19.1 7,1: 'Item 
~uidam an6At<>£~ sunt, hoe est sine civitate ... ' 
2 Inst. 1.5,3: ' ... et omnes libertos ... civitate Romana donavimus: multis additis modis, per 
quos possit libertas servis cum civitate Romana, quae sola in praesenti est, praestari.' D. 
48.1 9 .6,2: ' et sunt poenae, quae aut vitam adimant aut servitutem iniungant aut civitatem 
auferant...' 
23 E.g. Accursius: to D.1.6.3 , col. 12: 'Cives Romanos '; to D.48.20.5, col. 1423: 'in civitate 
existentem'; to Inst.1.12,2, col. 58: 'civitatem retinet'; to D.2.4.10,6, col. 127: 'civitatem 
amisit per deportationem'; to D.4.5.5,1, col. 455: 'cives esse desinunt'. Jacobus Butrigarius 
to D.1.5.18, p. 17b: 'pater, qui remansit in civitate'. Rainerius de Forlivio to D.48.1.5 , fol. 
118va: 'si amisit civitatem'. Albericus de Rosate to D.4.5.5,1, fol. 263vb: 'quod civitatem 
amittant'. Bartolus to C.6.24.1, fol. 18va: 'sed deportatus perdit civitatem'. Baldus to 
D.35.1.59(58), fol. 165vb: 'perditur tota civitas'. 
38 
which the modem words citizenship, cittadinanza, or Staatsangehorigkeit commonly 
tend to be associated and which justifies their use in this historic context as well. 
One way in which the texts of the Corpus iuris describe the loss of civitas in the 
context of deportation is to associate it with a loss of so-called 'things ' derived from 
the Roman civil law, and a retention of other 'things' derived from the law of 
nations. The most important passage in this regard is D.48.19.17,1 (1. Sunt quidam): 
Again_, some persons are without a po/is, that is without a city: as are ... those deported 
to an tsl_and, so that they do not have those things which are derived from the civil law 
but retam those things which are derived from the law of nations. ' 
Ite?1 quidam ~7t~At<>_£~ ~unt, hoe est sine civitate: ut sunt... in insulam deportati , ut ea 
qmdem, quae mns c1v1lts sunt, non habeant, quae vero iuris gentium sunt, habeant. 
This dichotomy between ea quae sunt iuris civil is and ea quae sunt iuris gentium ( or 
n~turalis) is the dominant way in which the jurists, from Accursius to Baldus, 
discussed the effects of capitis deminutio media. In the words of the Gloss: ' the 
de~orted person only loses his civitas, not his liberty, and thus he loses those things 
which are of the civil,. not the natural law' .24 The civilians also used a number of 
?.ther expressions. Depending on context, one could either be or not be 'deprived of , 
i~c~pable of , or ' excluded from' , the ius civile or the ius naturale/gentium. 
Surularly, these may or may not be ' taken away from' or 'lost by' somebody. In 
these contexts ius denotes objective law rather than subjective right. Although these 
phrases sound strange to our ears, the medieval civilians were not at all averse to 
describing the loss-retention dynamic of capitis deminutio in such objective terms 
that is, i~ terms of law or laws rather than in terms of rights. 25 The principle: 
however, is clear: a deported person moves from the sphere of the civil to the sphere 
of the natural. At face value, this seems to be a straightforward expression of Roman 
jurisprudence. Those who lose their civitas are removed from the positive legal 
24 A . 
. ccursms to _D.36.1.18(1 7),6, col. 1473: 'Nam deportatus civitatem solam amittit, non 
hbertatem, e! s1~ ea quae sunt iuris _civilis amittit, non naturalis, ut [D.48.19.17]. .. ' Cf. 
Jacobus Butnganus to C.6.24.1; Albencus de Rosate to D.48.19.17; Bartolus to D.48.19.17· 
Baldus to D.17.1.22,5. ' 
25 B_aldus_ provides a g~od ran~e of examples at: D.4.5.10, fol. 250rb (' .. .in iure naturae, quo 
captte mmutus non pnva°:1r ... ); _D.17: 1.22, fol. 113v (' ... quia deportatio tollit ius civile ... ); 
D.32.1.1,2, fol. _16lrb ('lbt, apohdes, td est, privati iure civili .'); C.6.24.1, fol. 69rb (' ... quis 
exclusus est a mre Romano.'); C.9.49.2, fol. 228va (' .. . quia deportatus retinet Iibertatem 
e~g?. e~t capax de iure gentium. sed non retinet civitatem, ergo non est capax de iur; 
c1vtl1... ). 
39 
system particular to that civitas (the Roman ius civile) but retain what is guaranteed 
by the transcendental norms common to all humankind. 
We should note that in the context of exile the civilians did not differentiate or in any 
way comment on the relationship between the natural law and the law of nations. 
They used them interchangeably, tacitly based on the idea that the law of nations was 
a form of natural law because it contained the precepts of natural reason. Even 
Accursius, whose exposition of the law, as a glossator, was closely tied to the text, 
speaks of ea quae sunt iuris natural is when the law refers to the ius gentium. 26 In 
other words, the jurists treated the ius naturale and the ius gentium as a single set of 
higher norms in the discourse on exile; their precise relationship did not matter. 
Important was the fact that they existed on a higher plane than the civil law and 
therefore had to be treated accordingly, affording a greater degree of protection from 
the interference of the prince and the positive ius civile, in line with the general 
precepts of Roman jurisprudence that we discussed in the previous chapter.27 
We will further discuss the natural law side of deportation below. For now, let us 
return to the meaning of civitas and the associated ea quae sunt iuris civilis. Phrasing 
the argument in terms of things reveals a view of capitis deminutio media that seems 
to lie half way between a subjective and an objective conception. One neither loses a 
law on the one hand (objective), nor rights on the other (subjective). Rather things 
derived from a law - things that a person possesses (habere, in the l. Sunt quidam) -
are taken away. As we shall see at length throughout this chapter, these things, which 
are lost when Roman civitas is lost, can broadly be described as the personal legal 
faculties exclusive to Roman citizens, such as active and passive testamentary 
powers. The possession of these faculties sets the civis apart from the peregrinus.28 A 
modem translator would at this point be inclined to render these subjective legal 
properties as 'the rights due to a citizen under the civil law', as 'rights of 
citizenship', or simply as 'civil rights' .29 
26 Accursius to C.6.24.1, fol. 179r: ' ... ea quae sunt iuris civil is perdunt, sicut est civitas et 
similia, ut est hie et... [D.48.19.17] ... sed ea quae sunt iuris naturalis non ... ' Cf. Accursius to 
D.36.1.18( 17),6, n. 24 above. 
27 Ch. l, pp. 27-8. 
28 E.g. Accursius to D.2.4.10,6, col. 127: 'civitatem amisit per deportationem, unde 
peregrinus dicitur'. For further references seep. 50, ns 57-8, below. 
29 E.g. Watson (ed.), The Digest of Justinian, vol. 4, p. 364 (D.48.19.17,1): 'Again some 
people are "stateless," that is, without a civitas, like ... those deported to an island, so that 
they have no rights under the jus civile but retain those which are of the jus gentium.' 
40 
If we are to remain faithful to the language of the sources, however, we should 
question whether ea is actually equivalent to iura in this context. There is one 
passage in the Roman law that would suggest as much - D.48.13.3 (l. Peculatus): 
The penalty for embezzlement of public money includes interdiction from fire and 
water, which has today been succeeded by deportation. In addition, anyone who is 
reduced to this status forfeits his property as well as all his former rights [omnia 
pristina iura].30 
Although there is a strong logical implication that these pristina iura are the ea quae 
sunt iuris civilis lost by the deported person, the Ordinary Gloss does not make that 
link explicit.31 We shall see in a moment that later jurists did just that. Even so, it is 
already relatively clear in the Gloss, from the way Accursius connects and reconciles 
the texts of the Corpus iuris, that there is indeed a general equivalence between the 
things derived from the civil law at the l. Sunt quidam and the iura mentioned here. 
The reason for this is that Accursius associates, whether directly or indirectly, the ea 
quae sunt iuris civilis lost by the deported person with a number of legal faculties 
described as iura in other contexts. Those contexts also make clear that possession of 
these iura is dependent on being a Roman citizen. For example, Accursius lists the 
ability to make a testament as a thing derived from civil law in his gloss at the l. Sunt 
quidam, but with reference to another law describing deported persons as not having 
(habere) a ius testamenti faciendi due to their lack of civitas.32 Also at the l. Sunt 
quidam, Accursius describes patriapotestas as a power derived from the civil law 
held by Roman citizens, 33 and in another gloss states that a deported father no longer 
has (habere) the ius patriaepotestatis.34 The best example is Accursius' gloss to a 
30 
'Peculatus poena aquae et ignis interdictionem, in quam hodie successit deportatio, 
continet. porro qui in eum statum deducitur, sicut omnia pristina iura, ita et bona amittit.' 
The translation is adapted from Watson (ed.), The Digest of Justinian, vol. 4, p. 345. The 
interdiction from water and fire (aquae et ignis interdictio) is an older form of exile which 
Justinian completely expunged from the law books except to say that it is no longer in force. 
Until the early Principate, interdictio aquae et ignis denotes a loss of legal protection (akin to 
hostis-transfuga status) suffered by a convicted and confined criminal in case he broke the 
bounds of his exile and returned to those lands forbidden to him. Cf. Mommsen, Romisches 
Strafrecht, p. 68 ff. and p. 964 ff. 
31 Accursius to D.48.13.3 . 
32 Accursius to D.48.19.17,1 , col. 1411: 'Non habeant. ut testari , ut [D.32. l.l,2: Hi, quibus 
aqua et igni interdictum est, item deportati fideicommissum relinquere non possunt, quia nee 
testamenti faciendi ius habent, cum sint a1t6AtbE<;.]. Item, nee eis legari .. .' 
33 Ibid.: 'Item, nee filios retinent in potestate ... quae potestas est iuris civilis .. .' 
34 Accursius to C.5.17.1, col. 891: 'Non ergo habet ius patriae potestatis? Respondeo ad hoe 
quad est specialis casus ubi natus ex matrimonio non est in potestate, et quod generaliter 
dicatur in potestate nasci, intelligitur quando nullus mutavit conditionem.' 
41 
passage in the Institutes, from which it transpires, as we shall discuss in some detail 
later, that the rights of agnates (iura agnatorum) are iura civilia held (again, habere) 
by Roman citizens and are lost by deportation. These rights essentially constituted 
the ability to succeed to inheritances and to acquire guardianships over related 
minors (ius succedendi et ad tutelam vocandi).35 
On the other hand, it is true that in the Ordinary Gloss not every civil faculty, not 
every thing lost by the deported person is explicitly described as someone's ius by 
Accursius or by his sources, at least in the context of exile. Nor can we find a 
statement in the Gloss that provides a parallel general definition of capitis deminutio 
media literally phrased in terms of iura civilia and iura naturalia/gentium rather than 
in terms of ea quae sunt iuris civilis/naturalis. Then again, we should not expect too 
much fluency, too many broader explanatory statements from the Gloss, given its 
habit of using contracted phrases and unexplained references to other parts of the 
law. As a result, Accursius' thought is often not immediately obvious but has to be 
reconstructed by the reader. As a glossator, he is still very faithful to the language of 
his sources. Nevertheless, the general statement of the l. Peculatus, along with the 
fact that ea quae sunt iuris civilis become linked to a number of specific iura in the 
Gloss, means that we can safely assume an equivalence between civil rights and 
things derived from the civil law. As I have indicated, these civil rights are things, or 
civil faculties, possessed by a citizen, and they are lost when the citizen forfeits his 
civitas and becomes a peregrinus through deportation. In other words, already in the 
thought of Accursius, we can observe a language in which civitas appears as a 
personal status of the citizen associated with a number of subjective rights. 
While this language of citizenship and civil rights remains at least partially implicit 
in the Gloss, the jurists working in the second half of the thirteenth century and 
beyond leave us in little doubt about its place in the civilian discourse on deportation. 
Before analysing their language of ius, we should note that the lawyers sometime 
used another term as a synonym for ea. Both Jacobus de Arena and Jacobus 
Butrigarius argue that through deportation 'civil acts [ civiles actus] are taken away 
from someone. ' 36 This is a good way of describing what were essentially the civil 
faculties of a citizen. It points to the fact that, for the most part, as we shall see in due 
35 Accursius to Inst.1.15,3. See pp. 75-8 below. 
36 Jacobus de Arena to D.12.6.13, fol. 78rb-va: ' ... cui civiles actus sunt adempti, ut deportato 
in insulam .. .' Cf. Jacobus Butrigarius to C.9.49.2, fol. l 12rb: ' ... possum dicere quod perdit 
actus iuris civilis .. .' 
42 
course, these constituted the ability of a citizen to do something which a peregrinus 
could not, such as the ability to make a testament. 
Nevertheless, it was not the language of things or acts, but the language of rights that 
proved increasingly popular amongst the jurists as a general way of describing the 
effects of capitis deminutio media. This is most noticeable in their commentaries to 
the l. Peculatus, which stated, as we saw above, that the deported person 'lost all his 
former rights [ iura]'. They explained that these rights were 'civil rights' ( iura civilia) 
and linked them to ea quae sunt iuris civilis of the l. Sunt quidam.37 Furthermore, 
they were 'proper to Roman citizens' ('propria civium Romanorum'), and thus 
depended on the possession of civitas.38 The best example of the nexus ea-iura-
civitas is a comment by Dinus de Mugello, recorded by Albericus de Rosate: 
There [where it says], 'former rights', that is, citizenship and all those things derived 
from the civil law, which are proper to Roman citizens. [ ... ] Dinus. 
ibi, pristina iura, id est, civitatem et omnia quae sunt de iure civili, quae sunt propria 
· · R 39 c1vmm omanorum. [ ... ] Dy. 
It should be emphasised that civitas here is itself described as someone's ius as well 
as the precondition for a whole set of other iura held by citizens. This is why Jacobus 
de Arena can say that 'deportation is to take away from someone the ius civitatis 
Romanae' . 40 This ius does not, at least in the first instance, refer to the law of the city 
of Rome, the ius civile (though that is of course implied), but to the subjective sense 
of civitas Romana, which signifies the quality of belonging. As Baldus notes, in this 
sense, civitas can be said to be a person's status, which is lost upon deportation.41 
Hence, we should have no qualms as translating Jacobus de Arena's ius civitatis 
Romanae as the 'right of Roman citizenship' - or, more naturally in other languages, 
as Burgerrecht or diritto di cittadinanza. 
37 Rainerius de Forlivio to D.48.13 .3, fol. 125va: 'ibi. iura. civilia, non iurisgentium, ut 
[D.48.19.17).' 
38 Jacobus de Arena to D.48.13.3, fol. 225rb: 'Peculatus. Pristina iura. que sunt propria 
civium Romanorum.' 
39 Albericus de Rosate to D.48.13.3, fol. l 92ra: 'ibi, pristina iura, id est, civitatem et omnia 
~uae sunt de iure civili, quae sunt propria civium Romanorum. [ ... ] Dy.' 
4 Jacobus de Arena to C.6.24.1, fol. 40ra: ' ... deportare est alicui ius civitatis Romanae 
adimere ... ' 
41 Baldus to C.9.49.2, fol. 228va: ' ... aut quis perdit bona et statum, id est, civitatem, et ita 
loquitur hie.' 
43 
ii. /us naturale vel gentium, ea quae sunt iuris naturalis/gentium 
When it came to the civil side of deportation, we saw that the Gloss preferred to 
speak of things derived from the civil law ( ea quae sunt iuris civil is) rather than in 
terms of civil rights (iura civilia). Although Accursius did deploy a language of 
rights, this remained partially implicit and was only brought out in its fullest extent 
by subsequent jurists. The same can be said for the natural side of deportation. The 
Gloss normally speaks in terms of things rather than ius. We will see in the course of 
this chapter that, just as in the case of civil things, these natural things possessed by 
the deported person - again, habere, at the l. Sunt quidam - can broadly be described 
as legal faculties; for this reason, Jacobus Butrigarius described them as natural 
'acts' (actus). 42 They were the legal faculties of a human being, who was no longer a 
citizen, under natural law. Here too there is strong evidence to suggest that these 
natural ea were generally equivalent to iura in Accursius' thought. In the context of 
deportation, the glossator associates the things derived from the natural law at the l. 
Sunt quidam with a number of legal faculties described as iura elsewhere. For 
example, the deported woman's action for dowry and the deported person's claim for 
a subsistence legacy are labelled by Accursius as ea quae sunt iuris naturalis in one 
gloss, and as iura naturalia in another. 43 The best evidence comes once again from 
the passage of the Institutes concerning the right of agnation and cognation. As we 
shall see in detail below, Accursius treats the rights of cognation (iura cognationis) 
and the rights of blood (iura sanguinis) as subjective iura naturalia that are not lost 
by the deported person.44 Already in the Ordinary Gloss, therefore, we can observe 
enough evidence with which to demonstrate at least the beginnings of a subjective 
natural rights language. According to Accursius, legal faculties granted by the 
natural law or the law of nations are the same as things possessed by a deported 
person and are also described as that person's iura. 
42 Jacobus Butrigarius to C.9.49.2, fol. l 12rb: 'Sed non perdit actus iuris naturalis vel 
gentium .. .' 
43 D.4.5.8 (I. Eas) is the basis for the action for dowry, and D.4.5.10 (I. Legatum) for the 
subsistence legacy. This legacy is given by the Gloss as a thing derived from the natural law 
at D.48.19.17,1, col. 1411 (' ... et alia quae iuris gentium sunt bene possunt, ut... [D.4.5.10].'). 
Both are natural things at D.36.1.18(17),6, cols 14 73-4 ('Nam deportatus civitatem so lam 
amittit, non libertatem, et sic ea quae sunt iuris civilis amittit, non naturalis, ut [D.48.19.17], 
et [D.4.5.8], et [D.4.5.10]. . .'). They are called iura naturalia in Accursius' gloss to 
Inst.1.15,3, col. 70 ('id est, certe non tolluntur capitis diminutione cum sint immutabilia haec 
iura ... Item, nee ea quae naturalem aequitatem habent, ut [D.4.5 .8] et [D.4.5.10].'). 
44 Accursius to Inst.1.15,3. See pp. 75-8 below. 
44 
The glossators, therefore, as represented by Accursius, provided the starting point for 
an interpretative shift that set the way clear for the development of a more 
perspicuous natural rights language in the civilian tradition. Although the 
commentators usually continued to speak in terms of 'things' in the manner of the 
Gloss, they were more comfortable than Accursius in referring also to ius. Hence, we 
will continue to see examples of ea quae sunt iuris naturalis/gentium that are also 
labelled iura. Sometimes it is possible to find broader statements framing the effects 
of deportation in terms of natural rights. Rainerius de Forlivio, for instance, writing 
in the first half of the fourteenth century, clearly established a general equivalence 
between the natural things of the l. Sunt quidam and natural iura. Commenting on the 
phrase of the l. Peculatus which states that the deported person loses 'all his former 
rights', Rainerius explains: 
There, [where it says] rights, [these are] civil rights, not those derived from the law of 
nations, as at [D.48.19.17]. 
ibi, iura, civilia, non iurisgentium, ut [D.48 .1 9.17].45 
In other words, a deported person loses his iura civilia, but not the iura derived from 
the law of nations. And these iura iuris gentium are equivalent to the personal legal 
faculties, the natural things possessed by the deportatus, at the l. Sunt quidam (cited 
here). In the second half of the fourteenth century, Baldus also provides a general 
explanation of the deported person's legal position in terms of ius: 
... subsequent acquisition is derived from the law of nations, of which he is not 
deprived, because deportation takes away the civil law .. . but the law of nations is 
common to all nations, and it is called the natural law, and natural rights are not taken 
away in that case ... 
. .. acquisitio subsequens est de iure gentium, quo non est privatus, quia deportatio toll it 
ius civile .. . sed ius gentium est commune omnium gentium, et dicitur ius naturale, et 
iura naturalia non sunt ibi adempta ... 46 
There appears to be an interesting shift in this passage from an objective to a 
subjective sense of ius. The civil side of deportation is unambiguously explained in 
terms of objective law: the ius civile, as a whole, is lost. But the natural side, at least 
towards the very end, seems to be described in terms of subjective rights. The 
passage is somewhat ambiguous · since we know that the jurists were equally 
45 Rainerius de Forlivio to D.48.13.3, fol. 125va. 
46 Baldus to D.17.1.22,1, fol. 113va. 
45 
confident in describing the dynamics of deportation in terms of law and laws. 
Arguably, the iura here might simply be considered as the natural law and the law of 
nations taken as two separate entities. Given what else we know about deportation, 
however, it seems more probable that the choice of plural is designed to evoke the ea 
quae sunt iuris naturalis of the l. Sunt quidam, in other words, the natural legal 
faculties possessed by a deported person. Such an interpretation is suggested by the 
fact that in this passage Baldus is commenting on the ability of the deported person 
to acquire property after his sentence. And once iura naturalia become associated 
with personal legal faculties, a subjective interpretation makes better sense. 
As Baldus' commentary indicates, however, the civil law tradition on deportation 
does not provide the very best examples of broader, generalising statements 
describing the effects of exile in terms of natural rights. In the next chapter, when 
discussing the thought of the jurists on the other instance of capitis deminutio media 
- the reduction of a citizen to the status of hostis (enemy) and transfuga (deserter) -
we shall see how Bartolus in particular reaches more readily for the language of 
subjective ius to explain the position of those exiles.47 
iii. Subjective ius as a Person's Mediate and Immediate Obligation 
While it is perfectly legitimate to couch our discussion of deportation in terms of 
civil and natural rights, we have to be careful not to modernise the discourse too 
much. The civil rights of the citizen and the natural rights of the exile are not merely 
rights in the sense of entitlement (claim, power, liberty). They cannot be 
automatically opposed to duties as is commonly done in modem debates about 
rights. We saw in chapter 1 that a person 's ius in the legal Latin of the ancient 
sources can also denote immediate obligation: my ius as my duty. This can be 
distinguished from a mediate obligation, which refers in the first instance to an 
entitlement, a faculty, through which someone can perform a duty; an immediate 
obligation simply is a duty in itself. 48 In a small number of instances in the context of 
deportation, the medieval civilians use ius in both of these ways. We will come 
across four cases. Firstly, the civil right of guardianship (ius tutelae) is a mediate 
obligation: although in the first instance a power (potestas) over the ward, it is the 
means of performing a duty that is portrayed as an undesirable burden in the law. 
47 Ch. 3, pp. 165-71. 
48 Ch. 1, pp. 25-6. 
46 
The other three cases are immediate obligations. The civil ius patriaepotestatis of the 
unemancipated son simply is the obligation toward his father. The natural right of 
consanguinity (ius consanguinitatis) simply is a sexual and marital prohibition. 
Finally, the right of patronage (ius patronatus) of the deported freedman simply is 
his obligation towards his patron. The medieval material therefore provides further 
evidence for the argument that we should redefine the meaning of subjective ius in 
the Roman law tradition. Subjective ius does not simply denote subjective right in 
the modem sense. In medieval Roman law, my ius can be both my entitlement as 
well as my immediate obligation. As I argued in the last chapter, since the language 
of ius is no more or less subjective when it refers to a person's immediate obligation, 
we should continue to speak of rights, but only if we understand that these rights can 
be twofold in the manner just described. We will return to this point towards the end 
of chapter 3. 49 
iv. Peregrinus, civis, subiectus 
The way the jurists discussed deportation would suggest that the deported person 
moved from the sphere of the civil to the sphere of the natural in a straightforward 
and complete fashion - that he possessed no legal capacity in respect of the civil, but 
every legal capacity in respect of the natural. As we shall see, that was an imperfect 
way of describing the mechanism of capitis deminutio media. In a general sense, it 
was of course right, but it requires two major qualifications. 
Firstly, the civil was never totally relinquished; not all aspects of the civil law, 
understood broadly as the Roman legal system, were really forfeited by the criminal 
ex-citizen. With the exception of Bartolus - who confronts this tension when 
discussing the imperial hostis-transfuga50 - the jurists never explicitly drew their 
readers' attention to it. As we would logically expect, the Roman law applies on 
Roman soil and is imposed, albeit in a differentiated manner, on citizens and non-
citizens alike. As a result, the deportatus, and even more so the hostis-transfuga, are 
subject to diverse regulations foreseen by the Roman criminal law in their respective 
circumstances. Most visible are a range of punitive measures beyond the loss of 
citizenship and civil rights, continually enforced against former citizens after their 
sentence. Immediately obvious in the case of deportation is the confinement of the 
49 Ch. 3, pp. 165-71. 
50 Ibid. 
47 
deported person to an island or an oasis for the rest of his natural life. Of course these 
aspects of the Roman legal system, its criminal law, are still binding. How could 
such sentences have any traction otherwise? 
'Civil', in this context, should therefore be understood in two ways. Firstly, it has a 
narrow sense which encompasses that part of the civil law pertaining exclusively to 
citizens and their civil rights. Secondly, it has a wider sense which refers to the 
apparatus of the prevailing Roman legal system applicable to all persons under 
Roman jurisdiction, without regard in the first instance to their statuses as citizens or 
otherwise. The second, wider sense receives scant attention both in the Corpus iuris 
and in medieval glosses and commentaries to it. The reason for this is that the 
category of peaceful foreigner, as we saw above, has for the most part disappeared 
from Justinian's universalising compilation. Traditionally, when membership of the 
Roman civitas was more exclusive, Roman law had a whole branch dealing with 
peregrini on Roman territory. Above all, the Romans were forced to regulate the 
interaction between cives and peregrini, and to consider how to make effective and 
accessible to foreigners what they considered to be norms derived from the ius 
gentium. But since Justinian considers all those free individuals subject to his 
authority to be Roman citizens, what remains in the Corpus iuris is merely a handful 
of passing references that alert us to this wider application of the Roman law.51 
One important text, which recurs in the context of exile, is the I. Non dubito 
(D.49.15.7), especially its second paragraph: 
I have no doubt that although free peoples and those bound to us by treaty [foederati et 
liberi] are foreigners to us, there is no postliminium between us and them; for what 
need is there of postliminium between us and them, since they retain their freedom 
[libertatem suam] and ownership of their own property [dominium rerum suarum] in 
our country just as in their own, and the same applies to us in their country? [ ... ] 2. But 
persons from civitates foederatae may be charged in our courts, and we inflict 
punishments on them if condemned. 52 
51 Cf. Catalano, '/us Romanum', who highlights the diverse senses of the terminology 
surrounding the civil law and charts their historical development from the Republic to 
Justinian, with a particular focus on the notion of ius Romanum. 
52 
'Non dubito, quin foederati et Iiberi nobis extemi sint, nee inter nos atque eos 
postliminium esse: etenim quid inter nos atque eos postliminio opus est, cum et illi apud nos 
et libertatem suam et dominium rerum suarum aeque atque apud se retineant et eadem nobis 
apud eos contingant? [ ... ] 2. At fiunt apud nos rei ex civitatibus foederatis et in eos damnatos 
animadvertimus.' The translation is adapted from Watson (ed.), The Digest of Justinian, vol. 
4, p. 400. 
48 
This law comes in the title of the Digest concernmg the rules of captivity and 
postliminium. As mentioned above, in a time of war between Rome and a hostile 
foreign people (hastes), individuals captured by either side become the slaves of their 
captors. If a Roman escapes or otherwise returns from enemy captivity, by right of 
postliminium, he regains his freedom and citizenship upon entering Roman territory. 
As the Gloss explains, these rules also apply to the case in which Rome has no 
customary relationship (usus) with an extraneous people. But once customary good 
relations are established, especially once a treaty (foedus) has formalised them, the 
rules of captivity and postliminium no longer take effect.53 For this reason the I. Non 
dubito states that foreigners from such peoples have their liberty as well as 
ownership of their property on Roman soil. Importantly - and here we return to that 
broader sense of ius civile - despite not being citizens, they are subject to the 
jurisdiction of Roman courts within the Roman empire. Bartolus, in his commentary 
to the I. Non dubito, links this idea with the general rule that a defendant in delictual 
cases is liable to the courts of a place for three reasons: in technical terms, forum 
arises either because he has his origin there (ratione origin is), or his domicile 
(ratione domicilii), or he has committed the offence in the territory over which the 
courts have jurisdiction (ratione delicti). 54 The peregrinus or forensis from a civitas 
foederata evidently does not have his origin in the empire and might not have a 
domicile there either. Nonetheless, he is subject to imperial jurisdiction ratione 
delicti. 55 
Unsurprisingly, given their source texts, it is characteristic of the medieval civilians 
not to be interested in the figure of the friendly, non-Roman foreigner. I cannot find 
any further discussion of the legal status of this person in their glosses and 
commentaries. As we saw above, the term peregrinus in the Corpus iuris normally 
refers to strangers from a different part of the Roman empire. In the Middle Ages, 
53 Accursius to D.49.15.24, cols 1517-8: 'quinque sunt genera gentium. Primum, hostes, de 
quibus hie <licit, in quibus, scilicet, postliminium locum habet. item, alii, cum quibus non 
habemus usum, et in his idem, ut [D.49.15.5,2]. Tertium, populi Iiberi, qui nee nobis nee 
alicui sunt subditi, sive sint nobis foederati, cum prius erant forte hostes, sive non. et in his 
non est opus postliminii, ut [D.49.15. 7]. et idem dico de iis qui non sunt liberi, sed subsunt 
regibus non subditis tamen imperio, si modo cum eis amicitiam habuimus, ut in [D.49.15.7] 
et [D.49.15.5].' 
54 Cf. D.5.1; C.3.13; Berger, Encyclopedic Dictionary, p. 476. 
55 Bartolus to D.49.15.7, fol. 227ra-b: 'Sed quaero, quid velit dicere hie textus, quod ex 
civitatibus foederatis rei apud nos fiunt? Aut, nota, illi de civitatibus foederatis delinquunt in 
territorio nostro, et non est mirum, quia ratione delicti sortiuntur forum nostrum undecumque 
sint. Aut non delinquunt apud nos, et videtur quod apud nos non possunt puniri, quia non est 
aliqua causa propter quam hie so11iantur forum .. .' 
49 
the dominant sense of peregrinus became pilgrim, as in the imperial constitution 
Omnes peregrini, promulgated by Frederick II in 1220 in order to ensure the safety 
of those travelling throughout Christendom for religious reasons. 56 This dominant 
sense of peregrinus forced the jurists to make a distinction in the context of 
deportation, since the law was equally clear that the civis who had been deported was 
reduced to that status.57 The lawyers therefore described the citizen pilgrim as a 
'good' foreigner (bona peregrinatio, peregrinatio in bona parte) and contrasted him 
to the 'bad' criminal ex-citizen (ma/a peregrinatio, peregrinatio in ma/a parte). 
They were clear that only the latter lost his Roman citizenship and suffered the 
effects of capitis deminutio media. Once again it is a testament to the anomalous 
nature of the friendly, non-Roman peregrinus that the lawyers never associated this 
bona peregrinatio with him. 58 
Nevertheless, it is important to recogmse that the Roman law does know such 
ordinary non-Roman peregrini, even if the civilians seem uninterested in them most 
of the time, as these foreigners provide an informative point of contrast to the 
criminal ex-citizen foreigners: the deportatus and hostis-transfuga. Two facts are 
again worth underlining in this context. Firstly, free foreigners can enjoy liberty and 
ownership of property in the empire without being subject to the restrictions, which 
could be severe, that we will encounter in the case of criminal ex-citizens. Secondly, 
a person did not have to be a citizen (nor at one time have been one) in order to feel 
the effects of the Roman law within the empire. 
The existence within the empire of peregrini, whether they were formerly citizens or 
not, also highlights that the legal dictum, Roman citizens are all those subject to the 
emperor, is not entirely accurate. It is not a bad rule of thumb given that these 
peregrini barely feature in the law, which is presumably why the medieval civilians 
continued to repeat this phrase.59 For all intents and purposes, a civis was also a 
subiectus. In the legal discourse, these terms alone do not provide any evidence for 
Ullmann's well-known claim that late-medieval political thought oversaw the 
56 
auth. Omnes peregrini, post C.6.59(58).10. This authentica formed part of Frederick's 
Constitutiones de statutis et consuetudinibus contra libertatem ecclesiae - cf. eh. 3, pp. 106-
7. 
57 D.2.4. l 0,6: 'Sed si per poenam deportationis ad peregrinitatem redactus sit patronus ... ' Cf. 
D.28.5.6,2; C.6.24.1. 
58 Accursius to Inst.2.14,12, C.6.24.1; Albericus de Rosate to D.2.4.10,6; Cinus de Pistoia, 
Jacobus Butrigarius, Bartolus, and Baldus to C.6.24.1. 
59 E.g. Albericus de Rosate to D.1.6.3, fol. 52ra: 'Nota Glossa ... quae <licit, et bene, quod hie 
cives Romani accipiuntur omnes subiecti Romano imperio.' 
50 
emancipation of the active 'citizen' from the passive 'subject'. These terms contain 
no such conceptual dichotomy.60 All cives are subiecti in that they are subject to the 
jurisdiction of their civitas, in this case the Roman empire. But not all subiecti are 
citizens, as we have seen, since foreigners are also subject to the jurisdiction of a 
foreign civitas within its territory - hence the provision of the I. Non dubito. When 
speaking of citizens, the choice of referring to them as cives or subiecti depends on 
whether one wishes to emphasise the element of subjection to the city's jurisdiction 
or the element of membership of that city. 
v. A Note on Method 
We should be aware of the potential pitfalls of deducing the content of civil and 
natural rights through a study of capitis deminutio media. The deported person and 
the hostis-transfuga lost their rights of citizenship, but were still subject to the civil 
law in that broader sense which allowed the sentence, with its diverse punitive 
measures, to be enforced against them. It is worth emphasising that these punitive 
measures placed sanctions on such individuals over and above what would have been 
the case had the penalty only resulted in the forfeiture of their citizenship. The 
effects of capitis deminutio media had two sides. On the one side, citizenship is lost, 
and because of that loss there are certain consequences. Working back from these 
subsequent restrictions in legal capacity, it is possible to establish the civil rights 
reserved for citizens. On the other hand, there are certain other penalties restricting 
the individual's scope of action independently of that loss of citizenship. Sometimes 
the difficulty lies in keeping the two sides separate and not to confuse the latter with 
the former. The ever-present danger is to read back incorrectly from an independent 
restriction that it resulted from the loss of citizenship and therefore conclude that the 
reverse of the restriction is a characteristic of citizenship. To give a concrete and 
rather obvious example: the deportatus was confined to an island or an oasis. This 
does not mean that general freedom of movement was a civil right reserved for 
Roman citizens. The medieval glosses and commentaries are difficult sources that do 
not tend to deal in those neat categories sought by the modem reader. But despite 
60 Ullmann, 'The Rebirth of the Citizen'. Already Post, 'Review', criticised Ullmann's thesis 
from this perspective. A more subtle critique is offered by Canning, Baldus de Ubaldis, pp. 
168-9. 
51 
certain grey areas, the jurists normally indicate ( or at least strongly imply) if there is 
a connection between loss of citizenship and loss ofright.61 
We have a similar problem with natural rights, and here we must make that second 
qualification to the lawyers' general statement that a deported person moves from the 
sphere of the civil to the sphere of the natural. Capitis deminutio media - especially 
in the case of the hostis-transfuga - does not reveal an unadulterated list of iura 
naturalia that an ordinary peregrinus might enjoy. This is because it does not create 
an ordinary peregrinus but rather a criminal ex-citizen. For the sentence to have any 
traction, the civil must necessarily continue to infringe upon the natural. The punitive 
measures of the sentence, independent of the loss of citizenship, suspend or make 
inaccessible - often severely - rights usually due to a person under natural law, in 
line with the Romanist doctrines concerning the powers of the civil law to derogate 
from (though not abrogate) higher norms. We will see how these higher norms are 
varyingly effective in confrontation with the city's authorities and sometimes barely 
effective at all. This is why it will be helpful to keep the figure of the peaceful 
stranger always in the back of our minds as we focus on those who were once 
citizens. 
4. The Basics of Deportation and Ordinary Roman Criminal Procedure 
With these caveats in mind, let us begin our discussion of the legal detail with some 
of the basic features of deportation. 62 As the Gloss explains, only the emperor 
himself or his immediate delegates, the urban prefect and the praetorian prefect (high 
imperial magistrates), and the imperial senate can deport a criminal. 63 It is the result 
61 I am thus more optimistic than Goria, 'Romani, cittadinanza', pp. 289-90, who, working 
with the late-antique Roman law texts only, emphasises the problems with deducing 
backwards the contents of citizenship from instances of capitis deminutio media. 
62 The two titles of the Digest most relevant for deportation are D.48. l 9 (De poenis) and 
D.48.22 (De interdictis et relegatis et deportatis), although aspects of this penalty are 
scattered throughout the Corpus iuris. The following paragraph is largely a summary of 
Mommsen, Romisches Strafrecht, p. 956 ff. and p. 964 ff. 
63 Accursius to Inst. l .12. Amongst the medieval civilians, there was an extensive debate over 
whether the provincial governor can also deport, and if not, why not. In their theoretical 
expositions of the Digest, the jurists were split on this question. In the context of banishment, 
however, as we shall see in eh. 4, pp. 225-64, they agreed that the provincial governor did 
not have the ius deportandi. Cf. Accursius to D.1.18.6,8, D.48.22.6,l; Rainerius de Forlivio 
52 
of a sentence passed on a guilty defendant at the end of a capital trial. Capital does 
not, as in modem usage, refer specifically to the loss of life, but also refers to the loss 
of liberty (i.e. penal slavery) or citizenship. This is because caput (gen. capitis) 
denotes the civil or legal capacity of a person. Hence deportation, with its 
concomitant loss of citizenship, constitutes a reduction (deminutio) of caput. 64 
Deportation was a penalty usually imposed on high-class defendants as a substitute 
for life-long penal labour in the imperial mines.65 It resulted from the commission of 
a serious crime, such as the forgery of documents, armed violence, or the 
embezzlement of public funds. Although the discourse was usually framed in terms 
of the masculine deportatus, it was a punishment for both men and women. A small 
number of laws specifically discuss the implications of the deportation of a female 
citizen. As we shall see, these normally concerned questions of marriage, dowry, and 
the status of children. 
Deportation was the result of an ordinary criminal trial. In Roman law, capital 
proceedings could not be brought to a conclusion, the court could not impose a final 
sentence - what the medieval jurists called a 'definitive sentence' (sententia 
diffinitiva) 66 - if the defendant remained absent. It was also incumbent on the 
authorities officially to summon the defendant to court. This part of the proceedings 
was called the citation, and without it the trial became unlawful. If the defendant 
refused to answer his summons he became contumacious, leaving the courts with 
two remedies. Firstly, the judge could apply the procedure of annotation (adnotatio 
bonorum). This meant that the defendant's goods would be registered and frozen 
(annotated) for one year, after which they were confiscated by the imperial treasury 
(the fisc). If the defendant returned within the one-year period, his goods would be 
unfrozen and proceedings would continue as usual. If he returned after the year was 
up, he could no longer recover his goods but was still able to contest the principal 
charge. The greatest criminal penalty a judge could impose on an absent defendant 
to D.48.22.6; Albericus de Rosate to D.48.19 .2, C.9.2.6; Bartolus to D.48.22. 7; Baldus to 
D.32.l.3. 
64 On the Roman concept of capital punishment generally, see Mommsen, Romisches 
Strafrecht, p. 907. 
65 The opus publicum is a less well-known instance of capitis deminutio media and was the 
preferred form for plebeians - see ibid., pp. 952, 958-9. 
66 According to Berger, Encyclopedic Dictionary, p. 429, 'definitiva sententia ' was a 
'postclassical term for the final judgment in a civil trial, to be distinguished from 
interlocutory, preliminary decisions.' As we shall see in chs 3 and 4, the medieval jurists 
used this distinction in criminal proceedings as well ; dif.finitiva rather definitiva was the 
common medieval spelling. 
53 
was relegation (relegatio ), a less severe form of exile. Generally speaking, the 
Romans of late antiquity expected to get their man.67 
Relegation was a form of exile that resulted in exclusion from, or confinement to, a 
specific place within the Roman empire. As it was not accompanied by loss of 
Roman citizenship, it was not considered a capital penalty. We should note that 
relegation does not pertain to local, municipal citizenship as deportation does to 
Roman citizenship. Relegation was a sentence designed to restrict the criminal's 
movement and might (although usually it did not) include the confiscation of his 
goods. The loss of local citizenship, however, formed no part of it.68 In fact, it 
appears that loss of the latter did not constitute a penalty in Justinian's criminal law 
at all, which is unsurprising, given that it was not an advantageous status to hold.69 
As a result, there is no direct equivalent in the Roman law texts to medieval urban 
exclusion. This is presumably one of the reasons for the difficulties ( discussed in 
chapter 4) which the civilians experienced in finding an appropriate analogy for the 
medieval Italian bannitus, and why they had to resort to drawing comparisons with 
mechanisms such as deportation that referred to the empire as a whole rather than to 
a given city within it.70 
Since one of the functions of the medieval ban was to act as a coercive procedural 
mechanism,71 and questions of procedure formed an important part of the medieval 
discourse on banishment, we need to be aware of ordinary criminal procedure in 
Roman law, which was a continual reference point for the medieval jurists in that 
context. The first thing to note is that the medieval civilians overwhelmingly 
accepted the view of ordinary criminal procedure they found in their sources.72 For 
them, it was the archetype of criminal procedure: how it ought to be.73 Of course it 
was possible to point out exceptions to the rule: one law in the Code states that 
67 Cf. Mommsen, Romisches Strafrecht, p. 333 ff. 
68 On relegation, see ibid., p. 964 ff. The relevant titles are again D.49.18 and D.48.22. 
69 Cf.ch. 1,pp. 7-10. 
7° Ch. 4, pp. 225-64. 
71 For the distinction between bannum pro contumacia and bannum pro delicto, see eh. 1, pp. 
19-21. 
72 The point has been noted by Cavalca, II bando, p. 85. 
73 See any medieval gloss or commentary to D.48.17 (De requirendis vel absentibus 
damnandis), D. 48.19.5 (I. Absentem), and C.9.40 (De requirendis). More specific examples 
include: Accursius to D.48.17.1, D.48.19.5; Cinus de Pistoia to C.9.2.7, C.9.40. l; Jacobus 
Butrigarius to C.9.40.1; Rainerius de Forlivio to D.48.3.6, D.48.17.4; Albericus de Rosate to 
D.48.17, D. 48.19.5, C.9.40; Bartolus to D.48.17.1, D.48.3.6, D.48.19.5; Baldus to C.9.2.6, 
C.9.40.1, C.9.47.21. 
54 
certain imperial officials can be condemned with a capital penalty in their absence, 
and another one in the Digest threatens the same for those who are engaged in the 
crime of making eunuchs.74 As Bartolus points out, however, these are isolated, 
special cases, and in no way alter the general position of the ius commune.75 As we 
shall discuss in more detail in chapter 4, one of the aspects of the medieval ban that 
the jurists viewed with disapproval was its divergence from the authoritative 
procedural principles of the ius commune. The fact that an absent person could be 
subjected to the severe effects of the criminal ban, a capital penalty, was a derogation 
of the Roman law and had to be justified in relation to the law-giving powers of the 
Italian cities. 76 
5. Confinement and Confiscation of Property 
Besides losing their civil rights, deported persons were confined to a remote island or 
an oasis somewhere in the empire, which they were forbidden to leave on pain of 
death. 77 The reason for this, according to Baldus, was that 'they may die there 
quickly on account of the bad air'. 78 Note that ancient forms of exile, whether 
deportation or relegation, never exiled somebody from the Roman empire in a 
geographical sense; they always confined him somewhere within it, by specifying an 
area, varyingly large or small, that he was forbidden to enter. In the case of the 
deported person, this was everywhere except his island or oasis. In other words, the 
deported person was exiled from the Roman civitas in a legal sense but not in a 
geographical sense.79 Like those non-Roman peregrini who came from beyond the 
empire, deported persons did not lose the protection of the law. Unlike medieval 
banniti, as Rainerius de Forlivio, Bartolus, and Baldus all point out, they could not 
be injured i11 person or property with impunity by other private individuals. Because 
74 C.9.47.21; D.48.8.4,2. 
75 Bartolus to D.48.17.1, fol. 189ra: 'Sed opponitur [C.9.47.21]. Solutio: regulariter absens 
crimine capitali non damnatur. fallit ut ibi. Item, est alius casus in [D.48 .8.4,2].' Baldus also 
accepts C.9.47.21 as an exception in his commentary to that law. 
76 Ch.4,pp. 193-212. 
nD 
.48.19.4; D.48.22 . 7 ,5. 
78 Baldus to C.4.59.1, fol. 131 va: 'Insulae autem assignabantur deportatis ut ibi propter 
malum aerem cito morerentur... ' 
79 In the words of Mommsen, Romisches Strafrecht, p. 973: ' Immer aber ist daran 
festgehalten worden, <lass die Ausweisung sich niemals auf das Gesamtreich erstrecken darf. 
Landesverweisung kennt das romische Strafrecht nicht.' 
55 
they possessed ea quae sunt iuris naturalis, the famous provision of D.1.1.3 also 
applies to them - 'since nature has established a certain fellowship amongst us, it 
follows that it is a grave wrong for one human being to encompass the life of 
another. ' 80 
To his island or oasis the deported person went 'stripped of everything' ('nudus 
exulat').81 Before Justinian, the entire estate of the deported person was confiscated 
by the fisc. Justinian changed the law in a novel that became known in the Middle 
Ages as Bona damnatorum. 82 Here the emperor established that in cases in which the 
old laws laid down confiscation for crimes resulting in death or deportation, property 
was no longer to be confiscated but to be transferred to relatives in the first instance, 
and only if these were lacking, to be taken by the fisc. Significantly, this rule was not 
to apply in cases of Iese majesty. Although the medieval civilians accepted 
Justinian's innovation,83 they nevertheless discussed in some depth the effects of 
property confiscation according to the old laws. There are two reasons for this. 
Firstly, much of the law on deportation in the Digest and Code concerns questions 
over the confiscation and possible restitution of the deported person's estate: 
specifically, the resulting relationship between the estate, now held by the fisc, and 
the deported person's previous debtors and creditors. 84 Secondly, as we shall see in 
chapter 4, in the context of medieval banishment the lawyers most commonly used 
deportation as an analogy with which to solve problems thrown up by the city's 
confiscation of the banished person's estate. 85 In other words, the law on deportation 
had much practical use. It will therefore be worth establishing the basic 
jurisprudence of this field. 
80 Rainerius de Forlivio to D.48.19.17, fol. 128va: 'ibi, habeant. et, nota, non possunt offendi 
impune, ut [D.1 .1.3: ... et cum inter nos cognationem quandam natura constituit, consequens 
est hominem ho mini insidiari nefas esse.].' The translation is adapted from Watson (ed.), 
TheDigestofJustinian, vol. 1, pp. 1-2. Bartolus to D.1.1.3, fol. 6va: 'Sed ego video quod 
deportati non perdunt ea quae sunt iurisgentium, licet perdant ea quae sunt iuris civilis ... Sed 
si hoe est, hie est eis reservata defensio, ergo est illicita offensio.' Cf. Baldus to C.6.24.1. 
81 D.4.5.7,3: 'Item cum civitas amissa est, nulla restitutionis aequitas est adversus eum, qui 
amiss is bonis et civitate relicta nudus exulat. ' 
82 Coll.9.9, § quoniam and§ ut autem (=Nov.134.4/13; § ut autem is inserted as auth. Bona 
damnatorum, post C.9.49.10). 
83 E.g. Accursius to D.28.1.8,1, D.48.20.1; Bartolus to D.4.5.7,3 , D.48.20.1; Albericus de 
Rosate and Baldus to auth. Bona damnatorum, post C.9.49.10. 
84 Especially D.48.23 and C.9.51 (both De sententiam pass is et restitutis). 
85 Ch. 4, pp. 229-30. 
56 
Given that ownership of property as well as contractual obligations associated with 
an individual's estate were derived from the law of nations and the natural law the 
' 
confiscation associated with capiti~ deminutio media was undoubtedly an 
infringement of higher norms. This issue comes to the fore in a law concerning the 
effects of deportation on the contract of deposit. 86 If the depositor is deported and his 
property confiscated, should the depositary hand the property back to the depositor, 
which he would be bound to do under normal circumstances, or should it fall to the 
fisc? 
If we look only to the law of nature and nations, the money should be restored to the 
man who gave it [i.e. the deportatus ]; if we look to the civil law and the legal order, it 
should rather be surrendered to the commonwealth; for a person who has offended 
against the commonwealth ought also to suffer indigence so that by his example to 
others he serves to deter wrongdoing.87 
In other words, the rationale, or causa, for derogating from natural law lies in public 
utility, portrayed here as the benefit gained from the deterrence of future crime. As 
we have seen, public utility was the usual justification in the Roman law tradition for 
transgressing higher norms. 88 
It is striking that the medieval jurists themselves never truly addressed the tension 
resulting from the conflict of civil-law confiscation and the demands of natural law. 
The · authority of the ius commune seems to have been such, and the principle of 
property confiscation in criminal cases so firmly and unquestionably established 
within it, that this derogation of the natural law apparently did not require 
justification. Nicolini's excellent study of confiscation by the prince in the medieval 
civilian tradition is very telling in this regard. In his detailed presentation of the 
jurists' discussion of causa - when the prince was indeed allowed to confiscate the 
property of his subjects and thus derogate from higher norms - it becomes evident 
that they did not ever feel the need to consider the case of penal confiscation. 89 As a 
cornerstone of the Roman law it was simply assumed to be just; causa was assumed. 
In this and the following chapter, we will see that this attitude of the jurists was not 
restricted to the case of property confiscation. Other instances in which the Roman 
criminal law derogated from higher norms were treated as equally unproblematic. 
86 Cf. Berger, Encyclopedic Dictionary, p. 432. 
87 
D.16.3.31. The translation is adapted from Watson (ed.), The Digest of Justinian, vol. 2, p. 
19. 
88 Cf. eh. 1, pp. 27-8. 
89 N. 1· . L ' 1co m1, a proprieta. 
57 
In the context of deportation, the jurists mostly restricted themselves to explaining 
the mechanisms of confiscation, explaining how the deported person's goods, 
including existing obligations of debt and credit, were transferred to the fisc. The 
main points of juristic opinion remained constant from Accursius to Baldus. 
Following the sentence, the fisc possessed the deported person's property as well as 
the legal actions to pursue his previous debtors, but could in tum be sued by his 
previous creditors. If the deportatus received a judicial pardon that included the 
restitution of his estate, he again became liable to his creditors and his debtors once 
again became liable to him unless the obligation had been fulfilled in the meantime.
90 
The only point at which higher norms became an obstacle was in a very technical 
discussion that had little impact on the rules just described. In his Digest 
commentary, Albericus de Rosate posits the obvious problem: given that the 
sentence of deportation is a civil act, it ought not to be able to transfer the estate of 
the deported person to the fisc - in particular his property and contractual obligations 
- given that these are rooted in the law of nations. In his solution, Albericus relies on 
the opinion of Dinus de Mugello and Rainerius de Forlivio that, while confiscation 
could take away the civil part of the deported person's obligation, he would still be 
bound by the obligation's natural part. But since this natural aspect is without effect 
('sine effectu'), it would do nothing to change the fact that the fisc remains both the 
effectual creditor and debtor in respect of the obligations connected to the deported 
person's estate.91 Although Albericus seems to favour this exercise in hair splitting at 
90 Accursius to D.4.5. 7 ,3, D.28.1.8, D.48.20.1 ( confiscation) and to D.48.23 .2/3, 
C.9.51.1/3/4/11/12 (restitution). The agreement of later jurists can be observed in any 
subsequent commentary (or lack thereof) to the titles D.48.23 and C.9.51 - e.g. Cinus de 
Pistoia, Jacobus Butrigarius, Albericus de Rosate, Baldus to C.9.51; Bartolus to D.48.23, 
C.9.51. One question that does not seem to have been addressed by the Gloss concerned 
problems arising from the alienation of property while the fisc possessed the estate. If the 
estate is restored to the former owner, does he now have a claim against a third party to 
whom the fisc had sold a portion of his former property? At C.9.51.13, fol. 123ra, Bartolus 
argues that he does not: ' ... quod deportatus restitutus recuperat bona, an hoe sit verum in 
bonis quae sunt alienata? [ ... ] Certe casus videtur quod non, quia ille qui emit a fisco nullo 
modo potest inquietari ... Nam constat quod civitas interim ... legitime possidebat bona et 
legitime vendidit, et habeatur interim tanquam haeres ... Sed ille qui emit ab haerede, vel 
saltem a bonaefidei possessore, non po test inquietari... Bona vero, quae in eadem causa 
remanent, non alienata, ista recuperat per restitutionem ... ' In eh. 3, pp. 118-21 , we will see 
that Jacobus de Belviso shared this opinion in the context of confiscation resulting from 
imperial banishment. 
91 Albericus de Rosate to D.45.2.19, fol. 104vb: 'vel die, manente, in deportato, scilicet 
obligatione naturali, sed sine effectu, quae in fiscum translata non est... Dy. [ ... ] pro prima 
facit quia sententia deportationis est actus civilis, unde non debet posse tollere obligationem 
58 
one point in his commentaries, on two other occasions he says the opposite, once by 
again citing Dinus (!) and once in his own voice, arguing that both the natural as well 
as the civil aspects of the deported person's obligation are indeed taken over by the 
fi 92 1sc. Whatever the case, the end result - the fisc as effective debtor and creditor -
remains the same. The discussion was settled by Bartolus and Baldus, who agreed 
that the deported person's goods and his obligations, both in their civil and natural 
capacity, were fully transferred to the fisc as legal successor. 93 
6. Cancellation of the Sentence: indulgentia and restitutio 
We will shortly consider in detail the impact of the sentence on the legal position of 
the deported person, in particular on his civil and natural rights. Given that the 
moment of loss of citizenship is revealing in this regard, one might assume that the 
cancellation of the sentence and the retrieval of Roman citizenship would be an 
equally fruitful area of study. Unfortunately, this is not the case. The first thing we 
should note, however, is that the judicial pardon of a criminal sentence is 
fundamentally different to postliminium. It is true that both judicial pardon, at least in 
the case of deportation, and postliminium result in the retrieval of Roman citizenship. 
Postliminium, however, results in the recovery of liberty as well as citizenship 
following enslavement by an enemy, whereas the pardoned deportatus was already 
free and only recovers his citizenship. Postliminium regulates the rules of captivity 
between warring enemies, whereas judicial pardon is an aspect of the Roman 
criminal law. It is probably for these reasons that the medieval jurists did not relate 
their discussions of judicial pardon to the institution of postliminium. Nor does it 
play a part in the discourse on the hostis-transfuga or that concerning medieval 
banniti. 
naturalem, quae est de iure gentium, sicut nee dominium ... Contrarium est verum, scilicet 
~uod per sententiam deportationis tollitur utraque obligatio civilis et naturalis, et transfertur 
m fiscum ... h~c. ~idetur fa I sum, ut no ta tum per Dynum ... quare prim am videtur verius, quod 
~~llatur ~ola clV!hs et rem an eat natural is ... sed sine effectu ... Ray.' 
Albencus de Rosate to D.46.1.4 7( 48), fol. l 13vb: 'ibi, quasi tota, scilicet civilis et 
natur~lis, qua tenebatur ante deportationem. Dy. ibi, extincta sit, et in fiscum translata. Dyn.' 
Alb~r_1~us to D.48.19.17, fol. 203va: 'Item, deportatus liberatur ab omni obligatione naturali 
;} c1v1h qua tenebatur... quia etiam natural is est sublata et in fiscum translata ... ' 
Bartolus to D.4.5.7,3, D.45.2.19, D.46.1.47(48), D.46.2.14,1; Baldus to D.17.1.22,1, 
D.46.1.47(48). 
59 
In the context of deportation, the jurists mostly restricted themselves to explaining 
the mechanisms of confiscation, explaining how the deported person's goods, 
including existing obligations of debt and credit, were transferred to the fisc. The 
main points of juristic opinion remained constant from Accursius to Baldus. 
Following the sentence, the fisc possessed the deported person's property as well as 
the legal actions to pursue his previous debtors, but could in tum be sued by his 
previous creditors. If the deportatus received a judicial pardon that included the 
restitution of his estate, he again became liable to his creditors and his debtors once 
again became liable to him unless the obligation had been fulfilled in the meantime. 
90 
The only point at which higher norms became an obstacle was in a very technical 
discussion that had little impact on the rules just described. In his Digest 
commentary, Albericus de Rosate posits the obvious problem: given that the 
sentence of deportation is a civil act, it ought not to be able to transfer the estate of 
the deported person to the fisc - in particular his property and contractual obligations 
- given that these are rooted in the law of nations. In his solution, Albericus relies on 
the opinion of Dinus de Mugello and Rainerius de Forlivio that, while confiscation 
could take away the civil part of the deported person's obligation, he would still be 
bound by the obligation's natural part. But since this natural aspect is without effect 
('sine effectu'), it would do nothing to change the fact that the fisc remains both the 
effectual creditor and debtor in respect of the obligations connected to the deported 
person's estate.91 Although Albericus seems to favour this exercise in hair splitting at 
90 Accursius to D.4.5.7,3, D.28.l.8, D.48 .20.1 (confiscation) and to D.48 .23 .2/3, 
C.9.51.1/3/4/11/12 (restitution). The agreement of later jurists can be observed in any 
subsequent commentary (or lack thereof) to the titles D.48.23 and C.9.51 - e.g. Cinus de 
Pistoia, Jacobus Butrigarius, Albericus de Rosate, Baldus to C.9 .51; Bartol us to D.48.23, 
C.9.5 l. One question that does not seem to have been addressed by the Gloss concerned 
problems arising from the alienation of property while the fisc possessed the estate. If the 
estate is restored to the former owner, does he now have a claim against a third party to 
whom the fisc had sold a portion of his former property? At C.9.51.13, fol. l 23ra, Bartolus 
argues that he does not: ' ... quod deportatus restitutus recuperat bona, an hoe sit verum in 
bonis quae sunt alienata? [ ... ] Certe casus videtur quod non, quia ille qui emit a fisco nullo 
modo potest inquietari .. . Nam constat quod civitas interim ... legitime possidebat bona et 
legitime vendidit, et habeatur interim tanquam haeres ... Sed ille qui emit ab haerede, vel 
saltem a bonaefidei possessore, non po test inquietari. .. Bona vero, quae in eadem causa 
remanent, non alienata, ista recuperat per restitutionem .. . ' In eh. 3, pp. 118-21, we will see 
that Jacobus de Belviso shared this opinion in the context of confiscation resulting from 
imperial banishment. 
91 Albericus de Rosate to D.45.2.19, fol. I 04vb: 'vel die, manente, in deportato, scilicet 
obligatione naturali, sed sine effectu, quae in fiscum translata non est... Dy. [ ... ] pro prima 
facit quia sententia deportationis est actus civilis, unde non debet posse tollere obligationem 
58 
one point in his commentaries, on two other occasions he says the opposite, once by 
again citing Dinus (!) and once in his own voice, arguing that both the natural as well 
as the civil aspects of the deported person's obligation are indeed taken over by the 
fi ~ h . 1sc. W atever the case, the end result - the fisc as effective debtor and creditor -
remains the same. The discussion was settled by Bartolus and Baldus, who agreed 
that the deported person's goods and his obligations, both in their civil and natural 
capacity, were fully transferred to the fisc as legal successor. 93 
6. Cancellation of the Sentence: indulgentia and restitutio 
We will shortly consider in detail the impact of the sentence on the legal position of 
the deported person, in particular on his civil and natural rights. Given that the 
moment of loss of citizenship is revealing in this regard, one might assume that the 
cancellation of the sentence and the retrieval of Roman citizenship would be an 
equally fruitful area of study. Unfortunately, this is not the case. The first thing we 
should note, however, is that the judicial pardon of a criminal sentence is 
fundamentally different to postliminium. It is true that both judicial pardon, at least in 
the case of deportation, and postliminium result in the retrieval of Roman citizenship. 
Postliminium, however, results in the recovery of liberty as well as citizenship 
following enslavement by an enemy, whereas the pardoned deportatus was already 
free and only recovers his citizenship. Postliminium regulates the rules of captivity 
between warring enemies, whereas judicial pardon is an aspect of the Roman 
criminal law. It is probably for these reasons that the medieval jurists did not relate 
their discussions of judicial pardon to the institution of postliminium. Nor does it 
play a part in the discourse on the hostis-transfuga or that concerning medieval 
banniti. 
naturalem, quae est de iure gentium, sicut nee dominium ... Contrarium est verum, scilicet 
quod per sententiam deportationis tollitur utraque obligatio civilis et naturalis et transfertur 
in fiscum ... h~c. ~idetur falsum, ut no ta tum per Dynum ... quare prim am videru'r verius, quod 
!~llatur ~ola c1v1Its et remaneat naturalis ... sed sine effectu ... Ray.' 
Albencus de Rosate to D.46. l.47( 48), fol. l l 3vb: 'ibi, quasi tota, scilicet civil is et 
natur~lis, qua tenebatur ante deportationem. Dy. ibi, extincta sit, et in .fiscum translata. Dyn.' 
Alb~r.1~us to D.48.19.17, fol. 203va: 'Item, deportatus liberatur ab omni obligatione naturali 
et c1v1It qua tenebatur... quia etiam naturalis est sublata et in fiscum translata ' 93 ••. 
Bartolus to D.4.5.7,3, D.45.2.19, D.46.l.47(48), D.46.2.14,1; Baldus to D.17.1.22,1, 
D.46.1.47(48). 
59 
There are two main titles relating to the cancellation of the sentence: D.48.23 and 
C.9.51 (both De sententiam passis et restitutis). The principal task which the lawyers 
set themselves was to explain the difference between the notions of mere indulgence 
(indulgentia) and full restitution (restitutio in integrum). The most important aspects 
of their thinking remained the same since the days of the Gloss.94 Accursius explains 
that indulgence simply constitutes a remission of the penalty, while restitution 
restores someone completely to the position occupied before the sentence. 95 In 
theory, the mechanism of indulgentia should be highly interesting for our purposes. 
As a major part of the penalty was the loss of civil rights, a focus on its remission 
might reveal what these were. Unfortunately, the jurists never listed the remitted 
penalties but rather appear to assume (as does the law) that their readers would have 
a working knowledge of them. We are in a somewhat better position concerning 
restitutio. Both the law and the jurists draw attention to some of the things which the 
pardoned deported person might regain by restitution though not by mere 
indulgence: good name and standing (Jama); social rank (ordo, dignitas); previous 
relations of patriapotestas; patronage and inheritance rights over former freedmen 
(ius patronatus); higher public offices (honores); and finally, all or part of the 
96 
confiscated estate. 
While the lawyers commonly speak of specific things returned or not returned to the 
pardoned deportatus, Cinus de Pistoia makes us aware that those things can also be 
described as iura. 97 This is another indication that we should not shy away . from 
generally describing the effects of capitis deminutio media in terms of rights. The 
problem, however, is that the jurists do not discuss the relationship between these 
regained rights and citizenship, and thus we do not have any evidence which would 
allow us to place them in the category of iura civilia or otherwise. In other words, we 
do not know whether they are linked to citizenship or to some other status in the law 
of persons which the deported person possessed prior to his deportation. If we know 
that they are associated with civitas, then this is because the lawyers make that 
connection clear in their discussion of the loss of citizenship. 
94 See any subsequent commentary to D.48.23 or C.9.51 - e.g. Cinus de Pistoia, Jacobus 
Butrigarius, Albericus de Rosate, Baldus to C.9.51; Bartolus to D.48.23, C.9.51. 
95 Accursius to Inst.1.12,1, col. 58: 'Die ergo interesse utrum princeps dicat haec verba, 
"restituo te", et tune ad omnia restitutus videtur... an haec, "indulgeo tibi", quo casu de 
poena tantum remittenda cogitasse videtur ... arg. [C.9 .51. 7] , et [C.9.51.4], et [C.9.51.l O] .. .' 
96 Law and Gloss at D.48.23 and C.9.51. 
97 Cinus de Pistoia to C.9.51.13, fol. 570vb: 'aut princeps <licit, "restituo", et tune pristinus 
status et omnia iura sunt restituta.' 
60 
The second problem is that restitution does not refer to a number of general 
capacities restored to someone and open to them in principle by virtue of regaining 
citizenship, but rather to a list of specific things held by an individual in his personal 
position prior to the sentence of deportation. This becomes particularly clear in the 
rules concerning the restitution of patriapotestas. The question is not whether the 
general legal capacity to hold or be subject to this power is returned to the pardoned 
deportatus, but whether those relations of paternal control by which he was 
previously affected are restored.98 Only if we understand this, does the rule that 
patriapotestas is not returned by mere indulgence make any sense.99 This surely does 
not mean that the general capacity is not returned, but merely that the deported 
person is not placed again into those specific familial relations of which he was a 
part prior to the sentence. Given that patriapotestas is a right of citizenship, and 
assuming that indulgentia returns citizenship, then the ability to exercise this right 
must also, at least in principle, have been returned. 
7. Civil and Natural Rights 
i. The Law of Inheritance 
The jurists argue that the deported person loses a number of civil rights upon 
deportation, all of which fall under the heading of private law. The most important 
right, or at least the one most discussed by the jurists, was that of inheritance. The 
Roman law provided a number of ways in which a person could pass on or receive an 
inheritance. The first method was by appointment, and included the legal instruments 
of testament (testamentum ), legacy (legatum ), and trust (fideicommisum ). A typical 
use of trusts and legacies was to ensure that besides the principal heir, who was 
instituted as_ such in the deceased's testament, a third party could benefit from the 
98 
Cf. Law and Gloss at C.9.51.6/9/13. The wording of C.9.51.13 is a good example: 'In 
quaestione testamenti, quad deportati filius remeante patre fecisset, remotis Ulpiani atque 
Pauli notis Papiniani placet valere sententiam, ut in patris sit filius potestate, cui dignitas ac 
bona restituta sunt.' 
99 
C.9.51.6: 'In insulam filio deportato hacque ratione vinculo paternae potestatis exempto, si 
P?St~a ex in?ulge~tia divi Alexandri, ut proponis, reditus in patrium solum praecedensque 
d1gmtas rest1tuta sit, potestas tamen patria repetita non videtur.' Accursius comments, col. 
1842: 'Tu die verius: hie fuit indulgentia, ibi restitutio .. .' 
61 
estate. 100 Although, strictly speaking, legacies and trusts were separate from 
testament, the jurists often referred to the possession of all three institutions jointly as 
a person's testamentary powers (testamenti factio ). These powers were labelled 
'active' when someone passed on an inheritance, and 'passive' if he received one. 
The point that emerges clearly from the discourse on deportation ever since the Gloss 
is that the deportatus is barred from all of these mechanisms, and that this 
prohibition is based on his loss of civil rights.
101 In the words of Jacobus de Arena, 
h . d . ' 102 Th
 
the deported person 'loses bot active an passive testamentary powers . e 
commentators in the fourteenth century especially left little doubt that these powers 
were intrinsically bound up with citizenship. Baldus, for example, states very 
categorically that 'every last will and testament, and every acquisition through them, 
is derived from the civil law' .
103 We should be aware that the law also refers to 
testamentary powers as a right: as the ius testamentifaciendi.
104 Albericus de Rosate 
shows that passive testamentary powers could be separately described as the ius 
succedendi (the right of succession). According to Albericus, the deported person's 
loss of this ius succedendi explains his inability to receive legacies. 
105 
The Roman law contained a second category of ways in which a person could pass 
on or receive an inheritance. These were not dependent on the voluntary appointment 
100 For a useful guide on testaments, trusts, and legacies: Berger, Encyclopedic Dictionary, 
pp. 732-3, pp. 470- 1, and pp. 539-40. According to Berger (p. 471), 'the differences between 
fideicommissa and !egata gradually disappeared and under Justinian both institutions were 
considered equal (per omnia exaequata sunt, D.30.1).' 
'°
1 D.29.1.13,2 (testamenti factio/institution as heir); D.32.1.1,2 (active testamentary 
powers/institution of fideicommissum); D.32.1.7 (beneficiary of fideicommissum); 
D.35 .1.59, 1 (beneficiary of legacy); C.6.24.1 (institution as heir). Accursius thus includes 
testaments, fideicommissa , and legacies in his gloss to D.48.19.17, 1, col. 1411: 'Non 
habeant. ut testari, ut [D.32.1. 1,2]. Item, nee eis legari, ut [D.32.1.7,1].' Cf. Accursius to 
C.6.24.1, col. 1111 , explaining the phrase, 'Qui deportantur, si heredes scribantur, tamquam 
peregrini capere non possunt': 'Non possunt. quia ea quae sunt iuris civilis perdunt, sicu
t est 
civitas et similia, ut hie, et [D.29.1.13 ,2] et [D.48.19.1 7].' 
102 Jacobus de Arena to D.12.6.13 , fol. 78rb-va: ' ... tamen amittat tarn active quam passive
 
testamenti factionem .. .' Cf. Jacobus Butrigarius to C.9.49.2, fol. l 12rb: ' ... possum dic
ere 
quod perdit actus iuris civilis quia non potest testari, ut hie, vel succedere alicui.' 
103 Baldus to C.6.24.1, fol. 69va: ' ... nam omnis ultima voluntas est de iure civili, et acquisitio
 
ex ipsa .. .' Bartolus is equally emphatic at C.6.24.1, fol. l 8va: ' ... nam haereditas de iure ci
vili 
est... sed deportatus perdit civitatem et ea quae sunt iuris civilis, ergo haereditatem 
non 
capit.' 
104 D .32.1.1,2: ' .. .item deportati fideicommissum relinquere non possunt, quia nee testament
i 
faciendi ius habent...' 
105 Albericus to D.35.1.59(58): ' ... a lege per respectum ad personam deportati ius succedendi 
extinguitur. ' 
62 
by a testator and fell under the institution of praetorian possession (bonorum 
possessio). Praetorian possession granted the beneficiary the deceased's estate in its 
entirety or a part thereof, and governed a wide range of inheritance scenarios. For 
example, it established an order of succession in the case of intestacy, and it could be 
used to enforce certain claims against the provisions of the testator's will.
106 One can 
find nothing but agreement amongst the medieval civilians with the clear statements 
of the law that barred deported persons from this type of inheritance as well.
107 In the 
context of deportation, we find the following typical scenario. A father has a son and 
by him a grandson. The son is deported and, since he is therefore held to be dead 
(pro mortuo), loses his praetorian rights of succession in relation to his father's 
estate, that is, they pass over him and now lie in the hands of the grandson without 
any prejudice to the latter's legal position. 108 The deported person's 'death' in this 
sense, which is also described as a 'civil death' (mors civilis), means that he is taken 
out of any familial lines of succession.
109 This includes the usual praetorian rights of 
succession that a patron enjoyed over the estate of his freedman, namely, the right 
(ius) of seeking bonorum possessio against the provisions of the freedman's will. 110 
In this case, too, the deported patron is passed over in favour of his immediate 
descendants.
111 Similarly, in one of his glosses to the Institutes, Accursius explains 
that by deportation 'the right of succession' (ius succedendi) is forfeited, which was 
introduced by the praetor to regulate the order of inheritance amongst relatives in the 
106 
For the diverse types of praetorian possession, their historical development and relation to
 
succession governed by the ius civile, see Berger, Encyclopedic Dictionary, pp. 375-7. Note 
~hat by the time of Justinian praetorian and ius civile succession had been completely fused 
mto one system. 
107 
The most general prohibition occurs at D.37.1.13: 'Edicto praetoris bonorum possessio
 
his denegatur, qui rei capitalis damnati sunt neque in integrum restituti sunt. rei au
tem 
capitalis damnatus intellegitur is, cui poena mors aut aquae et ignis interdictio sit.' Accur
sius 
here ( col. 1635) fully accepts this prohibition and explains that aquae et ignis interdictio is 
'today' replaced by deportation. 
108 
D.37.4.1,8: 'Filium habuit et ex eo nepotem: filius emancipatus vel in potestate manen
s 
deportatus est: quaeritur, an nepoti noceat. et verius est in utroque casu nepo
tem 
admittendum: deportatos enim mortuorum loco habendos.' The same rule is given at D.1.
6.7. 
Cf. Accursius and Bartolus to D.l.6.7, D.37.4.1,8; Baldus to D.l.6.7. 
109 
E.g. Accursius to D.48.19.2, fol. 241v, commenting on 'civitatis amissio': 'Amissio. quae
 
est mors civilis, ut [D.37.4.1,8].' Cf. Baldus to D.35.1.59(58): ' ... de morte civili, per quam 
perditur tota civitas.' 
110 
D.3 7.14.21: 'Sive patronus sive libertus deportetur et post restituatur, amiss um patronatu
s 
et petendae contra tabulas bonorum possessionis ius recipitur: quod ius servatur et s
i in 
metallum patronus vel libertus damnatus restituatur.' 
111 
D.38.2.4,2: ' Si deportatus patronus sit, filio eius competit bonorum possessio in boni
s 
liberti nee impedimento est ei talis patronus, qui mortui loco habetur. ' 
63 
case of intestacy. 112 While Accursius strongly implies that these praetorian rights of 
succession fall under the aegis of citizenship and civil rights, that connection is made 
explicit by Albericus de Rosate. 
113 Albericus affirms the opinion of Dinus de 
Mugello, while discussing the inability of a deportatus to claim bonorum possessio, 
that this ius succedendi is part of ea quae sunt iuris civilis and for this reason lost 
d · 114 upon eportat10n. 
It seems clear, therefore, that non-testamentary forms of inheritance - at least in their 
passive sense - were dependent on citizenship and thus lost by the deported person: 
he could no longer succeed by praetorian possession. The question remains whether 
he himself, although intestate, was able to pass on an inheritance if he died (in a 
natural rather than civil sense) during a state of deportation. This question is raised 
since, as we shall discuss below, the deported person was capable of acquiring and 
possessing property after his sentence while he was deported. One law in the Code 
forecloses any such possibility by stating that a deported person cannot have an heir 
in those things acquired after the sentence and that they too are confiscated by the 
fisc upon his death.11 5 The Gloss on the whole agrees with this argument but does not 
provide any rationale for it.
116 Most of the commentators neither oppose Accursius' 
opinion nor explore it in more depth. 
117 While it is therefore clear that the deportatus 
was not in any way able to pass on an inheritance, we do not know why this was the 
case. In the initial sentence of deportation, property confiscation was a separate 
aspect of the penalty from the loss of civil rights. As we saw above, it is also true that 
a non-Roman peregrinus (rather than a criminal ex-citizen) was able to own property 
112 Inst.1.16,6: 'sed et si in insulam deportatus quis sit, cognatio solvitur.' Accursius 
comments, col. 72 : 'Cognatio. id est, ius cognationis a praetore inductum, ex quo ad 
bonorum possessionem vocantur per unde cognati .. . Cognationis. idest, ius succedendi et ad 
tutelam vocandi perit...' 
113 Accursius to C.6.24.1, col. 1111: 'Non possunt. quia ea quae sunt iuris civil is perdunt, 
sicut est civitas et similia ut hie ... quia pro mortuis habentur, ut [D.37.4.1.8) .. .' Accursius 
also indicates a link between bonorum possessio and civitas at D.48.19.2 - seen. 109 above. 
114 Albericus de Rosate to D.37.4.1,8, fol. 173vb: 'ibi, deportatus est, et sic perdit ea quae 
sunt iuris civilis. sed successio est de iure civili, ergo amittit aliam successionem per 
deportationem. Dyn. [ ... ] ibi, deportati sunt, et sic perdiderunt ius succed_endi, ut supra dixi_.' 
115 C.9.49.2: 'Deportati nee earum quidem rerum, quas post poenam mrogatam habuermt, 
heredem habere possunt, sed hae publicantur.' 
116 Accursius to D.17.1.22,5 and C.9.49.2. 
117 E.g. Albericus de Rosate to D.48.19.17, fol. 203rb, with reference to the opinion o!' Dinus 
de Mugello: 'Respondeo, verum est quod publicatur quicquid habet tempore sentent1ae. sed 
id, quod postea acquirit, non publicatur donec vivit. [ ... ] sed post morte_m eius eti~m _illud 
[publicatur], ut [C.9.49.2), [D.17.1.22,5), et ibi notatum. Dy.' Cf. Cmus de P1stoia to 
C.9.49.2; Jacobus Butrigarius to D.17.1.22,5, C.9.49.2; Bartolus to D.48.20.1. 
64 
within the Roman empire. Should we see the inheritance restriction and confiscation 
r~quired by t~e. C_ode as~ re~ult of the deported person's loss of citizenship (and thus 
his gener~l civil mcapac1ty m the realm of inheritance) or as an additional punitive 
mea~ure _md~pe~dent of _this loss? If it resulted purely from the loss of citizenship, 
the ~mphca~10n IS that, 1f any non-Roman peregrini died during their stay in the 
empire, t~e1r property would be confiscated by the fisc. Although a rather harsh 
~tance, this appears to be the most likely interpretation given that the deported person 
IS excluded from all other forms of inheritance because of his loss of civil rights. 
Indeed, such a conclusion is suggested by Baldus, the only jurist to discuss this 
pr?blem in more detail. He affirms the provision of the Code - the deported person ' s 
th~ngs _fall to ~he fisc as vacant goods (bona vacantia) upon his death - explaining 
this with the idea that the deportatus is pro mortuo, citing a number of laws that 
explain the deported person's loss of testamentary powers and praetorian 
· 118 . 
possess10n. And as we have Just seen, the jurists firmly established these within 
the realm of civil rights and citizenship. 
We should therefore conclude that, as a general rule, the deported person was barred 
from every right of inheritance, and that this prohibition resulted from his loss of 
citizenship. The Roman law of inheritance, however, is highly complex and 
encompasses a vast number of rules and regulations, exceptions and special cases. In 
the c~ntext of deportation there are three exceptions that we need to discuss. The first 
cons_titutes a special favour granted to a soldier (miles). The deported person retained 
~as~1ve testamentary powers in respect of the will of a miles, i.e. he could be 
mstituted as heir in it, or receive a legacy or jideicommissum through it. 11 9 The law 
states ~hat, while this dispensation applies in respect of a deported person, it does not 
apply m respect of a slave. Bartolus explains this distinction with the enticing idea 
118 
Baldus to D.46.1.4 7~ 48), fol. 28va: 'Et per hoe patet quod quaesita post confiscationem 
non quaer~mtur. fisco v1gor~ confiscation is, sed forte tanquam bona vacantia si non habet 
haeredem. While t?e conditional 'si' is odd here, Baldus is emphatic that the deportatus 
does not have an h~1r at ~?49.2, fol. 228va: 'Sed in naturali acquisitione deportatus accipit 
cum effectu et ~ap1t ~estJtut10nem donec vivit, sed in morte fiscus occupat, cum haeredem 
habere non poss1t. et 1ta loquit_ur Jex nostra, et hoe puto verum. accedat ad hoe [D .3 7 .11.1,8]. 
~ 9am pro mortuo rep~tatur, qma s~ntentiam passus est, ut [D.37.4.1,8), [D.35.1.59]. .. ' 
. J?.29.1.13(1~),2_: _Et deportatJ et fere omnes, qui testamenti factionem non habent, a 
~mhte heredes mstJtm possunt.' The Gloss, col. 535, accepts that normally 'hi non capiunt 
mre commu?e, ut [C.?.24.1] e_t [D.37.1.13], sed ex testamento militis sic, ut hie et 
[D.32.1.7,1). Hence this exception applies also to trusts D32 I 71· 's· ·1 d fid · · . · · · · , . 1 m1 es eportato 
1 e1comm1ssum reh~uerit, verf us est, quod et Marcellus probat, capere eum posse., The 
Gloss'. col. 890, agam recogmses that this occurs 'iure singulari, ut [D.29.1.13 2]'. Cf 
Albencus de Rosate to D.29.1.13(14),2 and D.32.1.7,1. ' · 
65 
that the deported person's ability in this regard is somehow derived from the law of 
nations - he is capax de iure gentium while the slave is not.
120 Unfortunately, 
Bartolus does not go on to explain how a deportatus can be capax de iure gentium. 
Baldus, in a singular comment, takes this idea one step further and declares that 'the 
testament of a deported miles is derived from the law of nations, not the civil law, 
and therefore pertains actively and passively to the deported person.' 
121 In other 
words, Baldus implies that the ius gentium also contains some modes of inheritance 
available to peregrini. Although this argument clearly goes against the entire 
tradition of viewing inheritance as an institution of the ius civile in the discourse on 
deportation, Baldus does not explore it further. 
The other two exceptions in the law of inheritance take into account the deported 
person's continuing life on his island or in his oasis - the fact that he died a civil 
rather than a natural death. This first situation arose from the conditions of a 
fideicommissum which had specified that a given person, who now happened to be 
deported, had to choose a third party as its beneficiary. The deportatus retained this 
faculty of selection (facultas eligendi) until his natural death.122 This is what the 
jurists called arbitration (arbitratio) ,123 and as Bartolus and Baldus make clear, such 
a faculty could also be described as someone's right, as the ius eligendi.
124 Despite 
being someone 's ius, the jurists, as early as Accursius, also placed this faculty into 
the realm of fact (factum)- consistere infacto. 125 
120 Bartolus to D.29.1. 13(14),2, fo l. 152ra: ' ... per quae militibus omnis solemnitas iuris est 
remissa, videtur concessum quod possunt institui omnes quae sunt capaces de iure gentium. 
Sed si non sunt capaces de iure gentium, nee de iure civili, tune non possunt institui.' 
121 Baldus to D.29. I . I 3(1 4),2, fo l. 78va: ' Deportati testamentum militis non est de iure civili, 
sed de iure gentium, et ideo competit deportato active et passive. h.d. notabiliter.' 
122 D.3 I. I. 77(79),4: 'Hereditatem filius cum moreretur fili is suis vel cui ex his voluisset 
restituere fuerat rogatus: quo interea in insulam deportato eligendi facultatem non esse poena 
peremptam placuit nee fideicommissi condicionem ante mortem fili i heredis exsistere .. .' The 
Gloss, col. 845, clarifies: 'Mortem. naturalem.' There is a similar scenario at D.36. 1.18,6. 
Cf. Albericus de Rosate to D.31.1.77(79),4 and D.36.1.1 8(1 7),6. 
123 E.g. Bartolus to D.36.1.18(1 7),6, fol. 145ra: 'sed video quod deportatus potest esse 
arbitrator, ut hie, et [D.3 1. I. 77(79),4]. .. ' 
124 Ibid.: 'Per deportationem non perditur ius eligendi fideicommissum .. .' Cf. Bartolus and 
Baldus to D.31.1. 77(79),4. 
125 Accursius to D.36.1.18(17),6, col. 1474: 'Haec autem electio, quae datur, consistit in 
facto, ut hie et [D.31.1.77(79),4].' Baldus to D.31.1.77(79),4, fol. I 56vb: 'Ius eligendi 
deportatus non perdit... Nota quod ea quae sunt facultatis et consistunt in facto, non 
amittuntur per deportationem.' 
66 
The second scenario follows this pattern. D.4.5 .10 (1. Legatum) allows deported 
persons to receive an annual or monthly legacy 'because such a legacy consists more 
in fact than in law' . 
126 Accursius explains that this legacy is not really part of the 
realm of law because it does not matter whether the deportatus 'can receive an 
inheritance by testament or not'; all that matters is the 'fact' that he can eat and 
drink. It is thus 'a special favour of sustenance' designed to ensure that the deported 
person has a certain minimum level of housing and food. 127 As Baldus explains, it is 
not right that someone, who has not been condemned to death, should be deprived of 
the basic means of life.
128 Again, although this legacy exists in the sphere of factum, 
it is also ius: Accursius describes it both as a 'thing' derived from the natural law and 
as one of a deported person's iura naturalia. 129 Baldus too explains that this legacy 
consists 'in fact, i.e. in the right of nature [iure naturae], of which the deported 
person is not deprived'. 130 
In these two cases the jurists establish a conceptually difficult (and not entirely clear) 
relationship between ius and factum. One set coordinate is that both the capacity to 
arbitrate a fideicommissum as well as the capacity to receive a subsistence legacy 
form part of the realm offactum. The jurists seem to base this language offactum on 
the idea that these capacities are basic human functions: the ability to chose one 
person over another and the ability to eat and drink. They are factual because they 
relate to what humans can naturally do as human beings, physically and mentally, 
without any legal artifice . This naturalness, however, also provides their connection 
to ius. This ius cannot be civil because they have no capacity for the civil - hence the 
opposition ius (civile) -factum. Their ius is natural instead because it encompasses 
their natural human faculties. Only in this sense is ius (naturale) also factum . In other 
126 
D.4.5.10: 'Legatum in annos singulos vel menses singulos relictum vel si habitatio 
legetur, morte quidem legatarii legatum intercidit, capitis deminutione tamen interveniente 
perseverat: videlicet quia tale legatum in facto potius quam in iure consistit.' 
127 
Accursius !o D.4.5.10, col. 457: 'Quam in iure. quia non consideratur an possit capere ex 
testamento vel non, sed hoe factu, an possit comedere, vel non, vel bibere. nam et si non sit 
cap~x admi~itur ... Et est hoe speciale favore alimentorum. idem in habitatione, quia nee sine 
ea v1vere qms potest.' 
128 
Baldus to D.4.5. I 0, fol. 250rb: 'Et nota quod non licet aliquem condemnatum occidere 
cum non possumus ei alimentorum subtrahere nisi esset condemnatus ad mortem ' ' 129 · ' •.. 
Accursms to D.48.19.17,1, col. 1411: ' ... et alia quae iuris gentium sunt bene possunt, ut... 
[D.4.5.10].' To D.36.1.18(17),6, cols 1473-4: ' ... et sic ea quae sunt iuris civilis amittit non 
n~~rali~, ut... [D.4_.5.10]. . .' To Inst. I. I 5,3, col. 70: 'id est, certe non tolluntur c~pitis 
d1mmut1one cum smt immutabilia haec iura... Item, nee ea quae natural em aequitatem 
habent, ut... [D.4.5 .10].' 
13
~ Baldus to ~.4.5.1 O! fol. 250rb: 'Legatum. in facto, id est, in iure naturae, a quo capite 
mmutus non pnvatur, 1d est, esu et potu.' 
67 
I 
I 
words, the jurists aligned the categories of factum and ius naturale and opposed them 
to positive ius civile. 
Students of medieval political thought will notice a striking similarity between these 
ideas and William of Ockham's thought on the Franciscan poverty dispute. In his 
Opus nonaginta dierum, Ockham justified the notion that the friars' usus facti was 
entirely removed from the realm of positive law with the argument that it was a 
natural right of use (ius utendi naturale). 13 1 In the discourse on deportation, we can 
see that the medieval civilians, both prior and contemporary to Ockham, also 
described certain factual abilities of the deported person, including his ability to use 
the consumables necessary for human life, as his natural rights. The parallels did not 
escape the commentators of the fourteenth century. They discussed a number of 
issues arising from the legal status of the Friars Minor in the context of arbitration 
and the 1. Legatum. Albericus de Rosate and Bartolus, for example, pointed out that 
the 1. Legatum could be used to argue that the Franciscans are equally capable of 
receiving such subsistence legacies. 132 Baldus, while not mentioning the Franciscans 
explicitly, must have had them in mind when he associated the ius naturae of the 1. 
Legatum with the term usus facti. 133 We should not, however, overstate the 
similarities between the civilians' ideas and Ockham's. As we will see below, one of 
the defining features of the civilians' natural rights was that they gave rise to actions 
which could be enforced in the courts of the civitas. This was precisely one of the 
outcomes that Ockham wanted to avoid. 
Up to this point, we have focused on the problem of inheritance. If the discussion 
seemed extensive at points, it is worth recording that this issue has pride of place 
both in the law and in its medieval glosses and commentaries. We should not 
underestimate the importance of inheritance practices, and hence family power and 
strategy, to both late-Roman and late-medieval society. Lines of kinship were 
carefully defined, cultivated, and exploited, especially among the ruling classes for 
131 For references to the literature on Ockham's theory of natural rights, see eh. I, p. 26, n. 
86. 
132 Albericus de Rosate and Bartolus to D.4.5.1 0. 
133 Baldus to D.4.5.10, fol. 250rb: 'Legatum ad alimentum vel usum facti in facto consistit et 
plurimum facti habet, et ideo capitis diminutio non tollitur, etiam mero iure. h.d. lex bona, 
quae quotidie allegatur. Nota quod in legato alimentorum consideratur usus facti, et ideo 
etiam cuicunque incapaci possunt legari alimenta, nee est impedimentum ius civile, quia 
habet illud legatum plus facti quam iuris. ' 
68 
the sake of economic and political advantages over their rivals.134 So far we have 
observed that the deported person, through loss of citizenship and its associated civil 
rights, was lifted out of the Roman system of inheritance and thereby lost the ability 
to benefit from existing family lineages. The next section concerns not so much the 
way property was passed along these lineages but how these complex lines of 
relation were established in the first place. In this subject matter, too, Roman 
citizenship opened up the way to significant rights unavailable to foreigners . 
ii. Family law 
An important institution of Roman family law was patriapotestas, which constituted 
a set of legal prerogatives or powers that the male head of the family (paterfamilias) 
possessed over his children born in legitimate wedlock, and which usually lasted 
until he died or emancipated them. For example, a son-in-power (filiusfamilias) 
could not make his own testament, marry without his father's consent, and, strictly 
k. 1 · 13s spea mg, cou d not have property of his own. In the context of deportation, the 
medieval jurists accepted the axiom of the law that patriapotestas 'is a right [ius] 
proper to Roman citizens' . 136 As a result it had to be lost through deportation and the 
concomitant loss of citizenship. In the words of Accursius, deported persons ' do not 
retain power over their children ' since ' this power [potestas] is derived from the civil 
1 , 137 v · d d aw . ice versa, eporte sons and daughters were removed from the control of 
their fathers had they not been emancipated by them previously. 138 
134 
In lieu of a large literature, specifically in the context of northern Ita ly, see Lansing, The 
Florentine Magnates. With particular relevance to the theme of banishment and exile: 
Baxendale, 'Exiles in Practice '; Ganz, 'Paying the Price' · Kent 'I Medici in esilio' 
135 • . . ' ' • 
Cf. Berger, Encyclopedzc Dzctwnary, p. 621; Buckland and Stein, A Text-Book of Roman 
Law, pp. I 02-4. 
1
~
6 
~ccursius to C.5.17. 1, col. 891 : 'Item, quia potestas patria propria est Romanorum 
c1vmm, ut_ [D.1.6.3: Item in potestate nostra sunt liberi nostri , quos ex iustis nuptiis 
procreavenmus: quod ius proprium civium Romanorum est.].' 
131 A . 4 ccursms to D. 8.19.17,1, col. 1411 : 'Non habeant... Item, nee filios retinent in potestate, 
ut [Inst: 1.12'. I] ,_ quae potestas est iuris civilis ... ' Inst. I.! 2, I: 'Cum autem is, qui ob aliquod 
maleficmm m msulam deportatur, civitatem amittit, sequitur ut, quia eo modo ex numero 
civium Romanorum tollitur, perinde acsi mortuo eo desinant liberi in potestate eius esse.' 
138 
~n~t. .1.12, I : ' pari ratione et is, qui in potestate parentis sit, in insulam deportatus fuerit, 
desm1t m potestate parentis esse.' 
69 
The issue of patriapotestas leads to a number of complications since it intersects 
with the areas of marriage and legitimacy of children. Several laws make it very 
clear that a marriage is not automatically dissolved by the deportation of either 
spouse - a principle accepted by all the jurists. 139 The marriage is only dissolved if 
the spouse of a deported person changes their 'marital affection' after the sentence 
and goes on to seek a divorce. 140 This already suggests that the institution of 
marriage was not dependent on the possession of citizenship and civil rights. Indeed, 
Accursius argues that in this respect the deported person 'is not held to be dead', 
141 
and explicitly names marriage as one of the ea quae sunt iuris naturalis not lost by 
the deportatus. 142 In this he was probably influenced by the general medieval 
conception of marriage as a divine institution, and by the close connection between 
divine and natural law in civilian jurisprudence. 143 
Questions arose over the legal position of children born from a marriage that 
withstood a sentence of deportation. The jurists agreed that children conceived by a 
deported father after his sentence would be free and legitimate ( as they were born in 
lawful matrimony), but not Roman citizens, nor subject to patriapotestas. 144 The 
deportatus would be their 'natural' rather than their 'civil' father.
145 Accursius 
argued that this would also be the case if the child were conceived after the mother 
had been deported even though the father remained a Roman citizen.
146 The Gloss, 
however, was roundly reproved by subsequent jurists in this regard. Jacobus de 
Arena was the first to point out that Accursius' opinion does not accord with the 
principle of the law that a child conceived in lawful marriage should follow the 
139 D.24.1.13,1; D.48.20.5 ,1; C.5.16.24,2; C.5.17.1. Dinus de Mugello's comment to 
D.24.3.56(57) could be viewed as an exception, although a degree of ambiguity remains. 
14° C.5.16.24,2; C.5.17.1. 
141 Accursius to D.35 .1.59(58), fol. 180r: 'non ergo habetur pro mortuo deportatus, nee 
solvitur matrimonium ... ' 
142 Accursius to C.6.24.1, col. 1111: 'sed ea quae sunt iuris naturalis non, ut matrimonium ... ' 
143 Cf. eh. 1, pp. 27-8. 
144 Bartolus to D.36.1.18(17),5, fol. 144vb: 'Quaero an post deportationem concepti et nati 
sint legitimi? Respondeo, sic, quia matrimonium durat, [C.5.17.l]. Quaero an nascantur in 
potestate? Respondeo, per deportationem perdit ea quae sunt iuris civilis ... sed patriapotestas 
est de iure civili ... ergo filii non sunt in potestate patris. Et sic habes casum in quo filii 
nascuntur legitimi et non in potestate, secundum la. de Are.' Cf. Accursius to D.1.5.18, 
C.5.17.1; Jacobus de Arena and Albericus de Rosate to D.36.1.18(17),5. 
145 Jacobus de Arena to D.36.1.18(1 7),5 , fol. 127vb: 'vel die naturalem potius eum patrem, 
civilem non ... ' 
146 Accursius to D.1.5 .18, col. 44, with reference to a deported female: 'sed quid si concepit 
et peperit post [ deportationem ]? Respondeo, erit liber et non est in potestate patris ... sed non 
civis ... ' 
70 
condition of his father rather than his mother. 147 Instead, the jurists argued, the child 
will always be a Roman citizen and in .his father's power as long as the father retains 
Roman citizenship. Again, only if a child is conceived following the father's 
deportation, will that child not be a Roman citizen. 148 We should therefore note that 
Roman citizenship was patrilineal in the thought of the medieval civilians. Civitas 
was acquired by birth in the context of a lawful marriage when the child's father was 
a citizen. Secondly, we should note that children conceived before their father's 
deportation, although born afterwards, are treated as if they had been born before, 
and, though evidently not under the deported father's power, become citizens and are 
integrated into the broader legal structures of their family. This is especially 
important in the field of inheritance, where these children could benefit from existing 
family lineages, especially by praetorian possession. On the other hand, those 
children conceived after their father's sentence, who thus do not possess citizenship, 
are also not in any way integrated into the broader kin group; it is as if they had been 
born to someone outside the family. 149 · 
As I mentioned in the introduction to this chapter, we will come across a number of 
rights, a number of subjective iura, that do not fit the modem notion of subjective 
right as entitlement. Patriapotestas is an example where subjective ius can refer 
equally to entitlement as well as to duty. As the name indicates, this institution 
represented in the first instance the power of the paterfamilias over his children. This 
ius was lost by the deportatus upon his sentence along with his other iura civilia. The 
deported person, however, did not have to be a paterfamilias; he could also be a 
filiusfamilias. The ius patriaepotestatis lost by the deported son-in-power was 
147 
Jacobus de Arena to D.1.5.18, fol. 65rb: 'In glossa, "id est deportata" ... vel intellige 
g!o~sam quando concepit post damnationem, sed contradicit [D.1.5.19]. et ideo non hie bene 
d1c1t. Eadem glossa, ibi "ex facto", sed ibi in patre deportato, hie in matre.' D.1.5.19: 'Cum 
legitimae nuptiae factae sint, patrem liberi sequuntur: volgo quaesitus matrem sequitur.' 148 
The cleare~t summary of this matter is given by Jacobus Butrigarius to D.1.5.18, p. 17b: 
'In ?l,?ss~ quae incip_it . "id ~st dep~rtata", ibi "respondeo, erit liber et non est in potestate 
patns ... 11lae leges mh1I facmnt, qma loquuntur pro damnato, hie est contra in matre. [ ... ] si 
au~em mater ~it damnata, ut hie, et non ad mortem, tune aut pater, qui remansit in civitate, 
retme~ affechonem et uxorem, et tune filii semper erunt liberi et in potestate, ut hie ... ' 
Albencus de Rosate, at D.1.5.18, quotes in full Jacobus Butrigarius ' commentary to this law. 149 
The locus classicus for this argument is a complicated scenario from the law of 
fideicommissa, in which a deported male, whose children had been conceived after 
deportation, is considered to have died without children that would need to be taken into 
account in the _succession of property amongst the family group - see D.36.1.18(17),5 
( ... conceptos _qmdem ante deportationem, licet postea edantur, efficere, ut condicio deficiat, 
post deportat10nem vero susceptos quasi ab alio non prodesse ... ) with its Ordinary Gloss and 
subsequent commentaries by Jacobus de Arena, Albericus de Rosate, and Bartolus. 
71 
therefore qualitatively different to that of his father. At one point, Accursius 
describes patriapotestas as a form of 'love', introduced on account of the 'affection' 
the Romans had for their children. 150 He explains that this sets the Romans apart, 
since some peoples 'keep their children like slaves' while others keep them 
'completely free' .151 So, although Roman children might not have to suffer as much 
as those of more barbaric nations, patriapotestas is nevertheless a limit upon their 
freedom, a restriction rather than a legal advantage. This is why, in the case of 
judicial pardon, when a deported son who has received restitution falls again under 
the power of his father, this is said to happen to the son's detriment - patriapotestas 
is described in this context as a chain or bond. 152 In other words, the ius 
patriaepotestatis of an unemancipated son was a duty in the sense of an immediate 
obligation towards his father. 
Another area where deportation intersects with family law is that of dowry. We will 
recall from our discussion of D.4.5.8 (1. Eas) in chapter 1 that this law concerned the 
ability of a wife to reclaim her dowry from her husband using the action for dowry 
(actio dotis). We saw there how the 1. Eas associated the actio dotis with equity, 
natural obligation, and incorruptible iura naturalia, giving this connection as the 
reason why a woman retains possession of this action even after capitis deminutio. 
The implication was that the wife's claim to her dowry was her natural right - her ius 
naturale. 153 We will now see that this was the interpretation of the medieval jurists. 
An interesting development was that the latter not only viewed as a natural right the 
wife's underlying claim, but also her ability to enforce this claim through the actio 
dotis. 
The 1. Eas, however, is not the only law concerned with this theme. Others show that 
the Romans generally viewed it as equitable and socially desirable that women 
150 Accursius to D.49.15.19,7, col. 1514: 'Charitas. id est, patria potestas, quae fuit inducta 
propter affectionem liberorum iure civili Romanorum, ut [Inst.1.9,2].' 
151 Accursius to Inst.1.9,2, cols 40-1 : ' Romanorum. id est, omnium qui sunt de Romano 
imperio. aliae vero gentes quaedem ut servos tenent filios , ut Sclavi, aliae ut prorsus 
absolutos, ut Francigenae.' 
152 C.9.51.6: ' In insulam filio deportato hacque ratione vinculo paternae potestatis 
exempto ... ' Accursius, col. 1842, points out in his gloss to this law that some consider a son 
who has received restitution not again to fall under previous patriapotestas 'quia sic ei 
restitutio noceret'. 
153 D.4.5.8: 'Eas obligationes, quae naturalem praestationem habere intelleguntur, palam est 
capitis deminutione non perire, quia civilis ratio naturalia iura corrumpere non potest. itaque 
de dote actio, quia in bonum et aequum concepta est, nihilo minus durat etiam post capitis 
deminutionem,' Cf. eh. 1, p. 30. 
72 
should not be left without a dowry. 154 Hence, the law institutes a number of rules to 
the effect that a wife can recover her dowry with the actio dotis from her husband or 
from the fisc in case either spouse is deported and the marriage is subsequently 
dissolved. 155 For our purposes, the most interesting case concerns the deported and 
divorced wife. If she had been unemancipated at the time of deportation then her 
father will bring the action to recover the dowry as he would in any other scenario 
involving the actio dotis . If, however, she had been emancipated (materfamilias) and 
would normally herself bring the actio, the law insists that this capacity of hers is not 
lost upon deportation but granted 'from considerations of humanity'. 156 
It is surely this equity inherent in the actio dotis that convinces Accursius to list it 
both as one of ea quae sunt iuris naturalis and as one of the iura naturalia not lost 
by the deported person. 157 The law to which Accursius refers in these contexts is the 
1. Eas, which, as we know, links this actio to iura naturalia. The commentators bring 
out even more clearly what was already evident in the Gloss. Bartolus explains the 
logical corollary of the idea that the actio dotis is a ius naturale. If the action is 
natural, then the underlying obligation which it follows must be so too: hence, he 
writes, 'the contract of dowry is derived from the law of nations.' 158 Baldus, in his 
commentary to the 1. Eas, rephrases this argument in terms of subjective of ius. The 
ability of the woman to bring the actio dotis for a contract of dowry, which is derived 
from the law of nations, is her ius dot is. 159 
154 
E.g. C.5.17.1: 'ideoque dotis actio ipso iure non competit, sed indotatam esse earn, cuius 
laudandum propositum est, nee ratio aequitatis nee exempla permittunt. ' 
155 If h . . . 
. t e mamage 1s not dissolved upon deportation and if the wife is deported, then the 
dowry remains in the hands of her husband as before. If the husband is deported and his 
estate is confiscated as a consequence, including his wife's dowry, then the wife can make a 
petition to recover it from the fisc. The Gloss explains these rules at D.48 .20.5 and C.5.17.1. 
Cf. Jacobus de Arena, to D.48.20.5,1. 
156 
D.48.20.5,1: 'sed si mater familias sit et interim constante matrimonio fuerit deportata, 
dotem penes i:naritum remanere: postea vero dissoluto matrimonio posse earn agere, quasi 
humanitatis intuitu hodie nata actione.' (The translation is from Watson (ed.), The Digest of 
Justinian, vol. 4, p. 369.) Accursius, col. 1423, confirms: 'sed <licit potest humanitatem esse 
in hoe, quod per poenam non solvitur [matrimonium] vel quod non perdat dotem ' 157 , ' . 
Acc~rsms to D.36.1.18(17),6, cols 1473-4: ' ... et sic ea quae sunt iuris civilis amittit, non 
naturahs, ut... [D.4.5 .8]. .. ' To Inst.1.15,3, col. 70: 'id est, certe non tolluntur capitis 
diminutione cum sint immutabilia haec iura. .. Item, nee ea quae natural em aequitatem 
habent, ut [D.4.5.8]. .. ' 
158 
Bartolus to D.4.5.8, fol. 141 vb: 'Nota quod contractus dotis est iuris gentium ' I~ . 
Baldus to D.4.5.8, fol. 250rb: 'et allegatur haec lex quod contractus dotis et etiam actio de 
?ote e~t de iuregentium. [ ... ] Nota quod per emancipationem filiae dotatae non corrumpitur 
ms dotls ... et est ratio, quia ius dotis consistit in iuregentium ... ' 
73 
Given this strong emphasis on the natural equity of granting even divorced criminal 
women their dowries and on the inviolability of the action to recover them, it seems 
remarkable that the jurists accepted a major exception to this rule without explaining 
its rationale or the tension it causes. The law states that when a woman was 
sentenced for a small number of particularly severe crimes, such as Iese majesty or 
parricide, her dowry was confiscated as part of the sentence and she could not bring 
the actio dotis to recover it. 160 Although the jurists discussed various intricate issues 
surrounding the confiscation of a dowry, we do not find out how it is that the type of 
crime causes possession of the dowry and actio to be equitable, natural, and inviolate 
in one case but not the other. 161 Presumably, some crimes were just so severe that 
their punishment warranted derogation from higher norms in this regard. The tacit 
assumption seems to have been that the greater the crime, the less protection higher 
norms afford against punitive measures of the civil law. The fact that we cannot go 
beyond these rather vague inferences is another illustration of what we already 
observed in the case of property confiscation: the jurists did not explain or justify the 
ability of the Roman criminal law to transgress higher norms. 
We should also note that the civilians normally considered actions to be derived from 
the civil law. Yet here we have an example of an action that is somehow rooted in 
the natural law or the law of nations. 162 Kenneth Pennington has demonstrated that, 
from the later thirteenth century onwards, the civilians increasingly placed acti_ons in 
the realm of natural law if these were based on claims that were derived from higher 
norms, and that these actions received a greater degree of protection from 
mechanisms of the civil law as a result. 163 The evidence from deportation, broadly 
speaking, fits this narrative, although Accursius, who composed his Ordinary Gloss 
mostly in the fust half of the thirteenth century, would be an early outlier if 
Pennington's periodisation is accurate. Already in the Gloss, it was the natural right 
160 D.48.20.3: 'Quinque legibus damnatae mulieri dos publicatur: maiestatis, vis publicae, 
parricidii, venefici, de sicariis.' Hence Accursius to D.4.5.9, col. 457: 'Et quandoque 
emancipata. Idem videtur etiam si maximam vel mediam capitis diminutionem sit passa, nisi 
una de quinque legibus fuit damnata, ut [D.48.20.3].' 
161 Cf. Albericus de Rosate to D.48.20.3/5; Bartolus to D.48.20.3. 
162 For Bartolus, part of the equity inherent in the actio dotis was that this action, which was 
civil, was granted for a contract derived from the law of nations - at D.48.20.3, fol. 204ra: 
'Quaero in quo est ista humanitas? Quicquid dicat glossa. die quod ista humanitas est hie in 
hoe, quod mulieri damnatae quaeritur actio de dote, quae est de iure civili, ipse tamen 
contractus dotis est de iure gentium, [D.4.5.8].' 
163 Pennington, The Prince and the Law, p. 152 ff., discusses actions and higher norms in 
relation to aspects of judicial procedure. For the same considerations, see also his subsequent 
articles 'Due Process, Community and the Prince', and 'Law, Criminal Procedure'. 
74 
of the deported woman to bring an action in the courts for the recovery of her dowry. 
When we discuss the deported personf contractual capacities below, we will see that 
this was not an isolated case, but that they possessed a broader judicial capacity to 
pursue their natural rights. 
Our discussion of family law so far should serve well as an introduction to the most 
complex topic linked to deportation within that field: the rights of agnation and 
cognation, which intersect with the material on inheritance and patriapotestas. Given 
how technical the subject matter is, a short outline of the basic legal ideas will be 
worthwhile. Broadly speaking, in Roman law members of a family group are 
cognates to each other either by virtue of their natural blood relations or through the 
establishment of a civil cognation through adoption. Agnates become such through 
falling under the patriapotestas of the same head of a family or who would do so if a 
common male ancestor were still alive. Agnates are normally also cognates and thus 
form a restricted circle based on a common male line within the broader kin group. 164 
The Ordinary Gloss to the Institutes explains the significance of these diverse 
familial ties and how they are affected by capitis deminutio. In essence, the position 
as agnate or cognate granted a certain 'right of succession and of being called to take 
up legal guardianships of minors' when these had to be distributed amongst family 
members according to a complex system of priority, particularly in cases of intestacy 
- ius succedendi et ad tutelam vocandi.165 Accursius clarifies that before Justinian 
formally equalised the position of agnates and cognates, agnates were prioritised in 
the order of distribution of inheritances and guardianships. 166 
At I.1.15,3, the law establishes the legal effects of capitis deminutio on agnation and 
cognation. Note that my translation already reflects Accursius' interpretation of this 
difficult passage. 
164 
Berger, En_cyclopedic Dictionary, p. 358 (agnation) and p. 393 (cognation). 
165 
Accursius, to Inst.1.16,6, col. 72: 'Cognatio. id est, ius cognationis a praetore inductum ex 
quo ad bonorum possessionem vocantur per unde cognati... Cognationis. id est, ius 
succedendi et ad tutelam vocandi ... ' 
166 A . I 
cc.urs_ms_to nst.1.15,3, cols 69-70, a.v. sedagnationis: ' Est autem ius agnationis et ius 
cognat1001s stve cognatorum. Agnati enim habent ius praerogativum ad cognatos, scilicet, ut 
praeferantur in haereditatibus agnatorum mortuorum... Item, praeferuntur in tutelis 
agnatorum parvulorum ... his au tern deficientibus, in haereditatibus et tutelis cognati suo iure 
admittuntur, inter quos etiam agnati vocantur qui desierunt esse · in iure agnatorum vel 
desierunt habere iura agnatorum, ut per capitis diminutionem .. . ' Accursius acknowledges 
later developments further on in his remarks to the same law, col. 70, a.v. utique: 'Hodie 
vero istam differentiam agnationis quaerere non oportet, cum agnati ante cognatos nihil 
habeant, ut [Coll.9.1 § ex his (=Nov.118.5)].' 
75 
Sed adgnationis quidem ius omnibus modis capitis deminutio?e pleru~que perimitur: 
nam adgnatio iuris est nomen. cognationis vero ius non omn~bus modts c~mmutatur, 
quia civilis ratio civilia quidem iura corrumpere potest, naturaha vero non utique. 
But the right of agnation is indeed commonly lost by all forms ?f capitis demi~utio, 
for agnation is the name of a [civil] right. The right of cognation, however, is ?~t 
affected by all forms [ of capitis deminutio ], since a civil rule can indeed destroy c1v1l 
rights but natural ones not at all. 
In his gloss, Accursius explains that 'agnation is the name of a civil right', which 
could thus ' easily be taken away' by another provision of the positive law. 167 Hence 
he claims 'there is no doubt' that rights of agnation are lost 'through every kind of 
capitis deminutio', including deportation.168 This should not surprise us, given that 
the institution on which agnation depended, patriapotestas, was also a civil right. 
The matter becomes intricate when Accursius turns his attention to cognation. 
Initially, his treatment is unproblematic. Accursius explains that by 'naturalia' in this 
passage we should understand the 'rights of cognation ' (iura cognationis), which 
have this status because their name ' is imposed by reason of blood' , that is, 'because 
someone obtains them through birth ' .169 Commenting on the phrase 'not. at all' at the 
end of the passage, he confirms the law's categorical insistence: 
That is, certainly they are not taken away by capitis deminutio since these _rights are 
immutable ... nor [can] those [be taken away] which have natural eqmty, as at 
] 170 [D.4.5 .8) and [D.4.5.10 . 
We will recognise the two cited laws here from our discussion above. D.4.5.8 (I. 
Eas) grants the actio dotis to a deported woman, and D.4.5.10 (I. Legatum) describes 
the deported person' s claim for a subsistence legacy. Whereas these faculties of the 
deported persons are described as their 'things ' derived from the natural law 
elsewhere, here they are unambiguously presented as their iura naturalia. It should 
167 Ibid., col. 70, a.v. agnatio: ' Agnatio. iuris civilis est nomen ... unde et unum ius civile 
potest per aliud fac ile toll i ... ' . . . . 
168 Ibid., a.v. sed adgnationis: 'Illud autem ms praerogatrvum quod habent_ agnat1... f~crle 
perdunt... per omnem capitis diminutionem, ut per maximam vel per medram, de qmbus 
duabus non est dubium ... ' 
169 Ibid., a.v. agnatio: ' sed ius cognationis est naturale nomen, nam rat_ion_e s~ngu_inis h~c 
nomen impositum est.' And a.v. naturalia: 'Naturalia. id est, iura cognat10ms, sic dicta qma 
ea quis consequitur per nativitatem ... ' . . . . . . . .. 
170 Ibid. , a.v. utique: 'id est, certe non tolluntur caprtls drmmutlone cum smt 1mmutab1ha 
haec iura ... Item, nee ea quae naturalem aequitatem habent, ut [D.4.5.8) et [D.4.5.10).' 
76 
be emphasised that ius is subjective in this context, given that these iura naturalia 
refer to certain legal faculties or ' things ' variously obtained, lost, or possessed by the 
deportatus. 
Before discussing further the effects of capitis deminutio on the rights of cognation, 
Accursius makes a number of distinctions to help the reader's understanding of 
cognation itself. He highlights that a certain form of cognation is purely civil: 
adoption. Another form is purely natural: that between parents and natural children. 
A third form is civil as well as natural : the relationship between parents and their 
legitimate children. 171 Accursius then recapitulates the question and provides a 
remarkable answer: 'Can, therefore, all these [rights of cognation] be easily lost 
through every kind of capitis deminutio. I reply, yes in the first two cases, but not in 
the third .. . ' 172 One of these first two cases is capitis deminutio media, i.e. 
deportation. Given that Accursius has just associated rights of cognation with 
immutable iura naturalia, this statement is at first sight contradictory to the entire 
flow of his gloss. Where does it come from? It seems that Accursius has to 
accommodate an axiom of the Roman criminal law given at 1. 1.16,6, where there is a 
simple and clear affirmation that the ties of cognation are dis&olved upon 
deportation. 173 Right at the end of his gloss, Accursius qualifies his answer even 
further by reminding the reader that, although rights of cognation are taken away by 
capitis deminutio, 'consanguinity itself' is not, since no positive law can undo 'rights 
of blood' (iura sanguinis). 174 But what should we understand by consanguinity, this 
right of blood, granted to the deported in extremis? Explaining this rule again at 
1.1.16,6, Accursius states that, while deportation takes away cognation in the sense of 
the ius succedendi et ad tutelam vocandi, it cannot take away natural cognation, that 
is, consanguinity, which consists of an impediment to marriage and sexual 
relationships amongst blood relatives. 175 
171 
Ibid. 'Item nota quia quaedam cognatio est civilis tantum, ut per adoptionem ... Item, 
naturalis tanttim, ut ad filios naturales tantum ... Item, naturalis et civilis, ut ad eos qui 
legitimi sunt...' 
172 
Ibid.: 'Nunquid ergo haec omnia facile perduntur per omnem capitis diminutionem? 
Respondeo, sic in duo bus primis tan tum, non in tertio ... ' 
173 
Inst.1.16,6: 'sed et si in insulam deportatus quis sit, cognatio solvitur. ' 
174 
Accursius to Inst.1.15,3, col. 70, a. v. utique: ' Item, nota quod, licet praedicta iura 
cognationis tollantur maxima et media capitis diminutione, ipsa tamen consanguinitas non 
toll itur, ut [D.50.1 7.8: Iura sanguinis nullo iure civi li dirimi possunt.].' 
175 
Accursius to Inst. 1. 16,6, cols 72-3: 'Cognationis. idest ius succedendi et ad tutelam 
vocandi perit, sed ipsa consanguinitas non, ut [D.50.17.8), cum sit de iure natural i, ut... 
[Inst.I . I 0, I 0: Illud certum est serviles quoque cognationes impedimento esse nuptiis, si forte 
pater et fil ia aut frater et soror manumissi fuerint.]. .. ' 
77 
While this passage may appear strikingly inconsistent to the modem reader, I would 
suggest that Accursius is not so much contradictory as arguing scholastically, 
defining the two outer limits of the argument first before coming to settle on a 
compromise. His solution, in summary, is the following one. Rights of agnation are 
derived from the civil law and are lost when civil rights are lost; that is 
uncontroversial. Only citizens could hold patriapotestas or be subject to it. As a 
result, only they had the legal capacity of establishing agnatic male lines within the 
broader kin group that were presided over by a paterfamilias and which, during a 
previous age, had certain prerogatives in the distribution of inheritances and 
guardianships amongst the family. Rights of cognation, on the other hand, can be 
civil as well as natural. Their civil aspects, i.e. adoptive relationships and rights of 
succession, are also lost upon deportation. Only Roman citizens could benefit from 
cognatic lines in the succession of inheritances and guardianships. This fits nicely 
with the jurists' broader exposition of inheritance as a civil right. The deported, 
however, retain the natural aspects of the rights of cognation. These rights of blood 
boil down to consanguinity, a prohibition in the context of marital relationships. In 
other words, the civil law, even in criminal cases, cannot un-relate family members 
who are linked by blood, although it can dissolve merely civil connections. Finally, 
we should note that just as in the case of patriapotestas, the right of consanguinity is 
not a right in the modem sense. While consanguinity is a subjective ius - a ius 
cognationis and a ius sanguinis retained rather than lost by the deported person_ - this 
ius constitutes an immediate obligation or a restriction. 
Accursius ' discussion of agnation and cognation also implies that the ability to hold 
guardianships of related minors was derived from the civil law in the same way that 
the ability to inherit was. The jurists do not much discuss guardianships in the 
context of deportation, but they do make it clear that guardianships end once the 
guardian is deported. 176 What remains unclear is the reason for this, whether it is the 
result of the loss of civil rights or an independent penalty associated with 
deportation. From the civilians' discussion of the hostis-transfuga, considered in the 
following chapter, we will see that they indeed linked the ability to hold 
guardianships with citizenship.177 For now, let us consider the nature of this ability. 
One the one hand, it is clearly described in the law as a person's right and power (ius 
ac potestas) over a ward.178 On the other hand, as a passage in the Institutes clarifies, 
inheritances and guardianships are passed along familial lines together because 
176 Cf. Bartolus to D.26.7.32(33),7; Baldus to D.31.1.77(79),4. 
177 Ch. 3, pp. 101-2. 
178 E.g. Inst.l.13 , l ; D.26. l.l. 
78 
'where there is the benefit [emolumentum] of succession, there also ought to be the 
burden [onus] of guardianship' .179 Revealingly, there is a whole title in the Digest 
(De excusationibus) devoted to regulating the excuses for not taking up 
guardianships. 
180 
The right of guardianship is therefore a duty-right. It does not 
constitute an immediate obligation (like the right of patriapotestatis of the son-in-
power or the right of consanguinity) since, in the first instance, it is a legal power. 
Rather, it is a mediate obligation: the means to the performance of a duty in the 
family context. 181 
Although not a part of family law, we should briefly note that the jurists mentioned 
one other right lost by the deported person. This was the ability to be an arbiter a 
. 
' 
pnvate person appointed by two parties to make a binding ruling in lieu of an official 
judge. 182 (Note that this arbiter is different from the arbitrator discussed above in the 
context of fideicommissa and the ius eligendi.) Both Bartolus and Albericus de 
Rosate reproduced Dinus de Mugello's opinion that an arbiter had to be in 
possession of iura civilia, and that the ability to be one was therefore lost upon 
deportation. 
183 
Albericus informs us that this was also the opinion of Guido de 
Suzarra.184 
iii. Property, Contract, and Slavery 
While discussing the law of inheritance above, we saw that the deported person 's 
?wnership of property acquired after his sentence was a working principle. This idea 
1s based on three laws. C.9.49.2 is the provision of the Code that requires any 
property acquired by the deportatus after his sentence to be confiscated by the fisc 
179 I t I 17 ' b' · · ns . . : ... u 1 success10ms est emolumentum, ibi et tutelae onus esse debet.' 180 D.27. l . . 
181 
Berger, Encycl~pedic Dictionary, p. 747: 'Tutela is not only a right; it created on the part 
of th~ tut~r duties ~or the fulfillment of which he was responsible. Consequently 
guard1~nsh1p_ was c?ns1d~red a _munus (a charge) ... ' This is why the medieval j urists included 
tutela m !heir deta!led d1scuss1ons of munera - cf. Menzinger, 'Verso la costruzione ', pp. 
clxv-clxv1 and clxxxvii. 
182 
Berger, Encyclopaedic Dictionary pp. 365-6 
183 ' ' 
B~rtolus to D.48.1_3.3, fol. 184vb: 'Dicit Dyn. quod ista lex est argumentum quod 
banmtus e~ eo male~c10, _ex ~uo deberet sequitur deportatio, non possit esse arbiter. Nam ilia 
verba~ arb1tratu etc., ms s1gmficant... sed deportatus amittit pristina iura, ut hie, ergo etc.' Cf. 
Albencus de Rosate to D.48.13 .3. 
184 Albericus de Rosate to D.4.8.7. 
79 
I 
I 
upon his natural death, and thus assumes that he can have property in the first 
place. 185 D.17.1.22,5 states that a deportatus can enter a contract of mandate with 
another person in order to have the latter purchase property for him; if the mandatary 
does not keep faith, the deportatus can bring an action for mandate.
186 Lastly, 
D.32.1. 7 speaks of a deportatus transferring legal title over a slave to another 
person. 187 
Given that property (dominium) and contracts (obligationes) are well-known features 
of the ius gentium in medieval civilian jurisprudence, it should not come as a surprise 
that Accursius based the ability of the deportatus to own property and make 
contracts on his possession of ea quae sunt iuris gentium/naturalis .
188 The Gloss 
repeatedly clarifies that since the deportatus is a free person, he can purchase goods 
h l l · h" 189 (including slaves), own them, sell them, and enter ot er contractua re at10ns 1ps. 
Interestingly, Accursius once again associates these natural faculties of the 
deportatus with the sphere of factum, by citing a law stating that 'payment, sale, 
delivery pertain more to fact than to law'. 190 Just as in the case of the ius eligendi and 
the subsistence legacy, we can observe an alignment of the realm of factum with that 
of ius naturale/gentium in opposition to ius civile. Unfortunately, Accursius does not 
at this point provide a more detailed explanation for this move. 
Although the contract of purchase and sale ( emptio venditio) dominates the 
discussion of the deported person's contractual abilities, Accursius explains that the 
deportatus is generally bound by natural obligations after his sentence and that for 
these he can even offer a surety (jideiussor) - a person who undertakes to pay the 
185 C.9 .49.2: ' Deportati nee earum quidem rerum, quas post poenam inrogatam habuerint, 
heredem habere possunt, sed hae publicantur.' 
186 D.1 7.1.22,5: ' Is cuius bona publicata sunt mandare alicui potest, ut ea emat, et, si emerit, 
utilis erit mandati actio, si non praestet fidem: quod ideo receptum est, quia publicatis bonis 
quidquid postea adquiritur, non sequitur fiscum.' 
187 D.32.1.7: ' Si deportati servo fideicommissum fuerit adscriptum, ad fiscum pertinere 
dicendum est, nisi si eum deportatus vivo testatore alienaverit vel fuerit restitutus: tune enim 
ad ipsum debebit pertinere.' 
188 Accursius to D.48.19.17, col. 1411: 'at emere, et alia quae iurisgentium sunt, bene 
possunt, ut [D.27.10.7,2]. . .' 
189 Accursius to C.9.49.2, col. I 835: 'Deportati. Habuerint. Bene enim potest habere et 
vendere et contrahere, cum sit Iiber, ut.... [D.32.1.7). Again, to D.32.1.7, col. 890: 'unde die 
quod hunc servum post deportationem quaesierat, ex aliquo contractum forte, et sit suus, non 
fisci, [C9.49.2].' 
190 D.27.10.7,3: ' ... solutio venditio traditio facti magis quam iuris sunt...' 
80 
debt of the principal should the latter default on his obligation.191 It would have been 
clear to the medieval civilians that the deportatus had access to all those forms of 
contract associated with the ius gentiu~ at the 1. Ex hoe iure (D.1.1 .5), including the 
contract of letting and hiring (locatio conductio), except 'certain contractual 
elements established through the ius civile' .192 Hence, the jurists would have also 
been aware that the deportatus was barred from a range of civil contracts, such as 
stipulation (stipulatio) for example. 193 Furthermore, Accursius accepts the provision 
of the law that the deportatus can enter a contract of mandate, and confirms that any 
resulting action for mandate will be both 'usable' ('utilis') as well as 'legally valid' 
('efficax'). 194 Given that a deported woman could also use the actio dotis , the 
broader implication is that the deported person could generally sue for those rights 
due to him under natural law. Indeed, we will now see that this is the position of 
Accursius' successors. 
Generally speaking, later jurists confirmed the arguments already found in the 
Gloss. 
195 
Sometimes they brought out more explicitly and more fluently the ideas 
that were left implicit, or at least partially implicit, by Accursius. Jacobus de Arena, 
for example, clarified that a deported person can go to court ('in iudicio esse ') when 
bringing an action for mandate or an action for dowry.196 One of the weaknesses of 
the Gloss, however, was that it never faced a problem that should have been obvious 
to any medieval civilian. As mentioned above, it was the common teaching in the 
191 A . 
ccursms to D.46.1.47( 48), col. 976: 'post poenam autem naturaliter obl igatur ex 
contractu facto post deportationem ... et sic potest accipi pro ea natural i fideiussor ... ' For 
fide iussio, see Berger, Encyclopedic Dictionary, pp. 350-1 . 
192 D 1 1 5 ' . d . · 
· _- . :. . .. empt10nes ven 1tJones, Iocationes conductiones, obligationes insti tutae: 
exceptJs qmbusdam quae iure civili introductae sunt.' The translation is from Watson (ed.) 
The Digest of Justinian, vol. 1, p. 2. ' 
193
• A_lthough stipulation was open to peregrini in later classical law - Berger, Encyclopedic 
Dzctwnary, p. _ 716 - the medieval civilians viewed it as part of the ius civile. Thus Bartolus 
Lucanae civitatis, pp. 205b-6a: ' ... stipulatio est de iure civili publico Romanorum inducta ~ 
Gallo Aquilio .. .' Cf. Accurs ius D.2.14.27,2. Note that some classically civil contracts such 
as mancipatio or cessio in iure had disappeared by the time of Justinian in favour of traditio 
For a comprehensive survey of the Roman law of obligations, see Buckland and Stein, A 
Text-Book of Roman Law pp. 405-603. 
194 • ' 
195 Accurs!us to D.17.1.22,5, fol. 256v: 'Is cuius. alicui potest... Utilis. idest efficax ...
 
Very httle novelty can thus be found in the commentaries of Dinus de Mugello to 
D.3~.1 .. 7, Jacobus de Arena to D.17.1.22,5, Cinus de Pistoia to C.9.49.2, and Rainerius de 
Forhv10 to D.48.19.17. 
196 
Jacobus de Arena to D.17.1.22,5, fol. 85va: 'Eadem glossa, in fine. et nota confiscatum 
agere in iudicio esse ... et ad hoe [D.48.20.5].' The cited law is concerned with the actio 
dotis. 
81 
first half of the thirteenth century that actions were derived from the civil law. Yet 
the deported person lost ea quae sunt iuris civilis, so how could he possibly have an 
action?197 Jacobus Butrigarius' major contribution to the discourse on deportation 
was that he faced this problem head-on in the context of the action for mandate: 
And if you ask how he has an action for mandate when actions are civil, I reply that, 
since he can acquire property through an obligation that is derived from the law of 
nations, and an action arises from an obligation, he has the action by necessary 
198 
consequence. 
Jacobus does not question that actions are derived from the civil law. He merely 
argues that natural obligations must be logically substantiated in the form of a ( civil) 
action. So far, Jacobus has accepted the traditional teaching and refined it. He then 
introduces a distinction that constitutes a quite radical break from the past. Granted 
that the deportatus has the action in principle, he does not have the 'execution' 
('executionem') of it. 199 In other words, the deported person 'cannot bring actions to 
court' ('in iudicio deducere').200 This seems to be his way of reconciling that the law 
grants actions to the deported persons on the one hand, while also denying them all 
things civil on the other. In sum, although Jacobus considers the deported person to 
have rights under the natural law - such as the ability to acquire property by contract 
- if these rights are challenged by another, there is no way he can enforce them in 
court. 
It is important to stress that in this opinion Jacobus Butrigarius stood alone. A long 
passage from Albericus de Rosate's commentary to the 1. Sunt quidam, which is 
worth quoting in full, best explains the communis opinio of the fourteenth century.
201 
Albericus restates the fundamental principles found in the Gloss but also takes into 
197 In the words of Jacobus Butrigarius to C.9.49.2, fol. 112rb: 'Sed contra hoe occurrit 
dubium. nam nos dicimus quod actiones sunt iuris civilis, et omnia quae sunt iuris civilis 
denegantur deportato, et tarn en actio ei conceditur. .. ' 
198 Jacobus Butrigarius to D.17.1.22,5, p. 191 b: 'et si quaeras, quomodo habet actionem 
mandati, cum actiones sint civiles, respondeo quod, cum habeat acquisitionem obligationis 
quae est de iure gentium, et ex obligatione nascatur actio, habet earn ex necessaria 
consequentia ... ' 
199 Ibid., p. 192a: 'vel die quod, licet habeat actionem, non forte habebit executionem .. .' 
Similarly, to C.6.24.1, fol. I 5rb: ' nam ista deportatio tollit ea quae sunt iuris civilis, sed non 
ea quae sunt iuris gentium, scilicet contractum, nee actionem quam ex eo nascitur, licet 
exercitium non habeat...' 
200 Jacobus Butrigarius to C.9.49.2, fol. 112rb: 'Sed actiones non poterit in iudicio deducere.' 
201 Cf. Bartolus to D.4.5.7,3, D.17.1.22,5 , D.42.1.14, D.46.2.14,1, D.48.19.17; Baldus to 
D.17.1.22,5, D.32.l.7, D.46.1.47(48), C.6.24.l, C.9.42.2. 
82 
account subsequent developments in the discourse, particularly contributions by 
Dinus de Mugello and Ricardus Malumbra: 
Concerning the final paragraph, however, those things which are derived from the law 
of nations are listed at D.1.1.5, such as buying and letting and similar things. But how 
can he buy, since he possesses nothing, given that all his goods have been confiscated? 
I reply, it is true that whatever he had at the time of the sentence was confiscated but 
that ~hich he acquires afterwards is not confiscated as long as he lives. And in re;pect 
of this he can make contracts. [ ... ] But after his death even that [is confiscated]. .. 
Dinus. 
But of :what use is it to him to be able to make contracts or to retain those things which 
are denved from the law of nations, since he loses those things which are derived from 
the civil law, and all actions are derived from the civil law ... and no one can sue 
without an action ... ? I reply that he retains and possesses the ability to enforce actions 
[exercitium actionum] even though they are derived from the civil law. For by virtue of 
the fact that he is granted the ability to make contracts and other things derived from 
the law of nations, he is [also] granted that without which he cannot pursue or enforce 
such things ... Ricardus. 
It would also be possible to say that actions are derived from the law of nations in 
respect of their origin, since they arise from an obligation ... but an obligation is derived 
froi_n the law of nations ... hence also an action, which arises from it. [ ... ] That they are 
?enved from th~ civil la~. should be understood not in respect of their origin, but only 
m respect of their formalities and enforcement. This is noted by Dinus ... •202 
Although Albericus does not mention Jacobus Butrigarius by name here, it is clear 
that he rejects the latter's analysis by following Ricardus Malumbra's opinion. 
Albericus' argument is reinforced at the end by connecting it to Dinus' opinion that 
actions must in some sense be rooted in the law of nations for claims based on the 
latter. As Pennington has demonstrated, this idea had become dominant by the 
~
02 
_Alberi~us de Rosate to D.48.19.17, fol. 203ra-b: 'In finali, quae autem sunt ea quae sunt 
mns g~~tmm numer~ntur [D.1.1.5], ut emere, locare, et similia. sed quomodo potest emere, 
cui_n m_hil habeat, qu1a omnia bona sua publicantur? Respondeo, verum est quod publicatur 
qmcqmd habet tempore sententiae. sed id, quod postea acquirit, non publicatur donec vivit. 
et ~uper eo p~t~st contrahere. [ ... ] sed post mortem eius etiam illud [publicatur]. .. Dy. Sed 
qmd prodes! s1_b1 P?S~~ contrahere vel retinere ea quae sunt iurisgentium sunt, cum perdat ea 
quae sunt mns c1v1hs, et omnes actiones sunt de iure civili.. . et nemo sine actione 
experitur.. .? Respondeo quod retinet et habet exercitium actionum, licet sint de iure civili. 
?am e~ quo ~o?ceduntur ei contrahere et alia quae sunt iuris gentium, esse [sic - etiam?] 
il!u_d v1detur ~1b1 concessum sine quo illud exequi vel exercere non posset... Ric. posset etiam 
die~ qu?d act10n~s sunt de iure gentium, quo ad originem, cum oriantur ab obligatione ... sed 
o?l!~~tlo e~t de mre gentium ... ergo et actio, quae descendit ab ea. [ ... ] quod sunt de iure 
c1v1h mtelhgatur non quo ad originem, sed solum quo ad solemnitatem et exercitium earum. 
de hoe notatum per Dy ... .' 
83 
opening of the fourteenth century. 203 While the jurists therefore permitted the 
deportatus to pursue his natural rights in court, Baldus highlighted that he normally 
must send a legal representative (a procurator) rather than be himself allowed to 
leave his confinement for the sake of a lawsuit; the deportatus thus possesses the ius 
d. t 204 agen z per procura orem. 
As we have seen, the deported person was capable of owning slaves acquired after 
his sentence. On the other hand, if he owned any slaves previously, those were lost 
when the rest of his property was confiscated. We also saw that the deportatus lost 
his praetorian rights of succession over his previous freedman's estate, which made 
sense given that praetorian possession was a civil right. The law contains two further 
provisions regarding slavery and manumission. Firstly, the deportatus loses his other 
patronage rights over his previous freedman, in particular the privilege whereby the 
freedman was unable to summon his patron to court without the permission of the 
205 . d t praetor (a Roman magistrate). Secondly, the deportatus was not perm1tte o 
manumit the slaves he owned while deported (though he could, of course, sell 
them).206 The Gloss accepts these provisions of the law without much comment.
207 
This should strike us as somewhat problematic given that the institution of slavery 
and the mechanisms surrounding it were introduced by the law of nations. In fact, 
Accursius himself writes elsewhere that the 'reverence ' due from freedman to patron 
is established by natural law. Yet he did not feel the need to reconcile this with the 
principle of the 1. Sunt quidam that deported persons retain rights derived from the 
law of nations.208 Ordinarily, a peregrinus should be able to manumit his slaves and 
enjoy certain privileges over his freedmen. The deportatus , due to his criminal 
sentence, could not do so. In effect, we can once again observe an unexplained 
suspension of higher norms at the hands of the Roman criminal law, similar to the 
cases of the actio dotis and the confiscation of property. 
203 Seep. 74, n. 163 above. . 
204 Baldus to D.2.4.10,6, fol. 89ra: 'Hie notat Guil. quod deportatus per se non potest esse to 
iudicio, nee potest exire insulam causa litigandi. . .' The same to D.4.6.40,1, fo l. 255rb: 
' Secundo nota quod pro bonis ... licet sit deportatus ... habet ius agendi per procuratorem .. .' 
Cavalca, JI bando, p. 219, is therefore incorrect when he argues that t~e ~ ability to brio~ 
actions 'era riservata dal diritto comune a chi aveva perduto la c1ttadmanza, come I 
deportati. .. ' . 
205 D.2.4. l 0,6: 'Sed si per poenam deportationis ad peregrinitatem redactus sit patronus, 
putat Pomponius eum amisisse honorem.' Cf. D.2.4.4. 
206 D.48.22.2: 'Manumittere deportatum non posse divus Pius rescripsit. ' 
207 Accursius to D.2.4. l 0,6 and D.48.22.2. 
208 Accursius to D.48.22.3, col. 1428: 'vel ibi loquitur secundum ius naturale, scilicet, 
secundum hoe, quod libertus de bet praestare reverentiam patrono .. .' 
84 
While Accursius ' successors accepted the law's requirement that the deportatus lost 
the ius patronatus over his freedmen, there seems to have been some debate over 
how that loss should be understood.209 This debate is recorded by Albericus de 
Rosate, whose commentary represents real progress from the days of the Gloss. 
Credit, however, should be given less to Albericus than to the jurists whose ideas he 
is using. Noting the problem raised by the 1. Sunt quidam, he first introduces a 
distinction by Rainerius de Forlivio, according to which the deported person's lost 
ius patronatus only refers to an impediment of fact: the patron who is stuck on an 
island cannot physically be in contact with his freedmen; but according to natural 
law, the freedman is still bound to honour his patron.210 Albericus then initiates a 
contradictory line of argument. According to the Gloss, although the freedman could 
now, after deportation, theoretically sue his patron, this is not possible because 
actions are not granted against the deported, since the fisc is now the bearer of their 
estate and any inherent obligations.211 As we should expect, Albericus clarifies that 
this principle only applies to actions that had arisen before the sentence of 
deportation, but that deported persons can indeed be sued for property acquired or 
contracts made under the ius gentium afterwards.212 Although some, Albericus goes 
on, such as Jacobus de Arena, follow the position of the Gloss and thus fully accept 
the argument that the freedman can now, after deportation, indeed sue his patron in 
respect of such property or contracts even without permission of the praetor, he 
remains faithful to the opinion of Rainerius de Forlivio but qualifies it with a 
distinction by Guilelmus de Cuneo: the reverence still owed to the patron by natural 
law only prevents the freedman from summoning him to court personally, even after 
deportation, but does not stop him from summoning the deported patron's 
procurator. In agreement with Baldus, Albericus argues that it is through a procurator 
that a deportatus ought to respond to lawsuits initiated against him after his sentence. 
For if the deported could respond to actions personally, they would feign frivolous 
209E J b B . . 
. .g. aco us utnganus to D.2.4. l 0,6, pp. 71 b-2a: 'Media capitis diminutio primo tollit 
ms patronatus· ... ' 
~,o Albericu~ de Rosate to D.2.4.10,6, fo l. 11 Ova-b: 'Opponitur, quia ista reverentia debetur 
mre naturah, [C.2.2.2: .. . cum naturali ratione honor eiusmodi personis debeatur.] , ergo 
debe°:1r patr~no deportato, [D.48.19.17) . Solutio: die quod non obstat impedimentum iuris 
quommu~ et1am deportato reverentia debeatur, ut in contrario, sed obstat impedimentum 
f~cto'. quia non potest ab insula, in qua deportatus est, recedere .. . Ray. ' 
lb~d.'. fo_l. . 11,0vb: 'Et ca~e, quia glossa dicitur quod deportatus non potest conveniri nee 
esse m md1c10. Cf. Accurs1us to D.2.4.10,6, col. 127: ' In ius autem nee nunc ab eo vocari 
fi~terit, q_uia aliud impedit, ut puta deportatio, ut [D.4.5.7,3) .' 
Albencus de ~os~te to D.2.4.10,6, fol. 11 Ovb: 'Item in contractibus et bonis quae habebat 
tempore dep_o~at1001s non convenitur ipse, sed fiscus , cui bona sunt applicanda, sed in 
postea quaes1tis vel celebratis bene po test deportatus conveniri. .. ' 
85 
lawsuits in order to leave their confinements and thereby avoid punishment.
213 
This 
is a neat solution and fits well with the broader principles governing the deported 
person's status under natural law. It is interesting that, whereas some jurists seem to 
have had no compunction about allowing the civil law, through deportation, to break 
the natural bond fully between patron and freedman, Rainerius de Forlivio, 
Guilelmus de Cuneo, and Albericus thought this to be impossible by virtue of its 
very naturalness. 
Albericus' commentary also reveals something fundamental about the ius patronatus 
in medieval civilian thought. He writes: 'Capitis deminutio media of a patron takes 
away the right of patronage [ius patronatus] and restitution returns it... And you 
should understand the same concerning the freedman who has been deported and [the 
one] who has received restitution .. .'2 14 This is a prime example of the ambiguous 
subjective sense of ius . In the first instance, ius denotes the entitlement of the patron 
regarding his freedman, what we would call a right in modem English. In the second 
instance, ius is still subjective - Albericus is speaking of the ius patronatus lost and 
regained by the freedman - but now ius no longer denotes a personal entitlement. 
Just as the ius patriapotestatis of the deported son-in-power and the ius 
consanguinitatis, this ius denotes an immediate personal obligation: the reverence 
the freedman owes his patron. 
8. Conclusions 
The medieval civilian discourse on deportation reveals that the jurists possessed a 
concept of citizenship and civil rights on the one hand, and a concept of natural 
rights on the other. Through deportation, an individual moved from the realm of the 
civil to the realm of the natural. Deported persons lost their citizenship (civitas) and 
civil rights but retained their natural rights. This transition, however, was neither 
213 Ibid.: ' ... et pro illis poterit in ius vocari a liberto sine venia, secundum lac. de Are .... Quia 
tamen <licit Gu!. non esse verum quo ad hoe, ut possit venire ad iudicium personaliter ... sed 
per procuratorem bene poterit interesse in casibus suprascriptis ... illi tam~n p.rocuratori non 
debebitur reverentia, ut [D.2.4.10, 13). Item, deportatus non debet exire msulam causa 
Jitigandi, quin hoe esset inconveniens, et posset fingi lites, et sic occurrit legibus per 
indirectum. Gu!.' . 
214 Ibid., fol. 11 Ova: 'Media capitis diminutio patroni tollit ius patronatus et restitutio reddit... 
Et idem intelligas in liberto deportato et restituto ... ' 
86 
straightforward nor complete. Although the deportatus lost those iura civilia 
associated with citizenship, he remained subject to the civil law understood as the 
Roman legal system more broadly; this made the criminal sentence binding upon the 
former citizen. In other words, the deportatus was subject to Roman jurisdiction on 
Roman soil as any other non-Roman peregrinus. Yet unlike other peregrini, the 
effects of his criminal sentence, which contained certain penalties beyond the loss of 
citizenship, did not allow him to enjoy fully the legal position ordinarily granted to a 
free person under natural law. Here, positive law suspended natural rights in 
accordance with the jurists' general theory concerning the relationship between the 
civil law and higher norms: the former could derogate from but not generally 
abrogate the latter. 
The concept of Roman citizenship revealed in this discourse is substantially different 
to the way historians have previously characterised the civilian idea of municipal 
citizenship. Roman citizenship was not contractual, it conferred no public duties, and 
it was not defined by political participation. Instead, Roman citizenship was 
conceived as a subjective legal status entirely derived from the ius civile and 
associated principally with a number of entitlements in private law that were 
inaccessible to foreigners. Through their iura civilia, Roman citizens possessed 
active and passive inheritance rights - both those based on testament and on 
praetorian possession. Only they could make civil law contracts, exercise or be 
subject to patriapotestas, and form family lineages through agnation and civil 
cognation. Only they could hold guardianships of related minors and be arbiters in 
judicial disputes. We saw that two of these subjective iura actually constituted 
obligations. The ius patriaepotestatis of the son-in-power was his immediate 
obligation towards his father, and the ius tutelae was a mediate obligation, a power 
through which to exercise a duty. That Roman citizenship was largely a basket of 
entitlements should not really surprise us. Why else would its loss constitute part of a 
severe punis~ment for serious crimes? Logically speaking, it would not make any 
sense if the criminal stood to gain through his criminality. If the loss of citizenship 
were a reward, it would fall foul of the principle in Roman criminal law that 'no one 
ought to suffer a penalty and a reward for the same deed' . 215 
While one could perhaps argue that there is a propensity in much of the existing 
historiography to underestimate the modernity of medieval civilian ideas in this 
21s A . D 4 ccursms to . 8.22.8, col. 1432: 'nota quod ex eodem facto poenam et praemium non 
debet quis pati.' D.48.22.7,22 is another good example: 'sed muneribus si quis poena causa 
fuerit prohibitus, nihil valebit sententia: neque enim immunitatem poena tribuere debet.' 
87 
context, Pietro Costa, in his relatively recent grand history of European citizenship, 
has failed to understand a crucial element that sets them apart from later, especially 
post-Enlightenment discourses. 216 Costa's central theme is that of gradation, of 
diverse modalities of belonging: medieval thought on citizenship, whether juristic or 
otherwise, 'never brought into relief individuals "as such", free from a network of 
associations and obligations'. 217 There was thus no concise or abstract legal 
definition of the citizen. Rights and duties were variously bundled according to time, 
place, wealth, social class, age, gender etc. In contrast to modernity, the Middle Ages 
were a fundamentally communitarian period, an age of belonging and hierarchy, 
when it was difficult to see the individual part without its connection to the whole. 
Furthermore, although medieval lawyers had a conception of natural rights, he 
argues, juristic science nevertheless failed to imagine a self-sufficient, abstract 
individual capable of bearing such rights independently of modes of belonging to the 
community.218 Interestingly, the author himself never expressly differentiates in his 
own analysis between the individual as citizen and the individual as non-citizen. As a 
result, the reader is sometimes left wondering to which type of person his argument 
refers. 
The two issues are probably interrelated and seem to stem from a flawed 
understanding of the medieval jurists' conception of civil rights and natural rights. 
First of all, as should have become clear during the course of this chapter, rights of 
citizenship were not natural rights. The former one had by virtue of membership of a 
city, the latter by virtue of being human: they were different kinds of rights. If this is 
an obvious point, it is still worth reiterating since too much is written about medieval 
citizenship which fails to take it on board. Sara Blanshei's monograph on the 
citizenry and politics of justice in communal Bologna is a good example. Deriving 
her ideas from the existing literature, she claims that an important characteristic 
distinguishing the medieval jurists' natural rights from modem human rights is the 
former's 'lodging in and relationship to citizenship'.
219 On the contrary, this is 
perhaps the most important sense in which the civilians' natural rights could be 
called modem: in essence, they were rights due to an individual independent of civic 
status. One suspects that the tacit conflation between natural and civic rights 
underlying Costa's analysis springs from a view to the future, specifically to the later 
216 Costa, Civitas, eh. 1, pp. 1-50. 
217 Costa, Civitas, pp. 48-9: ' .. .ii discorso della cittadinanza non mette mai in scena i soggetti 
"come tali", svincolati da una rete di appartenenze e di obbedienze.' 
218 Ibid., p. 34 ff. 
219 Blanshei, Politics and Justice, p. 316. 
88 
eighteenth-century, when the two types of rights were indeed intertwined, when the 
rights of citizens were due to them because of natural law, as encapsulated nicely by 
the title of the French revolutionaries' Declaration des droits de l'homme et du 
citoyen. 
If we avoid this anachronistic conflation and distinguish clearly between civil rights 
on the one hand and natural rights on the other, the weaknesses of Costa's grand 
narrative become visible, both in relation to civic man and in relation to natural man. 
Firstly, because rights of citizenship cannot originate from anywhere else but the 
civil law, Costa's search for a civic status that is completely undetermined by the city 
will always end in vain. Nevertheless, in place of hierarchy and different kinds of 
belonging - diverse citizenships if you like - the jurists have a concept that is 
basically unitary: a status shared by all citizens that, for the most part, grants them 
the same rights inaccessible to foreigners. In their discussion of deportation, the 
civilians normally consider the citizen and his rights as such, independent of other 
personal qualities, save the odd exception, such as the case of patriapotestas, which 
depends on age and gender. I am not arguing that the Roman law of persons is not 
highly stratified and complex. I am merely proposing that all who are citizens share 
an identical underlying characteristic with a certain content that makes them citizens 
220 
. 
Secondly, this chapter demonstrates that the jurists could indeed imagine individuals 
understood as fundamentally separate from their political community, those who had 
lost their citizenship, yet were nonetheless bearers of certain subjective natural 
rights. In so doing, they went a long way - certainly longer than Costa maintains - in 
providing us with a view of the human individual 'as such'. At least as far as the 
jurists are concerned, the Middle Ages were in this regard more individualistic than 
older narratives of community and hierarchy would suggest. We have to be careful, 
however, not to overstate the case. The fact that deported persons were criminals 
meant that th~ jurists inevitably still conceived of these foreigners in relation to their 
former city at least to some degree. After all, deported persons remained bound by 
the punitive measures associated with their new status and could not fully enjoy their 
natural rights. Occasionally we catch a glimpse of the peaceful stranger, yet the 
jurists never cared to elaborate on the idea. 
2
~
0 
_Asche_ri, . 'Nella citta me~ievale italiana', has criticised Costa in a similar way regarding 
c1tJzensh1p m the legal practice and socio-political realities of the Italian cities. 
89 
The possession of natural rights granted deported persons the following legal status. 
They were free persons and under the protection of the law; they were not enslaved 
by the sentence nor could they be injured in person and property by other private 
individuals. They could acquire, own, and sell property, including slaves, and make 
other contracts (such as mandate and lease and hire) that were derived from the law 
of nations. These contractual relationships could be reinforced through fideiussio. 
They could marry and have legitimate children. A deported woman could possess a 
dowry and bring the actio dotis to recover it. Although normally excluded from 
inheritance, deported persons could receive a subsistence legacy and select the 
beneficiary of afideicommissum. Lastly, they possessed the right of consanguinity. It 
should be emphasised that these rights were not merely formal: the deported could 
bring actions in the courts of their former civitas in order to substantiate their natural 
claims. 
Samuel Moyn has recently argued that natural rights, unlike the human rights of the 
second half of the twentieth century, were primarily used in the construction of the 
state and citizenship, rather than constituting claims made outside and against the 
state.221 Although this is an insightful point about much of the western natural rights 
tradition ( one only needs to think of Hobbes and Locke), it does not apply to the 
medieval civilian conception of natural rights. The natural rights of the deported 
person might not have been used to make claims against the state in the mode of 
today's human rights, but they certainly transcended the state (or rather, the civitas). 
Rather than being used for the construction of the city and citizenship, these rights 
essentially acted as a kind of safety net. They were the residual rights due to a human 
being who could no longer draw rights from any civic source. 
The medieval civilians did not merely possess a substantive concept of natural rights, 
but they also articulated their theory of natural rights using a language of subjective 
ius naturale. In this way, they substantially expanded a language of which, as we 
saw in chapter 1, there are at best traces in their ancient authorities. This means, 
contrary to the current scholarly consensus, that the medieval canonists, 
philosophers, and theologians should no longer be considered to stand alone as the 
inventors of natural rights theories expressed in terms of subjective ius. The 
221 Moyn, The Last Utopia, e.g. p. 20: 'Unlike later human rights,_ ho~ever, they [natural 
rights] were deeply bound up with the construction, through revolution If nec~ssary, of stat_e 
and nation. It is now the order of the day to transcend that state forum of nghts, but until 
recently the state was their essential crucible.' 
90 
medieval civilian tradition constitutes a significant third point of origin. 222 Just as the 
civ~l rights of citizens, the iura naturalia of deported persons were mostly 
entitlements and thus rights in the modem sense. There were two exceptions. Both 
the right of consanguinity and the freedman's right of patronage were immediate 
obligations - a marital and sexual restriction in the case of the former, and the 
re~erence due from the freedman to his patron in the latter. Sometimes the jurists 
aligned the realm of ius naturale with that offactum, and opposed it to the realm of 
positive ius civile. While it is William of Ockham who is best known in medieval 
political thought for making this move, the idea can already be observed in 
Accursius' Ordinary Gloss. Unlike Ockham's ius naturale, however, the jurists' 
natural rights gave rise to actions that were capable of being pursued in the courts of 
the civitas. 
Finally, we should note the casuistry involved with the derogation of higher norms 
evident in the case of deportation. The deportatus had his estate confiscated he 
could not manumit his slaves, and in certain instances the deported woman ~as 
denied the retrieval of her dowry. All of these cases represent a prominent 
suspension of natural rights. Yet the civilians never really explained or justified this 
su_s~en~ion. Where was iusta causa? Besides the law's brief reference to public 
utility m the case of property confiscation, we are left in the dark. In other words in 
the discourse on deportation, the civilians never developed a theory explaining ;hy 
a~d to what extent a mechanism of the positive criminal law could derogate from 
higher norms. They never truly questioned the examples of the Roman law and 
assumed that they were just, remaining faithful to this maxim in the Digest: 'It is not 
possible to find an underlying reason for everything which was settled by our 
forebears. [ ... ] Accordingly, it is not right to go ferreting after the motives behind the 
things which are settled as law. To do otherwise is to subvert many present 
certainties. '223 This problem will become more pressing in the case of the hostis-
transfuga discussed in the following chapter, since, as we shall see there, that 
peregrinus suffered from a much greater suspension of higher norms than the 
depo:1ed person. In both cases, it is impossible to pinpoint any exact rationale, any 
consistent notion concerning the boundary between natural right and the city of 
Rome. 
222 f C. eh. 1, pp. 23-31. 
223 
_D.1.3.20-1: 'Non omni~m, quae a maioribus constituta sunt, ratio reddi potest [ ... ] et ideo 
rat10nes eorur,n, quae const_Itu~ntur, inquiri non oportet: alioquin multa ex his quae certa sunt 
subvertuntur. The translat10n IS from Watson (ed.), The Digest of Justinian, vol. 1, p. 13. 
91 
92 
Chapter 3 
The Imperial Ban 
As long as the state is a political entity this requirement for internal peace compels it in 
critical situations to decide also upon the domestic enemy. Every state provides, 
therefore, some kind of formula for the declaration of an internal enemy. The noMµto<; 
declaration in the public law of the Greek republics and the hostis declaration in 
Roman public law are but two examples. Whether the form is sharper or milder, 
explicit or implicit, whether ostracism, expulsion, proscription, or outlawry are 
provided for in special laws or in explicit or general descriptions, the aim is always the 
same, namely to declare an enemy. 
Carl Schmitt, The Concept of the Political (1932) 
1. Introduction 
In the previous chapter we discussed the ordinary criminal procedure of Roman law 
by which Roman citizens, following a capital trial, lost their citizenship. In the 
Corpus iuris civilis, deportation was the usual form of capitis deminutio media. The 
present chapter, on the other hand, is mainly concerned with the second, much rarer 
instance of this legal mechanism: whereby a citizen acquired the status of hostis 
(enemy) and transfuga (deserter to the enemy) through extraordinary means rather 
than by the normal procedures of Roman criminal law. Like the deportatus, the 
hostis lost his Roman citizenship and, in principle, retained rights derived from the 
laws of nations and nature. The punitive sanctions of the civil law associated with 
this status, however, severely restricted the enjoyment of those natural rights, much 
more so than in the case of deportation. Most pressingly, the hostis was an outlaw 
who could be injured in person and property, even murdered, with impunity by any 
private person. The evident similarities between the hostis in Roman law and the 
medieval imperial bannitus meant that the former came to be established as the 
central figure in the civilian discourse with which the legal status of the latter was 
understood. Corresponding to the state of the field on deportation, this form of exile 
has also received little in the way of proper scholarly attention. The following 
93 
chapter therefore represents the first in-depth study of the hostis, the empire's 
internal public enemy, in medieval Roman law. 
As in the case of deportation, an analysis of rights lost and rights retained by the 
hostis will provide insights into the meaning of Roman citizenship and natural rights. 
The civilian discourse on this topic will both confirm the conclusions drawn in the 
previous chapter and deepen our understanding of these concepts. An important 
natural right that will appear in this context is the right to be summoned to court in 
capital trials for the purpose of presenting a defence. We will explore the tension 
between the demands of this right and the demands of extraordinary procedure, 
particularly the idea of ipso iure sentencing involved in the acquisition of hostis 
status. We will also continue to explore more generally the problematic relationship 
between natural rights and the punitive provisions of the civil law, the conflict of 
which resulted in the former being at the same time mutable in one sense but 
immutable in another. One of the questions raised by the hostis is why the natural 
rights of this person were so much less efficacious than the natural rights of the 
deportatus, who, as we saw, was able to live peaceably on his island and own 
property there under the protection of the law. Moreover, we will see further 
evidence for a language of subjective civil and natural rights, and I will expand the 
argument that ius should in certain contexts be understood as a person's duty while 
remaining nonetheless subjective. This will lead to a number of broader conclusions 
about the concept of ius in medieval legal thought. 
2. Treason and the hostis in Late Antiquity 
If we used modem categories, we would say that the hostis is an essentially political 
figure. Rather than being the creation of ordinary criminal law, that status is 
conferred in special circumstances upon the commission of treason - the political 
crime tout court - through an administrative act of the authorities; as such, the hostis 
could be said to fall under public rather than criminal law. While the distinction 
between ordinary and criminal on the one hand, and extraordinary, political, and 
public on the other, might be helpful to us, we have to acknowledge that neither the 
medieval jurists nor their sources spoke of the hostis in terms of those categories. As 
we shall see in this chapter, the medieval civilians did not consciously separate that 
figure from other areas of the criminal law on treason by conceptualising it 
94 
specifically as a political phenomenon. But it is also true that the status of the hostis 
and the way in which it was acquired were nonetheless truly exceptional and difficult 
to reconcile with basic precepts of criminal jurisprudence. 
The best evidence for the exceptional nature of the hostis is the fact that it is possible 
to find only a scattered handful of relevant passages amongst the vast texts of 
Justinian's compilation. 1 The most important law can be found at D.4.5.5,1 (1. 
Amissione, § Qui deficiunt): 
Those who defect suffer capitis deminutio, and they are said to defect who abandon 
those under whose command they are and join the ranks of the enemy, but also those 
w~om the senate has judged to be enemies or are deemed so by a law, certainly to the 
pomt that they lose their citizenship.2 
This passage establishes three ways in which a citizen can become a hostis. The first 
is by military desertion. This links to other laws that treat the soldier who has 
deserted to the enemy as one of the latter - transfuga pro haste habetur: 
Traitors and de~erters t? the enemy generally suffer capital punishment... for they are 
counted as pubhc enemies, not as soldiers. 3 
The other two ways of becoming a hostis are, broadly speaking, legislative 
enactments. The l. Amissione first speaks of a resolution passed by the imperial 
-senate, and secondly of a law in general, although it is not clear in the first instance 
to which law-giving body this power is granted. The phrase used (' lege lata'), 
however, has a core technical meaning that implies a law passed by one of Rome's 
popular assemblies.4 
Whatever form of procedure is applied, the effects of capitis deminutio media in this 
instance, just as in the case of deportation, are basically twofold. Firstly, citizenship 
I 
D.4.5.5, l ; D.4.6.14; D.11.7.35 ; D.26.1.15; D.41.1.51; D.48.8.3 ,6; D.48.19.8,2/38, l ; 
D.49.15.19,4/7/8; D.49.16.5,8; D.49.16.7. 
2 
D.~.5.5 , l_: 'Qui defi~iunt, cap!te min_uuntur (deficere autem dicuntur, qui ab his, quorum 
sub 1mper_10 s~nt, des1stunt et _m hostmm numerum se conferunt): sed et hi , quos senatus 
~ostes md1cav1t vel lege lata: utique usque eo, ut civitatem amittant.' 
1?·_49.16. 7: 'Proditores transfugae plerumque capite puniuntur... nam pro hoste, non pro 
m1hte habentur.' The translation is from Watson (ed.), The Digest of Justinian, vol. 4, p. 409. 
Cf. D.49.15 .1 9,4: ' ... nam qui malo consilio et proditoris animo patriam reliquit, hostium 
numero habendus est.' 
4 
Cf. Mommsen, Romisches Strafrecht, pp. 256-7 (n. 4). This is also the view of the Ordinary 
Gloss to D.4.5.5, 1, col. 455 : ' ... lege lata, scil icet, a populo .. .' 
95 
and its associated rights are lost. Secondly, there are a number of penalties beyond 
the loss of citizenship. Most importantly: 
It is lawful to kill deserters to the enemy wherever they are met with, as though they 
. 5 
were enemies. 
In other words, the former citizen fully acquires the status of a foreign enemy with 
whom Rome is at war - we would say 'enemy combatant' today. Given that the ex-
citizen can now be treated as if he were a foreign enemy, it is not unreasonable to 
assume that the provisions of the Roman law which generally govern the relationship 
between Rome and her war-time enemies now also apply to this hostis. We shall see 
below that this is how the medieval civilians understood the material. 
For now I would like to address the issue of procedure. Besides the perfunctory 
comments contained in the 1. Amissione, there is no indication in the Corpus iuris as 
to how precisely a citizen becomes a hostis. Yet there appears to be a potential 
conflict between the provisions regarding the latter and the general procedural rules 
of Roman criminal law, in particular the idea that an absent defendant cannot suffer a 
capital penalty. This question is particularly pressing in respect of the alleged 
military deserter who, by the very fact of his desertion, is not within the power of the 
authorities. 
To understand the hostis, we need to situate that figure within the broader context of 
the Roman law of treason. Mommsen's Romisches Strafrecht remains the best guide 
to the material in the Corpus iuris.6 In classical Roman law, the crime of treason or 
lese majesty - crimen (laesae) maiestatis - had a very wide ambit. In the words of 
Ulpian, it encompasses all 'that is done against the Roman people or against their 
security' .7 Instances of lese majesty thus included acts that to us would not appear as 
public-political in nature. For example, knowingly falsifying public records, 
pretending to hold office as a private citizen, releasing a condemned criminal from 
prison, melting down consecrated statutes of the emperor, and even the possession of 
purple clothing, all fell under the scope of the crimen maiestatis. 8 An extreme 
subcategory of this crime was perduellio, treasonous activity 'animated by a hostile 
5 D.48.8.3,6: 'Transfugas licet, ubicumque inventi fuerint, quasi hostes interficere.' The 
translation is from Watson (ed.), The Digest of Justinian, vol. 4, p. 334. 
6 Mommsen, Romisches Strafrecht, pp. 537-94. 
7 D.48.4. l: 'Maiestatis autem crimen illud est, quod adversus populum Romanum vel 
adversus securitatem eius committitur.' 
8 D.48.4.2; D.48.4.3; D.48.4.4; D.48.4.6; C.11.9.4. 
96 
spirit against the commonwealth or emperor'. 9 Perduellio is therefore normally 
translated as high treason. In essence it referred to acts of war or war-like acts against 
the community. It is intrinsically linked to the notion of desertion to the enemy, and 
the term for a person who commits such an act - perduellis - is basically 
synonymous with the term hostis. 10 
According to Mommsen, the penalty for acts of perduellio was imposed ipso iure ( or 
ipso facto), that is, at the very moment when the act was committed. But even in 
these cases, Mommsen argues, the formal establishment of guilt through an ordinary 
trial was normally required before the defendant could be treated as a perduellis and 
thus lawfully killed. The ipso iure provision meant that the sentence was declaratory 
rather than condemnatory in nature, which had the result that the confiscation of the 
defendant's estate was dated back to the time of the crime rather than commencing 
from the time of the sentence. 11 Note, however, that Mommsen makes an exception 
for notorious instances of desertion to the enemy where apparently no trial and 
declaration were necessary. While contentious cases would have had to go to court, 
the notorious transfuga could be killed as and when he was encountered. 12 The 
second exception is very briefly alluded to at the 1. Amissione. The political 
upheavals of the later Roman republic resulted in the development of an 
extraordinary and controversial form of procedure - famously used by Cicero against 
the Catalinarian conspirators - whereby the senate acquired the power to declare 
(through a senatusconsultum), on the instigation of the consuls (later the emperors), 
any citizen to be a hostis, even if the alleged offender was absent and undefended. 
This senatorial prerogative formally remained in place under the emperors and came 
to be used by the latter in order to eliminate their rivals, especially pretenders and 
their followers; the emperors had the senate, either at Rome or Constantinople, 
9 D.48.4.11: ' ... sed qui perduellionis reus est, hostili animo adversus rem publicam vel 
principem animatus ... ' The translation is adapted from Watson (ed.), The Digest of Justinian, 
vol. 4, p. 318. 
10 D.50.16.234: 'Quos nos hostes appellamus, eos veteres "perduelles" appellabant, per earn 
adiectionem indicantes cum quibus helium esset.' 
11 Mommsen, Romisches Strafrecht, pp. 590, 592, 957, I 006-8 . Legal acts undertaken by the 
defendant in relation to his estate (e.g. gifts and testament) in the period between the 
commission of the crime and the trial were therefore void. This allowed the fisc to proceed 
against those who came into the possession of the defendant's property after the crime. 
12 Ibid., p. 623. Although this seems to be intelligent conjecture based only on the rather 
scarce evidence of D.48.8.3,6 - seep. 96, n. 5 above. 
97 
declare them hastes without the benefit of any sort of criminal trial and its 
established procedures. 13 
Military desertion and the declaration of hostis status therefore provided the 
medieval civilians with at least some (though not much) material that could be used 
in an attempt to circumvent the requirements of ordinary criminal procedure. One of 
the themes of this chapter - and of chapter 4, when we revisit this problem in the 
context of the communal ban 14 - is an exploration of the extent to which the 
medieval jurists confronted the tension inherent in their authorities between ordinary 
and extraordinary procedure, and the extent to which they reconciled the hostis with, 
or in fact isolated him from, the general precepts of criminal jurisprudence. This 
issue is important for two reasons. Firstly, as we shall see in due course, after 
Bartolus both the communal and the imperial bannitus were considered to be modem 
versions of the hostis. Secondly, as we saw in chapter 1, not wholly unlike the 
ancient Roman senate and emperors, the medieval Italian cities and the German 
emperors resorted to outlawry in dealing with their political opponents. Sometimes 
they sought to bypass normal criminal procedure in order to do so; ipso iure 
sentencing for treasonous acts can thus be found in municipal law. 15 As will becoi:ne 
clear in this chapter and in chapter 4, the medieval civilians did not, in procedural 
terms, embrace the ancient Roman legacy of the hostis, but generally tried to limit 
the impact of concepts of extraordinary procedure. The controversy caused by 
Emperor Henry VIl's banishment of King Robert of Naples towards the beginning of 
the fourteenth century resulted in a consensus that key procedural rights of 
defendants were guaranteed by natural law, which meant that even the prince could 
not ignore them. 
3. The hostis in Medieval Thought: Theoretical Considerations 
The medieval civilian material on the hostis can be split (somewhat artificially) into 
two. Firstly, there are the more academic considerations of the jurists contained in 
13 Mommsen, Romisches Strajrecht, pp. 256-9. The latest examples given by Mommsen are 
Stilicho and Gilda, who were both declared hastes in 397 by the senates of Constantinople 
and Rome respectively. For further details, see Momrnsen, Romisches Staatsrecht, vol. 3, pp. 
1240-51. 
14 Ch. 4, pp. 213-21. 
15 Ch. 1, pp. 18-23. 
98 
their glosses and commentaries to the Roman law in which they discussed the hostis 
in the abstract, without the immediate intention of using this figure as an analogy for 
contemporary legal phenomena. Secondly, we have Bartolus' application and 
development of hostis jurisprudence in the more practically orientated context of his 
discussion of the medieval imperial bannitus. For the sake of clarity, and in order to 
acquire a sense of how the discourse developed over time, it makes sense to begin 
with the former. 
i. Procedure 
As ever, the Glossa Ordinaria is the best starting point. 16 In his gloss to the I. 
Amissione, Accursius begins by affirming that the subject matter is indeed a case of 
capitis deminutio media. 17 He then elaborates the three ways mentioned in that law 
by which a person can become a hostis. Firstly, he explains that the opening of the 
passage, which speaks of those who 'join the ranks of the enemy', refers to 
transfugae, 18 and that these 'cease to be citizens' but retain their liberty and do not 
become penal slaves. 19 Secondly, he confirms that the senate can judge persons to be 
hastes, and that these enemies did not originate from beyond the empire but rather 
used to be its own subjects (subditi), that is, Roman citizens.20 Thirdly, he clarifies 
the tricky phrase 'lege lata' in the following way: 
... deemed so by a law, that is, a law of the people, which either designates a specific 
person as a hostis, or which generally states that whoever commits a certain act suffers 
capitis deminutio, that is, loses his citizenship, as is the case whenever someone falls 
d lty . . 21 un er a pena zpso zure. 
In other words, the people (presumably the Roman people) can pass a special law for 
the sake of designating a certain person as a hostis, or pass a general law which sets 
down that any person becomes a hostis upon the commission of a certain criminal act 
16 The Ordinary Gloss to D.4.5 .5,1 is briefly summarised by Cavalca, II bando, p. 89. 
17 Accursius to D.4.5 .5,1 , col. 455: 'Minuuntur. media capitis diminutione.' 
18 Ibid.: ' Et in hostium. ut transfugae. Accursius.' 
19 Ibid.: 'Conferunt. quia cives esse desinunt, ut [D.49.15.19,7]. non tamen est servus, ut 
[D.26.1.15].' 
20 Ibid.: 'Hastes. qui erant de subditis. Accursius.' 
21 Ibid.: ' .. .lege lata, scilicet a populo, vel exprimendo personam alicuius, vel generaliter, ut 
quicunque hoe fecerit, sit capite minutus, vel civitatem amittat, ut et dicitur quis ipso iure 
cadere in canonem latae sententiae quandoque. Accursius.' 
99 
ipso iure. Regarding the latter, Accursius may have had in mind the case of the 
transfuga, even if he never explicitly states that this status is acquired ipso iure. 
Elsewhere, Accursius indicates that the acquisition of hostis status is equivalent to a 
declaration of war. Glossing the well-known l. Hastes (D.49.15.24), which states that 
enemies 'are those on whom the Roman people has publicly declared war, or who 
themselves [declare war] on the Roman people', Accursius explains that the hastes in 
question could also be subditi, understood as in the I. Amissione. 
22 In effect, 
Accursius dissolves any boundary between internal and external warfare. Public war, 
properly speaking, can also be waged against the domestic enemy. The Roman 
people can declare war on any private citizen within the empire, and conversely, any 
private citizen can declare war on the community; and the methods by which the 
Roman people can declare war on a citizen are those outlined in the l. Amissione. It is 
also a logical inference that desertion to the enemy is a way in which a citizen can in 
tum declare war on the Roman people. We should bear in mind that while the 
discussion might be couched in terms of the people, it was established jurisprudence 
by the time of Accursius that the Roman people had transferred its jurisdiction to the 
emperor through the /ex regia. 23 This means, as we shall see very clearly when 
discussing Bartolus' thought on the matter, that the powers relating to the hostis 
declaration are lodged in the Roman emperor. 
This is the extent of Accursius' procedural considerations on the hostis. They 
constitute a logical if conservative interpretation of his sources. What is missing is 
detail about how precisely hostis status is acquired through the various ways 
described above, and in particular, how these relate to the norms of ordinary criminal 
procedure. It is remarkable that the Ordinary Gloss, which normally cites fastidiously 
every law that could possibly relate to a given text under discussion, never once 
references the hostis as a potential exception to normal procedure. In this, however, 
Accursius is no different to his successors, including Albericus de Rosate, Bartolus, 
and Baldus. In their Digest commentaries, when they discussed the hostis in abstract 
terms, the fourteenth-century jurists did not develop the procedural points raised by 
22 D.49.15.24: 'Hostes sunt, quibus bellum publice populus Romanus decrevit vel ipsi 
populo Romano ... ' The translation is from Watson (ed.), The Digest of Justinian, vol. 4, p. 
406. Accursius comments, col. 1517: 'Decrevit. etiam subditi, ut [D.4.5.5,1]. ' 
23 As stated in Justinian's Constitutio Deo Auctore (tit. De conceptione Digestorum), 
D.1.4.1, and Inst.1.2.6. The precise nature of that transfer was, of course, a matter of debate 
and is one of the key questions in the history of Roman law and political thought - for useful 
recent summaries, see Ryan, 'Bartolus of Sassoferrato', pp. 66-71 , and his 'Political 
Thought', p. 423-33. 
100 
the Gloss, nor did they cite the l. Amissione as an exception m their detailed 
theoretical discussions of ordinary criminal procedure. 24 
ii. Legal Status 
As in the case of procedure, abstract considerations on the substantive legal status of 
the hostis remained largely unchanged from the days of the Gloss; very little 
conceptual development can be found in the Digest commentaries of subsequent 
jurists. In fact, commentary on the relevant laws - granted that there are very few of 
them - is scarce and reveals a certain lack of interest in developing the law 
connected to this figure. Generally speaking, the jurists did little more than accept 
the provisions regarding the hostis and the transfuga contained in the Corpus iuris. 
Despite the scarcity of evidence, it is possible to learn a number of things. It is clear 
that the hostis can be generally equated to the deportatus insofar as both lose their 
citizenship and civil rights. We have already seen above how the former ceases to be 
a citizen upon acquiring hostis status. Albericus de Rosate points out that as a result 
' ' 
he also loses ea quae sunt iuris civilis, such as active and passive testamentary 
powers.
25 
In this context we can find two other civil rights already discussed in 
chapter 2. Firstly, one passage states that patriapotestas is lost by a transfuga. 
Accursius explains that we should understand this loss just as in the case of 
deportation.
26 Secondly, the law considers the effect of acquiring hostis status upon 
24 
Cf. Albericus de Rosate, Bartolus, and Baldus to D.4.5.5, D.41.1.51, D.48.8.3,6, and 
D.49.16. 7. Albericus is a partial exception: he develops the procedural jurisprudence of the 
Gloss at D.4.5.5, 1 in the context of communal banishment. Since the communal bannitus is a 
different figure to the imperial hostis, it makes better sense to discuss Albericus' thought in 
eh. 4 (pp. 215-18). For references to the jurists' discussion of ordinary criminal procedure, 
see eh. 2, p. 54, n. 73 . 
25 · -
Albencus de Rosate to D.4.5.5, 1 fol. 263vb: 'Notabilis est iste § et glossae hie positae ad 
materiam ... transferentium se ad hostes, quod civitatem amittant, id est, ea quae sunt iuris 
civilis, sicut testamenti factionem in activa et passiva significatione, et alia quae competunt a 
iure civili.' 
26 
D.49.l~ .19,7: 'Filius quoque familias transfuga non potest postliminio reverti neque vivo 
patre, qma pater sic ilium amisit, quemadmodum patria, et quia disciplina castrorum 
antiquior fuit parentibus Romanis quam caritas liberorum.' Accursius comments, col. 1514: 
'Ami sit. quo ad patriam potestatem, ut per deportationem, ut [Inst.1.12, 1].' No disagreement 
with the Gloss' interpretation of the law can be found amongst later jurists - see Jacobus de 
Arena, Rainerius de Forlivio, Albericus de Rosate, and Bartolus to D.49.15.19,7. Cf. eh. 2, 
pp. 69-72. 
101 
one's position as guardian. We saw in the previous chapter that guardianships were 
lost through capitis deminutio media, although there remained an ambiguity over 
whether this was connected to the loss of civil rights. 
27 In the context of the 
transfuga, the matter is initially complicated by one law (D.26.1.15) which denies 
that a transfuga loses his guardianships because he remains a free person, i.e. he only 
loses his citizenship; by implication, the ability to hold guardianships is not a civil 
right nor is it lost through capitis deminutio media. 
28 Accursius provides two 
possible solutions to this tension. According to the first, the transfuga remains a 
guardian only to his 'detriment' ('praeiudicium') but 'not truly' ('non vere'). 
According to the second solution, which Accursius views as 'more truthful ' 
('verius'), the transfuga mentioned at D.26.1.15 is not a real deserter because he did 
not go over to the enemy 'animated by treachery' ('proditorio animo'). Rather, he 
only pretended to desert in order to scout out the enemy. But because he did this 
against the wishes of his superiors, he is called a transfuga nonetheless. He does not, 
however, lose his citizenship and therefore remains a guardian. If he were a real 
deserter, then he would indeed forfeit his citizenship, and as a consequence, also his 
guardianships.29 In other words, according to this solution, which is the one favoured 
by the commentators as well, the ability to hold guardianships is a right dependent 
upon the possession of citizenship. 30 
As has already been indicated above, beyond the loss of citizenship and civil rights, 
the hostis suffered a set of penalties that were far harsher than those suffered by the 
deportatus . Firstly, the law contains a number of punitive provisions specifically 
regarding the citizen who had become a hostis. The medieval civilians acknowledged 
these and fully accepted the idea that the transfuga could be killed with impunity by 
27 Ch. 2, pp. 75-9. 
28 D.26.1.1 5: ' Si quis tutor non sit captus ab hostibus, sed missus ad eos quasi legatus, aut 
etiam receptus ab eis, aut transfugerit, quia servus non efficitur, tutor manet, sed interim a 
~raesidibus alius tutor dabitur.' 
9 Accursius to D.49.15.19,7, col. 1514: ' desinit ergo esse civis, ut [D.4.5.5, 1]. et sic est 
media capitis diminutio .. . ergo debet desinere esse tutor... sed contra [D.26.1.1 5]. Solutio: ibi 
remanet tutor in sui praeiudicium, non vere. vel verius, ibi non animo proditorio, hie sic, ideo 
ibi non desinit esse civis, hie sic. vel die ut ibi.' Accursius to D.26.1. 15, col. 107: 'Sed nunc 
transfuga desinit esse civis, ut [D.49.15.19,4] et [D.4.5.5,1 ], ergo desinit esse tutor. .. et hie 
<licit tutor manet. Solutio: hie non proditorio animo, sed sic fingebat ut videret statum 
hostium, tamen transfuga est quia facit contra voluntatem populi Romani, et hie non desinit 
esse civis, et sic tutor manet. secus est cum proditoris animo, ut [D.49.15.19,4].' 
3° Cf. Albericus de Rosate, Bartolus, and Baldus to D.26.1.15. The only exception is 
Rainerius de Forlivio, who thinks that Accursius' first solution is the correct one - to 
D.49.15.19, 7, fol. 135vb: 'glossa opponit [D.26.1.15]. respondet tribus modis: prima vera.' 
102 
anyone. 
31 
Both Albericus de Rosate and Bartol us highlight that even a son and father 
may kill each other should one of them desert to the enemy; at least in this sense, a 
son ought to be more obedient to his fatherland than to his father. 32 Furthermore, the 
deserter's voluntary association with the enemy meant that he did not possess the 
right of postliminium, which was only granted to those captured by the enemy, nor 
could he receive restitutio in integrum. 33
 If he is captured by Romans and not killed 
on the spot, the law allows him to be tortured, hanged, or burned alive.34 The only 
way he might be spared punishment is if he betrays and thereby causes the capture of 
a large number of other deserters and brigands. 35 
Secondly, the hostis is subject to the provisions of the Roman law which generally 
regulate the relationship between Romans and their war-time enemies, according to 
the dictum of D.41.1.51 that '[w]e receive a transfuga by the law of war'. 36 In their 
commentaries, the civilians clarified that this law or right of war (ius belli), grants 
Roman citizens the same powers over the hostis-transfuga as over any other foreign 
enemy. 
37 
These powers, as Albericus de Rosate points out, are set down in an 
important passage earlier in the same title of the Digest,38 which highlights that in 
Roman jurisprudence the fundamental aspects of the law of war were derived from 
the ius gentium: 
31 
Cf. Accursius, Jacobus de Arena, Rainerius de Forlivio, Albericus de Rosate, and Bartolus 
to D.48.8.3,6. 
32 
Al?ericus . to D.49.15.1_9, 7; fol. 222va: 'Item, allegatur iste § quod fil ius magis tenetur 
ob~dire patn ae qua1!1 patn. [ ... ] et quod filius magis teneatur patriae quam patri multum facit, 
gm~ films potest mterficere patrem contra patriam pugnantem, ut [D. I I. 7.35: Minime 
~a10re~ lugendum putaverunt eum, qui ad patriam delendam et parentes et liberos 
mterfic1endos venerit: quern si filius patrem aut pater fili um occidisset sine scelere etiam 
ffaemio adficiendum omnes constituerunt.].' Cf. Bartolus to D.49.15.19, 7. ' 
~ .49.1 5. ! 9;4: ?ra?sfuga~ nullum postliminium est: nam qui malo consilio et proditoris 
ammo patnam rehqmt, hostmm numero habendus est.' Accursius adds col. 1514: 'Nullum 
postliminium. nee restituitur in integrum, licet laesus, ut [D.4.6.14]. ' Cf. Jacobus de Arena 
!ainerius de Forlivio, and Albericus de Rosate to D.49.15.19,4. ' 
35 D.49.16.7, D.48.19.8,2, and D.48.19.38,1 w
ith the Ordinary Gloss. 
D.49.16.5,8 and Ordinary Gloss. 
36 
D.41.1.51: 'Transfugam iure belli recipimus.' 
37 
J_aco?us de Arena to D.41.1.51, fol. I68vb: 'Ibi, "iure belli", concesso nobis in hostes.' 
Ramenus de Forlivio, ibid., fol. 37rb: 'Transfugam. Ibi, "belli", concesso nobis in hostes 
secundum Ja. de Are.' ' 
38 
Albericus de Rosate, ibid., fol. 74rb: 'Transfugam. Ibi, "iure belli", id est, iure nobis in 
hostes concesso. Ja.[ ... ] hoe intellige ut dixi [D.41.1.5,7].' 
103 
Again, property taken from the enemy is forthwith the property of the taker under the 
law of nations ... so that also free men are reduced to slavery but those who escape the 
power of the enemy retain their original freedom. 39 
In other words, by the law of nations I can acquire legal title over the property of my 
vanquished enemy - hence D.41.1.51, 1 states that 'property of the enemy, which is 
on our territory, becomes not public property, but that of the first taker. '
40 
Bartolus 
further explains that in respect of a just war - and this can be a war within the 
commonwealth, as 'between the church and its subjects' - I cannot be sued, after 
peace has been made, by my former enemies for property taken from them during the 
time of war, since I have lawfully gained title over it by the ius gentium.
41 
It is also 
by the law of nations that I can acquire legal title over enemy persons. Thus it is 
equally possible for Roman soldiers captured in war to be enslaved by the enemy, 
which in turn gives rise to the institution of postliminium - the reacquisition of 
freedom and citizenship upon return to Roman territory from enemy captivity. 
Because these rules apply to the transfuga, Accursius, explaining D.41.1.51, 
comments that he 'will become our slave if before he was a free man' .
42 
The Roman law contained one other important regulation regarding the hostis: fraud 
and malice (do/us) are morally good as well as lawful if committed by a private 
39 D.41.1.5,7-D.41.l.7: 'Item quae ex hostibus capiuntur, iure gentium statim capientium 
fiunt... adeo quidem, ut et liberi homines in servitutem deducantur: qui tamen, si evaserint 
hostium potestatem, recipiunt pristinam libertatem.' The translation is from Watson (ed.), 
The Digest of Justinian, vol. 4, p. 2. 
40 D.41.1.51, 1: 'Et quae res hostiles apud nos sunt, non publicae, sed occupantium fiunt.' 
The translation is from Watson (ed.), The Digest of Justinian, vol. 4, p. 12. While the ability 
to acquire enemy property was rooted in the law of nations, the Roman law further refined 
this rule with a number of qualifications. The medieval civilians understood these rules in 
the following way - cf. Rainerius de Forlivio, Albericus de Rosate, and Bartolus to 
D.41.1.5,7. Only movables (mobilia) fall to private persons: booty is collected by the war-
time leader, such as a duke or a king, and then distributed by him to his soldiers according to 
their merit. Captured real estate (immobilia), on the other hand, falls to the fisc unless it used 
to be owned by a Roman citizen, in which case it would be returned to the previous owner. 
41 Bartolus to D.41.1.5,7, fol. 76vb: 'Facio unam quaestionem. quidam de quodam castro 
habuerunt guerram cum duce. Mariscallus ibi concurrit et habuit conflictum. modo, facta 
pace, volebat sibi restitui damnum. an debet repetere? Et dico quod non, quia de iure 
gentium erat effectum suum quod cepit, et sic de iure civili non tenentur depraedantes, cum 
licitum fuit bellum inter ecclesiam et suos subditos.' 
42 Accursius to D.41.1.51, col. 357: 'Recipimus. id est, noster sit servus si prius fuerit liber, 
et etiam interficere licet, ut [D.48.8.3,6). ' Cf. Jacobus de Arena to D.49.15 .19,7, fol. 230rb: 
' .. . transfuga captus servus fiat admodum hostis capti, ut [D.41.1.51] .' 
104 
. b · 43 person agamst pu he enemies. According to Baldus, this means that one may 
theref?r_e lawfully ensnare, ambush, betray, and capture an enemy.44 While all this 
was _hc1t ~or a private person, the jurists qualified this rule in respect of public 
officials, smce the law also states that a war-time leader has to keep faith with th 
.f e enemy I the two sides have entered into an agreement such as a truce. 45 
4. Imperial Banishment before Bartolus 
Havin~ covered the jurists' abstract deliberations on the hostis , we will now move on 
to their more practically orientated thought concerning the imperial bannitus. In 
contrast to the vast amount of material on the communal ban, civilian considerations 
on the imperial ban are thin on the ground, and the ones that exist tend to be brief. 
Before Bartolus, I can only find two notable contributions to this topic: first the 
Ordinary Gl~ss, followed by the commentary of Jacobus de Belviso (d. 1335) to the 
rele~ant sect10n of the Libri feudorum. Later, Baldus went on to discuss the imperial 
ban m a number of places, but for the most part, he follows the interpretation of his 
master, Bartolu~, and repeats the material of his predecessors; while it will prove 
u~eful _to ~ent10n aspects of his work at times, it does not merit a separate 
~1scuss1on. The state of modem historical scholarship reflects the lack of evidence 
m the sources: civilian discussions of imperial banishment before Bartolus even the 
more significant ones in the Ordinary Gloss and by Jacobus de Belviso,' have not 
been examined by historians. 
43 D.4.3.1,3 : ·~o~ fuit autem c?ntentus praetor dolum dicere, sed adiecit malum, quoniam 
veteres dolum etJam bonum d1cebant et pro sollertia hoe nomen accipiebant maxime si 
!idversus hostem latronemve quis machinetur.' ' 
. ~aldus to D.4.3.1,3, fol. 230rb: 'Nota quod licet ponere insidias hosti et latroni quia 
hc1tum est eos capere.' 
45 B~rtolus to D.4.3.~'~' fol. 129va: 'Bonus est dolus qui committitur adversus hostes 
pu~hcos. h. d. No_ta, hc1tum est decipere hostem, quod intellige in privato. Nam dux belli, si 
fac1tyac~m host!, debet ill_ud servari, [D.2.14.5: publica conventio est, quae fit per pacem, 
quotlens mter se duces bellt quaedam paciscuntur].' For the same distinction see Accursius 
and Baldus to D.4.3.1,3. ' 
46 Baldus discusses or at least mentions the imperial ban at: auth. Cassa, post C.1.2(5).12; 
auth. Item nu/la, post C.1.3(6).2; auth. Item quaecunque, post C.1.3(6).13; C.1.3(6).32(31) · 
C.7 .53.8; LF.2.28 c. Domino guerram. ' 
105 
One can only speculate as to why the jurists did not discuss the imperial ban more 
than they did. In the case of the glossators, this could be explained by their 
preoccupation with explaining and reconciling the texts within the Corpus iuris; we 
would not expect them to devote much space to contemporary legal problems. But 
given the growing civilian interest in these following the practical tum of the later 
thirteenth century, it is remarkable that the commentators did not devote more 
attention to this issue. The implication is that the emperor's ban, at least after the 
middle of the thirteenth century, did not have much traction and was therefore not a 
pressing legal problem for the Italians, despite the fact that emperors - notably Henry 
VII and Louis the Bavarian - continued to descend into Italy and banish their 
opponents well into the fourteenth century. Perhaps, at a broader level, this is a 
reflection of the decline of the empire's grip on Italy following the death of Frederick 
II (d. 1250). 
i. The Legislation of Frederick II 
One potential reason why imperial banishment was not discussed at greater length is 
the fact that this medieval legal practice left merely traces in the jurists' collection of 
authorities, being mentioned only in a handful of texts in the Corpus iuris; naturally, 
these texts are medieval rather than classical. Most of them come from Frederick II's 
1220 coronation constitutions, which were promulgated jointly with Pope Horiorius 
III and designed to protect the liberties of the church from infringements by local 
legislation and custom. At the emperor' s request, the Bolognese law faculty 
incorporated this legislation into Justinian 's Code, where one can find it broken up 
into eleven separate authenticae.47 Frederick's constitutions were part of a wider 
program of imperial legislation that reaches back to the twelfth century, whose 
purpose was to defend the interests of the church, especially by punishing spiritual 
47 Cf. Coing, Handbuch, pp. 156 and 161; Lange, Die Glossatoren, p. 79. For an edited text, 
see Weiland (ed.), MGH, Leg. sec. IV, Tom. II, pp. 106-9. While Frederick's legislation is 
also known as the Constitutio ad decus or the Constitutio in basilica beati Petri, the 
medieval jurists viewed it as a group of separate laws with the name Constitutiones de 
statutis et consuetudinibus contra libertatem ecclesiae. Under this title, the full text of 
Frederick's legislation is also presented in the Volumen parvum of the medieval Corpus 
iuris, as part of the tenth collation, immediately after the Librifeudorum. 
106 
offences (such as heresy) and by giving additional force to ecclesiastical sentences of 
excommunication. 48 
For a proper understanding of the coronation constitutions, we need to be aware of 
the main features of excommunication in canon law in the period following the 
publication of Gratian's Decretum (ea. 1140). 49 There were two basic types of 
excommunication: excommunicatio minor and excommunicatio maior. The former 
led to a lighter set of penalties, such as denial of the sacraments and a prohibition on 
social interaction with other Christians, while the latter caused a severe loss in the 
civil capacity of the excommunicate (though the extent of that loss varied depending, 
for instance, on the nature of the excommunicate's offence). Generally speaking, 
excommunicatio maior barred an individual from substantial parts of the legal and 
public sphere. The contents of that sphere changed as canonistic doctrine developed 
over time, but could include legal acts of a more private nature (actus legitimi), such 
as the making of testaments and contracts, as well as legal acts of a more public 
nature (actus forenses), thus preventing the excommunicate from being in court as 
plaintiff, judge, or advocate, holding public offices or electing others to them, or 
exercising the notarial profession. An excommunicated vassal was liable to lose his 
fief, and an excommunicated lord could no longer demand feudal services from his 
vassal. Much in the same way as the communal and the imperial ban in civil law, 
excommunication had two broad functions. It was used as a punishment for crimes, 
but also as a coercive procedural device with the purpose of bringing a contumacious 
defendant back to obedience to the ecclesiastical authorities. 
Let us now examine Frederick 's legislation in detail. His first constitution subjects 
any place to ' the ban of a thousand marks ' if its officials presume to make any 
statutes infringing upon the liberties of the church. 5
0 It is possible that this ban 
merely constituted a monetary penalty - bannum in its older sense. The choice of 
words (bannum subiacere, rather than bannum solvere, componere), however, 
implies not so much a fine as a sanction upon the status of a person. If the meaning is 
48 
See Eichman?, Acht u~d Bann, pp. 27-63 (part 2), 70-3, 80-7, 95-7, and 111 -46 (part 4), 
but note that his focus 1s on the German rather than the Italian parts of the empire. Cf. 
Poetsch, Die Reichsacht, pp. 36-41. 
49 
The following summary is based on Eichmann, Acht und Bann, parts 2-4, pp. 27-146; the 
test rr.10dem Anglophone guide remains Vodola, Excommunication. 
Weiland (ed), MGH, L~g. sec. IV, Tom. II, p. 108: ' ... necnon Iocum ubi deinceps talia 
presumpta fuermt banno mille marcarum precipimus subiacere. [ ... ] Qui si per annum huius 
nostre cons~itutionis inven~i fuerint contemptores, bona eorum per totum nostrum imperium 
mandamus 1mpune ab ommbus occupari ... ' 
107 
outlawry, then the monetary penalty refers to a condition for the cancellation of the 
ban. We should note that the version ( or rather, paraphrase) of this law which came 
to be inserted into the Code as the authentica Cassa did not include this phrase. 
51 
Just as the original, however, Cassa did include the penalty that those who continue 
to contravene its demands will, after one year, suffer loss of legal protection in 
respect of their goods, which would be yielded to the first occupier. 
52 
It is hard to see 
this as anything else than an aspect of the imperial ban. 
There are four other relevant constitutions. Item nulla threatens the imperial ban if a 
person or community imposes taxes on the Church and remains contumacious in the 
face of imperial or ecclesiastical summons to court. 53 Item quaecunque imposes the 
imperial ban ipso iure if a person or community continues to remain 
excommunicated after one year for violating the liberties of the church.
54 
Both leave 
open the possibility for the cancellation of the ban as long as the required satisfaction 
has been made - monetary satisfaction in Item nulla, and ecclesiastical absolution 
from excommunication in Item quaecunque. The third authcntica is Gazaros and has 
two parts. In the first, the emperor declares that all heretics fall under the ban and 
suffer a number of other penalties, such as perpetual infamy and the confiscation of 
their estate - presumably this applies to those who have already been condemned as 
heretics by an ecclesiastical court. In the second part, the ban appears as a measure 
against those who are under suspicion of heresy but refuse to face the charge: they 
51 The civilians were aware of this difference - Odofredus to auth. Cassa, post C.1.2(5).12, 
fol. 1 Ovb: ' ... et in longa constitutione de qua est extracta hec constitutio, que est sub rubrica 
"De statutis et consuetudinibus contra libertatem ecclesie conditis", § "cunctis", in verbo 
"hac edictali", x. coll. in corpore authentici.' Cf. Albericus de Ro sate and Baldus to ibid. 
52 post C.1.2(5).12 - e.g. in the 1558-60 Lyon version of the Corpus iuris civilis, cols 2~-5: 
'Cassa et irrita esse denuntiari per totam Italiam praecipimus omnia statuta et consuetudmes 
contra libertatem ecclesiae eiusque personas inductas adversus canonicas et imperiales 
sanctiones, et ea de capitularibus penitus aboleri mandat nova constitutio; et de caetero 
similia attemptata, ipso iure nulla esse decernimus. si quid contra fiat, poenae quae statutae 
sunt imminebunt. Sed si per annum huius novellae constitutionis aliqui inventi fuerint 
contemptores, bona eorum per totum nostrum imperium impune ab omnibus occupentur.' 
53 post C.1.3(6).2. Weiland (ed.), MGH, Leg. sec. IV, Tom. II, p. 108: 'Item nulla 
communitas vel persona, publica vel privata, collectas sive exactiones, angarias vel 
parangarias ecclesiis aliisque piis locis aut ecclesiasticis personis imponant aut invadere 
ecclesiastica bona presumant; quod si fecerint et requisiti ab ecclesia vel imperio emendare 
contempserint, triplum refundant, et nichilominus banno imperiali subiaceant, quod absque 
satisfactione debita nullatenus remittatur.' 
54post C.1.3(6).13. MGH, ibid.: 'Item quaecunque communitas vel persona per annum in 
excommunicatione propter libertatem ecclesie facta perstiterit, ipso iure imperiali banno 
subiaceat, a quo nullatenus extrahatur, nisi prius ab ecclesia beneficio absolutionis obtento.' 
108 
are subject to the ban for one year, after which time they are considered as 
condemned for heresy.55 Lastly, Credentes subjects those to the imperial ban who 
have been excommunicated for supporting heretics and are not absolved by the 
church after one year. In this case, the ban seems to be connected to a number of 
specific sanctions which mirror the effects of excommunicatio maior in canon law: 
they become infamous and lose active and passive testamentary powers; they can no 
longer hold or elect others to public office, act as witnesses, bring an action in court 
(while remaining liable to being sued by others and appear as defendents), or act as 
judges, advocates, and notaries.56 We should note, however, that, similar to the case 
of Cassa, the word 'ban' ('bannimus') does not seem to appear in the version of 
Credentes incorporated into the Code, where it is replaced with 'damnamus'.57 If 
'damnamus' is used, the meaning of the passage changes: for then the emperor no 
longer imposes his ban, nor are the penalties in this law associated with the latter but 
' 
are imposed as free-standing secular penalties for persistent excommunication; or 
rather, they constitute in large part a restatement in civil law of the effects of 
excommunicatio maior in canon law. 
Although we cannot gather details from these highly rhetorical passages about the 
precise procedural and enforcement mechanisms that pope and emperor may have 
had in mind, it is clear that the imperial ban in these constitutions follows the 
55 17ost c_. 1.5(8).10. MGH, ibid.: 'Chataros, Patarenos, Leonistas, Speronistas, Arnaldistas, 
C1rcumc1sos _et o~nes _hereticos utriusque sexus, quocumque nomine censeantur, perpetua 
dampnamus mfamia, d1ffidamus atque bannimus, censentes ut bona talium confiscentur nee 
ad eos ~lteriu~ revertantur, ita quod filii ad successionem eorum pervenire non possint, cum 
longe sit gravms etemam quam temporalem offendere maiestatem. Qui autem inventi fuerint 
sola. suspitione notabiles, nisi ad mandatum ecclesie iuxta considerationem suspitionis 
~uahtatemque p_e~sone pr~priam innocentiam congrua purgatione monstraverint, tamquam 
mfames et banmtl ab ommbus habeantur, ita quod sic per annum permanserint ex tune eos 
sicut hereticos condempnamus.' ' 
56 po~t C.1.5(8).4. MGH, ibid.: 'Credentes preterea, receptatores, defensores et fautores 
heretlcorum bannimus, firrniter statuentes, ut si postquam quilibet talium fuerit 
~xco~municatione n?tatus, satisfacere contempserint infra annum, extunc ipso iure sit factus 
mfam,s nee ad pubhca officia seu consilia vel ad eligendos aliquos ad huiusmodi nee ad 
testimonium admittatur. Sit etiam intestabilis, ut nee testamenti liberam habeat factionem 
?ec ad. ?ereditatis successionem accedat. Nullus preterea ei super quocumque negotio, set 
1pse alu respondere cogatur. Quod si forte iudex extiterit, eius sententia nullam obtineat 
firmitatem, nee cause alique ad eius audientiam perferantur. Si fuerit advocatus eius 
p~trocinium _nullatenus admittatur; si tabellio, instrumenta confecta per ipsum nullius p~nitus 
smt moment1.' 
57 
auth. Credentes, post C.1.5(8).4 - in the 1558-60 Lyon version of the Corpus iuris civilis 
col. 96: 'Credentes praeterea, receptatores, et defensores haereticorum damnamus, firmite; 
statuentes ... ' 
109 
patterns introduced in chapter 1, that is, it can be divided conceptually into coercive 
procedural and permanent punitive elements.58 For the most part, what Frederick did 
was to provide the church with another weapon against those who refused to obey its 
authorities. In certain circumstances, their status was now further aggravated with the 
imperial ban, which, like excommunication, appears in these laws primarily as a 
coercive procedural device. Hence the repeated provisions that, if a malefactor (more 
broadly) or excommunicate (more narrowly) remains unrepentant and does not seek 
ecclesiastical absolution, he should also be burdened with the imperial ban until, as is 
mentioned in two of the authenticae, he has submitted once again to the church. In a 
secondary sense, the imperial ban is a punishment in and of itself, particularly for 
those who have already been condemned by the church for the crime of heresy. 
The Ordinary Gloss provides very little help for understanding these constitutions. 
They are barely glossed, and if they are, Accursius never explains what is meant by 
the imperial ban.59 His contemporary rival, Odofredus de Denariis (d. 1265), glosses 
three of the authenticae in his Lectura on the Code, and although his remarks are 
more expansive than Accursius' glosses, they again tell us next to nothing about the 
imperial ban itself. In this regard, Odofredus restricts himself to a close paraphrase of 
the law under discussion. 60 Odofredus briefly alludes to the imperial ban in one other 
place: in his commentary to the 1183 Peace of Constance (between Frederick I and 
the Lombard cities), where he states that a city subject to the ban of the emperor or 
the pope suffers a form of capitis deminutio and hence loses any privileges which it 
had previously received.61 
58 Ch. 1, p. 19. 
59 While there are a number of glosses at Cassa and Item nu/la, Credentes only has one brief 
gloss, and Item quaecunque and Gazaros are not glossed at all. 
60 Odofredus to: auth. Cassa, post C.1.2(5).12, fol. lOvb ('Tertio, imponit penam contra 
facientibus, et dicit quod, si aliqui statuta fiant contra libertates ecclesiarum, etiam civitates 
de statutis per annum debent auferre, alias bona eorum per universum orbem publicentur. '); 
auth. Item nulla, post C.1.3(6).2, fol. 21 vb ('Secundo dicit, si aliquis contra fecerit, tenetur 
reddere acceptum in triplum et banno imperiali subijcietur, a quo non extrahetur nisi 
satisfecerit. '). Odofredus also mentions the imperial ban at auth. Omnes peregrini, post 
C.6.59.10, fol. 89ra: 'Et qui contra hanc constitutionem fecerint banno imperiali 
subponitur.. .' 
61 Vienna, Osterreichische Nationalbibliothek, MS. 2094, fol. 24vb, a.v. In perpetuum (as 
quoted in Ryan, 'Bartolus of Sassoferrato', p. 69, n. 13): 'Item si posita fuerit civitas in 
banno imperatoris quasi patiatur capitis diminutionem [ .. . ] vel in banno domini pape posita 
fuerit. his autem casibus potest dici privilegia cessare .. .' 
110 
The only other mention of Frederick II's ban in the age of the glossators can be 
found in a treatises on procedure by Roffredus (d. ea. 1243), his Libelli iuris civilis et 
canonici. In the seventh part of the Libelli iuris canonici (de accusationibus, 
inquisitionibus et denunciationibus ), Roffredus explains criminal procedure 
acc~rding to the ius commune and, in passing, voices his distaste for contemporary 
bamshment as a form of coercion against the contumacious defendant: 'Nor do I 
admit that he is placed under the ban, since this would be very harsh, that banished 
thus, he could perchance be punished and harmed by anyone. ' 62 Shortly afterwards, 
however, Roffredus leaves aside criminal procedure according to the ius commune in 
favour of a short digression on procedure in Sicily according to the constitutions 
promulgated by Frederick II for that kingdom in 1231 (known as the Constitutions of 
Melfi or the Liber Augustalis).63 Roffredus is careful to make clear that although 
these laws were promulgated by the emperor, they are not general laws applicable 
eve~here as part of the ius commune, but particular laws only valid in the kingdom 
of S1c1ly - promulgated, in other words, by Frederick in his function as king of Sicily 
M ' 
not as Roman emperor. When Roffredus arrives at the problem of the 
contumacious defendant in criminal matters, he proceeds to outline over the course 
of a relatively brief rubric the basic provisions of the royal ban as it can be found in 
Frederick's constitutions.
65 Although different in parts, for our purposes the Sicilian 
ban is similar enough to its imperial equivalent in the north of the peninsula that we 
· do not need to discuss it in detail here. 66 Roffredus takes us through the different 
stages of the ban, its increasing severity following the defendant's persistent 
contumacy, and its final stage when, one year after the beginning of proceedings, the 
defendant is finally considered forbannitus with the usual disabilities: his goods are 
co~fiscat~d; he is denied access to the courts; he is considered to be legally dead and 
- mterestmgly, although Roffredus in no way develops the point - to be a 'public 
62 
Roffredus, Libelli iuris _canonici, fol. 44va: 'nee admitto quod ponatur in banno, quia hoe 
esset valde durum, ut pumretur forte a quolibet et offenderetur sic exbannitus ' 63 
Cf. Dilcher, Die Sizilische Gesetzgebung. · 
64 
Roffr_ed~s, Libelli iw:is . cano~ici, _fol. 45ra: 'Qualiter autem in regno Sicilie in 
accus~hombus per constltutiones 1mpenales procedatur, pro illis de regno aliquid tangam ... 
u_nde 1st_a que tract~ modo non sunt generalia iura, licet sint iura principis, sed sunt iura 
smgulana, que sunt m regno Sicilie tantum ' 
65 · . 
~~Id., f~I. 45r~, De accusato contumace qualiter puniatur. In the printed 1500 Avignon 
~6
d1t1on, this rubric takes up about three quarters of a column. 
Both Ficker, F?rschun?~fl_ I, pp. 2_ 18-23, . as well as Mooney, 'The Legal Ban', pp. 76-7 
and 124-5, explam the S1c1han ban m detail. A shorter description is given by Cavalca JI 
bando, pp. 69-70. ' 
111 
I I 
enemy' ('hostis publicus'); as such, he 1s liable to be harmed with impunity by 
anyone.67 
As we might expect, the commentators had a greater interest in the imperial ban than 
the glossators. Nevertheless, their engagement with this legal institution, at least in 
their commentaries to Frederick's authenticae, remained very limited, particularly 
until about the middle of the fourteenth century.68
 There was a general tendency to 
comment little or even not at all on Credentes and Gazaros, while their longer 
commentaries on Cassa, Item nulla, and Item quaecunque, were more concerned 
with other, related topics - notably excommunication and questions arising from 
communal statutes - than with an explanation of imperial banishment. Remarkably, 
Bartolus did not mention the imperial ban once in this context (although he did, of 
course, discuss it in detail elsewhere). His contemporary, Albericus de Rosate, 
touches on a few of the ban's aspects, and Baldus, in the second half of the 
fourteenth century, went on to discuss it at somewhat greater length, though, as 
mentioned above, mainly through the prism ofBartolus' ideas.
69 
Albericus' commentary raises three points of interest. 
7° Firstly, he claims that 
Odofredus associated the meaning of bannum in these authenticae exclusively with 
the monetary fine mentioned in the original, longer version of Cassa: that, for him, 
bannum mi/le marcarum was synonymous with poena mi/le marcarum. Albericus 
67 Roffredus, Libelli iuris canonici, fol. 45rb: 'et sic anno transacto, incipiente currere 
postquam per litteras annotationis fuerit publice nunciatum, bona ilia fisco quaeruntur.:. Si 
autem annus transierit, nee legitimam causam poterit sue absentie pretendere, tune pennde 
habebitur acsi esset per sententiam condemnatus ita quod sibi aditus vel iudicii imina 
precludantur, et nee appellare nee supplicare poterit forbannitus ita quod quo ad omnia fere 
pro mortuo habeatur, hostis publicus reputetur sic ut ab omnibus offendatur, et qui eum 
offenderit sit immunis ... ' It seems that Roffredus did not follow the precise nomenclature of 
the law in every respect. Ficker and Mooney (see n. 66 above) agree that the defendant is 
only called (for)bannitus within the one-year grace period after the ~eginning of 
proceedings, butforjudicatus once this period is over and the ban has been finahsed. 
68 Cinus de Pistoia (d. 1336), in his famous Lectura on the Code, commented on Cassa, Item 
nulla, and Gazaros but did not discuss the imperial ban. 
69 Albericus de Rosate and Baldus to auth. Cassa, post C.1.2(5).12, auth. Item nulla, post 
C.1.3(6).2, and auth. Item quaecunque, post C.1.3(6). I 3. 
70 Albericus' commentary to auth. Item quaecunque, post C.1.3(6).13, fol. 26ra, merely 
paraphrases the law in respect of the ban: 'Ibi, banno imperial i. de effectu huius banni dixi 
[auth. Cassa, post C.1.2(5).12]. et nota quod qui stat in excommunicatione per annum 
propter libertatem ecclesiae, subiacet banno imperiali. qui autem occasione haeresis stat in 
excommunicatione per annum, ut haereticus condemnatur, ut [auth. Gazaros, post 
C.1.5(8). l O].' 
112 
seems to concur with this view - at least he does not disagree with it - and although 
clearly aware of the loss of legal protection for property threatened at the end of 
Cassa, he does not link this, or any further sanctions, with the imperial ban. 71 
Moreover, Albericus summarily reports the conclusions of two quaestiones on this 
topic, which he also attributes to Odofredus. 72 The first concerns the cessation of the 
penalties of Cassa. If a malefactor returns to the emperor's grace and favour, can he 
then reclaim the property which was previously taken away from him by another 
when his goods did not enjoy the protection of the law? According to Albericus, 
Odofredus argued that he only recovers the said goods if they are specifically 
returned to him by the king or if he is generally granted restitutio in integrum. 73 The 
second quaestio examines whether the penalties of Item nulla - subjection to the 
imperial ban and the requirement to repay three times the amount taken from the 
church through illegal taxation - apply to the criminal even if he has not been 
summoned to court by the church. Odofredus is said to have answered in the 
affirmative regarding the repayment obligation, since this is a punishment for theft in 
the ius commune, but that the defendant would indeed have to be summoned for the 
ban (i.e. the fine of a thousand marks) to take effect. 74 
;~ Albericus de Rosate _to auth. Item nulla, post C.1 .3(6).2 - seen. 74 below. 
As can be observed m ns 73-4 below, Albericus repeatedly hints at the existence of a gloss 
(gl~ssa/apparatus/commentum) by Odofredus to Frederick's coronation constitutions from 
which ~e draw~ his mate~i~l. Albe_ricus ' descr!ption of this work in his commentary to auth. 
Cassa ( glossa m corpore I quaesho formatur m corpore, unde haec sumitur, per Odo.'), and 
to auth. Creden_tes,,Post C.1.5(8).4, _fo!. 41 vb ('et vide quod notatum per Odof. in originali 
unde haec sum1tur ), suggests that 1t 1s a gloss to Frederick's constitutions as included in 
Coll.. X . of the Authenticum; i.e. he is not referring to Odofredus' Lectura on these 
constJtut10ns as authenticae in the Code. Albericus' choice of words ('ut dicitur'/'dicitur 
commentasse'), ho;"'ever, also implies that he is not certain of the work's provenance. 
Ind~ed, Odofredus Lectura does not mention any such gloss nor contain the material 
ascnbed to the latter by Albericus - see p. 110, n. 60 above. Furthermore I have not found 
any traces of this work in the reference literature, and the quaestiones' discussed below 
~3annot be found under Odofredus' name in Bellomo, Quaestiones. 
Alb_e~i~us de Rosate to auth. Cassa, post C.1.2(5).12, fol. I2vb: 'lsta est nova constitutio 
Fedenc1 1mperatoris, q~i, se~undum ?ct., ~oronatus fuit a Papa Honorio Mccxx ... In glossa in 
corpore ~- Oldra .. [szc]... Sect qmd s1 post occupationem aliqui redeant ad gratiam 
lmp~ratons, nunqmd_ bona recuperent? Haec quaestio formatur in corpore, unde haec 
sum1tur, per Odo., _q~1, ut dici~r, fecit ibi apparatum. Ubi, post multa in effectu, tenet quod 
non recuperent, ms1 ad ea smt per principem restituti per restitutionem autem sic ut 
[C.9.51.?]. .. ' ' ' 
74 
Alb~ricus de Rosate to auth. Item nulla, post C.1.3(6).2, fol. 23ra: 'An autem si ab 
Eccles1a n_on requir~tur'. in~urrat poenas huius constitutionis? et videtur quod non, a c;ntrario 
sensu_ h~ms constJtutJoms. Sect, nota, dominus Od., qui dicitur comrnentasse istas 
const1~t1?nes Feder!ci, tenet quod etiam non requisiti teneantur ad poenam tripli, cum sit 
poena mns comrnums contra raptorem, ad poenam autem banni non teneantur nisi requisiti . 
113 
One of Baldus' arguments in the context of Frederick's constitutions is worth 
mentioning here since it confirms the interpretation of the imperial ban offered thus 
far, and, in particular, demonstrates that the medieval jurists themselves used the 
analytic distinction between its coercive procedural and permanent punitive elements 
(granted that they did not use those terms). In his commentary to Item nu/la, Baldus 
describes the imperial ban mentioned there as 'interlocutory' (interlocutoria), i.e. 
provisional, rather than 'definitive' (diffinitiva), i.e. permanent, since it may be 
cancelled for a due cause: the return of the defendant to obedience and his payment 
of the monetary penalty. 75 Baldus develops this distinction at Item quaecunque, 
where he argues that the ban may only be cancelled in this way if it is imposed for 
present and continuing contumacy, such as the persistent excommunication 
mentioned in Item quaecunque, and if it is in the power of the defendant to take 
himself out of this contumacious state by submitting to the authorities, which always 
remains a possibility for the excommunicate in canon law. In this specific case, the 
ban can be said to assume the provisional nature of the sentence of excommunication 
on which it depends. But if the ban is imposed for a limited period of contumacy in 
the past, then it is clearly no longer in the power of the defendant to end this 
contumacy by returning to the fold in the present, and he therefore cannot effect the 
cancellation of his ban. In other words, in this scenario, the defendant has missed the 
deadline for submitting to the authorities, and in respect of this prior period of 
contumacy he is now given a lasting punishment. Baldus compares this tQ the 
principles of annotation in Roman law, according to which, as we saw in chapter 2, 
the criminal defendant faces permanent sanctions (i.e. confiscation of his estate) if he 
does not appear at court after a one-year deadline.
76 
et istud bannum, de quo hie Ioquitur, est poena mille marcharum, secundum Od. in dicto 
commento. De isto banno imperiali habetur etiam [auth. Gazaros, post C.1.5(8).10].' 
75 Baldus to auth. Item nulla, post C.1.3(6).2, fol. 35rb: 'Ultimo, notatur hie quando bannum 
relaxatur ex causa. ergo patet quod bannum est interlocutoria quaedam, quia diffinitiva non 
posset relaxari.' 
76 Baldus to auth. Item quaecunque, post C.1.3(6).13, 38vb-9ra: 'Ultimo, notatur quod de 
banno potest quis extrahi, praemissa tamen satisfactione et habita concordia ab offenso. 
Nunquid autem indistincte sit verum? Respondeo, quandoque bannum imponitur alicui 
ratione contumaciae quae semper est in fieri, ut quia contumaciter perseverat in 
excommunicatione, tune bannum potest dimitti ut hie; assumit enim naturam 
excommunicationis cui cohaeret, in quo nunquam clauditur gremium redeunti. Si autem 
bannum respicit contumaciam de praeterito, et non perseverat in contumacia de futuro, tune 
non relaxatur bannum, postquam secum trahit suum effectum, ut [D.48.3.3/6] et 
[D.48.17.1].' Cf. eh. 2, p. 53 . 
114 
ii. The Libri feudorum 
Besides Frederick II's coronation constitutions, the imperial ban is explicitly 
mentioned only in one other part of the medieval Corpus iuris: in the books on 
feudal law, the Librifeudorum, appended as a tenth collation of the Authenticum in 
the Volumen parvum. The relevant passage appears in book 2, section 28 (Hie finitur 
!ex. Deinde consuetudines regni incipiunt. ), c. Domino guerram: 
But Obertus and Gerardus use this argument: just as a vassal is not bound to aid or 
~rovide se~ices to a !or? who has been excommunicated or banished by the king -
mde~d, bem~ a?solved m the meantime from his oath of fealty until his lord has 
rec~1ved restitution from the church or the king - in the same way, a vassal is not 
obligated to a lord who is unjustly waging war against another. 77 
It seems that in the feudal literature of the northern Italian jurists there never really 
developed a tradition of exegesis concerning the imperial ban mentioned in this 
passage - imperial, because the king under discussion was the emperor, who was 
also the king of the regnum Italicum (northern Italy). Indeed, the most important 
summae feudorum in circulation during the thirteenth century - by Jacobus de 
Ardizione ( d. ea. 1240), 78 Accursius/Jacobus Columbi (floruit ea. 1199-1244) 79 and 
w ' 
Odofredus - merely restated or paraphrased the argument of c. Domino guerram. 
77 
From the critical edition of Lehmann, Das Langobardische Lehnrecht, p. 158: 'Sed 
Obertus ~t Gerardus eo utuntur argumento, quod, quemadmodum dominum 
exc~~mum_catum vel a r~ge bannitum non est obligatus vasallus ad adjuvandum vel 
serv1tmm e1 praestandum, 1mmo solutus est interim sacramento fidelitatis nisi ab ecclesia 
vel a rege fuerit restitutus, ita nee istum injuste guerram alicui facientem.' ' 78 
Ja~obus de Ardizi_one, Summa feudorum, fol. 13rb-va: ' ... sicut non cogitur adiuvare 
dommum ex~ommumcatum vel bannitum vel hereticum vel eum dominum a quo vasallus a 
Sacramento sit absolutus ... Verumtamen, domino excommunicato vel bannito vel heretico ad 
fidem vel satis_factionem veniente, recuperatur pristinus ius, ut inde de medio tempore non 
conqueratur. .. ' 
79 
A~cursius?acobus ~olumbi, Summa feudorum, p. 189b: 'Et hec vera sunt quod tenetur 
dom1_num adm_vare. s1c~t supra dicitur, nisi dominus sit excommunicatus, vel a Rege 
~anmtus, q~omam mtenm vasallus est a sacramento fidelitatis absolutus, ut infra predicto 
tltulo et cap1te (Feud. II, 28).' The authorship of this summa, edited by Palmerius in BIMA II 
~nd errone~~sly att_ri?uted by him to Hugolinus (d. ea. 1233), is not entirely clear. For some 
time, preva1hng ~p~mon assigned it to Jacobus Columbi, who was thought to have revised an 
older work by Ptlhus (d. ea. 1213). Most recently, Weimar has proposed Accursius as the 
f~al author. Cf. Lange, Die Glossatoren, pp. 285-6 and 343. 
Odofredus, Summa in usus feudorum, fol. 71 v: 'Item, non tenetur vasallus adiuvare 
dominum vel ei servitium praebere, si dominus excommunicatus vel bannitus, imo in totum 
est absolutus sacramento fidelitatis, do nee absolvetur vel restituetur ... ' 
115 
The earliest exception to this trend appears to be the Ord
inary Gloss to the Libri 
Jeudorum, 81 which discusses at somewhat greater length 
one issue arising from 
imperial banishment: namely, whether the vassal, absolved
 from his oath of fealty, 
can sell the fief without his lord's consent such that the lor
d cannot revoke this sale 
even after he has received restitution from the king. The Gl
oss' preferred solution is 
that the vassal cannot alienate it permanently, granted that t
he lord has no remedy as 
long as he remains under the ban. 
82 The discussion raises a number of interesting 
points. Firstly, the author mentions that it is possible t
o damage the banished 
person's goods with impunity. 
83 Secondly, the lord's ownership (dominium 
directum) of the fief is not automatically lost by the ban; this penalty wou
ld have to 
be expressly stated in the sentence. 
84 But even if the lord lost his goods, the author 
concludes, the vassal could not alienate the fief with las
ting effect. It should be 
emphasised that it is the figure of the deportatus who 
provides the operative 
analogies for this argument. For just as the deported person retrieves his estat
e upon 
restitutio in integrum, so too the fief, if confiscated by the f
isc, would be returned to 
the imperial bannitus after he has received restitution from
 the emperor. Besides, 
according to Justinian' s novel Bona damnatorum, deported p
ersons' goods no longer 
pass to the fisc but devolve onto their relatives. Thus, if the
 bannitus has sons, they 
would acquire his rights over the fief during his time of ba
nishment, and the vassal 
would have to seek their consent for a sale. 
85 
81 It remains debatable to what extent Accursius was the real a
uthor of the Ordinary Gloss to 
the feudal law. While it seems clear that the author heavily
 depended on an older apparatus 
to the Libri feudorum by Pillius, opinions part over the varying degrees to wh
ich Accursius, 
Jacobus Columbi, and an anonymous contributed to the 
reworking and expansion of the 
older material into what eventually became known as the O
rdinary Gloss around the middle 
of the thirteenth century. Cf. Lange, Die Glossatoren, pp. 
283-5 and 348-9. For ease of 
reference, I will refer to the author as Accursius. 
82 Accursius to LF.2.28 c. Domino guerram, col. 62: 'Domino.
 Solutus est. Nunquid vasallus 
interim poterit vendere sine consensu ipsius domini sui ba
nniti, ita quad teneat venditio et 
quad dominus non possit earn vendicationem rescindere
 postquam fuerit restitutus? Et 
videtur quad sic ... Sed dicas, et forte melius, quad non pos
sit interim vendere, licet interim 
non possit venditio retractari.' 
83 Ibid.: ' ... nam et interim, dum est bannitus, posset damnum im
pune in bonis eius dari .. . ' 
84 Ibid.: 'Nee enim dominus amisit dominium directum, quad
 retinuerat tempore investiture, 
propter bannum, nisi forte bona eius essent expressim public
ata .. .' 
85 Ibid.: ' .. . cum etiam tune forte, eo restituto, et bona ei
us intelligerentur restituta, ut 
[C.9.51.1] et [C.9.51.13,3]. et maxime si dominus habet filios, in
terim feudum ad eos 
pertinet, cum hodie bona deportati non publicentur, ut 
[auth. Bona damnatorum, post 
C.9.49.10].' 
116 
It was not until some decades later, in Jacobus de Belvis
o's Lectura on the Libri 
feudorum ( composed some time before 1309), 
86 that it is possible to observe a real 
advance in the civilian tradition on imperial banishment.
 87 The Jack of doctrinal 
development in the interim is indicated by the fact that, besi
des the text of c. Domino 
guerram itself, it is chiefly the Ordinary Gloss that provide
s the points of departure 
for Jacobus' commentary; no other civilians or canonists are
 cited. 
At the opening of his discussion, Jacobus affirms the dictum
 of the law that a vassal 
is not bound to obey and come to the aid of a lord excommu
nicated by the church or 
banished by the king or emperor ('a rege vel ab imperatore') until th
e lord has 
rec~ived restitution. 
88 Furthermore, although the vassal can lawfully disobey h
is 
banished lord, he may not harm him through deliberate act
s of aggression. Granted 
that banniti can generally be killed with impunity by anyone,
 a vassal cannot kill his 
banished lord. Jacobus adduces the analogy of the freed
man, who has a special 
obligation to his patron not shared by others. For this reason
, the freedman, in respect 
of his patron, is excluded from the rule of Roman law t
hat allows a husband in 
certain cases to kill another man caught in jlagrante committing adultery
 with his 
wife.89 
Jacobus _then discusses a number of interrelated questions a
rising from the Ordinary 
Gloss. Firstly, can the vassal validly sell the fief without his 
lord's consent during the 
latter's time of banishment? While Jacobus fully accepts th
e basic arguments of the 
Gloss, he refines and develops them. Initially, Jacobus emp
hasises that the royal ban 
does not as a matter of course result in the confiscation of g
oods by citing a number 
of passages from the Digest and Code that describe th
e ordinary Roman Jaw 
procedure regarding contumacious defendants in capital tri
als. The principle to 
which he is referring is that within the one year period fo
llowing the initiation of 
proceedings, the defendant suffers no permanent detrime
nt while the authorities 
search for him: rather than being confiscated, his property
 is registered and placed 
:~ Lange and Kriechba~~· Die Kommentatoren, p. 618. 
The 1511 Lyon ed1t1on of Jacobus' Lectura is reprinted
 in OIR 12 under the title 
fsommentarii in au~henticum et consuetudines feudorum. 
Jacobus de Belv1so to LF.2.28, c. Domino guerram, fol. 9
7rb: ' ... secundum Obertum et 
Gera~dum, et utuntur eo argumento, quad, si dominus 
est excommunicatus vel a rege 
banmtus, n?n tene~r vasall~s eum adiuvare, nisi fuerit res
titutus ... Nota quod vasallus non 
de bet ?bed1re d?mmo ban~1to a rege vel ab imperatore .. .'
 The same basic position can be 
found m Baldus more detailed commentary to c. Domino gu
erram. 
89 
J~cobus de B~lviso to ibid.: 'Queritur quid si dominus sit bannitus, an vas
allus poterit eum 
occ1dere, cum hceat bannitum occidere. respondeo quad non
, arg. [D.48.5.39(38),9]. .. ' 
117 
into custody. Here we can see Jacobus' desire to shape the contemporary ban as 
much as possible in the mould of the old Roman law procedure of annotation. 
90 By 
default, he goes on, the ban (and excommunication) do not take away the vassal's 
obligation to the lord; they only suspend 'the effect of the obligation', and, therefore, 
when the lord is restored to his position, his vassal is bound to obey him.
91 All this 
applies unless the ban is given on the basis of medieval legislation that specifically 
includes a provision for the confiscation of goods ('legem specialem'), or the ius 
commune is used throughout the proceedings, since the latter stipulates that the 
property of an annotated defendant is confiscated after one year. In these cases, the 
lord loses his dominium over his fief, as well as the feudal obligation owed by the 
vassal, since both are transferred to the fisc. 92 
Next, Jacobus notes that the Gloss touches on the issue of restitution and once again 
agrees with its author, namely, that confiscated goods, including any fiefs, are 
returned to an imperial bannitus who has received restitution. He bases his 
conclusion on the argument of c. Domino guerram - i.e. the idea that the vassal is 
once again obligated to his lord following the latter's restitution - and on two 
passages from the Code that insist on the comprehensiveness of restitutio in integrum 
in respect of formerly deported persons. Jacobus, however, makes one important 
qualification: 'This I consider true by virtue of the said laws only if the things under 
consideration remain in the same state and in the hands of him to whom. they 
originally came at the time of the ban and on account of the ban, and if the things 
have not been alienated and have not been given as another fief. '
93 
90 Ibid.: 'Queritur in glossa quod, si dominus est a rege bannitus, an vasallus poterit interim 
vendere feudum sine eius consensu. Arguit primo glossa quod sic, triplici ratione faciliter, 
quod die ut in glossa. respondet glossa et bene quod non, quia non perdit dominus directum 
dominium feudi propter bannum. pro glossa facit, quia nihil allegat, [D.46.3.41], et 
[D.48.3.6,1 ], et [D.48.1 7.1].' 
91 Ibid.: 'Nam excommunicatio vel bannum non tollit obligationern qua vasallus est obligatus 
domino, sed tamen effectum obligationis. et ideo domino absoluto et restituto statim tenetur 
ei obedire, ut hie ... ' Baldus, to c. Domino guerram, affirms this argument, which he 
(erroneously) attributes to the Gloss. 
92 Jacobus de Belviso to ibid.: ' ... nisi bona sint publicata per legem specialem vel bona 
annotata de iure communi anni circulo elapso, ut [C.9.40.1/2], et [D.48.17.1/5]. tune perditur 
dominium et tollitur obligatio inter eos et transeunt in fiscum, ut [D.49.14.1]. .. ' 
93 Ibid.: 'Tangit glossa incidenter questionem aliam. Quid si bona domini fuerint publicata et 
postea restitutus est, an feuda vasallorum et omnia bona intelliguntur sibi restituta esse. hie 
est argument quod sic. [C.9.51.1/1 3]. quod puto verum per dicta iura manentibus rebus in 
eodem statu apud eum ad quern originaliter tempore banni et propter bannum pervenerunt, 
non alienatis rebus nee translatis in alium feudum.' 
118 
Jacobus explains this qualification in a lengthy quaestio disputata, which forms the 
remaining and greatest part of his COffi\Tientary to c. Domino guerram.94 He reports 
that although he debated this quaestio at Bologna, he observed a real case of the 
same type when he stayed in Naples at the court of King Charles II (d. 1309).95 The 
facts are the following. The king gave a castle to one of his counts as a fief. The 
count committed treason due to which he was 'banished or condemned'. Later, the 
king gave the same castle as a fief to a knight. The king then granted restitutio in 
integrum to the count with a specific reference to his lost goods. The question is 
whether the count can now sue the knight for the castle.96 The majority of the legal 
detail pro and contra need not trouble us. It is an extensive exposition on the 
mechanism of restitutio in integrum that does not tell us anything about the legal 
status of the banished person, his citizenship, or rights (whether civil or natural). One 
aspect, however, is worth picking out: while Jacobus adduces a great number of 
diverse (and often far-fetched) analogies in his discussion, there is a tendency for any 
direct mention of restitutio in Roman law to refer to the case of deportation; I can 
count six separate references of this kind. 97 
Jacobus' solution to the original question relies on a number of distinctions and sub-
distinctions. I will confine myself to a brief summary of his position. The first 
distinction is this: either the count is (A) banished ('bannitus') or he is (B) 
condemned ('condemnatus'). If he is (A) banished, then he is either: (i) banished and 
94 
Extant in Rome, Biblioteca Apostolica Vaticana, MS. Archivio S. Pietro A. 29, fol. 222ra-
vb (anonymous), and MS. Chigi, E.VIII. 245, fol. 268rb-vb (Jacobus de Belviso) - incipit: 
Dominus rex dedit comiti castrum in feudum - and Florence, Biblioteca Nazionale 11.1.344, 
fols 59rb-60rb (Jacobus de Belviso) - incipit: Rex dedit comiti in feudum castrum. Cf. 
Bellomo, Quaestiones, p. 158 (no. 309) and p. 402 (no. 451); Martino, 'Iacopo Belvisi', p. 
65 (no. 18). The solution of this quaestio is summarised by Baldus in his commentary to c. 
Domino guerram. 
95 
On Jacobus'_sojoums in Naples, see Martino, 'lacopo Belvisi', esp. pp. 68-71, and Lange 
and Kriechbaum, Die Kommentatoren, p. 614. 
96 
Jacobus de Belviso to LF.2.28 c. Domino guerram, fol. 97rb: 'Sed queritur, et bane 
quaestionem disputavi in civitate Bononie ego, Jacobus de Belviso, quam vidi de facto in 
curia dominis regis Karoli dum essem in regno Sicilie. Et est questio talis. dominus rex dedit 
castrum comiti in feudum. comes commisit proditionem, propter quam est a rege bannitus 
vel damnatus. rex postea concessit idem castrum militi in feudum. rex deinde restituit in 
integrum comitem ad gratiam et ad bona. Queritur nunquid comes possit a milite castrum 
advocare et quo iure pro comite.' · 
97 
To pick one good example, ibid.: 'Tertio, quia, licet mutato statu persone ius et dominium 
competens ei amissum sit, alii quesitum, tamen, statu in integrum restituto, totum restituitur 
in ~ri~tinum sta~m, ut patet in patria potestate dissoluta patris deportatione, que patri 
restltmtur eo reshtuto, ut [C.9.51.13]. idem in iure patronatus, [D.37.14.21].' 
119 
restored according to customary law ('ius consuetudinarium'); (ii) banished and 
annotated according to the ius commune; (iii) banished and his goods are confiscated 
according to a/the law of the kingdom ('legem regni'); (iv) banished simply speaking 
(' simpliciter ' ).98 In the first and second cases (i & ii), he loses his fief after one year 
has elapsed. These cases proceed according to the model of Roman law annotation. 
The customary law under consideration might refer to Frederick II's legislation for 
Sicily, which emulated this procedure to some degree: thus, the Liber Augusta/is 
allows the (for)bannitus one year to appear at court before his ban is finalised, that is, 
before he is considered to beforjudicatus and his goods are confiscated.99 In the third 
case (iii), the count loses his fief immediately. This seems to be a repetition of 
Jacobus' earlier argument that a ban involves confiscation if it is issued on the basis 
of a !ex specialis, i.e. one which specifically includes a provision for the confiscation 
of goods. 100 In the fourth case (iv), a so-called ' simple' ban will not have that effect. 
This again reflects Jacobus' earlier argument (and Accursius' position) that 
banishment does not automatically include property confiscation (in Accursius' 
view, not unless this has been expressly stated in the sentence). Jacobus adds that a 
ban given simpliciter has a preliminary nature and, lacking an examination of the 
matter and a determination of guilt, takes the form of annotation. If we read this in 
the context of the Liber Augusta/is, this simple ban might refer to that initial year in 
which the defendant was a (for)bannitus but did not suffer confiscation.
101 This 
would also explain why, if (B) the king formally condemned the count, presumably 
in his absence, the latter 'undoubtedly' and 'legitimately' lost his fief. As we saw 
above, in Sicily, after one year had elapsed, the bannitus was declaredforjudicatus, 
that is, treated as if he had confessed and had been condemned by a definitive 
sentence. 102 In those cases in which the count does not lose his fief, Jacobus 
concludes, he can of course reclaim it from the knight by means of an action. In 
those cases, however, in which the count does lose his fief, restitutio in integrum will 
not return it to him. As Jacobus stated at the outset, restitutio only restores what is 
acquired by someone else immediately at the time of and due to the sentence of 
98 Ibid., fol. 97vb: 'Solutio: aut comes est bannitus aut condemnatus. si bannitus, aut est 
bannitus et restitutus secundum ius consuetudinarium ut vasallus, aut bannitus et annotatus 
secundum ius commune, aut bannitus et bona publicata secundum legem regni, aut est 
simpliciter bannitus.' 
99 Ibid.: ' Primo casu et Secundo, amittit feudum anni circulo revoluto ... ut [D.48.17.1/2/5].' 
For Frederick's Sicilian legislation, see pp. 111-12 above. 
100 Jacobus de Belviso to LF.2.28 c. Domino guerram, fol. 97vb: ' Terto etiam casu perdit 
feudum et alia bona, ut [C.9.49/0.48.20].' 
101 Ibid. : 'Quarto casu [non] perdit feudum nee bona alia, quia feudum regulariter non perdit 
nisi per sententiam, culpa cognita et convicta, ut [D.48.3.6, I]. :. 
102 Ibid.: 'Si est condemnatus, legitime indubitanter perdit feudum .. . ' 
120 
con~scatio~, and remains in that same state. But if it is acquired by someone 
mediately, 1.e. through the king, who at the time of the sentence became the real 
owner of the confiscated property, it is not restored.103 
We should draw two general conclusions from the pre-Bartolist discourse on 
im~~rial banishment. Firstly, despite the fact that banishment could be eminently 
political and exceptional in nature - the case related by Jacobus de Belviso was one 
?f high-level treason (proditio) against the king - there is a clear tendency to 
mterpret the ban restrictively and to limit its impact where possible. On the whole 
the jurists achieve this by comparing banishment to the ordinary institutions of th~ 
Roman criminal law. There is no mention of the exceptional forms procedure in the I. 
!missione, of t~e crimen laesae maiestatis, perduellio, the hostis, or the transfuga; 
mstead, annotat10n and deportation are recurring analogies. As we shall see in the 
next chapter, a similar pattern can also be observed regarding the communal ban: 
before Bartolus, one cannot find a jurist who establishes a fundamental analogical 
link between bannitus and hostis. 
104 Secondly, the jurists do not provide an 
overarc~ing . conception with which to analyse imperial banishment. Although 
d~portatlon 1s clearly an important point of reference, the bannitus is certainly not 
viewed as a modem-day deportatus; there is no direct or straightforward comparison 
between the two figures. In other words, the legal position of the bannitus is not 
established with reference to some underlying status, but through the use of diverse 
analogies - such as ?eportation and annotation - and rules taken from medieval legal 
sources, the most important of which were of course Frederick II' s coronation 
constitutions and c. Domino guerram. What we will see below is that Bartolus' 
co~tribu~ion did not lie in linking the bannitus to the hostis in the looser way in 
which his predecessors compared the bannitus to the deportatus. He went much 
furt~er than that: for Bartolus, in a very meaningful sense, the bannitus simply is the 
hostzs, such that the legal position of the former can be derived almost completely 
from that of the latter. 
103 lb.d 'S d .b · 
t .: e qu1. ~s cas1bus comes amisit feudum, non puto communiter restitutum esse ad 
ca_s~rum contra m1htem. sed quo _cas_u ~on_ amisit feudum, tune suo iure utili in rem petit a 
?11hte castru~·· ·. Non obstat restttutto m mtegrum indulta comiti a rege, quia restitutio in 
mte~run:1 r~stttu1t omne quod amittitur et alii immediate, idest tempore sententie sive 
pubhcattoms bonorum vel amissionis feudorum, non [sic] est quesitum et in eodem statu 
ma~et._.. sed quod alii mediate, idest per regem, qui vere est dominus, acquiritur et non 
restttuttur ... ' 
104 Ch. 4, pp. 200-12 and 225-64. 
121 
I 
11 
I 
I 
iii. Thoughts on the Relationship between Theory and Practice 
German kings and emperors routinely employed the terms hostis and c
rimen laesae 
maiestatis ( or variations thereof) in their sentences of banishment. The use of these 
terms in relation to the royal ban seems to have arisen at the same ti
me as the ban 
itself: in Germany, around the tum of the millennium; in Italy, afte
r the ban had 
migrated into that country about a century later.
105 As we have seen, these are 
technical terms in Roman law. The extent of romanising language 
in the sources, 
especially in the literary ones, is remarkable. Already in the elevent
h century, the 
German Annales Altahenses described Henry IV's banishment of the duk
e of Bavaria 
in the following terms: in 1070, the king ('Caesar' ) convened a 'senate' of his Saxon 
magnates and had them pass a senatusconsultum to declare the duke a h
ostis. 106 This 
demonstrates, at least on the part of the author, a working knowledg
e of the hostis 
declaration in the politics of the classical empire. Given this awaren
ess of Roman 
concepts amongst learned circles as well as in the imperial chance
ry, should we 
therefore assume that Roman law had a hand in the creation and deve
lopment of the 
ban? 
For a long time, when most historians viewed the ban and its atten
dant outlawry 
(Friedlosigkeit) as ur-Germanic institutions, the answer to that question was no. 
Siuts, who convincingly refuted that older narrative, argues that some
 knowledge of 
Roman legal institutions and ideas regarding exile survived throug
hout the early 
Middle Ages - terms such as proscriptio, relegatio, and deportatio, amo
ngst others, 
can be found in the legal sources of that time - and that these coales
ced with other 
traditions - ecclesiastical excommunication as well as indigenou
s 'Germanic ' 
precursors - to form the type of outlawry in its archetypal late-mediev
al form, which 
fu st emerged in Germany during the central Middle Ages (that is, not before the 
tenth century). As Siuts himself admits, however, much remains unclear about the 
105 Ficker, Forschungen I, pp. 81-2, 166-8 and 170-1 ; cf. Poetsch, Die Reic
hsacht, pp. 11-15. 
106 Annales Altahenses, quoted in Franklin, Reichshofgericht, p. 33 (n. 1): 'Caesar - subito 
frequenti senatu a scriba legem Saliam, qua capitale est in principem
 coniurare, recitari iubet 
et universos confestim iureiurandi, quo cuncti pro salute reipublicae
 atque principis adstricti 
erant admonet sententiam rogat. Omnes in sententiam legis ped
ibus (ut ajunt) eunt: fit 
sena~sconsultum, Otto hostis iudicatis, proscribitur, bona eiu
s confiscantur, praedia 
publicantur, Boiaria caesari adiucatur.' The rather puzzling refere
nce (presumably) to the 
Lex Salica should not distract from the romanising language of this pas
sage. Other examples 
from the eleventh and early twelfth century in the German literary 
sources can be found in 
Franklin, Reichshofgericht, pp. 27, 33-4, and 45-6. 
122 
development and ongms of the ban. 
107 Nevertheless, his theory is not an 
unreasonable one. Given that a general, if largely unscientifi c, famil
iarity with the 
Roman law - including Justinian 's Institutes, Codex, and Novels (in the form of the 
Epitome Juliani) - endured throughout the early Middle Ages, it is certainly possible 
that the notions of the hostis declaration and Iese majesty were transmitted through 
this channel. 108 
Although unaware of Siuts ' work ( despite being published almost two decades 
before his own monograph), this was precisely the argument of Cavalca. Since the 
latter does not engage with Siuts, be accepts that part of the traditi
onal narrative 
according to which an archetypal Germanic Friedlosigkeit existed thr
oughout the 
early Middle Ages. On the other hand, Cavalca argues against 
the traditional 
conception in that he stresses (more so than Siuts) the survival, even if it was partial, 
of the Roman legal tradition on exile in the early medieval period, a
nd claims that 
this was as an important constitutive element, alongside Germanic Fri
edlosigkeit, in 
~he formation of the ban in its late-medieval form.
109 The main basis for his argument 
Is the undoubtedly romanistic influence in the Italian sources after ea
. 1100, 110 and 
some (though not much) evidence that the concepts of maiestas and hostis can also 
be found in sources before the tenth century. 111 
The_ ~istoriographical emphasis on the continuity or otherwise of th
e Roman legal 
tr~d1t10n should not, however, obscure the potential relevance of extra
-legal sources. 
It Is equally reasonable to imagine that the classical concepts of the hos
tis and /aesa 
maiestas regained currency in the empire during the eleventh an
d the twelfth 
centuries through a reading of Roman literary material. Perhaps the
 author of the 
Annales Altahenses mentioned above drew inspiration from Suetoni
us' De Vita 
101 s· 
mts, Bann -und Acht esp. pp. 14-79 
108 · ' . . 
A go~d ove1:1ew of the history of Roman law from the early Middle
 Ages to the eleventh 
century, 1s provided by Lange, Die Glossatoren eh I pp 1-34 
109 
' . ' . . 
Cavalca, // bando, eh. I, pp. 17-100. 
11 0 
A point emphatically made in ibid pp 65-75 Il l . ., . · 
Cavalc~ provides a numb_e~ of convincing examples regarding Iese 
majesty, mostly from 
the barbarian law_ codes - 1b1d., pp. 75-7. However, he can only
 point to two concrete 
examples, con~ernmg th~ hosti~. [I.] A si~g_Ie late-Carolingian capitulary which employs th
e 
phrases . pubhcus regm hostJs et dom1m sui imperatoris inimi
cus' (p. 69). [2 .J The 
declarat10n of the duke of Benevento as a hostis publicus by the Rom
an senate in 872 (p 
77). Ca~al~a also pro~ides some evidence for knowledge of the distinction r
elegatio~ 
deportatzo m early-medieval and pre-Imerian legal sources (pp. 56 and 79). 
123 
'I 
Caesarum, a work widely read in the Middle Ages and which recounts the pertinent 
periods of Roman history in a sufficient amount of detail. 
112 
While it is therefore probable that Roman legal and political ideas played some part, 
through one channel or another, in the development of the ban as a form of outlawry 
during the central Middle Ages, it seems relatively clear that the concepts of the 
hostis declaration and Iese majesty did not enter imperial legal practice, at least not at 
first, from the academic civilian tradition. The dates alone preclude this: Roman 
terminology in relation to banishment can be found in the German sources decades 
before the revival of Roman law gained pace in Bologna during the second half of 
the eleventh century. Furthermore, the terminology of the imperial chancery, even in 
Italy, where one could reasonably expect the influence of the law schools, reveals a 
lack of technical usage alien to learned jurisprudence. Thus, inimicus and rebellis are 
used as synonyms for hostis, and proscriptio describes the institution of outlawry as 
a synonym for bannum, even though inimicus, rebel/is, and proscriptio do not have 
these meanings in Roman law.
113 Proscriptio brings to mind the triumviral and 
Sullan proscriptions of the late republican period, when this word came to be 
associated with a type of outlawry whose effects resembled that of hostis status. As 
Mommsen has pointed out, however, Sullan proscription did not become a part of 
Roman criminal law, nor did the word proscriptio, as a technical legal term, ever 
acquire the sense of outlawry; it certainly cannot be found with that meaning in the 
texts of Justinian's compilation. 
114 In fact, the use of proscriptio suggests the 
influence of classical literary texts, especially Roman histories, on this aspect of 
medieval legal culture. 
Further evidence that the entry of the concept of the hostis into imperial legal 
practice did not depend on the revival of Roman law at the universities, is the 
striking fact that this figure, as we have just seen, played no role in the civilian 
discourse on imperial banishment before Bartolus, that is, before the middle of the 
112 E.g. Suetonius, The Lives of the Caesars, vol. 1, Caligula 7.1 , p. 412 , ('Neronem et 
Drusum senatus Tiberio criminante hastes iudicavit. ') and vol. 2, Nero 49.2, p. 176 ('Inter 
moras perlatos a cursore Phaonti codicillos praeripuit legitque se hostem a senatu
 
iudicatum .. . '). 
113 Ficker, Forschungen I, pp. 81-2, 166-8 and 170-1; cf. Poetsch, Die Reichsacht, pp. 11-15. 
114 Mommsen, Romisches Strafrecht, p. 938: 'Eine technische Bezeichnung der Aechtung 
findet sich im ordentlichen Strafrecht der Romer nicht. Proscribere ist... an sich indifferent 
und fiir jede offentliche Bekanntrnachung verwendbar. Die proscriptio hominis ... ist <lurch 
Sulla zum Schreckenswort geworden, aber nicht zum Rechtswort. In der Rechtssprache
 
bezeichnet proscriptio ohne Beisatz regelmassig die proscriptio bonorum, den Concurs, auch 
die Verrnogensconfiscation, nicht die Aechtung.' 
124 
fourteenth century. As Roffredus' discussion of the Sicilian hostis publicus reveals 
the jurists must have been aware of contemporary terminology and practice. Indeed: 
it would have been very remarkable had that not been the case. Yet despite the 
terminological and conceptual overlap, for a long time they did not establish a 
connection between the medieval imperial hostis and the hostis of their classical 
authorities. As has been mentioned previously, chapter 4 will demonstrate that this 
basic chronology is reflected in the development of the discourse on communal 
b~nishment. Although the communal ban, unlike its imperial equivalent, was widely 
discussed by the jurists already from around the middle of the thirteenth century, it 
was only with Bartolus that the hostis became the central analogy for the communal 
bannitus. 
115 (That fact seems less stark in the municipal context since the cities as 
indicated chapter 1, did not deploy the term hostis to describe the bannitus; they u'sed 
rebel/is, inimicus, and proditor instead.) 116 It appears, therefore, that just as civilian 
theory did not export the hostis into imperial legal practice, so it did not cause the 
development of banishment as outlawry in the statutes of the Italian cities where 
. 
, 
withdrawal of legal protection can already be found towards the beginning of the 
thirteenth century. As Ficker argues, it is likely that the appearance of this type of 
ban can be put down to the influence of imperial practice, which had brought it to 
Italy about a century earlier. 117 
The argument that civilian scholarship was not responsible for the creation of the 
institution of banishment is supported by the civilians' view of the bannitus as 
something inherently alien to Roman law.
118 This is why the jurists commonly 
referred to those banished from the Italian cities as 'the banished persons of our time' 
(banniti nostri temporis), and why Albericus de Rosate, for example, introduces the 
· ban in his Quaestiones statutorum by explaining that this legal institution is not 
mentioned 'in the old laws' but only in new medieval ones, such as the legislation of 
Frederick II and the Libri feudorum. 119 This view of the ban mainly springs from the 
unease that the jurists felt about its procedural aspects. As we shall see in chapter 4, 
115 h C . 4, pp. 200-12 and 225-64. 
116 h C. l,p.21. 
117 s 'b'd ee 1 1 ., p. 18. 
i1s A h. . h 
s 1stonans ave generally acknowledged: cf. Ghisalberti, 'La condanna', p. 31; 
Cavalca, II bando, pp. 11 and 83; Canning, Baldus de Ubaldis, p. 128; Mooney, 'The Legal 
Ban' , p. 88. 
119 
Alber_icus, de Ro~a:e, Qua~stiones statutorum IV, q. 1, fol. 66ra (quoting Nicolaus de 
Ma_tarelhs):. ~t ~e 1stis ~anms non fit mentio nee nominatur in antiquis legibus ... Sed in 
leg1bus nov1ss1m1s Federici secundi imperatoris de isto banno fit mentio .. . et communiter
 
habetur in fine libri usus feudorum ... ' 
125 
the point that emerges very clearly in their discussions of the communal ban is that, 
in procedural terms, it constituted a deviation from the norms of good jurisprudence. 
The penalties imposed on the disobedient defendant by the bannum pro contumacia 
differed from, and far exceeded, those of adnotatio bonorum, the ordinary Roman-
law procedure for contumacy in capital cases. The ban also fell foul of the major 
precept that an absent defendant cannot be sentenced to a capital penalty, which 
banishment certainly was, not least because outlawry effectively amounted to a death 
sentence. Hence, the jurists viewed banishment as a statutory derogation of the ius 
commune. It is notable that this understanding remained in place after Bartolus had 
generally aligned the bannitus with the hostis. 1
20 Even if the bannitus and the hostis 
could be fruitfully compared in terms of substantive legal status (i.e. both are public 
enemies liable to be killed with impunity), the hostis declaration did not serve as a 
general theoretical model for the procedure of banning in the municipal context.
121 
All this is not to say that the academic revival of Roman law did not have any impact 
on the development of the ban after its establishment in legal practice. Quite the 
contrary: the civilians, as legal practitioners and advisers at the imperial court and in 
the Italian cities, would have tried as far as possible to shape the medieval institution 
of banishment, as they saw it, according to the sounder Roman principles of the ius 
commune. Already in the twelfth century, the German chronicler Rahewin, reports an 
episode in the Gesta Friderici for the year 1158, during which certain legal experts 
advised Frederick I not to banish the city of Milan before he had formally summoned 
its representatives in the manner set out by the Roman law.
122 Frederick II's Sicilian 
Constitutions are another good example, given that these, as we saw above, emulated 
the Roman law procedure of adnotatio bonorum. Regarding the Italian cities, 
Cavalca describes a number of the Romanistic influences on their statutory law 
concerning the ban, at least some of which must have resulted from the practical 
120 Baldus thus continues to describe the ban as a novel institution at C.6.24.1, fol. 70va: 
'Apparet igitur ex praedictis bannitos nostri temp~ris. non habere propriam et, totalem 
similitudinem nee cum deportatis nee cum excommumcatls, sed esse novum genus ... 
121 Cf. eh. 4, pp. 193-221. . . 
122 Ottonis et Rahewini gesta Friderici, p. 204: 'At sapientes et legum pent! persuadent, 
Mediolanenses, Iicet improbos et infames, iudicis tamen officio per legitti~as !nducias 
citandos esse ne violentia eis illata vel contra ius in absentes prolata sententia v1deretur. 
Legittimas ve~o inducias dicunt iudicis edictum unum, mox alterum et tertium, se~ unum_ pro 
omnibus, quod peremptorium nominatur. Quod et factum est. [ .. ._J 11?perator, ast1pulant1_bus 
iudicibus et primis de Italia, contra Mediolanenses condempnattoms proferens se~tentiar:i 
bostes eos iudicat omnique apparatu ad obsidionem civitatis accingitur.' The episode 1s 
recorded by Ficker, Forschungen, p. 186. 
126 
activities of the university jurists: from their role in the redaction, interpretation and 
judicial application of the statutes. 123 ' 
Despite the jurists' view of the bannitus as foreign to Roman law, the classical hostis 
did of course bear.a relatively close resemblance to the former. This is why Bartolus' 
establishment of a general equivalence between the two figures made sense and 
became civilian communis opinio thereafter. Credit here must be given to Cavalca, 
who was the first amongst modern historians to emphasise the fact that the Roman 
legal tradition possessed, in the form of the hostis, a parallel to medieval (and 
Germanic, as he put it) forms of outlawry, and that, as a result, Bartolus' argument 
was not founded on forced analogies but rather on a shrewd sense of legal reality. 124 
Cavalca's revisionism was in part aimed against Ghisalberti, whose ground-breaking 
study of the civilian theory of banishment (published in 1960) supported the then 
?ominant, exclusively Germanic interpretation of the ban by overemphasising the 
madequacy of the jurists' initial comparisons of banishment to relegation and 
deportation, and thus deducing from that a lack of affinity between medieval 
('Germanic') and Roman forms of exile. 125 
The question remains why no jurist before Bartolus considered using the hostis as a 
general analogy for the imperial bannitus. Although it is impossible to be certain on 
this_roint, one likely factor was the imperial ban's dual use as a coercive procedural 
device and as a permanent punishment. As a procedural device, the ban was a 
mechanism employed against contumacious defendants in various ordinary civil and 
cr~minal matters with the purpose of forcing the latter to submit to imperial authority; 
this would explain the jurists' comparisons to adnotatio bonorum, the equivalent 
mechanism in Roman criminal law. In contrast, the ipso iure acquisition and 
declar~tion of hostis status was an extraordinary, purely political measure designed 
to pumsh extreme cases of treason. Although, as we saw in chapter 1, this was also 
the function of the permanent imperial ban, the figure of the hostis therefore only 
truly corresp?nded to one part of the ban's range of application. 126 Another plausible 
123 f C . Cavalca, II bando pp. 65-75 
124 ' • 
Ibid., eh. l, pp. 17-100, esp. pp. 58-65 and 90-6. 125 Gh. lb . 'L ' 1sa ert1, a cond_anna , esp. pp. 6-10, 33, and 69. Mooney, 'The Legal Ban', pp. xi, 3, 
~8 ~n~ 177, agree,d w_it_h. Ghisalbe~tj in ~iewing the ban as a purely 'Germanic' legal 
mst1~ut1??· ,cavalca s cntic1~m of Gh1salbert1 followed that of Crifo, Ricerche, p. 36 (n. J 35), 
~nd Esil!o,, p. 721, who pomted to aquae et ignis interdictio as an earlier Roman parallel to 
. Ge:°1~mc forms of outlawry - see eh. 2, p. 41, n. 30, for further details on this legal 
mstltut10n. 
126 Ch. 1, p. 19. 
127 
reason for the civilians' initial reluctance to relate the imperial bannitus to the hostis 
is the latter's exceptionalness and the consequent lack of relevant material in the 
Corpus iuris. It was difficult to extract many legal ideas and principles from a few 
scattered passages. If, for example, there were questions concerning property 
confiscation or restitution, the hostis was of no help at all. In cases such as these, the 
jurists had no alternative but to look elsewhere, and as we saw in our discussion of 
Accursius and Jacobus de Belviso, they habitually turned to deportation. This was a 
sensible choice given that out of the two major forms of exile in the Corpus iuris -
relegation and deportation - it was the latter that most closely resembled the severity 
of imperial banishment. 
5. The Quarrel between Emperor Henry VII and King Robert of Naples 
Before we examine Bartol us' thought on imperial banishment, it is necessary to 
introduce the political controversy that provided the context for Bartol us' work. The 
controversy in question was stirred up by Emperor Henry VII's proceedings against 
King Robert of Naples at the opening of the fourteenth century. This episode 
constitutes a cause celebre in the modem literature chiefly for two reasons. The first 
concerns the nature of the medieval empire, particularly its universality, the 
relationship between the emperor and other princes, as well as the relationship 
between empire and papacy. The second is more arcane, concerning developments in 
procedural law, particularly surrounding the technicalities of judicial summons, but 
also the appearance of the idea that the possibility of defending oneself against legal 
prosecution, and by extension the right to be summoned to court, was a right rooted 
in natural law. 
The outline of the case is as follows. 127 Robert was king of Sicily, a kingdom that 
extended over the south of the Italian peninsula (but excluded the island of Sicily 
itself); 128 in practice, it was independent from the empire but a fief of the Roman 
church. Robert, however, also held the county of Provence as a fief from the empire. 
127 For a good general account of Henry's Romzug, see Bowsky, Henry VII in Italy . Better on 
the legal details of the quarrel between Henry and Robert is Pennington, The Prince and the 
Law, eh. 5, pp. 165-201. 
. . . . 
128 This confusing situation resulted from the split in 1302 of the old kmgdom of Sicily mto 
two parts: the island of Sicily (kingdom of Trinacria) and its peninsular portion (kingdom of 
Sicily). 
128 
After Henry VII's crossing of the Alps in 1310, he was in constant conflict with 
Robert over influence in northern Italy . right up to his death in 1313. In September 
1312, Henry began proceedings for high treason against Robert. Remaining 
contumacious in the face of imperial summons to appear before the emperor's court 
at Pisa, Robert was eventually banished in April 1313, declared guilty of the crimen 
laesae maiestatis, and a rebel/is and hostis imperii. In his sentence, Henry deprived 
Robert of all his lands and titles and mentioned the usual penalties associated with 
imperial banishment. The sentence was worded in a comprehensive manner but it 
did not specifically include or except the kingdom of Sicily. Although Henry' had in 
fact summoned Robert, the emperor also stated that this, strictly speaking, had not 
been necessary, a position made possible by the alleged notoriety of Robert's crimes 
and because the emperor, as the supreme civil authority, was not bound by the civil 
law (legibus solutus).129 In March 1314, Pope Clement V, being an ally of Robert 
officially annulled the sentence with his bull Pastora/is cura, claiming jurisdictionai 
superiority over the emperor due to papal plenitude of power. In this bull, Clement 
explained at length all the reasons why Henry's sentence was null. 
Two important legal documents emerged from the papal camp during this 
controversy that had a lasting impact on the civilian tradition on banishment. The 
first is the bull Pastora/is cura, which entered the body of canon law as a chapter in 
the
1
~lementi~es (2.11.2). The s_econd is Oldradus de Ponte's (d. ea. 1335) consilium 
43. According to Eduard Will, who has analysed the latter in detail, Clement's 
arguments were based on Oldradus' consilium as well as that of another unknown 
jurist; it is certainly the case that many of the principal ideas in cons ilium 43 can also 
b £ · d · p 131 e oun m astoralis cura. Composed in the context of Oldradus' service to the 
papal curia in A vignon, his cons ilium displays a predictably pro-papal and anti-
129 F H , or enry s sentence, see Schwalm (ed.), MGH, Leg. sec. IV, Tom. IV, Pars II, pp. 985-
90 (no. 946). 
130 E~i~ed by Schwalm in ibid., pp. 1373-8 (no. 1254); I use the version in Oldradus' 
Cons1lza et quaestiones (Rome, 1472) - NB. volume unfoliated. Albericus de Rosate to 
C.9.8.5, quotes this consilium at length following a transcription of Henry's sentence. Note 
tha~ a number o~ other pro-papal as well as pro-imperial consilia and tracts related to this 
topic have survived, although they did not enter the ius commune. These as well as 
~
1
ldra_dus: c?nsilium, are discussed by Pennington, The Prince and the Law, pp. '171-85. 
_Will, D1e Gutachten des Oldradus', esp. pp. 40-1 and 50. The anonymous consilium is 
e~1ted by Schwalm, MGH, Lee. s_ec. IV, Tom. IV, Pars II, pp. 1320Al (no. 1250). Will's 
view h~s been subsequently quest10ned by Meijers, Le droit romain, p. 190 ff. Pennington, 
Th: Prine~ and the Law, p. 179, agrees with Will, arguing that these two consilia were part 
of a dossier put together at the papal court to provide a juristic basis for the legislation that 
Clement planned.' 
129 
II 
imperial stance. 132 Since Pastora/is cura in particular served Bartolus as continual 
point of reference in his discussion of Robert's banishment, and since he also repeats 
some of the opinions expressed by Oldradus (though without ever citing the latter), 
we should examine these documents here. Given that the two texts are closely 
related, it makes sense to examine them in conjunction with each other; we will use 
Oldradus' cons ilium as a more detailed and academic explanation of the points raised 
by the papal bull. 
Clement and Oldradus put forward a number of arguments, not all of which are of 
equal relevance to our discussion. The one that has aroused the greatest interest 
amongst historians of political thought is the pope's claim that the kingdom of Sicily 
lay outside the jurisdiction of the Roman empire ('extra districtum imperii'). This 
idea, which caused a wave of contemporary polemic concerning the empire's 
universality or otherwise, 133 appears in the immediate context of the bull merely as 
an assumed premise allowing Clement to construct two arguments. The first was his 
argument for the invalidity of Henry's sentence due to errors in the legal 
technicalities surrounding Robert's summons. Since, according to Clement, Sicily 
was Robert's regular domicile as well as his region of origin, despite his feudal 
holding of Provence, the defendant (Robert) was located beyond the territory (extra 
territorium) of the summoning authority (the emperor).134 The doctors of law, as 
Oldradus noted in another consilium specifically devoted to this topic, varied in their 
opinions on whether, and how precisely, a judge could summon a defendant extra 
territorium. 135 In what seemed to have been a rather acerbic debate, some jurists 
argued that a judge is indeed able to summon a defendant in this way, whereas others 
argued that he is not. 136 Pastora/is cura was itself an intervention in this debate and, 
132 Cf. Lange and Kriechbaum, Die Kommentatoren, pp. 602-12, esp. pp. 607-9. 
133 Examples in Bowsky, Henry VII in Italy, p. 185 ff. 
134 Clem.2.1 1.2: 'Ut illud, tanquam notissimum, nullatenus omittamus, quod regem, extra 
districtum imperii, in regno scilicet Siciliae, notorie ac continue tempore supra dicta 
morantem ... Denique licet rex ipse terras aliquas ab imperatore tenere dicatur in feudum: non 
tamen in eis, sed in regno praefato domicilium suuam fovebat. [ ... ] Rursus non est silentio 
relinquendum, quod ipse rex noster et ecclesiae Romanae ratio~e . regn_i ~raedi~ti notorie 
subditus, homoque et vasallus, ac in eodem regno, non, ut praemittitur, m impeno, certum 
continuum et notorium suum habens domicilium, ac originem inde trahens ... ' On the 
concepts of origin and domicile, see eh. 1, pp. 8-10. Forum is di~cussed_ in eh. 2, ~- 49. . 
135 Oldradus de Ponte, Consilia et quaestiones, cons. 88: 'Tercio quentur de tah questlone. 
Iudex intra territorium existens mandavit executori quod citaret quendam suum subditum 
extra existentem; quod fecit. queritur an talis citatio arctet citatum. Et advertatis quia in hoe 
variant doctores.' 
136 Consider this report by Bartolus, to D.48.17 .1, fol. 189r: 'Cynus disputavit istam 
quaestionem Senis et <licit quod ista citatio, quae fit verbo vel per edicta, potest fieri de eo 
130 
unsurprisingly taking the side of the latter, declared that a judge could in no sense 
summon extra territorium - tellingly, this is how the import of Pastora/is cura is 
summarised in its titular dicta in the printed editions of the Clementines. 137 Rather, 
the pope argued, a request has to be made by the summoning authority to the 
ordinary judge of the province in which the defendant resides so that the latter might 
'remit' (i.e. extradite) him to the former. But in this case, Henry had not requested 
that the Pope (being Robert's feudal lord for Sicily and hence ordinary judge) remit 
Robert to him. As a result, Henry's sentence was null. 138 There is no need to become 
entangled in this very technical debate on summonses. Suffice it to say that in one 
way or another posterity accepted the papal line on this point: both Bartolus and 
Baldus agreed with Pastora/is cura that Henry's summons of Robert was 
unlawful. 139 
The premise that Sicily was extra districtum imperii also meant Clement could deny 
that Robert was guilty of the crimen laesae maiestatis, while granting that Robert 
had indeed committed hostilities against the empire. As a vassal of the emperor for 
the county of Provence, Robert owed Henry only that which was due to him because 
of the feudal bond between them, but by this alone Robert was not subject (subditus) 
qui est in ~lieno territori_o ... Sed ad illam Decretalem [i.e. Pastora/is cura], <licit ipse, non 
potes~ dan responsum m pace, sed pertranseat cum aliis erroribus Canonistarum.' As 
Pennmgton, T~e Prince and the Law, pp. 197-8, points out, Bartolus is referring to Cinus' 
fam?us q~aestz_o Rec~or civitatis ( edited by Monti, Le 'Quaestiones' e i 'Cons ilia', pp. 55-
74) m which Cmus discusses the problem of summonses extra territorium. Cf. Leicht 'Cino 
da Pistoia e la citazione di Re Roberto'. ' 
· 
137 
~he _ Richter_ and Friedberg edition states: 'Sententia, lata contra citatum, extra citantis 
te~i_tonum exsistentem, nulla est ipso iure. II . d. loann. Andr.' Similarly, the 1582 Rome 
edition (In Aedibus Populi Romani), col. 128: 'Sententia lata contra citatum extra citantis 
territorium existentem, nulla est ipso iure. Hoe <licit. ' 
138 Cl 2 11 2 ' . · · . ~?1· . . _: .... ~itare non potmt imperator, nee citatio, si quam forsan de ipso infra 
~mpem fines feci!, c1tatum extra imperium constitutum arctavit. Nos quoque, regis ordinarius 
mdex, quod ad imperatoris iudicium citaremus vel citari aut remitti faceremus eundem 
nequaquam fuimus requisiti.' ' 
139 Bartol us to D.48.17.1, fol. l 89r: 'Fae it ad quaestionem illam: utrum iudex istius civitatis 
possit ci~are_ ~liquem extra territorium suum existentem per suum nuncium vel per literas. 
D~cretah_s dici~ quod n?n, [~lem.2.11.2]. .. Ego consuevi tenere illam Decretalem, tanquam 
~xis~ens m terris ecclesiae, dicens earn esse veram de iure.' The same position can be found 
m his T~actatus super constitutione Ad reprimendum, CQT (Basel) - henceforth: Bartolus, 
Ad reprzmendum .- p. 280a. Note that Baldus' justification of the papal position, at 
C.1.3(6).32(31), differs from Bartolus'. Cf. Woolf Barto/us oifSasso'errato pp 75-8 and 90-4. ' 'J' ' . 
131 
. h. h 140 Th 
to Henry's power (potestas) in such a way that he could comrmt 1g treason. e 
corresponding section in Oldradus ' consilium provides a more detailed explanation. 
One could only commit the crimen laesae maiestatis by injuring a greater power 
(maior potestas) . But Henry did not have this power over Robert, firstly because 
Sicily was beyond imperial jurisdiction, and secondly, because a feudal bond does 
not grant it to the feudal lord, since a feudal bond does not give imperium over a 
vassal. Rather the feudal bond is a contract with obligations on both sides.
141 
Robert 
is not an internal rebel or traitor, but an external enemy waging war against the 
empire; he was a foreigner who happened to have a fief in imperial lands.
142 
In other 
words, Clement and Oldradus strategically demote Robert from the status of citizen 
or subject to that of foreign landholder. 
While historians of political thought have traditionally been interested in Pastora/is 
cura for its impact on debates about the nature of the medieval empire and national 
sovereignty, Kenneth Pennington, in a relatively recent contribution, has stressed the 
bull's importance in the development of procedural law in the ius commune. 
According to Pennington, the most important facet of the quarrel between Henry and 
Robert was that it acted as a catalyst for the creation of an inalienable natural right to 
judicial defence, which minimally consisted in the right to be summoned to court and 
143 . h ld b . 
thus in the opportunity to present a defence. Although due credit s ou e given 
to Pennington for highlighting this issue, he does tend to overstate his case. The key 
14° Clem.2.11.2: ' Unde imperator in ipsius personam nullam ratione terrarum huiusmodi, ni_si 
ex natura feudorum, debitam superioritatem habebat. [ ... ] imperatori eo modo non subfmt, 
quod in eum crimen laesae maiestatis commiserit, quantumlibet omnia in _se~tentia supra 
dicta contenta veritatem indubiam continerent. [ ... ] quum nullam super praem1ss1s potestatem 
haberet imperator in re gem, sibi alias, quam ut praefertur, non subditum ... '. . . 
141 Oldradus de Ponte, Consilia et quaestiones, cons. 43 : 'non comm1tt1tur cnmen Iese 
maiestatis nisi cum leditur maior potestas. oportet ergo quod sit maior potestas, sed hie nulla 
est potestas in leso respectu ledentis loquendo de potestate ... Et _si dic~tur feudum ab ~o 
tenet, respondeo, hoe non dat domino imperium in vasallo. Est emm qmdam con~ractu~, m 
quo quis alteri obligatur... Non autem imperium conceditur ... Si non habet merum tmpenum, 
non est sub potestate sua ... ' . . . . . . 
142 Ibid.: 'Et hoe est etiam videre in hoste, qm comm1ttens m prmctpem, s1ve et helium 
indicens non dicitur committere crimen Iese maiestatis, ut [D .49 .15 .24].' 
143 Pennington, The Prince and the Law, esp. eh. 5, pp. 165-201 - _at p. 188: 'Fund~mentally, 
the importance of Pastora/is lies outside the framework of_ pu_bhc law. Clements decretal 
established that a key element of judicial process was mv10Iable b~cause natural la':" 
demanded that people be given an opportunity to defend th~mselves m ~ourt. Past~ralz~ 
established an essential element of due process and protected tt from the will of the pnnce. 
See also his subsequent articles 'Due Process, Community and the Prince', and ' Law, 
Criminal Procedure', which summarise the arguments of his book. 
132 
question facing the jurists was indeed 'whether Henry had extended Robert due 
proce_ss of law' (if that modem concept is not anachronistic), 144 but the more specific 
question, whether a defendant had an immovable right to be summoned for the 
purpose of a defence, was a relatively minor part of the debate. As we have seen, the 
crux of the pope's procedural argument was that Henry 's sentence was invalid due to 
technical legal faults in Henry's actual summons of Robert, in particular, because a 
defendant could not be summoned extra territorium. For this reason, it is only at the 
very end of his bull that Clement mentions, almost in passing, that the emperor 
cannot deprive a defendant of a judicial defence since this is derived from natural 
law over which the prince has no cognizance.145 It was not absolutely necessary for 
Clement to do this given that Henry had in fact summoned Robert. It seems that the 
pope wanted to refute all of the claims in Henry's sentence exhaustively and 
preclude the potential counterargument that technical deficiencies in the summons 
could be discounted by the fact that the prince was legibus solutus and therefore did 
not need to issue summons in notorious cases anyway. 
Again, Oldradus' consilium 43 provides some theoretical background for Clement's 
~ove. The purpose of the summons is to provide the defendant with the knowledge, 
either real or presumed, of the charges against him.146
 Without this knowledge, it 
ld b . "bl 147 · wou e 1mposs1 e to mount a defence. This f acultas or possibilitas defendendi 
in judicial affairs is derived from the natural law because the latter also grants each 
~~d every person the ability to defend himself physically against unlawful violence. 
And because iura naturalia are immutable, the prince cannot take this faculty 
away from the defendant. 149 We should note that Oldradus was not the first jurist 
144 p . 'D 
145 
ennmgton, ue Process, Community, and the Prince' . 
Cl~m.2.11 .2: 'Nee praedicta suppletio circa subditum etiam ad ea potuisset de ratione 
refern, . per quae de c~in:1ine, ~raesertim sic gravi delato defensionis, ( quae a iure proven it 
natural~,) fa~ultas adtmt valmsset, quum illa imperatori tollere non licuerit, quae iuris 
naturalts exs1stunt.' 
146 Oldradus _de Ponte, cons. 43: 'Effectus intentus est scientia citati et facultas defendendi. 
Et hoe non v1detur posse omitti, quia interveniet vere vel interpretative .. . ' 
147 I?id.: ' sed ~uia defensio sine scientia esse non potest, necessaria est omnino scientia vera 
vel mterpretatlva.' 
148 lb 'd 's· · · 
. t : : 1cut emm mre nature permissa est unicuique defensio contra extraiudicialem 
v1~lenttam, ut [D.1.1.3: u~ vi~ atque; in_iuriam propulsemus: nam iure hoe evenit, ut quod 
q~1sque o~ tutelam corpons sm fe~ent, mre fecisse existimetur.. .], et [D.9.2.45,4: Qui, cum 
ahter tuen se non possent, damm culpam dederint, innoxii sunt: vim enim vi defendere 
omnes leges ornniaque iura permittunt.]. [ ... ] Sic et adversus violentiam iudicialem ' 
149 Ibid.: 'et cum princeps hac non precedente iudicat, subtrahit homini defensi~~em iure 
nature concessa. Sed hoe non potest princeps, ut [Inst.1.2, 11: Sed naturalia quidem iura, 
133 
(although he seems to be the first known civilian) to argue that summonses were 
derived from the natural law .150 Pennington has found one earlier instance in the 
commentary of the French canonist Johannes Monachus ( d. August 1313 - so before 
the promulgation of Pastora/is cura) 15 1 to Pope Boniface VIII's 1303 decretal Rem 
non novam. 152 In Johannes' commentary, the argument is aimed at the papal 
· 153 pnnce. 
By making the right to be summoned a natural right, one could draw on a whole 
tradition concerning the immutability of iura naturalia and the emperor's limited 
power to operate on these higher norms. It added an extra layer of defence against 
the highest secular judicial authority in Christendom. This strategy, however, was not 
watertight, since it did not address the obvious counter argument: the ability of the 
prince to derogate from higher norms with iusta causa. Of course, it was Oldradus ' 
and Clement's clear intent to deny that the emperor could in any way deprive 
someone of judicial defence. Pennington' s examination of the reception of Pastora/is 
cura in the legal tradition of the fourteenth century reveals widespread agreement, 
both amongst canonists and civilians, with the bull's arguments and especially the 
idea that the emperor could not dispense with judicial summons. But again, there is 
no suggestion, at least in the evidence presented by Pennington, that the jurists 
confronted the challenge posed by iusta causa. 154 He points to one exception, 
Simone da Borsano (d. 1381), who argued that the pope 's plenitude of power 
allowed him to deprive a defendant of judicial defence with just cause, while denying 
that the emperor could do the same. 155 Johannes Andreae ( d. 1348) 'may' have held 
the same position.156 This leads Pennington to conclude that although the provisions 
of Pastora/is cura had been 'established as the regulatory norms of judicial 
quae apud omnes gentes peraeque servantur, divina quadam providentia constituta semper 
firmas atque immutabilia permanent... ]. .. ' . . . . 
150 One of the anonymous cons ilia in the A v1gnonese dossier which, accordmg to 
Pennington, informed Clement's legislation (and which also contains Oldra_du~' co~silium 
43) employed the same argument as Oldradus regarding summonses - this 1s edited by 
Schwalm, MGH, Leg. sec. IV, Tom. IV, Pars 2, pp. 1378-98 (no . 1255) and discussed by 
Pennington, The Prince and the Law, pp. 183-5. Cf. p. 129, n. 131 above. . 
15 1 Cf. s. Stelling-Michaud, 'Jean Le Moine ou Johannes Monachus', Dictionnaire de drozt 
canonique, vol. 6, cols 112-3. 
152 Ext. corn. 2.3.1. 
153 For further details, see Pennington, The Prince and the Law, pp. 160-3. 
154 Ibid., pp. 190-201 . 
155 Cf. 'Simone da Borsano', DBGI, vol. 2, pp. 1869-70. 
156 Pennington, The Prince and the Law, pp. 195-6 and 191 respectively. 
134 
procedure in Roman and canon law' by the end of the fourteenth century, 
'nevertheless the papal "prince" was not yet bound by his own legislation.' 157 
Ac~ordi~g to Pennington, this situation had changed by the end of the 151h century, at 
which time ' [d]efendants' rights to present their cases in court had become so 
embedded in the ius commune that even the prince's absolute power could not 
dislodge them '. 158 Pennington' s evidence, however, does not substantiate this claim. 
His discussion of the Pazzi conspiracy, the case study on which his argument is 
based, reveals an apparent lack of agreement within the later ius commune whether 
the prince (the pope in the Pazzi case) could derogate ex iusta causa from higher 
. h 159 norms m t e context of procedure. If one reads Pennington' s footnotes carefully, it 
seems that two out of the four cons ilia examined - written on behalf of Lorenzo dei 
Medici, who had been summarily excommunicated by Pope Sixtus IV in 1478 -
acc~pt that the prince could in theory proceed without a summons. Essentially, they 
demed that the necessary factual circumstances (total notoriety) had been met in this 
particular case; in other words, the pope had not acted with just cause. 160 
Of course, any discussion of the prince in the form of emperor or pope was by its 
ve~ nature exceptional given their special positions in the legal hierarchy of western 
Christendom, particularly so in the pope ' s case due to his status as the Vicar of 
Christ. Indeed, as we will see in chapter 4, what emerges clearly from the civil law 
_tradition on communal banishment ( even before Pastora/is cura had been absorbed 
into civilian jurisprudence) is that defendants, including those who had committed 
157 Ib"d I ., p. 201. 
158 Ib"d 1 ., p. 267. 
159 Ibid., eh. 7, pp. 238-68. 
160 Ib"d 1 ., P; 24 7 (n. 41 ), ~ro?1 ~ joint cons ilium written by the leading doctores legum of 
Fl~renc~: . A~ secundam hm1tat10nem que est de substantia cause respondemus quod ad hoe 
ut 1sta l~m1tat10 procedat oportet duo concurrere copulative: primum quod utatur plenitudine 
potestatls, secun~um quo_d moveatur ex iusta causa subsistente. Probatur: papa et imperator 
n?~ sunt supr~ JUS gen_tmn:i et natural e... et si quandoque per eos dispensari concedatur 
d1c1t~r co need~ _de plemtudme potestatis. .. Sed numquam conceditur talis dispensatio nisi 
subs!stente leg1t1ma_causa .... ~rgo ad dispensandum quod non requiritur citatio que est de iure 
gentmm vel naturah ... requmtur copulative plenitudo potestatis.' From the same cons ilium 
p. 248 (n. 42t ·~ed _ in is~o processu exprimitur una causa que esset sufficiens ad 
procedendum sme c1tatlo~e, s_1 esset vera, videlicet totalis notorietas. [ ... ] Sed in hoe processu 
est exp~essa causa notonetatls que est falsa et insufficiens, ergo etc. ' The other consilium 
was :"'T1tte~ by Gi~olami Torti and used the same logic - p. 266 (n. 120): ' ... quod ubi 
not?num sit ~efe~s10nem non competere, tune procedi possit sine citatione, quia tune non 
tolhtur defen_s10 s1 non subest... ubi vero apparet defensionem competere, ut in casu nostro 
vel esset dubmm, procedat opinio tenentium quo requiritur citatio ... ' ' 
135 
notorious crimes, had to be summoned in criminal proceedings before they could be 
condemned in absentia; they had to be given at least the opportunity to present their 
defence. Amongst the commentators, with the exception of Albericus de Rosate, 
there was nothing but enthusiastic approval in the communal context for the 
argument of Pastora/is cura that this right was derived from natural law.
161 
Despite the exceptional position of the emperor, much the same can be said for the 
imperial ban. In practice, as we saw in chapter 1, emperors were careful to summon 
defendants before banishing them, 162 and when Henry Vil exceptionally claimed that 
he was not bound by this rule, he was firmly reprimanded by the pope, who used 
Henry's deviation from established procedural norms for his own political ends. In 
the legal theory on imperial banishment, granted that before Bartolus there is little 
material on this topic, we have not seen any indication so far that the jurists used the 
l. Amissione, that is, the idea of the hostis declaration, to frame the imperial ban as an 
exceptional form of procedure that could dispense with the legal formalities that 
were usually required. Quite the opposite: the ordinary mechanisms of Roman 
criminal law, such as annotation and deportation, are recurring points of reference in 
their discussions. And as we shall now go on to see, Bartolus in his extensive 
considerations on the imperial bannitus fully agreed with Clement and Oldradus that 
the right to judicial defence, rooted as it was in natural law, has to be respected even 
in imperial proceedings for lese majesty. At the same time, his embrace of the hostis 
as the key analogy for the imperial bannitus, as well as the idea of ipso iure 
sentencing, went quite a long way in negating the provisions of Pastora/is cura. 
6. Bartolus on Imperial Banishment 
i. Introductory Remarks 
Bartolus is the first and only medieval jurist to provide an in-depth and theoretically 
satisfying account of the medieval imperial ban. His thought introduced two 
conceptual shifts in the discourse. Previous jurists, such as Accursius and Jacobus de 
Belviso, defined the legal position of the imperial bannitus by the casuistic use of 
161 Ch. 4, pp. 213-221. 
162 Ch. 1, p. 19. 
136 
analogy. They often compared the bannitus to the deportatus but they did not 
consider these to be identical legal figures; they established a looser kind of 
~onnection. For Bartolus, in contrast, the modem imperial bannitus or rebel/is simply 
1s the same as the ancient imperial hostis. By closely assimilating the former to the 
latter, he provides an overarching conception. The position of the bannitus is no 
longer predominantly established on a case-by-case basis, but by a systematic and 
creative application of the position of the hostis found in the texts of the Corpus 
iuris. Specific questions about the imperial bannitus are answered by reference to a 
general, underlying legal status. Secondly, Bartolus moves the discussion away from 
a relatively restrictive reading of the imperial ban, which associated it with the 
ordinary criminal procedures of Roman law, to one more detrimental to the bannitus 
through employing the concepts of Iese majesty and perduellio, and in consequence, 
notions of extraordinary political procedure. 
As we saw above, neither Accursius' nor Jacobus de Belviso's work on the imperial 
ban has been analysed by historians. With reference to later commentators 
especially Bartolus, the most comprehensive histories of the legal theory of the ba~ 
by Ghisalberti and Cavalca tend merely to mention without much further discussion 
the existence of an imperial ban, and at best highlight that it deprived the , bannitus of 
th . h f h . . 163 e ng ts o t e c1v1l law. Some aspects of Bartolus' treatise on Henry VII's 
~onsti~tion _Quoniam nuper est, which contains his most important thought on 
1m~~nal b~mshment, have been examined in Italian legal histories of the concept of 
pohtical cr~me (ii reato politico). These, however, focus on notions of Iese majesty 
and sovereignty, and tell us very little about procedure or the legal position of the 
163 Gh. lb . 'L ' 1sa ertJ, a condanna, pp. 42-4; Cavalca, I! bando, pp. 51-5. The accounts of both 
~utho~s are, h_owever, ~ery_ ~on~sed. Ghisalberti makes a cardinal error in arguing that the 
1mpenal hostls loses_ his civil nghts as well as the rights derived from the ius gentium. He 
seems to be led to this erroneous conclusion by a corrupted passage in the printed editions of 
Bartolus' Trac(atus bannitorum!de bannitis which states just that: see, p. 166, n. 256 below. 
Cavalca, -~· 52, ~orrectly says that the imperial bannitus loses his civil rights (and hence 
~om~n c1~1zens?1p, as well as the ability to reside in the empire, make any contracts, and to 
mhe~1t); his claim about contrac~s,_ ho~ever, is wrong, as we shall see shortly. On p. 54, he 
provides further ~xamples of c1v1l nghts lost, including patriapotestas and testamentary 
~owers, and ment10ns a number of other features of what he calls an imperial ban or a ban de 
1~re ~ommune. The problem is that these features form an odd mixture ostensibly drawn 
Uudgmg_ by the footnotes) fro?1 various jurists' discussions on annotation, relegation, 
de~ortat10n, and communal ba~1shment. In fact, it is very difficult to see how any of the 
evidence he employs at that p~mt r~lates t~ the imperial ban. Note that Woolf, Barto/us of 
Sass~ferrato , pp. 200-1, ment10ns m passmg that the imperial bannitus retains the iura 
gentlum. 
137 
imperial bannitus, particularly about civil rights lost and natural rights retained.
164 
Mooney, in his 1976 PhD dissertation, is the only historian to have approached 
Bartolus' commentary specifically from the perspective of banishment. 
Unsurprisingly, however, given the focus of his thesis on the communal ban in the 
statutory law of Florence and Nello da San Gimignano's fifteenth-century treatise De 
bannitis, Mooney's analysis remains very limited and confines itself to a short 
summary of Bartolus' main points.
165 Nevertheless, Bartolus' work, especially his 
commentary on Quoniam nuper est, provides a much richer harvest than the lack of 
historical interest might suggest. 
ii. Ad reprimendum: Ordinary Procedure 
Before we examine Bartolus' thought in detail, let us for a moment turn back to the 
quarrel between Henry and Robert. In order to bolster his legal grounds for 
summarily proceeding against his enemies in Italy, Henry issued two imperial · 
h d I . . th . 166 H ' constitutions in April 1313 that a a astmg impact on e zus commune. . enry s 
constitutions, Ad reprimendum and Quoniam nuper est, came to be appended to the 
end of the Corpus iuris civilis as Extravagantes Henrici VII under the titles 
Quomodo in crimine laesae maiestatis procedatur (tit. 1) and Qui sint rebelles (tit. 2) 
- Bartolus viewed these as the eleventh collation of the Authenticum. During the last 
years of his life, some time after 1354, he wrote extensive glosses, commonly 
referred to as treatises, on both constitutions. According to a report of the fifteenth-
century jurist Bartolomeo Cipolla, Bartolus was authorised to do so by 'the 
emperors'. This may stand in connection with his mission, as part of a delegation 
from the city of Perugia, to the court of Emperor Charles IV in 1355.
167 Whatever the 
case may be, Bartolus' works did indeed become the ordinary glosses to Henry's 
164 Ghisalberti, 'Sulla teoria dei delitti di lesa maesta ', pp. 113-15 and pp. 133-4. Sbriccoli, 
Crimen laesae maiestatis, regularly cites Bartolus' commentary on Quoniam nuper, but 
there is little actual discussion of its content, especially in respect of the legal position of the 
imperial bannitus - e.g. pp. 187-8 and 225-6 discuss the case of Henry and Robert t? stress 
the idea that only a subditus can commit the crimen laesae maiestatis. Note that this work 
focuses on the early-modem period. 
165 Mooney, 'The Legal Ban' , pp. 123-4, 130, and 216. 
166 Edited by Schwalm, MGH, Leg. sec. IV, Tom. IV, Pars II, pp. 965-7 (nos. 929 and 931). 
167 Bartolus certainly devotes his works to Henry VII and Charles IV - Ad reprimendum, p. 
261: ' ... ipsas glossare volui ego, Bartol us a Saxoferrato... et etiam a_d lau_d~m d~vi?a~ 
recordationis domini Henrici Imperatoris, ipsarum constitutionum auctons, av1 1llustnss1m1 
domini d. Caroli IIII. Imperatoris nunc regnantis, cui debito fidelitatis adstringor ... ' 
138 
laws and were printed together with the latter in the early modern editions of the 
Corpus iuris. 168 . 
Ad reprimendum declared that in cases of Iese majesty the imperial authorities could 
proceed summarily ('summarie et de piano sine strepitu et figura iudicii') and come 
to a definitive sentence even in the absence of the defendant, as long as the latter had 
been legitimately summoned and could therefore be deemed contumacious. 169 In 
~ffect: what ~enry did was to update the ius commune to reflect long-established 
1mp~nal practice. In his commentary to Ad reprimendum, 170 Bartol us fully accepts 
the 1de~ that a contumacious person could be condemned in absentia, 171 but he also 
emphasises the papal argument in Pastora/is cura that the necessity for a lawful 
summons was guaranteed by natural law and thus had to be included in any form of 
abbreviated or summary procedure before a sentence could be passed. 172 His pupil 
Baldus shared this position. 173 
168 Cf S . 
. av1gny, Geschichte des Romisches Rechts, vol. 6, pp. 175-7; van de Kamp, Bartolo 
da Sassoferra_to, pp. 43-5; F. Calasso, 'Bartolo da Sassoferrato', DBI, vol. 6, pp. 659-60; 
~:nge and Knechbaum, Die Kommentatoren, pp. 189, 688, and 728. 
. Sch_w~lm (ed.), MGH, Leg. sec. IV, Tom. IV, Pars II, pp. 965-6 (no. 929): ' .. . ne ex eorum 
ab~ent1a m detestando le~e ma_iest~tis_ crimine processus et sententia retardentur ... sancimus, 
ut m qu_oc~nque Jes~ maiest~tls cnmme et maxime ubi contra Romanorum imperatores vel 
reges ahq~1d quad _dictum cnmen tangat asseratur commissum, possit procedi ... summarie et 
d~ piano s1?e strep1tu et fi~ura iudicii ... Ad hoe si qua communitas, corpus vel collegium vel 
al~a _quaev1s per~ona, cumscu~que status, . dignitatis vel condicionis existat, super dicta 
c~1mme per nun~m?1 vel ~er hteras seu et1am per edictum publice propositum citata vel 
c1ta_tum neg!exent m ~ermmo _pers~naliter vel legitime comparere, contra ipsam vel ipsum 
pr01nd~, acs1 p~rsonahter ommbus mteresset et lis foret legitime contestata ... ad diffinitivam 
~~ntenham et em~ exec_utionem procedatur ... ' 
F_or, a good _d1scuss1on of Ad reprimendum as well as Bartolus' commentary on it, see 
Beth, La dottrma ·~ Bartol us' commentary is also discussed by Pennington, The Prince and f~1e Law, pp. 196-200, and by Mooney, 'The Legal Ban', pp. 121-l. 
E.g. Bartolus,_ Ad r~pr_ime~dum_, p. ~~2b: 'Diffinitivam. Vides ergo casum in quo, absente 
reo, potest negotmm hqu1dan et d1ffimtive pronuntiari in criminalibus ' 
172 lb ' · 
id., p. 273a-?: 'T~lluntur substantialia iudicii de iure civili introducta, non autem ea 
quae su_nt naturaha, a mre n~turali seu gentium inducta naturali ratione... Quaero igitur, 
utrum _s1~ necesse _quad pa~s c1tetur? ~espondeo, sic, ut in hac lege innuitur. Item, quia hoe 
est de i_me naturah. nam pnmum hommem delinquentem citavit Deus, dicens: Adam, ubi es? 
Hoe etiam probatur [Clem.~.11.2], ubi sententia domini Imperatoris Henrici, qui fecit hanc 
legem et ~os~ea cond~mnav1t Regem Robertum, Regem Hierusalem et Siciliae, cassa fuit ex 
eo quod c1tat10 non fu1t facta legitime ' ID ' 
Baldus to C. I .14( 17).11, quoted and discussed in Pennington The Prince and the Law p 
201 . ' ' . 
139 
The emperor was emphatic in Ad reprimendum that the provisions of his constitution 
should apply to all cases of Iese majesty. 174 Bartolus explains that the constitution 
was widely framed in this way since (as we saw at the beginning of this chapter) 
there are many offences that fall under the !ex Julia maiestatis. 175 Logically, that 
would include perduellio, the extreme subcategory of Iese majesty - treasonous 
activity 'animated by a hostile spirit against the commonwealth or emperor' , 176 
which Bartolus will go on to link with the hostis and rebel/is in his commentary on 
Quoniam nuper. The important point is that Ad reprimendum, and by extension 
Pastora/is cura, cover even these acts of high treason, or as the constitution itself 
states, acts of rebellion directed immediately 'against the Roman emperor or kings '; 
the preface of the constitution specifically employs the term rebellio. 177 As Bartolus 
explains, the crimen laesae maiestatis is the broader category, the genus, while such 
immediate acts of rebellion are its species. 178 It is important to remember this point 
as the procedural safeguards introduced in Ad reprimendum and Pastora/is cura -
particularly the natural right to a summons and a defence - appear to be in conflict 
with the key provision of Quoniam nuper. The latter explains who should be 
considered a rebel of the empire and states that rebel, i.e. hostis, status is acquired 
ipso iure immediately upon the commission of the crime. 179 But how can this be 
squared with the natural right to judicial defence? We shall return to this question in 
due course. 
174 See n. 169 above. 
175 Bartolus, Ad reprimendum, p. 264a: 'In quocunque. Cum enim Jex lulia maiestatis habeat 
multa capitula, ut [D.48.4.1-4]. in quocunque crimine cui competit aliquod dictorum 
capitulorum habet locum haec Jex.' 
176 D.48.4.11 - seep. 97, n. 9 above. 
177 Schwalm (ed.), MGH, Leg. sec. IV, Tom. IV, Pars II, p. 965 (no. 929): 'Ad reprimenda 
multorum facinora, qui .. . hostili animo animati conantur humana, verum etiam divina 
precepta, quibus iubetur quod omnis anima Romano principi sit subiecta, sceleratissimis 
facinoribus et assiduis rebellionibus demoliri ... ' Commenting on the word 'rebellionibus', 
Bartolus, p. 263a, notes: 'Quid sit, dicetur infra proxime, etc.' This refers to his treatise on 
Quoniam nuper which immediately follows his treatise on Ad reprimendum. 
178 Bartolus, Ad reprimendum, p. 264a: 'Et maxime. Hie praecedit genus, scilicet omne 
crimen quod incidit in legem luliam maiestatis, et additur species, quando cornmittitur in 
lmperatorem vel Regem.' 
179 For the text of Quoniam nu per, see p. 146, n. 194 below. 
140 
iii. The Distinction between bannitus imperii and rebellis imperii 
First we should note that despite his close alignment of the medieval imperial 
bannitus with the ancient hostis, Bartolus argues that there remains one important 
conceptual distinction between the two figures : bannitus is a broader term that can be 
divided into the subcategories used throughout this chapter; it can refer both to the 
coercive procedural ban as well as the permanent punitive one. For this reason, 
Bartolus explains in his commentary to Quoniam nuper that, strictly speaking, one 
should distinguish between a rebel of the empire - whom he fully equates with the 
hostis - and a bannitus of the empire. Whereas to be a rebel was 'fatal' (mortalis), to 
be a banished person was 'medicinal' (medicinalis) , since the penalty of the former 
'cannot be remitted except by a full judicial pardon ' , while the penalty of the latter 
'is remitted when the contumacy has been cleansed, that is, when satisfaction has 
been made', as described in Frederick II' s laws Item null a and Item quaecunque. 180 
This medicinal ban should be compared to annotation in Roman law, since both are 
judicial remedies against the contumacious defendant; they are medicinal because it 
is in the power of the defendant to return to obedience. While Bartol us seems to view 
Roman law annotation as the default interpretation of the medieval term bannum 
imperiale, he clarifies that the effects of this ban are different from those of 
annotation, and are mainly established by regional laws and customs. He does, 
however, specify one general consequence: just as excommunication removes a 
person from ecclesiastical offices, so the ban removes someone from the 'benefices ' 
(beneficia) of the empire. 181 In his Digest commentary, where Bartolus also speaks of 
180 
Bartolus, Tractatus super constitutionem Qui sint rebelles, CQT (Basel) - henceforth : 
Bartolus, . Qui sint rebelles - p. 287a: 'Item, nota quod aliud est esse rebellem lmperi i, aliud 
est ~s~e ~n. banno lmperii: primum, scilicet rebellio, est mortale, ut hac lege, nee potest 
rem1tt1 ms1 ex mera gratia. secundum, scilicet bannum, est medicinale, quia remittitur 
purgata contumacia seu facta satisfactione, ut [auth. Item nulla, post C. l.3(6).2] et [auth. 
Item quaecunque, post C. l.3(6).13]. ' 
181 
lbi~., p. 287a-b: 'et sic videtur quod quatenus excommunicatio removet quern ab officiis 
ecclesiae, eatenus bannum a beneficiis lmperi i, ut dictis iuribus. et tangit per Gul[elmus de 
~uneo] in d. ~uthentica item quaecunque. facit [Sext.5 . l 1.1]. facit [D.48.17. J ], ubi glossa 
1llam adnotationem appellat bannum, et est medicinale quia, cum venit intra annum, 
recuperat statum. secundum vero statuta et consuetudines locorum, quern esse bannitum, 
habet diversos effectos, de quo aliquid infra dicam in glossa magna, etc.' This last comment 
refers to Bartolus long gloss a. v. rebellando (discussed below), which examines the effects 
of hostis status in detail. His reference to the Ordinary Gloss points out Accursius' 
description of annotation as a bannum fisci - at D.48.17.1, col. 1384: 'Secundo, bona 
annotantur, dicendo, omnia bona pono in banno fisci, nisi venerit intra annum ... ' Bartolus' 
reference to the Liber sextus suggests that he borrowed the terminology of 
141 
I 
I I 
11 
the ban in connection to Frederick's legislation - and therefore of the imperial ban in 
its medicinal sense - he indicates that these beneficia are synonymous with iura 
civilia and ea quae sunt iuris civilis, given that the bannitus loses these; in that sense, 
the bannitus can even be compared to the deportatus. 182 
Although Bartolus strictly separates the terms rebel/is imperii and bannitus imperii 
towards the beginning of his commentary to Quoniam nuper, later on, when 
discussing the legal position of the rebel/is, he also calls the latter an exbannitus. 183 
This is to be expected and falls entirely in line with the terminology of imperial 
practice. Henry, in his sentence on Robert, as well as describing his foe as a rebel/is, 
proditor, and hostis imperii, employed the terms exbannire and exbannitio; 184 and it 
is that sentence which Bartolus is primarily discussing in his commentary on 
Quoniam nuper. Bartol us' initial distinction is there to emphasise the differences 
between a rebel-bannitus, who is subject to the permanent punitive ('fatal') ban for 
the crime of treason, and the bannitus who is subject to the coercive procedural 
('medicinal') ban for contumacy. This distinction had one further and quite 
fundamental procedural consequence. As mentioned above, a rebel-bannitus became 
such ipso iure upon the commission of his act of rebellion without the need for a 
prior judicial sentence or declaration. Conversely, when someone was banished for 
contumacy, the effects of banishment did not apply until after a period of persistent 
disobedience and usually had to await a sentence by an appropriate authority. 
Procedural considerations aside, there is nothing to suggest that in Bartolus' thought 
the substantive legal statuses of these two types of imperial banniti were significantly 
mortalis/medicinalis from the canon law on excommunication - Sext.5.11. I : 'Quum 
medicinalis sit excommunicatio, non mortalis, disciplinans, non eradicans, dum tamen is, in 
quern lata fuerit, non contemnat...' The idea that the medicinal ban removes a person from 
the beneficia of the empire is supported by Guilelmus de Cuneo's commentary to auth. Item 
quaecunque, fol. l 3vb: 'Et potest dici quod loquatur de personali quando bannitur per 
principem, et privet eum a communione illorum de imperio ... ' This comment is also cited by 
Baldus in his commentary to the same authentica, fol. 39ra: 'et per istud bannum imperiale 
privatur quis iuris civilis beneficio et conversatione fidelium imperii, secundum Guil.' 
182 Bartolus to D.4.5 .5, 1, fol. 141 rb: 'Et determinavi sic, quod ille qui est bannitus imperii, 
aequiparatur deportato et perdit ea quae sunt iuris civilis, ut hie. Et ita intelliguntur iura nova, 
quae dicunt, banno imperiali subiaceant, hoe est, iura civilia perdant, ut in [auth. Item nul/a, 
post C .1.3(6).2], et in multis aliis.' 
183 Bartolus, Qui sint rebelles, p. 290b: ' ... sicut exbannitus et condemnatus ab Imperatore 
non perdit ea quae sunt iuris civilis Papalis ... ' 
184 Schwalm (ed.), MGH, Leg. sec. IV, Tom. IV, Pars II, pp. 989-90 (no. 946): ' .. . privamus 
eundemque Robertum rebellem, proditorem et hostem imperii dicteque maiestatis crimine 
reum de toto predicto imperio exbannimus et diffidamus ... Huiusmodi au tern privationum, 
confiscationis, exbanitionis, diffidationis, condempnationis ... ' 
142 
different during their time of banishment. Such an interpretation would be odd given 
that in practice their positions were basically identical once they were fully banished 
- both were thus subject to the liability of being injured in person and property with 
impunity by any private person. 
iv. The I. Hostes: Banishment as a Declaration of War 
The best introduction to Bartolus' view on the imperial rebel is not his treatise on 
Quoniam nuper but rather a section of his famous commentary to the I. Hastes 
(D.49.15.24), where he is mainly concerned with describing the legal relationships 
between the empire and its various peoples, and between the empire and foreign 
l 18s H . peop es. avmg argued that practically all the diverse peoples who obey the 
Roman Catholic church form a part of the Roman people and are Roman citizens, 
Bartolus qualifies this in one important respect: they are citizens unless the emperor 
has declared a public war on them, on account of which they become hostes.186 
Bartolus explains this in the following terms: 
The ~hird way of declaring a public war is when it is declared by the Roman people, 
that is, by th~ e_mperor, to whom all jurisdiction of the Roman people has been 
transferred ... Similarly, [a public war is declared] when someone declares war on the 
emperor... And theref?re I thi~ that the cities of Italy against whom the emperor has 
decl~red war, as agamst the city of Florence and others, truly are enemies of the 
empire, and captured persons are enslaved, and the things that are captured there 
become the property of the captors ... 187 
In this passage, Bartolus succinctly applies established civilian jurisprudence on the 
hostis to contemporary rebellion against the empire. As Accursius before him he 
' combines the idea of the I. Amissione with the idea of the I. Hastes: the community 
(through the emperor) can declare war on a private citizen and vice versa. Some 
185 
186 
Cf. Woolf, Barto/us ofSassoferrato, p. 25 ff. and pp. 198-9. 
Bartolus t~ D.49.1_5.24, fol. 228ra: 'Et secundum hoe quasi omnes gentes quae obediunt 
Sanct~e ~atn Eccles1ae sunt de populo Romano. [ ... ] Praedicta vera nisi helium publicum 
esset md1ctum contra aliquos ex istis. Tune, nota, efficerentur hostes propter indictionem 
belli, ut infra dicam.' 
187 
Ibid., fol. 228rb: ' Tertio _modo indicitur helium publicum quando indicitur a populo 
Romano, :el _ab_ I~p_eratore, m quern translata est omnis iurisdictio populi Romani ... Item, 
quando ah_qm_s _md1c1t bell um Imper~t?ri .. . Et ideo puto quod civitates Italiae contra quas 
lmper~tor m?1x1t bellu~, ut contra c~v~tate~ Florentiae et similes, sunt vere hostes Jmperii, 
et capt1 efficmntur serv1, et res, quae 1b1 capmntur, efficiuntur capientium ... ' 
143 
. , 'I 
I' I 
, I 
jurists, such as Baldus, associated the emperor's (and any other a
uthority's) internal 
declaration of war with the term dif.fidatio, which seems to hav
e migrated into the 
civilian tradition from canon law.
188 Once a citizen has acquired hostis status, there is 
nothing to differentiate him from any other forei
gn enemy of Rome; hence, it is 
possible to acquire lawful title over his person and p
roperty. 
As Bartolus explains in his commentary to Quoniam nuper,
 the terms hostis and 
perduellis encompass both the external enemy - tho
se who were never subject to the 
emperor - as well as those who used to be his fa
ithful subjects but then withdrew 
their obedience. For this reason, the term rebel/is 
is used 'today' to demarcate the 
domestic host is. 189
 As we shall see below, the only way in which the
 domestic and 
the foreign enemy could be said to differ is proced
ural. Medieval legal and political 
developments, such as the quarrel between Henry 
and Robert, forced contemporary 
legislators and lawyers to devise certain mechani
sms (Ad reprimendum, Quoniam 
nuper), including safeguards (Pastora/is cura), to regulate th
e process by which a 
citizen became an enemy, mechanisms that were n
owhere to be found in the texts of 
the Corpus iuris. 
As is well known, in his commentary to the l. Hast
es, Bartolus accepts the validity of 
the Donation of Constantine and, consequently, 
that the church holds temporal 
jurisdiction in those central Italian lands granted to it.
190 This, however, makes the 
inhabitants of church lands no less Roman. Thes
e provinces are still part of the· 
Roman empire, except that the pope now take
s the place of the emperor and 
188 Baldus to auth. Item nu/la, post 1.3(6).2, fol. 35rb: '!stud 
tamen bannum est quaedam 
diffidatio, quam non potest facere par in parem, 
sed superior in inferiorem, et ista diffidatio 
est quodammodo initium belli. Et dicunt Host[iensis], et
 Jo[hannes] And[reae] post eum, 
quod Christianus Christianum non debet offender
e nisi prius eum diffidaverit, vel iudicis vel 
iuris authoritatem habuerit. alias est proditor, u
t in [Sext.5.4.1 ], in verbo "diffidatus", in 
Novellae, quod nota.' Johannes Andreae's refe
renced Novella commentaria to Sext.5.4.1 
contains the principle reported by Baldus as part
 of a brief discussion of just war amongst 
Christians. 
189 Bartolus, Qui sint rebel/es, p. 288b: 'Sed quaero, in quo diff
ert hoe nomen rebellionis ab 
eo quod antiqui dicebant duelles, quos nos voca
mus hostes, ut [D.50.16.234)? Respondeo, 
illud nomen quo ad quid generalius est. Nam com
petit etiam his qui non consueverunt esse 
sub obedientia, qui indixerunt nobis helium vel
 nos eis, ut [D.49.15.24). Sed hoe nomen 
rebellionis proprie convenit his qui defecerunt cum
 essent subditi, [D.4.5.5 , 1 ] .. .' This point is 
picked up by Ghisalberti, 'Sulla teoria dei delitti d
i lesa maesta', pp. 113-15. 
19° Cf. Woolf, Barto/us of Sassoferrato, pp. 26 and 72-107; Cann
ing, Baldus de Ubaldis, pp. 
30-55; Ryan, 'Bartol us of Sassoferrato', pp. 73-6. 
144 
exercises the same jurisdiction as he once did.
191 This means that when someone 
rebels against the pope - or even against his rect
ors, as long as such rebellion is 
directed ultimately at the pope rather than at the
 rectors per se - such a person 
becomes a true enemy and rebel of the Roman e
mpire, with all the ensuing legal 
consequences. 
192 In the next chapter we will see that Bartolus doe
s not grant this 
power to local political entities, that is, cities 
and kingdoms within the Roman 
empire, understood broadly as encompassing we
stern Christendom; these cannot 
banish someone from the empire and their banniti r
emain Roman citizens. Naturally, 
this rule also applies to the ban of political entit
ies within church lands, such as 
Bartolus' own city of Perugia. 
193 
v. Quoniam nuper est: Extraordinary Procedure and The L
egal Status of the 
Rebel 
Let us now turn _our full attention to Henry VII'
s constitution Quoniam nuper and 
Bartol us' commentary to it. The opening of the c
onstitution notes that doubts have 
arisen amongst the emperor's subjects in northern Italy about wh
ether someone can 
officially be considered a rebel of the empire befor
e being sentenced for Iese majesty 
~y the_ authorities. Henry's response is to cla
rify that this status is acquired 
immediately upon the criminal act - i.e. ipso iure o
r ipso facto - and encompasses all 
persons 'who, in whatever way, publically or se
cretly, commit acts of rebellion 
against Our honour and against their due allegian
ce to Us, and who, through their 
191 
Bartolus to D.49. ~ 5.24, fol. 228ra: 'Quidam sunt populi qui non obediunt Princ
ipi, tamen 
asseru?t se habere hbertatem ab ipso ex contrac
tu aliquo, ut provinciae quae tenentur ab 
ecc~esiae Romana,. quae fuerunt donatae ab Im
peratore Constantino ecclesiae Romanae, 
pos1to pro constant!, quod donatio tenuerit, quodq
ue revocari non possit, adhuc dico istos de 
popul~ Romano esse. ~am ecclesia Romana ex
ercet illas in terras iurisdictionem quae erat 
Imp~n.o R?~ano, _ et 1s~d _fatentur, non ergo
 desinunt esse de populo Romano, sed 
adm101strat10 1starum provmciarum est alteri conc
essa.' 
192 
Ibi~., fol. 228r~: 'Hie tamen videtur quod si 
quis est rebellis officiali vel Rectori 
eccle~iae, fit rebelh~ t~ti ~ccles!a~, quae tenet 
locum Imperii in istis partibus. [ ... ] Circa 
praed1~ta qu~er~, q_u1d s1 ~hq_ua c1~1tas vel terra re
sistit alicui officiali domini Imperatoris vel 
summ1 ponhfic1s 10 tems 10 qu1bus tenet Imp
erium, an ex hoe dicantur hostes vere? 
~espon~eo, quandoque q~is reb_ell~t _officiali non
 propter officialem, sed propter Principem, 
1deo qma non_ vu~t obedire pr10c1p~, tune princ
ipi videtur rebellare, et principi indicere 
bel!um ... _S~d s1 qms rebellaret recton propter fact
um ipsius rectoris, quia eos male tractat, ut 
fac1~nt _c1v1ta!es. Ducatus et Marchiae, et tune 
non dicerentur hostes principis vel summi 
pontific1, sed 1psms rectoris.' 
193 Ch. 4, pp. 253-64. 
145 
rebellion, plot anything to the detriment of the prosperity of Our Empire, either 
against Us or against Our officials regarding those matters pertaining to the office 
with which they are entrusted'. 194 
In his commentary, Bartolus explains the ultimate rationale for Quoniam nuper. The 
ancient law-makers did not use the word rebellis and hence Emperor Henry 
promulgated this constitution to clarify the term. 
195 The most basic meaning of 
rebellare is 'to resist', and this resisting can consist in action as well as passive 
disobedience. 196 In the old laws, when someone deserted from the territory of his 
superiors and joined the ranks of the enemy in their territory, they called him a 
transfuga. 197 When subjects rebelled, not by joining the enemy in their territory, but 
by resisting the prince in the territory where they stood, the old laws did not have a 
special name for them, but simply called them 'those who defect', as at the 1. 
194 Schwalm (ed.), MGH, Leg. sec. IV, Tom. IV, Pars II, pp. 966-7 (no. 931): 'Quoniam 
nuper est ad auditmn nostrum deductum, quod inter nonnullos fideles et subditos nostros 
provincie Lombardie aliorumque locorum ltalie disceptationes et dubia sepius oriuntur, an 
infidelis et rebellis imperii quisquam reputari debeat, nisi prius per nostre rnaiestatis 
sententiam condempnatus appareat, nos attendentes, quod acta prava malorum potius quam 
verba sententiarum ipsos faciunt pena condignos, et eo ipso quod quis peccat, correctionem 
meretur et quanto plus differtur punitio, tanto magis culpa crescit et gentibus in exemplum. 
perniciosius transit, tenore presentium declaramus, decernimus et pronuntiamus, quod illi 
omnes et singuli sunt rebelles et infideles nostri et imperii, qui quomodocunque publice vel 
occulte contra nostrum honorem infidelitatis vel rebellionis opera faciunt et in nostram seu 
imperii prosperitatem aliquid machinantur contra nos sive nostros officiales in hiis, que ad 
commissum eis offitium pertinent, rebellando ... ' 
195 Bartolus, Qui sint rebelles, p. 287a: 'Hoe modo assumpto vocabulo vocantur rebelles, quo 
vocabulo antiqui legum conditores non utebantur, et ideo facta est haec constitutio ad hoe 
declarandum.' Strictly speaking, this is not true. The word rebellio does appear in classical 
Roman texts (cf. Glare (ed.), Oxford Latin Dictionary, vol. 2, pp. 1738-9), although Bartolus 
is right in the sense that it does not seem to have been a technical legal term and is thus not 
generally used in juristic texts. Nevertheless, in the Corpus iuris it appears at least once, at 
C.12.45(46).2 (promulgated AD 403 by the Emperors Arcadius, Honorius, and Theodosius), 
which describes deserters (though not deserters to the enemy) as rebels. Again, the sense 
seems to be non-technical: 'Si desertores inventi resistendum atque armis obtinendum 
putaverint, tamquam re belles in ipsis temeritatis suae conatibus opprimantur. . .' The word 
'rebelles ' is also used in the vulgate, but not glossed by Accursius. 
196 Bartolus, Qui sint rebelles, p. 287a: ' ... sciendum est quod rebellare idem est quod 
resistere... et hoe resistere potest fieri faciendo aliquid contra, vel non faciendo et non 
obediendo ... ' 
197 Ibid. : 'licet hoe nomine "rebellis" iura antiqua non utantur, sed si aliquis rebellat mutando 
locum, hoe est, recedendo de terris subiectis et eundo ad hostes, talis vocabatur transfuga, ut 
[D.48.19.30,1], et [D.49.15.19,4/7], et [D.49.16.7].' 
146 
Amissione.
198 And if, by rebelling in this way, these rebels actively made war on the 
prince, they were punished ipso iure and thereby lost their civil rights as stated by the 
1. Amissione.
199 
But in the old laws, according to Bartolus, when subjects rebelled 
against the prince merely through passive disobedience, as by not receiving imperial 
officials for example, they were not punished ipso iure, and did not lose their civil 
rights at the time of the offence. They only lost these if, and at the time when, the 
prince or the senate declared them to be hostes, again as described by the 1. 
Amissione.200 
According to Bartolus, the main import of Henry's constitution is that it extends the 
ipso iure acquisition of rebel status in two ways. Firstly, since active rebellion can be 
manifest or hidden, and since the old laws did not emphasise this, Henry makes it 
absolutely clear that both types of resistance count as rebellion.
201 Secondly, rebel 
status is now acquired also by those who resist the prince through passive 
disobedience, such as the above-mentioned unwillingness to receive his officials, or 
by scheming to delay them unduly, or by making illicit leagues and societies to the 
detriment of imperial interests. Such actions, even if they are not properly acts of 
war, are nonetheless rebellious. 
202 In other words, this medieval constitution is a 
direct continuation and expansion of the law at the 1. Amissione.203 
198 
Ib_id. ,. p. 2~7~-b: _' Qua?~o.que rebellant, retinendo locum in quo stabant sub Principis 
obed1entia, et 1stJ de mre c1v1h antiquo non habebant speciale nomen, sed vocabantur "illi qui 
deficiunt", ut [D.4.5.5, 1]. 
'.
99 
Ib_id., P: 2_87b: 'Et isti, si quidem rebellabant faciendo bellum Principi, tanquam hostes 
ipso mre pumebantur et perdebant ea quae iuris civil is sunt, ut ibi.' 
200 
Ibid., pp. 287b-8a: 'Sed si aliquis solum rebellasset non obediendo Principi vel non 
r~c~ptando gentem suam, non puniebatur ipso iure nee perdebat ea quae iuris civilis erant, 
IllSI 
Princeps vel senatus indiceret, [D.4.5.5]. 
201 
Ibid., p. 287b: 'sed quia hoe rebellare potest publice fieri et occulte, et ibi non 
declarabatur, ideo haec !ex declarat.' 
202 
Ibid., p. 288a-b: 'et in hoe suppletur per hanc legem. Nam quocunque modo machinatur 
aliquid contra Principis honorem et rebellando est rebellis ... scilicet, ut dicantur rebelles et si 
ope_ra hostilia non fa~ia~t, sed s~lum si _non ~bediant. [ ... ] Machinantur. Id est, aliquid dolose 
ord~nant, cum 1?~chmat10 propne cons1stat m arte verborum ... ut faciendo inter se ligam et 
s?c1etates_ proh1b1tas sub aliquod colore licito .. . vel pone quod, adveniente Imperatore vel 
em~ n~nc10, volente ingredi civitatem pro recuperandis iuribus Imperii, tarn ibi quam alibi, 
[a]h~u1 respondent quod nolunt intromittere tantam gentem eius, vel eum. esse periculosum 
statu1 -~orum, e_t tenent eum in verbis, et sic labitur tempus quo multa posset facere pro 
Impem prospentate. Istae sunt machinationes quaedam, licet ad actus bellicos non deveniant. 
~~}a n:iachina, sec~ta i~obedientia, ~acit eos re_belles, etc.' 
Ibid., p. 288a: et sic haec Jex d1recto tend1t ad declarandum [D.4.5.5,1] .' 
147 
At the end of Quoniam nuper, Henry declared that the constitution's 
provisions also 
apply to those who rebel against his officials. Bart
olus argues for a limited 
understanding of what it means to be an imperial offic
ial.204 This constitution, he 
states, only protects the emperor himself or his immedia
te subordinates (such as his 
legates and vicars), but it does not cover local political entities: cit
ies, kings, and 
other princes. Although these, broadly speaking, derive
 their jurisdiction from the 
universal empire, they are nonetheless independent from
 it in practice by imperial 
grant or long-standing custom, and therefore exercise 
jurisdiction principally for 
their own sake and not for the sake of the empire.
205 Since resistance against them 
does not directly impinge upon imperial interests, the c
onstitution ceases to apply. 
For a more detailed discussion of rebellion against loca
l political entities, Bartolus 
refers us to his work on banniti nostri temporis, that is,
 on the communal ban. As 
mentioned above, these considerations do not pertain to 
the papacy since it assumes 
the position of the empire within church lands. 
206 
Bartolus argues that the meaning of imperial official s
hould be limited in another 
way. For the constitution to take effect, rebellion against
 aii imperial official must be 
directed immediately against the wellbeing or state of t
he empire. The constitution 
does not refer to those that are contumacious (and, by implicati
on, subject to 
banishment proceedings) in ordinary civil or criminal matters brou
ght before an 
imperial court. Granted that, improperly or broadly spe
aking, such persons can_ be 
said to rebel (since they resist imperial authorities), they are not on 
account of this 
rebels of the empire in the sense of Quoniam nuper and therefore do n
ot fall under its 
204 There is a good discussion of this issue in Mooney, 'The
 Legal Ban', pp. 214- I 7. 
205 For the relationship of local and universal political
 entities in the thought of the 
commentators, see Woolf, Barto/us of Sassoferrato, and Canning, Bald
us de Ubaldis. 
206 Bartolus, Qui sint rebelles, p. 289a: 'Sed an haec constitutio habet 
locum contra eum qui 
rebellat contra aliquem Regem vel Principem, vel c
ontra aliquam propriam civitatem? 
Videtur cum hie dicitur "contra officiates nostros".
 sed omnes praedicti possunt dici 
official;s Imperatoris, quia vel ab eo habent iurisdiction
em ex constitutione, vel concessione, 
vel longa consuetudine ... Contrarium dico, nam tune n
on fit contra prosperitatem Principis 
vel Imperii principaliter... Et quod dicitur de officialib
us Imperatoris intelligitur de _ h_is qui 
principaliter ac directo pro eo sunt, ut Legati et Vicarii 
Imperatoris ... R_eg~s v~ro et c1v1tates, 
esto quod habeant iurisdictionem a Principe, tamen ea
rn exercent prmc1pahter propter ~e, 
non propter Principem, vel eius officiales, vel Imp
erii utilitate?1. se~ babe~ l?c~m. ms 
antiquum, scilicet [D.4.5.5, I], quae generaliter loquitur de his . ~m 
defi~mnt a 1~nsd1ctJo~e 
eorum sub quorum imperio sunt. et dicam in tractatum 
de banmt1s nostn tempons. Sed qmd 
si aliquis rebellat contra Rectores per Romanam eccles
iam positos in partibus ltaliae? Puto 
huic constitutioni locum fore, cum eius Imperium sit e
cclesiae ... quod apparet, quia rebellat 
contra prosperitatem Imperii.' 
148 
· · 
207 · 
prov1s10ns. This seems to reflect Bartolus' earlie
r distinction between the 
'medicinal' and 'fatal' ban. A contumacious defendant
 banished for crimes other 
than rebellion is not fully a rebel or hosti:S, nor does he 
acquire that status ipso iure, 
b~t he s~ould nonetheless be compared to that figure, as 
it provides the best analogy 
with which to determine his legal status. 
Ne~t, Bartolus introduces the specific and thorny cas
e of Henry and Robert by 
posmg two separate questions. Firstly, can a person f
rom the lands of the church 
become a rebel of the empire? Secondly, can this perso
n commit the crimen /aesae 
. t . 208 B l d I · 
mazes atzs. arto us ea s with the second question f
irst and completely accepts 
the _papal logic o_f Pastora/is cura. Robert cannot comm
it the crime of Iese majesty 
agamst Henry, smce Robert's domicile and origin are 
beyond the confines of the 
empire, for which reason Robert is not subject to the imperial majesty; he
 cites the I. 
Non dubito (D.49.15.7), which, as we saw in the previous chapter
, describes the 
position of peregrini in the Roman empire. In other wor
ds, foreigners cannot commit 
this crime. 209 
2
~
7 
~bi~.: 'In e~ ve~o quod loquitur in eo qui rebellat
 contra officiales suos, loquitur 
h~1tatJ~e ... Et . mtelhge quando rebellat pro his quae 
petuntur pro statu Imperii. Secus si 
ahqu~ s1_ngu_lans pe~~ona conveniretur coram Principe
 vel eius iudice in aliqua causa civili 
~el cnmmah ad pe_t1tionem alt~rius, vel esset contumax
, aut alio modo resisteret. tune enim, 
~1cet largo mo~o ~1catur reb~lhs ... tamen propter hoe n
on dicitur rebellis Imperii nee incidit 
m. h~nc _constitut10~em, qma non rebellat contra pro
speritatem Imperatoris vel Imperii 
~r~nc1pahter, quod m hac constitutione requiritur. .. s
ed in damnum adversarii cum quo 
ht1gat.' 
20
_
8 
Ibid., p. 289b: _'Quid si al_iquis homo de terris ecclesiae aliquid facia
t contra Principem de 
his quae ad rebelhonem pertment, an rebellis est et in cr
imen laesae maiestatis incidat?' 
209 
Ibid.: 'Et Henricus Imperator huius legis auctor determ
inavit quod sic, condemnando de 
h?c Robertum_ Regem, [~lem.2. ~ 1.2]. sed Papa in § "rursus" determinavit contrarium, 
dtcens_ eum ~m est de terns ecclesiae Romanae originis 
et domicilii non posse dictum crimen 
comm1ttere m Imperatorem ._ et propter hoe cassavit I
mperatoris sententiam contra Regem 
~obe~~ lat_am. [ .. . ] Nam dicta verba, "commisit crimen laesae m
aiestatis in eum", possunt 
mtelhgt sic: m ~um, i~ est, in_ ipsi~s Prin~ip!s maiesta
tem. et tune est vera sententia Papae. 
Nam cum non sit subd1tus ma1estatJ Impenah, non debe
bat sibi illas fidelitates quae dicuntur 
supra [I. Ad_r~prim_endum_] ci~ca pri?cipium, nee eum tenetur revereri. Se
d quo ad eum Iiber 
est et non d1c1tur drrecto m ems ma1estatem committere
, ut [D.49.15. 7j.' For a discussion of 
the I. Non dubito, see eh: 2, PP: 48-9. Baldus, at C. 1.3(6).32(31), ch
ooses to justify this part 
of the papal argument m a different way. Since his 
thought on the matter is brief and 
undeveloped, I have omitted it here. 
149 
i/ 
I 
Bartolus then poses the separate question whether Robert, given that he cannot 
1 f h · 210 commit the crime of lese majesty, can nonetheless become a rebe o t e emp1re. 
This is an odd distinction since lese majesty, as we have seen, encompasses 
rebellion. Indeed, the kind of rebellion undertaken by Robert - acts of war against 
the empire - clearly falls under the category of perduellio. It is therefore difficult to 
see how one could meaningfully separate rebellion from lese majesty. To confuse 
matters further, Bartolus tacitly reintroduces lese majesty as the crime under 
discussion in the considerations that follow. We will return to this problem shortly. 
For now, let us follow Bartolus' reasoning. 
Initially, Bartolus points to the problematic status of Robert. The latter is not one of 
Henry's subjects (subditus) by reason of origin or domicile, as the pope claims, but 
arguably he is one of Henry subjects by reason of the fief (the county of Provence) 
which he holds from the empire.211 Bartolus leaves this specific question for now and 
asks again: can the emperor proceed against a person from the lands of the church 
who has engaged in the kind of hostilities committed by Robert?212 Bartolus argues 
that the emperor can indeed do this, since the people of the church's lands are an 
allied people ('foederati ') and, as the l. Non dubito explains, imperial authorities 
have jurisdiction over such peregrini when they visit the lands of the empire; if they 
commit crimes on imperial soil, as Robert did, then the emperor may launch a 
prosecution ratione delicti.213 But it is also true, as the pope states in Pastora/is cura, 
that the emperor could only have passed a definitive sentence on Robert and 
implemented the appropriate penalty if Robert had been in his power or had been 
extradited to him by the pope.214 What Clement's decretal does is to follow the rule 
of Roman law that a definitive sentence cannot be passed on an absent defendant in 
210 Bartolus, Qui sint rebelles, p. 289b 'et hoe de crimine laesae maiestatis. An autem sit 
rebellis?' 
211 Ibid.: 'Videtur quod non, quia non est subditus, ut dixi. In contrarium videtur in 
[Clem.2.11.2], ubi Imperator appellat Regem Robertum hostem et rebellem quia subditus est 
ratione comitatus Provinciae, quam tenebat ab Imperio.' 
212 Ibid.: 'Sedan contra hominem de terris Ecclesiae sic delinquentem possit Imperator, vel 
habens auctoritatem ab eo, procedere?' 
213 Ibid.: 'Quod videtur, quia homines ecclesiae sunt Imperii foederati , ergo apud eum 
possunt fieri rei et in eos animadverti, [D.49.15.7,2], maxime cum eorum forum s?rtiant~r 
ratione delicti commissi, in immittendo gentes contra Principem ad terras Impem. et sic 
Imperator fecit contra dictum Regem Robertum.' On ratio delicti, see eh. 2, p. 49. . 
214 Ibid.: 'sed Papa determinavit contrarium, nisi sit inventus vel ad eum ex more rem1ssus, ut 
[Clem.2.11.2], § "nos quoque", ibi, "quod si punitio".' 
150 
capital matters.
215 
Ad reprimendum, which removes this obstacle in cases of lese 
majesty, does not apply to Robert since this law was promulgated after Henry's 
imperial coronation and therefore does .not bind the church's subjects (though it 
binds all others!). This is justified by the idea that Henry's coronation should be 
interpreted as a treaty between empire and church, by which the emperor obligated 
himself not to impose harsher laws on the church's subjects than were already in 
existence at the time when the treaty was made. 216 At this point, Bartolus' thought is 
as ingenious as it is sophistic. 
There seem to be two underlying problems with which Bartolus is struggling. It is 
difficult for him to provide a comprehensive explanation, one that solves the 
problems thrown up by the complicated jurisdictional patterns of medieval Europe, 
with the relatively neat categories of Roman law and imperial administration in the 
late-antique Corpus iuris. Robert is somehow both a foreigner and not a foreigner: he 
holds the imperial fief of Provence but his domicile and origin lie beyond the empire. 
In the end, Bartolus decides that Robert is a peregrinus from the lands of the church 
despite his imperial feudal holdings; he accepts Clement's and Oldradus' logic that 
Robert is a foreign landholder but not a citizen or subject. The lack of clarity over 
Robert's status is compounded by the fact that the civil law tradition, at least after 
Pastora/is cura, lacks the vocabulary with which to describe those criminal acts of a 
foreigner (whose origin and domicile is not in the empire) that would come under 
Iese majesty if he were a citizen or a subject: such a foreigner cannot commit Iese 
majesty, but nor is being a foreign enemy a separate delictual category in Roman 
law; as a result, Bartolus introduces the awkward distinction between Iese majesty 
and rebellion, and argues that Robert commits the latter without committing the 
former, even though one is actually a subcategory of the other. This move looks 
particularly odd given that Bartolus earlier explained that the whole point of the term 
rebel/is is to demarcate the domestic hostis. One way to make sense of this part of 
Bartol us' thought is to introduce a scholastic distinction found in Baldus' work. Only 
a citizen or subject can commit acts of Iese majesty or rebellion properly speaking 
('proprie'), whereas a foreigner, such as Robert, can only commit them improperly 
215 
Ibid., ~P· 289b-90a:_ 'Se~ _ce~~ verum est quad de isto crimine, quad est capitale, non 
posset de mre communt punm nts1 praesens esset et illud consideravit ilia Decretalis ' 216 b" , ' · I id., ~- 290a: _ ~ec obstat [I. Ad reprimendum] quia illa fuit facta post coronationem 
Imperatons praed1ct1. Unde Ecclesia non admittit istam in praeiudicium subditorum suorum. 
~t sic videtur inisse foedus cum eo quod contra homines ecclesiae leges graviores non 
1mponat: Unde ~atio illius capituli [Pastora/is cura] non habet locum in hominibus Imperii 
vel altenus regm, sed solum in hominibus ecclesiae.' 
151 
11 I 
Ill 
11 
I I 
I 
·I• I 
speaking ('large'); m this broader sense, a rebel 1s anybody who disobeys the 
authorities. 217 
At this point in his commentary, Bartolus has established two things. Firstly, Robert 
did commit the crime of rebellion, at least broadly speaking, against Henry. 
Secondly, Henry's sentence on Robert for lese majesty was null for the reasons 
explained above. In the next section of his commentary, Bartolus explains in detail 
the consequences of Robert's rebel status; we find out, that this was acquired ipso 
iure upon the commission of his crimes. It appears, therefore, that Bartolus separates 
judicial proceedings for lese majesty, and the resulting sentence, from the ipso iure 
acquisition of rebel status. This interpretation is also suggested by an earlier section 
of his treatise on Quoniam nuper in which Bartolus lays out three ways in which 
someone becomes a hostis or rebel of the empire. At that point, he is still explaining 
the rationale for Henry's legislation: in particular, the section where the emperor 
remarks that doubts have arisen amongst his subjects concerning the legal position of 
rebels before an official sentence has been passed on them. Bartolus states that there 
can be no doubt that someone is a rebel after having been sentenced as such 
following proceedings for lese majesty ('per sententiam'). Similarly, there is no 
doubt if the emperor (or senate) has declared war on a citizen ('indiceret bellum'), as 
according to the 1. Amissione and the 1. Hostis . Doubt, however, has arisen over the 
ipso iure acquisition of this status, hence the emperor has promulgated the present 
law for the sake of clarification.218 It follows from Bartolus ' thought, given that a 
citizen becomes a rebel ipso iure, that both judicial sentence and the hostis 
declaration, strictly speaking, do not bestow that status but declare its prior 
acquisition with the deed. Nevertheless, the distinction between these three modes is 
important; as we shall now go on to see, there are key differences in procedure as 
well as in the substantive legal statuses acquired. The main dividing line lies between 
a judicial sentence on the one hand, and the hostis declaration and ipso iure 
acquisition of rebel status on the other. 
217 Baldus to D.4.5.5,1, fol. 249ra: 'Item nota quod proprie dicitur rebellis qui primo erat 
civis vel subditus ... et iste proprie incidit in crimen laesae maiestatis ... hoe tamen nomen 
rebellis quandoque ponitur large pro quolibet inobediente ... ' Cf. Canning, Baldus de 
Ubaldis, pp. 121-2. 
. . . 
218 Bartolus, Qui sin! rebelles, p. 287b: 'Condemnatus. Hie nullum ~tt dubmm. n~m. per 
sententiam transit in notorium iuris, quod pro veritate habetur... Item, s1 Imperator ahqmbus 
indiceret bellum, ut [D.49.15 .24] et [D.4.5 .5,l] etc.' 
152 
Before discussing the specific penalties for rebellion, we should examine the section 
of Bartolus ' treatise on Quoniam nuper. that explains the acquisition of rebel status in 
the most general terms. Although the passage is lengthy, it is worth quoting in full: 
But since he loses those things derived from the civil law, what of those things he 
?olds from ca?~n law, that is: from the highest pontiff? I reply that a certain law [ius] 
1~ ~bove the c1v1l law, a certam one below the civil law, and a certain one equal to the 
c.1v1l law. In the first case, by rebellion alone, the natural law, that is, the law of nations 
[ms natural~ vel ge~tium] is not lost, for he retains those [ea], [D.4.5.5]. But someone 
can be deprived of 1t ~eo] by a judicial sentence, as is evident in the case of a person 
condemned to death, smce by that sentence citizenship and liberty are taken away from 
him, and the latter is a natural faculty, as at [D.1.5.4]. 
And. in the same wa~ I consider those things that someone holds from the highest 
P?~t1ff. For by rebelhon alone he does not lose those. For his authority is above the 
c1v~l_ law. Nevertheless, someone can be deprived of that law [illo iure] by a sentence 
leg1ttmately. passed even by a secular judge. I speak of course concerning lay people. 
Wherefore, _if the emperor had passed a sentence on the said king [Robert] as a present 
defendant, 1t wou_ld have been valid, [Clem.2.11.2]. .. But since he passed it on an 
absent defendant, 1t was not valid, as I have explained above. 
Concerning that law [iure], however, which is beneath the civil law as are the laws 
[iura] p~rtaining to the cities and other lords constituted beneath the 'empire, those he 
loses without doubt. For those laws, if they are made by the authority of the emperor 
s~em to be his laws... ' 
And_ concerning that law [iure] which is equal to our civil law, it is not possible de iure 
to give an example, since everything belongs to the emperor except those things which 
the church holds... B~t I presuppose, as is observed de facto, that some kings and 
peoples do not recogmse the emperor as their lord. It should, however, be said that a 
rebel ?oes ~o.t _lose th~se laws [iura] by his rebellion. For by this they lose the civil 
laws [1ura c1v1ha], that 1s, [the rights] of Roman citizens ... just as a person banished and 
c?~dernne~ ~y t?~ ~mpero~ does not_ lo_se those things that are derived from the papal 
c1V1l law [mns c1v1hs papahs] ; and this 1s the case in [Clem.2.11.2]. ' 219 
219 Ibid., p. 290b: '~ed cum perdat ea quae sunt iuris civilis, quid de his quae a iure canonico 
~eu ~ s_ummo Ponttfice habet? Respondeo, quoddam ius est super ius civile, quoddam infra 
1us c1v1l~, quodda~ aeq~ale illi. Primo casu per solam rebellionem non perditur ius naturale 
vel gentmm; ea emm rett?et'. [D.4.5.5] . sed per sententiam potest quis privari eo, ut patet in 
damnato ad mortem, qma 1psa sententia privatur civitate et libertate, quae est naturalis 
facultas, ut_ [D.1.5.4]. et eodem modo dico de his quae quis habet a summo Pontifice. Nam ea 
sol~. rebel hone non perdit. Nam eius auctoritas super ius civile est. sed per sententiam 
leg1tlm~ latam posset qu~s illo iure ~rivari, etiam a iudice seculari; de laicis vero loquor. 
Unde, s1 lmperator pr~tuhsset sententtam contra dictum regem praesentem, valuisset, [Clem. 
~.11.2r .. ta~~n, qma m absentem protulit, non valuit, ut dictum est. De eo vero iure quod est 
mfra . ms c1v1l~, ut_ sunt !ura p~opria ci~itat_um e: aliorum dominorum sub Imperio 
constttutorum, ilia sme dubto perd1t. Illa emm mra, st cum auctoritate Principis sint facta, 
153 
The first thing to observe is that Bartolus establishes a framework of legal and 
territorial spaces that closely follows Roman jurisprudence on the hierarchy of laws 
as well as the general tenets of civilian political thought. Just as the prince cannot 
simply abrogate the laws of nations and nature, so a punitive mechanism of the civil 
law, the acquisition of hostis status, cannot deprive someone generally of these 
higher norms. The prince does not have complete authority over these spheres of 
law, whereas he does have complete control over the ius civile - it is 'his' law - such 
that he can take away from the rebel every 'thing' derived from it. Note the 
objectivity of Bartolus' language here: as was so often the case in the material on 
deportation, the discussion is held in terms of laws, and things derived from laws, 
rather than rights. 
Bartolus goes further than traditional Roman jurisprudence in placing what he calls 
the 'papal civil law' into the sphere of higher norms as well; by this he means the 
temporal law of the lands of the church. His argument reflects the unique position of 
these lands and the authority of the pope in the commentators' thought. Here as 
elsewhere, Bartolus accepts the papacy's hierocratic claims of superiority in respect 
of the empire. The emperor has no authority in the pope's lands, since these form a 
jurisdiction irrevocably reserved to the latter by the Donation of Constantine and in 
his capacity as vicar of Christ.
220 
When we move down the hierarchy to lower civil norms, Bartolus highlights .that 
through loss of the Roman civil law, the rebel also loses, at least de iure, the laws of 
those local political entities constituted beneath the empire. This argument stems 
from the conviction that all political authority within the empire, taken in its 
universal sense, must in one way or another derive from it, whether by privilege, 
concession, prescription, or long-standing custom.
221 As we would expect, Bartolus 
takes into account the realities of his day: it is only logical that the effects of imperial 
banishment cannot be felt in those parts of the empire that are de facto independent 
ipsius iura esse videntur ... De eo vero iure quod est aequale iuri nostro civili, exemplum poni 
non potest de iure, cum Imperatoris omnia sint exceptis his quae habet ecclesia... Sed 
praesuppono, prout de facto observatur, quod aliqui reges vel populi [non] recognoscunt 
Imperatorem in dominum. tamen dicendum est quod ilia iura non perdit rebellis ex sua 
rebellione. Nam ex hoe perdunt iura civilia, scilicet, civium Romanorum ... sicut exbannitus 
et condemnatus ab Imperatore non perdit ea quae sunt iuris civilis papalis, et sic est casus in 
[Clem.2.11.2].' The CQT (Basel) edition erroneously misses an important 'non'; this is 
present in CQT (Venice), fol. 105vb. 
22° Cf. Woolf, Barto/us of Sassoferrato, pp. 72-107; Canning, Baldus de Ubaldis , pp. 30-55. 
221 A good expression of this idea in Bartolus' thought can be found in his commentary to the 
I. Hastes (D.49.15.24), discussed by Woolf, Barto/us ofSassoferrato, p. 25 ff. 
154 
from it: the cities and kingdoms, such as the city of Florence or the kingdom of 
Sicily, that do not recognise the emperor as lord. 
In this part of Bartolus' commentary, we can also observe his distinction between, on 
the one hand, the ipso iure acquisition of rebel status and its declaration (which he 
seems to associate with the term exbannire at the end of the passage), and on the 
other hand, a judicial sentence for Iese majesty. Confusingly, Bartolus appears to use 
this distinction to argue that, whereas the declaration and ipso iure acquisition of 
rebel status do not cause someone to be deprived either of the laws of nations and 
nature or of the papal civil law, a judicial sentence legitimately passed does indeed 
have the power to do so. (Naturally, he restricts this to lay persons, since civil judges 
have no jurisdiction over ecclesiastics.) I would suggest, however, that Bartolus is 
not arguing that the prince, through a judicial sentence, could completely rob a 
person of these higher norms, since this would fall foul of the jurists' precept that 
higher norms are fundamentally immutable and cannot be abrogated by a civil 
authority. Instead, Bartolus means that a judicial sentence has powers of derogation 
in respect of higher norms that the declaration and ipso iure acquisition of rebel 
status do not have. Bartol us' example of a criminal sentenced to death provides a 
good illustration. In Roman law, the condemned first becomes a penal slave (servus 
poenae) until the time of his execution.222 Since he is thus deprived of his liberty, 
which, as the cited law states, 
223 is a faculty granted to all free humans under natural 
law, this sentence constitutes a derogation of higher norms.
224 Slaves, however, are 
not generally deprived of all the provisions of the natural law. 
225 Thus, even a penal 
slave is capable of receiving a subsistence legacy, which, as we saw in the previous 
chapter, constituted a natural right. 226 
It is striking that Bartolus offers no rationale concerning why a judicial sentence for 
Iese majesty has greater powers of derogation in respect of higher norms than the 
222 
223 
On penal sl~very, see Mommsen, Romisches Strafrecht, pp. 495-8. 
. D.1.5.~: 'L1bertas est naturalis facultas eius quod cuique facere libet, nisi si quid vi aut 
mre proh1betur.' 
224 It Id b . · · 
. w?u ~ mterestmg to examme systematically the relationship between capitis 
demmutw maxima and . natural law in the thought of the medieval civilians. To my 
knowledge, a study of this type, which lies well beyond the confines of this thesis does not 
currently exist. ' 
225 
~.g. Inst._ I. I 0, 10 states that slaves are bound by the iura cognationis, which prohibit 
~antal relations between blood relatives; as we saw in eh. 2, pp. 75-8, the medieval civilians 
viewed these as iura naturalia. 
226 
D.3~.8.3: 'Si in metallum damnato quid extra causam alimentorum relictum fuerit, pro 
non scnpto est nee ad fiscum pertinet: nam poenae servus est.. . ' Cf. eh. 2, p. 67 . 
155 
'I 
II I 
I; 
II 
other modes of acquiring rebel status. This is probably linked to the civilians' habit 
of not discussing causa regarding the penalties of the Roman criminal law. As we 
saw in the last chapter, they did not generally feel the need to justify those aspects of 
deportation (such as property confiscation) that infringed upon higher norms, nor did 
they have any rule about the extent to which higher norms could be infringed; the 
authoritative models of the Roman law were assumed to be inherently reasonable. 
Much the same could be said about this case. It was clear to Bartolus that in Roman 
law some penal sentences, such as a death sentence, could derogate from higher 
norms to a much greater extent than the acquisition of hostis status: the rationale for 
this state of affairs did not have be discussed. 
Bartolus opens his discussion of the individual consequences of rebel status by 
applying the full force of the provisions in the Digest concerning those who had 
become hastes and transfugae of the empire: 
Concerning the penalty of rebellion, say that one is inflicted ipso iure because they lose 
all those things derived from the civil law, [D.4.5.5]. Another penalty can be inflicted 
upon them by any person, for he can be killed with impunity, as at [D.48.8.3,6] a~d 
[D.49.16.7], and they can be captured and detained as slaves,_ as _a~ [D.41.1.51]. A th_ird 
penalty is imposed because it is possible to sentence such an md1v1dual so that a capital 
penalty is imposed on him, as at [D.48.19.38,1).227 
· 
Although Bartolus uses a tripartite division, the penalties under consideration really 
separate into two categories. Firstly, there are those that are acquired ipso iure: the 
loss of civil rights, as well as the liability of being killed or enslaved as any other 
enemy of Rome. Secondly, a rebel can be sentenced to death; the final law cited by 
Bartolus states that the punishment for a transfuga is to be burned alive or hanged. 
Once again, Bartolus seems to follow his distinction between a sentence for Iese 
majesty and the ipso iure acquisition of rebel status. While anyone can kill a rebel 
with impunity ipso iure, the latter can only be formally sentenced to death by the 
authorities following proceedings for lese majesty; the benefit to the rebel is that he 
avoids falling into penal slavery. 
What about the rebel's property? If he is condemned by judicial sentence for the 
crime of Iese majesty, then all his goods are confiscated by the fisc, as is usual for 
227 Bartolus, Qui sint rebelles, p. 290a: 'De poena vero rebellium, die ~uod quaedam 
infligitur ipso iure, quia perdunt omnia quae iuris civilis sunt, [D.4.5.5]. Aha poena potest 
infligi eis a quolibet homine, nam potest impune occidi, ut [D.48.8.3,6] et [D.49.16.7], et 
possunt capi et detineri ut servi, ut [D.41.1.51] . Tertio poena imponitur quia potest talem 
damnare ut capite puniatur, ut [D.48.19.38, 1 ].' 
156 
cases of this kind. 228 The same, however, does not apply to the ipso iure acquisition 
of rebel status. In medieval Roman law, ownership (dominium) was divided into 
dominium directum, denoting the feudal lord's ultimate ownership of a thing, and 
into dominium utile, denoting the quotidian ownership of it granted to his vassal.229 
Bartolus begins by arguing that the rebel retains dominium directum because this is 
derived from the law of nations. 230 And even though dominium utile was, strictly 
speaking, introduced by the civil law, the rebel does not lose this either· he retains it 
' ' albeit under the name of dominium directum. The reason is that 'use and similar 
things', with which dominium utile is associated and which constitute the 'effects' of 
,J • • 1 d . d 23 1 uommzum, are a so enve from the law of nations. Having strengthened the 
rebel's legal position, Bartolus proceeds to make a crucial qualification: 
But granted that these [goods] are retained and not confiscated, they are, however, 
conceded to the person who occupies them. The case is in [D.41.1.51]. For if a person 
can b~ killed and captured and held as a slave, much more so does this apply to a 
thing. 32 
In this passage, Bartolus is reiterating established civilian opinion on the property of 
the hostis. As we saw towards the beginning of this chapter, it is by the law of 
nations that one can lawfully acquire legal title over property captured from the 
enemy; the jurists did not differentiate between the foreign and the domestic hostis in 
this regard. The import of Bartol us' commentary here is to highlight that, while 
enemy persons have legitimate dominium over their property by the ius gentium, it is 
by the same ius gentium that this legal title lasts only until the moment at which their 
property is forcibly taken away from them. Bartolus' position may appear 
paradoxical but it is the only logical interpretation of his sources, which state that 
both dominium and the ability to seize enemy property are rooted in the law of 
228 lb"d 'S d . · · 1 .: e puto, SI sequeretur condemnatio per mdicem, et si condemnaretur crimine 
laesa~ maiest~ti~, quod tune omnia publicarentur hodie, per [C.9.8.7/8].' As noted in eh. 2, p. 
56, ~mce Justmian's auth. Bona damnatorum did not apply to cases of Iese majesty, the fisc 
contmued to confiscate the estates of those condemned for this crime. 
~:: Cf. Meynial, 'Notes sur la formation'; Feenstra, 'Dominium and ius in re aliena'. 
Cf. Accursius and Bartolus to D.1.1.5. 
~
31 Bart_olus, (!ui sint rebelles, p. 290a: 'Item quaero, quid de dominio bonorum? Respondeo: 
Illud, SI est directum, est de iure gentium, ergo retinet. sed si haberet utile dominium, illud 
non retinet sub isto nomine, quod puto hie a iure civili inductum. sed quia hoe idem sonat 
quod effectus dominii ... ideo puto quod hoe retineat sicut directum, quia hie effectus dominii 
est de iure gentium, arg. [Inst.2.5 ,6], ubi dicitur usum et similia, . quae sunt de effectu 
dominii, esse de iure gentium.' 
232 Ibid. : ' sed licet retineantur nee publica efficiantur, tamen occupanti conceduntur. Casus 
est in [D.41 .1.51]. Si enim potest occidi persona et capi et haberi ut servus, multo magis res .' 
157 
nations. We should note that the same principle is at work in respect of the rebel's 
person. By virtue of the ius gentium the rebel retains his natural faculty of liberty 
upon the acquisition of rebel status and does not lose it unless he is physically 
captured and thereby enslaved, again according to the rules of the ius gentium, which 
instituted the laws of captivity and postliminium. 
Both here and in his commentary to the 1. Hastes that we discussed above, Bartolus 
claims that when individuals become enemies of the Roman empire - which, by 
extension, includes rebels against the church in its lands - they can be captured as 
slaves and their property can be lawfully taken away from them. While he makes this 
point without qualification in respect of the universal powers of empire and papacy, 
right at the end of his commentary to the 1. Hastes, he limits the impact of this idea at 
least in a local Italian context. 233 Bartolus posits the case of two warring Italian 
cities, such as Florence and Pisa, which do not recognise a superior. To put matters 
beyond doubt, one of them is said to be an enemy of the empire. According to the 
ancient customs introduced by the ius gentium, the laws of captivity and 
postliminium should indeed apply. But according to modem usage and the long-
established custom amongst Christians, these laws are no longer observed in respect 
of persons. No one, therefore, should be taken as a slave and sold in such a conflict. 
In respect of property, however, the old rules are followed, and enemy property 
continues to be ceded to its captors even amongst Christians. Bartolus emphasises 
that this modem custom should be followed.234 
Next, Bartolus considers the effect of rebel status on jurisdiction, that is, on the 
ability to hold public offices: 
Likewise, what of the positions of jurisdiction that such a rebel might hold? I reply that 
jurisdiction is derived from the civil law because it was introduced by the public law, 
as noted at [D .2.1.l] ... That is to say, its origin lies in the law of nations, by which 
there were also kings, [D.1.1.5]. But as many things subsequently augmented 
233 Cf. Woolf, Barto/us of Sassoferrato, p. 199. 
234 Bartolus to D.49.15.24, fol. 228rb-va: 'Ulterius quaero, quid de istis bellis quae fiunt inter 
civitates et civitates, an dicantur invicem hostes ut sit locus iuris captivitatis et postliminii? 
[ ... ] Quandoque est contentio inter duas civitates quae superiorem non recognoscunt, ut inter 
civitatem Florentiae et civitatem Pisana. et pone, ut tollam omnem dubitationem, quod 
quaelibet istarum sit hostis lmperii. certe, de iure gentium antiquis moribus introducto 
deberet esse ius captivitatis et postliminii... sed secundum mores moderni temporis et 
consuetudinis antiquitus observatae inter Christianos, quantum ad personas hominum, non 
observamus iura captivitatis et postliminii, nee venduntur nee habentur servi capti . sed 
quantum ad res, iura ista servamus. Cui consuetudini est standum ... ' 
158 
jurisdiction, it is for this reason said to be part of the civil law, [D.1.1.6]. And granted 
that jurisdiction does not pertain to him who is a rebel as a certain right introduced by 
the civil law, it remains nonetheless as a certain royal power introduced by the law of 
· 235 · 
nations. 
Although the institutional core of jurisdiction - the power of a public authority to 
dispense justice - was first introduced by the law of nations, the public law of Rome 
(as the Gloss to D.2.1.1 indicates)236 subsequently supplemented and adapted this 
original institution in so many respects that today it should properly be assigned to 
the sphere of the civil law. Only royal power maintains a direct link to the original 
institution; D.1 .1.5 (1. Ex hoe iure) declares that by the law of nations 'kingdoms 
were founded'. Bartol us' basic point is that the rebel loses all types of jurisdiction 
except royal authority, which is, of course, another implicit argument for Robert: 
even if the effects of the acquisition ofrebel status extended to Sicily, these could not 
deprive Robert of his kingship. We should bear in mind that at one point in his 
Digest commentary, Bartolus states briefly that 'statutes are derived from the civil 
law', arguing that imperial banniti could not institute them as a result. 237 By civil 
jurisdiction in the passage quoted, it is therefore likely that Bartolus also had in mind 
the statutory powers of northern Italian cities, which lacked royal authority, and 
which could therefore be deprived of these powers ipso iure upon becoming hastes 
of the empire. 
The contractual abilities of the rebel follow the pattern established in Bartolus' 
discussion of his dominium. The rebel remains capable of making contracts derived 
from the law of nations, while he could no longer make civil contracts. Although not 
entirely clear on this point, Bartolus' wording also suggests that pre-existing civil 
contracts are dissolved upon the acquisition of rebel status while equivalent ius 
gentium contracts are not - if, indeed, that is what he means by 'losing' or 'not 
235 Bartolus, Qui sint rebelles, p. 290a: 'Item, quid de iurisdictionibus quas haberet talis 
rebellis? Respondeo, iurisdictio est iuris civilis quia de iure publico introducta est, ut 
notatum [D.2.1.1] ... Origo enim fuit de iure gentium, quo etiam fuerunt Reges, [D.1.1.5]. sed 
quia circa iurisdictionem multa sunt addita, ideo dicitur de iure civili, [D.1.1.6]. et licet in 
eum qui est rebellis non cadat iurisdictio tanquam ius quoddam a iure civili inductum, 
remanet tamen tanquam quaedam potestas regia inducta de iure gentium.' 
236 Accursius to D.2.1.1 (a. v. potest), col. 105: 'Est enim iurisdictio potestas de publico 
introducta cum necessitate iuris dicendi et aequitatis statuendae.' 
237 Bartolus to D.1.1.9, fol. 1 Ora: 'Septimo, quaero nunquid exconimunicati et interdicti 
possint facere statuta. Et dico quod non .. . Idem credo de bannitis imperii, ut dictum statutum 
contra libertatem ecclesiae, § i. cum, nota, non obediunt principi, perdunt ea quae iuris civilis 
sunt, [D.4.5.5]. sed statuta sunt iuris civilis, ut hie, ergo etc.' 
159 
I 
I . 
II ' 
I 
nations. We should note _that the same principle is at work in respect of the rebel's 
person. By virtue of the ius gentium the rebel retains his natural faculty of liberty 
upon the acquisition of rebel status and does not lose it unless he is physically 
captured and thereby enslaved, again according to the rules of the ius gentium, which 
instituted the laws of captivity and postliminium. 
Both here and in his commentary to the l. Hastes that we discussed above, Bartolus 
claims that when individuals become enemies of the Roman empire - which, by 
extension, includes rebels against the church in its lands - they can be captured as 
slaves and their property can be lawfully taken away from them. While he makes this 
point without qualification in respect of the universal powers of empire and papacy, 
right at the end of his commentary to the l. Hastes, he limits the impact of this idea at 
least in a local Italian context.233 Bartolus posits the case of two warring Italian 
cities, such as Florence and Pisa, which do not recognise a superior. To put matters 
beyond doubt, one of them is said to be an enemy of the empire. According to the 
ancient customs introduced by the ius gentium, the laws of captivity and 
postliminium should indeed apply. But according to modern usage and the long-
established custom amongst Christians, these laws are no longer observed in respect 
of persons. No one, therefore, should be taken as a slave and sold in such a conflict. 
In respect of property, however, the old rules are followed, and enemy property 
continues to be ceded to its captors even amongst Christians. Bartolus emphas_ises 
that this modern custom should be followed.234 
Next, Bartolus considers the effect of rebel status on jurisdiction, that is, on the 
ability to hold public offices: 
Likewise, what of the positions of jurisdiction that such a rebel might hold? I reply that 
jurisdiction is derived from the civil law because it was introduced by the public law, 
as noted at [D.2.1. l] .. . That is to say, its origin lies in the law of nations, by which 
there were also kings, [D.1.1.5]. But as many things subsequently augmented 
233 Cf. Woolf, Barto/us ofSassoferrato, p. 199. 
234 Bartolus to D.49.1 5.24, fo l. 228rb-va: 'Ulterius quaero, quid de istis bellis quae fiunt inter 
civitates et civitates, an dicantur invicem hostes ut sit locus iuris captivitatis et postliminii? 
[ ... ] Quandoque est contentio inter duas civitates quae superiorem non recognoscunt, ut inter 
civitatem Florentiae et civitatem Pisana. et pone, ut tollam omnem dubitationem, quod 
quaelibet istarum sit hostis Imperii. certe, de iure gentium antiquis moribus introducto 
deberet esse ius captivitatis et postliminii.. . sed secundum mores moderni temporis et 
consuetudinis antiquitus observatae inter Christianos, quantum ad personas hominum, non 
observamus iura captivitatis et postliminii, nee venduntur nee habentur servi capti. sed 
quantum ad res, iura ista servamus. Cui consuetudini est standum ... ' 
158 
jurisdiction, it is for this reason said to be part of the civil law, [D.1.1.6]. And granted 
that jurisdiction does not pertain to him who is a rebel as a certain right introduced by 
the civil law, it remains nonetheless as a certain royal power introduced by the law of 
nations. 235 
Although the institutional core of jurisdiction - the power of a public authority to 
dispense justice - was first introduced by the law of nations, the public law of Rome 
(as the Gloss to D.2.1.1 indicates)236 subsequently supplemented and adapted this 
original institution in so many respects that today it should properly be assigned to 
the sphere of the civil law. Only royal power maintains a direct link to the original 
institution; D.1.1.5 (l. Ex hoe iure) declares that by the law of nations 'kingdoms 
were founded'. Bartolus' basic point is that the rebel loses all types of jurisdiction 
except royal authority, which is, of course, another implicit argument for Robert: 
even if the effects of the acquisition ofrebel status extended to Sicily, these could not 
deprive Robert of his kingship. We should bear in mind that at one point in his 
Digest commentary, Bartolus states briefly that 'statutes are derived from the civil 
law' , arguing that imperial banniti could not institute them as a result. 237 By civil 
jurisdiction in the passage quoted, it is therefore likely that Bartolus also had in mind 
the statutory powers of northern Italian cities, which lacked royal authority, and 
which could therefore be deprived of these powers ipso iure upon becoming hastes 
of the empire. 
The contractual abilities of the rebel follow the pattern established in Bartolus ' 
discussion of his dominium. The rebel remains capable of making contracts derived 
from the law of nations, while he could no longer make civil contracts. Although not 
entirely clear on this point, Bartolus ' wording also suggests that pre-existing civil 
contracts are dissolved upon the acquisition of rebel status while equivalent ius 
gentium contracts are not - if, indeed, that is what he means by 'losing' or 'not 
235 Bartolus, Qui sint rebelles, p. 290a: 'Item, quid de iurisdictionibus quas haberet talis 
rebell is? Respondeo, iurisdictio est iuris civilis quia de iure publico introducta est, ut 
notatum [D.2.1.1 ]. .. Ori go enim fuit de iure gentium, quo etiam fuerunt Reges, [D.1.1.5]. sed 
quia circa iurisdictionem multa sunt addita, ideo dicitur de iure civili, [D.1.1.6]. et licet in 
eum qui est rebellis non cadat iurisdictio tanquam ius quoddam a iure civili inductum, 
remanet tamen tanquam quaedam potestas regia inducta de iure gentium.' 
236 Accursius to D.2 .1.1 (a. v. potest), col. 105: 'Est enim iurisdictio potestas de publico 
introducta cum necessitate iuris dicendi et aequitatis statuendae.' 
237 Bartolus to D .1.1. 9, fol. 1 Ora: 'Septimo, quaero nunquid excommunicati et interdicti 
possint facere statuta. Et dico quod non ... Idem credo de bannitis imperii, ut dictum statutum 
contra libertatem ecclesiae, § i. cum, nota, non obediunt principi, perdunt ea quae iuris civilis 
sunt, [D.4.5.5] . sed statuta sunt iuris civilis, ut hie, ergo etc.' 
159 
I I 
losing' contracts already made ('promissionibus factis'). 
238 While ius gentium 
contracts may have survived the acquisition of rebel status, it is difficult to see their 
practical use during the time of the ban, given that the rebel could be freely killed 
and enslaved; similarly, as we saw above, it is licit for a private person (though not a 
public official) to defraud the hostis. If we briefly look ahead to the communal ban, 
Bartolus maintains that even though the bannitus was a hostis of his city, contractual 
obligations (with the exceptions of peace agreements) were thereby not automatically 
rendered null. And while the bannitus could not enforce valid contracts during his 
time of banishment, he could claim for them after having received restitution. It 
would not be unreasonable to assume that the imperial rebel is in a similar position 
regarding ius gentium contracts should he ever receive a judicial pardon. 
Although the civilians commonly considered the feudal contract to be based in the 
law of nations and natural law,
239 Bartolus claims that it is nonetheless rendered 
ineffective ipso iure by the acquisition of rebel status. His way around the resulting 
tension is by arguing that this situation is not brought about by the rebel's loss of 
civil rights, but by the fact that it was one of the conditions of the original feudal 
contract that it would not be binding upon the vassal if the lord became a rebel; 
therefore, the vassal is not bound to obey his lord during the latter's banishment. This 
is how Bartolus integrates the established jurisprudence of LF.2.28 c. Domino 
guerram into his exposition of imperial banishment. 
240 If we combine this argument 
with the material on dominium above, the broader implication for Robert's case is 
that, although he does not lose his lordship, his dominium utile, over the county of 
Provence (at least until someone physically robs him of it), his own vassals are, on 
the other hand, not legally bound to aid their lord in his rebellion against the 
emperor. 
Bartolus seems to have been the first and only jurist to analyse in some degree of 
detail the banishment of entire cities. As we saw above, Odofredus makes a passing 
allusion to the idea that a city banished by the emperor suffers capitis deminutio and 
238 Bartolus, Qui sint rebelles, p. 290b: 'Item, quid de obligationibus et promissionibus factis 
his rebellibus? Respondeo, si quidem sunt iuris civilis, ipso iure perdunt; si iurisgentium
, 
tune ea non perdunt, ut dictum est.' 
239 Cf. Canning, Baldus de Ubaldis , pp. 82-6; Pennington, The Prince and the Law, p. 126 ff. 
240 Bartolus, Qui sint rebelles, p. 290b: 'sed in aliquibus promittentes ipso iure non tenentur, 
ut si esset debitor fidelitatis. Nam tune, cum esset rebellis, ei obedire non tenetur, no
n 
tanquam liberatus per rebellionem, sed tanquam ab ipso initio promissione conditionata, 
ut 
sit conditio, si non rebellis superiori sit. Et ita intelligo [LF.2.28 c. Domino guerram] et quod 
ibi notatum.' 
160 
thus a loss of its privileges. Baldus, in his treatment of the medicinal imperial ban of 
the authenticae Item nulla and Item quaecunque, mentions the possibility that a 
whole city might be banished in this way but does not explore the issue further.
241 
Bartolus, in contrast, discusses the issue only from the viewpoint of rebellion and 
Iese majesty. In his Digest commentary he argues that a city, which, as a corporation 
(universitas), is a fictive person distinct from its members, can both commit and be 
punished for the crime of Iese majesty. 242 Bartolus points to the example of ancient 
Carthage, which was put to the plough, and also recalls Henry VII' s sentence on 
Brescia, which the emperor ordered to be destroyed, although he later remitted this 
sentence out of mercy. 
243 In his treatise on Quoniam nuper, Bartolus explains that a 
city can properly be called a rebel if its citizens, or the 'greater part' thereof, rebel. In 
that case, the city, as a corporate body, ipso iure loses the same civil rights as a 
private person (Bartolus particularly mentions jurisdiction in this regard), and the 
individual citizens of that city can themselves be fully treated as hastes. Hence 
anyone can plunder, pillage, and capture property in the territory of the rebel city.
244 
If the city as a whole is captured, unlike a private person, its people is not enslaved; 
this rule was established by Justinian, who, according to one of his Novels, granted 
freedom to the African cities that he recaptured form the Vandals. 
245Furthermore, the 
241 
Baldu_s to a_uth. It~m nul'.a, post C.1.3(6).2, fol. 35rb ('Quinto nota quod ita requiritur 
commumtas s1cut smgulans persona, et ita ponitur in banno. '), and to auth. Item 
quaecunque, post C.1.3(6). I 3, fol. 38vb ('Tertio nota excommunicatum banniri banno 
Imperiali, in quo banno poni potest non solum singularis persona sed civitas, ut [auth. Item 
nulla, post C.1.3(6).2]. '). 
242 
Cavalc~, II band?, p. I 18,_ briefly discusses Bartolus' argument in his Digest 
commentanes that a city corporation can commit crimes and be banished as a result but h
e 
~~es not analyse this issue in detail nor examine Bartolus' commentary to Qui sint rebelles. 
. Bartol~s to D.48.19 .16, I 0, fol. 200va: 'In quo dico sic: quaedam sunt delicta propter quas 
fih.us p~mtur_ p~opt~r patre~, ut in. crimine legis Iuliae maiestatis ... tune isto casu po
test 
u?1~ers1tas ~nmm.ahte.r .pumr~, ut fu1t factum de Carthagine, quae passa est aratrum ... Et eg
o 
v1?1 ~e~tentJa~ . d!ffirutiva.m 1mp~ratori~ J:Ienrici quam dedit contra civitatem Brixiae, qu
ae 
fu~t s1.b1 re?elhs, m 9ua d1cebat 11lam c1v1tatem esse subiiciendam aratro, quam poenam e
x 
m1sencord1a relaxav1t.. . ' Bartolus' commentary is briefly discussed by Betti, 'La dottrina
', 
pp. 43-4. 
244 
Bartolus, Qui si~t rebelles, p. 29Ia: 'Item, an istae poenae cadant in rebellem civitatem? 
[.-) Quaedam propne motu civium vel maioris partis rebellant, et isti sunt proprie rebelles. in 
qmbus, quantum ad ai:nissionem iurium et iurisdictionis, dicendum est idem quod in privat
o. 
9u~ntu~ ~d poenam .mf~rendam per hominem privatum, dicendum est quod quilibet pote
st 
m e.m~ c1v1tate et terntono capere, praedari, comburere, et similia facere, quae consueveru
nt 
fien his contra quos bell um habemus ... ' 
2
~
5 
Ibid.: 'Sed .si capiatur una civitas, cuius populus sit rebellis, an populus efficiatur servus 
s1cut un.us pnvatus captus.7 .Respondeo, non. imo libertatem recipiunt, ut in [Coll. I. I § 
occupatzs (=Nov. I pr.)J , et 1b1 notatum in glossa super verbo "antiquam".' 
161 
city's jurisdiction automatically reverts to the emperor. 246 Only he, and not any 
private person, has the authority to pass a judicial sentence on the city in order to put 
it to the plough; alternatively, he may condemn the entire population to be 
expelled.247 
Before we finish our discussion of Quoniam nuper, let us tum back for a moment to 
reconsider the problem of procedure. We observed throughout this section that 
Bartolus' distinction between a judicial sentence for Iese majesty on the one hand, 
and the declaration and ipso iure acquisition of rebel status on the other, has a 
substantial effect on the rebel's legal position. By a judicial sentence the rebel can be 
condemned to death, resulting in an initial period of penal slavery, and the fisc can 
confiscate his estate. In contrast, the ipso iure acquisition of rebel status (which may 
or may not be formally declared afterwards) only results in a form of capitis 
deminutio media, and the rebel thus retains his freedom and property until he is 
physically deprived of them or until a later sentence for Iese majesty follows. If, in 
order to pass a sentence, the emperor proceeds according to ordinary procedure, even 
if he only holds a summary trial, then the provisions of Ad reprimendum and 
Pastora/is cura apply and the defendant would have to be summoned and be given 
the chance to present a defence. Only after remaining contumacious can the 
defendant be condemned in absentia (although this does not apply to the church's 
subjects, who would have to be present). Given that rebel status is acquired ipso iure, 
a sentence would necessarily be declaratory in nature, and presumably specify the 
prior moment of the rebellious act. The other result of the ipso iure provision is that 
the procedural safeguards of Ad reprimendum and Pastora/is cura cannot apply at 
that initial point when rebel status is acquired, from which time the rebel can be 
lawfully treated as such. 
The ipso iure idea therefore does much to undermine the right to judicial defence. 
Both public officials as well as private persons may make war on an imperial rebel 
even before he has been officially declared as such, let alone condemned for this 
crime in court. Similarly, given that Bartolus separates the hostis declaration from 
246 Ibid.: 'Item, an iurisdictio erit privati capientis? Et respondeo, non, cum iure antiquo illam 
ante iam perdidit, et sic est Principis, et sic tune, cum occupatur, non habet iurisdictionem.' 
247 Ibid., p. 291a-b: 'De poena vero inferenda per sententiam iudicis, dico quod nullus potest 
condemnare civitatem subiiciendam aratro nisi solus Princeps. Nam, licet domus particulares 
sint privatorum, ipsa civitas universaliter considerata est ~rincipis ... sicut Tr?ia et _Carthago 
passae sunt aratrum ex praecepto eius qui tenebat Impenum, [D.7.-4.21]. ~1co etia~ quod 
nee potest cogere homines civitatis captae illam dimittere cum ommbus boms et ad aham se 
transferre, sed sol us Princeps po test...' 
162 
proceedings for Iese majesty, it seems reasonable to assume that when the emperor 
declares war on a citizen he is also not bound by Ad reprimendum and Pastora/is 
cura. In fact, according to Baldus, this · is one of the arguments that could be put 
forward for the validity of Henry's sentence on Robert. Granted that Henry's 
sentence is invalid as a sentence, since this would require a legitimate summons, it is 
valid as a declaration of war ('bellica diffidatio et hostis declaratio'), which is not 
subject to this requirement. 248 Such a declaration of war could also be interpreted as 
a legislative rather than judicial act. Baldus cites the Gloss' explanation of the l. 
Amissione, which, as we saw, states that a citizen can be declared a hostis through 
the promulgation of a special law ('lege lata') designed specifically for this purpose. 
As a legislative act, Baldus goes on, it is not bound by the need for a summons.249 
These are both good points; unfortunately, they remain unelaborated as part of a long 
list of merely potential arguments for Henry' side, and Baldus does not confirm or 
even engage with them when he provides his own solution to the legal problems 
created by the emperor's quarrel with Robert. Baldus' conclusion, which rationalises 
Clement's position mainly through papal plenitude of power, is rather blunt and does 
not engage with Bartolus' more subtle thought on the matter.250 
Although Bartol us significantly undermines defendants' rights, he does not leave the 
suspected rebel without remedies. In the same way that the emperor can declare a 
. citizen to be an enemy, if a citizen feels unjustly suspected for rebellion, he can 
petition the authorities to declare the truth about his status and thus clear his name: 
'for just as a city can lay claim to one of its citizens, by the same reason a citizen can 
declare himself a citizen of his city and consequently not a rebel'. Bartolus does not 
state which authorities he has in mind here, though it would be reasonable to assume 
that these include the imperial court. 
251 Of course, nothing prevents the citizen from 
248 
. Baldus to C. y.s~.8, fol. 76vb: 'praeterea, et ad istud videtur posse responderi. nam 
P_nnceps potest -md1cere bellum etiam _non citato, nee est de consuetudine generali quod 
c1tentur h~stes ... ut [D.49.15.24]. suffic1t, nota, sola diffidatio, ut notatum in [Sext.5.4.1] in 
~ovell~. hcet ergo sententia imperatoris non valeret ut sententia, valet tamen ut bellica 
d1ffidat10 et hostis declaratio.' The latter reference is to Johannes Andreae's Novella on the 
Liber sextus - see p. 144 n. 188 above. 
249 ' 
Baldus to C. 7.53.8, fol. 76vb, fol. 77ra: 'Item, contra notorie delinquentem potest fieri lex 
J}0 sta~tum, ut notatum [D.4.5.5, I], et in legis promulgatione non requiritur citatio alicuius ... ' Ibid., fol. 77ra-b. 
251 
B_artolus, Q~i ~int rebelles, p. 290ra: 'quia, cum talem poenam ipso iure incurrant, an peti 
poss1t pronuncian ~al~m esse rebellem? puto quod sic ... et est expressum [Clem.2.11.2], ubi 
Imper_ato~ p~on~nc1av1t Regem Robertum esse rebellem. Et eodem modo, qui se putaret esse 
de tah cnmme mfamatum, posset petere se pronunciari non esse talem ... Nam sicut civitas 
163 
I 
}I 
11 I 
I 
I/ 
111 
; I 
being defrauded, injured, and even killed in the meantime. If he suffers injury fro
m a 
private person, however, he can take the latter to court, whe
re the issue of his rebel 
status, and consequently the lawfulness of the injury, will be decided by a judg
e.252 
Bartolus seems to have thought that the prospect of subseque
nt judicial punishment 
for unlawful injury would act as a sufficient deterrent to prevent individ
uals 
attacking each other capriciously. Baldus, in his commentary
 to Item nu/la, reports 
the opinion of Innocent IV that a person unlawfully banished b
y a king or by another 
lord can even bring an action for subsequent damages 
against the banishing 
authority; unfortunately, he does not provide further details.
253 
Clearly these remedies were an imperfect way of keeping t
he spirit of Pastora/is 
cura alive. On the other hand, it is difficult to see what Bar
tolus could have done 
given that he had to work with an ipso iure idea according to
 which some suspects 
are definitely guilty before their guilt has been proven in cou
rt. His provision of at 
least some remedies for rebel suspects, however, speaks to the
 conviction with which 
the civilians generally attempted to limit forms of extraord
inary procedure. One 
might claim that the idea of petitioning the imperial authorities
 to clear one' s name is 
inherently flawed given that it was the emperor and his
 court that launched 
proceedings against suspected rebels such as Robert in the f
irst place. To this it is 
possible to reply that the same consideration applies to ordina
ry procedure: how fair 
a hearing would Robert have received had he in fact appear
ed at Henry's court ·at 
Pisa to make his defence? This is simply to point to the absur
dities of a legal system 
that does not separate executive and judicial functions . We can hardly b
lame 
Bartolus for not inventing the separation of powers. 
posset vendicare civem suum, eadem ratione civis dicere s
e civem civitatis illius, et sic per 
consequens non rebellem .. . ' 
252 Ibid., p. 291 a: 'Circa secundum, scilicet de poena quae pot
est infligi a quol ibet homine, 
quaeritur an eum qui est rebellis occulte liceat occidere. 
Et die quod sic. nam ipso facto 
poenam hanc incurrit, ut hie, et in iudicio quo proceditur de
 homicidio liquidabitur an fuerit 
rebellis ... ' 
253 Baldus to auth. Item nu/la, post C.1 .3(6)2, fol. 35ra: 'Et adde quod diffida
tus vel bannitus 
iniuste potest agere contra diffidantem, et damna probabit p
er sacramentum, ut in [X.1.40. 7] 
per Inn[ocentem IV]. ' Cf. Innocent to X.1.40.7, fol. 175ra: 'Item n
obis videtur quod, si 
aliquis rex vel dominus terrae iniuste banniverit, vel diffam
averit, vel diffidaverit aliquem, 
qui multa mobilia et immobilia in eius districtu habebat, qu
ae sunt per alios occupata, quod 
dominus ei tenetur de omnibus amissis, actione quod metus 
causa ... ' 
164 
7. Bartolus' quaestio Lucanae civitatis: A Language of Subjective ius 
As we saw above, Bartolus, when discussing in general term
s the effects of capitis 
deminutio on the rebel in his commentary to Quoniam nuper, predominant
ly uses a 
language of objective ius - of laws and things derived from those laws. Th
is is 
because one of his aims is to explain the relationship betwee
n the Roman civil law 
and other legal spheres (the natural law/law of nations, the papal civil law, 
and local 
law). It is only at times that the perspective shifts to the point of view of 
the rebel 
and the discussion becomes about what the latter retains and l
oses. As a result, ius is 
very difficult to translate in this passage: in a connected grou
p of sentences, or even 
in the very same sentence, ius as the subject can shift from denoting a legal sp
here 
( the ius civile, for example) to a personal quality of the rebel that is retained
 or lost. 
We can observe this particularly well at the end of the passa
ge, where ius becomes 
increasingly subjective in sense. Bartolus begins by discussing civil legal sys
tems 
equivalent de facto to the Roman civil law, but ends by speaking of iura civili
a that 
are the property of Roman citizens and equivalent to ea qua
e sunt iuris civilis. At 
that point, we are moving in the same subjective conceptual and linguistic categ
ories 
as in the law of deportation. So-called things derived from th
e civil law - which, in 
the context of Quoniam nuper, are the ability to make certain contracts an
d to hold 
certain public offices - are iura possessed and lost by Ro
man citizens, and can 
therefore properly be described as their civil rights or rights of
 citizenship. 
The objectivity of Bartolus ' language in his treatise on Quoniam nuper stand
s in 
contrast to the subjectivity of an equivalent section in his quaestio Lucanae civ
itatis 
statuto cavetur (quaestio I in the printed editions), debated at Pisa in 
February 
1342. 
254 
This quaestio is devoted to the topic of communal banishmen
t, and as such 
we will discuss it in detail in the next chapter. 
255 One important passage, however, 
concerns the imperial hostis: 
254 
Its authenticity is confirmed by Bellomo, Quaestiones, p. 427 (no. 1 ). Bar
tolus' Tractatus 
bannitorum/de bannitis appears to be an excerpt from this qu
aestio. Lange and Kriechbaum 
IJ_ie Kommentatoren, p. 727, are less certain: ' .. . der Traktat
 De bannitis, moglicherweise nu; 
em Auszug aus Bartolus' Quaestio Lucanae civitatis.' Their reluctance se
ems to be based on 
Ca_Ia~so, 'Bart~lo ~a Sassoferrato', DBI, vol. 6, p. 653, 
who refers to Diplovataccius' 
opm10n, cont~med m an additio to the tractatus (CQT (Basel), p. 355a
), that the treatise, 
rather than bemg an excerpt from the quaestio, 'est insertus
 de verbo ad verbum per eundem 
Bartolum ' into Lucanae civitatis. Note that the tractatus 
is much less comprehensive and 
contains at least one major textual error (both in the 1588 Basel and the 1596
 Venice edition 
at my disposal) - seen. 256 below. 
255 Ch. 4, p. 253 ff. 
165 
I 
I 
And it remains to be seen what such a banished person loses on ac
count of his 
banishment. To this I reply: you ought to know that a transfuga and hostis of the city 
of Rome loses Roman citizenship [civitatem Romanam] and all the rights of Roman
 
citizenship [iura civitatis Romanae]; he retains only the rights derived from the law of
 
nations [iura gentium], as at [D.4.5.5] and by the argument of [D.48 .19.17].
256 
Here the viewpoint is entirely that of the hostis; the only aim is to show what
 this 
individual retains and loses. The possession of civitas Romana is linked to 
the 
possession of iura civilia, which, as the citation of the 1. Sunt quidam (D.48.19.17) 
indicates, are synonymous with ea ('things') - as we have seen in this chapter and 
the previous one, these things are mostly favourable legal faculties unavailabl
e to 
foreigners. In this passage, we can therefore confidently translate civitas Rom
ana 
as 'Roman citizenship' and its concomitant iura as 'rights'. The same applie
s to 
the iura gentium: these iura are also things (1. Sunt quidam) possessed by a person, 
and once again constitute, for the most part, a set of favourable legal faculties 
- in 
Bartolus commentary to Quoniam nuper, the ability to possess liberty, hold 
dominium, make ius gentium contracts, and exercise royal authority. These iu
ra, 
therefore, can be properly called the rebel's natural rights. 
We observed in the case of deportation that although the jurists repeatedly stated 
without qualification that the deported person loses ea quae sunt iuris civilis 
I iura 
civilia and retains ea quae sunt iuris naturalis I iura naturalia - just as Bartolus does 
regarding the rebel in his commentary to Quoniam nuper - it is also the case that the 
Roman civil law still fully binds the deported person, especially in the sense t
hat he 
remains affected by other punitive measures beyond his loss of civil right
s (i.e. 
confinement to an island and the inability to manumit slaves), which restrict the 
rights ordinarily enjoyed by a free and peaceful peregrinus on Roman territory under 
natural law and the law of nations. As we have seen in this chapter, the same p
attern 
can be observed in the civilians' discussion of the imperial rebel or hostis; in fa
ct, the 
256 Bartolus, Quaestio I (Lucanae civitatis), CQT (Basel) - henceforth, Bartolus, Lucanae 
civitatis - p. 205b: 'Restat videre quae perdat talis exbannitus propter tal
e exbannimentum. 
Ad quod respondeo: vos debetis scire quod transfuga et hostis civitat
is Romanae perdit 
civitatem Romanam et omnia iura civitatis Romanae; retinet tantum
 iura gentium, ut 
[D.4.5.5] et arg. [D.48.19.17].' The equivalent section in the printed editions of Bar
tolus' 
Tractatus bannitorum/de bannitis reads - CQT (Basel), p. 356b / CQT (Venice), fol. 130
rb 
(my italics): 'Restat videre quid bannitus amittit pro tali banno. Ad quod vos debetis sc
ire 
quod transfugae et hastes populi Romani, ultra alias civitates, perdunt civita
tem Romanam 
et omnia iura gentium, ut [D.4.5.5] et [D.48 .19.1 7].' There is an obvious textual error
 here, 
since the idea that the hostis loses the iura gentium is not only contrary to
 Bartolus' own 
thought but to the entire civilian tradition on capitis deminutio media. This g
arbled text led 
Ghisalberti to his faulty interpretation of the hostis - seep. 13 7, n. 163 above.
 
166 
latter suffers additional punitive measures beyond the loss of citizenship that a
re far 
~~rsher ~han those suffered by the deported person - most notably, the liability
 to be 
tnJured m ~erson and property. To Bartolus' credit, he is the first and only jurist to 
confront this conceptual problem. In Lucanae civitatis, having stated that the 
hostis 
loses his iura civilia but retains the iura gentium, he makes the foll
owing 
qualification: 
This is true, name)~, that he loses the iura of the city of Rome in respect of his benefit 
[commo~um], ?ut m respect of his detriment [incommodum], the iura of the city of
 
Rome bmd him on Roman territory, as the Gloss notes in one
 solution at 
[D.~9.15.1 ~".]. And this is also proven because, by making a contract with us, he is
 
obligated civilly, as ~oted by the Gloss at [D.4.5. 7,3]. And this is also clear because the
 
law [Jex] of the city of Rome, which permits them to be killed is binding 
[D.48.8.3,6].257 ' '
 
~et _us. expand the references of this tightly argued passage. Firstly, Ba
rtolus' 
msptrat1on for the commodum-incommodum distinction rests on one of Accu
rsius' 
soluti?ns a~ D .49 .15 .19, 7 regarding an inconsistency in the law on the transfuga and 
guard1ansh1ps. As we saw in our discussion of this problem above, Acc
ursius 
suggests that a transfuga might remain a guardian only to his detriment 
(praeiudicium) but not truly. Although this solution is the basis of Bartolus' 
argument here, we should recall that it was not favoured by Accursius nor ever
 taken 
up b~ · subsequent_ juris_ts, including Bartolus himself, when discussing the specific 
~uestion of guard1ansh1ps; the civilians consistently argued that the latter wer
e civil 
Ill nature and therefore lost upon capitis deminutio media. Secondly, the m
iddle 
reference to the Glos~ at D.4.5.7,3 might seem cryptic, but it is merely a
 very 
contracted way of statmg a well-established principle in the law of deportatio
n: the 
d~portatus, after his sentence, can make ius gentium contracts; he can ob
ligate 
himself, as ~ccu_rsius mentions there. 
258 And even though these contracts only create 
a_ natural_ ~bhga~10n betw~e_n the parties, as Bartolus himself argues, they coul
d give 
nse to c1vI1 a~t10ns a~d htI~ation in Roman courts. In other words, the depor
tatus, 
and by extension the 1mpenal hostis, are bound by the iura civilia of Rome 
in the 
257 
Bartolu_s, Lucanae ci~it~tis, p. 205b: 'hoe verum, scilicet, quod perdi
t iura civitatis 
~o~a~ae mquantum resp1c1t suum commodum, sed prout respicit incom
modum ipsius iura 
c1v1tahs Romanae ~um ligant apud nos, ut notat glossa secundum 
unam soluti;nem 
[D.49.15.19, 7]. et eham prob~tur, quia promittendo nobis, obligatur civiliter, ut notatu
m pe; 
~lossam [D.4.5.7,3]. Et hoe etiam patet quia Jex civitatis Romanae, quae permittit eos o
ccidi 
hgat, [D.48.8 .3,6~.' Mooney 'The Legal Ban', p. 118, mentions this passage in passin
g bu; 
does not develop its arguments. 
258 A . 
ccursms t_o D.4.5. 7,3 , col. 457: 'sed nonne qui solam civitatem amisit ob
ligari potest? 
Respondeo, uhque, cum sit Iiber.' 
· 
167 
I 
;IJ 
sense that they are liable to be sued in Rome's courts for ius gentiu
m contracts. 
Bartolus wants to emphasise the idea that an ex-citizen is subject to the provisions of 
Roman law on Roman territory. The best example is, of course, the law
 cited at the 
end, D.48.8.3,6, according to which a transfuga can be killed as a hostis by any 
private person. 
Through the commodum-incommodum distinction Bartolus attempts 
to provide a 
better, systematic explanation of the way in which iura civilia are lo
st by capitis 
deminutio media. One obstacle to this was the ambiguous nature of th
e Latin legal 
terminology: civitas, ius, ius civile, iura civilia, ea quae sunt iuris civili
s. In fact, it is 
not possible to give a consistent and accurate translation of the word
 iura in this 
passage. The preponderant sense of the first set of iura is subjective. They are 
'things' that a Roman citizen previously possessed and has now lost upo
n his loss of 
citizenship; for the most part, these constitute entitlements, particularly
 in the sphere 
of private law, unavailable to a foreigner. Although Bartolus' charac
terisation of 
them as advantageous is therefore a good rule of thumb, it is, however,
 not entirely 
accurate. The ex-citizen also loses iura that are disadvantageous: the re
quirement of 
holding guardianships and the duties of patriapotestas for children-in
-power. We 
saw in chapter 2 that the ability to hold guardianships was a mediate ob
ligation ( and 
thus a legal entitlement - potestas - in the first instance), while the child's ius 
patriaepotestatis was an immediate obligation (a duty in itself). 
259 As I have 
emphasised in chapters 1 and 2, a person's ius, in legal Latin does not
 only denote 
what we, in modem English, would properly call a right - i.e. an entitl
ement - but 
also an immediate obligation.
260 In the present context, both types of iura are civil 
'things' possessed and then lost by Roman citizens. This means that 
we can only 
accurately translate iura as 'rights' in its first occurrence in the pass
age above if, 
counterintuitively, the word 'right' is allowed to possess this double mea
ning. 
Even when a person's ius refers to an immediate obligation, we should
 nevertheless 
be confident in categorising this ius as a subjective right and in using the terminology 
of rights in our explanation of the jurists' thought. The English word 'right' is closest 
to the Latin word ius and nicely conveys the latter's connection to th
e notions of 
justice and right in an objective sense, i.e. that which is right. One might contend, 
however, as Charles Donahue does (with reference to Michel Villey's argument), 
that ius in this instance is not actually subjective because of its sense of obligation. 
Since this ius does not denote subjective right in the modem sense, it should be 
259 Cf. eh. 2, pp. 71-2 (patriapotestas) and pp. 78-9 (guardianships). 
26° Ch. I , pp. 25-6; eh. 2, p. 46-7. 
168 
placed into the realm of objective right and interpreted as someone's 'due' - that 
which justice bestows upon that person in the given situation.
261 If we take the 
argument to its logical conclusion, given that a person 's ius is not alw
ays his right 
(though it can be), we should translate Ulpian's famous description of justice not as 
'the constant and perpetual will to render to each his right [ius suum ]', but as ' the 
constant and perpetual will to render to each his due '.
262 On one level, this line of 
reasoning is perfectly valid. My ius is, of course, always my due. It is 
what the law 
and justice in any given situation grant to me or demand from me. Nevertheless, as 
should have become clear from the evidence presented in this and 
the previous 
chapter, the problem with Donahue' s position is that it does not do justice to the fact 
that in the sources ius, as immediate obligation, is described and conc
eptualised in 
highly subjective terms. In the passage quoted above, it is thus difficult to see how 
one could separate the son's ius patriaepotestatis from the other iura, an
d accurately 
describe it as 'objective due' or 'that which is just' without losing the sense of the 
Latin; such a perspective cannot properly take into account the concept
ion of iura of 
this kind as things possessed or lost by a person. 
In all this, it would be amiss to draw strict dividing lines between 
the different 
subjective and objective categories of ius. 263 While it is possible to accept that ius 
signifies objective due on one level, on another level this objective sense dissolves 
into two parts: what is due to me (my personal entitlement) and what is due from me 
(my personal obligation). To employ a scholastic distinction: one might say that ius 
in the objective sense as due is the genus, whereas ius as personal entitlement and ius 
as personal obligation are two separate species. Rather than exclude each
 other these 
' 
three different senses mutually imply each other. But on this second lev
el, where ius 
denotes personal entitlement or personal obligation, it often becomes
 difficult to 
translate this word as 'due': given the personal nature of these iura, it 
makes much 
better sense to translate them as rights, as long as we reme
mber, very 
counterintuitively, that a right can also be an immediate obligation. If w
e can accept 
261 Cf. eh. 1, pp. 25-6. 
~
62 
D:1.1. 10: 'Iustitia est c~nstans et perpetua voluntas ius suum cuique
 tribuendi.' That 'ius' 
m this senten~e should be mterpreted and translated as someone's 'd
ue' and not 'right' was, 
of c~urse, V1lley's contention, since, in his view, ius did not den
ote subjective right in 
cl~ss1cal .Roman law. See Villey's wor.ks cited in eh. 1, p. 23, n. 
74, esp. his 'Suum jus 
cmque tnbuens'. 
263 A . d . h ( 
s mentlone m c . 1 p. 25, n. 80), Kaser ('Zurn "Ius"-Begriff der Romer', p. 69), 
has 
noted the flui?ity of ius in classical Roman law, in that ius could be 
subjective and objective 
at the same time. My argument is a broader development of that p
oint with regard to the 
medieval civilian tradition. 
169 
I 
J!J 
!I 
:./ 
II 
that, then we could translate Ulpian's dictum equally as 'to render to each his due' or 
'to render to each his right'. One translation takes the objective perspective, the other 
one the subjective perspective, but fundamentally they mean the same thing. 
Returning to our passage, we can see that the preponderant meaning of the second set 
of iura (those that are detrimental and are retained by the hostis) is that of law(s). 
This is why Bartolus, at the end of the passage, employs the word lex, the specific 
term for law. The law of Rome, the Roman legal system, is still binding and 
applicable to ex-citizens in two ways. Firstly, it is binding in the sense that it applies 
to all peregrini on Roman soil. Hence the deportatus can be sued in the courts of 
Rome. But note that once again, while Bartolus' rule of thumb is generally good, 
strictly speaking, the deportatus does not only derive disadvantages from the civil 
law. The deportatus might be sued, but he himself can equally sue another for ius 
gentium contracts in Roman courts using civil actions. Bartolus uses this rule of 
thumb because of the second sense in which the laws of Rome still bind ex-citizens. 
Since the deportatus and the hostis are criminals, the Roman law establishes a 
number of punitive sanctions not suffered by an ordinary peregrinus. Especially for 
the hostis, that is his predominant experience of the Roman law on Roman territory. 
In other words, to explain the dynamic behind the loss and retention of iura civilia in 
the case of deportation and imperial banishment, Bartolus divides ius between its 
subjective and objective senses. The subjective sense is associated with the benefits 
(and, to a lesser extent, liabilities) of Roman citizenship, and the objective sense 
denotes the law, the Roman legal system as a whole, whose provisions, especially its 
punitive provisions, are still binding on the former citizen. We should be aware, 
however, that in legal Latin this division is hardly ever neat. As we have just seen, 
the term ius encompasses a range of concepts that we, in modern English, would 
separately call rights, duties, justice, and law. To make sense of this, it might help to 
imagine the concept of ius as a fluid continuum on a vertical plane, where the 
subjective sense is at the bottom, and the objective sense at the top. !us as right, in 
the sense of personal entitlements and obligations, would be at the lowest end, and 
ius as justice at the highest. The point where they connect in the middle is the notion 
of ius as law (whether positive or natural) and, slightly below that, the notion of a 
person's objective due. Justice determines the content of the law, the law determines 
the content of someone's due, and someone's due dissolves into personal 
entitlements and obligations. One of the reasons why the jurists could so easily 
switch between one sense of ius and another, and why sometimes ius is so 
ambiguous and implies a number of senses, is because of this interconnectivity 
170 
between the subjective and the objective idea. It is impossible to have one without 
the other: ius as personal entitlement and obligation is a function of ius as objective 
due, which is a function of ius as law, which is a function of ius as justice. This is 
why at times the jurists are happy to present the effects of capitis deminutio media in 
unambiguously objective terms, such that one cannot translate them otherwise. 
Hence, we are left with the awkward phrase in English of 'losing' or 'retaining' a 
law or laws. The point is that for the medieval civilians it did not fundamentally 
matter whether one spoke subjectively or objectively. Loss of the civil law implies 
loss of the rights granted to a person by the provisions of the civil law. Loss of civil 
rights implies that the provisions of the civil law that grant these rights to a person 
are no longer applicable. That is the beautifully and frustratingly flexible nature of 
ius. To translate that word properly in English, we would require a word that denotes 
all these different senses at the same time. 
8. Further Conclusions 
i. The Contents of Civil and Natural Rights 
The civilians' discussion of the hostis, especially Bartolus' treatise on Quoniam 
nuper, largely confirms the contents of civil and natural rights established in the last 
chapter. In the context of deportation, the jurists portrayed Roman citizenship largely 
as a bundle of entitlements, with a small number of obligations, in the private-law 
spheres of inheritance, family, and contracts. Most of these - testamentary powers, 
patriapotestas, and civil law contracts - reappeared, if only briefly, in their 
discussion of the hgstis; the civilians also affirmed the idea, only indicated in the 
material on deportation, that the ability to hold guardianships was derived from the 
civil law. Bartolus' treatise on Quoniam nuper, however, is unique in pointing to an 
important public element of Roman citizenship that was otherwise left unmentioned: 
all types of jurisdiction, except those pertaining to royal authority, were civil in 
nature, and thus only Roman citizens possessed the ability to hold and exercise 
public offices; as a concrete example, Bartolus mentions the capacity of making 
municipal statutes. For Bartolus, therefore, just as Roman citizenship was an 
overwhelmingly advantageous attribute in private law, so too it was in public law, 
for it is difficult to construe as anything else the blanket ability to hold public offices. 
171 
I I 
When considering the hostis, the civilians also confirmed that liberty, do
minium, and 
ius gentium contracts were natural rights; according to Bartolus, bo
th dominium 
directum as well as dominium utile were ultimately rooted in natura
l law. Two 
natural rights appeared in this context that were not mentioned in the d
iscussion of 
deportation: firstly, the right to hold and exercise jurisdiction connected to royal 
authority; secondly, the right to be summoned to court in capital procee
dings for the 
purpose of presenting a defence. 
ii. Derogation from Higher Norms 
In the previous chapter, we observed the casuistry with which the civilia
ns permitted 
higher, natural norms to be suspended by deportation, a punitive mech
anism of the 
civil law. The same casuistry is evident in their discussion of the hostis. 
The fact that 
the latter can be killed with impunity is a clear derogation from the
 natural law 
recorded at D.1.1.3, that one human being should not encompass the lif
e of another. 
The enslavement of the hostis takes away his natural faculty of liberty
, and seizing 
his property violates his natural rights of dominium. Yet the jurists never felt the 
need to justify these infringements; they assumed the presence of just cause in this 
context. This is particularly clear in Bartol us' treatise on Quoniam nuper, where he_ 
explains, again without any justification, that a judicial sentence of the emperor has 
greater powers of derogation in respect of higher norms than the hostis
 declaration. 
As we saw, this is a fair exposition of the Roman law, which, in th
e eyes of the 
jurists, implicitly provided iusta causa such that there was no need for a discussion 
of that principle. 
One interesting if somewhat paradoxical feature of the process of dero
gation in the 
context of the hostis is that, on one level, it is not the civil law that is th
e derogating 
agent. The civil law, strictly speaking, through its procedural mecha
nisms - the 
declaration and ipso iure acquisition of hostis status - only places a
 person in a 
position, i.e. that of the hostis, whereby the ius gentium itself derogates 
from aspects 
of the ius gentium or the natural law. (As we have seen, the jurists did not 
differentiate between the two in the context of exile.) By the natural law and the law 
of nations the hostis possesses his liberty and dominium, but it is als
o by the ius 
gentium that these things can be taken away from him. This is bec
ause the ius 
gentium regulates the law of war as well as the laws of captivity and post
liminium. 
172 
As we concluded in chapter 2, the casuistry with which the jurists allowed the civil 
law to derogate from higher norms means that it is impossible to lo
cate a fixed 
boundary between civil and natural right in medieval civilian thought. A
s a result, it 
is equally impossible to find any measure intrinsic to the natural rights w
e have been 
discussing that could determine the extent of their potential violability. 
While higher 
laws - the ius naturale vel gentium - might be immutable in princip
le and thus 
unable to be taken away by human ordinance, the rights they conf
er upon the 
individual are not, since the civil law only permits them to operate to th
e extent that 
it sees fit. The appropriateness of the restriction is measured by some oth
er yardstick: 
in the case of capitis deminutio media, the accepted models of the Rom
an criminal 
law, established by ethical, social, or political norms that at one poin
t found their 
way into the legal system. When the jurists did discuss cause elsewhere their ' 
considerations are very revealing in this regard. As we saw in chapter 1,
 the civilians 
would allow the civil law to derogate from higher norms if they consid
ered the case 
to be reasonable, just, and directed towards public utility, so long as it did not violate 
their notions of decency and morality. But if these external criteria were 
not fulfilled 
' 
the civil law lost this authority.
264 Notions of reasonableness, justice, public utility, 
decency, and morality, however, are conditioned by their specific
 social and 
historical contexts; they are, and will always remain, fluid and open t
o (radically) 
diverse interpretations. 
ii. The hostis in the Middle Ages 
In his 1932 treatise The Concept of the Political, Carl Schmitt drew inspiration from 
the Roman law tradition of the hostis. The state, he argued, must be ab
le to declare 
the internal enemy if it is under attack from within. This also ap
plies to the 
constitutional Rechtsstaat, whose legal norms are only binding in norm
al situations, 
but are suspended in states of emergency: when the constitution itself is 
under attack, 
'the battle must then be waged outside the constitution and the law, henc
e decided by 
the power of weapons'.
265 Schmitt's reading of the hostis in Roman law portrays the 
latter as a fundamentally extra-normal and extra-legal figure. While this 
is probably a 
264 Ch. 1, pp. 27-8. 
265 
Schmitt, The Co~cept of_the _Political, pp. 46-7, at p. 47 (quoting Lon;nz von Stein). In a 
Ion? foot~ote _to this section m the original German edition, Schm
itt claims to provide 
vanous h1stoncal examples of the internal enemy declaration. H
is references include 
M?mmse?'s R?misch~s Staatsrecht and Eichmann's Acht und Bann, w
hich are supposed to 
evidence its existence m the Roman and German traditions of public la
w. 
173 
fair assessment of the classical tradition, it does contrast with the way in which the 
concept of the hostis was received in medieval civilian thought. 
For a long time, the hostis did not provide the general analogy with which the 
medieval bannitus was understood. Before Bartolus, the jurists looked to ordinary 
Roman criminal law, particularly annotation and deportation. Bartolus' subsequent 
establishment of a general equivalence between bannitus and hostis resulted in a 
modification of the classical concept by which the hostis was moved away from the 
sphere of the exceptional and connected to the realm of the ordinary, such that the 
severity of the original idea was considerably attenuated. Rather than being extra-
legal, the imperial hostis-bannitus was embedded within higher, non-civil legal 
norms. Bartolus thus created a fairly detailed framework that defmed the position of 
the hostis by outlining both his rights as well as his liabilities. Ironically, in the 
medieval legal tradition, outlawry did not place the outlaw beyond the law. In terms 
of procedure, Bartolus and the other civilians held fast to the idea that the validity of 
a judicial sentence in capital trials depended on the defendant being granted his 
natural right to be summoned and to present a defence. Of course, this natural right, 
as indeed any other, was never fully safe from an argument that the prince could 
suspend it ex iusta causa. The idea of the declaration and ipso iure acquisition of 
rebel status also did much to undermine the natural right to a defence. But even in 
this case, Bartolus limited the exercise of arbitrary power by providing the innocent 
defendant with remedies that were reasonable in their historical context. 
It seems likely that these remedies, limited though they were, constituted an 
expression of the evident value that the jurists placed on procedural safeguards, 
particularly on the opportunity to present a defence. This was probably reflective of a 
wider medieval mindset concerning the legitimacy of judicial proceedings that was 
itself influenced by the ideals of civilian jurisprudence; in practice, as we have seen, 
the imperial court made sure to summon defendants even in political cases. The fact 
that the vast majority of the relevant material in the Corpus iuris describes normal 
Roman criminal law - which was of course the main source of the civilians' 
procedural ideals - must have been another factor contributing to the medieval 
modification of the hostis-transfuga. Since only a tiny number of texts deal with the 
latter, the jurists had very little authoritative law with which to describe this figure 
and develop exceptional procedure along classical lines. Indeed, it seems no accident 
that it took until a jurist of Bartolus' calibre to create a comprehensive theory of the 
internal hostis. In this area of legal thought Bartolus certainly vindicated his position 
as 'prince of the jurists'. 
174 
Chapter 4 
The Communal Ban 
1. Introduction 
In the previou~ _two c~apters, we examined the legal mechanisms by which a person 
los~ Roman c1ttzensh1p, or citizenship of the Roman empire, and retained rights 
denved ~rom th~ ~~tural law and the law of nations. According to the legal theory of 
the med1ev~l c1v1hans, de iure the Roman empire extended over the entirety of 
weste~ Christendom, although it was acknowledged that de facto those kings, lords, 
and c1~1es beyond the effective realm of the German rulers no longer obeyed the 
authority of the Roman emperor. In this chapter, we will examine the loss of 
citizens~i~ of_ local political entities within the Roman empire, that is, the citizenship 
of the_ cities m central and northern Italy, which were not only theoretically, but 
sometm~es also ac~ally, subject to the German emperors during our period. The 
mechamsm b_y _which a person lost the citizenship of an Italian city was the 
co_m~unal cnmmal ban (bannum pro maleficio) in its most severe form, whose 
pnnc1pal characteristics, as we saw in chapter I, mirrored those of the imperial ban. 1 
As shall become clear in this chapter, one important exception was that the effects of 
the com~unal ban were only felt within the territory of the city from which someone 
was bamshed; the communal ban was local, the imperial ban universal. 
The ~i~ilians were well aware that the term bannum had a number of meanings in the 
municipal context, inc_luding 'order of a public authority' and 'monetary fine', and 
tha~ t~e ban, as_ a s~nct10n ~mployed against contumacious defendants, was subject to 
vanat10ns, dependmg, for mstance, on whether it was issued in criminal or in civil 
cases.
2 
The criminal ban for the greater crimes typically allowed the city's fisc to 
1 Ch. l,pp.18-23. 
2 
Baidu~ to D.12.6.19,_ fol. 67ra: 'Dicitur autem bannum quatuor modis. Primo, idest, vox 
praecoms ... S~cunda, 1dest, mulcta .. . Tertio, annotatio bonorum ... Quarto idest exilium 
unde_e;ules d1cuntur banniti ... ' _Al~e~ic_us de Rosate, Quaestiones statutor~m 1v,' q. 1, fol'. 
6~~~- .. . ?ann~~ est quaedam iud1ciahs sententia poenalis contra contumacem prolata in 
c1v1h vel m cnmmali ... ' Cf. Albericus de Rosate to C.6.24.1 (p 239 n 238 b I )· B rt I 
toD.1.18.6,9. . , . eow , aous 
175 
111 
confiscate the estates of banniti and permitted private individuals to in
jure them in 
person and property, even kill them, with legal impunity. The vast ma
jority of the 
juristic discourse on banishment concerns this type of ban; it was what the civilians 
referred to when they spoke of 'the banished persons of our time' (banniti nostri 
temporis).3 
Unlike the medieval civilian discourses on deportation and the impe
rial ban, the 
jurists' thought on the criminal ban of the Italian cities has been subject to a small 
number of detailed studies of legal history, the most comprehensiv
e of which 
remains Desiderio Cavalca's ll bando nella prassi e nella dottrin
a giuridica 
medievale, published in 1978. 
4 Taken together, these studies cover much of the 
civilian discourse and, for the most part, provide a fair, if at time
s imprecise, 
exegesis of juristic theory. For this reason, readers familiar with the modern literature 
will already have encountered a good amount - though by no means
 all - of the 
source material discussed below. The aim of this chapter is not so muc
h to critique 
existing accounts, but to reassess the civilian tradition on communal
 banishment 
from the perspective of political thought, using the same method and 
pursuing the 
same themes as in chapters 2 and 3.
5 We will continue to explore the process through 
which a citizen became an outlaw and examine the personal legal prope
rties lost and 
retained as a result. This conceptually focused analysis, with emphasi
s on judicial 
procedure, citizenship, civil rights, and natural rights, as well as the lega
l norms from 
3 In the words of Nello da San Gimignano, Tractatus de bannitis, fol. 1
12ra-b: 'Et ut huius 
materia opusculi clarius dignoscatur, premittendum putavi plures ess
e species bannitorum .. . 
Quandoque diffinitive ponuntur in banno ex forma statutorum, quibus cavetur q
uod in 
criminalibus contra absentes contumaces possit ad diffinitivam senten
tiam perveniri, et quod 
citati certo modo non comparentes habeantur pro confessis... de h
is ultimis bannitis, qui 
dicuntur banniti nostri temporis, erit huius tractatus materia princip
alis ... ' Cf. Ghisalberti, 
'La condanna' , pp. 30-2; Mooney, 'The Legal Ban', pp. 2, 17, 29
, 88, 167; Cavalca, II 
bando, pp. 17-22, 42, 184. 
4 This followed a single article, by Ghisalberti, 'La condanna' 
(1960). Two recent 
contributions are Milani, 'Giuristi, giudici e fuoriusciti ' (2007), and Vallerani, 'D
iritti di 
cittadinanza' (2013). Mooney's 1976 PhD dissertation, 'The Legal Ban', although f
ocused 
on Florentine statutory law and Nello da San Gimignano's fifteent
h-century Tractatus de 
bannitis, is interspersed with some limited discussion of legal theo
ry from our period. 
Pazzaglini, The Criminal Ban, a good exposition of the Sienese statutes
, also contains a little 
civilian material from the thirteenth century. 
5 It should be acknowledged that two existing studies of political thoug
ht briefly touch on the 
communal ban. Some aspects of Bartolus' theory are mentioned 
by Woolf, Barto/us of 
Sassoferrato, at p. 118 and pp. 200-2. Canning, Baldus de Ubaldis, pp. 127-8, uses Baldu
s' 
view of banishment as a case study of the territorial limitation of city 
statutes. 
176 
which rights were derived, will demonstrate the precise character of t
he legal and 
territorial spaces occupied by banniti. 
i. City and Commune 
As is the case here, modem scholarship generally refers to the crimina
l ban of the 
Italian cities as the 'communal ban'. This label derives from historia
ns' habit of 
using the term 'commune' as a synonym for 'Italian city' in this pe
riod, but also 
from the fact that it was normally the podesta, the city commune's chi
ef executive 
and judicial officer, exercising the jurisdiction established by the commune's 
statutes, who issued the criminal ban. This designation, however
, might be 
considered imprecise or even misleading given the fractious natu
re of civic 
government for much of this period. The rise of factions within the 
citizen body, 
violent conflict, mass expulsions, and the partisan monopolisation of po
litical power 
meant that 'commune' was a contested concept; one should not as
sume that it 
referred to the city as a collective in any straightforward way.
6 Just how complicated 
the situation could be is illustrated by an agreement reached at Man
tua in 1208 
between Cremona, Verona, Azo of Este, and the Counts of Sanboni
facio: which 
referred to certain obligations of Verona to 'the knights of Brescia
, who have 
withdrawn [i.e. been expelled] from Brescia and whom we take to be the commune of 
Brescia'.
7 This example comes from a world in which the rule of the commun
e, 
associated with the old noble elites, was challenged by the rising popula
r faction (the 
popolo) in many Italian cities during the course of the thirteenth century. The 
formalisation of popular power meant that, strictly speaking, the com
mune became 
only one of several governmental and legislative institutions with
in the city. 
Consider the following consilium by Cinus de Pistoia ( d. 1336/7), in which he 
commented on the institutional relationship between commune and
 popolo in 
Florence: 
Although in the ius commune the term 'people' [populus] also includes n
oble 
!11agnat~s, however, according to the customary law of almost a
ll Italy, the term 
people only refers to the commoners. For this reason, when we sp
eak of 'commune 
and people' , the term 'commune' in fact signifies the magnates, and 
the term 'people' 
6 
A point convincingly made by Ryan, 'Bartolus of Sassoferrato', 
pp. 79-83, where the 
following examples can be found. 
· 
7 
Fick.er, F?~schungen, IV, pp. 266-8, at p. 268: ' ... et militium Brixie, 
qui exierunt Brixie, 
quod mtelhgtmus esse commune Brixie.' 
177 
merely the commoners. Hence, there are different councils, classes of officials, and 
types of legislation in the city. For the council and legislation of the commune are one 
thing, and the council and legislation of the people are another.
8 
According to Cinus, the commune and popolo are only the two separate parts that 
make up the city as a legal whole - in the background here is the example of ancient 
Rome, which had consisted of the Senate and People of Rome. 9 But even though the 
ban was normally issued by a communal officer, I have seen no evidence in the legal 
practice of the Italian cities nor in civilian theory to suggest that it ever referred to ( or 
was meant to refer to) anything but the city in its entirety; a citizen banished by the 
podesta, was banished from the city, not just from the commune. In other words, the 
commune's statutes on banishment had a universal application within the city, and 
the podesta represented the whole community in this matter. Indeed, one of the 
arguments to which the jurists occasionally referred in order to justify the punitive 
measures associated with the ban, is that the bannitus, through his contumacy, has 
injured the respublica itself, i.e. the city as a collective whole, because he has injured 
the magistrate who represents the respublica. 10 Bartolus is even clearer when he 
compares the prerogatives of the medieval commune to those of the ancient Roman 
senate: just as the Roman senate once had the power to declare Rome's citizens to be 
public enemies (hastes) of Rome, today the commune of an Italian city, through the 
criminal ban, can declare the citizens of that city to be the city's hostes.
11 
8 Rome, Biblioteca Apostolica Vaticana, MS. Chigi E.VIII, fol. 137a: 'Licet de iure 
communi appellatione populi etiam magnates nobiles complectantur, de iure tamen 
consuetudinario quasi totius ytalie appellatio populi non continet nisi plebeios. Et hie venit 
quando dicimus comune et populo, quod appellatio Communis significat etiam magnates et 
populi plebeios tantum. hinc distincta sunt consilia civitatum et ordines [ et] decreta 
decurionum. nam aliud consilium et decretum communis et aliud consilium et decretum 
populi.' Magnus Ryan has kindly made this transcription available to me; he discusses 
Cinus' consilium in 'Bartolus of Sassoferrato', p. 82, and 'Political Thought', pp. 440-1. Cf. 
Bellomo, Quaestiones, p. 299 (no. 197). 
9 In the words of Ryan, 'Political Thought', p. 441 : 'Florence acted, when it acted, as the 
Commune and People of Florence, just as Rome had acted as the Senate and People of 
Rome.' 
10 E.g. Jacobus de Arena, Tractatus de bannitis, fol. 356va, justifying the statutory 
prohibition on banniti holding public offices: 'Secundo obstat reipublicae offensio. nam per 
magi stratum ipsa respublica repraesentatur... Sed cons tat quod non faciendo, id est, non 
parendo ipsi magistratui, videtur contra eum facere et offendere ... et contra faciendo videtur 
eum offendere, ergo et ipsam rempublicam offendit...' Cf. Baldus to D.4.5.5 (p. 209, n. 119) 
and Cinus de Pistoia, in Monti, Le "quaestiones" e i "cons ilia" (p. 218, n. 157). 
11 Bartol us, Lucanae civitatis - p. 211, n. 127 below. 
178 
This state of affairs was probably the reason why the civilians were uninterested in 
defining the city or the relationship between its diverse organs of government in their 
considerations on the communal ban. If a specific judicial officer is mentioned it is 
unsurprisingly the commune's podesta, but the discussion of the ban itself and the 
sta~tes from which it arose is always framed broadly, with reference to the city in its 
e~tuety. The point is embodied by the terminological vagueness regarding the city in 
this context: the terms commune/communale, municipiumlmunicipale, and civitas are 
used ~nterchangeably to describe the source of statutory law and the exiling 
authonty. Because the commune had this wider jurisdiction in the context of 
banishment, it makes sense to use the words 'communal', 'municipal', and 'civic' as 
general synonyms to describe the ban of the Italian cities, as long as one bears in 
mind that, strictly speaking, the city and the commune were not the same thing. The 
first two ~erms _are often preferable to 'civic', since the latter can refer to the city in 
gen~r~l, mcludmg the universal city of Rome, but which, in this chapter, unless 
exphcitly stated otherwise, will refer exclusively to the local Italian civitas. 
ii. Roman and Local Citizenship 
By the opening of the fourteenth century, the jurists came to understand the 
communal ban as a local, restricted form of capitis deminutio media. 12 It is therefore 
possible to identify many of the same patterns and jurisprudential rules evident in the 
cases of deportation and imperial banishment. One fundamental difference between 
local _and universal. exile, however, was that the communal ban could, generally 
speak~ng, only depnve a person of the citizenship of an Italian city along with the 
assoc~ated benefits of its statutory law, but not of Roman citizenship and the 
associated benefits of the ius commune. In this way, the jurists further developed the 
notio~ of a verti~a!ly integrated dual citizenship found in the Corpus iuris, according 
to which every citizen of a city within the empire was also a Roman citizen. 13 Given 
that the communal ban primarily operated at the interface between statutory law and 
the ius commune, an analysis of the legal position of banniti will be fruitful regarding 
the two types of citizenship derived from these legal spheres. Firstly, it will entirely 
confirm what we have learned about Roman citizenship in the previous two chapters. 
12 A . . pomt s~c.cmctly put by Baldus at D.4.5.5, 1, fol. 249rb: 'Nee obstat quod dicatur, quod 
pot~states c1v1ta~m non ~ossunt deportare, quia verum est vera deportatione. sed bannum 
f
3
ap1tale est quasi deportat10 quantum ad iura illius civitatis ... ' 
Cf. Ch. 1, pp. 7-8. 
179 
Secondly, it will demonstrate that municipal citizenship, contrary to how it has been 
portrayed in the modem literature, was not fundamentally contractual, nor in essence 
defined by the performance of public duties or by political participation. Much like 
Roman citizenship, it was a personal legal status entirely derived from the city's 
positive law, which primarily (though not exclusively) conferred a number of 
advantageous rights in the private sphere, especially in the law of inheritance and 
dowry. 
iii. The Systematic and Casuistic Traditions 
The primary source material on the communal ban is enormous and, in terms of 
volume, far outstrips what the jurists wrote on deportation and imperial banishment. 
This is to be expected given its relevance to the everyday life of the Italian cities - in 
the words of Jacobus de Arena ( d. ea. 1296), the material is both 'utilis et 
quotidiana' .14 Broadly speaking, the discourse can be separated into two, using a 
distinction between what I have labelled the systematic and the casuistic parts of the 
tradition. This distinction rests on different modes of legal argumentation. When the 
jurists argued systematically, they referred to some broader or overarching concept 
or analogy establishing a general rule to determine the answer to specific questions 
about the legal position of banniti. Such a rule might stipulate that banniti have a · 
fundamental status - e.g. banniti are Roman citizens - from which flow certain 
entitlements and obligations. When the jurists argued casuistically, they answered 
one specific question about the status of banniti using a number ( often a very large 
number) of diverse analogies and principles drawn from the Roman law, adduced for 
the one particular case at hand. Typically, this material comes in the form of 
extensive quaestiones that contain long pro, contra, and solutio sections; reading 
these in the original involves joining battle with a massive slough of technical detail 
and endless citations of Roman law. Most of the time, the analogies and principles 
under consideration appear far-fetched to modem readers (if they are not completely 
incomprehensible) and hold very little explanatory value for those who do not have 
the benefit of a medieval legal education. This means that the majority of the detail 
becomes irrelevant for our purposes since it does not explain why the author has 
chosen a given side in a legal dispute. Ultimately, one analogy or principle is deemed 
14 Jacobus de Arena, Tractatus de bannitis, fol. 355va: 'Quoniam bannitorum materia utilis 
et quotidiana est, et circa earn plurima sunt advertenda, ideo circa illam plures quaestiones
 
subiicam .. .' Cf. Ghisalberti, 'La condanna', p. 6; Mooney, 'The Legal Ban', p. 88. 
180 
to be more convincing or applicable than another, which, to a twenty-first century 
reader, normally amounts to little more than a tautological restatement of the chosen 
solution. For these reasons, questions frarried in a casuistic manner usually only 
require the briefest of summaries. 
This distinction is related to the two senses of Roman law, or ius commune, that we 
will encounter in this chapter, both of which defined the legal position of the 
bannitus in their own way. The first sense is associated with systematic 
argumentation: it is Roman law as a binding legal norm valid in all the lands of the 
Roman empire; this allows banniti to be Roman citizens and to access the benefits of 
the ius civile. It is a layer of law that exists just as much as the Italian cities' statutory 
law, the natural law, and the law of nations exist. The second sense is associated with 
casuistic argumentation: it is Roman law as a universal and authoritative set of law 
books containing a repertoire of legal standards, methods, principles, and analogies, 
capable of providing the solution to any particular problem thrown up by 
contemporary legal practice on a case-by-case basis. In this way, the Roman law 
could be used to ameliorate the situation of banniti, but it could also rationalise 
statutory provisions to their detriment. Either way, banniti were never completely 
right-less; as we observed in our discussion of the imperial ban, outlawry in the 
civilian tradition never placed the outlaw beyond the law. 15 
vi. A Chronology of the Discourse 
Since the material will be divided thematically, it makes sense at this point to give a 
rough chronological overview of the discourse.
16 Although the evidence is scarce, 
the first indications for learned juristic reflection on civic exclusion come from the 
late twelfth and early thirteenth century, around the same time as criminal 
'.
5 ~ allerani has ide~tified this second, casuistic use of the Roman law as a way in which the 
Junsts d~~med. an? 1mpr~ved .th~ lega! status of banniti - 'Diritti di cittadinanza', p. 1 I: 'In 
~nt~ambt_ 1 _ ~as1, I eseges1 cas1st1ca det doctores ha individuato un terreno comune di diritti 
md1spomb1h, non soggetti alla diminutio iuris imposta dal formulario dei bandi. II bando in 
sostanza, diminuiva, fino quasi ad annullarla, la partecipazione, ma non sradicava de! t~tto
 
l'~pp~rtenenza a una rete di diritti che legeva l'individuo ai suoi vicini, ai parenti, agli altri
 
ctves m generale.' 
16 
Besides the exceptions that will become evident in what follows, the broad brush strokes 
of my chronology are also those of Ghisalberti, 'La condanna', pp. l 0-20, Cavalca, JI bando 
P~- 78- I 00, and Milani, 'Giuristi, giudici e fuoriusciti' ( which is especially useful for th; 
thirteenth century). 
181 
banishment, in its characteristic form of outlawry, firs
t appeared in the communal 
statute books. Perhaps the earliest mention of the ban co
mes from a procedural work 
traditionally ascribed to Pillius de Medicina ( d. ea. 1213). 
17 One relevant quaestio 
conventionally attributed to Azo (d. ea. 1220/9), and edited by Ernst
 Landsberg, also 
survives, 18 and we already saw in the last chapter th
at Roffredus (d. ea. 1243) 
discussed the Sicilian royal ban in his Libelli iuris civilis
 et canonici. 19 One can find 
a slight increase in references to the ban (both communal and imper
ial) in the works 
of the final generation of glossators. I can count six sep
arate loci containing the term 
bannum or bannitus in the Ordinary Gloss of Accursius
 (d. 1260/3),20 and seven in 
the glosses of his contemporary Odofredus ( d. 1265).
21 Furthermore, a handful of 
short quaestiones on the ban are contained in Alber
tus Galeotti's (d. ea. 1272) 
Margarita, a collection of quaestiones circulating i
n the law schools that was 
composed around the year 1250.
22 
While there was clearly at least some discussion of ban
ishment amongst civilians in 
the first half of the thirteenth century, the discourse 
appears to be fragmentary. 
Although a lack of surviving evidence is presumably 
one reason for this, another 
must be the glossators' focus on the theoretical expos
ition of the Corpus iuris. A 
proper examination of the communal ban therefore only
 began with the practical tum 
in the second half of the thirteenth century, when the 
jurists of the post-Accursian 
generation started to display a significantly greater inter
est in the statutory law of the . 
17 Libellus de preparatoriis litium et earum preambuli
s, (ed.) Palmerius, BIMA III 
(henceforth, Pillius (BIMA), Libellus), rub. IIII, Si quis in ius vocatus 
non ierit, p. 24. Lange, 
Die Glossatoren, p. 236, views the authorship of Pillius
 as unlikely; although the question 
remains inconclusive, more recently, the Bolognese jurist Guizardinus d
e Porta Esteria has 
been considered as a potential author, allegedly composi
ng the Libellus in the years 1220-22. 
Cf. Ghisalberti, 'La condanna', p. 15; Cavalca, II ba
ndo, pp. 12, 83; Pazzaglini, The 
Criminal Ban, pp. 17-18; Milani, 'Giuristi, giudici e fuori
usciti ', p. 600. 
18 Landsberg, Die Quaestiones des Azo, pp. 65-8, quaestio VIII, Titiu
s commisit delictum 
(henceforth Azo (Landsberg), Titius commisit delictum). Cf. Kantorow
icz, 'The quaestiones 
disputatae of the Glossators', p. 48; Belloni, Le question
i civilistiche, pp. 90 (Azo A 9) and 
92 (Azo B 2/7); Lange, Die Glossatoren, p. 268. Milani, 'Giuristi, g
iudici e fuoriusciti', pp. 
600-2, contains a good discussion of this quaestio. 
19 Ch. 3, pp. 111-12. 
2° Communal ban: lnst. l.12,2, D.42.4.13; Communal an
d imperial ban: LF.2.28 c. Domino 
guerram; Annotation as the ban of the fisc: D.48.1
7.1,1; Bannum as fine: D.25.4.1,3, 
C.5.43.3. 
21 Communal ban: C.1.5(8).7, C.6.7.2; Imperial ban: auth. Cassa, po
st C.1.2(5).12, auth. 
Item nulla, post C.1.3(6).2, auth. Omnes peregrini, post C.6.59.1 O; A
nnotation as ban of the 
fisc: C.9.40.1; Either communal or imperial ban: C.6.22.8
. 
22 R. Isotton, 'Galeotti, Alberto', DBGI, vol. 1, pp. 929-30. 
182 
Italian cities. This change in legal culture was particul
arly fruitful regarding the ban 
since it coincided with the height of the system of party
-political expulsions, forcing 
the jurists to grapple with pressing contemporary legal needs that arose n
ot only from 
the ordinary administration of justice but also from civic politics. The re
sult was the 
proliferation of the discourse. 
23 
The predominant legal literary genre of the period be
tween ea. 1260 and 1330 in 
relation to the ban was the quaestio de facto or the quaestio ex facto em
ergens, that 
is, the analysis of a disputed question arising from legal
 practice (rather than theory), 
d. d. h 
24 
1scusse m t e set format pro-contra-solutio. Most of 
the quaestiones on the ban 
from this time were not printed separately in the early 
modem period and we have 
them at first hand only in manuscript form. We do, ho
wever, have them at second 
hand in early modem editions, primarily in the wor
ks of the great synthesisers 
writing between the later thirteenth and mid fourteenth 
century - notably Guilelmus 
Duranti ( d. 1296), 
25 Albertus Gandinus ( d. ea. 1311 ), 26 Johannes Andreae ( d. 
1348),
27 
and Albericus de Rosate (d. 1360)2
8 
- who reported the relevant quaestiones 
at 'length. Their collections incorporate the thought of 
the most famous jurists of the 
later thirteenth century, such as Guido de Suzarra (d. ea. 1293), Jacob
us de Arena (d. 
2
~ ~s Cavalca, II bando, p. 2, already noted, '[i]n quel periodo, ii 
bando assunse particolare 
nltevo come strumento repressivo non solo e non tan
to dei reati comuni, quanto dei reati 
potitici.' 
24 
Cf. Bellomo, Quaestiones, 'Introduzione' , pp. vii-xxxviii. This bo
ok is an invaluable 
reference work for the civilian quaestiones literature
 from the mid-twelfth to the early 
thirteenth century. 
25 
Speculum iudiciale, of which there are two recensions:
 the first was completed before 
1276 and did not circulate widely; the second, the basis 
for later printed editions, some time 
in the late 1280s up to 1291. As is usual, the edition 
I use is based on Duranti 's second 
rec_ension and _also contains Johannes Andreae's ex
tensive additiones. Cf. Lange and 
~nec~baum, Die Kommentatoren, pp. 481-6. For more d
etails on the Speculum, particularly 
its datmg, textual layers, and sources, see R. Naz, 'Guill
aume Durand', Dictionnaire de droit 
canonique, vol. 5, cols 1030-52. 
26 
Tractatus de maleficiis, rub. Quid agendum sit reo absente et contumac
e: et de bannitis 
pro maleficio, and rub. De bannitis pro maleficio, expertly edited by Ka
ntorowicz, Albertus 
Gandinus, pp. 128-51. Albertus' Quaestiones statutorum are edited by
 Solmius in BIMA III. 
Note that Solmius' edition is based on a text dated to 
1289 and does not take into account 
Albertus' second recension of 1301/2. Cf. Lange and K
riechbaum, Die Kommentatoren, pp. 
4 71-7; Kantorowicz, 'Leben und Schriften'. 
27 See n. 25 above. 
28 
Albericus wrote lengthy commentaries as well as his Quaestiones sta
tutorum (also known 
as OP_us statutorum, Commentarium de statutis, and Trac
tatus de statutis), part IV of which 
contains 114 quaestiones purely devoted to banniti! 
Cf. Lange and Kriechbaum, Die 
Kommentatoren, p. 681. For a good introduction spe
cifically to Albericus' Quaestiones 
statutorum, see Romano, Le 'quaestiones disputatae ', pp. 
47-70. 
183 
ea. 1296), and Dinus de Mugello (d. ea. 1298/1303), little of whose work has 
survived (whether in print or otherwise),29 as well as the quaestiones of lawyers who 
are not generally well known but who made significant contributions in the field of 
banishment: most importantly, Fredericus de Scalis (d. ea. 1289) and Nicolaus de 
Matarellis (ea. 1314).30 Although, as compilers, they reported the opinions of others, 
they did not merely create impartial records. Rather, they reviewed and critically 
evaluated the existing literature in light of civilian communis opinio at the time of 
writing. They were supposed to provide practising lawyers with those solutions that 
were most likely going to lead to success in court. They therefore give us a good 
indication of the state and development of scholarship on the ban during the later 
thirteenth and early fourteenth centuries.
31 
Although the importance of the scholarship of this period has been recognised for 
some time now,32 and although there has been a relatively recent impetus to study the 
quaestio literature of the later thirteenth century in the context of banishment,
33 there 
is a marked tendency in the existing historiography to shy away from properly 
reconstructing (insofar as this is possible) the thought of certain key members of this 
generation of jurists through those quaestiones of theirs that have only survived in 
the reports of others. 34 In some important respects, this has resulted in a 
misrepresentation, mostly in the form of an undervaluation, of the contribution of 
late-thirteenth century jurists such as Fredericus de Scalis and Nicolaus de Matarellis 
29 E.g. in the first manuscript recension of Gandinus ' Tractatus de maleficiis, almost half of 
the paragraphs are signed with 'G.', indicating their origin in the Suppletiones ad gloss am of 
Guido de Suzzara - Kantorowicz, 'Leben und Schriften', p. 298. On Guido's suppletiones, 
see Martino, Ricerche. 
3° Cf. S. Menzinger, 'Federico della Scala', DBGI, vol. 1, p. 883, and M. Semeraro, 'Niccolo 
Mattarelli ', DBGI, vol. 2, pp. 1425-6. 
31 Cf. Lange and Kriechbaum, Die Kommentatoren, p. 475 (Albertus Gandinus), p. 482 
(Guilelmus Duranti), p. 662 (Johannes Andreae), p. 677 (Albericus de Rosate). 
32 Cavalca, II bando, p. 79: 'In realta, gli Statuti comunal i furono invece studiati dai giuristi 
della seconda meta del XIII secolo con un'intensita ed una profondita speculativa che non 
teme confronti. Le opinioni di Dino del Mugello, Iacopo d' Arena, Guido da Suzzara, Alberto 
da Gandino, Alberto Galeotti , Ugolini Fontana, Guglielmo Durante, Lambertino Ramponi, 
Martino da Fano, dello stesso Accursio e di molti altri costituirono la base del pensiero di 
Cina, Bartolo e Baldo.' 
33 Milani, 'Guidici, giuristi e fuo riusciti'; Vallerani, 'Diritti di cittadinanza'. 
34 This tendency is most evident in the older accounts of Ghisalberti, 'La condanna', and 
Cavalca, II bando. Ghisalberti's reliance on printed authors is particularly evident in his 
comment (pp. 4-5) that '[l]e tappe dell'evoluzione e dello sviluppo della teoria del bando 
criminale sono definite dalle successive trattazioni scientifiche dell'istituto che Jacopo 
d 'Arena, Alberto Gandino, Bartolo da Sassoferrato, [ e] Nella di San Gimignano... ci 
lasciarono ... ' 
184 
in favour of those authors whose works have survived in print and who used and 
reported their quaestiones.
35 The fact that we regularly have multiple later reports of 
older quaestiones allows us to reconstruct their contents with a relative degree of 
confidence. The benefit is the ability to trace the progress of the discourse in more 
certain terms, and to give credit where credit is due. 
We will see that most of the fundamental arguments and concepts developed 
between ea. 1260 and 1330; especially the thought of the period up to about 1300 
stands out for its originality. While different schools of thought were still competing 
for advantage during the late thirteenth century, this debate gave way to a set 
communis opinio on almost all questions concerning the ban during the opening two 
or three decades of the fourteenth century. With the exception of Bartolus' 
contribution to the systematic tradition, the role of the commentators writing after 
that time mainly lay in reporting and refining extant opinion. 
36 Bartolus' systematic 
thought on the ban was both impressively original and profoundly rooted in the 
discourse as it had developed up to his day. He provided the most sophisticated and 
intellectually satisfying theoretical framework in which the existing tradition could 
be accommodated. Indeed, one of the aims of this chapter is to demonstrate the 
quality and originality of Bartolus ' work, which has not been adequately recognised 
by historians to date. Following Bartolus ' death, his interpretation was faithfully 
transmitted to posterity by his pupil Baldus. 
2. Banishment and Factionalism: The Rule of Law? 
Given the mass of material we have concerning communal banishment, it is striking 
that there is so little discussion of the lawfulness of the factional struggles within the 
Italian cities. As- we shall see below, the jurists certainly assessed the validity of civic 
35 
For three specific cases in which material is attributed to Albericus de Rosate rather than 
to the jurist whose thought he is reporting, see below p. 188, n. 47 (Thomas de Piperata), p. 
239, n. 237, and p. 243, n. 251 (both Nicolaus de Matarellis). 
36 
Although I would emphasise the originality of Bartolus' thought, broadly speaking I agree 
with Ghisal~erti ' s characterisation of the fourteenth century - 'La condanna ' , p~. 26-7: 
' ... l'opera ~e, co~~entatori tende piu ad integrare con l'esemplificazione delle sempre piu 
numerose 1potes1 d1 bando la gia abbastanza larga casistica ricavabile dalla lettura dei 
postaccursiani e dai primi Tractatus bannitorum che ad arricchire di spunti veramente 
originali la dottrina sull'argomento .. .' 
185 
legislation on the ban, but they rarely examined the conditions under w
hich it was 
lawful to apply this to political adversaries. There was, in other words, 
no concerted 
effort to provide a legal order to the partisan conflicts between G
uelphs and 
Ghibellines, magnates and popolo, which were such a prominent featur
e of the age. 
Linked to this was an avoidance of fundamental, we would say c
onstitutional, 
questions about the nature of a city and the status of exiled communitie
s in relation 
to these upheavals. One might ask, using the example given above, how
 the knights 
of Brescia, who purported to be the commune, related to the party within
 the physical 
city of Brescia, who no doubt did the same. Instead, the focus of the jurists was on 
the person of the bannitus - his individual legal status - in the abstract. T
hey operated 
under the assumption (technical considerations aside) that his expulsion had been 
lawful without regard to the broader political context. There are only a s
mall number 
of exceptions to this trend. In one of his comments to the Digest, Alberic
us de Rosate 
seems to have recorded such debate as there was on this topic by the m
iddle of the 
fourteenth century. Tellingly, he notes that he does 'not recall any other
 place where 
this material is discussed', despite the fact that 'in miserable Italy ...
 one faction 
continually expels the other from cities and castles' .
37 
Albericus posits three interrelated questions and raises a number of po
ints pro .and 
contra for each. Firstly, can a faction expelled from a city set 
up its own 
representatives or magistrates?
38 One could argue for the proposition by considering 
the expelled faction to be a corporate body (universitas) and therefore capable of 
37 Albericus de Rosate to D.49.15.21 ,1, fol. 222vb: 'Et iste § est aureus et singular
is ad 
materiam quae saepe occurrit in miserabili Italia, ubi continue una 
pars expellit aliam de 
civitatibus et castris. et non memini alibi de ista materia tractari nisi 
hie.' Milani, 'Giuristi, 
giudici e fuoriusciti ', discusses a handful of quaestiones and one co
ns ilium from the later 
thirteenth century that refer to banishment as a result of party politic
s: Albertus Gandinus, 
Quaestiones statutorum, qq. 24-6; Nicolini, 'Un "consilium" inedito'; Caprioli,
 'La 
Miscellanea Romana', pp. 138-9 (no. 21.7 - fol. 47rb-va). Besides the material provi
ded by 
Albericus, these are the only instances in the entire discourse from ea.
 1200-1400 where this 
is the case. Importantly (as already noted by Cavalca, II bando, p. 116, regarding Gan
dinus), 
the authors do not discuss the underlying legality of the political ban
, but focus on certain 
issues (such as debt and credit) surrounding the legal status of banniti. Furthermore, 
as we 
shall see in this chapter, discussions of the ban as a penalty for treaso
n and rebellion outside 
the immediate context of civic factionalism are also exceptional a
nd all come from the 
fourteenth century. There is therefore little to justify Milani's conclusion that 'lungo tutto 
ii 
corso de! Duecento ii bando politico e largamente presente nella rifl
essione della scienza 
giuridica' (p. 624). 
38 Albericus de Rosate to D.49.15.21, 1, fol. 222vb: 'Et ideo, quia saepe in
 Italia occurrit, pro 
declaratione materiae quaeramus tres quaestiones. Primo, si una p
ars expellit aliam de 
civitate, villa, vel castro, nunquid pars extrinseca possit constituere syn
dicum?' 
186 
electing officials and making its own laws. This follows the exampl
e of ancient 
Rome, recorded at the opening of the Digest, where the plebs defied t
he senatorial 
elite and seceded in order to set up separate legal institutions; hence the c
ity of Rome 
came to be composed of two . corporate bodies - the senate and t
he people. 39 
Albericus mentions that Jacobus de Arena used this section of the Dig
est to argue 
that an expelled faction can indeed act as a corporate body, elect officia
ls, and make 
statutes; furthermore, he adds, such is the modem custom of all Italy.
40 On the other 
hand, it could be said that it is legally impossible for an expelled facti
on to form a 
corporate body since it was lawfully expelled for criminal, seditious beh
aviour, from 
which it should draw no benefit.41 
Secondly, can the magistrate of the faction within the city summon to
 court those 
who have been expelled, and if they do not appear, place them under the
 ban? On the 
one hand, it is the duty of every person to respond to a court summ
ons and the 
prerogative of the magistrate to punish those who do not. On the oth
er hand, one 
could remain absent due to justified fear aroused by the enmity and partiality of the 
other faction. Besides, the podesta of the internal faction has no jurisdiction over the 
external faction, since the latter had no part in his election, and for the 
same reason, 
that election could even be considered void.
42 
39 D 
.1.2.2,8 (adapted from Watson (ed.), TheDigestofJustinian, vol. 1, p. 4): ' .. . it came to 
pass that the plebs fell at odds with the members of the senatorial clas
s, and seceded, and set 
up laws for itself, which laws are called plebiscites. Soon after the ple
bs had been wheedled 
back, because these plebiscites were giving rise to many disputes, the
 decision was made in 
the !ex Hortensia that they were to be deemed to have the force of laws
.' 
40 
Albericus de Rosate to D.49.15 .21, 1, fols 222vb-3ra: 'Super prima qu
aestione, quod pars 
extrinseca possit constituere syndicum, videtur quia est quaedam univ
ersitas .. . Item, quia per 
iura videntur esse in civitatibus duae partes, scilicet patricii et plebei
i ... et patricii dicuntur 
una universitas et plebeii alia. et hoe optime probatur [D.1.2.2,8], ubi dicitur quod 
pars quae 
secessit de Roma sibi iura constituebat, et [D.1.2.2, 15]. ex quibus §§ tenet ibi lac. 
de Are. 
quod pars expulsa potest sibi rectorem eligere, statuta facere, et per
 consequens syndicum 
constituere. .. et istud de consuetudine et de facto servat in Ital
ia, quod tales partes 
extrinsecae constituurit syndicum et eligunt sibi rectorem, et statuta
 et alia faciunt, sicut 
universitas.' This is an accurate report of Jacobus' opinion at D.1.2.2. 
41 
Albericus de Rosate to D.49.15.21 , I, fol. 223ra: 'In contrarium vi
detur, quia omnes 
universitates videntur illicitae et reprobatae nisi reperiantur a iure app
robatae ... et quia tales 
expulsi videntur seditiosi et propter seditionem expulsi ... cum e
rgo ex hoe poenam 
mereantur, non de bent praemium consequi ... nota quod omnem seditio
nem vel emulationem 
continet a iura reprobatur .. . ' . 
42 
Ibid.: 'Ad secundam quaestionem, quod pars extrinseca possit citari, e
t si contumax fuerit, 
puniri, videtur primo quod omnis vocatus in ius venire debet... et
 si no~ veniat, puniri 
potest... In contrarium est, quia isti iusto metu ducti propter i
nimicitias non tenent 
comparere. .. Item, quia potestas intrinsecorum non videtur hab
ere iurisdictionem in 
187 
Thirdly, does the faction within the city represent the legitimate public authority 
(respublica) of the city, such that the city council can lawfully make statutes and 
conduct other official business in the name of the whole city, which would be 
binding on both factions?43 It is possible to say against the proposition that decisions 
of the city council are only valid and binding on the entire community if both 
factions of councillors are present. Moreover, a city where one faction has been 
expelled has suffered capitis deminutio, similar to one that has been put to the 
plough. Therefore, the former universitas of the city is split into two separate ones, 
which have jurisdiction only over themselves. 44 On the other hand, one could argue 
that the faction within retains the right to act in the name of the whole civic 
corporation (the ius universitatis), since those expelled are rebels who have attacked 
their homeland and have been punished due to their contumacy.
45 
The solution to which Albericus seems to have attached particular weight - since it 
appears both in his commentaries as well as in his Quaestiones statutorum (where it 
is given as the only approach to this topic )46 - comes from a quaestio attributed to the 
Bolognese jurist Thomas de Piperata ( d. ea. 1282). 47 His argument rests on a 
distinction based on the idea of legal fault (culpa). If the expulsion was the fault of 
the internal faction, then the arguments for the expelled faction apply. But if this was 
extrinsecos, quia non est electus per eos ... imo, quod plus est, non videtur tenuisse electio 
potestatis, quia in eius electione consentire debent universi nobiles civitatis ... ' 
43 Ibid., fol. 222vb: 'Tertio, an pars intrinseca repraesentet rempublicam illius civitatis, ut sic 
maiori parte decurionum possint statuta condere, obligare bona civitatis, et intrinsecos et 
extrinsecos, et alia agere quae spectent ad regimen et administrationem civitatis.' 
44 Ibid., fol. 223ra: 'Ad tertiam quaestionem, utrum intrinseci cum maiori parte decurionum 
possint obligare extrinsecos et bona civitatis, videtur quod non, quia decreta decurionum non 
valent nisi adsint duae partes decurionum ... Item, quia civitas per expulsionem unius partis 
videtur passa capitis diminutionem ad similitudinem civitatis quae passa est aratrum ... et 
videtur quasi dissoluta universitas ... et videtur quod ex una universitate factae sunt duae ... ' 
45 Ibid.: 'Contrarium facit, quia intrinseci videntur retinere ius universitatis ... quia expulsi 
dicuntur rebelles civitatis, quia videntur oppugnare patriam ... et quia videntur absentes ex 
quadam conturnacia ... ' 
46 Albericus de Rosate, Quaestiones statutorum IV, q. 83, fol. 81 va. 
47 Cf. A. Padovani, 'Tommaso da Piperata', DBGI, vol. 2, pp. 1959-60. It is possible that the 
author is Albertus Odofredus (d. 1299/1300), son of the famous glossator Odofredus de 
Denariis - cf. C. Natalini, 'Alberto di Odofredo', DBGI, vol. 1, p. 27. The incipit of the 
quaestio is In aliqua civitate, and it is extant in Rome, Biblioteca Apostolica Vaticana, MS. 
Archivio S. Pietro A. 29, fols 138vb-39va (Thomas de Piperata), and MS. Chigi, E.VIII. 245, 
fol. 92ra-va (Albertus Odofredus) - cf. Bellomo, Quaestiones, pp. 92-3 (no. 177) and p. 264 
(no. 118). This quaestio is briefly discussed by Cavalca, II bando, p. 115, but incorrectly 
attributed to Albericus. · 
188 
the fault of the expelled faction, then the arguments for the internal faction apply.48 
Although this appears to be a very vague solution, culpa was an established concept 
in Roman law, and would have given practising jurists some ideas and guidelines 
about how to interpret the facts of each given scenario. 
49 On the other hand, unless 
defined more narrowly, culpa was vague enough to legitimate just about any faction 
in the messy world of Italian politics - perhaps this too endeared it to practising 
jurists. 
The best, and only other, sustained treatment of Italian factionalism comes from 
Bartolus of Sassoferrato in his well-known treatise On Guelphs and Ghibellines, 
where he analyses a number of legal questions arising from the competition between 
civic factions, including the problem of violent conflict and expulsion. 50 Under 
normal circumstances, Bartolus argues, it is seditious and therefore illegal for one 
faction to rise up against the governing faction in order to remove it from power. But 
if the government that it aims to depose is tyrannical and of the worst kind 
('pessimum '), then this would be licit as long as two conditions are met. Firstly, it is 
impossible to have recourse to a superior (such as emperor or pope), who has the 
capacity of deposing the tyrannical government without unreasonable difficulty. 
Secondly, the actions of the rebelling faction have to be directed towards public 
utility and restore the city to good order; it would be unlawful to rise up merely for 
48
_ Albericus de Rosate to D.49.15.21, 1, fol. 223ra-b: 'has quaestiones disputavit d. Tho. de 
Ptperata, et solvendo <licit de iure civili scripto distinguendum utrum pars extrinseca sit 
expu!sa culpa ~ua, an culpa partis intrinsecae. Primo casu <licit... partem extrinsecam 
synd1cum constttuere ~o? posse, c~m sit universitas illicita et seditiosa, ut supra allegatum 
~st.. et eodem ca~u d1c1t quod st extrinseci contemnantur, citati ad mandata potestatis 
mtrmsecorum vemre, possunt condemnari et mulctari secundum formam iuris et statutorum 
et consuetudinu1? c!vitatis per iura _sup_ra hoe allegata. Item, in eundem casum <licit quod 
contractus et ob!1gat10nes factae per mtrmsecos valere et extrinsecos ad hoe teneri ... Secundo 
vero casu, quando culpa partis intrinsecae extrinseca fuit expulsa, <licit partem extrinsecam 
posse syndicum constituere, non ut parte sed ut congregationem hominum pro violentia 
propulsanda et iu~t!ti~ consequenda, argumentum iurium supra ad hoe allegata. [ ... ] Item, in 
eundem ~asum d1c1t 1llos de parte extrinseca non posse banniri, condemnari, mulctari, vel 
aggravan per ~o~estatem intrinsecorum, cum iusto metu absint, ut praedictum est. [ ... ] Et 
eod~m c~su d1~1t contractus et alia gesta per intrinsecos non valere, nee praeiudicare 
e~tnnse~t~ pe~, tura ad ho_c allegat~. ~t ita determinat dictus d. Tho. et incipit quaestio "in 
:J1qua c1v1tate , et est 41. m quaest10mbus Doctorum Bononiae.' 
Cf. Berger, Encyclopedic Dictionary, pp. 419-20. 
50 
For a critical edition, see Quaglioni (ed.), Politica e diritto, pp. 129-46 (henceforth 
Bartolus (Quaglioni), De Gue/phis et Gebellinis). 
189 
I 1J 
, I 
I 
I 
111 
one's private utility and institute another tyrannical regime after expelling the other 
faction. 51 
The requirement that there be no superior lord capable of removing the tyrannical 
faction from power dovetails with Bartolus' argument in his treatise On the Tyrant, 
where he stresses the role of the superior judge in deposing a lower tyrannical office-
holder and foregoes any mention of the people ( or faction thereof) in this regard. 
52 
He justifies popular execution in On Guelphs and Ghibellines with reference to the 
Roman law dictum that it is licit to declare law unto oneself by one's own authority 
when there is no recourse to public authority.
53 Bartolus bases the second condition 
on the idea that in a city which has no superior lord, a tyrannical government detains 
the commonwealth by unlawful force 'from the commonwealth itself ('tenent rem 
publicam ... ab ipsa re publica').54 As Magnus Ryan has highlighted, this awkwardly 
formulated idea is the internal aspect of Bartol us' famous theory of a free city that is 
'a prince unto itself (civitas sibi princeps). 55 The city, as a corporate entity, is its 
own superior in the absence of any other lord (i.e. emperor or pope), and this legal 
abstraction is the locus of political obligation as well as the injured party in the case 
of political crime. And just as the existing tyrannical faction causes injury to the city-
prince through its tyranny, so too a rebelling faction intent on securing the 
government merely for its own interest rather than the common good - which is one 
51 Bartolus (Quaglioni), De Gue/phis et Gebellinis, pp. 137-8: 'Quandoque non solum vult 
una pars resistere, sed vult deponere de regimine alios regentes: tune si quidem insurgere
nt 
adversus istud regimen istud esset simpliciter illicitum ... Si vero regimen quod voleba
nt 
deponere erat tyrannicum et pessimum, tune ad hoe esse unius partialitatis et unius nomin
is 
licitum est duobus concurrentibus. Primo, quod habendo recursum ad superiorem non poss
it 
illa tyrannides deponi sine magna difficultate. Secundo, quod ipsi hoe faciant propt
er 
utilitatem publicam, ut status civitatis restauretur. Secus si hoe facerent ut ipsi nova
m 
tyrannidem inchoarent, aliis expulsis.' 
52 For a critical edition of this treatise, see again Quaglioni ( ed.), Political e diritto, pp. 171-
213 (henceforth Bartolus (Quaglioni), De tyranno) - at p. 202 : 'Nono quero: si aliquis dux, 
marchio, comes vel baro, qui habet iustum titulum probatur tyrannus exercitio, quid deb
et 
facere superior? Respondeo: de bet eum deponere .. .' 
53 Bartolus (Quaglioni), De Gue/phis et Gebellinis, p. 138: 'Primum probatur, quia licitum 
est sibi ius die ere propria auctoritate quando non habetur copia iudicis .. ' 
54 Ibid., pp. 138-9: 'Secundum probatur, quod ad propriam utilitatem non liceat, quia illi qui 
occupatam per tyrannidem tenent rem publicam, vi earn possident ab ipsa re publica v
el 
domino superiori, non ab aliquo privato. Aliter ergo qui talem tyrannidem deponere conat
ur 
propter se vim facit illlicitam ... ' 
55 Ryan, 'Bartolus of Sassoferrato', esp. p. 79 ff. 
190 
of Bartolus' definitions of tyranny
56 
- causes the same injury. Rebellion would be 
unlawful not because of injury to the tyrannical faction, which takes the place of a 
private person, but because it would constitute the use of illicit force against the city 
itself, a public person and the lawful superior. 
According to Bartolus, the tyrannical government of a faction is not the only factor 
legalising its expulsion. He clarifies in his commentaries that the overarching 
requirement is just causa (iusta causa): as long as one faction expels the other with 
just cause, it can subsequently undertake legal acts, such as legislating, that bind the 
whole city.57 While tyrannical government is no doubt one instance where causa is 
present, Bartolus' treatise On the Tyrant illustrates that causa is a very broad 
justification, which also applies when the expelled faction has committed lesser 
offences: there, he argues that no one can doubt the just cause inherent in exiling 
'quarrelsome' and 'seditious' magnates.
58 (All too evident is Bartolus' sympathy for 
the nexus of Guelph and popular causes that dominated the political settlement of his 
own Perugia.) Fundamentally, therefore, Bartolus' solution is the same as Thomas de 
Piperata's since it rests on the culpa or otherwise of the factions - their behaviour -
in the conflict resulting in expulsion: one faction had cause to expel the other if the 
latter was in some way culpable. Although both solutions are ultimately as vague as 
each other because neither the boundaries of causa nor those of culpa are well 
56 
Bartolus (Quaglioni), De tyranno, p. 196: 'Dico quod ille tyrannus est ex parte exercitii, 
qui opera tyrannica facit, hoe est, opera eius non tendunt ad bonum commune, sed propriu
m 
ipsius tyranni. Istud enim est non iure principari.' 
57 
Bartolus to D.1. 1.9, fol. 1 Ora: 'Secundo quaero ... quid si una pars expulit aliam de civitate? 
[ ... ] aut sunt expulsi ex iusta causa, et adhuc in remanentes remanet potestas condendi legem, 
quia cum non obediant civitati suae, perdunt omnia iura propria civitatis suae, ut [D.4.5.5] , et 
sic perdunt ius condendi statuta, quare potestas remanet apud praesentes. Aut sunt expul
si 
sine iusta causa, per invidiam forte, et tune, licet statuta facta per remanentes serventur for
te 
quantum ad eos .. . tamen in praeiudicium expulsorum non possunt...' We should note th
e 
tension between this argument and Bartolus' short comment to D.30.1.32,2, where he simp
ly 
states that the expulsion of one faction means that neither faction can undertake an a
ct 
(submission of the city to a lord) binding the whole city; there is no consideration as to 
whether there was iusta causa in the expulsion. Cf. Woolf, Barto/us of Sassoferrato, pp. 
189-95. One can observe a similar tension in Baldus' thought. He repeats Bartolus' iusta
 
causa distinction in his commentary to D.1.1.9. At D.3.4.1 and D.30.1.32,2, however, th
e 
only criterion determining the ability of one faction to submit the whole city to a lord 
is 
whether it constitutes the maior pars. Thus, even if the maior pars has been expelled, it ca
n 
act for the whole city, without a consideration, it seems, whether the expulsion occurred 
ex 
iusta causa. Cf. Canning, Baldus de Ubaldis, pp. 130 and 153-4. 
58 
Bartolus (Quaglioni), De tyranno, p. 199: 'Quis enim dubitat, quod si aliquis esset in 
civitate potens, rixosus et seditiosus, iste debeat de civitate expelli a quolibet iusto iudic
e? 
[ ... ] Tune cum iusta causa fieret, non esset actus tyrannicus.' 
191 
I 
defined, Bartolus does at least provide relatively detailed guidelines to the case of 
tyranny. On the Tyrant describes under what circumstances rulers should be 
considered tyrannical, to what extent their official acts have legal force, and how 
they can be deposed (though his thought is by no means without its weaknesses).
59 
While this is not the place for a full review of the treatise, we should note that 
Bartolus considers any judicial proceedings initiated by a proven tyrant against 
rebels to be null and void ipso iure. He refers to the same idea as Thomas de 
Piperata: a defendant is not compelled to appear in a court which is in a notoriously 
dangerous place and whose judge is notoriously hostile and partial.6
0 
The question remains why the civilians did comparatively little to analyse the role of 
faction in the expulsions of the thirteenth and fourteenth centuries. The biographies 
of the jurists reveal that they were deeply involved in the political life of the Italian 
cities. The fact that we cannot find more arguments undermining the authority of 
civic governments in this area with some reference to their partiality and illegitimacy 
could be taken to imply that the jurists only worked for the winners. This was 
evidently not the case: everybody needed lawyers, and not uncommonly the lawyers 
themselves were on the losing side. There is some evidence to suggest that Jacobus 
de Arena was caught up in the massive expulsions of the Lambertazzi in Bologna 
during the 1270s and 1280s; we know this with more certainty in the case of 
Fredericus de Scalis.61 These two jurists were instrumental in creating what became 
the dominant theory of banishment, yet, as we shall see below, they did much · to 
mitigate the effects of the ban without ever questioning the legitimacy of sectarian 
governments to issue it. One could reasonably speculate, therefore, that it was not 
good legal strategy for the defeated to base their claims on the partiality of the civic 
governments. This, in tum, points to a general assumption in favour of the winning 
faction in the law and politics of this period: that, in other words, even if the rule of a 
faction was partisan, by having taken control of the physical city and its 
59 Cf. Quaglioni (ed.), Political e diritto, pp. 7-71 ('Introduzione'). Ryan, 'Bartolus of 
Sassoferrato', pp. 86-7, points out some of the problems of Bartolus ' definition of tyrannical 
government in a civitas sibi princeps. 
60 Ibid., p. 189, regarding a tyrant by defect of title: 'Dico sic, quad aliqui processus fiunt 
contra rebelles seu inimicos tyranni et tales processus sunt ipso iure nulli . Non enim 
debuerunt comparere coram iudice sibi notorio inimico vet in loco sibi notorie suspecto ... ' 
Again, at pp. 205-6, regarding a tyrant by conduct: 'Undecimo quero an facta per supradictos 
tyrannos qui iustum habent titulum valeant? Et dico quad aut [tyrannus] fecit processus 
contra suos exitios et rebelles et non valent, quia non debuerunt comparere coram iudice sibi 
notorie inimico, ut dictum est in precedentibus. ' 
61 Cf. D. Quaglioni, 'Iacopo d'Arena', and S. Menzinger, 'Federico della Scala' , in DBGI, 
vol. l, pp. 1099-1100 and p. 833 respectively. 
192 
administrative organs, that faction's government was taken to represent the genuine 
public authority of the civitas. Perhaps for this reason the jurists did not challenge the 
foundations of the system of expulsion but tried to affect it from within, provided 
practising lawyers with a set of working arguments that could be used in the 
inevitable litigation that came about as a result. It was more realistic to grant what 
happened in practice the force of law and channel this in ways that were deemed 
legally and morally acceptable. 
3. The Lawfulness of the Criminal Ban: Statute and ius commune 
As became clear in chapters 2 and 3, according to Roman criminal procedure, an 
absent defendant in capital trials could not receive a final ('definitive ') sentence (loss 
of citizenship, slavery, or death), but was subject to annotation, which resulted in the 
confiscation of his estate after one year. The highest criminal penalty for an absentee 
was relegation (i.e. geographical confinement, a non-capital penalty). Statutory law 
on the criminal ban was therefore in clear violation of the ius commune. After the 
defendant had not appeared at court within a certain period of time, the judge took 
him to have confessed (pro confesso) and placed him under the ban, whose effects 
- clearly amounted to a definitive capital sentence. If the bannitus was captured by 
communal officials, he was subject to immediate punishment - not untypically the 
administration of the death sentence - without any further verification of the alleged 
offence.6
2 The jurists therefore took it upon themselves to justify two core aspects of 
contemporary legal practice as legitimate derogations from the ius commune: firstly, 
that in criminal procedure a capital penalty could be imposed on an absent defendant 
without a full hearing; and secondly, that this penalty could consist in outlawry. 
Although the two points are closely related, and the justification of one implies the 
justification o_f the other, we will analyse them separately below for the sake of 
conceptual clarity. 63 
The jurists rationalised communal practice with reference to their well-known 
broader arguments about the relationship between municipal law and ius commune 
that vindicated the legislative powers of the Italian cities as they had developed by 
62 Cf. eh. 1, pp. 19-21. 
63 Questions of legitimacy are discussed by Ghisalberti, 'La condanna', pp. 19-24, and in 
more detail by Cavalca, II bando, pp. 10-12 and I 75-87. 
193 
the later Middle Ages. 64 The civilians pursued two strategies. 65 One interpreted civic 
statutes as the written expression of long-established custom, which, as one key 
passage in the Corpus Juris explained, possessed the force of law.66 The second 
located the law-giving powers of the cities in imperial concession. According to the 
interpretation of the jurists, at D.1.1.9 (l. Omnes populi) the emperor had granted 
each civitas within the empire the capacity to legislate for itself. 67 Generally 
speaking, the jurists considered municipal law to be valid only within the legislating 
city's territory, where it was supposed to be complementary to the ius commune. 
Statute could, however, also derogate from, though never abrogate, Roman law (as 
well as higher natural norms) under certain defined circumstances. In this respect, the 
relationship between statute and ius commune was equivalent to the relationship 
between ius commune and natural law. Where statute clashed with the ius commune, 
the jurists interpreted statute restrictively, which reflected their belief in the 
fundamental authority of the Corpus iuris. We shall see that their analysis of the ban 
came to exemplify the application of these broader jurisprudential rules. In fact, 
Joseph Canning has already used Baldus' thought on banishment as a case study for 
the territorial limitation of statute. 68 
i. Procedure 
Right from the beginning of the thirteenth century, for the next one hundred years, 
the jurists justified the procedural deviations from the ius commune inherent in 
banishment through references to the authority of custom, in this case, the general 
64 The literature on this topic is large. For an overview, see Lange and Kriechbaum, Die 
Kommentatoren, pp. 224-63. Important Anglophone studies include Woolf, Barto/us of 
Sassoferrato, Canning, Baldus de Ubaldis, and Pennington, The Prince and the Law. 
65 The following summary of the role of custom and D. l. l.9 is based on the succinct, up-to-
date analysis provided by Ryan in 'Bartolus of Sassoferrato', and more comprehensively, in 
'Political Thought'. 
66 D.l.3.32,1 (translation from Ryan, 'Political Thought', p. 445): 'Ancient custom is not 
without reason observed as if it were law, and this is the law which is said to be established 
by mores.' 
67 D.l.l.9 (adapted from Watson (ed.), Digest, vol. I, p. 2): 'All peoples who are governed 
by laws and customs observe in part their own particular law and in part law common to all 
mankind. For that which each people has established as law for itself is particular to that 
civitas and is called ius civile, as the law proper to that civitas.' 
68 Canning, Baldus de Ubaldis, pp. 127-8. 
194 
custom of all Italy;69 in the fourteenth century, they preferred to cite D.1.1.9. 70 It is 
noteworthy that the jurists merely alluded to the sources of the cities' authority to 
make law (which they did not do very often) and that these allusions remained 
largely unexplained. Although the jurists assessed the validity of, challenged, and 
sometimes even struck down civic legislation surrounding the ban, there existed a 
broad assumption that the Italian cities possessed the fundamental capacity to make 
law, and in doing so, to diverge from the ius commune. Hence it was also assumed 
that they legitimately possessed the criminal-law powers to institute capital penalties, 
a feature of merum imperium, which had accrued to them by that time but which was 
not granted to them in the scheme of the Corpus iuris. 71 These were not the 
controversial parts of the discourse on banishment. 72 
Albertus Gandinus, introducing the criminal ban in his Tractatus de maleficiis 
( composed during the late thirteenth century) provides the first sustained explanation 
69 
The earliest surviving argument for the validity of the communal ban is based on custom. 
Pillius (BIMA), Libel/us, p. 24: 'Sed quid si ille contumax, etiam multa indicta, non veniat, 
nee in ea prestanda pareat? Respondeo quod ponat eum in banno sue civitatis secundum loci 
consuetudinem que pro lege habetur, ut [D.1.3.32] ... ' Cf. p. 182, n. 17 above. 
7
° Cavalca, II bando, pp. 175-87, fails to appreciate that the jurists clearly distinguished 
between Roman law and statutory procedure, and that they maintained a strict interpretation 
of the ius commune position while allowing statute to derogate from the former. The result is 
- a tendency to conflate the jurists' discussion of statute and ius commune, and thus to 
overstate their resistance to contemporary practice. Unsurprisingly, he comes to the 
(unwarranted) conclusion that in this regard 'la dottrina di diritto comune era perplessa e 
discorde' (p. 178). 
71 
For medieval juristic debates on whether the Italian cities possessed merum imperium, see 
Pennington, The Prince and the Law, eh. 1, pp. 8-37. On the definition of imperium, see 
Berger, Encyclopedic Dictionary, pp. 493-4. 
72 
A notable exception is the early quaestio by Azo (Landsberg), Titius commisit delictum 
( cf. p. 182, n. 18 above), pp. 65-8: 'Titius commisit delictum, propter quod deportatus est a 
potestate huius civitatis ... Solutio. Dominus Azo dicit, quod nisi esset de consuetudine 
Lombardiae vel privilegio speciali principis, nulla fuerit deportatio. Sed quia ita obtinere 
hodie per totatn Lombardiam ex privilegio imperatoris constat, deportatio et bonorum 
ademptio de iure tenuit...' Strictly speaking, the question is not civic procedural deviation 
from the ius commune, but whether Italian cities possess the jurisdiction in principle to give 
capital penalties such as deportation (which here surely is an analogy for the ban). As 
Landsberg notes (p. I 05), the imperial privilege under discussion refers to the Peace of 
Constance. The only other exception is Baldus, at LF.2.28 c. Domino guerram, fol. 54va, 
who covers similar themes but rejects any argument from the Peace of Constance: 
' ... potestates terrarum possunt alios forbannire quia habent merum imperium ... et hoe de 
consuetudine generali. Alii dicunt per pacem Constantiae, sed male dicunt, quia per formam 
illius pacis requiritur quod rectores terrarum confirmentur ab Imperatore, quod de facto non 
observatur, et ideo sola consuetudo patrocinatur civitatibus ... ' 
195 
,/: 
'' 
I I 
of civic practice in terms of custom and its relationship to criminal procedure m 
Roman law: 
But today, according to general custom, what is stated concerning the annotation of 
goods and [the rule] that the accused is awaited for one year, is not observed in respect 
of contumacious persons. Instead, after someone has been officially summoned, he is 
placed under the ban of the commune and the podesta, and afterwards he is condemned 
on the basis of that accusation and ban; and this is observed in Lombardy, the Marches, 
Tuscany, and elsewhere throughout all Italy. And even though it seems that without 
joinder of issue a definitive sentence cannot be passed on the accused, the opposite is 
in fact the case, because every podesta and judge can enforce his administration of 
justice through a punitive judgement... The best reason is the opinion of Accursius, 
who notes that even without joinder of issue it is possible to proceed to the 
condemnation of an absentee by imposing the penalty of relegation or a lower one ... 
An absentee, however, cannot be subject to a penalty higher than relegation, according 
to. .. [D.48.17 .1]. Indeed, this ban, which is applied in this manner, and the 
condemnation that follows, take the place of the penalty of relegation, since such 
penalties are no longer in use today .. . And it is good that this custom is observed.
73 
For Gandinus, the rationale of modern customary deviation from Roman law 
procedure is the penalisation of the defendant's contumacy to ensure that the city · 
courts' administration of justice is not without effect in the face of such 
disobedience. While others in the thirteenth century, including Guido de Suzzara, 
still seemed to have felt at least some need to assess or question the widespread 
procedure of placing absentee defendants under the criminal ban, only to justify it on 
the basis of Italian custom, the legitimacy of that part of legal practice, as a purely 
procedural point, appears to have been largely taken for granted by the fourteenth 
73 Albertus Gandinus, Tractatus de maleficiis, rub. Quid agendum sit reo absente et 
contumace, pp. 129-30: ' Sed hodie de consuetudine general i non observatur contra 
contumaces, quod dicitur de annotatione bonorum et quod accusatus expectetur per annum, 
immo, cum fuerit citatus solemniter, ponitur in banno communis et potestatis et postea super 
ilia accusa et banno comdemnatur, et ita servatur in Lombardia, Marchia, Tuscia et alibi per 
totam Italiam, et licet videatur, quod lite non contestata non possit contra accusatum 
sententia diffinitiva proferri, tamen contra est, quia quilibet potestas et iudex potest 
iurisdictionem suam defendere penali iudicio ... et maxime secundum Accursium, qui notat, 
quod etiam lite non contestata potest procedi contra absentem ad condemnationem, 
imponendo penam relegationis et infra ... Ultra autem relegationem in absentem pena non 
infertur, ut... [D.48.17.1]. Hoe enim bannum, quod sic datur, et condemnatio, que 
subsequitur, loco pene relegationis succedit, cum ipse pene hodie non sint in usu ... et hoe 
bene de consuetudine observatur.' The same basic position can be found in Albertus' 
Quaestiones statutorum, q. 58. Cf. Ghisalberti, 'La condanna' , pp. 20-1 ; Cavalca, II bando, 
p. 179; Pazzaglini, The Criminal Ban, pp. 19, 43 , 84; Milani, ' Giuristi, giudici e fuoriusciti ' , 
pp. 622-3. Albericus de Rosate provides a similar introduction to the ban in his Quaestiones 
statutorum IV, q. 1. 
196 
century.74 This tendency can already be spotted in Guilelmus Duranti' s Speculum 
iudiciale, where the procedural principles of Roman law are brushed aside with a 
. fi 75 l fi . passmg re erence to custom. Barto us, or his part, occasionally notes the clash 
between statute and ius commune, but provides no justification for the statutory 
provisions, which he deems to be lawful. 76 In one brief comment, Baldus bases 
communal procedure on D.1.1.9 and notes that even in Roman law (as we saw in 
chapter 2) there are certain exceptional cases where an absent defendant may receive 
a capital penalty. 77 
Although the civilians generally accepted the core features of communal procedure, 
there were some initial doubts over the usual practice of immediately sending a 
bannitus, taken to be pro confesso, to be executed without the benefit of a hearing 
after he had been captured by the officials of the city. Basing their objections mainly 
on Roman procedural law, some jurists in the thirteenth century, such as Albertus 
Galeotti and Guilelmus Duranti, viewed this custom as unlawful. 78 Others, such as 
Guido de Suzzara and Albertus Gandinus, were at best ambivalent towards it, noting 
74 Caprioli, 'La Miscellanea Romana ' , has partially edited a number of Guido's quaestiones 
from the Archivio di Stato di Roma, MS. 1004 - at p. 163 (no. 43 - fols 52vb-3vb): 'Set 
numquid ban(n)um perpetuum lite non contestata poterit dari alicui, si de aliquo crimine 
fueri t accusatus; et fuerit citatus et non venerit ad precepta pot(estatis)? Et videtur quod non, 
~um ... Econtra videtur quod sic, quia .. . et maxime secundum Ac(cursium), qui dixit et 
no(tavit} quod ... Et hoe bene observat consuetudo in Lonbardia [!] tota. Guido de Suzaria. ' 
For Accursius' position, see pp. 225-6 below. 
75 Guilelmus Duranti, Speculum iudiciale, de accusatione, § 6 sequitur, in princ. §, p. 13a: 
'Et secundum leges nunquam super crimine contra absentem proceditur lite non contestata ... 
Verum tamen de consuetudine plerunque locorum absens contumaciter pro confesso 
habetur... ' 
76 Bartolus to D.48.19.5, fo l. l 98ra: 'et hoe de iure communi. sed secundum statuta Ital iae ex 
sola contumacia habetur pro confesso et habetur contumacia pro probatione legitima. Unde 
ubi sunt statuta, statuta servabis.' Cf. Bartol us to D.48.17 .1. 
77 Baldus to C.9.2.6, fol. 207va: 'Secundo opponitur [C.9.47.21]. Solutia: quod illud est 
speciale, et per illam legem possunt fundari statuta terrarum quae disponunt quod absens 
possit capitaliter condemnari, nam tale statutum est solummodo ius positivum et imo tenet 
' ' 
ut [D.1.1.9].' Cf. eh. 2, pp. 54-5. 
78 Guilelmus Duranti, Speculum iudiciale, de accusatione, § 6 sequitur, v. sed pone, p. 15a: 
'Sed pone, A. fuit accusatus de morte T. citatus non venit se defendere, propter quod 
bannitus est, et demum propter contumaciam condemnatus. nunc captus est per curiam. 
quaeritur an de beat moriri? Et videtur quod sic ... de bet eum propter contumaciam haberi pro 
confesso ... Argumentum contra, quod non debeat moriri, [D.48.19.5] , [D.48.17.1], 
[C.9.40.1] , [C.1.3(6).53(54)]. et hoe verum est. Debet ergo iudex facere quod de crimine 
constet, ut in praedicta lege [C.1.3(6).53(54)], ibi, "post legimitas probationes iuri cognitas". 
The same quaestio is found in Albertus Galeotti ' s Margarita, c. 31 De criminibus, v. pone 
quad quidam f uit accusatus de morte Titii. Cf. Cavalca, II bando, p. 235 . 
197 
a de iure/de facto divergence.79 None of the commentators in the fourteenth century 
had these qualms, and if they addressed the issue, they justified civic practice. Both 
Albericus de Rosate and Bartolus repeat the argument of Dinus de Mugello that a 
bannitus must be given a hearing upon his capture by officials unless, that is, statute 
specifically states otherwise, in which case such a derogation of the ius commune is 
valid; Albericus cites D.1.1.9 in support of this position. 80 Baldus justifies this with 
analogies from Roman law illustrating the idea that someone can be barred from 
judicial recourse on account of his contumacy.81 
Jacobus Butrigarius was the only commentator who felt the need to address this 
statutory derogation of the ius commune in greater depth. 82 He acknowledged that to 
some great doctors of the law it was 'very repugnant that someone is immediately 
decapitated after having been captured, even if the crime has not been definitively 
79 Johannes Andreae, additio to Speculum iudiciale, de accusatione, § 6 sequitur, v. quid si 
bannitus, p. 15b: 'Guid. de Suza. in quaestione statutorum mihi [sic). .. Ponit autem formam 
banni ... quod potestas contra illum contumacem tulit sententiam ad capitis truncationem ... 
condemnando ipsum pro confesso de illo crimine. ille postea capitur. adversarius petit 
executionem sententiae et banni ... captus allegans quod ultra relegationem non debuit puniri 
propter contumaciam ... solvit... eum de iure audiendum per leges hie allegatas, licet ltalica 
consuetudo servet contrarium.' Albertus Gandinus, Quaestiones statutorum, q. 59, contains a 
shorter version of the same position. It appears that Guido had some success in court with 
the de iure argument. Caprioli, 'La Miscellanea Romana', pp. 162-3 (no. 43 - fols 52vb-
3vb), includes an abridged version of the quaestio reproduced by Johannes Andreae, with the 
following closing passage: 'Dominus Guido dice bat quod bene debebat audiri .. . et 
(re)ferebat dominus Guido se quadam vice patrocinium prestitisse in civit(ate) Mantue 
cuidam, qui fuerat ban(n)itus bene per decen(n)ium... ita quod // non perdidit. ' Cf. 
Ghisalberti, 'La condanna', pp. 62-3; Cavalca, II bando, p. 12, I 79; Pazzaglini, The Criminal 
Ban, p. 53; Milani, 'Giuristi, giudici e fuoriusciti', p. 624; Vallerani, 'Diritti di cittadinanza', 
pp. 5-6. 
80 Albericus de Rosate to D.48.3.6, fol. l 75ra: 'Nota ex hac littera quod numquam inveni nee 
invenies alibi ita clare, scilicet, quod banniti non statim, cum capti sunt, puniuntur nee pro 
confessis habentur, sed quasi re integra etc., nisi aliud lege municipali caveretur. [ ... ] Dy.' 
And again, Quaestiones statutorum IV, q. 54, fol. 75ra: 'Si autem non solum bannitus sed 
per sententiam condemnatus, quod, si venerit in fortiam communis vel potestatis, debeat 
decapitari, sicut hodie fit, <licit Odof[redus], quando si talis sententia processit ex forma iuris 
municipalis, tune statim executioni mandatur, quia ius municipale derogat iuri communi, ut 
[D.1.1.9].' Dinus' position is also affirmed by Bartolus at D.48.3.6. Cf. Cavalca, JI bando, 
pp. 73 and 181 (where he incorrectly summarises the import of Albericus' q. 54). 
81 Cf. Baldus to D.24.3.47, fol. 19rb: 'contumaci potest audientia denegari, [D.4.6.26,6], et 
ratione contumaciae clauditur iudiciorum copia ... hinc est quod ille, qui fuit contumax in 
defendendo, non auditur post terminum ... hoe tenet Din. in [D.48.3.6].' Cf. Baldus to 
D.1.1.3. 
82 Cf. Cavalca, JI bando, pp. 72-3, 181-2. 
198 
ascertained'. 83 Jacobus' way of justifying this practice was to view the ban as an 
interlocutory (i.e. provisional or preliminary) ruling, which can be imposed on a 
contumacious defendant in Roman law, rather than a definitive capital sentence, 
which cannot. 84 This interlocutory ruling prohibits the bannitus from entering the 
city's territory under threat of the statutory penalty for banniti pro maleficio who are 
captured, i.e. the immediate execution of the death sentence. Therefore, if the 
bannitus returns under his own volition, he will be punished not because he had been 
subject to a previous definitive capital sentence given in absentia, which would be 
illegitimate according to the ius commune, but because he disobeyed the obligations 
of his ban. According to Jacobus, this conforms to the principle of Roman law that 
even capital penalties can be imposed on exiles who break the conditions of their 
confinement without the necessity for a trial. Thus, he who is relegated to an island 
for life, having been caught after escaping, is deported to an island; an escaped 
deportatus is executed.85 For Jacobus, the ban is therefore a form of annotation. 
Their essential similarity is that both are interlocutory measures deployed against the 
contumacious rather than definitive penal sentences. 86 
ii. Outlawry 
A central issue considered by the jurists was whether civic statutes could withdraw 
legal protection from banniti such that they were liable to be murdered with legal 
impunity by private individuals. All of the civilians who discussed this question, with 
the possible (but very unlikely) exception of Jacobus de Arena, accepted and 
83 Jacobus Butrigarius to C.9.40. l, fol. 11 Ova: 'et certe vidi magnos doctores .. . quibus 
multum repugnat quis statim eo capto decapitatur etiam crimine non liquidato. ' 84 Jacobus Butrigarius to auth. item nulla, post C.1.3(6).2, fol. 16vb: ' ... bannum imponitur 
per interlocutoriam et ex contumacia, ut [C.7.57.4]. Sed relegationes et deportationes fiunt 
per diffinitivam sententiam et ex crimine, et ideo non equiparantur.' 
85 Ibid.: ' ego puto quod operetur hoe in casu isto, quod, si contingat eum non servare 
bannum, quod possit affici maiori pena, nam non debet venire in territorio unde bannitus est. 
et si veniat, tanquam inobediens potest affici maiori pena, ut dicimus in relegato, ut 
[D.48.19.8, 7]. sed si veniret super districtu isto contra formam banni, puniretur secundum 
statutum, non ut de crimine convictus, sed quia bannum non servavit.' Cf. D.48.19.4; 
D.48.19.28,13. 
86 Jacobus Butrigarius to C.9.40.1, fol. 11 Ovb: 'Et ideo equipara bannum annotationi.' To 
C.7.66.2, fol. 60rb: 'Et attende quod annotatio non est pena criminis sed contumacie, sicut et 
bannum non est pena criminis .. .' 
199 
justified contemporary practice.87 The problem of Jacobus is a thorny one because 
doubts have been raised over the extent to which material can be attributed to him 
('lac. de Are.') rather than to another mysterious jurist with the alleged signature of 
'lac. de Aret.' (Jacobus de Aretino, i.e. Arezzo ), which appears in fourteenth-century 
reports of older juristic opinion. While most historians have followed Savigny's 
opinion that these signatures refer to the same person, in Cavalca's view, the 
attribution of material by later jurists in the field of banishment is sufficiently 
consistent that it is possible to associate two distinct sets of thought with Jacobus de 
Arena and Jacobus de Arezzo.88 It seems, however, that Cavalca's evidentiary base 
was insufficiently broad, since a wider examination of the sources reveals no such 
consistency, at least in the printed editions at my disposal. 89 In the material on the 
ban, the signatures 'lac. de Are.' and 'lac. de Aret.' are used too often by too many 
different jurists to describe the same body of thought, including that which has been 
traditionally ascribed to Jacobus de Arena and for which he is well known, to 
warrant the conclusion that these are two different jurists. 90 In what follows, it is 
therefore assumed that 'lac. de Are.' and 'lac. de Aret.' refer to the same person. 
It is usually one section of Jacobus de Arena's Tractatus de bannitis that is adduced 
for his alleged opinion that statutes permitting the killing of banniti are unlawful.
91 
87 Cf. Ghisalberti, 'La condanna', pp. 21-2, 69-74; Cavalca, JI bando, pp. 97-8, 206-13; 
Pazzaglini, The Criminal Ban, p. 60; Vallerani, Diritti di cittadinanza', pp. 7-8. 
88 Cavalca, JI bando, pp. 13-15 and p. 223 ff. Cf. Savigny, Geschichte des Romischen Rechts, 
vol. 5, p. 407 (n. f); Lange and Kriechbaum, Die Kommentatoren, p. 435. 
89 Complete certainty will not be had until someone examines the manuscripts. 
90 E.g. Jacobus' quaestio Statutum est in civitate quod bannito ratio non fiat is attributed by 
Johannes Andreae to lac. de Aret. (additio in Speculum iudiciale, de accusatione, § 6 
sequitur, v. quid si bannitus) but to lac. de Are. by Albericus de Rosate (Quaestiones 
statutorum IV, q. 89) and Baldus (D.4.5.5) - see pp. 234-6 below for further details on this 
quaestio. Jacobus is famous for his criticism of the Gloss at D.36.1.18(17),5 (I. Ex facto), 
where he argues that banniti should not be compared to deportati. This opinion is attributed 
to lac. de Are. by Albericus de Rosate (D.36.1.18(17),5; C.6.24.1), Bartolus (D.4.5.5; 
D.36.1.18(17),5), and Baldus (D.4.5.5; D.28.2.29,5), but to lac. de Aret. by Johannes 
Andreae (additio to Speculum iudiciale, de cessione actionis, § 1 sciendum, v. illud etiam 
quaeritur) and Baldus (C.6.24.1 ). Bartolus associates this opinion with both lac. de Are. and 
lac. de Aret. in a single section of his quaestio Lucanae civitatis (p. 205b); this is also the 
case in his discussion at D.24.3.47 of Jacobus' view on judicial defence. Baldus attributes 
Jacobus' Tractatus de bannitis to lac. de Aret. at C.6.24.1. It is clear from his report that this 
is indeed the same treatise. 
91 Ghisalberti, 'La condanna', pp. 70-1; Cavalca, JI bando, pp. 14, 212-3; Pazzaglini, The 
Criminal Ban, p. 60; Milani, 'Giuristi, giudici e fuoriusciti', p. 620. 
200 
Although the printed edition is subject to at least one interpolation,92 the authenticity 
of most of the treatise seems beyond doubt given that the vast majority of the 
material and the opinions found therein are · reproduced or attributed to Jacobus by 
other jurists.93 Jacobus opens the discussion of this particular issue by stating that, 
while statutory law frequently allows the murder of banniti, 'de iure I say that a 
banished person cannot be harmed'. The crux of his argument is that no matter 
whether banniti are considered analogous to deported or relegated persons, neither 
lose legal protection in Roman law.94 Moreover, civic statutes that expressly permit 
this 'are not in accordance with the rigour of the law' ('non valent de iuris 
subtilitate'); he cites laws conveying the idea that such statutes would be immoderate 
and beyond the scope of what can rightly be decreed by a city council. 95 Jacobus' 
argument appears to have been primarily aimed at Guido de Suzzara, whom he 
reproves in the following section. It was Guido's rather extreme contention that 
banniti can be killed with impunity merely because they lose the citizenship and 
92 Jacobus de Arena, Tractatus de bannitis, fol. 356ra: 'Solutio: dicunt Rainerius, Oldradus, 
et quamplures moderni quod ... ' The reference to these later jurists caused Marcello, 'Opere', 
p. 856, to raise doubts over the treatise's authenticity. Cortese, II diritto, p. 280, Ghisalberti, 
'La condanna', and Milani, 'Giuristi, giudici e fuoriusciti', all accept the treatise as authentic 
without a specific discussion of the issue. 
93 The opening of Jacobus' treatise on the false comparison bannitus-deportatus, and the 
disagreement with Guido de Suzzara on the licit killing of banniti, are summarised and 
attributed to Jacobus by Baldus (C.6.24.1). Bartolus (D.24.3.47) and Baldus (C.6.24.1) 
attribute to Jacobus the opinion on judicial defence found in the treatise. The quaestiones 
whether a bannitus can effectually cede his rights and actions, whether a bannitus can sue 
hereditario nomine for a non-bannitus and vice versa, and whether a bannitus can act as a 
procurator for a non-bannitus and vice versa, are all closely paraphrased (without the 
anachronistic reference to later jurists) and attributed to Jacobus by Albericus de Rosate 
(Quaestiones statutorum IV, qq. 4, 50, and 63). The largest section of the treatise consists of 
Jacobus' quaestio Statutum est in civitate quod nullus bannitus possit esse consul, 
reproduced extensively and attributed to Jacobus by Johannes Andreae (additio in Speculum 
iudiciale, de accusatione, § 6 sequitur, v. quid si bannitus) and Albericus de Rosate 
(Quaestiones statutorum IV, q. 91 ). It is not recorded by Bellomo, Quaestiones. 
94 Jacobus de Arena, Tractatus de bannitis, fol. 355va: 'dico ut plurimum per leges 
municipales permistum quod impune possint offendi... et ita solent dicere statuta. De iure 
vero dico bannitum offendi non posse. .. quia, sive comparentur relegatis sive etiam 
deportatis, impune non poterunt offendi, quia relegatos vel deportatos non Iicet occidere, 
cum non sit de illis casibus in quibus possunt impune occidi ... ' 
95 Ibid., fol. 356ra: 'Imo, quod plus est, talia statuta hoe expresse permitentia non valent de 
iuris subtilitate, [D.50.9.4: 'Ambitiosa decreta decurionum rescindi debent, sive aliquem 
debitorem dimiserint sive largiti sunt. '], [C.1.3(6)26: 'ideo patientiam et modestiam suam, 
quam leges nostrae et publica disciplina ... exposcit... ' ]. ' 
201 
i I I 
Ii 
civic rights of their city akin to deported persons.
96 Jacobus does well to point out 
that this is an incorrect interpretation of the law on deportatio
n. 
For Cavalca, the most important piece of evidence for the 
authenticity of Jacobus' 
treatise, and for the distinction between Jacobus de Arezzo
 and Jacobus de Arena, 
was that Jacobus de Arena's position on the killing of bann
iti in his treatise could 
also be found both in his surviving other work and amongst t
he opinions attributed to 
him by later jurists, whereas Jacobus de Arezzo was consistently associated wi
th the 
opposing side in this debate. 
97 In fact, the contrary is the case. The argument of the 
treatise is not repeated anywhere (which is striking, given that it was argu
ably such 
an exceptional deviation from communis opinio ). Instead, one can find at 
least two 
examples in Jacobus' commentaries where he accepts the
 lawfulness of statutes 
permitting the killing of banniti,
98 and there are several reports from the fourteenth 
century that associate 'lac. de Aret.' as well as 'lac. de Are.' 
with that position.
99 We 
will come across some of these in the following discussion. 
96 Ibid.: 'Alii tamen dicunt, et male, quod bannitos licet impun
e occidere quasi sint apostatae, 
arg. [D.48.19.17] et [D.32.1.1,3], et apostata idem est sine civitate 
et iurisdictione, quasi 
civilia sint eis denegata, et ita consuluit Guido de Suza
. , et male, ut dixi. '. The word 
'apostatae ' is surely a corruption of the Greek word a1t6A10E<
; found both at D.48.19.17 and 
D.32.1.1 ,3 to describe deported persons; this is confirmed 
by Baldus' rendering of Guido's 
opinion at C.6.24.1 - see p. 203, n. 100 below. 
97 Cavalca, II bando, pp. 13-15 and p. 223 ff. Cavalca provides 
two alleged instances outside 
the treatise where Jacobus argues, or is said to argue, again
st the notion that banniti can be 
killed with impunity; these are, however, wholly inaccurat
e. Firstly, Jacobus grants banniti 
the capacity of judicial defence. Secondly, banniti cannot be killed by priva
te persons after 
the city's officials have captured them. These, however, are 
completely separate issues. 
98 Jacobus de Arena to D.48.8.3,6, fol. 224rb (granted that th is is 
non-committal): 
'Transfugas. argumentum contra bannitos, ut ubique licite 
occidantur.' His commentary to 
C.9.2.3, fol. 59va, however, leaves little doubt: 'Si
 vero opponit peremptoriam 
[exceptionem], ut quia dicit se interfecisse bannitum ... non debet admi
tti accusatio, quia data 
est cuilibet interficiendi licentia, quia idem est bannire quam
 dicere quod quilibet possit eum 
offendere. [ ... ] lac. de Are.' Cavalca, II bando, pp. 42 and 92, is 
fully aware of these 
passages but does not view them as problematic for his i
nterpretation. NB. Although it is 
likely that Jacobus' surviving commentaries - printed 
under the title Commentarij in 
universum ius civile - are largely authentic, they are also e
vidently interpolated at times, 
although foreign materials are often marked as such. C
f. Lange and Kriechbaum, Die 
Kommentatoren, pp. 441-3. 
99 Most importantly, Jacobus' quaestio Statutum est in civitat
e quod bannito ratio non fiat 
(attributed by different jurists both to lac. de Are. and lac. de Aret. - seep. 20
0, n. 90 above) 
affirms the lawfulness of statutes permitting the killing 
of banniti - see p. 206, n. 112, 
below. For further details on this quaestio, see pp. 234-6
 below. Baldus (at D.4.5 .5 and 
D.24.3.47) cites lac. de Are. in support of that argument - see p. 208-9
 below. In another 
quaestio (Jn statuto vel consuetudine alicuius civitatis), Jacobus argues 
that the exceptio 
202 
The result could be interpreted as a major tension in Jacobus' thought. A 
more 
convincing interpretation would be to understand his empha
sis on the argument de 
iure or de iuris subtilitate to refer to the position of the i
us commune. De iure 
( communi), a bannitus cannot be killed because there is no analogous figu
re in the 
Roman law that could justify this, and the whole notion that contuma
cious 
defendants are punished in this way is alien to it. Therefore
, judging purely on the 
basis of the ius commune, if there were no statutory law regar
ding this matter, banniti 
could not be harmed, but that does not preclude the lawfulnes
s of statutes to sanction 
such a penalty in express terms. This would explain why la
ter civilians never cited 
Jacobus as the one jurist who opposed the communis opinio on the licitness of k
illing 
banniti. And indeed, this is how Baldus understood Jacob
us' position when he 
reported the controversy between the latter and Guido de Suz
zara. 100 
Baldus' interpretation dovetails with the consensus of the fo
urteenth century that by 
default, according to the ius commune, banniti could not be h
armed without express 
statutory regulations to that effect.
101 Albericus de Rosate and Bartolus emphasise 
that even if the bannitus is compared to the deportatus, and
 he loses Roman civil 
rights as a result, he is still protected by the injunction of natural law (at D.1. 1.3
) that 
human beings should not cause harm to each other. The only
 analogy in Roman law 
potentially detrimental to banniti in this regard is the hostis-
transfuga. That figure, 
however, is too exceptional on account of the nature of hi
s crime - perduellio, the 
most extreme case of Iese majesty - to serve as a default point of compariso
n for 
banniti in general.
102 This makes sense: as will become increasingly clear during t
he 
banni nullifies an accusation of murder, which reflects his 
commentary to C.9.2.3 (n. 98 
immediately above) - seep. 212, n. 132, below. 
100 
Baldus to C.6.24.1, fol. 70vb: 'Sed hie quaeritur nunquid ba
nniti possint impune offendi. 
dixit Gui. de Suza. quod sic, ex etymologia vocabuli. Nam 
sunt apolides, id est, sine civitate 
et sic eis auxilia civilia interdicta sunt... Sed contrarium es
t verum, quia nee iure communi 
nee municipali hoe cavetur. Item nee deportati possunt imp
une offendi, cum non perdunt ea 
quae sunt de iure naturali, et offensio personae est contra iu
s naturale ... secundum lacob. de 
Are. Sed si statutum diceret expresse quod possent imp
une offendi, tune tales banniti 
censentur transfugae et hostes civitatis unde sunt bann
iti ... ' This is how one should 
understand Baldus' brief comment at D.24.3.47, fol. l 9rb: 'S
edan de iure communi bannitus 
po test impune offendi? die quod non, nisi sit diffidatus ... ' 
101 Cf. Ghisalberti, 'La condanna', p. 73; Cavalca, II bando, pp. 9
7-8, 213. 
102 
Bartol us to D.1.1.3: 'Nunquid de iure comm uni licet nostro
s bannitos offendere. Pro hoe 
facit quia banniti nostri temporis aequiparantur deportatis.
. . Sed ego video quod deportati 
non perdunt ea quae iurisgentium sunt, licet perdant ea qu
ae sunt iuris civilis ... Sed si hoe 
est, hie est eis reservata defensio, ergo est illicita offens
io. In contrarium videtur facere 
[D.48.8.3,6], ubi dicitur quod ille qui transtulit se apud hostes de iure c
ommuni potest occidi . 
[ ... ] quia nil aliud sunt banniti nisi hostes civitatis suae. [ ... ] Sed e
go dico quod de iure 
203 
course of this chapter, although the criminal ban, like the imposition of hostis status, 
could be used as a penalty for treason (pro delicto ), it was primarily a coercive 
procedural device employed by the courts against contumacious defendants in 
ordinary criminal matters (pro contumacia); hence, those banished for political 
crimes were only a subgroup of banniti pro maleficio. And because this was the main 
purpose of the ban, Bartolus further argues that if statute does not regulate the legal 
status of banniti pro maleficio, one should by default revert to the mechanism of 
annotation, the ius commune procedure for contumacy in criminal cases.
103 Another 
mismatch between the category of bannitus and hostis was that there were various 
types of banniti in statutory law who were not subject to the measures of the bannum 
pro maleficio in its most severe form, and who could thus not be harmed with 
impunity like the hostis (e.g. those banished for minor crimes). 104 For this reason, as 
we shall see shortly, Bartolus argues that those banniti whom statute expressly 
permits to be killed are indeed hastes, but this does not mean that all contumacious 
defendants are hastes; those subject to a less severe ban should rather be compared to 
'vile persons'. 105 
Almost every argument concerning the lawfulness of statutes expressly permitting 
the killing of banniti was already in play by the later thirteenth century, and very 
little change can de detected until Bartol us' reassessment of the tradition. Before 
Bartolus, this quaestio went as follows. 106 The central argument opposmg 
communi non licet, quia sunt duae quaestiones, quia aliud est dicere nunquid bannitus possit 
occidi, aliud est dicere nunquid eo ipso quod est bannitus lex possit inducere occisionem. et 
haec non est quaestio mea, sed prima. et dico quod non obstat § ille, quia ibi est casus 
specialissimus propter maximum praeiudicium quod infert propter revelationem secretorum, 
etc.' Cf. Albericus de Rosate to D.48.19.1 7 and Quaestiones statutorum IV, q. 16. 
103 Bartolus, Lucanae civitatis, p. 206b: 'Sed pone statuto nihil continetur de exbannitis, quid 
erit de iure communi? dico quod de iure communi potest quis exbanniri. Nam annotatio, 
quae olim fiebat, erat quoddam bannum.' 
104 f h C . c . 1, p. 20. 
105 Bartolus, Lucanae civitatis, pp. 205a-6b: 'Dico talem exbannitum, qui secundum formam 
statuti impune offenditur, quod vere, recte, et proprie posset dici transfuga civitatis suae. [ ... ] 
Quid ergo dices de exbannito qui non potest impune occidi .. . ? Respondeo quod talis 
exbannitus non est transfuga nee hostis civitatis. [ ... ] quia non contulit se in numero hostium, 
quod patet, quia civitas non habet eum pro hoste ... aequiparatur ergo talis personis vilibus ... 
Et si aliud diceres, oportet te dicere omnem contumacem esse transfugam et hostem civitatis 
suae, quod esset iniquissimum.' Cf. Ghisalberti, 'La condanna', p. 43; Cavalca, II bando, p. 
96. 
106 The most significant contributions to this topic are: Albertus Gandinus, Tractatus de 
maleficiis, rub. De bannitis, qq. 2 and 3; Johannes Andreae, additio to Speculum iudiciale, 
de constitutionibus, v. quid si extraneus; Albericus de Rosate to D.48.19.17 and Quaestiones 
204 
contemporary statutes was that the natural law prohibited murder at D.1.1.3 and so 
did God's equivalent commandment in the Decalogue. These are eternal, higher legal 
norms and it is beyond the capacity of any civil law, such as a city's statute, to 
change them in a fundamental way. 107 Nevertheless, the jurists provided a range of 
counterarguments to demonstrate that the cities both could and should authorise the 
murder of banniti in their legislation. In this context, the cities' general capacity to 
legislate in deviation from Roman procedural law was sometimes based on the force 
of custom, but more often the jurists referred to the efficacy of D.1 .1.9: the emperor, 
since he himself conceded in this law the authority to every Italian city to legislate, 
has sanctioned their statutes, which take priority over the ius commune within the 
legislating city's territory. 108 Next, the jurists argued that, while civil laws, including 
local statutes, could not abrogate natural and divine norms, they could of course 
derogate from them ex iusta causa - in the words of Albericus de Rosate: 
Wherefore, from what has been said, some also want to argue that even a statute by 
which a banished person can be killed with impunity is not lawful, as if it were against 
the divine and natural law, which may not be abolished by the civil law. But one 
should argue, granted that a statute cannot completely invalidate the divine and natural 
law, that it can, however, distinguish and interpret it by excluding the killer from the 
penalty and sentence on account of public utility and in order to strike fear into the 
hearts of criminals. [ ... ] Therefore, such a statute is valid ... not least because the ius 
commune grants this faculty in many instances ... 109 
statutorum IV, q. 17; Bartolus, Lucanae civitatis; Baldus to D.4.5.5, C.1.1 ( 4).1, and 
C.6.24.1. 
107 Albertus Gandinus, Tractatus de maleficiis, rub. De bannitis, q. 2, p. 131: 'Sed pone 
contineri in statuto civitatis, quod banniti pro maleficio impune possint offendi et specialiter, 
quod possint occidi; queritur, an valeat tale statutum? Et primo videtur quod non, tamquam 
factum contra ius divinum. Continetur enim in divino precepto: "non occides, non 
mechaberis". Et ius divinum, quod est excellentius, non debet posse tolli per dictum 
statutum ... quod est debilius. [ ... ] Praeterea tale statutum est contra ius gentium et naturali 
instinctu introductum, ut unus homo alii non insidiatur, ut [D.1.1.3], et ius civile, quod est 
mutabile, mu tare vel toll ere iura naturalia seu gentium non po test, que sunt immutabilia ... ' 
108 Ibid.: 'E contra videtur, quod tale statutum valeat et teneat de iure ... Nam cum princeps 
concesserit unicuique civitati iura posse condere inter suos, ut [D.1.1.9], ipse imperator 
dictam facere legem videtur...' Ibid., q. 3, p. 416: ' ... cum statutum vincat legem in eo loco, 
ubi factum est, et maioris auctoritatis sit, quam lex general is in eo municipio ... ' 
109 Albericus de Rosate to D.48.19.17, fol. 203rb: 'Unde ex praedictis etiam quidam dicere 
voluerunt quod etiam statutum quod bannitus possit impune occidi non valet, quasi sit contra 
legem divinam et naturalem, quae tolli non potest per ius civile. sed dicendum est quod, licet 
non possit statutum tollere ius divinum vel naturalem directo, potest tamen distinguere et 
interpretari, tollendo poenam et sententiam occidenti propter publicam utilitatem et terrorem 
malefactorum. [ ... ] valet ergo tale statutum ... et quia hoe Jex civilis in multis casibus 
concedit... ' 
205 
Albericus correctly points out that the Roman law itself, in fact, provides a range of 
examples in which it derogates from higher norms by granting to a private person the 
right to kill another with legal impunity. As we know, a military deserter to the 
enemy (transfuga) could be killed by anyone, and a husband, in certain defined 
circumstances, could kill an adulterer caught in jlagrante with his wife. The same 
rule applies to lethal acts of self-defence, especially against robbers, brigands, and 
deserters, as well as to a master who kills his fugitive slave if the latter resists 
capture. 110 And if the Roman law deems it reasonable to derogate from higher norms 
in these situations, given its exemplary authority, then, in principle, it is also 
reasonable for statute to do so if it has just cause. 111 The jurists, including Jacobus de 
Arena, emphasised causa inherent in the idea that disobedience to the courts should 
be punished in this way. 11 2 As we saw above, Albertus Gandinus explains that the 
rationale of the ban lay in ensuring that the administration of justice is not without 
effect in the face of contumacy. The ban thus serves as a substitute penalty to prevent 
criminals profiting from the 'defect in public retribution' caused by the inability of 
the courts to impose the ordinary penalty. 113 This is what Albericus de Rosate refers 
to when, in the passage quoted above, he speaks of striking fear into the hearts of 
criminals: through the ban, they are duly punished, potential criminals deterred, and 
110 Albertus Gandinus, Tractatus de maleficiis, rub. De bannitis, q. 2, p. 131: 'Quid plura? 
Ipse etiam imperator concedit, quosdam homines posse impune occidi variis de causis: ut 
[D.48.8.3,6] et [C.9.9.4] et [C.3.27 .1 ] et [C.1.12(15).4(3)].' 
111 Ibid., q. 3, p. 416: ' ... nam certo certius est quod lege dari potest privato occidendi 
auctoritas, sicut pro hoe sunt multe leges ... Cum igitur Jex municipalis vim legis habeat et ius 
civile vocetur, ut [D.1.1.9] ... quare hanc auctoritatem inter suos prestare non potest non 
video ... ' 
112 Johannes Andreae, additio to Speculum iudiciale, de constitutionibus, v. quid si 
extraneus, p. 77a: 'Ad contrarium inducit lac. de Aret. quod hoe fit in poenam contumacium. 
praecipue autem est omnis legis intentio magistratibus et iudicibus obedire ... Rationi igitur 
congruit tale statutum in poenam contumaciae et ad terrorem delinquentium .. .' Johannes 
Andreae explains the origin of this material at rub. de accusatione, § 6 sequitur, v. quid si 
bannitus, p. 19b: 'Habetur alia et Jaco. de Aret. quadrimembris quaestio. [ ... ] Prosequitur 
autem prius an talia statuta valeant, de qua dicam infra, de constitutionibus, ante finem.' This 
is Jacobus' quaestio Statutum est in civitate quad bannito ratio non fiat - see pp. 234-6 
below for further details on this quaestio. Martinus de Fano (d. ea. 1272) in a quaestio -
recorded by Albertus Gandinus, Tractatus de maleficiis, rub. De bannitis, q. 3, and by 
Johannes Andreae, additio to Speculum iudiciale, de constitutionibus, v. quid si extraneus -
used the element of disobedience as causa to justify that a bannitus pro maleficio can be 
killed even if the ordinary penalty for his alleged crime was less severe. NB. No jurist 
provides an incipit, nor is it given as one of Martinus' quaestiones by Bellomo, Quaestiones. 
Cf. M. Semeraro, 'Martino de! Cassem da Fano', DBGI, vol. 2, pp. 1291-2. 
113 Albertus Gandinus, Tractatus de maleficiis, q. 17, p. 145: ' ... causa, quare ille bannitus 
impune poterat occidi, que erat propter defectum publice vindicte, que in eum irrogari non 
poterat...' 
206 
respect for the law is upheld. Ultimately, iusta causa is present because public utility 
(publica utilitas) - Johannes Andreae speaks. of the common good ( bonum commune) 
and public safety (salus communitatis) - is served in this way. 11 4 
This is a notable attempt by the civilians to justify the suspension of natural norms 
and to give some definition to just cause and public utility, something that was 
almost entirely absent from their discussion of deportation and the imperial hostis. 
As they stand, their considerations are coherent enough, yet remain relatively 
shallow because there is no explanation as to why the extreme penalties of the 
criminal ban are reasonable measures in the fight against contumacious criminals. 
Their argument - outlawry is justified in the name of public utility because the 
effective administration of justice requires the contumacious to receive a substitute 
punishment - could be applied to any measure, whether more lenient ( e.g. ius 
commune annotation) or even more extreme. Here too we can observe the same 
conceptual weakness in civilian natural rights theory that became evident in the 
previous two chapters: there is no rule specifying the severity of measures to be 
taken in the service of public utility, and therefore no rule concerning the extent to 
which higher norms can be violated. Publica utilitas and iusta causa are such vague 
concepts that we remain in the realm of casuistry. 
The quaestio tradition on the licit killing of banniti also makes clear that the civilians 
were aware of the relevance of the hostis-transfuga right from the later thirteenth 
century. This was one of several instances in the Roman law where private persons 
could kill an individual with legal impunity, and therefore, the argument went, it was 
also perfectly legal for statutes to make similar arrangements regarding banniti. The 
important point to remember is that in this argument the hostis was merely one 
authoritative example used to support the general notion that civil laws could 
derogate from higher norms in this way; the argument was not that the bannitus 
could be killed be_cause he was, or was analogous to, the hostis. 
Prior to Bartolus, one can only find a few scattered hints pointing towards this latter 
use of the hostis analogy. One quaestio included in Albertus Gandinus' Tractatus de 
maleficiis asks whether a private person can kill a bannitus after communal officials 
114 Johannes Andreae, additio to Speculum iudiciale, de constitutionibus, v. quid si 
extraneus, p. 77b: ' ... quod occisio banniti vel malefactoris licite fit in quantum ordinatur ad 
salutem communitatis ... et valet ad bonum commune.' 
207 
have captured him. 115 Gandinus twice uses the analogy of the hostis - amongst many
 
others - in order to argue against the proposition. Firstly, he argues (correctly) that in 
Roman law a hostis who has been captured may no longer be killed by a private
 
person. 11 6 Secondly, the bannitus is only treated as a hostis because he does not
 
submit to the authorities; but once he is captured, that substitute penalty is no longe
r 
necessary, since the authorities can now administer justice themselves. 
117 Although it 
is true that the analogical link to the hostis could be taken to justify the banished 
person's lack of legal protection in general, the specific terms of the quaest
io 
actually address the distinct point about his legal position following capture. The
 
general force of the analogy is therefore questionable: strikingly, Gandinus does no
t 
use the hostis in this way when he is defending statutory provisions permitting the
 
killing of banniti. 
Before Bartolus, I can only find a single comment that answers the specific question
 
whether, statute permitting, a bannitus can be killed with impunity (prior to capture) 
by comparing him to the hostis.
118 If a report by Baldus at the L Amissione (D.4.5.5) 
is to be believed, it was the contention of Jacobus de Arena that the bannitus, 
because he injures the respublica through his contumacy, is a hostis and rebellis of 
the respublica, and could for this reason be killed by anyone. 
119 Although it is 
115 Cf. Cavalca, II bando, pp. 90-1; Pazzaglini, The Criminal Ban, p. 61 . Cavalca, II bando
, at 
p. 92, erroneously asserts that an equivalent quaestio can be found in Jacobus d
e Arena 's 
Tractatus de bannitis. 
11 6 Albertus Gandinus, Tractatus de maleficiis, rub. De bannitis, q. 17, p. 144-5: 'Nam Jex 
<licit, quod hostem licet interficere, ut [D.48.8.3,6], tamen postquam captus est, non potest 
occidi, ut [Inst.1.3 ,3] et [D.1.5.4,2] ... ' Kantorowicz believes this and the following section of 
the quaestio to be taken from the work of Johannes Fasolus (d. ea. 1286) - cf. C. Bukowska 
Gorgoni, 'Fagioli, Giovanni', DBI, pp. 166-70. 
117 Albertus Gandinus, Tractatus de maleficiis, rub. De bannitis, q. 17, p. 14 7-8 : 'Nam qui 
est in banno et in sua libertate, est inobediens in non veniendo et est factus hostis n
on parens 
nee se capi perrnittens; hunc talem perrnissum est capere et occidere, quia
 talis sua 
contumacia auget penam ... Sed cum est apprehensus et in communi potestate, c
essat talis 
contumacia ... cuius causa perrnittebatur occidi...' 
118 According to Cavalca, II bando, p. 93, Baldus, at D.4.5.5, reports that Martinus de F
ano 
justified the lawfulness of killing banniti by comparing them to hastes. This report refers to a 
quaestio extensively reproduced by Albertus Gandinus, Tractatus de maleficiis, rub. De 
bannitis, q. 3 (see p. 206, n. 112 above), where it is clear that Martinus did not use this 
analogy. Baldus therefore made this point in his own voice. 
11 9 Baldus to D.4.5 .5, 1, fol. 249vb: 'Item, quia magistratus repraesentat rempublicam,
 et ille 
qui offendit magistratum offendit rempublicam, quia magistratus est persona rep
raesentata. 
unde, quia fuit contumax, est rebellis et hostis reipublicae, ut hie, et proprie iste qu
i interficit 
bannitum videtur exequi poenam in statuto contentam, et commissa est sibi
 executio, 
secundum la. de Are.' 
208 
reasonable to doubt the accuracy of Baldus' report - not least due to the passage's
 
suspiciously overt Bartolist tone - it does receive some degree of corroboration
 
through Johannes Andreae's report of Jacobus' quaestio Statutum est in civitate quod 
bannito ratio non fiat. 120 The question is whether private individuals may damage the 
banished person's property with legal impunity. Amongst a number of analogies and
 
arguments for the proposition, Jacobus briefly states that because of his
 
disobedience, the bannitus is disavowed ('diffidatus') as a hostis, by example of the 
I. Amissione; and just as a hostis can be injured in person, so he can be injured in his 
property. 121 I suspect that Baldus took this argument from its narrower context in
 
Jacobus' quaestio and presented it in isolation at the L Amissione, emphasising its
 
broader significance in order to adduce an early authority for the Bartolist position
. 
Further arguments for this explanation will be provided in our discussion of Jacobus
' 
systematic thought below. 122 
One should therefore not make too much of these three examples. Their use of the
 
hostis - saving the possible (though by no means certain) exception of Jacobus de 
Arena's argument as reported by Baldus - is purely instrumental, adduced as part o
f 
a range of analogies for the solution of certain narrowly defined legal problems tha
t 
do not address the principal question whether it is generally lawful to kill banniti. To 
use a distinction introduced at the beginning of this chapter, this is a casuistic rathe
r 
than systematic use of analogy; it is not the basis of a broader argument about the
 
underlying status of banniti. Indeed, during the course of this chapter we will 
encounter further casuistic applications of the hostis analogy in other contexts.
 
Importantly, before Bartolus, these applications remained casuistic, and the civilians
 
did not defend the lawfulness of contemporary legal practice because they viewed
 
the bannitus as generally or fundamentally analogous to the hostis. We will return to 
this point in due course. 
A general and fun~amental (that is, systematic) equation between bannitus and hostis 
can first be found in Bartolus' own thought on this topic, which is best presented in
 
his quaestio Lucanae civitatis, part of which we already discussed in respect o
f 
120 See pp. 234-6 below for details on this quaestio. 
121 Johannes Andreae, additio to Speculum tudiciale, de accusatione, § 6 sequitur, v. quid si 
bannitus, p. 19b: 'Sed ad hoe respondetur quod bannitus, qui non obtemperat ius d
icenti, 
videtur diffidatus ut hostis, [D.4.4.5,1]. .. Et si diffidatus in persona, fortius in rebus ... Et hae 
poenae personales et reales recte contumacibus infliguntur. .. ' Cf. Cavalca, II bando, p
p. 92-
3. 
122 See pp. 234-6 below. 
209 
imperial banishment. 123 We will discuss this quaestio fully in the n
ext section of this 
chapter. For now, we will only consider that part in which Bar
tolus provides the 
basis for the institution of the communal ban, and in particular
 the idea that the 
bannitus may be lawfully killed with impunity. In this section, B
artolus first states 
that a bannitus pro maleficio should properly be considered a hostis-transfuga of h
is 
city because they share two essential characteristics. Firstly, both 
are disobedient to 
those who have jurisdiction over them. 1
24 But because there are different types of 
banniti in ordinary criminal and civil proceedings, Bartolus clari
fies, as explained 
above, that not every contumacious defendant, not every type of b
annitus, is a hostis 
of his city. For that to be the case, another condition has to hold tru
e, namely, that the 
city has declared war on the bannitus by allowing him to be kill
ed with impunity; 
only then is he truly a hostis. 
125 But one might say in opposition that hastes are only 
those who are enemies of the Roman people, not of any city 
within the empire. 
Properly speaking, Bartolus replies, this is true. Nevertheless, eac
h city may have its 
own hostes. 126 He imports the concept of the internal declaration
 of war from the 
universal to the local level: 
For just as those subjects whom the [Roman] senate judges to be hastes are enemies
 of 
the Roman people, as at [D.4.5.5,l] ... so too those subjects whom the commune judg
es 
to be hastes are the enemies of the city. And by banishing them, [the commu
ne] judges 
them to be its hastes ... 
127 
123 Ch. 3, pp. 165-71. See p. 165, n. 254 for details on the quaestio. 
124 Bartolus, Lucanae civitatis, p. 205a: ' Dico talem exbannitum, 
qui secundum formam 
statuti impune offenditur, quod vere, recte, et proprie posset d
ici transfuga civitatis suae. 
Item, potest dici hostis civitatis suae. et hoe probo ex forma co
nstituendi sic. Illi dicuntur 
transfugae qui ab his sub quorum imperio sunt, desistunt et in num
ero hostium se transferunt, 
ut [D.4.5.5,1 ] ... Sed exbannitus ideo ponitur in banno quia inobediens sit 
his sub quorum 
imperio et iurisdictione est, et in numero aliorum exbannitorum
 ad hostes civitatis suae se 
transfert.' 
125 Bartolus, Lucanae civitatis, p. 205a: 'Item hoe probatur ex effectu
 statuti, nam licitum est 
transfugam impune occidere, ut [D.48.8.3,6], et eodem modo licet iuste exba
nnitum. [ ... ] Sed 
talibus exbannitis civitas indixit helium, permittendo eos im
pune occidi, et statuendo 
officiales et alios ad eos capiendos et puniendos. ergo sunt hostes
 civitatis.' 
126 Ibid.: 'Sed tu potes instare, hostes sunt quibus populus Romanus
 helium publicum indixit 
et ipsi populo Romano, ut [D.49.15.24]. sed exbanniti non sunt huiusmo
di ... Respondeo, 
hostes inter quos habent locum iura captivitatis et postliminii 
sunt hi de quibus loquitur 
[D.49.15.24]. Sed alio modo loquendo, quaelibet civitas potest habere su
os hostes, ut si 
ipsum helium indixit...' 
127 Ibid.: 'nam sicut illi etiam subditi, quos senatus hostes iudicat, ho
stes sunt populi Romani , 
ut [D.4.5 .5, l ] ... ita illi subditi, quos commune iudicat hostes suos, sunt hos
tes civitatis. sed 
exbanniendo eos, iudicat suos hostes ... ' 
210 
The institutional and legal arrangements of Rome provide an auth
oritative blueprint 
that can be implemented on the level of an Italian city; at least in
 this instance, the 
latter was, so to speak, a Rome in miniature. · Perhaps this is an in
direct reflection of 
Bartolus' dictum of the civitas sibi princeps. Certainly no jurist before him was so 
bold in establishing such a direct link between what Rome can do a
nd what an Italian 
city can do. And if Rome and its law are the model, their authorit
y means that little 
further justification is required. I suspect this is why one cannot find in Bartol us' 
work the existing quaestio tradition on the licitness of statute gran
ting the killing of 
banniti; at best, there are a few scattered statements basing the
 rationale of this 
statutory permission on public utility, that is, the increase in public
 safety caused by 
'purging' the city's territory of 'evil men' .
128 Baldus, too, does not devote much 
space to this topic. The argument that is repeated most often
 is Bartol us': the 
bannitus can be killed because he is a hostis of his city.
129 
The civilians not only discussed the lawfulness of contemporary s
tatutes permitting 
the killing of banniti on an abstract, general level, but they also co
nsidered a number 
of legal issues from the perspective of the killer and the banis
hed victim. Most 
importantly, the jurists agreed that the killer of a bannitus was safe from prosecution 
because he had committed an act authorised by the law; 
130 not only that, he was the 
minister or executor of that justice which the public authorities could not administer 
themselves upon the absentee.
131 For these reasons, as was already argued by 
Jacobus de Arena in a quaestio much valued by posterity, the kil
ler could raise the 
'defence' or 'objection' of banishment (exceptio banni) - i.e. the claim that the 
victim was a bannitus - in order to nullify any proceedings fo
r murder initiated 
against him. 132 And even if the exceptio banni was not raised 
by the defendant 
128 The fullest expression of this idea is in ibid., p. 202b: 'Se
d istud statutum, quod 
exbannitus possit impune occidi, respicit utilitatem publicam, 
ut provincia purgata malis 
hominibus maneat in quiete .. .' Cf. Bartol us ' cons. I.I 07 and 1.188
. 
129 E.g. Baldus to _C.6.24.1, fol. 70vb: 'Sed si statutum diceret expres
se quod possent impune 
offendi, tune tales banniti censentur transfugae et hostes civitati
s unde sunt banniti, quia de 
numero eorum sunt quibus indictum est helium, arg. [D.4.5.5, l], secundum B
ar., qui ita notat 
in disputatione, ubi caute <licit, hostes suae civitatis, et non <licit
, hostes principis vel populi 
Romani .. .' Cf. Baldus to D.4.5.5 and C.1.1(4).1. 
130 Johannes Andreae to Sext.5.12.24 and additio to Speculum iudicia
le, de constitutionibus, 
v. quid si extraneus; Baldus to D.2.7.1, D.12.6.19, and C.7.48
.l; Jacobus Butrigarius to 
D.1.6.1,1. 
131 Id Ba us to D.1.1.3, D.4.3.1,3, D.4.5.5 , and C.7.48.1. 
132 (In) statuto vel consuetudine alicuius civitatis, reported by: Albertus Gandinus, T
ractatus 
de maleficiis, rub. De bannitis, qq. 17-18; Johannes Andreae, additio to Speculum i
udiciale, 
§ 6 sequitur, v. quid si bannitus; Albericus de Rosate to C.1.12.
4 and Quaestiones 
statutorum IV, qq. 36-8; Baldus to D.4.5.5. Bellomo, Quaestiones, p. 460 (no. 1
58), is only 
211 
himself, as the same Jacobus argued in another quaestio, any sentence in favour of a 
b · · · II d "d 133 anmtus was lpso lure nu an v01 . 
4. Aspects of the Casuistic Tradition 
The jurists produced a vast amount of casuistic quaestiones on the ban. The present 
chapter is not supposed to provide a comprehensive synopsis of all areas of juristic 
discussion. Some topics have been reduced to the briefest of summaries, others even 
cut out completely, since the issues raised - although undoubtedly important for 
those whose lives were actually affected by the criminal ban - are relatively 
peripheral to the fundamental legal ideas at play or because they belong better to the 
realm of legal history than in a study of political thought. Furthermore, in order to 
avoid repetition, I have included casuistic quaestiones (if they have not already been 
covered in this chapter thus far) in my exegesis of the systematic part of the 
discourse whenever these intersect. This is why the following section will only cover 
the central topics of procedure, judicial defence, property, and contract. 134 
i. Procedure and Judicial Defence 
When the jurists discussed the lawfulness of statutory procedure and outlawry, we 
saw that they had in mind the criminal ban as a coercive procedural mechanism - the 
aware of Johannes Andreae's report of this quaestio. Cf. Albertus Gandinus, Quaestiones 
statutorum, q. 96. 
133 Si lata sententia pro bannito, reported by: Johannes Andreae, additio to Speculum 
iudiciale, de sententia et de his quae ipsam sequuntur, § 8 iuxta, v. item est nu/la; Albericus 
de Rosate, Quaestiones statutorum IV, q. 60. It is not recorded by Bellomo, Quaestiones. 
Baldus, too, follows Jacobus' opinion at C.6.24.1. This issue had already been briefly, but 
inconclusively, discussed by Albertus Galeotti, Margarita, c. 24 De sententiis, v. sed pone 
quod sententia lata est pro bannito. 
134 I have excluded inter alia certain quaestiones surrounding the licit killing of banniti and 
the cancellation of the ban, as well as the following topics: bannitus agat contra bannitum; 
the banishment of clerics; office-holding; and marriage. Once again, I refer the interested 
reader to Ghisalberti, ' La condanna', and especially Cavalca, II bando, the most 
comprehensive study to date. The question of marriage and dowry has been covered by 
Kirshner, ' Wives ' Claims ', p. 275. 
212 
bannum pro contumacia, whose core aims were to punish the bannitus for his 
disobedience and his alleged crime, to induce h1m to submit to the authorities, and to 
ensure that he stayed away if he did not choose to do so. 135 This view of the ban is 
also evident in their discussion of its procedural aspects in other contexts. Both 
Jacobus de Arena, in his Tractatus de bannitis, 136 and Guilelmus Duranti, in the 
Speculum iudiciale, exclusively presented the ban as bannum pro contumacia. The 
latter provides a standard formula of the criminal ban for use by the legal 
practitioner. The formula represents a final warning: having emphasised that the 
defendant has been repeatedly and lawfully summoned for a given crime, it states 
that he will fall under the ban if he does not appear within a certain period of time. 
Once the time is up, the bannitus is pro confesso and becomes an outlaw, who can 
only rid himself of this status if he returns to obedience by reconciling with the 
offended party and paying a monetary fine to the court. 137 Strikingly, for such a 
standard-setting, practice-orientated treatise, the Speculum iudiciale does not 
mention the ban as a permanent punitive measure - the bannum pro delicto, which 
normally constituted the ordinary punishment for the alleged crime (such as treason) 
and could only be lifted by a judicial pardon. 138 It is true, as we shall see below, that 
this version of the ban was discussed in the context of procedure by Nicolaus de 
Matarellis, notably in his Lectura on the Code, and by some of the commentators in 
135 h Cf. c . 1, pp. 20-1. 
136 Jacobus de Arena, Tractatus de bannitis, fol. 356ra: ' .. . bannum est contumatia partis .. . ' 
Jacobus also brings out the coercive purpose of the ban, fol. 356va: ' ... nam bannum est 
quaedam mulcta, ut [D.25.4. 1,3 : 'Quid ergo, si non responderit aut non veniat ad praetorem? 
[ ... ] cogenda igitur erit remediis praetoris et in ius venire et, si venit, respondere ... ]. .. ' Cf. 
Pazzaglini, The Criminal Ban, p. 19. 
137 Guilelmus Duranti, Speculum iudiciale, de accusatione, § 6 sequitur, v. in locis quoque, 
p. 15a: 'In locis quoque et casibus in quibus accusatus absens in bannum ponitur, sic banni 
forma scribetur. FORMA. P. accusatus a tali crimine commisisse inferendo totam libelli 
formam, quo posito, dicas, quia legitime citatus fuit tali die per talem cursorem, ut ipsa die 
super ipsa accusatione coram tali iudice compareret. Item, et alia vice citatus die tali per 
talem cursorem, ut super eodem negocio coram eodem iudice eadem die compareret. 
Insuper, altera die tali fuit de mandato eiusdem iudicis per talem cursorem incontrata 
eiusdem P., scilicet in tali parochia, vel castro, vel villa, praesentibus tali et tali testibus ad 
hoe adhibitis et rogatis, voce praeconia publice proclamatum quad, si quis dictum P. vellet 
defendere super accusatione praefata, usque ad diem tertiam coram ipso iudice compareret. 
Ipse autem nee venit, nee alius se eius defensioni obtulit pro eodem. Idcirco; nisi dictus P. 
aut alter pro eo hinc ad quinque dies venerit, stare et obedire super his et ipsius iudicis 
mandatis legitime, ex tune sit in banno talis curie pro maleficio supra dicta, a quo liberari 
non possit nisi secundum formam iuris, et nisi solverit eidem curiae centum solidos et 
concordaverit cum dicta T. ' Cf. Cavalca, 11 bando, p. 172. 
138 f C . eh. 1, pp. 20-3. 
213 
their examination of ipso iure sentencing. 139 Although the civilians were therefore 
clearly aware of the bannum pro delicto and must have considered it an important 
part of the ban as a legal institution, it does seem that they attached greater weight to 
its procedural function. This interpretation is further corroborated if we consider that 
treason and rebellion, whose punishment was a central purpose of the bannum pro 
delicto, barely feature amongst those juristic sources in which the cause of 
banishment is revealed. We have already seen that there was very little discussion of 
the ban as a result of party politics; references to political crimes beyond this 
immediate context remain equally exceptional. 140 The broader implication is that the 
ban was, above all, the means by which the civic courts proceeded against 
contumacious defendants in ordinary criminal rather than extraordinary political 
matters. Although the ban was of course used for eminently political purposes 
( especially at the height of the factional expulsions in the later thirteenth and early 
fourteenth centuries), it was not at heart a political institution, much unlike, it should 
be stressed, the senatorial hostis declaration of ancient Rome. 
The emphasis in the Speculum iudiciale on correct summons - and thus the right of 
the defendant to present a defence - is reflective of an indisputable rule that had been 
established at least by the second half of the thirteenth century. 141 In the words of 
Jacobus de Arena: 'so that someone may be placed under the ban, he ought to be 
lawfully summoned, otherwise he cannot be lawfully banished' . 142 In fact, the jurists 
139 For Nicolaus de Matarellis, seep. 238 ff. below. He is the only jurist who clearly explains 
the difference in the cancellation of the two types of ban - as recorded by Baldus at C.6.24.1, 
fol. 70rb: 'et intelligo per contumaciam bannitum ilium qui ideo bannitus est quod 
inobediens fuit iudici intendenti ad diffinitivam negocii procedere ... Unde habet naturam 
cuiusdam interlocutoriae, sicut... remedia praetoris ad cogendum inobedientes... et ideo 
revocari potest per iudicem quando videbitur iudici et mutari ... et conditio similiter apponi, 
dicendo ... liberari non possunt nisi solvant centum, vel nisi hoe et istud fecerint, et similia. 
[ ... ] Sed si bannitus velit obedire iudici et de banno exire, debebit conditionem appositam 
adimplere ... Alioquin ilia remedia, quae inducta sunt ad revocandum et evitandum suam 
diffinitivam et eius poneam, eadem locum habent ad revocandum bannum datum pro delicto, 
et non alias, cum istud bannum fuerit tanquam sententia diffinitiva .. .' 
140 Aside from the instances in the discussion of ipso iure banishment, pp. 215-18 below, I 
can only find two relevant consilia by Baldus: III.96 and III.263-4. 
141 Cf. Ghisalberti, 'La condanna', pp. 60-1; Cavalca, I! bando, pp. 170-3; Pazzaglini, The 
Criminal Ban, pp. 22-3. 
142 Jacobus de Arena, Tractatus de bannitis, fol. 355vb: ' ... ut quis ponatur in banno, debet 
legitime citari, alias non posset legitime banniri.' 
214 
of this period described in considerable detail what counted as a lawful summons. 143 
Guilelmus Duranti explained that one was required even in manifest and notorious 
crimes, which normally warranted fewer procedural formalities. 144 Both Jacobus and 
Guilelmus based their arguments on Roman criminal procedure, which, as we know, 
considered summons to be a necessity; 145 Guilelmus also cites canonistic precedents 
and authorities (including Innocent IV and Hostiensis, the most famous canonists of 
the thirteenth century) to justify his insistence that 'whatever else may be the case, it 
is certain that according to the law no one ought to be denied a legitimate 
defence'. 146 If a summons was omitted, or there were any other procedural 
deficiencies, the jurists concluded that there were grounds for appeal and that the ban 
was null and void from the outset. 147 
We saw in chapter 3 that in the case of imperial banishment the jurists fully 
embraced these procedural safeguards in respect of judicial sentences, but went quite 
a long way in undermining defendants' rights with the notion of ipso iure 
banishment. We will recall Bartolus' argument that an imperial hostis became such, 
and could be treated as such, right from the moment of the commission of the 
rebellious act without the need for a prior judicial sentence or declaration, - a 
143 Ibid., fol. 355va-b; Albertus Gandinus, Quaestiones statutorum, q. 100. Much of the same 
material can be found in Albericus de Rosate's commentaries to D.46.2.19 and C.6.24.1, and 
in his Quaestiones statutorum IV, qq. 49 and 51. 
144 Guilelmus Duranti, Speculum iudiciale, de notoriis criminibus, § 1 scias, v. dixit quoque, 
p. 42b: 'Dixit quoque idem Innocentius quod ... in omni casu tamen requiritur quod citetur 
reus, et ei veuienti exponatur factum propter quod citatur, et ius defensio audiatur ... alioquin 
non tenet sententia super notoriis lata ... ' On Innocent ( d. 1254 ), see A. Melloni, 'Sinibaldo 
Fieschi', DBGI, vol. 2, pp. 1872-4. Note that I have deliberately excluded the repetitio on 
notorious crimes commonly printed in Bartol us' commentaries at D.11. 7.4. While Lange and 
Kriechbaum, Die Kommentatoren, p. 713, attribute this work to Bartolus (though without 
providing any evidence), it does not read like Bartolus to me. An early-modem additio in the 
1574 Turin edition agrees (fol. 202rb ): 'haec n. repetitio non habetur in lecturis antiquis, et 
ideo puto quod non sit Bart., quod etiam ex stylo, et tot capitulis suadetur, quot Bar. non 
so let allegare.' 
145 Cf. eh. 2, p. 53. 
146 Guilelmus Duranti, Speculum iudiciale, de notoriis criminibus, § 1 scias, v. hodie autem, 
p. 43a: 'Sed certe, quicquid fiat, nulli est de iure legitima defensio deneganda ... ' Johannes 
Andreae's additio informs us that these are the '[v]erba Hostiensis'. On Hostiensis (d. 1271), 
see K. Pennington, 'Enrico da Susa, CardinaleOstiense', DBGI, vol. 1, pp. 795-8 . 
147 Jacobus de Arena, Tractatus de bannitis, fol. 355vb; Albertus Gandinus, Tractatus de 
maleficiis, rub. De bannitis, qq. 4-6; Rainerius de Forlivio to D.48 .17.4,2 and D.49.7.1; 
Albericus de Rosate, Quaestiones statutorum IV, qq. 23-4; Bartolus to D.42.1.3; Baldus to 
C.1.4.2, C.7.66.2, C.9.2.3, C.9.40.1 , and C.9.47.2. On the issue of appeal, see Ghisalberti, 
'La condanna', p. 73, and Cavalca, II bando, pp. 175,209, 224-5. 
215 
II II 
particularly severe instance of the bannum pro delicto. 148 At the I. Amissione 
(D.4.5 .5), the fourteenth-century commentators acknowledged that the procedural 
logic regarding the hostis could theoretically also be applied to the communal 
bannitus. Bartolus and Baldus, building on Accursius' exposition of this classic 
passage, 149 briefly mention the possibility of ipso iure banishment, as well as the idea 
that an individual could be banished through a unique legislative enactment, but do 
not develop these thoughts or go into procedural details there. 
150 According to 
Albericus de Rosate, the I. Amissione and its gloss do indeed support Bartolus' 
interpretation of ipso iure sentencing. On the other hand, Albericus points out, one 
could argue for the requirement of a prior judicial sentence with reference to an 
important decretal (Cum secundum leges) promulgated by Boniface VIII (d. 1303). 
This asserts that in the case of ipso iure property confiscation for heresy a sentence 
must precede the act of confiscation by the fisc; the ipso iure provision only means 
that a sentence would be declaratory rather than condemnatory, its effects dated back 
to the date of the crime. 151 As Jacobus de Belviso explains, this has the result that 
any legal acts by the defendant in respect of his estate, which were undertaken 
between the commission of the crime and the sentence (such as alienation and 
testament), are void, since the fisc is taken to have been the real owner during that 
period. 152 Baldus too appears to support this interpretation: in passing, he alludes to a 
148 Cf. eh. 3, pp. 128-64. 
149 Cf. ibid. , pp. 99- 100. 
150 Bartolus to D.4.5.5,1, fol. 14lva: 'Et idem dico per omnia in his, qui sunt effecti hostes ex 
vigore alicuius reformationis vel ipsa lege lata, ut hac lege dicitur. ' Baldus to ibid., fo l. 
249rb: 'Nota quod rebellando quis efficitur hostis suae civitatis, et statim quod est 
pronunciatus rebellis per iudicem vel per reformationem ex tune desinit esse civis .. . Secundo 
nota quod aequiparantur sententia iudicis vel legis reformationi vel statuti, ut hie, secundum 
unam lecturam. [ ... ] In glossa "antiqua littera", ibi "alicuius", nota quomodo per legem 
specialem persona potest iudicari hostis et rebellis, et banniri, et privari civilitate ob 
del ictum.' The same to LF.2.28 c. Domino guerram, fo l. 53vb: ' Sed nunquid !ex possit ipso 
facto aliquem forbannire vel ad mortem damnare. Respondeo, sic. glossa est Accur. 
[D.4.5.5 ,1 ] .. .' 
151 Albericus de Rosate to D.4.5 .5, 1, fo l. 263vb: 'Ultimo nota quod <licit glossa, quod !ex et 
statutum potest introducere quod quis sit bannitus ipso facto vel incidat in talem poenam. 
nam hoe ipso quod facit contra legem vel statutum incurrit ipsam poenam sine aliqua 
sententia vel declaratione iudicis . [ ... ] Sed quod exigatur nihilominus declaratio iudicis, facit 
[Sext.5.2.19]. . .' The same to C.3.26.1, fol. 159ra: ' Ex hac littera fuit decretalis [Sext.5.2.19], 
ubi bona haereticorum non apprehenduntur nisi lata sententia, licet confiscata sint ex die 
commissi criminis.' 
152 Jacobus de Belviso to Coll.2 .6 (de incestis et nefariis nuptis) § sancimus (=Nov.12.1), 
which concerns the case of ipso iure confiscation of property for illicit marriage, fol. 2 lrb-
va: 'Queritur secundo nunquid ipso iure vel per sententiam bona publicentur. Et glossa primo 
arguit quod statim et ipso iure. [ ... ] Sed quidam dicunt sententiam necessariam fore ... Prima 
216 
case in which statute condemns a traitor ipso iure, and indicates that a trial, at which 
the accused can present his defence, is necessary before it is lawful to implement the 
statutory penalty through a declaratory sentence. 153 While, in theory, this may have 
been the jurists ' favoured stance (at least in a communal context), Albericus reports 
that at Bergamo he once successfully defended an accused murderer, who had killed 
someone immediately after the latter had committed a crime whose penalty was the 
ipso iure application of the criminal ban (and thus the liability to be killed with 
impunity), before any judicial sentence of banishment. Albericus knows, however, 
that he is on shaky ground, noting that this position clashes with Boniface's decretal. 
This confirms that if the requirement for a declaratory sentence was accepted, the 
liability to be killed with impunity, unlike property confiscation, was not dated back; 
otherwise there would have been no difference between the two positions. 
154 
In the communal context, the civilians did not create a proper theory of ipso iure 
banishment, nor did they achieve a consensus on crucial procedural details. It is 
interesting that the jurists did not discuss this mechanism more than they did, given 
its presence in statutory law as a way of punishing certain major crimes that were 
usually of a political nature. This suggests that ipso iure banishment was less 
significant and commonly applied than references to it in the statute books imply. 
This also suggests that ipso iure sentencing was less legally controversial than it 
appears at face value, which would explain the notable lack of references to 
Pastora/is cura in this context. If one accepts the notion that declaratory trials are 
necessary, then the requirements of Pastora/is cura are met, since the defendant 
would have to be summoned and granted a hearing before he could be treated as a 
opinio placet et hodie approbatur in decretali Bonifacii [Sext.5.2.1 9] ... Queritur tertio quid 
prodest scire cum bona non auferantur nisi lata sententia? Respondet glossa quo ad fructus 
medii temporis .. . Item, propter alienationem rei et propter supremam voluntatem ... ' 
153 Baldus to C.9.2.3, fol. 204va: 'Pone quod statutum civitatis dicit quod proditor sit ipso 
iure condemnatus. modo dicitur quod quidam est proditor et ipso iure condemnatus. petitur 
executio. an isto iudicio executionis poterit venire ad defensionem procurator? Breviter dico 
quod non, quia disputatur de crimine et super crimine oportet declarari . unde, dato quod non 
feratur alia sententia, scilicet, condemnatoria, tamen fert aequipollens, scilicet, 
declaratoria .. .' The implication is that the defendant himself has to appear according to the 
rule that procurators are not admitted in capital trials - see pp. 220-1 below. 
154 Albericus de Rosate, Quaestiones statutorum IV, q. 94, fol. 85ra: 'Duo sunt statuta 
Pergami . Uno quorum cavetur quod faciens feritam sanguinolentiam cum armis vetitis ipso 
facto sit in banno de maleficio. alio cavetur quod offensa illata bannito de maleficio sit 
impunita. Quidam fecit feritam praedictam et antequam banniretur per sententiam fuit 
occisus. quaeritur an impune occidi potuerit. Obtinui quod sic ... Sed quod expectari de beat 
sententia seu declaratio, bene facit... [Sext.5.2.19]. ' Albericus relates the same case at 
D.40.5.30,6 and C.3 .26.1 . 
217 
bannitus (granted that property confiscation is dated back to the date of the crime). If 
we think back to Albericus de Rosate's considerations on political banishment, it was 
assumed there that the podesta of the faction in power would summon members of 
the opposing faction before banishing them. This supports what we saw in chapter 1, 
namely, that in legal practice the courts were normally careful to summon 
defendants. 155 It appears likely, therefore, that the cities did not typically rely on ipso 
iure provisions to circumvent procedural safeguards in political (or any other) cases, 
although Albericus' account indicates that Bartolus' interpretation of ipso iure 
sentencing must have been followed in practice at least some of the time. 
Most of the jurists' discussion of procedural questions related to the defendant who 
had already been banished. Minor exceptions aside, the civilians fully embraced the 
contemporary practice of not permitting the bannitus to appear as a plaintiff (actor) 
in court, either in person or through a legal representative (procurator); most felt little 
need to justify this. 156 Interestingly, as it foreshadows Bartolus' interpretation of the 
ban, Cinus de Pistoia, in a consilium, rationalised the banished person's judicial 
incapacity through an idea that had been used by Jacobus de Arena in a different 
context some years before: the bannitus is a rebel of the respublica, since, through 
his contumacy, he did not obey the judge in whom the respublica is represented; 
because of his disobedience he is unable to sue and disavowed (diffidatus) as a 
hostis. 157 The comparison to the hostis, however, seems to be more decorative than 
anything else, as Cinus does not explain what it contributes to the main argument 
that disobedience to the courts precludes judicial recourse. 
155 Ch. 1, pp. 22-3 . 
156 Jacobus de Arena, Tractatus de bannitis, fol. 356ra-b, at 356ra: 'Dico ergo quod bannito 
agenti et litem moventi omnis audientia denegatur.' Albertus Gandinus, Quaestiones 
statutorum, q. 96; Johannes Andreae, additiones to Speculum iudiciale, de cessione actionis, 
§ 1 sciendum, v. illud etiam quaeritur I de accusatore, super rubrica; Albericus de Rosate to 
C.6.24.1 and D.46 .2.19, and his Quaestiones statutorum IV, qq. 50 and 63. Guilelmus 
Duranti, Speculum iudiciale, de procuratore, § 2 ratione, in princ. § - v. item quod, made a 
misguided comparison to the deportatus, seeing that the latter, as we know, was not actually 
barred from the courts. Cf. Cavalca, II bando, pp. 219-22; Vallerani, 'Diritti di cittadinanza', 
p. 3 f. 
157 Cinus' consilium is edited by Monti, Le "quaestiones " e i "cons ilia ", pp. 111-12. The 
relevant passage appears at p. 112: ' .. . quia bannitus reputatur rebellis reipublice eo quod non 
paruit iusdicenti in quo respublica representatur et sic factus inhabilis vitio suo et diffidatus 
ut hostis ex quo prohibetur esse in iudicio .. .' Cf. Cavalca, II bando, p. 93 . 
218 
While the jurists fully accepted the banished person's incapacity as plaintiff, they all 
rejected the notion that he, sued or accused again in his former city, 158 could not 
defend himself, notwithstanding statutes that broadly prohibited the bannitus from 
being granted justice (non reddatur ius) or explicitly forbade him a hearing as 
defendant (non audiatur in defendendo). 159 Already in the thirteenth century, well 
before Pastora/is cura, there was a general conviction that even the Devil deserves a 
defence. 16° Communis opinio was established by Jacobus de Arena, who employed a 
number of Roman law analogies ( e.g. from the law of slaves, freedmen, and fraud), 
but who relied especially on the comparison between the bannitus and the 
excommunicate: in canon as well as civil law (after Frederick II's authentica 
Credentes) the latter was granted the ability to defend himself. 161 The jurists writing 
in the first decades of the fourteenth century accepted Jacobus' position and found 
plenty of evidence for it in the ius commune, though (oddly) they did not link the 
issue to Pastora/is cura (promulgated in 1314). 162 It was only with Bartolus and 
Baldus that the banished person's right to a summons and defence was presented 
with reference to the papal bull as a completely inviolable rule of natural law. They 
never explained why Pastora/is cura was entirely safe from statutory derogation; 
158 For the question of forum, see pp. 246-7 below. 
159 Woolf, Barto/us of Sassoferrato, p. 202, mentions Bartolus ' view that judicial defence 
was derived from natural law. Ghisalberti, 'La condanna', pp. 40 and 62, and Milani, 
'Giuristi, giudici e fuoriusciti ', pp. 620-1, cover this topic with reference to Jacobus de 
Arena. Cavalca's somewhat broader, if still very limited review of the source material (// 
bando, pp. 223-4), leads him to the erroneous conclusion that 'la dottrina non era unanime 
nel concedere al bandito ii diritto alla difesa'; the opposite was in fact the case. 
160 The saying is recorded by Albericus de Rosate at C.1.3(6).32, fol. 30rb: ' ... et hoe est quod 
dici consuevit quod etiam Diabolo non est deneganda defensio.' Pennington, The Prince and 
the Law, p. 144, therefore exaggerates when he writes (my italics): 'Only after the jurists 
concluded that parts of the judicial process were protected by natural law did they clearly 
articulate a subjective, almost inalienable, right of defendants to have their day in court.' 
161 Jacobus de Arena, Tractatus de bannitis, fol. 356ra: 'sed si ipse ab alio conveniatur... tune 
in omnibus suis defensionibus audiatur. sic, et alias prohibitus agere, tamen conventus potest 
se defendere, [D.2.4.14], [D.49.1.15], [D.48.19.19]. Sic etiam dicitur de haeretico, quod ipse 
non agit, tamen conventus potest defendere se, [auth. credentes, post C.1.5(8).4], et in 
excommunicato, qui, licet non agat, tamen potest se defendere per alium, ne quis in 
periculum animae suae communicare ei cogatur, ut [X.2.1.7] et [X.2.25.5].' The same 
position can be found in his commentary to D.24.3.47 . For Credentes, see eh. 3, p. 109. The 
two decretals (X.2.1.7 - Intelleximus) and (X.2.25.5 - Cum inter [dilectos filios] priorem) 
were issued by Alexander III ( d. 1181) and Innocent III ( d. 1216) respectively. Cf. Albertus 
Gandinus, Quaestiones statutorum, q. 97. · 
162 Johannes Andreae, additio to Speculum iudiciale, de accusatione, § 6 sequitur, v. quid si 
bannitus I de cessione actionis, § 1 sciendum, v. illud etiam quaeritur; Rainerius de Forlivio 
to D.48.1.5; Albericus de Rosate extensively reports the discourse as it had developed up to 
that point in his Quaestiones statutorum IV, qq. 5, 32, 39-41 , 52-4, and 89. 
219 
bannitus (granted that property confiscation is dated back to the date of the crime). If 
we think back to Albericus de Rosate's considerations on political banishment, it was 
assumed there that the podesta of the faction in power would summon members of 
the opposing faction before banishing them. This supports what we saw in chapter 1, 
namely, that in legal practice the courts were normally careful to summon 
defendants.155 It appears likely, therefore, that the cities did not typically rely on ipso 
iure provisions to circumvent procedural safeguards in political ( or any other) cases, 
although Albericus' account indicates that Bartol us' interpretation of ipso iure 
sentencing must have been followed in practice at least some of the time. 
Most of the jurists' discussion of procedural questions related to the defendant who 
had already been banished. Minor exceptions aside, the civilians fully embraced the 
contemporary practice of not permitting the bannitus to appear as a plaintiff (actor) 
in court, either in person or through a legal representative (procurator); most felt little 
need to justify this. 156 Interestingly, as it foreshadows Bartolus' interpretation of the 
ban, Cinus de Pistoia, in a consilium, rationalised the banished person's judicial 
incapacity through an idea that had been used by Jacobus de Arena in a different 
context some years before: the bannitus is a rebel of the respublica, since, through 
his contumacy, he did not obey the judge in whom the respublica is represented; 
because of his disobedience he is unable to sue and disavowed (difjidatus) as a 
hostis. 157 The comparison to the hostis, however, seems to be more decorative than 
anything else, as Cinus does not explain what it contributes to the main argument 
that disobedience to the courts precludes judicial recourse. 
155 Ch. I, pp. 22-3. 
156 Jacobus de Arena, Tractatus de bannitis, fol. 356ra-b, at 356ra: 'Dico ergo quod bannito 
agenti et !item moventi omnis audientia denegatur.' Albertus Gandinus, Quaestiones 
statutorum, q. 96; Johannes Andreae, additiones to Speculum iudiciale, de cessione actionis, 
§ I sciendum, v. illud etiam quaeritur I de accusatore, super rubrica; Albericus de Rosate to 
C.6.24.1 and D.46.2.19, and his Quaestiones statutorum IV, qq. 50 and 63. Guilelmus 
Duranti, Speculum iudiciale, de procuratore, § 2 ratione, in princ. § - v. item quad, made a 
misguided comparison to the deportatus, seeing that the latter, as we know, was not actually 
barred from the courts. Cf. Cavalca, II bando, pp. 219-22; Vallerani, 'Diritti di cittadinanza', 
p. 3 f. 
157 Cinus' consilium is edited by Monti, Le "quaestiones" e i "consilia", pp. 111-12. The 
relevant passage appears at p. 112: ' ... quia bannitus reputatur rebellis reipublice eo quod non 
paruit iusdicenti in quo respublica representatur et sic factus inhabilis vitio suo et diffidatus 
ut hostis ex quo prohibetur esse in iudicio ... ' Cf. Cavalca, II bando, p. 93 . 
218 
While the jurists fully accepted the banished person's incapacity as plaintiff, they all 
rejected the notion that he, sued or accused again in his former city, 158 could not 
defend himself, notwithstanding statutes that broadly prohibited the bannitus from 
being granted justice (non reddatur ius) or explicitly forbade him a hearing as 
defendant (non audiatur in defendendo ). 159 Already in the thirteenth century, well 
before Pastora/is cura, there was a general conviction that even the Devil deserves a 
defence. 16° Communis opinio was established by Jacobus de Arena, who employed a 
number of Roman law analogies ( e.g. from the law of slaves, freedmen, and fraud), 
but who relied especially on the comparison between the bannitus and the 
excommunicate: in canon as well as civil law (after Frederick II's authentica 
Credentes) the latter was granted the ability to defend himself. 161 The jurists writing 
in the first decades of the fourteenth century accepted Jacobus' position and found 
plenty of evidence for it in the ius commune, though (oddly) they did not link the 
issue to Pastora/is cura (promulgated in 1314).162 It was only with Bartolus and 
Baldus that the banished person's right to a summons and defence was presented 
with reference to the papal bull as a completely inviolable rule of natural law. They 
never explained why Pastora/is cura was entirely safe from statutory derogation; 
158 For the question of forum, see pp. 246-7 below. 
159 Woolf, Barto/us of Sasso/errata, p. 202, mentions Bartolus' view that judicial defence 
was derived from natural law. Ghisalberti , 'La condanna', pp. 40 and 62, and Milani, 
'Giuristi, giudici e fuoriusciti', pp. 620-1, cover this topic with reference to Jacobus de 
Arena. Cavalca's somewhat broader, if still very limited review of the source material (II 
bando, pp. 223-4), leads him to the erroneous conclusion that 'la dottrina non era unanime 
nel concedere al bandito ii diritto alla difesa'; the opposite was in fact the case. 
160 The saying is recorded by Albericus de Rosate at C.1.3(6).32, fol. 30rb: ' ... et hoe est quod 
dici consuevit quod etiam Diabolo non est deneganda defensio.' Pennington, The Prince and 
the Law, p. 144, therefore exaggerates when he writes (my italics): 'Only after the jurists 
concluded that parts of the judicial process were protected by natural law did they clearly 
articulate a subjective, almost inalienable, right of defendants to have their day in court.' 
161 Jacobus de Arena, Tractatus de bannitis, fol. 356ra: 'sed si ipse ab alio conveniatur... tune 
in omnibus suis defensionibus audiatur. sic, et alias prohibitus agere, tamen conventus potest 
se defendere, [D.2.4.14], [D.49.1.15], [D.48.19.19]. Sic etiam dicitur de haeretico, quod ipse 
non agit, tamen conventus potest defendere se, [auth. credentes, post C.1.5(8).4), et in 
excommunicato, qui, licet non agat, tamen potest se defendere per alium, ne quis in 
periculum animae suae communicare ei cogatur, ut [X.2.1.7] et [X.2.25.5].' The same 
position can be found in his commentary to D.24.3.47. For Credentes, see eh. 3, p. 109. The 
two decretals (X.2.1. 7 - Intelleximus) and (X.2.25 .5 - Cum inter [ dilectos .filios J priorem) 
were issued by Alexander III ( d. 1181) and Innocent III ( d. 1216) respectively. Cf. Albertus 
Gandinus, Quaestiones statutorum, q. 97. 
162 Johannes Andreae, additio to Speculum iudiciale, de accusatione, § 6 seq~itur, v. quid si 
bannitus I de cessione actionis, § 1 sciendum, v. illud etiam quaeritur; Rainerius de Forlivio 
to D.48.1.5; Albericus de Rosate extensively reports the discourse as it had developed up to 
that point in his Quaestiones statutorum IV, qq. 5, 32, 39-41, 52-4, and 89. 
219 
evidently, for whatever reason, they did not consider statute to possess sufficient 
causa. 163 Albericus de Rosate was alone in questioning this consensus. He boldly 
rejected the pope's assertion in Pastora/is cura by rejecting the logic that 
underpinned it: one cannot extrapolate a natural right to judicial defence from the 
natural right to bodily defence at D.1.1.3; the two cases are fundamentally 
different. 164 Therefore, statute is free to deprive the bannitus of judicial defence if it 
contains express non audiatur provisions to that effect.
165 Ever the realist, Albericus 
mentions that his view, while harsh, was followed in practice 'everywhere', an 
indication that the civic courts were probably less obedient to dominant legal opinion 
than the university jurists liked. 166 But not even Albericus thought that this meant 
that the bannitus is not granted any hearing whatsoever. He can always deny the 
facts as presented by the plaintiff and claim that the latter has no action against him. 
According to Albericus, it would only be a defence properly speaking if the plaintiff 
had an action and the defendant raised a formal, extenuating exceptio that would 
negate it ( e.g. the exceptio banni). 167 The jurists, including Albericus, furthermore 
163 Baldus is clearest at D. l .1.3, fol. 9vb: 'Unde Papa nee lmperator possunt derogare 
naturali defensioni, ut [Clem.2.11.2]. Et ideo unicuique est dandus terminus ad suam 
defensionem faciendum ... Ex praedictis apparet quod non valet statutum quod bannitus non 
admittatur ad aliquam defensionem, nisi statutum respiciat causam in qua est contumax.' Cf. 
Bartolus to D.2.4.14; Baldus to D.1.1.3, D.2.7.1, C.6.24.1, C.9.9.25(26); and both to 
D.24.3.47. Note that if statute explicitly forbade the bannitus to be heard in defendendo, they 
varyingly limited defences - in Baldus' case, severely so - to certain essential elements that 
they considered natural rather than civil. Baldus' view is, in effect, therefore not so different 
to Albericus de Rosate's. 
164 Albericus de Rosate to C.1.3(6).32, fol. 30rb-va: 'immo, quidam dicere voluerunt quod 
talis defensio non posset tolli per Papam vel Principem vel legem communem vel 
municipalem, quasi talis defensio a iure naturali procedat. et videtur hoe expressum in 
[Clem.2.11.2]. .. Nee obstat quod defensio est de iure naturali, quia verum est de defensione 
manuali, qui concessum est cuique se defendere, de qua loquitur [D.1.1.3]. .. et de ista non 
loquitur hie. Sed de defensione danda reo ... et ista non est de iure naturali, sed iure gentium 
vel civili, et immo per aliud ius tolli potest. et hanc opinionem veriorem puto.' The same 
position is found in his Quaestiones statutorum IV, q. 53 . The association of the ius gentium 
with the sphere of the civil rather than the natural is as peculiar as it remains unexplained. 
165 The phrase non reddatur ius was too general to have the same effect and only prevented 
the bannitus from appearing as plaintiff - cf. Albericus to C.1.3(6).32 and Quaestiones 
statutorum IV, q. 52. 
166 Albericus de Rosate, Quaestiones statutorum IV, q. 53, fols 74vb-5ra: 'et sic videtur tale 
statutum valere, et sic communiter servatur. [ ... ] et haec opinio, licet dura, ubique servatur.' 
167 Ibid., q. 41 , fol. 72va: 'Item quaero, si statuto caveatur quod bannitus non audiatur etiam 
in defendendo, et aliquis, cui vere contra ipsum nulla actio compete bat, ipsum conveniat... an 
audiatur volens se defendere? Et videtur quod sic, quasi non videatur defendere, cum actori 
nulla actio competebat... quod puto verum ubi nulla ex toto actio actori competeret, quia tune 
etiam nulla facta defensione per eum, actore nihil probante, absolvetur... Sed si aliquando 
220 
agreed that the bannitus could engage a procurator to make his legitimate 
representations. 168 Since Roman law did not permit this in capital cases, they argued 
that the authorities should in this instance grant a guarantee of safe passage to allow 
the bannitus to appear personally. They considered this guarantee as binding upon 
the city and private citizens alike. Ever since an influential quaestio by Oldradus de 
Ponte, one of the arguments used to justify this was the Roman law idea that peace 
agreements between hastes and public officials may not be broken; naturally, after 
Bartolus, this became the dominant argument. 
169 
ii. Property and Contract 
On the whole, the jurists felt no need to discuss or justify the banished person's 
liability to be injured in respect of his property (theft or other damage) that had not 
been confiscated by the fisc. One of Jacobus de Arena's quaestiones (Statutum est in 
civitate quad bannito ratio non fiat), 170 and a very similar quaestio by Nicolaus 
habuit actionem, licet sit exceptione sublata, tune, quia reus habet necesse de exceptione 
pro bare, obstaret sibi statutum, quia tune opus esset quod defensionem faceret...' 
168 Cf. Cavalca, I! bando, pp. 190 and 221, who failed to note the establishment of this 
communis opinio. 
169 Albericus de Rosate, Quaestiones statutorum IV, q. 39, fols 72vb-3rb: 'Quaero etiam 
aliam quaestionem. bannitus ob maleficium accusatus de alio maleficio volens se defendere 
petit fidantiam. quaeritur an sibi dari possit et debeat, et data, serv;ri. [ ... ] Sed latius e~ 
subtilius in alia quaestione disputata per dominus Oldr. in studio Paduano, quae incipit 
"potestas Paduae de crimine capitali" ... Secundo, ex pacti observatione. pacta, nota, belli 
ducis hosti fidantiam praebent... Sed potestas dux civitatis et princeps [!] est... et bannitus 
quodammodo est hostis civitatis, eo quod ab eius obedentia se subtrahit, [D.4.5.5, 1]. [ ... ] 
Concludendo et determinando sic dixi[t]. Aut iste bannitus accusatur de crimine ex quo 
deberet imponi poena relegationis vel maior, et in quo per procuratorem se non posset 
defendere ... ut [D.48.1.13,l], [D.3 .3.33 ,2]. .. et hoe casu absque dubio est fidantia 
concedenda et concessa servanda. [ ... ] Aut iste bannitus accusatur de crimine ex quo debetur 
poena relegationis vel infra, vel civiliter convenitur, et tune licentia de necessitate danda non 
est, quia se defendere potest per procuratorem, [D.4.4.20(21)].' Extant in Cordoba, 
Biblioteca De La Catedral, MS. 108, fols 226va-7vb - Bellomo, Quaestiones, pp. 602-3 (no. 
36). Cf. Bartolus to D.2.14.5, D.4.3.1,3, D.4.4.20(21), D.48.1.5; Baldus to D.1.5.4, 
D.2.4.10,6, D.4.3.1,3, D.4.4.20(21). 
170 Reported by Johannes Andreae, additio to Speculum iudiciale, de accusatione, § 6 
sequitur, v. quid si bannitus. See pp. 234-6 below for further details on this quaestio. 
221 
I II 
I I 
111 
I 
Capocius de Urbe (d. 1368),171 are the only two exceptions. They argue that such 
injury is lawful, even if statue only very generally states bannito non fiat ratio, 
bannito non reddatur ius, or bannitus non audiatur. Therefore, the bannitus does not 
acquire an action for damages that he would be able to bring to court later should his 
ban ever be cancelled. 172 Among a substantial list of explanations, one can find the 
argument that this is the general custom of all Italy, 173 and, as mentioned above, 
Jacobus uses the hostis as an analogy for the bannitus: given that such individuals 
can be injured in person, he writes, it is only logical that they can be injured in 
respect of their property. 174 Jacobus also cites a law on prescription - the acquisition 
of ownership of property belonging to another through possession for a fixed period -
which implies, however imperfectly, that one can legally acquire ownership of the 
banished person's property by taking it away from him. 175 In effect, this is the same 
argument Bartolus later went on to make regarding the property of the imperial 
hostis. 
The jurists were more interested in whether the same rules applied if banniti suffered 
contractual injuries, and thus we have a number of quaestiones on this topic. 176 They 
171 Statuto civitatis Romanae - reproduced by Albericus de Rosate, Quaestiones statutor~m 
IV, q. 89, but not recorded by Bellomo, Quaestiones. Cf. E. Cortese, 'Capocci dei Monti, 
Niccolo, DBGI, vol. 1, p. 427. 
172 Johannes Andreae, additio to Speculum iudiciale, de accusatione, § 6 sequitur, v. quid si 
bannitus, p. 19b: 'constat etiam regulariter prohibita causaliter ex Jege Jicere. Patet in furto et 
damno causaliter permissis ... Ubi autem talia permittuntur, non est locum querelae ... Nee 
obstat exceptio banni, cum a principio actio non nata fuerit... ' 
173 Ibid.: 'Item generalis consuetudo habet bannitos talia pati ... et consuetudo facit licitum 
quod alias non liceret...' 
174 See pp. 208-9 above. 
175 Johannes Andreae, additio to Speculum iudiciale, de accusatione, § 6 sequitur, v. quid si 
bannitus, p. 19b: 'Non obstat praesens accusatio in furto, ex quo saltem videtur tenere ... quia 
ibi illicitum initium, hie licitum, [C. 7.31.1,3: Hoe tantummodo observando, ut in his 
omnibus casibus ab initio bona fide earn capiat, secundum quod exigit longi temporis 
praescriptio ... ] ... ' 
176 Cf. Cavalca, JI bando, pp. 222 and 234; Milani, 'Diritti di cittadinanza', p. 5. My 
summary is based on the following quaestiones: Guilelmus Accursius, Statutum est in 
civitate Bononie, reported by Albertus Gandinus, Quaestiones statutorum, q. 103, Johannes 
Andreae, additio to Speculum iudiciale, de sententia, § 8 iuxta, v. item est nulla, Albericus 
de Rosate, Quaestiones statutorum IV, q. 44 - cf. Martino, "'Quaestiones" civilistiche', pp. 
269-71, and Bellomo, Quaestiones, p. 80 (no. 156), p. 318 (no. 243), pp. 612-3 (no. 6); 
Thomas de Piperata, Continetur in statuto civitatis Bononie, reported by Albericus de Rosate 
to C.4.44.2 and Quaestiones statutorum IV, q. 45 - cf. Bellomo, Quaestiones, pp. 81-2 (no. 
158), pp. 264-5 (no. 119), p. 319 (no. 247), p. 397 (no. 442); Lambertinus Ramponi, 
Continetur in statuto, reported by Johannes Andreae, additio to Speculum iudiciale, de 
rescindendis venditionibus, § nunc dicendum, v. ubi autem, p. 237b; Jacobus de Arena, 
222 
------ -
all agreed, or rather assumed, that banniti possessed the fundamental capacity to 
make binding contracts. The basis of this assumption may have been the idea, 
highlighted by Jacobus de Arena, that contracts are ultimately derived from the ius 
gentium, whose provisions still apply to banniti. 177 The jurists also accepted (without 
debate) that statute could limit this capacity; a common sanction forbade banniti to 
make contracts to their advantage (ad sui utilitatem contrahere). This did not mean 
that banniti were barred from certain categories of contracts, but that the facts of 
each case had to be examined to see if they received any benefit: if they did, the 
contract was void. 178 The jurists' usual example of a valid disadvantageous contract 
is a sale in which the bannitus is deceived into being paid less than half of the just 
· 179 S · pnce. ome contention arose over whether the bannitus, if he suffered contractual 
injury regarding a contract validly made ( e.g. when he was cheated in price), 
acquired an action, that is, whether the contract not only obligated the bannitus but 
also the other party. A minority -Thomas de Piperata and, following him, Albericus 
de Rosate - argued that the statutory liability to be injured in property (in rebus) 
extended to the contractual sphere, and that no action arose as a result. 180 Most 
jurists, including Jacobus de Arena, Albertus Gandinus, Lambertinus Ramponi ( d. 
1304), 181 Johannes Arldreae, and Nicolaus Capocius, rejected this interpretation, 
Statutum est in civitate quad bannito ratio non fiat - cf. p. 221, n. 170 above; Nicolaus 
Capocius de Urbe, Statuto civitatis Romanae - cf. p. 222, n. 171 above; Albertus Gandinus, 
Quaestiones statutorum, q. 113 . 
177 Johannes Andreae, additio to Speculum iudiciale, de accusatione, § 6 sequitur, v. quid si 
bannitus, p. 19b: 'Item, ex vi banni non fuit impedita origo actionis, quae est de iure 
gentium, [D.2.14.7: Juris gentium conventiones quaedam actiones pariunt, quaedam 
exceptiones.].' 
178 Albertus Gandinus, Quaestiones statutorum, q. 103, p. 206: 'Statutum est quod bannitus 
non possit ad sui utilitatem contrahere. Titius locavit domum ad annum cuidam: in processu 
temporis, hie Titius cepit esse in bampno: bane domum vendidit cuidam, eo pacto ut emptor 
permitteret ilium conductorem ibi habitare usque ad finem anni. Queritur an tale pactum 
teneat. [ ... ] pactum est iniquum quia si non posset expelli inquilinus, prodesset bampnito, 
quod statutum non patitur... ' Cavalca, 11 bando, p. 234, wrongly considers this contract to be 
valid. 
179 Albericus de Rosate, Quaestiones statutorum IV, q. 45, fol. 73vb: 'Sed quid si statuto 
caveatur quod bannitis non fiat ratio et eorum querelae et petitiones non audiantur. quidam 
bannitus vendidit praedia sua et deceptus fuit ultra dimidiam iusti pretii. deinde exemptus est 
et cancellatus de banno. nunquid agere potest contra emptorem ... ?' 
180 Ibid., fol. 74ra: 'In contrarium, quia ex forma statuti potuit impune darnnificari, ergo 
fortius in contractu, cum auctoritate legis fiat.. . Item, cum auctoritate statuti licuerit tune 
decipere et eum damnificari, ergo nulla actio nata fuit. Ergo absurdum esset quod per 
exemptionem de banno nasceretur quae tune nata non fuit.. .' Cavalca's small source base (/l 
bando, p. 222) leads him to portray this incorrectly as communis opinio. 
181 Cf. G. Mazzanti, 'Ramponi, Lambertino', DBGI, vol. 2, pp. 1651-2. 
223 
11 
,,II 
maintaining that when statute refers, or is interpreted to refer, to things (res ), it is 
interpreted restrictively to exclude contracts. 
182 Jacobus and Nicolaus further 
emphasised the natural justice in remaining faithful to one's contractual obligations, 
even in respect of public enemies and rebels. 
183 Hence, the contractual injury was 
unlawful, resulting in an action for the bannitus that he could of course only bring to 
court after the cancellation of his ban. 
184 We will see that this basic position was later 
also adopted by Bartolus. Strangely perhaps, even after Bartolus established the 
hostis as a general analogy for the bannitus, the jurists never considered this position 
to clash with the rule of Roman law that it is lawful for private persons to defraud 
hastes. It appears that they only considered this rule to invalidate private peace 
agreements but not any other type of contract. 
As we will see shortly, the casuistic discourse on contract illustrates certain 
principles that were beginning to gain ground in the civilians' systematic 
interpretation of the ban during the later thirteenth century, and which were to be 
established as formal rules in Bartolus' synthesis of the tradition. According to these 
principles, the bannitus retained all legal capacities - by implication also those of the 
ius commune and Roman citizenship - unless statute had expressly taken them away. 
And if the bannitus possessed a right, granted that he could not sue for it while the 
ban lasted, he could do so after its cancellation. Although these principles were 
already established in the quaestiones on contract by the late thirteenth century, they 
did not receive universal application in the discourse on banishment at that time. In 
their way stood, above all, the widely respected opinion of the Ordinary Gloss, to 
which we shall now turn. 
182 Johannes Andreae, additio to Speculum iudiciale, de accusatione, § 6 sequitur, v. quid si 
bannitus, p. 19b: 'De secundo articulo tenet quod, licet ex commissis in bannitum et eius res 
non agat exemptus, ex contractu tamen recte agit, ex quo exemptus ... ' Ibid., de rescindendis 
venditionibus, § nunc dicendum, v. ubi autem, p. 237b: 'Item, non obstat secundum, de 
damnis et iniuriis non emendandis. [ ... ] patet hoe in lege Aquilia, quod non competit in 
damno procedente ex contractu. Damnum enim dicitur in rebus datum ... ' 
183 Ibid., de accusatione, § 6 sequitur, v. quid si bannitus, p. 19b: 'Est igitur fides servanda .. . 
etiam hosti et rebelli ... nam nihil magis fidei humanae congruit quam conventa servare ... ' 
184 Ibid.: 'Quae exceptio, cum competat propter contumaciam non parentis, est temporalis, 
quamdiu durat contumacia. et ideo, ea cessante quia factus est obediens et exemptus est de 
banno et eius poena, agere po test...' 
224 
5. The Systematic Tradition 
i. Accursius 
It was already noted in the last chapter that the jurists generally viewed the ban as 
something alien to Roman law, mainly because it deviated from ius commune 
measures against contumacious criminal defendants. This is why the jurists regularly 
pointed out that the 'banished persons of our time' could not be found in the Corpus 
iuris, save in the 'modem' legislation that was appended to it, namely, Frederick II's 
coronation constitutions and the Libri feudorum. 185 As Bartol us rightly argued in the 
middle of the fourteenth century, the Roman law did, in fact, have a figure quite 
similar to the bannitus: the hostis-transfuga. Nevertheless, before Bartolus, as we 
shall see once again in this section, no jurist used the ancient hostis as a general (that 
is, systematic) analogy for the medieval bannitus. The civilians' view of the ban as a 
relatively recent creation was mirrored in the fact that, for the first 150 years of 
juristic discussion of banishment, they did not compare the bannitus to the figure in 
Roman law that most closely resembled him. It was thus the dominant opinion of the 
thirteenth century that the deportatus should serve as a general analogy. We can 
already witness this in Azo's quaestio on the ban, where there is an assumption that 
civic exile is equivalent to deportation, and one of the main issues was whether 
Italian cities possessed the jurisdiction to deport. 186 
The surviving evidence suggests that the earliest considerations on the fundamental 
nature of the communal bannitus come from the first decades of the thirteenth 
century. Jacobus de Arena reports the opinion of Ubertus de Bobbio (d. ea. 1245),
187 
according to whom banniti should be compared to deported persons if deportation 
was the penalty for their crime in Roman law.
188 Accursius, in a famous gloss to 
1.1.12,2 (§ Relegati), _used a variation of the same argument: 
185 E.g. Albericus de Rosate, Quaestiones statutorum IV, q. 1, quoted in eh. 3, p. 125 (n. 
119). A further example is Jacobus Butrigarius to auth. Item quaecunque,post C.1.3(6).13, 
fol. 20ra: 'Sect quid est bannum? certe per leges veteres nil dicebatur de banno, sect hec !ex 
nova facit mentionem.' See also Baldus' comments (quoting Nicolaus de Matarellis) to 
C.6.24.1, on p. 240, n. 239 below. 
186 Azo (Landsberg), Titius commisit - seep. 195, n. 72 above. 
187 Cf. N. Sarti, 'Uberto da Bobbio', DBGI, vol. 2, pp. 1989-90. 
188 Jacobus de Arena to C.6.24.1, fol. 40ra: 'Dominus Ubertus de Bobio dixit quod si crimen 
est tale de quo pena deportationis sequeretur, tune habeantur pro deportatis ... ' 
225 
But to whom shall we compare the banished persons of our time? I reply, if the crime 
is such that he loses his citizenship [civitate] as well as his goods, either de iure or 
according to the custom of the cities, then I will compare him to the deported person, 
otherwise I will compare him to the relegated person ... 189 
According to Accursius, banniti should be treated as deportati if their crime results 
in the loss of citizenship and the confiscation of goods in the ius commune - which 
would be equivalent to a penalty of deportation, hence Accursius' test subsumes 
Ubertus' - or in the municipal law of their cities, which corresponds to the penalties 
normally suffered by banniti pro maleficio. The opinion that the bannitus is 
analogous to the deportatus, commonly accredited to the Gloss by later jurists, was 
shared by Guido de Suzzara190 and Dinus de Mugello, 191 two of the most famous 
civilians of the later thirteenth century. Its inclusion in the great works of synthesis 
of that period - Albertus Galeotti' s Margarita, 192 Guilelmus Duranti' s Speculum 
iudiciale, 193 and Albertus Gandinus' Tractatus de malefic its 194 - reveals its standing 
189 Accursius to Inst.1.12,2, col. 58: 'Sed bannitos nostri temporis quibus comparabimus? 
Respondeo: si tantum est maleficium ut etiam cum civitate bona sua publicentur, vel de iure 
vel de consuetudine civitatum, tune deportato comparabo, alias relegato .. .' Cf. Ghisalberti, 
'La condanna', pp. 15-16; Mooney, 'The Legal Ban', p. 161; Cavalca, II bando, pp. 83-5; 
Pazzaglini, The Criminal Ban, p. 18; Milani, 'Giudici, giuristi e fuoriusciti ', pp. 602-3. 
19
° Caprioli, 'La Miscellanea Romana', p. 163 (no. 43 - fols 52vb-3vb) - seep. 197, n. 74 
above; Nicolini, ' Un "consilium" inedito' (summarised expertly by Cavalca, II bando, p. 
198, and also discussed by Milani, 'Giuristi, giudici et fuoriusciti' , pp. 618-9), at p. 354: 
' ... predicti dominus Sensus et alii tres banniantur perpetuo secundum penam deportationis 
que eis imponitur qui in le gem Iuliam de vi publica ... inciderunt...' 
191 Bartolus, Lucanae civitatis, p. 202a, reports: 'Exbanniti nostri temporis aequiparantur 
deportatis, secundum glossam [Inst.1.12,2]. .. quod tenet Dy. in [D.28.1.8, 1 ].' Bartol us also 
recalls Dinus' opinion in his commentary to D.28.1.8,1. 
192 Albertus Galeotti, Margarita, c. 32 De testamentis, pp. 221-2: 'Pone quod bannitus fecit 
testamentum, nunquid valet? die, aut est bannitus ex tali causa, ex qua deportabatur olim, aut 
non. primo casu, non valet testamentum, quia tune bannum succedit loco deportationis ... in 
secundo casu valet, videlicet, si est causa ex qua non deportabatur. .. ' 
193 Guilelmus Duranti, Speculum iudiciale, de arbitro et arbitratore, § 4 differt, v. quid de 
bannito, p. 1 OOa: 'Quid de bannito nostri temporis, quern deportato comparamus si ob grave 
delictum sit bannitus, ut notatum [Inst.1 .12,2] .' 
194 Albertus Gandinus, Tractatus de maleficiis, rub. De bannitis, pp. 130-1: 'Et circa hanc 
materiam nota, quod dictos bannitos equiparamus relegatis, si fuerint banniti pro maleficio, 
pro quo pena relegationis imponitur, si autem fuerint banniti pro maleficio, pro quo 
imponatur pena deportationis, eos equiparamus deportatis, ut notat Accursius [Inst. 1.12,2].' 
Cf. Ghisalberti , 'La condanna', pp. 15, 47; Mooney, 'The Legal Ban', p. 161. 
226 
as communis opinio. 195 Unsurprisingly, Albericus de Rosate reports that 'many' of 
the 'old' doctors and advocates followed the po.sition of the Gloss. 196 
In light of Bartol us' later position, we should ask why the jurists initially considered 
the deportatus a good analogy for the bannitus pro maleficio even though on the face 
of it the latter has a much closer resemblance to the hostis-transfuga. There are three 
plausible reasons for the jurists' choice; these are very similar to the reasons outlined 
in chapter 3 to explain the absence of the hostis analogy in the discourse on imperial 
banishment before Bartolus. 197 Firstly, as Accursius' argument highlights, the 
bannitus does share some central characteristics with the deportatus: exile, loss of 
citizenship, and confiscation of property. Secondly, despite the fact that both are 
outlaws, there are important differences between the bannitus and the hostis which 
complicate a straightforward comparison between the two. As we have seen, the 
criminal ban was, for the most part, a mechanism for dealing with contumacious 
defendants in ordinary criminal proceedings, granted that it could also be used as a 
penalty for political crimes and serve partisan-political purposes. The senatorial 
hostis declaration, on the other hand, was a purely political mechanism outside 
ordinary Roman criminal law, and the hostis-transfuga could only be treated as such 
because he had committed the most extreme form of lese majesty. Thirdly, the sparse 
treatment of the hostis-transfuga in the Corpus iuris, which of course reflected his 
exceptionalness, meant that this figure, although useful in certain quaestiones 
(mainly those surrounding the lawful killing of the bannitus), proved to be of little 
concrete help to the jurists in their discussion of the manifold, quotidian problems 
thrown up by banishment. There was, however, plenty of law on deportation, which 
yielded a number of relevant legal principles. 
This is the reason that the analogy bannitus-deportatus was used both in a systematic 
and casuistic mode during the thirteenth century. If the analogy was granted 
systematic force, and the status of the bannitus was deduced from the status of the 
deportatus , then, as· Bartolus later pointed out, it was only logical that the bannitus 
lost all civil rights. 198 Hence, those jurists following Accursius' opinion argued that 
195 Cavalca, II bando, p. 86, shows that many of the post-Accursians maintained the position 
of the Gloss but does not go as far as calling it communis opinio. 
196 Albericus de Rosate, Quaestiones statutorum IV, q. 13, fol. 67va: 'et hanc glossam secuti 
fuerunt multi antiqui doctores et advocati .' 
197 h C . 3, pp. 127-8. 
198 Bartolus, Lucanae civitatis, p. 240b: 'Idem tenet Dy. in [D.28.1.8,1], et secundum hoe 
perderet exbannitus omnia quae iuris civilis sunt, ut [D.48.19 .17].' 
227 
banniti did not possess testamentary powers. 199 According to Dinus de Mugello and 
Guido de Suzzara, it was another corollary that the bannitus could not become a 
judicial arbiter. 200 As a systematic analogy, deportation was therefore highly 
detrimental to banniti: as we know, loss of civil rights and Roman citizenship meant 
far-reaching disadvantages, especially in family, inheritance, and contract law. Since 
we have no record of the jurists who favoured the deportation analogy discussing this 
issue, it is unclear whether they envisaged the loss of civil rights to exceed the 
territory of the banishing city. A rigorous application of the analogy certainly left the 
way open to such a conclusion, given that the penalty of deportation referred to the 
entire Roman empire. On the other hand, Baldus briefly recalls that the jurist 
Fredericus de Senis (d. ea. 1348),201 a late supporter of the Accursian position, 
considered the effects of the ban limited to the city's territory. 202 
While deportation, as a systematic analogy, was certainly very damaging to banniti, 
we have to remember that the jurists of the thirteenth century only solved a minority 
of questions surrounding the ban on this systematic level. Besides testamentary 
powers and the ability to be an arbiter, they did not explicitly mention any other 
capacities lost through the loss of civil rights. Many questions were solved in a 
199 Albertus Galeotti, Margarita, c. 32 De testamentis, pp. 221-2 - see p. 226, n. 192 above. 
Guilelmus Duranti, Speculum iudiciale, de instrumentorum editione, § 12 compendiose, v. 
quid de bannito, p. 320a: 'Quid de bannito? Die quod, si bannitus est propter levem causam, 
puta pecuniariam, potest testari. si propter criminalem, puta propter aliquid maleficium ex 
quo olim deportabatur, bannitus est perpetuum, non potest... quia tune bannum loco 
deportationis habetur ... de hoe notatum [Inst.1.12,2].' Bartolus to D.28.1.8,1, fol. 91 va: 
'Quaero an exbanniti nostri temporis possint testari. Quidam habent istam quaestionem hie 
formatam per Dy., et distinguunt, aut est exbannitus de crimine propter quod perdit bona, aut 
non, ut notat glossa [Inst.1.12,2].' Cf. Ghisalberti, 'La condanna', pp. 51-2; Cavalca, II 
bando, p. 235. 
200 Bartolus to D.48.13.3, fol. l 84vb: 'Dicit Dyn. quod ista lex est argumentum quod 
bannitus ex eo maleficio ex quo deberet sequitur deportatio, non possit esse arbiter. Nam illa 
verba arbitratu etc. ius significant... sed deportatus amittit pristina iura, ut hie, ergo etc.' 
Albericus also reports Dynus' opinion there. For Guido de Suzzara's opinion, see Guilelmus 
Duranti, Speculum iudiciale, de arbitro et arbitratore, § 4 differ!, v. quid de bannito (with 
Johannes Andreae's additiones), and Albericus de Rosate to D.4.8.7. Cf. Rainerius de 
Forlivio to D.48.13.3. 
201 A Sienese canonist and one of Baldus' teachers at Perugia - cf. L. Bianchin, 'Petrucci ( di 
Petruccio), Federico', DBGI, vol. 2, p. 1566. 
202 Baldus to D.1.18.6,8, fol. 68va: 'sed dominus Fed. de Se. disputando tenuit glossam 
[Inst.1.12,2], scilicet, quod perdat iura civilia tanquam deportatus saltem in civitate sua .. . ' 
Cf. Baldus to D.28.2.29,5 and C.6.24. l (where there is an obvious misprint attribuiing the 
opinion of 'Fed. de Se[nis]' to 'Fed. de Scal[is]'!). Ghisalberti, 'La condanna', p. 45, 
erroneously attributes Fredericus de Senis' opinion to Baldus. 
228 
casuistic fashion that did not involve deportation. As in the case of contract, this 
could redound to the benefit of banniti, since a comparison to deported persons 
would have severely restricted their rights by categorically excluding them from civil 
law contracts. This is a good illustration ·of the fact that the jurists' systematic 
analogies never possessed such force that they governed or structured every part of 
the discourse. As a result, the systematic and casuistic traditions did not always 
combine very well. 
As a casuistic analogy, deportation was mainly employed in the solution of problems 
arising from the confiscation of the banished person's estate. Despite the fact that 
they were not uncommonly adduced as part of a series of diverse analogies and 
principles, references to the law of deportation and the rules contained therein tended 
to take on a central role. This was true right from beginning of the thirteenth century 
until the end of our period, long after the jurists had ceased to use the deportatus as a 
general analogy for the bannitus. As Baldus explained in one of his cons ilia: 
Granted that banished persons are not truly deported persons ... however, by the fact 
that they are deprived of all their goods, they are treated as deported persons in respect 
of their goods, if not in respect of their loss of civil capacity [capitis 
d · · ] 203 emmut10nem ... 
After some initial debate, it thus became communis opinio that the principles of 
property confiscation in the law of deportation decided questions of debt and credit 
in relation to the banished person's estate.204 If his estate had been confiscated, the 
banished person's debtors were now liable to the fisc and the fisc was liable to his 
previous creditors. 205 The jurists applied the same rules to the question of tax. Back 
203 Baldus, cons. 1.284, fol. 86ra: 'licet banniti non sunt vere deportati ... tamen ex quo 
omnibus bonis expoliantur, ad bona, licet non quo ad capitis diminutionem, habentur loco 
deportatorum .. .' A similar comment can be found in Johannes Andreae, additio to Speculum 
iudiciale, de cessione actionis, § 1 sciendum, v. illud etiam quaeritur, and Albericus de 
Rosate, Quaestiones statutorum IV, q. 68, both with reference to the opinion of Hugolinus de 
Fontana - as noted by Cavalca, II bando, p. 86. Cf. H. Angiolini, 'Ugolino Fontana', DBI, 
vol. 48, pp. 723-4. 
204 f h C . c . 2, pp. 56-9. 
205 Azo (Landsberg), Titius commisit delictum; Caprioli, 'La Miscellanea Romana', pp. 138-
9 (no. 21.7 - fol. 47rb-va); Albertus Gandinus, Quaestiones statutorum, qq. 24-6; Guilelmus 
Duranti, Speculum iudiciale, de competentis iudicis aditione, § I generaliter, v. sed pone 
(with Johannes Andreae's additiones); Jacobus Butrigarius to D.17.2.65(66),12; Albericus 
de Rosate to D.46.3.102; Bartolus to D.48.23.1 and D.46.3.102; Baldus to C.4.30.12 and 
C.9.12.2. An important quaestio by Thomas de Piperata on this topic (Jn civitate Pistorii seu 
Florentie - reproduced by Albericus de Rosate at D.46.3.102) is extant in Rome, Biblioteca 
229 
in the first half of the thirteenth century, Odofredus had made a blanket statement 
that banniti should be compelled to pay taxes and render personal services ( collectae, 
publicae functiones) to the city, since otherwise they would benefit from their 
punishment. 206 The commentators argued that this was true if the bannitus retained 
his estate; by implication, this excluded banniti pro maleficio. If the estate was 
confiscated, however, then any public obligations upon it had to me made good by 
the fisc. 207 This not only followed logically from the law of deportation, but also 
made sense in the context of the criminal ban. After all, how could these banniti pay 
taxes and perform personal services when their estates had been confiscated and 
when anyone could injure them with legal impunity? 
ii. Fredericus de Scalis 
Although the position of the Gloss was dominant in the later thirteenth century, 
resistance against it began to arise at that time. Three jurists were at the forefront of 
this reaction: Fredericus de Scalis (d. ea. 1289), Jacobus de Arena (d. ea. 1296), and 
Nicolaus de Matarellis ( d. ea. 1314 ). 208 These jurists were in agreement that the 
general comparison between the bannitus and the deportatus was not intellectually 
viable and too detrimental to the former. They therefore defined the legal status of 
the bannitus in novel and less punitive terms. It is possible that they were induced to 
reach this position because of their own political involvement. As mentioned above, 
we know that Fredericus de Scalis was subject to the repressive measures taken 
against the Ghibelline Lambertazzi faction in Bologna during the 1270s and 1280s; 
the same may have applied to Jacobus de Arena. On the other hand, there is no 
Apostol ica Vaticana, MS. Archivio S. Pietro A. 29, fols I 01 vb-2ra - Bellomo, Quaestiones, 
p. 54 (no. I 13 ). Cf. Mi lani, ' Giuristi, giudici e fuoriusciti ', pp. 605-62 I ; Vallerani , ' Diritti di 
cittadinanza', pp. 6-7. 
206 Odofredus to C.1.5(8). 7, fo l. 32rb: ' Unde colligitur hie argumentum quod non debent esse 
melioris conditionis banniti quam non banniti, quia si banniti non solverent collectas et non 
facerent alias publicas functiones, et non banniti solverent, iam essent melioris conditionis 
banniti quam non banniti...' Cf. Ghisalberti, 'La condanna', pp. 55-6; Cavalca, II bando, p. 
43 . 
207 Bartolus to D.39.2.39,2, fol. 43ra: 'Quare die, si quidem bannitus retinet possessionem, 
tenetur solvere collectas, quia est capax obligationis passivae, ut [D.4.5. 7,3]. Si vero non 
retinet possessionem .. . quia possessiones devenerunt in fiscum ... tune non tenetur solvere, 
quia fiscus habet res quarum nomine tenetur solvere.' Cf. Bartolus to C.1.2(5).5; Rainerius 
de Forlivio and Albericus de Rosate to D.39.2.39,2; Baldus to D.27.3.9,5-6. 
208 For biographical references, see p. 184, n. 30, and p. 192, n. 61, above. 
230 
evidence to suggest that Nicolaus de Matarellis suffered a similar fate. Nor would an 
explanation purely in terms of faction account for the general popularity that their 
views enjoyed during the fourteenth century. It is therefore far from clear to what 
extent social and political background factors , rather than reasons intrinsic to the law, 
facilitated the shift away from the Gloss. 
We will deal with each of the jurists in tum, and begin with Fredericus de Scalis' 
quaestio Titius filiusfamilias, debated in Bologna in April 1289. 209 Its lasting 
importance is demonstrated by the fact that Cinus de Pistoia, Johannes Andreae, 
Albericus de Rosate, Bartol us, and Baldus all report or use Fredericus' quaestio in 
their discussion of the ban; it appears to be reproduced in full by Baldus at D.4.5.5,1 , 
which is the version we will use here. 210 The case is the following. Titius, an 
unemancipated son (filiusfamilias ), was placed under the criminal ban for murder. 
The father of Titius nevertheless instituted Titius in his testament as his official heir, 
and then died. Titius, while still under the ban, transferred his rights (iura) in the 
estate as a gift (in technical terms: cessio causa donationis) to the procurator of his 
grandmother Maria. Following their father's death, Titius' sisters, the daughters of 
the testator, took possession of the estate in Titius' absence and now claim it for 
themselves. The sisters base their claim on the provision of municipal law that 
daughters may (only) inherit the family estate if there is no living, legitimate male 
heir, a condition, they argue, which has been fulfilled by Titius' ban.211 Maria, the 
grandmother, does not accept this and is suing for the estate through her procurator. 
The crux of the matter is whether Titius, while under the ban, was able to inherit the 
estate and, if so, whether he could cede his rights to his grandmother. If Titius was 
unable to inherit, then the estate falls to his sisters. It is assumed that the bannitus is 
209 Extant in: Rome, Biblioteca Apostolica Vaticana, MS. Archivio S. P ietro A. 29 (fols 
169va-70vb ); B ibliote_ca Centrale Nazionale, MS. Vittorio Emanuele 1511 (fols 220rb-1 ra); 
Bologna, Biblioteca de! Reale Collegio di Spagna, MS. 109 (fols 208vb-10ra). Cf. Bellomo, 
Quaestiones, p. 125 (no. 239), pp. 5 10-1 (no. 291), p. 559 (no. 43). 
21
° Cinus de Pistoia to C .6.24. 1; Johannes Andreae, additio to Speculum iudiciale, de 
cessione actionis, § 1 sciendum, v. illud etiam quaeritur; Albericus de Rosate to C.6.24.1 and 
Quaestiones statutorum IV, q. 13; Bartolus, Lucanae civitatis, p. 204b; Baldus to D.4.5.5. 
Despite its importance, Cavalca, II bando, p. 86, only briefly discusses Albericus' report of 
this quaestio. 
211 Lansing, The Florentine Magnates, shows how the increasing patrilineal emphasis in the 
inheritance practices of the Florentines noble elites during the twelfth and thirteenth 
centuries considerably weakened women's inheritance claims. Assuming this was part of a 
broader trend in Italian urban society, the statute under discussion seems to be an expression 
of the latter. 
231 
considered pro confesso, denied all access to justice, and would be immediately 
decapitated after having been captured by communal officials. 212 
The central argument against Maria's case is that Titius, taken to be a deportatus, 
cannot inherit and therefore cannot cede rights that he does not possess. Fredericus, 
having noted the position of the Gloss, provides several reasons why a bannitus 
should not be compared to the deportatus, stressing key dissimilarities between the 
two figures. As we know, the Gloss states that a bannitus is like a deportatus if he 
loses his citizenship as well as his goods, either according to the ius commune or 
according to municipal law. But, Fredericus argues, the bannitus does not lose his 
goods according to the ius commune because Justinian's authentica Bona 
damnatorum has repealed property confiscation for crimes resulting in deportation; 
nor does he lose his property according to local statute, because in this case the latter 
(at least according to Fredericus) does not impose this penalty on banniti. Secondly, 
the ban runs contrary to the principles of Roman criminal procedure and therefore 
ought to be interpreted restrictively. Given that statute imposes certain penalties on 
the bannitus, which do not include deportation, no interpretation of statute in this 
matter should infer what is not explicitly stated. Thirdly, the bannitus is condemned 
by the commune's podesta, who is normally compared to a municipal magistrate or a 
provincial governor in Roman law. These officials, however, do not have sufficient 
jurisdiction to deport; as we saw in chapter 2, this right is restricted to the highest 
imperial dignitaries.213 Finally, the bannitus is by definition absent, whereas in the 
ius commune an absentee cannot be deported, but is subject to annotation instead. 
For these reasons, the bannitus cannot be treated as a deportatus. Ergo, Titius 
possesses testamentary powers and, as the lawful inheritor of his father's estate, can 
cede his rights as a gift to Maria, who is able to sue for them in court. 214 
2 12 Baldus to D.4.5.5, 1, fol. 249va: 'Quaestio talis est. Titius filiusfamilias solemniter positus 
est in banno tanquam homicida. pater instituit sibi haeredem ilium Titium et mortuus est. ille 
filius... causa donation is cessit iura procuratori Mariae suae aviae quae ha be bat in dicta 
haereditate et in bonis. filiae testatoris tenent bona. Maria vult experiri de illis bonis contra 
filias testatoris. et scias quad ex lege municipali extante filio filia non succedit... Item, scias 
ex lege municipali bannitus habetur pro confesso, et veniens in fortiam communis 
decapitatur, et querimoniae eius non audiuntur.' 
213 Ch. 2, p. 52. 
214 Baldus to D.4.5.5, 1, fol. 249va-vb: 'Unde contra Mariam excipitur primo quad ille filius 
bannitus non potest esse haeres quia non capax quasi deportatus. Item quad non valet cessio 
quia filius non se immiscuerat, unde nullum ius habebat, et idea nullum transtulit. [ ... ] 
solutio: salvo meliori consilio, mihi, Federico de Scalis, legum doctori, videtur quad iste 
Titius, qua ad veritatem, licet sit bannitus pro homicidio, non sit deportatus nee habeatur pro 
deportato, et quad inducitur [Inst.1.12,2], pari ratione, dico quad aperte demonstrat 
232 
This is an extremely lenient interpretation of the effects of the ban. As we shall see in 
due course, although the new communis opinio of the fourteenth century took on 
board Fredericus' arguments against using the· deportatus as a general analogy for 
the bannitus, it rejected the idea that the latter could benefit from municipal statutes 
that granted favourable rights to the city's citizens - in the scenario of Titius 
filiusfamilias, the prerogative of a male heir to inherit to the exclusion of female 
relatives. Throughout our period, communis opinio also diverged from Fredericus' 
view on the cession of rights.2 15 As Albericus de Rosate explained, it was the opinion 
of almost every jurist that a bannitus could only cede the rights that he did possess in 
certain defined ways: 
The doctors of law have become accustomed to distinguish whether the person to 
whom the cession was made is merely a procurator or acting in his own name. In the 
first case, the objection [of banishment] stands, since the procurator represents the 
person of the principal, and therefore he ought to be of the same condition as the 
latter .. . In the second case, when the cessionary acts in his own name, the doctors 
distinguish whether the cession was made ex causa lucrativa, in which case the 
objection stands, or whether it was made ex causa onerosa, in which case it does not, 
as in [D.44.4.4,27-9].216 
contrarium. nam ibidem dicitur quad ille comparatur deportato qui bannitus est de tali 
maleficio de qua, secundum legem, deberet perdere civitatem [et] bona, vel secundum 
consuetudinem vel statuta civitatis. et ex hac lege non perdit bona, [auth. Bona damnatorum, 
post C.9.49.10]. Item, ex statuto non perdit bona, nam poenae bannitorum ex forma statuti 
sunt certae et limitatae, inter quas non est amissio bonorum. [ ... ] nam cum iste bannitus sit 
contra iura communia, non est ad consequentias protrahendum .. . at in statuto non continetur 
quad habeatur pro deportato, sed aliae poenae sunt in eo statutae, unde stabo contentus. Item, 
ipse bannitus est a defensore civitatis vel a praeside provinciae, qui deportare non potest, ut 
[D.31.1.1 ,3] . quomodo ergo posset deportatis assimilari? Item, quia certum est absentem 
neminem deportari vel in tantam poenam damnari... Item, quia manifestissime legitur eum 
esse requisitum et annotatum ... Cessant ergo in eo omnia iura quae loquuntur de capitalibus 
criminibus et de non capacibus, quia iste capax est, cum non sit deportatus nee assimiletur 
deportato. Potuit ergo cedere iura sua et transferre in quamcunque personam. [ ... ] Nee dicas 
tu quad per procuratorem huius aviae nihil potuit ei quaeri.. . Nee dicas tu quad avia, quae 
agit ex titulo lucrativo, repellatur propter vitium autoris ... ' 
215 Cf. Ghisalberti, 'La condanna', pp. 56-60 (a rather muddled analysis); Cavalca, II bando, 
fiP· 230-5; Vallerani, 'Diritti di cittadinanza', p. 5. 
16 Albericus de Rosate, Quaestiones statutorum IV, q. 4, fol. 66vb: 'Doctores consueverunt 
distinguere, aut ille, cui cessio est facta, est simplex procurator, aut in rem suam. Primo casu, 
obstet sibi exceptio [banni], quia repraesentat personam domini, et immo eiusdem debet esse 
conditionis ... Secundo casu, quando in rem suam, distinguunt, aut habet cessionem ex causa 
lucrativa, et sibi obstet exceptio, aut onerosa, et tune non, per [D.44.4.4,27-29].' Cf. Albertus 
Galeotti, Margarita, c. 4 De procuratore in rem suam, p. 32; Guilelmus Duranti, Speculum 
iudiciale, de cessione actionis, § 1 sciendum, v. illud etiam quaeritur; Albertus Gandinus, 
Quaestiones statutorum, qq. 93-4; Bartolus to D.46.2.19 and D.44.4.4, 18; Baldus to 
233 
The first distinction was based on common sense. A bannitus could not simply have 
a representative sue for his rights in court and thereby circumvent the statutory 
prohibition on banniti receiving justice. But if the bannitus had ceded his rights to 
another person such that the latter was now suing in his own name, then the jurists 
applied the causa lucrativa/onerosa distinction from the Roman law of fraud (dolus 
malus), taking the exceptio doli as analogous to the exceptio banni. Lucrative cause 
occurs when rights are ceded for free, as a gift for example.2 17 Onerous cause applies 
when the cessionary purchases those rights. In the first case, the exceptio banni 
defeats the cessionary's action, in the second it does not. If we apply this scheme 
back to the case of Titius .filiusfamilias and assume that Titius lawfully inherited the 
estate, this means that while his gift to Maria was perfectly valid, Maria cannot sue 
for her rights in court because Titius' sisters can block her action through the 
exceptio banni. Had Titius sold his rights to Maria, however, then she would be able 
to sue. 
iii. Jacobus de Arena 
The contribution of Fredericus de Scalis, Jacobus de Arena, and Nicolaus de 
Matarellis to the civilian discourse on banishment has been undervalued by the 
modem historiography. The most discernible cause for this is the fact that their 
quaestiones, with the exception of the material in Jacobus' Tractatus de bannitis, 
were not printed separately during the early modem period; in the printed tradition, 
we are therefore almost exclusively reliant on reports contained in the works of 
others. One of the reasons why each of these jurists is discussed separately here is to 
give their thought the prominence that it deserves. The evidentiary problem is 
particularly complex regarding Jacobus de Arena. It has been established thus far 
that 'lac. de Aret.' does not refer to a different jurist and that his Tractatus de 
bannitis is authentic. Jacobus also authored five distinct quaestiones disputatae on 
the ban, citations and reports of which are scattered across the works of Albertus 
Gandinus, Johannes Andreae, Albericus de Rosate, and Baldus. We have already 
discussed some of these, including Jacobus' most important, and at the same time 
C.2.12.9. Besides Fredericus de Scalis, Jacobus de Arena, Tractatus de bannitis, fol. 356rb, 
is the only other jurist to reject communis opinio by providing a similar, if subtly different, 
scheme. 
217 Berger, Encyclopedic Dictionary, p. 383, is a useful guide to causa lucrativa. 
234 
most illusive, quaestio Statutum est in civitate quod bannito ratio non.fiat.218 Later 
reports make it relatively clear that this covers the following two sub-questions: 
1. Are statutes permitting the killing of banniti lawful? 
2. Can a bannitus sue for property and contractual damages suffered during his time of 
banishment after his ban has been cancelled? 
The contents of these sub-questions only survive in Johannes Andreae's additiones 
to the Speculum iudiciale, which formed the basis for our discussion of them above. 
According to Baldus, however, the quaestio also contains a further three sub-
questions, although he does not report their contents: 
3. What do the terms bannitus,Jorbannitus, and dijjidatus signify? 
4. Can these individuals be compared to the hostis? 
5. Can the bannitus be called a rebellis?219 
These sub-questions are tantalising, and if Jacobus did indeed conclude that the 
bannitus was analogous to the hostis, this quaestio may constitute an important line 
of influence from Jacobus to Bartolus. That possibility is supported by Baldus' 
association of Jacobus' name with the Barto list position in a comment at the I. 
218 [1.] Statutum est in civitate quod nullus bannitus possit esse consul (p. 201 , n. 93). [2 .] In 
statuto vel consuetudine alicuius civitatis (p. 212, n. 132). [3 .] Si lata sententia pro bannito 
(p. 212, n. 133). [4.] Statutum est in civitate quad bannitus potest impune offendi (concerning 
the unlawful killing of a banished cleric), reproduced by Albericus de Rosate, Quaestiones 
statutorum IV, q. 29. Bellomo, Quaestiones, p. 177-8 (no. 345), notes that Albericus cites the 
same quaestio at D.2.6.2 . [5 .] Statutum est in civitate quad bannito ratio non.fiat (p. 200, n. 
90; p. 202, n. 99; p. 206, n. 112; pp. 208-9; pp. 221-4) - seen. 219 immediately below for 
further details. 
2 19 Johannes Andreae, additio to Speculum iudiciale, de accusatione, § 6 sequitur, v. quid si 
bannitus, p. 19b: 'Habetur alia et laco. de Aret. quadrimembris quaestio. Primo, qua ponitur 
thema de statuto civitatis quo prohibetur ratio fieri bannito pro maleficio nisi pro homicidio 
vel mutilatione. Caius bannitus mutuavit Titio. damnum passus est, vel furtum et rapinam. 
quaeritur an, exemptus de banno, agere possit de praedictis damnis quae passus est de 
mutuo? Item, eadem quaestiones formantur de haerede illo mortuo in banno vel post 
exemptionem. Prosequitur autem prius an talia statuta valeant...' Baldus to D.4.5.5, fol. 
249vb: 'An exemptus de banno possit de iniuria sibi interim facta, vel ex contractu interim 
inito, agere, et quid est dicere bannitus, quid defidatus, quid forbannitus, et an hosti 
comparantur, vide in quaestione disputata per lac. de Are., quae incipit "statutum". ibi etiam 
ponit an exbannitus possit dici rebellis.' Albericus de Rosate, Quaestiones statutorum IV, q. 
2, fol. 66ra: 'et vide quaestionem disputatam per la. de Are., quae incipit "statutum est in 
civitate quod bannito ratio non fiat" ... ' In q. 89, fol. 82va, Albericus lists this quaestio as no. 
284 in the Liber magnus quaestionum disputatarum. It is unknown to Bellomo, Quaestiones. 
235 
11111 
Amissione that we discussed above.220 On the other hand, it is striking that Bartolus 
himself never attributed to Jacobus the idea of using the hostis as an analogy for the 
bannitus, even though (as we will see) he cited Jacobus generously regarding other 
aspects of his thought on banniti. Given how obsessed Roman lawyers were with 
scholarly authority, it seems highly unlikely that Bartolus would not have cited 
Jacobus in support of his argument had he in fact been able to do so. Once again, I 
consider the explanation given above for Baldus' comment at the l. Amissione to be 
the more likely one. What Baldus presents as three proper sub-questions is merely a 
reference to Jacobus' casuistic employment of the hostis analogy in the context of 
property damages, so that he can present Jacobus as an authority for Bartolus ' 
position. 
Jacobus' surviving systematic thought on banishment is contained in the opening 
section of his Tractatus de bannitis and in his commentary to D.36.1.18(17),5 (I. Ex 
facto) and C.6.24.1 (I. Qui deportantur). 22 1 There, he rejects the Gloss' comparison 
between deportatus and bannitus for two reasons. Firstly, as Fredericus de Scalis had 
pointed out, in Roman law, municipal magistrates and provincial governors cannot 
deport, hence neither can today's podesta, who are usually compared to the 
former.222 Secondly, deportation and banishment are too dissimilar, for: 
... the whole world is forbidden to a deported or relegated person except that island to 
which he is deported or relegated. But the opposite applies to the banished person, 
since he can remain and live everywhere except in the city from which he was 
banished .. . 223 
This acknowledgement of what was no doubt a natural result of legal practice is the 
first clear indication in the theoretical discourse that the ban was territorially limited 
to the city that issued it; the idea formed a set part of communis opinio thereafter. But 
220 See pp. 208-9 above. 
221 Cf. Ghisalberti, 'La condanna', pp. 16-17, 20, 47; Mooney, ' The Legal Ban', p. xi; 
Pazzaglini , The Criminal Ban, pp. 18-19; Cavalca, Il bando, pp. 53-5, 87, 236; Milani, 
'Giuristi, giudici e fuoriusciti', p. 620. 
222 Jacobus de Arena to D.36.1.18( 17),5, fol. l 27vb: 'Quid de bannitis nostri temporis? 
dicunt quidam idem quod in deportato... Sed contra videtur quia sententia deportationis 
etiam a presidibus dari non potest... multo minus a civitatum rectoribus.' The same idea is 
found in his comment to C.6.24.1. 
223 Jacobus de Arena, Tractatus de bannitis, fol. 355va: 'Verum quidam dicunt, et non male, 
quod ilia glossa non est vera, quia maxima dissonantia est inter eos, quia deportato vel 
relegato totus mundus interdicitur excepta insula in quam deportatur vel relegatur. sed in 
bannito est econtra, quia ubique potest stare et habitare excepta civitate ex qua exbannitur ... ' 
236 
even in the banishing city's territory, the bannitus should not really be considered a 
deportatus, since: 
.. . he does not lose those th ings that are derived from the civil law of the Romans on 
account of the ban, just as a relegated person .. . I believe, however, that he loses those 
things proper to the city from which was banished, since he is a stranger [peregrinus] in 
respect of the latter, as at [C.6.24.1 ] .. .' 224 
Jacobus appears to be struggling to find a suitable analogy for the bannitus in Roman 
law. Whereas in one respect the relegatus is an inappropriate analogy due to 
Jacobus ' view on the geographical dimensions of relegation, that figure is a good 
analogy in the sense that he is an exile who does not lose his Roman citizenship. The 
distinction that follows was as obvious as it was original. It was obvious because, as 
we saw in chapter 1, it is a core tenet in the Corpus iuris that each person possesses a 
vertically integrated dual citizenship: universal Roman citizenship and the 
citizenship of a local city.225 It made perfect sense to say that through the ban of a 
local city, a bannitus lost the citizenship of the latter while retaining Roman 
citizenship. It was original because there is no such thing as withdrawal of local 
citizenship in Roman law. This is why, ironically, Jacobus refers again to deportation 
towards the end of his comment with his reference to C.6.24.1, which states that 
deportati lose their testamentary powers because they become peregrini. Jacobus' 
argument is that the same principle should apply in respect of the banished person 's 
former local city. Just as the deportatus loses the rights of Roman citizenship, so the 
bannitus loses the rights of citizenship of a city within the empire. Later jurists 
widely credited Jacobus with this idea and, as we shall see, it became a central pillar 
of the new communis opinio of the fourteenth century.226 Jacobus himself concludes 
from this that the bannitus retains the right of patriapotestas227 as well as his 
testamentary powers.228 Unfortunately, he does not given any indications as to the 
contents of local citizenship. 
224 Jacobus de Arena to D.36.1.18(17),5, fol. 127vb: 'Non ergo perdit que sunt iuris civilis 
Romanorum propter bannum, sicut relegatus ... Ea autem que sunt propria civitatis a qua est 
bannitus amisisse credo, quia quantum ad illam peregrinus est, ut [C.6.24.1 ] ... ' 
225 Ch. 1, pp. 7-8 . 
226 E.g. Albericus de Rosate and Bartolus to D.36.1.18(17),5; Bartolus, Lucanae civitatis, p. 
205b; Baldus to D.4.5.5,1 and C.6.24.1. 
227 Jacobus de Arena, Tractatus de bannitis, · fol. 355va: 'Item, bannitus filios in potestate 
retinet, quia certi sunt modi quibus patriapotestas solvitur, inter quos bannum non 
numeratur.. .' 
228 Jacobus de Arena to C.6.24.1, fol. 40ra: ' ego dico quod bannitus potest institui, nee est 
similis deportato .. . unde dicam bannitum posse institui, cum non inveniam prohibitum. lac. 
de Are.' We should note Jacobus' comment to D.37.1.13 (which states that the deportatus is 
237 
111111
1 
iv. Nicolaus de Matarellis 
Nicolaus de Matarellis' thought survives in a number of forms. His most important 
work on the ban is the quaestio Quidam inculpatus, debated in Padua on an unknown 
date. 229 It is reported by Johannes Andreae in his additiones to the Speculum 
iudiciale 230 and by Albericus de Rosate, both in his commentaries and in his 
Quaestiones statutorum.23 1 Many of the same ideas as in the quaestio can be found in 
the relevant sections of Nicolaus' Lectura on the Code, which Baldus directly quotes 
at length as part of his commentary to C.6.24.1 . 232 
Quidam inculpatus posits the following scenario. Someone was banished for murder 
and all his goods were confiscated. Afterwards, the bannitus was instituted as 
another person's heir, took possession of the estate, but then died. Now the banished 
person's own heir wants to sue for that estate, including those goods located beyond 
the territory of the city from which the deceased was banished. The principal 
question is whether or not the whole estate passes to the banished person's heir.233 
excluded from bonorum possessio), fol. 132va: 'Argumentum huius legis quod si aliquis est 
bannitus, aut perpetuo, et tune non potest heres institui, aut ad tempus, et tune potest institui.' 
Given that this would be in direct contradiction to what is unambiguously Jacobus' thought, 
and given that the remark is non-committal, it most likely constitutes a report of the opinion 
of others (if it is not an interpolation). Cavalca, I! bando, p. 236, nevertheless considers this 
comment to represent Jacobus' own thought. Ghisalberti, 'La condanna', p. 52, also 
erroneously argues that Jacobus, in his treatise, followed Accursius' position in the case of 
testament, by confusing the question whether the bannitus possesses testamentary powers 
with the procedural question whether he, or his heir, is able to sue for an inheritance when 
faced with the exceptio banni. 
229 Extant in Rome, Biblioteca Apostolica Vaticana, MS. Archivio S. Pietro A. 29 (fols 
273vb-6ra), MS. Chigi, E.VIII. 245 (fols 108vb-9rb) - with Matarellis as the author in both 
cases - and MS. Chigi (fols 61 rb-2vb ), where it is assigned to Jacobus Butrigarius. Cf. 
Bellomo, Quaestiones, p. 181 (no. 351), p. 245 (no. 70), and p. 276 (no. 146). Given that 
both Johannes Andreae and Albericus de Rosate attribute the quaestio to Matarellis, the 
attribution to Butrigarius in MS. Chigi is surely erroneous. 
230 Johannes Andreae, additio to Speculum iudiciale, de accusatione, § 6 sequitur, v. quid si 
bannitus, p. 18b: 'Nico. de Mata. in quaestione quae incipit "quidam inculpatus" ... ' 
231 Albericus de Rosate to C.6.24.1; Quaestiones statutorum III, q. 2, and IV, qq. 1 and 13. 
232 Baldus to C.6.24.1, fols 69va-70rb: 'Ex praedictis habes quod est bannire sive bannitio ... 
secundum Nico. de Ma. hie et [C .8.40.1] in sua lectura, unde circa effectus bannitionis ibi sic 
ait. [ ... ] secundum Nie. de Mat., cuius sunt haec verba.' 
233 The background of the quaestio is most clearly set out by Albericus de Rosate, 
Quaestiones statutorum IV, q. 13, fol. 67vb: 'Et quaestio est talis. Quidam bannitus fuit de 
homicidio et eius bona omnia publicata. postquam ipse ab aliquo haeres institutus. adita 
haereditate, decessit. nunc haeres eius contendit de bonis, quae ipse bannitus in alia civitate 
238 
Nicolaus separates the quaestio into a number of subsections that discuss the 
following problems. Firstly, does the bannitus possess testamentary powers so that 
he can receive the estate and then pass it on to his heir?234 Secondly, does property 
confiscation apply only to the goods that the bannitus had at the time of his sentence, 
or also to the goods that he received in his subsequent inheritance?235 Thirdly, does 
property confiscation extend beyond the territory of the banishing city?236 
Before answering these specific questions, Nicolaus spends some time discussing in 
depth the nature of the ban as well as some of its procedural facets; he appears to be 
the first, and only, jurist to have done so, which partly explains his later 
popularity. 237 As this introductory section is for the most part a summary of 
arguments more clearly set out in his Lectura on the Code, we will mainly rely on 
the latter as reported by Baldus. Nicolaus starts by offering his interpretation of the 
etymology of the word bannum, -i, and its cognates (bannio, -ire, -ivi, -itum; 
bannitio, -ionis), which, according to him, derives from banderia, a military standard 
used in battle, and the verb 'to go' (ea, ire, ivi, itum). This combination embodies the 
notions of a public announcement and public proceedings. Hence, bannum can 
signify a judicial proclamation or order, much like edicts in Roman law. Nicolaus, 
however, swiftly moves on to the related concept of bannum as a coercive device and 
penalty in criminal proceedings. 238 He notes that the ban is not contained in the old 
laws_ of Justinian, but only in the new laws of the medieval emperors and cities. It 
arose because of the fragmentation of the ancient Roman empire, which made the 
administration of the old criminal law too difficult. Its purpose, however, is exactly 
habebat et de aliis bonis praedictae haereditatis ei delatae. quaeritur an ipsa bona ad ipsum 
haeredem debeant pertinere an ad alium?' 
234 Albericus de Rosate to C.6.24.1, fol. 32va: 'Quod bona praedicta spectent ad haeredem 
praedictum .. .' 
235 Ibid.: 'Deinde discutit utrum publicatio facta in sententia banni extendatur ad illam 
haereditatem postea delatam bannito.' 
236 Ibid.: 'Quod ad alium articulum, de bonis quae habebat bannitus in alia civitate, quod non 
includantur in publicatione ... ' 
237 Briefly covered by Ghisalberti, 'La condanna', pp. 31 and 41, and by Cavalca, I! bando, 
pp. 19-22, 88, and 102-3. Both authors, except Cavalca, Il bando, on pp. 19-22, attribute the 
ideas to Albericus de Rosate rather than Nicolaus. 
238 Albericus de Rosate to C.6.24.1, fol. 32vb-ra: 'Ad prim um dicit quod bannum dicitur a 
"banderia", quod est parvum vexillum, et a verbo "eo, is", unde "bannio", idest, publice 
contra alium "eo", idest, procedo, quia publici processus fiunt contra contumaces ... et ideo 
talis processus, cum publice fit, maxime per publicam praeconizationem, appellatum est 
bannum ... si intendatur ad notificationem vel comminationem publicam, habeat vim edicti ... 
sect si intendatur ad poenae impositionem, tune intelligatur publica deiectio .. .' 
239 
the same: firstly, to force the contumacious to obey, and secondly, to punish 
criminals. 239 
Nicolaus rejects the Accursian deportation analogy but does not compare the 
bannitus to the relegatus either. Granted that the bannitus has some similarities with 
both of these Roman law figures, the former is too different for any direct 
equivalence. 240 According to Nicolaus, the best analogy for banishment is 
excommunication in canon law. There are two reasons for this. Firstly, their effects 
are similar: 'just as excommunication excludes a man from the communion of the 
church in respect of spiritual matters, thus banishment excludes a man from the 
public good'. They are both outlaws, barred from invoking the aid of the courts and 
officials in their respective legal spheres. 241 Banishment is, in sum, 'public 
expulsion' (publica deiectio) in the form of 'secular excommunication' (saecularis 
excommunicatio ). 242 This is an astute observation; there seems little doubt that 
banishment and excommunication, generally speaking, were equivalent mechanisms 
in civil and canon law. As we saw in chapter 3, they served very similar purposes, 
which is why they both possessed coercive procedural and permanent punitive 
elements.243 In fact, procedural similarities are Nicolaus' second reason for choosing 
the excommunicatus as his preferred analogy. Comparing the imperial ban in the 
Frederician constitutions to excommunication in the Decretals, he correctly 
highlights that both can be conferred by a judicial sentence or ipso iure, and that both 
can result from contumacy ( coercion) or be the proper penalty for a crime 
239 Baldus to C.6.24.1, fol. 69vb: ' ... notandum est quod, cum in legibus antiquis inducti sint 
diversi modi cogendi ad obedientiam contumaces et puniendi delinquentes ... et bannitiones 
ad eundem finem sint inductae ... sciendum est quod bannitiones in legibus Iustiniani non 
nominantur, sed a legibus novissimis, ut in usib. feud., de statu. contra libe. eccl., et deinde a 
legibus municipalibus et consuetudinibus civitatum. [ ... ] Et quia ... non poterant facile 
adimpleri, specialiter propter divisionem imperii et diversitatis iurisdictionem, inventae sunt 
bannitiones .. . ' 
240 Ibid., fol. 69va: 'Sed ego circa hoe dico, quod nee deportatis nee relegatis sunt 
comparandi in totum. Est, nota, bannum quaedam poena per se, quae nee est deportatio nee 
relegatio, licet aliqua habeat similia cum utraque .. . ' 
241 Ibid. : ' ... videmus quod assimilatur excommunicationi, ut sicut excommunicatio eiicit 
hominem extra communionem ecclesiae, quo ad spiritualia, ita bannitio eiicit hominem extra 
bonum publicum, id est, gaudere non debeat auxilio reipublicae nee officiis publicis nee 
iurisdictione iudicis adiuvetur... ' 
242 Ibid., fol. 69vb: ' ... bannitio, quae est quaedam secularis excommunicatio ... ' Albericus de 
Rosate, Quaestiones statutorum IV, q. 1, fol. 65vb: ' ... bannum est publica deiectio .. .' 
243 Cf. eh. 3, pp. 106-110. Ghisalberti, 'La condanna' , pp. 37-41, considers some parallels 
between excommunication and banishment. 
240 
(punishment).244 Hence, as Nicolaus sensibly points out, the law speaks of them 
jointly at LF.2.28, c. Domino guerram.245 
To explain the sanctions that the ban minimally ought to impose, Nicolaus employs 
the language of canon law and essentially describes the effects of excommunicatio 
maior as we encountered them in chapter 3: exclusion from offices (officia, honores, 
dignitates), the courts (iurisdictio iudicantis), and public legal acts (actus 
forenses). 246 To illustrate this point in Quidam inculpatus, Nicolaus cites canon law 
as well as Frederick's constitution Credentes, which, as we have seen, gives force to 
many of the effects of excommunicatio maior in civil law. 
247 In other words, 
according to Nicolaus, the ban results in those sanctions in the secular sphere that the 
excommunicate suffers in the ecclesiastical sphere. We should bear in mind that this 
does not mean that the bannitus automatically becomes an excommunicate ( or vice 
versa); in Nicolaus' thought, they only happen to be subject to the same penalties in 
their respective, but separate, legal spheres. 
Nicolaus makes clear that secular excommunication is only the default interpretation 
to fall back on should local statute and custom not regulate the ban; but if the latter 
do so (which, in practice, of course they did), then they should be followed - a point 
that Nicolaus does not find necessary to justify. He provides the following 
guidelines. Firstly, if statute or custom does not regulate the procedure of banning, 
then, if the ban is pro contumacia, the rules of annotation in Roman law and the 
procedural rules of excommunication in canon law should apply - Nicolaus 
emphasises the need for a lawful summons. If the ban is pro delicto, then it is simply 
the penalty imposed on a present or absent defendant at the end of an ordinary 
244 Baldus to C.6.24.1, fol. 69va: 'unde, sicut excommunicatio quandoque a iure, quandoque 
a iudice infertur, ut [X.5 .39.49/58], et infertur quandoque propter contumaciam, quandoque 
propter offensam, ut [X.5.40.23], sic exbannitio quandoque fit a lege, ut in usi. feu., in rub. 
de statutis contra libertatem eccl. c. fi. § item quaecunque, et quandoque ab homine, ut infra 
eo. tit. § item nulla.:. Item, communiter fit quandoque propter contumaciam, quandoque 
propter delictum, ut de facto videmus fieri. argumentum est in dictis §§ et in § gazaros .. .' 
245 Ibid. , fol. 69va: 'et hanc similitudinem posuit textum in [LF.2.28 c. Domino guerram].' 
246 Ibid., fol. ?Ora: ' ... sicut excommunicatus repellitur a bonis ecclesiasticis, ita per 
bannitionem intelligo bannitum. .. esse deiectum ab officiis, honoribus, et dignitatibus 
territorii iudicantis, et interdictam ei iurisdictionem iudicantis ad similitudinem 
excommunicationis ... et ab omnibus forensibus actibus esse in illo territorio remotum .. .' Cf. 
eh. 3, p. 107. 
247 Albericus de Rosate to C.6.24.1, fol. 33ra: 'Cum igitur bannum sit deiectio a publico 
bono, et sit simile excommunicationi, intelligo bannitum deiectum a publicis bonis, officiis, 
honoribus, et actibus, ad exemplum excommunicati, ut [Sext.5 .1 1.8]. .. et [auth. Credentes, 
post C.1.5(8).4].' 
241 
I 
criminal trial. 248 Secondly, if statute or custom does not specify the punitive effects 
of the ban, and it is pro contumacia, then it simply takes the form of secular 
excommunication as explained above. A ban should be understood as pro 
contumacia if it is given during proceedings that could not be brought to their 
ordinary conclusion, in lieu of a definitive judgement that would normally result in a 
different penalty. If the ban is pro delicto, and if the ius commune has a penalty for 
the crime under consideration, then the ban simply takes the form of that penalty. 
Interestingly, this may include deportation (including the loss of Roman 
citizenship!), but its effects are restricted to the territory of the city. If the ius 
commune has no penalty for the crime, then the ban takes the form of secular 
excommunication with the additional penalty of relegation from the territory of the 
city. Relegation is a good default penalty to fall back on since it can be given to 
present as well as absent defendants in Roman law, and thus covers the ban's range 
of application.249 Just as in the case of relegation - and this applies both to the 
bannum pro contumacia and bannum pro delicto - confiscation of goods is not an 
assumed part of the penalty but must be expressly included in the sentence. 250 It is 
248 Baldus to C.6.24.1, fol. 69vb: 'De forma autem tenenda in processu bannitionis faciendae 
per iudicem, puto ante omnia tenendum quod consideretur quod, si aliqua forma super hoe 
data est a lege vel consuetudine loci, ubi bannitio fit, illa servanda est... quandoque fit 
propter delictum, ut cum <licit iudex vel lex, bannimus latrones omnes vel assassinos, et tune 
ita demum poenam et effectum bannitiones patiuntur aliqui si sunt de tali facto inculpati per 
sententiam ... quandoque propter contumaciam ... sed si null a forma reperitur ... tune poterit 
fieri bannitio, si legitime citatus, non venerit ad iudicem, sicut dicitur in excommunicatione ... 
et in annotatione bonorum ... ' Cf. Albericus de Rosate to C.6.24.1, fol. 33rb: 'Si pro delicto, 
sive in abs en tern sive praesentem lata sit sententia banni ... ' 
249 Baldus to C.6.24.l, fol. 70ra-b: 'Si autem non est certus effectus declaratus a lege vel 
consuetudine loci, nee in sententia bannitionis ... et est bannitio propter poenam delicti ... tune 
intelligitur talem habere effectum qualem deberet habere condemnatio facta ex illo delicto 
secundum legis communis ordinationem, ut... bannitus ex delicto, cuius poena secundum 
communem le gem est deportatio ... deportari de beat... et perdit commoda suae civitatis, et 
quaecunque a iure civili sint inducta in territorio suae civitatis perdat... sed si per legem 
communem in tali delicto non est certa poena declarata, tune erit talis effectus simplicis 
relegationis, quae tarn in praesentem quam in absentem proferri potest, et propter earn 
eiicitur a bono publico ... sicut excommunicatus ... quasi haec sit secularis excommunicatio ... 
Si autem bannitio pro contumacia facta est, et... si super tali banno nullus effectus est 
inductus per legem vel consuetudinem loci, intelligo talem bannitionem censendam esse 
quandam secularem excommunicationem ... et intelligo per contumaciam bannitum ill um, qui 
ideo bannitus est quod inobediens fuit iudici intendenti ad diffinitivam negocii procedere ... ' 
Cf. Albericus de Rosate to C.6.24.1, fol. 33rb: ' ... et intellige pro contumacia quando alia est 
imponenda poena pro delicto ex forma statuti ... ' 
250 Albericus de Rosate to C.6.24. l, fol. 33rb: 'et hoe sive interveniat expressa publicatio 
bonorum, quod quandoque fit etiam in relegatione. .. sive non, nee intelligitur tacita 
publicatio, si expresse non fiet, ut dicitur in relegato, [D.48.22.4].' 
242 
striking that Nicolaus imbues the canon law and excommunication with such 
normative power in these considerations. As we have seen, it was usual for the 
civilians to fall back purely on Roman law mechanisms, such as annotation and 
relegation, given the authority of the Corpus iuris in their thought. 
Following his introductory remarks, Nicolaus returns to the specific questions of 
Quidam inculpatus. He reiterates his agreement with Fredericus de Scalis and 
Jacobus de Arena (though he does not cite them) that the bannitus is not like the 
deportatus, setting out all the reasons already adduced by the other two jurists.25 1 
Even though Nicolaus does not view the relegatus as the best general analogy for the 
bannitus, as we have seen, in questions of testament, however, the latter has more in 
common with the former than with any other figure. It seems that Nicolaus did not 
reach for the analogy of excommunication in this context since, as Albericus notes in 
one of his additiones to this quaestio, canonistic communis opinio denied the 
excommunicate testamentary powers. 252 Nicolaus provides four key reasons for 
comparing the bannitus to the relegatus. Firstly, just as property confiscation may 
result from the ban, the same penalty may result from relegation if this is specifically 
included in the sentence. Secondly, like banishment, relegation can be a penalty for 
present as well as absent defendants, whereas an absentee cannot be deported. 
Thirdly, as in the case of banishment, someone can be relegated from a certain 
geographical area, whereas deportation can only assign that person to an island.253 
Finally, using relegation as an analogy lessens the severity of the sentence, which 
one should always aim to do in the juristic interpretation of criminal matters. Hence, 
just as the relegatus, the bannitus retains his testamentary powers and only loses 
251 · Cf. Albericus de Rosate to C.6.24.1, fol. 32va; Quaestiones statutorum IV, q. 13, fol. 
67vb. Cavalca, II bando, p. 84, erroneously argues, by confusing the contra section of q. 13 
with its solutio, that Albericus (not Nicolaus!) accepted the comparison bannitus-deportatus 
in his Quaestiones. 
252 Albericus de Ros~te, Quaestiones statutorum IV, q. 13, fol. 68rb: 'Sed quia supra dixi 
bannum et excommunicatione parificari .. . quaero utrum excommunicatus habeat testamenti 
factionem? [ ... ] Canonistae communiter tenent quod non sit testabilis.' 
253 Albericus de Rosate to C.6.24.1, fol. 32va: 'sed fortior est in banno natura relegationis 
quam deportationis. .. Item, licet bonorum publicatio sit de natura deportationis... tarn en 
expressa publicatio, quae fuit in casu proposito, etiam relegationem quandoque sequitur, ut... 
et [D.48.22.7,4]. .. cum relegatione autem habet bannum multa similia quia utrunque potest 
fieri in praesentem et absentem. Item, ad certum locum non astringunt... Item talis 
assimilatio et interpretatio mitigationem poenae inducit, ergo amplectenda .. . nam relegatus 
habet testamenti factionem in significatione activa et passiva, ut [D.48.22.7,3] et 
[D.28.1.8,3].' 
243 
those goods that have been expressly taken away from him. 254 In the context of 
Quidam inculpatus, this means that the bannitus lawfully inherited the estate and was 
able to pass it on to his own heir. Furthermore, the ban's confiscation provision only 
referred to the goods owned by the bannitus at the time of the sentence, but does not 
extend to his inheritance, which he received afterwards. Ironically, the key analogy is 
deportation, that is, the principle that property acquired by the deportatus after his 
sentence does not fall to the fisc during the period of deportation. The cases are 
considerably different, but the analogical point is clear.25 5 
Finally, Nicolaus comes to the question whether the ban's confiscation provision 
extends to goods located beyond the territory of the city.256 He uses this opportunity 
to establish a general rule about the territorial extent of the ban as a whole, both in 
respect of the person and property of the bannitus. His rule rests on the following 
distinction. If the judge who passes the sentence derives his jurisdiction from, and 
proceeds according to, the ius commune, then the effects of the sentence extend 
throughout the empire, and the banished person's property is confiscated by the 
authorities in whose territory it is located. In the old days, such proceeds used to go 
to the imperial fisc, but now the Italian cities claim them as they have acquired these 
fiscal rights. If, however, the judge either derives his jurisdiction from, or proceeds 
according to, local municipal law, then the sentence does not extend outside the area 
in which that law applies. Given that the ban is a creation of statute, its effects, 
whether personal or real, cannot reach beyond the territory of the banishing city. This 
means that the banished person ' s heir in Quidam inculpatus is free to inherit all of 
the deceased's goods, no matter where they are located.257 Nicolaus provides a long 
254 Ibid., fol. 33rb: 'In casu ergo propositae quaestionis, sive dicamus bannum fuisse poenam 
delicti sive contumaciae, dicemus hunc bannitum testabilem fuisse et potuisse haeredem 
institui, et omnia retinuisse quae sibi expresse adempta non fuerunt, ut [D.48.22.4 ], 
[D.28. 1.8,3]. ' 
255 Ibid: ' Super articulo utrum bannum extendatur ad bona postea quaesita, die quod non, ut 
praedixi per [D.1 7.1.22,5]. ' 
256 f C . Cavalca, /l bando, p. 202. 
257 Albericus de Rosate to C.6.24.1, fol. 33rb: ' An autem bona alterius territorii, die aut iudex 
qui publicavit habebat iurisdictionem a iure communi et secundum ordinem iuris communis 
poenam imposuit, aut utrunque vel alterum procedit a iure municipali. Primo casu, extenditur 
sententia ad bona ubicunque posita, non quod iudex ille possit occupare extra suam 
iurisdictionem, sed officiales illius territorii in quo sunt ista bona ... Si vero publicatio facta 
est auctoritate legis municipalis dantis iurisdictionem vel statuentis poenam, vel utrunque, 
tune publicatio non extenditur ad bona non subiecta iurisdictioni illius iudicis .. . et hoe verum 
quantum ad bona ... De poena autem personali ... servanda est praedicta distinctio utrum a 
lege communi vel municipali processit iurisdictio et poenae impositio. ex quibus patet quod, 
cum in quaestione proposita fuerit publicatio iure municipali, non extenditur ad bona alterius 
244 
and involved list of reasons for this argument, but the central idea seems to be that 
the force of statute simply 'does not extend to those who are not subject to it; 
therefore neither does the sentence ' .258 This . is common sense: why should, and 
indeed how could, the statutes and customs of one Italian city have the force of law 
in territory that it does not control? The ius commune, on the other hand, is valid in 
all the lands of the empire and even a local judge can activate it, so to speak, under 
the circumstances described. 
We should note at this point that the fourteenth-century commentators shared 
Nicolaus ' view on the territorial extent of the communal ban, which is one of the 
reasons why they quoted his work extensively. Bartolus is the only jurist who re-
examined the issue of property confiscation at length, but his conclusion only 
constitutes a minor variation of Nicolaus' opinion.259 Bartolus affirms the principle 
that confiscation on the basis of local medieval law only extends to the territory 
subject to that law, but that confiscation on the basis of the ius commune takes effect 
everywhere, and that in this latter case the goods are seized by the civitas in whose 
territory these are located. For Bartolus, however, the test to decide from which law 
the confiscation derives does not take into account the origin of the judge's 
jurisdiction, as is the case in Nicolaus' scheme. According to Bartolus, the. only 
relevant factor is whether the judge proceeds according to the provisions of local law 
or those of the ius commune.260 Bartolus further explains that if smaller political 
entities have the same overlord, such that they are subject to that law of the superior 
by virtue of which confiscation occurs, then the latter is valid in their territories as 
well.261 This basic idea is reflected in Cinus de Pistoia' s argument that the legal 
territorii. et sic concluditur haeredem huius banniti debere habere bona ... et in alia civitate 
posita.' Shortly before, fol. 32vb, Albericus explained: ' constat autem quod civitates ltaliae 
sibi vendicare iura fiscalia, et iurisdictionem divisam esse per civitates, sicut erat olim per 
provincias.' 
258 Ibid., fol. 32vb: 'Statutum autem non extenditur ad non subditos, ergo nee sententia ... ' 
259 See also Baldus' brief remarks at C.1.1(4).1, C.6.24.1, and D.28.6.10,6, and Jacobus 
Butrigarius to auth. item nulla, post C.1.3(6).2. 
260 Bartolus to C.1.1(4).1, fol. 5va: 'Alii dicunt quod Nie. Mata. in quadam sua disputatione 
dicit. [ ... ] Hoe praemisso, in quaestione praecedenti sic dico. [ ... ] Secundo casu, quando 
iurisdictiones sunt distinctae et bursae fiscales distinctae. Aut publicatio non fit de iure 
communi, et tune non extenditur ad alia bona, etiam alibi sita ... Aut fit de iure communi, et 
tune ad omnia bona etiam alibi posita extenditur. tamen quaelibet [civitas] habebit bona in 
suo territorio sita ... Nee placet ut distinguas an iudex habeat iurisdictionem a lege communi 
vel a lege municipali, quia non curo unde iurisdictionem habeat, sed. hoe solum, an 
pronuncietur super eo quod de iure communi est dispositum, an vero de novo disponat.. .' 
261 Ibid. : 'Aut iurisdictiones sunt separatae sed bursa fiscalis est una in effectu in utroque 
loco ... publicatio fit secundum constitutiones vel leges speciales, et... illae leges speciales 
245 
effects of a ban extend to new territory acquired by the banishing city if this new 
territory is now subject to that city's law.262 In this light, one could also understand 
Albericus de Rosate's report of the legislation of the Milanese Visconti lords, who 
attempted to make the ban of each city within their dominion effectual throughout all 
their lands.263 When two independent cities were merely allied, however, the jurists 
did not generally think that this automatically caused the ban of one to possess the 
force of law in the territory of the other; Bartolus noted that this rule was commonly 
observed in legal practice.264 
An issue related to territory is that of forum - court liability: the rules about where 
the bannitus can be accused of crimes and sued in civil matters.265 As we saw in 
chapter 2, forum arises by reason of origin, domicile, and offence (ratione originis, 
domicilii, delicti). 266 Communis opinio, which was strongly influenced by Dinus de 
Mugello, was that a bannitus could potentially be taken to court in all three 
depending on his situation.267 If he was banished from his city of origin he was still 
sunt in utroque loco ... Exemplum: plures iudices sunt deputati ab uno Rege per diversa 
territoria regni . unus publicat secundum constitutionem regalem, et tune omnia bona quae 
sunt in regno essent publicata ... ' 
262 Cinus de Pistoia to C.5.12.8, fol. 305va: 'Item, facit ad quaestionem de facto . Statuto 
civitatis Paduae continetur quod banniti de maleficio in civitate vel eius districtu habitare 
non possint. Quidam banniti habitant post istud statutum Lendenariae, quae non erat de 
districtu Paduano tempore conditi statuti, sed postea est adiecta dicta terra districtui Paduano. 
quaeritur an illi banniti ibi possint habitare impune ... ? Et dixit Ricar. Mai. quod non ... Et per 
praedicta dicimus quod Jex et consuetudo civitatis vel regni porrigitur ad loca sive territoria 
sibi postea adiuncta ... argumentum contra ... videtur distingui an ita sint adiuncti ut sub una 
consuetudine vel legibus sint, an non ... ' The same comment is printed in Jacobus de Arena, 
Commentarii, where the final distinction is attributed to Rainerius de Forlivio. Ricardus 
Malumbra's quaestio survives in Rome, Biblioteca Apostolica Vaticana, MS. Chigi, E.VIII. 
245, fol. lOlrb-cf. Bellomo, Quaestiones, p. 271 (no. 133), where we can see that only the 
first half of the text is actually Ricardus' - everything from 'et per praedicta' seems to be a 
later addition. 
263 Albericus de Rosate to D.49.15.7, fol. 221vb: 'Et idem posset dici de pluribus civitatibus 
quae subsunt eidem domino vel regno. et de hoe erat statutum in civitatibus subiectis 
dominio dominorum Mediolanensum, quod bannitus in una intelligeretur bannitus in 
omnibus.' Cf. Canning, Baldus de Ubaldis, p. 128. 
264 Bartolus to D.49.15.7, fol. 227ra: 'Ultimo nota quod dicunt quidam, quod hie est casus 
quod, si aliquae civitates foederatae sunt, quod exbannitus in una sit exbannitus in alia, quod 
non placet et usus non approbat.' Cf. Rainerius de Forlivio and Albericus de Rosate to 
D.49.15.7; Ghisalberti, 'La condanna' , p. 44; Canning, Baldus de Ubaldis, p. 128. 
265 Cf. Ghisalberti, 'La condanna', p. 66; Cavalca, Il bando, pp. 228-30. 
266 Ch. 2, p. 49. 
267 Besides the references below, see also: Guilelmus Duranti, Speculum iudiciale, de 
accusatione, § 6 sequitur, v. quid si bannitus I de competentis iudicis, § 1 generaliter, v. sed 
246 
liable there by reason of that origin. As we have seen, origin was an immutable fact 
of nature resulting from one's birth; therefore, it was retained despite the fact that 
domicile and citizenship were lost through the ban. 268 Of course, the bannitus was 
able to establish a new domicile in another city and thus also become liable there.269 
As Bartolus explains very clearly, the city (or cities) of origin and domicile have 
jurisdiction no matter where the offence was committed.27° Finally, the bannitus was 
also liable in the place in which he had committed an offence.27 1 These rules meant 
that a defendant could potentially be placed under multiple bans of the same city for 
different offences, 272 and, as Baldus highlights, he could even be banished on 
account of the same offence in more than one jurisdiction; he adds, however, that if 
such a bannitus is caught and subject to subsequent punishment, he cannot be 
. h d . l h 273 pums e agarn e sew ere. 
pone (with Johannes Andreae's additiones); Albertus Gandinus, Tractatus de male.ficiis, rub. 
De bannitis, q. 8; Bartolus to D.48.5.20(19). 
268 Albericus de Rosate, Quaestiones statutorum IV, q. 26, fol. 70ra: 'Item quaeritur. .. an 
bannitus de uno maleficio possit de alio maleficio accusari et banniri. Et videtur quod non, 
quia per primum bannum perdidit civitatem et domicilium, quod ibi habebat... Solvendo, 
dicit, secundum Dyn., quod is per primum bannum perdidit domicilium quod in ilia civitate 
habebat ratione incolatus, tune non potest ibi iterum ... banniri. Si autem perdidit domicilium 
quod habebat ibi ratione originis, tune poteri iterum ... banniri.' 
269 Ibid., q. 80, fol. 81 ra: 'Quaero an banniti de maleficio reputentur vagabundi, ut sic ubique 
possint convenire ... Communiter doctores tenent quod ex solo banno non reputentur 
vagabundi .. . cum domicilium habere possint in aliis civitatibus, unde banniti non sunt...' 
270 Bartol us to C.1.1 ( 4).1, fol. 5ra: 'Certum est quod de iure comm uni potest delinquens 
runiri in loco domiciiii vel originis de delicto alibi commisso.' 
71 Albericus de Rosate, Quaestiones statutorum IV, q. 26, fol. 70b: 'Quod dictum est supra 
secundum Dyn. de amissione domicilii, incolatus, et originis, puto intelligendum quando 
accusaretur ratione domicilii tantum. si enim accusaretur quia ibi deliquisset, tune ratione 
delicti ibi sortiretur forum, etiam si ibi domicilium non haberet...' 
272 E.g. ibid., q. 98, fol. 85ra: 'aliquis, qui bannitus erat pluribus bannis, praesentavit 
bannitum, nunquid eximi debeat de omnibus bannis, an de uno tantum?' 
273 Baldus to C.3.15.1, fol. 190rb: 'Tertio quaero, quid si unus est bannitus de uno istorum 
locorum, puta, de loco delicti, utrum ex eodem delicto possit banniri de loco originis et loco 
domicilii? Jaco. de Bel. hie disputando tenuit quod sic. si tamen in uno loco punitur, in alio 
accusari non poterit. .. Ex hoe apparet quod, licet aliquis sit bannitus, non tamen ex hoe 
dicitur punitus, cum non sit facta executio.' 
247 
v. The Early Fourteenth Century: Cinus de Pistoia, Jacobus Butrigarius, 
Johannes Andreae, and Albericus de Rosate 
By the opening decades of the fourteenth century, a strong communis opinio had 
formed on the systematic interpretation of the ban. The jurists of this period were 
essentially synthesisers, who presented the bannitus through the combined thought of 
Fredericus de Scalis, Jacobus de Arena, and Nicolaus de Matarellis. One can find 
particularly extensive reproduction of their writings in the works of Johannes 
Andreae and Albericus de Rosate; as we saw above, it is often only through their 
compendia that we have good knowledge of this material at all. 274 Cinus de Pistoia, 
too, in his widely circulating Lectura super Cadice, summarised the import of 
Fredericus de Scalis' quaestio Titius filiusfamilias together with Jacobus de Arena's 
commentary to the l. Ex facto in order to provide his readers with the fundamentals 
of banishment.275 On more than one occasion we are informed that the 'modem' 
doctors of law commonly refute the Accursian analogy bannitus-deportatus. 
Johannes Andreae and Albericus de Rosate also name Ricardus Malumbra and 
Oldradus de Ponte as adherents of the new position. 276 Albericus mentions that 
Nicolaus de Matarellis' quaestio Quidam inculpatus was especially popular amongst 
the modems ('continua et habita est mentio').277 
What we have learned about the new communis opmzo can be summarised as 
follows. The bannitus lost the iura of his local city while retaining the iura of Rome: 
he lost his local citizenship and its associated rights but remained a Roman citizen. 
Although the jurists never said so explicitly, the bannitus was, of course, still bound 
274 Johannes Andreae: additiones to Speculum iudiciale, de cessione actionis, § l sciendum, 
v. illud etiam quaeritur (Fredericus de Scalis/Jacobus de Arena), and de accusatione, § 6 
sequitur, v. quid si bannitus (Nicolaus de Matarellis). Albericus de Rosate: to C.6.24.1 and 
Quaestiones statutorum IV, q. 13 (Fredericus de Scalis); to D.36.1.18(17),5 (Jacobus de 
Arena); to C.6.24.1 and Quaestiones statutorum III, q. 2, and IV, qq. 1 and 13 (Nicolaus de 
Matarellis). 
275 Cinus de Pistoia to C.6.24.1 - briefly, though somewhat confusingly, discussed by 
Cavalca, Il bando, pp. 236-7. 
276 Albericus de Rosate to C.6.24.1, fol. 32rb: ' ... Ric[ardus Malumbra], Old[radus de 
Ponte]. .. et communiter Modernes Doctores reprobant dictam glossam, et dicunt bannitos 
non aequiparari deportatis .. .' Cf. Albericus de Rosate, Quaestiones statutorum IV, q. 13; 
Johannes Andreae, additio to Speculum iudiciale, de cessione actionis, § 1 sciendum, v. illud 
etiam quaeritur. 
277 Albericus de Rosate, Quaestiones statutorum IV, q. 13, fol. 67vb: 'Et quia ista sunt utilia, 
et continua et habita est mentio de quaestione domini Nie. de Mata., quae multum declarat 
praesentem materiam, ideo eius effectum subiico ... ' 
248 
by the iura of his former city in the sense that he was subject to the punitive 
sanctions of the ban, which had to be expressly contained in the city's statutory law 
and were restricted to its territory. Although,· in principle, the bannitus remained a 
Roman citizen, the penalties of banishment - such as the liability to be injured in 
person and property, and the inability to seek legal redress and to engage in certain 
legal acts (e.g. advantageous contracts) - took away a number of crucial rights that 
Roman citizens ordinarily enjoyed under the ius commune. In other words, by the 
first decades of the fourteenth century, the bannitus possessed all the rights of a 
Roman citizen in the territory from which he was banished unless statute, as 
interpreted and moderated by the jurists, expressly took these away. 
As we have seen, the jurists commonly mentioned testamentary powers and 
patriapotestas as rights of Roman citizenship of which the bannitus was not 
deprived. It would therefore be logical to assume that the bannitus also continued to 
possess the ability to hold guardianships;278 indeed, this was the position of Jacobus 
Butrigarius.279 Communis opinio, however, which goes back to Jacobus de Arena and 
which was shared by Bartolus and Baldus, was that the bannitus could not do so. 
This was because Roman law closely linked the ability to hold guardianships to the 
possession of one's estate. When the latter was confiscated (as was usually the case 
in banishment pro maleficio ), guardianships ceased, and the ward was granted the 
actio tutelae against the fisc, which became liable for claims resulting from the 
guardian's · former administration.280 Hence, the jurists viewed this issue primarily as 
a question related to the banished person's estate, and in these matters, as we saw in 
the cases of debt and tax, they continued to compare the bannitus to the 
d 281 eportatus. 
278 As Baldus comments at D.4.5.5,1, fol. 249: 'Si praedicta theorica Iacobi [de Arena] est 
vera, sequitur quod... duraret legitimus administrator vel tutor. Sed contra notatur in 
[D.27.3.9] per Ia[cobum de Arena].' Seen. 281 below. 
279 Jacobus Butrigarius to C.9.40.1, fol. 11 Ova: 'nunquam ergo bannitus perdit patriam 
potestatem vel tutelam vel curam.' 
280 D.27.3.9,6; C.9.49.5; C.9.12.2. Cf. Berger, Encyclopedic Dictionary, p. 747. 
28 1 Jacobus de Arena to D.27.3.9,5-6, fol. 1 OOra: 'argumentum quod si tutor alicuius 
banniatur et eius bona confiscentur, tutela sit finita et tutele agi possit...' Bartolus to 
D.26.7.33(32),7, fol. 59rb: 'Quaero an bannitus desinat esse tutor?[ ... ] Communiter videtur, 
aut per bannimenta bona sunt publicata, aut non, ut [Inst.1.12,2] et ibi notatum.' Cf. 
Albericus de Rosate, Quaestiones statutorum IV, q. 14, and Baldus to D.27.3.9,5-6. Cf. 
Ghisalberti, 'La condanna', p. 49. 
249 
II I 
II I 
Jacobus Butrigarius also debated an important quaestio on whether the bannitus 
became legally infamous.282 In Roman law, this infamia de iure ordinarily resulted 
from criminal convictions (amongst other causes) and triggered serious impediments 
in civil capacity, such as the inability to hold public office or to make criminal 
accusations. 283 Jacobus argues that because the ban is not a definitive sentence 
according to the rules of the ius commune, the bannitus cannot be considered legally 
infamous if statute does not expressly contain this penalty. 284 At best, such a 
bannitus suffers infamia de facto, which is less of a legal status than an 
acknowledgement of the fact that criminal proceedings, even if not properly 
concluded, impair someone 's reputation in practice. 285 As Albericus de Rosate 
reports, this was the communis opinio, which was later also followed by Bartolus and 
Baldus. 286 Jacobus Butrigarius does acknowledge, however, that a city could, in 
theory, legislate to have infamia result from the ban, though presumably, such 
282 Albericus de Rosate, Quaestiones statutorum IV, q. 72, fol. 77rb-vb: 'Inde sumpta 
occasione, quaero quaestionem disputatam per lac. Bu. aliquis fuit bannitus de crimine 
publico vel alias famoso, eo habito pro confesso ex forma statuti. superveniente abolitione 
generali, vel ex alia causa fuit de banno cancellatus, quaeritur an a testimonio repelletur.' 
Extant in Rome, Biblioteca Centrale Nazionale, MS. Vittorio Emanuele 1511, fols 26lra-2rb 
- cf. Bellomo, Quaestiones, p. 541 (no. 376). lrifamia is discussed by Ghisalberti, 'La 
condanna', p. 45, Mooney, 'The Legal Ban', p. 117, Vallerani, 'Diritti di cittadinanza', p . 8. 
283 For details, see Mommsen, Romisches Strafrecht, p. 986 ff. ('Btirgerliche 
Zurticksetzung'). 
284 Albericus de Rosate, Quaestiones statutorum IV, q. 72, fol. 77rb: ' ... ex tali banno non est 
infamis, nam in his quae contra ius commune introducuntur, non debet fieri extensio sed 
arctatio ... ergo non debet ad infamiam vel aliud non expressum extendi. [ ... ] Leges ergo 
dicentes damnatum de crimine publico vel alias famoso effici infamem ... de vero confesso 
vel convicto intelligi debent, non de interpretativo per bannum vel contumaciam. Nam ex 
lege municipali non, quia non est expressum, et iure communi non, quia de vero, non 
interpretativo intelligi de bent...' 
285 Ibid.: ' ... tales banniti apud bonos et graves infamantur infamia facti...' Cf. Johannes 
Andreae, additio to Speculum iudiciale, de cessione actionis, § 1 sciendum, v. illud etiam 
quaeritur, p. 33a: 'Ergo concluditur talem bannitum pro deportato non haberi, nee etiam 
infamia notari, licet eius opinio sit gravata, [C.9.40.3].' 
286 Albericus de Rosate, Quaestiones statutorum IV, q. 72, fol. 77rb: 'quod non efficiatur 
infamis tenuit ipse Ia., et idem tenuit Ia. de Bel... et bane etiam opinionem Dy. tenuit... et 
haec opinio communiter approbatur et servatur.' Cf. Jacobus de Arena to D.1.2.2,4; Bartolus 
to D.3.2.22 and Lucanae civitatis, p. 204b; Baldus to D.1.2.2,6, D.24.3.47; auth. Item 
quaecunque,post C.1.3(6).1, C.6.24.1, and C.9.40.3. Strangely, when Bartolus (at D.3.2.22) 
and Baldus (at D.24.3.47) report Jacobus' quaestio, they relate the same material, but say 
that Jacobus came down on the exact opposite side of the argument, i.e. that the bannitus 
does suffer infamia de iure. Albericus' interpretation seems more likely, however, since 
Jacobus himself argues twice in his commentaries (at C.9.40.2 and auth. item quaecunque) 
that the bannitus only becomes infamous de facto. 
250 
infamia would not extend beyond that city's territory. 287 As Baldus states in a 
memorable phrase, since banishment results from local statute or custom and not the 
ius commune, 'those banished from England are not infamous in our lands'. This 
c. k f,c:: . h . . 288 leads Baldus to conclude that banniti are 1ree to ta e up o 11ces m ot er c1t1es. 
Albericus de Rosate, on the other hand, is inconsistent on this point. In one comment 
he claims to have observed banniti taking up offices in foreign cities, but argues in 
another comment that even infamia de facto, which (unlike infamia de iure) follows 
banniti wherever they go, prevents them from doing so, 'according to the custom of 
all Italy'.289 
Most importantly, the jurists writing in the first half of the fourteenth century fleshed 
out Jacobus de Arena's argument that the bannitus loses the 'things' derived from 
municipal law and local citizenship. As Albericus de Rosate explains: 
The aforesaid doctors of law argue, however, that, if by virtue of a civic statute a 
bannitus is granted the inheritance of an entire estate or some part thereof, or some 
other right [aliud ius], such as the widower's entitlement to his wife's dowry or a part 
thereof, he is deprived of such rights [iuribus] by the ban, since a benefit [beneficium] 
of this type granted by statute can be taken away from him by another statute.290 
287 This is evident from the second part of his quaestio, where it is assumed that the bannitus 
became infamous, and the principal question is whether a city could lift this infamy through 
its cancellation of the ban - Albericus de Rosate, Quaestiones statutorum IV, q. 72, fol. 
77va-b: 'Ia. But., qui bane quaestionem disputavit, tenuit quod per cancellationem tollatur 
infamia si foret ex banno secuta... hie autem loquimur de ea quae oritur per le gem 
municipalem, quae potest per aliam legem municipalem tolli. Nam qui potest inducere, 
po test toll ere .. .' 
288 Baldus to D.1.2.2,6, fol. 17va: ' ... banniti de una patria quandoque honorantur in alia ... et 
sic banniti non sunt in fames, quia non sunt banniti de iure Romano, sed ... per statutum vel 
consuetudinem ... et sic banniti ex Anglia apud nos non sunt infames ... ' 
289 Albericus de Rosate to Sext.5.13.87, fol. 300rb: 'cum tales bannitos communiter, saltem 
facti, sequatur infamia apud bonos et graves, maxime quando ex publicis et magnis 
criminibus banniuntur ... de more totius Italiae non possint ad dignitates promoveri ... ' But in 
Quaestiones statutorum IV, q. I 00, fol. 85ra: 'Item, quaero an banniti nostri temporis possint 
eligi ad honores et officia in aliis civitatibus in quibus non sunt banniti ... Vidi servari quod in 
aliis civitatibus admittuntur...' NB. A commentary to the Liber sextus under Albericus' name 
is appended at the end of his Digest commentary in the 1585 Venice edition (repr. OIR 26). 
Lange and Kriechbaum, Die Kommentatoren, p. 680, note that the authorship of this work is 
not entirely clear. 
290 Albericus de Rosate to C.6.24.1, fol. 32rb: 'Dicunt tamen praedicti Doctores quod, si per 
statutum daretur haereditas tota vel pro parte huic bannito, vel aliquod aliud ius, puta dotis 
lucrandae in totum vel pro parte, per bannum a talibus iuribus privaretur, quia tale 
beneficium a statuto datum potest sibi auferri per aliud statutum. ' An analogous passage can 
251 
ill 
I I 
I 
In other words, the 'things' lost by the bannitus are rights (iura), which consist in the 
benefits (beneficia) that his former city's statutes confer upon its citizens.291 In the 
passage quoted, Albericus mentions two such rights of citizenship, both of which are 
prerogatives in the field of inheritance. The first refers to the type of provision 
mentioned in Fredericus de Scalis' quaestio Titius filiusfamilias: the male inheritor is 
granted a more advantageous position in relation to female kin than he holds in the 
ius commune. In a similar way, statutory law strengthened the right of the widower to 
acquire his deceased wife's dowry (ius dotis lucrandae), which prevented the latter 
reverting to her family. 292 Such statutory provisions seem to have been common. 
Johannes Andreae also mentions the ius dotis lucrandae. 293 Later, Bartolus and 
Baldus would give as examples the preferential treatment of male heirs as well as the 
right to make a testament with fewer witnesses than required by Roman law.294 The 
view of local citizenship in this context as a set of entitlements in the private sphere, 
specifically in the law of inheritance, bears a striking resemblance to the jurist' 
understanding of Roman citizenship. As we saw in the previous two chapters, the 
latter too was primarily conceived as a set of private-law entitlements, of which 
rights of inheritance formed a crucial part. 
Before moving on to Bartolus, I would like to emphasise that the hostis had not 
become a point of reference at all in the systematic discourse on banishment by the 
opening decades of the fourteenth century. This point is important because Cavalca 
mistakenly claimed that the hostis analogy became 'the key to a new interpretation of 
the ban' already in later thirteenth century.295 As we have seen, at best - and this is 
be found in Johannes Andreae, additio to Speculum iudiciale, de cessione actionis, § 1 
sciendum, v. illud etiam quaeritur. 
291 The word used by Albertus Gandinus, in an early expression of this idea, is praemium. 
Tractatus de maleficiis, rub. De bannitis, q. 3, p. 416: ' ... huiusmodi exbanniti exceptati sunt 
ab omnium civium premiis, et cives esse desinunt...' 
292 Cf. Kirshner, 'Wives' Claims', p. 261. 
293 See n. 290 above. 
294 Bartol us, Lucanae civitatis - see p. 25 8 n. 316 below. Cf B Id t D 1 18 6 8 d . a us o . . . , an 
D.28.2.29,5. 
29s I Cava ea, JI bando, esp. pp. 88-93, at pp. 89-90: 'Neppure l'altro passo de! Digesto 
relativo ai capite minuti aveva sollecitato nei Glossatori diretti richiami al bando; si erano 
limitati ad osservare che ii transfaga perdeva ii diritto di cittadinanza, ed era considerato 
hostis dal senato. Piu tardi pen'>, nella seconda meta de! '200, questo passo diverra la chiave 
per una nuova interpretazione de! bando e sara poi approfondito dai Commentatori nel XIV 
secolo ... La dottrina inizio a considerare ii bando, non piu sotto ii profilo dell 'esilio-
intemamento ( deportazione ), ma dell' aqua et ignis interdictio, verso la fine del '200.' This 
comment can only be understood by bearing in mind Cavalca's erroneous conflation 
(throughout his book) of the hostis declaration with aqua et igni interdictio. Although 
252 
the only real evidence that Cavalca presented - the jurists used the hostis 
casuistically in the solution of certain circumscribed legal problems.
296 Clearly, the 
raw idea that the bannitus could be compared .to the hostis - due to his disobedience 
to public authorities and the resulting lack of legal protection - was circulating in the 
law schools. No one before Bartolus, however, made the hostis the defining concept 
through which to understand the ban. His move constituted a substantial shift in the 
systematic discourse. In this regard, the civilian traditions on imperial and communal 
banishment developed hand-in-hand. In both cases, it was Bartolus who redirected 
the central focus away from the ordinary institutions of Roman criminal law 
(annotation, relegation, deportation) to the extraordinary figure of the hostis-
transfuga. 
vi. Bartolus 
Bartolus was the last medieval civilian to have a defining impact on the systematic 
part of the discourse on communal banishment. What we find in Baldus' 
commentaries is an edited report of the consensus of the early fourteenth century, 
adjusted to take into account his master's interpretation; Baldus' consilia are a 
testament to the application of the Bartolist system in legal practice.
297 As was 
remarked above, the best expression of Bartol us' thought on the communal ban can 
be found in his quaestio Lucanae civitatis. The Tractatus bannitorum - which seems 
to be merely an excerpt from Lucanae civitatis298 - and his commentaries, although 
they cover most of the same fundamental arguments, are considerably less 
outlawry was a central penalty associated with both, these were profoundly different legal 
institutions - cf. eh. 2, p. 41, n. 30. 
296 Perhaps the clearest indication prior to Bartolus that there might be a general equivalence 
between bannitus and hostis occurs in Albericus de Rosate's commentary to D.4.5.5,1, fol. 
263vb: 'Notabilis est iste § et glossae hie positae ad materiam bannitorum et transferentium 
se ad hostes .. .' But as we have seen, Albericus in no way develops that line of thought in his 
systematic material on the ban. 
297 Baldus to D.1.18.6,8; D.4.5.5,1 ; D.12.6.19; D.28.2.29,5; C.6.24.1; cons. I.284, I.401-2, 
11.55, IJ.145, III.9, III.96, III.263-4, IV.13, IV.142, V.54. A good, if very brief, exposition of 
Baldus' thought is Canning, Baldus de Ubaldis, pp. 127-8. As Canning remarks (p. 128): 
'There is nothing new in Baldus' position on banishment: it accords with the ideas of, for 
instance, Jacobus Butrigarius and Bartolus.' In agreement: Cavalca, JI bando, p. 98. 
298 See eh. 3, p. 165, n. 254 for details on the quaestio. 
253 
exhaustive. 299 Since the ex1stmg historiography relies on the treatise and the 
commentaries, 300 what follows is the first full exposition of Bartol us' thought. 
The actual terms of Lucanae civitatis concern the following problem. One statute in 
Lucca imposes capital punishment on anybody who breaks a peace agreement by 
killing someone. Another statute permits those who have been banished for murder 
to be killed with legal impunity. Titius, who had been banished for murder, 
concluded a contractual agreement (stipulatio )301 with his enemy Sempronius, a 
private citizen not under the ban, that they would not harm each other under threat of 
a stiff financial penalty. Sempronius then killed Titius anyway. The question is 
whether Sempronius is subject to the respective statutory and contractual 
penalties. 302 
After extensive remarks pro et contra, Bartolus provides a comprehensive 
framework with which to understand the ban before moving on to the solution of this 
specific question. He notes that 'all modems' agree with Fredericus de Scalis' 
opinion that the bannitus is not analogous to the deportatus, listing some of the 
299 Bartolus to D.4.5.5,1 and C.6.24. l. Note that some ascribe to Bartolus also a Tractatus 
exbannitorum!de exbannitis, which is not to be confused with his Tractatus bannitorumide 
bannitis. Ghisalberti (p. 22) accepts its authenticity because it is listed as one of Bartolus' 
works by van de Kamp, 'Bartolo da Sassoferrato', p . 54. Lange and Kriechbaum, Die 
Kommentatoren, pp. 726-7, also regard it as authentic. This is based on the view of Calasso, 
'Bartolo da Sassoferrato', DBI, vol. 6, p. 655, which in turn rests on the opinion of 
Diplovataccius, who lists this treatise under Bartolus' name, and claims in an additio to it 
(CQT (Basel), p. 357a / CQT (Venice), fol. 130vb) that Bartolus himself referred to this 
treatise in his Tractus super constitutione Qui sin! rebelles, 'in ver. quid si aliquis ' (no. 7). 
However, neither in the Basel (p. 289b) nor in the Venice edition (fol. 105ra) of the CQT can 
this reference be found. I regard the treatise to be a forgery because the relegatus is given as 
the best analogy for the bannitus, which is entirely contradictory to Bartolus' thought. 
Feenstra, 'Bartole dans le Pays-Bas' , pp. 213-4, already raised this point. His view is 
accepted by Mooney 'The Legal Ban', p. 74 (n. 49). 
300 With the exception of Woolf, Barto/us of Sassoferrato, p. 200. Cf. Ghisalberti, 'La 
condanna', pp. 23-4, 40-44, 48, 52-4; Mooney, 'The Legal Ban', pp. xi, 72-4, 118, 161, 165; 
Cavalca, Il bando, pp. 94-100, 237-8. 
301 St· l · f . zpu aflo was a common type o contract m Roman law - cf. Berger, Encyclopedic 
Dictionary, pp. 716-17. 
302 Bartolus, Lucanae civitatis, p. 202a: 'Lucanae civitatis statuto cavetur quod frangens 
pacem occidendo capite puniatur. Alio statuto cavetur quod exbannitus pro homicidio possit 
impune occidi. Accidit quod Titius, qui erat pro homicidio bannitus ... facit pacem cum 
Sempronio suo inimico non bannito, et promiserunt se non offendere sub poena statuti et 
poena mille lib. hincinde stipulatione promissa. demum dictus Sempronius interfecit dictum 
Titium exbannitum. Quaeritur an incidat in poenam statuti? et an dicta poena pecuniaria sit 
commissa an non?' 
254 
reasons with which we are familiar and a number of other technical dissimilarities.303 
For similar reasons, however, Bartolus rejects the comparison to the relegatus. Most 
importantly, unlike the ban, relegation is a definitive sentence, which, although it can 
be passed on an absent defendant, only follows a full judicial examination of the 
defendant's guilt. The ban, on the other hand, is given on the basis of contumacy 
alone without any examination of the principal matter. And unlike the bannitus, the 
relegatus does not lose the protection of the law.304 Nor, Bartolus goes on, should the 
bannitus be compared to the excommunicate. Although banishment and 
excommunication have obvious parallels - both are measures against contumacious 
defendants, and just as the excommunicate 'is separated from the communion of the 
faithful', so the bannitus 'is separated from the communion of citizens' - the 
dissimilarities are such that one cannot make a general comparison between the 
two.305 Firstly, it is common practice in Italy that the bannitus, after he has missed 
the final deadline for appearing at court, is considered pro confesso: he can no longer 
be heard concerning the alleged crime and is thus unable to establish his innocence. 
Excommunication, on the other hand, never forecloses an examination of the alleged 
original offence; the excommunicate always remains in a position to effect the 
cancellation of his sentence through his obedience to the ecclesiastical judge and by 
providing guarantees that he will obey the latter's resolution of the principal 
matter. 306 Secondly, a sentence of excommunication, even if passed by a local 
ecclesiastical judge, is recognised by the church everywhere, whereas the effects of 
303 Ibid., fol. 204b: 'Contrarium credo verum, scilicet, quod non aequiparetur deportato. [ ... ] 
et sic fuit determinatum Bononiae in publica disputatio per dominum Fredericum de Scalis, 
et in hac sententia sunt communiter omnes Modemes.' 
304 Ibid., pp. 204b-5a: 'Tertio est videndum utrum exbannitus aequiparetur relegato? [ ... ] 
Contrarium credo verum, nam exbannitur aliquis propter solam contumaciam, nulla de 
delicto praecedente cognitione, secundum statuta et consuetudines ltaliae. sed per sententiam 
relegationis, licet posset fieri in absentem ... tamen debet de crimine cognosci ... Praeterea, 
non reperitur quod relegatus possit impune occidi, sed exbannitus sic ... ' 
305 Bartolus, Lucanae civitatis, p. 203a: ' ... propter quod bannum quis separatur a 
communione civiu~, ita excommunicatio, propter quam quis separatur a communione 
fidelium .. .' Ibid., p. 205a: 'Contrarium credo, scilicet, quod non aequiparatur 
excommunicato in omnibus, ita quod de uno ad alium arguatur. [ ... ] Non obstat [LF.2.28 c. 
Domino guerram], quia fateor quod quantum ad illud aequiparatur propter contumaciam. in 
multis aliis posset reperiri concordantia et discordantia, quod Canonistis relinquo.' Baldus, at 
C.6.24.1, lists a number of further dissimilarities between the ban and excommunication. 
306 Bartolus, Lucanae civitatis, p. 205a: 'nam isti exbanniti, statim lapso termino banni, 
habentur pro confessis de maleficio, nee possunt innocentiam suam purgare, secundum 
statuta et consuetudines Italiae. Sed excommunicatus, praestita satisdatione de parendo 
mandato iudicis, absolvitur, ut [X. l.40.1], [X.5.39.29/48].' This accords with Baldus' 
interpretation of excommunication at auth. Item quaecunque, post C.1.3(6).13 - cf. eh. 3, p. 
114. 
255 
I I 
the ban are restricted to the territory of the banishing city. And once again, like the 
relegatus, the excommunicate cannot be killed with legal impunity.307 As we saw 
earlier in this chapter, Bartolus concludes that the bannitus should rather be 
compared to the hostis and the transfuga: it was the element of rebellion against 
public authority and the banished person's outlaw status that convinced Bartol us of 
the efficacy of this analogy. 
Bartolus then proceeds to explore the legal status of the bannitus. Given that the 
hostis-transfuga is his guiding analogy, he summarises the latter's position in Roman 
law in a passage that we have discussed at length in the previous chapter: the hostis-
transfuga loses the iura of Rome insofar as they pertain to his advantage 
( commodum ), but loses them insofar as they pertain to his disadvantage 
(incommodum). 308 According to Bartolus, this same distinction should apply as a rule 
when local citizenship is lost: 
And I say the following concerning such a banished person. Given that this man is a 
transfuga and hostis of his city from which he is banished ... I say that through his 
banishment he loses that citizenship [civitas] of his and all the iura proper to his city 
from which he is banished insofar as they are favourable to him; but insofar as they are 
unfavourable to him, he remains bound to these iura as much as [he remains bound] to 
that city and its men, just as before [his banishment]. He does not, however, lose the 
iura of the law of nations and the common iura of the city of Rome, that is, the Roman 
empire. For he is not a deserter of the whole empire, but only of his city ... 309 
Bartolus illustrates this point with references to various privileged groups in the 
Roman law of persons (e.g. clergy, soldiers, decurions). We should understand the 
city as one such privileged, corporate body of individuals (universitas) from which 
307 
Bartolus, Lucanae civitatis, p. 205a: 'Praeterea, sententia excommunicationis a suo 
iudicie lata ligat apud omnes, ut [X.5.39.59]. Sed exbannimentum non facit quern videri pro 
exbannito nisi in ill a civitate unde exbannitur. .. Praeterea, non reperitur quod 
excommunicatus possit impune occidi...' 
308 h 3 C . , pp. 165-71. 
309 
Bartolus, Lucanae civitatis, p. 205b: 'et ita dico de tali exbannito. nam cum ipse sit 
transfuga et hostis illius civitatis, unde exbannitus est... dico quod per exbannimentum perdit 
illam suam civitatem et omnia iura propria civitatis illius, und exbannitus est, prout faciunt 
pro se; sed prout faciunt contra se, illis iuribus remanet ligatus, quantum ad illam civitatem 
et eius homines, sicut primo. iura vero gentium et iura communia civitatis Romanae, seu 
Romani imperii, non perdit. nam non est transfuga totius Imperii, sed illius civitatis 
tantum ... ' Mooney, 'The Legal Ban', p. 118, is therefore incorrect in arguing that Bartolus 
only applied the commodum!incommodum distinction to the loss of Roman citizenship. This 
distinction, in respect of the communal bannitus, is mentioned in passing by Kirshner, 
'Mulier alibi nupta ' , p. 163. 
256 
--
the bannitus is removed: hence, he loses the 'rights and privileges' of members of 
that corporation but remains under the disposition of the ius commune.310 In his 
commentaries, Bartolus reminds us that this argument is not restricted to Italian 
cities, but applies equally well to kingdoms within the empire. 311 
For further support, Bartolus quotes Jacobus de Arena's comment to the 1. Exfacto. 
Clearly, he viewed his argument as a development of the basic idea expressed 
there.312 And it is true that Bartolus' scheme is a direct continuation of the early-
fourteenth-century consensus that the bannitus retains Roman citizenship while 
losing the iura of his local city. What Bartolus contributes is conceptual precision by 
explaining explicitly what had remained implicit in the discourse up to that point: the 
bannitus lost citizenship and its associated benefits; however, just as the hostis of 
Rome, who has lost Roman citizenship, is still bound by the Roman legal system -
i.e. the punitive measures of its criminal law - in the same way, the bannitus remains 
bound to the legal system (the iura) of his former city. This makes perfect sense. For 
how else could he be subject to the statutory sanctions of banishment within its 
territory? The hostis and bannitus are peregrini in mala parte, not peaceful 
i': • 313 1ore1gners. 
There is, however, one important conceptual difference between these two cases. As 
we saw in chapter 3, Bartolus' commodum/incommodum distinction served as an apt 
characterisation of the loss of Roman citizenship because most of its associated 
rights were entitlements, and because the hostis and deportatus primarily 
experienced Roman law as a set of punitive measures. Nevertheless, the distinction 
was not flawless since Roman citizenship also contained a small number of 
unfavourable elements (e.g. guardianships), which were lost as well. Similarly, the 
Roman legal system was not only disadvantageous to those who had suffered capitis 
deminutio media; the deportatus, for example, could sue in Roman courts for ius 
310 Bartolus, Lucana<: civitatis, p. 205b: 'Praeterea, hoe esse verum probatur quia, si aliquis 
transfugit vel est reprobatus ab uno collegio, iura propria illius collegii perdit, et remanet in 
dispositione iuris communis, ut [D.27.1.6], [C.1.3(6).51], [C.9.41.8], [C.10.71.2], 
[D.50.2.14]. Ita, cum iste exbannitus sit ex delicto remotus ab universitate illius civitatis, 
perdit iura et privilegia ipsius. pro hoe facit [C.10.38.1].' 
311 Bartolus to D.4.5 .5,1 fol. 14lrb: 'Sed ille qui est bannitus alicuius civitatis vel regni, non 
perdit ea quae sunt iuris civilis Romanorum, sed ea quae sunt propria iura illius civitatis vel 
regni ... ' 
312 Bartolus, Lucanae civitatis, p. 205b: 'ad istam opinionem confirmandaminduco in testem 
lac. de Aret... qui hoe tenet expresse [D.36.1 .18(17),5] , ubi, inter caetera, his utitur verbis ... ' 
313 Costa, Civitas, pp. 45-6, nicely brings out the contrast between the bannitus, the foreigner 
who is a hostile ex-citizen, and the original foreigner, who is friendly and peaceful. 
257 
I I 
I 
111 
l 
gentium contracts. In other words, the distinction was a limited and purely 
descriptive rule of thumb. 314 In the context oflocal citizenship, on the other hand, the 
distinction appears to be strictly prescriptive: every statute favourable to the bannitus 
no longer applies to him, whereas every unfavourable statute continues to do so. 
Coming back to the original question of Lucanae civitatis, this is the reason why 
Bartolus argues that the banished person's killer, Sempronius, is not liable for the 
statutory penalty, as otherwise the bannitus, Titius, would benefit from the statute 
. h. h h 315 pums mg t ose w o break peace agreements. For the same reason, banniti are 
unable to make testaments, be instituted as heirs, or succeed through intestacy, if the 
operation of any of these is dependent on the provisions of statutory law. 316 
According to Bartolus' logic, the bannitus, even though no longer a citizen, would 
continue to be liable for the duties of civic membership, unlike the deportatus and 
hostis, who lost the burdens of citizenship along with its advantages. Unfortunately, 
we have no real evidence with which to explore this counterintuitive point. The only 
major civic duty that was discussed in the context of banishment is the liability to 
muneralonera.317 But as we have seen - contrary to Bartolus' logic in Lucanae 
civitatis - the jurists, including Bartolus himself, argued that the bannitus is no 
314 h C . 3, pp. 165-71. 
315 Ibid.: 'Ad solutionem igitur nostrae quaestionis descendo, et dico quod occisor non incidit 
in poenam statuti. Nam cum statutum sit proprium ius civitatis ... illud ius perdit exbannitus 
per bannum, cum ad ipsius exbanniti commodum pertinet. et ideo occisor non tenetur de 
poena iuris.' 
316 Ibid., p. 206a-b: 'Tu distingue: aut fecit testamentum secundum ius commune, et potest, 
aut fecit testamentum secundum ius proprium, sive civitatis, puta quia erat statutum quod 
possit fieri cum minori numero testium, et tune tale testamentum non valet, quia tale ius 
perdit... Item, consuevit dubitari utrum exbannitus posset haeres institui. [ ... ] Tu distingue: 
aut instituitur in testamento quod de iure communi sortitur effectum, et potest, aut de iure 
proprio civitatis, unde est exbannitus, et tune non potest, quia illius iuris non est capax ... 
Item posset dubitari utrum exbannitus posset succedere ab intestato. et videtur dicendum 
quod aut loquimur de successione quae venit de iure communi, et succedit, aut de 
successione ab intestato quae venit ex forma statuti, ut si per statutum diceretur quod 
masculus praeferatur mulieri proximiori in gradu, et tune non succedit, quia huius iuris est 
omnino incapax .. .' Baldus, at C.6.24.1, fol. 7lra, very briefly voices one disagreement with 
his master's scheme. While Baldus fully accepts that the bannitus himself cannot inherit on 
the basis of statute, he argues that the banished person's relatives can nevertheless claim an 
inheritance from the bannitus on that basis. Baldus does not specify whether this refers only 
to rights ab intestato or also to the banished persons' testaments. Either way, this position is 
considerably more favourable to banniti and especially to their families. 
317 Questions of forum were not resolved on the level of statutory citizenship but according 
to the rules of Roman law, using the concepts of origin, domicile, and ratio delicti - see pp. 
246-7 above. 
258 
longer obligated in this way by example of the rules of confiscation in the law of 
deportation. 318 Interestingly, at one point, Bartolus briefly mentions that banniti 
continue to be liable for reprisals because they are still bound by the unfavourable 
iura of their city.319 The cities of Italy commonly granted their citizens the right to 
undertake reprisals against the whole citizen body of another city, after other legal 
remedies had been exhausted, in order to satisfy claims that the other city's courts 
had refused to recognise.320 In this way, one citizen and his property became liable to 
be seized for the unpaid debt of his fellow citizen. 321 Unfortunately, Bartolus 
nowhere elaborates this argument, not even in his Tractatus represaliarum. 322 
Baldus, for his part, briefly comments that banniti are not subject to reprisals granted 
against the citizens of their former cities precisely because they are no longer 
, 323 
czves. 
In Lucanae civitatis, Bartolus, having covered Sempronius' liability to the statutory 
penalty, goes on to explain why he is not liable for the contractual penalty either. The 
contract is not invalid because the bannitus was incapable of making a stipulatio. 
This type of contract forms part of the Roman civil law, and in principle the 
bannitus, as a Roman citizen, has access to all ius commune contracts. 324 It is rather 
that this specific stipulatio is invalid: firstly, peace agreements with banniti 
318 See pp. 229-30 above. 
319 B~rtolus to D.4.5.7,3, fol. 141 vb: 'Videtur ergo quod, licet bannitus perdat iura civitatis 
suae quantum ad suum commodum, non tamen hoe ad suum incommodum, quin obligetur ex 
his quae postea facit. et sic facit ad quaestionem an bannitus possit capi pro represaliis, et de 
hoe dixi plene in tractatu represaliarum.' 
32° Cf. Bartolus, Tractatus represaliarum, CQT (Basel), pp. 327-40. The relationship of 
citizenship and reprisals in the thought of Bartolus and Baldus is discussed by Kirshner, 
'Civitas sibifaciat civem', pp. 709-10, and Canning, Baldus de Ubaldis, p. 183-4. 
32 1 Bartolus, Tractatus represaliarum, CQT (Basel), p. 327b: ' ... quando conceditur licentia 
capiendi personam vel res alterius pro altero. et ista sunt quae in vulgari nostro vocantur 
represaliae ... ' 
322 That includes part 7 (' ... si represaliae simpliciter concedantur contra homines et personas 
alicuius terrae, contra.quas personas possint exerceri?'), ibid., pp. 336-8, where Bartolus says 
many other interesting things about citizenship that have yet to the subject of a full 
examination. 
323 Baldus to D.4.5.5, 1, fol. 249vb: ' ... bannitus desinat esse civis suae civitatis et sic 
represaliae concesse contra cives non ligant bannitos. ' He repeats this position at C.4.60.1. 
324 Bartolus, Lucanae civitatis, pp. 205b-6a: 'de poena vero conventionali est magis dubium. 
nam, ut proponitur, fuit per stipulationem promissa, quae stipulatio est de iure civili publico 
Romanorum inducta... unde videtur quod talis bannitus sit capax talis obligationis. dico 
tamen quod poenalis stipulatio non est commissa. hoe tamen non accidit ex eo quod talis 
exbannitus non sit capax talis obligationis. nam capax est omnis obligationis quae descendit 
. . ' a mre commum ... 
259 
I 
I i 
11 
contravene the statutes of the city, which, on account of public utility, require that 
the bannitus not be given any legal protection; secondly, in Roman law, it is lawful 
for a private person to break peace agreements with hostes.325 Bartolus, however, 
makes sure to explain that if statute, analogical arguments from Roman law, and 
considerations of public utility do not intervene, the bannitus can effectually enter 
every contractual arrangement according to the ius commune, such that an action 
would arise if the other party breached the contract.326 Naturally, the bannitus would 
be unable to bring the action in court while his ban lasted: in Baldus' words, he 
would 'acquire' a right but be unable to 'exercise' it.327 The bannitus is, however, 
able to cede rights to another, and if he does so without lucrative cause (through sale, 
for example), the exceptio banni will not hinder the cessionary in court; the bannitus 
himself would only be able to sue should his ban ever be cancelled.328 As a Roman 
citizen, Bartolus goes on, the bannitus also possesses all active and passive 
testamentary powers of the ius commune, including in the territory of the city from 
which he is banished.329 He agrees with Nicolaus de Matarellis that the city has no 
claim to a subsequent inheritance even if the sentence had initially included the 
confiscation of the banished person's goods.330 Baldus adds that the city could only 
325 Ibid., p. 206a: 'sed procedit defectus ex parte eius qui promittit et eius quod promittitur 
mixtim ... et istud, quod talis exbannitus possit occidi, est inductum pro quiete communi suae 
civitatis propter utilitatem publicam ... cum ergo tale pactum sit contra publicam utilitatem 
illius civitatis, non valet principalis nee poenalis stipulatio ... praeterea ... iste exbannitus est 
transfuga et hostis suae civitatis. Sed licitum est privato venire contra fidem datam hosti a 
se .. . ' This opinion goes back to Oldradus de Ponte's quaestio Titius fecit pacem: Bellomo, 
Quaestiones, pp. 209-10 (no. 419). Jacobus de Belviso 's quaestio on this topic is reproduced 
by Albericus de Rosate, Quaestiones statutorum IV, q. 28 (no incipit). See also: Jacobus 
Butrigarius to D.3.3.45; Bartolus to D.2.14.5 and D.4.3.1,3; Baldus to D.4.3 .1 , D.4.3 . l ,3, and 
C.7.48.1. 
326 Bartolus, Lucanae civitatis, p. 206a: 'Sed si aliquis promitteret alicui tali exbannito 
pecuniam ex causa donationis vel ex alia causa, quae non venerit contra publicam utilitatem, 
nee esset alias improbata de iure communi vel ex statuto, tune tali exbannito competeret 
obligatio et actio, cum, ut dixi, sit capax omnis obligationis iuris communis.' 327 Ibid.: 'Credo tamen quod, si ageret coram iudice illius civitatis unde exbannitus est, quod 
sibi denegatur iurisdictio, cum sit extra protectionem illius civitatis.' Baldus to C.6.24. l, fol. 
70vb: ' ... saepe ius quaeritur, tamen exerceri non potest...' 
328 Bartolus, Lucanae civitatis, p. 206a: 'Et si dices, quid prodest habere actionem et non 
posse agere? Respondeo, prodest si rebanniretur. Item prodest quia potest alteri vendere 
actionem praedictam vel cedere ex causa non lucrativa, et dicta cessionario non obstabit 
exceptio .. .' Cf. Baldus to D.4.6.40,1. 
329 Seep. 258, n. 316 above. Bartolus to C. l 0.1.4, fol. 4ra-b: 'tamen retinet ius succedendi de 
iure communi in civitate ilia ... ' 
330 Bartolus to D.1.6.7. Baldus agrees with his master in his commentary to the same law, 
albeit through a confusing comparison to deportation, which is inconsistent with the Bartolist 
principles on the banished person's testamentary powers that he otherwise espouses. We 
260 
confiscate a later inheritance if there were express statutory prov1s10ns to that 
effect.331 
Bartol us' depiction of the bannitus as a Roman citizen in Lucanae civitatis is a lucid 
synthesis of the communis opinio that had developed by his time. As we have seen, it 
was generally agreed that the bannitus possessed all the rights of a Roman citizen 
unless statute had expressly taken them away. Baldus went on to acknowledge this 
very unambiguously when he explained that 'in its territory, statute can deprive a 
bannitus not only of the benefit of statute, but also of the ius commune' .332 Since the 
ban ran counter to criminal procedure in Roman law, the jurists interpreted statute 
restrictively (stricti iuris/33 and, as Bartolus did in Lucanae civitatis, defined the 
extent to which it was allowed to operate through the principles of civilian 
jurisprudence - above all, through the authoritative analogies in Roman law (such as 
the hostis) and the concept of public utility, which was, of course, itself defined by 
the exemplars in the Corpus iuris. The reasonableness, and hence legality or 
otherwise, of each statutory penalty associated with the ban, from the liability to be 
killed with impunity to the right to present a defence, whether it was analysed in a 
systematic or casuistic manner, was established in this way. As our discussion of 
Fredericus de Scalis ' quaestio Titius filiusfamilias and the casuistic discourse on 
contract demonstrates, it is also clear that Bartolus followed his predecessors in 
permitting the bannitus to transfer rights and to sue for them after the cancellation of 
his ban. While it is therefore certainly the case that Bartol us' arguments concerning 
the bannitus as Roman citizen were already part of the civilian tradition on 
banishment, it is equally true that no jurist before Bartolus drew the different strands 
should note that, contrary to his own position, Bartolus, at C.6.24.1, fol. l 8va, tentatively 
posits that, although the bannitus can accede to an inheritance de iure communi, 'forsan 
tamen posset commune sibi auferre tanquam ab indigno'. In Roman law, the indignus 
possess testamentary powers, but the inheritance is nonetheless taken away from him 
(usually by the fisc) because of his ungrateful conduct towards the testator - Berger, 
Encyclopedic Dictionary, p. 499. Bartolus nowhere develops this idea. Baldus reproves his 
master in this respect at C.6.24.1 - seen. 331 immediately below. 
331 Baldus to C.6.24.1, fols 70vb-lra: 'Tertio quaeritur, si bannitus ... succedit... an fiscus 
illius civitatis ... possit auferre ei illam haereditatem tanquam indigno? Bar. sentit quod sic. 
Sed contrarium est verum, quia nemo est indignus nisi in casibus expressis in iure .. . Sed 
nullo iure cavetur quod bannitus .. . sit indignus. ergo non de bet auferri ei tanquam indigno 
nisi statutum hoe diceret expresse .. . ' 
332 Baldus to D.1.6.7, fol. 36vb: ' illud pro constanti teneas, quod statutum potest in suo 
territorio auferre bannito non solum beneficium statuti, sed etiam ius commune ... ' 
333 Baldus, at C.6.24.1, fol. ?Ova, went on to summarise this rule more clearly than any other 
jurist: ' ... et valde stricti iuris, ita quod quicquid expresse non reperitur infertum in banno et 
condemnatione intelligitur omnino omissum ... ' 
261 
'ill 
111111 
II 
11:11
11 
111 
of this part of the discourse together as explicitly and succinctly as Bartolus did in 
Lucanae civitatis. 
Moreover, in many respects, Bartolus' account of the communal ban in Lucanae 
civitatis brings to the fore the general principles that governed the communis opinio 
of the fourteenth century. For the municipal bannitus, the first safety net was not 
natural law, as was the case for the deportatus and imperial bannitus, but the Roman 
civil law. The communal ban primarily operated at the interface between local 
statutory law and the ius commune, not between the ius commune and natural law -
or to put it into more abstracts terms, the interface between the local and the 
universal city, not between the universal city and natural right. For this reason, the 
banished person's legal position was an expression of the jurists' broader conception 
of the relationship between municipal statute and ius commune. Since the ius 
commune was a universal legal norm valid in each local city, their citizens, including 
their banniti, were by default also Roman citizens. When statute clashed with the ius 
commune, it was interpreted restrictively: hence the ban was interpreted restrictively 
and only took away certain circumscribed rights in ways deemed acceptable by the 
jurists. Statute could not abrogate the ius commune but only derogate from it within 
its territory: hence the ban could not generally deprive the bannitus of Roman 
citizenship and was territorially restricted. 
Of course, as a human being, the communal bannitus was also protected by the 
natural law; this acted as a secondary safety net. The relationship between municipal 
statute and natural law was, generally speaking, equivalent to the relationship 
between the ius commune and natural law. Statute could therefore derogate from the 
natural law ex iusta causa and, for example, permit the bannitus to be killed with 
impunity. We observed in chapters 2 and 3 that there was no set rule about the extent 
to which the punitive mechanisms of the ius commune could derogate from natural 
law. Cause remained wedded to relatively vague notions of public utility and the 
authoritative examples of the Roman law. The same casuistry is evident in the 
municipal context. Ultimately, what could be done, and not done, in the name of the 
public good was never clearly defined. This explains the fact that the civilians did 
not rationalise the inviolability of the banished person's natural right to judicial 
defence. As we saw in chapter 3, the jurists continued to argue that the papal prince 
could derogate from the provisions of Pastora/is cura with just cause; 334 save 
Albericus de Rosate, who rejected the entire argument of Clement's bull, they did 
not allow statute to do the same. That the pope had greater powers to suspend natural 
334 Ch. 3, pp. 134-6. 
262 
law than the Italian cities should not come as a surprise, given his exceptional status 
in medieval legal thought. 
The fundamental casuistry governmg the relationship between civil and natural 
norms was directly reflected in the relationship between local and universal civil 
norms, that is, municipal statute and ius commune. The jurists allowed the statutory 
mechanism of banishment to take away certain rights that the bannitus, as Roman 
citizen, derived from the ius commune, but they did not provide a rule about how 
much of the ius commune, under what circumstances, could be suspended by statute. 
Ultimately, statute operated to the extent that the jurists deemed reasonable, and that 
reasonableness was, above all, based on the authority of analogical arguments from 
Roman law. In other words, just as there was no clear boundary between nature and 
the universal city of Rome, so there was no clear boundary between the local and the 
universal city in the medieval civilian tradition. 
The final point of this section concerns the role of the Roman empire. As Magnus 
Ryan has observed, historians have traditionally regarded the empire as a problem in 
civilian political thought. Given that the medieval empire was a mere shadow of its 
classical self, the argument goes, the all-powerful empire of the Corpus iuris was a 
troublesome legal fiction that had to be overcome in order to justify medieval Italian 
realities: 'as if the crisis unleashed by Pope Gregory VII at Canossa would only be 
laid to theoretical rest in the lecture-rooms of Perugia' .335 More recently, Kenneth 
Penn.ington has questioned the practical value of the jurists' tenacious insistence on 
the continued de iure universality of the empire, ' a theoretical overlordship with few 
if any concrete consequences' . 336 Although both claims contain a kernel of truth, 
they do not do justice to the much greater role that the empire actually played in 
civilian thought. In the discourse on banishment, both imperial and communal, 
granted that the clash between statute and ius commune was subject to discussion, the 
existence of the empire itself was recognised as an unproblematic part of legal 
reality. Rather than being an obstacle that had to be overcome, it served as tool in 
defining the legal status of banniti. In imperial banishment, it tended to aggravate the 
position of the bannitus because of the draconian measures in Roman law directed 
against those who rebel against the emperor. In communal banishment, by way of 
contrast, it tended to do the opposite. It is only because of the empire's continued 
existence that the bannitus of the fourteenth century could draw on Roman 
citizenship as an invaluable source of rights. If Baldus' cons ilia are reflective of the 
335 Ryan, Barto/us of Sassoferrato , p. 66. 
336 Pennington, The Prince and the Law, p. 183. 
263 
II 
Ill, 
impact of juristic communis opinio on legal practice, then not a few medieval banniti 
would have been grateful to this legal fiction. 
6. Conclusions 
i. Medieval Citizenship( s) 
The discourse on communal banishment demonstrates that entitlements in the private 
sphere, specifically in the law of inheritance, were at the very heart of the concept of 
municipal citizenship. For when the jurists explained the loss of local citizenship, 
they pointed to the loss of these entitlements as the foremost consequence. It was 
only logical that citizenship mainly consisted in a collection of advantageous rights 
given that its forfeiture was one of the penalties of banishment. Had citizenship been 
predominantly a collection of obligations, the absurd situation would have arisen in 
which the bannitus benefited from his ban. The same logic explains why loss of 
Roman citizenship was a punishment in Roman law. That is not to deny that duties 
were an important part of municipal citizenship, and there is plenty of evidence that 
the liability to pay civic taxes (munera) was central amongst these. It is also clear 
that citizens possessed certain rights in the public-political realm not shared by 
foreigners, i.e. the ability to hold public offices (honores). What the civilian 
discourse on the ban does, is to put the place of munera and honores into 
perspective, which allows us to correct the skewed interpretation of municipal 
citizenship that has resulted from a focus on these aspects in the existing 
historiography. 
This more balanced understanding of municipal citizenship also militates against 
aligning it closely with the Aristotelian ideas of contemporary scholastic 
philosophers. In fact, Roman and local citizenship in the civilian tradition had much 
more in common with each other than with a conception that defined civic 
membership along a scale of judicial and political participation (simpliciter, 
secundum quid); they were much richer concepts than that. Both Roman and local 
citizenship were purely civil statuses, meaning that they were entirely dependent on 
the positive law of their respective city. Both were conceived as valuable assets, 
principally bestowing rights rather than obligations. These rights were primarily, 
264 
though not exclusively, located in private law. It is true, as Canning has 
demonstrated, that Baldus imported the Aristotelian concept of natural political man 
when he discussed citizenship outside the discourse on banishment. In this respect, 
however, Baldus was unique amongst the civilians. Remarkably, his Aristotelianism 
did not influence his theory of the ban: rather than imbuing citizenship with a natural 
quality, he followed his predecessors in presenting it as a purely civil status. 
Similarly, Baldus' argument that office-holding is the defining feature of citizenship 
does not reappear in this context, and the status continues to be coupled to private-
law rights. It seems that the discourse on banishment was too well established by 
Baldus' time for him to integrate it effectively with other parts of his thought. 
The discourse on banishment allows us to critique two further arguments in the 
historiography. The first is that municipal citizenship was a contract. Although 
banishment resulted in the loss of citizenship, the assessment that this constituted a 
breach of contract is nowhere to be found. This indicates that the rare instances in 
which the jurists applied the language of contract to the relationship between citizen 
and community are not expressions of a fundamental theory. Either the jurists did not 
view citizenship as a contract in a meaningful technical sense, or the idea did not 
have a very wide sphere of application in their thought. Secondly, it is worth 
reiterating one of the points raised in relation to Costa' s grand narrative of medieval 
citizenship in chapter 2.337 It was clearly the case that the citizens of the Italian cities 
were not equal in the eyes of the law; in that respect, Costa's argument about 
hierarchy and diverse forms of belonging applies. The specialist literature has 
demonstrated that the jurists allowed for different levels of political participation,338 
and granted distinct sets of rights and duties depending on whether a citizen was 
male or female, 339 a layman or a cleric, 340 a citizen of the city itself or of its contado 
(civis comitatensis). 341 Nevertheless, despite the fact that there were diverse citizens, 
it is clear that, just as in the case of Roman citizenship, there was such a thing as an 
underlying legal status shared by all citizens, which bestowed a common set of rights 
337 Ch. 2, pp. 88-90. 
338 See eh. 1, pp. 14-15 . 
339 Kirshner, 'Mulier alibi nupta '; Menzinger, 'Ladonna medievale ' . 
34
° Canning, Baldus de Ubaldis , p. 134 ff. 
341 Kirshner, 'Civitas sibifaciat civem' , pp. 702-5 ; Meier, Mensch und Burger, pp. 176-9; 
Gilli, 'Comment cesser d'etre etranger', pp. 71-2. 
265 
differentiating them from foreigners. 342 It was this shared status that was lost through 
the ban. 
ii. The Reception of the hostis in the Municipal Context 
When it came to the reception of the hostis, broadly speaking, the discourses on 
communal and imperial banishment developed in the same manner. It was only with 
Bartolus that the hostis replaced the ordinary figures of Roman criminal law as the 
central analogy for the bannitus. On the imperial level, as we observed in the 
previous chapter, this resulted in a modification of the classical concept. Bartolus 
moved the hostis away from the sphere of the exceptional and towards the sphere of 
the normal by placing the hostis within a defined legal framework and by limiting 
extraordinary forms of procedure; he thereby weakened the force of the original idea. 
An equivalent process is apparent on the municipal level, where the jurists went even 
further in reducing the import of the hostis analogy. The position. of the communal 
hostis-bannitus was thus governed not only by natural but also by civil legal norms, 
which ensured that he possessed a wider range of rights than his imperial 
counterpart. Moreover, the jurists did not embrace the ancient hostis as a procedural 
model. It was firmly established communis opinio that a summons, and therefore the 
opportunity to present a defence, must precede a court's banishment of a defendant, 
and that the bannitus, if accused of another crime or sued in a civil matter, must be 
heard in his legitimate defences. Insofar as the jurists discussed this issue at all, 
Bartolus' view of ipso iure sentencing in the case of imperial banishment, which 
went quite a long way in undermining defendants' rights, was not widely shared in 
the municipal context. The favoured opinion was that even ipso iure banishment 
required a declaratory trial and sentence before the defendant could be treated as an 
outlaw. This variation may partly be explained by the tacit assumption that offences 
against the emperor, the supreme temporal lord of western Christendom, were of 
much greater consequence than offences against an Italian city. 
The most plausible reason for this modification of the classical concept of the hostis 
on the municipal level is the jurists ' principal view of the ban not as a political 
device to be used in emergencies, like the ancient hostis declaration, but as the 
342 A point nicely brought out (prior to Costa's book) by Canning, Baldus de Ubaldis, pp. 
l 77-81 (in the context of Baldus' thought on the acquisition of citizenship by grant), and 
then by Meier, Mensch und Burger, pp. 191-3. 
266 
ordinary legal mechanism employed against contumacious defendants in criminal 
trials. In ordinary procedure for contumacy - part of whose function was to give the 
defendant the chance to obey before his final banishment and to return to obedience 
afterwards - it made no sense to employ a mechanism akin to the hostis declaration. 
Undoubtedly, the ban, particularly in its guise as bannum pro delicto, was also used 
for political purposes, but that was not the primary function of the institution in 
communal practice. Even in cases of treason and rebellion, however, it seems likely 
that the cities were normally unwilling to circumvent established procedural 
safeguards (i.e. summonses) with ipso iure provisions, which would be the most 
reasonable explanation for the lack of a real civilian theory of exceptional procedure. 
In this regard, the standard practice of the communal courts appears to have been the 
same as that of the imperial court, revealing once again a medieval mindset 
(probably itself influenced by civilian procedural ideals) that viewed the opportunity 
to present a defence as a prerequisite for the validity of judicial proceedings. 
267 
II' 
: I 
I 
I I 
I 
268 
Abbreviations 
AG 
ASI 
BIMA 
BSSD 
CGJC 
CHMPT 
DRTR 
HPT 
MEFRM 
MGH 
OIR 
RIDC 
SQCD 
Bibliography 
Archivio giuridico 
Archivio storico italiano 
Gaudentius, A. (ed.) Bibliotheca iuridica medii aevi, Scripta 
anecdota glossatorum, 3 vols (Bologna: [various publishers], 
1888-1901). 
Segoloni, D., (ed.), Bartolo da Sasso/errata: Studi e 
documenti per il VI centenario, 2 vols (Milan: Giuffre, 1962). 
Viora, M. (ed.), Corpus glossatorumjuris civilis, 11 vols 
(Turin: ex officina Erasmiana, 1966-73). 
Bums, J. H. (ed.), The Cambridge History of Medieval 
Political Thought, ea. 350-1450 (Cambridge: Cambridge 
University Press, 1988). 
La nozione di 'romano ' tra cittadinanza e univers italita: 
Proceedings of the Second International Seminar on 
Historical Studies 'Da Roma alla terza Roma', April 21-23, 
1982 (Naples: Edizioni scientifiche italiane, 1982). 
History of Political Thought 
Melanges de I 'Ecole fram;aise de Rome, Mayen Age 
Monumenta Germaniae Historica 
Maffei, D., Cortese, E., Rossi, G. (eds), Opera iuridica 
rariora, 30 vols (Bologna: Forni, 1968-84). 
Rivista internazionale di diritto comune 
D'Amelio, G., Campitelli, A., Caprioli, S., Martino, F., Studi 
sulle 'quaestiones' civilistiche disputate nelle universita 
medievali, (Catania: Tringale, 1980). 
269 
I 
I! 
I
I! 
,1, 
II 
I 
11 
TR Tijdschrift voor Rechtsgeschiedenis 
Tractatus (Lyon) Tractatus ex variis iuris interpretibus collecti, 18 vols (Lyon, 
1549). 
Tractatus (Venice) Tractatus illustrium in utraque tum pontificii tum caesarei 
iuris facultate iurisconsultorum, 28 vols (Venice: 1584-6). 
TRHS Transactions of the Royal Historical Society 
Primary Manuscript Sources 
Rome, Biblioteca Apostolica Vaticana, MS. Chigi E.VIII 
Vienna, Osterreichische Nationalbibliothek, MS. 2094 
Primary Printed Sources 
Accursius/Jacobus Columbi (Hugolinus), Summa super usibus feudorum, (ed.) I. B. 
Palmeri us, in A. Gaudentius et al. (eds), BIMA, vol. 2, Scrip ta anecdota 
antiquissimorum glossatorum (Bologna: in aedibus Petri Virano olim fratrum 
Treves, 1892), pp. 181-94. 
Albericus de Rosate, Commentarii [to the Digest and Code], 8 vols (Venice, 1585-6), 
repr. OIR 21-28. 
- Quaestiones statutorum (=Commentarium de statutis), Tractatus (Venice), 
vol. 2 (1584), fols 2r-85v. 
Albertus Galeotti, Aurea margarita ac pene divina (Cologne: apud Ioannem 
Gymnicum, 1595). 
Albertus Gandinus, Quaestiones statutorum, (ed.) H. Solmius, in I. B. Palmerius et 
al. (eds), BIMA, vol. 3, Scripta anecdota glossatorum vel glossatorum aetate 
composita (Bologna: in aedibus successorum Monti, 1901), pp. 154-214. 
- Tractatus de maleficiis, (ed.) H. Kantorowicz, in Albertus Gandinus und das 
Strafrecht der Scholastik, vol. 2 (Berlin, Leipzig: W. de Gruyter & Co., 1926). 
Azo, Quaestiones, (ed.) E. Landsberg, Die Quaestiones des Azo (Freiburg: Mohr, 1888). 
Baldus de Ubaldis, Commentaria [to the Digest and Code], 8 vols (Venice: apud 
Iuntas, 1615-16). 
270 
- Consiliorum sive responsorum volumina I-V (Venice: Polus, 1575), repr. 
Turin: Bottega d'Erasmo, 1970. 
-In ususfeudorum commentaria (Lyon, 1585). 
Bartolus of Sassoferrato, Commentaria [to the Digest and Code], 9 vols (Turin: apud 
Nicolaum Beuilaquam, 1574). 
- Consilia, quaestiones, et tractatus (Basel: ex officina episcopiana, 1588). 
- Consilia, quaestiones, et tractatus (Venice, 1596). 
- De Gue/phis et Gebellinis, De tyranno, (ed.) D. Quaglioni, Politica e diritto 
nel Trecento italiano: II 'De tyranno' di Bartolo da Sassoferrato (1314-
1357), con l'edizione dei trattati 'De Gue/phis et Gebellinis ', 'De regimine 
civitatis' e 'De tyranno' (Firenze: Olschki, 1983). 
Bellomo, M., Quaestiones in iure civili disputatae: Didattica e prassi colta nel 
sistema del diritto comune fra Duecento e Trecento (Roma: Istituto storico 
italiano per ii Medio Evo, 2008). 
Belloni, A., Le questioni civilistiche del secolo XII: Da Bulgaro a Pillio da Medicina 
e Azzone (Frankfurt a. M.: V. Klostermann, 1989). 
Caprioli, S., 'La miscellanea romana dell'archivio di stato (MS. 1004)', in SQCD, 
pp. 11 7-214. 
Cinus de Pistoia, In Codicem et aliquot titulos primi Pandectorum tomi, id est, 
Digesti veteris, doctissima commentaria (Frankfurt a. M.: S. 
Feyerabendt, 1578). 
Corpus iuris canonici, 4 vols (Rome: in aedibus populi Romani, 1582). 
Corpus iuris civilis, [printed with the Glossa Ordinaria of Accursius], 5 vols (Lyon: 
Hugues de la Porte, 1558-60). 
Dinus de Mugello, Super lnfortiato et Digesto nova (Lyon: Jacob Myt, 1513), repr. 
OIR 17. 
Doenniges, W. (ed.), Acta Henrici VII imperatoris Romanorum et monumenta 
quaedam alia medii aevi (Berlin: Nicolai, 1839). 
Guilelmus de Cuneo, Lectura super Cadice (Lyon, 1513), repr. OIR 8. 
Guilelmus Duranti, Speculum iudiciale (=Speculum iuris), [printed with the 
· additiones of Johannes Andreae ], 2 vols, (Frankfurt: sumtibus heredum 
Andreae Wecheli & Ioannis Gymnici, 1592). 
Hobbes, T., Leviathan, (ed.) R. Tuck (Cambridge: Cambridge University Press, 
1996). 
Jacobus Butrigarius, In primam et secundam veteris Digesti partem, 2 vols (Rome, 
1606), repr. OIR 14. 
-Lectura super Cadice (Paris, 1516), repr. OIR 13. 
Jacobus de Ardizione, Summafeudorum (Asti: F. Silva, 1518), repr. CGJC 5. 
Jacobus de Arena, Commentarii in universum ius civile (Lyon, 1541 ), repr. OIR 16. 
- Tractatus de bannitis, in Tractatus (Venice), vol. 11, part 1 (1584), fols 
355r-7r. 
Jacobus de Belviso, Commentarii in authenticum et consuetudines feudorum (Lyon, 
1511), repr. OIR 12. 
Johannes Andreae, Novella super Sexto Decretalium (Venice: A. Torresanus, 1491). 
271 
I 
11 
,11 
I 
11 
I 
I 
I 
Ill 
I 
I 
Krueger, P., Mommsen, T., Scholl, R. (eds), Corpus iuris civilis, 3 vols (Berlin: 
Weidmann, 1886-1895). 
Lehmann, K., Das Langobardische Lehnrecht (Gottingen: Dieterich, 1896). 
Martino, F., '"Quaestiones" civilistiche disputate a Bologna negli ultimi decenni del 
secolo XIII: Studio sui manoscritto Leipzig, U.B., 992 e Paris, B.N., Lat. 
4489', in SQCD, pp. 223-96. 
Monti, G. M., Cina da Pistoia: Le "Quaestiones" e i "Cons ilia" (Milan: Vita e 
pensiero, 1942. 
Nello da San Gimignano, Tractatus de bannitis, in Tractatus (Lyon), vol. 10, fols 
l 12v-161 V. 
Nicolini, U., 'Un "consilium" inedito di Guido da Suzzara e la lotta politica a 
Perugia al tempo di Corradino', Annali di storia de! diritto 8 ( 1964 ), pp. 349-
55. 
Odofredus, Lectura super Cadice, 2 vols (Lyon, 1552), repr. OIR 5. 
- Summa in usus feudorum (Complutum [Alcala de Henares]: Q. Gerardus, 
1584). 
Oldradus de Ponte, Consilia et quaestiones (Rome: A Rot, 1472). 
Ottonis et Rahewini gesta Friderici I. imperatoris, (eds) G. Waitz and B. de Simson, 
3rd ed. (Hanover, Leipzig: impensis bibliopolii Hahniani, 1912). 
Pillius, Libellus de preparatoriis litium et earum preambulis, ( ed.) I. B. 
Palmerius, in Palmerius et al. (eds), BIMA, vol. 3, Scripta anecdota 
glossatorum vel glossatorum aetate composita (Bologna: In aedibus 
successorum Monti, 1901), pp. 14-68. 
Rainerius de Forlivio, Lectura super Digesto nova (Lyon, 1523), repr. OIR 9. 
Richter, E. L. and Friedberg, E. (eds), Corpus iuris canonici, 2nd ed., 2 vols (Leipzig: 
B. Tauchnitz, 1879-1881). 
Roffredus, Libelli iuris civilis, Libelli iuris canonici, Quaestiones sabbatinae 
(Avignon, 1500), repr. CGJC 6. 
Schwalm, J., (ed.), MGH, Legum sectio IV Constitutiones et acta publica 
imperatorum et regum, Tomus IV 1298-1313, Pars II (Hanover and Leipzig: 
impensis bibliopolii Hahniani, 1909-11 ). 
Scott, S. P. (trans.), The Civil Law, 17 vols (Cincinnati: The Central Trust Company, 
1932), online at http://www.constitution.org/sps/sps.htm, uploaded by the 
Constitution Society between 19.06.2002 and 10.01.2003, last modified 
10.06.2016. 
Suetonius, The Lives of the Caesars, The Lives of Illustrious Men, (trans.) J. C. 
Rolfe, Loeb Classical Library, 2 vols (London: W. Heinemann/ Cambridge, 
MA: Harvard University Press, 1970). 
Watson, A. (ed.), The Digest of Justinian, rev. English-language ed. , 4 vols 
(Philadelphia: University of Pennsylvania Press, 1998). 
Weiland, L. (ed.), MGH, Legum sectio IV Constitutiones et acta publica 
imperatorum et regum, To mus II: 1198-12 72 (Hanover: impensis bibliopolii 
Hahniani, 1896). 
272 
Secondary Sources 
Abbott, F. F., and Johnson, A. C., Municipal Administration in the Roman Empire 
(Princeton: Princeton University Press, 1926). 
Ascheri, M., 'Nella citta medievale italiana: la cittadinanza o le cittadinanze?', 
Initium: Revista catalana d'historia de! dret 16 (2011), pp. 299-312. 
Balsdon, J.P. V. D., Romans and Aliens (London: Duckworth, 1979). 
Bami, G. , "'Civis" e "civitas" nel "De statutis" di Alberico da Rosate' , in Studi in 
onore di Giuseppe Grosso, vol. 4 (Turin: G. Giappichelli, 1971), pp. 497-513. 
Baxendale, S. F., 'Exiles in Practice: The Alberti Family In and Out of Florence, 
1401-1428', Renaissance Quarterly 44 (1991 ), pp. 720-56. 
Berger, A., Encyclopedic Dictionary of Roman Law, Transactions of the American 
Philosophical Society, new series, 43 (1953), repr. 1991. 
Betti, E., 'La dottrina costruita da Bartolo sulla constitutio "Ad reprimendus", in 
BSSD, vol. 2, pp. 37-47. 
Bizzarri, D., 'Ricerche sul diritto di cittadinanza nella costituzione comunale', Studi 
senesi 32 (1916), pp. 19-136. 
Blanshei, S., 'Crime and Law Enforcement in Medieval Bologna', Journal of Social 
History 16 (1982). 
- Politics and Justice in Late Medieval Bologna (Leiden, Boston: Brill, 2010). 
Bowsky, W. M., 'A New Consilium of Cino of Pistoia (1324): Citizenship, 
Residence, and Taxation' , Speculum 42 ( 1967), pp. 431-41. 
- Henry VII in Italy: The Conflicts of Empire and City-State, 1310-1313 
(Lincoln: University of Nebraska Press, 1960). 
Brett; A. S.; Changes of State: Nature and the Limits of the City in Early Modern 
Natural Law (Princeton, Oxford: Princeton University Press, 2011 ). 
- Liberty, Right and Nature: Individual Rights in Later Scholastic Thought 
(Cambridge: Cambridge University Press, 1997). 
Brown, A., 'Insiders and Outsiders: The Changing Boundaries of Exile ', in W. J. 
Connell, (ed.), Society and Individual in Renaissance Florence (Berkeley: 
University of California Press, 2002), pp. 37-83. 
Buckland, W. W., and Stein, P., A Text-Book of Roman Law from Augustus to 
Justinian, 3rd ed. (Cambridge: Cambridge University Press, 1963). 
Calisse, C., Storia de! diritto penale italiano dal secolo VI al XIX (Firenze: G. 
Barbera, 1895). 
Canning, J., 'A Fourteenth-Century Contribution to the Theory of Citizenship: 
Political Man and the Problem of Created Citizenship in the Thought of 
Baldus de Ubaldis' , in B. Tierney, and P. A. Linehan (eds), Authority and 
Power: Studies on Medieval Law and Government Presented to Walter 
Ullmann on his Seventieth Birthday (Cambridge: Cambridge University 
Press, 1980), pp. 197-212. 
- 'Ideas of the State in the 13th and 14th Century Commentators on the 
Roman Law' , TRHS, 5th series, 33 (1983), pp. 1-27. 
273 
- 'Italian Juristic Thought and the Realities of Power in the Fourteenth 
Century', in Canning and O. G. Oexle (eds), Political Thought and the 
Realities of Power in the Middle Ages (Gottingen: Vandenhoeck & Ruprecht, 
1998), pp. 228-39. 
- 'Law, Sovereignty, and Corporation theory, 1300-1450', in CHMPT, pp. 
454-76. 
- The Political Thought of Baldus de Ubaldis (Cambridge: Cambridge 
University Press, 1987). 
Carlyle, R. W., and Carlyle, A. J., A History of Mediaeval Political Theory in the 
West, 6 vols (repr. Edinburgh: Blackwood & Sons, 1970). 
Catalano, P., '!us Romanum: Note sulla fondazione del concerto', in DRTR, pp. 531-
57. 
Cavalca, D., II bando nella prassi e nella dottrina giuridica medievale (Milan: 
Giuffre, 1978). 
Coing, H., 'Zur Geschichte des Begriffs "subjektives Recht'", in H. Coing, F. H. 
Lawson, and K. Gronfors (eds), Das subjektive Recht und der Rechtsschutz 
der Personlichkeit (Frankfurt a. M., Berlin: A. Metzner, 1959), pp. 7-23. 
- (ed.), Handbuch der Quellen und Literatur der neueren europaischen 
Privatrechtsgeschichte, vol. 1, Mittelalter (1100-1500) (Munich: Beck, 
1973). 
Coleman, J., 'Medieval Discussions of Human Rights' , in W. Schmale (ed.), Human 
Rights and Cultural Diversity: Europe, Arabic-Islamic World, Africa, China 
(Goldbach: Keip, 1993), pp. 103-19. 
Coli, H., 'Civitas', in A. Azara and E. Eula (eds), Novissimo digesto italiano, vol. 3 
(Turin: Unione Tipografico-Editrice Torinese, 1967), pp. 338-43. 
Cortese, E., 'Cittadinanza: Diritto lntermedio', in Enciclopedia de! diritto, vol. 7 
(Milan: Giuffre, 1960), pp. 132-9. 
- La norma giuridica: Spunti teorici nel diritto comune classico, 2 vols 
(Milan: Giuffre, 1962-4). 
Costa, P., Civitas: Storia della cittadinanza in Europa, vol. 1, Dalla civilta comunale 
al settecento (Roma, Bari: Laterza, 1999). 
Crifo, G., 'Esilio', in Enciclopedia de! diritto, vol. 15 (Milan: Giuffre, 1966), pp. 
720-1. 
- Ricerche sull"'exilium" nel periodo repubblicano (Milan: Giuffre, 1961 ). 
Dilcher, H., Die sizilische Gesetzgebung Kaiser Friedrichs JI. Quellen der 
Constitutionen von Melfi und ihrer Novellen, (Koln, Wien: Bohlau, 1975). 
Dizionario biografico degli italiani, 80 vols (Roma: Istituto della Enciclopedia 
Italiana, 1960-2014). 
Dizionario biografico dei giuristi italiani, (eds) I. Birocchi, E. Cortese, A. Martone, 
M . N. Milerti, 2 vols (Bologna: Societa editrice il Mulino, 2013). 
Donahue, C., '!us in the Subjective Sense in Roman Law: Reflections on Villey and 
Tierney', in D. Maffei, I. Birocchi, M. Caravale, E. Conte, and U. Petronio 
(eds), A Ennio Cortese, vol. 1 (Rome: II Cigno, 2001), pp. 506-35. 
274 
Donati Giacomini, P., and Poma, G., Cittadini e non cittadini nel mondo romano: 
guida ai testi e ai documenti (Bologna: CLUEB, 1996). 
Eichmann, E., Acht und Bann im Reichsrecht des Mittelalters (Paderborn: F. 
Schoningh, 1909). 
Feenstra, R., 'Bartole dans le Pays-Bas (anciens et modernes)', in BSSD, vol. 1, pp. 
173-81. 
- 'Dominium and ius in re aliena: The Origins of a Civil Law 
Distinction', in P. Birks (ed.), New Perspectives in the Roman Law of 
Property: Essays for Barry Nicholas (Oxford: Clarendon Press, 1989), pp. 
111-22. 
Ficker, J., Forschungen zur Reichs- und Rechtsgeschichte Italiens, 4 vols (Innsbruck: 
Verlag der Wagner'schen Universitats-Buchhandlung, 1868-74). 
Franklin, 0 ., Das Reichshofgericht im Mittelalter, vol. 1 (Weimar: H. Bohlau, 1867). 
Ganz, M. A. , 'Paying the Price for Political Failure: Florentine Women in the 
Aftermath of 1466 ', Rinascimento 34 (1994), pp. 237-57. 
Gardner, J. F., Being a Roman Citizen (London, New York: Routledge, 1993). 
Garnsey, P., 'Roman Citizenship and Roman Law in the Late Empire', in S. Swain 
and M. Edwards (eds), Approaching Late Antiquity: The Transformation 
from Early to Late Empire (Oxford: Oxford University Press, 2004), pp. 133-
56. 
- Thinking about Property: From Antiquity to the Age of Revolution 
(Cambridge: Cambridge University Press, 2007). 
Gaulin, J.-L., 'Les registres de bannis pour derte a Bologne au Xllle siecle: une 
nouvelle source pour l 'histoire de l 'endertement' , MEFRM 109 ( 1997), pp. 
479-99. 
Ghisalberti, C., 'La condanna al bando nel diritto comune', AG 158 (1960), pp. 3-75. 
- 'Sulla teoria dei delirti di laesa maesta nel dirirto comune', AG 149 (1955), 
pp. 100-79. 
Gilli, P., 'Comment cesser d'etre etranger: citoyens et non,-citoyens de la fin ~du 
Moy en Age', in P. Lardin, and J .-L. Roche (eds), L 'Etranger au Mayen Age 
(Paris: Publications de la Sorbonne, 2000), pp. 59-77. 
Glare, P. G. W. (ed.), Oxford Latin Dictionary, 2°d ed., 2 vols (Oxford: Oxford 
University Press, 2012). 
Goria, F., 'Romani, cittadinanza ed estensione della legislazione imperiale nelle 
costituzioni di Giustiniano', in DRTR, pp. 277-342. 
Grignaschi, M., 'La definition du 'civis' dans la scholastique', Anciens Pays et 
Assemblees d'Etat 35 (1966), pp. 71-100. 
Heers, J., and Bee, C. (eds), Exil et civilisation en Jtalie: X!Ie-XV!e siecles (Paris: 
Presses Universitaires de Nancy, 1990). 
Heers, J., L 'esilio, la vita politica, e la societa nel Medioevo (Naples: Liguori, 1997). 
Parties and Political Life in the Medieval West, (trans.) D. Nicholas, 
(Amsterdam, Oxford: North-Holland Publishing Co., 1977). 
Hohfeld, W. N., Fundamental Legal Conceptions as Applied in Judicial Reasoning, 
(eds) D. Campbell and P. Thomas (Aldershot: Ashgate, 2001). 
275 
Hopfl, H., 'Fundamental Law and the Constitution in Sixteenth-Century France', in 
R. Schnur (ed.), Die Rolle der Juristen bei der Entstehung des modernen 
Staates (Berlin: Duncker & Humblot, 1986), pp. 327-56. 
Jones, A. H. M., The Later Roman Empire, 284-602: A Social, Economic, and 
Administrative Survey, 4 vols (Oxford: Blackwell, 1964). 
Kantorowicz, E. H., 'Kingship under the Impact of Scientific Jurisprudence', in M. 
Clagett, G. Post, and R. L. Reynolds (eds), Twelfth-Century Europe 
and the Foundations of Modern Society (Madison: University of Wisconsin 
Press, 1961 ), pp. 89-111. 
- 'Leben und Schriften des Albertus Gandinus', Zeitschrift der Savigny-
Stiftung fur Rechtsgeschichte, Romanistische Abteilung 44 (1923), pp. 224-
358. 
- 'The Quaestiones disputatae of the Glossators ', TR 16 ( 1939), pp. 1-67. 
Kaser, M., 'Zurn "Ius"-Begriff der Romer', Acta Juridica 1977 (Essays in Honor of 
Ben Reinart II), pp. 63-81. 
Kent, D., 'I Medici in esilio: una vittoria di famiglia ed una disfatta personale', ASI 
132 (1974), pp. 3-63. 
Kilcullen, J., 'Medieval Theories of Natural Rights', accessed online 08.11.2013 at 
http://www.mq.edu.au/about us/faculties and departments/faculty of arts/ 
rnhpir/staff/staff-politics and international relations/john kilcullen/ 
medieval theories of natural rights. 
Kirshner, J., 'A consilium of Rosello dei Roselli on the meaning of "Florentinus", 
"de Florentia", and "de populo"', Bulletin of Medieval Canon Law 6 (1976), 
pp. 87-91. 
- 'Ars Imitatur Naturam: A Consilium of Baldus on Naturalization in 
Florence', Viator 5 (1974), pp. 289-331. 
- 'Between Nature and Culture: An Opinion of Baldus on Venetian 
Citizenship', Journal of Medieval and Renaissance Studies 9 (1979), pp. I 79-
208. 
- 'Cittadinanza come genere nelle citta-stato del Medioevo e del 
Rinascimento', paper given at the conference 'Innesti. Storia delle donne, 
storia di genere, storia sociale', Siena, 7-9 Feb. 2003, accessed online at Reti 
Medievali, http://www.rmoa.unina.it/id/eprint/832, uploaded 26.11.2014, last 
modified 02.05.2015. 
- 'Civitas sibi facial civem: Bartolus of Sassoferrato's Doctrine on the 
Making of a Citizen', Speculum 48 (1973), pp. 694-713. 
- 'Donne maritate altrove: genere e cittadinanza in Italia', in S. Seidel 
Menchi, A. Jacobson Schutte, T. Kuehn (eds), Tempie spazi di vitafemminile 
tra medioevo ed eta moderna (Bologna: II Mulino, 1999), pp. 377-429. 
- 'Mulier alibi nupta', in I. Baumgartner (ed.), Consilia im spaten Mittelalter: 
Zurn historischen Aussagewert einer Quellengattung (Sigmaringen: J. 
Thorbecke Verlag, 1995), pp. 147-75. 
- 'Paolo di Castro on Cives ex Privilegio: A Controversy over the Legal 
Qualifications for Public Office in Early Fifteenth-Century Florence', in A. 
276 
Molho and J. A. Tedeschi (eds), Renaissance Studies in Honor of Hans 
Baron (Florence: G. C. Sansoni, 1971), pp. 227-64. 
- 'Wives' Claims against Insolvent Husbands in Late Medieval Italy', in J. 
Kirshner and S. F. Wemple (eds), Women of the Medieval World: Essays in 
Honor of John H. Mundy (Oxford: Basil Blackwell, 1985), pp. 256-303. 
Korkman, P., and Makinen, V. (eds), Transformations in Medieval and Early-
Modern Rights Discourse (Dordrecht: Springer, 2006). 
Kriechbaum, M., Actio, !us und Dominium in den Rechtslehren des 13. und 14. 
Jahrhunderts (Ebelsbach: Aktiv, 1996). 
Lange, H., Romisches Recht im Mittelalter, vol. 1, Die Glossatoren (Munich: Beck, 
1997). 
Lange, H., and Kriechbaum, M., Romisches Recht im Mittelalter, vol. 2, Die 
Kommentatoren (Munich: Beck, 2007). 
Langeli, A. B., and Corbucci, M. P., 'I "libri dei banditi" del comune di Perugia 
(1246-1262)', Bolletino de/la deputazione di storia patria per l'Umbria 75 
(1978), pp. 123-341. 
Lansing, C., The Florentine Magnates: Lineage and Faction in a Medieval 
Commune (Princeton: Princeton University Press, 1991). 
Laurent, J. K., 'The Exiles and the Signory: The Case of Ferrara', The Journal of 
Medieval and Renaissance Studies I I (1981), pp. 281-97. 
Leicht, P. S., 'Cino da Pistoia e la citazione di Re Roberto di Arrigo VII', ASI 112 
(1954),pp. 313-20. 
Liebenam, W., Stadteverwaltung im romischen Kaiserreiche (Leipzig: Duncker & 
Humblot, 1900). 
Liebeschuetz, W., 'Citizen Status and Law in the Roman Empire and the Visigothic 
Kingdom', in W. Pohl and H. Reimitz (eds), Strategies of Distinction: The 
Construction of Ethnic Communities, 3 00-800 (Leiden: Brill, 1998), pp. 131-
52. 
Martino, F., 'Iacopo Belvisi in Provenza ea Napoli', RIDC 17 (2006), pp. 61-93. 
- Ricerche sull'opera di Guido da Suzzara: Le "Supleciones" (Catania: 
Tringale, 1981 ). 
Mathisen, R. W., 'Concepts of Citizenship', in S. F. Johnson (ed.), The Oxford 
Handbook of Late Antiquity (Oxford: Oxford University Press, 2012), pp. 
744-65. 
- 'Peregrini, Barbari, and Cives Romani: Concepts of Citizenship and the 
Legal Identity of Barbarians in the Later Roman Empire', American 
Historical Review 111 (2006), pp. 1011-1040. 
Mazzoni, V., 'Note sulla confisca dei beni dei ghibellini a Firenze nel 1267 e sul 
ruolo della Parte Guelfa', ASI 158 (2000), pp. 3-28. 
Makinen, V., 'The Evolution of Natural Rights, 1100-1500', in V. Makinen and P. 
Korkman (eds), Universalism in International Law and Political Philosophy 
(Helsinki: COLLEGIUM, 2008), pp. 105-19. 
277 
- (ed.), The Nature of Rights: Moral and Political Aspects of Rights in Late 
Medieval and Early Modern Philosophy (Helsinki: Philosophical Society of 
Finland, 2010). 
Mehu, D., 'Structure et utilisation des registres de bannis pour dettes a Bologne au 
XIIIe siecle', MEFRM 109 (1997), pp. 545-67. 
Meier, U., Mensch und Burger: Die Stadt im Denken spiitmittelalterlicher 
Theologen, Philosophen und Juris ten (Munich: Oldenbourg, 1994 ). 
Meijers, E. M., Le droit romain au moyen age, part 2, Etudes d'histoire du droit, vol. 
4 (Leiden, Universitaire Pers Leiden: 1966). 
Menzinger, S., 'Diritti di cittadinanza nelle quaestiones giuridiche duecentesche e 
inizio trecentesche (I)', MEFRM 125 (2013), accessed online at 
http://mefrm.revues.org/l 468, uploaded 09.12.2013. 
- 'Fisco, giurisdizione e cittadinanza nel pensiero dei giuristi comunali 
italiani tra la fine del XII secolo e l'inizio del XIII', Quellen und 
Forschungen aus italienischen Archiven und Bibliotheken 85 (2005), pp. 36-
73. 
- 'La donna medievale nella sfera pubblica: Alcune riflessioni in tema di 
cittadinanza nel panorama degli studi storico-giuridici ', in M. Davide (ed.), 
La condizione giuridica de/le donne nel medioevo: Atti de/ convegno di 
studio, Trieste, 23 novembre 2010 (Trieste: CERM, 2012), pp. 117-43. 
- 'Verso la costruzione di un diritto pubblico cittadino ', in E. Conte and S. 
Menzinger (eds), La Summa Trium Librorum di Rolando da Lucca (1195-
. 1234): Fisco, politica, scientia iuris (Rome: Viella, 2012), pp. cxxv-ccxvii. 
Meymal, E., 'Notes sur la formation de la theorie du domaine divise ( domaine direct 
et domaine utile) du Xlle au XIV siecle dans les romanistes', in Melanges 
Fitting, vol. 2 (Montpellier: Societe Anonyme de l 'Imprimerie Generale du 
Midi, 1908), pp. 409-61. 
Milani, G., 'Giuristi, giudici e fuoriusciti nelle citta italiane del duecento: note sul 
reato politico comunale ', in J. Chiffoleau, C. Gauvard, and A. Zorzi (eds), 
Pratiques sociales et politiques judiciaires dans le villes de / 'Occident a la 
fin du Mayen Age (Rome: Ecole frarn;aise de Rome, 2007), pp. 595-642. 
- L 'esclusione dal comune: Conjlitti e bandi politici a Bologna e in altre citta 
tra XII e XIV secolo (Rome: Istituto storico italiano peril Medio Evo, 2003). 
- 'Prime note su disciplina e pratica del bando a Bologna attorno alla meta del 
XIII secolo', MEFRM 109 (1997), pp. 501-23. 
Mommsen, T., Romisches Strafrecht (Leipzig: Duncker & Humblot 1899). 
- Romisches Staatsrecht, 3 vols (Leipzig: S. Hirzel, 1871-88). 
Montorsi, W., 'Plebiscitae Bononiae: II perduto Statutum Populi Bononie ed una 
raccolta di leggi sui beni dei banditi ', Bulletino dell 'Istituto storico italiano 
per ii Media Eva e Archivio muratoriano 70 (1958), pp. 181-98. 
Mooney, A. M. C., 'The Legal Ban in Florentine Statutory Law and the De bannitis 
of Nello da San Gimignano (13 73-1430)', unpublished Ph.D. dissertation, 
University of California, Los Angeles (1976). 
278 
Moyn, S., The Last Utopia: Human Rights in History (Cambridge, MA, London: 
Belknap, 2010). 
Mueller, R. C., Immigrazione e cittadinanza nella Venezia medievale (Rome: Viella, 
2010). 
Nicolini, U., La proprieta, ii principe e l'espropriazione per pubblica utilita (Milan: 
Giuffre, 1952). 
Noy, D., Foreigners at Rome: Citizens and Strangers (London: Duckworth, 2000). 
Norr, D., 'Origo: Studien zur Orts-, Stadt- und Reichszugehorigkeit in der Antike', 
TR 31 (1963), pp. 525-600. 
Norr, K. W., 'Zur Frage des subjektiven Rechts in der mittelalterlichen 
Rechtswissenschaft', in D. Medicus, H.-J. Mertens, Norr, and W. Zollner 
(eds), Festschrift fiir Hermann Lange zum 70. Geburtstag am 24. Januar 
1992 (Stuttgart, Berlin, Koln: W. Kohlhammer, 1992), pp. 193-204. 
Oakley, F., Natural Law, Laws of Nature, Natural Rights: Continuity and 
Discontinuity in the History of Ideas (New York: Continuum, 2005). 
Pazzaglini, P. R., The Criminal Ban of the Sienese Commune, 1225-1310 (Milan: 
Giuffre, 1979). 
Pennington, K., 'Due Process, Community and the Prince in the Evolution of the 
Ordo iudiciarius ', RIDC 9 (1998), pp. 9-47. 
- 'Law, Criminal Procedure', in W. C. Jordan (ed.), Dictionary of the Middle 
Ages: Supplement 1 (New York: Charles Scribner's Sons, 2004), pp. 309-20. 
- 'Law, Legislative Authority, and Theories of Government, 1150-1300', in 
CHMPT, pp. 424-53. 
- 'Review Essay: The History of Rights in Western Thought', Emory Law 
Journal 47 (1998), pp. 237-52. 
- The Prince and the Law, 1200-1600: Sovereignty and Rights in the Western 
Legal Tradition (Berkeley, Oxford: University of California Press, 1993). 
Pertile, A., Storia del diritto italiano dalla caduta dell 'impero alla codijicazione, vol. 
5, Storia del diritto penale (Padua: premiato stabilimento tipografico alla 
Minerva dei Fratelli Salmin, 1876). 
Piergiovanni, V., 'Alcuni consigli in tema di forestieri a Genova nel Medioevo', in 
M. del Treppo (ed.), Sistema di rapporti ed elites economiche in Europa 
(secoli XII-XV//), (Naples: Liguori, 1994), pp. 1-10. 
Planitz, H., 'Der Schuldbann in Italien', Zeitschrift der Savigny-Stiftung fiir 
Rechtsgeschichte, Germanistische Abtheilung 52 (1932), pp. 134-259. 
- Deutsche Rechtsgeschichte (Graz: H. Bohlaus, 1950). 
Poetsch, J., Die Reichsacht im Mittelalter und besonders in der neueren Zeit 
(Breslau: M. & H. Marcus, 1911 ). 
Post, G., 'Review: Walter Ullmann, The Individual and Society in the Middle 
Ages', Speculum 43 (1968), pp. 387-390. 
Pugliese, G., Actio e diritto subiettivo (Milan: Jovene, 1939). 
- "'Res corporales", "res incorporales" e il problema del diritto · soggettivo', in 
Studi in onore di Vicenzo Arangio-Ruiz, vol. 3 (Naples: Jovene, 1953), pp. 
223-60. 
279 
Ill 
Quaglioni, D., 'Le radici teoriche della dottrina bartoliana della cittadinanza' , in his 
"Civilis Sapientia ": Dottrine giuridiche e dottrine politiche fra medioevo ed 
eta moderna (Rimini: Maggioli, 1989), pp. 127-44. 
- 'The Legal Definition of Citizenship in the Late Middle Ages', in A. Molho, 
K. Raaflaub, and J. Emlen (eds), City States in Classical Antiquity and 
Medieval Italy (Stuttgart: F. Steiner, 1991), pp. 155-67. 
Reid, C. J., 'The Canonistic Contribution to the Western Rights Tradition: An 
Historical Inquiry', Boston College Law Review 33 (1991), pp. 37-92. 
- 'Thirteenth-Century Canon Law and Rights: The Word !us and its Range of 
Subjective Meanings', Studia Canonica 30 (1996), pp. 295-342. 
Reid, J. S., The Municipalities of the Roman Empire (Cambridge: Cambridge 
University Press, 1913) 
Ricciardelli, F., The Politics of Exclusion in Early Renaissance Florence (Turnhout: 
Brepols, 2007). 
Riesenberg, P., 'Citizenship and Equality in Late Medieval Italy', Studia Gratiana 
15 (1972), pp. 423-39. 
- 'Citizenship at Law in Late Medieval Italy', Viator 5 (1974), pp. 333-46. 
- Citizenship in the Western Tradition: Plato to Rousseau (Chapel Hill, 
London: University of North Carolina Press, 1992). 
- 'Civism and Roman Law in Fourteenth-Century Italian Society', in D. 
Herlihy, R. S. Lopez, and V. Slessarev (eds), Economy, Society, and 
Government in Medieval Italy (Kent: Kent State University Press, 1969), pp. 
237-54. 
Romano, A., Aspetti dell 'insegnamento giuridico nelle universita medievali. IV Le 
'quaestiones disputatae' (Reggio Calabria: Parallelo 3 8, 197 5). 
Rota, A., 'II diritto naturale come insieme di norme regolatrice dei rapporti della 
societa umana nella dottrina della prima rinascenza giuridica', Rivista 
internazionale difilosofia de! diritto 25 (1948), pp. 125-63. 
Rummer, J., 'A Fourteenth-Century Legal Opinion', The Quarterly Journal of the 
Library of Congress 25 (1968), pp. 178-93. 
Ryan, M. J., 'Bartolus of Sassoferrato and Free Cities', TRHS, 6th series, 10 (2000), 
pp. 65-89. 
- 'Freedom, Law, and the Medieval State', in Q. Skinner and B. Strath (eds), 
States and Citizens: History, Theory, Prospects (Cambridge: Cambridge 
University Press, 2003), pp. 51-62. 
- 'Political Thought', in D. Johnston (ed.), The Cambridge Companion to 
Roman Law (New York: Cambridge University Press, 2015), pp. 423-51. 
- 'Rulers and Justice, 1200-1500', in P. A. Linehan and J. L. Nelson (eds), 
The Medieval World (London: Routledge, 2001 ), pp. 503-15. 
Savigny, F. C. von, Geschichte des romischen Rechts im Mittelalter, 6 vols 
(Heidelberg: Mohr, 1815-31 ). 
Sbriccoli, M., Crimen laesae maiestatis: II problemo de! reato politico a/le soglie 
de/la scienza penalistica moderna (Giuffre: Milan, 1974). 
280 
Schmitt, C., The Concept of the Political, (trans.) G. Schwab (Chicago, London: 
University of Chicago Press, 2007). 
Shaw, C., The Politics of Exile in Renaissance Italy (Cambridge: Cambridge 
University Press, 2000). 
Sherwin-White, A. N., The Roman Citizenship, 2nd ed. (Oxford: Clarendon Press, 
1973). 
Siuts, H., Bann und Acht und ihre Grundlagen im Totenglauben (Berlin: Walter de 
Gruyter & Co., 1959). 
Stam, R., Contrary Commonwealth: The Theme of Exile in Medieval and 
Renaissance Italy (Berkely, London: University of California Press, 1982). 
Stein, P., Roman Law in European History (Cambridge: Cambridge University Press, 
1999). 
Tierney, B., '!us and Metonymy in Rufinus', in R. J. Castillo Lara (ed.), Studia in 
honorem eminentissimi cardinalis Alphonsi M Stickler (Rome: LAS, 1992), 
pp. 549-58. 
- 'Medieval Rights and Powers: On a Recent Interpretation', HPT 21 (2000), 
pp. 327-38. 
- 'Origins ofNatural Rights Language: Texts and Contexts, 1150-1250', HPT 
10 (1989), pp. 615-46. 
- Religion, Law, and the Growth of Constitutional Thought, 1150-1650 
(Cambridge: Cambridge University Press, 1982). 
- Rights, Laws and Infallibility in Medieval Thought (Aldershot: Variorum, 
1997). 
- The Idea of Natural Rights: Studies on Natural Rights, Natural Law and 
Church Law, 1150-1625 (Atlanta: Scholars Press, 1997). 
Tuck, R., Natural Rights Theories: Their Origin and Development (Cambridge: 
Cambridge University Press, 1981 ). 
Ullmann, W., 'De Bartoli sententia: Concilium repraesentat mentem populi', in his 
The Papacy and Political Ideas in the Middle Ages (London: Variorum 
Reprints, 1976). 
- 'The Rebirth of the Citizen on the Eve of the Renaissance Period', in A. R. 
Lewis (ed.), Aspects of the Renaissance: A Symposium. Papers Presented at a 
Conference on the Meaning of the Renaissance (Austin, London: University 
of Texas Press, 1967), pp. 5-25. 
Vallerani, M., 'Diritti di cittadinanza nelle quaestiones giuridiche duecentesche (II). 
Limiti dell'appartenenza e forme di esclusione', MEFRM 125 (2013), 
accessed online at http: //mefrm.revues.org/1446, uploaded 04.12.2013. 
- II sistema giudizario de/ comune di Perugia. Conflitti, reati e processi nella 
seconda meta de! XIII secolo (Perugia: Deputazione di storia patria per 
l'Umbria, 1991). 
- 'L'amministrazione della giustizia a Bologna in eta podestarile' , Atti e 
memorie de/la Deputazione di storia patria per le province di Romagna 43 
(1992), pp. 291-316. 
281 
- 'Logica della documentazione e logica dell'istituzione. Per una rilettura dei 
documenti in forma di lista nei communi italiani della prima meta del XIII 
secolo', in I. Lazzarini and G. Gardoni (eds), Notariato e medievistica. Per i 
cento anni di 'Studi e ricerche di diplomatica comunale' di Pietro Torelli 
(Rome: Istituto storico italiano peril Medio Evo, 2013), pp. 109-145. 
- '"Ursus in hoe disco te coget solvere fisco": Evasione fiscale, giustizia e 
cittadinanza a Bologna fra Due e Trecento', in E. C. Pia (ed.), Credito e 
cittadinanza nell'Europa mediterranea dal medioevo all'eta moderna (Asti: 
Centro studi Renato Bordone sui Lombardi, sul credito e sulla banca, 2014), 
pp. 39-50. 
Van de Kamp, J. L. J., Bartolo da Sassoferrato, (trans.) I. Hess-Lutomirski, (Urbino: 
Stabilimento tipografico editoriale urbinate, 1935). 
Villey, M., 'Du sens de !'expression jus in re en droit romain classique', Melanges 
Fernand de Visscher II. Revue internationale des droits de l'antiquite 2 
(1949), pp. 417-36. 
- La formation de la pensee juridique moderne, 4th ed. (Paris: Montchretien, 
1975). 
- 'La genese du droit subjectif chez Guillaume d'Occam'; Archives de 
philosophie du droit 9 (1964), pp. 97-127. 
- 'Le "jus in re" du droit romain classique au droit moderne', in Conferences 
faites a l'Institut de Droit Romain en 1947 (Paris: Recueil Sirey, 1950), pp. 
187-225. 
- 'Les origines de la notion du droit subjectif', in his Le9ons d'histoire de la 
philosophie du droit, 2nd ed. (Paris: Dalloz, 1962), pp. 221-50. 
- 'L'idee du droit subjectif et les systemes juridiques romains', Revue 
historique de droitfrant;ais et etranger, 4th series, 24 (1946), pp. 201-27. 
- 'Suum jus cuique tribuens', in Studi in onore di Pietro de Francisci, vol. 2 
(Milan: Giuffre, 1956), pp. 361-71. 
Villien, A. et al. (eds), Dictionnaire de droit canonique, 7 vols (Paris: Letouzey et 
Ane, 1924-65). 
Vodola, E., Excommunication in the Middle Ages (Berkely: University of California 
Press, 1986). 
Washburn, D. A., Banishment in the later Roman Empire, 284-476 CE (New York: 
Routledge, 2013). 
Weigand, R., Die Naturrechtslehre der Legisten und Dekretisten van Irnerius bis 
Accursius und van Gratian bis Johannes Teutonicus (Munich: Hueber, 1967). 
Will, E., 'Die Gutachten des Oldradus de Ponte zum Proze8 Heinrichs VII. gegen 
Robert von Neapel, nebst der Biographie des Oldradus', Abhandlungen zur 
Mittleren und Neueren Geschichte 65 (1917), pp. 1-63. 
Woolf, C. N. S., Barto/us of Sassoferrato: His Position in the History of Medieval 
Political Thought (Cambridge: Cambridge University Press, 1913). 
Zaremska, H., Les bannis au Mayen Age, (trans.) T. Douchy, (Paris: Aubier, 1996). 
282 
CAMBRIDGE 
UNIVERSITY 
LI BRARY 
11 
I