The Creation of the Ius Commune: From Casus to Regula 9780748642922

Studies the transformation of Roman legal rules into the 'common law' of Western Europe in the period 1100–140

157 118 2MB

English Pages 320 [318] Year 2010

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

The Creation of the Ius Commune: From Casus to Regula
 9780748642922

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

The Creation of the Ius Commune

EDINBURGH STUDIES IN LAW Series Editor Elspeth Reid (University of Edinburgh) Editorial Board David L Carey Miller (University of Aberdeen) George L Gretton (University of Edinburgh) Hector L MacQueen (University of Edinburgh) Kenneth G C Reid (University of Edinburgh) Reinhard Zimmermann (Max-Planck-Institute of Comparative and ­International Private Law, Hamburg) Volumes in the series: Elspeth Reid and David L Carey Miller (eds), A Mixed Legal System in ­Transition: T B Smith and the Progress of Scots Law (2005) Hector MacQueen and Reinhard Zimmermann (eds), European Contract Law: Scots and South African Perspectives (2006) John W Cairns and Paul du Plessis (eds), Beyond Dogmatics: Law and S ­ ociety in the Roman World (2007) William M Gordon, Roman Law, Scots Law and Legal History: Selected Essays (2007) Kenneth G C Reid, Marius J de Waal and Reinhard Zimmermann (eds), Exploring the Law of Succession: Studies National, Historical and Comparative (2007) Vernon Valentine Palmer and Elspeth Christie Reid (eds), Mixed Jurisdictions Compared: Private Law in Louisiana and Scotland (2009)

EDINBURGH STUDIES IN LAW VOLUME 7

The Creation of the Ius Commune From Casus to Regula Edited by John W Cairns and Paul J du Plessis

EDINBURGH UNIVERSITY PRESS

© The Edinburgh Law Review Trust and the Contributors, 2010 Edinburgh University Press Ltd 22 George Square, Edinburgh www.euppublishing.com Typeset in New Caledonia by Koinonia, Manchester, and printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne A CIP record for this book is available from the British Library ISBN 978 0 7486 3897 0 (hardback) The right of the contributors to be identified as authors of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988.

Contents



Preface List of Contributors List of Abbreviations

vii ix xi

Introduction John W Cairns and Paul J du Plessis

1

  1 The Sources of Medieval Learned Law Harry Dondorp and Eltjo J H Schrage

7

  2 The Infrastructure of the Early Ius Commune: The Formation of Regulae, or its Failure Kees Bezemer   3 Ius Quaerens Intellectum: The Method of the Medieval Civilians James Gordley

57 77

  4 Medieval Family and Marriage Law: From Actions of Status to Legal Doctrine Laurent L J M Waelkens

103

  5 The Roman Concept of Ownership and the Medieval Doctrine of Dominium Utile Thomas Rüfner

127

  6 Succession to Fiefs: A Ius Commune Feudorum? Magnus Ryan

143

  7 Towards the Medieval Law of Hypothec Paul J du Plessis

159

  8 The Ignorant Seller’s Liability for Latent Defects: One Regula or Various Sets of Rules? Jan Hallebeek   9 The Glossators’ Monetary Law Wolfgang Ernst

175 219

vi

the creation of the ius commune

10 Citations and the Construction of Procedural Law in the Ius Commune Richard H Helmholz

247

11 Doctoribus bona dona danda sunt: Actions to Recover Unpaid Legal Fees James A Brundage

277



295

Index

Preface

The papers collected in this volume were originally delivered on 12-13 December 2008 at a conference organised in Old College by the Centre for Legal History of the University of Edinburgh and the Edinburgh Roman Law Group. It is worth remembering that the Edinburgh Roman Law Group was established by the late Professor Peter Birks – sometime Professor of Civil Law in the University of Edinburgh – as an interdisciplinary forum dedicated to the study of Civil law and the Civilian tradition in its broadest sense. Its regular meetings attract a wide audience of students, specialists, and members of the public interested in its aims. Usually meeting three times each academic year, it is addressed by distinguished scholars in the field. The conference out of which this volume has grown was entitled “From Casus to Regula: The Creation of the Ius Commune”. All the papers delivered are included here in a revised form. The conference allowed fruitful discussion between the speakers and between the speakers and the audience, and this discussion is reflected in the versions of the papers presented as chapters here. The theme of the conference was chosen because there is little detailed work published in English on the development of the historical ius commune beyond grand narratives of considerable generality. If there are valuable individual studies and even monographs in English, the conference and consequent volume allowed a more general detailed assessment, and an assessment that involved proper consideration not only of the interpretation of the Roman-law texts, but also of those of canon law and feudal law, ranging over a wide area of legal practice and scholarship in the Middle Ages. The results allow and encourage development of a more nuanced account of how the medieval jurists took the texts of the Roman and canon laws and moved from casuistic jurisprudence towards the development of what would now be recognised as legal doctrine.

viii

the creation of the ius commune

The conference of December 2008 was the second mini-conference of those planned by the Roman Law Group and Centre for Legal History on a roughly four-year cycle. The first conference, held in 2005, led to the publication of J W Cairns and Paul du Plessis (eds), Beyond Dogmatics: Law and Society in the Roman World (2007) [= Edinburgh Studies in Law III]. That publication promised (at vii-viii) that the “next gathering will focus on the development of the ius commune in its historical sense in the period 1100-1400 and the rise of legal doctrine” and that “the collection of papers from this conference will form the second volume produced by the ERLG”. This volume fulfils that promise. The third mini-conference will be devoted to humanism and law. This volume, and the conference that preceded it, would not have been possible without the generous financial support of the Edinburgh Legal Educational Trust and the School of Law of the University of Edinburgh. The Edinburgh Roman Law Group and the Centre for Legal History thank both for their generosity and commitment to legal history. The School of Law’s Research Office, in particular Mrs Alison Stirling and Mrs Myra Reid, worked hard to make the conference happen and deserve our gratitude. Finally, the editors of this volume wish to thank the Editorial Board of the Edinburgh Studies in Law in general and Ms Elspeth Reid in particular for accepting this volume. JWC PJduP Old College, Edinburgh October 2009

List of Contributors

kees bezemer is Associate Professor, Faculty of Law, University of Leiden. james a brundage is Ahmanson-Murphy Distinguished Professor Emer­­ itus of History and Law, University of Kansas. john w cairns is Professor of Legal History, University of Edinburgh. harry dondorp is Associate Professor, Faculty of Law, VU University Amsterdam. wolfgang ernst is Professor of Roman Law, University of Zurich. james gordley is W R Irby Professor of Law, Tulane University School of Law. jan hallebeek is Professor of European Legal History, VU University Amsterdam. richard h helmholz is Ruth Wyatt Rosenson Distinguished Service Professor of Law, University of Chicago School of Law. paul j du plessis is Senior Lecturer in Law, University of Edinburgh. thomas rüfner is Professor of Law, University of Trier, Judge, Higher Regional Court at Koblenz. magnus ryan is Fellow of Peterhouse, and University Lecturer in History, Cambridge. eltjo j h schrage is Honorary Professor, Nelson Mandela Metropolitan University, Port Elizabeth and sometime Professor of Law, University of Amsterdam. laurent l j m waelkens is Professor of Roman Law, Catholic University of Leuven.

List of Abbreviations

Auth BamSB BIDR BIMAE

Authenticum Staatsbibliothek Bamberg Bulletino dell’Istituto di diritto romano Bibliotheca iuridica medii aevi: Scripta anecdota glossatorum, ed A Guadenzi, G B Palmieri et al 3 vols (1888-1901; 2nd edn 1913; repr 1962) BMCL Bulletin of Medieval Canon Law BML Bibliotheca Medicea Laurenziana (Florence) BN Bibliothèque Nationale; Biblioteca Nacional; Biblioteca Nazionale BSB Bayerische Staatsbibliothek (Munich) BV Biblioteca Vaticana (Rome) c contra c. capitulum C Code (of Justinian) C. causa ca circa CGIC Corpus Glossatorum Iuris Civilis, Curante Juris Italici Historiae Instituto Taurinensis Universitatis (1966–1973) CIC Corpus iuris canonici Clem Constitutiones Clementis V Coing, Handbuch  H Coing (ed), Handbuch der Quellen und Literatur der neueren Europäischen Privatrechtsgeschichte, 3 vols (19731988) Coll Collatio Comp. Compilatio CTh Codex Theodosianus d died d. dictum

xii

D D. DDC Dec DEC

the creation of the ius commune

Digest Distinctio R Naz (ed), Dictionnaire de droit canonique, 7 vols (1935-1965) Decisio G Alberigo et al (eds), Decrees of the Ecumenical Councils, trans N P Tanner et al, 2 vols (1990) d.p.(a.) dictum post (ante) EdTh Edictum Theodorici fl floruit Gai Inst Institutes of Gaius gl gloss Glos ord Glossa ordinaria i.f. in fine i.m. in medio J Inst Institutes of Justinian 2 Kaser, RPR M Kaser, Das römische Privatrecht. 1. Das altrömische, das vorklassische unde klassische Recht, 2nd edn [= Handbuch der Altertumswissenschaft, ser 10, Rechtsgeschichte des Altertums, part 3, vol 3] (1971 and 1975) 2 Kaser, ZPR M Kaser, Das römische Zivilprozessrecht, 2nd edn, rev by K Hackl [= Handbuch der Altertumswissenschaft, ser 10, Rechtsgeschichte des Altertums, part 3, vol 4] (1996) Lange, RRM H Lange and M Kriechbaum, Römisches Recht im Mittelalter, 2 vols (1997-2007) LF Libri feudorum LMA Lexikon des Mittelalters vol. I (München: Deutscher Taschenbuch Verlag, 2002) LQR Law Quarterly Review Meijers, Études  E M Meijers, Études d’histoire du droit, ed R Feenstra and H F W Fischer, 4 vols (1956-1966) MGH Monumenta Germaniae Historica MIC Monumenta iuris canonici NJW Neue Juristische Wochenschrift Nov Novels of Justinian OIR Opera Iuridica Rariora, selecta cura et studio Domenico Maffei et al, (1967- ) Part Las Siete Partidas del Don Alfonso el Sabio (cited by individual Partida) q. quaestio

list of abbreviations

QGRKP

xiii

Quellen zur Geschichte des römisch-kanonischen Prozesses im Mittelalter, ed L Wahrmund (1905-1931; repr 1962) r reigned RHD Revue historique du droit français et étranger RIDA Revue internationale des droits de l’antiquité RIDC Rivista internazionale di diritto commune RO Record Office RSDI Rivista del storio del diritto romano SDHI Studia et Documenta Historiae et Iuris Sext Liber Sextus SG Studia Gratiana sv sub voce TUJ Tractatus universi juris, ed Franciscus Zilettus, 22 vols in 28 (1584-1586) TvR Tijdschrift voor Rechtsgeschiedenis X Liber Extra (Decretales Gregorii IX) Zimmermann, Obligations  R Zimmermann, The Law of Obligations: Roman Foundations of the Civilian Tradition (1990; repr with rev 1996) ZRG, GA Zeitschrift der Savigny-Stiftung für Rechtsgeschichte, Germanistische Abteilung ZRG, KA Zeitschrift der Savigny-Stiftung für Rechtsgeschichte, Kanonistische Abteilung ZRG, RA Zeitschrift der Savigny-Stiftung für Rechtsgeschichte, Römanistische Abteilung

Introduction John W Cairns and Paul J du Plessis The jurist Paul wrote: “The law may not be derived from a rule, but a rule must arise from the law as it is.”1 This fragment is found in D 50.17, entitled De diversis regulis iuris antiqui, which consists of 211 such rules or regulae culled from the works of jurists of the classical era. The quotation from Paul emphasises the extent to which the Roman lawyers thought that regulae were extrapolated from cases, casus or causae. When the medieval jurists were confronted by the Roman texts, with their concrete discussion of particular cases, they looked at this title and this text for guidance and started to develop the casuistic Roman jurisprudence found in the Digest, Code, and Novels into more general rules or regulae. In the English-language literature Peter Stein explored this in his 1966 study, which focused on the general trends.2 Stein thereby contributed to a larger, well-known narrative about the development of a pan-European ius commune or universal common law, ca 1100-1400. Though there are antecedents, this grand narrative can for our purposes be dated to the epochal and brilliant work of Savigny on the medieval rediscovery of Roman law.3 What made Savigny’s study particularly valuable was his discussion of individual jurists and their works, bringing to light once more the significance of these early scholars. He did not especially focus on the juristic content of their labours. This set the agenda for much future scholarship, and the history of European law has tended since then to be told as the story of succeeding schools of jurists and their differing approaches to the texts without paying much attention to the details of their work. For an Anglophone audience, the short work of Vinogradoff, Roman Law in Medieval Europe (reprinted as recently as 1968) provided an influential account.4 Vinogradoff’s study was first published in 1909. Scholarship has developed considerably since then, retelling, refining and developing this traditional narrative. In particular, research started to take new directions, in ­important   1 D 50.17.1.   2 P Stein, Regulae Iuris: From Juristic Rules to Legal Maxims (1966) 131-152.   3 F C von Savigny, Geschichte des römischen Rechts im Mittelalter, 6 vols (1815–1831).   4 P Vinogradoff, Roman Law in Medieval Europe, with a new foreword by P Stein (1968).

1

2

the creation of the ius commune

works such as Paul Koschaker’s Europa und das römische Recht (1947), developing – broadly speaking – two branches of literature in this area of legal history. First, much more refined versions of the traditional narrative, sometimes with a national focus, appeared. Examples of this genre include Franz Wieacker’s highly influential Privatrechtsgeschichte der Neuzeit: unter besonderer Berücksichtigung der deutschen Entwicklung, 2nd ed (1967); Manlio Bellomo’s polemical The Common Legal Past of Europe 1000–1800 (translated from the Italian by Lydia Cochrane) (1995); and Peter Stein’s short but elegant Roman Law in European History (1999). It is perhaps fair to say that, despite the focus on Germany, Wieacker has provided a particularly seductive and powerful version of the grand narrative that still dominates modern thinking. The second type of literature has been much more focused on particular and specialised issues, such as canon law, the sources of law, political theory and the like. In the English-language scholarship, the work of Joseph Canning, for example, has been especially prominent in pursuing the impact of the learned laws on the political thought of the Middle Ages,5 a topic made prominent in the United Kingdom in the second half of the twentieth century by Walter Ullmann.6 The field in which the most important and profound work has been done, however, is that of canon law. Charles Donahue in 1984 explained why “the history of medieval canon law is not written”. In particular, he focused on the lack of study of the application of canon law in the courts.7 In the twenty-five years since, there has been significant scholarship in the English language on this and other aspects of canon law. This has recently been surveyed and assessed by Wolfgang Müller in his introduction to the Festschrift for Kenneth Pennington.8 The issue raised particularly by Donahue has been covered by Dick Helmholz in his study of the spiritual jurisdiction in England.9 Donahue   5 See, e.g., Joseph Canning, The Political Thought of Baldus de Ubaldis [= Cambridge Studies in Medieval Life and Thought VI] (1987).   6 See, e.g., Walter Ullmann, Law and Politics in the Middle Ages (1975). See also K Pennington, The Prince and the Law, 1200-1600: Sovereignty and Rights in the Western Legal Tradition (1993).   7 C Donahue, Why the History of Canon Law is Not Written [= Selden Society Lecture 1984] (1986).   8 W P Müller, “Introduction: Medieval Church Law as a Field of Historical Inquiry” in W P Müller and M E Sommar (eds), Medieval Church Law and the Origins of the Western Legal Tradition: A Tribute to Kenneth Pennington (2006) 1.   9 R H Helmholz, The Canon Law and Ecclesiastical Jurisdiction from 597 to the 1640s [= The Oxford History of the Laws of England I] (2004).

introduction

3

has himself edited volumes dealing with court.10 One may suspect that the emigration of Stephan Kuttner to the USA has played a prominent part in the development of such studies. If canon-law scholarship is strong – indeed flourishing in the English language – the same is not true (at least to the same extent) for the study of medieval Roman and learned feudal law. Much of what has been published in English has been written by Continental scholars. Examples would include Kees Bezemer’s studies of Jacques de Revigny and Pierre de Belleperche, and Tammo Wallinga’s edition of the Casus Codicis of Wilhelmus de Cabriano.11 To these could be added the published proceedings of the Danish Carlsberg Academy conferences on medieval legal history. These, however, are studies very much aimed at a specialist audience coming from a scholarly tradition going back to, for example, Emil Seckel in Germany and E M Meijers in the Netherlands. They can pose difficulties for the novice. The aims of this volume are therefore twofold: first, to flesh out the grand narrative with detailed studies covering the learned laws – Civil, canon and feudal; and secondly, to provide greater access to the sources and conventions of this field of study. The volume opens with Harry Dondorp and Eltjo Schrage’s introduction to the sources and editions of medieval learned law. As well as guidance on the texts, this provides advice on how to approach the study in more general terms. This is of profound importance. There is no comparable account available in the English language. It draws on and updates an earlier introductory booklet which appeared in Dutch and German under the title Utrumque Ius.12 The next two chapters focus in broad terms on issues of methodology. Kees Bezemer explores how and why the medieval jurists developed regulae, and assesses the central importance of the term in medieval legal thought. The broad sweep of Jim Gordley’s chapter allows the reader to appreciate 10 C Donahue (ed), The Records of the Medieval Ecclesiastical Courts: Reports of the Working Group on Church Courts Records [= Comparative Studies in Continental and Anglo-American Legal History VI and VII] (1989-1994). 11 K Bezemer, What Jacques Saw: Thirteenth-Century France through the Eyes of Jacques de Revigny [= Studien zur europäischen Rechtsgeschichte xcix] (1997); K Bezemer, Pierre de Belleperche: Portrait of a Legal Puritan [= Studien zur europäischen Rechtsgeschichte CXCIV] (2005); T Wallinga, The Casus Codicis of Wilhelmus de Cabriano [= Studien zur europäischen Rechtsgeschichte CLXXXII] (2005). 12 E J H Schrage and J H Dondorp, Utrumque ius: Een inleiding tot de studie van de bronnen van het middeleeuwse geleerde recht [= Juridische Reeks Vrije Universiteit 4] (1987); E J H Schrage and J H Dondorp, Utrumque ius: eine Einführung in das Studium der Quellen des mittelalterlichen gelehrten Rechts [= Schriften zur Europäischen Rechts- und Verfassungsgeschichte VIII] (1992).

4

the creation of the ius commune

the different methods that medieval jurists used to understand and interpret the texts. Sometimes, their deductive approach can seem modern; but he identifies a very different sensibility in the way they linked texts together to form a pattern. The remainder of the chapters fall into four groups. Laurent Waelkens provides a challenging account of how the medieval scholars created a law reflecting their Judeo-Christian ideas of family and marriage out of Roman legal texts which were founded on uniquely Roman ideas of marriage and the family. He shows how important medieval theology and canon law were in this process. A second group is constituted by the chapters of Thomas Rüfner, Magnus Ryan, and Paul du Plessis. Rüfner analyses Bartolus’ definition of ownership in context showing the anachronism of many modern interpretations of it. He makes it clear that, properly understood, Bartolus’ definition is perfectly compatible with feudal realities and the medieval concept of dominium utile. He concludes that the path of medieval scholarship was from regula to casus. In an exploration of succession to fiefs, Ryan explores the way in which the ius commune feudorum became recognised and discussed as a significant category of its own leading the glossators to develop the concept of the natura feudi. Du Plessis analyses the problems the medieval jurists faced in dealing with the inconsistencies of the Roman texts in discussing the law of pignus and hypothec. The chapters of Jan Hallebeek and Wolfgang Ernst focus on commercial issues. Hallebeek, in a substantial piece, discusses the liability of the ignorant seller for latent defects in the object sold. The differing remedies in the Roman texts allowed the medieval jurists to develop contrasting solutions to complex intellectual problems as they weighed up alternate advantages and disadvantages. The problems posed by the Roman texts’ references to money and payment are the subject of Ernst’s chapter. The glossators operated in a world of metal currency and bullion where currency could be devalued through reduction of its silver content. They used a variety of Roman- and canon-law texts to resolve the problems posed. The fourth group examines the learned procedure of the ius commune and the actions of the lawyers needed to operate it. Through a discussion of citation in the ius commune, Dick Helmholz shows how the jurists drew on Roman- and canon-law texts to develop a working system. Securing payment of fees for legal work done often proves tricky for lawyers. Jim Brundage’s elegant chapter shows how the medieval lawyers negotiated the quicksand to develop actions for payment of fees based on Roman legal foundations.

introduction

5

In 1938, Hermann Kantorowicz observed that: “There is scarcely any type of legal literature, any legal method, any legal doctrine or concept, that was not created or foreshadowed in the successive medieval law schools.” He also pointed out that the school of glossators, “though the most famous, [is] the least known and the most difficult to know, for in this case, and in this case alone, many of its most valuable works are still hidden in the manuscripts”.13 As Dondorp and Schrage show, modern technology is making manuscripts much easier to access; but although great progress has been made since 1938, much remains to be done. The chapters in this book show the rich possibilities of detailed study. Further development of our understanding of this central area of European legal history will be dependent on an iterative process as focused, detailed research refines grand narratives and is in turn refined by them.

13 H Kantorowicz, Studies in the Glossators of the Roman Law (1938) 1.

1  The Sources of Medieval Learned Law Harry Dondorp and Eltjo J H Schrage* A. PROLOGUE (1) Websites (2) New bibliographical tools (3) A short guide B. THE TEXT OF THE CORPUS IURIS CIVILIS (1) Digest (2) Codex Justinianus (3) Justinian’s novellae (4) Volumen (5) References to Roman law texts C. TEACHING AND WRITING (1) Glossae (2) Disputationes (3) Apparatus (4) Summae (5) Commentum and lectura (6) Repetitio (7) Consilia D. ROMAN AND CANON LAW E. THE CORPUS IURIS CANONICI (1) Gratian’s Decretum (a) Palea (b) System of citation (c) Recensions (d) Roman law in the Decretum Gratiani (2) Early glosses to the Decretum (3) Early summae

* We would like to thank our friends and colleagues (in alphabetical order) Kees Bezemer (Leiden), Gero Dolezalek (Aberdeen) and Olivia Robinson (Glasgow) for their numerous valuable remarks. Olivia Robinson even took upon herself the tedious task of correcting the English.

7

8

the creation of the ius commune

(4)  Ius novum: decretals  (a) System of citation  (5)  Compilationes antique  (a) System of citation  (6)  Liber Extra (a) System of citation  (7)  Liber Sextus, Clementinae, and Extravagantes  (a) System of citation  F.  PROCEDURAL LAW  G.  FEUDAL LAW  H.  A STRATEGY FOR RESEARCH  I.   CONCLUSION  A. PROLOGUE More than twenty years have passed since we published in Dutch an introduction  to  the  foresta selvatica  of  the  medieval  learned  law.  A  German  edition  followed  in  1992,1  and  the  book  is  still  in  print.  During  the  years  that have passed, however, an important number of new developments have  taken place, speci fically the expansion of the number and quality of the bibliographical  tools  available.  This  expansion  is  due  partly  to  the  appearance  of a great number of new important books and partly to the growth of the  internet.  We  will  discuss  these  bibliographical  tools  more  fully,  but  before  doing so we will mention a few useful websites. (1) Websites There exist nowadays a number of very useful websites. For Roman law by  far the best portal is www.iuscivile.com, maintained by the Glasgow scholar  Ernest Metz ger. It not only includes a directory of historians of ancient law  (which is useful in itself), but also a unique collection of links to sources of  Roman law (in the proper sense of the word: le droit romain romain) available on-line, and a list of emendations to the Watson-edited translation of  the Digest. The most comprehensive set of on-line Roman legal texts is maintained by  Alexandr Koptev, partly on his Russian Roman law-library site, partly on the  now more extensive Grenoble site.2 These include most of the Justinianic and    1  E J H Schrage and J H Dondorp, Utrumque Ius. Eine Einführung in das Studium der Quellen des mittelalterlichen gelehrten Rechts [= Schriften zur Europäischen Rechts- und Verfassungsgeschichte VIII] (1992).   2  www.ancientrome.ru/ius/library or: http://web.upmf-grenoble.fr/Haiti/Cours/Ak/index.htm

the sources of medieval learned law

9

pre-Justinianic texts.3 For the student of medieval ius commune, however, a  caveat is necessary, which will be discussed more fully below. It is not selfevident (or rather, it is unlikely) that these modern editions of the texts are  identical with the texts used by the medieval scholars. Texts, commentaries and the like of the glossators are not to be found on  these web sites. References to older editions of law books are indeed, generally speaking, very rare. Moreover, they also require specialised knowledge of  the field. The personal website of Otto Vervaart, because it contains several  pages  devoted  to  both  medieval  Roman  and  canon  law,  is,  however,  very  helpful, particularly for beginners.4 For medieval canon law the website of the Kuttner Institute, maintained  by Jörg Müller, should be mentioned.5 The website of Kenneth Pennington  not only gives access to many of his own publications, but also a “Guide to  Medieval Canon Law Jurists and Collections” up to 1500.6 (2) New bibliographical tools During  the  past  two  decades  many  new  bibliographical  tools  have  seen  the  light,  some  in  print,  some  on  the  internet.  The  Bulletin of Medieval Canon Law seems to have ceded its place to the Rivista internazionale di diritto comune, but  the  Tijdschrift voor rechtsgeschiedenis continues  to  publish valuable bibliographic al surveys. At www.bautz.de/bbkl the German  Kirchenlexikon gives detailed information about the life and works of many  canonists  and  legists;  hence,  it  can  be  used  to  supplement  the  bio-bibliographical  Guide  of  Pen nington  mentioned  above.  Hermann  Lange  and  Maximiliane  Kriechbaum  have  published  Römisches Recht im Mittelalter,  their magnum opus in two volumes.7 The book provides many prosographical  details  and  bibliographies  on  the  glossators  (vol  1)  and  commentators  (vol 2), and is indispensable to any modern legal historian. When confronted  with the richness of data contained in these two volumes, the reader should  nonetheless take care to retain a critical sense. Verification of the data may  be necesssary. Additional material concerning French and Spanish jurists can    3 Apart from the late antique codifications, Koptev covers republican material such as the Twelve  Tables and Cicero’s De Legibus as well as major surviving legal works from the second to the sixth  century ce. These works are of very limited use for the Middle Ages, but the Grenoble site also  offers some trans lations in English, among which S P Scott, The Civil Law Including ... the Enactments of Justinian and the Constitutions of Leo, 17 vols in 7 (1932; repr 1973), also accessible via  www.constitution.org/sps/sps.htm   4  http://home.hetnet.nl/~otto.vervaart/medieval_law.htm   5  www.kuttner-institute.jura.uni-muenchen.de   6  http//faculty.cua.edu/Pennington   7  Lange, RRM.

10

the creation of the ius commune

be found in the  Dictionnaire historique des juristes français and in Juristas universales.8 Among  these  bibliographical  tools  Douglas  Osler’s  forthcoming  bibliography  of  European  legal  literature  to  1800  deserves  special  mention.  The  project, pursued at the Max Planck Institut für europäische Rechtsgeschichte in  Frankfurt,  aims  to  establish  a  databank  in  which  surviving  copies  of  an  individual printed edition are recorded in a census on a strictly systematic  basis. The catalogues Osler has made of the holdings of an important number  of libraries9 have laid the foundation for the whole project. His forthcoming  survey will show that the works of the Italian jurists had already been printed  many times before 1520. The database will describe these editions in detail  and,  as  a  bibliographical  tool  for  scholars  with  an  interest  in  the  works  concerned, will surpass all hitherto known. In this context the website of the Bibliothèque Nationale in Paris deserves  to  be  mentioned,  since  it  makes  some  80,000  (!)  old  editions  freely  available via the internet: www.gallica.bnf.fr. Some additions can be found in the  Bibliotheca Glossatorum.10 Though the website of Gero Dolezalek at the website of the University of  Leipzig was last updated in 2002, it is, nonetheless, still a useful bibliographical tool.11 It contains an important number of references to manuscripts of  both medieval canon and Roman law, including a link to the list of incipits of  legal literature in manuscript, compiled by Giovanna Murano, and a link to  Dolezalek’s comprehensive cumulative inventory of all manuscripts of canon  and Roman law which have so far been mentioned in catalogues or in legalhistorical publications.12 It will eventually replace the famous Handschriftenverzeichnis and Repertorium;13 and will tell who published what on which  manuscript, and where. That bibliographical file will soon comprise all the  data on manuscripts until now compiled from periodicals such as the Bulletin of Medieval Canon Law, the Archiv für katholisches Kirchenrecht, the Österreichisches Archiv für Kirchenrecht, the Zeitschrift der Savigny-Stiftung für   8 P Arabeyre et al (eds), Dictionnaire historique des juristes français, XIIe-XXe siècle (2007) gives  bio-  and  bibliographical  details  about  1,278  French  jurists  from  the  middle  ages  onwards;  R  Domingo (ed), Juristas universales (2004) I, 239-562 (Juristas Antiguos).   9  Max  Planck  Institut  in  Frankfurt,  Biblioteca  di  Giurisprudenza  of  the  University  of  Florence;  Biblio teca  Volterra  in  the  École  Française  in  Rome;  Robbins  Collection  of  the  University  of  California at Berkeley; and the Van Zyl Collection of the University of Cape Town. 10  See www.labinf-digips.unisi.it/glossatorum 11  See www.uni-leipzig.de/~jurarom 12  Murano’s list of incipits and Dolezalek’s database are made accessable via http://www.uni-leipzig. de/~jurarom/manuscr/index.htm 13  G Dolezalek and H van der Wouw (eds), Verzeichnis der Handschriften zum römischen Recht bis 1600 (1972); S Kuttner, Repertorium der Kanonistik. Prodromus corporis glossarum (1937).

the sources of medieval learned law

11

Rechtsgeschichte (kanonistische Abteilung), and the Proceedings of the International Congress of Medieval Canon Law. Furthermore, important textbooks have been published. Ennio Cortese  produced  a  second  edition  of  his  classic,  Il diritto nella storia medievale,  under  the  amended  title  Le grandi linee della storia giuridica medievale  (2000).  Earlier  Manlio  Bellomo  made  his  work  accessible  to  the  Englishspeaking  reader  as  The Common Legal Past of Europe, 1000-1800 (1995).  Two  introductions  to  medieval  canon  law  have  ap peared.  That  by  Jim  Brundage also contains valuable bibliographical data on the main canonists.14  In 1996 Richard Helmholz published his Spirit of Medieval Canon Law.15 In 2008 the long-awaited History of Medieval Canon Law in the Classical Period was published, with superb contributions from Peter Landau, Rudolf  Weigand, Charles Duggan and Kenneth Pennington on Gratian’s Decretum  and the commentaries thereto, the early decretal collections, the compilationes antique and the works of the decretalists up to 1234.16 (3) A short guide Despite the existence of these aids, the editors of this volume expressed the  view  that  a  short  introduction  to  the  sources  of  the  medieval  learned  law  might be useful for the legal historian who is making first steps in this field.  Such a legal historian will need answers to questions as to which texts should  be consulted and how to find details about their origins and their connections.  What  specific  knowledge  will  be  needed  to  interpret  the  medieval  texts? Which traps will need to be avoided? This chapter provides instructions on where to find the medieval source material, which kind of source to  select and to use, and for which purpose. We  have  chosen  a  practical  approach  that  also  outlines  the  relationship  between  this  survey  and  the  older  literature.  It  goes  without  saying  that  what is contained in the older literature retains its value and its study should  not  be  omitted.  We  thus  have  made  extensive  use  of  the  immensely  rich  documentation and information enclosed in the various contributions to the  first volume (“Mittelalter (1100-1500)”) of Helmut Coing’s Handbuch,17 but  we will not repeat the basic information it contains in the contributions by  Peter Weimar on the works of the glossators, by Norbert Horn on the works  14 J A Brundage, Medieval Canon Law, 3rd impr edn (1997). 15  R H Helmholz, The Spirit of Classical Canon Law (1996). 16  W  Hartmann  and  K  Penning ton  (eds),  The History of Medieval Canon Law in the Classical Period, 1140-1234: From Gratian to the Decretals of Pope Gregory IX (2008). 17  Coing, Handbuch.

12

the creation of the ius commune

of the  commentators, and by K W Nörr on the works on legal procedure. With regard to the text of law books, modern scholars face a specific problem:  they have to make sure that they are dealing with the same text as the medieval  scholars had on their desks. We will go into further detail when dealing with  the various parts of the Corpus iuris civilis and the Corpus iuris canonici. Similar  difficulties  are  on  hand  when  it  comes  to  Latin  dictionaries.  Heumann-Seckel18 is the product of nineteenth-century scholarship and stands  in the German pandectistic tradition. Useful as it is, it might sometimes miss  the point as far as medieval learning is concerned. There exist a number of  very early vocabularies, which are very useful. One of the oldest is the Vocabularius utriusque iuris. The first printed edition was published in 1474, but its  content is based on much earlier material. It went through more than seventy  editions in the next 150 years. The Vocabularius was reputedly compiled by a  jurist of Erfurt University, named Jodocus, who signed some manuscripts of  the text.19 More easily found – because a reprint of the Venice 1571 edition  ap peared in 1971 – is Albericus de Rosate (1290-1360), Dictionarium juris (1st edn, Bologna, 1481). Also useful are the humanist vocabularies of Jacob  Spiegel,20 Calvinus (Johan Kahl),21 and Brissonius (Barnabe Brisson).22 B. THE TEXT OF THE CORPUS IURIS CIVILIS The Corpus iuris civilis, as the medieval jurists knew and read it, differed  in  many  respects  –  division,  content,  textual  tradition  –  from  Justinian’s  legislation  as  found  in  the  editions  used  today.  The  latter  are  based  upon  the  Krüger-Mommsen  edition,  a  nineteenth-century  reconstruction  of  the  original Justinianic text.23 This edition is easily accessible on the internet as  a fully-searchable text.24 There is a recent English translation of the Digest  edited by Alan Watson, and two translations of the Institutes: one by J A C  18 Heumanns Handlexikon zu den Quellen des römischen Rechts neubearbeit von E Seckel (1914). 19 For ample data about the Vocabularius and related legal encyclopedias of the Middle Ages, see  E Seckel (ed), Beiträge zur Geschichte beider Rechte im Mittelalter (1898). 20  J Spiegel, Lexicon iuris civilis, 1st edn (1538). 21  J  Kahl  (Calvinus),  Lexicon juridicum: Juris Caesarei simul et Canonici: feudalis item, civilis, criminalis, theoretici, ac practici, 1st edn (1600). 22  B Brisson (Brissonius), De verborum quae ad ius civile pertinent significatione, 1st edn (1557). 23  P Krüger, Th Mommsen, R Schöll and G Kroll (eds), Corpus iuris ciuilis, 3 vols (1872-1895);  there are many re prints, the most recent being, 2000-2005. The merits (and deficiences) of this  edition are discussed in F Wieacker, “Mommsens digestorum editio maior. Aspekte und Aporien”,  in E Spagnesi (ed), Le Pandette di Giustiniano. Storia e fortuna di un codice illustre. Due giornate di studio, Firenze 23-24 giugno 1983 (1986) 199-214. 24  www.upmf-grenoble.fr/Haiti/Cours/Ak/corpjurciv.htm (also as pdf); www.archive.org provides a  reprographical edition of the 1872-1895 edn.

the sources of medieval learned law

13

Thomas,  the  other  by  Peter  Birks  and  Grant  McLeod.25  S  P  Scott’s  1932  translation of Roman law texts is faulty in many ways, but it contains the only  printed English translation of the Code and Novels and is available on the  internet.26 Good translations into English of the Code and Novels by F H  Blume, as well as of the rest of the Corpus iuris civilis, are available on-line  at the website of the University of Wyoming’s College of Law.27 (1) Digest With regard to the text of the Digest the central question is the relationship  between the Florentina and the vulgata–tradition. The Littera Florentina (or  Codex Florentinus) has since the sixteenth century been the name given to  a manuscript in two volumes, 907 leaves in total, which were written in the  Byzantine-Ravenna uncials, characteristic of Constantinople, Alexandria and  the Levant.28 Recent research dates the manuscript to between 533, when  the Digest was officially promulgated, and 557.29 It is noted in a margin of  the  manuscript  that  in  1135  it  was  robbed  from  Amalfi  –  which  was  then  still part of the Byzantine territory – by the citizens of Pisa: hence its older  name, Littera Pisana. There it was looted and brought to Florence after the  war  of  1406,  where  it  is  still  kept  in  the  Biblioteca  Medicea  Laurenziana.  Nowadays  two  reprographical  editions  of  the  manuscript  exist.30  It  is  the  oldest manuscript of the Digest known; the humanists therefore considered it  the best, and editions from the seventeenth century onwards are accordingly  based mainly upon it,31 as indeed is the modern Krüger-Mommsen edition. 25 P Krüger and Th Mommsen (eds), trans A Watson (ed), The Digest of Justinian, 4 vols (1985); J  A C Thomas, The Institutes of Justinian, Text, Translation and Commentary (1975); Justinian’s Institutes. Translated with an Introduction by P Birks and G McLeod (1987). 26  Scott, Civil Law (n 3). 27  http://uwacadweb.uwyo.edu/blume&justinian/default.asp 28  See A Belloni, “Un’ipotesi per le Pandette fiorentine”, in V Colli and E Conte (eds), Iuris historie:  Liber amicorum Gero Dolezalek (2008) 1. 29  M Cavallo, “Libri e scritture del diritto nell’età di Giustiniano” (1987) 15 Index 103; B Stolte,  “The Partes of the Digest in the Codex Florentinus” (1984) 1 Subseciva Groningana 69. D Osler,  “Flash light on the Florentine” (1984) 3 Rechtshistorisches Journal 18 provides critique. See also  W Kaiser, “Zur Herkunft des Codex Florentinus. Zugleich zur Florentiner Digestenhandschrift  als Erkentnisquelle für die Re dak tion der Digesten”, in A Schmidt et al (eds), Sachsen im Spiegel des Rechts (2001) 39. 30  Iustiniani Augusti Digestorum seu Pandectarum codex Florentinus olim Pisanus phototypice expressus (1902-1910); A Corbino and B Santalucia (eds), Justiniani Augusti Pandectarum Codex Florentinus, 2 vols (1988). 31  For  example,  the  Torelli  edition  (1553),  the  unfinished  Brenkman  edition  and  the  proposed  Cunningham edition. See G C J J van den Bergh and B H Stolte, “The Unfinished Digest Edition  of  Henrik  Brenkman”  (1977)  45  TvR  227;  J  W  Cairns,  “Alexander  Cunningham’s  Proposed  Edition of the Digest. An Episode in the History of the Dutch Elegant School of Roman Law  (Part  II)”  (2001)  69  TvR  307  at  340-347;  H  E  Troje,  “Ubi  in  libro  Florentino  duae  lectiones 

14

the creation of the ius commune

The  first  generation  of  glossators  already  knew  of  the  existence  of  the  Florentina, as the textual variations they mention prove. It is speculated that  a copy of the Florentina saw the light in the course of the twelfth century.  This (now lost) copy is generally called the Codex secundus, which laid the  foundat ion  of  a  very  rich  tradition  of  manuscripts  of  the  Digest,  generally  referred  to  as  the  Littera Bononiensis or  Vulgata.  How ever,  no  real  facts  are known. It may well be that non-Florentine readings derive from several  other old manuscripts, each comprising only parts or fragments of the Digest.  This tradition is of Bolognese origin and was the text the glossators and their  successors taught and commented upon. It is trans mitted in more than 1,000  manuscripts (with parts or fragments of the Digest) and in numerous printed  editions of the sixteenth and seventeenth centuries. The  medieval  system  of  reproduction  of  these  manuscripts  was  strictly  re gulated  and  stood  under  university  control.32  Nevertheless,  within  the  tradition of the vulgate textual variations are far from rare. Specific features  which distinguish this tradit ion from the Florentina immediately stand out:  for example, the vulgata lacks Greek quotations and replaces them sometimes  with a translation from the hand of Burgundio of Pisa (d 1193),33 but mostly  by the words “Graeca, non leguntur”. The  Littera Bononiensis or  vulgata is  the  text  which  is  found  in  the  manuscript and earliest printed editions of the Corpus iuris civilis.34 There  is  no  modern  critical  edition.35  The  most  accessible  older  prints  are  those  published by Hugo à Porta in Lyon between 1538 and 1627. Earlier editions  are printed in a gothic font, which those who are unfamiliar with its forms can  inveniuntur ... Zur Geschichte der Digesteneditionen (16.–19. Jh.)” (2004) 72 TvR 61; H E Troje,  Graeca leguntur: Die Aneignung des byzantinischen Rechts und die Entstehung eines humanistischen Corpus Iuris Civilis in der Jurisprudenz des 16. Jahrhunderts  [= Forschungen  zur  neueren Privatrechtsgeschichte XVIII] (1971). 32  F P W Soetermeer, De pecia in juridische handschriften (1990). His Leiden dissertation is translated into Italian and German: Utrumque jus in peciis. Aspetti della produzione libraria a Bologna fra Due e Trecento [=  Orbis  Academicus  VII]  (1997);  Utrumque ius in peciis. Die Produktion juristischer Bücher an italienischen und französischen Universitäten des 13. und 14. Jahrhunderts [= Ius Commune Sonderhefte CL] (2002); there are additions in G Murano, Opere diffuse per exemplar e pecia (2005). On related problems, see F P W Soetermeer, Livres et juristes au moyen âge [= Bibliotheca eruditorum XXVI] (1999). 33  P Classen, Burgundio von Pisa: Richter – Gesandter – Übersetzer [= Sitzungsberichte der Heidelberger Akademie IV] (1974) 39-50. 34  For a chronological survey of the printed editions, see E Spangenberg, Einleitung in das römischJustinianeische Rechtsbuch oder Corpus Iuris Civilis romani (1817; repr 1970) 650-926. His work  will be superseded by the forthcoming bibliography of Osler. 35  There exists an older critical edition of the first four titles of the Digest, with the gloss, by J G  Claussen, Denuo edendae Accursianae glossae specimen (1828 or 1829). On the website of the  University of Saarbrücken there is an edition of D 1.1.1 and D 12.1.1 with the gloss (maintained by  Thomas Rüfner): http://archiv.jura.uni-saarland.de/Rechtsgeschichte/Ius.Romanum/english.html

the sources of medieval learned law

15

find difficult to read. Furthermore, the texts of the early editions are full of  special abbreviations. Consequently these early editions are rather difficult to  use, as consultation of the 1969 reprint of the 1488 edition with the Accursian  gloss will demonstrate.36 There is, however, a tool to decipher these abbreviations – the famous lexicon by Adriano Cappelli, nowadays accessible through  the internet.37 The seventeenth-century editions are much more accessible  than the older ones. The 1627 edition is the last printed edition to include  the complete Accursian gloss.38 Several reprints of it have been published,  most  recently  in  Osnabruck  (1965-1966)  and  Frankfurt  (2000-2005).  The  sixth volume of this edition is particularly important as it contains Dayoz’s  index to the Corpus and gloss. The easiest way to get a reasonably reliable impression of the variations  between the Florentina and the Vulgata is by consulting the critical apparatus  of the Gebauer-Spangenberg edition, which stands in the humanistic tradition.39 Humanism placed considerable emphasis on the philological aspects  of the text. A specific feature of the medieval manuscripts and early printed editions  is the partition into three parts of the Digest: Digestum Vetus (D 1.1-D 24.2),  Infortiatum (D 24.3-D. 38.17), and Digestum Novum (D 39.1-D 50.17). In  early manuscripts the Infortiatum ends abruptly in D 35.2.82; the rest (called  the Tres Partes) was then added to the Digestum Novum. The text and its  divisions had reached their final form by the end of the thirteenth century.40 (2) Codex Justinianus In  the  medieval  editions  of  the  Corpus iuris  in  manuscript,  the  first  nine  books  of  the  Code  were  published  as  one  volume.  The  last  three  books  36 In CGIC VII-IX. 37 www.hist.msu.ru/Departments/Medieval/Capelli/index.html  gives  a  reprograhical  copy  of  A  Capelli, Lexicon abbreviaturarum: dizionario di abbreviature latine ed italiane, 6th  edn  (1929;  many repr). Additions in A Pelzer, Abréviations latines médiévales, supplément au ‘Dizionario di abbreviature latine ed italiane’ de Adriano Cappelli, 2nd edn (1966; repr 1982, 1995, 2002). 38  Regarding the text of the gloss, however, the 1627 edition has some disadvantages, as the ordinary  gloss is printed with later additions. While additions bearing the name of Bartolus or Cujacius are  easily recognised as not belonging to the gloss, some of the additions are much less obvious. 39  Corpus Iuris Civilis, ed G Ch Gebauer and G A Spangenberg I (Institutiones et Digesta) (1776);  ed G A Spangenberg II (Codex, Novellae et Consuetudines Feudorum) (1797). Useful additional  remarks are found in P Pescani, “Origine delle lezioni della Litera Bononiensis superiori a quelle  della Litera Florentina” (1982) 85 BIDR 283. For criticism of the quality of the apparatus criticus,  see Van den Bergh and Stolte, “Unfinished Digest Edition” (n 31) 244. 40  J Q Whitman, “A Note on the Medieval Division of the Digest” (1991) 59 TvR 269; J A C J van de  Wouw, “Zur Textgeschichte des Infortiatum und zu seiner Glossierung durch die frühen Bologneser Glossatoren” (1984) 11 Ius Commune 231; W P Müller, “The Recovery of Justinian’s Digest  in the Middle Ages” (1990) 20 BMCL 1.

16

the creation of the ius commune

contained public law and were included in another volume (Volumen) under  the title, Tres Libri Codicis. The  problem  of  the  relationship  between  the  Littera vulgata and  the  Florentina is perennial in studies involving the Digest; there is no equivalent  for those of Justinian’s Code. Promulgated in 534, the Code re mained well  known,  even  in  the  West  after  the  fall  of  that  part  of  the  Empire.  Copies  were already produced as early as the sixth and seventh centuries, though  sometimes  in  abbreviated  form.  The  subscriptiones,  which  gave  the  origin  and date of the various constitutions, were omitted; so too were the constitutions in Greek. From the twelfth century onwards the glossators added abbreviat ions of  Justinian’s novellae, known as autenticae, on the same subject in the margin  of the Code.41 In time they became inserted in the text. The collection itself,  called the Authenticum, from which they took the autenticae, was included in  the fifth volume of the Corpus iuris civilis, the Volumen. Three enactments  of the medieval Holy Roman emperors were also inserted: Habita  of 1155  was inserted after C 4.13.5;42 Sacramenta puberum, also of 1155, was inserted  after C 2.27 (28).1;43 and Ad decus of 1220 – divided into eleven parts – was  inserted at various places.44 Since Paul Krüger – with his aims of approaching  the original text as far as possible – eliminated the autenticae, it is necessary  to consult an older printed edition than the Krüger-Mommsen to read the  text of the autentica that the medieval jurists had on their desk. (3) Justinian’s novellae There  was  no  official  and  definitive  collection  of  the  legislation  Justinian  promulgated  after  534.  Only  a  number  of  private  collections  existed.  One  of them, known as the Authenticum from the twelfth century onwards, was  41 An analytical description of all known manuscripts of Justinian’s Code is given by G Dolezalek,  Repertorium manuscriptorum veterum Codicis Iustiniani, 2 vols [= Ius Commune Sonderhefte  XXIII] (1985). The books also contain a critical edition of the glosses to C 1 and C 5.1 and of the  lecturae and commenta on C 2.1. See further on the Code, C Tort-Martorell, Tradición textual del Codex Iustinianus. Un estudio del libro 2 [= Ius Commune Sonderhefte XLV] (1989); on the  Tres Libri, see n 49 below. 42  Edition  in  W  Stelzer,  “Zum  Scholarenprivileg  Friedrich  Barbarossas  (Authentica  Habita),  mit  einer Edition der ältesten Überlieferung nach Hs. Harvard Law School Library 64” (1978) 34  Deutsches Archiv für Erforschung des Mittelalters 123. There is further literature in C Arnold  (ed), Bibliographie zur Geschichte Kaiser Friedrichs II. und der letzten Staufer [= MGH Hilfsmittel VIII] (1986); and there are additions in F Martino, Frederico II: il legislatore e gli interpreti  (1989). 43  Edition in L Sorrenti, “L’Autentica ‘Sacramenta puberum’ nell’esegesi dei dottori bolognesi del  Duecento: Guizzardino e Iacopo Baldovini” (1991) 2 RIDC 69. On its content, see J J Hallebeek,  “A Commentary of Azo upon Authentica Sacramenta puberum” (1992) 60 TvR 289. 44  The constitution itself found a place in the Libri Feudorum.

the sources of medieval learned law

17

used  by  the  glossators.  This  collection  consisted  initially  of  134  novellae (Greek  novellae in  translation),  but  the  glossators  omitted  a  quarter  of  them. Following the example of the nine books of the Code, they divided  the  remaining  97 into  nine  parts,  collationes,  in  order  to  facilitate  quotations.  This  collection  is  published  in  another  volume,  the  Volumen,  of  the  manuscript and early printed editions of the Corpus iuris civilis. From this  collection, known as the Autenticum, the glossators inserted abstracts, called  autenticae, in the Code, initially in the margin.45 Another collection, the Epitome Iuliani, was also known to the glossators  and  canonists,  but  became  obsolete  in  the  fourteenth  century.  Around  50  re ferences to texts of this collection are found in the ordinary gloss.46 (4) Volumen Bound  together  in  a  separate  volume  of  the  manuscript  and  early  printed  editions of the Corpus iuris civilis are the Institutes, Authenticum, Tres Libri (the  last  three  books  of  the  Code)  and  (from  the  middle  of  the  thirteenth  century) often also the Libri Feudorum. The original text of the Institutes remained, generally speaking, unaltered  throughout  the  ages.  Its  text  presented  neither  the  problems  arising  from  different textual traditions that affected the text of the Digest nor those from  the epitomisation that the Code and Novels had suffered.47 The  date  of  birth  of  Authenticum remains  uncertain;  but  the  glosses  of  Irnerius  prove  that  from  the  early  days  of  the  school  of  Bologna  a  good  manu script had circulated there. The influence of other collections of Justinian’s novellae (such as the Epitome Juliani) remained very limited. A critical  edition of the Authen ticum was produced by G E Heimbach.48 The Tres Libri Codicis, the tenth to twelfth books of the Code, belonged,  45 See section B(2) above. 46 See W Kaiser, Die Epitome Iuliani. Beiträge zum römischen Recht im frühen Mittelalter und zum byzantinischen Rechtsunterricht (2004). 47 There is a trial edition of the Institutes with the gloss in the Bologna-published Accursii Florentini glossa ad Institutiones Justiniani imperatoris Liber I, Corpus Juris Civilis cum glossa magna Accursii Florentini auspiciis et consilio Regiae Academiae Italicae editum [1939]; pre-Accursian  glosses are published in P Torelli, “Glosse preaccursiani alle Istituzioni” (1942) 15 RSDI 3. For  the older textual tradition of the Institutes, the following older works are indispensable: Th von  Dydynski, Beiträge zur handschriftlichen Überlieferung der justinianischen Rechtsquellen (1891);  E Schrader, Prodromus corporis juris civilis (1823); P Krüger, Geschichte der Quellen und Litteratur des römischen Rechts (1912). 48  G  E  Heimbach  (ed),  Authenticum. Novellarum constitutionum Justiniani versio vulgata,  2  vols  (1851; repr 1974). For bibliography and historiography, see F A Biener, Geschichte der Novellen Justinians (1824; repr 1970). There are use ful additions in G Lanata, “Le Novelle giustinianee e la  traduzione dell’autentico” (1979) 49 Byzantion 239.

18

the creation of the ius commune

according to med ieval opinion, in the Volumen, not in the Code.49 Since these  books cover mostly issues of public law they have an intrinsic unity. The genesis of the Libri Feudorum is complicated.50 The early glossators  sometimes referred to feudal law, but it was the legists of the thirteenth century  who  inserted  feudal  law  into  the  Corpus iuris,  adding  a  collection  of  the  feudal laws of Lombardy to the Authenticum: a letter from bishop Fulbert of  Chartres on the oath of fealty (ca 1020); a compilation of Lombard customary  law compos ed in 1155 by Obertus de Orto; and imperial constitutions. The  last to be incorp orated was the constitution Ad decus of 12 December 1220,  issued at the coronation of Frederick II and confirmed by pope Honorius III. Some  sixteenth-century  editions  of  the  Corpus iuris  with  the  Glossa ordinaria  include  the  Lombarda  with  the  gloss  of  Carolo  de  Tocco.51  The  Lombarda or Leges Longobardorum is a collection of Lombard laws.52 These  were taught in Pavia and Naples (whether or not they were taught in Bologna  is disputed); but their gloss by Carolo de Tocco, completed around 1215, was  known in Bologna. While Accursius sometimes referred to these laws,53 the  medieval jurists never considered them to be part of the Corpus iuris civilis. (5) References to Roman law texts The modern mode of referring to specific sources within the Digest, Institutes  and Code is by (the number of) book, title, lex (constitutio) and paragraph  (of which the first one is called “pr” for “principium”; counting starts from  then on). Numbers appeared for the first time in an edition of 1510. Before  49 On  the  Tres Libri before  Accursius,  see  E  Conte,  Tres Libri Codicis. La ricomparsa del testo e l’esegesi scolastica prima di Accursio  [=  Ius  Commune  Sonderhefte  XLVI]  (1990).  He  first  analyses, on the basis of the eldest manuscripts, the division into two parts, then gives a survey of  the manuscripts of the Tres libri with glosses, and eventually he edits for the first time some 561  glosses of Iacobus, Cyprianus, Rogerius, Placentinus, Pilius, Hugolinus, dating back to the twelfth  and early thirteenth centuries. 50  See section G. below. 51  A critical edition of the Lombarda is lacking; an accessible old one is that of Frankfurt 1613 (by  Lin den berg) or of Venice 1537 with the gloss of Carolus de Tocco: Leges Longobardorum cum argutissimis glosis Caroli de Tocco (1537; repr 1964) (with an introduction by G Astuti). The first  edition of the Corpus iuris to include the Lombarda is that of Lyon, 1656. 52  The  Lombarda  is  an  eleventh-century  merger  of  two  collections  (the  Liber Papiensis and  the  Capitulare Italicum) with laws of the Lombard kings (643-755) and their successors. For editions,  see F Bluhme (ed), Leges longobardorum libri tres [MGH Legum sectio VI] (1868); reprinted  in  G  Pade letti  (ed),  Fontes iuris italici medii aevi  (1877)  I.  On  Lombard  law,  see  G  Diurni,  L’Expositio ad Librum papiensem e la scienza giuridica preirneriana [= Biblioteca della Rivista  di storia del diritto italiano XXIII] (1976); G d’Amelio, “Una fal sa continuità: il tardo diritto longobardo”, in Per Francesco Calasso. Studi degli allievi (1978) 369. 53  For references in the ordinary gloss, see C Lefebvre, “La glose d’Accurse, le décret et les décretales”, in Atti Studi Accursiani (1968) I 257 note 54; LF 1.10, LF 2.22 and LF 2.58 also refer to  the Lombarda.

the sources of medieval learned law

19

that date, references were built up as follows: citations in glosses began (if  referring to texts in other parts of the Corpus iuris) with the letters which  stood for Digest, Institutes or Code. Novellae were not referred to as such.54  Around 1140 the note for the Digest was a capital “D” with a line through the  centre of the letter. By the late-twelfth century jurists and scribes uniformly  used  “ff.”  to  signal  a  reference  to  the  Digest.55  “C.”  for  Code  and  “Inst.”  for Institutes speak for them selves. Numbers being unknown, the name of  the  title  gives  a  necessary  refer ence,  followed  by  the  first  words  (incipit)  of the lex and paragraph. Thus, “ff. loc. (l.) Si merces, Qui fundum” stands  for D 19.2.25.1: “ff.” being read as Digest, “loc.” standing for the title locati conducti, “Si merces” for lex 25, and the paragraph “Qui fundum” being the  first one after the principium. There is also a title in the Code which pertains  to  the  contract  of  lease:  C  4.65.  Consequently  “C. loc. Emptorem” stands  for the lex Emptorem in the title locati conducti and refers to C 4.65.9. For  the last or penultimate paragraph or lex, one sometimes finds in place of the  incipit the abbreviation “ult.” or “fin.”, respectively “pe.”, “penult.” or even  “antepenult.”. Since these words stand in the ablative case their meaning is  ultima/ultimo (viz lege or paragrapho), finali, (ante)penultima/o. If the title  contains a single lex, then the in dication is often una or un(ica). Books thirty  to thirty-two of the Digest bear the same title de legatis et fideicommissis. The medieval reference is de leg. i., ii., or iii., immediately followed by the  incipit of the lex and (if necessary) of the paragraph. References to the Institutes (and the other parts of the Volumen) are similar. Slightly different are  the  references  to  the  autenticae inserted  in  the  Code.  Modern  quotations  consist of “C” (for Code), the number of the book and the title, followed by  “aut. Post” and then the number of the constitution after which the autentica  found its place. If there are more autenticae after the same con stitution, they  are numbered “C 1.3 aut. 3 post 33” (or: “aut. 3 post C 1.3.33”, alternatively “aut. Statuimus post C 1.3.33”)  which  means:  the  third  autentica  inserted  after  the  33rd  constitution  in  the  third  title  of  the  first  book  of  the  Code.  For the convenience of the reader a reference to the novella is added.56 The  medieval reference would be: “aut. Statuimus, C. de episc. et cler”.57 54  References to a “novella” occur sometimes, but seldom. What was meant was  a text from the  Epitome Iuliani. 55  G Dolezalek, “A Series of Papal Decretals from the Late Twelfth Century and its Usefulness for  the Dating of Manuscripts of Roman Law” (2004) 15 RIDC 77 at 79-81. 56  These may be found through the indices of Nicolini and d’Amico or Ochoa and Diez: see n 59  below. 57  H U Kantorowicz, “Die Allegationen im späteren Mittelalter” (1935) 13 Archiv für Urkundenforschung 159; repr in E J H Schrage (ed), Das römische Recht im Mittelalter. [= Wege der  Forschung  DCXXXV] (1987) 71.

20

the creation of the ius commune

When transcribing such a citation (allegatio) the modern reader will write  the whole title with small letters and start the lex and the paragraph with a  capital,  even  when  the  scribe  used  another  type  of  writing.  Abbreviations  of  the  title  remain  unaltered.  Where  the  medieval  scribe  erred  and  wrote  “ff.”, instead of “C.” (or vice versa), the modern reader corrects this in his  transcription.58 The transposition tables in the indices of Nicolini and d’Amico  or Ochoa and Diez are indispensable for the solution of the references of the  medieval jurists.59 C. TEACHiNG AND WRiTiNG The  revival  of  Roman  law  in  Italy  in  the  course  of  the  twelfth  century  is  intimately connected with the rise of universities, and scholarship with that  of teaching.60 The reliability of the stories told by Odofredus de De nariis (d  1265) about Pepo, his successor Irnerius, and the Four Doctors (Quattuor Doctores:  Bulgarus,  Martinus,  Iacobus  and  Hugo)61  and  the  origins  of  the  Bolognese  law  school  will  not  be  further  considered  here.62  The  accuracy  of these thirteenth-century accounts is, of course, questionable. They have  all the hallmarks of the typical stories that students have been telling about  58 See S Kuttner, “Notes on the presentation of text and apparatus in editing works of the Decretists  and Decretalists” (1959) 15 Traditio 452; R Feenstra and G Rossi, Index abbreviationum et de modo citandi fontes [= Ius romanum medii aevi I.1c] (1961). 59  H Nicolini and F S d’Amico, Indices corporis iuris cvilis iuxta vetustiores editiones cum criticis collatas [= Ius romanum medii aevi, Subsidia 1] 4 vols (1964-1970). Slightly shorter but also useful  is X Ochoa and A Diez, Indices titulorum et legum Corporis iuris civilis [= Universa biblioteca  iuris Subsidia II] (1965). For the index to the Authenticum, see [S Kuttner], “Index titulorum  Authentici in novem collationes digesti” (1944) 2 Seminar, Annual extraordinary number of The Jurist 82; repr (1963) 75 Studi Senesi 191. Humanist editions of the Corpus iuris civilis with the  Gloss (the last one, Lyon,1627) also contain indices. 60  An easily accessible introduction is O F Robinson, T D Fergus and W M Gordon, European Legal History. Sources and Institutions, 3rd edn (2005) ch 3. See furthermore P Landau, “The Development of the Law”, in The New Cambridge Medieval History (2004) IV 113. 61  A detailed bibliography of these jurists appears in Lange, RRM I 151-192; a brilliant analysis of  the Bolognese school of law is given by E Cortese, Le grandi linee della storia giuridica medievale (2000) 251-344. Some additions appear in F Roversi-Monaco, “Il’Circolo giuridico di Matilde, da  Bonizone a Irnerio”, in O Capitani (ed), Bologna nel Medioevo [= Storia di Bologna II] (2007)  387-409. 62  Traditionally Pepo’s and Irnerius’ teaching is dated 1075-1125. Recently Winroth suggested that  the teaching of Roman law was in its infancy in the 1130s. Gouron and Pennington opposed this  view. Cf A Winroth, The Making of Gratian’s Decretum (2000) 173; A Winroth, “The Teaching  of Law in the Twelfth Century”, in M Münster-Swendsen (ed), Law and Learning in the Middle Ages  [Proceedings  of  the  Second  Carlsberg  Academy  Conference  on  Medieval  Legal  History  2005] (2006) 41; A Gouron, “Sur un moine bénédictin en avance ou en retard sur son temps”  (2007)  85  RHD 315;  A  Gouron,  “Le  droit  romain  a-t-il  été  la  ‘servante’  du  droit  canonique?”  (2007) 12 Initium: Revista catalana d’història del dret 231; K Pennington, “The Big Bang, Roman  Law in the Early Twelfth Century” (2007) 18 RIDC 43.

the sources of medieval learned law

21

their teachers and their teachers’ teachers from the time of the Talmud to the  present day, as the Harvard professor Charles Donahue has put it.63 After  the  thorough  research  done  by  Weimar  there  is  no  doubt  left  that  the writings of the glossators mirror the structure of their lectures.64 An introductory survey of the various types of literature is given by Bellomo. He also  describes the differences between glossa, lectura, summa, repetitio, and so on.65 In class the libri legales were read. First, before the individual constitutions  of  the  Code  (or  leges of  the  Digest)  were  explained,  the  content  of  the  title  was  introduced  (materia tituli  or  continuatio  tituli).  Thereupon  followed a short de scription of the case discussed in the individual constitution  or  lex  (casus).  Then  the  text  was  read  (lectio),  difficult  words  were  explained  (expositio verborum)  and  texts  of  similar  (similia) or  opposite  content (contraria) were mentioned. The disciplines of trivium (grammatica, dialectica and rhetorica) were helpful for harmonising the texts, which led to  distinctiones and brocardica (generales or notabilia), legal arguments or rules  deduced from the casuistic texts. Eventually the teacher tested the student’s  understanding by asking questions. These related to the interpretation of the  leges in the Corpus iuris (questiones legitime) or to the technique of pleading  (questiones disputate). The latter were discussed out side the lectures: disputations  were  held  on  Friday,  Saturday  or  Sunday.  Hence,  these  questions  were also called questiones veneriales, sabbatine, or dominicales. The works of the early glossators consist mainly of glosses added to the  text of the Corpus iuris, and summaries of Justinian’s Institutes and Code,  which circulated separately. About 170 manuscripts of the Digest, Code and  Institutes with pre-Accursian glosses exist, but there are no printed editions  – except for some trial editions.66 Printed editions of almost all summae exist:  from the anon ymous summa “Iustiniani est in hoc opera” up to the famous  summa aurea of Azo, completed around 1210. (1) Glossae Glossae are short elucidations attached to important words in the text. The  theologians glossed the Bible, the jurists the law books. The first medieval  63 C Donahue, Book Review (2007) 25 Law and History Review 401. 64 P Weimar, “Die legistische Literatur und die Methode des Rechtsunterrichts der Glossatorenzeit” (1969) 2 Ius Commune 43; Coing, Handbuch, 140-146. 65  M Bellomo, The Common Legal Past of Europe (1000-1800), trans. L G Cochrane (1995) 129-148.  See also K Pennington, “The Decretists. The Italian School”, in Hartmann and Pennington (eds),  History of Medieval Canon Law (n 16) 161. On the glossators, see Lange, RRM I 118-150; on the  commentators, see Kriechbaum in Lange, RRM II 355-434. 66  See  the  references  in  notes  2,  6  and  7  above.  S  Caprioli  et  al  (eds),  Glosse preaccursiani alle Istituzioni. Strato azzoniano libro primo (1984).

22

the creation of the ius commune

jurists thus derived their name “glossators” from the dominant form of their  literary  work:  glosses  to  the  individ ual  leges  and  constitutions.  Brief  notes  were  inserted  between  the  lines  and  added  in  the  margin.  Later  on  they  became  copious  enough  to  form  a  framework  within  which  the  original  text was enshrined. The glosses explain difficult words, refer to similar and  conflicting texts, add distinctiones to solve the contrarieties, pose questions,  and note notabilia or brocardica. Distinctions,  notabilia  and  questions  contained  in  the  manuscripts  with  glosses  were  also  collected  and  published  separately.  The  first  collections  date back to the period of the Four Doctors; Azo Porcius (d ca 1230) and  Hugo linus  com posed  the  last.  Two  early  collections  with  distinctions  have  been edited, small parts of those of Pilius, Azo and Hugolin.67 Otto Papiensis  composed a collect ion of notabilia (brocardica), which Azo later revised and  abridged.68  Many questions of the Four Doctors (Bulgarus, Jacobus, Hugo  and Mar tinus) have been published. The collections not only contain quaestiones legitimae,69 but  some quaestiones disputatae as  well:  five  of  Jacobus  and fourteen of Hugo. They pose a case, give arguments pro and contra, and  propose a solution.

67 Still fundamental is E Seckel, “Distinctiones glossatorum, Studien zur Distinktionenliteratur der  ro ma nistischen Glossatorenschule”, in Festschrift der Berliner Juristischen Fakultät für Ferdinand von Martitz (1911; repr 1956) 277; important additions are in G Pesca tore, Book Review  (1912) 33 ZRG, RA 519. There are editions by G B Palmerio (ed), Scripta anecdota glossatorum  (1892;  repr  1962)  II  139-174;  G  Pescatore  (ed),  Die Distinctionensammlung des Ms. Bonon. Colleg. Hisp. Nr. 73 (1913; repr 1968). 68  On brocardica, see P Weimar, “Brocarda, Brocardica”, in LMA (1980) 707-708; D V Dal brenta,  Brocardica. Un’introduzione allo studio e all’uso dei brocardi. Principì di filosofia forense (2007).  Several printed editions exist of the collection of Otto Papiensis, revised and abridged by Azo  (with the additions of his student Cacciavillanus). The Naples, 1568 edition was reprinted in 1967:  Azonis Brocarda (CGIC IV.3). For the relationship between the work of Otto Papiensis’ and Azo,  see M Schwaibold, “Wer such, der findet” (1985) 4 Rechtshistorisches Journal 202. Whether the  oldest part of this collection – beginning with: dolum per subsequentia purgari – is written by  Pa piensis, is uncertain. See M Schwaibold, Brocardica “Dolum per subsequentia purgari”. Eine englische Sammlung von Argumenten des römischen Rechts aus dem 12. Jhdt. [= Ius Commune  Sonderhefte XXV] (1985). 69  Two  anonymous  twelfth-century  collections  with  quaestiones legitimae are:  H  Fitting  (ed),  Quaestiones de iuris subtilitatibus des Irnerius (1894; repr 1977), and E Seckel (ed), Die summa Vindocinensis, aus dem handschriftlichen Nachlass herausgegeben von E Genzmer (1939). The  first col lection is erroneously attributed to Irnerius by its editor. The second collection is wrongly  characterised as summa. A Padoa Schioppa published sixty questions of Pilius in his “Le quaestiones super Codice di Pillio da Medicina” (1973) 39 SDHI 262.

the sources of medieval learned law

23

(2) Disputationes In the 1190s, using older material,70 Pilius Medicinensis com posed a collection of questions discussed at disputations that was to be used for centuries.71  Of the fifty-nine questions of Azo preserved in manu scripts, forty-one have  been edited by Ernest Landsberg and Annalisa Belloni.72 Roffredus Beneventanus published a collection of quaestiones de facto emergentes around 1220,  which  was  printed  in  the  sixteenth  century  under  the  title  Quaestiones sabbatinae.73 The advantage of his collection, as Roffredus point ed out in his  preface, was that, while Pilius had used fictitious or hypothetical cases, his  questions derived from practice. Disputations  were  a  regular  academic  exercise  in  the  thirteenth  and  four teenth centuries. Statutes prescribed that they were put in writing, and  the  station ers  lent  out  quires  with  questions.74  The  professors  referred  to  them in their lectures as, for example, Jacques de Revigny did.75 Fourteenthcentury  col lections  composed  by  Jacobus  Butrigarius,  Albericus  de  Rosate  and Bartolus were printed in the fifteenth and sixteenth centuries. Many of  the cases posed concern ecclesiastical and feudal matters. Thus, Roffredus’  collection opens with a question concerning the election of a canon to the  chapter of a cathedral church. (3) Apparatus Deciding when glosses have reached the density that justifies their characterisation as an appar  atus is an issue of judgement and terminology. In any  case,  in  recent  times,  apparatus  written  by  the  Four  Doctors  have  been  identified,  such  as,  for  example,  that  of  Bul ga rus  on  the  title  D  50.17  (de regulis iuris). Beckhaus published it in 1856, together with additions made  70 See  U  Nicolini,  Pilii Medicinensis Quaestiones sabbattinae. Saggio di edizione  [=  Pubbl.  della  Facoltà di Giurisprudenza della Università di Modena LXVII] (1946). Parts of the early collections of Bulgarus, Martinus, Hugo and Johannes Bassianus have been edited in Scripta anecdota glossatorum (1892; repr 1962) I 235-266; II 195-209; H Kantorowicz, Studies in the Glossators of Roman Law (1938; repr 1969) 246-263; E Seckel, “Die Quaestiones Vindobonenses des Johannes  Bassianus”  (1935)  55  ZRG, RA  338;  and  A  Belloni,  Le questioni civilistiche del secolo XII. Da Bulgaro a Pilio da Medicina e Azzone [= Ius Commune Sonderhefte XLIII] (1989). 71  The edition of Rome, 1560, has been reprinted: Pilei Medicinensis quaestiones aureae (1967). 72  E Landsberg, Die Quaestiones des Azo (1888); Belloni, Questioni (n 70) 125-172. For a comp arison  with Pilius’s questions, see Belloni, Questioni (n 70) 173-208. 73  The edition Lyon 1500 was reprinted in 1968 (CGIC VI). 74  Still  fundamental  is  H  Kantorowicz,  “The  quaestiones  disputatae  of  the  glos sators”  (1939)  16  TvR 1, repr in H Kantorowicz, Rechtshistorische Schriften (1970). On disputations, see now M  Bellomo, I fatti e il diritto tra le certezze e i dubbi dei giuristi medievali (secoli xiii-xiv) [I libri di  Erice XXVII] (2000). 75  The questions referred to in his works are edited in C H Bezemer, “Les quaestiones disputatae  Orléanaises dans les commentaires de Jacques de Révigny” (1990) 58 TvR 5 at 18-38.

24

the creation of the ius commune

by  Placentinus.76 Martinus glossed the Digest (Digestum novum) so densely,  that it became characterised as his apparatus to this particular volume of the  Corpus iuris civilis.77 Glosses of Rogerius and Placentinus have been preserved  in  manu scripts,  but  they  have  left  no  apparatus.  Bulgarus’  pupil  Johannes  Bassianus (d 1197) composed a gloss-appa ratus to the Institutes, Digest and  Code, which his pupil Azo used and abridged. Severino Caprioli has edited  Johannes Bassianus’ commentary to D 50.17, which was thought to have been  lost.78 Of the apparatus on the Institutes, Digest and Code composed before  the Glossa Ordinaria, however, the only one available in print is Hugolinus’  apparatus to the Tres Libri (books ten to twelve of the Code).79 By  far  the  most  important  gloss-apparatus  was  composed  by  Accursius  (1185-1263), a student of Azo (d 1230). His work – in total 96,940 glosses –  covers all parts of the Corpus iuris. It achieved a uniquely important status,  because it was added to the standard manuscripts of the Corpus iuris used  in  the  universities,  1,200  of  which  still  exist.  Numerous  printed  editions  appeared in the sixteenth and seventeenth centuries, of which the Venice,  1487-1489, and Lyon, 1627, editions have been reprinted recently.80 On the  one hand, it is the product of two centuries of teaching, because Accursius  inserted  an  abundance  of  glosses  of  earlier  glossators.  On  the  other  hand,  it  laid  the  foundation  for  future  developments,  since  it  became  in  its  turn  the  basis  of  interpretation  by  later  jurists.  It  immediately  superseded  the  ap paratus  of  Azo  –  because  it  was  a  revision  thereof  according  to  H  H  Jakobs81 – and around 1350 also those of Hugolinus and Jacobus Balduinus.  Many manuscripts contain additions to the gloss. Some of them have been  published separately, namely those of Jacobus de Arena, who taught in Padua,  and of Dinus de Mugello, who taught in Bologna from 1284.82 The casus in  76 F C G Beckaus (ed), Bulgarus ad Digestorum titulum de diversis regulis iuris antiqui commentarius et Placentini ad eum additiones sive exceptiones (1856; repr 1967). On regulae iuris, see P  G Stein, Regulae iuris. From Juristic Rules to Legal Maxims (1966) 132-137. 77  G Dolezalek, “Der Glossenapparat des Martinus zu Digestum novum” (1967) 84 ZRG, RA 245. 78  S Caprioli (ed), “Quem Cuiacius Johanni tribuerat” (1963) 7 Annali di Storia del diritto 133. 79  In the printed editions of Azo’s lectura codicis it supplements the lectures on to C 1-C 9. The  edition Paris, 1577 was reprinted in 1966 (CGIC III). 80  Compared with the Venice edition (repr 1967), the text of the gloss in the Lyon edition (repr  1965-1966 and 2000-2008) is mingled with additions of a later date. For a survey of the printed  editions, see Spangenberg, Einleitung (n 34). 81  H  H  Jakobs,  Magna  Glossa, Textstufen der legistischen glossa ordinaria  [=  Rechts-  und  Staatswissen schaft liche Veröffentlichungen der Görres Gesellschaft NF CXIV] (2006). 82  Jacobus’  Commentarii in universum ius civile  (1541;  repr  1967)  contains  his  ad ditions  to  the  gloss  to  the  Digest  –  with  further  additions  by  later  authors  deriving  from  references  to  later  commentators – and a lectura codicis. Dinus’ additions to the Infortiatum and Digestum Novum  were printed in Lyon 1514 (repr 1971); a few manuscripts contain his additions to the Digestum Vetus. He also wrote the first collection of Consilia, of which several printed editions appeared.

the sources of medieval learned law

25

the printed editions of the Glossa ordinaria are com posed by Accursius’ sons  Francis and William, by Vivianus Tuscus, Guido de Cumis and others. In about forty years Accursius managed to compose a gloss to the whole of  the Corpus iuris, in which hardly any internal contradictions could be found.  In 1515 Antonio Nicelli found only 121.83  Accursius began his work before  1220; it still contains some references to decretals in collections before the  1234 Liber Extra of pope Gregory IX.84 Accursius’ gloss to the Digestum Vetus  first circulated in 1235, but he kept improving and revising his work until his  death. In all parts one finds cross-references to glosses in other parts.85 (4) Summae These discuss the content of a book – such as the Institutes, the Code (C 1–C  9)  –  in  the  order  of  its  titles,  preceded  by  a  general  introduction  (materia  codicis).86 Of the Digest, only certain titles have been summarised. The early summae of the Institutes and Code were written outside Bologna.  The  anonymous  summa,  Justinianus est in hoc opere,  written  in  southern  France,  follows  the  order  of  the  Institutes.87  André  Gouron  has  recently  suggested a French origin for the summa Trecensis,88 which follows the order  of  the  Code,  written  in  1140-1160,  again  certainly  outside  Bo logna.89  The  summa adopts some material from both Bulgarus and Martinus (e.g. Martinus’  distinction  Interesse quandoque  in  VII.31).90  It  is  the  basis  for  Rogerius’  83 Antonio  Nicelli,  De concordia glossarum seu de glossis contrariis et de eorum concordantiis  (1515), also printed in TUJ XVIII 187-211. 84 See  Ch  Lefebvre,  “La  glose  d’Accurse,  le  décret  et  les  décretales”,  in  Atti Studi Accursiani  (1968) I 282-284. He counted 260 references to canon law (in 96,940 glosses). They occur where  questions of criminal law, the law of procedure, marriage, interest and pre scription are discussed. 85  There is a scholarly discussion as to whether Accursius wrote two successive re cens ions of his work,  the first before 1235. In 1939 Nicolini distinguished between them in his critical edition of the  gloss to Inst 1. Cf P Torelli, Accursii Florentini glossa ad Institutiones Justiniani imperatoris Liber I (n d). This does not apply to the Codex, Digestum Vetus and Authenticum, as Feenstra, Dolezalek  and Martini established. Weimar distinguishes two recensions of the gloss to the Libri Feudorum.  See the reference in note 188 below. See also F P W Soetermeer, “L’ordre chronologique des  apparatus  d’Accurse  sur  les  libri  ordinarii”,  in  M  J  Peláez  (ed),  Historia del derecho privado. Trabajos en homenaje a Ferran Valls i Taberner (1989) 28; G Diurni, “La Glossa accursiana: stato  della questione” (1991) 64 RSDI 341; F Mancuso, “Per la datazione e l’ordine di realizzazione  degli apparati accursiani. Due testimonianze” (2000) 112 Studi Senesi 350. 86  Analysis of the development of this genre of medieval legal literature appeares in: L Loschiavo,  Summa Codicis Berolinensis. Studio ed edizione di una composizione ‘a mosaico’ [= Studien zur  europäischen Rechtsgeschichte LXXXIX] (1996). 87  P Legendre (ed), La Summa Iustiniani est in hoc opere [= Ius Commune Sonderhefte II] (1973). 88  A Gouron, “L’auteur e la patrie de la Summa Trecensis” (1984) 12 Ius Commune 1; repr in his  Études sur la diffusion des doctrines juridiques médiévales (1987) III. 89  Because Fitting (erroneously) attributed the Summa Trecensis to Irnerius, the edition is entitled  H Fitting (ed), Summa Codicis des Irnerius (1894; repr 1971). 90  It also adopted Martinus’ Tractatus de iure dotium in V.8. Kantorowicz’s ascription of the title 

26

the creation of the ius commune

unfinished summa codicis (until C 4.58).91 In the 1170s, when he taught in  Montpellier, Placentinus supplemented Rogerius’ summa. He subsequently  wrote a summa institutionum,92 and then revised Rogerius’ work, so that he  would be remembered for his own summa of the Code.93 The early glossators in Bologna had only written on certain topics (summulae): Bulgarus, for  instance, on de iure et facti ignorantia.94 Johannes Bassianus wrote a summa  of  the  Authenticum,  which  in  its  revised  version  (by  Accursius)  is  printed  together with Azo’s Summa Aurea of the Code.95 For the composition of his  summae,  Azo  used  material  drawn  from  his  teacher  Johannes  Bassianus  –  how much is, however, debated. As Azo intended, his Summa of the Code  soon superseded the older ones; its authority equalled that of the Accursian  gloss. Many editions of it were published in the course of the fifteenth and  sixteenth centuries, often in one and the same binding with a summa institutionum, also by Azo, a summa of the Tres Libri (by Placentinus and Pilius),96  and a summa of several titles of the Digest.97 (5) Commentum and lectura These  terms  refer  to  lecture  notes  edited  by  the  professor  himself  or  to  an account of his lectures written by a student. It is difficult to distinguish  be tween the two, but a commentum is also used to describe a commentary  written by someone who did not teach. The emphasis shifted in the course of  the thirteenth century from gloss-apparatus copied in the margins of the text  of the Corpus iuris civilis to commentaries published separately. The later  legists  also  derive  their  name  “commentators”  from  these,  but  the  earliest  commenta or lecturae date back to the twelfth century.98 Lecture  notes  of  Johannes  Bassianus’  reading  (lecturae)  of  the  Instide acquirenda uel amittenda possessione to Martinus has been put in doubt by Gouron. See his  “L’élaboration de la Summa Trecensis”, in V Giuffrè (ed), Sodalitas. Scritti in onore di Antonio Guarino (1984) VIII 3681; repr in Gouron, Études (n 88). 91  G B Palmerio (ed), Scripta anecdota glossatorum, 2nd edn (1914; repr 1962) II 47-233. From C  5.1 on, the anonymous author of the Summa Tubigensis sup plemented his own text (ad C 5.1 de sponsalibus), parts of the Summa Trecensis, and some of the summa codicis of Placentinus. 92  Summa Institutionum (1535; repr 1973) (CGIC I). 93  Placentini Summa Codicis (1536; repr 1962) (with a preface taken from another edition). 94  Edited by Kantorowicz, Studies (n 70) 245-246. 95  Azo, Summa Codicis, many editions. A good one is that printed in Pavia in 1506, Azonis summa super codicem (repr 1966) in CGIC II; or that of Lyon, 1557 (repr in 1968): Azonis summa aurea. 96  Summa C. 10.1-C. 10.37 is written by Placentinus, C 10.38-C 11.40 by Pilius. 97  Parts are additions of Hugolinus. Cf P Weimar, “Zur Entstehung der Azoschen Digestensumme”,  in J A Ankum et al (eds), Satura Roberto Feenstra sexagesima quintum annum aetatis complenti ab alumnis collegis amicis oblata (1985) 371. 98  F P W Soetermeer, “Une catégorie de commentaires peu connue. Les commenta ou lecturae  inédits des précurseurs d’Odofrède” (1991) 2 RIDC 47.

the sources of medieval learned law

27

tutes  and  of  several  titles  of  the  Code  and  Digest  have  been  preserved  in  manu scripts.99 These were written by Nicolaus Furius, one of his students.  In 1557 Contius discovered a manuscript with a report of Azo’s lectures on  the Code, written around 1220 by Alexander de Sancto Aegidio. Together  with  the  gloss-apparatus  of  Hugolinus  to  the  Tres Libri  (the  final  three  books  of  the  Code)  they  appeared  in  print  in  1577  under  the  title  Azonis ad singulas leges xii librorum Codicis Justinianei commentarius.100 Notes of  Accursius’  lectures  have  not  survived.  An  unknown  student  –  or  students,  because the manuscript copies vary from each other – “reported” the lectures  of  Odofredus  de  Denariis  (d  1265),  Accursius’  contemporary  and  rival  in  Bologna. His lectura reportata covers all parts of the Digest and Code. A few  manuscript and printed editions exist. One of them (Lyon, 1550-1552) has  been reprinted, 1967-1969.101 (6) Repetitio In the course of the thirteenth century it became obligatory that the lectures  reached a certain point in the Digest or Code every twelve days (fifteen in  winter),  and  covered  all  the  leges  that  had  been  selected  by  the  professor  and  his  students.  As  a  consequence,  there  was  no  longer  time  to  expound  all  the  texts.  Once  a  week,  however,  from  St  Luke’s  Day  (18  October)  to  Christmas, and from Easter to August, the university organized a repetitio, a  lecture outside the normal curriculum. It was obligatory to give one repetitio  each year on one of the leges of the Digest or Code. The structure differed  little from a lecture, but as such the repetitiones laid the foundation for later  treatises and monographic studies on specific subjects of law. The lectura of  Jacobus de Arena, who taught in Padua, of Jacques de Revigny and Pierre  de Belleperche, who lectured in Orleans, of Guillaume de Cunh, professor  in Toulouse, and of Bartolus and Baldus contain many of their repetitiones,  but not all.102   99 Two early lectures on the Institutes have been published. The first is the In nomine domine  composed in England, ed with trans by F de Zulueta and P Stein, The Teaching of Roman Law in England in 1200 (1990) [= Selden Society Supplementary Ser VIII]. The other, the Lectura Vindobonensis,  is  probably  of  French  provenance;  edn  in  G  B  Palmerio,  Scripta anecdota glossatorum, 2nd edn (1914; repr 1966) I. His attribution to Irnerius is no longer accepted. 100  The Paris edition of 1577 was reprinted in 1966 (CGIC III). 101  In OIR II-V; VI contains the index. 102  Still known are 130 repetitiones of Jacques de Revigny, 125 of Pierre de Belleperche, fifty-eight  of Bar tolus, several of Raimund de Forlivio, and numerous of Baldus. See C H Bezemer, Les répétitions de Jacques de Révigny (1987). Forty-nine repetitones of Pierre de Belleperche are  printed (1571; repr 1968): see note 105 below. See E M Meijers, “L’université d’Orléans au xiii  siècle”, in Meijers, Études III 128 and C Bezemer, Pierre de Belleperche. Portrait of a Legal

28

the creation of the ius commune

Of  the  lectures  Jacques  de  Revigny  gave  between  1260  and  1280,  only  his  lectura codicis  has  been  printed  –  under  the  name  of  Pierre  de  Belleperche103 – and parts of his lectura institutionum.104 The lectures on the Code  of Pierre de Belleperche are only preserved in manuscript. Of his lectures  on  the  Digestum novum,  one  printed  edition  exists:  Frankfurt,  1571.  It  contains a commentary on D 43-46 and 49.105 A “report” written by William  de Brandestone on his lectures on J Inst 4.6 (De actionibus) forms the main  part  of  his  lectura institutionum.106  There  also  exists  a  report  (made  by  an  unknown student) of the lectures Guillaume de Cunh gave in Toulouse in  1315/1316 and 1316/1317.107 This lectura reportata was known to Cinus da  Pistoia in Bologna, and to Rainerius de Forlivio in Pisa. Jacobus Butrigarius’  lectures  –  he  began  to  teach  in  Bologna  around  1307  –  on  the  Digestum Vetus and Code were printed in the sixteenth and seventeenth centuries.108  In his lectures on the Digestum vetus, written about 1325, Cinus da Pistoia  frequently cited the French professors Jacques de Revigny and Guillaume de  Cunh; Pierre de Belleperche is mentioned in the additions made by Rainerius  de Forlivio. The printed editions also contain his commentary on the Code,  written between 1312 and 1314.109 His lectures on the Digestum vetus only 

103 

104 

105  106 

107  108 

109 

Puritan (2005) 171-180. On Bartolus’ repetitiones see S Lepsius, “Bartolus de Saxoferrato”, in  Compendium Auctorum Latinorum Medii Aevi (2004) II.1 101-156. See H Kiefner, “Zur gedruckten Codexlectura des Jacques de Revigny” (1963) 31 TvR 5. The  Paris 1519 edition was reprinted in 1967 and 1968. On the relationship be tween the manuscript  and printed editions, see Bezemer, Répétitions (n 102) 159-163. Wael kens has established that  the lectura super C 8.53 derived from his master Simon de Paris. Cf L Waelkens, La théorie de la coutume chez Jacques de Révigny (1984). For an edition of his lectura super Inst 4.6, see L J Van Soest-Zuurdeeg, La lectura sur le titre De actionibus (Inst. 4.6) de Jacques de Révigny (1989). The Paris 1514, and Lyon 1536 (repr;  1972) editions of the lectura institutionum of Pierre de Belleperche contain Jacques’ lectures  on J Inst 3.14, 3.15, 3.21, 3.23, 3.24, 3.25 and on J Inst 4.3-4.5. See P Weimar, “Die Erstausgabe  der so ge nann ten Lectura Institutionum des Pierre de Belle perche” (1967) 35 TvR 284. Meijers  attributed the lectura Institutionum of Bartolus to Jacques de Revigny, Waelkens to his pupil  Raoul d’Harcourt. See L Waelkens, “La lectura institutionum de Raoul d’Harcourt” (1992) 3  RIDC 79. Reprinted in Bologna in 1968 in OIR X, together with his repetitiones. The first edition was printed in Paris in 1512; the Lyon edition of 1536 was reprinted in 1972.  The lectura covers J Inst 1.1-1.10. Several lectures on J Inst. 3 and 4 were adopted from Jacques  de Revigny (see n 104). Guillelmus de Cugno, Lectura super codice (1513; repr 1968) in OIR VIII. There is no printed  edition of his lectures on the Digestum novum, given in 1315/1316. Both lectures are short in length, covering only parts of the Digestum vetus and the Code. The  first, originally printed in Rome, Jacobi Butrigarii in primam e secundam veteris digesti partem  (1606) was reprinted in 1978 in OIR XIV; the latter, first printed in Paris (1516), was reprinted  in 1973 in OIR XIII. The Frankfurt edition of 1578 was reprinted in 1964. On Cinus and Jacques de Revigny see C  H Bezemer, “Word for Word or Not. On the Track of the Orléans Sources of Cinus’ Lecture on  the Code” (2000) 68 TvR 433.

the sources of medieval learned law

29

cover some titles; he began writing another, very elaborate commentary, but  left it unfinished when he died in 1336.110 About 1340, Albericus de Rosate,  an ad vocate and judge in Bergamo, finished his commentary on the Code –  which he entitled lectura although he never lectured at a university – and  subsequently a commentary on all parts of the Digest.111 Bartolus’ works are  numerous and extensive, including commentaria on all parts of the Digest  and of the Code.112 A commentary on the Institutes has been printed under  his name, but the attri bution is incorrect.113 Baldus de Ubaldis (1327-1400)  wrote  a  volum inous  com mentary  on  the  Code  and  on  the  first  part  of  the  Digest. His com mentaries on the other parts of the Digest (Infortiatum and  Digestum novum)  and  the  Tres Libri  are  short  and  patchy.114  The  printed  editions of Baldus’ work include a commentary on the Institutes, which in  fact is written by Bartolomeus de Novara.115 (7) Consilia The  earliest  glossator  whose  consilia  have  survived  is  Johannes  Bassianus.  His  advice  in  a  feudal  law  case,  published  by  Eduard  Meijers  in  1934,  is  well  known.  Five  more  of  his  consilia  have  recently  been  published  by  Belloni.116 Luigi Chiapelli edited one of Azo from 1205.117 The archives in  Bologna  contain  numerous  consilia  of  Odofredus,  very  few  of  which  have  been published. Of the commentators’ collections, those of Dinus, Oldradus  de Ponte, Bartolus and Baldus were printed in the sixteenth and seventeenth  century.

110 See D Maffei, La lectura super digesto veteri di Cino di Pistoia (1963). 111 The Venice edition of the 1585 edition was reprinted in 1974-1982 in OIR XXI-XXIX. 112 There are numerous printed editions of his commentary on all parts of the Corpus iuris. The  edition  of  Venice,  1525,  was  reprinted  in  1996  (9  vols).  The  attribution  to  Bartolus  of  the  commentary on the Institutes is incorrect (see n 113 below), that of the commentary on the  Authenticum is disputed. Bartolus left his commentary on the Tres Libri unfinished (it ends at  C 11.34; the repetitio on C 12.1.1. is also his). It is supplemented with a commentary by Contes  de Perusio, a contemporary of Baldus. 113  It is disputed whether this commentary is written by Jacques de Revigny or Raoul d’Harcourt.  See  n  104  above.  The  commentary  on  J  Inst  3.21  is  written  by  Guido  de  Cumis.  See  E  M  Meijers, “La théorie des ultramontani concernant la force obligatoire et la force probante des  actes sous seing privé” (1931) 12 TvR 38 at 40 (repr Meijers, Études IV). 114  On the early printed editions, see V Colli, Incunabula operum Baldi de Ubaldis [= Ius Commune  Son derhefte XXVII] (2000). The Venice, 1599 edition of the works of Baldus was reprinted in  2004 (8 vols). 115  See D Maffei, “Bartolomeo de Navarra (d 1408) autore della lectura institutionum attribuita a  Baldo degli Ubaldi” (1990) 63 RSDI 5. 116  Meijers, Études III 273; A Belloni, “Giovanni Bassiano Consulente” (1994) 21 Ius Commune 78. 117  L Chiapelli and L Zdekauer (eds), Un consulto d’Azzone dell’anno 1205 (1888).

30

the creation of the ius commune

D. ROmAN AND CANON LAW The glossators viewed Justinian’s codification as sufficient in itself to solve all  legal problems.118 The canon lawyers, however, though their interests were  mostly ecclesiastical, could not ignore Roman law, for the Church itself laid  claim to be governed by Roman law. Roman law was ap plied in questions for  which no rules of canon law existed. Differences between the two laws were  listed in differentiae legum et canonum.119 Of course, many canons dealt with subjects of religious and family life, not  covered by Roman law. Rules regarding baptism and penitence, for instance,  had  no  counterpart  in  Roman  law.  There  were  also  contradictions.  Thus,  Roman law allowed consensual divorce. Some of these contradictions have  been  decisive  in  the  development  of  private  law.  The  famous  adage  pacta sunt servanda, for instance, derives from a canon of the Council of Carthage  (348). Sometimes, the differences between the leges and canones were less  obvious. An example thereof is the phrasing of an oath, “I promise to you and  through you to St Peter to henceforth follow the teachings of the Church”,  which the canonists considered contrary to privity of contract. How can one  be bound to St Peter, a third party, who is not present at the moment the  contract is concluded? Hence, to find out whether on a specific subject the teaching of the canon  lawyers varied from the glossators, one should not only consult the text of the  Corpus iuris canonici, but also the commentaries of the canonists. For the  twelfth century the first work that should be consulted is the summa of Simon  de Bisignano, of which a searchable full text is available on the internet.120  Thus, especially when dealing with Gratian’s Decretum, the first part of the  Corpus iuris canonici, other texts may prove a better starting point.

118 Lange, RRM I 102-103 gives a detailed survey of the use of Gratian’s Decretum and the decretals  by the various glossators. See also, for the commentators, Lange, RRM II 205-207. 119  Hieronymus  de  Zanetti  (d  1493)  listed  356  differences;  his  Contrarietates inter ius civile et canonicum are printed in TUJ I 197. See, for an extensive list of the differences, J Portener,  Recherches sur les differentiae iuris civilis et canonici au temps du droit classique de l’Eglise  (1946) 136-151. Lange, RRM II 216-218 gives several examples; see also H Mohnhaupt, “Die  Differentienliteratur  als  Ausdruck  eines  metho dischen  Prinzips  frü her  Rechtsvergleichung”,  in B Durand and L Mayali (eds), Excerptiones iuris: Studies in Honor of André Gouron (2000)  439. 120  See www.unifr.ch/cdc/summa_simonis_de.php

the sources of medieval learned law

31

E. THE CORPUS IURIS CanOnICI In 1500 Jean Chappuis published “the body of canon law” – a counterpart of  the Corpus iuris civilis. As collections, however, they vary in many respects.  The  first  part  of  the  Corpus iuris canonici,  Gratian’s  Decretum,  is  quite  different in character. Justinian had aimed at codifying Roman law, and the  compilers tried to avoid internal contradictions. The Decretum, on the other  hand, is a twelfth-century collection of often contradictory ecclesiastical texts  on  legal  issues,  which  Gratian  sought  to  reconcile  through  his  dicta. The  texts date from ancient Rome to the Second Late ran Council of 1139. The  second part of the Corpus iuris canonici consists of the ius novum, twelfth-  and  thirteenth-cen tury  papal  decretals.  Private  and  authorised  collections  followed.  While  the  body  of  civil  law  was  firmly  established  in  the  twelfth  century, that of canon law only attained its final form in the fourteenth century. (1) Gratian’s Decretum In Bologna, at roughly the same time as the Four Doctors taught Roman law,  Gratian composed a textbook of canon law that was to become fundamental  for the teaching of canon law in the Middle Ages. Gratian entitled his work  Concordia discordantium canonum – “The Harmony of Conflicting Canons”.  It  soon  became  generally  known  as  the  Decretum  Gratiani.  Gratian’s  title  indicates that he intended, through his dicta, a reconciliation of the different  canonical  rules  into  an  intellectually  consistent  system.  Thus  his  work  was  more  than  a  compilation  of  the  various  legal  norms  in  church  matters,  retrieved  from  patristic  texts,  papal  letters  and  conciliar  canons,  although  such compilations (made by Ivo of Chartres, Anselm of Lucca and others)  were his primary sources. The most complete critical edition of Gratian’s Decretum appeared in 1879  as the first volume of E Friedberg (ed), Corpus Iuris Canonici (reprinted in  1955).  The  Bayerische  Staatsbibliothek  has  now  put  a  searchable  full  text  on  the  internet.121  The  con cordance  to  Gratian’s  Decretum,  published  in  1990, is also based upon the Fried berg edition.122 This work represents the  vulgate  recension  of  the  Decretum  with  additions  (paleae)  that  were  later  inserted. The edition is based upon a small number (eight) of several hundred  medieval manuscripts. Anders Winroth has recently esta blish ed that four of 

121 At http://mdz.bib-bvb.de/digbib/gratian/text 122 T Reuter and G Silagi (eds), Wortkonkordanz zum Decretum Gratiani [= MGH Hilfsmittel X]  (1990).

32

the creation of the ius commune

these  contain  an  earlier  and  shorter  version  of  Gratian’s  Decretum.123  The  great  majority  of  the  manuscripts,  however,  that  were  used  in  the  schools  of  canon  law  contain  the  second  or  vulgate  recension,  which  includes  the  canons of the Second Lateran Council of 1139, and have a varying number  of additions (paleae) that were inserted circa 1150-1180 by later canonists.  The number of these additions varies. Thus, several manuscripts of French  origin contain paleae that are not found in the printed editions of Gratian’s  Decretum.124  It  is  clear  that  the  transmontane  manuscript-tradition  differs  from that of Bologna. Since the first incunabula appeared in Strasbourg in 1471, the Decretum  has been printed frequently, either on its own or as part of the Corpus iuris canonici. In 1566 pope Pius V appointed a commission to correct and emend  the printed text of the Corpus iuris canonici. The result of their work, the  editio Romana, appeared in 1582 as the official edition of the Roman Catholic  Church.  It  was  reprinted  many  times  and  editions  deviating  from  it  were  excluded from the courts and univers ities of the Catholic church. The first  editor to go back to the manuscript tradition was Just Henning Böhmer, a  German protestant. The four manuscripts he used for his edition, published  in Halle in 1747, are, however, not always reliable. Emil Richter returned in  1839 to the editio Romana for his edition, which is easily available, because it  was reprinted in the Patrologia latina (vol CLXXXVII). The Young Research  Library (UCLA) has put a photographic reproduction of its set of the 1582  edition on the internet.125 The first two volumes contain Gratian’s Decretum,  together with the Glossa ordinaria (as revised by Bartholomew of Brescia).  Because  the  correctores  aimed,  in  editing  the  canons,  to  reproduce  the  original text (not  that  found  in  the  compilations  Gratian  used,  such  as  the  Panormia and Collectio tripartita of Ivo of Chartres) the text varies from the  Decretum of the manuscript tradition. Gratian’s  Decretum  is  divided  into  three  principal  parts.  The  individual  sections contain two kinds of texts: the canons and Gratian’s comments, dicta Gratiana. 123  A Winroth, The Making of Gratian’s Decretum (2000). The first recens ion is preserved in MSS:  Florence BN MS Conv Soppr A.1 402 (from D.28 d. p. c13); Admont Stiftsbibliothek MSS 23  and 43; Barcelona Arch C de Aragon MS Ripoll 78 (until C.13); Paris BN MS nouv acq lat 1761  (until C.12 q.2 c.39). 124  On the paleae, see J Rambaut-Buhot, “Les paleae dans le Décret de Gratien”, in S Kuttner and J  J Ryan (eds), Proceedings of the Second International Congress of Medieval Canon Law: Boston 1963 [= MIC Ser C Subsidia I] (1965) 23; R Weigand, “Versuch einer neuen, differenzierten  Liste  der  Paleae  und  Dubletten  im  Dekret  Gratians”  (1999)  23  BMCL  114;  J  Buchner,  Die Paleae im Dekret Gratians, Untersuchungen ihrer Echtheit (2000). 125  http://digital.library.ucla.edu/canonlaw

the sources of medieval learned law

33

Pars I is a treatise on the ordination and election of clerks. The division  into 101 distinctiones was not made by Gratian himself. The medieval canonists ascribe this division, which was already common in the 1160s, to Paucapalea.126 The first twenty distinctions deal with the sources of law. They have  recently  been  trans lated  (together  with  the  ordinary  gloss)  by  Augustine  Thompson and James Gordley as The Treatise on the Laws with the Ordinary Gloss (1993). Their translation is based upon the text of the editio Romana.  Pars II consists of thirty-six moot cases. In each causa Gratian outlined a case  and various legal questions evolving from it; he then discussed each question  separately, citing  the  authorities pro  and  contra.127  Causa  33, quaestio  3  is  a treatise on penance, known as De paenitentia. Pars III was added in the  second recension of Gratian’s Decretum. It deals primarily with sacramental  and liturgical law. (a) Palea This  term signifies  either  that  Gratian  adopted  the  canon  twice  from  his  sources, or that the canon has been added to the Decretum after its completion. Until around 1180 numerous texts were inserted in manuscripts with  Gratian’s work. Jacqueline Rambaud listed 149 of them, Rudolph Weigand  169. Many of these paleae are earlier canons that Gratian himself had not  adopted. D.5 c.1 for instance is a papal letter of 601. Some paleae are texts of  Roman law. In several manuscripts of French origin decretals written after  1150 were also inserted. Thus, the decretal Relatum (JL 7793; X 3.26.11) of  pope Alexander III was attached to C.13 q.2 d.p.c.7. (b) System of citations Nowadays the individual canons are referred to by their number in the  Friedberg  edition.  The  citation  does,  however,  vary  from  part  to  part  of  the  Decretum,  because  the  causae  of  the  second  part  are  subdivided  into  questiones. Canons in the first part, which contains 101 distinctions, are cited  as, for instance, “D.1 c.1”. This is the first capitulum (c.1) of the first distinction  (D.1).  Canons  in  the  second  part  are  cited  as,  for  instance,  “C.2.  q.6  c.40”. This is the fortieth canon (c.40) in the sixth question (q.6) of Causa 2   (C.2). The canons of C.33 q.3, however, are referred to differently. They are  cited as canons of a treatise on penance divided into distinctions, for instance  126  See T C McLaughlin, Summa Parisiensis (1952) x-xv. 127  Their content is described (shortly) by Peter Landau in P Landau, “Gratian and the Decretum Gratiani”, in Hartmann and Pennington (eds), History of Medieval Canon Law (n 16) 22 at  36-37.

34

the creation of the ius commune

“D.3 de pen. c.24”. The canons in the third part, a treatise on consecration,  are cited in a similar way, for instance “D.2 de cons. c.82”. The comments  of Gratian (dicta) are cited according to their place in the Friedberg edition,  thus before (ante) or after (post) an individual canon, for instance: “D.76 d.  post c.8”. This is a dictum of Gratian (d. or dict. or d.G.) after D.76 c.8. This  citation  system  only  became  common  in  the  twentieth  century.  From the seventeenth century onwards citations began with the number of  the canon, for instance, “c.1 D.I and c.40 C.II q.6”. Gratian’s dictum after  D.76  c.8  was  cited  as  “dictum  post  c.8  D.LXXVI”.  The  numbering  of  the  canons,  however,  only  became  uni form  in  Renaissance  editions.  Hence,  until relatively recently the canons and dicta of Gratian were generally cited  by their opening words, the first canons of a distinction or question also by  number  (“i.”,  “ii.”,  “iii.”),  and  the  last  two  canons  as  next  to  last  (“pen.”  or  “penult.”) and last (“ult.”). Hence, D.1 c.1 was cited as “di.i. c.i” (or simply  “i. c.i”). Similarly, C.2 q.6 c.40 was cited as “ii. q.vi. c. Si quando”. Gratian’s  dicta were also cited by their opening words. A paragraph sign was used to  distinguish them from a canon/capitulum. D.76 p.c.8 was referred to as “di.  lxxvi § necessario ergo”. (c) Recensions Until recently, the date of completion of Gratian’s Decretum was set around  1140,  because  the  latest  sources  he  used  were  the  canons  of  the  Second  Lateran Council of 1139. It is now generally accepted, however, that (at least)  two vers ions should be distinguished, since Anders Winroth has established  that four manu scripts contain a reference to this council (“sicut in generali  sinodo Innocentii pape Rome habita”), but not the canons themselves. The  authorities cited in this earlier version all date from before 1120. A description of the contents of this earlier recension is given in the appendix to his  The Making of Gratian’s Decretum (2000). A summary of recent scholarship  on this subject with a complete biblio graphy until 2006 is given by Winroth.128  Additional literature can be found on his website Domus Gratiani.

128 A Winroth, “Recent work on the making of Gratian’s Decretum” (2004-6) 26 BMCL 1. Recently  Pennington adhered to Larrainzar’s proposal that St Gallen Stiftsbibliothek MS 673 contains an  even earlier version, a first draft of Gratian’s work. Winroth characterises the work as an abbreviation of Gratian’s Decretum. Larrainzar dates this draft to 1142, Pennington to much earlier  (1120s). See K Penning ton, “The Big Bang, Roman law in the early twelfth century” (2007) 18  RIDC 43.

the sources of medieval learned law

35

(d) Roman law in the Decretum Gratiani More than 200 texts of Roman law were added to the Decretum in the second  recension.  Some  of  them  were  adopted  from  earlier  canonical  collections  and contain pre-Justinianic law: for example, the source of C.2 q.1 c.2 is the  collection of Anselm of Lucca. The vast majority, however, of these texts was  directly taken from the Digest and Code of Justinian.129 Also, around fifteen  of the later additions to the Decretum consist of texts of Roman law. Various  manuscripts of French provenance, for instance, insert Code 1.3.30 at the  end  of  C.1  q.7  c.26.  Weigand  has  noted  them  in  his  list  of  paleae.130  The  first recension contains only three passages with quotations from Justinian’s  codification. In C.2 q.6 c.28 Gratian cited a Novel of 523, in which Justinian  allowed ten days for appeal.131 In C.15 q.15 c.2-3 and in c.4 he included seven  fragments  from  the  Code  and  Digest,  which  permitted  women  to  accuse  someone of specific crimes. (2) Early glosses to the Decretum From the 1150s onward Gratian’s Decretum became the central text in the  canonists’ teaching. Its first glosses not only ex plained words in the text, but  also quoted parallel or contrary texts. There were few references to Roman  law in these early glosses. Guibert de Bornado was an early jurist who, more  than his contemporaries, applied Roman law to problems of canon law; he  also functioned as a judge in the court of Frederick Barbarossa between 1159  and 1178. In 1983, Weigand edited thirty of Guibert’s glosses.132 There is even  occasional reference to individual glossators, as in an early anonymous gloss  on D.84 c.5: “contra Martinus papa supra Lector si uiduam (D.34 c.18)”.133  As with many glosses written in about 1150, it does not yet use the division of  the first part of Gratian’s Decretum into distinctions. The work of Weigand has established that there are various “gloss-compositions”, that is groups of glosses transmitted in several manu scripts which  only partially or sporadically explain the text. This partial nature dis tinguishes  129 See  A  Vetulani,  “Gratien  et  le  droit  romain”  (1945/47)  24/25  RHD  11,  repr  in  A  Vetulani,  Sur Gratien et les decrétales (1990); S Kuttner, “New Studies on the Roman law in Gratian’s  Decretum” (1953) 11 Seminar 12, repr in S Kuttner, Gratian and the Schools of Law, 1140-1234  (1983); J M Viejo-Ximénez, “La ricezione del diritto romano nel diritto canonico”, in La cultura giuridico-canonica medioevale: Premesse per un dialogo ecumenico (2003) 157. 130  Weigand, “Liste der Paleae und Dubletten” (n 124) 126-128. 131  Nov 23.1; Authenticum Coll 4.2. The glossators inserted parts of the novel after C 7.62.5. 132  R Weigand, “Romanisierungstendenzen im frühen kanonischen Recht” (1983) 69 ZRG, KA 200  at 202. 133  R Weigand, “Glossen zum Dekret Gratians” (1992) 26 SG 750 note 11.

36

the creation of the ius commune

them from a gloss-apparatus, such as Ordinaturus magister, composed shortly  after 1180 in the school of Huguccio and revised at the end of the 1180s. The  majority of these gloss-compositions originate in Bologna, but one is characterised by Weigand as  being of  French provenance and another of  AngloNorman pro venance.134 The manuscripts contain glosses of many decretists,  among them Paucapalea, Rolandus, Rufinus, Stephan of Tournay, Cardinalis,  Gandulf and John of Faenza. A list of the sigla used in the manuscripts has  been  published  by  Weigand.135  He  has  also  given  a  cumulative  edition  of  early glosses on a section of Gratian’s Decretum and published several glosses  of individual authors.136 The vast majority of the early glosses are, however,  not  available  in  a  modern  edition.  The  same  holds  true  for  the  first  glossapparatus Ordinaturus magister, composed in the school of Huguccio. (3) Early summae The earliest works on the Decretum fall mainly into two types: glosses and  summae.137 The canonists combined both elements in their summae. Several  that circul ated in Bologna were published in the nineteenth century. The first  summa to the Decretum, composed by Paucapalea ca 1150, was published  by J F von Schulte in 1890 (reprinted in 1963). In this short and descriptive  summa, Paucapalea fre quently made use of Roman law, for instance to explain  the concept of prescription (D.1 c.12). In his commentary to Causa 2-7, he  laid out the basics of judicial procedure in ecclesiastical courts. The  works  of  Roland  and  Rufinus  were  the  most  influential  of  those  of  the Bolognese teachers in the 1150s. In 1874 Friedrich Thaner published a  summa of Roland (reprinted in 1962), of which the third part (C.27-C.36) is  now considered to have been a separate tract on marriage. Weigand’s analysis  in 1980 revealed that the manuscripts contain several recensions, written in  the 1150s. The edition follows the last (fifth) recension, finished in around  134 For a detailed description of which gloss-composition can be found in which manuscript, and  whose  glosses  they  contain,  see  R  Weigand,  Die Glossen zum Dekret Gratians: Studien zu den frühen Glossen und Glossenkomposition, 2 vols [= (1991-92) 25-26 SG] (1991) I 393; R  Weigand, “The Development of the Glossa ordinaria to Gratian’s Decretum”, in Hartmann and  Pennington (eds), History of Medieval Canon Law (n 16) 55. 135  Weigand, “Development of the Glossa ordinaria” (n 134) 95-97. 136  D.11 pr.-c.6; D.12 pr.-c.6; C.1 q.3 c.4, 13-15; C.27 q.2 pr.-c.11; C.30 q.4 and de cons. D.4 c. 1-5  are edited in Weigand, Glossen zum Dekret Gratians (n 134) I 1-392. An edition of a number  of glosses of in divid ual authors is given at Weigand, Glossen zum Dekret Gratians (n 134) I  569-659 and in R Weigand, Glossatoren des Dekret Gratians (1997). 137  The canonists also adopted other genres of the glossators – distinctiones, notabilia and brocardica.  Most of these collections are anonymous and short. Modern editions exist of the Distinctiones Monacensis (see n 142 below) and Ricardus Anglicus’ distinctions: G Silano, “The ‘distinctiones  decre to rum’ of Richardus Anglicus” (unpubl PhD diss, University of Toronto 1981).

the sources of medieval learned law

37

1160. In around 1164 Rufinus finished a summa, which in length and detail  surpassed the earlier works. Singer published this summa, which became the  most  influential  commentary  on  Gratian’s  Decretum  in  the  1170s,  in  1902  (repr 1963). Rufin’s student Stephan of Tournay – so designated because he  became bishop of Tournay in 1191, having actually been born in Orleans –  wrote a summa to Gratian’s Decretum between 1165 and 1167 after he had  return ed to France. Schulte published this summa in 1892, but left out the  parts Stephan had borrowed from Paucapalea, Roland and Rufin. His work  was  not  only  influential  in  France,  but  also  in  Bologna,  where  it  was  used  by John of Faenza, whose summa was a compilation of those of Rufin and  Stephan. This summa has, however, never been printed. Even though Bologna set the standard, both Roman and canon law were  taught very early in southern France.138 An abbreviation of Gratian’s Decretum  was composed in Provence as early as 1150.139 Stephan of Tournay, Raymond  Arène  (Cardinalis),  and  perhaps  also  Roland,  worked  in  southern  France.  In  the  early  1160s,  Gerard  de  Pucelle  taught  in  Paris  and  he  later  taught  in  Cologne  (1165-1168  and  1180)  and  England.  Only  three  works  of  the  early, northern canonists have yet been published. In 1952 T P McLaughlin  published the French summa, Magister Gratianus in hoc opere, composed  in  Paris  in  the  late  1160s.140 Stephan  Kuttner  and  Gérard  Fransen  edited  the summa Elegantius in iure divino, composed about 1169 in Cologne. Its  author, probably Bertram of Metz, only loosely fol lowed the general pattern  of Gratian’s Decretum.141 The third work, a collection of distinctiones, is of  minor value.142 The majority of the canonists’ works composed in 1160-1180,  however, can only be consulted in manu scripts.143 (4) Ius novum: decretals Like  the  Roman  emperor  in  antiquity,  the  pope  was  consulted  by  judges  on matters of law, and clerks and private persons would turn to the pope as  138 See A Gouron, “Die Entstehung der französischen Rechtschule” (1976) 93 ZRG, RA 138; repr  with additions in A Gouron, La science du droit dans le Midi de la France au moyen âge (1984). 139  See R Weigand, “Die Dekretabbreviatio Quoniam egestas und ihre Glos sen”, in W Aymans et  al (eds), Fides et ius: Festschrift Georg May (1991) 249. 140  T P McLaughlin, The summa parisiensis on the Decretum Gratiani (1952). 141  See P Landau, Die Kölner Kanonistik des 12. Jahrhunderts: Ein Höhepunkt der europäischen Rechtswissenschaft [= Kölner Rechtsge schicht liche Vorträge I] (2008). 142  Two editions exist: A J de Groot, Distinctiones ‘Si mulier eadem hora’ seu Monacensis (1996);  and R Sorice, Distinctiones ‘Si mulier eadem hora’ seu Monacensis [= MIC series A Corpus  Glossatorum IV] (2002). 143  See, for a detailed account of the works of the schools in France, Germany and England, R  Weigand,  “The  Transmontane  Decretists”,  in  Hartmann  and  Pennington  (eds),  History of Medieval Canon Law (n 16) 174.

38

the creation of the ius commune

Roman  citizens  had  appealed  to  the  emperor.  His  written  answers,  decretales, though not meant as legislation, were authoritative statements of law,  and were for this reason collected by canon lawyers. First, they occur in some  appendices to Gratian’s Decretum. In the 1170s the new law – papal letters  and canons of the recent council of Tours (1163) – was compiled independently.144  Bernard  Balbi  –  later  bishop  of  Faenza  (1191)  and  Pavia  (1198)  – composed the first sys tematic collection, known as the Breviarium extravagantium (survey of authoritative texts outside Gratian’s Decretum), which he  finished  between  1177  and  1179.145  His  com pilation  does  not  include  the  canons  of  the  Third  Vatican  Council  of  1179.  He  revised  and  updated  his  collection in the late 1180s – three recensions have been pre served in the  manuscripts, two of Italian provenance completed in 1187 and 1191, and a  northern European recension, which includes decretals of 1193.146 Bernard  arranged his Breviarium in five books, divided into titles, arranged after the  order of the Digest and Code. Other collections cir culat ed, but his Breviarium was used in the law schools. The summary of this col lect ion, which he  composed as bishop of Faenza, was published by Ernst Laspeyres, Bernardi papiensis summa decretalium (1860; repr 1956). Simon de Bisignano was the first canonist to make extensive use of texts  outside Gratian’s Decretum. In his summa, written at the end of the 1170s, he  cited extravagantes on some 175 occasions, amounting in all to about eighty  decre tal letters.147 In the 1160s, Rufin, Stephan of Tournay and the anonymous author of the summa Parisiensis had referred to only one. In 2007, P V  Aimone put his edition on the internet.148 Parts of three related summae have  recently been printed: a third of the summa, Et est sciendum, written in the  province of Sens around 1182, and the first part (until C.1) of two summae  written in Paris around 1186 by two Anglo-Norman canonists, Honorius and  an  unknown  author.149  Modern  editions  of  the  other  works  of  French  and  144 On the decretal collections, see G Fransen, Les décrétales et collections de décrétales [= Typ ologie des sources du moyen âge occidental II] (1972); P Landau, “Die Entstehung der systematischen Dekretalen sammlungen und die europäische Kanonistik des 12. Jahrhunderts” (1979)  65  ZRG, KA  120,  repr  in  P  Landau,  Kanones und Dekretalen  (1997);  C  Duggan,  “De cretal  Collections  from  Gratian’s  Decretum  to  the  Compilationes  Antique”,  in  Hartmann  and  Pennington (eds), History of Medieval Canon Law (n 16) 246. 145  Several other systematic collections of decretals are known. One of them is the French Collectio francofurtana, published by P Landau and G Droschbach (2007). 146  G Fransen, “Les diverses formes de la Compilatio Prima”, in Scrinium Lovaniense: Mélanges historiques Etienne van Cauwenbergh (1961) 235. 147  See  T  P  McLaughlin,  “The  Extravagantes  in  the  Summa  of  Simon  de  Bisignano”  (1958)  20  Medieval Studies 167. 148  http://www.unifr.ch/cdc/summa_simonis_de.php 149  F  Gillman,  “Die  Dekretglossen  des  Codex  Stuttgart  Hist.  f.419”  (1927)  107  Archiv für

the sources of medieval learned law

39

Anglo-Norman  provenance  are  lacking.  Only  the  first  twenty  distinctions  of Huguccio’s summa – the culmination of fifty years of Bolognese learning  since Gratian – have yet been published.150 Huguccio left his work unfinished when he became bishop of Ferrara in  1190 – just before the introduction of Bernard’s Breviarium into the canonistic curriculum. He wrote his summa in five stages: first pars I (D.1-D.101),  Causa 2-22 and 27-36 except De penitentia (C.33 q.3), subsequently Causa  1, then pars III (on the sacraments) and C.33 q.3, finally C.23-C.23 q.4 c.33. Several manuscripts contain passages from other summae to supplement the  text of the first stage. Wolfgang Müller has described in detail which summae  were used.151 As Kuttner pointed out, the intense activity of interpreting the  Decretum came to a standstill in about 1190.152 Study of the papal decretals  took its place. (a) System of citation The canonists of the twelfth century referred to the papal letters and conciliar  canons outside Gratian’s Decretum (known as extravagantes) by their opening  words, as, for instance, “ut in Extra. Sicut dignum”. Many, but not all, of these  decretals and canons are also included in the Corpus iuris canonici. Some of  them were only included in the compilationes antique (the old compilations  used in the law schools) of which Bernard’s was the first; a few were excluded  from both. In  works  composed  before  1190,  the  canonists  seldom  quoted extravagantes by title. It is therefore not always certain which collection of decretals  an author used. Simon de Bisignano probably used a primitive Italian collection  (without  a  division  in  titles);153  Honorius  certainly  used  the  Appendix concilii Lateranensis, because he used its titles in his citations.154 Nowadays 

150  151  152  153 

154 

katholische Kirchenrecht  192;  repr  with  additions  in  Gesammelte Schriften zur klassischen Kanonistik von Franz Gillman  (1988);  R  Weigand,  P  Landau  et  al  (eds),  Magistri Honorii summa ‘de iure canonico tractaturus’ [= MIC series A Corpus Glossatorum VIII] (2004) I; R  Weigand, P Landau et al (eds), Summa ‘omnis qui iuste iudicat’ sive Lipsiensis [= MIC series A  Corpus Glossatorum VII] (2007) I. O Přerovský (ed), Huguccio Pisanus summa decretorum [= MIC Series A Corpus Glossatorum  VI] (2006) I. W P Müller, Huguccio: The Life, Works, and Thought of a Twelfth-Century Jurist (1994) 76-80. S  Kuttner,  “Bernardus  Compostellanus:  A  Study  in  the  Glossators  of  Canon  Law”  (1943)  1  Traditio 283. See W Holtzmann, “Zu den Dekretalen bei Simon von Bisignano” (1962) 18 Traditio 451 at  453. Holtzmann listed four ma nu scripts belonging to the group of the Collectio Berolinensis,  ed by J Juncker, “Die Collectio berolinensis” (1924) 13 ZRG, KA 284. See Duggan, “De cretal  Collections” (n 144) 268-269. See S Kuttner and E Rathbone, “Anglo-Norman Canonists of the Twelfth Cen t ury” (1949-1951) 

40

the creation of the ius commune

the decretals are cited with their number as found in Philip Jaffé, Regesta pontificum romanorum, 2nd edn by S Löwenfeld et al (1885-1888). When the  decretal is included in one of the Compilationes Antique or the Liber Extra,  this  is  also  mentioned.  Nowadays  the  decretal  Sicut dignum, mentioned  above, would be cited as JL 12180 (= 1Comp. 5.10.7; X 5.12.5). (5) Compilationes antique The five collections that form the basis for the codification of pope Gregory  IX in 1234 are known as the Compilationes antique – the old compilations.  They were edited by Emil Friedberg as Quinque compilationes antique nec non collectio canonum Lipsiensis  (1882;  repr  1952).  This  edition  contains  the  decretals  that  are  not  included  in  the  Corpus iuris canonici.  For  the  numerous decretals that were adopted, Friedberg referred to his edition of  the Liber Extra. The compiler of the Liber Extra re-edited the text, but these  changes are not marked. The parts of the text in the old compilations that  were left out, Friedberg has added, written in italics, and marked with “†”.  Because  he  aimed  to  reconstruct  the  original  text  of  the  decretal,  he  also  added material that is not found in the compilationes antique. This is text,  written in italics, that he placed between parenthesis: “[ ]”. Bernard’s Breviarium is the first collection, or Compilatio prima, a compilation  of  conciliar  canons  and  papal  letters  until  1190.  Early  papal  letters,  which  had  not  been  entered  in  Bernard’s  collection,  and  decretals  issued  between 1190 and 1200 form the Compilatio secunda. They were compiled  by Johannes Galensis in around 1212 after pope Innocent III had promulgated a collection of his decretals in 1210. Canonists began to put together collections of Innocent’s decretals early in  his pontificate. In 1203, Rainer of Pomposa selected 123 of Innocent’s decretals from the papal registers. During his  stay  in  Rome  in  1208,  Bernardus  Compostellanus  had  also  consulted  the  registers.  His  collection,  however,  contained some decre tals that the Bolognese schools attributed to Innocent.  These could not be found in the papal registers and thus were rejected by the  curia. Two years later, Petrus Beneventanus compiled a collection of which  the pope approved. Innocent accordingly sent it to Bologna with a letter of  approval. This is known as the Compilatio tertia (though, in fact, it was the  second to be compiled). Sixteen decretals, which Petrus and Innocent knew  not to have been registered, were added to the registers in 1210. The wording  7 Traditio 279 at 283-284; repr in Kuttner, Gratian and the Schools of Law (1983). The collection is published in J D Mansi, Sacrorum conciliorum nova et amplissima collectio (1767; repr  1962) XXII.

the sources of medieval learned law

41

of the text of these sixteen decretals varies between the Compilatio tertia and  the registers. This implies that the entries in the registers were not made to  secure the correct text, but to prevent forgeries entering the decretal collections. In the fourth collection, Compilatio quarta, Johannes Teutonicus combined  the canons of the Fourth Lateran Council of 1215 with decretals, from the  last  years  of  Innocent’s  pontificate,  that  circulated  outside  the  Compilatio tertia. An English translation of the canons of the Fourth Lateran Council  can be found on the internet.155 Though this new compilation failed to get  the papal authentication that Johannes had sought, it was used in the schools  of canon law around 1220. The fifth collection, Compilatio quinta, is the last  compilation before the Liber Extra. It contains decretals of pope Honorius,  at whose request the collection had been assembled by Tancred. With the pontificate of Innocent III (1198-1216), who had studied theology  in  Paris  and  law  in  Bologna,  the  focus  of  the  canonists  shifted  definitively  towards the ius novum, contemporary papal letters and conciliar canons. Pope Celestin (1191-1198) did not issue many decretals, and the canonists  used  the  time  of  his  pontificate  to  consolidate  their  teaching.  Along  with  Gratian’s  Decretum,  Bernard’s  Breviarium  now  became  the  standard  textbook in the schools of law. None  of  the  main  works  on  the  Decretum  written  around  1200  by  the  canonists in Bologna has been printed. Even the gloss of Johannes Teutonicus (Johann Zemecke), the Glossa ordinaria, has only been printed in the  revised version necessitated by the publication in 1234 of the Liber Extra of  Gregory IX. It was edited by Bartholomew of Brescia (Bartolomeus Brixiensis) who incorporated refer ences to new decretals and offered new solutions  to old questions. He thus thoroughly updated the 1217 version of Johannes  Teutonicus,  correcting  or  modifying  Johannes’  positions.156  Also  available  only in manuscript are the main sources used by Johannes Teutonicus – the  Glossa Palatina of Laurentius Hispanus (who in turn took many glosses from  Ordinaturus Magister) and the summa of Huguccio. The same is true for the  works of the other Bolognese and northern canonists of that time.157 Ricardus Anglicus (Richard de Mores), who taught in Bologna from 1191  155 www.papalencyclicals.net/councils 156 See J F Schulte, Glosse zum Dekret Gratians von ihren Anfängen bis auf die jüngsten Ausgaben  (1872) 79-82; R Weigand, “The glossa ordinaria”, in Hartmann and Pennington (eds), History of Medieval Canon Law (n 16) 82. 157  A beginning has been made with the edition of the Paris apparatus, Animal est substantia, which  has a strong romanising tendency. Coppens published his edition of D.1-D.29: http://web.mac. com/eccoppens/Animal_est_ substantia/Introduction.html

42

the creation of the ius commune

to  1202,  was  probably  the  first  to  use  Bernard’s  Breviarium as  a  textbook.  He  composed  the  first  casus decretalium,  a  résumé  of  the  decretals  that  Bernard used in writing his own. The edition by Ernst Laspeyres of Bernardi Papiensis summa decretalium (1860; repr 1956) 327-357 contains excerpts of  both casus. Richard’s apparatus, written when he taught in Bologna (11911202) was very influential, as was the commentary composed between 1202  and  1210  by  his  compatriot  Alanus.  Other  canonists,  like  Petrus  Hispanus  and Bernardus Compostellanus, also glossed the first compilation. None of  these apparatus is printed. Two of the commentaries to the Compilatio tertia  (1210),  however,  were  recently  printed.  Laurentius  Hispanus’  apparatus  was  edited  in  1991  by  Brendan  McManus.158  A  critical  edition  of  the  first  two  books  of  Johannes  Teutonicus’  commentary  appeared  in  1981,  edited  by  Kenneth  Pennington,159  who  has  supplemented  this  with  three  further  books available on the internet.160 Glosses to the compilatio secunda (1212)  are  only  preserved  in  manuscript.  Antonio  García  y  García  published  a  critical edition of the canons of the Fourth Lateran Council (1215) together  with the gloss-apparatus of Johannes Teutonicus, Vincentius Hispanus, and  Damasus Hongarus.161 Johannes Teutonicus was the only canon ist who wrote  an apparatus to his Compilatio quarta.162 It was edited in 1576 by Antonio  Agustín  in  his  Antique decretalium collectiones commentarii  (also  in  his  Opera omnia, published in Paris, 1621). Tancred did not gloss the collection  he made in 1227 of Honorius’ decretals in 1227. Tancred, who learned Roman law from Azo and canon law from Laurentius,  changed  the  style  of  apparatus  to  the  decretals.  He  incorporated  the  glosses  of  other  canonists  with  little  change,  providing  the  sigla  of  their  authors. Until then canonists had incorporated glosses of others in their own  commentary. Consequently, his apparatus functions as a survey of opinions.  His  apparatus  to  the  compilatio prima–tertia, written  between  1210  and  1215, and revised in the 1120s, became the Glossa ordinaria to these collections. It has, however, never been printed. 158 B J McManus, “The Ecclesiology of Laurentius Hispanus (ca 1180-1248) and his Contri bu t ions  to the Romanization of Canon Law Jurisprudence: With an Edition of the Apparatus glos sa  rum  Laurentii Hispanii in com pi la tionem tertiam” (unpubl PhD diss, University of Syracuse 1991). 159  Johannes Teutonicus, Apparatus in Compilationem tertiam, ed Kenneth Pennington [= MIC,  Corpus glossatorum III] (1981) I. 160  http://faculty.cua.edu/pennington 161  A  García  y  García  (ed),  Constitutiones Concilii quarti Lateransis [=  MIC  series  A  Corpus  Glossatorum II] (1981). 162  He used his own and Vincentius’ glosses to the Canons of the Fourth Lateran Council for his  ap pa ratus to the Compilatio quarta. Many of his glosses are the same; some are changed or  augmented.

the sources of medieval learned law

43

(a) System of citation Around  1200,  the  canonists  referred  to  the  extravagantes  by  giving  the  number of the compilation, then the title in an abbreviated form, followed by  the opening words of the capitulum, for instance “(extra) ii. de elect. Suffrageneis”. This chapter in the third title of the Compilatio secunda is now cited  as “2Comp. 1.3.1” with an added reference to its place in the Liber Extra in  parenthesis: “(X 1.6.11)”. A chapter not found in the Liber Extra is cited as:  “1Comp. 2.20.34 (X -)”. (6) Liber Extra The  Liber Extra  or  Decretales Gregorii IX  is  a  collection  of  almost  2,000  chapt ers that includes all relevant papal and conciliar law since Gratian. In  1230,  pope  Gregory  IX  (1227-1241)  commissioned  Raymond  of  Peñafort  to  com pose  a  col lection  that  would  replace  the  various  decretal  collections  in  circulation.  The  collection  is  arranged  according  to  the  pattern  of  the compilationes antique that Bernard of Pavia had invented. The collection  is  divided  into  five  books  and  each  book  into  titles.  To  memorise  the  subjects of the books, the students used the fol lowing verse: “iudex, iudicium,  clerus, connubia, crimen” (“judge, judgement, cler gy, marriage, crime”). The  chapters in each title are placed in chronological order: thus, the oldest papal  letters and canons come first. Raymond used the material of the compilationes antique, but not all. He  eliminated many of the letters of Honorius III (1217-1227) – about 40% of  the compilatio quinta. Following the example of the compilers of the older  collections, he re-edited the text of the papal letters he used, he shortened  or deleted the description of the facts of the case, and changed the wor ding  when needed. He also included papal letters from the beginning of Grego ry’s  pontificate and phrased new rules, for instance chapter X 5.36.9: “If damage  or injuries are inflicted through your fault, either through your incompetence  or negligence, you have to make satisfaction...”163 Pope  Gregory  IX  promulgated  this  collection  in  September  1234  and  sent  it  to  the  law  schools  in  Bologna  and  Paris,  to  be  taught  as  the  law  of  the Church. It re placed all earlier collections of decretals. The old universal  law  of  the  church,  namely  the  canons  until  Gratian’s  Decretum  –  whether  Gratian had included them in his work or not – were abolished insofar they  conflicted with the new law. The ius novum – papal decretals and conciliar  163 “Si culpa tua datum est damnum or iniuria irrogata, seu aliis interrogantibus opem forte tu listi,  aut haec imperitia tua siue negligentia evenerunt: iure super his satisfacere te oportet...”

44

the creation of the ius commune

canons of the period between Gratian’s Decretum and the Liber Extra – was  abolished, even when in con form ity with the Liber Extra. Therefore, though  pope  Gregory  spoke  of  this  work  as  a  collection  (compilatio nova),  it  can  nonetheless be properly characterised as a codification.164 The modern edition of the Liber Extra appeared in 1881 as the second  vol ume of the Corpus iuris canonici (reprinted in 1955), edited by Emil Friedberg. The text is the same as in the editio Romana of 1582. Variations from  the text of Raymond de Peñafort of which Friedberg was sure are marked  with “*”. The text written in italics is not part of the Liber Extra. It is a reconstruction of the original text of the decretal. The Law Faculty of the University of Catania has put a full searchable text of the Friedberg edition on the  internet.165 As already noted, the editio Romana is made available for viewing  on the UCLA Library website.166 The index to the decretals (Margarita) is  published along with an index to the gloss by Andy Kelly on his website.167 William  Naso’s  lectures  on  the  new  Liber Extra  date  back  to  around  1234. It is the earliest set of lectures that has been preserved. Like the other  commentaries  writ ten  during  Gregory’s  pontificate,  it  is  only  available  in  manuscript. The new codification prompted Bartholomew of Brescia (Brixiensis) to revise the ordina ry gloss to the Decretum. Bernard of Parma published  a first recension of his ap paratus (which would in its final form become the  ordinary  gloss)  in  1241,  the  year  of  Gregory’s  death.  Goffredus  Tranensis,  auditor litterarum contradictarum under Gregory IX and Innocent IV, wrote  a  summa decretalium  (1241-1243)  after  he  completed  his  apparatus.  This  influential  summa  achieved  many  manuscript  and  printed  editions.  The  Lyon  1519  edition  was  reprinted  in  1968.  Bernard  revised  his  gloss  at  the  beginning  of  the  pontificate  of  Innocent  IV  (1243-1254),  and  did  so  again  after the Council of Lyon (1245), where thirty-eight new constitutions were  issued. Sinebald Fliscus, who was elected to the papacy in 1243 as Innocent  IV, comple ted his gloss-apparatus to the codification of his predecessor, and  also glossed his own Novellae. The Frankfurt 1570 edition of his apparatus  was reprint ed in 1968. Bernard of Parma continually revised his gloss until  his  death  in  1266.168  Henry  of  Susa  (1194-1271)  –  known  as  Hostiensis  as  164 M Bertram, “Die Dekretalen Gregors IX (1234): Kompilation oder Kodifikation?”, in Magister Raymundus: Atti del Convegno per il IV centenario della canonizzazione di san Raimondo de Penyafort (2002) 61. 165  www.lex.unict.it/liber/accedi.asp 166  http://digital.library.ucla.edu/canonlaw 167  http://www.english.ucla.edu/faculty/kelly 168  See  S  Kuttner  and  B  Smalley,  “The  Glossa  ordinaria  to  the  Gregorian  Decretals”  (1945)  40  English Historical Review 97; repr in S Kuttner, Studies in the History of Medieval Canon Law  (1990).

the sources of medieval learned law

45

he became cardinal-bishop of Ostia in 1262 – updated his commentary on  the  decretals  until  1270.  By  chance,  a  manuscript  of  the  first  recension,  composed  between  1254  and  1265,  has  been  preserved  in  Oxford.169  The  printed  editions  contain  the  version  of  1270.170  Two  indep endent  printed  first  editions  appeared  in  1512,  one  in  Paris,  the  other  in  Strasbourg.  His  Lectura super quinque libros decretalium is vast, but later jurists praised its  clarity.  According  to  Hostiensis,  no  one  would  have  understood  him,  if  he  had written tersely. Both editions included his commentary on the Novellae  of pope Innocent IV of 1251 to 1253. Around that time he also completed  his Summa aurea, which follows the order of the Liber Extra.171 Hostiensis  mentions that he had to rewrite his summa because the first version went up  in flames. The first printed edition appeared in 1473. Bernard’s gloss became  the Glossa ordinaria to the Liber Extra.172 (a) System of citation The citation form is similar to those used for the compilationes antique. Until  the nineteenth century, canonists cited decretals in the Liber Extra with: X or  Extra (within the Liber Extra with supra and infra), the title (in an abbreviated form), followed by the beginning words of the chapter. An example is:  “ut X de appel. Super eo”. Nowadays the number of the book, of the title, and  of the chapter are given: “X 2.28.10”. (7) Liber Sextus, Clementinae, and Extravagantes Pope Gregory did not stop writing papal letters on questions of law in 1234, nor  did his successors. The canonists collected these extravagantes and referred  to them in their commentaries. The popes promulgated several collections  them selves, pope Innocent IV (1243-1254) published Novellae (1245, 1247,  1253) and pope Gregory X (1271-1276) published Novissimae (1276). After  a request from the law school in Bologna, pope Boniface VIII (1294-1303)  replaced these decretals with a new exclusive collection, the Liber Sextus.  The Liber Sextus incorporates the canons of the Council of Lyon (1245 and  169 K Pennington, “An Earlier Recension of Hostiensis’ Lectura on the Decretals” (1987) 17 BMCL  77;  M  Bertram,  “Hand schrif ten  und  Drucke  des  Dekreta len kommentars  (sog.  Lec tura)  des  Hostiensis” (1989) 75 ZRG, KA 177. 170  For  the  manuscript  and  printed  editions,  see  M  Bertram,  “Handschriften  und  Drucke  des  De kre ta len kom mentars” (n 169). 171  On the medieval editions, see F P W Soetermeer, “Summa Archiepiscopi alias summa copiosa”  (1999) 26 Ius Commune 1. 172  On the editions of his apparatus, see S Kuttner, “Notes on the Glossa ordinaria of Bernard of  Parma” (1981) 11 BMCL 86.

46

the creation of the ius commune

1274) and most of the decretals prom ulgated by his predecessors combined  with some of his own. It follows the struct ure of the Liber Extra: both the  order of the books and of the titles. Early in the fourteenth century, pope Clemens (1305-1314) ordered the  compilation of the canons of the council of Vienne (1311-1312) and some of  his decretals. His successor, pope John XII, sent this collection, the Constitutiones Clementinae, to the law schools. Unlike the Liber Extra and the Liber Sextus,  the  new  collection  was  not  given  a  sole  exclusive  authority.  Papal  letters written after 1298 could still be used in court to argue a case. Several  private  collections  were  made,  of  which  the  correctores Romani  in  1580  authenticated two: a collection of twenty decretals, the Extravagantes Joannis XXII  composed  in  1325,173  and  a  collection  of  sixty-nine  (or  seventy-four)  decretals,  entitled  the  Extravagantes communes.  These  were  divided  into  titles at the end of the fifteenth century by Jean Chappuis when he prepared  them for the Paris edition of the Corpus iuris canoni that appeared in 1500. Johannes Monachus wrote the first apparatus on the Liber Sextus, of which  the Paris 1535 edition was reprinted in 1968. Johannes Andreae, his pupil  Guido de Baysio, and Zenzelinus de Cassanis also glossed the Liber Sextus.  The apparatus of Guido de Baysio was printed several times in the sixteenth  century. The apparatus of Johannes Andreae became the ordinary gloss. The  sixteenth-century  editions  of  the  Corpus iuris canonici  with  the  ordinary gloss contain the apparatus of Johannes Andreae to the Liber Sextus  (1306) and to the Clementinae (1326), together with the glosses of Zenzelinus  de Cassanis to the Extravagantes Joanni XXII. The 1582 editio romana with  the ordinary gloss is available on the web,174 as is the index to the gloss.175  Glosses of Johannes Andreae to decretals of Johannes XXII can be found as  additiones in his gloss to the Liber Sextus. (a) System of citation The Liber Sextus is designated as VI and the Clementinae as Clem. The citation  forms are similar to that of the compilationes antique and the Liber Extra.  The medieval and renaissance canonists gave the title and the beg inning of  the decretal. In the editions of the sixteenth century and later, decretals in  both collections of extravagantes are cited as, for instance, “c.1 de conces.  preb. tit. IV. in Extravag. Jo. XXII”, and nowadays as “Extrav. Jo. 4.1”. 173 J Tarrant, Extravagantes Iohannis XXII [= MIC Series B Corpus Collectium VI] (1983). 174 http://digital.library.ucla.edu/canonlaw 175 http://www.english.ucla.edu/faculty/kelly

the sources of medieval learned law

47

F. PROCEDURAL LAW In the twelfth century neither the Corpus iuris civilis nor Gratian’s Decretum  provided a consistent description of procedure. The question of how a trial  must be conducted was, however, of great importance for the Church, because  a holder of a church-office could not be deposed without an orderly trial. The first works on the ordo iudiciarius (or ordo iudiciorum) were written  by legists – at the request of Church prelates.176 The Roman concept of a trial  con ducted by a single judge, if necessary assisted by professionals, had the  advant age that it strengthened their powers. Until Roman law was revived,  ecclesiastical councils rendered judgment. The early ordines iudiciarii were based upon the Corpus iuris civilis, but  the  use  of  canon  law  increased  in  the  course  of  the  twelfth  century.  The  anonymous  French  ordo  Tractaturi  (1160s),  for  instance,  adopted  parts  of  the  Code-summa Trecensis, parts  of  the  Decretum-summa  of  Stephan  of  Tournay  (on  ap peal  and  prescription),  and  perhaps  made  use  of  Rogerius’  lectures on actions.177 The English ordo Quia iudiciorum (1180s) was the first to list differences  be tween Roman and canon law in procedural matters. Like other works of  English provenance, it regularly referred to decretals of Alexander III. Such  citations were sparse in Italian works.178 A copy of an Anglo-Norman ordo,  which the canonist Richard de Mores (Ricardus Anglicus) wrote shortly before  1190, circulated in Bologna.179 Tancred remarked that it was one of the two  works he had on his desk when he was writing his Ordo iudiciorum (on the  basis of notes he made in 1210-1215). The other he erroneously attributed  to Pilius.180 Bartholomeus Brixiensis revised his ordo after the promulgation  of the Liber Extra. Many copies circulated, but his ordo was superseded by  Guillaume Durand’s Speculum iudiciale. This work, completed around 1290  (the  earlier  1275  recension  contains  some  variations),  was  to  become  the  176 With the exception of a work on criminal trials: Rhetorica ecclesiastica / Tam veteris quam novi  (ca 1160), ed L Wahrmund [= QGRKP I pt 4] (1906; repr 1962). 177  See L Fowler-Magerl, Ordo iudiciorum vel ordo iudiciarius [= Ius Commune Sonderhefte XIX]  (1984) 58-64 and 173. On both the summa Trecensis and Stephan of Tournay, see section C.(3)  above. 178  The  reason  was  probably  that  his  decisions  were  often  taken  from  Roman  law  or  from  the  writings of legists. See K W Nörr, “Päpstliche Dekretalen und römisch kanonischer Zivilprozess”, in W Wilhelm (ed), Studien zur europäischen Rechtsgeschichte (1972) 60. 179  Ed L Wahrmund [= QGRKP II pt 3] (1915; repr 1962). 180  This was the ordo known as Invocato Christi nomine, dependent on Pilius, but written shortly  after  1198  by  Ben ci venne  of  Siena,  ed  L  Wahrmund  [=  QGRKP  V  pt  1  1935;  repr  1962).  Tancred’s ordo has been published in Pilli, Tancredi, Gratii libri de iudiciorum ordine, ed  F  Bergmann (1842; repr 1965) 89-314. This edition also contains two French (Parisian) revisions  (1225, 1234) called Reformatio I and II.

48

the creation of the ius commune

standard work on the Roman-canon law of procedure. Important additions  – with references to the opinions of later legists and canonists – were made  around  1350  by  Johannes  Andreae,  and  later  by  Baldus  de  Ubaldis.  Many  printed editions include these additions, as for instance the edition of Basel  1574, reprinted in 1975. The glossators and canonists not only published ordines iudiciarii on civil  procedure, they wrote treatises on specific questions as well. They dealt with  delays and contumacy, declinatory exceptions and the refusal of judges, proof  and witnesses, sentences and appeal and the like. Very influential were two  works  on  claims:  a  summa de varietate actionum,  written  by  Placentinus  during his stay in Mantua, and the arbor actionum of Johannes Bassianus.181 All ordines and procedural treatises until 1234 are described extensively  with a list of all manuscript and printed editions by Linda Fowler-Magerl.182  She con sequently uses the incipit of these works instead of the title given by  the sixt eenth-century printers. Placentinus’ work is, for instance, listed under  Cum essem Mantue, Tancred’s work is listed under Assiduis postulationibus. Recently  the  Polish  scholar  Wiesław  Litewski  published  in  two  volumes  a  German translation of his textbook on the law of procedure until 1234: Der römisch-kanonische Zivilprozess nach den älteren ordines iudicarii (1999).  The book provides detailed information on all aspects of the law of procedure  and the references to the sources are very specific. It is an indispensable tool  for those who are interested in procedural matters. Litewski  rightfully  emphasises  the  importance  of  Si considerarum ingenium  that  Roffredus  Beneventanus  published  in  the  1230s.  Printed  under  the  title  Libelli iuris civilis,183  it  provides  examples  for  all  kinds  of  petitions  and discusses many procedural details. From the Authenticum had  derived the rule that in trials before a judge ordinary – who had jurisdiction  because of his office – the plaintiff had to prepare a libellus describing his  complaint.  Early  ordines  had  not  yet  phrased  a  claim;  the  plaintiff  simply  requested  that  justice  be  done.  Separate  treatises  on  libelli,  which  began  to  appear  in  the  early  thirteenth  century,  also  discussed  what  is  claimed:  damages  or  specific  performance.  The  models  Roffredus  published  were  181 Placentinus, Summa de varietate actionum, ed L Wahrmund [= QGRKP IV pt 3] (1925; repr  1962): Rhodius’ edition of Mainz 1530 and its many reprints are unreliable. A von Brinz edited  Johannes Bassianus’ Arbor actionum (1854). 182  L Fowler-Magerl, Ordines iudiciarii and libelli de ordine iudiciorum from the Middle of the Twelfth to the End of the Fifteenth Century [= Typologie des sources du moyen âge occidental  LXIII]  (1994);  L  Fowler-Magerl,  Ordo iudiciorum vel ordo iudiciarius  [=  Ius  Commune  Sonderhefte XIX] (1984). 183  The edition of Avignon, 1500, has been re printed in CGIC VI in 1968 along with his Libelli de iure canonico.

the sources of medieval learned law

49

popular  among   practitioners  and  remained  in  use  well  into  early  modern  times. He structured his work according to the arbor actionum of Johannes  Bassianus. Roffredus also referred to this work, but a source he may have  concealed is the first treatise on libelli, written by Bernard Dorna in 1213.184 G. FEUDAL LAW Around  1020,  Guillaume  V,  count  of  Poitou  and  duke  of  Aquitaine,  asked  Fulbert, bishop of Chartres, what duties the vassal owed to his lord. In his  reply Fulbert described six duties as being central to the oath of fealty: to  keep his lord safe, to protect him from harm, to preserve the lord’s justice, to  prevent damage to his possessions, and not to prevent the lord from carrying  out his duties. Gratian in cluded Fulbert’s letter in his Decretum (C.22 q.5  c.18),  where  it  became  the  starting  point  for  the  decretists  to  discuss  the  relationship between a vassal and his lord.185 Around 1155 Obertus de Orto, a judge in Milan at the time that a compilation  of  customary  law  was  undertaken  by  the  city’s  commune,  described  Lombard  feudal  law  –  in  the  form  of  two  letters  to  his  son,  a  student  in  Bologna who had reported (he said) that feudal law was not taught. These  two letters, together with imperial statutes of the emperors Conrad II (10241039) and Lothair II (1125-1135), form the first recension, called recensio antiqua or  Obertina, of  a  collection  of  feudal  laws  that  entered  into  the  Corpus iuris civilis in the thirteenth century.186 Early manuscripts contain as  additions various imperial statutes of the emperors Frederick I (1152-1190)  and  Henry  VI  (1190-1197).  Some  com pilers  also  included  Fulbert’s  letter.  Until ca 1250 there was no standard collection of the Libri feudorum. This collection of feudal law gradually entered into the schools of law. Early  glossators discussed feudal matters;187 but the first to gloss this  collection was  184 Published by L Wahrmund in QGRKP I pt 1 (1905; repr 1962). 185 On Fulbert’s letter, see G Giordanengo, “Epistula Philiberti”, in G Giordanengo, Féodalités et droits savants dans le Midi médiéval [= Collected Studies Series CCCLXXIII] (1992). On the  feudal oath of fealty, see M Ryan, “The Oath of Fealty and the Lawyers”, in J Canning and O  G Oexle (eds), Politisches Denken und die Wirklichkeit der Macht im Mittelalter (1998) 211,  criticised in K Pen ning ton, “The Formation of Jurisprudence and the Feudal Oath of Fealty”  (2004) 15 RIDC 57. 186  See E A Laspeyres, Über die Entstehung und älteste Bearbeitung der Libri feudorum (1830).  The collection ends at LF 2.24; it includes the Capitula Ugonis de Gambolado (after LF 2.2)  but ex cludes Fulbert’s letter (LF 2.6 in the recensio vulgata and LF 2.7pr). An edition with  apparatus criticus  is  K  Leh mann,  Consuetudines feudorum [Libri feudorum, Ius feudale Langobardorum], Compilatio antiqua (1892) I, reprinted in K Lehmann (ed), Consuetudines feudorum editio altera, curavit K A Eckhardt (1971). 187  See E M Meijers, “Les glossateurs et le droit féodal” (1934) 13 TvR 129; repr in Meijers, Études  III with an edition of a consilium of Johannes Bassianus.

50

the creation of the ius commune

Pilius, who, from around 1182, worked and taught in Modena. Six manu scripts  contain his (unfinished) gloss-apparatus, written in the 1190s. It gets as far  as LF 2.38. The text of one of those manuscripts has been edited.188 Pilius  glossed the collection in its second stage, known as the recensio Ardizona – so  called because it was supposed to be the collection Jacobus Ardizone had on  his desk while writing his summa feudorum between 1220 and 1240.189 His  summa is preserved in seven manuscripts, and the edition published in Asti  in 1518 has recently been reprinted.190 About twenty manuscripts contain the  recensio Ardizona, all with additional imperial laws.191 In  the  course  of  the  thirteenth  century  the  collection  of  feudal  laws  reached its final version and entered into the manuscript and printed editions  of the Corpus iuris civilis: the vulgata-recension. The recensio Ardizona was  enlarged  with  constitutions  which  the  jurists  had  till  then  characterised  as  extravagantia: a constitution of Henry III (1039-1056); three of Lothair III  (1125-1137); and three promulgated by Frederick I at Roncaglia in 1158. The  collection ends with the constitution Ad decus. Fulbert’s letter (already incorporated in Gratian’s Decretum) became part of the Libri feudorum as well.192  The collection of feud al law then also acquired its place in the manuscripts  of the Corpus iuris.193 The genesis of the ordinary gloss is as complicated as the history of the  Libri feudorum.  On  the  one  hand,  it  is  said  that  Accursius  plagiarised  the  188 A Rota, “L’apparato di Pillio alle Constitutiones Feudorum e il Ms. 1004 dell’ Archivio di Stato  di Roma”, in Studi e memorie per la storia dell’ Universita di Bologna (1938) XIV 1. The glosses  to LF 2.27 (nos 202-204), LF 2.54 (nos 245-255), and LF 2.55 are later additions. See P Weimar,  “Die Handschriften des Liber Feu dorum und seiner Glossen” (1992) 1 RIDC 72. Alvarot tus  (d 1453) asserted that Pilius also wrote a summa feudorum. This summa, which has not been  preserved, is probably revised and enlarged in the summa edited by Palmerio (see n 197 below). 189  Weimar’s research has established that Jacobus’ summary is also based upon an additonal collection of feudal laws (extravagantia) he had compiled. See Weimar, “Handschriften” (n 188) 35. 190  Iacobus de Ardizone, Summa super usibus feudorum composita (1518) in CGIC. 191  In this version, the collection ends with LF 2.51, still includes the Capitula Ugonis de Gambolado  and ex clud es Fulbert’s letter. All manuscripts with the Ardizona recension contain supplements,  called extravagantia, but these vary. See Weimar, “Handschriften” (n 188) 36-41. Many of them  became  part  of  the  vulgata recension  found  in  the  manuscript  and  printed  editions  of  the  Corpus iuris civilis. 192  Weimar  further  distinguishes  a  recensio protovulgata  (which  excludes  the  Capi tula  Ugo nis  and  incorporates  Fulbert’s  letter;  at  this  stage  Accursius  treated  LF  2.52-2.57  not  as  part  of  the compilation, but as extravagantes) and a recensio Accursiana (which includes the Capitula Ugonis after LF 2.22) regularly as the tenth collatio of the Authenticum in manuscripts of the  Volumen. See Weimar, “Handschriften” (n 188) 49. 193  The Libri Feudorum with the ordinary gloss have been reprinted (with a commentary) in M  Mon tor zi, Diritto feudale nel Basso medioevo: Materiali di lavoro e strumenti critici per l’esegesi della glossa ordinaria ai Libri feudorum (1991). An English translation of the Libri feudorum  is found as an appendix to T Craig, Jus feudale, with an Appendix Containing the Books of the Feus, trans Lord Clyde (1934) II 1079-1183.

the sources of medieval learned law

51

(now lost) apparatus of Jacobus Columbi, of whom Alvarottus (d 1453) said  that he had glossed the Libri feudorum so elegantly, that after him no one  dared to do so.194 Weimar on the other hand distinguishes between Accursius’  apparatus – in twenty of the many manuscripts195 – and the Glossa ordinaria, a  revised version thereof. Additions of ca 1282 refer to a “liber Jacobi Columbi  de  Reg(g)io”.  According  to  Weimar,  this  should  be  understood  as  a  reference to a manuscript owned by Jacobus Columbi that contained the original  version of the Accursian gloss, not as a reference to an apparatus, which could  be plagiarised.196 For the same reason, the summa that the editor G B Palmerio attributed to  Hugolinus is ascribed either to Jacobus Columbi or to Accursius.197 Odofredus’  Summa, of which parts were copied verbatim by Baldus, was printed in 1584  in Alcalá de Henares.198 Jean de Blanot wrote a famous commentary on J Inst  4.6 (de actionibus), a part of which was published separately as Tractatus de feudis et homagiis, which Guillaume Durand used in his Speculum iudiciale,  IV.3, de feudis.199 Whether Jacques de Revigny wrote the Summa feudorum attributed to him has been questioned.200 Kees Bezemer recent ly proposed  Jacobus Balduini as its author.201 A  tractatus de feudis,  ascribed  to  Pierre  de  Belleperche  in  the  printed  editions,  was  in  fact  written  by  Johannes  Fasolus  (1233-1286)  who  taught  in Naples.202 The text of his work shows strong similarities with that of the  Summa  ascribed  to  Mar tinus  Syllimanus  (ca  1245-1306).203  Around  1300  194 Laspeyres, Entstehung (n 186) 363-375, 399. 195 Weimar, “Handschriften” (n 188) 75-81 distinguishes between two recensions: a gloss to the  protovulgata recension, incorporating Pilius’ unfinished gloss, and adding his (Accursius’) own  glosses  up  to  LF  2.56.  Subse quently  Accursius  enlarged  his  apparatus  incorporating  Pilius’  introduction and adding a gloss to LF 2.57. The Glossa ordinaria is a revision of the first recension, as it does not contain Accursius’ gloss to LF 2.57. 196  Cf Weimar, “Handschriften” (n 188) 84. At 76 n 87 Weimar explains that the glosses ad LF  2.54 with the siglum “co.”, are copied from a commentary to the Authenticum by Columbus, a  contemporary of Azo, preserved in Munich BSB MS Clm 13018 (which ends with a constitution  of Frederick I, which was later incorporated in the vulgata-recension of the LF). 197  Hugolinus [sic], Summa in BIMAE II 181-194. The summa is attributed to Jacobus Colombi  by E Seckel, “Ueber neuere Editionen jurister Schriften aus dem Mittelalter I” (1900) 21 ZRG,  RA 212 at 255-271; and attributed to Accursius by Weimar. 198  Odofredus Bononiensis, Summa in usus feudorum (1584). There has been no reprint. 199  In J Archer, “Notes sur le droit savant du moyen âge” (1906) 30 RHD 146. 200  Jacobus de Ravanis, Summa feudorum, ed C Pecorella, 2nd edn (1959). 201  C Bezemer, “Jacobus Balduini: Probably the Author of the Summa feudorum parmensis” (2006)  74 TvR 325. 202  Meijers, Études III 159; D Maffei, Giuristi medievali e falsificazioni editoriali del primo Cinquecento [= Ius Commune Sonderhefte X] (1979) 58-59. The Lyon, 1517 edition was reprinted in  1970. 203  The Summa is printed in TUJ X.1 1. Meijers doubted the authorship of Mar tinus Syllimanus:  Meijers, Études III 178.

52

the creation of the ius commune

Andreas de Iserna, a professor and advocate in Naples, whom later jurists  describe  as  the  champion  of  the  feudal  lawyers,  completed  his  extensive  Commentaria in usus feudorum  (with  a  long  introduction  ending  with  a  discourse  on  the  authority  of  the  Libri feudorum).  Several  sixteenth-  and  seventeenth-century  printed  editions  exist  of  this  work.  Andreas’  collegue  Jacobus de Belviso wrote a short lecture on the Libri feudorum.204 Bartolus  and Baldus (who criticised the style of Andreas de Isernia) used it extensively  for their commentaries on the Libri feudorum.205 On  feudal-law  problems  the  commentaries  to  the  Libri feudorum  are  not the only works to be consulted. There are other loci classici to discuss  feudal  matters,  for  instance  D  41.2.3.5  (Paul,  Edict  54)  and  J  Inst  4.6  (de  actionibus).206  Some  of  the  quaestiones and  many  of  the  commentators’  consilia concern feudal law. H. A STRATEGy FOR RESEARCH For the glossators and canonists there existed loci ordinarii (or leges solemnes)  to discuss questions of law. In their glosses the medieval jurists referred to  similar and contrary texts in the Corpus iuris civilis and Canonici, but as a  whole  these  references  turn  out  to  be  limited  in  number.  Cross-referring  shows a limited, more or less fixed number of texts. By reading all these references, one eventually returns to the same principal texts, the loci communes, the  sedes materiae  where  a  specific  question  of  law  is  discussed.  The  loci communi turn out to be the same in the works of the glossa tors and commentators, though the emphasis may shift from one of these texts to another. For  the canonists the sedes materiae shifted in the thirteenth century from the  Decretum  to  the  decretals.  Once  the  modern  reader  is  aware  of  the  sedes materiae, he can relatively easily follow the development of the legal doc trine  throughout the ages. The fact that there is a specific sedes materiae to be found for each legal  pro blem does not, however, preclude the collision of doctrines. The casuistic  texts allowed various interpretations. Questions such as whether transfer of  204 Printed in Lyon in 1511 along with his commentary to the Authenticum, it is now available in  a reprint: Jacobi de Belviso Aurea lectura ... summam autenticorum consuetudinesque et usus feudorum elucidans in OIR XII. 205  The printed editions of Bartolus works include a lectura authenticorum (of which his authorship is put in doubt); those of Baldus include a lectura feudorum. On the latter, see C Danusso,  Ricerche sulla Lectura feudorum di Baldo degli Ubaldi (1991). 206  On D 41.2.3.5, see Meijers, Études III 3, 261 (Johannes Bassianus); on J Inst 4.6, see Van SoestZuurdeeg, La lectura (n 104) 344-347 (Jacques de Revigny).

the sources of medieval learned law

53

ownership requires a causa, whether Roman law knows of a general remedy  for unjust enrichment or whether canon law grants the third party a remedy  when  a  contract  is  concluded  in  his  favour  remain  questions  for  all  times.  In  their  glosses  and  commentaries  the  jurists  discussed  the  various  views  expressed by their prede cessors and contemporaries. There is even a special  type of literature devoted to the controversies among the glossators: dissensiones dominorum. Gustav Hänel publish ed three twelfth-century collections  of issues on which the Four Doctors differed in their opinions.207 Two other  collections date from the be ginning of the thir teenth century – one of which  (completed  shortly  after  1216)  is  ascribed  to  Hugolinus.208  A  seventeenthcentury pendant was published in Venice by Antonius Fachinaeus, Controversium iuris.209 Many of the dissensiones referred to a dissenting view by Martinus. He  contradicted Bulgarus’ strict interpretation of Roman law, invoking natural  equity as his main argument, for instance granting a claim in case of unjustified  enrichment through an extensive interpretation of D 3.5.5.5 (a case of mala fide unauthorised management of another’s affairs). Of the next generation  of glos sators, as the collections composed around 1200 show, some followed  Martinus and others Bulgarus. This fact gives rise to the hypothesis that there  were two “schools” among the glossators: one going back to Bulgarus, to whom  Justinian’s codification in itself was sufficient to solve all legal problems, and  another beginning with Martinus, who sought to harmonise Roman law with  the principles of canon law.210 Indeed the collection ascribed to Hugolinus  spoke of Martinus “et eius sequaces”. Azo and Accursius referred to “Gosiani”.  What was meant is not certain: glossators who adopted Martinus’ interpretation of a specifi c text, or followers of Martinus, who adhered to his view in  general? And who were they? Of the younger generation, Johannes Bassianus  unreservedly sided with his teacher Bulgarus, his contemporaries Aldricus,  207 The two oldest collections date back to the times of the Four Doctors: an anonymous collection  (Vetus collectio) and one ascribed to Rogerius in the sixteenth century (collectio Rogerii). The  author of the third collection (collectio codicis Chisiani) also referred to glossators of the next  generation, such as Pilius and Placentinus, but not to Azo. All three are edited by G Hänel,  Dissensiones dominorum sive controversiae veterum iuris romani interpretum qui glossatores vocantur (1834; repr 1964). Of the collection attrib uted to Rogerius, several sixteenth-century  printed editions exist and one by Haubold (1821). 208  V  Scialoja  (ed),  “Di  una  nuova  collezione  delle  dissensiones  dominorum”  (1888)  9  Studi e documenti di storia e di diritto 249; (1890) 11 Studi e documenti di storia e di diritto 417; (1891)  12 Studi e documenti di storia e di diritto 241, reprinted in V Scialoja, Studi giuridici (1934) II  327-413. The collection ascribed to Hugolinus is edited in Hänel, Dissensiones (n 207). 209  A  Fachinaeus,  Controversarium juris libri tredecim. In quibus omnes fere quaestiones praecipuae. Nunc primum in Italia excusi (1609). 210  See Cortese, Grandi linee (n 61) 262.

54

the creation of the ius commune

Lothair, Albericus de porta Ravenate are referred to as “Martini sequaces”.211  As to whether Rogerius (d ca 1170) sided with Martinus, Gouron has recently  explained that this is the case, where he used the summa Trecensis212 (and Lo codi)213 in his summa on the Code.214 Pilius and Placentinus often preferred  Martinus’ interpretation of texts, but not always. The boundaries between the  two “schools” were not very sharp. Consequently,  the  student  who  wants  to  study  the  medieval  history  of  a certain doctrine should first establish the sedes materiae. Coing provides  a  useful  introduction  to  the  relevant  Roman-  and  canon-law  sources  on  each  subject  he  dis cusses  in  his  Europäisches Privatrecht  (1500-1800).215  Fifteenth-century indices and repertories provide more detail. The indices to  the ordinary gloss, Odofredus’ lecturae, and to the works of Durand, Bartolus  and  Bal dus  have  already  been  mentioned.  A  classical  tool  is  the  Repertorium iuris copiosum of Johannes Bertachi nus (1448-1497) on both Roman  and canon law.216 Much older, but not always reliable is Albericus de Rosate  (1290-1360), Dictionarium utriusque iuris.217 The second step is reading the ordinary gloss to the particular text. Reading  the text implies solving the allegationes, which refer to similar or contrary texts,  and reading the glosses to those texts too. If the glosses happen to contain  references to earlier writers, one can almost be certain that a reconstruction  of the genesis of the doctrine in the twelfth and early thirteenth centuries  is  possible.  At  the  time  Accursius  wrote  his  gloss-apparatus  to  all  parts  of  the  Corpus iuris,  Bartholo meus  Brixiensis  revised  the  glossa ordinaria  to  Gratian’s Decretum, and Ber nardus Parmensis composed a commentary to  the new Liber Extra, which would become the ordinary gloss to this collection of decretals. It  is  usually  helpful  to  compare  the  gloss  with  works  of  contemporaries:  Odo fredus,  Innocent  IV  and  Hostiensis.  The  lectures  of  Odofredus,  for  211 See Collectio Hugolini (n 208) no 53. 212 See text at C.(3) above. 213 This summa was composed in the Provence around 1260. The Latin translation is published by  H Fitting (ed), Lo Codi in der lateinischen Übersetzung des Ricardus Pisanus: die provenzalische Fassung der Handschrift A (Sorbonne 632.) Vorarbeiten zu einer kritischen Textausgabe, (1906; repr 1974). The text is available via www.uni-tuebingen.de/kabatek/codi/index.html 214  A Gouron, “Lo Codi, Source de la somme au Code de Rogerius”, in Satura Feenstra (n 95) 309  repr in his Études (n 88). Whether Rogerius’ summa copies parts of Lo Codi or vice versa is not  certain. 215  H Coing, Europäisches Privatrecht (1985) I: Älteres Gemeines Recht, 1500 bis 1800. 216  There  are  many  similar  repertories,  for  instance  M  Boldewan,  Loci communes iuris civilis  (1551); Gemellus, Sedes materiarum totius iuris utriusque (1598); A Barbosa, Loci communes iurisprudentiae (1737). 217  The Lyon, 1539 edition was reprinted in 1984.

the sources of medieval learned law

55

instance,  cover  all  parts  of  the  Digest  and  Code.  Though  the  works  of  Odofredus never reached the authority of the Accursian gloss – its academic  status became almost exclusive – a comparison of the two is fruitful, because  he referred to earlier glossators far more than Accursius did. A  third  step  is  reading  the  sedes materiae  in  the  commentaries  of  their  pre decessors:  Azo  and  Hugolinus,  Johannes  Teutonicus,  Laurentius  Hispanus  and  Tancred.  Unfortunately,  the  printed  editions  of  their  works  only cover parts of the Corpus iuris civilis and canonici: the lectures of Azo  – of a later date than his famous summa Codicis – followed the order of the  Code. Commentaries to the decretals of pope Innocent III – to the Compilatio III and to the canons of the Fourth Lateran Council – have been printed  recently. Commentaries to earlier decretal collections are con tained only in  manuscripts. Johannes’ gloss to the Decretum, completed in 1216/1217, was  revised because of the promulgation of the Liber Extra, but gives many references to the early decretists. With regard to the glossators, the fourth step is verifying whether the sedes materiae appears in the dissensiones dominorum and in another book, not  yet mentioned, the Casus Codicis of Guillelmus de Cabriano.218 Both take  us back to the middle of the twelfth century, the days of the Four Doctors,  and offer an insight into the jurisprudence of those days. The Casus Codicis are probably based on Bulgarus’ lectures on the Code.219 They show which  texts were treated, thus giving an impression not only of what Bulgarus found  most important or interesting for his students, but also of the opinion held  in Bulgarus’ school and of his way of reasoning. Starting with the ordinary  gloss to the Corpus iuris civilis and canonici, it is thus often possible to read  legal history backwards. It should be noted, however, that the twelfth-century  sources are often hidden in manuscripts and rare editions. A final caveat is necessary. In manuscripts some care was taken to indicate  the original author of a gloss by placing a siglum after his gloss – such as “bul.”  for Bul garus, “Mar.” for Martinus, and “az.” for Azo.220 The sigla, however,  could easily be mis read or attributed wrongly,221 and, a gloss, after revision,  218 T Wallinga (ed), The Casus Codicis of Wilhelmus de Cabriano [= Studien zur Europäischen  Rechts ge schichte CLXXXII] (2005). 219  T Wallinga, “Bulgarus, un des quattuor doctores”, in Les grand juristes [= Actes des journées  inter nat ionales d’histoire du droit Aix-en-Provence, 22-25 mai 2003] (2006) 21. 220  W H Bryson, Dictionary of Sigla and Abbreviations to and in Law Books before 1607 [Virginia  legal studies series] 2nd edn (1996). 221  In  his  edition  of  Irnerius’  glosses  to  the  Digestum Vetus,  Besta  attributed  the  sigle  Yr  (of  Henricus de Bayla) to Irnerius/Guarnerius (regularly: Y or G). Consequently he missed a lot of  material. E Besta, L’opera d’Irnerio (1896; repr 1980) II “Glosse inedite d’Irnerio al Digestum  Vetus”.

56

the creation of the ius commune

may bear the reviser’s siglum instead of that of the original author; glosses  may also have been copied without the author’s “siglum”.222 i. CONCLUSiON The editors of this volume were of the view that the cupida legum iuventus  might feel the need for a short, updated guide through the foresta selvetica of  the learned law of the Middle Ages. We now have reached the end of our tour  through the wild woods. It only remains to conclude. The sum of it all is that  medieval Roman law, like beauty itself, is a joy forever. We feel sympathy with  Alexander the Great, who, according to Aulus Gellius (Attic Nights xx, 5) told  Aristotle that he would rather be first in learning than in wealth and power.  Had he lived in the twenty-first century he would without doubt have loved  Roman law. From the other root of medieval scholarship, the Biblical tradition, we find an answer to the question why “Happy is the man that findeth  wisdom and the man that getteth understanding. For the merchandise of it  is better than the merchandise of silver, and the gain thereof than fine gold”  (Proverbs 3, 13-14). We hope the reader will share with us the eternal beauty  of the Roman and canon law of the Middle Ages. .

222 See Dolezalek, Repertorium (n 41) 460-524; G Dolezalek, “Libri magistrorum and the transmission of glosses in legal textbooks”, in V Colli (ed), Juristische Buchproduktion im Mittelalter  (2002) 315.

2  The Infrastructure of the Early Ius Commune: The Formation of Regulae, or its Failure Kees Bezemer A. INTRODUCTION B. FROM CASUS TO REGULAE: THE CONCURRENCE OF ACTIONS C. FROM THE LEX AQUILIA TO THE CONTRACT OF SALE D. THREE DIFFERENT ARGUMENTS FOR A CASE OF TOLL COLLECTION E. LIMITS TO THE FREEDOM TO ARGUE F. THE IMPACT OF REVIGNY AND BELLEPERCHE IN FRANCE: JEAN FAURE G. ITALY DIVIDED: CINUS VS DINUS H. REGULAE AS CONCLUSION OF A COMMENTARY: JACOBUS BUTRIGARIUS I. CONCLUSION A.  INTRODUCTION All interpretative activity necessarily begins with the explanation of words. This is what the medieval jurists first did when they started to explore the corpora of Roman and canon law. Next to this “literal” approach, they soon applied other exegetic methods, such as indicating texts containing identical (similia) or contrary (contraria) solutions to a legal problem or casus. These similia and contraria, originally listed in the margins of the manuscripts at a fixed distance from the main text, presented a challenge to inventive legal minds: was it possible to explain the discrepancies and reduce these to apparent contradictions? The method chosen for this end was usually the distinction (distinctio). Harmony was attained by a reduction of all the relevant texts to a “tree” of casus, of variable complexity, in which all the 57

58

the creation of the ius commune

aspects of a specific subject matter were supposed to have found their place. An alternative method, also aimed at reduction of the texts pertaining to a  specific matter, was to develop a regula or regulae. In this case the stress was  less on the analytical process of explaining (apparent) differences, and more  on achieving a synthesis that found a rule or rules that connected casus at a  higher level. This way it was sometimes possible to create more order in areas  of law that had not – or not sufficiently – been the object of this technique  of  generalisation.  There  was  a  title  in  the  Digest  on  the  point  (D  50.17),  consisting of 211 regulae iuris antiqui (rules of the old law, that is, the law  of the so-called classical period). This title received a lot of attention from  the glossators, and served both as an example and as an invitation to develop  new regulae.1 It is the development of the latter that will be discussed in this  contribution. To clarify, this chapter will discuss the designation with the word “regula”  of  rules  not  found  described  as  such  in  the  authoritative  texts.2  Examples  will  be  taken  from  different  areas  of  private  law.  The  emphasis  will  be  on  Roman  law  and  its  interpreters  between  1200  and  1400.  The  year  1200  is  chosen because around this date the academic jurists became aware of the  need  to  integrate  law  other  than  the  Roman  into  a  “system”,  which  later  was to become known as the ius commune. It should also be observed that  in  around  1300  canon  law  provided  new  impetus  to  the  formation  of  new  regulae  through  the  promulgation  of  the  Liber Sextus  in  1298.  This  was  a  codification of canon law, and it contains eighty-eight regulae iuris at the end  of its fifth and last book. Of course, only a few examples of the formation of  regulae (or the failure to form regulae) in the period here examined can be  given, and there can be no claim of being complete. The  development  of  regulae  was  closely  related  to  the  emergence  of  separate  subject  matters  (materiae),  and  it  is  important  to  pay  particular  attention to the role of technical legal citations. The way in which individual  jurists  dealt  with  this  important,  but  rather  neglected,  aspect  of  medieval  legal literature would seem to have had an effect on their readiness to accept  the existence of a regula. Finally, a few words should be said about the situation that emerged after  a new regula had been developed. This was not the end of the debate. The  inductive process leading to the formation of a regula having been completed,  a new phase was entered upon in which the exact scope of the regula had to  be fixed. This happened by listing the casus where it did not apply. In other    1 See, in particular, P Stein, Regulae Iuris: From Juristic Rules to Legal Maxims (1966) 131-152.   2 Cf Stein, Regulae (n 1) 153 (“so-called doctrinal rules”).

the infrastructure of the early ius commune 

59

words, its ambit was defined by formulating the exceptions to its application.  This (deductive) phase generally falls outside the reach of this contribution.  It is important to be aware that this process meant that the texts that had been  instrumental to the formation of the regula acquired a different status. From  this moment on, they were seen in the light of the regula they had helped to  create, and – in this context – lost their separate status as casus, continuing  their existence in the shadow of the regula.  B. FROM CASUS TO REGULAE:  THE CONCURRENCE OF ACTIONS The  first  example  illustrating  the  theme  of  this  volume  concerns  an  area  that received its form between 1200 and 1400, and which shows how casus  can produce regulae. The process can be described as the birth, youth and  adulthood of a doctrine – in this case the doctrine of concurrence of actions.  Tracing  the  development  of  ideas  about  this  subject,  I  have  been  able  to  determine the moment when casus – I mean the cases to be found in the  appropriate  Roman  law  texts  –  have  led  to  the  formation  of  regulae,  and  when, in this case almost simultaneously, the subject matter has become a  separate materia. I have even been able to pinpoint the place of birth of the  doctrine. This provides a good starting-point; it is also an example of the use  of a “search engine” of exceptional quality and rarity. The author who first named the doctrine of concurrence of actions was  Azo (d after 1220). He was the first to speak of cumulatio (literally “accumulation”) in this context. I discovered this by consulting Dolezalek’s Repertorium  at C 2.1.3 Edita actio, a text which at a later stage was to become the home  base (sedes) of the subject. In Azo’s gloss on the words “decernit aequitas”,  the word cumulatio is used to describe the situation of someone who starts  a lawsuit about something that was already the object of legal proceedings.3  Dolezalek found this gloss in twenty manuscripts. In Azo’s (printed) Summa  and Lectura the two instances when cumulatio is forbidden are mentioned,  but  the  word  cumulatio  is  missing.4  This  shows  the  importance  of  the    3 See G Dolezalek, Repertorium manuscriptorum veterum Codicis Iustiniani (1985) 600 no 338:  “Tunc dicitur mutari actio, cum quis vetere relicta aliam (i.e. actionem) intendit. Quod quando  liceat, videamus. Et quidam ante litem contestatam potest, post litem contestatam non potest.  Tunc enim non mutat, quia necesse habet in veteri querela persistere, ut ff. de iudic. De qua re [D  5.1.74] i. resp., et l. Non potest [D 5.1.23]. Non tamen diffiteor quin de alia vel de re eadem agere  possit alio modo, etiam post litem contestatam. Set illa non erit mutatio, set cumulatio actionis”  (italics added).   4  See Summa Azonis (1506; repr 1966) 20B (ad tit C 2.1), and Azonis ad singulas leges XII librorum  Codicis (1577; repr 1966) fol 59r (nos 23-25, ad C 2.1.3).

60

the creation of the ius commune

laborious research in the manuscripts. Azo says that concurrence of actions  is prohibited in the following cases: first, if another action is blocked by the  defence of res judicata (exceptio rei iudicatae); and secondly, if a choice has  to be made between the available actions (elective concurrence). Accursius  (d 1263) repeats the essentials of Azo’s view, as he often did, and more than  once uses the word cumulatio, which marks the beginning of a success story.5 In this story the word regula appears for the first time in the work of Jacques  de Revigny. He distinguishes three situations when concurrence of actions is  not allowed, which he derives from a theory (ars) devised by the Accursian  gloss.6 These situations correspond with the two cases mentioned by Azo. The  third situation – actions contradicting each other – later became separately  known as Dinus’ third regula (see below), but Revigny still attaches the word  regula to the theory of Accursius as a whole, a theory which he considers to  be  most  true,  and  which  he  defends  against  some  doctors  who  wanted  to  phrase differently the first part about the exceptio rei iudicatae.7 The identification of this regula created another mind set. The texts, which formerly  were  separately  mentioned  as  cases  in  which  concurrence  was  forbidden,  have now become the supporting texts of the regula, and connected to each    5 It concerns the gloss “Emendari” ad C 2.1.3 Edita actio.   6 See Revigny’s Lectura super Codice, published under the name of Pierre de Belleperche, repetitio  ad C 2.1.3 Edita actio (1519; repr 1967) fol 50rb: “Sed supposito quod sit admittenda cumulatio,  queritur qualiter sciemus ubi est admittenda, et ubi non. Glosa dat talem artem. Ubicumque actio  primo intentata sic se habet ad cumulandam vel intentandam [i.e. actionem] quod lata sententia  absolutoria in primo intentata non obstaret exceptio in cumulanda vel intentanda [i.e. actione],  nec  sunt  ille  due  actiones  sibi  contrarie  nec  tolluntur  invicem  electione,  tunc  est  admittenda  cumulatio et alias non.” I have checked the three manuscript versions of this repetitio. For this  passage they contain no variants worth mentioning.   7  Revigny ad C 2.1.3 (n 6) fol 50rb: “Supposito quod cumulatio sit admittenda, dico artem istam  verissimam. Doctores quidam regulam cumulationis predictam volunt [ed: nolunt] dare sub aliis  verbis sic. Ubi actiones sic se habent quod lata sententia in una obstat exceptio rei iud. in alia,  non procedit cumulatio. Regula [ed: regule] sub istis verbis proposita sic [ed: si] intelligatur sicut  dictum est quod ubi sunt due actiones sic se habentes [ed. habent] quod lata sententia in prima,  obstaret exceptio rei iud. in secunda. Verum est quod non admitteretur cumulatio. Sed converte:  prima actio sic se habet ad secundam quod lata sententia in secunda obstaret exceptio in prima.  Numquid ipsa admitteretur cumulatio? Dico quod sic, quia prima [i.e. actio] non esset preiudicialis  secunde,  immo  [ed:  ideo]  secunda  prime.  Unde  dico  quod  sic  debet  dari  regula:  ubi  sunt due actiones, et prima sic se habet ad secundam quod, si esset lata sententia absolutoria in  prima obstaret exceptio rei iud. in secunda, non admittetur cumulatio per l. preall. ff. de excep. l.  Fundum et l. Fundi [D 44.1.16 and D 44.1.18]” (italics added). The emendations are based on the  manuscript versions of this passage. See fol 50vb: “Sed numquid econverso actioni [ed. actione]  ex confessione prima intentate possit cumulari actio prima ex vendito? Dico quod secundum [ed:  secunda] regulam cumulationis est quod ubi sententia absoluta lata in prima obstaret exceptio in  secunda, non valet cumulatio quia prima est preiudicialis secunde , ut ff. de excep. l. Fundum et l.  Fundi [D 44.1.16 and D 44.1.18]” (italics added). See also L J van Soest-Zuurdeeg, La lectura sur le titre De actionibus (Inst. 4,6) de Jacques de Révigny (1989) 190 (lines 125-128 and 140-143),  199 (lines 416-421), 341 (lines 162-177), and 342 (lines 193-197).

the infrastructure of the early ius commune 

61

other. It was a matter of time before someone formalised this aggregate of  texts on a higher level. The  man  who  took  the  step  was  Dinus  de  Mugello  (d  1303?).  With  hindsight this step may seem obvious and unsurprising,8 but it was indeed a  significant advance. In his treatise On the Rules of Law (De regulis iuris), on  the title of that name in the Liber Sextus, Dinus discusses our subject matter  at the twentieth rule: Nullus pluribus uti defensionibus prohibetur (“No one  should be forbidden to use more than one defence”).9 Here, Dinus speaks of  a materia cumulationis.10 He appears to be the first to do so. From then on  the subject matter had achieved a status that made it fit for separate treatment in teaching and writing. After Dinus we see that the new status of what initially were a couple of  casus  now  triggered  questions  about  the  regulae  that  previously  had  been  posed about these casus. The question about the ratio legis, once asked about  the single texts that constituted the basis of a rule, was now put on a different  level: “what is the reason underlying this rule?”11 Another effect of the new situation was that some authors increased the  number of rules in an attempt to reach an exhaustive treatment of the subject.  Bartolus (1314-1357) did so. He arrived at a total of six rules.12 Closer exami  8 The influence exerted by Revigny on Dinus is still largely unexplored. Meijers states that both  Dinus’ treatise De actionibus and his treatise De regulis iuris contain opinions of Revigny, usually  anonymously. See Meijers, Études III 118-119 (and the “note des éditeurs” at n 419).   9  This treatise can be consulted on the internet on the website of the French National Library. See  http://gallica.bnf.fr, Tractatus Dyni super regulis juris. There is one drawback: the edition, which  is that of Paris, 1498, is not foliated. 10  Before  he  starts  with  his  lengthy  treatment  of  the  subject  matter  Dinus  makes  the  following  remark:  “Sed  an  per  viam  cumulationis  proponere  poterit  plures  actiones?  Quidam  non  intelligentes  cumulationis  materiam  dicunt  simpliciter  quod  non,  ut  in  predicto  §  Quotiens  [D  50.17.43.1]; et ideo tractatus [ed: male contractus] de cumulatione communiter valeat et sit utilis  apud canonistas sicut apud legistas, quia secundum canones debet exprimi res et causa, licet non  actio. Numquid pendente primo iudicio licebit eandem rem ex alia causa petere vel licebit aliam  rem petere ex eadem causa. Hec enim est cumulationis natura. Ad huius declarationem et intellectum querendum est an cumulatio sit admittenda, et in quibus casibus” (italics added). 11  See,  e.g.,  Jacobus  Butrigarius,  Lectura super Codice  ad  C  2.1.3  (1516;  repr  1973)  fol  42vb:  “Expedita materia mutationis modo venio ad cumulationem, [et quero que sint regule impedientes cumulationem (omitted in original owing to homoioteleuton)] et ponam exempla earum.  Secundo que sint rationes illarum regularum. Tertio que regule cumulationem firment et defendant, distinguam ergo. Et primo quero que sint regule impedientes cumulationem? Respondeo  sunt tres.”  12  Bartolus de Saxoferrato, Commentaria in primam Codicis partem ad C 2.1.3 Edita actio (1574)  fol 39vb: “Dat [i.e. Dynus] tamen tres regulas in quibus cumulatio est prohibita, de quibus duas  regulas  hic  [ed:  haec]  glosa  ponit.  Prima  est  …  Secunda  regula  est  …  Addit  Dynus  tertiam  regulam: quando sunt plures actiones contrarie, non admittitur cumulatio, ut l.i. infra de fur. [C  6.2.1], l. Eum qui & l. Contra maiores infra de inoffic. test. [C 3.28.14 and 16]. Dices tu: ista non  est nova regula. Nam si sunt contrarie electione tolluntur, ut d. l.i. [fol 40ra] infra de fur. [C 6.2.1].  Respondeo etc … Iste sunt regule Dy. & gl. Ego addo alias tres regulas impedientes.”

62

the creation of the ius commune

nation of the additional three reveals that they are merely refinements of the  others. In this respect, therefore, I agree with Bartolus’ contemporary, the  practician  Albericus  de  Rosate  (d  1360),  who  wrote  that  Dinus’  treatment  is adequate for practice (in northern Italy), and that the rest belongs to the  world of hallucinations.13 Bartolus’ pupil Baldus (1327-1400) maintained the  three rules of Dinus, but in response to the refinements of Bartolus devised  a new classification of the types of concurrence of actions.14 Surveying the development of the ideas about concurrence of actions from  Azo to Baldus, we see above all the formation of a separate segment within  the  ius commune,  with  its  own  rules  based  on  a  limited  number  of  texts.  Future discussion was now to be centred around the rules, pushing the texts  that had been at their origin into the background. We enter the world of the  limitationes and extensiones of the rules, which requires a different analysis  of the texts because their weight has been reduced in comparison with the  rules that emerged from them. C. FROM THE LEX AQUILIA TO THE CONTRACT OF SALE Now I shall give some examples where a detailed analysis of the citations used  has proved to be quite illuminating, although the development of a regula did  not take place. These have been selected from Revigny’s lectura on the Code.  This is partly because it is a source with which I am very familiar; but also,  and more significantly, because studies of legal development in the middle  ages have shown Revigny to be an author who introduced new arguments and  texts to the discussion of issues that previously seemed to have been settled  by the Accursian gloss.15 The examples concern the duty of disclosure: is a seller obliged to inform  his  buyer  about  certain  defects  of  the  object  sold?  Revigny  developed  a  13 See Albericus de Rosate, Commentarii in primam Codicis partem ad C 2.1.3 (1586; repr 1979)  fol  73ra  no  15  (i.m.):  “de  cumulatione  antiqui  et  Moderni  Doctores  longas  filaterias  fecerunt,  sed tandem vidi teneri quod plene notatur per dominum Dynum extra de reg. iur. cap. Nullus  pluribus, libro vj. Et ibi etiam per Ioannem Andr. et de elect. cap. Ut quis duas [Sext 1.6.25],  et per Azonem in summa de ordi. iudi. § i. Et licet cumulationis materia sit subtilis, raro tamen  vidi contingere in practica, quia advocati nolentes intricare processus communiter procedunt in  prima, demum in alia instantia proponunt quod cumulari posset.” 14  Cf. Baldus de Ubaldis, In primum, secundum, & tertium librum Codicis commentaria, repetitio  ad C 2.1.3 Edita actio (1599; repr 2004) fol 95vb no 78: “Et licet Dom. Dyn. tractet hanc materiam  in c. Nullus pluribus de reg. iu. lib. 6, alto et pulchro stylo: ego tamen istam breviter trado, dicens  quod triplex est cumulatio, sc. permissa, non permissa, et mixta.” According to this edition, the  repetitio was given at Padua in 1380 (fol 92rb). 15  See, e.g., T Repgen, Vertragstreue und Erfüllungszwang in der mittelalterlichen Rechtswissenschaft (1994) 137-148, and W Ernst, Die Einrede des nichterfüllten Vertrages (2000) 88-90.

the infrastructure of the early ius commune 

63

theory on the subject that reveals a way of arguing which is not immediately  obvious without a thorough analysis of the citations. During a lecture Revigny said that he witnessed a dispute on the question  “what is a patent defect?” He at once gave a detailed answer, in which two  things stand out: first, the importance of the social position of the contracting  parties; and secondly, the knowledge of each party about the defect. Revigny  first  discusses  a  cloth  merchant.  If  he  sells  cloth  to  a  simple  student,  the  information he should give to the buyer is different from when he sells to  a professional buyer of cloth. And, Revigny adds: “I also consider how the  buyer has touched and inspected the object bought.” To illustrate this he tells  a story from his own experience: Once I saw that someone had bought a bird with a wound under its wing. The  buyer said to the seller: “My friend, you have deceived me, you knew it had this  wound under its wing.” Says the seller: “Yet I am not liable to you, because the  defect was patent.” I [i.e. Revigny] say that a seller who knows of the defect is  liable. Therefore in case of a sale I consider how the buyer has inspected the bird,  whether he has looked under its wing. And if the seller has seen that the buyer did  not inspect the bird under its wing, it should be clear to him that the buyer was  ignorant of the defect. If, in addition, the buyer has not spoken of a defect, it is  clear he was in error. Therefore the seller is liable: he was obliged to inform him  from the moment he was aware of the defect.16

This  story  was  not  only  a  means  to  keep  the  attention  of  his  audience.  The detailed description of what happened was necessary to show that the  seller in this case had acted deceitfully, whereas the buyer was only negligent in his inspection of the bird. These facts resulted in the application of  the “rule” that the dolus of the seller outweighed the culpa of the buyer. As  16 Revigny, Lectura super Codice ad C 4.58.1 Si non simpliciter (n 6) fol 206vb: “Sequitur videre ex  parte emptoris. Emptor pro vitio quod scivit non aget, nisi fuerit stipulatus. Si autem ignoravit  vitium, agit nisi fuerit error supinus, ut ff. de contrahenda emptione l. Si ea que [D 18.1.43]. Dico  quod si venditor scivit vitium et emptor ignoravit quantumcunque supine, venditor tenetur ei ex  quo sciebat; alias scire debebat emptorem hoc ignorare. Unde ex quo patitur eum errare, tenetur  si  instruxit eum, quia in dolo est venditor. Dolus autem preponderat culpe quantecumque,  ut ff. ad leg. aquil. Idem si obstetrix in fine [D 9.2.9.4] et ff. de actio. emptio. l. Ex empto § Si  quis virginem [D 19.1.11.5]. Sic ergo si vitium sit patens, [fol 207ra] non tenetur venditor. Ego  vidi  dubitari  quid  diceretur  vitium  patens.  Dico  quod  ego  considero  conditionem  contrahentium; unde dico quod venditor pannorum alio modo debet certiorare scolarem simplicem, alio  modo debet certiorare illum qui expertus est in emptione pannorum. Et etiam considero modum  tangendi et videndi rem que emitur. Ego vidi semel quod quidam emerat avem habentem sub ala  plagam. Dicebat emptor venditori: ‘Amice, tu me decepisti; tu sciebas eam habere plagam sub  ala.’ Dicit ille: ‘Immo ego non teneor vobis, quia vitium erat patens.’ Dico quod venditor sciens  tenetur. Unde in venditione ego considero quomodo inspexerit, utrum viderit eam sub ala. Unde  si venditor vidit quod emptor non inspiciebat avem sub ala, debuit scire eum ignorare vitium.  Unde si non dixerit vitium, patens fuit eum errare. Unde tenetur: debuit enim eum certiorare ex  quo sciebat vitium.”

64

the creation of the ius commune

a  consequence  the  seller  had  to  inform  the  buyer  about  the  wound.  This  seems fair enough. But it is important to look at the texts cited. The first is  the famous text about the javelin throwers from the Digest’s title on the lex Aquilia (D 9.2.9.4). Revigny probably had in mind the last sentence about  a  slave  who  carelessly  crossed  a  field  where  people  were  exercising  at  the  javelin, and someone deliberately aimed at the slave and killed him. In that  case the dolus of the javelin thrower outweighed the negligence of the slave,  making the javelin thrower liable to the owner of the slave. It is important to  notice that this text is about the law of delict, not about the contract of sale.  Revigny could transplant it because the boundaries between these areas of  law had not yet attained the impermeability of later times. The second text  cited  (D  19.1.11.5),  about  a  seller  who  knowingly  allows  the  buyer  to  err  about a defect of the object sold (a woman slave is not a virgin as the buyer  thinks) is indeed a good example of dolus on the side of a seller.17 This freedom to cite texts from apparently any part of the legal corpus is  typical of Revigny. But not all authors felt as free as he did. I shall return to  this issue below (see section E.). The preceding cases should be compared to a similar case about wine sellers  who sold wine to students in Orleans. Not every seller told his customers that  the wine had to be drunk before Christmas. At the time, methods for the  preservation of wine were still primitive. A seller had to inform the students  about the nature of the wine, Revigny says.18 In this case, there was no need  for subtle ways of reasoning. There is a text (D 18.6.16(15)) which says that a  wine seller is obliged to warn his buyer that after a specific date the quality of  the wine will deteriorate. If he does not, it is considered to be dolus. There  is no mention of culpa on the side of the buyer, but we would place this case  also under the heading “duty of disclosure”. Revigny does not make a connection with the other case. Anyhow, his way of arguing in that case did not make  a connection easy. “Duty of disclosure” had not yet developed into a distinct  subject matter (materia) with its own rules.

17 Revigny will have cited it for the passage “et sciens errare eum venditor passus sit”. 18 Revigny, Lectura super Codice ad C 4.48.3 Dolum auctoris (n 6) fol 203ra: “Isti rustici vendunt  quotidie vina scholaribus; isti rustici bene sciunt naturam vini quod non potest servari ultra natale,  et si servetur ultra natale non valet aquam. Scholaris tale vinum emit a rustico bona fide. Rusticus  tacuit naturam vini. Scholaris servavit vinum suum, deinde invenitur penitus inutile. Iste dolus  rustici  non  nocet  scolari,  rusticus  enim  debuit  ei  dicere  naturam  vini,  ut  ff.  e.  Si  vina  que  [D  18.6.16].”

the infrastructure of the early ius commune 

65

D. THREE DIFFERENT ARGUMENTS FOR A CASE OF TOLL  COLLECTION The next example has been selected to show that in the art of arguing, not  only  (modern)  boundaries  in  the  domain  of  private  law  were  ignored,  but  also those between private and public law. It is another example drawn from  the colourful cases mentioned in the commentaries of Jacques de Revigny. The abbot of a French monastery was staying in England.19 He charged  his prior to dispatch some wine to him for his personal use. While the wine  was already at sea, the abbot apparently found another way to cater for his  needs, so he saw the opportunity of making a good sale in England (in another  commentary Revigny introduces a wine seller who says that in England wine  is very expensive, “as there is nothing but beer”).20 The messenger, who found  out that the wine was already at sea, said it was to be sold. Having heard this,  the French tax collector said: “You will have to pay the tolls, as it is no longer  carried to your abbot.” The messenger responded: Initially we intended the wine to be shipped for the personal use of the abbot, yet  because of a supervening event we are forced to sell it, to prevent any harm being  done. So we are not obliged to pay the tolls.

Revigny sets out three arguments. The first is taken from the lex Caesar (D  39.4.15), a text from the title on tax collectors, duties and confiscations, which  he had seen cited. This states that a ship loaded with stones had left Crete  19 See his commentary on D 39.4.16 Interdum in Paris BN MS lat 14350 fol 53vb: “Legem istam  Cesar [D 39.4.15] vidi allegare [allegari?]. Lex dicit quod de rebus quas ducit quis ad usum suum,  non solvit pedagium, C. e. l. Universi [C 4.61.5]. Abbas qui erat in anglia mandavit priori qui erat  in francia mittere sibi vinum, et sic ad usum suum vehebatur. Deinde, cum vina in mari essent,  vina  illa  emerunt  .  Habuerat  vina,  unde  expediebat  sibi  quod  vina  illa  venderentur.  Nuncius invenit vinum in mari et dixit: ‘Vendamus illa vina.’ Dicit exactor pedagiorum: ‘Solvatis  pedagia,  ex  quo  non  portatis  ad  illum  abbatem.’  Dicit  nuncius  ideo  illius:  ‘Propositi  sumus  ab  inicio ut ad usum abbatis duceremus ista vina, set ex supervenienti causa propter necessitatem  dampni vitandi oportet quod nos vendamus ea; unde non tenemur ad pedagium.’ Argumentum  est  lex  illa  Cesar  quia  ab  inicio  ceperunt  navigare  pro  causa  questus,  scilicet  propter  usum  abbatis.  Ex  facto  tamen  mutaverunt  consilium  propter  necessitatem  dampni  vitandi.  Est  ergo  inspiciendum propositum quod ab inicio habuerunt, non propositum quod post habuerunt, quia  hoc fecit necessitas dampni vitandi, sicut in lege illa Cesar spectandum est inicium; et sic fuit  iudicatum per legem aliam que non multum facit C. de coll. do. l. Si filia fuerit [D 37.7.3]. Lex  videtur  dicere  oppositum.  Quidam  habebat  concubinam.  Causa  illius  paraverat  vestes.  Habuit  servam. Tradidit ei vestes prioris. Deinde dicit de secunda: ‘Concubine mee lego vestes paratas  causa  eius.’  Numquid  habebit  vestes  prioris  concubine?  Videtur  quod  non,  quia  non  fuerunt  parate  ad  usum  suum,  set  ad  usum  prime.  Dicitur  quod  ymo  non  inspicietur  inicium.  Sic(ut)  videtur, licet essent parata vina ad usum abbatis, quod non spectetur illud inicium, set illud quod  postea venalia sunt, supra de l. iii. l. Qui concubinam [D 32.1.29].” 20  The remark about the price of wine in England can be found in his repetitio on D 12.1.22 Vinum.  See K Bezemer, What Jacques Saw. Thirteenth-Century France through the Eyes of Jacques de Revigny, Professor of Roman Law at Orleans (1997) 121 and 125.

66

the creation of the ius commune

before the coming into force of a decree of Caesar restricting such transports  to leaseholders. It was then driven back to the island by the wind, before later  setting out once more after the decree had come into force. The text rules  that the time of the first departure was the material one and the decree did  not apply to its cargo. This text could be analogously applied to the case of  the abbot: in both cases the original intention decided the matter. “And so  it was judged,” states Revigny, “by another text which has little force.” The  other text is D 37.7.3 Si filia fuerit, which deals with collation of dowry. It is  impossible to find a satisfactory explanation of the argument contained in this  text, though any suggestions would be welcome. That Revigny, never hesitant  about daring analogies, considered the text to be of little value to our problem  indicates that it will not be easy to find the reasoning underlying it.21 Yet,  Revigny  would  not  be  Revigny  if  he  did  not  have  another  text  up  his sleeve, and in this case he provides an argument against the wine-loving  abbot taken from a text on the law of legacies. It is the lex Qui concubinam  (D 32.1.29). This deals with the position when a testator, who had once had a  concubine for whom he had ordered clothes to be made, had afterwards given  these clothes to her successor, and then wrote in his last will: “To my concubine I bequeath the clothes made for her.” The text states that the clothes  had to remain with the second concubine, because what was important was  the intention of the testator, not the original destination of the clothes. Transposed to the case of the abbot, this means that what was relevant was the  eventual destination (sale) of the wine, not its original destination (personal  use of the abbot). It is an argument. Most of us nowadays would probably  not accept that a text from the law of legacies could be used to solve a fiscal  problem. To Revigny and some of his contemporaries it was a sound argument  by analogy, although it takes some time to work out how. It is obvious that this  free way of reasoning is not very apt to produce internally coherent doctrines,  let alone regulae based on Roman law texts. Revigny must have realised this. E. LIMITS TO THE FREEDOM TO ARGUE Revigny’s  very  free  way  with  the  texts  soon  provoked  criticism  and  inevitable reaction. It started with his younger colleague Pierre de Belleperche (d  1308), who more than once complained about the liberties Revigny permitted  himself in handling the texts. 21 It is interesting to  consider  whether  the  abbot’s  case  was  a  real  case,  or  simply  a  hypothetical  object of an academic dispute? I am inclined to say it was a real case because certain terms typical  of Revigny when he describes academic disputes (“questio”, “disputari”) are not present.

the infrastructure of the early ius commune 

67

A good example is what Belleperche said about one of the many cases in  which Revigny dealt with the custom of retrait lignager, the right of the kin of  a seller of family property to retrieve the property in exchange for payment of  the price and any expenses incurred. It was during an academic dispute that  the following problem was posed. A usurer buys a family property for 100 pounds. At the same time he lends  1,000 pounds to the seller. One of the latter’s kin offers 100 pounds to the  usurer in order to get the property back. The usurer replies that the money  lent  has  to  be  returned  too.  Some  argued  that  the  loan  was  connected  to  the sale by means of a pactum adiectum, and as such was a part of the price.  They based this argument on the lex Fundi (D 18.1.79), which states that, if  someone sells a piece of land on the condition that half of it should be leased  to him for a period of ten years, he can use the actio venditi to enforce the  contract of lease, at least if the land was sold at a lower price to make the  condition acceptable. Revigny, however, used this text to prove that, in retrait lignager, it was  necessary simply to offer the interest on such a linked loan, as it was only  the  advantage  (utilitas)  to  be  derived  from  the  loan  of  money  that  was  to  be considered part of the price. It is quite clear that in practice, the custom  of retrait lignager  would  become  impossible  to  exercise  if  the  other  view  prevailed.  It  is  probable  that  the  development  of  Revigny’s  argument  was  influenced by the desire to defend this custom, which he considered reasonable.22 He could not detach the loan from the sale; but in this way he was able  to reduce its impact on the amount of money to be returned to the buyer by  the kinsfolk of the seller. Belleperche was deaf to arguments of this nature. In his commentary on  another text put forward by Revigny (D 45.1.68), he mentioned the question  and Revigny’s answer to it.23 He began by stating, “I do not know what he  wants to say”, and he concluded that the account of Revigny’s lecture must  have been deficient, which, of course, is not the case. Then he gave further  examples to demonstrate that Revigny’s view was incorrect. His deafness to  the  background  to  that  view  is  epitomised  in  the  comment:  “You  will  see  questions like these disputed by professors. They are not problematic, their  solution lies in the texts (‘casus sunt legis’).” In other words, these questions  had no value for an academic dispute as they can be answered without any  22 Compare his argument against buyers who tried to frustrate the custom by making unnecessary  expenditure on the estates they had bought. See Bezemer, Jacques (n 20) 7 and 12. 23  See K Bezemer, “Ne res exeat de genere, or how a French custom was introduced into the ius  commune” (2000) 11 Rivista internazionale di diritto comune 67 at 79-80 and n 29, where the  entire passage is reproduced.

68

the creation of the ius commune

debate. Belleperche refused to see that Revigny had intentions beyond simply  applying Roman law texts blindly to cases for which a solution was at hand in  customary law. In the long run Revigny triumphed, not because of his subtle  way  of  arguing,  but  because  of  the  number  of  places  where  he  discussed  retrait lignager. The sheer quantity made it possible for later French authors  to develop it into a separate subject matter, with rules based not on Roman  law texts but on customary law. The clash between their views became apparent when Belleperche was  still a baccalarius. Revigny presided over an academic dispute for which he  had invented the following question. A debtor dies before paying his debt,  leaving behind two heirs. One of the heirs wants to pay his part of the (divisible) debt. Is the creditor obliged to accept? According to Roman law, the  solution would seem too simple to be worthy of a dispute. Revigny defended  the opposite by saying that the original debtor would not have been able to  settle his debt by instalments, and that the position of the creditor should  not  deteriorate  because  of  the  emergence  of  more  than  one  heir.  Belleperche, probably dreaming of an easy victory over the formidable debater  Revigny, protested and put forward the Roman law solution. Unfortunately  for  him,  Revigny  had  rallied  some  supporters,  because  an  important  man,  probably one of his former pupils, addressed Belleperche in a rather patronising manner, and explained that some debtors have to pay their debt in its  entirety and others not. It was the end of the debate. Belleperche’s attempt  to corner Revigny had failed. A couple of years later, when Belleperche had  his own school, he said to his students: “At the time, I wished to have paid a  penny to have had this text at my disposal.” He was indicating how little value  he attached to the text(s) adduced by the important man, and indirectly by  Revigny.24 Since I do not believe that Revigny allowed the use of arguments that had  no significance for the question asked, there must be some connection. In  this case, it again seems that a rule of customary law was at the base of the  argument. In France the liability of heirs for debts was linked to the possession (saisine) of the moveables of an inheritance. Only the heir who possessed  these  had  to  pay  the  debts,  and  he  had  to  pay  them  in  their  entirety,  not  according to his proportion of the estate. The text of the important man (C  2.18.3) concerns an heir who was forced to pay the complete debt for which  a charge had been given. This is the sum total of the similarity. Belleperche  24 For the commentary of Belleperche containing the story of the dispute, see K Bezemer, Pierre de Belleperche. Portrait of a Legal Puritan (2005) 40 note 61. The other text cited by the important  man was D 10.2.25.12, and not D 10.2.25.16, as mentioned by mistake.

the infrastructure of the early ius commune 

69

had little patience with arguments of this nature. He wanted more substantial  analogies; but his was an approach that left little room for local law, unless it  was very similar to Roman law. Because Belleperche wanted texts of Roman law to have a substantial and  not just a formal connection to a problem posed, he consequently wanted to  restrict freedom of argument, pulling up the bridges, so to speak, between  certain areas of law. A good example is provided by what he says about an  opinion of his contemporary Jacques de Boulogne (1248-1301), a colleague  for about five years. In a discussion on the way a (royal) decree (rescriptum)  should be formulated if it could affect existing rights or obligations, Boulogne  had  used  a  text  about  legacies  (D  32.1.22)  as  an  argument.  Although  the  analogy between a prince granting a privilege and a testator making a legacy  is not far-fetched, Belleperche points out an important difference: It is not good to argue from last wills to privileges, because the things a sovereign  does are for the benefit of his subjects, whereas a testator acts in his own interests  and his soul’s.

In  this  case  it  meant  that  the  king  of  France  was  not  obliged  to  declare  expressly that he wanted any other rights or obligations to be abolished.25 In  another case it was again Revigny who had to be chided for taking too many  liberties in his way of arguing. The pope had ordered a bishop to provide a  cleric with a benefice on the recommendation of the abbot of Saint-Euverte,  a monastery in Orleans. The abbot had died before any recommendation had  been made. Could the bishop still make a valid provision? Revigny used texts  from the law of contract, because he regarded the recommendation as a kind  of agreement. Belleperche rejected the argument, certainly because of its too  “horizontal” nature. A papal order does not depend primarily on somebody’s  recommendation.26 There  are  more  restrictions  of  this  type.  Some  texts  in  the  corpus  iuris  permit  argument  from  contracts  to  last  wills,  and  vice  versa.  As  a  legalist,  Belleperche could not contradict this. Yet even in this context he stated that  arguing from last wills to contracts is not allowed if, at issue, are aspects of 

25 For the preceding question, see Bezemer, Portrait (n 24) 19-21 and n 50. It is important to note  that Belleperche approved of Revigny’s opinion: the latter argued with a text about the decree of  a prince (D 27.1.12.1). 26  The case is described in K Bezemer, “The Law School of Orleans as School of Public Administration” (1998) 66 TvR 247 at 268-269. Additional material in K Bezemer, “Pierre de Belleperche:  An Early Attempt to Make Public Law a Separate Branch of Legal Science?”, in J Krynen and M  Stolleis (eds), Science politique et droit public dans les facultés de droit européennes (XIIIe-XVIIIe siècle) (2008) 41 at 45 and n 15.

70

the creation of the ius commune

last wills that go beyond the essential elements of a contract.27 Here again  Belleperche is found acting as the guardian of an art of arguing founded on  substantial (Roman law) elements, not on free analogies. Belleperche’s view  became dominant in the sixteenth century, which is beyond the scope of this  chapter. This stress on the essential elements sometimes led to unintended effects  of another nature. In many parts of France, especially in the north, a wooden  house  was  considered  to  be  a  moveable.  Revigny  mentioned  this  layman’s  opinion,  although  he  did  not  accept  it.28  Belleperche  entirely  ignored  this  aspect and instead asked about the nature of the possessorium of a wooden  house, the right to protect one’s possession (saisine). To determine its nature,  he used a text about the things that belong to a house, even should they not be  physically connected to it (D 19.1.17), and are therefore considered immoveable. As possessorium is not mentioned in this text, Belleperche concluded,  by means of an argument a contrario, that the possessorium is a moveable  “because the soil does not belong to the essence (esse) of the possessorium”.29  This scholastic manoeuvre resulted in a solution that French laymen would  also have understood, as they saw the rights connected with wooden houses  also as moveables. F. THE IMPACT OF REVIGNY AND BELLEPERCHE IN  FRANCE: JEAN FAURE It  has  been  necessary  to  devote  so  much  attention  to  demonstrating  the  differences between Revigny and Belleperche because each of them exerted  27 See Bezemer, Portrait (n 24) 156 and n 4. 28 As usual his argument is very much based on common sense. A wooden house cannot be removed  without dismantling it. Therefore it has to be seen as an immoveable. See Revigny’s commentary  on D 41.1.60 in Paris BN MS lat 14350 fol 75rb: “Ego vidi rusticos quosdam qui dicebant quod  domus lignea esset mobilis, nam, dicunt ipsi, domus potest separari et ferri alio loco. Ego dico est  immobilis et inseparabilis, nam illud inseparabile est quod [MS qui] non potest removeri in statu  in quo est. Unde domus lignea potest removeri: hoc est verum si dissolvatur. Sed certe in statu in  quo est non potest removeri. Unde videamus  lignea est res immobilis, et est casus legis,  infra de vi et vi ar. l. Si plane si quis [D 43.16.1.8].” 29  Petrus  de  Bellapertica,  Commentaria in Digestum novum  ad  D  43.16.1.8  Si quis  (1571;  repr  1968) 27: “Dixi possessorium est res mobilis, et est casus legis (casus est), supra de action. empt.  l. Fundi nichil aliud est, in principio [D 19.1.17pr]. § iste videtur contra me. Dicit quod domus  ligneae, quia solo affixae sunt, immobiles sunt, ideo etc. Non facio vim domino [de eo?], non est  simile de domibus, quia aedes sunt res immobiles, quin ipsum solum, cui aedes inhaerent, faciunt  unum totum, ut supra de usuc. l. Eum qui [D 41.3.23]; et ideo, quin ad essentiam aedium solum  est necessarium, ideo etc. Sed possessorium, licet affixum sit solo, non est res immobilis, quia  solum non est de esse possessorii, quia possessorium mobile remanet. Sic intelligo l. alle. Fundi  [D 19.1.17], et dixi loco suo.”

the infrastructure of the early ius commune 

71

considerable  influence  on  what  was  to  follow  in  Italy.  Revigny  must  have  influenced Dinus de Mugello in respect of the formation of regulae. Belleperche  certainly  inspired  Cinus  de  Pistoia  to  use  another  method  for  the  development  of  doctrines,  by  which  I  mean  his  reliance  on  distinctions  (distinctiones). Before turning to the Italian sequel, a few words should be devoted to the  influential French author Jean Faure (d 1340). He was well informed about  the ideas of Revigny and Belleperche, and also about legal practice, secular  and ecclesiastical, from his days in the south of France. This can easily be  illustrated by a remark of his about the concurrence of actions: These days this problem will hardly ever occur in practice, because canon law does  not require that the name of the action is expressed, and neither does the custom  of secular courts.30

By  this  Faure  meant  that  many  of  the  theoretical  problems  about  concurrence of actions had no counterpart in legal practice. Equally  interesting  is  what  Faure  wrote  about  the  nature  of  regulae.  A  couple  of  passages  selected  from  his  commentary  on  the  Institutes  reveal  that Faure was very much aware of the power (potestas) and the limitations  of legal rules. He clearly stood in the tradition of Dinus whom he mentioned  on various occasions. He realised that regulae are, in essence, based on casus.  In a discussion on actiones noxales, for example, he concluded: About this (subject matter) a firm rule can hardly be given, but it can be said that  on average there are no noxal actions on the basis of contracts or quasi-contracts.31

In this case this was simply because the few texts concerned were too different  and specific to be generalised into a rule. When there was an established rule,  Faure was anxious to separate it from texts that did not fall within its scope.  His  discussion  of  penal  actions  provides  a  good  example.  The  text  (J  Inst  4.12.1) says that if the litis contestatio has been achieved, and the offender  30 Joannes  Faber,  In quatuor Institutionum Libros Commentarii  (1582)  fol  159rb:  “Sed  haec  quaestio vix potest cadere in facto hodie, cum nomen actionis non exprimatur de iure canonico,  ut extra de iud. c. Dilecto [X 2.1.6], nec de consuetudine curiae laicalis.”  31  Faber ad J Inst 4.8 (n 30) fol 165ra: “Quomodo ergo intelligemus hanc materiam? Super hoc vix  potest dari certa regula, tamen potest dici quod regulariter ex contractibus, vel quasi contractibus  non agitur noxaliter, ut patet ff. de actio. & obli. l. Ex contractibus [D 44.7.49], de nego. gest. l.  Si filius [D 3.5.13] & d. l. Si oleum § penul. [D 4.3.9.4a], ff. de minoribus l. Quod si minor § i.  [D 4.4.24.1], & hoc verum pro dolo commisso in contractu, vel extra contractum. Sed interdum  agitur noxali actione ex contractu, pro delicto commisso extra contractum quando virtus actionis  potest se ad hoc extendere, ut patet ff. de pig. act. l. Si sevus [D 13.7.31], man. l. Inter causas  § Sed cum servus [D 17.1.26.7], de fur. l. Si servus communis § i. cum seq. [D 47.2.62.1-2]. Pro  delicto vero per servum commisso agitur noxaliter etc.” 

72

the creation of the ius commune

dies, the proceedings will be continued, also against his heir or heirs. The  Accursian gloss mentions some texts to underpin this rule. Faure commented: These casus have nothing to do with the regula which is given here, and which is  about penal actions resulting from a wrongdoing. The actions mentioned in these  casus are different, so the exceptions given here (by Accursius) are useless.32 

Because Faure had a thorough knowledge of the Orleans jurists Revigny  and Belleperche, it is interesting to see if and how he responded to the liberties Revigny allowed himself both in his method of teaching and in his way of  arguing. Faure’s view on Revigny’s way of teaching has more or less become  a classic among legal historians. I quote: Professor Jacques de Revigny continues in this place and adds the subject of the  summons (libelli), that usually is treated at another text. I will omit it here, as it is  not the proper subject (to discuss here).33

Faure ascribes Revigny’s behaviour to a need for ostentation. And,  what  did  Faure  make  of  the  less  visible  subtleties  in  Revigny’s  technical citations? Had he seen them? Did he side with Belleperche in his  fight against the abusive use of texts? If I have not yet found a telling example  to  demonstrate  Faure’s  views  on  these  issues,  it  is  nonetheless  possible  to  point to an indication of his opinion. There are many passages where Faure  discussed  the  custom  of  retrait lignager;  few  show  any  inspiration  from  the discussions of Revigny, who had put so much of his ingenuity into the  subject.34 G. ITALY DIVIDED: CINUS VS DINUS About the same period, matters took a different turn in Italy. Dinus’ pupil  Cinus de Pistoia (1270-1336/7) did not share his master’s interest in regulae,  old or new (i.e. those still to be discovered). This may have been caused by  Cinus’ initial aversion to the canon law branch of legal studies and practice.  This was an attitude he had in common with Belleperche, and it is a plausible  inference that it may have encouraged him to keep away from the regulae 32  Faber ad J Inst 4.12.1 in medio (n 30) fol 173ra: “Hi casus nil faciunt ad regulam quae hic traditur,  quae  loquitur  in  actionibus  poenalibus  quae  descendunt  ex  maleficio;  et  patet  quod  actiones  quae competunt in illis casibus non sunt huiusmodi, & ideo frustra fuerunt hic fallentiae positae.  Redeamus ad materiam huius § quaerendo ultra glosam aliqua quae faciunt ad intellectum § etc.” 33  Faber  ad  J  Inst  4.6  pr.  (n  30)  fol  115va:  “Haec  sufficiunt  pro  lectura  huius  paragraphi.  Sed  dominus Iaco. de Ra. altius exorditur, et cumulat hic materiam libellorum, quae consuevit tractari  in authent. Offeratur [aut. C 3.9.1]. Ego vero hic omitto quia non est propria materia et si de hoc  videre vis: vide in Spec. de libel. concept.; ibi reperies satis.” 34  See K Bezemer, “Ne res exeat de genere” (n 23) 107-109.

the infrastructure of the early ius commune 

73

iuris of the Liber Sextus, and whatever else concerned the formation of (new)  rules. However this may be, Cinus did not continue Dinus’ search for regulae,  whether in canon or Roman law. Kriechbaum has noted this aspect of Cinus’ work, and, in this connection,  has pointed to Cinus’ interest in distinctiones, which is certainly due to the  influence of Belleperche.35 Comparison of the treatises by Dinus and Cinus  on  intestate  succession  provides  a  striking  example  of  this.  Dinus’  treatise  ends  with  some  regularia  and  casus  about  the  innovations  brought  by  the  Code and the Novels. He mentions five regulae, sometimes with their exceptions.36 Cinus’ treatise (if the attribution is correct) ends with casus breves  about  descendant,  ascendant,  and  collateral  heirs.  Only  in  the  last  case  is  the word regula mentioned.37 This might be viewed as a relapse into former  practice,  but  it  seems  much  more  likely  to  be  a  statement  aimed  against  unwarranted generalisations without textual support. Cinus’  pupil  Bartolus  de  Saxoferrato  (1313-1357)  is  supposed  to  have  followed  his  master  in  this  respect.  There  is,  however,  at  least  one  exception  to  be  mentioned.  In  his  discussion  of  the  subject  of  concurrence  of  35 Cf M Kriechbaum in Lange, RRM II 294-295 (without the connection to Belleperche). 36 I quote from the collection of treatises available on the website of the French National Library.  See http://gallica.bnf.fr. Tractatus illustrium in utraque, Tomus VIII, Pars I (1584) fol 319rb no  24:  “Postremo  videtur  quod  quaedam  innovata  sunt,  &  quaedam  irregularia  in  successione  ab  intestato. Nota primo quaedam regularia, & casus, per quae sunt innovata ab intestato C. com.  de success. l. Vitrico et l. Affinitas [C 6.59.3 and 7]. Dicuntur autem affines omnes illi qui per  uxorem mihi contingunt, ff. de grad. l. Non facile § Affines [D 38.10.4.3 in medio]. […] Item et  alio casu: si uxor decedat relictis filiis emancipatis, vel nepotibus, pater una cum eis habet utilem  usumfructum, C. de bo. mat. l. 3 [C 6.60.3.1]. Item est & alia regula quod servus non potest esse  successor, ut l. Servus C. comm. de succes. [C. 6.59.4]. Fallit [i.e. regula] cum mora est facta in  libertate praestanda, C. ad Orfi. l. fi. [C 6.57.6], de his quib. ut indig. l. fi. [C 6.35.12]. Item fallit  cum decessit apud hostes, C. de suis et legiti. l. Apud hostes [C 6.55.8]. Item est alia regula: donec  potest habere locum successio ex testamento, cessat ab intestato, ff. de acqui. haer. l. Quandiu  [D 29.2.3], C. comm. de success. l. Antequam [C 6.59.8]. Fallit ista regula si sic dicit testator: “Te  haeredem instituo, si legitimus noluerit”, ff. de haere. insti. l. Si ita quis haeredem instituat [D  28.5.83]. Item est alia regula: qui proximior est in gradu, proximior & potior est in successione …  Item est regula quod nullus hospes potest etiam ab intestato succedere alumno sive vitrico, sive  hospiti … Ultimo restat quae sunt per ius novissimum [i.e. the Novels] immutata etc.” 37  The treatise attributed to Cinus can be found in the collection mentioned in the preceding note.  It  follows  Dinus’  treatise.  For  the  doubts  about  its  attribution,  see  M  Kriechbaum,  in  Lange,  RRM II 657-658. The passage mentioned in the text is on fol 321vb: “Casus breves Trium partium  tractatus successionum Cyni. De descendentibus. Si sunt descendentes, cunctis aliis praeferuntur,  etiam si gradu sint remotiores. Et succedunt primi gradus aequaliter, ulteriores vero in stirpes  cum eis qui primi sunt gradus. Ulteriores autem cum ulterioribus, sive aequalis sive inaequalis  gradus sint. Sive paris sive disparis numeri, succedunt pariter in stirpes in infinitum. Hic deservit  Authen. in successione [authentica after C 6.55.12]. De ascendentibus. … De collateralibus …  Tertio haec regula obtinet ut qui prior est gradu, potior sit, remotiores excludens … Hic deserviunt authenticae Itaque [authentica after C 6.59.11], Cessante [authentica after C 6.58.3], Post  fratres [authentica after C 6.58.3], in hoc ordine aptissimo.” 

74

the creation of the ius commune

actions, Bartolus pays tribute to both Dinus and Cinus. Dinus’ three rules  are  extended  with  three  more  rules.38  Cinus’  four  genera cumulationis  are  enriched with a fifth (same reason, same case, against different persons).39  Thus, with Bartolus, both approaches have found their place. H. REGULAE AS CONCLUSION OF A COMMENTARY:  JACOBUS BUTRIGARIUS At the same time, Jacobus Butrigarius (d 1348), one of Bartolus’ teachers, was  another Italian under the influence of the Ultramontani, but in a different  way from Dinus and Cinus. His work showed that regulae were considered  so important that they deserved a special place in lectures and commentaries. Butrigarius’  commentary  (repetitio?)  on  the  lex Precibus  (C  6.26.8)  provides an example of the way regulae can find expression in the structure  of a commentary or lecture. After a discussion of “vulgar” substitution, he  wrote: Having discussed the contrary texts and the questions, I come to its regulae. The  first is that a “vulgar” substitution is expressed by negative words (“if x is not my  heir, it will be y”).40

From this rule he inferred three consequences. What we see here is that once  a rule had been established, a new (deductive) phase was entered, in which  its scope, its exceptions, and its limitations had to be fixed. I. CONCLUSION The examples in the preceding sections are intended to show, first, how, in  the course of the thirteenth and fourteenth centuries, regulae were developed  for  certain  legal  subject  matters,  and  secondly,  why,  for  others,  this  development did not happen. One should not forget that, even if an author  38 See above, section B. and note 12.  39 See  Bartolus  de  Saxoferrato,  Commentaria in primam Codicis partem  ad  C  2.1.3  Edita actio (1574) fol 40rb no 22: “Quintum genus cumulationis est quando sunt plures actiones ex eadem  causa, & ad idem contra diversos.” 40  Jacobus Butrigarius, Lectura super Codice ad C 6.26.8 Precibus (1516; repr 1973) fol 18ra: “Visis  contrariis et questionibus, venio ad eius regulas. Et prima regula est quod vulgaris expressa fit  verbis  negativis,  et  tacita  fit  verbis  negativis  tacite  intellectis;  et  fit  hoc  modo:  ‘Si  talis  non  est  heres, talis sit mihi heres.’ Ex quo infertur quod si adeatur hereditas, expiret, quia venit casus  eius contrarius. Secundo infertur quod per eam succedit quis testatori, non pupillo, ut l.i. ff. e. [D  28.6.1.3]. Et ex hoc infertur aliud, scilicet quod substitutus vulgariter numquam ex ea substitutione capit bona impuberis, sed bona testatoris tantum; nisi caperet ex substitutione tacita pupillari, ut dixi. Modo venio ad pupillarem [i.e. substitutionem] etc.”

the infrastructure of the early ius commune 

75

saw reasons to accept the existence of a regula in a particular area, another  author might disagree. The detection of (new) regulae remained a matter of  interpretation. There is one thing I must emphasise above all: the use of the word “regula”,  or a similar expression, is an important marker of the development described  above. This word had a well defined meaning (see D 50.17.1) that very much  stressed its inductive character. A regula was to be extracted from the law,  not the other way around. In consequence, every medieval jurist knew what  it meant if somebody claimed to have discovered a (new) regula. And if the  new regula was indeed accepted, it necessarily engendered discussions about  its  exact  scope  and  its  exceptions.  This  meant  that  a  new  regula  could  be  eventually wrecked and deprived of utility by the number of exceptions to  which it was subject. If I may put my point about the word “regula” in another way: we should  not  assume  too  quickly  the  existence  of  a  regula  if  the  word  itself,  or  a  synonym, was not used. Much more than nowadays, in the medieval world  the use of a word carried a lot of weight, because – for many people – words  had as much reality as the things to which they referred. Naming a thing was  acknowledging the existence of the concept. Therefore, it was highly relevant  whether or not the word “regula” was used. In other words, a regula was a  regula if it was called a regula.

3  Ius Quaerens Intellectum: The Method of the Medieval Civilians James Gordley A. INTRODUCTION B. THE APPLICATION OF TEXTS TO NEW SITUATIONS C. SEEKING ORDER (1) Order through the juxtaposition of texts (2) Order through ascribing explanatory power to terms in certain texts (3) Order through a schema D. CONCLUSION A. Introduction In the West, since the time of the Roman jurists, the task of understanding and interpreting the law has usually belonged to a distinct class which a person can enter only after a long period of specialised study. Its members are united, not so much by agreement as to their conclusions, but by a common understanding of how conclusions are to be reached. They share a method which enables each jurist to arrive at opinions of his own and to explain why he agrees or disagrees with those of other jurists. This chapter is about the method of the medieval civilians who expounded Roman law from the twelfth century, when the study of Roman law was revived, through the fourteenth century, when they produced their last great works. Their task, as they conceived it, was to explain the texts compiled by the emperor Justinian in the sixth century and which became known as the Corpus iuris civilis. The largest part of the compilation, the Digest, was made up of short extracts from Roman jurists. Nevertheless, the method of the medieval civilians was unlike that of the Roman jurists. It will be helpful to say a few words about the Roman jurists in order to see how it was different. Some scholars, such as Fritz Schultz and Peter Stein, believe that the 77

78

the creation of the ius commune

Roman  jurists  learned  to  think  systematically  from  the  Greeks,  and  in  particular  from  philosophers  such  as  Aristotle.1  While  the  jurists’  method  was systematic, nevertheless it did not work in the same way. Unlike Greek  philosophers,  the  jurists  did  not  abstract  a  concept  from  ordinary  experience, define it, and then explore the logical consequences of the definition.  Commonly, the jurists took a concept that was familiar from ordinary experience and, instead of defining it, they refined and clarified it by putting a series  of particular cases. Some of their most basic concepts, such as possession,  negligence (culpa in the narrow sense) or consent, are familiar in the ordinary  experience of people almost anywhere. Wherever objects belong to people,  a person will sometimes possess an object that does not belong to him, such  as another’s blanket or spear. He will have been taught in childhood not to be  careless with sharp objects or fire. It will matter, at least sometimes, whether  he consented to a course of action or not. The genius of the Roman jurists was not to discover concepts such as these  but  to  see  that  they  were  legally  significant  and  to  refine  them  by  putting  particular cases. Could a man be in possession of a farm without going round  every  part  of  it?  Certainly.2  Would  he  retain  possession  even  if  he  left  on  an errand leaving none of his household at home? He would.3 What did it  mean to be negligent? As Reinhard Zimmermann has observed, the jurists  did not “try to subsume the facts of an individual case under an abstract rule  or formula” but proceeded “casuistically”.4 A pruner is negligent if he harms  someone by cutting off a branch over a public way without calling out.5 So  is  a  javelin  thrower  who  kills  someone  by  throwing  it  in  an  inappropriate  place,6 a farmer who burns stubble on a windy day and cannot control the    1   Their evidence seems to me to be thin, although I cannot review it here. I can note that these  authors  disagreed  with  each  other.  Schulz  and  Stein  claimed  the  Romans  used  the  methods  which,  according  to  Aristotle,  characterise  a  true  science.  They  disagreed  as  to  when.  Schulz  thought that Greek methods were used during the late Republic in what he calls a “Hellenistic  age”, and that classical jurists lost track of them: F Schulz, History of Roman Legal Science (1953)  67. Stein thought that Greek methods were first used by classical jurists who initiated a “scientific  revolution”: P Stein, Regulae Iuris: From Juristic Rules to Legal Maxims (1966) 33-35. Theodore  Viehweg claimed that rather than using the method Aristotle ascribed to a true science, the jurists  had used the more tentative method that Aristotle described in the Topics: T Viehweg, Topik und Juisprudenz, 4th edn (1969). But it is hard to see the resemblance, except that neither method  looks like the one Aristotle believed to characterise a true science. These disagreements show that  it is difficult to find evidence of a Greek influence on the Roman jurists even when one is looking  hard for it.   2  D 41.2.3.1.   3  D 43.16.1.24.   4  Zimmermann, Obligations 1008.   5  D 9.2.31.   6  D 9.2.9.4.

ius quaerens intellectum: the method of medieval civilians 

79

fire,7 and a barber who shaves a customer out of doors near a playing field and  cuts him when a ball strikes the hand that is holding the razor.8 Similarly, in  considering when the parties consented to a sale, the jurists did not discuss  the nature of consent. They put cases about copper sold as gold or vinegar  sold as wine.9 Their manner of stating these cases was in one sense abstract:  they left out all the facts that might be encountered in real life that had no  bearing on the meaning of the concept they were testing. For that reason, it  is hard to tell whether their cases actually occurred and whether they were  hypothetical. Nevertheless, each statement of a case was concrete in that it  described a factual situation that might occur. Watching the Roman jurists leap from concept to conclusion can be unsettling  for  those  familiar  with  other  Western  intellectual  traditions,  such  as  ancient Greek philosophy, or, for that matter, modern physics and economics.  A  philosopher,  a  physicist  or  an  economist  defines  his  concepts  abstractly,  and then works out their consequences by logical steps. The Roman jurists  explained their concepts by putting particular cases, moving from a concept  to a conclusion at once, without a logical explanation of how they got from  the one to the other. Like the Roman jurists, and unlike philosophers, physicists or economists,  the medieval jurists did not work by defining concepts and logically working  out their implications. For that reason, their method can also seem unsettling,  like that of the Romans. Despite this similarity, however, their method was  not that of the Roman jurists. Their starting point was the texts of the Corpus iuris civilis. While a Roman jurist felt free to jump from concept to conclusion on his own authority,  a  medieval  civilian  presented  his  conclusions  as  grounded on the authority of the texts, even though the texts did not resolve  the matter explicitly. Another difference concerned the matters about which  they drew conclusions. The Roman jurists worked by testing their concepts  against  particular  cases.  To  do  so,  they  needed  a  good  stock  of  concepts,  but not much structural organisation beyond the concepts themselves. They  were not much concerned with how the various parts of their law were interrelated. The medieval jurists were concerned with the relationship of each  text to every other. They identified ordering principles that related one part  of the law to another that was embedded in these texts. We  will  consider,  first,  how  the  medieval  civilians  applied  their  texts  to  new situations which the texts did not explicitly consider. We will then see    7 D 9.2.30.3.   8 D 9.2.11pr.   9 D 18.1.9.

80

the creation of the ius commune

how they found an order in their texts with which the Romans had been less  concerned. b. the ApplIcAtIon of texts to new sItuAtIons The medieval civilians applied their texts to new situations in three ways. One  was to generalise the scope of a text that deals with one situation so that it  could be applied to a new one. Another way was to limit the scope of a text  framed in general terms by distinguishing those situations to which it will be  applied from those to which it will not. The third way was to move directly  and by analogy from the situations expressly dealt with in the texts to a new  situation. None of these ways of applying authoritative texts to new situations  is distinctly medieval. Modern jurists proceed in the same three ways when  they apply a code, a statute or a precedent. They begin with a text and then  move up, down, or sideways. Movement  in  each  of  these  three  directions  can  be  illustrated  by  the  medieval civilians’ discussion of the relief to be given if a price was disproportionately low or high. They generalised a post-classical text that dealt with the sale of a land: If you or your father part with an estate for a lesser price when it is worth a greater,  it is equitable that you return the price to the buyer and recover the land by the  authority of the court, or, if the buyer chooses, that you recover the amount of the  deficiency in the just price. The price is considered to be too low if less than half  of the true price was paid.10

At an early date, the medieval civilians took this text to protect sellers of things  other than land, and buyers as well as sellers.11 The result was the general  remedy for a one-sided contract that became known as laesio enormis. In order to generalise this text, the medieval civilians limited the scope of  another one, this time a classical text: “Pomponius says with regard to the price  10 C 4.44.2. 11 The  Corpus legum,  better  known  as  the  Brachylogus,  written  at  the  beginning  of  the  twelfth  century, does not speak of land but of objects sold: H Böcking (ed), Corpus legum sive Brachylogus iuris civilis  iii.xiii.8  (1829).  The  Dissensiones dominorum  of  the  early  thirteenth  century  reports  a  dispute  in  which  all  participants  take  it  for  granted  that  the  buyer  has  a  remedy.  The  disputed question is whether, for him to have it, the sales price must be twice or one-and-a-half  times the just price. The participants are said to be Placentinus and Albericus, who wrote in the  twelfth century, and Martinus, a student of the legendary jurist Irnerius to whom the revival of  Roman legal studies was traditionally ascribed: G  F  Haenel  (ed),  Diversitates sive dissensiones dominorum, Hugolinus § 253 (1834). According to Accursius, it was agreed among the jurists that  the remedy would be available generally in contracts bonae fidei: Glos ord at C 4.44.2 sv auctoritate iudicis.

ius quaerens intellectum: the method of medieval civilians 

81

in sale that the contracting parties are naturally permitted to take advantage  of each other.”12 Accursius explained, “unless he is deceived beyond half the  just price”.13 The medieval civilians applied other texts by analogy to explain when a  price is just. These texts described how prices were to be determined in other  situations. They concluded that the just price is the price generally prevailing  at the time and place the contract was made. According to Accursius,14 that  conclusion was supported by two texts that said that the value of a slave is  not his value to individuals but to people in general. One text concerned a  slave who was to be redeemed by paying his value.15 The other dealt with the  amount to be paid for a slave who was killed.16 The value of land, according to  Accursius, is to be determined by its quality and rents. He cited two imperial  rescripts and a text calling into question the sale by a guardian of land that  was “infertile, stony and pestilential”.17 The value of chattels was their price as  of the time and place of sale, although, in opposition to his own view, Accursius cited a text concerning the difficulties of valuing a stolen cow.18 Although the medieval jurists’ conclusions were supposed to be logically  consistent with the texts, nevertheless, whether they were generalising a text,  limiting its scope, or applying the text by analogy, the texts might be logically  consistent with more than one conclusion. For example, in discussing laesio enormis, the medieval jurists generalised the text that gave the seller a remedy  when he had received less than half the just price. They limited the one that  allowed the buyer and seller to take advantage of each other. It would have  been as logically consistent to have limited the first text and generalised the  second. It can be difficult to tell why the jurists found one conclusion more  compelling than another. For example, in the view of some modern scholars, the medieval jurists  regarded the classical text allowing the parties to take advantage of each other  as repugnant to their Christian values.19 That view may go beyond what we  can  know.  Accursius  merely  said  that,  “it  is  permitted  for  the  contracting  parties  to  deceive  each  other  up  to  half  the  just  price”.  He  did  not  say  anything about whether it was contrary to Christian values for them to do so,  12 13 14 15 16 17 18 19

D 4.4.16.4; see also D 19.2.22.3. Gl ord at D 44.16.4 sv naturaliter licere. Gl ord at C 4.44.2 sv autoritate iudicis. D 35.2.63. D 8.2.33. D 27.9.13pr. D 13.1.14.3. J Q Whitman, “Long Live the Hatred of Roman Law!” (2003) 2 Rechtsgeschichte 40 at 48-49.

82

the creation of the ius commune

and, if it was, why they should be allowed to deceive each other at all. His  contemporaries did not seem to find this position troubling. A papal decretal  provided that, in canon law, as in civil law, relief would not be given unless  the contract price deviated by half from the just price.20 Although France was  a Christian country, the French never went as far as the medieval civilians.  They maintained into modern times that the remedy should only be given  to sellers of land.21 England was a Christian country and never developed a  similar remedy. In the nineteenth century, in Germany, which still considered itself a Christian country, jurists were still arguing about how to read  these two texts: to provide a remedy only for sellers of land,22 or for buyers  as well and parties to similar transactions.23 In considering why the medieval  civilians reached their conclusions, we must go by what they tell us. Here,  they tell us very little. One reason they tell us little is that, as mentioned, in one respect, their  method resembled that of the Roman jurists. As the Romans jumped from a  concept to its application without a logical explanation of how they got from  the  one  to  the  other,  the  medievals  jumped  from  a  text  to  its  application.  Nevertheless, just as the Romans did not think any conclusion was equally  sound, the medieval jurists did not think that any application of the texts was  equally compelling. They often disagreed over the best solution, or tried to  work  toward  a  better  one.  By  looking  at  instances  in  which  the  jurists  did  disagree or changed their views over time, we can get a sense of what made  one solution more attractive to them than another. We will look at how they  approached three problems: whether a person is liable if he bothers a neighbour by making smoke on his own premises; whether he is liable if he makes  a true statement that injures the reputation of another; and whether he is  liable if he builds a water mill that cuts off the flow of current to a nearby mill  owner on the same stream. To resolve the first problem, the medieval civilians generalised a text; to resolve the second, they limited one; and to resolve  the third, they applied their texts by analogy. We have seen that there was  nothing distinctively medieval in applying a text to a new situation in these  three ways. We will now see that in each of these instances, there was nothing  20 X 3.17.3. 21 C de Ferrière, Dictionnaire de droit et de pratique II.v sv “lézion d’outre moité de juste prix”,  nouv edn (1769) 135, 137; H Lacombe de Prezel, Dictionnaire portatif de jurisprudence et de pratique II.v sv “lesion” (1763) 430. 22  See, e.g., R von Holzschuher, Theorie und Casuistik des gemeinen Civilrechts (1864) III 729-30; K  von Vangerow, Leitfaden für Pandekten-Vorlesungen (1847) III § 611; C von Wächter, Pandekten  (1881) III § 207. 23  See, e.g., J Seuffert, Praktisches Pandektenrecht (1852) II § 272.

ius quaerens intellectum: the method of medieval civilians 

83

distinctively medieval about the way the jurists looked for the best solution.  In each case, there are direct parallels in the work of later jurists. The problem of the person who bothers his neighbour by making smoke was  governed chiefly by two short Roman texts. One text reported that the jurist  Aristo “does not think that smoke can lawfully be discharged from a cheese  shop onto the buildings above it”. The text added: He also holds that it is not permissible to discharge water or any other substance  from the upper onto the lower property, as a man is only permitted to carry out  operations on his own premises to this extent, that he discharge nothing onto those  of another.24

The other text suggested that, nevertheless, a person could “create a moderate  amount of smoke on his own premises, for example, smoke from a hearth”.25  So  the  question  arose,  what  general  rule  could  explain  why  there  was  an  action for the smoke of the cheese shop but none for that of the hearth? Accursius distinguished the discharge of smoke from that of water on the  ground that “smoke naturally disperses”. Having done so, he was unable to  distinguish the smoke from a cheese shop from the smoke from a hearth. He  concluded that the text about the cheese shop could not mean what it said.  Instead, “the person with the upper premises is required to bear the smoke,  and  on  that  account  not  to  have  windows”.26  His  inveterate  critic,  Jacques  de Revigny (Iacobus de Ravanis), had little difficulty pointing out why this  solution was unsatisfactory: the text “says the complete opposite of what the  Gloss [of Accursius] says”.27 According to Odofredus, the owner of the cheese shop was liable because  one cannot use one’s land in a way that harms another: “unusquisque debet  facere  in  suo  quod  non  officiat  alieno”.28  That  rule  appeared  to  rest  on  a  straightforward principle, and perhaps, for that reason, it lived on. Rephrased  by Blackstone – sic utere tuo ut neminem laedas29 –  it passed into the common  law. Nevertheless, it did not explain why one person could annoy another with  the smoke from his hearth. It has been criticised by modern jurists because  it does not explain why one can harm another in cases like that of the hearth  but not like that of the cheese shop.30 24 D 8.5.8.5. 25  D 8.5.8.6. 26  Gl ord to D 8.5.8.5 sv ad iure. 27  Albericus de Rosate Bergamensis, In primam ff. veter. part. commentarii at D 8.5.8.5 no. 5 (1585)  in OIR XXI (describing the opinion of Jacques de Revigny). 28  Odofredus, Lectura super digesto veteri at D 8.5.8.5 (1550) in OIR II. 29  W Blackstone, Commentaries on the Laws of England , 4 vols, (1766-69; repr 1979) III 217.  30  W V H Rogers, Winfield and Jolowicz on Tort, 14th edn (1994) 404. 

84

the creation of the ius commune

Jacques  de  Revigny  proposed  a  different  distinction:  a  person  cannot  discharge  anything  onto  another’s  property.31  That  solution  was  suggested  by Aristo’s remark that one must “discharge nothing onto [the premises] of  another”. It sounded like a clear rule. Perhaps for that reason it was accepted  by the great medieval jurist Baldus degli Ubaldi,32 and became popular among  the Dutch and German jurists of the seventeenth and eighteenth centuries.33  But, again, it does not explain why one is not liable for the hearth, which does  discharge smoke onto another’s property. Bartolus of Saxoferrato found another solution: I think the following is to be said: Sometimes the owner of the lower premises  makes fire in the usual way for the ordering of his family, and then he may do it  lawfully, and he is not liable if the smoke ascends unless he acts with an intention  to injure. In the same way, if the owner of the upper premises lets water flow in  some  way  that  is  normal,  for  his  water  clock,  he  is  not  liable  if  some  descends  unless he acts with an intention to injure. But if the owner of the lower premises  wants to make a shop or inn where he is continually making a fire and a great deal  of smoke, he is not allowed to do so, as in this text. In the same way, if the owner  of the upper premises lets water flow beyond what is normal, he is not allowed to  do so, as this text says.34

For Bartolus, then, two things mattered: the extent of the interference, and  whether the interfering use of the land was usual. This is a solution to which  many modern jurists have returned.35 Bartolus’ solution does not seem to rest on a straightforward principle, like  the rule of Odofredus that one landowner cannot harm another. It is not as clear  as the rule of Jacques de Revigny that no one can discharge anything on another  person’s land. To apply Bartolus’ rule, one must decide how much smoke is a  “great  deal”  and  what  activities  are  “usual”.  But  his  rule  did  distinguish  the  hearth  from  the  cheese  shop.  Presumably,  the  jurists  would  have  agreed  on  a single rule if they had found one that did seem to rest on a straightforward  principle that was clear and which also could distinguish the hearth from the  cheese shop. Since they could not find a rule that was satisfactory in all these  ways, they disagreed as to what the proper rule should be. Because they disagreed, we can get a sense of the sort of rule that they would  have regarded as satisfactory. We can also see that, in this instance, there was  31 Albericus, Commentarii at D 8.5.8.5 no. 5 (n 27) (giving the opinion of Jacques de Revigny). 32  Baldus degli Ubaldi, Commentaria Corpus iuris civilis at D 8.5.8.5 (1577). 33  See, e.g., D Gothofredus, Corpus iuris civilis . . . cum notis integris Dionysii Gothofredi, Antonii Anselmo, Simonis von Leuwen at D 8.5.8.5 no 37 (1726); J Voet, Commentarius ad Pandectas 8.5 §  5 (1726). 34  Bartolus de Saxoferrato, Commentaria Corpus iuris civilis at D 8.5.8.5 (1615). 35  J Gordley, Foundations of Private Law (2006) 68-69.

ius quaerens intellectum: the method of medieval civilians 

85

nothing distinctively medieval about their ideas about what a satisfactory rule  would look like. There is nothing distinctively medieval about wanting a rule  that seems to rest on a straightforward principle, that is clear, and that distinguishes cases like the hearth from cases like the cheese shop. Since modern  jurists did not find a rule that was satisfactory in all of these ways, they continued  to disagree over which of the three rules proposed by the medieval jurists was  the best one. In the example just discussed, the jurists started with texts that dealt with a  specific situation and tried to formulate a general rule. At other times, they tried  to limit the scope of a general text. An example is the way they dealt with the  question of whether a person is liable for making a statement that is truthful but  detracts from another’s honour or reputation. As we have seen, if the statement  were untruthful, the person who made it would be liable in an action for iniuria.  One text suggested that he would not be liable if the statement were true: “It  would not be fair (bonum aequum) for one who defames a pernicious person  to be condemned, for the wrongs of such a person ought to be observed and  made known.”36 The medieval civilians were faced with two alternatives: to  conclude that a person who reveals another’s wrongs is never liable, or to find  some way to limit this text. Accursius  mentioned  one  possible  limitation  only  to  reject  it:  the  text  might apply only to statements made in judicial proceedings.37 That limitation  was too strict if “wrongs … ought to be observed and made known”. Cinus  de  Pistoia  cited  another  text  that  said  it  would  be  “inhumane”  to  disclose  the extent of another’s wealth.38 He suggested that it would be wrong and  inhumane (iniuriosus et inhumanus) to disclose another person’s defects.39  But could one disclose another’s moral as well as his physical defects? Could  one never disclose a physical defect? Jacques de Revigny thought that one  could disclose conduct for which the plaintiff would be legally answerable. In  that event, he said, the defendant should not be liable since there was a public  interest in the revelation.40 This position was taken by Pierre de Belleperche  in his Lectura Institutionum,41 and later adopted by Dinus de Mugello.42 But  Petrus refined it in a comment on C 9.35.5. The important question, indeed, is  whether the disclosure is of aid or interest to the commonwealth.  Therefore,  36 37 38 39 40 41 42

D 47.10.18pr. Gl ord to D 47.10.18pr sv eum qui. C 9.35.2. Cinus de Pistoia, In Codicem commentaria at C 9.35.5 no 10 (1578). Cinus, Commentaria at C 9.35.5 no 10 (n 39). Petrus de Bellapertica, Lectura Institutionum at I.4 de iniuriis (1536). According to Cinus, Commentaria at C 9.35.5 no 12 (n 39).

86

the creation of the ius commune

the defendant should be able to reveal matters of public interest even when  the plaintiff would not be legally answerable: for example, the plaintiff had  leprosy  and  might  infect  the  city.43  This  solution  became  standard.  It  was  accepted by Cinus himself 44 and by Bartolus.45 Here, although the jurists finally settled on a rule, they did so only after  several  different  rules  had  been  suggested  and  found  wanting.  Cinus’  proposal, that one could not reveal another’s defect, was ambiguous. Cinus  himself must have thought that, standing alone, it was inadequate since he  accepted  the  rule  that  became  standard.  Other  rules  seemed  narrow:  that  one could only reveal another’s conduct in legal proceedings, or if that person  would be legally answerable for it. The jurists finally concluded that there  must be a public interest in the disclosure. Once again, there was nothing distinctively medieval about the way the  medieval  jurists  arrived  at  this  solution.  Modern  jurists  have  walked  the  same path. They have faced the same alternatives: to exclude liability for any  truthful  statement,  or  to  impose  liability  but  to  limit  it.  The  common  law,  traditionally, excluded liability. Most jurisdictions today impose liability but  limit it in the same way as the medieval jurists: the defendant is not liable  for disclosures in the public interest.46 The defendant is liable in the United  States  for  the  new  tort  of  “intentional  disclosure  of  embarrassing  private  facts” except for information “of legitimate concern to the public”.47 He is  now liable in England for “breach of confidence” even no information had  been confided to him unless “there is a sufficient public interest ... to justify  curtailment of the conflicting right”.48 In France, a defendant will be liable  for revealing facts about a celebrity’s private life in a tabloid when he would  not be if the same information appeared in a historical work.49 In Germany,  a defendant is liable for disclosing the crimes a person committed in the past  but not if they  were  war  crimes  since  the  press  has  a  “legitimate  interest”  in  “informing,  instructing  and  ...  shaping  public  opinion”.50  According  to  the European Court of Human Rights, there is “a fundamental distinction”  between  merely  “reporting  details  of  the  private  life  of  an  individual”  and  “impart[ing] information and ideas on matters of public interest”.51 43 Petrus de Bellapertica, Super IX libros Codicis at C 9.35.5 no 16 (1569). 44 Cinus, Commentaria at C 9.35.5 no 12 (n 39). 45 Bartolus, Commentaria at D 47.10.18 no 1 (n 36). 46  Gordley, Foundations (n 35) 250-252. 47  Restatement (Second) of Torts § 652(D) (1976). 48  Campbell v MGN Ltd [2004] 2 AC 457. 49  Cass, 2e ch civ, 14 Nov 1975, arrêt no 729, pourvoi no 74-11.278 (unpublished). 50  OLG, Frankfurt 6 Sept 1979 NJW 1980, 597. 51  Von Hannover v Germany, 24 June 2004 (Application no 59320/00) no 63.

ius quaerens intellectum: the method of medieval civilians 

87

There have been different views of what information is a matter of public  interest. The view in the United States is notoriously broader than that in  continental European countries. The view is necessarily broader in modern  democracies informed by mass media than in medieval society. Nevertheless,  it is not surprising that jurists, medieval and modern, have limited liability for  disclosing truthful information by asking whether there is a public interest in  the disclosure. Any narrower limit seems arbitrarily to prohibit some disclosures while permitting others, which is the objection Accursius made against  permitting  only  disclosures  in  legal  proceedings,  and  Jacques  de  Revigny  made against permitting only disclosures about conduct for which the plaintiff  was  legally  answerable.  There  is  nothing  distinctly  medieval  about  the  discomfort they felt about stopping short of a rule that protected disclosures  in the public interest.  The third way in which the medieval civilians applied their texts to new  situations was by drawing analogies to situations that were dealt with by their  texts. An example is their discussion of a question put by Bartolus: “if I have  a mill below, and others make a mill above so that they impede me, may they  do this?”52 As Alan Watson has pointed out, no Roman text expressly dealt  with this problem.53 Bartolus said it was an old question that had been argued  by “Franciscus Accursius (not the author of the Ordinary Gloss) and other  doctors of Bologna”.54 It had become important in the Middle Ages because  of a revolution in mill technology.55 When mills became more efficient and  useful for a greater variety of tasks, rivers became crowded with them. Accursius, the author of the ordinary gloss, cited one text in favour of the builder  of the first mill.56 It said that usage should be respected.57 But, he added, the  opposite result was supported by three other texts which allowed a proprietor  to divert water which had its source on his own land,58 and by a fourth which  dealt with a more remote problem.59  Bartolus concluded that “if, indeed, the river was public, as is every river  that flows perpetually60 ... [and] the first person had his mill lawfully ... the  52 Bartolus Commentaria at D 43.12(11).2 no 2 (n 36). 53  A  Watson,  “The  Transformation  of  American  Property  Law:  A  Comparative  Law  Approach”  (1990) 24 Georgia Law Review 163 at 175. 54 Watson, “Transformation” (n 53) 175. 55 J Gimpel, The Medieval Machine: The Industrial Revolution of the Middle Ages (1977) 7-15. 56 Gl ord at C 3.34.7 sv procurator. Accursius mentioned the problem without resolving it in his  Gl ord at D 2.13.6.6 sv rationis. 57  C 3.34.7. 58  D 39.3.21; D 38.2.24.12; D 38.2.26. 59  D 50.17.61(62) (the problem of repairing one’s own house). 60  Citing D 43.12.1.3.

88

the creation of the ius commune

second  person  cannot  act  or  take  away  from  him  the  advantage  that  he  lawfully began [to derive] from what is public”.61 The result should be the  same, he added, “if the second person builds below the first mill and backs  up the water”.62 The key to the problem, he said, is that the river is “public”.  He relied on a text that stated, “whenever anything is done in a public place,  it should be permitted on condition that it causes no injury to anyone”.63 He  supported his conclusion by citing texts that dealt with things done in public  places. The building of the mills was like an encounter between carts on a  public road: the first one to enter prevails. It was like two people who want to  build on a strip of beach where anyone was entitled to build: the first to begin  building prevails.64 Similarly, the first to draw water from a public river may  do so if it is “without injury to those nearby”.65  Again, there is nothing distinctly medieval about Bartolus’ use of analogy  to  resolve  a  new  problem.  Blackstone  did  so  when  he  discussed  the  same  problem: the rights of the first person to build a mill. Like Bartolus, he thought  that the first person was protected against late comers. He drew an analogy  to a wild animal which belongs to the first person to capture it.66 In the early  nineteenth  century,  Justice  Shaw  defended  the  same  position  by  drawing  the same analogy as Bartolus to an encounter on a public road between the  first to occupy a spot and one who came later.67 Ultimately, American courts  rejected Shaw’s position. An important reason, I have argued elsewhere, is  that Roman law was a richer source than the common law of analogies that  would  support  it.68  Bartolus  could  cite  texts  about  people  who  build  on  a  public beach and those who draw water from a public river. In contrast, it  is hard to think of a common law action to vindicate a right to use a public  place.69  Like  Bartolus,  Shaw  drew  an  analogy  to  an  encounter  on  a  public  road,  but  the  case  was  hypothetical  and  to  common  lawyers  might  have  seemed  recherché.  If  a  person  were  pushed  out  of  a  spot  he  occupied  on  a road, he would recover at common law for battery simply because he was  pushed, and not by asserting a quasi-proprietary right to the spot on which he  was standing. When Bartolus drew the analogy, he could cite a text. 61 Bartolus, Commentaria at D. 43.12(11).2 no 2 (n 36). 62  Bartolus, Commentaria at D 43.12(11).2 no 8 (n 36). 63  D 43.8.2.10. 64  Bartolus, Commentaria at D 43.12(11).2 no 10 (n 36). 65  Bartolus, Commentaria at D 43.12(11).2 no 10 (n 36). 66  Blackstone, Commentaries (n 29) II 403 67  Carey v Daniels 49 Mass (8 Met) 466, 478 (1839). 68  Gordley, Foundations (n 35) 116-119. 69  The  only  one  that  comes  readily  to  mind  is  an  action  for  public  nuisance,  and  it  can  only  be  brought privately when the plaintiff has been specially affected.

ius quaerens intellectum: the method of medieval civilians 

89

In these instances, then, there was nothing distinctively medieval in the  way  the  medieval  jurists  worked  from  their  texts  to  a  conclusion,  whether  they were generalising a text, limiting it, or applying it by analogy. Modern  jurists have worked from their authoritative texts in much the same way. They  are still using the method of applying these texts that the medieval jurists first  developed. It would be wrong to conclude, however, that modern and medieval jurists  work in the same way. One difference is that modern jurists use not only the  method that the medieval jurists had developed, but other methods as well.  Another  difference  is  that  the  medieval  jurists  had  to  work  with  a  closed  corpus of texts compiled centuries before their own time. To begin with, modern jurists use other methods as well. At times, one of  these newer methods has taken the limelight, leaving the older one in the  shadows. We remember the nineteenth century as an age of conceptualism or  Begriffsjurisprudenz. The conceptualist method was to define legal concepts  such as contract or property abstractly and then to try to extract conclusions  from the definitions. For example, contract was defined in terms of the will  of the parties. It was thought to follow that a contract, even at an unfair price,  was valid in principle if it was freely entered into. Property was defined in  terms of the will of the owner: he could use his property as he chose. The  nineteenth-century  conceptualists  then  encountered  problems  explaining  why sometimes an owner could not use his property as he wished, as in cases  like the cheese shop. Since the late nineteenth and early twentieth  centuries,  when  jurists  became  aware  of  the  limits  of  conceptualism,  the  favoured  method  has  been  to  reason  functionally,  in  terms  of  the  policies  that  the  law should serve. Some of the time modern jurists continued to apply texts  in much the same way as medieval jurists, but other times they used these  newer methods as well. The reason is not that the medieval jurists were incapable of making what  we  would  call  conceptualist  or  functionalist  arguments.  Occasionally,  they  did so. But they did so sporadically, without regarding these arguments as  the  basis  for  an  alternative  method  for  applying  their  texts.  For  example,  Bartolus  made  what  we  would  call  a  conceptualist  argument  in  discussing  the rights of the person who builds the first mill along a river. He considered  what we would call a conceptualist objection of a hypothetical interrogator. If  the first to build were protected against late comers, he would injure them,  for “is it not always an injury to those nearby since then one cannot make  his own building?” It is an injury to deprive another of a right, and the first  builder  had  deprived  them  of  the  right  to  build.  Bartolus  made  what  we 

90

the creation of the ius commune

would call a conceptualist response: “I answer that he does not suffer injury  or harm but he does not acquire an advantage; and these are not equivalent  or the same in meaning (ratio).”70 His argument was that it is not an injury to  prevent another from acquiring an advantage; therefore the late comer is not  injured. In the nineteenth century, Joseph Story made the same argument  as  Bartolus’  hypothetical  interrogator:  to  take  possession  of  a  wild  animal  “presupposes no ownership already existing”, but if he who builds the first  mill can prevent others from doing likewise, he would deprive them of a right  they previously possessed.71  Similarly,  the  medieval  jurists  occasionally  made  what  we  would  call  arguments of public policy. In discussing what we call the conflict of laws,  Bartolus  considered  whether  a  city  or  principality  could  punish  its  own  citizens  for  crimes  committed  elsewhere.  Bartolus  answered  that  it  could  “for it is in the interest of the republic to have good subjects … but men are  made good by punishments imposed for wrongs”.72  Presumably,  medieval  jurists  such  as  Bartolus  would  not  have  made  such  arguments  if  they  had  regarded  them  as  improper  or  antithetical  to  their usual method. The reason they made them so rarely seems to be that  they thought their usual method was quite satisfactory, and, indeed, that it  epitomised the proper way to apply an authoritative text. Ernst Gombrich  described a similar phenomenon in the history of art. In ancient Egyptian  relief sculpture, one can find, on rare occasions, touches of realism: the bent  back and the tension in the muscles of a man pulling on a rope. But these  were  “isolated  instances”  that  “remained  without  consequence”.  “They  do  not become part of the tradition to be improved and extended ...” Gombrich  likened them to “random mutations ... weeded out by a principle of natural  selection”.73 That may be a bad simile. They were not random. The Egyptian  sculptor and the medieval jurists must have regarded them as nice touches  although not the basis for a new method. The principle of natural selection  did not work against them, but neither did it work for them so that they could  be developed and refined. Another difference between the medieval and modern jurists is that, in  modern law, jurists apply texts that are comparatively recent, or, at least, not  as remote from their own time as the Roman texts were from the time of  the medieval civilians. Moreover, when new problems arise, modern jurists  70 Bartolus, Commentaria at D 43.12(11).2 no 5 (n 36). 71  Tyler v Wilkinson 24 Fed Cas 472, 474 (1827) (Case no 14312). 72  Bartolus, Commentaria at C 1.4 no 47 (n 36). 73  E M Gombrich, Art and Illusion: A Study in the Psychology of Pictorial Representation, 2nd edn  (1961) 143.

ius quaerens intellectum: the method of medieval civilians 

91

can often turn to new cases and statutes that address them and have become  authoritative  texts.  In  contrast,  the  medieval  jurists  worked  with  a  closed  body of ancient texts, those of the Corpus iuris. Supposedly, the holy Roman  emperors,  as  successors  of  the  Caesars,  could  alter  or  add  to  these  texts,  but  they  rarely  did  so.74  Much  of  medieval  legislation  did  deal  with  new  problems, but it was the work of princes or municipal authorities. It did not  become part of the ius commune, the law based on the Corpus iuris, which,  in the eyes of the civilians, was common to all principalities and cities. Consequently, the medieval jurists were more likely to find themselves confronted  with a problem that had no parallel at the time their authoritative texts were  written, and that the authors of these texts could not have imagined, let alone  addressed. When the medieval jurists tried to resolve such a problem, they  became creative in the extreme. They had to cite texts that had nothing to do  with the problem. A  good  illustration  is  their  discussion  of  what  we  call  conflict  of  laws.  Suppose  the  statutes  or  customs  of  Bologna,  Padua  and  Modena  differ.  Which law should apply when suit was brought in one city over an event that  had occurred in a different city or between citizens of different cities? The  Roman jurists had never addressed the problem. They had distinguished the  ius gentium,  which  governed  matters  such  as  sales  and  applied  to  people  regardless  of  their  nationality,  from  the  ius civile  which  governed  matters  such  as  marriage  and  applied  only  to  Romans.  They  had  allowed  subject  peoples in their Empire to keep much of their own law. But they had never  addressed  the  question  of  whether  a  Roman  court  should  apply  someone  else’s law instead of its own. The medieval jurists believed this question should  be resolved by the ius commune, a law common to all cities and principalities,  rather than by a statute or custom of any one of them. Iacobus  de  Arena  proposed  some  basic  rules  that  were  accepted  by  Cinus of Pistoia and refined by his pupil Bartolus. They still sound familiar.  According to Iacobus, if a question concerns procedure, it is to be resolved by  the law of the forum; if it concerns contract, it is to be resolved by the law of  the place of contracting; if delict, meaning tort or crime, by that of the place  where the wrong was done; and if property, by the law of the place where  the  property  was  located.  As  a  medieval  civilian,  Iacobus  was  required  by  his method to cite a Roman text for each of these propositions. The texts he  cited show the lengths to which medieval jurists might go when there were  no good texts on hand.  74 M Bellomo, The Common Legal Past of Europe 1000-1800 (1995) 67-68.

92

the creation of the ius commune

For the proposition that, in matters of procedure, the judge should apply  the  law  of  his  own  court,  he  cited  a  text  that  advises  a  Roman  judge  that  “although witnesses should not be lightly summoned from long distances”,  “he should find out what the practice is in the province of which he is the  judge”, and summon witnesses from another city if that is the practice.75 To  show that a contract should be governed by the law of the place where the  contract was made, he cited a text that said that the warranty against eviction  in a sale of land should be adapted to the custom of the region.76 To show  that,  in  delict,  one  should  look  to  the  place  the  wrong  was  committed,  he  cited a text that discussed what a master must pay when his slave cuts down  a neighbour’s tree, and a law imposes a “particular fine” on a man who does  so.77 Here he was helped by a gloss of Accursius which said that the fine was  imposed by the law “of the place where the trees were”.78 Jacobus also cited  a text which said that some crimes which occur in one province do not occur  in others. For example, in Arabia, the enemies of a person commit a crime  called σκοπελίσμον by piling stones on a person’s land as a sign that anyone  who cultivates it will die horribly. That crime should be punished, according  to the text, because it violates the principle that it is wrong to threaten another’s life.79 On the law that applies to suits over property, Iacobus cited one  text that concerns what a magistrate should do when a house is in a ruinous  state,80  and  another  that  deals  with  a  guardian  who  paid  taxes  on  a  ward’s  estate that were not legally due but are paid by the land owner, rather than  his tenants, by the custom of the region.81 These texts merely show that, in  Roman times, local practice mattered when it was a question of summoning  witnesses, construing implied warranties, fining people who cut down trees,  keeping buildings in good repair, and paying taxes that are customary but not  legally due. Moreover, vengeance can take odd forms in places like Arabia.  None of these texts concerns the conflict of laws. By  the  fourteenth  century,  the  creativity  of  the  civilians  was  drawing   criticism.  Richardus  Malumbra  complained  that  “our  older  fathers  and  doctors ... argued from the witness of our laws which were close to the matter  in question”. Now, he said, instead of “sticking to the text and the gloss and  the  opinions  of  the  most  respected  doctors”,  the  jurists  “turn  to  fables  or  75 76 77 78 79 80 81

D 22.5.3.6. D 21.2.6. C 3.41.2. Gl ord at C 3.14.2 sv saltui. D 47.11.9. C 8.10.3. D 26.7.37.6.

ius quaerens intellectum: the method of medieval civilians 

93

make  arguments  so  logistic  and  sophistic  that  they  have  no  truth  but  only  its  appearance”.82  Perhaps  he  had  in  mind  imaginative  uses  of  the  Roman  texts like the one we have just seen. If so, the criticism may have been a bit  unfair. There were only so many Roman texts, and no legal problem could be  resolved without citing them. c. seekIng order As mentioned earlier, the concern of the medieval jurists was not solely to  apply  their  texts  to  situations  that  the  texts  did  not  expressly  cover.  They  wanted to explain each text of the Corpus iuris in terms of every other. That  effort  led  them  to  impose  an  order  on  the  texts  that  the  Romans  had  not  contemplated. Their method of finding order in the law was neither Roman nor modern.  As mentioned already, the Roman jurists worked by testing their concepts  against  particular  cases.  To  do  so,  they  needed  a  good  stock  of  concepts  but  not  much  structure  beyond  them.  Since  early  modern  times,  jurists  who  have  sought  order  have  tried  to  explain  rules  by  higher-level  principles  or  concepts.  I  have  described  elsewhere  how  this  approach  began  in  the sixteenth century with the work of the late scholastics, who, for the first  time, organised Roman law into doctrines that they tried to explain by higher  principles.83 Their approach was taken up in the seventeenth and eighteenth  centuries by members of the northern natural-law school founded by Hugo  Grotius and Samuel Pufendorf. Nineteenth-century jurists tried to explain  rules in terms of basic concepts which, in turn, they explained by still more  basic concepts. Medieval jurists sought order but not in this way. They sought order by  linking related texts. Sometimes they juxtaposed Roman texts that used the  same word or expression. The meaning of a word in one text was identified  or distinguished from the meaning of the same word in another. Sometimes  they ascribed explanatory power to a term found in one text. They treated  it as the key to understanding other texts that may not have used the same  term.  Sometimes  they  created  a  schema.  Differences  or  similarities  in  results were explained by identifying an element that they had in common  – an element that might or might not be found in a text — and then identifying  a  circumstance  that  supposedly  accounted  for  the  difference  or  the  similarity. 82 Quoted in Albericus, Commentaria, Proemium (n 27). 83 J Gordley, The Philosophical Origins of Modern Contract Doctrine (1991) 69-111.

94

the creation of the ius commune

Sometimes the linkage of texts illuminated an area of law better than if the  texts were read singly. In these instances, a modern reader is apt to imagine  that  the  medieval  jurists  were  groping  towards  a  principled  explanation  of  an area of law. That is the sort of explanation he is used to. He may regard  their work as a primitive attempt at a principled explanation. At other times,  the medieval jurists did little more than cross-reference and classify the texts  that bore upon the same area of law but were dispersed in the Corpus iuris.  Reading the texts together was no more illuminating than reading them singly.  The modern reader is apt to think that the medieval jurists were attempting  to make something like a concordance or a card file. It is true that the order  the medieval jurists found in their texts served the purposes for them that  would  be  served,  for  a  modern  jurist,  by  a  principled  explanation  or  by  a  card file: in the one case, illuminating an area of law; in the other, labelling  and  cross-referencing.  Nevertheless,  the  medieval  jurists  themselves  were  neither searching for principled explanations nor merely trying to label and  cross-reference. Their goal was to assign each text a place in the scheme of  things by relating it so far as possible to every other text. (1) order through the juxtaposition of texts One way the medieval jurists found order was by explaining one text by juxtaposing others that used the same word or expression. One of the other texts  might use a different word or expression that appeared in still a further text;  this could then be linked to the second text and through it back to the first.  The end result was a chain of interconnections among texts in which each text  was supposed to shed light on the meaning of every other. Sometimes the juxtaposition of texts led to a deeper insight than one might  have had reading the texts in isolation. An example is the jurists’ discussion  of  the  role  of  consent  in  contract  formation.  In  Roman  law,  only  certain  contracts such as sale and lease were legally binding upon consent. Others,  such  as  gratuitous  loans  for  use  and  consumption,  were  binding  when  a  certain object was handed over. Others were binding upon completion of a  formality involving the use of certain words. All of these contracts required  consent. As Ulpian said, quoting Pedius, “there is no contract, no obligation  that does not consist of consent, whether it is formed by the handing over of  something or by the use of certain words …”84 But each contract had its own  rules as to when it became binding. By  juxtaposing  Roman  texts,  the  medieval  civilians  arrived  at  or  sharp84 D 2.14.1.3.

ius quaerens intellectum: the method of medieval civilians 

95

ened the insight that, in principle or by nature, all contracts are binding by  consent although not all are legally enforceable. That insight may have been  glimpsed  by  the  Roman  jurists.  When  the  texts  were  linked,  it  appeared  more clearly. Accursius explained: There are two roots of an effective obligation. One is natural. The natural root  arises out of consent which can be given by all men, even slaves. D 2.14.1.3 ...  Nevertheless  this  natural  root  does  not  have  the  force  to  compel  performance  of the obligation. It does prevent a person who has performed from taking back  his performance. D 46.1.16.3. To this natural root a civil root is sometimes joined  which gives efficacy or form or clothing to the natural root so that it can produce  an obligation.85

In this passage, Accursius pulled together two lines of Roman authority. The  first  is  the  text  he  cited  as  an  example  of  a  natural  obligation,  D  2.14.1.3.  This is the text just mentioned in which Ulpian said that all contracts require  consent.  The  second  is  the  Roman  distinction  between  natural  and  civil  obligations. Natural obligations were created by certain contracts that were  not directly enforceable, such as a contract made by a slave or by a father with  his  son.  These  contracts  nevertheless  had  certain  legal  consequences,  two  of which are mentioned in D 46.1.16, the last text Accursius cited: a natural  obligation could be the subject of a guarantee, and it could be raised as a  defence if a party performed and later sued to recover his performance. In  Accursius’ view, then, consent to real and formal contracts produces a natural  obligation that must then be “clothed” by the delivery of the object or the  formality to produce a civil obligation. Another  Roman  text  said  that  “an  obligation  is  a  legal  bond  which  ties  us to the necessity of delivering some thing in accordance with the laws of  our state”.86 Accursius thought that the text applied to both natural and civil  obligations. Since the civil law did not compel performance of a natural obligation, Accursius concluded that “whenever a person is said to be bound by a  natural obligation, the statement should be understood of the ius gentium”. According  to  other  texts,  the  ius gentium  was  a  law  governing  matters  such  as  sales  that  applied  to  people  regardless  of  their  nationality.  It  was  distinguished from the ius civile, which governed such matters as marriage  and applied only to Romans. In several texts, the distinction is dressed up  with some philosophical speculation. The ius gentium is the law “established 

85 Gl ord at J Inst 3.14pr sv necessitate. 86 J Inst 3.13pr.

96

the creation of the ius commune

among all men by natural reason”.87 The ius gentium emerged in response to  human necessities; to it pertain such things as war, captivity and slavery, since  by the law of nature all were born free. “And by the ius gentium, nearly all  contracts were introduced such as sale and purchase, lease and hire, partnership, loan for consumption and others without number.”88 Thus, by pulling  all these texts together, Accursius concluded that, according to the law established by natural reason, all contracts give rise to an obligation upon consent,  although the obligation is natural and not civil. One cannot say whether, by  linking these texts, he derived a new insight or merely clarified the insights of  the Roman jurists. In any case, he had reached a conclusion that, at the most,  was implicit in the texts. At other times,  by  juxtaposing  their  texts,  the  medieval  jurists  classified  them without coming any closer to an insight that would explain them. An  example is a distinction they drew between “natural” and “civil possession”.  In  one  text,  Ulpian  said  that  “if  someone  has  gone  away  from  his  field  or  his house leaving none of his household there”, he does not lose possession  by  going  away,  and  so  he  can  bring  an  action  for  possession  if  he  finds  an  intruder upon his return.89 In another text, Paul said that a person “naturally”  possesses what he stands upon, and that “the ownership of things originated  in  natural  possession  …  a  relic  of  which  survives  in  those  things  that  are  captured on land, sea and in the air, for they become the property of those  who  first  take  possession  of  them”.90  Ulpian  said  that  both  “natural”  and  “civil” possession was protected, although in the context of a quite different  problem: protecting a possessor who does not have legal title.91 Putting their  statements  together,  the  twelfth-century  jurist  Hugo  concluded  that  there  were two kinds of possession: natural, when one was in physical control: and  civil, as in the case Ulpian put where “someone has gone away from his field  or house”, a conclusion that was adopted by Accursius.92  The distinction between natural and civil possession appeared in leading  treatises such as that of Johannes Voet on into the eighteenth century.93 It was  useful as a label for the obvious point made by Ulpian: that one does not lose  possession whenever one leaves one’s house. But it is merely a label. It does  not shed light on why possession ought to be protected. 87 88 89 90 91 92 93

J Inst 1.2.1. The same phrase appears in D 1.1.9. J Inst 1.2.1. D 43.16.1.24. D 41.2.1pr; D 41.2.1.1. D 43.16.1.9. Gl ord at D 43.16(15).1.9 sv civiliter. See, e.g., Voet, Commentarius 41.2 § 3 (n 33).

ius quaerens intellectum: the method of medieval civilians 

97

(2) order through ascribing explanatory power to terms in certain texts Another way in which the medieval civilians found order in their texts was to  ascribe explanatory power to a term found in one text, treating it as the key to  understanding other texts. Again, some of the conclusions they reached were  more illuminating than others. A success, to judge  by  its  later  history,  was  the  formulation  by  Bartolus  of what came to be called the doctrine of causa. According to this doctrine,  for the parties to enter into a contract that the law will respect, they must  have done so for one of two causae or reasons: either out of “liberality”, or to  receive something in return. The word causa appeared in three key Roman  texts. One text used that term to explain that, in Roman law, not all contracts  were legally actionable: “When there is no causa, it is accepted that no obligation can be constituted by an agreement; therefore a naked agreement does  not give rise to an action although it does give rise to a defense (exceptio).”94  Accursius  and  Jacques  de  Revigny  had  described  the  causa referred  to  by  the first text as simply “something given or done”.95 The other two texts said  that the formal contract of stipulatio had to have a causa.96 The causa in the  second, according to Accursius, was re or spe, a thing or the hope of a thing.97  According to Jacques de Revigny and Pierre de Belleperche, the texts meant  that  a  stipulatio  is  not  valid  if  a  party  mistakenly  believes  there  is  a  causa  or reason for obligating himself.98 These texts might seem an unpromising  source for a general ordering principle of contract law. Yet Bartolus found  one. When the first text spoke of an agreement without a causa, he said, it  meant “no causa fulfilled (impleta), but the agreement was certainly made  for a causa”.99 Consequently, one sort of causa was the receipt of something  in return. In dealing with the texts concerning stipulatio, Bartolus explained  that the causa might be the receipt of something in return. That alone was  not enough, however, since a stipulatio was binding even if a promise was  gratuitous. In that event, Bartolus explained, the causa was “liberality”. There  were, then, two kinds of causa.100   94 D 2.14.7.4.   95 Gl ord at D 2.14.7.4 sv causa; Jacques de Revigny, Lectura super Codice, published under the  name of Petrus de Bellapertica, at C 2.3.10 (1519; repr 1967) in OIR I.   96  D 12.7.1; D 44.4.2.3.   97  Gl ord at D 44.4.2.3 sv idoneum.   98  Jaques de Revigny, Lectura to C 2.3.5 (n 95); Petrus de Bellapertica, Commentaria in Digestum Novum to D 44.4.2 (1571; repr 1968).   99  Bartolus, Commentaria at D 12.4.7.2 (n 36). 100  Bartolus, Commentaria at D 44.4.2.3 (n 36).

98

the creation of the ius commune

Baldus provided the doctrine with a rationale: “Without a causa equity will  not say that an action arises lest one party use his substance badly and the  other be unjustifiably enriched.”101 He seems to have meant that in the case of  a party who is not using his substance badly. If he were, as he said elsewhere,  it would be “foolishness” rather than “liberality”. 102 If he receives something in  return, neither party is unjustifiably enriched. As he explained elsewhere, “in  contracts, equity or equality must be served”,103 a requirement he identified  both with the remedy for laesio enormis104 and with the principle of a famous  text that said, “it is equitable that no one should be enriched by another’s loss or  injury”.105 Since equity required that there be a causa, Baldus read the doctrine  into canon law in which non-formal agreements were binding without regard  to the Roman rules. That was so, he explained, provided they had a causa.106 Less  successful  in  retrospect  was  their  attempt  to  find  a  basis  in  the  ius gentium for the Roman distinction between contracts that were and were not  binding  upon  consent.  Bartolus  did  not  reject  Accursius’  view  that  consent  gives rise to a natural obligation by the ius gentium. Nevertheless, he tried to  find a basis in the ius gentium for the distinction between consensual and real  contracts. He claimed that the ius gentium gave these contracts their “name”.  The  “name”  made  these  contracts  actionable,  for  “nominate  contracts  give  rise to an action by this alone: that they exist and have a name”.107 Consensual  contracts were binding on consent and real contracts only on delivery of an  object,  because  of  a  difference  in  their  names.  Consensual  contracts  such  as sale took their names from an act a party performs by agreeing; I can sell  you my house today by so agreeing even if I do not deliver it to you until next  month. Real contracts such as deposit took their names from an act a party  performs by delivering: I cannot say I am depositing an object with you unless I  am actually depositing it right now.108 Baldus not only accepted this explanation  but concluded that innominate contracts were unenforceable in canon law.109 101 Baldus degli Ubaldi, In Decretalium volumen commentaria at X 1.4.11 no 30 (1595). 102 According to Baldus, if the causa were not stated expressly in the notarial document that had  become the accepted substitute for the Roman stipulatio, one should presume the contract was  made out of “foolishness” rather than “liberality”: Baldus, Commentaria at C 34.30.13 no 14 (n  32). 103  Baldus, Commentaria at C 4.44.2 no 48 (n 32). 104  Baldus, Commentaria at C 4.44.2 no 48 (n 32). 105  D 50.17.206. See Gordley, Philosophical Origins (n 83) 54-57. 106  Baldus, Commentaria at C 3.36.15 no 3 (n 32); see Baldus, In Decretalium volumen at X 1.35.1  (n 101).  107  Bartolus, Commentaria at D 2.14.7.5 nos 14-16 and at D 2.14.7, no 2 (n 36). 108  Bartolus, Commentaria at D 2.14.7 no 2 (n 36). 109  Baldus, In Decretalium volumen at X. 1.4.11 no 30 (n 101); Baldus, Commentaria at C 3.36.15  no 3 (n 32).

ius quaerens intellectum: the method of medieval civilians 

99

(3) order through a schema Another way in which the medieval civilians found structure in their texts was  by making a schema. Differences or similarities in results were explained by  identifying an element which they had in common and then a circumstance  which supposedly accounted for the difference or the similarity. Sometimes, the result was a mere list. An example is Accursius’ account of  the mistakes that vitiate consent to a sale. Ulpian had said: The  sale  is  invalid  if  there  is  disagreement  either  as  to  the  fact  of  sale  (in ipsa emptione) or the price or any other matter. If, therefore, I thought I was buying  the Cornelian estate and you that you were selling the Sempronian, the sale is void  on the ground that we were not at one as to the physical thing sold (in corpore) …  Of course, if we are merely in disagreement over the name but at one on the actual  thing, there is no doubt that the sale is good; for if the thing is identified, a mistake  over its name is irrelevant. Next comes the question whether there is a good sale  if there is no mistake as to the physical thing (in corpore), but there is in regard to  its substance (in substantia), for example where vinegar is sold for wine, or copper  for gold, or lead or something else resembling silver for silver.110

Ulpian concluded that consent had not validly been given at least in the case  of the copper sold for gold and the lead sold for silver. Accursius noted that  consent was lacking if there was an error about (1) whether there was a sale,  (2) price, (3) corpore, (4) substantia, and (5) materia.111 The schema does not  tell us much more than we could see from Ulpian’s examples. Sometimes  the  schema  explained  nothing.  For  example,  Accursius  proposed a three-fold classification of fault (culpa): “the fault is sometimes  within the act; sometimes it precedes the act; sometimes it follows the act”.112  He found instances of each in his texts. Fault is within the act when a man  throws another off a bridge,113 or a doctor operates negligently on a slave.114  It precedes the act when someone digs a pit to catch bears or deer in a public  place and something falls in,115 or when a person who has borrowed silver for  a dinner party takes it aboard ship where it is lost to shipwreck or pirates.116  It follows the act when a doctor operates correctly but neglects after-care.117  110  This refers to D 18.1.9. 111  Gl ord at D 18.1.9 sv aliquo alio. He had a sixth category as well: error in “sex”. That category  was based on D 18.1.11.1, which said that the buyer of a slave could void the sale if he mistakenly thought that the slave was male but not if he knew she was female but mistakenly thought  she was a virgin. 112  Gl ord at D 9.2.8pr sv sed culpa reus. 113  D 9.2.7.7. 114  D 9.2.7.8. 115  D 9.2.28pr. 116  D 44.7.1.4. 117  D 9.2.8pr.

100

the creation of the ius commune

He classified his texts into these three categories, but the classification does  not tell us much about fault. Nevertheless, sometimes the schema suggested a reason why results should  be similar or different. An example is a series of distinctions concerning fraud  that  seem  to  have  first  appeared  in  a  gloss  to  the  Liber pauperum  of  the  eleventh-century jurist Vacarius,118 and were subsequently picked up by Azo  and  Accursius.119  Fraud  in  the  causa,  which  led  a  person  to  contract  who  otherwise  would  not  have  contracted,  was  distinguished  from  incidental  fraud, which led a person to contract on worse terms than he otherwise would  have accepted. Incidental fraud was distinguished into fraud ex proposito if  it was practised intentionally, and fraud ex re ipsa if a party paid too much or  too little though the other party had done nothing intentionally to defraud  him. In the first case, the victim had an action for the difference between  the amount he paid and the amount he should have paid, however small that  difference might be. In the case of fraud ex re ipsa, he had an action only if  the difference exceeded half the just price. They cited a text for each branch  of  these  distinctions.  Here,  the  schema  was  not  merely  a  list.  It  identified  the evil to be remedied. In the case of intentional incidental fraud and laesio enormis the evil was the same: an unfair price. However, in the one case, the  unfairness was brought about intentionally, and would be remedied however  small it was, while in the other, it was remedied only if it was greater than  half the just price. d. conclusIon We  have  seen  that  the  medieval  jurists  faced  two  tasks  which  the  Roman  jurists did not. One was to apply authoritative texts to new situations. The  Romans had few texts, and their opinions, for the most part, were not based  on the authority of texts but on their own authority or that of other jurists.  The second task was to find order in the way that different areas of law were  related to each other. The Romans had not felt the need to do so. In carrying out the first of these tasks, the medieval jurists developed ways  of applying texts to new situations that have been used by jurists ever since.  Sometimes they generalised a text to apply to a wider number of  situations;  118 Vacarius, Liber pauperum, ed F de Zuleta, at D 19.2.23.3 [= Selden Society XLIV] (1927; repr  1972) 51. 119  Azo Portius, Summa Codicis at C 2.20 no 9 (1557). See H Fitting (ed), Summa trecensis (1894)  [published under the title Summa Codicis des Irnerius] lib 4 tit 41; A Gaudentius (ed), Rogerius,  Summa Codicis at C 4.44, in BIMAE I; Hostiensis, Summa aurea, (1574) iii. 17, par. 7. See also  G Fransen, Le Dol dans la conclusion des actes juridiques (1946) 49-55.

ius quaerens intellectum: the method of medieval civilians 

101

sometimes they limited it by distinguishing situations in which the text would  apply from situations in which it would not; and sometimes they drew analogies  with  a  different  situation  from  those  their  texts  already  considered.  Modern jurists do the same. The principal differences are that modern jurists  use  other  methods  as  well,  and  that  the  medieval  jurists  were  sometimes  forced  to  use  their  texts  more  imaginatively  to  apply  to  problems  that  the  authors of their texts had never considered. In  carrying  out  the  second  of  these  tasks,  the  medieval  jurists  found  a  larger order in the law in a way that was neither Roman nor modern. The  Roman jurists were not much concerned with a larger order. When modern  jurists have sought one, they have tried to explain rules by larger principles  or concepts. The medieval jurists sought order by linking texts to each other,  sometimes  juxtaposing  those  that  used  the  same  terms,  sometimes  taking  a term used in one text as an explanation for the rules found in others, and  sometimes making a schema in which differences or similarities among texts  were  explained  by  identifying  an  element  that  the  texts  had  in  common  and another that supposedly accounted for the difference or the similarity.  Sometimes through the use of these methods, the medieval jurists seemed to  have achieved the sort of insight that moderns have sought through a principled explanation; at other times, they seemed to have achieved only the sort  of order that moderns obtain with a concordance or a card file. Of  course,  it  is  only  in  looking  backwards  that  one  can  say  that,  with  a  few exceptions, there was nothing distinctively medieval about the way these  jurists  carried  out  the  first  task,  and  nothing  modern  about  the  way  they  carried out the second. It is only in looking backwards that one can distinguish the one task from the other. For the medieval jurists, it was all a matter  of understanding their texts, each text in terms of every other.

4  Medieval Family and Marriage Law: From Actions of Status to Legal Doctrine Laurent L J M Waelkens A. INTRODUCTION B. THE ROMAN LAW OF PERSONS C. ROMAN FAMILY LAW D. MARRIAGE (1) Roman case law (a) Coniunctio and coniugium (b) Vir et uxor (c) Concubinage (d) Nuptiae (e) Matrimonium (2) Early medieval legal texts about matrimony (3) Twelfth-century case law about matrimony E. MARRIAGE AND FAMILY: FROM CASUS TO REGULA A. INTRODUCTION The theme “from casus to regula” is particularly interesting in the law of persons. The sources of classical Roman law contain only case law (casus) about family matters and personal relations between house-mates and relatives. There were no rules (regulae). The way in which, at the end of the twelfth century, this case law was shaped into rules and systematic doctrine is significant for the history of medieval law.

103

104

the creation of the ius commune

B. THE ROMAN LAW OF PERSONS In  classical  Latin  a  persona  was  a  theatre  mask,  hence  a  character  and  a  theatre player. In legal texts the word was used for the actors in a law suit.  When the word occurs in the Corpus iuris civilis, it has thus to be understood  as referring to one of the men present in court. It is accordingly found used  for a plaintiff or pursuer, for a defender or defendant, and for the judge. An  action in personam was not understood as an action against a person, but as  an action against the defendant whose name was mentioned in its formula. Gaius commenced his Institutes with the law of persons. In the first part  of  his  book  he  described  the  situation  of  men  in  court.  His  reflections  on  slaves and family members are related to their representation in court. It is  impossible to understand the first part of his Institutes without realising that  the contents are about civil courts. Contemporary Romanists have to fight  the  tendency  to  find  in  Gaius  rules  of  modern-time  civil  institutions.  The  example of tutela illustrates this well. The relevant passages (Gai Inst 1.142200) do not explain the institution of guardianship, but rather the way women  and  under-age  patresfamilias  were  represented  in  the  military  civil  court.  In  the  early  Republic  civis  meant  brother-in-arms  and  civil  legal  proceedings were military. Civil courts were only accessible to soldiers. Women and  children, however, could appear in the civil court, as they were recognised as  successors to deceased soldiers. When they appeared in the civil court, they  were represented by a soldier. This procedural representation remained when citizenship was no longer  reserved  to  fighting  men.  For  women  it  existed  until  the  Imperial  procedure by cognitio for lawsuits was developed, in which women received free  access to the court according to the rights of all freemen (ius liberorum).1 In  Republican civil courts the representative of a woman played only a formal  role. If she arrived alone in court, the praetor appointed a tutor from among  the bystanders. The latter had nothing to fear, because the woman could not  sue her tutor. On the other hand, by Gaius’ time, the tutor of children was  liable for his actions in representing a pupil. This liability developed into the  institution of guardianship for children. Inspired by this institution, modern  Romanists  have  translated  tutela  as  guardianship  and  understood  tutela mulierum as  guardianship  of  women,  transplanting  medieval  and  modern    1 That is, access to the Imperial court in the manner of all free people (liberi). It is necessary to dismiss completely from one’s mind the idea that women gained independent access to the court through having children, as found expounded in Kaser, RPR2 320-321 and some recent literature based on this old interpretation. Gai Inst 1.44 has to be understood as case law.

medieval family and marriage law 

105

rules  into  the  old  case  law.2  This  kind  of  transplant  of  modern  notions  to  antiquity undermines the proper legal understanding of Roman case law. In  Gaius’ Institutes, tutela is in fact a mere aspect of an appearance in court. The meaning of the persona in the Roman law as “a player in a lawsuit”  was still present in the thirteenth centu ry. Discussions about the question of  whether witnesses are persons are still to be found in glosses and comments.  Most glossators accepted that they were.3 In his famous theory about persona representata, Sinibaldus Fliscus sought to explain how a deceased body could  still be liable or present in court.4 There was nothing about his identity, his  personality or his legal personality! Hostiensis taught us that, in opposition to  the legists, canonists were not only resolving the conflicts of people in court,  but that they were also concerned with the care of their souls.5 Nevertheless,  for Hostiensis, personae appeared in court as litigants, judges and witnesses.6  In the fourteenth century the idea of persons as beings with human dignity  appeared.7 But it was not until the early modern period that legal doctrine  provided that all people received rights at birth (iura naturalia), which meant  that  everyone  achieved  the  status  of  person,  without  it  being  necessary  to  appear in court.8  This therefore is the first obstacle to understanding the medieval law of  persons. That law dealt with legal procedure, not with the inalienable rights  of persons. In a short chapter such as this, it is impossible to review how the  Roman law about persons was transformed into the modern law of persons,  from old case law to modern rules. The focus will accordingly be on family  and marriage.

  2 The  argument  is  more  fully  developed  in  L  Waelkens,  “Tutela  mulierum  herbekeken”,  in  D  Heirbaut  (ed),  Festschrift Monique Van Melkebeek  [=  Iuris  Scripta  Historica  XXV]  (2009)  (in  press).    3  Y  Mausen,  Veritatis adiutor, La procédure du témoignage dans le droit savant et la pratique française (XIIe–XIVe siècles) (2006) 25-28.   4  For  bibliography  on  this  question,  see  R  Feenstra,  “L’histoire  des  fondations:  À  propos  de  quelques études récentes” (1956) 24 TvR 381 at 414-419.   5  Consider his theory about forum internum or forum conscientiae.   6  See  the  passages  about  persona  mentioned  in  [Hostiensis],  Index locupletissimus rerum ac verborum notabilium ex Hostiensis lecturis super quique libros Decretalium (1631; repr 1965) 86.    7  A Lefebvre-Teillard, Introduction historique au droit des personnes et de la famille (1996) 41-43.   8  Most famously found in H Grotius, De iure belli ac pacis I.iv.4, ed B J A de Kanter-van Hettinga  Tromp (1939; new edn with notes by R Feenstra, 1993) 31.

106

the creation of the ius commune

C. ROMAN FAMILY LAW Familia  constituted  an  important  element  of  the  Roman  civil  law.9  It  was  essential  to  participation  in  a  civil  lawsuit.  The  civil  law  originated  in  the  army. Roman soldiers obtained special rights or privileges as an inducement  to  participate  in  military  campaigns.  All  these  privileges  were  attached  to  the “status of liberty”.10 Most veterans kept this status on leaving the army.  From the fourth century bc on, peace treaties (foedera) granted aspects of  the  status  of  liberty  to  other  peoples.  Provinces  that  promised  loyalty  and  payment of taxes to the senate received a partial status of liberty. The extent  of the grant depended on what they had to offer to Rome. Liberi  (later  understood  as  freemen)  had  a  series  of  political  rights,  of  which the most important was a special procedure to settle mutual conflicts.  Roman  citizens,  who  were  brothers-in-arms  (cives),  or  veterans  who  kept  their military privileges, brought mutual conflicts before the military praetor,  who took into consideration their status of liberty. Veterans and those peoples  who  had  acquired  liberties  under  their  treaty  (foedus)  went  to  their  own  peregrine praetor for an initial application. The praetors checked the liberties, and gave instructions about the law to apply to the judges who had to  decide civil cases. This law made by the praetors was the civil law.11 The  idea  of  familia  was  central  to  the  civil  lawsuit.  It  is  found  as  early  as the Twelve Tables in connection with the military comitia centuriata. A  citizen could invoke his civil rights in favour of his entourage. He could therefore  appear  in  the  praetor’s  court  for  his  relatives,  who  could  thus  benefit  from  his  status  of  liberty.  On  the  other  hand,  when  he  lost  his  case  –  and  one party loses in every case – he became liable himself. The civil law was  accordingly developed through citizens acting in court for their friends and  for famuli – those people sharing their home. Those whom a citizen, using  his  own  civil  rights,  defended  in  court  constituted  his  familia  (the  word  is 

  9 A wonderful study providing a critical analysis of the Roman family is J F Gardner, Family and Familia in Roman Law and Life (1998).  10  The origin of the word libertas is difficult to retrace, because we have very few old Latin texts,  and most scholars are influenced by the meaning it had in classical literature. It is obvious that  it had to do with military service. The word liber could have indicated an able-bodied, fighting  man. On Rome’s Aventine Hill stood the temple of Liberty, which housed the administration of  the censors. The census-tables were kept there.  11  On this interpretation of the Republican formulary procedure, see L Waelkens, Civium causa, Handboek Romeins recht  (2008)  48  and  123-134;  on  the  theory  of  liberties,  see  L  Waelkens,  “De Romeinse oorsprong van de fundamentele rechten en vrijheden” (2003) 71 TvR 187; and L  Waelkens, “The Status of Liberty, from Roman Civil Law to Human Rights”, in B Raymaekers  (ed), Lectures for the XXIst Century (2008) 187.

medieval family and marriage law 

107

probably derived from famuli, “close people”).12 One main significance of a  litigant’s familia was his liability for the individuals who constituted it. Familia  was thus devel oped in civil law as a structure of liability. For the judges and  praetor who supervised execution of judicial decrees, a citizen who pleaded  the cause of a member of his familia was liable not only with his own assets,  but also with the assets of all family members for whom he could appear in  court. From the point of view of the court and according to civil procedure,  a litigant who became liable under a decree had only one estate: the community of goods that was made up of his own estate and those of the members  of his family. This community of goods made it necessary to establish clearly  who was a member of which familia. In Gaius’ Institutes and the other sources of classical law the familia can  thus be understood as an ascertained group of people acting together in civil  law, represented in court by a single paterfamilias, and liable according to the  value of the common estate. The sources contain case law about membership  of and release from the family. These show that the central feature of the  familia remained unchanged in Gaius’ Institutes from the Twelve Tables. The  familia was about neither kinship nor acquaintance: rather, it was concerned  with  acting  in  court.  The  family  did  not  originate  in  marriage,  but  in  civil  liability, in dealing with problems about assets and liabilities. Assets and liabilities can be expressed in book-keeping, and double-entry  book-keeping  is  a  remnant  of  the  old  Roman  family.  The  daybook  (codex expensi et accepti) expressed the position of the paterfamilias as debtor and  creditor  as  regards  other  citizens.  The  ledger  (codex rationum)  expressed  the  position  of  the  family  members  in  their  relationship  with  the  paterfamilias.  Since  the  latter  was  also  a  family  member,  he  too  had  his  account  in the ledger. What the paterfamilias owed to another citizen according to  the daybook was expressed in the ledger as due to him by a family member.  What he claimed as due to him in the daybook was expressed in the ledger as  owed to a family member. This feature of double-entry book-keeping – that  credits and debts are opposite in daybook and ledger – is due to its origin in  Roman family law.13 Family members created “natural” obligations between  themselves;14 but these could not be enforced in court. In classical Roman  law the paterfamilias could only be sued by other patresfamilias and not by  citizens from his own family. 12 Lewis and Short, A Latin Dictionary (many edns), sv familia. 13 About Roman double-entry book-keeping, see L Waelkens, “Gaius IV, 73: debet ou debetur?”  (2000) 68 TvR 347. 14  The word naturalis in combination with obligatio either developed from the meaning of “origin”  or was a reference to the naturales, an ancient word for children.

108

the creation of the ius commune

Book-keeping is what underpins a business enterprise, the sole word for  which in Latin became familia. Roman texts provide the examples of familia stationis, familia aquarum, familia gladiatoria, familia monetalis, familia thermensis, familia mansionis, familia publicanorum, familia vectigalis, and  so on.15 Familia never means a loving family bringing together wife, husband  and children. D 50.16.195 stated that a single person had a familia and that a  woman who managed her own business had a familia on her own, independent of that of her husband. A woman acting in court was called domina. In  some texts the term materfamilias is used.16 The  Church  Fathers  changed  the  scope  of  familia  by  focusing  on  the  Judeo-Christian  concept  of  kidushin.  Biblical  views  of  the  relationship  between wife and husband spread a new focus on cohabitation. New couples  were  expected  to  leave  the  house  of  their  parents  and  to  establish  a  new  household.17 In both Jewish and early Christian communities this establishment of a new household “under the huppa” was common practice; but it  is completely absent from the case law of the Corpus iuris.18 Nonetheless,  the  fact  that,  in  late  classical  Roman  law,  the  paterfamilias  is  often  called  dominus – the word used in the ius gentium for a litigant – indicates that the  legal concepts of domus and familia remained, as in the old civil law, closely  related to each other. This meant that Jewish and Christian fathers of a family  who respected kidushin had explicitly to emancipate children who married.  In contrast to medieval canon law, classical Roman law did not recognise tacit  emancipation by wedding. Tacit  emancipation  by  cohabitation  starts  to  appear  in  legal  sources  in  the sixth century. Thus, it is explicitly mentioned in texts from Spain19 and  Rhetia.20 In Roman law not only the emancipation of a man, but also that of a  woman, led to the status of dominus or domina, with the capacity to acquire  15 See J-J Aubert, Business Managers in Ancient Rome: A Social and Economic Study of Institores, 200 bc–ad 250 (1994) index at 509. 16  See, e.g., C 1.48.1 or C 6.61.5pr. 17  For the texts, see J Gaudemet, Le mariage en Occident (1987) 43-48. For texts and context of this  new vision, see J Gaudemet, L’Eglise dans l’Empire romain (IVe-Ve siècles) [= Histoire du Droit  et des Institutions de l’Eglise en Occident, ed  G Le Bras, III] (1958) 514-561 (“L’Eglise et la vie  familiale”). When consulting the rich legacy of Gaudemet, it is important to avoid his confusion  of family and marriage. On “kidushin”, see M Meiselman, Jewish Women in Jewish Law (1978). 18  Nonetheless, Vismara saw traces of the Christian family in the law of Constantine: G Vismara, “I  rapporti patrimoniali tra coniugi nell’alto medioevo”, in Il matrimonio nella società altomedievale  (1977) 634-635. 19  A Jordà Fernandez, “Pàtria poestat i emacipació per causa de matrimoni: Del dret romà a la seva  concreció als costums de Tortosa e als furs de Valencia” (2001) 1 Revista de Dret històric Català  93. 20  D Illmer, “Zum Problem der Emanzipationsgewohnheiten im Merovingischen Frankreich”, in  L’Enfant [= Recueils de la Société Jean Bodin XXXVI] (1976) 144.

medieval family and marriage law 

109

rights  and  liability  in  court.  Contemporary  scholars  of  Roman  law  often  forget the first rule of interpretation in the Corpus iuris: all legal texts about  men are applicable to women.21 Christian theology also supported the equal  treatment of man and woman.22 The consequences of tacit emancipation by  cohabitation seem obvious. Under Roman law it would have led to two new  families, one for each spouse; this was the current situation in classical Roman  law.23 In the early Middle Ages, however, this had changed, so that a married  and  emancipated  couple  no  longer  had  two  separate  families,  but  instead  together  constituted  a  single  new  familia.  In  kidushin  husband  and  wife  became duo in carne uno. This caro unus was expressed in a single family for  both. Canonists taught that such tacit emancipation led to the husband and  wife being one family. Likewise, the idea of community of goods appeared  quite early in Christianised regions.24 Thus, in Carolingian times the idea that  a single family was created by cohabitation was common in Europe.25 Canonists promoted this style of family.26 In a study of the so-called “Ligurian law”,  Meijers’ analysis of the twelfth-century jurisprudence led to identification of  a family with a community of goods, which was to be found from Flanders  to Liguria and from Saintonge to Salzburg. At the death of one spouse half  of this community fell open for succession.27 This was clearly the new family  of the early Middle Ages: a family in community with two spouses who were  emancipated by cohabitation and who had equal rights in the family. The canonists and legists of the twelfth century dramatically changed their  views  about  the  family.  At  university  they  studied  the  Corpus iuris.  They  deduced rules about familia from classical Roman case law. A Roman familia  had only one head, one paterfamilias. The Corpus iuris contained case law  about  the  relations  between  the  dominus  and  his  family  members,  mostly  cases about potestas, that is, the authority of a paterfamilias over the family  members  for  whom  he  was  held  liable  and  had  to  pay. These  universityeducated lawyers distilled rules from the cases about the Roman family, and  21 D 50.16.1. 22 See, for instance, J de Churruca, “Egalité et inégalité des conjoints dans le mariage des premiers  siècles”, in J Ganghofer (ed), Le droit de la famille en Europe (1992) 241 with bibliography. 23  As observed by Gardner, Family and Familia (n 9) passim. 24  See, for instance, Vismara, “Rapporti patrmoniali” (n 18) 674-675. 25  Set out, e.g., in K Schmid, “Heirat, Familienfolge, Geslechterbewusstsein”, in Matrimonio nella società altomedievale (n 18) 103. 26  It  was,  e.g.,  introduced  in  the  law  of  Bremen  by  Bishop  Adaldag  of  Bremen:  W  Barkhausen,  “Erzbischof  Adaldag  und  König  Harald  Gormsson,  Über  das  bremische  eheliche  Güter-  und  Erbrecht und seine Beziehungen zu den dänische Landschaftsrechten” (1994) 111 ZRG, GA 363. 27  E M Meijers, Le droit ligurien de succession 3 vols (1928-1936). Meijers studied only the law of  succession, but in Roman law and its aftermath, succession is family law, namely about the liquidation of a family. 

110

the creation of the ius commune

transplanted  them  into  the  relations  of  husband  and  wife.  Case  law  about  appearance in court and liability was transplanted to a new family that originated in cohabitation. In the Middle Ages this cohabitation was marriage. In  order to evaluate properly the evolution of family law from casus to regula,  we  also  have  to  analyse  the  evolution  of  marriage  during  the  reception  of  Roman law. This is the second aspect of this chapter. D. MARRIAGE In classical Roman law marriage had a limited character as a legal institution.  From  the  Middle  Ages  onwards,  however,  it  has  been  common  to  read  of  Roman matrimonial law in the Corpus iuris. In fact, the Corpus iuris contains  case law about relations between individuals in which is discussed coniugium, coniunctio, consortium, concubinatus, contubernium, nuptiae  and  matrimonium. All of these words express different types of sexual relationships.  Every  specific  term  occurs  in  case  law  which  deals  with  other  problems.  There was no total vision about sexual relationships; and no Roman lawyer  developed discussion of their different aspects in an attempt to understand  them as constituting a single legal institution. Thus, when reading the Roman  sources, it is vital to distinguish these different types of relationships clearly.  An interesting fragment in this context is C 5.27.5, the modern discussion of  which is very confused. The Latin text explains a problem of succession in a  complex relationship. The praetorian prefect has to decide who are a couple’s  legitimi, that is their heirs on intestacy. In the case, all the above-mentioned  terminology  is  used.  All  translators  have  confused  the  terminology  about  relationships found in this text, leading to its interpretation as dealing with  the  legitimacy  of  children  born  before  marriage.28  Intestacy,  however,  is  a  question  of  liquidation  of  a  familia,  which,  as  shown  above,  is  a  means  of  structuring liability. The common view that coniugium, nuptiae, and matrimonium leads to a family, which is what modern Romanists understand the  text as about, is typically medieval. (1) Roman case law It is first important to examine the Roman terminology as preserved in the  case law.

28 See L Waelkens, “La légitimité de C. 5, 27, 5”, in F Roumy et al (eds), Mélanges en l’honneur d’Anne Lefebvre-Teillard (2009) 1031.

medieval family and marriage law 

111

(a) Coniunctio and coniugium The terms coniunctio and coniugium both referred to all lasting sexual relationships. There was, however, an important difference between them. A coniugium was a relationship that still subsisted; a coniunctio was a relationship  that had ended. Contubernium, a relationship between slaves, was a coniugium.29 Incestuous partnership was considered a coniunctio or coniugium;30  but it could not be nuptiae.31 Concubinage was coniugium.32 Matrimonium  was not necessarily coniugium: in case of imprisonment there was no coniugium,  but  a  matrimonium  could  continue.33  Coniugium  was  unthinkable  after death; but the Code mentions a case where heirs, claiming goods given  in  dowry,  argued  that  the  matrimonium  of  the  widow  continued  after  her  husband’s death.34 A couple within the prohibited degrees of consanguinity  who lived in coniugium had to be separated;35 this was because incest had to  do with coniugium.36 Saint Augustine wrote of sexual relationships in such  works as De bono coniugali and De coniugiis adulterinis. He never discussed  matrimonium.37 De coniugiis adulterinis contains his celebrated theory that  adultery and bigamy had to do with coniugium.  (b) Vir et uxor The normal terminology for husband and wife, for a couple living together, for  spouses, was vir et uxor. The phrases uxorem ducere and maritum habere were  used in discussing citizens, pilgrims and people whose status was unknown.  The  civil  consequences  of  the  actions  indicated  by  these  terms  depended  on the family position of both spouses and on their status of conubium. The  latter  was  an  old  privilege  in  peace  treaties.  Descendants  of  spouses  with  conubium acquired civil status. The sources show that a woman described as  an uxor was not necessarily in matrimonio.38 Children who contracted nuptiae  were not spouses, but became so when they reached they required age.39 But  couples who lived in an incestuous relationship never became spouses.40 It  29 30 31 32 33 34 35 36 37

D 23.3.39pr. D 48.5.39(38).4. Gai Inst 1.59. Conc. Toledo, can 17; Ovid, Metamorphoses 14.298 and 10.295. C 5.16.24.2; C 5.17.1. C 5.15.2.  D 48.5.39(38).4. J Inst 1.10.4; D 23.2.8. See www.augustinus.it searching sv matrimoni*. All quotations from the works of Saint Augustine  containing the word matrimonium date from after 1100. 38  See, e.g., D 49.15.12.4; D 48.5.7. 39  D 23.2.4. 40  D 48.5.14(13).4.

112

the creation of the ius commune

was necessary to be classed as vir et uxor for the actio rerum amotarum to be  applicable.41 The general term for marriage in the Middle Ages, maritagium,  was derived from the Roman term maritus.  (c) Concubinage Concubinage is a difficult relationship to understand. It is dealt with in the  Digest title, De concubinis (D 25.7). It could only occur where nuptiae or  matrimonium  were  impossible.  Concubinage  was  not  possible  when  the  relationship  was  lewd  and  it  required  monogamy.  The  judge  determined  the  status  of  concubinage  from  the  intention  of  the  woman  (ex sola animi destinatione).42 He discerned the affection of spouses from the affection of  concubines.43  Concubinage  with  a  child  was  impossible.44  If  nuptiae  were  presumed, concubinage was again impossible.45 During the council of Toledo  (400 ce) the bishops agreed that people living in concubinage would not be  excluded from communion, at least if the relationship was monogamous. As  the marital bond was also informal in nature, judges would have to consider  the  aims  and  intentions  of  the  couple  to  decide  their  status.  Concubinage  continued as an accepted relationship until the twelfth century.46 (d) Nuptiae Nuptiae meant wedding with an exchange of vows. D 23.2 de ritu nuptiarum  contains the relevant texts and case law. The verb nubere applied to women,  while men were understood uxorem ducere. The plural nuptiae referred to  both promises. Nuptiae  implied  a  sexual  relationship  (coniunctio)  and  a  relationship  involving  assets  (consortium).47  It  required  the  consent  of  both  partners:  nuptias enim non concubitus, sed consensus facit.48 Because of the patrimonial consequences, it required the approval of any paterfamilias, but assent or  approval could be replaced by a court decision.49 In the late Empire it was the  parents, not the paterfamilias, who had to approve. Justinian also required  approval from the parents of minors, even if they lacked parental authority.  41 42 43 44 45 46 47 48 49

D 25.2: De actione rerum amotarum. D 25.7.4. D 24.1.3.1. D 25.7.1.4. D 23.2.24. Council of Toledo in Dist. 34 c. 4. D 23.2.1. D 50.17.30 and D 35.1.15. D 23.2.2.

medieval family and marriage law 

113

Consummation was not essential for nuptiae;50 but a wedding normally led  to coniugium, so it was impossible for couples within the prohibited degrees.  Financial considerations led to the creation of other impediments, such as  the prohibition of the wedding of tutor and pupil.51 That a marriage had taken place could be proved in any way, including by  presumptions. A promise to procreate children would suffice.52 In classical  law  written  evidence  of  marriage  could  be  provided  by  tabulae nuptiales.  Leading  a  woman  to  a  man’s  house  gave  a  very  important  proof  that  they  were wed.53 Confarreatio was a special version of nuptiae. Its main impact was on the  familia, which was about book-keeping. After a wedding, newborn children  belonged  to  the  family  of  their  father.  Only  after  nuptiae  could  a  dowry  account  be  opened.  Dowry  was  a  proof  of  nuptiae.  A  woman  who  proved  there had been a wedding was entitled to the actio dotis at the end of the  relationship. Nuptiae also had penal aspects. Adultery could only be committed with  a  woman  who  was  wed  in  this  way,  not  with  one  in  matrimonium.54  Also,  betrothal was an institution connected to nuptiae. People were never obliged  to wed, but to provide compensation for wasted expenses.55 (e) Matrimonium The  existing  confusion  in  the  literature  of  Roman  law  and  legal  history  means that matrimonium is a difficult relationship to understand. It has been  confused with coniugium and nuptiae. The Roman sources, however, clearly  distinguish  them.  Matrimonium  did  not  require  the  consensus  needed  for  nuptiae.56 Adultery was a breach of nuptiae, but not of matrimonium.57 Matrimonium could persist after liquidation of the family.58 A female slave could  live in matrimonium with her dominus; but there could never be nuptiae.59 Much of the confusion is due to the use of the term of contrahere matrimonium that is found in Gaius, the Digest, and the Institutes. Most Romanists  understand this as about a contract; but contrahere in this fragment means  50  D 23.2.7. 51  D 23.2.59. 52  Aulus Gellius, Noctes Atticae 4.3.2 53  D 23.2.5. 54  D 48.5.35. 55  D 23.1.1; D 23.1.2; C 5.17.2. 56  D 23.2.22. 57  D 48.5.35(34). 58  C 5.16.24.2. 59  C 6.4.4.3.

114

the creation of the ius commune

something of the nature of “bringing along”, “involving”. Investigation of the  occurrence of the word matrimonium in classical Roman law reveals its use  in four categories: first, the status of liberty for children; secondly, the restitution of dowry; thirdly, the restitution of gifts and donations at the end of a  relationship; and fourthly, the decreeing of divorce. When dealing with these  matters, the courts had to investigate and decide whether the relationship  between the couple was in fact one of such a high quality. Whether or not matrimonium existed was a matter to be decided by the  assessment of a judge. He checked whether or not there was affectio maritalis, the intention of the couple to stay together for life in a monogamous  relationship, caring for each other and bringing up children. Affectio maritalis was not some kind of poetic moment, but an important legal concept. It  survived separation. Matrimonium was most certainly a civil law concept.60  The judge had also to ascertain whether the spouses had conubium: the civil  privilege of living in matrimonio. Those who were straightforward citizens  always  had.  Peregrines  had  conubium  when  it  had  been  provided  by  the  peace treaty relating to their province. After 212 ce all free inhabitants of  the Empire had conubium. The  first  category  of  matrimonial  cases  concerned  liberty.  Gaius  linked  the  status  of  liberty  to  matrimonium.  Children  born  and  educated  in  a  high-quality relationship received liberty at birth. Further, the children of a  slave living in a high-quality matrimonial relationship with her dominus also  received the status of liberty.61 The three other categories all required the judge to deal with the end of  matrimonium. Three titles of the Digest contain case law about recovery of  dowry  goods.62  When  confronted  with  an  actio rei uxoratae,  later  an  actio dotis, the judge had to establish if, first, there were nuptiae with the establishment of a dowry account, and secondly, if there was no longer matrimonium.  If  matrimonium still  subsisted,  no  restitution  could  follow.  For  instance,  when a paterfamilias was imprisoned, his succession fell open; but if his wife’s  affectio maritalis continued, the heirs did not have to return the dowry. D  24.1,  De donationibus inter virum et uxorem,  deals  with  gifts  and  donations  between  cohabitating  couples.  After  the  ending  of  the  relationship,  gifts  had  to  be  given  back,  if  it  was  shown  that  what  had  ended  was   matrimonium;  this  was  not  the  case,  however,  if there  had  never  been   matrimony. 60 See, e.g., D 24.1.32.13; D 24.1.3.1; D 39.5.31pr; D 48.5.14(13).4; C 5.17.1. 61 C 6.4.4.3. 62 D 23.3 (De jure dotium); D 23.4 (De pactis dotalibus); D. 23.5 (De fundo dotali).

medieval family and marriage law 

115

Finally, the existence of matrimonium was always at issue in cases involving  divorce. When a litigant claimed that divorce meant that the consequences  of a wedding were at an end because of a divorce, the judge had to establish whether or not there had been matrimony. There could be no divorce  without a breach of matrimony. A debate over this type of issue could arise  after the death of one or both spouses. Augustus introduced high inheritance  taxes for people without three children born in wedlock. It is easy to illustrate  this with an example. Consider the fiscus suing a woman with three children  for inheritance taxes. If the fiscus proved the third child was born after the  end of matrimony, the widow would have to pay the maximum rate; if she  proved all three were born in matrimony, she would fall under the provision  for a reduced rate! Of course, most people never had to establish or even know whether or  not they lived in matrimony. Outside a court, it was of little or no importance.  Hence the frustration of the Pandectists: what was this matrimony without  forms and without legal control? As Ludwig Mitteis said: “Eine verwirklichte  Lebensgemeinschaft!”63 (2) Early medieval legal texts about matrimony After  the  division  of  the  Western  Empire  in  476  ce,  this  Roman  case  law  continued in the jurisdictions of the new kingdoms.64 In the sixth century, the  Edict of Theodoric seems to have preserved unaltered the meaning of the  Roman terminology about relationships.65 In general, the Edict of Theodoric  favoured  Christian  interests.  This  becomes  clear,  for  example,  through  reading the juris prudence about the goods of abbeys or bishoprics. But in  dealing  with  relationships,  the  Theodoric’s  courts  strictly  followed  Roman  case law. As yet, the idea of a single family with a common estate was not  accepted. Matrimony did not compel a paterfamilias to share his familia.66  Further north, the same continuity is also found in the case law collected in 

63 This celebrated expression is found quoted in Kaser, RPR2 73.  64  D  Claude,  “Niedergang,  Renaissance  und  Ende  der  Präfekturverwaltung  im  Westen  des  römischen Reichs (5.-8. Jh.)” (1997) 114 ZRG, GA 352. 65  For coniugium, see EdTh 59; for uxor/maritus, see EdTh 24, 37, 54, 59 and 153; for nuptiae, see  EdTh 36-37; for matrimony, see EdTh 20, 36 and 54. The cases mentioning matrimonium belong  to classical Roman case law: EdTh 20 = CTh 9.24.3; EdTh 36 = C 5.5.6 (396); EdTh 54 = CTh  3.16.1-2. On the meaning of matrimonio in EdTh 20, it may be noted that the text contains de hoc matrimonio and this hoc refers to the fact that a matrimony has been proven in the case. See my  remarks about the use of personal pronouns in Latin, in L Waelkens, “La cause de D. 44, 4, 2, 3”  (2007) 75 TvR 199 at 206-207. 66  EdTh 93.

116

the creation of the ius commune

the Lex Romana Burgundionum.67  Later sources, however, show that things have changed. Medieval relationships  were  influenced  by  Christianity.  In  the  West  the  current  text  of  the  Bible  was  the  Vulgate  of  Saint  Jerome.68  A  keen  lawyer,  Jerome  imported  legal  terms  into  his  theology.69  In  the  Vulgate  Bible,  the  terminology  used  for relationships is very close to the terminology of Roman law. Jerome used  the  term  coniugium  for  sexual  relationships  in  general,70  and  nuptiae  for  weddings or kidushin.71 He was sparing with the word matrimonium, which  occurs only four times in the Old Testament and twice in the New.72 Each of  these instances is clearly to be understood as representing the high-quality  relationship found in Roman law. The matrimony of Roman law does not seem initially to have presented  a  problem  for  Christian  doctrine.  As  noted,  St  Augustine  did  not  use  the  word.73 In his theology and his moral writings he dealt mainly with coniugium: he did not concern himself with the legal consequences of relationships. The idea of matrimonium as such is absent from his thinking. The same  spirit is also found in the later penitentiaries of the Christian moralists: again,  no  matrimony.  Historians,  however,  are  apt  to  cause  confusion  by  eagerly  transplanting into the early Middle Ages the later mentioned twelfth-century  identification of coniugium, nuptiae and matrimony.74 Reading the penitentiaries  in  manuscript  provides  considerable  –  often  comical  –  information  about sexual relations; but they never disclose anything about matrimonium.  Furthermore, in Regino of Prüm’s published books about causes before the  67 Uxor/maritus  in  21.2;  24;  25;  nuptiae  in  1.2;  16.1-2;  21.1;  22.3;  22.6;  22.9;  26.1;  27.1-2;  36.4;  37.1-2; 37.4-5; matrimonium in 21; 22.2; 27.2. 68  The  fragments  from  the  Bible  in  the  Collatio legum Romanarum et Mosaicarum  lead  to  the  conclusion that another Latin translation was known in Ancient Rome.  69  J Gaudemet, Le droit romain dans la littérature chrétienne occidentale du IIIe au Ve siècle [= Ius  romanum medii aevi I.3.b] (1978) 112-126. 70  Gen 26.7; 28.1; 30.4; 31.17; 34.4; 39.19; 45.19; 49.31; Lev 20.10; Deut. 7.3; 25.7; Judg 1.13; Ruth  1.12 (vinculo coniugali); 4.10; 1 Kings 4.15; 2 Chron 29.9; Tob 7.13; Esth 1.18; 3.13f; 8.11; Mt  1.20 and 24. 71  Gen 29.22; 29.26; Lev 21.3; 22.12; Judg 14.15; Tob 6.13; 9.2; 9.5-7; 9.18; 12.1; Is 54.1; 62.4; 1  Macc 9.37; 9.41; 10.58; Mt 9.15; 9.22; 22.3-4; 22.8-12; 24.38; 25.10; Mk 2.19; Lk 5.24; 12.36; Jn  2.1-2; 1 Cor 7.8; 7.11; 7.34; Apoc 19.7.  72  Ruth 13; 2 Chron 9.14; Prov 30.23; Eccl 23.22; 1 Cor 7.10; 7.38. 73  See above section D. (1) (a) at n 37. 74  Thus, it is necessary to be cautious in using the two volumes of Matrimonio nella società altomedievale (n  18),  which  contain  extensive  material  about  relationships  in  the  early  Middle  Ages,  because most of the authors analyse everything using modern concepts. If, however, the reader  is sensitive to the presence of this twentieth-century conceptualisation, a lot of information may  be found in the articles of P Toubert, “La théorie du mariage chez les moralistes carolingiens”  (at 233); G Picasso, “Il fondamenti del matrimonio nelle collezioni canoniche” (at 199); and R  Manselli, “Il matrimonio nei penitenziali” (at 287).  

medieval family and marriage law 

117

synods,75 there is much about relation ships, but discussed in terms of uxor et maritus, coniugium, sponsa, copulatio or nubere. Nubere is even used in  the sense of having sexual intercourse.76 But the word matrimonium occurs  only three times: twice in a quotation from the Pauli Sententiae,77 and once  in a quotation from Rabanus Maurus, who also understood matrimonium as  a legal concept when he wrote “de his qui in matrimonio sunt, sed nubere  non possunt”.78 Collections  of  canons  often  mention  sexual  relationships,  but  rarely  matrimonium. This can easily be confirmed by checking Gratian’s Decretum  (1132/1140),  which  drew  on  collections  of  canons  from  the  previous  few  centuries. Gratian included excerpts from case law, decretals and conciliar  decisions about coniugium and nuptiae. The word matrimonium, however,  occurs only ninety-one times, and is used in a distinctive way. It is first found in quotations from ancient Roman sources: the Novels;79  the Digest;80 the Code;81 from Ambrose;82 Jerome;83 the Vulgate; 84 the Roman  councils of Carthage85 and Ancyra;86 Isidore of Seville;87 John Chrisostom;88  Symachus;89  the  popes  of  Roman  Antiquity;90  and  finally  the  lawyer  pope  Gregory the Great.91 The  second  series,  which  consists  of  eight  fragments,  is  early  medieval.  Underlying these seems to be an understanding of the term matrimonium 

75 Das Sendhandbuch des Regino von Prüm,  ed  and  trans  F  W  H  Wasserschleben  and  Wilfried  Hartmann [= Ausgewählte Quellen zur deutschen Geschichte des Mittelalters XLII] (2004).  76  Regino of Prüm, Libri synodales 2.250 in Sendhandbuch (n 75) 368-369. 77  Regino  of  Prüm,  Libri synodales  2.128-129  in  Sendhandbuch  (n  75)  310-311:  matrimonium contrahere quoted twice e lege Romana; cf Pauli Sententiae 2.20.4 in Lex Romana Visigothorum,  ed G Haenel (1849; repr 1962) 368. 78  Regino of Prüm, Libri synodales 2.243 in Sendhandbuch (n 75) 366-367. 79  D. 78 c. 2. 80  C. 3 q. 7 c. 2; C. 30 q. 3 c. 6. 81  C. 29 q. 2 c. 3; C. 36 q. 2 c. 3.  82  C. 27 q. 2 c. 36; C. 28 q. 1 c. 14 §3; C. 32 q. 4 c. 9. 83  D. 31, c. 7; C. 32 q. 1 cc. 10 and 11; C. 32 q. 4 c. 2; C. 32 q. 4 c. 14. 84  From 1 Cor 7.11: C. 32 q. 5 c. 16 and C. 33 q. 1. 85  D. 23 c. 2; C. 33 q. 2 c. 1. 86  D. 27 c. 1: Pope Martin quoted by the council of Ancyra; see Friedberg in CIC I col 98 n 5. 87  D. 23 c. 3; D. 40 c. 7; C. 32 q. 4 c. 9; C. 32 q. 5 c. 15; C. 35 q. 4 c. 1. 88  C. 27 q. 2 c. 45; C. 32, q. 1, c. 4. 89  C. 36 q. 2 c. 2. 90  Fabianus (236-250): C. 32 q. 7 c. 25; C. 35 q. 2-3 c. 3; Siricius (394-398): C. 27 q. 2 c. 50; C. 33 q.  2 c. 12; Innocentius I (402-417): D. 26 c.3; C. 28 q. 1 c. 1; Leo (440-461): C. 27 qu. 2 c. 17; C. 32  qu. 2 c. 12; Pelagius (555-560): D. 28 c. 13; C. 32 q. 1 c. 6. 91  D. 10 c. 1; D. 33, c. 2; C. 20 q. 1 c. 2; C. 27 q. 2 c. 12; C. 27 q. 2 c. 29; C. 28 q. 2 c. 2 (ius matrimonii); C. 32 q. 1 c. 10; C. 33 q. 2 c. 1; C. 33 q. 4 c. 7; C. 35 q. 2-3 c. 20; C. 35 q. 5 c. 2; C. 35 q.  8 c. 1-2.

118

the creation of the ius commune

equivalent to that of Rabanus Maurus. Popes Zachary (741-752),92 Stephen  V  (816-817),93  Eugene  II  (824-827),94  and  Nicholas  I  (858-867)95  pursued  German princes who, with the aim of merging and consolidating  territories,  lived in incest with heiresses. These popes accordingly initiated a series of  lawsuits in which the lawyers went back to the classical Roman law to find  texts for arguments about forbidden relationships.96 These eighth- and ninthcentury quotations in the Decretum are contemporary with the ninth-century  revival of matrimony in legal texts. For instance, the term  matrimonium is  found in the Carolingian comments on CTh 4.6.4 that Haenel published.97  Thus,  seven  commentators  close  to  the  Carolingian  chanceries,  from  the  interpretator  of  the  Theodosian  Code  to  the  author  of  the  Epitome  of  Sankt-Gallen,98 commented on the same ancient Roman casus. Thus, they  used  a  case  dealing  with  the  inheritance  rights  of  naturales liberi  (innate  freemen) in a matrimonium legitimum (a matrimony recognised by a court)  to develop a theory about the lawful relationship which give succession rights  to  the  children.  As  a  result,  children  born  outside  such  a  lawful  relationship became progressively understood to be “natural children”, who did not  have full inherit ance rights. Succession entailed the ending of a family. From  these comments on CTh 4.6.4 can be inferred both the early medieval idea  that  a  family  arises  through  relationships,  and  also  the  growing  conviction  in the legal world  that  the  right  type  of  relationship  to  bring  a  family  into  existence is Roman matrimony. This developing view of the lawyers still did  not penetrate into the world of bishops and confessors. Thus, the word matrimonium never occurs in Rather of Verona’s Libri praeloquiorum, which were  written in 934-936 and dealt with cohabitation, coniugium and sexuality.99  The  third  series  of  texts  in  the  Decretum  that  contain  the  word  matrimonium dates from the twelfth-century. It includes Gratian’s extracts from 

92 C. 30 q. 3 c. 2 and 3. 93 C. 33 q. 2 c. 1. 94 C. 27 q. 2 c. 23: this text may also have been altered by Gratian: see Friedberg in CIC I col 1069  n 253. 95  C. 27 q. 2 c. 26; C. 30 q. 3 c. 1; C. 30 q. 4 c. 1; C. 32 q. 7 c. 25. 96  On this episode, see P Corbet, Autour de Burchard de Worms, L’Eglise allemande et les interdits de parenté (IXe-XIIe siècle) [= Ius Commune, Sonderhefte, Studien zur Europäischer Rechtsgeschichte CXLII] (2001). 97  Lex Romana Visigothorum (n 77) 110-111.     98  On  the  Carolingian  comments  published  by  Haenel,  see  now  the  excellent  study  of  D  Liebs,  Römische Jurisprudenz in Gallien (2. bis 8. Jahrhundert)  [=  Freiburger  Rechtsgeschichtliche  Abhandlungen, Neue Folge XXXVIII] (2002). 99  Ratherii Veronenis, Praeloquiorum Libri VI, ed L D Reid [= Corpus christianorum, Continuatio  medievalis XLVIA] (1984) 46-59.

medieval family and marriage law 

119

contemporary canonical sources.100 Gratian himself used the word nineteen  times.101 In some fragments, he inserted the word into a quotation in which  it had not originally occurred.102 It is obvious that in early medieval sources matrimonium was an exclusively  legal term, only to be found in the context of law suits. From the Carolingian lawsuits on consanguinity and incest onwards, it also became clear that  matrimonium no longer had the exclusive meaning it had had in Rome of a  high-quality relationship, but instead also referred to what the Romans called  coniugium.  Nonetheless,  the  original  meaning  remained  familiar.  Around  1100, the word matrimonium appeared in moral philosophy. It occurs twice  in  the  Sentences  of  Anselm  of  Laon  (d  1117);  for  him  it  had  the  Roman  meaning of a high-quality relationship, and was not to be confused with the  Christian institution of coniugium.103 (3) Twelfth-century case law about matrimony The sudden rise in significance of matrimonium in Gratian’s Decretum was  not  fortuitous.  In  1122,  the  Investiture  Controversy  had  been  concluded  by  the  Concordat  of  Worms.  The  Concordat  established  the  jurisdiction  of  the  Church,  but  did  not  define  its  competence.  One  thing  was  beyond  doubt: as in Roman times, ecclesiastical judges had no authority to enforce  their decrees. Execution  of  their  decisions  was  therefore  submitted  to  the  imperial  judges,  those  lords  who  acquired  sovereign  competence  through  feudal  delegation.  Shortly  after  1122,  Gratian  composed  the  Decretum  in  order to sustain the new legal order. He recalled the old principle of Roman  law that the sentences of episcopal courts had to be executed by the imperial  administration.104 The jurisdictional competence of the new courts was vitally  100 Polycarpus: C. 35 q. 2-3 c. 11; Urbanus II (1088-1099): C. 35 q. 2-3 c. 3; C. 35 q. 6 c. 4; Paschalis  II (1099-1118): C. 30 q. 3 c. 5; C. 30 q. 4 c. 5; C. 35 q. 2-3 c. 22; Callixtus II (1119-1124): D. 27  c. 8; Celestinus II (1134-1144): C. 35 q. 6 c. 2; Palea : C. 27 q. 2 c. 18. 101  C. 27 q. 2 and c. 1-2, c. 28, c. 39; C. 28 q. 1 title; C. 29 q. 1 and c. 1; C. 30 q. 5 c. 8 ; C. 91 q. 1  c. 1; c. 2; c. 3; c. 7; C. 33 q. 2 c. 9; c. 11 and c. 13; C. 34 q. 1-2 c. 8; C. 35 title and q. 1; C. 35 q.  2-3 c. 21. 102  D. 31 c. 14: see Friedberg in CIC I col 115 n 147; C. 29 q. 2 c. 2: see Friedberg in CIC I col  1093 n 15; C. 27 q. 2 c. 23: see Friedberg in CIC I col 1069 n 253. 103  Anselm of Laon, Systematische Sentenzen, ed F P Bliemetzrieder [= Beiträge zur Geschichte  der  Philosophie  des  Mittelalters,  Texte  und  Untersuchungen  XVIII.3]  (1919)  129-151.  He  replaced (at 139) even nuptiae with coniugium: “consensus enim facit coniugium, non coitus”.  He wrote (at 146): “adulterium ante matrimonium cum coniuge commissum”; but stated (at 146  in fine): “copulentur in matrimonio […] attamen non ob hoc coniuges habendi sunt”, because  the promissio coniugii is for him the central issue. 104  In C. 11 q. 1 c. 36. About the audiences, see G Vismara, La giurisdizione civile dei vescovi (secoli I-IX) (1995). Cf A Dumas, “Histoire de la juridiction ecclésiastique”, in DDC VI 326-383.

120

the creation of the ius commune

important. It is logical to assume it was Gratian’s first concern. Indeed the  only  reason  for  Gratian  to  collect  all  this  material  immediately  after  1122  must have been for the sake of the new ecclesiastical courts. In the Decretum Gratian set out the enforceable canon law. The oldest  version of the Decretum was divided in two parts. The first consisted of the  distinctions  about  the  sources  of  canon  law  and  the  status  of  clerics;  the  second dealt with causes relating to all the matters about which clerics were  striving  to  enforce  their  rules,  including,  for  instance,  relationships,  discipline, penitential matters and consecration.105 This  bipartite  structure  reflected  the  jurisdiction  of  the  older  episcopal  courts. The last decision of a Western Emperor about their competence had  been a Novel (35) of Valentine III (452 ce). This Novel had restricted the  competence of the episcopal courts.106 In the early Middle Ages, however,  it had been reinterpreted. Vismara noticed it was invoked in the episcopal  courts.107  A  ninth-century  Interpretatio  of  Novel  35  demonstrates  that  Majoranus  had  amended  the  Novel  in  favour  of  the  bishops.108  Valentine  III  would  have  established  the  jurisdiction  of  the  episcopal  courts:  first,  ratione personae  in  all  cases  with  clergymen  as  litigants,  and  secondly,  ratione materiae in all typical church matters. Study of the position in Italy  and Gaul suggests that these two guiding principles survived in ecclesiastical  practice.109  They  must  have  inspired  Lanfranc  du  Bec  when  he  asked  for  the restoration of independent episcopal courts in England in the 1070s. In  the Decretum, Gratian stated explicitly that all cases involving clergymen as  litigants had to be decided in the new courts, ratione personae.110 The issues  brought together in the causae of the Decretum were also reserved for the  ecclesiastical courts, this time ratione materiae.  The  new  courts,  which  at  the  end  of  the  century  were  organised  into  consistory  courts,  achieved  great  success  in  many  regions.  They  inspired  more confidence than the feudal courts. From the beginning, they accepted  a procedure of appeal and, from the 1140s, they enjoyed the support of the  school of Bologna and the Faculté du Décret in Paris. All over Europe, people  tried to bring their conflicts before the ecclesiastical courts. 105 New research on the composition of the Decretum is summarised by C Larrainzar, “La investigacion actual sobre el Decreto de Graciano” (2004) 90 ZRG, KA 27; but still see the synthesis  of older views in G Mollat, “Corpus iuris canonici, Décret”, in DDC IV 611-627. 106  Codex Theodosianus, II, Leges novellae, ed P M Meyers and Th Mommsen (1990) 142-148. 107  Vismara, Giurisdizione civile (n 104) 173 ff. 108  See Leges novellae (note 106) 148-149; the Novel of Majoranus is itself lost. 109  Vismara, Giurisdizione civile (n 104) 206-207. 110  C. 2 q. 11.

medieval family and marriage law 

121

There  were  different  ways  in  which  lay  litigants  could  bring  their  civil  conflicts before the ecclesiastical judges. They could empower them to act by  bringing in a cleric as party to a juridical act. A classic example was provided  by the reintroduction of the Roman testament in many regions. By leaving a  bequest to a parish priest or a local church in a last will, testators could make  the ecclesiastical judges competent. To make sure the church or the priest  received the bequest, the judges had to apply testamentary law. They reintroduced the Roman testament, because their jurisprudence had to be accepted  by the feudal judges. As the latter were imperial delegates, they could not  refuse  to  apply  Roman  law.  Thus,  the  introduction  of  the  testament  –  not  to be confused with the older donation pro anima – was due to the twelfthcentury ecclesias tical judges in different regions.111 This raised the question that if lay people could successfully invoke the  new courts ratione personae, why should they not also try to do so ratione materiae? Thus, if cohabitation was a church matter, why should the many  disputes that originated in cohabitation not be brought before the ecclesiastical courts? These conflicts included not only disputes about relationships,  but also those about such costs as housing, living and childcare, and about  the status of children, the inheritance rights of spouses and children, parental  authority,  protection  of  women  and  children,  community  property,  dowry  and the like. The resolution of conflicts arising from cohabitation also introduced contract law, delict or penal liability into the new courts. There  accordingly  was  a  considerable  body  of  case  law  in  the  twelfth  century about the competence of ecclesiastical judges in issues of cohabitation.  When  questioned  by  defendants,  the  jurisdiction  of  the  ecclesiastical  court had to be established in a preiudicium. Lawsuits leading to preiudicia  were introduced by actions of status. These actions were explained thoroughly  by the thirteenth-century canonist Guillaume Durand and by the glossators,  but they were already very common in twelfth-century canonical procedure.  When a case was introduced in a canon-law court, the first discussion often  dealt with competency. A preliminary hearing was introduced by an action of  status. The preiudicium, the provisional decision, was called possessio status,  as according to usage in Roman law, the term “possession” typically indicated  a provisional situation. For  the  legists,  preiudicia  did  not  have  the  force  of  res iudicata,  and  111 See, for instance, Acts of Last Will, ed L Waelkens, vols II and III [= Recueils de la Société Jean  Bodin LX-LXI] (1993) II 10-12 (Denmark); II 38 (England, where the executor had to resort  to the ecclesiastical court); II 97 (France); II 156 (Northern Netherlands); II 163 (Southern  Netherlands); III 82 (Zurich); III 144 (Alsace); III 162 (Scotland).

122

the creation of the ius commune

the possessio status was confirmed by the final sentence. On their part, the  canonists were confronted with the problem that every preiudicium about   competence was potentially an attack on the secular judges. They introduced  two  ways  of  strengthening  their  preiudicia.  First,  they  admitted  appeals  against  interlocutory  decisions,  relying  on  arguments  that  sometimes  were  hardly congruent with Roman law.112 This meant that decisions were checked  by a higher ecclesiastical court, or eventually by Rome, before finally being  submitted to the exequatur of the less learned local lord. Secondly, the canonists drew on the imperial Roman law to provide arguments about jurisdiction.  As imperial delegates, the lords derived their jurisdiction from Roman law. If  canonical preiudicia were made according to Roman law, they had in theory  to accept them. Many actions of status dealt with the nature of sexual relationships and  whether  they  qualified  for  jurisdiction.  It  is  therefore  a  necessary  conclusion  that  cohabitation  was  accepted  as  a  criterion  for  jurisdiction.  The  actions of status about cohabitation have recently been studied by Florence  Demoulin-Auzary. Her brilliant and insightful work is based on an impressive  know ledge of legal manuscripts, and results in an excellent and sophisticated  book. 113 Nonetheless, while writing about twelfth-century preiudicia establishing “marriage”, she does not appreciate that she is in fact recounting its  birth. In these actions of status, the canonists used Roman-law arguments  about cohabitation with the single aim of establishing their jurisdiction. No  longer  was  the  issue  that  of  resolving  problems  of  liability,  liberty,  inheritance, donations or payments. To realise their objective and to convince the  secular judges, the twelfth-century ecclesiastical courts discovered an arsenal  of arguments in Justinian’s compilations. When quoting them, the canonists  had no reason to distinguish between coniugium, consortium, matrimonium, nuptiae, uxoratus, and maritagium. Examination of the case law studied by  Demoulin-Auzary  reveals  that  there  was  a  new  criterion  of  the  canonical  jurisdiction: matrimony. Central to these actions of status was proof of, first,  consensual cohabitation, and secondly, tractatus (sexual intercourse). In the  terminology of classical Roman law, consensual cohabitation was nuptiae and  intercourse was coniugium. If both were proved, the result was  matrimony  and  it  came  under  the  authority  of  the  ecclesiastical  courts.  Matrimony  was  therefore  developed  as  a  matter  of  jurisdiction.  The  development  of  112 A Padoa Schioppa, Ricerche sull’appello nel diritto intermedio (1970) II 55-65. 113 F Demoulin-Auzary, Les actions d’état en droit romano-canonique: mariage et filiation (XIIeXVe siècles) [= Bibliothèque de droit privé CCCLXXXVII] (2004). All of the medieval sources  are quoted in footnotes, making half of the book consist of editions of the sources.

medieval family and marriage law 

123

matrimony in the case law may be followed in the Quinque Compilationes Antiquae, where it is found in the fourth book.114 From  the  moment  cohabitation  became  a  matter  for  the  ecclesiastical  courts, that of the status of children inevitably followed. As noted, Christian  cohabitation  led  to  a  family.  Only  the  ecclesiastical  judges  could  therefore  decide whether any children were the legitimi of the Roman law, the right  heirs on intestacy. By recognising and constructing the family, the ecclesiastical judges indicated how estates were to be realised and the property transmitted. They embedded a vision of a particular way of life into a progressively  developing case law about proof of cohabitation, from presumption of concubinage to the elements proving with certainty the possession of matrimonial  status.115 This would inspire the theologians. E. MARRIAGE AND FAMILY: FROM CASUS TO REGULA That cohabitation gave access to the canon-law courts because it was a Christian matter now provided an important and refined subject for theologians.  Twelfth-century  theologians  and  moralists  recycled  the  coniugium  and  matrimonium of the canonists to develop simple maritagium into a complex   institution. Gaudemet was of the view that the first lengthy comment on “marriage” was  written by Hughes of Saint-Victor (d 1141) in his De sacramentis.116 Gaudemet  was,  however,  too  keen  to  identify  the  developed  institution.  Hughes  still  distinguished between coniugium and matrimonium. He thought that the first  was a natural and divine institution, while the latter was the creation of lawyers,  secundum iudicium ecclesie.  He  considered  that  coniugium  was  sometimes  possible  when  matrimonium  was  prohibited.117  The  consensus  that  led  to  coniugium was strengthened by the contract of matrimony.118 Matrimonial law  was unstable;119 but it resolved problems left by the Creator.120 Robert Pullus,  who taught in Oxford (1133-1138) and in Paris (1142-1144), was another who  also still distinguished between the legal term of matrimonium and the old  114 See E Friedberg (ed), Quinque compilationes antiquae, nec non collectio canonum Lipsiensis  (1882;  repr  1956)  44-53,  91-96,  126-130,  144-146,  and  180-181.  Most  of  this  case  law  was  adopted into the Decretals of Gregory IX. 115  Demoulin-Auzary, Actions d’état (n 113) 151-188 (“la vie maritale, de la présomption de concubinage à la possession matrimoniale”), 189-234 (“la possession matrimoniale”). 116  Gaudemet, Mariage en Occident (n 17) 146-147. 117  Hughes de Saint-Victor, De sacramentis XI [= Patrologia Latina CLXXVI] (1854), cols 483-484. 118  Hughes, De sacramentis (n 117) col 487. 119  Hughes, De sacramentis (n 117) col 490. 120  Hughes, De sacramentis (n 117) col 506 in fine and 508 in fine.

124

the creation of the ius commune

idea of coniugium, so that the consensus to a coniugium could be in contradiction to a matrimonium.121 Nevertheless, Pullus mentioned, for the first time,  coitus matrimonii.122  The  twelfth-century  theologians  were  largely  inspired  by St Augustine’s doctrine of conugium and by the way Ambrose and Augustine understood kidushin. Influenced by contemporary ecclesiastical jurisdiction, and drawing on the arguments of the Church Fathers, they developed a  theological understanding of how couples should live. Peter Lombard (bishop of Paris, 1158-1160) adopted in his Sententiae (IV  26-42) the new relationship and the new family developed by the canonists.123  He  was  the  first  moral  philosopher  and  theologian  to  do  so.  He  confused  coniugium  and  matrimonium.124  In  his  work  he  analysed  creation  of  the  bond  of  marriage,  consensus ad idem,  marital  relations,  and  impediments  to marriage. His moral analysis of the relationship fits with the image of the  relationship to be found in the case law of the Roman court under Alexander  III (1159-1181).125 Matrimony (matrimonium) was the regular term for this  new institution. In the famous decision Tanta est vis matrimonii, Alexander  III’s court ruled that the force of matrimony was such that even children born  before the marriage of their parents acquired the status of heirs following the  marriage.126 If justified by Roman law, this was nonetheless a typical decision  of the twelfth century, and far from classical Roman law.127 Intestacy had to  do with the ending of a family and realisation of its assets; so when family  was established by matrimony, it was no longer established by cohabitation. From the 1160s onwards, theologians transformed what had been case law  about relationships into a regulated ideal of a way of life. They transmitted  this to scholars and confessors, who then returned this interpretation to the  canonists.  Huguccio,128  Magister  Honorius  of  the  school  of  Kent,129  or  the  121  122  123  124 

125  126  127  128 

129 

Robertus Pullus, Sententiae [= Patrologia Latina CLXXXVI] (1854), cols 947 and 952. Pullus, Sententiae (n 121) col 948. Petrus Lombardus, Sententiae in IV Libris Distinctae, II, Liber III et IV (1981).  See, e.g., Petrus, Sententiae (n 123) 420 (he refers to matrimony in the works of Saint Augustine;  it is not mentioned in them); 421-422 (quae sint consideranda in coniugio? Sunt et alia plura in matrimonio consideranda …); 422 (quid sit coniugium? Sunt igitur nuptaie vel matrimonium viri mulierisque coniunctio maritalis, inter legitimas personas, individuam vitae consuetudinis retinens); 422 (efficiens autem causa matrimonii est consensus per verba expressus, nec de futuro sed de presenti); 423 (consensus solus facit matrimonium – consensus leads to matrimony and  matrimony is a coniugium, even without copula carnalis). See X 4.1 De sponsalibus et matrimoniis. X 4.4.6. See Waelkens, “Légitimité” (n 28) 1045. For  the  next  few  centuries  it  will  probably  still  be  necesssary  to  read  the  quaestiones  about  matrimony  in  manuscript,  because  the  edition  of  Huguccio,  Summa decretorum,  I,  Distinctiones I-XX, ed O Prerowsky (2006) contains only the first 7% of his Summa, first edited in 1949.  Magister Honorius, Summa de iure canonico tractaturus, I, ed R Weigand et al (2004) also ends  

medieval family and marriage law 

125

Distinctiones Monacenses of the French-Rhenish School130 all comment on  Gratian’s case law against the background of doctrine about matrimony. The  consistory courts seem to have had to apply rules matching with a theological  theory.  A  trend  had  now  been  set.  In  the  thirteenth  century,  major  academic  comments about  a  unique  matrimony–coniugium–nuptiae  were  developed  in  the  works  of  scholars  such  as  William  of  Auxerre,  Alexander  of  Hales,  Albert  Magnus,  Saint  Bonaventura  or  Thomas  Aquinas.  These  definitely  influenced parish priests, preachers and confessors. Their vision of relationships  was  followed  in  daily  life  and  regulated  it.  In  thirteenth-century  law  books,  maritagium  (marriage)  –  the  term  derived  from  Roman  uxor et maritus – was understood as the common word in daily use for the new matrimony. This new understanding is also found in the works of the thirteenthcentury canonists. Hostiensis, in his comments on the Liber Extra’s title De matrimonio,131 applied the thirteenth-century vision of matrimony to the old  case law on relationships. The  radical  novelty  of  this  is  illustrated  by  a  significant  problem  that  the lawyers would never have admitted. If, in an action of status, the court  decided  there  was  no  matrimony,  innocent  children  fell  into  a  legal  black  hole.  They  no  longer  had  a  family:  there  was  no  access  either  to  the  old  Roman case law or to the new family law and rules on intestacy; there was  no access to the consistory courts. To deal with the problem, the canonists  for some cases developed a requirement of provision of aliment of children,  inventing an actio ex lege based on D 25.3.5.1. The problem would finally  lead to the acceptance of divorce in the early modern era, in order to save  the situation of children when there were marital problems. In the thirteenth  century, however, it presented case lawyers with problems arising out of the  regulae developed by theologians. 

provisionally with C. 1 q. 7; but remarks on matrimony may be found passim. See, e.g., at 131  on the duas uniones in matrimonio, animorum et corporum. 130  Distinctiones ‘Si mulier eadem hora’ seu Monacenses, ed R Sorice (2002) 130-145. 131  Henrici de Segusio [Hostiensis], In quartum Decretalium librum Commentaria to X 4 (1581;  repr 1965).

5  The Roman Concept of Ownership and the Medieval Doctrine of Dominium Utile Thomas Rüfner A. BARTOLUS AND HIS DEFINITION OF OWNERSHIP B. THE DEVELOPMENT OF THE DOMINIUM UTILE C. DUO NON POSSUNT HABERE DOMINIUM EIUSDEM REI IN SOLIDUM (1) The glossa ordinaria on D 13.6.5.15 (2) Odofredus and the notion of habere dominium alio iure (3) Different modi of ownership in the canon law (4) Albericus de Rosate’s synthesis (5) Bartolus and the doctor de Aurelianis D. THE MEANING OF BARTOLUS’ DEFINITION OF OWNERSHIP E. FROM REGULA TO CHAOS? A. Bartolus and his definition of ownership Commenting on D 41.2.17.1 – a text from the Digest dealing with the difference between possession and ownership (dominium) – the celebrated jurist Bartolus de Saxoferrato gives the following definition of ownership: “What, then, is ownership? Answer: it is the right of complete disposal over a corporeal thing, as long as it is not prohibited by law.”1 This definition has influenced conceptions of property law for many centuries.2 It is, for example, repeated almost verbatim in the French Code civil, article 544:

  1 Bartolus a Saxoferrato, In primam Digesti Novi partem Commentaria ad D 41.2.17.1 no 4 (1574; electronic edn by A J B Sirks 2004) fol 73va: “Quid ergo est dominium? Responde, est ius de re corporali perfecte disponendi, nisi lege prohibeatur.”   2 On its influence on later authors of the ius commune, see H Coing, Europäisches Privatrecht, 1500-1800 (1985) I 292.

127

128

the creation of the ius commune

Ownership  is  the  right  to  enjoy  and  to  dispose  of  things  in  the  most  absolute  manner provided they are not used in a way prohibited by statutes or regulations.3

The definition of Bartolus is thus apparently not only an adequate expression of the classical Roman law but also a forerunner of the modern, liberal  concept of ownership. The modern continental concept of ownership is that of an absolute right  of use and disposal. There can be only one owner (or group of co-owners).  The possibility of a vertical division of ownership is excluded.4 Article 544 of  the French Code was drafted as a clear expression of the legislative will to end  all feudal institutions.5 In feudal law, the right of ownership is vertically split  between the feudal lord and the tenant: neither has the full right of disposal,  because each has to respect the position of the other. Thus, neither lord nor  tenant has the right to dispose de la manière la plus absolue (“in the most  absolute manner”). It would thus seem that the definition of Bartolus is as  incompatible with feudalism as article 544 of the Code civil was designed to  be. Just like the right to dispose de la manière la plus absolue, the ius perfecte disponendi can be attributed to neither party in a feudal relationship.6 Most scholars agree that in Roman law, too, ownership was an absolute  and indivisible right.7 It is therefore not surprising that Max Kaser gives a  definition of Roman ownership which is almost as close to the words used by  Bartolus as article 544 of the Code civil: Even though the sources contain no definition of ownership, it is possible for this  period  [i.e.  the  classical  period  of  Roman  law]  to  define  ownership  as  absolute  power under private law, which – subject to restrictions imposed by law or party  autonomy – allows every legal or factual disposition over the thing.8    3  “La propriété est le droit de jouir et disposer des choses de la manière la plus absolue, pourvu  qu’on n’en fasse un usage prohibé par les lois ou par les règlements.” English translation by G  Rouhette and A Rouhette-Berton available via Légifrance at http://195.83.177.9/code/liste.phtml ?lang=uk&c=22&r=360#art3537 (accessed on 3 May 2009).   4  The term “vertical division” is used to refer to the division of property rights between a superior  owner and an inferior owner as in the case of the relationship between feudal lord and tenant. By  using this term, the present writer does not wish to express an opinion on the question whether it is  really appropriate to explain the existence of dominium utile and dominium directum in medieval  law as a division of ownership; on this issue, see M Kriechbaum, Actio, ius und dominium in den Rechtslehren des 13. Und 14. Jahrhunderts (1996) 335-339.   5  See J-L Halpérin, Histoire du droit des biens (2008) 194.   6  For the incompatibility of this conception of ownership with feudal institutions, see P Garnsey,  Thinking about Property  (2007)  200;  H  Coing,  “Zur  Eigentumslehre  des  Bartolus”  (1953)  70  ZRG, RA 348 at 354. Cf also the references in D Hecker, Eigentum als Sachherrschaft (1990) 41  n 51.   7  See, however, the critical remarks by R Feenstra, “Duplex Dominium”, in J A Ankum, R Feenstra  and W F Leemans (eds), Symbolae iuridicae et historicae Martino David dedicatae (1968) I 55-71.   8  Kaser,  RPR2 I 400: “Obschon uns die Quellen keine Definition des Eigentums bieten, läßt es sich für diese Stufe bestimmen als die privatrechtliche Vollherrschaft, die innerhalb der von der

the roman concept of ownership 

129

If Kaser’s view is correct, Bartolus coined a definition of ownership which  the Romans, knowing that “each definition in Civil law is dangerous”,9 never  spelled out, but which was an accurate explanation of the Roman concept of  dominium. However,  the  fact  that  Bartolus  gave  a  definition  of  ownership  which  appears  to  be  an  adequate  reflection  of  both  the  classical  Roman  and  the  modern liberal concept of ownership gives rise to an obvious question:10 how  could a medieval lawyer, whose primary goal was to find workable legal rules  for a feudal society, come up with such a definition?11 More specifically, how  could Bartolus give this definition at the outset of a short treatise on property,  which – just a few lines below the passage just quoted – contains a comprehensive exposition of the doctrine of dominium utile? As  is  well  known,  the  doctrine  of  domnium utile (or  duplex dominium) is the device through which medieval lawyers integrated feudalism into the  conceptual  framework  of  Roman  law.  According  to  this  doctrine,  both  the  feudal lord and his vassal are owners of the fief. The lord has the dominium directum or superior ownership, but the vassal has the dominium utile. This  doctrine  introduces  the  concept  of  vertically  split  ownership  into  Roman  law. It is thus in flat contradiction with the principles of Roman and modern  property law that have just been explained. This chapter will explore how Bartolus – and other medieval legal scholars12  – could at the same time come up with a definition of ownership that appears  true  to  Roman  principle  and  also  develop  the  doctrine  of  dominium utile.  The study is undertaken with the hope that it will also help to explain the  relationship between general rules and specific cases in the ius commune.

Rechtsordnung und der Privatautonomie gezogenen Grenzen jede rechtliche oder tatsächliche  Verfügung über die Sache gestattet.”   9  D 50.17.202: “omnis definitio in iure civili periculosa est”. 10  This apparent continuity is noted (and questioned) by D Willoweit, “Dominium und proprietas.  Zur Entwicklung des Eigentumsbegriffs in der mittelalterlichen und neuzeitlichen Rechtswissenschaft” (1974) 94 Historisches Jahrbuch 131 at 131-132. 11  The question was already asked by Fernando Vazquez de Menchaca in the sixteenth century: cf  K Seelmann, Die Lehre des Fernando Vazquez de Menchaca vom Dominium (1978) 57. 12  While the present writer is confident that he has traced the development discussed in this article  with some accuracy, the conclusions presented are not based on a full survey of all sources. Therefore, no claim is made that the medieval jurists quoted were actually the first to hold the views  attributed to them. Some of the doctrines and ideas explained below may be older than appears  from the sources cited for them.

130

the creation of the ius commune

B. the development of the dominium utile It is not necessary for the present purpose to trace the history of the concept  of dominium utile in every detail, but a few basic facts need to be recounted.13  The  concept  of  dominium utile was  not  used  exclusively  to  explain  the  position of the vassal within a feudal relationship. It occurs in connection with  a variety of legal positions which were not ownership in the Roman sense, but  somehow seemed to resemble ownership. Among these positions was that of someone who had completed the longi temporis praescriptio.14  The  Justinianic  texts  left  some  room  for  doubts  whether the position of the possessor of a piece of land, after the period of  ten or twenty years had lapsed, was really that of an owner in the full sense.  Some  jurists,  therefore,  held  that  the  longi temporis  praescriptio  only  led  to  the  acquisition  of  dominium utile for  the  possessor.15  Equally,  the  term  dominium utile was employed with reference to the position of certain longterm users of land such as the superficiary, the emphyteuta, and – according  to  some  –  the  usufructuary.  In  contrast  to  feudal  tenure,  these  types  of  tenancy were known to Roman law and subject to extensive provisions in the  Justinianic sources. Some of these sources, especially those from the postclassical period, did not follow classical principle and contained ambiguous  statements regarding the legal position of the tenant. This led the medieval  lawyers to classify superficiaries, emphyteutae and the like as domini utiles of the land they held.16 However, as Feenstra has shown in a very detailed  study, it is likely that the concept of dominium utile was originally developed in order to explain the position of the vassal in feudal law. The first scholar to  employ the term may have been the glossator Pilius.17 The doctrine became  part of the generally accepted institutions of the ius commune – at the very  latest – when it was incorporated into the Gloss of Accursius.18 It is Bartolus,  13 On the doctrine of dominium utile, see generally E Landsberg, Die Glosse des Accursius und ihre Lehre vom Eigenthum (1883) 92-101; E Meynial, “Notes sur la formation de la théorie du  domaine divisé (domaine directe et domaine utile) du XIIe au XIVe siècle dans les romanistes”,  in Mélanges Fitting (1908; repr 1969) II 409. 14  Cf Kaser, RPR2 II 285-288. 15  On this, see R Feenstra, “Les origines du dominium utile chez les glossateurs (avec un appendice  concernant l’opinion des ultramontani)”, in R Feenstra, J H A Lokin, and N Van Der Wal (eds),  Flores legum H J Scheltema Antecessori Groningano oblati (1971) 49 at 58-62.  16  D Strauch, “Das geteilte Eigentum in Geschichte und Gegenwart”, in G Baumgärtel et al (eds),  Festschrift für Heinz Hübner zum 70. Geburtstag (1984) 273 at 277. 17  Feenstra, “Origines du dominium utile” (n 15) 84. 18  See  Feenstra,  “Origines  du  dominium  utile”  (n  15) 60  and  84  for  references  to  the  relevant  glosses. On the dissenting opinions of Jacques de Revigny and other ultramontani, see Meynial,  “Théorie du domaine divisé” (n 13) 427; Feenstra, “Origines du dominium utile” (n 15) 87-93;  R  Feenstra,  “Dominium  utile  est  chimaera”  (1998)  66  TvR  381  at  389-397.  As  Feenstra  has 

the roman concept of ownership 

131

however, who is credited with being the first jurist to give a systematic and  comprehensive overview of all instances of dominium utile.19 Although the famous definition of property is almost immediately followed  by  the  systematic  explanation  of  the  various  types  of  dominium utile  just  mentioned, Bartolus does not seem to see any contradiction. 20 This does not  mean, however, that the medieval jurists were generally unable to perceive  the possible conflict between the doctrine of dominium utile and the principles of Roman property law. C. duo non possunt habere dominium eiusdem rei in solidum D 13.6.5.15 was one place in the sources where the contradiction was too  obvious  to  be  overlooked.  The  text  deals  with  the  case  of  a  vehicle  that  is  given  on  loan  or  hired  out  to  two  people.  Is  each  of  them  severally  liable  for the return of the car to its owner? Or is the obligation to return the car  divided  between  the  two?  Ulpian  reports  the  opinion  of  Celsus  filius  and  quotes him as having said – obiter – that two people cannot own or possess  the  same  thing  in solidum.  The  expression  in solidum is  probably  used  to  indicate  that  Celsus  does  not  exclude  the  possibility  of  partial  ownership.  There can very well be several co-owners of one thing, but no two persons  can own the entirety of a thing at the same time.21 (1) the glossa ordinaria on d 13.6.5.15 Accursius  seems  to  think  D  13.6.5.15  requires  little  explanation  as  far  as  ownership is concerned: This much is clear that two people cannot be owners of the entirety of a thing as  a whole or some part thereof as is stated here and below de castrensi peculio l. hereditate §. pater [D 49.17.19.3] ...22 shown, the ultramontani did not dispute that emphyteutae and feudal tenants had a position that  amounted to ownership. Rather, they tended to view the position of the emphyteuta or tenant as  the only true dominium and therefore to view the assumption that there were two distinct types  of  dominium (utile and  directum) as  unnecessary.  Thus,  they  anticipated  the  development  of  French law in the eighteenth century: see Coing, Europäisches Privatrecht (n 2) I 293; Feenstra,  “Origines du dominium utile” (n 15) 93. 19  Meynial, “Théorie du domaine divisé” (n 13) 442-443; Feenstra, “Origines du dominium utile” (n 15) 84; for a detailed analysis of this system, see Coing, “Eigentumslehre des Bartolus” (n 6)  355-364. 20  Bartolus, In primam Digesti Novi ad D 41.2.17.1 nos 5-8 (n 1) fol 92va. 21  Cf the translation by P Birks in Th Mommsen, P Krüger, A Watson (eds), The Digest of Justinian (1985) I 403. 22  Accursius, Glossa in Digestum Vetus, gl dominium ad D 13.6.5.15 (1487; repr 1969) fol 220va: 

132

the creation of the ius commune

Accursius confines himself to the remark that the maxim that there cannot be  two owners of the entirety of a thing is clear enough.23 With that, Accursius  moves on to discuss the more problematic question whether possession can  be acquired by two people in solidum. (2) odofredus and the notion of habere dominium alio iure A few years after Accursius, Odofredus24 deems it necessary to address the  problem  of  duplex dominium  in  his  lecture  on  D  13.6.5.15,  although  he  expressly states that the matter does not properly belong there: Now,  Gentlemen,  we  would  not  have  to  go  any  further  here;  rather,  we  would  have to move on to other leges, which follow. However, as some issues are being  touched upon here, which need to be touched upon elsewhere, I will go further  and  discuss  shortly  what  is  written  here  by  Johannes  and  Azo.  In  this  section,  you have the rule that two people cannot have full ownership of the same thing.  The  same  follows  from  the  provisions  below  in  the  title  de castrensi peculio l. hereditatem §. pater [D 49.17.19.3] and below in the title de acquirenda possessione l. possideri §. ex contrario [D 41.2.3.5]. But contrary to this rule it seems  in the case of a thing which belongs to the dowry that the husband is the owner,  because the rei vindicatio is granted to him, which is not given to anyone except  the owner, as can be seen in the text C. de rei vendicatione l. doce [C 3.32.9], and  similarly, the wife is the owner, as is shown by the text in C. de iure dotium l. in rebus [C 5.12.30]. To this, some respond by saying that two people cannot have  ownership of the same thing under the same law, but they can have it under two  different sets of laws. And they give a bad example for this, because they say that  the husband is the owner – under the law of nations and not under the civil law  because he would already have acquired ownership by (adverse) possession. And  this is clear from the fact that he has the vindicatio as provided in l. doce [C 3.32.9].  And to that other l. in rebus [C 5.12.30] the followers of this doctrine respond that  while it is true that the husband is the owner and the wife is the owner, she is not  the owner in reality, but only in expectation, because the dowry has to be returned  to  her  upon  dissolution of the marriage. – But you should use the examples as  follows: Two people can have ownership of the same thing under different sets of  laws, for example one person can have the superior ownership, like one who owns  property directly, and the other can have the inferior ownership, like a superficiary, as seen above si ager vectigalis vel emphyteuticarius petatur l. ii. [D 6.3.2].  But two people cannot have full ownership of the same thing in such a way that  both are full superior or inferior owners.25 “Hoc est planum quod duo non possunt esse domini insolidum nec totius nec partis rei ut hic et  infra de castrensi peculio l. hereditate §. Pater ...” 23  There is no further discussion at D 49.17.19.3 either: see Accursius, Glossa in Digestum novum (1487; repr 1968) fol 265vb. 24  On his life, see Lange, RRM I 323-334. 25  Odofredus, In secundam Digesti Veteris partem Praelectiones ad D 13.6.5.15 (1552; repr 1968)  fol  54vb:  “Or  Signori,  hic  non  esset  procedendum,  sed  esset  procedendum  ad  alias  leges  que  sequuntur, tamen quia hic tanguntur quedam que alibi sunt tangenda, ultra procedam et breviter  tangam vobis que hic scripta sunt per Johannem et Azonem. Vos habetis in §. isto si duobus in 

the roman concept of ownership 

133

Odofredus starts his explanations by citing the glosses on the Digestum vetus written by Azo (and his teacher Johannes Bassianus).26 The contents of these  glosses force Odofredus to spend more time explaining the text than would  otherwise be warranted. After  mentioning  two  other  texts  in  Justinian’s  compilation  that  contain  the  principle  duo non possunt habere dominium eiusdem rei in solidum,27  Odofredus comes to the problem of duplex dominium. He does not, however,  cite  the  case  of  a  vassal  and  his  lord,  who  both  have  dominium under  the  doctrine of dominium utile. Rather, Odofredus refers to the case of dowry,  which seems to contradict the rule given by Celsus: C 3.32.9, a rescriptum of  the emperors Carus Carinus and Numerianus from 283 ce, states clearly that  the husband is the owner of all things belonging to his wife’s dowry. Another  imperial enactment, C 5.12.30, made by Justinian in 529 ce, provides that  the  wife  shall  have  an  actio in rem  to  recover  her  dowry.  In  this  context,  Justinian uses the word vindicare although he does not actually say that the  wife’s remedy is a rei vindicatio. Even so, the two texts can be read as saying  that both the wife and the husband are owners of the things belonging to the  dowry. Next, Odofredus cites the opinion of some unnamed scholars (quidam) who  try to solve the apparent contradiction between the maxim duo non possunt habere dominium eiusdem rei in solidum and the texts on dowry. According to  these quidam, the rule that there can only be one owner in solidum does not  principio quod duo non possunt habere dominium eiusdem rei in solidum: et ad hoc concordat  infra de castrensi peculio l. hereditatem §. pater et infra de acquirenda posessione l. possideri §.  ex contrario. sed videtur contra quod maritus est dominus rei dotalis, quia datur ei rei vendicatio  que non datur nisi domino ut C. de rei vendicatione l. doce, item uxor similiter est domina rei  dotalis ut C. de iure dotium l. in rebus. Ad quod respondent quidam duo non possunt habere  dominium  eiusdem  rei  eodem  iure.  sed  alio  et  alio  iure  sic.  et  subiiciunt  exemplum  malum,  quia dicunt quod maritus est dominus iure gentium et non civili, quia iam usucaperet: quod ex  eo apparet quia maritus habet rei vendicationem ut in l. doce et ad l. illam in rebus respondent  licet  maritus  in  veritate  sit  dominus  et  uxor  sit  domina,  verum  est  non  re  sed  spe  quia  soluto  matrimonio reddenda est ei dos. Sed vos ponatis ita exempla quod duo possunt habere dominium  eiusdem rei et alio iure, puta quod unus potest habere directum dominium sicut ille qui habet  directam proprietatem: alius potest habere utile dominium ut superficiarius ut supra si ager vectigalis vel emphyteuticarius petatur l. ii. Sed duo non possunt habere dominium eiusdem rei in  solidum, ita quod ambo directi domini sint in solidum vel utiliter.” 26  On Azo’s gloss in general, see F C von Savigny, Geschichte des römischen Rechts im Mittelalter  (1850; repr 1961) V 12-14; P Weimar, “Die legistische Literatur und die Methode des Rechtsunterrichts in der Glossatorenzeit” (1969) 2 Ius Commune 43 at 71-72; Lange, RRM I 266-267.  According to H H Jakobs, Glossa magna (2006) 16, Azo’s apparatus magnus on the Digestum vetus was written by Azo in continuation of the work of Johannes Bassianus. See also H H Jakobs,  “Odofredus  und  die  Glossa  ordinaria”,  in  F  Dorn  and  J  Schröder  (eds),  Festschrift für Gerd Kleinheyer (2001) 271 at 343. 27  D 41.2.3.5 actually refers to possessio not dominium.

134

the creation of the ius commune

apply where the concurring owners are owners by virtue of different sets of  laws (alio iure). As becomes apparent later in the text, Odofredus agrees with  this formula, but gives it a meaning very different from the one preferred by  the nameless quidam. For them, only the dominium of the husband consists  in re, while the wife’s ownership only exists in spe. It only becomes real when  the  marriage  is  dissolved  and  the  husband  has  to  give  the  dowry  back  to  the wife. Thus, there is no real contradiction to the maxim duo non possunt habere dominium eiusdem rei in solidum – there is only one owner at a time.  The husband is the only real owner until the time of dissolution, when ownership returns to the wife. According to Odofredus, the quidam stress that the husband’s ownership is  based on the ius gentium and not the ius civile. The purpose of this somewhat  enigmatic remark seems to be the exclusion of another possible solution –  according to C 5.12.30.2, the wife’s remedies are subject to usucapio. If the  husband remains in possession after the dissolution of the marriage, he can  acquire permanent ownership through the lapse of time. One could assume  that C 3.32.9, which treats the husband as owner, refers to the time after the  husband  has  become  owner  through  usucapio.  This  would  of  course  be  a  weak method of explaining away the contradiction. The quidam – with whom  Odofredus seems to agree on this point – deny it by insisting that the husband  is  owner  by  virtue  of  the  ius gentium. According  to  the  Roman  sources,  usucapio is an institution of the ius civile,28 whereas traditio – the standard  mode to transfer property in Justinianic law – belongs to the domain of the ius gentium.29 If the husband is owner by virtue of the ius gentium, his ownership  is based on his receipt of the dowry and not on subsequent adverse possession. The words Sed vos ponatis ita exempla introduce the position of Odofredus  himself.  According  to  him,  the  rule  duo non possunt habere dominium eiusdem rei in solidum applies only within one of the two categories of ownership. No two persons can have the dominium directum, and no two persons  can have the dominium utile. However, the existence of a dominus directus and a dominus utilis, as in the case of the emphyteuta, is permissible. For  Odofredus, dominium utile and dominium directum are forms of ownership  based on different iura. He takes over the formula of the quidam and uses  it to bring the institution of dominium utile in line with the classical Roman  rule contained in D 13.6.5.15. Even  though  questions  of  textual  history  are  not  the  main  focus  of  this  paper, a few remarks on the citation of Johannes and Azo may be in order.  28  J Inst 2.6pr. 29  D 41.1.9.3.

the roman concept of ownership 

135

As  Horst  Heinrich  Jakobs  has  shown,  when  the  apparatus  of  Johannes  and  Azo  is  mentioned  in  the  printed  edition  of  Odofredus’  lecture  on  the  Digestum vetus, the reference is often to a remark which cannot be found in  the apparatus of Azo and Johannes Bassianus, but in the glossa ordinaria of  Accursius.30 However, in this case, at least the printed version of the glossa ordinaria  (quoted  above)  does  not  have  a  discussion  of  duplex dominium,  while at least one manuscript of Azo’s apparatus contains a short text which  seems to allude to the views discussed by Odofredus.31 A manuscript located  in Munich contains the following gloss which refers to the word dominium vel possessionem in D 13.6.5.15: Under the same law, but under different sets of laws, it is possible as in the case of  a husband and wife, as above qui sat. da. co. sciendum [D 2.8.15] and C. de iure do. in rebus [C 5.12.30] – which I deny.32

The gloss qualifies the statement duorum quidem in solidum dominium vel possessionem esse non posse by saying that this is only true eodem iure, while  there can be two possessors or two owners diverso iure. The final clause quod nego makes it clear that the glossator is only reporting the view of someone  else, from which he personally takes his distance. The  allegation  of  D  2.8.15  probably  refers  to  the  third  section  of  the  fragment which states that both the husband and the wife are possessors of  the dos. The short gloss makes the point that possession and ownership of  the dowry are assigned to the husband and the wife at the same time, and  mentions (but in the end does not follow) the explanation that the two own  (and possess) under different sets of laws. This gloss is apparently the starting point for Odofredus’ discussion. He  explains to his students that the view rejected in the apparatus of Azo (and  Johannes) is held by quidam. In the way that it is conceived by those quidam,  Odofredus does not follow it either, but he agrees with it if ius diversum (or  30  H H Jakobs, “Petitorium et possessorium in uno libello intendere” (1998) 115 ZRG, RA 323; H  H Jakobs, “Studien zur Geschichte der glossa ordinaria”, in H H Jakobs et al (eds), Festgabe für Werner Flume (1998) 99 (revised version of “Petitorium et posessorium”); H H Jakobs, “Or signori!  Die accursische Glosse als apparatus joannis et azonis in Odofredus’ Lectura super Digesto veteri”,  (2000) 117 ZRG, RA 311; Jakobs, “Odofredus und die Glossa ordinaria” (n 26) 271-352. All these  articles are now in H H Jakobs, Digesten – Glosse – Savigny. Kleine Schriften zur Wissenschaft vom Römischen Recht, ed W Ernst (2004). See also Jakobs, Glossa magna (n  26) 17. 31  Jakobs, Glossa magna (n 26) 17 stresses that Odofredus may refer to the Accursian gloss when he  cites Johannes and Azo. Instances where Odofredus refers to material that is only present in Azo’s  apparatus, but absent from the glossa ordinaria thus do not contradict Jakobs’ view. 32  Munich BSB MS Clm 14028: “eodem iure, diverso enim iure potest ut in viro et uxore ut supra  qui sat. da. co. sciendum et C. de iure do. in rebus. quod nego”. The present writer has not been  able to see the manuscript himself and wishes to express his gratitude to Horst Heinrich Jakobs  who kindly consulted his collection of microfilms and passed on the text of this gloss.

136

the creation of the ius commune

alium) is taken to refer to the dominium utile, which can co-exist with the  dominium directum of another person. Of course, it would be of interest to  know where the view of the quidam, which is cited by Odofredus and which  seems to be at the basis of the gloss in the apparatus of Azo, originated, and  if  it  is  accurately  reported  by  Odofredus.  To  answer  this  question,  further  research on the manuscript sources is needed. (3) different modi of ownership in the canon law The lectures of Odofredus can be dated to between 1231 and 1263.33 In 1234,  pope Gregory IX published his Decretals (or Liber Extra). Only a few years  later, Bernardus Parmensis de Botone wrote a commentary on the Decretals which came to be regarded as the glossa ordinaria on the Liber Extra.34  Both works, the Decretals and the gloss by Bernardus, seem to have been  important  for  the  question  at  hand.  The  Liber Extra contains  X  2.19.9,  a  text taken from a Decretal of pope Innocent III, written in 1207, which –  in a context not otherwise of interest in the present study – repeats almost  verbally the rule contained in D 13.6.5.15: “since two people cannot at the  same time possess the same thing as a whole in the same way …”35 The gloss  of Bernardus de Botone comments on this phrase as follows: The  text  does  well  to  say  “in  the  same  way”.  See  ff. de acquirenda possessione possideri §. econtrario [D 41.2.3.5]. For one person can have civil possession, the  other natural possession. See ff. de precario. et habet §. eum qui [D 43.26.15.4].  Similarly ff. commodati si ut certo §. ultima [D 13.6.5.15], but in opposite sense ff. uti possidetis l. 3 in principio [D 43.17.3pr], where it is said that one has just possession and the other unjust possession. As in ff. qui satisdare cogantur sciendum §. si fundus [D 2.8.15.3] and ff. de iure dotium quamvis [D 23.3.75] and C. de iure dotium in rebus [C 5.12.30] from which it can be deduced that the husband is the  owner of the things that belong to the dowry, but the wife as well. The words “in  the same way” resolve the contradiction.36 33 According to Lange, RRM I 330, the lectures can be dated between 1236 and 1263. However,  as Jakobs, “Odofredus und die Glossa ordinaria” (n 26) 272 n 5 has shown, Odofredus may have  started his teaching career a few years before 1236. 34  Kriechbaum in Lange, RRM II 208; H Zapp, “Bernardus de Botone”, in LMA I col 1976; K W  Nörr, “Die kanonistische Literatur”, in Coing, Handbuch I 365 at 376-377. 35  “[Q]uum duo insimul eandem rem et eodem modo in solidum possidere non possint …” 36  Bernardus de Botone, gl eodem modo ad X 2.19.9 in Decretales D. Gregorii Papae IX, I-II 19  (1613; electronic edn by A J B Sirks, 2005) col. 691: “Bene dicit, eodem modo ff. de acquirenda  possessione possideri §. econtrario. nam unus potest habere civilem, et alius naturalem possessionem. ff. de precario. et habet §. eum qui sic ff. commodati si ut certo §. ultimus. Sed contra ff.  uti possidetis l. 3 in principio ubi dicitur, quod unus iustam habet possessionem, et alius iniustam:  ut ff. qui satisdare cogantur sciendum §. si fundus et ff. de iure dotium quamvis et C. de iure  dotium in rebus. Ex quibus colligtur, quod maritus est dominus rerum dotalium, et uxor similiter;  hoc verbum, eodem modo, solvit contrarietatem.”

the roman concept of ownership 

137

The  passage  from  the  Liber Extra differs  from  D  13.6.5.15  in  that  it  only  mentions possession and not ownership. Bernard’s gloss however, uses the  text to discuss both possession and ownership. It lists texts from the Digest  which  restate  the  principle  that  possession  (D  41.2.3.5)  or  possession  and  ownership (D 13.6.5.15) cannot belong to more than one person at the same  time and contrasts these statements of principle with other Justinianic texts  that seem to prove that under some circumstances divided possession and  divided ownership can occur. The  final  part  of  the  gloss  is  most  important  for  the  present  purpose.  Bernard uses the example of dowry, which Odofredus calls bad; but it is not  clear  whether  he  follows  the  nameless  quidam quoted  by  Odofredus  who  consider  the  wife’s  ownership  as  consisting  in spe only.  Rather,  Bernard  is  content to say that both the husband and the wife own the things belonging  to the dowry. He concludes that the insertion of eodem modo in the text of  X 2.19.9 resolves the apparent contradiction between the rigid rule duo non possunt habere dominium eiusdem rei in solidum and the instances of divided  ownership and/or possession that can be perceived in other texts. It is obvious that the position of the canon law as explained by Bernardus  de  Botone  is  very  similar  to  the  solution  proposed  by  the  quidam and  by  Odofredus  himself.  According  to  Odofredus,  two  people  can  concurrently  own the same thing as long as each one of them can base his position on an  alium ius. Bernardus requires that ownership be of a different kind. While it is  not clear if Bernardus alludes to domnium directum and dominium utile when  he mentions different modi of ownership, it is obvious that his doctrine is just  as apt to reconcile the maxim duo non possunt habere dominium eiusdem rei in solidum with the reality of divided ownership as the position of Odofredus. The peculiar strength of Bernard’s conception is – of course – that it is  based on the textual authority of X 2.19.9, while Odofredus has no textual  argument in support of his claim that two people can own the same thing  in solidum as long as each one of them owns alio iure. (4) Albericus de rosate’s synthesis The  fact  that  the  sources  of  canon  law  lend  themselves  more  easily  to  an  interpretation which leaves some room for exceptions to the rule duo non possunt habere dominium eiusdem rei in solidum than do the Justinianic texts  may be the reason why Albericus de Rosate (ca 1290-1360)37 bases his treatment of the issue primarily on canon law authorities: 37 On his life, see Kriechbaum in Lange, RRM II 665-669.

138

the creation of the ius commune

After the text has been explained, add another point for which this paragraph is  often cited, namely that full ownership and possession of a thing cannot belong  to several people. The same is stated in extra de probationibus c. licet [X 2.19.9]  and see what is noted there and see Innocent extra de praescriptionibus c. auditis  [X 2.26.15] in the end of the text and the same is also stated below de castrensi peculio l. haereditatis § pater [D 49.17.19.3] de concessione praebendarum c. post electionem [X 3.8.7] and see Roffredus in the Libellus on the rei vindicatio utilis  which is granted to the superficiary. For on the basis of different kinds of ownership, say superior and inferior ownership, there can be several owners of the same  thing as Hostiensis notes there in his Summa under the same title §. quibus detur.  And see other authors of summae and the Speculum.38

Albericus begins by reiterating quod dominium et possessio eiusdem rei non potest esse apud plures in solidum. At first sight it may seem that X 2.19.9 et quod ibi notatur are cited only in support of this basic rule. In reality however,  these citations39 already indicate the important qualification which Albericus  adds to the basic rule: it is not applicable if two owners base their claim on  diversa dominia.  With  this,  Albericus  takes  over  the  doctrine  of  the  gloss  to  X  2.19.9.  Immediately  afterwards  he  identifies  dominium directum and  dominium utile as diversa dominia which can co-exist in spite of the maxim  duo non possunt habere dominium eiusdem rei in solidum.  Albericus  thus  makes the connection which is already present in the lecture of Odofredus  but absent from the gloss of Bernardus de Botone. The importance of this  connection is underlined by the citations at the end of the passage which all  confirm the existence of dominium utile as a second form of ownership.40 38 Albericus de Rosate, In secundam Digesti Veteris partem Commentarii ad D 13.6.5.5 no 16 i.f.  (1585; repr 1977) fol 78rab: “Exposita litera collige unum ad quod saepe allegatur iste §. quod  dominium et possessio eiusdem rei non potest esse apud plures insolidum. Concordat extra de  probationibus c. licet et vide quod ibi notatur et vide Innocentium extra de praescriptionibus c.  auditis circa finem et concordat etiam infra de castrensi peculio l. haereditatis §. pater; de concessione praebendarum c. post electionem et vide Roffredum in libello de utili rei vendicatione quae  datur  superficiario:  nam  ex  diversis  dominiis,  puta  directo  et  utili  possunt  plures  esse  domini  eiusdem rei ut ibi notat Hostiensis in summa eodem titulo §. quibus detur, et alios sumistas, et  Speculum”. Cf also no 19 i.pr fol 78va. 39  The commentary on X 2.26.15 by Sinibaldus Fliscus (pope Innocent IV), which is cited immediately after X 2.19.9, does not contain an explicit statement on the issues discussed by Albericus.  It is probably cited for the remark that several church officials may at the same time exercise the  same rights over a certain group of believers, e.g. the archidiacon, the parson, the bishop and the  archbishop can excommunicate or visit the same subjects (“... sicut apparet in visitatione ecclesiasiae et aliis similibus, quae non prohibentur simul esse apud plures, nam archidiaconus, plebanus,  Episcopus et archiepiscopus eosdem subditos excommunicare et visitare possunt”), Innocent IV,  Apparatus in V libros decretalium, ad X 2.16.15 no 6 (1579; repr 1968) 307. 40  Cf  Roffredus  Beneventanus,  Libelli iuris civilis,  (1500;  repr  1968)  24-26.  The  reference  to  Hostiensis is difficult to understand. If Hostiensis eodem is taken to refer to the title De commodato in the Summa Hostiensis, the citation seems to be erroneous. At least in the printed edition  accessible  to  the  present  writer,  the  title  De commodato in  book  3 only  contains  a  §  Quibus dentur, which does not mention the dominium utile or anything else that might be pertinent to 

the roman concept of ownership 

139

(5) Bartolus and the doctor de aurelianis Bartolus de Saxoferrato does not mention the possible conflict between the  doctrine of dominium utile and the rule duo non possunt habere dominium eiusdem rei in solidum in  his  extensive  commentary  on  D  13.6.5.15.41  As  has  already  been  mentioned,  he  betrays  no  doubts  regarding  the  admissibility of dominium utile under the rule in D 13.6.5.15 in the small treatise on  dominium utile that is contained in the commentary on D 41.2.17.1. However,  Bartolus comes to mention D 13.6.5.15 and the rule contained therein in his  commentary on D 41.2.17.1 at the outset of his discussion of dominium utile: I ask: how many types of ownership are there? A certain Doctor from Orleans says,  as that German Doctor, who yesterday held a repetition here, has reported, that  there is only one type of ownership. But there are two. And I prove this through  the l. possessores in fin. C. de fundis patrimonialibus lib. 11 [C 11.62.12.1] where  it is said that the emphyteuta is the owner (of his land). And yet someone else  remains the owner, viz. the lessor, l. i et ii C. de iure emphyteutico [C 4.66.1-2].  And if there are two owners, the types of ownership must be different, because  the same ownership cannot belong to two persons, as is stated in l. si ut certo § si duobus vehiculum supra commodati [D 13.6.5.15].42

Bartolus  discusses  the  opinion  of  an  unknown  French  jurist,  of  which  he  was informed by an equally anonymous German. Ennio Cortese has recently  shown that the French scholar was probably Jean Nicot.43 He taught law at the  the issues discussed by Albericus. See Henricus de Segusio, Summa (1537; repr 1962) fol 146ra.  However, Hostiensis acknowledges the existence of the utilis rei vindicatio and the dominium utile elsewhere in his work, see book 3, title De locato et conducto, §  Que actio ex hoc contractu nascatur, fol 151va (concerning the emphyteuta) and title De feudis, § Quod ius acquiratur vasallo,  fol 153vb-fol 154ra (concerning the vasallus). The text in the title De feudis is commonly cited for  the final sentence Sed dominus meus dicit quod utile dominium est chimera. Unlike his teacher  Jacobus Balduini, however, Hostiensis himself accepts the notion of dominium utile as is apparent from  the  sentence  immediately  preceding  the  one  just  quoted:  “quidquid  tamen  superius  sit  allegatum satis videtur dici posse quod habeat [sc. vasallus] dominium utile ad exemplum predictorum.” Cf Meynial, “Théorie du domaine divisé” (n 13) 427 who writes that Hostiensis reports  “l’opinion de son maître sans s’y rallier”. The Speculum Iudiciale of Gulielmus Durantis contains  similar statements signalling acceptance of the doctrine of dominium utile; cf Wilhelm Durantis,  Speculum Iudiciale bk 4 pt 2 De causa possessionis et proprietatis § 4 no 10 and bk 4 pt 3 De locato  § 5 no 29 (1574; repr 1975) II 149 (concerning the superficiary) and 270. 41  Bartolus a Saxoferrato, In secundam Digesti Veteris partem Commentaria ad D 13.6.5.15 (1574;  electronic edn by A J B Sirks, 2004) fols 82vb–83vb. 42  Bartolus, In primam Digesti Novi, ad D 42.2.17.1 no 5 (n 1) fol 92va: “Quaero, quotuplex est  dominium?  Quidam  doctor  de  Aurelianis,  ut  retulit  ille  doctor  Theutonicus,  qui  pridie  hic  repetiit, dicit quod unum solum est dominium. Sed duo sunt. et probo per l. possessores in fin. C.  de fundis patrimonialibus lib. 11, ubi dicit, quod emphyteuta est dominus, et tamen alius remanet  dominus scilicet concedens l. i et ii C. de iure emphyteutico. et si sint duo domini, diversa dominia  sunt, quia non potest idem dominium esse apud duos: ut l. si ut certo § si duobus vehiculum supra  commodati.” 43  E Cortese, “Controversie medievali sul dominium utile: Bartolo e il quidam doctor de Aurelianis”,  in  A  Padoa  Schioppa,  G  di  Renzo  Villata  and  G  P  Massetto  (eds),  Amicitiae Pignus. Studi in

140

the creation of the ius commune

university of Orleans for many years before becoming bishop of Orleans in  1357.44 In the university, he was the doctor assigned to the natio germanica.45  Thus,  it  is  very  plausible  that  a  German  jurist  should  have  propagated  his  point of view when visiting the university of Perugia and holding a repetitio  there around 1350.46 The  Frenchman  maintains  that  there  is  only  one  dominium,  and  thus  denies the doctrine of dominium utile. Since a lecture by Nicot has recently  been discovered,47 we know that he continued and exacerbated the critical  attitude of earlier French jurists towards the doctrine of dominium utile.48 In  the  context  of  his  argument  against  the  position  of  the  doctor de Aurelianis,  Bartolus  cites  D  13.6.5.15  and  transforms  the  text  –  which  at  first sight contains a strong argument against the doctrine of dominim utile  –  into  an  authority  supporting  the  same  doctrine.  The  line  of  reasoning  is  simple. Bartolus obviously agrees with the position of Albericus de Rosate,  according to which the maxim duo non possunt habere dominium eiusdem rei in solidum does not apply where different types of property are involved.  Bartolus now takes one of the texts where the Justinianic law itself does not  stick to the classical principle of indivisible ownership: C 11.62.12.1, which  indeed calls the emphyteutae owners of their land (fundorum domini). Now  Bartolus confronts this text with two sources where, in accordance with the  classical principle, the dominus fundi is the lessor who gives out his land to  the emphyteuta. Bartolus concludes that the rule in D 13.6.5.15 can only be  satisfied if it is accepted that the dominium of the lessor is of a different type  than  the  dominium of  the  emphyteuta.  Thus  the  maxim  duo non possunt habere dominium eiusdem rei in solidum becomes an argument in favour of  the recognition of the dominium utile. Bartolus’  statement  demonstrates  that  there  was  no  longer  any  doubt  ricordo di Adriano Cavanna (2003)  I  613  at  29.  On  the  identity  of  these  two  jurists  and  the  question whether – at the time of Bartolus – French jurists opposed the concept of dominium utile, compare also Feenstra, “Origines du dominium utile” (n 15) 87-92 and Kriechbaum, Actio (n 4) 315 n 96. 44  On  Nicot’s  life,  see  M  Duynstee,  “An  Unknown  Fourteenth-Century  Lecture  of  the  Orléans  School: Jean Nicot on Book VI of the Code” (1992) 60 TvR 363 at 365-367. 45  Cortese, “Bartolo e il quidam doctor de Aurelianis” (n 43) 619; Duynstee, “Jean Nicot” (n 44) 366  n 19. 46  For the place and date of the repetitio mentioned by Bartolus, see Cortese, “Bartolo e il quidam doctor de Aurelianis” (n 43) 613. It was customary for doctores from abroad to hold a repetitio at  a university they were visiting; cf K Bezemer, What Jacques Saw (1997) 54. A famous example is  the repetitio held by Pierre de Belleperche at Bologna in 1300 when he was on his way to Rome  for the jubilee; see Savigny, Geschichte (n 26) VI 28; Kriechbaum in Lange, RRM II 548. 47  Duynstee, “Jean Nicot” (n 44) 363-375. 48  Cortese, “Bartolo e il quidam doctor de Aurelianis” (n 43) 628 and (at 633) cf the excerpts from  Nicot’s lecture. See above n 18 for the attitude of the earlier French jurists.

the roman concept of ownership 

141

regarding the compatibility of the doctrine of dominium utile with the rule  duo non possunt habere dominium eiusdem rei in solidum.49 d. the meAning of BArtolus’ definition of ownership So  far,  we  have  seen  how  the  medieval  lawyers  found  a  way  around  the  conclusion that the maxim duo non possunt habere dominium eiusdem rei in solidum,  which  is  so  clearly  stated  in  D  13.6.5.15,  could  not  be  reconciled with the doctrine of dominium utile. The remaining question is whether  this  has  any  consequences  for  our  understanding  of  the  famous  definition  of ownership. If Bartolus shared the conviction of medieval jurists (at least)  since Odofredus that D 13.6.5.15 was no obstacle to the integration of feudal  tenure in the ius commune, how could he have created a definition of ownership which seems to exclude all forms of split ownership and feudal tenure?  There is a strong suspicion that the definition must be read differently in the  light of its originator’s stance on dominium utile. This suspicion is confirmed if one turns to the reasons which Bartolus gives  for the inclusion of the several elements of his definition. Only the justification for the words perfecte disponendi is of interest here, because these words  seem to go against the possibility of split ownership: Why do I say that ownership is the right of complete disposal? Because of the l. in re mandata C. mandati [C 4.35.21] and with respect to the difference between  ownership and possession which is the right to stand on a thing.50

Bartolus thus gives two reasons. First, he cites a text in Justinian’s Code which  states that, whereas the contract of mandate binds the mandatarius to carry  out  the  aliena negotia with  the  utmost  care,  each  man  is  free  to  act  as  he  pleases in most – though not all! – of his own affairs (non omnia negotia, sed pleraque). The citation makes it clear that Bartolus wants to emphasise the  character of ownership as a right existing for the sake of the owner. However,  the text that Bartolus chooses explicitly states that, even in one’s own affairs,  some restrictions exist. This already makes it likely that, for Bartolus, ownership does not really presuppose the possibility to dispose de la manière la plus 49 The same appears from the treatment of the issue by Paulus de Castro. Paulus devotes only a few  lines in his commentary on D 13.6.5.15 to dominium utile: Pauli Castrensis, In secundam Digesti Veteris partem Commentaria ad D 13.6.5.15 no 2 (1582; electronic edn by A J B Sirks, 2004)  fol 76va. 50  Bartolus, In primam Digesti Novi ad D 41.2.17.1 no 4 (n 1) fol 92va: “Quare dico perfecte disponendi? per l. in re mandata C. mandati et quo ad differentiam possessionis, quae est ius insistendi  rei.”

142

the creation of the ius commune

absolue.51  Secondly,  Bartolus  states  that  he  uses  the  words  perfecte disponendi to distinguish ownership from possession.52 Since possession does not  entail any right of disposition, but only – as Bartolus puts it – a ius insistendi,  this, too, shows that Bartolus is far from requiring absolute, unencumbered  power.53 We can conclude that it is at least highly likely that the definition of ownership given by Bartolus was less Roman and less modern than it sounds to us.54  It was, in truth, a definition that was perfectly compatible with the doctrine  of dominium utile and the feudal realities of the time. e. from reGula to ChAos? Hopefully, this paper has shed some light on the ways in which the medieval  lawyers adapted the Roman sources to the needs of their time. Unfortunately,  though, the treatment of dominium utile is hardly an example for the development of clear rules from the vast mass of cases and opinions contained in  the  Justinianic  sources.  The  new  rule  which  emerged  from  the  labours  of  the  jurists  –  the  general  definition  of  ownership  coined  by  Bartolus  –  was  probably  less  meaningful  than  one  might  have  expected.  Moreover,  one  might say that the jurists of the Middle Ages destroyed a clear rule, which  was  contained  in  D  13.6.5.15.  At  the  end  of  the  period  of  the  commentators, the maxim duo non possunt habere dominium eiusdem rei in solidum had become almost meaningless. The exact meaning of the terms dominium directum and dominium utile and the extent to which the Roman rules on  dominium were applicable to either type had to be worked out anew. In that  sense, the path of medieval property law did not lead from casus to regula,  but rather from a rule that proved to be too rigid to many difficult cases.

51 Thus contrary to Coing’s interpretation (“Eigentumslehre des Bartolus” (n 6) 354), the reference  to C 4.35.21 does not imply that the owner of a thing enjoys a sovereignty similar to that which  every person has over their own actions, but rather that the owner is free within the limits of his  or her position – just as every person is free in their actions only within the limits of the human  condition.  See  also  P  Grossi,  Le situazioni reali nell’esperienza giuridica medievale. Corso di storia del diritto (1968) 155. On the continuing use of C 4.35.21 in the context of the concept of  ownership by later authors of the ius commune, see Coing, Europäisches Privatrecht (n 2) 292. 52  Coing, “Eigentumslehre des Bartolus” (n 6) 353. 53  The verb insistere is used in D 41.2.1pr to characterise possession: “Possessio appellata est, ut et Labeo ait, a sedibus quasi positio, quia naturaliter tenetur ab eo qui ei insistit ...” 54  Grossi, Situazioni reali (n 51) 155; Willoweit, “Dominium und proprietas” (n 10) 146; Seelmann,  Lehre des Fernando Vazquez (n 11) 65; Hecker, Eigentum (n 6) 40-41 have all come to similar  conclusions.

6  Succession to Fiefs: A Ius Commune Feudorum? Magnus Ryan A. B. C. D. E. F.

INTRODUCTION THE LIBRI FEUDORUM AND THE GLOSSATORS SUCCESSION IN THE LIBRI FEUDORUM LATER INTERPRETATION OF THE ACCURSIAN GLOSS NATURA FEUDI AND RECTUM FEUDUM CONCLUSION A. INTRODUCTION

Succession was one of the most heavily represented themes in later medieval and early-modern juristic discussion of fiefs. In this, ius commune feudorum merely reflected the wider trends of practical and academic jurisprudence. However, the frequency with which the phrase ius commune feudorum occurs in their surviving works indicates that many of the glossators thought that the law of fiefs constituted a significant category of its own within the broader ius commune, and this was indeed the case. The fief was thought to be sui generis. This is of some significance because the hermeneutic bias of the glossators from the outset and, at Bologna at least, with few deviations, had been to assume that most if not all phenomena could be described and analysed by means of the tools Justinian gave them. It took until the last decade or so of the twelfth century before a lawyer explicitly stated that it was forbidden to quote as authorities texts outside the Corpus Iuris (that lawyer being Azo: non licet juristis allegare nisi Justiniani leges). However, at Bologna at least, the principle was obviously enshrined in the practice of teaching and commentary long before that and was probably the aspect of glossatorial method which did more than anything else to create an autonomous discipline of law during the twelfth century.1 Although glossators such as the mid-thirteenth-century Jacobus de Ardi­zone observed that there was nothing about the vassal in the Roman   1 E Landsberg, Die Quaestiones des Azo (1888) 74.

143

144

the creation of the ius commune

law,2 the schools had in fact found at least one fertile stretch of the Corpus iuris to help them answer questions about vassals: the ius patronatus – that is,  the law governing relations between freedmen and their former owners, now  their patroni. Seemingly the earliest glossator to do this, Johannes Bassianus,  who taught Azo at Bologna in the later twelfth century, had even argued in  a text which shows some marks of having been prompted by a practical case  that “by the custom of the kingdom what we read in the laws about freedmen  is applied by everyone to vassals”.3 These details about the vassal matter because they are in marked contrast  to what happened with the fief; the glossators found no such parallels for the  fief. The fief as known to the glossators was not an analogue or a composite of  analogues drawn from the laws of Justinian; rather, it was the fief of the Libri  or Liber feudorum. This was the short and rather chaotic collection of treatises  composed in Pavia and Milan over at least fifty years of the twelfth century  and probably longer, seemingly in its earliest stratum a product of Lombardist  jurisprudence,  in  its  later  sections  almost  a  manual  by  and  for  communal  judges and advocates in Pavia and Milan.4 An apparatus of civilian glosses was  written on the Libri by Pillius de Medicina, which was left incomplete at his  death in 1208/13 and went on to form more than half of the eventual standard  gloss of Accursius.5 The technically accurate title of the Libri feudorum by the  end of the Middle Ages was the Decima collatio de feudis, the tenth and final  section of the Novels in the vulgate form used at the medieval schools known  as the Authenticum, but for most of the thirteenth and fourteenth centuries  the text could appear just about anywhere in the fifth and “short” volume of  the Corpus iuris (the Volumen parvum) alongside the Authenticum, Institutes  and last three books of the Code, and it went under a variety of titles.   2 Iacobus de Ardizone, Summa super usibus feudorum composita (1518) fol 3vb.   3 Bassianus: “Quod vero de libertis legitur in legibus ex regni consuetudine in vasallis ab omnibus  observatur”, quoted from Bassianus’ Summa de accusationibus by Ardizone in Summa feudorum  (n 2) fol 9ra. The kingdom concerned is Lombardy. I agree with Lange, RRM I 224: the text as we  have it probably originated in a professional consultation or consilium by Bassianus. For Bassianus’  summa, also known as Libellus accusatorius and Libellus accusationis, see E Seckel, “Über neuere  Editionen juristischer Schriften aus dem Mittelalter” (1900) 21 ZRG, RA 212 at 300-306; and S  Kuttner, “Zur neuesten Glossatorenforschung” (1940) 6 Studia et Documenta Historiae et Iuris  275 at 313; repr Kuttner, Studies in the History of Medieval Canon Law (1990) I.    4  K Lehmann, Das langobardische Lehnrecht. Handschriften, Textentwicklung, ältester Text und Vulgattext nebst den capitula extraordinaria (1896).    5  For Pillius’ death, see Lange, RRM I 229 and n 23. For an edition of his apparatus, see A Rota,  L’apparato di Pillio alle “Consuetudines Feudorum” e il Ms. 1004 dell’Archivio di Stato di Roma  [= Studi e Memorie per la Storia dell’Università di Bologna Ser Ia XIV] (1938) 1-170. A reprint  of  the  Glossa ordinaria  to  the  Libri feudorum  is  available  in  M  Montorzi,  Diritto feudale nel basso medioevo. Materiali di lavoro e strumenti critici per l’esegesi della glossa ordinaria ai Libri feudorum (1991).

succession to fiefs: a ius commune feudorum?

145

B. THE LIBRI FEUDORUM AND THE GLOSSATORS Ius commune feudorum begins with this text. This is not to imply that fiefs  and vassals had not been legally stable phenomena across Europe before the  appearance  of  the  Libri  and  their  adoption  by  the  civilians  and  canonists,  merely that the Libri feudorum or Lombard feudal law served as the foundation of sustained glossatorial meditation on fiefs. If one ignores the glosses  to the Libri themselves (which number about 636 by Professor Montorzi’s  reckoning), the massive 96,000 remaining separate annotations comprising  the  standard  gloss  to  the  Roman  law  contain  about  thirty-one  annotations  that so much as mention fiefs, lords, vassals or feudal concessions. In many  cases such observations are restricted to one or two words, and contain no  profound or detailed analysis. Twelfth-century  lawyers  –  particularly  at  Bologna,  whose  traditions  are  partly preserved, and better preserved in this Accursian aspic than those of  other  twelfth-century  law  schools  –  were  not  much  interested;  indeed,  of  these glosses, perhaps a handful reach back further than the generation of  Azo.  Whilst  it  is  true  that  glossators  as  far  back  as  Irnerius  had  made  the  occasional comment about vassals and fiefs,6 and although a trickle of glosses  and  comments  runs  thereafter  through  sources  such  as  the  Dissensiones dominorum  and  the  occasional  free-standing  summa,  no  coherent  body  of  jurisprudence  on  such  matters  took  shape  until  the  end  of  the  twelfth  century; ius commune feudorum was a latecomer. This meant that when the  Libri feudorum arrived at the schools, there was no strong tradition of jurisprudence to condition their exegesis. C. SUCCESSION IN THE LIBRI FEUDORUM A great deal of the Libri feudorum is concerned with succession, the other  main preoccupations of the text being forfeiture and procedure, especially  proof.  The  standard  position  may  be  reduced  to  the  following  brief  statements.  Fiefs,  formerly  called  benefices,  were  initially  not  even  lifetime  grants, let alone heritable. Later, the lord chose which son would inherit but  now it is divided between sons equally. Grandsons in the direct line were then  admitted (by Conrad II), as well as – in default of sons – brothers, so long as  the fief at stake was not in its first generation.7 Daughters may not inherit,    6 See E M Meijers, “Les glossateurs et le droit feudal” (1934) 13 TvR 129 [repr with additions in  Meijers, Études III 260] passim.   7  LF 1.1.1 and 2.

146

the creation of the ius commune

in contrast to successions regulated by the laws, unless specific provision has  been  made  for  them  at  the  original  enfeoffment.8  Lateral  succession  used  to  be  restricted  to  the  fourth  grade  according  to  the  Roman  computation  (fratres patrueles), but now it extends to males ad infinitum.9 Agnates have  a  well-protected  interest  in  any  fief  which  is  already  in  its  second  generation – that is, which has already been inherited once – and can even take a  fief forfeited by a vassal when his own son cannot.10 No alienation of such a  fief  (a  “paternal  fief”  or  feudum paternum)  is  valid  without  the  consent  of  the agnates.11 Sons of a morganatic second marriage by the vassal may not  succeed.12  No  illegitimate  or  adoptive  children  may  succeed.13  Ascendants  may not succeed.14 At the outset of civilian reflection on feudal succession, then, stood the  explicit contention that succession to fiefs was atypical in a number of ways.  The glossators took this as a given. The manuscript containing the Summa Trecensis, the earliest full summa on the Code, implicitly recognises this by  its very organisation, for it contains a short treatise on succession according  to Roman law, then according to Lombard law, and finally an extract from an  early recension of the Libri describing succession to fiefs.15 More explicitly,  and getting on for  a  century  later,  Accursius  noted  that  the  bar  on  female  succession was contrary to Roman law, explaining: Here,  therefore,  [the  text]  establishes  something  different  to  what  is  observed  according to law, nor is this surprising, for the authority of the Roman laws is not  negligible, but they do not extend their force to the point at which they overcome  custom or usage …16

It is striking that Accursius does not object to this, because the passage he  is  paraphrasing  is  of  course  an  inversion  by  Oberto  dall’Orto,  one  of  the    8  LF 1.1.2; 1.8; 2.2.3; 2.11; 2.17.   9  LF 1.1.3. 10  LF 2.26; LF 2.26.18 and LF 2.31. 11  LF 2.39 pr. 12  LF 2.29. 13  LF 2.26.9 and 11. 14  LF 2.50. 15  F Patetta, Il manoscritto 1317 della biblioteca di Troyes, in F Patetta, Studi sulle fonti giuridiche medievali, ed  G Astuti (1967) sec XV 561-577 at 568; see Troyes BM 1317 fol 71rb-va. 16  Glos ord to LF 1.1.3, gl a successione feudi [Montorzi, Diritto feudale (n 5) 16b]: “Hic ergo aliud  statuit  quam  in  legibus  observatur,  nimirum.  Legum  Romanarum  non  vilis  est  auctoritas,  non  tamen usque adeo vim suam extendunt, ut consuetudinem vincant, aut mores ut. j. de cog[nitione]  feu[di] c. i [LF 2.1]. In isto § notare licet deteriores esse conditiones mulierum quam virorum, et  hoc est quod dicitur ff. de sta[tu] ho[minum] l. in multis viris [sic; recte: iuris; D 1.5.9]. Sed quare  hoc contingat quero? Respon[deo] quia dominum iuvare non possunt, nec pugnam facere, quibus  insto, quia filii earum possunt, tamen removentur secundum quod hic dicitur, est ergo pro ratione  voluntas.”

succession to fiefs: a ius commune feudorum?

147

contributors to the Libri, of C 8.52(53).2, where Constantine accords custom  considerable force but subjects it to the twin restrictions of law and reason.  Accursius goes on to say that women cannot fight and therefore cannot assist  their  lord,  which  is  the  reason  given  in  the  Libri  at  several  points,  but  he  concludes on a note of mystification: sons could serve on their behalf, so why  should women not succeed after all? His response is a resigned est ergo pro ratione voluntas. Where the exclusion of women was concerned, the Libri  themselves  provided  plenty  of  escape  clauses.  The  standard  formulation  in the text is that women may not succeed to a fief unless they have been  expressly permitted to do so by the original charter of enfeoffment.17 That  allowed plenty of room for manoeuvre. D. LATER INTERPRETATION OF THE ACCURSIAN GLOSS This  eirenic  attitude  to  the  succession  mechanisms  in  the  Libri  was  fairly  common  but  not  universal  amongst  the  glossators.  We  can  see  this  in  the  fortunes of a gloss by Accursius to Code 3.36.14 (Familiae erciscundae) on  the  implications  of  eviction  after  a  division  between  co-heirs.  Accursius’  question is as follows. Two brothers have divided their paternal inheritance  in  a  judgment  of  familiae erciscundae,  the  one  taking  their  father’s  fiefs,  the other his alods. The former dies leaving only a daughter. Does she have  an action of eviction against her uncle? Accursius answers no: the defunct  co-heir knew when he accepted the fiefs that if he left a daughter she would  be  unable  to  inherit  them.18  Jacques  de  Revigny  at  Orleans  elaborates  by  providing an instructive mini-dialogue.19 The one brother says to the other: “I  am weak, so I shall have the alods and you, who are strong, shall have the fiefs”.  The additional detail that the division has been made in consideration of the  incapacity of one son to serve his fiefs adds nothing of substance to the discussion, and is only mentioned here because it incidentally reveals that Revigny  shared one of the major assumptions of the Italian glossators – itself barely  17 See above at n 8. 18 See Glos ord to C 3.36.14, gl pro parte: “Item, quid si uni assignetur per divisionem iudicis totum  allodium, alii feudum qui decedens reliquit tantum filiam haec filia evicto ei feudo agere possit,  quod videtur quasi non sit ei vitium quod foemina sit nata: ut j. de leg. her. l. lege [C 6.58.14]  et  de  lib.  praeter.  vel  exher.  l.  fi.  [C  6.28.4].  Econtra  quod  non:  sicut  si  res  empta  naturaliter  interiisset vel fato: ut ff. de evic. l. si servus [D 21.2.21pr]. Item quia pater causam praestitit ar.  ff. de evict. l. si servum [D 22.1.25]. Sed puto quod non, cum pater eius filie accipiens in partem  hoc  sciebat  posse  evenire,  nisi  fuisset  stipulatus  de  evictione:  ut  j.  de  evict.  l.  si  fundum  [C.  8.44(45).27].”  19  I follow the spelling suggested by L Waelkens, “La ‘Lectura Institutionum’ de Raoul d’Harcourt”  (1992) 3 Rivista Internazionale di Diritto Commune 79 at 79 n 2.

148

the creation of the ius commune

more than a paraphrase of a passage in the Libri feudorum – to the effect that  physical disability could nullify hereditary title to a fief.20 Revigny adds that so  long as the girl’s father knew he was taking fiefs, and knew the iura feudorum,  then she has no remedy. If he did not, then perhaps his ignorance should not  count against him and his daughter. Now this is interesting. It was a regula iuris that ignorance of the law prejudiced anyone, whereas ignorance of fact  need not.21 The strong implication of Revigny’s final remark is therefore that  the iura feudorum are not so well established that they come under that rule.  It is hard to imagine what iura feudorum might mean if not the Libri, which  Revigny demonstrably knew and by that very name.22 The varied fortunes of Accursius’ gloss to C 3.36.14 in the jurisprudence  inspired by the school of Orleans provide a clear index of the influence of  the Libri feudorum. In the next generation Pierre de Belleperche thought  the very question fatuous since eviction could only be pleaded against the  person from whom the evicted party had title. In his own treatment of the  question he cut out the circumstantial detail to an irreducible core, omitting  Revigny’s description of the circumstances which made Accursius’ question  more  accessible,  details  derived  from  the  dispositions  of  the  feudal  law.23  Cinus of Pistoia was content to repeat Belleperche’s charge of irrelevance,24  although  he  frequently  cited  the  Libri feudorum  in  other  contexts  where  Belleperche and Revigny had ignored them.  The  professor  of  Roman  law  at  Orleans  and  Toulouse,  Guillelmus  de 

20

LF 2.36 [Lehmann, Lehnrecht (n 4) 166]. See, for glossatorial discussion of this passage, the Glos

ord [Montorzi, Diritto feudale (n 5) 71b-72a); Jacobus de Ardizone, Summa feudorum (n 2) fol  28rb; Jacobus de Belvisio, Summa feudorum (below, n 50) fol 100vb.  21  Sext 5.12.13; cf D 22.6.9pr. 22  R Feenstra, “Quaestiones de materia feudorum de Jacques de Révigny” (1972) 84 Studi Senesi  379 at 386 [repr R Feenstra, Fata Iuris Romani. Études d’Histoire du Droit (1974) 298]. 23  See Florence BML Plut 6 Sin 6 fol 158rb, with variants listed from Cambridge Peterhouse 34  [hereafter “Ca”] fol 109vb: “Querit glo. ulterius po[ne] pater decessit relictis duobus heredibus.  habet  feudalia  et  quedam  censualia.  [Ca  omits:  dividit  quod  primogenitus  habeat  feudalia  et  alius  censualia].  primogenitus  decedit  relicta  filia.  illa  non  potest  feudalia  habere.  evincantur.  nunquid illa filia [Ca adds: non potest feudalia habere] regressum habet contra heredes patris  sui. videtur quod sic dicit glo., ar[gumento] huius legis, que videtur dicere expresse; alias esset  naturam vituperare, ut j. de li. pre. l. maximum [C 6.28.4]. videtur contra, quia debet scire istud,  ideo etc. ar. j. de evict. si fundum [C 8.44.27]. Ista questio fatua est. non habet regressum de  evictione nisi propter defectum actoris a quo causam habuit evincatur. ita non est hic et ideo non  habet regressum de evict. ut ff. de evict. l. hoc iure [D 21.2.27].” 24  See Cyni Pistoriensis in Codicem et aliquot titulos primi Pandectarum tomi ... doctissima Commentaria (1578; repr 1964) I fol 179vb: “Ultima gloss. querit de quadam quaestione circa feudum et  allodium, quae videtur Pet. fatua quaestio. Nam de evictione non agitur, nisi propter defectum  autoris, a quo causam habuit possessor, res evincatur, ut ff. de evict. l. hoc iure [D 21.2.27], ergo  etc.”

succession to fiefs: a ius commune feudorum?

149

Cuneo,25 a contemporary of Cinus, approached the gloss in a more charitable  manner.26  His  conclusion  was  the  same  as  that  of  Revigny.  If  the  son  who  accepted  the  fiefs  at  the  time  of  the  original  division  knew  that  they  were  fiefs, then his daughter would have no action against the other son who took  the alods. But the precise details of Guillelmus’ commentary take us much  further into the background of principle and general rule than any previous  gloss. Guillelmus began by explaining the difference between a fief and an alod.  The  holder  of  an  alod  might  be  under  the  jurisdiction  of  somebody  else,  but he owns his alod; he does not hold it in dependent tenure.27 The fief, in  contrast, is held from somebody else such that services are owed for it.28 The  Libri feudorum are used to exemplify both comments. The fief differs from  emphyteusis in so far as it is granted out of affection and not as a purely financial transaction.29 Finally, and most relevantly given the terms of the original  question, Guillelmus explains: And  a  fief  is  of  such  a  nature  that  daughters  cannot  succeed  to  a  regular  fief  [rectum feudum] and the reason is because they cannot serve like a man in a fief  [LF 1.38] and therefore it is contrary to the law that says daughters and sons ought  to succeed equally [C 6.58.14; 6.28.4].30 25 For biographical details on Guillelmus, see E M Meijers, “La première époque d’épanouissement  de l’enseignement de droit à l’université de Toulouse”, in Meijers, Études III 167 at 186; and now  Lange, RRM II 568-581. 26 See Guilhelmus de Cungno super Codice (1513) fol 54ra, with variants from the manuscript Basel  Universitätsbibliothek C. I. 6 [hereafter “B”] fol 53va-b. 27  Guillelmus de Cuneo, Super codice (n 26) fol 54ra: “Sed glosa movet postea talem questionem et  pro evidentia est sciendum quod nos habemus allodia et feuda. allodia vocantur illa que quis ut  propria possidet nec tenet ab aliquo. Sed ipse tanquam dominus possidet licet sit sub iurisditione  alterius in feu. de illo [B: allod.] c. uno. alias est tub ti. de prohi. alie. feud. per fre. c. hac edictali.  § ad hec de allodiis [LF 2.53.11].” The last three paragraphs of LF 2.53 were frequently given  their own rubric (De allodiis) in manuscripts of the Libri feudorum. Guillelmus allows for both  possibilities in giving the alternative reference to Hac edictali. See Lehmann, Lehnrecht (n 4)  179, and Montorzi, Diritto feudale (n 5) 92a-93a. The correct reference should be Ad hoc, qui allodium. 28  Guillelmus de Cuneo, Super codice (n 26) fol 54ra: immediately following: “Sed habemus etiam  feudum et dicitur feudum quando aliquis tenet rem aliquam pro qua tenetur prestare servitium  aliquod in feudum [sic; recte feudis,] de feu. cogni. c. i. [LF 2.1].” 29  Guillelmus  de  Cuneo,  Super codice  (n  26)  fol  54ra:  immediately  following:  “...  et  differt  ab  emphiteota quia feudum principaliter fit propter amorem magis quam propter lucrum in feu. de  feu. da. in vicem [B: adds sub] le. com. c. uno. [LF 1.26]. Sed emphiteosis fit principaliter propter  lucrum insti. loc. et conduc. § adeo [J Inst 3.24.3], et j. de iur. emphi. l. fina. [C 4.66.3].” The  distinction was standard. 30  “Et feudum est talis nature quod filie non succedunt in recto feudo et est ratio quia non possunt  servire sicut vir in feudum ff.a de suc. feu. c. uno. § filiab [LF 1.38] et ideoc et est contra ius quod  dicit filiasd et filios equaliter debere succedere, l. j. de le. here. l. lege xii. ta. [C 6.58.14] et j. de li.  p. et exher. l. maximum virum [C 6.28.4].” Notes: aB: omits ff bB: fila vero c B: omits et ideo d Ed:  filia. Guillelmus de Cuneo, Super codice (n 26) fol 54ra: immediately following.

150

the creation of the ius commune

These comments contain nothing out of step with Italian jurisprudence of the  thirteenth century. The fief is an independent phenomenon: it is sui generis  and not to be subsumed by emphyteusis. Some of the peculiar characteristics  of the fief are contrary to the Roman law. In this case, the point of disagreement is the inequality of male and female heirs. It is even more important  that Guillelmus refers in this context to the rectum feudum, a phrase to which  we will return shortly. This would have been impossible for Revigny, for all  that he relied on the feudal law as iura feudorum. By Guillaume de Cuneo’s  time, it seems clear, the matter was therefore closed: ignorance of the Libri  was  not  an  excuse.  With  that  implicit  recognition,  natura feudi  gained  an  enhanced traction and authority in legal argument. A  second  oddity  of  feudal  succession  provoked  a  disagreement  of  a  far  higher  order  and  concerned  what  exactly  the  word  “heir”  meant;  it  gives  a  precise  indication  of  the  differences  in  emphasis  between  jurisprudence  governed by the Libri and so natura feudi on the one hand, and jurisprudence  limited to the Corpus iuris on the other. In the course of his commentary on  C 6.3.6, Cinus relates the following question by Revigny and Belleperche: It once happened that the king of the Franks granted to a certain knight a castle  in fief in the following way: I give to you and to your heirs. The said knight died  without sons, whereupon the king sought the return of the fief, because in this case  heirs ought to be understood as sons.31

Cinus goes on to explain that Revigny and Belleperche disagreed with the  verdict, on the grounds that the phrase et haeredibus should not be taken to  refer to heirs of the body, but in the wider sense of any heir, so as to include  extranei. The immediate obstacle to such an argument is D 36.1.18(17).8 in  the title on the Senatus consultum Trebellianum, which in the Vulgate reads: If, moreover, an individual left a fideicommissum in this form – “My son, I grant  to you on your faith, that, if you were to die without an heir, you would restore  the inheritance to Seius” – the deified Pius replied in a rescript that he is to be  regarded as thinking of children.32 31 Cinus, In Codicem ad C 6.3.6 (n 24) fol 348ra (De operis libertorum l. Liberti): “De facto acciderat,  quod rex Francorum cuidam militi dedit castrum in feudum hoc modo: Do tibi et haeredibus tuis.  Miles iste decessit sine filiis, unde Rex petebat feudum, quia in casu huiusmodi haeredes debent  intelligi filii.” 32  See Accursii Glossa in Digestum Infortiatum in CGIC VIII fol 202vb: “Si quis autem ita fideicommissum reliquerit: fidei tue fili committo, ut si sine herede moriaris restituas Seio hereditatem,  videri eum de liberis sensisse Divus Pius rescripsit.” Mommsen’s edition reads “si alieno herede  moriaris”. As will become clear, had the Vulgate contained this reading, the debate about to be  examined would perhaps have taken a different direction. I am grateful to Bernhard Stolte for  discussing these variant readings with me, and for his reassurance that the text of the Basilica  provides no solution to this difficulty.

succession to fiefs: a ius commune feudorum?

151

Accursius observed in his gloss to this passage that si sine herede might refer  to any heir, that is, not simply to heirs of the body but to extranei as well; but  he rejected this interpretation on the grounds that it is extremely uncommon  for there to be no heir, even if the heir is merely the fisc.33 He continued: And it is following this argument that, if I grant to you and to your heirs a fief or a  precarious tenancy, it would seem to be understood as a grant to sons, otherwise  it would be perpetual ...34

Now this supports strongly the judgment against the claimants of the fief in  the case related by Cinus. He notes this, and says that Revigny and Belleperche argued against the application of D 36.1.18(17).8 to feudal succession. Their objection, he explained, was rooted in the very point Accursius  made when he argued that si sine herede must refer only to heirs of the body,  otherwise the condition of the fideicommissum would be extremely difficult  to fulfil, and no testator is likely to have imposed such a condition intentionally.35 In other words, there are obvious reasons for the restrictive reading in  the case described in the Digest, but no obvious reason why this should be  extended as a point of principle to feudal succession. Revigny’s discussion of  the matter has not been found;36 but Belleperche’s discussion has been. On  closer  inspection,  it  emerges  that  Cinus  did  not  report  Belleperche  quite  faithfully, for the precise term he used in his summary of the case was liberis rather  than  Cinus’  heredibus.  Beyond  this,  Belleperche’s  treatment  of  the  question corresponds in all important points with what Cinus tells us.37 The  33 Accursius, Glossa (n 32) fol 202vb: “sine herede, alias est sine herede et alias si sine alieno herede,  scilicet existente etc. id est extraneo, sic C. de li. pre. l. iii. [C 6.28.3], qualis est quilibet preter  liberos, sed prima verior et hec est ratio presumptionis, quia raro occurrat quod quis decedet sine  herede cum saltim fiscus sit heres ut C. unde vir et uxor l. i. [C 6.18.1].” 34  Accursius, Glossa (n 32) fol 202vb: “et est secundum hoc ar. quod si concedo feudum vel precarium tibi et tuis heredibus quod de filiis videatur intellectum alias duraret sine fine …” 35  Cinus, In Codicem (n 24) fol 348ra, immediately following the passage cited above, note 163: “Et  ad l. ex facto [D 36.1.18], respondent [scil. Revigny and Belleperche] quod ibi non extenditur  appellatio haeredum, sed restringitur ad filios, ideo quia inutile redderetur fideicommissum, cum  raro  vel  nunquam  conditio  posset  existere,  si  intelligeretur  de  haeredibus  quibuscunque,  cum  impossibile vel raro sit, quem non habere haeredem.” 36  The single manuscript containing his Lectura codicis (Paris BN MS lat 14350) breaks off in book  4, and the commentary in the printed edition (fols 257rb-258va) does not contain any reference  to the affair. The additiones by Revigny to the Infortiatum (where D 36.1.18.8 might have elicited  a response) listed by G Dolezalek, Verzeichnis der Handschriften zum römischen Recht bis 1600 (1972) III s.v. “Iacobus de Ravanis”, in Toledo Biblioteca de la Catedral 32-9 contain no identifiable addition at the relevant passage (fol 170va-b). No repetitio by Revigny is known on either  C 6.3.6 or D 36.1.18. See C H Bezemer, Les Répétitions de Jacques de Révigny (1987) 111-156,  183-235. 37  See  Florence,  BML  Plut  6  Sin  6  fol  228vb,  with  variants  listed  from  Ca  (n  23):  “et  per  istam  [legem?] sententia fuit determinata Parisiis [Ca: Parisius]. Rex concessit castrum alicui [Ca: alicui  alicui] et liberis suis. decessit non relictis liberis sed extraneis, non iudicantur [Ca: inducuntur] 

152

the creation of the ius commune

disagreement between Revigny and Belleperche on the one hand and Accursius on the other is a dispute about the autonomy of feudal succession. For in  restricting the application of D 36.1.18(17).8 to the single matter described  there, the two ultramontani were implicitly treating feudal succession as yet  another case where it is all but impossible to die without any heir at all. Cinus then refers to another jurist whose opinion supported that of the  gloss against Revigny and Belleperche. Dinus, he tells us, stressed that Accursius was talking about the feudal contract, which by its very nature does not  pass to extranei.38 Dinus indeed insists on the special nature of feudal succession in his additional gloss to D. 50.16.219:  To the gloss on D 36.1.18(17) ... where stipulation is discussed, but it does not  seem  to  be  understood  as  about  extraneous  heirs,  the  following  answer  can  be  made. It there discusses the feudal contract, which by its very nature has not been  accustomed to pass to an extraneous heir, whence it is said in feudal custom that,  whenever there is discussion of heirs, it is to be understood as concerning male  children and not women or extraneous heirs as in the Libri feudorum [LF 1.13pr].39

The reactions over the next two or three generations to Accursius’ gloss  to D. 36.1.18(17).8 are most instructive. First, and most obviously, the story  heredes ut castrum habere debeant. istud iudicium falsum est. ubi sibi stipuletur et heredibus  suis videtur stipulari omnibus heredibus ut ff. de probat. si pactum [D 22.3.9], ergo si dono tibi  et tuis heredibus non debet astringi iste favor quia illi extranei heredes habebunt. nam clausula  que imponitur ad augmentandum non debet restringi ff. de condi. et de. l. cum tale § ult. [D  35.1.72.8]. nec obstat l. allegata. verisimile est ubi condicionem apponit quod illam impleri velit  et ideo bene intelligitur [Ca: intelligatur] de heredibus propriis [sic; recte Ca: proprii] corporis  alias condicio non posset impleri, ut l. ex facto § ult. [D 36.1.18.8], et ideo iudicium malum fuit.”  Note that Belleperche does not mention Revigny here. 38  Cinus,  In Codicem  (n  24)  348ra:  “Dyn.  vero  videtur  sentire  cum  illa  glo.,  et  sic  secundum  eum bene fuit iudicatum quia ibi loquitur in contractu feudali, qui de sui natura non transit in  haeredem extraneum. Unde in usu feudorum dicitur quod quotiens de haerede tractatur, intelligitur de filiis masculis, et non de foeminis sive haeredibus extraneis, ut de ali. feudi. c. i. versic.  et sic [sic; recte si] clientulus [LF 1.12 pr]. Et ita scribit in d. § ultim. et de verborum signific. l. in  conventionalibus [sic; recte conventionibus, D. 50.16.219].” 39  This comment does not appear in Dynus de Muxello super Infortiato et Digesto novo (1513) in  OIR XVII, nor in BV MS Ottobon lat 1307, which contains Dinus’ additiones on the Infortiatum.  I have taken the quotation from BV MS Urb lat 156 fol 146vb: “ad glo. ex facto ad trebell. ... ubi  loquitur in stipulatione et tamen non videtur intelligi de hered. extran. sic potest responderi. ibi  loquitur de contractu feudi dotali [sic; recte feodotali] qui de sui natura non consuevit transire  ad heredem extraneum, unde in consuetudine feudi dicitur quod quotiens de herede tractatur,  intelligitur de filiis masculis et non feminabus [sic] sive de heredibus extraneis, [ut in § codicill.]  x. coll. de ali. feu. § fi. [LF 1.13pr].” At D 36.1.18(17).8, Dinus writes (at BV MS Urb lat 156 fol  94vb): “Sed op., quod in dubio videtur lex innuere quod appellatione heredum non solum filii sed  extranei heredes intelliguntur, ut infra, de ver. si. in omnibus [not found], et propterea dicas quod  aut sunt talia iura que solum transeunt ad filium, ut est feudum quod non transit ad extraneum,  ut in constitutione feudorum, de suc. feudi c. i. [LF 2.8], et tunc ap[pellatione] hered[um] soli  filii contineantur, alias si ad filium tanquam ad quemcunque successorem, tunc extranei heredes  continentur, ut preal. l. cum si[milibus].”

succession to fiefs: a ius commune feudorum?

153

shows that the Libri feudorum joined the debate late in the day. Juristic reflection had been prompted in the first place by an argumentum a similibus: how  appropriate was Accursius’ extension to feudal inheritance of the word haeres  in a Digest fragment? A discussion rooted entirely in the Roman law then  ensued, only to be concluded by the arrival (late in the thirteenth century) of  a text which confirmed the position of Accursius. Here, then, is a case where  the introduction of the Libri feudorum to the exclusively Justinianic theory  of the ultramontani resulted in a flat contradiction, and in the vindication of  the Libri feudorum. Secondly, the progression from Accursius’ gloss to Cinus’ commentary on  the Code (by way of Pierre de Belleperche, and Dinus’ Additiones) shows  how easy it was for later scholars to make the implicit explicit. The Glossa ordinaria is sparing in its citation of the Libri feudorum, but slightly richer in  examples of argument from Roman law to feudal tenure. It was a comparatively straightforward matter to fill out the numerous passages where Accursius had observed ut in feudo or ut in feudis with precise references to title  and chapter of the Libri feudorum. The only casualties of this development  were a few of the detailed arguments of the ultramontani: only a few, and  only the details, because Revigny’s picture of the fief agreed in most respects  with that of the Libri feudorum. Some jurisprudence, of the Italian glossators as well as the ultramontani,  was  rendered  obsolete  by  the  attachment  of  their  successors  to  the  Libri feudorum. The question of whether the ius accrescendi applied to the case  where a fief is granted to two brothers, one of whom then dies, had exercised  some civilians. Cinus, in his commentary to C 6.10.1, says that some of the  moderni argued that the ius accrescendi applied only to a fief granted by the  emperor.40 Andreas Bonellus de Barulo argued in precisely this manner in  his commentary on the Tres Libri (books 10-12 of the Code).41 Belleperche  denied  that  an  imperial  grant  made  any  difference.42  The  Libri feudorum  40 Cinus, In Codicem (n 24) fol 353rb. 41 See D. Andreae de Barulo ... Commentaria super tribus postremis libris Codicis ad C. 10.14.1  (1601)  in  OIR  XVIII  35:  “Donavit  Imperator  mihi,  et  tibi  feudum,  si  tu  decedis  sine  liberis  portio  tua  mihi  accrescit.”  Further  on,  Andreas  argues:  “Et  lex  ista  est  argumentum  quod  in  contractibus inter vivos sit locus iuri acrescendi, sed contrarium est verum, ut in usufructu, qui  accrescit proprietati etiam postquam alteri fuerit acquisitus, ut ff. de verb. obl. l. Si mihi et Titio  [D 45.1.110], et de usufr. accr. l. 1, 2 et 3.” The passage occurs with only the most minor variants in  Madrid, BN MS 825 fol 104va. For other treatments of this question, see the references provided  by A Romano, Aspetti dell insegnamento giuridico nelle università medievali IV. Le quaestiones disputatae” (1975) 35 (no 24).  42  See Florence, BML Plut 6 Sin 6 fol 230rb at C 6.10.1: “Queritur de feudis. conceditur feudum  duobus  simul  alter  sine  herede  moritur.  numquid  accrescit  alteri.  Dicunt  doc[tores]  aut  est  concessum a principe et tunc accrescit, si socius libera. prin. l. un[ica, C 10.14.1], aut a privato, 

154

the creation of the ius commune

make it more than clear that no such principle applies, and Cinus summed up  the issue with the observation that the question was only fit for disputations,  since the precise matter was covered by the Libri.43 Once again, the most  influential figure appears to have been Dinus: Cinus refers to him at the end  of his comments on C 6.10.1, and it emerges from Dinus’ own additio at D  45.1.110 that virtually the entire argument came from him.44  E. NATURA FEUDI AND RECTUM FEUDUM A fundamental characteristic of the future ius commune feudorum is already  discernible  in  these  stories.  This  is  flexibility.  The  Libri feudorum  make  it  clear  that  supplementary  agreements,  what  a  civilian  would  understand  as  “pacta apposita”, could be made between lord and vassal at the moment of  investiture, which, recorded in the charter of enfeoffment, would be definitive  in  any  future  litigation.  Thirteenth-century  disputations  in  particular  show that the glossators took this very seriously. In the first example, such  an agreement at the time of the iudicium familiae erciscundae would have  protected the daughter. The Libri feudorum do not contain much law anyway,  and much of what they do contain was of no use to the glossators or anyone  after them, concerned as it was with modes of proof which were no longer  current  or  would  soon  cease  to  be.  The  value  of  the  text  was  in  directing  attention not to rules but to questions: was there a charter of enfeoffment;  did  it  contain  any  supplementary  agreements?45  In  their  absence,  the  fief  was deemed to be a normal or rectum feudum, defined in turn by reference  to the second common quantity in both the examples we have just seen: the  et dico quod non quia quesita accrescunt.” Ca (n 23) gives an identical reading, but omits the  obviously correct sine. 43  Cinus, In Codicem  (n 24) fol 353rb: “Disputari posset, tamen non est quaestio, quia casus de  fratri. de no. bene. investi. c. i. col. 9 [sic; LF 2.12.]. Item in iure emphyteutico secundum Dy. ut  ipse scribit in dicta l. si mihi et Titio [D 45.1.110].”  44  This is another example to add to W M Gordon, “Cinus and Pierre de Belleperche”, in A Watson  (ed), Daube Noster. Essays in Legal History for David Daube (1974) 105 at 108, who notes Cinus’  implication that his debt to Belleperche for his commentary on the rubric of C 3.33 amounts to an  isolated comment, whereas three quarters of it is in fact borrowed. Dinus’ gloss to D 45.1.110 is  missing from the 1513 edition of the Additiones. See BV MS Urb lat 156 fol 135vb: “No. hic arg.  quod in contractibus non est locus iuri accrescendi, unde si concessi aliquibus feud., puta duobus  quod uno eorum mortuo debeat ad me redire.’ He concludes: ‘dico quod in feudo non est questio,  ut in capitulo tercio de fratribus novo [sic; recte de novo] beneficio investitis, coll. x [LF 2.12 ...].”  Dinus must mean the passage corresponding to Lehmann, Lehnrecht (n 4) 130; there is no third  section of LF 2.12. 45  M J Ryan, “Ius Commune Feudorum in the Thirteenth Century”, in A Romano (ed), … Colendo iustitiam et iura condendo …  Federico II Legislatore del Regno di Sicilia nell’Europa del Duecento. Per una storia comparata delle codificazioni europee (1997) 51 at 62-63.

succession to fiefs: a ius commune feudorum?

155

vital concept of natura feudi.46 Here the glossators were extremely creative.  The expression natura feudi appears only twice in the Libri feudorum, once  in connection with a vassal’s right to trial before peers, once as that which a  fief loses when it is granted with full freedom of alienation. Yet the glossators  regularly  mentioned  exclusion  of  women  and  the  restrictive  interpretation  of  who  constitutes  an  heir  as  belonging  to  “the  nature  of  a  fief”.  Neither  extension of the term was improper if judged by the main emphases of the  Libri feudorum. LF 2.2 mentions a pactum allowing inheritance by women  as at least “contra feudi consuetudinem”, and the passages in the Libri where  succession is restricted to males in the absence of any compelling reason to  admit females are legion. The edifice of doctrine had a firm foundation in  natura feudi, or the closely related concept of the rectum feudum or “regular  fief”.  Charters  of  enfeoffment  from  northern  Italy  in  the  twelfth  century  frequently employ the phrase rectum feudum47 and there is even a rubric,  possibly put there by the glossators themselves, describing the “Custom of  the regular fief” (De consuetudine recti feudi) in the Libri (LF 2.33). But the  rest  of  the  edifice  could  bend  with  the  wind  because  the  glossators  never  doubted that additional agreements could be made at the moment of enfeoffment even against natura feudi. The skill of the glossators in extracting from the Libri an idea of natura feudi or the standard fief prompts an important question. Is there any progression or story to be told here about feudal succession, and if so, when does it  stop? There can of course be no question of identifying a single jurist as solely  responsible for the propagation of an idea which we have already seen to be  general  or  implicit  in  the  jurisprudence  of  a  variety  of  thirteenth-century  lawyers, but one could make a very strong case for Dinus, who deploys the  notion of natura feudi freely in his work. We have seen that in the debates  between  Accursius  and  the  ultramontane  giants  Revigny  and  Belleperche,  the last word was that of Dinus. In this respect it is significant that Dinus is  also the first recorded jurist to make use of the concept of natura feudi in a  professional consultation.48 The occasion was a succession dispute which had  arisen in Verona, the lineaments of which were similar to those of the question  treated by Accursius and Revigny about one co-heir taking the father’s fiefs  and the other his alods. In Dinus’ consilium, a brother and sister have split  the inheritance, only for the brother to discover that some of what he now  46  Ryan, “Ius Commune Feudorum” (n 45) 55-65. 47  J  Ficker,  Forschungen zur Reichs- und Rechtsgeschichte Italiens  (1868-1874;  repr  1961)  IV  183-184 (no 141: Parma, 1167, here described as “per morem Theutonicum”); 191-192 (no 151:  Mombello, 1178), as two examples among many. 48  Ryan, “Ius Commune Feudorum” (n 45) 58-60.

156

the creation of the ius commune

has consists of fiefs. He asks for the fiefs to be taken out of the reckoning  and for what remains to be split equally between himself and his sister, for  sisters  cannot  inherit  fiefs  anyway.  Dinus’  arguments  here  are  very  similar  to his earlier comments as he adjudicated between Revigny and Accursius,  although they are more extensive. The key point of the consilium is that a  simple  agreement  to  divide  an  inheritance  can  only  relate  to  those  things  which can, by the ius commune, potentially pertain to both parties anyway.49  Such a claim requires a fairly robust notion of what a fief is; this was also the  broader issue behind the discussion of what constituted an heir for feudal  purposes. F. CONCLUSION As we see, then, it took most of the thirteenth century for jurisprudence to  stabilise on this point. At roughly the same time as Dinus wrote, the first full  defence of the validity of the Libri feudorum as a source of ius commune was  written by Iacobus de Belvisio, whose commentary on the Libri feudorum  would become standard-setting. The most significant point Jacobus makes in  this, one of the most widely-used and influential commentaries on the Libri feudorum, is that the customs contained therein should be treated as general  customs and therefore valid in the absence of specific local customs to the  contrary.50 Accordingly, whenever a fief is granted, it is deemed to be granted  according to the nature of a fief described by the Libri.51 A detailed consideration of the detailed and long-running debates between the medieval lawyers  over the validity of the Libri feudorum would be out of place here. Long after  the main outlines of feudal succession had achieved fixity in the works we have  seen and many more like them, jurists continued to challenge the validity of  the text, or parts of it, on the grounds that the Libri contained merely the  local customs of Milan and other Lombard cities – the opinion which Jacobus  de Belvisio explicitly sought to rebutt. According to this school of thought, the  only components of the text with general validity were the imperial constitutions of Lothar III and Frederick Barbarossa.52 On occasion, such arguments  49 Ryan, “Ius Commune Feudorum” (n 45) 59. 50 See Jacobus de Belvisio, Commentarii in authenticum et consuetudinibus feudorum in OIR XII  fols 81-102 at 82va: “... maxime si consuetudines de feudis non reperiuntur ergo istae tanquam  generales debent servari …” 51  Iacobus de Belvisio, Commentarii: (n 50) fol 82va: “... ergo quando feudum datur videtur dari  secundum naturam feudi in isto iure consuetudinario contenti”. 52  M J Ryan, “Zur Tradition des langobardischen Lehnrechts”, in G Dilcher and D Quaglioni (eds),  Gli inizi del diritto pubblico/Die Anfänge des öffentlichen Rechts  (2009)  II  225  at  233.  Note  Dinus’ usage above (n 39): LF 2.8 is not an imperial text, yet is termed by Dinus “constitutio”. 

succession to fiefs: a ius commune feudorum?

157

smack more of convenience than genuine scepticism. Nevertheless, in the  later Middle Ages it was all but impossible to analyse a feudal inheritance  case without reference to natura feudi, even if only as the starting-point for  an argument which would eventually contradict the relevance or applicability  of the concept to a given case.

The similarity, especially when abbreviations were used, between “consuetudo” and “constitutio”  must always be borne in mind, but the technical significance of calling the Libri feudorum “constitutiones” should not be underrated, all the less since many jurists did so.

7  Towards the Medieval Law of Hypothec Paul J du Plessis A. B. C. D.

INTRODUCTION THE ROMAN-LAW POSITION MEDIEVAL LEARNED LAW CONCLUSIONS A. INTRODUCTION

Accounts of the external history of the reception of Roman law and its contribution to the creation of the legal systems of Western Europe usually contain a number of stock narrative elements. Chief among these is the statement that the teaching of Roman law at medieval Italian universities from the twelfth century onwards and the associated scholarly endeavours of medieval jurists contributed significantly to the creation of a pan-European ius commune which, though it did not replace local customary law, in time proved instrumental in the formation of the legal systems of Western Europe.1 Although this statement continues to have broad academic support, it is often difficult to judge its accuracy since the study of medieval learned law has been hampered by the existence of obscure (unprinted) sources and fierce academic disagreements about methodology. In recent years, much progress has been made in this field of study. The publication of important bibliographical surveys such as those of Coing2 and Lange3 as well as the creation and dissemination of authoritative readings of important sources have greatly facilitated scholarly investigation into medieval learned law. Furthermore, individual scholars have made important contributions to modern understanding of the origins of specific legal doctrines in medieval legal scholarship.4 While it cannot be denied that these scholars have made significant advances, it seems fair to state that much of recent scholarship on medieval   1 See, for example, P Stein, Roman Law in European History (1999), ch 3.   2 Coing, Handbuch.   3 Lange, RRM.   4 To name but two: S Kuttner, The History of Ideas and Doctrines in Medieval Canon Law (1980); and R Feenstra, Le droit savant au Moyen Age et sa vulgarisation (1980).

159

160

the creation of the ius commune

learned law continues to focus on external factors (sources, changes to the  text, textual transmission) rather than the law itself. More specifically, there  is little scholarship on the intellectual methods employed by medieval jurists  to transform Roman-law texts into the ius commune of Western Europe. The  aim of this chapter is to make a modest contribution to modern understanding  of the law of hypothec by investigating a single aspect of this legal institution  and its transmission into medieval learned law. This investigation will focus  on the manner in which the medieval jurists engaged with a particular rule  of law and the way in which they developed it. The argument will be divided  into  two  sections.  First,  an  account  of  the  Roman  law  of  hypothec  will  be  provided in order to explain the context of the rule in question. Secondly, the  transmission of this rule into medieval learned law will be examined. B. THE ROMAN-LAW POSITION To understand the Roman law on hypothec, certain introductory observations  about its origin in the Roman law of pledge are required. Roman law knew two  forms of “security” which a debtor could employ to allay the creditor’s fears  about solvency. The form preferred by the Romans until the classical period  appears to have been personal surety whereby the debtor used solvent third  parties to guarantee that he would repay the debt.5 The other form, which  seems to have become dominant during the classical period, is that of real  security where the debtor used his assets (or those belonging to third parties)  to guarantee the repayment of the debt. Real security seems to have originated in the fourth century bce in the concept of fiducia. It was not a contract  as such; rather, it was an informal agreement (pactum) added to one of the  formal  modes  of  conveyance  such  as  mancipatio  or  cessio in iure.6  Under  fiducia, a debtor agreed to relinquish ownership of the asset to the creditor  until such time as the debt had been repaid. Fiducia benefited the creditor,  but had a number of significant drawbacks for the debtor, most notably the  loss of ownership. To counteract these, a different form of real security was  created in Roman law, namely pignus. Unlike fiducia, pignus was a contract  and would eventually be classified according to Justinian’s Institutes as one  of the contractus re which came into existence upon the handing over of the  object.7 Pignus had a number of advantages over fiducia, chief of those being  that the debtor did not relinquish ownership, only possession of the object.    5 B  Nicholas,  An Introduction to Roman Law  (1962;  repr  with  int  and  glossary  by  E  Metzger  (2008)) 151.   6  The standard work remains G Noordraven, Die Fiducia in het Romeins Recht (1988).   7  See Zimmermann, Obligations 220.

towards the medieval law of hypothec

161

Not  much  is  known  about  the  transition  from  the  earliest  form  of  real  security  to  pignus,  a  flexible  agreement  whereby  only  possession  of  the  pledged object was transferred. At best it may be speculated that the legal  requirements of fiducia (loss of ownership and formality) must have made  it less useful in an increasingly monetised Roman economy characteristic of  the third century bce.8 While pignus resolved many of the practical problems  surrounding fiducia, the debtor, who retained ownership of the object, still  had to relinquish  possession  of  it  to  the  creditor.  This  limited  the  debtor’s  use of his property and also meant that a potentially valuable object could  only really be pledged once, even for a small amount of debt. Thus, during  the  course  of  the  late  Republic,  a  variation  on  pledge  was  created  which  circumvented the problem of loss of possession. The pledge without possession (known in Justinianic law as hypotheca) represented the final stage in the  development of real security in Roman law.9 The third-century jurist Ulpian  defines a hypothec as follows: Strictly speaking we use the term pignus for the pledge which is handed over to  the creditor and hypothec for the case in which he does not get even possession.10

According to classical Roman law, the creditor obtained a preferential real  right over the object of the hypothec even though he did not have effective  control of it as in the case of pledges with possession.11 If the debtor defaulted  in his repayment of the debt, Roman law developed legal remedies to enable  the creditor to claim the object of the hypothec (the actio pigneraticia), thus  satisfying the requirement of possession, and to sell it to recover the debt in  whole or in part.  The transition from a pledge with possession to one without possession is    8 See J A C Thomas, Textbook of Roman Law (1976) 330-331.   9 While it was originally thought that the use of a Greek term suggested that hypotheca had its  origins in Greek law, this view has since been refuted; see M Kaser and R Knütel, Römisches Privatrecht § 31.15, 17th edn (2003) 189. Pace W Kossak, “The Reception of Roman Law Regulations on Mortgage in the Legislation of Ukraine”, in M Jonca and A Debinski, Roman Law and European Legal Culture (2008) 67 whose views on the Roman law of hypothec are clearly incorrect. 10  D 13.7.9.2: “Proprie pignus dicimus, quod ad creditorem transit, hypothecam, cum non transit  nec possessio ad creditorem.” Thomas, Roman Law (n 8) 332 postulates that the terms were not  really distinguished until the time of Justinian, which goes some way to explaining the terminological confusion in the texts. English translations of Digest passages are taken from the Watson  translation as adapted. 11  Marcian’s statement in D 20.1.5.1 – “inter pignus autem et hypothecam tantum nominis sonus  differ” – does not necessarily stand in opposition to Ulpian’s definition in D 13.7.9.2, nor does it  necessarily indicate that the two concepts were not clearly distinguished. When the context of  D 20.1.5.1 (especially the introductory paragraph and the subsequent paragraph) is examined,  it becomes clear that Marcian’s statement merely means that pledge and hypothec apply to the  same type of obligations. 

162

the creation of the ius commune

traditionally linked to the rise of agricultural tenancy.12 Tenancy was a particular manifestation of the contract of letting and hiring (locatio conductio),  classified in the Roman scheme of contracts as one which came into existence  through consensus on the essentials of the contract. Like its historical predecessor  pledge,  hypothec  operated  on  the  principle  of  subsidiarity.  There  could be no agreement of pledge/hypothec without the existence of a debt  created by another causa, mostly ex contractu (in this case, the contract of  letting  and  hiring).  A  contract  of  pledge  (and  thus  also  one  of  hypothec)  could be created with little formality. D 20.1.4 shows that only consent was  required by law to create the pledge/hypothec irrespective of the wording  used or whether it was recorded in writing.13 This made it especially useful  when used in conjunction with the agreement creating the debt, such as a  contract  of  tenancy.14  In  most  cases,  the  “clause”  regulating  the  hypothec  would have formed part of the contract. Additionally, a provision could be  included stating that the creditor would have the legal right to sell the goods  once he had obtained possession.15 The  contract  of  pledge/hypothec  was  particularly  useful  in  the  various  manifestations  of  the  contract  of  letting  and  hiring.  In  most  cases,  it  was  used to ensure payment of the rent either by handing over a specific movable  object as security (the slave Eros in D 20.4.9pr) or by agreeing that certain  objects  belonging  to  the  tenant  would  be  deemed  hypothecated  for  the  payment of rent even though they were not physically handed over to the  creditor. Initially, this seems to have been done by express agreement in the  contract of tenancy, but by the end of the classical period a certain category of  objects (collectively called invecta et illata) were deemed to be pledged ipso iure even if the parties did not expressly agree as much in the contract. At this  point it is important to note that the legal rule under discussion, namely that  the landlord had a hypothec over the invecta et illata of the tenant as security  for the payment of rent, operated both in urban and in agricultural tenancy.  Both were merely variants of the Roman contract of letting and hiring, but  the nature of the circumstances which these contracts catered for led to the  development of variants of the same rule to address specific situations. It is  12 See Thomas, Roman Law (n 8) 332. Compare D 20.6.14, D 47.2.62.8 and Gai Inst 4.147. 13 D 20.1.4: “contrahitur hypotheca per pactum conuentum, cum quis paciscatur, ut res eius propter  aliquam obligationem sint hypothecae nomine obligatae: nec ad rem pertinet, quibus fit verbis,  sicuti  est  et  in  his  obligationibus  quae  consensu  contrahuntur  ...”  (“a  mortgage  [read  pledge/ hypothec]  is  made  by  consent,  when  someone  agrees  that  his  property  will  be  bound  by  way  of mortgage for some obligation. As in consensual contracts, it does not matter what words are  used”). 14  See P J du Plessis, “The Interdictum de Migrando Revisited” (2007) 54 RIDA 1. 15  The legal position on this point varied over time: see Thomas, Roman Law (n 8) 331.

towards the medieval law of hypothec

163

conventionally assumed that agricultural tenancy is the older of the two types  of contract. While the consensual contract of letting and hiring is assumed to  have been created in the second half of the second century bce if not before,  the term colonus (tenant of agricultural land) does not appear in legal sources  before the first century bce.16  In agricultural tenancy, the landlord acquired a hypothec over at least two  categories of object. The texts are as follows: I let you some land and (as is customary) it was agreed that its produce would be a  pledge to me in respect of the rent. If you were stealthily to remove the produce,  [Julian] used to say that I can have the action for theft against you. Equally, if you  sold the standing crops to someone who took it away, we say that it becomes a  stolen thing. Crops, while in the soil, are part of the land so that a tenant makes  them his own, because he is regarded as having gathered them with the owner’s  consent.  One  certainly  cannot  say  the  same  in  the  case  just  put;  for  by  what  reasoning can they be held to become the tenant’s property when the purchase  reaps them from himself?17 We accept that property brought onto an urban leasehold is hypothecated, as if  it had been impliedly agreed. The opposite is true of rural tenancies.18

As the first text shows, the jurist Africanus (second century ce), quoting his  teacher Julian, states that the landlord had a hypothec over the crops gathered  by the tenant with the owner’s consent. While the text mentions an agreement (compare D 20.2.7pr) between the parties to this effect, it is generally  assumed by modern scholars that the hypothec over the fruits of the rented  property was an implied contractual term which operated ipso iure and to  which the parties did not need to agree explicitly.19 The same rule clearly did  not apply in respect of other goods brought onto the rented property by the  agricultural tenant, as the second text, from Quintus Neratius Priscus, a jurist  16 P W De Neeve, Colonus: Private Farm-Tenancy in Roman Italy during the Republic and Early Principate (1984) 39. On the dating of the contract of letting and hiring, see Th Mommsen, “Die  römische Anfängen von Kauf und Miete” (1885) 6 ZSS, RA 260; M Kaser, Das altrömische Ius  (1949) 297-382; and comprehensively H Kaufmann, Die altrömische Miete §§ 1-5, 6-27 (1964)  1-31. See also A Watson, The Contract of Mandate in Roman Law (1961) 9-10; A Watson, The Law of Obligations in the Later Roman Republic (1965) 100-101 where the author postulates that  locatio conductio operis was probably the earliest form of letting and hiring in Roman private law  to be distinguished, followed by locatio conductio operarum and locatio conductio rei. 17  D 47.2.61.8: “Locavi tibi fundum, et (ut adsolet) convenit, uti fructus ob mercedem pignori mihi  essent. Si eos clam deportavis, furti tecum agere posse aiebat. Sed et si tu alii fructus pendentes  vendideris  et  emptor  eos  deportaverit,  consequens  erit,  ut  in  furtivam  causam  eos  incidere  dicamus. Etenim fructus, quamdiu solo cohaereant, fundi esse et ideo colonum, quia voluntate  domini eos percipere videatur, suos fructus facere. Quod certe in proposito non aeque dicitur:  qua enim ratione coloni fieri possint, cum emptor eos suo nomine cogat?” 18  D 20.2.4pr: “Eo iure utimur, ut quae in praedia urbana inducta illata sunt pignori esse credantur,  quasi id tacite convenerit: in rusticis praediis contra observatur.” 19  De Neeve, Colonus (n 16) 9 n 34. Compare D 20.2.7pr.

164

the creation of the ius commune

of the same period, indicates. In this scenario, the implication of the latter  part of the text is that parties to the contract of agricultural tenancy had to  agree explicitly that property other than fruits was hypothecated as security  for the rent. These two texts also demonstrate one of the modes of classification of pledges (and hypothecs) used in classical Roman law, namely that  some only came into existence by express agreement, whereas others were  implied ex iure. The same classification was also used in urban tenancy, a variant of the  contract  of  letting  and  hiring  that  developed  in  the  first  century  bce.  An  example of an express hypothec from urban tenancy is where the parties to  the contract listed specific objects to be hypothecated for the payment of the  debt. In D  20.4.9pr  the  jurist  Africanus  recounts  a  case  of  a  specific  slave  called Eros being hypothecated to ensure the payment of rent for baths.20  Similarly,  in  D  20.3.4,  Paul  discusses  a  case  where  a  debtor  listed  specific  assets to be hypothecated to  ensure  the  repayment of  a  loan.21  The  listing  of specific objects had a number of advantages. First, given that the parties  agreed on the specific assets to be hypothecated, it provided a level of assurance for the creditor that the value of the hypothecated goods could sustain  the amount of the debt. Secondly, it also removed any uncertainty as to the  identity of  the  objects  as  they  could  be  easily  read  from  the  list.  The  only  major drawback of listing specific goods to be hypothecated relates to depreciation in value. If the value of a slave declined over time, it could expose  the creditor to added financial risk. This may be one of the reasons why a  second form of hypothec came into existence. Unlike the first, the second  form  of  hypothec  consisted  of  a  general  contractual  clause  hypothecating  some (not all) of the tenant’s remaining assets in the rented dwelling.22 This  20 D 20.4.9pr: “Qui balneum ex calendis proximis conduxerat, pactus erat, ut homo Eros pignori  locatori  esset,  donec  mercedes  solverentur:  idem  ante  calendas  Iulias  eundem  Erotem  alii  ob  pecuniam  creditam  pignori  dedit.  consultus,  an  adversus  hunc  creditorem  petentem  Erotem  locatorem praetor tueri deberet, respondit debere: licet enim eo tempore homo pignori datus  esset, quo nondum quicquam pro conductione deberetur, quoniam tamen iam tunc in ea causa  Eros esse coepisset, ut invito locatore ius pignoris in eo solui non posset, potiorem eius causam  habendam.”  (“A  man  who  rented  baths  from  the  first  of  the  following  month  agreed  that  a  slave Eros should be mortgaged to the lessor until the rent was paid. Before the first of July, he  mortgaged  Eros  to  another  creditor  for  a  loan.  Asked  whether  the  praetor  should  protect  the  landlord against the latter creditor in a suit for Eros, he answered that he should. Although, when  Eros was mortgaged, nothing was yet owing for rent, even then the position of Eros was that he  could not be released from the mortgage without the landlord’s consent. So the landlord should  have priority.”) 21  D  20.3.4:  “Titius  cum  mutuam  pecuniam  accipere  vellet  a  Maevio,  cavit  ei  et  quasdam  res  hypothecae nomine dare destinavit ...” (“As Titius wanted to borrow money from Maevius, he  promised to repay Maevius the amount, listing certain assets to be mortgaged”). 22  D 20.1.15.1.

towards the medieval law of hypothec

165

clause seems to have been created towards the latter part of the period ca  150-127 bce.23 Legal sources show that an explicit agreement was initially  required to create such a contractual clause, but eventually it also became an  implied term.24 The reason why this change occurred cannot be ascertained  with certainty, but it has been suggested that it was in some way related to  the  standardisation  of  the  law  of  pledge  and  the  declining  socio-economic  circumstances of urban tenants.25 The content of this clause was restricted to goods classified as invecta et illata, a circumscribed category which referred to valuable movable objects.26  Since the content of this category was fixed, it did not matter whether the  landlord  consented  to  them  being  brought  into  the  rented  property.  To  reinforce  the  different  positions  in  urban  and  in  agricultural  tenancy,  the  following  statement  occurs  in  a  constitution  dated  to  223  ce  extracted  in  Justinian’s Code: It is settled law that things which tenant farmers bring onto a farm, if their owners  have agreed, are bound by the law of pledge to the owners of the property. But  where a house is leased, the owner’s knowledge is not required in the case of things  brought in or moved in; for these things are held by the law of hypothec.27

Roman legal sources never explain the reason for this distinction between  urban and agricultural tenancy. This text will therefore be used as the sedes materiae  to  investigate  how  these  rules  on  hypothec  were  received  into  medieval learned law. C. MEDIEVAL LEARNED LAW This  section  will  explore  a  number  of  juristic  writings/anonymous  works  representative of the glossators, the ultramontani and the commentators, as  the intellectual endeavours of these groups have traditionally been considered  in  cumulative  combination  to  have  created  the  ius commune  out  of  the Roman, canon, feudal and customary laws in the period 1100-1400.28 In  23 24 25 26 27

Du Plessis (n 14) 8. Du Plessis (n 14) 14-17. Du Plessis (n 14) 16-17. Du Plessis (n 14) 9-11. C 4.65.5 (trans B W Frier, Landlords and Tenants in Imperial Rome (1980) 110): “Certi iuris est  ea, quae voluntate dominorum coloni in fundum conductum induxerint, pignoris iure dominis  praediorum teneri. Quando autem domus locatur, non est necessaria in rebus inductis vel illatis  scientia domini: nam ea quoque pignoris iure tenentur.” 28  The contract of letting and hiring seemingly does not feature prominently in classical canon law,  apart from statements on remissio mercedis and emphyteusis (X 3.18.3-4). While this may create  the impression that canon law was not concerned with letting and hiring, it seems more plausible 

166

the creation of the ius commune

fact, the majority of texts here discussed are derived from commentaries on  Justinian’s Code, initially a more fertile source for juristic commentaries than  the Digest, and the source of the most important text on the topic.29  Both  Lo Codi  and  the  Summa Trecensis  –  two  of  the  most  influential  twelfth-century  anonymous  summaries  of  Justinian’s  Code  –  mention  the  hypothec.30 The passage from Lo Codi (4.69.21-22) states: If someone takes an object for use [ie rents it], like a house, the person from whom  he accepted it [i.e. the landlord] ought to release all the goods which he [i.e. the  tenant] ferried into the house when he wishes to move out of the house, provided  he has paid the rent or does not [wish to] remain for the [remaining] period for  which he has not paid. But if he has not paid the rent and is not prepared to pay  it, he to whom the house belongs [i.e. the landlord] is able to retain or obtain the  goods in the house pro pignore until such time as it has been paid.31 

The author(s) of Lo Codi, an instruction manual for lay-judges, distinguished  between pledges with possession and those without possession, even though  the term hypothec does not appear in the text.32 This can be seen from the  use of the phrase “retinere vel petere” towards the end of the passage, which  suggests the existence of two different regimes. This text shows a familiarity  with the Roman-law concept of a hypothec; but since it is merely a summary  it  is  silent  on  a  number  of  important  issues.  First,  it  is  unclear  whether  a  distinction between urban and agricultural tenancy was made and whether  the passage in question applies to both, since only urban tenancy (domus) is  mentioned. Secondly, it does not state whether the hypothec arrangement is  that the Roman law of letting and hiring was incorporated en masse into canon law. Where canon  law differed from Roman law, specific mention was made, as in the case of the two institutions  listed above. Certainly a number of prominent medieval canonists, such as Goffredus de Trano,  Henricus de Segusio (Cardinal of Ostia (Hostiensis)) and Sinibaldus Fliscus (later pope Innocent  IV) discuss letting and hiring and by the seventeenth century, most of the Roman law of lease was  being used in ecclesiastical courts: see G B de Luca, Theatrum vertitatis et iustitiae sive decisivi discursus (1669) book 4 for a collection of court decisions on letting and hiring. 29  But see C M Radding, “Reviving Justinian’s Corpus: The Case of the Code”, in P Andersen et al  (eds), Law before Gratian (2007) 35. 30  The Corpus legum (Brachylogus) does not contain any significant information about hypothec  apart from a single note on the word [actio] serviana in Brachylogus in H Böcking (ed), Corpus legum sive Brachylogus iuris civilis 3.14.3 (1829) 100. 31  H Fitting (ed), Lo Codi: eine Summa Codicis in provenzalisher Sprache aus der Mitte des XII. Jahrhunderts in der lateinischen Übersetzung des Richardus Pisanus (1906) 143: “Si aliquis recipit  aliquid ad pisionem, sicuti domum, ille a quo accepit debet dimittere omnes res ei quas portavit  in domo quando vult exire de domo, si pagavit precium vel non remanet pro eo quod non paget.  Set [sic] si non pagavit precium nec est paratus quod paget, ille cuius est domus potest retinere vel  petere res quae sunt in domo pro pignore donec sit pagatus.” The Latin translation has been used  instead of the original Provençal because it was edited, standardised and corrected by Hermann  Fitting. It therefore provides a more accurate text. For general information and a bibliography on  this work, see Coing, Handbuch I 200-201. 32  On the target audience of Lo Codi, see Lange, RRM § 49 I 415-421.

towards the medieval law of hypothec

167

included in the contract of lease as an implied contractual term, and whether  the creditor (landlord) could sell the objects to recover the debt owed, though  there is evidence to suggest that the notion of implied terms was known to  the drafters of this treatise.33  The discussion of the hypothec in the Summa Trecensis (4.57.4-5) resembles that of Lo Codi, but with a few additional insights: In  fact,  some  [elements]  arise  from  the  nature  of  the  action,  even  though  it  is  neither  stated  nor  implied,  for  example  that  at  the  end  of  the  lease  and  once  the rent has been paid, the goods brought in by the tenant are restored by the  landlord. Similarly if some [element] of violence intervened, he ought to restore  it  using  the  same  remedy.  Necessary  or  useful  goods  brought  into  the  rented  property also have to be restored.34

The authors of this summary were familiar with the notion of implied terms,  though in this text the idea is linked to that of the nature of the action.35 The  phrasing of the sentence suggests that the pledge/hypothec arrangements in  lease were seen as an implied term. In contrast to Lo Codi, the authors of the  Summa Trecensis explicitly used the term invecta et illata, but its meaning  is not provided. The extent to which the Roman law is followed is evident  from  the  statement  that  goods  hypothecated  would  only  be  restored  once  the rent had been paid and that this term also included necessary and useful  goods, though once again the text does not state whether a distinction was  drawn between urban and agricultural tenancy and whether this rule applied  to both. Although this passage does not contain much information on legal  remedies,  the  authors  speculated  that  the  actio conducti  could  be  used  to  enforce the return of the pledge/hypothec. The works of three glossators will be used to contribute to the discussion of  hypothecs in tenancy. Azo (d ca 1230)36 distinguished between a pledge and  a hypothec: “It [the text] does not state through the law of hypothec, since  33 Lo Codi 4.69.7 (n 29) 140: “Locacio [sic] est talis naturae quod aliquis potest petere illud quod  neque dictum neque cogitatum est in principio negocii [sic], sicuti est in hoc placito ...” 34  H  Fitting,  Summa Codicis des Irnerius mit einer Einleitung  (1894)  132:  “Veniunt  quidem  ex  natura actionis [actio locati] quedam, licet [nec] dicta nec cogitata sint, veluti ut finita conductione et solutis pensionibus res a colono illatae item inductae a domino restituantur. Item si quid  vi abstulerit, eodem iudicio restituere debet. Et quod in re locata necessario seu utiliter impe[n] dit, et hoc ei restituendum est.” This work, wrongly identified by Hermann Fitting as the Summa Codicis of Irnerius, is now thought to have been the product of collaboration between a number  of medieval jurists, produced ca 1135–1140. On the authors and target audience of this work, see  Lange, RRM § 45 I 402-407. 35  On the nature of the action and its importance in medieval learned law, see L Capogrossi Colognesi, Remissio Mercedis (2005) 122. 36  On the life of Azo, see the entry on Azzo in LMA I 1317-1318, and the entry by P Weimar, “Die  legistische Literatur der Glossatorenzeit”, in Coing, Handbuch I 202-203.

168

the creation of the ius commune

it does not appear [to have arisen] through agreement, but indeed through  delivery ...”37 This statement is a classic example of how the medieval jurists  struggled  to  explain  some  of  the  inconsistencies  in  the  Roman  legal  texts  brought about by alterations by Justinian’s compilers. In the text in question  (C 4.65.5) the term pignus is used whereas hypothec would have been more  accurate. Azo tries to justify the use of this term in the manner set out above.  His grasp of the difference between a pledge and a hypothec, though incorrect, demonstrates a familiarity with the two concepts. It also shows that he  understood the notion of an implied term that could be read into a contract  of lease and which affected movable property “delivered” (brought in) to the  rented premises without the parties having to agree on it explicitly: Indeed the owner has the right of hypothecs by virtue of the lease and if someone  were to let out urban property, goods brought and moved into it will be obliged  by the law of pledge even if it was done without the knowledge of the owner. But  goods brought onto agricultural property are only then bound if they have been  brought in with the consent of their owners.38

Azo speculated on the reason why the landlord’s knowledge was required in  cases of agricultural tenancy: It is customary to ask why it is different in [cases of] urban tenancy than in agricultural tenancy. Johannes [Bassianus] replied that in agricultural land [goods] are  not brought in unless we know [about it]. For rusticity [is] sacred and nothing is  brought [onto the land] except cattle, farm implements or similar goods. The matter  is  different  in  urban  property,  since  much  is  introduced  [into  urban  property]  about which we do not know something or which we have not seen. Consequently  in that case the obligation is nonetheless created without the explicit knowledge  of the owner.39

It is not immediately apparent what Azo meant by the “sanctity of rusticity”.40  When viewed in context, however, he seems to be suggesting is that, in contrast  to urban tenancy, a close relationship between a tenant of agricultural land  37 Azo,  Lectura super Codicem  on  C  4.65.5,  in  CGIC  III  171:  “Pignoris  iure]  Non  dicit  iure  hypothecae, quia nedum videtur conventione, sed etiam traditione …” 38  Azo, Lectura super Codicem on C 4.65.5, in CGIC III 171: “Habet etiam dominus ex locatione  ius  hypothecarum  et  si  quidam  locet  urbanum  praedium  res  inductae  et  illatae  etiam  domino  ignorante pignoris iure tenentur. Ea vero quae in rusticum praedium locatum inducuntur iam  demum tenentur si voluntate dominorum fuerint inducta …” In the remainder of this passage,  Azo also comments on the position of the goods of the secondary tenant. 39  Azo, Lectura super Codicem on C 4.65.5, in CGIC III 171: “Non est necessarius] Solet quaeri  quare  aliud  sit  in  urbano  quam  in  rustico.  Respondit  Ioannis  quia  in  rusticum  praedium  non  inferuntur, nisi quae volumus quemlibet scire. Sancta enim rusticitas non inducit[ur] nisi boves  et arrata et similia. Aliud in urbano praedio, praeterea quia multa inducuntur quae nolumus scire  aliquem vel videre. Unde ibi non exquisita scientia domini nihilominus contrahitur obligatio.” 40  Johannes Bassianus was a student of one of the quattuor doctores (Bulgarus) in Bologna and the  teacher of Azo: see Lange, RRM § 25 I 215-226.

towards the medieval law of hypothec

169

and the landlord existed, which permitted scrutiny of the goods brought onto  the land. Whether this reflects social practice in thirteenth-century Italy is  difficult to ascertain, given the wide variety of agricultural practices visible  during this period.41 Further  observations  about  the  term  sancta rusticitas  are  required.  As  the text stands, it is unclear whether Azo is attributing the term to Bassianus  or whether he is putting forward the notion personally. While this may not  seem significant at first, the use of this term in fact reveals much about the  author as the term sancta rusticitas has an important intellectual history in  medieval scholarship. The term seems to have originated from a letter by St  Jerome, dated 394 ce, and addressed to Nepotian, the nephew of Heliodorus  who had abandoned a life in the military for one in the service of the church.  At the time of this letter, he was a presbyter at Altinum where his uncle was  bishop. The letter is concerned with the duties of the clergy: The church is committed to the bishop; let him take heed whom he appoints to be  his almoner. It is better for me to have no money to give away than shamelessly to  beg what I mean to hoard. It is arrogance too to wish to seem more liberal than he  who is Christ’s bishop. “All things are not open to us all.” [Virgil, Eclogues 8.63] In  the church, one is the eye, another the tongue, another the hand, another the foot,  others ears, belly and so on. Read Paul’s epistle to the Corinthians and learn how  the body is made up of different members. [1 Corinthians 12.12-27] The rude and  simple brother must not suppose himself a saint just because he knows nothing;  and he who is educated and eloquent must not measure his saintliness merely by  his fluency. Of two imperfect things holy rusticity (sancta rusticitas) is better than  sinful eloquence.42

This letter was widely circulated because of its content and it is therefore not  surprising to find it being cited by a medieval jurist. In this context, however,  the citation is not used to provide substantive force to the legal argument.  Rather, it serves as an embellishment (almost a figure of speech) to reinforce  the extent of control exercised by landlords in the countryside.43 The  influential  glossator  Odofredus  from  Cremona  (d  1265)44  also  paid  attention to the rule which operated in the case of urban tenancy:  41 D  Osheim,  in  his  chapter  on  rural  Italy,  in  D  Abulafia  (ed),  Italy in the Central Middle Ages 1000-1300  (2004)  176,  indicates  that  landlord  supervision  of  parts  of  an  estate  (especially  in  church land) was a feature of thirteenth-century Italy. 42  English translation from http://www.newadvent.org/fathers/3001052.htm 43  It is interesting to note that the term sancta rusticitas also had a surprising second meaning in the  works of the Humanists who equated it with outdated and ignorant medieval knowledge: see J D  Tracy, “Against the ‘Barbarians’: The Young Erasmus and His Humanist Contemporaries” (1980)  11 The Sixteenth Century Journal 3 at 11-12. 44  Odofredus, though very influential, is not ranked among the most original of the glossators. On  his life, see the entry by P Weimar, “Odofredus de Denariis”, in LMA VI 1361; and P Weimar,  “Odofredus”, in Coing, Handbuch I 176-181. 

170

the creation of the ius commune

But  what  is  the  reason  for  the  difference  and  I  respond  that  the  reason  is  the  following:  in  urban  property  tenants  bring  many  things  onto  the  property  of  which the landlord has no knowledge: yet they are obliged through tacit pledge  [D 2.14.4pr].45 But in agricultural land it is not true that the tenant brings certain  goods onto the land which are unknown to any man.46 

Although not phrased in the same manner, Odofredus (while not mentioning  the idea of “sainted rusticity”) arrives at a similar conclusion to that of Azo,  namely that it is very seldom that goods are brought onto agricultural property  without  the  knowledge  of  someone.  Again,  this  suggests  a  greater  level  of  super  vision in agricultural tenancy. Accursius’ glosses on C 4.65.5 follow Azo’s approach to this topic:47 Gl certi: [Goods] introduced in order to reside there permanently for as long as  they inhabited [the leased property]48 [D 20.1.3249 and D 20.2;50 contra D 20.2.4].51

45 D 2.14.4pr: “item quia conventiones etiam tacite valent, placet in urbanis habitationibus locandis  invecta illata pignori esse locatori, etiamsi nihil nominatim convenerit” (“Likewise, on the ground  that even agreements by implication are valid, it is settled that in the letting of urban dwellings,  the movables [of the tenant] are constituted a pledge for the landlord even though nothing is  expressly agreed”). 46  Odofredus,  Lectura super Codice  on  C  4.65.5,  in  OIR  V  fol  257v: “Certi  iuris]  …  Sed  quae  est ratio diversitatis et respondeo ratio est illa: quod in praedium urbanum conductores multa  inducunt, quae nolunt locatorem scire: sed tamen illa tacite pignori tenentur [D 2.14.4pr]. Sed  in praedium rusticum non est verisimile, quod conductor inducat aliqua, quae non velit omnem  hominem scire.” 47  On the life of Accursius, see the entry by P Weimar, “Accursius”, in LMA I 75-76. 48  Accursius, Glossa in Codicem C 4.65.5, in CGIC X 248: “Induxerint ut ibi perpetuo sint quamdiu  inhabitaverint.” 49  D 20.1.32: “debitor pactus est, ut quaecumque in praedia pignori data inducta invecta importata  ibi nata paratave essent, pignori essent: eorum praediorum pars sine colonis fuit eaque actori suo  colenda debitor ita tradidit adsignatis et servis culturae necessariis: quaeritur, an et Stichus vilicus  et ceteri servi ad culturam missi et Stichi vicarii obligati essent. respondit eos dumtaxat, qui hoc  animo a domino inducti essent, ut ibi perpetuo essent, non temporis causa accommodarentur,  obligatos” (“A debtor agreed that whatever was brought on the mortgaged land or there arose or  was produced should be subject to mortgage. Part of the land was untenanted, and the debtor  handed it to his managing slave to farm, assigning him the slaves needed for the purpose. The  question was whether the managing slave, the other slaves sent to the farm, and the manager’s  vicarii were subject to the mortgage. Scaevola replied that only those who were brought on by  their owner as permanent, not those lent temporarily, were mortgaged”). 50  The entire title is cited and will therefore not be reproduced. 51  D 20.2.4: “Eo iure utimur, ut quae in praedia urbana inducta illata sunt, pignori esse credantur,  quasi  id  tacite  convenerit:  in  rusticis  praediis  contra  observatur.  1.  Stabula,  quae  non  sunt  in  continentibus aedificiis, quorum praediorum ea numero habenda sint, dubitari potest? Et quidem  urbanorum sine dubio non sunt, cum a caeteris aedificiis separata sint. Quod ad causam tamen  talis taciti pignoris pertinent, non multum ab urbanis praediis different” (“We accept that property  brought on to an urban leasehold is hypothecated, as if this had already been impliedly agreed.  The opposite is true of rural tenancies. 1. The position of stables which are not directly adjacent  to urban land is doubtful. Since they are separate, they are certainly not urban property; but so  far as the implied hypothecation is concerned they hardly differ from it”). 

towards the medieval law of hypothec

171

Gl ea quoque:52 [That is goods] which are introduced without the knowledge of  the  owner.  And  it  is  true  that  such  things  are  frequently  introduced  which  the  tenant does not wish the landlord to know about. This does not happen often in  agricultural  lands,  since  nothing  may  be  introduced  other  than  hoes  and  farm  implements and similar goods, [certainly not] money which they dare not wish for  [C 11.48.5;53 D 20.2.1;54 contra D. 20.2.10;55 D. 2.14.4pr.;56 D. 20.2.7pr].57 What  if the goods belonging to another are brought into a rented house or onto rented  land and the owner of the object knows it and does not dispute it, it seems that  goods of this nature are not obliged as is argued in D 13.7.20pr.58

Accursius’ summary of the accepted glossatorial interpretation of this area of  law shows that jurists were generally unsure about the reason for the difference in hypothec arrangements between urban and agricultural tenancy. The  glossators postulated that the difference may be related to the level of supervision visible in agricultural tenancy, and Accursius’ statement on the matter  seems to imply that tenants of agricultural land were closely supervised by  landlords and could not bring anything (especially money) onto rented land  without the knowledge of the latter.59 52 Accursius, Glossa in Codicem on C 4.65.5 (n 48) 248: “Quae inducuntur domino ignorante. Et est  vero quod saepe talia inducuntur quae non vult conductor sciri a locatore quod de levi non accidit  in rusticis cum non nisi ligones et arrata et similia inferant non pecunias quas non auderent optare  [C 11.48.5; D 20.2.1; contra D 20.2.10; D 2.14.4pr; D 20.2.7pr]. Quid si in domum vel fundum  conductum rem alienam quod inferat domino rei sciente et non contradicente videtur quod non  sint obligate ut arg. D 13.7.20pr.” 53  C 11.48.5: “domini praediorum id quod terra praestat accipiant, pecuniam non requirant, quam  rustici optare non audent, nisi consuetudo praedii hoc exigat” (CE 366) (“The owners of an estate,  shall receive products of the soil as rent, and shall not ask for money, for which the rustics do not  dare to wish, unless the custom on the estate demands that” [Blume Translation]). 54  D 20.2.1: “senatus consulto quod sub marco imperatore factum est pignus insulae creditori datum,  qui pecuniam ob restitutionem aedificii exstruendi mutuam dedit, ad eum quoque pertinebit, qui  redemptori domino mandante nummos ministravit” (“By a resolution of the senate in the time  of the emperor Marcus, the mortgage of a block [of flats] to a creditor, who made a loan for a  building to be put up on the site and conveyed to him, enures in favour of the person who, at the  owner’s bidding, provided the ultimate owner with the money to lend”). 55  D  20.2.10:  “tutoris  heres  cum  herede  pupilli  transactione  facta,  cum  ex  ea  maiorem  partem  solvisset, in residuam quantitatem pignus obligavit: quaesitum est, an in veterem contractum iure  res obligata esset. respondit secundum ea quae proponerentur obligatam esse” (“The heir of a  guardian made a compromise with the heir of his ward, paid most of it, and mortgaged property  for the rest. The question asked was whether the property was bound for what was due under the  old contract. Scaevola answered that on the facts stated it was”).  56  Cf n 45 above. 57  D 20.2.7pr: “in praediis rusticis fructus qui ibi nascuntur tacite intelleguntur pignori esse domino  fundi locati, etiamsi nominatim id non convenerit” (“As regards rural land the crops are impliedly  taken to be hypothecated to the owner of the land, even if not agreed in so many words”).  58  D 13.7.20pr: “aliena res pignori dari voluntate domini potest: sed et si ignorante eo data sit et  ratum habuerit, pignus valebit” (“Property belonging to another can be given as a pignus with its  owner’s consent. Even if the pledge is made without his knowledge and he later ratifies, the pignus  is valid”). 59  For  one  of  the  later  glossators’  views  on  this  matter,  see  Jacobus  de  Arena,  Commentarii in universum ius civile on C 4.65.5, in OIR XVI fol 30r.

172

the creation of the ius commune

The  scientific  approach  of  the  school  of  Orleans  (ultramontani),  which  influenced legal science during the thirteenth century, was not exactly revolutionary,  but  their  approach  towards  textual  analysis  and  textual  authority  was novel. One of the prominent French jurists associated with the school  of Orleans, Jacques de Revigny (d second half of the thirteenth century)60  approached  the  matter  by  distinguishing  between  urban  and  agricultural  property: Certi iuris] ... The reason for the difference can be [said to be] twofold, namely  that sainted rusticity has everything in plain sight, accordingly among those things  which the agricultural tenant introduce into the rented land nothing may be said  to be harmful if the owner knows about it: since he does not introduce anything  which  is  not  agricultural  implements,  cattle  and  other  similar  things  [required]  for agriculture. Thus the knowledge of the owner in the countryside equates with  knowledge of those things introduced onto the rented land. But the [urban] tenant  introduces  many  things  into  the  house  which  no  one  sees  in  any  way.  Thus  in  matters of invecta et illata it does not require the knowledge of the owner, for it  may be grievous if the owner sees all.61

De  Revigny’s  view  on  the  reason  for  the  distinction  between  hypothecs  in  urban tenancy and those in agricultural tenancy is similar to that of Azo. In his  opinion it was related to the sanctitas of rustic land which caused everything  to be visible to others. Further on in this discussion, he also examined in this  passage  whether  the  innkeeper  had  a  hypothec  over  goods  brought  in  by  guests and whether this required his express consent. De Revigny concluded  that consent was not required since it could be perilous to the traveller to  show his goods to the innkeeper. Baldus de Ubaldis (1327-1400), one of the most prominent of the commentators, enquired in his commentary on Justinian’s Code how the landlord’s  hypothec over urban tenant’s goods could exist without consensus: The invecta et illata of the agricultural tenant are obliged for the rent with the  consent of the owner [of the rented property], those of the urban tenant even if  the owner does not know about it. [This is what the text states.] The opposition is:  that obligation of pledges was created by agreement [D 2.14.4pr],62 thus an obligation can never be created between those who do not know about it, since among  those  who  do  not  know  about  it  there  is  nothing  in  consensus,  which  was  not  first conceived of in intellect. But one who does not know understands nothing,  60 On the life of Jacques de Revigny, see Coing, Handbuch I 281-282. 61 Jacques de Revigny, Lectura super Codice on C 4.65.5, in OIR I fol 209r: “Ratio diversitatis potest  esse duplex: quod sancta rusticitas omnia palam habet: unde quam in his quae infert rusticus in  fundum conductum non est vilis si sciat dominus: quod non infert nisi aratra, boves et similia  intrat ad culturam ruris: ideo exigit scientia domini in rure quod eo sciente sint invecta in fundum.  Sed conductor multa infert in domum quod nollet videri ullo modo. Unde non exigitur in invectis  et illatis in domo scientia domini. Nam quia omnia videret dominus hoc esset grave.” 62  Cf n 45.

towards the medieval law of hypothec

173

as below [C 8.53.10].63 One solution is that the agreement is introduced for the  benefit of man and thus consensus is required as in the contrary lex. However, a  certain agreement is introduced by law and then consent is not required, unless it  is presumed by law, as in this case [D 13.7.11.5]64 and above [C. 2.3.2].65 But the  opposition is that the agricultural and urban tenant agree on the matter and thus  also in the arrangement of the law. The solution is not the same reason, since it is  easy for the owner [to spot] movable goods in his rustic property: for it is generally assumed that agricultural people have little in the way of movable goods and  thus it is all in the open. But the owner does not count or know of the goods of an  urban tenant.66 63 C 8.53.10: “nec ignorans nec invitus quisque donat. unde si de hoc fundo non cogitasti, cuius velut  donationi consensisse continetur instrumento, maiores veritate rei quam scriptura vires obtinente  intellegis, de quo non cogitasti nec specialiter subscripsisti, nihil te perdidisse” (CE 294) “No one  makes a gift unknowingly or unwillingly. Hence, if you were not thinking of the farm which is  mentioned in the document as though you had consented to make a gift thereof, you know that  the truth has greater force than a writing and you have not lost what you did not have in mind or  which you did not specially mention in the writing.” [Blume translation]) 64  D 13.7.11.5: “solutam autem pecuniam accipiendum non solum, si ipsi, cui obligata res est, sed  et si alii sit soluta voluntate eius, vel ei cui heres exstitit, vel procuratori eius, vel servo pecuniis  exigendis praeposito. unde si domum conduxeris et eius partem mihi locaveris egoque locatori tuo  pensionem solvero, pigneraticia adversus te potero experiri (nam iulianus scribit solvi ei posse):  et si partem tibi, partem ei solvero, tantundem erit dicendum. plane in eam dumtaxat summam  invecta mea et illata tenebuntur, in quam cenaculum conduxi: non enim credibile est hoc convenisse, ut ad universam pensionem insulae frivola mea tenebuntur. videtur autem tacite et cum  domino aedium hoc convenisse, ut non pactio cenacularii proficiat domino, sed sua propria” (“It  is correct to say that the money is paid not only where it is paid to the creditor himself to whom  the thing is charged but also when it is paid with his consent either to someone whose heir he is or  to his procurator or to a slave in charge of collecting debts. Hence, if you rent a house and sublet  part of it to me and I pay my rent to your lessor, I will have the action on pignus against you [for  Julian writes that it is permissible to pay him]. And if I pay part to you and part to him, the same  will be said pro tanto. It is clear that my own furniture and movables will be charged only with the  sum for which I took my lodging; for it is not to be believed that my odds and ends were agreed  to be charged for the rent of the whole block. However, this agreement is impliedly taken to have  been made with the owner of the building as well, so that it is not from the bargain of the tenant  sublessor that the owner derives advantage, but from his own”). 65  C  2.3.2:“post  venditionem  hereditatis  a  te  factam,  si  creditores  contra  emptores  actiones  suas  movisse probare potueris eosque eas spontanea suscepisse voluntate, exceptione taciti pacti non  inutiliter defenderis” (CE 202) (“If you are able to prove that after the sale of the inheritance  made  by  you,  the  creditors  [of  the  inheritance]  brought  their  actions  against  the  purchasers,  and that the later voluntarily submitted to them, you are not ineffectually defended [in actions  brought against you for the same debt] by a plea of an implied pact” [Blume translation]). 66  Baldus de Ubaldis, Commentaria in iiii et v codicis librum on C 4.65.5 (1615) fol 137r: “Certi iuris]  Invecta et illata coloni, obligantur pro mercede domino scienti, inquilini etiam ignoranti. [hoc  dicit]. Op[positio] ista obligatio pignorum oritur ex conventione [D 2.14.4pr] quia ergo nunquam  oritur obligatio inter ignorantes, quia inter ignorantes nihil est in consensu, quod prius non dsit  in intellectu. Sed ignorans nihil intelligit, ut infra [C 8.53.10]. So[lutio] quaedam est conventio  introducta  pro  hominem  et  tunc  est  necessarius  consensu,  ut  in  lege  contraria.  Quaedam  est  conventio tamen inducta per legem, et tunc non est necessarius consensus, nisi praesumptus a  iure, ut hic, & [D 13.7.11.5] et supra [C 2.3.2]. Sed oppo[sitio] colonus & inquilinus conveniunt  in ratione, ergo & in iuris dispositione. Solutio non est eadem ratione, quam facile est domino res  mobiles rustici sui: nam rustici parum habere creduntur in rebus mobilibus et illud palam habent.  Sed res inquilini scire et numerare non habet dominus.”

174

the creation of the ius commune

Baldus  thus  came  to  the  conclusion  that,  because  it  is  introduced  by  law,  consensus is not required. His wide-ranging discussion explored various key  questions, such as the distinction between urban and agricultural tenancy. One  of the distinctive features of Baldus’ treatment of the subject is his penchant  for  introducing  tangents  into  the  discussion.  Thus,  he  enquired  whether  this rule applied to the innkeeper and whether goods had to be shown to or  deposited with the innkeeper before liability would vest. Additional tangents  included matters such as the lease of a colonus patriarius, emphyteusis, the  property of others, minors, the church and of servants working for tenants. D. CONCLUSIONS The aim of this volume is to investigate the intellectual techniques used by  medieval jurists to transform the case-based Roman legal texts into general  rules  of  law.  Against  this  backdrop,  this  chapter  set  out  to  investigate  the  medieval treatment of a rule of law for which there appeared to be no justification in Justinian’s compilation of Roman law. This investigation into medieval  justifications for this rule of law demonstrated the various techniques used  by  the  jurists.  While  it  would  be  perilous  (and  indeed  scientifically  incorrect) to draw general conclusions, since only the works of a select group of  jurists (as opposed to all the jurists representative of a specific period) were  analysed, certain tentative conclusions will be put forward. First, it is clear  that this rule of law rose from slender beginnings in epitomes of Justinian’s  Code. For much of the twelfth century, medieval jurists struggled with the  confused terminology (pignus/hypothec) found in Roman legal sources. As  for the rationale underlying this rule of law, the glossators eventually settled  on the issue of “control”. This was achieved by interpreting the rule of law in  a manner familiar to their circumstances. If Jacques de Revigny’s analysis of  this rule is taken as being representative of the thirteenth-century approach,  it would seem that “control” continued to be the dominant justification of the  existence of the rule of law, combined with a secondary argument relating  to  safety  of  goods.  By  the  fourteenth  century,  a  different  justification  for  the  existence  of  this  rule  becomes  visible  based  on  the  sophisticated  legal  notion of implied consent. As I have mentioned elsewhere, these phases in  the development of a legal rule are not linear and it would be unwise to draw  too general conclusions from the opinions of individual jurists, but an analysis  of these techniques provides modern scholars with a greater appreciation of  the intellectual tools used by medieval jurists to transform Roman law into  the ius commune of Western Europe.67 67 P J du Plessis, “The Creation of Legal Principle”, (2008) 4 Roman Legal Tradition 46.

8  The Ignorant Seller’s Liability for Latent Defects: One Regula or Various Sets of Rules? Jan Hallebeek* A. INTRODUCTION B. ROMAN LAW (1) Traces of historical developments in the Corpus iuris (2) Justinianic law C. THE ERA OF THE GLOSSATORS (1) Early legal scholarship (2) Placentinus and Johannes Bassianus (3) Azo (4) The last generation of glossators (5) Conclusions (a) Actio empti; actio quanti minoris (b) Praetorian remedies (c) The view of Accursius D. FROM THE GLOSSA ORDINARIA TO THE FOURTEENTH CENTURY (1) Romano-canonical procedure (2) The school of Orleans and Cinus E. THE ERA OF THE COMMENTATORS (1) Liability of the ignorant seller in case of latent defects (2) The scope of the Aedilician actio quanti minoris (3) Claiming price reduction: one remedy or various sets of rules? F. CONCLUSIONS

* The author is grateful to Kees Bezemer (Leiden) and Wim Decock (Leuven) for their comments on a draft version of this paper and to Margaret Hewett (Cape Town) for correcting the English and for further advice.

175

176

the creation of the ius commune

A. INTRODUCTION In 1999 a European Directive was issued that required specific protection  for  the  buyer  of  consumer  goods.  This  was  subsequently  implemented  in  the  various  European  jurisdictions.  When  repair  or  replacement  of  these  goods, in the case of non-conformity, is impossible or cannot be demanded  or when the seller refuses to provide either of these solutions, the buyer has  the remedies of reduction in price and rescission.1 The corresponding liability  of the seller does not necessarily depend on explicit warranties, contractual  clauses or any kind of malicious intent (mens rea) on his side, but is simply  imposed by the law or based on an implied warranty. In the Netherlands, the  implementation of the Directive resulted in some changes in the civil code.  These were promulgated in 2003. The remedies of price reduction and rescission, just mentioned, were adopted in a specific provision for sale contracts  in book seven of the code (art 7:22 BW), next to the already existing general  rules on rescission in view of breach of contract in book six (art 6:265 ff. BW). Because of the many similarities between, on the one hand, these remedies  of  price  reduction  and  rescission  and,  on  the  other,  the  Roman  law  actio quanti minoris  and  actio redhibitoria,  it  was  argued  that  the  ties  between  contemporary Dutch private law and Roman law in this respect had again  become apparent. These ties had been severed in 1992, when, together with  the introduction of the new Dutch civil code, the specific rule on the seller’s  liability for latent defects of the previous civil code, dating from 1838 (art  1543 OBW), had lost the force of law.2 General concepts of liability for latent  defects  –  a  kind  of  precursor  of  the  provisions  imposed  by  the  European  Directive – can already be found in early modern times.3 The sources of Roman law, especially the Corpus iuris, are not characterised by general rules of law; but if the concepts mentioned above are substantially based on Roman law, some rules must have already developed out of the  casuistic texts of the Corpus iuris. Cases dealing with the seller’s liability for  latent defects can be found in the Digest and Code titles on the remedies of    1 See art 3 (2) of the EU Directive 99/44, dealing with warranties to consumers within the European market. See also R Sefton-Green, Mistake, Fraud and Duties to Inform in European Contract Law (2005); for the links between Roman and contemporary law, see the chapter by M Schermaier in that volume.   2 J J Verhoeven, “Verborgen gebreken bij koop” (1990) 39 Ars Aequi 495; K Krzeminski, “Ad Edictum Aedilium Curulium; De Romeinsrechtelijke grondslag van de aansprakelijkheid voor verborgen gebreken” (2005) 54 Ars Aequi 692; J J Verhoeven, “Het gelijk der curulische aedielen; naar een moderne uitleg van de regeling betreffende de conformiteit bij koop in geval van verborgen gebreken” (2006) 23 Groninger Opmerkingen en Mededelingen 17.   3 H Coing, Europäisches Privatrecht, Band I, Älteres Gemeines Rechts (1500 bis 1800) (1985) 452-453.

the ignorant seller’s liability for latent defects

177

sale (D 19.1, de actionibus empti et venditi and C 4.49, de actionibus empti et venditi), as well as those on the aedilician edicts (D 21.1, de aedilitio edicto et redhibitione et quanti minoris, and C 4.58, de aedilitiis actionibus). It is the  purpose of this contribution to investigate the way in which developments in  medieval learned law contributed to the formation of a more general doctrine  on the seller’s liability for latent defects, as far as such liability is imposed by  the law, in other words, not based on warranties or contractual clauses voluntarily agreed upon by the parties, or on the seller’s malicious intent. B. ROMAN LAW Before investigating the medieval commentaries on the texts of the Corpus iuris relevant to our issue, it is necessary first to come to grips with the texts of  the Corpus iuris itself. This has to be done from the appropriate perspective. From the beginning of the twentieth century, after the Corpus iuris had  lost its significance for legal practice even in the German territories, the focus  of the majority of Romanists has generally been directed towards historical  developments during Roman antiquity revealed by the texts of the Corpus iuris, with much attention being paid in particular to the classical period of  Roman law (from the beginning of our era until the end of the third century  ce).  As  will  be  explained  below,  this  approach  is  not  that  of  the  medieval  jurists  and,  if  we  are  not  sufficiently  conscious  of  this,  our  perception  of  medieval scholarship can be obscured. On the other hand, it is this historical  development prior to the compilation of the Corpus iuris that can explain the  problems  in  interpretation  that  occur  when  we  attempt  to  understand  the  texts of the Corpus iuris as provisions of consistent legislation. (1) Traces of historical developments in the Corpus iuris As  regards  the  question  of  the  ignorant  seller’s  liability  for  latent  defects,  we can roughly distinguish three stages of historical development in Roman  Antiquity. The first starts with the introduction of the aedilician edicts at the  beginning of the second century BCE. Before that time, the regular contractual remedy for the buyer – the actio empti – was already in existence; but it  is very doubtful whether this remedy, being a iudicium bonae fidei, could be  used against an ignorant seller. The aedilician edicts, promulgated by magistrates  with  jurisdiction  in  the  marketplace,  known  as  the  aediles curules,  offered the buyer remedies for rescission (the actio redhibitoria or aestimatoria) and price reduction (the actio quanti minoris) for the case where a slave  or beast of burden, purchased at the market, showed latent defects. The edicts 

178

the creation of the ius commune

imposed upon the seller a duty to reveal to the buyer possible defects in the  merchandise and to give him certain warranties. Moreover, under the edicts  the seller was liable even for defects of which he was unaware. The texts of the  edicts themselves were incorporated in the Digest of Justinian. The one for  slaves can be found in D 21.1.1.1, and that for beasts of burden in D 21.1.38pr  (with a later addition for all the other cattle in D 21.1.38.5). The actio quanti minoris was not mentioned explicitly in the edict for slaves, but nevertheless  appears to have been applied to the sale of defective slaves (cf D 21.1.55). The second stage of development we can trace in the Corpus iuris dates  back  to  the  period  of  classical  Roman  law.  In  view  of  the  social  need  to  grant  the  buyer  better  protection  beyond  the  sphere  of  the  marketplace,4  the regular contractual remedy of sale – the actio empti, which was actually  older than the aedilician edict – was applied by analogy with the aedilician  remedies. In the formulary procedure this could presumably be achieved by  interpreting the clause “ex fide bona” of the actio empti as covering the observance of aedilician principles.5 This implies that it began including a liability  of ignorant sellers. Some texts indicate that the seller who was unaware of  the defects, and thus acting in good faith, could be held liable under the actio empti  of  civil  law.  Moreover,  this  remedy  now  was  applied  to  claim  price  reduction (D 19.1.13pr) and rescission (D 19.1.11.3). The third and final stage of development originated in Justinian’s decision  to  retain  the  old  remedies  of  the  aedilician  edicts,  which  in  his  days  had  become more or less redundant, and to extend their application. This development is clearly visible in the Corpus iuris. The Digest and Code contain,  next to the titles on the remedies for sale (de actionibus empti et venditi – D  19.1 and C 4.49), separate titles on the aedilician edicts (D 21.1, de aedilitio edicto et redhibitione et quanti minoris, and C 4.58, de aedilitiis actionibus).  Moreover,  according  to  D  21.1.1pr  and  D  21.1.63,  the  aedilician  edict  for  the sale of slaves also applies to other things. These two texts are regarded as  containing Justinianic interpolations. However, only a few examples of goods    4 M Wlassak, Zur Geschichte der negotiorum gestio. Eine rechtshistorische Untersuchung (1879)  169-173.  This  opinion  was  followed,  amongst  others,  by  H  Honsell,  Quod interest im bonaefidei-iudicium. Studien zum römischen Schadensersatzrecht [= Münchener Beiträge zur Papyrusforschung  und  antiken  Rechtsgeschichte  LV]  (1969)  81-87;  and  Kaser,  RPR2  558.  In  some  of  the  older  literature  D  19.1.13pr  and  D  19.1.11.3  were  still  presumed  to  be  interpolated.  See  G Impallomeni, L’Editto degli edili curuli (1955) 245, 251; and P Stein, “Medieval discussions  of the buyer’s actions for physical defects”, in D Daube (ed), Studies in the Roman Law of Sale Dedicated to the Memory of Francis de Zulueta (1959) 102.   5  It may be argued, though, that it is not very obvious to grant a iudicium bonae fidei such as the  actio empti against a seller who is not to blame for anything. The liability is usually explained as  resulting from a breach of duty of care: being owner, the seller is the first to notice the defect and  should inform the buyer. 

the ignorant seller’s liability for latent defects

179

with latent defects, other than slaves or cattle, can be found in the Digest title  on the aedilician edicts (D 21.1, de aedilitio edicto et redhibitione et quanti minoris), containing replies from classical jurists. The same holds good for  the imperial constitutions adopted in the title de aedilitiis actionibus of the  Code (C 4.58).6 (2) Justinianic law To  understand  how  the  texts  of  the  Corpus iuris  were  interpreted  during  the Middle Ages and how they could serve as material to build up a doctrine  on liability for latent defects, it is important not to approach the texts in the  fashion of those many modern Romanists who concentrate on the developments in the law of antiquity prior to the compilation of the Corpus iuris.  This was by no means the approach of the medieval scholars. These jurists  were perfectly aware of the historical genesis of the Roman law embodied  in  the  Corpus iuris.  They  knew  exactly  when  the  classical  jurists  and  the  emperors had lived, but nonetheless they did not approach the Corpus iuris  as if it were a historical source. They considered all texts of the Corpus iuris  to be provisions of one and the same scheme of legislation, having an equal  force of law and promulgated at approximately the same time, namely in the  sixth century by Justinian. Accordingly, it did not matter whether a text originated from pre-classical, classical, or post-classical times, or whether a text  was interpolated either by Justinian or earlier in post-classical compilations  of  classical  replies.  It  is  necessary  to  adopt  this  perspective  to  understand  medieval legal scholarship. Reading  these  texts,  containing  such  traces  of  historic  development,  as  provisions  of  consistent  legislation  containing  no  contradictions  leads  to  problems in interpretation. The Corpus iuris appears to contain two separate  kinds of remedy for the buyer of goods with latent defects in order to hold the  seller liable: first, the actio empti, and secondly, the aedilician actions. Both  seem to be available for the same kinds of defects. The aedilician actions had  originally only applied to slaves and cattle with an illness (morbus) or a certain  (corporeal) defect (vitium); but, since Justinian extended the application of  the aedilician actions to the sale of things other than just slaves and cattle, the  range of defects should be understood as extended beyond those mentioned  specifically in the original edicts.7 Such defects, not mentioned in the edicts,    6 D 21.1.49 and C 4.58.4 (Diocletian) deal with sale of defective plots of land.    7 See also for the Justinianic reform, N Donadio, “Garanzia per i vizi della cosa e responsabilità  contrat tuale”, in E Jakab and W Ernst (eds), Kaufen nach Römischem Recht. Antikes Erbe in den europäischen Kaufrechtsordnungen (2008) 61 at 69-71.

180

the creation of the ius commune

included, for example, poisonous herbs growing in the plot of land one had  bought. Some defects, however, remained exempted from application of the  aedilian edicts. At the same time, there were certain defects where the actio empti could  not  be  used.  Furthermore,  both  kinds  of  remedy,  the  actio empti  and  the  aedilician  actions,  were  aimed  or  could  be  aimed  at  the  same  purpose,  namely price reduction or rescission. Finally, both kinds of remedies seem to  have been available against a seller in good faith, that is, someone who was  not aware of the defects. As regards the use of the actio empti against the  ignorant seller of defective goods, the texts in the Digest are not unequivocal. This resulted in a debate among the glossators. Moreover, in the cases  where ignorant sellers appear to be liable, no general pattern can be found  regarding the extent of their liability. The most striking difference between the two types of remedy is that the  actio empti originated in the ius civile or civil law (originally resulting from  legislation, plebiscites, decisions of the Senate, imperial constitutions, and so  on), whereas the actio quanti minoris and the actio redhibitoria belonged to  the ius honorarium or praetorian law (originally resulting from the edicts of  the magistrates). Though the distinction between civil and praetorian actions  was still preserved in the Corpus iuris (see J Inst 4.6.3), it had lost its procedural significance. Since the formulary procedure had come to an end (342  ce),  it  was  no  longer  necessary  in  civil  litigation  to  request  the  magistrate  for  an  action  that  derived  from  either  civil  law  or  praetorian  law.  But  one  important rule was preserved (J Inst 4.12pr): civil-law actions were perpetual,  praetorian actions temporal or time-limited. Before turning to the commentaries of the glossators and commentators, it  is first appropriate to present a general outline of the provisions of the Corpus iuris relating to the ignorant seller’s liability for latent defects. These provisions centred around the specific use of the actio empti for price reduction or  rescission in the titles D 19.1 and C 4.49, and the remedies of the aedilician  edicts in titles D 21.1 and C 4.58 – namely the actio quanti minoris, the actio redhibitoria,  and  the  actio in factum.  The  Corpus iuris  indicates  that  the  actio empti was the primary remedy, and the aedilician actions the additional  ones. This follows from the composition of the Digest, as explained in the  introductory  constitutions  by  Justinian  himself.  All  that  was  introduced  by  laws concerning purchase and sale should have a prominent position within  the books on patrimonial law (de rebus), that is, books twelve to nineteen of  the Digest. Provisions derived from the former aedilician edicts are compiled  in an additional book in the fourth part, to wit the twenty-first book of the 

the ignorant seller’s liability for latent defects

181

Digest, which necessarily has its place not far from the contracts of sale, as if  these provisions were their subordinates (ministrae).8 According to Justinian,  the aedilician remedies developed from their beginning as attendants (pedisequae) in the wake of the remedies for sale.9 This precedence of the civil law  actio empti over the praetorian actions of the aedilician edicts is, moreover,  confirmed by the idea, still preserved in the Corpus iuris, that it is the task  of the ius honorarium to support, supplement and rectify the ius civile (D  1.1.7.1). The main provisions dealing with the use of the actio empti as a remedy for  latent defects can be found in D 19.1.13pr and C 4.49.9.10 The opening lines  of D 19.1.13pr are generally phrased as if they were a rule of law not limited  to the sale of certain goods or the existence of certain defects, although some  specific  examples  follow.  When  the  seller  is  unaware  of  the  defects,  the  remedy (actio empti) can be used to claim price reduction. When the seller  knowingly sold the defective object, the buyer can, on the other hand, claim  compensation for his interest that amounts to full damages including consequential losses. As stated above, there are, however, some texts in the same  Digest title where a seller in good faith appears not to be liable. This is the  case where the slave he sold is a thief (D 19.1.13.1) or taxes (tributum) are  due for the plot of land he sold (D 19.1.21.1). Other fragments in the Digest  seem to confirm that D 19.1.13pr has a general purport. This possibility was  at least discussed in later times by the medieval jurists. D 19.1.13pr and C 4.49.9 also make clear the way in which assessment of  price reduction is supposed to take place. The buyer can claim the difference  between what he paid and what he would have paid if he had known of the  defect (quanto minoris empturus esset). Thus, it is not the objective difference in value that is owed, and certainly not as much as the buyer’s interest.  The generally-phrased rule of D 19.1.13pr does not mention the possibility  of claiming rescission, but other texts in the same title show that the actio empti can also be used for this purpose. In D 19.1.11.3 this is again phrased  in general terms: rescission in view of latent defects is also covered by the  actio empti. Other texts, however, have a mere casuistic character: the actio empti  is  used  for  rescission  of  the  sale  of  a  slave-woman  whom  the  buyer  erroneously considered to be virgin.11 There is no indication whatsoever that    8 Constitutio Omnem § 4.    9 Constitutio Tanta § 5. 10 The cases, however, are not identical. In D 19.1.13pr the seller did not give any specific information concerning the merchandise; in C 4.49.9 he did, namely the amount of capitatio due for the  plot of land, which was sold.  11  D 19.1.11.5.

182

the creation of the ius commune

the use of the actio empti for price reduction or for rescission was limited to  a certain period of time. Being a civil action, it was perpetual. The most important, generally-phrased lines, dealing with the aedilician  actions  –  the  actio  redhibitoria,  the  actio quanti minoris  and  the  actio in factum12 – can be found in D 21.1.1.1 (the edict for slaves) and D 21.1.38pr  (the edict for beasts of burden). The aedilician edicts primarily imposed a  duty  upon  the  seller  to  provide  the  buyer  with  the  necessary  information  concerning  the  merchandise  and  to  give  certain  warranties.  Moreover,  as  appears from D 21.1.1.2, the edict for the sale of slaves was applicable also  to cases where the seller was unaware of any defects. It was argued that this  was not unfair because the seller was in a position to know these defects, and  it made no difference to the buyer whether he was deceived by the seller’s  lack of knowledge (ignorantia) or by the latter’s slyness (calliditas). As stated  above, the edict for slaves applied, according to its own wording, also to the  sale  of  “immovables,  movables  as  well  as  living  things”,13  and,  according  to  another  text,  even  to  all  other  goods  (venditiones ... ceterarum quoque rerum).14 Accordingly, the provision was supposed to have a wide application,  making the other edict, that is the one for beasts of burden reproduced in D  21.1.38pr, more or less superfluous, unless one argued that the latter still had  some significance, since D 21.1.1.1 mentions only rescission and D 21.1.38pr  includes the remedy for price reduction. But, as seen above, other texts in  title D 21.1 make clear that, in the case of defective slaves, it was possible  to claim reduction of the selling price. The fact that the examples given in  the Digest, where the aedilician remedies applied, almost exclusively cover  cases of defective slaves and cattle has a historic reason and does not mean  that the application is limited to such merchandise. Accordingly, the defects  which resulted in liability should not have been seen as limited to illnesses  (morbi) and corporeal defects (vitia).15 There was only a limited number of  defects where the aedilician actions did not appear to be applicable, although  sometimes the actio empti could nevertheless be used. These included the  cases of slaves with certain mental defects16 and the sale of a slave-woman  erroneously considered to be a virgin.17 Other cases, where the Digest explicitly  and  in  more  general  terms  maintained  that  the  aedilician  liability  for  12 In some specific instances in title D 21.1, such an actio in factum is granted to claim reimbursement of the selling price or rescission (see D 21.1.31.17, 22). 13  D 21.1.1pr. 14  D. 21.1.63. 15  C 4.58.3 mentions a plot of land with poisonous herbs; cf also D 21.1.49.  16  D 21.1.4. 17  D 19.1.11.5.

the ignorant seller’s liability for latent defects

183

latent defects was excluded, included a purchase from the Imperial Treasury  (fiscus),18 and the emptio simplaria.19 The term simplaria probably referred  to daily purchases of minor importance.20 The main differences between, on the one hand, the actio empti used for  price reduction and rescission and, on the other, the aedilician actions, used  for the same purposes, exists in the fact that the praetorian remedies were  only available within a limited period of time. The actio redhibitoria had to be  brought within six months from the day the contract of sale was concluded,21  or  from  the  moment  the  defect  became  known  to  the  buyer  or  the  buyer  could have known of it.22 The actio quanti minoris had to be brought within  one  year  from  the  day  the  contract  of  sale  was  concluded.23  Finally,  there  was also a difference between the actio empti and the actio quanti minoris  as regards the assessment of price reduction. Whereas the first remedy was  aimed at the difference between the selling price and what the buyer would  have paid if he had known of the defect,24 the second was aimed at the difference between the selling price and the actual, that is, the objective market  value of the defective object.25 It may be questioned in which way the compilers considered the praetorian actions to add something to the civil remedy. In the Corpus iuris the  civil actio empti seems to be the primary remedy for latent defects, but its  applicability was somewhat wider than that of the aedilician actions. It can be  used for some defects that the aedilician edicts did not cover and its use was  not restricted to a limited period of time. If indeed praetorian law supported,  supplemented and rectified civil law, as D 1.1.7.1 stated, and Justinian wanted  the  aedilician  actions  to  function  as  remedies  additional  to  the  remedy  of  sale, what could these actions offer? The only obvious advantage is the fact  that the objective assessment of the price-reduction of praetorian law could  sometimes result in a higher amount than the subjective assessment of civil  law. It is possible, of course, to question whether this was sufficient ground  for not abrogating the aedilician edicts, and to ask what other reasons might  have induced Justinian to adopt in his legislation the two titles on the aedili18 See D 21.1.3. 19 D 21.1.48.8. 20 In H G Heumann and E Seckel, Handlexikon zu Quellen des römischen Rechts (many editions),  simplarius is described as geringfügig. 21  D 21.1.38pr; C 4.58.2; D 21.1.19.6. 22  D 21.1.55. 23  D 21.1.19.6. On the annus utilis, see D 21.1.38pr and C 4.58.2.  24  See quanti minoris empturus esset, si … scisset in D 19.1.13pr or quanto, si scisset emptor ab initio, minus daret pretii in C 4.49.9. 25 See quo minoris cum venirent fuerint, D 21.1.38 and D 21.1.31.5.

184

the creation of the ius commune

cian actions next to those on the actions of sale. Usually this is ascribed to  the traditionalistic tendencies of both Justinian himself and the Eastern law  schools.  No  matter  what  the  reason  may  have  been,  the  medieval  jurists  encountered two distinct remedies and two distinct sets of rules, and, as will  be seen, this fact was to cause serious problems in interpretation. C. THE ERA OF THE GLOSSATORS In discussing the era of the glossators, the focus will be on the actio empti  rather than on the aedilician actions, since the actio empti was the principal  remedy in Justinianic law and most of the time the aedilician actions were  applied in cases where the seller had given warranties, or was supposed to  have done so. It  is,  however,  first  appropriate  to  notice  that  the  extensive  interpretation given in the medieval period to the Roman rule on a fair price (iustum pretium) found in C 4.44.2, in some cases covering latent defects, meant the  buyer could claim rescission of the sale on the grounds that the selling price  had been too high. This provision had originally granted the seller an action  if something was sold for less than half the fair price. In such a case, he was  considered  to  suffer  an  extreme  prejudice  (laesio enormis).  In  the  Middle  Ages the rule on a fair price was generalised. It was used not just to protect  the seller, but also was applied when the buyer was deceived, in the sense that  the selling price was too high. The remedy was named the condictio ex lege rem maioris after the opening words of C 4.44.2. Among the early glossators  at Bologna, there was, however, no agreement as to in exactly which cases the  buyer suffered such laesio enormis. According to Martinus (d before 1166)  and some others, the actual selling price had to be at least twice as much as  the fair price. Azo was of a different opinion, arguing that the buyer could  claim  rescission  of  the  sale  because  of  laesio enormis  if  what  he  had  paid  on top of the fair price was at least half of that fair price.26 It was his view  which prevailed and which was adopted in the authoritative Accursian gloss.27  As  a  consequence  the  remedies  of  sale  and  the  aedilician  remedies  were  still necessary for cases of latent defects, where the difference between the  selling price and the fair price did not exceed half of the latter.

26 G F Haenel (ed), Dissensiones dominorum sive controversiae veterum iuris romani interpretum,  Hugolinus § 253 (1834; repr 1964) 426-427.  27  See the gloss iudicis ad C 4.44.2. 

the ignorant seller’s liability for latent defects

185

(1) Early legal scholarship One  of  the  first  issues  to  be  discussed  by  the  Bolognese  glossators  was  probably the extent of the liability of the seller who was unaware of the defects.  Although this is not always stated explicitly in the sources, this discussion did  not involve situations where the seller had given express warranties against  defects in his merchandise, but this clearly followed from the nature of the  Roman law texts adduced as arguments. The  author  of  Lo Codi,  an  anonymous  summa  on  the  Code,  originally  written in the Provençal language and dating from just after the middle of  the twelfth century, noted that sometimes in the Digest the seller who was  unaware  of  the  defects  was  also  held  liable  under  the  actio empti.  Some  examples were given (the leaky barrel of D 19.1.6.4 and the unsound beams  of D 19.1.13pr), without entering into the problem of whether or not these  examples  contained  the  general  rule,  namely  that  ignorant  sellers  were  always liable. The cases which did consider the ignorant seller liable did not  adopt,  however,  an  equal  extent  of  liability.  Most  of  the  time  the  ignorant  seller was liable for the balance between what the buyer paid and the lesser  sum he would have paid if he had known the defect, but some texts, such  as D 19.1.6.4 (sale of a leaky barrel) or D 19.2.19.1 (lease of a leaky barrel),  were understood to point in the direction of liability for the other party’s full  interest. In Lo Codi this was justified by the supposition that the vendor or  lessor of a leaky barrel might not have known of the defect, but that he was  to blame for such lack of knowledge.28 The  same  line  of  reasoning  was  adopted  and  further  elaborated  by  Rogerius  (d  between  1162  and  1166),  who  is  said  to  have  founded  a  law  school at Montpellier. This glossator drew a distinction between three types  of ignorant sellers. Some were not to blame for their ignorance and could  not be held liable. Some were ignorant when they could easily have known  of the defects, and these are liable quanto minoris “as in the case of selling  a slave who is inclined to run away or is otherwise defective”. In exceptional  cases, such as the sale of a leaky barrel or of an infertile slave woman,29 the  ignorant seller was liable for the buyer’s full interest. The reference for the  second category to the servus fugitivus and servus morbosus, combined with  an explicit reference to D 22.1.1, may indicate that, according to Rogerius,  the  actio empti  for  price  reduction  should  be  applied  in  conformity  with  the assessment rule of the aedilician edicts. If so, Rogerius also adopted the  28  H Fitting (ed), Lo Codi in der Lateinischen Übersetzung des Ricardus Pisanus 4.62.9 (1906) 129.  D 19.1.6.4 does not speak about interest or consequential damages.  29  D 19.1.6; D 19.1.1.

186

the creation of the ius commune

objective  standard  –  that  is,  the  difference  between  the  selling  price  and  the market value at the time of the contract – for the use of the actio empti,  although this is not in conformity with D 19.1.13pr and C 4.49.9.30 Rogerius’  contemporary Albericus rejected the former’s opinion that the vendor of a  leaky barrel (the case in D 19.1.6.4) was liable for the buyer’s full interest.  The text of D 19.1.6.4 stated that the seller was obliged to deliver a barrel  which  did  not  leak,  even  if  he  were  unaware  of  the  defect;  but  Albericus  applied to this case the rule of D 19.1.13 that limits the vendor’s liability to  the lesser amount the buyer would have paid if he had known of the defect.31 The underlying problem probably was that the Corpus iuris contains cases  where the ignorant seller is not liable at all, where he is only liable for quanti minoris, or where he is liable for the buyer’s interest. This prompted Rogerius  to discern three groups of ignorant buyers, each belonging to a certain category.  A similar distinction was drawn by his contemporary Henricus de Baila in a  gloss to the case in C 4.49.9. If the seller knew the exact amount of capitatio  due  for  the  plot  of  land  he  sold,  but  deliberately  provided  the  buyer  with  incorrect information, he was liable for all damages. The seller who gave the  wrong information out of ignorance was liable only for the balance between  what the buyer paid and the lesser sum he would have paid, whereas the seller  who did not say anything could not be blamed and was not liable at all.32 Later glossators preferred to acknowledge a principal rule for all ignorant  sellers, and it is possible that Albericus, in his rejection of Rogerius’s interpretation of D 19.1.6.4, was one of those. The Casus Codicis of Wilhelm of  Cabriano – said to be a faithful reproduction of the teachings of Wilhelm’s  master,  the  glossator  Bulgarus  (d  1166)  –  offered  two  explanations  for  the  fact that the ignorant seller was sometimes liable for defects and sometimes  not. It argued that he was liable when he had given warranties, and that he  was not liable, if he had not done so. Alternatively, one could argue that the  texts that said that the ignorant seller was not liable should be understood as  if he was not liable for the buyer’s interest, but that, despite his ignorance, he  was liable for the balance between what the buyer paid and the lesser sum he  30 See the gloss with siglum R ad D 19.1.13 in Paris BN MS lat 4450 fol 194v: “Venditor rei uendite,  detrimentum ignorans …” Cf Rogerius, Summa Codicis, IL (De actionibus empti et venditi), in  BIMAE I 128. Rejecting for the actio empti the subjective standard for calculating the quanti minoris – i.e. the amount the seller would have paid less had he known the defect – can explain  the occurance of Rogerius in the gloss essem empturus ad D 19.1.13pr of Accursius.  31  See the gloss by Albericus ad D 19.1.6.4 in Paris BN MS lat 4450 fol 193v, in F C von Savigny,  Geschichte des römischen Rechts im Mittelalter, 3rd edn (repr 1956) IV 428. 32  See the gloss with siglum yr. ad C 4.49.9 in Paris BN MS lat 4536 fol 80va: “Venditor sciens nisi  ueram capitationem predicat …” A similar gloss, but without siglum, can be found in München BSB  MS Clm 22 fol 84rb: “Venditor sciens nisi in capitatione …”

the ignorant seller’s liability for latent defects

187

would have paid had he known of the defects.33 This seems to be a first step  in the direction of a general rule for all ignorant sellers. The same approach  can be found in a gloss by Otto Papiensis.34 (2) Placentinus and Johannes Bassianus Placentinus (d 1192), a citizen of Piacenza, was one of the most important  glossators  of  the  twelfth  century.  He  taught  at  Mantua  where  ca  1160  he  wrote  a  treatise  on  legal  remedies,  the  Libellus de actionum varietatibus,  which begins with the words Cum essem Mantue. Probably around 1170 he  moved to Montpellier in France to succeed his master Rogerius. There he  wrote his Summa Codicis. In Cum essem Mantue, when discussing the remedies of the contract of  sale,  Placentinus  adopted  the  tripartite  division  of  Rogerius.  The  ignorant  seller was sometimes liable for all damages, as in the case of the leaky barrel;  sometimes only for a limited amount, as in the case where he had sold cattle  with  a  disease  or  a  runaway  slave;  and  sometimes  not  liable  at  all,  unless  he gave the buyer warranties, as in the cases where he had sold a thief or  had not mentioned the existence of a servitude. For the assessment of the  reduction in price, Placentinus clearly followed the subjective calculation of  D 19.1.13pr, which was the difference between the purchase price and the  price the buyer would have paid had he known of the defect.35 There is a  gloss by Lotharius (d after 1211) suggesting that Placentinus considered the  cases where the ignorant seller was liable to be exceptions to the principal  rule that ignorant sellers should not be liable. An argument to consider this  as the principal rule was found in the statement of D 19.2.22.3 (see also D  4.4.16.4) that, according to nature, seller and buyer may deceive each other.  Later glossators considered this text not applicable to cases where the buyer’s  disadvantage resulted from defects in the merchandise, except in the cases of  D 19.1.13.1 (sale of a fur) and D 19.1.21.1 (sale of a predium tributarium).36 If a first step towards one general rule for ignorant sellers can be traced  33 Ad  C  4.49.9  in  T  Wallinga  (ed),  The Casus  Codicis of Wilhelmus de Cabriano  [=  Studien  zur  Europäischen Rechts ge schichte CLXXXII] (2005) 314-315.  34  This  glossator  considered  the  cases  of  selling  and  leasing  out  of  leaky  barrels  (D  19.1.6.4  and  D 19.2.19.1) as exceptional, where no distinction should be made between the one who knows  the defect and the one unaware. Both are liable for the full damages. See the gloss with siglum  Ot. ad C 4.58.1 in Munich BSB MS Clm 22 fol 86va and Paris BN MS lat 4536 fol 82vb: “Set si  ignorasset  …” 35  See Placentinus, Summa ‘Cum essem Mantue’ siue de accionum uarietatibus, tit XXII no 235, ed  G Pescatore [= Beiträge zur mittelalterlichen Rechtsgeschichte V] (1897; repr 1967) 56. 36  See the gloss by Lotharius ad D 19.1.13 in Paris BN MS lat 4450 fol 194v, ed in Savigny, Geschichte  (n 31) IV 465.

188

the creation of the ius commune

in  the  Casus Codicis  of  Wilhelm  of  Cabriano,  a  similar  approach  is  found  in  the  Liber Pauperum,  a  well-structured  compendium  on  the  Digest  and  Code, composed by Vacarius (ca 1120-after 1198), the first scholar to teach  Roman law in England. This referred to an existing opinion that there was  a difference between leaking barrels and other things containing some kind  of defect, since it was very easy to check whether or not a barrel leaks. As  a  consequence,  the  sale  of  leaky  barrels  should  be  seen  as  an  exceptional  case, where it was justifiable to hold the ignorant seller liable for the buyer’s  full damages.37 Since Vacarius did not mention the cases where the ignorant  seller was not liable at all, this pointed in the direction of a single criterion for  the extent of the ignorant seller’s liability, namely quanti minoris in the sense  of the balance between what the buyer paid and the lesser amount he would  have paid, had he known of the latent defect. Johannes Bassianus (dates unknown) holds an important position within  the tradition of Bolognese glossators, as not only a student of Bulgarus but  also  the  master  of  Azo,  whose  influence  would  last  until  the  end  of  the  Middle Ages. Johannes Bassianus was the first to formulate a general rule  on the liability of ignorant sellers. This can be found in an extensive gloss to  C 4.4.9.9. He did not use the term regula, but the expression generaliter … extendimus: we extend the provision generally. The criterion of D 19.1.13,  namely  that  the  ignorant  seller  was  liable  for  the  balance  between  what  the  buyer  paid  and  the  lesser  amount  he  would  have  paid,  had  he  known  the  defect,  was  now  extended  to  all  other  cases  and  all  other  defects,  not  just those mentioned in that text. This construction of a rule on the basis of  one single text in the Digest was justified by appealing to D 39.2.30.2: what  had  been  written  down  as  an  example  could  also  be  taken  to  refer  to  other  things.38 Thus, the instances mentioned in D 19.1.13 should be taken as mere  examples. Only two cases were excepted: the sale of a thief and the sale of a  plot of land liable for tributum. In these cases, the ignorant seller could not be  held liable since, as Johannes Bassianus subsequently explained, there was no  slave who had never stolen something from his master. Thus every slave might  be presumed to be a thief. For this reason it was written: “Quod domini faciant,  audent cum talia fures.” The quotation is from Virgil, Eclogues 3.16, but this  source was not mentioned by the gloss. Furthermore, the seller of a plot of land  for which tributum was owed (which in antiquity applied only to provincial and  37 Vacarius, The Liber Pauperum, ed by F de Zuleta, lib IV tit 46 [= Selden Society XLIV] (1927;  repr 1972) 151. 38  “Quod dictum est ‘aquae ducendae causa’, exempli gratia scriptum est: ceterum ad omnia opera  stipulatio accommodabitur.”

the ignorant seller’s liability for latent defects

189

not to Italian land) could not be held liable, since everyone must know that,  according  to  the  ancient  laws, tributum  was  owed  for  provincial  tenements.  Thus the buyer’s supine ignorance could not work out to his advantage.39 The gloss of Johannes Bassianus continues stating that there was another  opinion, to the effect that the ignorant seller was never liable except in some  of the cases mentioned before and in the one concerning the sale of secondhand garments as new in D 18.1.45. According to this opinion, it was the text  on  the  sale  of  a  fundum tributarium  which  should  be  adopted  as  a  general  rule  and  applied  to  other  cases.  But  it  was  also  possible  to  argue,  Johannes  Bassianus stated, that the ignorant seller was always liable for the the balance  between what the buyer paid and the lesser amount he would have paid if  he had known of the defect. This meant that the ignorant seller of a fundum tributarium did not have to pay additional damages to the same level as the  seller who knew of the defect. The ignorant seller of a slave who was a thief did  not have to pay the same level of damages as a seller who knew he was selling a  runaway slave. The latter had to pay full damages; the seller of the thief had to  pay only as much as the buyer would have paid less.40 39 “Qui  pecus  morbosum  uel  tignum  uitiosum  uel  seruum  fugitiuum  aut  uas  non  integrum  sciens  uendidit, tenetur ad omne interesse, si ignorans quanto minoris empturus esset, ut ff. de act. empti.  et uen. Iul [D 19.1.13] et l. Tenetur § Si uas [D 19.1.6], quod ad alias res et ad alia uitia generaliter  extendimus, ut quod in predictis exemplis dicitur ad cetera porrigatur, ut ff. de damp. infec. Dampni  § Qui uor. [D. 39.2.30.2]. Excipiuntur duo casus. Nam qui ignorans furem uendidit uel tributarium  predium, cum deberet uendere non furem, non tributarium, nec etiam in quanto minoris tenetur,  ut ff. de act. emp. et uen. Iul. § Item qui furem [D 19.1.13.1] et l. Si sterilis § i [D 19.1.21.1], quippe  debet quilibet suspicari et credere seruum furem esse. Nullus enim est, qui non quandoque uel  modicum domino furetur, quare etiam fur est, ut ff. de edilic. e. Quod si nolit § Si uenditor [D  21.1.31.1]. Propterea scriptum est ‘quid domini faciant audent cum talia fures’ id est serui. Item  quilibet debet scire secundum uetera iura quod nullum predium ytalichum tributarium est. Omne  autem prouinciale tributarium est. Ergo emptoris supina ignorantia in fure et tributario predio est  quare non prodesset ei, ut ff. de contrahen. empt. Et si consensu [D 18.1.15] et infra instit. de rerum  di. § Per traditionem [J Inst 2.1.40].” For reference see next note. 40  Gloss with siglum Jo. and Job. ad C 4.49.9 in BamSB MS Jur 21 fol 80vab, Munich BSB MS  Clm 22 fol 84rab and Paris BN MS lat 4534, fol 74va-b: “Alii dicunt quod in nullo casu tenetur  ignorans, nisi in aliquo predictorum et in alio, scilicet ff. de contrahen. emp. et uen. Si uestimenta  [D 18.1.45]. Regulare est ut non teneatur ignorans, set excipiuntur casus predicti. Et quod dicitur  de iure tributario predicta porrigunt generaliter. Potest etiam dici quod generale est, ut in omni  casu teneatur ignorans quanto minoris empturus esset. Quod ergo dicitur quod ignorans predium  tributarium uenditor non tenetur supplere ad omne interesse, sicut tenetur qui sciens, et sic intellige l. ff. de act. emp. et uen. Si sterilis § i [D 19.1.21.1]. Quod uero dicitur quod ignorans furem  uenditor non tenetur supplere adeo ut tenetur qui sciens uendidit fugitiuum. Illum enim habere  non licet, ut ff. de ac. emp. et uen. Iul. § Item qui furem [D 19.1.13.1] et ideo uenditor ignorans per  quanto minoris tenetur ad totum pretium, set non ad totum in fure sed in id quod minoris empturus  esset, ut ff. de tridua. ac. Illud § Si cuius [D 14.4.7.2] et ff. de v. si. Minus [D 50.16.32] et ff. de  fideius. Amis. in prin. [D 46.1.52pr] et C. de her. insti. Pater [C 6.24.2]. jo[annes]b[assianus].” This  opinion of Johannes Bassianus is confirmed in the dissensiones dominorum handed down under  the name of Hugolinus: see Haenel (ed), Dissensiones, Hugolinus § 410 (n 26) 530-531. 

190

the creation of the ius commune

Thus, Johannes Bassianus seems to have adopted as a general rule the view  that the seller who was aware of the defects was liable for the full loss of the  buyer, whereas the ignorant seller was only liable for the balance between what  the buyer paid and the lesser amount he would have paid if he had known of  the defect. The cases of D 19.1.13.1 (sale of a fur) and D 19.1.21.1 (sale of a  predium tributarium) either could be taken as exceptions to this general rule,  where the ignorant seller was not liable, or they could, through interpretation, still be brought into conformity with this general rule. Johannes Bassianus discussed a further problem: the difference between  the actio empti to claim reduction in price and the actio quanti minoris of the  aedilician  edict,  used  for  the  same  purpose.  These  two  actions  show  many  similarities and can be granted in similar situations.41 In a gloss to D 19.1.13,  Johannes Bassianus explained that the first was a civil action and the second  praetorian.42 But was not the second, in fact, redundant? The answer was no.  The praetorian action was aimed at the difference between the selling price  and the objective value or common price (estimatio communis). The civil action  was aimed at the difference between the selling price and the lesser amount he  would have paid. One might use the remedy which was the most advantageous.  Suppose that someone, who knew very well how to handle horses, bought for  forty a horse which moved backwards. It was more in his interest to use the  praetorian action, because had he known of the defect, he still would have paid  at least five for the horse, whereas the average buyer would not have given  anything. With the praetorian action he could claim back the full forty.43 (3) Azo The writings of the glossator Azo (d 1220) reveal that, by the time he was  lecturing  in  Bologna,  there  had  developed  a  view  that  sellers,  unaware  of  41 The aedilician edict imposed on the seller the duty to give certain warranties, but liability under  the actio quanti minoris does not seem to have been dependent upon those warranties. 42  In  the  Middle  Ages  the  actio quanti minoris  and  the  actio redhibitoria were  characterised  as  praetorian actions, although they were introduced not by the praetor but by the aediles curules.  For this reason the use of the term “praetorian” was in this respect criticised by the Humanist  jurists of the sixteenth century. The remedies did belong, however, to the ius honorarium and  in  the  Corpus iuris  the  terms  ius praetorium  and  ius honorarium were  more  or  less  used  as  synonyms: see D 1.1.7.1. 43  Gloss with siglum Job ad D 19.1.13, in Paris, BN MS lat 4450 fol 194v: “Pone ergo comuni estimacione est deterior equs in v. Ergo in v. competit illa pretoria, ac si scisset, minoris empturus esset  xl. Ergo pocius interest hic ciuili, nam per hoc consequitur quis quanto minoris esset empturus, ut  hic. Pone ergo e contrario emit equm retrogradum quis qui bene nouerat usum equorum. Istius  interest magis in pretoria quam ciuili, nam non fuerit  minoris quam v, si millesies  scisset. Si ergo agat pretoria, poterit consequi xl. totum. Enim estimacione communi hanc differentiam inuenit. Jo[hannes] b[assianus].”

the ignorant seller’s liability for latent defects

191

defects  in  their  merchandise,  were  liable  under  the  actio empti  for  the  balance between what the buyer paid and the lesser amount he would have  paid, but, at the same time, there was still the dissenting view, mentioned by  Johannes Bassianus but not linked to any glossator by name, that ignorant  sellers should not be held liable at all. The most important texts in the Corpus iuris holding the ignorant seller liable for defects were D 18.1.45 (garments  which had been repaired were sold as new), D 19.1.6.4 (sale of a leaky barrel),  D 19.1.13.1 (sale of a runaway slave), D 19.2.19.1 (lease of a leaky barrel) and  C 4.49.9 (sale of a plot of land which owed a higher capitatio than the seller had  stated). The most important texts stating that the ignorant seller was not liable  for defects were D 19.1.13.1 (sale of a thief) and D 19.1.21.1 (sale of a plot of  land for which tributum appeared to be owed). As  a  matter  of  fact,  the  case  of  C  4.49.9  closely  resembled  that  of  D  19.1.21.1. Yet they differed in the answer given to the legal problem, which  was explained by the fact that in C 4.49.9 the seller had provided the buyer  with incorrect information, whereas in D 19.1.21.1 the seller had given no  information  at  all.44  It  was  uncertain  whether  the  five  cases,  declaring  the  ignorant seller liable, should be considered as reflecting the principal rule  or rather the exception. The same could be asked about the two cases that  declared the ignorant seller not to be liable. Azo’s line of thinking can be found in his Apparatus maior to the Digestum Vetus as well as in his Lectura Codicis and Summa Codicis; he followed the  approach of his master Johannes Bassianus. Azo considered that the five cases  contained the general rule, applicable also to other cases of latent defects,  namely that the ignorant seller was liable for the balance between what the  buyer paid and the lesser amount he would have paid. In the fragments in the  Apparatus maior the term regula was not used (Azo stated that the ignorant  seller was always (semper) liable); but in the Summa Codicis, and likewise in  the Apparatus maior, the terms used were regulare esse and regula, though  it is clear that we are dealing with a general rule of law. Azo argued that it  made more sense to consider two cases to be exceptional, rather than five.45  44  See the gloss with siglum p. (but it cannot have been Placentinus since it refers to the opinion of  Azo) ad C 4.49.9 in Paris BN MS lat 16910 fol 84rb: “ff. e. l. i Si sterilis [D 19.1.21.1] contra …”  45  “Tenetur ignorans quanto minoris erat empturus emptor in capitatione predii, C. e. l. Si minor  [C  4.49.9]  et  in  uase  non  integro  uendito  et  in  uase  uitioso  locato,  supra  e.  Tenetur  §  Quod  [D. 19.1.6.4 i.f.] et infra locati Set addes § i [D 19.2.19.1]. Hic tamen melius et apertius dico  ignorantem pro non integro teneri ad interesse, ut supra e. Tenetur [D 19.1.6], pro uase uitioso  locato, idem pro uendito quanto minoris, in uestimentis, supra de contrahen. empt. Labeo [D  18.1.45], in seruo fugitiuo ut hic. Set hii v casus, ut dicunt, speciales sunt. In aliis enim ignorans  non  tenetur.  Set  nos  contra,  nam  semper  tenetur,  nisi  in  casibus  duobus,  ut  in  fure  et  predio  tributario, ut hic et infra e. Si ste. § Si pre. [D 19.1.21.1]. Cum ipsi dicunt v specialia sunt, multo 

192

the creation of the ius commune

Azo stated that there was a perfect explanation for the two exceptional cases:  every slave might be presumed to be a thief and every man should know that  tributum was owed for provincial land.46 In the text corresponding to this gloss in his Summa Codicis, Azo followed  Johannes Bassianus in quoting Virgil, who was this time explicitly recognised  as  the  author  of  the  quotation.47  Azo  also  dealt  with  the  question  whether  or not the aedilician actions were in fact redundant in his Apparatus maior  to the Digestum Vetus. According to Johannes Bassianus, the actio quanto minoris  in  particular  was  still  useful,  because  in  some  cases  this  remedy  allowed the plaintiff to claim a higher amount than would be gained by using  the  civil  action.  Azo  explained  that  the  aedilician  actions  were  not  superfluous, because it was not possible to use the civil actio empti for rescission  on the ground of corporeal defects, only for damages or reduction in price.  Moreover, the civil actio empti could not be used if one had bought a thief or  a predium tributarium. Therefore it was necessary to introduce the aedilician  actions, for they served a different purpose.48 magis nos duos dicemus.” For the reference, see the following note. 46 Gloss with siglum az. ad D. 19.1.13.1 in BamSB MS Jur 11 fol 223rb; Paris BN MS lat 4451 fol  137rb; Paris BN MS lat 4459 fol 179vb; BV MS Vat Lat 1408 fol 210rb; BV MS Borgh Lat 225  fol 164rab: “Et hoc ideo, quia scire debuit eum furem esse eo ipso quod seruus erat, maxime si  cautius erat. Dominis enim suis, et si nil aliud possunt, cinerem tamen subripiunt. Et fur est non  solum qui extraneo, set etiam qui domino suo furtum facit, ut de aedilic. e. Quod si nolit § i. [D  21.1.31.1]. Item, si predium erat provinciale, scire debet tributarium. Ideoque, quia est quod ei  imputetur, non agit contra ignorantem uel quod dicit lex non tenetur subaudi adeo non agitur  quanto  minoris  res  est,  non  quanto  minoris  empturus.  Nec  obstat  quod  dicitur  licere  contrahentibus se decipere, quia hoc in pretii quantitate non in uitiis rei. Verum est. az[o].” An almost  identical gloss with siglum az. ad C 4.49.9 can be found in Prague, Knihovna Národního Musea, XVII  A.10 fol 92va. A similar discussion (which cases make the rule?) can be found in anonymous gloss  ad D 19.1.13 in BV MS Vat Lat 1408 fol 210rab: “Infra e. t. Si sterilis § i [D. 19.1.21.1] assignatur  contrarium …” Cf also Azo, Lectura Codicis ad C 4.49.9 nos 1-4 in Azo, Lectura super Codicem  (1577; repr 1966) 348-349 and Azo, Summa Codicis ad C 4.49 no 17 (1559) fol 111rb. 47  Azo, Summa Codicis ad 4.49 no 17 (n 46) fol 111rb. 48  Gloss with siglum az. ad D 21.1.19.2 in BamSB MS Jur 11 fol 247vb; Paris BN MS lat 4451 fol  192rb;  Paris  BN  MS  lat  4461 fol  198ra;  BV  MS  Vat  Lat  1408 fol  235r;  BV  MS  Vat  Lat  2512 fol 198rab; BV MS Borgh Lat 225 fol 182rab: “Puta si uendidisti seruum fugitiuum uel pecus  morbosum  cuius  nomine  agi  potest  etiam  de  iure  ciuili  uel  ad  interesse  uel  quanto  minoris  empturus erat, ut supra de acti. emp. Iulianus in prin. [D 19.1.13pr]. Et nichilominus agi reddibitoria potest de iure pretorio uel quanto minoris uel ex contractu uel ex cautione res est. Et caue  tibi quia propter corporis morbos non agebat ut de iure ciuili ad redhibitionem, set ad interesse  uel quanto minoris, uel ex contractu uel ex cautione, ut in pecore morboso et seruo. Set propter  uitium animi in seruo fugitiuo agebatur de iure ciuili, non in seruo fure, nec predio tributario,  ut  notaui  supra  de  act.  emp.  Iulianus  [D  19.1.13].  Vnde  exponebatur  cautio  super  fugitiuo  et  quibusdam aliis quibus aduocari potest res. Item de sanitate de iure ciuili, ut supra de act. emp. l.  Ex empto § Etiam [D 19.1.11?] et C. de act. emp. Emptor [C 4.49.14]. Set sine cautione agebatur,  ut dixi, in fugitiuo, supra e. l. Quis sit § Idem recte [D 21.1.17.10]. Nec obstat quod dicitur supra  de acc. emp. Ex empto § Redhibitionem [D 19.1.11.3], ut ibidem notaui. Fuit ergo necessarium  has edilicias acciones proponi, utpote cum iste ad aliud dantur quam superiores de iure ciuili. 

the ignorant seller’s liability for latent defects

193

A more elaborated view can be found in Azo’s Summa Codicis. The actio empti could in some specific instances be used for rescission;49 but when the  actio empti was used for reduction in price, this could also sometimes result  in rescission, such as when the buyer would not have bought the merchandise at all, if he had known of the defect, as appeared from D 44.2.25.1.50 In  conformity  with  Azo’s  view  that  the  actio empti,  used  for  price  reduction,  was clearly different from the aedilician actio quanti minoris, the Apparatus major indicates that there was a difference between the civil action, directed  at the balance between what the buyer paid and the lesser amount he would  have paid, and the praetorian, directed at the difference between the selling  price and the common estimation.51 In at least one manuscript, this gloss continues by stating that there were  two actiones quanti minoris, one of civil law and one of praetorian law.52 This  is exactly what was written in Azo’s Summa Codicis. There was a civil actio quanti minoris, directed at the balance between what the plaintiff paid and  the lesser sum he would have paid if he had known of the defect. This remedy  was  perpetual,  but,  as  explained  above,  could  not  be  used  for  all  defects.  There was also a praetorian actio quanti minoris, directed at the difference  between  selling  price  and  actual  value.  This  remedy  could  be  brought  for  only  one  year,  and  could  also  be  brought  in  certain  cases  where  the  civil  action was not available. Both were necessary.53 Azo  thus  considered  that  the  civil  actio empti  used  for  price  reduction  Quid enim si emptor quidem empturus erat, set minori precio? Subuenitur ei per hoc edictum  uel ad redhibitium, uel quanto minoris erat res. Item quandoque dantur iste et non ille, ut supra  dixi. az[o].” 49  D 19.1.11.3. 50  Azo, Summa Codicis ad C 4.49 nos 21-23 (n 46) fol 111va. The distinction between the buyer who  would have bought, but for a lower price, and the buyer who would not have bought at all, may be  derived from Aristotle. See Aristotle, Ethica Nicomachea lib III, cap 1-4 [= Aristoteles Latinus,  ed G A Gauthier, vol XXVI 1-3 Fasc tertius] (1972) 179-183. 51  A gloss with siglum az. ad D 19.1.13, in BamSB MS Jur 11 fol 223rb; Paris BN MS lat 4451 fol  137rb; Paris BN MS lat 4459 fol 179vb; BV MS Vat Lat 1408 fol 210rb; BV MS Borgh Lat 225 fol  164rb: “Alia est accio ciuilis quanto minoris empturus esset, alia pretoria quanto minoris res ualet  ut infra de edil e. l. Quod si nolit § Si quis [D. 21.1.31.16] et de except. rei iudi. Si is qui § Est in  po. [D. 44.2.25.1]. az[o].”  52  See the gloss in BV Vat Lat 1408 fol 210rb: “Due sunt acciones quanti minoris …” This continuation of the gloss also maintains that the praetorian action can sometimes be used to claim a higher  amount. For these reasons the aedilician remedies are not superfluous. 53  Azo,  Summa Codicis  ad  C  4.58  no  3  (n  46)  fol  114ra:  “Actio  autem  quanto  minoris  empturus  esset emptor ciuilis est, ut ff. de act. empt. l. Iulianus [D. 19.1.13] et ideo perpetua est. Haec  autem quanto minoris res est, competit usque ad annum et cum ad aliud datur quam superior,  neutra est superflua. Quid enim si emptor quidem empturus erat sed minori pretio? Subuenitur  ei per hoc edictum, uel ad redhibendum, uel quanto minoris res est. Item esse quod dantur iste  scilicet redhibitoriae, uel quanto minoris res est et non datur illa ciuilis, quia illa in casibus tantum  competit, ut diximus supra de act. empt. ...” 

194

the creation of the ius commune

resembled  the  aedilician  actio quanti minoris  to  such  an  extent  that  he  presented the two as distinct actions, but with similar names. The question  however  arises  as  to  whether  he  actually  wanted  the  actio quanti  minoris civilis to be a separate remedy, distinct from the regular actio empti.  This  question  was  of  importance  for  procedural  law,  as  will  appear  in  the  next  section. This does not seem to have been the case. A Summa de Actionibus,  sometimes ascribed to Azo,54 considered that the terms quanti minoris and  redhibitoria did not indicate independent remedies, but were just additions  (adiectiones) to, for example, the actio empti.55 A student of Azo maintained  that, according to his master (dominus meus), the words quanti minoris were  just an addition and did not refer to a separate action.56 Azo’s own master,  Johannes Bassianus, had composed an Arbor actionum. This “tree of actions”  set out the remedies of the Corpus iuris, totalling 169, in a scheme of 180  circles with the praetorian actions at the left and the civil actions at the right.  Above each circle there were letters, placing an action into a certain category.  This  Arbor actionum,  however,  contained  only  one  actio quanti minoris,  which was described as praetorian and temporal.57 (4) The last generation of glossators According to Hugolinus (d after 1233), a student of Johannes Bassianus and  one of the last generation of glossators, the general rule should be based on  the five cases where the ignorant seller was liable and on the entire Digest  title on the aedilician edicts. In the exceptional cases where the ignorant seller  was not liable, the buyer could himself be blamed for the loss he suffered.  Or one should argue that, in the case of selling a thief or a tenement owing tributum, the seller was not liable under the actio quanti minoris civilis but  rather under the actio quanti minoris pretoria.58 Hugolinus used the term  54 W Litewski, Der römisch-kanonische Zivilproze nach den älteren ordines iudiciarii (1999) I 188.  55  Summa de Actionibus § 78, in BIMAE III 6. 56  Roffredus  Beneventanus,  Libelli iuris civilis  (1500;  repr  1968)  fol  101va.  See  also  Jacques  de  Revigny (under the name of Petrus de Bella Pertica), Lectura super prima parte Codicis ad C  4.58.2 (1519; repr 1967) fol 207rb: “Azo dicit quod quanto minoris et redibitoria non sunt nomina  actionis, sed sunt adiectio actionis.” 57  A Brinz, Arbor actionum (1854) 19; for the Arbor actionum of Johannes Bassianus, see also A  Errera, Arbor actionum, Genere letterario e forma di classificazione delle azioni nella dottrina dei glossatori (1995); and Lange, RRM I 221-224. 58  A gloss with siglum h. ad D 19.1.13 in London, British Library MS Royal 11 C III, fol 188ra and  Paris BN MS lat 4461 fol 178rb: “Generaliter dixerunt quidam, numquam ignorantem uenditorem  propter rei uitium teneri, sumpta generali regula propter istum casum de fure et alium, infra e.  Si steri. § Si predium [D 19.1.21.1] et speciale esse in v casibus, ut teneatur ignorans, scilicet  in pecore morboso, tigno uitioso, seruo fugitiuo, uestimentis interpolis et uase non integro, ut  hic et supra de contrahen. empt. Si uestimenta interpola [D 18.1.45] et supra e. Teneri § Si uas 

the ignorant seller’s liability for latent defects

195

regula here. There was also an extensive quaestio on D 19.1.13 by Jacobus  Balduini (d 1235), a contemporary of Hugolinus and student of Azo, which  also reflected his master’s teachings. Balduini used the term regulariter and  spoke about regula nostra.59 Accursius (ca 1182-1263) and Odofredus (d 1265) did not add anything  new to the doctrines developed by Johannes Bassianus and Azo, which were  generally accepted and confirmed by the last generation of glossators. In fact,  they were mere compilers and the importance of Accursius is primarily that,  through his ordinary gloss, which gained enormous authority in centuries to  come, more than 150 years of legal scholarship was compiled and transmitted  to future generations of students being educated in Roman law. Azo’s teachings were accepted on which cases should be adopted as general and which as  exceptional. They can be found in the gloss quanto minoris on D 19.1.13.1 as  well as in the Lectura on the Digestum vetus of Odofredus on D 19.1.13pr.60  Here we also find the traditional justification for the fact that the ignorant  seller was not liable when a thief or a predium tributarium was sold. Even  the  quotation  from  Virgil  is  found.  Only  in  one  respect  did  the  Accursian  gloss differ from the writings of Azo: it recognised, beyond any doubt, two  distinct remedies for price reduction – one civil and one praetorian. The most  important gloss for this position was that essem empturus on D 19.1.13pr.61  [D 19.1.6.4]. Nos totum contrarium dicimus, sumpta a pluribus regula, et fere ex toto illo titulo  de edilic. e. [D 22.1] et ratione specialitatis assignamus, ut non teneatur in illis duobus casibus,  quoniam emptori imputandum est. Et idem se queri debet quare non petiit sibi caueri furem non  esse uel non quesiuit utrum fur esset, cum satis potuit putare nullum seruum facile inueniri nisi  furem, saltem qui aliquod subripuit domino suo uel solitus est subripere, nam talis fur dicitur,  infra de ed. ed. Quod si nolit § Si uenditor [D 21.1.31.1]. Et etiam omnes fures appellantur, ut ibi  dicitur. Vnde Virgilius in bucolicis in illo loco dicit Damoeta cuium pecus ‘Quid facient domini  audent cum talia fures!’. Secundus est in predio tributario. Quid enim si predium erat provinciale? Nonne debuit putare illud tributarium? Vel aliter dic quod in istis duobus casibus tenetur,  sed non adeo ut in aliis. Nam in aliis tenetur ignorans quanto minoris fuisset empturus, set hic  quanto minoris res est. Et est quidam differentia inter has acciones, quoniam prima est ciuilis et  perpetua, secunda temporale et pretoria, ut infra de ed. e. Quod si nolit § Si quis [D 21.1.31.16]  et infra de excep. r. iud. Si is qui § Est in potestate [D 44.2.25.1] et C. e. Si minor [C 4.49.9] et  infra de edil. Quod si nolit § Si plures [D 21.1.31.5] et l. Bouem § Aliquando [D 21.1.43.6] et l.  Hominem [D 21.1.47] et l. Sciendum § fi. [D 21.1.19.6]. h[ugolinus].” 59  Paris BN MS lat 4458 fol 171ra. 60  Odofredus, In secundam Digesti Veteris partem Praelectiones ad D 19.1.13pr (1552; repr 1968)  fol 109vb. 61  “essem empturus Nota hic differentiam inter actionem quanto minoris et ciuilem et praetoriam.  Nam  in  ciuili  agitur  quanto  minoris  esset  empturus,  si  scisset,  ut  hic.  Sed  in  praetoria  quanto  minoris ualuit tempore contractus propter uicium, ut infra de edil. edic. l. Quod si nolit § Si plures  [D 21.1.31.5]. Sed aliis, ut R[ogerio], haec differentia non placet, ut arg. infra e. l. § Per contrarium in fi. [D 19.1.13.5], ubi etiam ciuili agitur quanto minoris ualet. Sed ibi non ea ex empto, siue  quanto minoris, que propter uicium detur, agitur. Sed propter dolum emptoris, ubicumque tamen  dicitur quanto minoris pretoria teneri, ibidem habet locum redhibitoria, si non esset empturus,  si uicium sciuisset, ut infra de excep. rei iu. l. Si is qui § i. [D 44.2.25.1].” Cf also the gloss quanto

196

the creation of the ius commune

In this gloss Accursius described the difference between, on the one hand,  claiming quanto minoris esset empturus, si scisset with the civil action and, on  the other, claiming quanto minoris ualet with the praetorian action. Rogerius  and others were said to have disliked this distinction. Yet, as we saw above,  the distinction Rogerius disliked was the one between the two kinds of assessment of price reduction. He preferred to use the objective standard of the  aedilician  actio quanti minoris,  namely  the  difference  between  the  selling  price and the value of the object at the moment the contract was concluded  (quanto minoris ualet), and, in spite of what can be read in D 19.1.13, he  wanted to use this criterion when the actio empti was used for price reduction.  We  should  consider  it  to  be  improbable,  if  not  impossible,  that  the  difference  Rogerius  disliked  would  have  been  the  distinction  between  the  actio quanti minoris civilis and the actio quanti minoris praetoria, since these  concepts had not yet been developed in his day.62 Odofredus did not follow Accursius’ opinion that there were two separate  actiones quanti minoris, civil and praetorian. In his view, the words quanti minoris empturus essem si scissem did not establish an independent remedy,  but simply were a different name (cognomen) for, or an addition (adiectio)  to, the contractual actio empti. The phrasing quanti minoris res valet quam  empta indicated the aedilician action for price reduction which was available  during one year.63 (5) Conclusions For more than a century and a half the glossators had produced opinions on  the use of the actio empti for latent defects against an ignorant seller. Gradually some general concepts and rules of law had developed from the casuistry  found in the Corpus iuris. (a) Actio empti; actio quanti minoris The seller, aware of defects, was held liable for the buyer’s full interest. When  the seller was not aware, the buyer could use the actio empti to claim quanti minoris empturus esset, si scisset.  In  such  a  case,  the  seller  was  liable  for  the balance between what the buyer paid and the lesser sum he would have  paid,  had  he  known  of  the  defects,  a  claim  which  was  not  restricted  to  a  minoris ad D 19.1.13.1, the gloss homo sit ad D 21.1.31.5 and the gloss minus daret precii ad C  4.49.9. 62  Unfortunately the literature has missed this point and sometimes suggested the text of the gloss to  be corrupt. See P van Warmelo, Vrywaring teen gebreke by koop in Suid-Afrika (1941) 61 and 65. 63 Odofredus, Praelectiones ad D 19.1.13pr (n 60) fol 109va. 

the ignorant seller’s liability for latent defects

197

certain period of time. Accursius no longer regarded this action as the regular  contractual action for sale, and provided it with a name of its own, the actio quanti minoris civilis. Only in two exceptional cases could the ignorant seller  not be held liable under the actio empti: when he had sold either a thief or  a fundum tributarium. In these cases the buyer had to blame himself for his  losses. He should have realised that all slaves were thiefs and that tributum  was owed for all provincial land. In these cases, however, it was still possible  to use the aedilician action. According to the Digest, the actio empti could also be used for rescission.64  In contrast to what happened to the texts in D 19.1, de actionibus empti et venditi, where the actio empti was used to achieve reduction in price, the  few scattered texts that allowed the actio empti to be used for rescission were  not developed into a general rule of law. The medieval jurists considered it  possible in only a few exceptional cases to use the actio empti for rescission,  such as when the seller was aware of the defects, as in the case of D 19.1.11.5.65  The principal rule was that the actio empti could not be used for rescission  on  the  ground  of  latent  defects.66  It  is  questionable,  though,  whether  this  restricted applicability of the actio empti was seen as a problem. Since the  plaintiff’s actio empti, used for price reduction, was directed at quanti minoris empturus esset si scisset, he would surely be indemnified, and where he would  not have bought the goods at all had he known of the defect, the use of the  actio empti for price reduction would in any case result in rescission.67 (b) Praetorian remedies Since the actio empti could only be used for rescission in some exceptional  cases, the praetorian actio redhibitoria, which had to be brought within half  a year, was by no means superfluous. The praetorian remedy for price reduction, however, which had to be brought within one year, had lost much of its  significance. The glossators saw, as one of the few reasons that could justify  the latter’s existence as a separate remedy, the fact that quanti minoris valet,  based on the pretium commune, sometimes covers more than quanti minoris empturus esset si scisset,  based  on  the  pretium singulare.  Moreover,  the  praetorian action could be used where the civil action was not available, such  as against the seller of a thief or predium tributarium. 64 65 66 67

D 19.1.11.3. Azo, Summa Codicis ad C 4.49 no 21 (n 46) fol 111va. The gloss contineri ad D 19.1.11.3 and the gloss conueniri ad D 21.1.19.2. See Azo, Summa Codicis ad C 4.49 no 21 (n 46) fol 111va; Roffredus, Libelli (n 56) fol 101rb, and  the gloss essem empturus ad D 19.1.13pr.

198

the creation of the ius commune

(c) The view of Accursius As  stated  above,  Accursius  was  probably  the  first  to  acknowledge  the  existence of two separate remedies quanti minoris aimed at price reduction  due to defects, but this stand was not undisputed. According to Odofredus,  one  of  these  actions  is  the  regular  contractual  action  of  sale.  The  words  quanti minoris,  as,  for  example,  added  to  the  remedy  by  Azo,  are  just  an  addition, which indicates that the buyer is using the contractual actio empti  for  a  specific  purpose,  namely  to  achieve  a  reduction  in  price.  Moreover,  Accursius’ opinion seems to have been rejected by the compilers of the Siete Partidas,  who  codified  the  Roman  law  of  sale  for  the  kingdom  of  Castile  at  approximately  the  same  time  as  or  just  after  Accursius  was  teaching  in  Bologna and composing his ordinary gloss. The Partidas adopted a remedy  for price reduction only for the difference between the selling price and the  actual price, thus taking the pretium commune as a premise: “tanta parte del  precio, quanto fallassen en verdad, que valia menos por razon de la tacha, o  de la enfermedad que era en ella.”68 D. FROM THE GLOSSA ORDINARIA TO THE FOURTEENTH CENTURY From  the  twelfth  century  onwards,  the  remedies  of  Roman  law  –  or  at  least their names – were used in drawing up the statement of claim or libel  (libellus), which means that the law of procedure, in particular of Romanocanonical procedure, is important here. Thus, we have to consider what were  the consequences of Accursius’ teaching that there were two actiones quanti minoris: a civil and a praetorian. But  there  were  further  developments  in  legal  scholarship  after  the  era  of the glossators. Since Accursius’ ordinary gloss gained enormous authority  during the Middle Ages it was impossible to teach the Corpus iuris without  taking it into account. This did not mean, however, that doctrinal developments had come to an end. From about 1235 a law school developed at Orleans  in France that soon became an important centre for legal education, mainly  in Roman law. Its great scholars, who taught at the end of the thirteenth and  68  Part 5.5.65 (“so much of the price, as much as it was mistaken in truth, as it is worth less by reason  of the defect or disease”). On the one hand, the gloss que valia menos on Part 5.5.64 by Gregorio  López de Tovar (ca 1496-1560) states that this provision derives from D 19.1.13pr (pretium singulare), but the gloss fallassen en verdad on Part 5.5.65 rejects the opinion that there is a civil and a  praetorian actio quanti minoris. The gloss que valia menos on Part 5.5.64 states that the Partidas  incline to the later opinion of Pierre de Belleperche and Cinus of Pistoia that there is only one  actio quanti minoris.

the ignorant seller’s liability for latent defects

199

the beginning of the fourteenth centuries, no longer produced glosses, but  rather commentaries in the form of written elaborations of lectures given on  the Digest and Code (Lecturae). Their ideas, sometimes deviating from those  of Accursius, were to be important for the doctrines of the Italian and French  commentators of the later fourteenth and fifteenth centuries. (1) Romano-canonical procedure In Romano-canonical procedural law, the earliest ordines iudiciarii required  that  the  plaintiff’s  demand  was  formulated  in  writing,  in  what  was  called the  libellus  (conventionalis),  namely  the  statement  of  the  claim  or  libel.  According to some jurists, this libel had to contain the name of the action  (nomen actionis); according to others, it had only to contain the reason for  the  claim  (causa petendi).  Nevertheless,  the  name  of  the  action  had  to  be  announced to the magistrate at a later stage in the litigation. The wording  of the libel could be decisive especially when the facts of the case (sale of a  defective object) gave rise to various remedies (actio empti and actio quanti minoris) which had the same end: price reduction. Was the use of the remedy  set out in the libel restricted to a certain period of time? At what exactly was  the claim directed? What had to be proved by the plaintiff? To answer these  questions  it  was  important  to  know  whether  the  claim  for  price  reduction  was based on the civil actio empti or on the praetorian actio quanti minoris. If  the  remedy  brought  for  price  reduction  because  of  defects  was  the  actio empti (as found in D 19.1.13pr), its function was to claim the balance  between what the plaintiff paid and the lesser sum he would have paid had  he been aware of the defect. This action was thus based on the pretium singulare. It was perpetual, so that its use was not restricted to a certain period  of time and it only prescribed after thirty years. There were, however, two  instances in which it could not be used: sale of a thief or sale of land owing  tributum. If the remedy brought for price reduction because of defects was  the praetorian actio quanti minoris – that is, one of the aedilician edicts – the  aim was to claim the difference between the selling price and the actual value  at the time the sale was concluded. This action was thus based on the pretium commune. It was temporal or annual.69 It had to be used within one year; but  it was available for the sale of a thief or of a predium tributarium. As we have seen, Azo introduced the term actio quanto minoris civilis; but  he taught that the words quanto minoris, as used here, were just an addition  (adiectio) to the regular actio empti. Furthermore, Odofredus was of the view  69  See J Inst 4.12pr.

200

the creation of the ius commune

that the term quanto minoris essem empturus was only a different designation (cognomen) for the actio ex empto civilis. The fact, though, that Accursius in the gloss essem empturus ad D 19.1.13pr presented the actio quanto minoris civilis as a separate and independent remedy, distinct from the aedilician actio quanti minoris praetoria and distinct from the regular actio empti,  prompted the jurists who drafted the statements of claims for civil litigation  according to Romano-canonical procedure to formulate a separate libel for  this remedy, which in earlier times had been considered nothing other than  the regular actio empti. This appears to have become the common practice in medieval works on  procedural law. Probably Roffredus Beneventanus (ca 1170-after 1244), in  his famous treatise on the statement of claims, the Libelli iuris civilis, was  the first to present separate formulas for the libel of the actio quanti minoris praetoria and that of the actio quanti minoris civilis. The first is directed at  quanto minoris res est; the second, which according to Roffredus was missing  in the Arbor actionum of Johannes Bassianus is directed at quanto minoris empturus essem.70  Many  more  treatises  on  procedural  law  were  to  follow  the example of Roffredus.71 The most important of these was the Speculum iudiciale  of  the  canonist  Guillaume  Durand  (ca  1230/31-1296).  This  work  was  a  perceptive  compilation  of  procedural  rules,  intended  both  for  legal  practitioners as well as litigants. It was widely used during the Middle Ages  and gave separate libels for the actio quanti minoris praetoria (ideo illa x peto quod a illa tanto minoris est) and for the actio quanti minoris civilis (propter quod uitium si ego sciuissem, dedissem duo minus; unde ago contra eum, ut illa duo mihi praestet).72 The Aurea practica libellorum (1311-1329) of Petrus  Jacobi  (Pierre  Jame)  stated  that  the  phrasing  quanti minoris emptor esset empturus was an addition to the civil actio empti;73 but the Aurea practica  nevertheless contained a separate statement of claim for the remedy provided  with such an addition. In the libel of this remedy, termed the actio ex empto quanto minoris civilis, which only prescribed after thirty years, the plaintiff  claimed “as much as he would have paid less for the horse aforesaid, had he  known it was inclined to kick with the heels, namely that the aforesaid 40 

70 The two libelli can be found in Roffredus, Libelli (n 56) fols 17ra & 101vb, with the reference to  the Arbor actionum on fol 101rb.  71  Cf the one by Jean de Blanot (d 1281): Joannes de Blanasco, In titulum de actionibus in institutis commentaria, in De actionibus tractatus (1596) I fols 222-285 at fol 259. 72  Wilhelmus  Duranti,  Speculum Iuris  lib  IV  partic  3  (De rescindendis uenditionibus)  §§  4  &  6  (1574; repr 1975) II 250. 73  Petrus Iacobi Aurelianensis, Aurea practica libellorum rubr 86 no 4 (1575) 353.

the ignorant seller’s liability for latent defects

201

Tours shillings be restored”.74 Next to this remedy there was the one termed  the actio ex empto quanto minoris praetoria, which had to be brought within  one year. In the libel for this remedy the plaintiff claimed “as much as the  aforesaid horse lacked in value in view of its defect at the time of the contract,  which is 60 Tours shillings”.75 (2) The school of Orleans and Cinus The jurists from the school of Orleans, followed by Cinus of Pistoia, rejected  the  existence  of  two  different  actiones quanti minoris,  one  civil  and  one  praetorian. This rejection can be found, for example, in the Lectura on the  Digestum Vetus  of  Jacques  de  Revigny  (d  1296).  Revigny  referred  to  the  opinion of Rogerius, as recorded in the gloss essem empturus ad D 19.1.13pr;  the latter did not supplement the civil action with a praetorian action and did  not distinguish between two ways of estimating reduction in price. Revigny  subsequently maintained that, by using the civil action, one could claim as  much as the merchandise lacked in value at the time at which the contract  was entered into, thereby referring to D 19.1.13.5. Rogerius did not take into  account what the buyer would have been willing to pay. Quanti minoris was  an addition (adiectio) to an action, not a separate action in itself. This was,  of course, the opinion of Azo and Odofredus. Were it otherwise, it could be  argued that, if the seller was deceived in a similar way, he would have the  actio quanti pluris. Thus, rather than two actions, there was only one action,  based on the contract of sale, with the addition of quanto minoris, which can  be brought within the first year. In two respects this use of the actio empti by Revigny for price reduction  is  extraordinary.  The  action  was  annual  in  duration  rather  than  perpetual,  and  was  based  on  the  pretium commune  instead  of  the  pretium singulare.  Revigny provided one further argument to support his opinion. In C 4.58.2  the emperor declared – Revigny spoke about the jurist, but it was in fact the  emperor,  either  Gordian  or  Justinian  himself  –  that  he  could  not  think  of  grounds why, more than one year after the sale had taken place, it should still  be possible to sue the seller, when the slave he had sold had run away. But  the emperor did not say that another action was available. It could be argued  that civil actions were perpetual and that was indeed the principal rule; but  this case was different. The civil action for price reduction came to an end  74 Iacobi, Aurea practica rubr 86 no 1 (n 73) 353: “quanto minoris fuisset empturus dictum equum,  si eum sciuisset calcitrosum, scilicet quod reddat sibi praedictos xl solid. Turon.”  75  Iacobi, Aurea practica rubr 88 no 1 (n 73) 360-361: “quanto minoris ualebat dictus equus, propter  illud uitium, tempore contractus, quod est in lx. soli. Turo.”

202

the creation of the ius commune

after one year. There was indeed an action based on sale with the addition  quanto minoris that could be used over a longer period; but this action was  intended to achieve rescission, that is, rescission of the contract.76 The last  lines of the fragment are difficult to read and contain two corrupt allegations.  One of these, the text of D 44.7.35 (the lex In honorariis), is important for a  further argument in support of the the view that there was only one action  for price reduction. It was to be explained more fully by later generations of  learned jurists, as we shall see below. The same opinion as that recorded in the Lectura on the Digestum Vetus,  can be found in Revigny’s Lectura Codicis, which was edited in the sixteenth  century under the name of Petrus de Bella Perthica. Here a further argument  was given as to why it was unnecessary to distinguish between two ways of  calculating  the  quanti minoris.  The  difference  between  the  selling  price  and  the  actual  value  of  the  merchandise  at  the  moment  the  contract  was  entered into, may be presumed to coincide with the balance between what  the  buyer  paid  and  the  lesser  sum  he  would  have  paid.  This  would  follow  from D 19.1.13.4.77 Pierre de Belleperche (d 1308), another jurist of the school of Orleans,  defended an opinion similar to Revigny’s. In two fragments in his Lectura Codicis, Belleperche rejected the idea that there were two different remedies  for claiming price reduction. There was only one actio quanto minoris, which  could be brought within one year to claim the difference between the selling  price and the pretium commune.78 Another fragment in the Lectura refers  76 Jacques  de  Revigny,  Lectura Digesti Veteris  ad  D  19.1.13,  in  Leiden,  MS  d’Ablaing  2  fol  249vb: “Dominus Rogerius non suppleat istam directam, quia queret inter directam et pretoriam,  quia una fit quanto minoris esset empturus, alia quantum res ualuisset tempore contractus, quia  ego reperio de ciuili quanto minoris que competit ad quanti res ualuisset tempore contractus,  infra e. l. Per contrarium [D. 19.1.13.5]. Vnde non considerit ipse uoluntatem quanto minoris  esset empturus,  quanti minoris res ualet, infra e. § Per annum. Et ego dico quod quanto  minoris non est nomen actionis set adiectio. Aliter etiam dicimus quod uenditor eodem modo  deceptus  haberet  quanto  pluris.  Non  etiam dico  quod  sint  due  actiones,  immo  dico  quod  illa  quam ipsi dicunt quanto minoris pretoriam, est adiectio actionis. Non dico quod una sit pretoria,  alia  ciuilis,  immo  semper  competit  actio  ex  empto  cum  hac  adiectione  quanto  minoris  primo  anno. Et dicit iurisconsultus quod non animaduerto quod possit, ut C. de edilic e. l. ii [C 4.58.2],  set non diceret sic aliqua competeret. Et si dicas quod ciuiles actiones sunt perpetue, uerum est  regulariter, sed non hic. Et est ratio, quia ista actio ex empto cum adiectione quanti minoris est  redibitoria ad solucionem contractus et eius iura; uerior arg. infra de ac. et ob. l. In similibus [D  44.7.55]. Sic ergo dico quod non fuit quanti pluris, set adiectiones unius actionis; sic seruanda  retractauit pretor, arg. l. In honoriam [D 44.7.35].” 77  Petrus de Bella Perthica [Jacques de Revigny], Lectura super prima parte Codicis ad C 4.58.2  (1519; repr 1967) fol 207rb. 78  Pierre de Belleperche, Lectura Codicis ad C. 4.49.9, in Florence, BML MS Plut 6 Sin 6 fol 208ra;  Cambridge, Peterhouse MS 34 (unfoliated): “Dicit glossa, scire debetis, est quanto minoris, ut ff.  de edic. e. l. Quod si nolit § Si plures [D 21.1.31.5] et est effectus, quod pretoria est annalis. Est 

the ignorant seller’s liability for latent defects

203

to the gloss congredi on C 4.58.2, the case where the buyer wanted to sue  the seller because the slave whom he had bought more than one year before  had run away. As just stated, according to C 4.58.2, the emperor could not  think of any remedy for price reduction after one year; but, according to the  gloss congredi, the buyer would still have the civil actio quanti minoris at his  disposal.79 Belleperche rejected this opinion. The only possible remedy was  that based on D 19.1.13pr. In the view of Belleperche, however, this remedy  was not a civil one. If the emperor said in C 4.58.2 that he could not think  of  any  remedy  which  could  be  used  after  one  year  for  reduction  of  price,  then nobody could think of such a remedy. Therefore, prescribing various  methods  of  estimating  the  quanto minoris  would  be  “applying  a  provision  to  words”,  which  is  not  allowed.  Prices  should  be  established  through  an  objective  estimation.  For  this  opinion  Belleperche  referred  to  D  35.2.63,  which stated that estimating the value of things was not done on the basis of  individual  affection  or  interest.80  Thus,  while  Revigny  maintained  that  the  only possible action for price reduction was civil and that, unlike other civil  actions, it had to be brought within one year, Belleperche taught that there  was only a praetorian action which, as with all praetorian actions, was timelimited. Both jurists agreed in rejecting various ways of estimating the quanti minoris.  The  ignorant  seller  was  always  liable  for  an  objective  estimation  of  the  difference  in  price,  based  on  the  pretium commune,  as  had  already  been Rogerius’ opinion in the twelfth century. The innovation of the jurists  of Orleans was always to restrict to the period of one year the possibility of  claiming price reduction because of defects. Cinus  of  Pistoia  (1270-1336)  adopted  the  opinion  of  Pierre  de  Bellealia quanto minoris ciuilis et est perpetua, ut Inst. de perpe. act. in prin. [J Inst 4.12.1]. Credo  quod non sit nisi una actio quanto minoris que usque ad annum competit tantum, ut infra de edil.  act. l. ii in prin. [C 4.58.2], cum idem sit quanto minoris et quanto minoris, ut ff. ad l. fal. Precia  [D 35.2.63].”  79  The gloss congredi on C 4.58.2: “redhibitoria uel quanto minoris ex empto uero, ciuili cum sit  pertpetua …”  80  Pierre de Belleperche, Lectura Codicis ad C 4.58.2, in Florence, BML MS Plut 6 Sin 6 fol 212ra:  Cambridge, Peterhouse MS 34 (unfoliated): “Dico glossa non probat per legem. Breviter credo  quod non est reperire nisi unam accionem quanto minoris. Per actionem ex eo contractu agitur  quanto minoris, ut supra allegata Iul. [D. 19.1.13pr]. Sed non reperio id est de ciuili. Probo hoc  per legem istam. Imperator dicit non animaduerto quo remedio etc. Nullus potest animaduertere si princeps non potest, ut infra de test. l. Omnium [C 6.23.19]. Preterea hoc esset imponere  legem uerbis, et non debas, quanto minoris erat res uel quanto minoris erat empturus, set pretia  rerum estimacione attendi debent, ideo etc. arg. ff. ad l. falci. l. Pretia [D 35.2.63] et ff. ad l. ac. Si  seruum meum [D 9.2.33]. Per legem que de accione quanto minoris pretoria loquet, habet locum  quanto minoris empturus esset, sicut lege allegata Iul. [D 19.1.13] loquitur. Tunc glossam non  credo ueram licet in iudiciis seruare forte redibitoria proprie vi. mensibus durat. Est alia inpropria  redibitoria ut infra e. Si predium [C 4.58.4].” 

204

the creation of the ius commune

perche in his commentary on the Code. Cinus was an early commentator  who  often  referred  to  the  teachings  of  Belleperche.  He  described  the  emperor who had spoken in C 4.58.2 as “full of Jurisprudence and having  the  entire  law  in  his  mind”  (legalis philosophiae plenus et qui omnia iura in pectore suo habet).  Moreover,  in  his  opinion  there  was  no  difference  between  the  two  ways  of  estimating  the  quanti minoris.  To  allow  various  methods of estimating the quanto minoris, as the Accursian gloss did, would  be  “applying  a  provision  to  words”,  which,  according  to  the  last  line  of  C  6.43.2, was not allowed. Cinus presented one further and rather scholastic  argument  on  why  the  gloss  was  wrong.  According  to  the  last  line  of  D  19.1.13.14,  no  actio empti  was  possible  against  the  ignorant  seller,  and,  according to D 19.1.13.1, the ignorant seller was liable for quanti minoris empturus esset.  Thus,  the  praetorian  action  was  also  aimed  at  quanto minoris empturus esset. Cinus  concluded that the gloss was mistaken, no  matter  what  the  legal  practitioners  who  persisted  in  their  deeply  rooted  errors (advocati radicatis erroribus insistentes) might say.81 E. THE ERA OF THE COMMENTATORS The scholars belonging to the era of the commentators, living and working in  Italy and France, no longer produced glosses, but continuous commentaries  on  the  texts  of  the  Corpus iuris,  sometimes  lex  by  lex,  sometimes  on  only  a selection of titles or provisions. Moreover, in this period we can trace an  early reception of the learned law into legal practice. Although possible influence on the statutes of Italian cities and the extensive practice of producing  consilia fall beyond the scope of this chapter, some attention will be paid to  one of Baldus’ consilia, dealing with latent defects, since this consilium has  already featured in the secondary literature. The  commentators  frequently  acknowledged  the  rules  concerning  the  liability of the ignorant seller for latent defects as developed during the previous  centuries.  There  were,  however,  some  new  developments  concerning  the  ratio between the two kinds of actio quanti minoris and the question whether  one  or  the  other  is  superfluous.  The  jurists  were  confronted  with  a  difference of opinion concerning the very existence of two remedies. The gloss had  ruled that there was a civil remedy for quanto minoris esset empturus and a  praetorian remedy for quanto minoris res ualet, but according to Pierre de  Belleperche, followed by Cinus of Pistoia, there was only one action for price  81  Cynus Pistoriensis, In Codicem et aliquot titulos primi Pandectarum tomi, id est, Digesti Veteris doctissima commentaria ad C 4.58.2 no 3 (1578; repr 1963) I fol 272rb.

the ignorant seller’s liability for latent defects

205

reduction, namely the praetorian one, although aimed at quanto minoris res ualet. (1) Liability of the ignorant seller in case of latent defects The gloss quanto minoris ad D 19.1.13.1 confirmed the general rule, developed  during  the  era  of  the  glossators,  that  the  ignorant  seller  is  liable  for  reduction in price. We will return below to the rather different question of  the number of actions available: only one or two? In the writings of the glossators it was not always explicitly stated that the  rule just mentioned was restricted to cases where the ignorant seller had not  given any warranties. Most of the time, however, this was beyond dispute. It  simply followed from the texts adduced as arguments. In the majority of those  texts, mainly cases from Digest title D 19.1, de actionibus empti et venditi,  there was no mention of any warranties. The writings of the commentators  were  sometimes  more  specific.  Paulus  de  Castro  (ca  1360/62-1441),  who  lectured for some time at Avignon and later in various places in Italy, stated  that, in case of warranties, the seller was liable for the buyer’s entire interest.  Only when no warranties were given, should a distinction be drawn between  the seller who was aware of the defects and the ignorant seller. The latter  was  generally  liable  for  price  reduction.82  The  authors  who,  following  the  gloss,  assumed  that  there  were  two  actiones quanti minoris  acknowledged  that there were three situations where the rule that the buyer, by using the  civil action, could claim quanti minoris empturus fuisset did not apply. For  example, Bartolus de Saxoferrato (1313/14-1357), a student of Cinus, who  studied and later taught in Perugia, noted that, in the case of selling or leasing  out leaky barrels, the seller or lessor would not be liable for quanti minoris  but for the other party’s full interest. He further noted that in two cases the  ignorant seller was not liable at all: when he sold a thief or a predium tributarium. He justified the latter two exceptions on the same grounds as had the  glossators mentioned above: everyone should know that slaves are thieves,  and everyone should know that for provincial tenements tributum has to be  paid.83 It seems that there was not much dispute about the applicability of the civil  actio quanti minoris. Bartolus’ view can be found in the works, for example,  of Bartholomeus de Saliceto (d 1412), who lectured at Bologna, Padua and  82  Paulus Castrensis, In primam Codicis partem commentaria ad C 4.49.9 (1582) fol 232va.  83  Bartolus  de  Saxoferrato,  Digestum Vetus in primum totum Pandectarum commentaria,  ed   Jacobus Concenatius, ad D 19.1.13 no 5 (1562; repr 2007) fols 774vb-775ra.

206

the creation of the ius commune

Ferrara,84  and  of  Paulus  de  Castro.85  Saliceto  held  that  there  was  also  an  existing opinion – not the prevailing one – to the effect that ignorant sellers  could not be held liable at all, and that the buyer, when using the civil action,  thus claiming quanti minoris fuisset empturus, might say that he would not  have bought the object at all, had he known of the defect (with reference to  D 14.4.7.2). The latter was the prevailing opinion, which had already been  acknowledged by the gloss essem empturus to D 19.1.13pr.86 The commentators also noticed the similarities and differences between  D 19.1.21.1 and C 4.49.9. In both these cases the defect existed in the fact  that it appeared that either property tax was owed for the tenement which the  buyer had purchased (tributum D 19.1.21.1) or more tax than the buyer had  expected (capitatio in C 4.49.9). But in D 19.1.21.1 the buyer was refused the  civil action for price reduction, whereas in C 4.49.9 this action was granted.  As seen above, this difference was supposed to result from the fact that in D  19.1.21.1 the seller had given no information at all, whereas in C 4.49.9 he  had provided the buyer with incorrect information. Following the glossators,  some commentators, such as Angelus de Ubaldis (1328-1407), a brother of  the better-known Bal dus de Ubaldis, also explained the difference between  the texts in this way.87 (2) The scope of the Aedilician actio quanti minoris The existence in the Corpus iuris of two different remedies for price reduction  in  the  case  of  the  sale  of  defective  goods  gave  rise  to  a  number  of  questions.  This  inconsistency  compelled  the  glossators  to  think  of  reasons  why one of the two was not superfluous. It was not difficult to think of cases  where  the  actio empti  for  quanti minoris  empturus esset,  which  Accursius  had considered to be a distinct action, offered the best opportunities. This  claim, which took into account the subjective interest of the buyer (pretium singulare),  could  exceed  the  claim  based  on  the  objective  value  (pretium commune) and even result in rescission if the buyer maintained that he would  not have bought at all if he had known of the defect. In some instances the  84 Bartholomaeus à Salyceto, Commentaria in Digestum Vetus accuratissima ad D 19.1.13pr no 3  (1541) fol 113va.  85  Paulus Castrensis, In secundam Digesti Veteris partem commentaria ad D 19.1.13pr no 2 (1582)  fol 121ra.  86  See also Azo, Summa Codicis ad C 4.49 no 21 (n 46) fol 111va.  87  A different explanation could be that the taxes owed in D 19.1.21.1 were outstanding, and those  in the case of C 4.49.9 concerned the future. In the case of outstanding payments, it was important whether the seller was aware of their existence or not. See Angelus de Perusio, Super codice nuper reuisa cum multis adiunctis, uidelicet repetitio ad C 4.49.9 (1545) fol 102vab.

the ignorant seller’s liability for latent defects

207

civil actio empti was the only option for recovering part of the selling price.  In certain cases the praetorian actio quanti minoris could not be used: the  sale of a slave with mental defects;88 the sale of a slave-woman who was no  virgin;89 and the emptio simplaria.90 In the Middle Ages, the emptio simplaria  was  understood  as  the  contract  where  parties  had  agreed  not  to  give  the  warranties mentioned in the aedilician edicts,91 or as the sale of simple things  which needed no sustenance (alimentum vitale),92 or as the sale of incorporeal  things,  such  as  an  inheritance.93  Moreover,  the  praetorian  action  was  annual in duration, the civil one perpetual. It was more difficult to see what  role the additional action could play next to the civil actio empti. The glossators nevertheless found two reasons to justify the existence of  the aedilician actio quanti minoris. It could be used in the exceptional cases  where the actio empti was not available – the sale of a thief and the sale of  a  predium tributarium.  Secondly,  the  claim  directed  at  quanti minoris res  could exceed the claim quanti minoris empturus esset, because, for instance,  what  an  average  person  would  have  given  for  a  defective  horse  (pretium commune) was less than what a specific buyer – perhaps someone capable of  handling such horses – was prepared to pay (pretium singulare). For the rest,  the glossators considered that both remedies applied to the sale of practically  the same defective things. This  opinion  was  not  followed  by  the  commentators.  According  to  Bartolus, the praetorian actio quanti minoris was only given for slaves and  animals. That was what D 21.1.1 and D 21.1.38 dealt with. D 19.1.13, on the  other hand, speaks about the sale of an unsound beam. As a consequence the  latter text could not be dealing with the aedilician action. Here, the civil-law  actio quanti minoris  was  granted.  It  had  a  wider  application  and  could  be  used against the ignorant seller of any kind of defective goods.94 Since the  aedilician edicts, as a result of Justinianic interpolations in D 21.1.1pr and D  21.1.63, were no longer restricted to the sale of slaves and certain animals in  the Corpus iuris, what prompted Bartolus to defend such a position? How  could he maintain that they were? It is quite unlikely that Bartolus wanted to  interpret the Corpus iuris in conformity with classical law. As  a  matter  of  fact,  the  commentators  could  see  that  the  titles  on  the  88 89 90 91 92 93 94

D 21.1.4. D 19.1.11.5. D 21.1.48.8. Vacarius, Liber Pauperum lib 4 tit 56 (n 37) 164. See Azo, Summa Codicis ad C 4.58 no 6 (n 46) fol 114vb.  See the gloss simplarium ad D 21.1.48.8.  Bartolus, Digestum vetus ad D 19.1.13 nos 1 & 3 (n 83) fol 774vb. 

208

the creation of the ius commune

aedilician edicts concerned almost exclusively cases dealing with the sale of  defective slaves and animals, and only occasionally something else, such as a  plot of land; but a return to classical law was incompatible with their approach  to the Corpus iuris in general. It is more likely that we are dealing here with  a medieval misinterpretation. On the basis of the wording of the edicts and  the cases dealt with in the titles D 21.1, de aedilitio edicto et redhibitione et quanti minoris, and C 4.58, de aedilitiis actionibus, it was argued that the  use of aedilician actions was restricted to the sale of slaves and animals. This  was done for the sole purpose of developing a weapon to attack the view of  Pierre de Belleperche and Cinus of Pistoia, who had claimed that the only  action for price reduction was the praetorian, and that a civil action for the  same purpose did not exist. If the praetorian action only applied to animals  and slaves, there must be a civil action; otherwise, it was impossible to explain  all the texts which granted remedies for price reduction when other defective goods were sold. When Bartolus and his followers maintained that the  aedilician  edicts  were  intended  only  for  the  sale  of  slaves  and  animals,  or  that  these  edicts  acquired  wider  applicability  only  through  later  interpretation,  this  probably  does  not  point  at  a  sincere  interest  in  legal  developments  prior  to  the  compilation  of  the  Corpus iuris,  but  rather  at  creating  an  argument  for  contemporary  debate.  A  similar  view  to  that  of  Bartolus  was  defended  by  Raphael  Fulgosius  (1367-1427),  who  lectured  at  Pavia,  Siena and Padua.95 Albericus de Rosate (ca 1290-1360), who practised law  in Bergamo, had already explained that the praetorian actio quanti minoris  was composed first. It had a limited application because it spoke only about  defective animals. The civil action, however, mentioned any defective object  purchased.96  In  fact,  this  explanation  was  historically  incorrect,  as  the  civil  actio empti was older than the aedilician actions. (3) Claiming price reduction: one remedy or various sets of rules? As  we  have  seen,  the  ordinary  gloss  maintained  there  was  an  actio quanti minoris civilis,  distinct  from  the  regular  actio empti  and  distinct  from  the  actio quanti minoris praetoria of the aedilician edicts. This idea was obviously  accepted in procedural law, since each of the two actions for price reduction  95 Raphael Fulgosius, In primam Pandectarum partem commentariorum ... Tomus secundus ad D  19.1.13 nos 3-4 (1554) fol 143rb. 96  Albericus de Rosate, In secundam ff. Veteris partem commentarii ad D 19.1.13 no 3 (1585; repr  1977)  fol  148vb:  “Est  et  alia  differentia,  ut  dicamus,  quod  quanto  minoris  pretoria,  prius  fuit  in compositione et defectiua erat, quia loquebatur de morbosis animalibus emptis. Ciuilis uero  loquitur de omni re morbosa empta, ut emptor agere possit quantominus emisset, si fuisset (lege:  sciuisset) morbum uel uitium, ut colligere potes hic, dum dicit de tigno uitioso.”

the ignorant seller’s liability for latent defects

209

appears to have received its own libel. The view of the gloss was rejected,  however,  by  the  Orleans  jurists.  Pierre  de  Belleperche,  followed  by  Cinus  of Pistoia, acknowledged the existence of only one actio quanti minoris the  praetorian,  which  had  to  be  brought  within  one  year.  The  commentators,  facing this difference in interpretation, had to think of arguments to adopt a  position of their own. Bartolus clearly decided to follow the gloss, as did most of the commentators. First he explained that, according to the gloss essem empturus on D  19.1.13pr, there were two actiones quanti minoris and he then referred to  the deviating view, mentioned in this gloss, which appealed to D 19.1.13.5.  Bartolus  ascribed  this  dissenting  view  to  Bulgarus  rather  than  Rogerius.  This  may  indicate  that  in  the  fourteenth  century  in  some  manuscripts  of  the  Digestum Vetus  the  ordinary  gloss  to  D  19.1.13pr  may  have  had  a  reading  different  to  that  found  in  later  printed  editions.97  Bartolus  subsequently presented a number of arguments in favour of the view of Pierre de  Belleperche and Cinus of Pistoia. Some of them derived from these jurists  themselves. First, in C 4.58.2 the emperor declared he was not aware of any  action  which  could  be  granted  for  price  reduction  after  the  period  of  one  year had lapsed. It was argued that if there had been an actio quanti minoris civilis, which was perpetual, the emperor would have known. Secondly, the  two ways of estimating price reduction resulted in the same amount, because  it was very unlikely that the buyer, when he was aware of the defect, would  have been willing to pay anything else but the estimatio communis. These two  arguments  were  subsequently  supported  by  complicated  reasoning,  based  on  the  text  D  44.7.35,  which  Jacques  de  Revigny  had  already  discussed.  According  to  this  text,  praetorian  actions  which  were  reipersecutory  were  available for a longer period than just one year, if they were in conformity  with  civil  law.  Thus,  when  usucapio  was  interrupted  before  the  required  period of time had elapsed, the actio Publiciana would be available for only  one year, because it was not in conformity with the ius civile. Now, Bartolus  argued that, if two actiones quanti minoris were in existence, the praetorian  one had to be perpetual, since it was given in conformity with civil law. It  was, however, an annual action, because it was granted contrary to the civil  law. The praetorian action observed the interesse commune, the civil action  observed the interesse singulare. This reasoning was obviously intended to  lead to the conclusion (although this is not very clear) that it was impossible  for the two actions to co-exist. As just stated, Bartolus himself followed the  97 See Van Warmelo, Vrywaring teen gebreke (n 62) 65 where the dissenting view is ascribed to  Hugolinus, citing Leiden MS BPL n 6 C.

210

the creation of the ius commune

gloss.  He  maintained  that  the  civil  action  could  not  be  omitted,  since  the  praetorian action was given only for the defective slaves and animals which  the aedilician edicts spoke about in D 21.1.1 and D 21.1.38. In D 19.1.13 the  actio quanto minoris was given for an unsound beam, and thus the remedy  granted there could not possibly be praetorian. Bartolus next refuted the two  arguments of Pierre de Belleperche and Cinus of Pistoia. He adduced three  examples to demonstrate that there could be an enormous difference between  the precium commune and the precium singulare, since the latter depended  on the buyer’s personal circumstances. I would not consider buying a slave if  I knew he was a murderer, but someone else, a ruler (tyrannus) for example,  might possibly like to have him. A doctor would not be willing to buy a horse  if he knew it was defective, but a stable man (marescalus) might be willing  to buy it, albeit for a small price.98 Moreover, a student would not consider  buying a book containing some errors, whereas a scholar (doctor) might not  mind so much. Finally, according to Bartolus, the argument derived from C  4.58.2 was not convincing. From the context of the emperor’s statement it  appears that he was consulted on the availability of the actio quanti minoris praetoria. He was not asked whether in general an action was available.99 Baldus de Ubaldis (1327-1400), who taught in various Italian law schools,  summarised in his commentary on D 19.1.13 more or less what his master,  Bartolus, had already said.100 Baldus’ consilium concerning the sale of a defective horse is of more interest. The main problem discussed in this consilium  was not so much the difference between the praetorian and the civil actions  for reduction in price, but how to prove that the horse that was bought and  later died had already been carrying the lethal disease at the time it was sold.  It is incidentally worth noting that this problem is somewhat similar to one  described by the canonist Panormitanus (Nicolaus de Tudeschis, 1386-1445)  on a text of the Liber Extra (X 3.19.4), which dealt with a horse that died  three days after the sale because of what appeared to be heart trouble.101 This is not the place to go into details concerning the four different ways  in which the plaintiff could furnish the required proof; rather, I will discuss  the remedies considered advisable in the case of the consilium, which elucidated the practical consequences of Accursius’ doctrine on the two separate 

  98 This example is reminiscent of the one in the gloss of Johannes Bassianus on D 19.1.13 reproduced in n 43 above.   99  Bartolus, Digestum Vetus ad D 19.1.13 no 1 (n 83) fol 774vab.  100  Baldus Perusinus, Tomus secundus in Digestum Vetus ad D 19.1.13pr (1541) fol 112va.  101  See Panormitanus, In tertium decretalium librum interpretationes ad X 3.19.4 no 4 (1547) fol  114rb.

the ignorant seller’s liability for latent defects

211

actiones quanto minoris.102 Baldus mentioned three actions, each of which  he  characterised  as  a  bona via.  He  noted  first,  however,  that  when  a  pact  concerning  diseases  and  defects  was  concluded  between  seller  and  buyer,  this pact had to be observed. This followed from D 4.3.37, D 21.2.31 and D  21.1.19.2. He went on to mention the three remedies. If it could be proved  that  the  horse  had  the  disease  at  the  time  of  the  sale,  there  was  the  actio redhibitoria and the actio quanto minoris, which in this case would be aimed  at quanto minoris essem empturus. These two remedies included restitution  of the selling price. For the latter, this followed from D 21.1.43.6. The third  action  available  was  the  contractual  actio empti  based  on  the  pact  which  could  also  be  used  for  rescission,  as  followed  from  D  19.1.11.5.  The  only  action not mentioned here was the actio quanto minoris praetoria. Baldus  explained the preference for the civil remedy for price reduction by referring  to D 21.1.43.6. The civil actio quanto minoris could lead to rescission when  the buyer declared that he would not have bought the horse at all, had he  known of the defect, because it took into account the pretium singulare. This  was already laid down in the gloss essem empturus on D 19.1.13pr. In the case  under dispute, this was apparently what the buyer wanted, namely restitution  of the entire selling price, and not just the difference between selling price  and pretium commune.103 Some writers, such as Bartholomeus de Saliceto, maintained that only the  actio quanti minoris praetoria could be used to claim restitution of the entire  selling  price.  Merely  declaring,  however,  that  one  would  not  have  bought  the thing was insufficient for that purpose. The entire price could only be  claimed back with this remedy when, according to the estimatio communis,  the horse had no value at all at the time it was sold.104 This had probably not  been  the  case  in  the  example  in  Baldus’  consilium.  For  the  rest,  Saliceto  adhered to the opinion of the gloss on the two actions for price reduction  and the arguments produced by Bartolus against the teachings of Pierre de  102  For an entirely different opinion, namely that Baldus in this consilium did not see any difference  between the praetorian and the civilian way of assessment and merged the aedilician action with  the civil one, see Zimmermann, Obligations 323-324. 103  Baldus Ubaldus Perusinus, Consiliorum sive responsorum ... Volumen Quintum consilium 499  (1589) fol 122rb.  104  Bartholomaeus  à  Salyceto,  Commentariorum ... in tertium et quartum Iustiniani Codicis  ad  C 4.49.9 no 2 (1541) fol 214ra. To emphasise the difference between pretium singulare and  pretium commune Saliceto referred again to the three examples mentioned by Bartolus and  spoke about a doctor or physician (medicus) who would not have wanted to buy the defective  horse.  In  this  phrasing  the  case  recalls  one  of  the  classical  examples  used  in  the  sixteenthcentury school of Salamanca: see Domingo de Soto, De iustitia et iure lib VI q 3 art 2, introd by  V D Carro and trans M González Ordóñez (1968) 555.

212

the creation of the ius commune

Belleperche.105 The same can be said for Paulus de Castro. The only critical  remark he made was that Pierre de Belleperche was right in his opinion that  quanti minoris civilis  is not the title of a specific, independent action, but  an addition (adiectio) to the actio empti. On the other hand, he rejected the  opinion of Belleperche that the action for price reduction was only aimed  at quanto minoris res est.  It  might  be  more  advantageous  to  claim quanto minoris fuisset empturus,  as  was  shown  by  the  examples.106  According  to  Angelus de Ubaldis, legal practice observed the teachings of neither Pierre de  Belleperche nor Cinus. He referred to the remark, now ascribed to Belleperche, concerning “practitioners who insist in their deeply rooted errors”,  and said that it was the common opinion of the entire world to observe the  opposite and to allow the buyer to sue the seller for latent defects for many  years after the sale.107 The treatise De actione et eius natura of the Paduan professor (Giovanni)  Battista  da  Sanbiagio  (Baptista  a  Sancto  Blasio,  ca  1425-1492)  contains  an  extensive  commentary  (two  columns  in  folio)  on  the  difference  of  opinion  between, on the one hand, the Accursian gloss and, on the other, the Orleans  jurists and Cinus. This does not dramatically change our understanding of the  teachings of the commentators, but it demonstrates that, on the threshold  of  early-modern  times,  and  while  the  process  of  reception  of  Roman  law  was  already  taking  place,  there  was  a  lively  debate  on  the  seller’s  liability  for price reduction in view of latent defects. The main issue for Sanbiagio   was to consider what kinds of actiones quanti minoris existed. Eventually he  appeared to adopt a middle-course between the gloss and Pierre de Belleperche. Sanbiagio started with a discussion of three arguments derived from Pierre  de  Belleperche  and  Cinus  of  Pistoia.  First,  according  to  their  view,  there  was only one action for price reduction – the praetorian action for quanto minoris res, which had to be brought within one year. Afterwards there was  no action left for the buyer, but there would be, it was argued, if there were  an  actio quanto minoris  civilis,  because  that  action,  being  civil,  would  be  perpetual. Second, it was very likely that the amount the buyer would have  been willing to pay, had he known of the defect, would have coincided with  the pretium commune. According to D 35.2.63, the pretium singulare should  105 Salyceto, Commentariorum ad C 4.58.2 no 4 (n 104) fol 220rb. 106 Paulus Castrensis, In secundam Digesti Veteris partem commentaria ad D 19.1.13pr no 4 (1582)  fol 121rb.  107 Angelus, Super codice ad C 4.58.2 (n 87) fol 105rab: “et hoc dicit esse uerum quicquid teneant  aduocati, sed mundi consuetudo obseruat contrarium quod ubi res uitiosa uenditur ratione uitii  in perpetuum uenditorem agitur ad interesse.” 

the ignorant seller’s liability for latent defects

213

not be considered when estimating prices. Third, if there were two actions,  the  praetorian  one  had  to  be  in  conformity  with  civil  law  and,  because  of  D 44.7.35, be available for more than one year, although this would not be  the case if it was contrary to civil law. Sanbiagio subsequently brought some  more arguments into the dispute. Fourth, if there were an action directed at  quanto minoris empturus fuisset, it would be possible to claim back the entire  selling price and at the same time to retain the merchandise. But, according  to D 18.1.1, it was not fair if the seller were deprived of his goods without a  quid pro quo, although both Saliceto in his commentary to C 4.49.9 and the  gloss essem empturus to D 19.1.13pr seem to have accepted this. Fifth, if the  buyer would not have paid anything, had he known of the defect, it seems  that he should not use the actio quanti minoris, but the actio redhibitoria. On  the basis of all these arguments, Pierre de Belleperche and Cinus of Pistoia  came to the conclusion that there was only one actio quanto minoris, namely  the praetorian. After having expounded the doctrine of Pierre de Belleperche and Cinus  of  Pistoia,  Sanbiagio  put  forward  two  counter-arguments.  The  common  opinion did not agree with the idea that quanto minoris was just an addition  to the actio empti of D 19.1.13pr. The gloss empturus essem to D 19.1.13pr  had the correct view, because the aedilician action was only available for the  slaves, animals and other things, about which D 21.1.1 spoke. Secondly, in C  4.58.2, the emperor was only asked whether the praetorian action for price  reduction was available. The entire title C 4.58, de aedilitiis actionibus, dealt  with aedilician actions. Accordingly, C 4.58.2 did so too. This text should be  understood as referring solely to the praetorian action. It said nothing about  the civil action. This followed from D 18.2.16. Furthermore, in case of doubt,  it was necessary to stick to the rule that the name of the title (rubrum) was  determined by the texts adopted under this title (nigrum).108 Having  explained  this,  Sanbiagio  responded  to  the  second  and  third  arguments  of  Pierre  de  Belleperche  and  Cinus  of  Pistoia  that  he  had  just  paraphrased.  Although  the  provisions  in  the  Corpus iuris did  not  express  this unequivocally, if the buyer would not have bought at all, he had available the actio redhibitoria, not the actio quanti minoris civilis. Otherwise,  the inconvenient consequences just mentioned (the seller losing his goods  without counter-performance) might occur. Sanbiagio preferred the opinion  108 Baptista  a  Sancto  Blasio,  Tractatus utilissimus solemnissimusque de actione et eius natura  Vigesima prima actio (Quanto minoris est) no 45 in Volumen V Tractatuum ex variis iuris interpretibus collectorum (1549) fol 62vb: “et quod notat glossa in c. Bone de confirma. uti. uel inutil.  (X 2.30.3) que dicit quod in dubio tenendum est quod nigrum disponat id quod rubrum.”

214

the creation of the ius commune

of Pierre de Belleperche and Cinus of Pistoia that it was better not to regard  the actio quanto minoris civilis as an action of its own separated from the  contractual  action  of  sale.  D  19.1.13pr  spoke  about  in condemnatione ex empto, while the jurist in all the fragments from D 19.1.11 until D. 19.1.13pr  dealt  with  the  actio empti,  and  the  compilers  adopted  D  19.1.13pr  in  the  title  on  the  remedies  for  sale.  It  was  true  there  was  only  one  actio quanti minoris praetoria, namely the one mentioned by Johannes Bassianus in his  Arbor actionum and the one Roffredus dealt with in his Libelli iuris civilis.109  There  was  also,  however,  the  actio empti  which,  through  its  wide  applicability, could be used for the purposes indicated in D. 19.1.13pr. But Bartolus  and Paulus de Castro accepted the majority view that there was an actio ex empto ciuilis quanto minoris emptor empturus fuisset, which differed from  the actio pretoria quanto minoris res communiter ualebat. One of Bartolus’  examples was subsequently ridiculed. Suppose that a doctor, knowing a book  contained errors, had nevertheless bought it for a higher price than an unlettered person, such as a student. He would have had the civil action, not the  praetorian, because if the doctor wanted to declare under oath that he would  only have bought it for the actual value, that would be perjury. All this was  nonsense, Sanbiagio argued, because the one who knew of the defect did not  want to buy at all, and certainly not for more than its pretium commune. D  1.3.5 and the gloss essem empturus on D 19.1.13pr did not refer to such very  exceptional situations.110 It becomes clear that in some respects Sanbiagio followed Pierre de Belleperche and Cinus of Pistoia. There was no actio quanti minoris civilis, distinct  from the actio empti, in the Corpus iuris. There was the actio redhibitoria  for rescission and the actio quanti minoris for price reduction, based on the  pretium commune. Thus, price reductions should be estimated according to  the actual value of things and not according to the pretium singulare. Rescission should not be claimed by the actio quanti minoris. But whereas Pierre  de Belleperche and Cinus of Pistoia had taught that, after one year, the buyer  no longer had remedies for latent defects available, Sanbiagio still considered  it possible to use the actio empti for the purposes indicated in D 19.1.13pr.

109 It may be noted, though, that Roffredus, Libelli (n 56) – at least the edition printed in Avignon  1500 (reprinted in Turin 1968) – also contains a libellus for the actio quanti minoris civilis, as  was noted above. 110  Sancto  Blasio,  Tractatus  Vigesima  prima  actio  (Quanto  minoris  est)  nos  43-48  (n  108)  fols  62va-65ra.

the ignorant seller’s liability for latent defects

215

F. CONCLUSIONS In  summary,  we  can  say  that  developments  in  medieval  legal  doctrine  concerning the ignorant seller’s liability for latent defects took place in two  different stages. The era of the glossators was characterised by the formation  of  a  rule  of  law,  the  era  of  the  commentators  by  the  debate  over  whether  the principle this rule contained – namely that the ignorant seller is liable  for  price  reduction  –  offered  the  buyer  the  choice  between  two  separate  remedies, each with its own features. The  starting  point  of  all  scholarship  was  the  Corpus iuris civilis,  which  offered hardly any general rules or principles on liability for latent defects.  Apart from the aedilician edicts, the most generally phrased text in the Digest  can  be  found  at  the  beginning  of  D  19.1.13.  For  the  rest  there  were  only  cases, and in some of these the ignorant seller was considered to be liable,  such as in D 18.1.45 (garments which had been repaired were sold as new),  D 19.1.6.4 (sale of a leaky barrel), D 19.1.13.1 (sale of a runaway slave) and  C 4.49.9 (sale of a plot of land which owed a higher capitatio than the seller  had stated). In other cases the ignorant seller was regarded as not liable, such  as in D 19.1.13.1 (sale of a thief) and D 19.1.21.1 (sale of a plot of land which  appeared  to  be  a  predium tributarium).  It  had  to  be  decided  which  cases  reflected the rule and which reflected the exception. In that discussion, the  number of the cases played a role (four or five is more than two) as did the  question whether the deviation from the principal rule could be rationalised.  Indeed, the buyers in D 19.1.13.1 and D 19.1.21.1 could be considered to owe  their loss to their own negligence. Second, it had to be decided to what extent  the ignorant seller was liable – whether for the buyer’s full interest or merely  for reduction in price. Reduction in price appeared to become the standard,  and full compensation the exception. This could also be  rationalised, such as  by the fact that it was very simple for a seller of barrels to check whether or  not the barrel leaked. All the cases just mentioned were found in titles dealing with the remedies  for  sale.  As  a  consequence,  the  action  granted  was  the  contractual  actio empti; but when this remedy was used against the ignorant seller of defective  goods, it came very close to a different action, one which could be found in  the titles on the aedilician edicts. Both actions, the actio empti and the actio quanti minoris, were available for exactly the same situation (sale of  defective  goods  by  an  ignorant  seller)  and  for  exactly  the  same  purpose  (claiming  a  reduction in price).111 This made Johannes Bassianus and Azo emphasise the  111 Although the aedilician edicts imposed a duty on the seller to provide the buyer with informa-

216

the creation of the ius commune

differences. The actio empti, aimed at reduction in price, was civil and thus  perpetual, and because of some texts in D 19.1, de actionibus empti et venditi,  was used to claim the balance between what the buyer paid and the lesser  sum he would have paid had he known of the defect (pretium singulare). The  aedilician actio quanti minoris was praetorian and thus limited in time, and  because of some texts in D. 21.1, de aedilitio edicto et redhibitione et quanti minoris, it was used to claim the difference between the selling price and the  value of the thing at the time of the contract of sale (pretium commune). The  next step in development was made by Accursius who defined the actio empti  used for price reduction as an independent and separate action, distinct from  the regular actio empti and distinct from the aedilician action for price reduction in price termed the actio quanto minoris civilis. Accursius made the Roman law remedies more systematic than the sources  could actually justify, and for this he was criticised. The Siete Partidas did not  adopt  his  view,  as  embodied  in  the  gloss  essem empturus  to  D  19.1.13pr,  while the Orleans jurist Pierre de Belleperche, followed by Cinus of Pistoia,  rejected it. In their opinion there was only one action for reduction in price  available against the ignorant seller in the case of latent defects – the aedilician actio quanti minoris – which had to be brought within one year and was  aimed  at  the  difference  between  the  selling  price  and  the  actual  value  at  the time of the sale. In the course of the fourteenth and fifteenth centuries,  hardly any commentator opted for the opinion of Pierre de Belleperche and  Cinus  of  Pistoia  and  almost  all  followed  the  gloss.  However,  the  scholarly  sources we considered do show that the alternative opinion was very much  alive in the doctrinal debate, even at the end of the fifteenth century. Both  views  –  that  of  the  gloss  and  that  of  Pierre  de  Belleperche  and  Cinus  of  Pistoia – were provided with many arguments, most of them derived from the  Corpus iuris. At the end of the fifteenth century, Giovanni Crispo de Monti  qualified the debate concerning the difference in assessment as an old and  hackneyed story (antiqua et trita quaestio). He decided not to deal with it in  order to proceed without delay to more interesting issues.112 Around the year  1535,  it  appears  that  the  Imperial  Court  of  Justice  (Reichskammergericht)  still applied the prevailing view of the medieval doctores. Its assessor Viglius  of Aytta (1507-1577) opposed, however, the idea that the actio quanti minoris empturus erat was an independent action, and followed the view of Pierre  tion and to give warranties, liability under the actio quanti minoris was not dependent on the  fact whether such warranties were given or not.  112  Joannes Crispus de Montibus, Termini omnium actionum cum arbore ad Quanto minoris (1519)  [found added to Iason’s De actionibus in the edition of Lyons, 1546].

the ignorant seller’s liability for latent defects

217

de Belleperche and Cinus of Pistoia that price reduction should be based on  the objective value of the merchandise. In an extensive report he rejected the  doctrine of Bartolus and supported his own opinion with nine arguments.113  One  generation  later,  the  academic  jurists  would  definitively  abandon  the  teachings of the gloss essem empturus on D 19.1.13pr. The most important difference between the two approaches was not so  much the exact assessment of the reduction in price. This would only have  made a substantial difference in very exceptional situations. According to the  gloss, however, the actio quanto minoris civilis for price reduction could be  used up to thirty years after the sale had been concluded, whereas, according  to the alternative view, any action had to be brought within one year. When  Roman law was be received into the legal practice of early modern times,  this  issue  of  liability  for  defects  required  a  choice  to  be  made.  The  gloss  retained its authority; but rules of law were no longer exclusively based on the  authority of legal texts from a distant past; instead, they were developed by  new generations of scholarly jurists and legal practitioners who were increasingly sensitive to the demands of legal practice and reliance on human reason.

113 See R Sprenger, “L’autorité de la ‘communis opinio’ dans le pratique de la Chambre impériale  de justice (c. 1535): une application de l’actio quanti minoris”, in A Wijffels (ed), Miscellanea Consilii Magni III, Essays on the History of Forensic Practice [= Verzamelen en bewerken van  de jurisprudentie van de Grote Raad, Nieuwe reeks XII] (1988) 161; and R M Sprenger and  A Wijffels, “De actio quanti minoris in de praktijk van het rijkskamergerecht ca. 1535. Uit de  aantekeningen van Viglius van Aytta”, in D Lambrecht (ed), Lopend rechtshistorisch onderzoek. Handelingen van het tiende Belgisch-Nederlands rechtshistorisch colloquium [= Iuris Scripta  Historica III] (1990) 129.

9  The Glossators’ Monetary Law Wolfgang Ernst A. INTRODUCTION (1) Money in Justinian’s laws (2) The glossators’ monetary environment B. GLOSSATORS ON MONETARY ISSUES (OTHER THAN DEBASEMENT) (1) Introduction (2) On the nature of money: the lex origo emendi (D 18.1.1) (3) Coins for bullion? (4) Exchangeability of denarii from different cities? C. GLOSSATORS ON COIN DEBASEMENT (1) Bulgarus and the case of the Bolognese bushel (2) Pillius’ Quaestio “Quidam creditor Lucenses” (3) Azo’s brocard (4) Evaluating the glossators’ achievements D. CANON LAW ON COIN DEBASEMENT (1) Introduction (2) Decretals on instabilities in the fields of money and measurement (a) 2 Comp. 3.25.3 Querelam (later X 3.39.26) (b) 3 Comp. 2.15.4 Quanto personam tuam (later X 2.24.18) (c) 3 Comp. 3.37.3 ex parte Compostellani (later X 3.39.18) (d) 3 Comp. 3.37.5 olim causam (later X 3.39.20) (3) A nominalist minority position? (4) The canon law’s classical position E. CONCLUSION: THE HUMBLE BEGINNINGS OF LEARNED MONETARY DOCTRINES

219

220

the creation of the ius commune

A. INTRODUCTION This chapter tries to establish what was the learned law’s view of money in the  time of the glossators. Before going into the glossators’ texts, it is necessary to  consider what Justinian’s law books contained in this respect, and therefore to  turn briefly to the views that the Roman lawyers of antiquity held on the law  regarding money and currency. (1) Money in Justinian’s laws The Roman economy was fully monetarised.1 Money for the Romans meant  coins, nummi. Roman coins were “commodity money”, but circulated by tale.2  The Roman currency was state-controlled; coins were referred to as materia forma publica percussa.  The  minting  of  coins  was  meticulously  organised.  There was no free minting. Coins came in two – later three – denominations,  brought into an official, rigid relation. Bi- and tri-metallism, the parallel use  of copper, silver, and later also gold, required adjustments from time to time,  and  these  were  brought  about  in  an  orderly  way.  Public  finances,  taxation  and  the  administration  of  justice  relied  on  the  Roman  currency.  Penalties  were fixed in amounts of Roman money. Judgements in private law matters  must also be rendered in sums of Roman money (condemnatio pecuniaria). A  number of Roman contracts required the fixing of a specific amount of money.  Given the overwhelming importance of money for the public sector as well  as for everyday commerce, non-monetary transactions being marginalised,3  it  is  no  wonder  that  Roman  lawyers  dealt  with  innumerable  legal  issues  involving money. In a nutshell, the Roman lawyers’ approach was as follows:4  coins  were  considered  –  and  indeed  were  –  personal  chattels  (literally:  pecunia), res nec mancipi.  Coins were thus in principle subject to the law  applicable  to  chattels.  (Though  produced  by  official  mints,  coins  were  the  subject of individual property, not state property.) Coins were considered res quae numero pondero mensura constant – goods that get counted, weighed  or measured (res fungibiles). The Roman jurists considerably modified the    1 On Roman money see, most recently, the contributions to W V Harris (ed), The Monetary Systems of the Greeks and Romans (2008).   2  This has been ascertained by K Hasler, Studien zu Wesen und Wert des Geldes in der römischen Kaiserzeit von Augustus bis Severus Alexander [= Bochumer historische Studien: Alte Geschichte  IV] (1980).    3  The persisting field of non-monetary transactions, as seen from a legal perspective, has been dealt  with  by  A  Bürge,  “Geld-  und  Naturalwirtschaft  im  vorklassischen  und  klassischen  römischen  Recht” (1982) 99 ZRG, RA 128.   4  On money in Roman law, see recently the overview by L Winkel, “L’argent en droit romain”, in  Argent et Gestion, Neuvièmes rencontres 23 et 24 novembre 2000 [= Collection Histoire, Gestion,  Organisations IX] (2001) 97.

the glossators’ monetary law

221

law pertaining to personal chattels when applied to coins. These modifications  decisively  supported  the  circulation  of  money.  The  legal  concepts  of  commixtio and consumptio nummorum show a clear grasp of money circulating as a means of exchange. Roman lawyers on a number of occasions quite  stringently distinguished minted coins, basically standard pieces of silver –  commodity money – from silver as bullion.5 Likewise, a corpus nummorum,  a hoard of coins, was distinguished from a sum of money stated in abstract  terms (quantitas, summa).6 The Romans had a sharp concept of legal tender,  or “lawful money”, debating, for example, whether foreign coins were legal  tender.7 While there was something like transfer of bank credit, this was not  seen as “legal tender”.8  Turning to Justinian’s codification, we find a basic approach to money which  faithfully reflects and revives the approach of the Roman lawyers of antiquity. This was so simply because an extensive overhaul of money-related legal  issues was neither necessary nor feasible, given the way Justinian’s corpora  of law were compiled from older material. The Digest and Code therefore  maintain  the  notion  of  “money”,  coins  circulating  by  tale  as  the  standard  medium  of  exchange.  The  Digest  and  Code  also  contain  an  abundance  of  money-related  legal  problems,  great  and  small,  but  the  relevant  texts  are  scattered throughout: there is no specific title in either body of law devoted to  money. The Digest preserves an interesting piece on monetary theory by the  jurist Paul – D 18.1.1.1.9 Paul here supports a sharp distinction between sale  and barter (this being the majority opinion among Roman lawyers), relying  on the qualification of money as a universal means of exchange. In spite of the abundance of legal material addressing monetary problems,  the Digest and Code do not accurately reflect the historical development of  monetary issues during the times of the Roman Republic and the Principate    5 A striking example is the handling of legacies, by which “all silver” is bequeathed. Silver coins,  Ulpian  tells  us,  are  not  included,  D  34.2.27pr-1:  “non  facile  enim  quisquam  argenti  numero  nummos computat”.   6  Cf, e.g., D 30.34.1, 3-4: a bulk of coins (corpus nummorum) can be bequeathed but once, whereas  several legacies for amounts of money can co-exist.   7  Volusius Maecianus, Assis distr § 45, in E Seckel and B Kübler (eds), Iurisprudentiae anteiustinanae reliquiae (1908) 415; Plinius, Nat hist 33, 46, ed K Mayhoff (1897) 149-150.   8  Recent research has intensively turned around the issue to what extent the Romans used “bank  money”: cf W V Harris, “The Nature of Roman Money”, in Harris, Monetary Systems (n 1) 174;  J Andreau, “Some Observations on Banking and Roman Law”, in W J Zwalve and E Koops (eds),  The Past and Future of Money. Proceedings of the Leiden Congress, August 31st and September 1st, 2006 (2008) 6. It would be difficult, however, to vindicate the status of legal tender for Roman  “bank money”.   9  On this text, see P Blaho, “Abgrenzung zwischen Kauf und Tausch in der Dichtung des Homer”,  in E Jakab and W Ernst, Kaufen nach römischem Recht (2008) 53 at 55-56.

222

the creation of the ius commune

(and, of course, were not meant to). Most notably, we miss a discussion of  the  single  most  important  issue  recurring  in  the  history  of  monetary  law,  namely, the interplay between fluctuations of the currency, tectonic shifts in  the monetary system, and monetary obligations – an issue which must have  troubled the third and fourth centuries.10 Justinian’s compilation was meant to  serve as the basic law of an empire with a well-ordered, centralised monetary  system.11 Deliberations by the Roman jurists on currency crises (which we  know did occur) had no place in Justinian’s law books. Justinian’s compilers  adapted the money-related texts, which they used for the Digest, Code or  Institutes,  to  the  contemporary  Byzantine  monetary  environment.  This  is  most strikingly reflected in a thorough revision of the original Roman denominations.12 Amounts stated originally in sestertii were changed into amounts  of aurei, while the original use of denarii remained unaltered.13 Thus, texts  of the Digest and Code could be read in the light of the contemporary, sixthcentury, monetary environment. This was unavoidable, given the intention  of the Byzantine legislator to make the Digest and Code the law of the land.  Wherever in the Digest or Code sums of money are mentioned (for reasons  other than illustrative) to fix fines or alimony payments, the coins available  in real life had to be used. The Byzantine lawmakers had little reason to deal  with currency turbulences which were absent from the monetary environment  established  in  sixth-century  Byzantium;  and  anyhow,  in  grand-scale  codifications issued in aspiring times, the legislators will seldom provide for  the possibility of currency disasters or the total collapse of the economy.  The  other  issue  conspicuously  absent  from  the  monetary  discussions  in  both the Digest and Code is the mingling of money from different currencies.  While  in  the  Roman  Empire  one  did  have  to  deal,  economically  and  legally, with a multitude of currencies,14 Justinian’s law texts seem to be based  on the idea that, uniform throughout the empire, the environment was that  of a self-contained currency, an ideal to which the Byzantine sixth-century,  with its strong gold-based currency, tried to live up.  10 Of course, Diocletian’s rulings on laesio enormis (C 4.44.2 and 8) were a response to the inflationary crisis of the late third/early fourth centuries, as was his edict on tariffs.  11  On  money  in  Byzantium,  see  M  F  Hendy,  Studies in the Byzantine Monetary Economy, c. 300-1450 (1985). 12  S  Mrozek,  “Zur  Geldfrage  in  den  Digesten”  (1970)  18  Acta Antiqua Academiae Scientiarum Hungaricae 353. 13  Probably at a rate of 1000:1. 14  Interestingly, our key sources see (n 7 above) are not from the Justinianic texts. In the Digest one  can only occasionally assume that a text dealt with the parallel use of different currencies; see, for  the case of D 46.3.99, W Ernst, “Rationalia ad D. 46.3.99”, in P Pichonnaz et al (eds), Spuren des römischen Rechts. Festschrift für Bruno Huwiler zum 65. Geburtstag (2007) 233.

the glossators’ monetary law

223

All in all, the Digest and Code offer ample material for the legal aspect  of  money,  dealing  both  with  the  idea  of  money  and,  predominantly,  with  a multitude of micro-level issues. But both the Digest and Code fall short  when one looks for specific guidance with regard to legal problems caused  by tectonic shifts in the monetary environment or transactions involving coins  from different currencies.  (2) The glossators’ monetary environment While one may debate to what extent features of real life found their way  into the glossators’ work, with regard to money there can be little doubt that  the glossators knew very well what, at their own time, money was and what  money  did.  The  twelfth-  and  early  thirteenth-century  glossators  lived  in  a  single-coin environment.15 Only relatively small, silver coins of one type were  in circulation. It was only from the late 1230s on that the minting of large  silver or gold coins, grossi, like the Venetian ducat or the  Florentine florin,  began in Northern Italy. The additional difficulties that arise with bi-metallism  do  not  concern  us  here  because,  in  the  world  of  the  early  glossators,  there was only one coin, the denarius, and silver was the only metal used for  minting.16 To speak of a single-coin environment is not, however, entirely correct. As  a denomination, the Carolingian denarius had indeed survived the break-up  of the Carolingian empire. But the coin was now minted by a multitude of  local potentates, all using different standards of fineness and faces or designs  of their own choice. “Most coinage was struck by and for the profit of the  local nobility, lay or ecclesiastical, who proceeded to reduce the weight and  fineness  of  the  denier  at  differing  rates,  resulting  in  the  regionalisation  of  currency.”17  In  commercial  dealings  monetary  obligations  were  therefore  specified by stating the origin of the denarius, such as denarii from Pavia or  Cologne. The  denarius  was  a  relatively  small  coin,  a  “penny”.  As  a  medium  of  exchange it was not ideal for large-scale transactions. Therefore, large-scale  obligations were often phrased by stating certain weights of silver or gold,  15 On  money  in  the  twelfth  century,  see  T  J  Sargent  and  F  R  Velde,  The Big Problem of Small Change (2003) 45-53. 16  Digest texts dealing with the interchangeability of smaller and larger coins, e.g. D 45.1.65.1, the  latter part of the lex quae extrinsecus, could hardly have real-life meaning for the glossators. Gold  coins were not, however, altogether unknown, if only because foreign gold coins were imported.  17  M  Blackburn,  “Money  and  Coinage”,  in  R  McKitterick  (ed),  The New Cambridge Medieval History (1985; repr 2002, 2004) II 538 at 559.

224

the creation of the ius commune

relying on the use of bullion.18 Scales and standard weights were everyday  features of medieval commerce. Insofar as unminted gold or silver, in typified  weights  (standard  bars),  was  used  in  transactions,  one  speaks  of  a  bullion  currency.  For  storing  purposes,  coins  and  precious  metal  were  used  more  or less indiscriminately, often kept not in the form of standard bars but as  objets d’art for church and other use (chalices, goblets and the like). Since  the  work  of  silversmiths  was  cheap,  only  the  silver-content  counted.  The  twelfth-century economy thus used, side by side, two rather different means  of exchange: bullion and, as pecunia numerata, a single denomination in the  form of a small silver coin, albeit in reality coined by a multitude of regionally  controlled mints, their products greatly varying in style, face and fineness.  In  contemporary  deeds,  sums  of  money  are  often  stated  in  librae  and  solidi.  These  units  do  not  refer  to  real-life  coins.  These  units  were  mere  counting aids, denoting multiples of the denarius, the solidus being the term  for twelve denarii, the libra being the term for twenty solidi or 240 denarii.  Thus,  instead  of  using  the  decimal  system  throughout,  larger  numbers  of  denarii were simply stated in dozens (the solidus) and sets of twenty dozens  (the libra). Contemporaries were well aware that if one referred to sums of  money using the libra, this was the libra numeralis, mere moneta imaginaria,  “money  of  account”.19  The  libra  only  became  moneta realis  once  moneta grossa started to be minted. The legal issues resulting from the parallel use  of denarii and librae, introducing the intricate problems of bi-metallism to  the medieval monetary world, lie outside the scope of this paper.20 A characteristic feature of the currencies in the high Middle Ages was the  recurring  debasement  of  coins.  The  circulation  of  coins  was  often  limited  to a statutory period (typically a year), at the end of which all coins had to  be  returned  to  the  mint  (reprobatio nummorum, renovatio monetae).  The  incoming coins, having lost weight as a result of use or deliberate clipping,  were  used  to  mint  the  same  amount  of  coins  anew  (without  adding  new  silver), now all, of course, of diminished fineness. The new coins normally  carried the same name as the old ones; they were of the same denomination.  18 Looking  into  (somewhat  older)  medieval  collections  of  deed-formulas  for  contracts  of  sale,  H  Siems  encountered  a  characteristic  difficulty  in  determining  whether  prices  fixed  referred  to  coins or to bullion: H Siems, Handel und Wucher im Spiegel frühmittelalterlicher Rechtsquellen  (1992) 386 and passim. From a learned legal point of view, using bullion as pretium would have  lead to a classification of the contract as barter (permutatio). 19  Like today’s “Special Drawing Rights” (SDR) of the International Monetary Fund. 20  On the monetary doctrine to be found in thirteenth- to fifteenth-century consilia, see E Stampe,  Das Zahlkraftrecht der Postglossatorenzeit  [=  Abhandlungen  der  preussischen  Akademie  der  Wissenschaften Phil-Hist Klasse I] (1928). For an overview of the later developments, see also H  Coing, Europäisches Privatrecht, 1500-1800 (1985) I 471-478.

the glossators’ monetary law

225

Sometimes one could tell the difference owing to a change in the appearance of the coin; sometimes not. If the older coins were taken officially out of  circulation, they became pecunia reproba.  B. GLOSSATORS ON MONETARY ISSUES (OTHER THAN DEBASEMENT) (1) Introduction The  glossators’  monetary  doctrines  were  investigated  in  Walter  Taeuber’s  magisterial  study  of  1933.21  Before  Taeuber,  no  one  had  delved  so  deeply  into the unprinted sources – such as the annotated manuscripts especially of  the Digestum Vetus, the Infortiatum and the Digestum Novum – and it may  be some time before they are explored so deeply again.22 Apart from Taeuber,  several other authors have made considerable contributions to the field.23 It may be worthwhile to restate briefly Taeuber’s overall insight, gained  by years of focusing on money-related glosses, into the specific nature of the  legal literature produced in the twelfth and early thirteenth centuries.24 Before  Accursius  could  compile  the  Glossa ordinaria,  stone  had  to  be  piled upon stone. An evolutionary story could be written for each and every  gloss, even for single words within a gloss, covering its transmission from early  authors like Irnerius to Accursius (and beyond). The material for these “histories of glosses” has to be elicited from dozens of manuscripts. The earliest  glosses,  extremely  abbreviated  statements,  became  the  basis  of  all  further  developments. All sorts of thoughts, ideas and notions – smart and stupid alike  – were absorbed into the process of transmission, frequently without attribution to a specific author. The cross-references used to annotate the Romanlaw texts were sometimes triggered by association only, the underlying “train  of thought” more often than not obscure to a modern legal mind. Indeed,  a mere cross-reference could conceal a sophisticated argument, obvious to  contemporary scholars but on first sight meaningless for today’s reader. Older  glosses  and  later  additions  were  reverently  treated  as  a  common  heritage.  The Glossa ordinaria wove a mass of glosses into a single apparatus, wiping  21 W Taeuber, Geld und Kredit im Mittelalter (1933; repr 1968). 22 Older studies, still useful, include W Endemann, Studien in der romanisch-kanonistischen Wirthschafts- und Rechtslehre (1883) II 184-187. 23  P  Grossi,  Ricerche sulle Obbligazioni pecuniarie nel diritto comune  (1960);  G  Hubrecht,  “Quelques  observations  sur  l’évolution  des  doctrines  concernant  les  paiements  monétaires  du  XIIe au XVIIIe siècle”, in Aequitas und bona fides, Festgabe zum 70. Geburtstag von A Simonius  (1955) 133.  24  Taeuber, Geld und Kredit (n 21) 127-129.

226

the creation of the ius commune

out the intricate stories behind the individual glosses. The generation(s) that  followed Accursius still knew of – or had easy access to – the many subtle  argumentative  nuances  which,  by  reason  of  previous  developments,  came,  for the thirteenth-century reader, with each of the individual glosses. A legal  historian reading the Accursian Glossa ordinaria can only fully apprehend  the legal content of the glosses if he or she re-establishes, at least partly, the  textual stages of their pre-Accursian history.25  The  Roman  property-law  aspects  of  money,  especially  commixtio  and  consumptio nummorum,  could  be  taken  up  by  the  glossators  quite  easily.  Difficulties arose as to monetary obligations. It is easy to see, in retrospect,  that  the  twelfth-century  state  of  monetary  affairs  required  answers  to  the  following questions. Could one discharge a bullion debt by payment of coins?  Could one use denarii issued by different rulers for payment? What should  the effect of debasement be? All these issues, none of which was addressed  head-on in the Roman-law texts, were eventually settled.  (2) On the nature of money: the lex origo emendi (D 18.1.1) Paul’s musings in D 18.1.1, the lex origo emendi, on money as the defining  element of sale, as opposed to barter, were taken up.26 There was some flexibility in how to read the passage that use and property of money sprang non tam ex substantia praebet quam ex quantitate. If the use of money ex quantitate was taken to refer to money in circulation,27 and the use ex substantia  to  refer  to  the  use  of  the  silver  as  material  for  (to  give  some  examples)  melting and production of precious objects, none of these two uses was really  excluded, the use ex quantitate was just considered to be comparatively more  important. One thus read non tam / quam as “not so much ... as”;28 or even 

25 On  the  Accursian  gloss,  see  most  recently  H  H  Jakobs,  Glossa magna  (2006),  and  the  same  author’s collected articles, H H Jakobs, Digesten – Glosse – Savigny. Kleine Schriften zur Wissenschaft vom Römischen Recht, ed W Ernst (2004).  26  The Greek quote from Homer was lost on them. 27  Some of the glosses explaining the words ex quantitate already took quantitas as an abstract, referring to a number of interchangeable (fungible) objects, interlinear gloss Trier MS 838 fol 179r: “ex  quantitate: id est ex numero; y.” Other glosses seem to understand quantitas as a specific  (corporeal) amount of coins. For both views, see Taeuber, Geld und Kredit (n 21) 332 n  919. On  the medieval concept of quantitas, see W Ernst, “Die Konkretisierung in der Lehre vom Gattungskauf”, in W Schön (ed), Gedächtnisschrift für Brigitte Knobbe-Keuk (1999) 49 at 57-60; W Ernst,  “Kurze Rechtsgeschichte des Gattungskaufs” (1999) 7 Zeitschrift für Europäisches Privatrecht  583 at 612-615. On the concept of fungible objects in the Middle Ages, see T Rüfner, Vertretbare Sachen? Die Geschichte der res, quae pondere numero mensura contant (2000) 74-92.  28  Marginal gloss Stockholm MS B680 fol 192v: “Ergo ex utroque, sed magis ex quantitate quam ex  substantia […] p”; taken from Taeuber, Geld und Kredit (n 21) 332 n 922. 

the glossators’ monetary law

227

read it as “as well ... as”.29 The assumption of a dual use of coins was in line  with the real-life situation that saw coined silver easily converted into bullion  (or silverware of all sorts) and vice versa. It should be noted that in a fully  monetarised  economy,  such  a  dual  use  of  coins  is  an  oddity,  money  being  defined as a medium of exchange, circulating by tale and hence – ideally –  regardless of the inherent substantive value of the coin. It was only later that  glossators overcame the dual-use doctrine and began to define money by its  circulatory function, negating the use ex substantia and hence reading non tam / quam as “not as ... but as”.30  Azo held that the technical usability of a coin’s metal was as important as  the usability as a means of exchange.31 Accursius took Azo’s statement a step  further in relying, for the usability as a means of exchange, on the technical  usability of a coin’s metal, thus making one of the uses (ex substantia) the  precondition of the other (ex quantitate): gloss praebet to D 18.1.1 Note that a coin has two types of usefulness – one deriving from its inherent utility  and the other from its ownership. The first derives from the utility of its substance,  because a single coin has as much value as the quantity of silver it contains. The  second derives from its value, because the value of a coin is level with an equivalent thing, and thus equality in value arises through a coin.32

The  notion  that  “a  single  coin  has  as  much  value  as  the  quantity  of  silver  it  contains”  (tantum valet unus nummus quantum argenti tantundem in massa)  was  to  become  a  key  reference  for  the  glossators’  (and  postglossators’) approach to currency policies. There was a general mistrust of coins  functioning as money proper, that is, circulating by tale as units for payment.  Rather, coins were commodified, defined as publicly-certified pieces of silver.  Since coins were valued according to their inherent metal value, they could  not – the glossators held – be given a nominal value exceeding their metal  29 Azo’s marginal gloss “dominumque” ad D 18.1.1: “Id est, utilitatem sui ipsius dominii prebet ex  duobus. Hoc est ipsa substantia, quia tantum potest tibi esse carum aurum vel argentum quod in  numo, ac si esset in massa. Item potest esse carum, quia per pecuniam facile est omnia invenire.  Azo.” BamSB MS iur 11 fol 209r. For variants in other MSS see Taeuber, Geld und Kredit (n 21)  334 note 922. 30  Taeuber, Geld und Kredit (n 21) 332 n 921, with references to Jacobus de Puteo and Hieronymus  Butigella. 31  See n 29 above. 32  “Nota quod in duobus prebet utilitatem: sui usus, et sui dominii. Primo ex substantia quia tantum  valet unus nummus quantum argenti tantundem in massa. Secundo ex quantitate, quia aequiparatur  quantitas  nummi  aequivalentiae  rei,  et  sic  per  nummum  fit  aequalitas  in  quantitate.”  Wording as in the edn of Lyon, 1618. The wording is not substantially different in earlier MSS of  the Digestum Vetus, e.g. Basel, MS CI4. See, also for variants in other MSS, Taeuber, Geld und Kredit (n 21) 334 note 925.

228

the creation of the ius commune

value.  This  results  in  a  specific  demand  made  on  monetary  policy-makers  based on the reading of D 18.1.1: rulers may not deceive the public by minting  underweight coin.33 Ideally, from the legists’ point of view, there should be  no seignorage at all.34 Canonists were to support a slightly more liberal, or  perhaps more realistic, approach, allowing for a “modest” seignorage.35 (3) Coins for bullion? The glossators’ environment knew two different means of exchange, denarii  and  bullion.  Could  a  debt  stated  in  bullion  be  discharged  by  payment  in  coins?  A  dispute  can  be  established  from  the  glosses  to  D  12.1.2,  the  lex mutuum,  which  is  a  key  introductory  text  to  the  law  on  loans.36  One  thus  turned to the law of loans, even though the debt expressed in bullion need  not have been based on a loan.  Hugolinus  found  bullion  and  coins  to  be  two  different  genera.  Repayment by coins would be numerus pro pondere, hence inadmissible. Azo (and  Accursius quoting him), on the other hand, argued that bullion and coin were  of the same genus, provided that the fineness was the same. gloss per solutionem to D 12.1.2.1: If  therefore  I  shall  have  received  money,  will  I  have  the  authority  to  return  a  measurement of cereal or a weight of silver? I answer “No” following the argument  below in D 46.1.42. It is otherwise if I shall have received silver, for then I return  money, as argued below in D 34.2.1. But the silver in the money should not be  mixed with base metal. Azo.37

Having  first  excluded,  from  the  scope  of  loan  contracts,  an  agreement  to  pay  money  for  return  of  cereal  or  bullion  (unminted  silver),  Azo  allowed  the  repayment  in  coins  for  a  loan  made  out  in  bullion,  provided  that  the  33 Taeuber, Geld und Kredit (n 21) 334.  34 Seignorage is the difference between the costs of minting and the nominal, the face value of the  coin. It is the seignorage which made minting a profitable, often hugely profitable, business. On  seignorage, see Sargent and Velde, Big Problem (n 15) 50-63. 35  On X 2.24.18, see T Bisson, “Quanto personam tuam (X 2.24.18): Its original significance”, in S  Kuttner (ed), Proceedings of the 4th International Congress of Medieval Canon Law [= MIC Ser  C Subsidia 5] (1976) 229; F Wittreck, “Conservare monetam. Geldwertstabilität im hochmittelalterlichen Aragon im Lichte der Dekretale ‘Quanto personam tuam’ (1199)”, in A Weber (ed),  Währung und Wirtschaft. Das Geld im Recht. Festschrift für H J Hahn (1997) 103; F Wittreck,  Geld als Instrument der Gerechtigkeit. Die Geldrechtslehre des Hl. Thomas von Aquin in ihrem interkulturellen Kontext (2002) 103-104. 36  Taeuber, Geld und Kredit (n 21) 161 n 441, 196. 37  BamSB MS jur 11 fol 145r in margine (R61 m): “Si ergo pecuniam accepero, numquid frumentum  in  mensura  vel  argentum  ad  pondus  reddere  potero?  Respondeo,  non,  argumento  infra,  de fedeiussoribus l. si ita fedeiussorum accepero [D 46.1.42]. secus si argentum accepero tunc enim  reddo pecuniam, ut infra, de auro et argento legatis l. Ia [D 34.2.1]. argenteam tamen non cum  here  mixtam. Azo.”

the glossators’ monetary law

229

coins themselves were pure. Azo’s prevailing opinion is very much in line with  the general focus on silver content. The dissenting opinion of Hugolinus is  preserved in the gloss pondere as recorded by Accursius:  Likewise, what if I shall have received silver of a particular weight? Do I have the  authority to return money? I answer “Yes”: not, however, if it is corrupt, as below  at D 34.2.1.1, D 34.2.9, D 34.2.19.1 and C 8.53(54).35.1. But Hugolinus says “No”,  as below at D 46.3.99, and he says it may have been particularly provided for in  advance in a final will and C 8.53(54).35.1 speaks in particular favour of gift ...38

(4) Exchangeability of denarii from different cities? For the exchangeability of various coins, the glossators found a text in the  Digest which was “spot on”. D 46.3.99 – then referred to as the lex Paulus –  stated that one need not accept payment in different coins, should this be to  one’s detriment as recipient.39 There was a minor dispute over the issue of  whether an impending detriment needed to be shown in concreto in order  to repudiate the unwanted coins, or whether a different type of coin in itself  could simply be seen as detrimental. In the long run the position prevailed  that best matched the general approach of the glossators. This was that it is  the silver that counts, so that, provided the weight of silver is right, there is  no reason to object to a different minting of the coin. This position, however,  was to be fully developed only by the postglossators.40  C. GLOSSATORS ON COIN DEBASEMENT (1) Bulgarus and the case of the Bolognese bushel The issue of the Bolognese bushel, though not in itself concerned with money,  was to exert great influence on the history of medieval monetary doctrine.  There was a lack of uniform and stable standards of measurement. Regional  standards of measurement were fixed and often changed by local statutes.  One unit of measurement of capacity was the bushel. The obvious question  arose of how a governmental redefinition of the bushel affects the obligation  38 Wording as in the edn of Lyon, 1618 (n 32): “Item, quid si accepero argentum ad pondus: nunquid  potero  reddere  pecuniam?  Respondeo,  sic;  non  tamen  ere  contaminatam,  ut  infra,  de auro et argento legatis l. I § finali [D 34.2.1.1] et l. cum auri certum [D 34.2.9] et l. cum aurum § proinde  [D 34.2.19.1] et C. de donationibus l. si quis argentum § I [C 8.53.35.1]. Sed H dicit,  quod non, ut infra, de solutionibus l. Paulus respondit creditorem [D 46.3.99] et predicta in ultima  voluntate specialia dicit. et in 1. si quis argentum [C 8.53.35.1] speciale dicit favore donationis ...” 39  On this text, and its real-life background, see Ernst, “Rationalia ad D. 46.3.99” (n 14); but see,  contra Ernst, B Kupisch, “D. 46,3,99”, in H Altmeppen, I Reichard and M J Schermaier (eds),  Festschrift für R. Knütel (2010) 617. 40  Sargent and Velde, Big problem (n 15) 77; Taeuber, Geld und Kredit (n 21) 163-164.

230

the creation of the ius commune

to return a quantity in kind. Our earliest source for the quaestio concerning  the Bolognese bushel is the Quaestiones Dominorum Bononiensium;41 from  there, we can trace it to Pillius42 and to Albertus Gandinus.43 Quaestiones Dominorum Bononiensium, Quaestio 60 Titius granted his farm to Seius to be cultivated according to the agreement that  he,  Titius,  should  annually  receive  a  specific  amount  of  corn,  to  be  measured  according to Bologna standards. Afterwards, however, the Bolognese reduced the  size of their measure, and even enacted a penalty if anyone measured by other  standards of measurement. The question is whether Titius can sue for the agreed  amount of corn, to be measured according to the earlier measurement, and if he  were to sue for it, whether he would incur the penalty established. Bulgarus said it  ought to be decided in favour of the plaintiff. The action on lease establishes this. Defendant Plaintiff44  D 45.1.49   D 50.17.34 D 40.5.24  D 44.7.22 D 16.3.1.9  D 32.75 D 34.2.40  D 12.1.22 D 4.4.3.2  D 2.14.44 D 2.11.7  & D 2.14.10 D 31.88 D 18.1.77; D 18.1.69; D 18.1.34 Nov 115 praef45

To  paraphrase  the  text:  Seius  thus  contracted  to  cultivate  Titius’  land  on  the terms that Titius was to receive a certain quantity of the harvest, to be  measured  in  Bolognese  bushels.  Subsequently  the  Bolognese  reduced  the  41 Although  Bologna  is  referred  to,  we  do  not  know  whether  this  was  a  real-life  case  or  just  an  academic exercise.  42  Pillius, Quaestiones Sabbatinae (1560), in CGIC IV (1967) 145-147 (quaestio 79). 43  Albertus Gandinus, Quaestiones Statutorum, in BIMAE III 193. 44  The texts on which the plaintiff and defendant founded their arguments were originally thus in  columns. 45  Paris MS lat 4603, in BIMAE 2nd edn I 246: “Titius dedit agrum suum colendum Seio hoc pacto, ut  unoquoque anno certam modiationem frumenti acciperet, metiendo scilicet mensura bononiensi.  Postea vero Bononienses mensuram diminuerunt, et penam etiam statuerunt si quis de cetero alia  mensura metiretur. Queritur an Titius possit petere modiationem constitutam, metiendo scilicet  mensura prestita, et si petierit, an incidat in penam a talibus constitutam. BULGARUS ait pro  actore iudicandum. Hic proponitur actio ex locato./ A/ D. de verborum obligationibus l. cum filius [D 45.1.49]./ D. de fideicommissariis libertatibus l. generaliter [D 40.5.24]./ D. depositi l. I § si quis servum  [D  16.3.1.9]./  D. de auro et argento legatis l. medico  [D  34.2.40]./  D. de minoribus l. Denique § scio etiam [D 4.4.3.2]./ D. si quis cautionibus l. Si quis [D 2.11.7]./ D. de legatis II l. penult. [D 31.88]./ D. de contrahenda emptione et venditione l. in lege [D 18.1.77], l. Rutilia [D 18.1.69], l. si in emptione [D 18.1.34]./ In Auth. tit. ut cum de appellatione cognoscitur,  constitutio pervenit [Nov 115 praef]./ R/ D. de regulis iuris l. Semper in stipulationibus [D  50.17.34]./ D. de actionibus et obligationibus l. cum quis in diem [D 44.7.22]./ D. de legatis III l. Nummis [D 32.75]./ D. si certum petatur l. vinum [D 12.1.22]./ D. de pactis l. cum in eo esset [D  2.14.44] et l. Rescriptum [D 2.14.10].” (The oblique strokes correspond to line-breaks in the main  text).Cf also Grenoble MS no 626, in BIMAE I (supplementary volume) 21.

the glossators’ monetary law

231

quantity  of  the  bushel,  and  made  it  a  statutory  offence  to  use  any  other  standard of measurement. The question is whether Titius can demand the  agreed  quantity  in  old  bushels,  and  if  he  does  so,  whether  he  suffers  the  punishment laid down. Bulgarus favoured judgment for the plaintiff, thereby  deciding that the old bushel must be used.  The  argument  advanced  for  this  conclusion  needs  to  be  extracted  from  the  cited  leges.  Most  of  these  texts  address  issues  resulting  from  changes  over time. We select some references to get the general idea. D 18.1.69 deals  with the sale of a pond including a strip of land of a certain width around it.  It was held that an increase in the surface of the pond owing to a rise in the  water level did not affect the extent of the land sold. In D 34.2.40 a bequest  of  the  testator’s  “travelling  silver”  is  taken  to  mean  the  silver  used  by  the  testator while travelling at the time the last will was made. D 50.17.34 and  D  32.75  deal  with  issues  of  interpretation  arising  from  terms  understood  differently  in  different  locations.  The  idea  to  be  taken  from  these  texts  is  that  the  place  where  the  contract  was  concluded  controls  the  language  of  the contract. Taken together with the previously cited texts the general idea  is that the time (and place) of contract making must be taken into account.  It is remarkable that the case of the changed bushel is tackled as an issue of  interpretation of a contract, given that these cases could equally well be seen  as raising such questions as of distributive justice or change of circumstances.  (2) Pillius’ Quaestio “Quidam creditor Lucenses” Bulgarus’  solution  for  the  case  of  the  changed  bushel  was  applied  to  the  case  of  debased  coin.  The  issue  is  elaborately  discussed  in  one  of  Pillius’  quaestiones, dating from around 1180, often referred to as Quidam Creditor Lucenses.46 It is one of his quaestiones Sabbatinae – classes given on a Saturday.47 This quaestio is the central achievement of the glossators’ efforts in the  field of money, the pinnacle of early glossators’ work, decisively shaping the  further development of the legists’ doctrine. The case had arisen in Lucca.48 The cities in Tuscany used the denarius (or  denaro), a coin which was still modelled after the Carolingian denarius. At  first, the coin was minted only in Lucca, until, sometime shortly after 1150,  Pisa,  which  was  in  an  economic  upswing,  started  minting  the  same  coin.  46 Text with all variants given by Taeuber, Geld und Kredit (n 21) 311; Taeuber has also most meticulously analysed the quaestio and its prehistory: Geld und Kredit (n 21) 119-148, 216-224. For all  details not addressed here, the reader may therefore wish to consult Taeuber. 47  On the Quaestiones Sabbatinae, see Lange, RRM I 234-235.  48  On the economic context, see Sargent and Velde, Big Problem (n 15) 139.

232

the creation of the ius commune

From then on, both cities competed for dominance of the regional Tuscan  coin market, some cities opting for the Tuscan (or Luccan) denarius, some for  that of Pisa. The coins of both cities suffered a debasement by half or more.  The following question was posed by Pillius. If coins from Lucca are given as  a loan and, before the repayment is due, debasement occurs, can the creditor  demand payment in old denarii or must he accept new ones? Pillius opted for  payment in the old coins. Many leges are referred to and arguments advanced  for creditor and debtor alike. While the whole set of arguments cannot be  laid out here, we shall analyse Pillius’ solution and the key arguments used  in the quaestio.  Pillius, in his solutio, relies first and foremost on the opinion advanced by  Placentinus49 on the basis that the depreciated coins are reproba in partem.  This referred to a text of the Digest stating that pecunia reproba cannot be  used for discharging debts, D 13.7.24, the lex eleganter.50 This is supported  by the rule that the debtor is not discharged by tendering only a part of what  is due, D 50.16.114, the lex solvendo. Pillius does not seem convinced that the  decision depends simply on choosing between two controlling moments in  time, that of the datio and that of repayment of the loan,51 the two opposing  rules being tempus dationis inspicitur52 and tempus solutionis inspicitur.  The key argument advanced for the creditor was D 12.1.3; then known  as the lex cum quid. Pomponius stated that, in the case of a loan, the debtor  may not return things of lesser quality (out of the same genus): for example,  he may not return new wine for old. This holds, Pomponius says, even in the  absence of a particular stipulation to this effect (a cautio) – this was in fact  the very legal issue at hand – because, he says, whatever seems to be agreed  upon can be seen as being stipulated.53 And it seems, Pomponius concludes,  that parties (typically) agree ut eiusdem generis et eadem bonitate solvatur, qua datum sit: items must be returned of the same genus and of the same  quality as had been given.  49 “Dominus p. dicit.” 50 As to the legists, this was already noted in an early gloss of Martinus ad D 13.7.24 found in BamSB  MS jur 12 fol 139r: “Reprobam pecuniam non liberare solventem. M(artinus).” Reading this gloss  narrowly, it only stated that the old coins, when officially out of circulation, can no longer be used  for payment. One does not learn, from D 13.7.24.1 or from Martinus’ gloss, how the debtor can  discharge the obligation instead. Could he pay with new, typically debased coins, and was it sufficient to pay the initial number of this coins or did he have to make up for the debasement? 51  A concluding remark of doubtful authorship (“dominus tamen – fin”) seems to hint at a distinction  depending on whether interest was agreed upon or not, perhaps assuming that the risk of debasement is taken into account in the rate of interest. 52  Or: “tempora primitiva servari debent”. 53  On the original meaning of D 12.1.3, see, most recently, U Babusiaux, Id quod actum est (2006)  61.

the glossators’ monetary law

233

The debtor invoked the absence of a coinage clause by which the creditor  could have insisted on repayment being made in old coins. He could try (or  did try) to argue, one learns,54 that he was entitled to use the money in circulation at the time of payment, because he had to render objects belonging  presently to the genus agreed upon (relying on sit instead of fuit in D 12.1.3).  Payment  in  today’s  denarii  would  hence  meet  both  requirements  of  same  genus and same qualitas, since, within the genus of the day, the quality of the  denarii was perfectly good. Indeed, another Roman-law text, D 17.1.52, here  also referred to, allowed the debtor, in the absence of a specific stipulation,  to use less-than-average quality in order to discharge a generic obligation.  The detailed discussion is largely devoted to the proper understanding of  eiusdem generis et eadem bonitate. The first issue addressed was whether the  old and new coins were from the same genus. A similar issue, in the interpretation of D 12.1.3, had previously been dealt with extensively, and one  can assume that the older discussion was known to Pillius and his students.55  The  issue  dealt  with  in  earlier  glosses  to  D  12.1.3  turned  on  the  example  given by Pomponius, namely the inadmissibility of tendering new wine for  old. The glossators had asked in what respect new wine was of lesser quality,  and whether, inversely, one could tender old wine for new.56 Most important,  perhaps,  was  the  question  whether  the  new/old  issue  was  one  of  genus  or  one  of  quality  (within  the  genus).  Some  glossators  tended  to  assume  that  wine  of  different  ages  constituted  different  genera;57  but  Pillius  does  not  seem to have subscribed to this view, which was so obviously out of tune with  Pomponius’ text.  In  another  quaestio,  Pillius  had  also  dealt  with  the  case  of  the  changed  bushel, and it is there that he addressed the public-law aspect of the problem.58  He approves Bulgarus’ solution, and separates the issue at public law from  that at civil law. The statutory regulations were future-oriented, hence not  retroactive, and anyway the issue was not about what actually happened in  54 Quidam creditor Lucenses, from “respondit debitor” on. 55 Taeuber, Geld und Kredit (n 21) 129-133.  56 We only give Azo’s gloss, which is already based on an extensive prior discussion, BV MS lat 1408  32m: “Quid si pro novo reddere vetus velit? R Job non liberatur, quia  non videtur eque bonum reddere, ut infra Institutionibus de actionibus § huic autem qui [J Inst  4.6.33d].  Sed  contrarium  est  argumentum  supra, si servitus vendicetur l. si forte § ultimo  [D  8.5.6.7] et infra, de verborum obligationibus ubi non [D 45.1.75]. Nam ibi dicit (iureconsultus),  quod est optimum, ipsum quoque bonum est. Hoc tamen, si eiusdem qualitatis est, ut utrumque  novum vel vetus. Secus si datum est vetus, redditur novum. Azo.” See Taeuber, Geld und Kredit  (n 21) 136 n 369 with variants. 57  In  the  later  words  of  Baldus:  “illud  bonum  non  subalternatur  illi  optimo,  et  est  ac  si  essent  diversorum generum”. See Taeuber, Geld und Kredit (n 21) 136.  58  Quaestio 79 in Pillius, Quaestiones Sabbatinae (n 42) 146.

234

the creation of the ius commune

the  city  (state),  but  about  what  ought  to  happen.  This  is  also  a  nice  little  contribution to the delineation of statutory law and ius commune. (3) Azo’s brocard The position first established by Pillius became part and parcel of the glossators’ monetary law. This was due to the authority attaching to the name of  Azo, to whom a brocard is attributed in the Brocardica sive generalia iuris.59  This work was published under Azo’s name, and may be traced, at least for  the greater part, to Otto Papiniensis, and dated around the end of the twelfth  century.60 We may here disregard the disputed issue of the editorship of this  collection, since the specific brocard we are looking into is assigned to Azo.  The brocard starts with the juxtaposition of two opposing statements, each  accompanied by a host of references, mostly to texts of the Digest and Code,  but  also  to  other  brocards  within  the  same  volume:  “the  same  measurement  or  coined  money  is  owed,  as  was  owed  at  the  time  of  contracting”  (eadem mensura vel moneta debetur, quae erat tempore contractus) and “the  reformed money or measurement is owed, whether it be greater or lesser”  (superveniens moneta vel mensura debetur, sive maior sive minor sit).  In  his concluding argument, Azo without hesitation opts for the rule of eadem mensura.  He  holds  (as  Pillius  had  held)  that  intervening  statutes  have  no  retroactive effect on contracts already operative.61 Texts referring to legacies  are eliminated by Azo on the ground that they depend on the wording of the  last will. Azo’s last, and perhaps most troublesome, point concerns D 12.1.22,  the  lex vinum,  with  Julian’s  statement  “the  time  of  payment  is  examined”  (tempus solutionis inspicitur). The glossators’ dealings with this Digest text  deserve a closer look.62  D  12.1.22,  the  lex vinum,  is  about  the  Roman  condemnatio pecuniaria.  It addresses the question of the moment in time on which the judge must  base  his  money  estimate  of  performance  which  the  defendant  has  unlawfully  denied  the  plaintiff  (aestimatio).  Julian  discusses  a  loan  of  wine,  the  borrower being sued for redelivery. The question arose whether the value  of the wine should be determined at the time its redelivery was due, or at  the time of consolidating the law suit (the litis contestatio), or at the time of  59 Brocardica sive generalia iuris (1567) 136, rubrica XII “De Contractibus”. For a critical edition,  recording the variant readings of the different MSS, see Taeuber, Geld und Kredit (n 21) 328-330.  The solutio of the brocard can also be found (anonymously) in Turin MS F II 14 (AA) fol 109r;  Taeuber, Geld und Kredit (n 21) 330. 60  For details on the issue of authorship, see Lange, RRM I 145.  61  “Statuta post contractum facta futuris contractibus dant formam, non praeteritis.”  62  For more details, see Taeuber, Geld und Kredit (n 21) 184-193, 229.

the glossators’ monetary law

235

the judge’s ruling. If a fixed date was agreed upon for the time of delivery,  according to Julian this should also be the point in time for the aestimatio.  (The moment of litis contestatio was only to be decisive if no time had been  agreed upon.) How  was  tempus solutionis inspicitur to  be  reconciled  with  the doctrine based on cum quid that the quality was fixed by (and thus at the  time of) the making of the loan? The lex vinum was quite a stumbling block,  not least because the glossators did not understand the original context of  aestimatio, since they did not subscribe to the strict necessity of condemnatio pecuniaria.63 We shall not go into the details of the way in which the glossators tackled this contradiction.64 Azo distinguished the lex vinum from the lex cum quid on the ground that, in the case of the lex vinum, the measure was  unaltered. While D 12.1.3 confirmed that the object matter of the datio – with regard  to genus and qualitas – determines the proper object matter of repayment,  Azo’s brocard refers to the time the contract was concluded as controlling  the issue of the coins to be used for repayment.65 With regard to loans this  makes  no  difference,  since  the  Roman  loan,  as  a  so-called  real  contract,  became binding only on the datio.66 However, the applicability of the test is  now considerably widened, and the rule could be relied upon with regard to  other contracts as well. Shifting from “payment” to “contract” was vital for  extending the scope of the rule to contracts other than loans. The idea is now:  the contract determines – perhaps indirectly, by reference to the items made  over for the “real” contract – the amount of silver the creditor can reclaim. Azo’s brocard explicitly equates the issues of debased coinage and measurements  altered  by  statute.  The  brocard  addresses  and  solves  both  types  of  case in a uniform fashion. There is no distinction between diminutions and  enlargements  of  measurements,  and  thus  the  brocard  does  not  invariably  favour  the  creditor;  but  given  that  changes  of  measurements  or  coinage  63 On  the  issue  of  specific  performance  (as  opposed  to  condemnatio pecuniaria),  see  T  Repgen,  Vertragstreue und Erfüllungszwang in der mittelalterlichen Rechtswissenschaft (1994).  64  It should be noted that this text, like D 12.1.3 (lex cum quid), did not deal with money. As with  D 12.1.3 (lex cum quid), it would not have made sense at all, for a Roman lawyer, to apply D  12.1.22, the lex vinum, to a monetary obligation, since coins then were not subject to free aestimatio, “since the value of coins is fixed” (cum certa sit nummorum aestimatio), as D 12.3.3, the  lex nummis depositis,  explained.  In  Roman  times,  a  number  of  coins  could  well  be  subject  to  aestimatio, by using another coin with a different face-value, but then the aestimatio was indeed  certa, because the Roman currency system defined the various face-values of its different coins as  handy multiples/fractions of each other. For the medieval lawyers, the lack of a currency system  with coins of different face-values meant that little could be made out of D 12.1.3 (lex cum quid).  65  This idea goes back to Pillius. See section C.(2) above.  66  On Roman financial loans as real contracts, see W J Zwalve, “The Past and Future of ‘mutuum’”,  in Past and Future of Money (n 8) 17.

236

the creation of the ius commune

normally worked against the creditor, the effect of Azo’s brocard may have  been more important for the creditor than for the debtor. Later printed editions inserted another element of Azo’s position, which  had initially only been part of the solution, into the brocard, which came to  be referred to in this language: The  same  measurement  or  coined  money  is  owed  as  was  owed  at  the  time  of  contracting; or, if other money is tendered, the loss must be accounted for.67

Azo’s brocard was to feature as the controlling formula for the coming centuries. From 1250 onwards, we find Azo’s doctrine confirmed in the Bolognese  statutes: Of quality of money tendered: We enact that, if money is owed on account of a  dowry or of any other legitimate past event, let he, to whom the money is owed,  receive money of the same quality that it will have had at the time of the contract.68

(4) Evaluating the glossators’ achievements In the ius commune the issue of depreciation of coinage was linked to the lex cum quid, D 12.1.3. How appropriate was this text for the issue at hand? It is  disturbing that D 12.1.3 became a cornerstone of the monetary doctrine of the  medieval learned lawyers. The text bears no relation to money at all. True, the  text is about mutuum, but money was by no means the only possible subject  matter of the Roman mutuum, which covered financial loans and non-cash  loans alike. The quality issue dealt with by Pomponius could arise with regard  to all possible objects of mutuum, except money. In a well-ordered monetary  environment, with coins circulating by tale, the question of the quality of the  substance used for discharging the loan, can hardly be raised. Pomponius’  question was meaningful, in the Roman context, only for non-cash loans, for  which Pomponius gives a loan of wine as an example. Of course, there was  nothing wrong with extending the scope of D 12.1.3 to financial loans as well,  in times when coins indeed varied in quality.  Nor did D 12.1.3 really deal with changes over time.69 Rather, the text is  about properly selecting, from a variety of presently available commodities,  67 “Eadem mensura vel moneta debetur, quae erat tempore contractus; vel si alia moneta praestetur, habeatur ratio damni.” For the editions, see n 59 above. 68  Statuta communis Bononiae (1250) lib IV rub 25, in L Frati (ed), Statuti di Bologna dall’anno 1245 all’anno 1267  (1869)  I  406:  “Cuius  bonitatis  pecunia  solve  debeat.  Statuimus  quod  si  ex  causa dotis vel cuiuscumque alio casu legittimo ex preterito tempore peccunia debeatur, cuius  bonitatis fuerit contractus eiusdem bonitatis fuit et accipiat ille, cui peccunia debetur.” Cf also,  for Bolognese rulings predating this statute, Taeuber, Geld und Kredit (n 21) 243-244. 69  Pillius perhaps understood this better than his successors.

the glossators’ monetary law

237

those items which are fit to be used for discharging the borrower. However,  the  example  given  by  Pomponius  –  that  of  old  wine/new  wine  –  made  it  possible  to  read  a  temporal  aspect  into  the  text,  although  this  was  hardly  more than an associative link triggered by the terms “new” and “old”. For  Pomponius,  old  and  new  wine  were  different  commodities  available  at  the same time: new wine was not the “debased” successor of the old wine.  Reading the temporal aspect into D 12.1.3 the formula arrived at was simple:  tempus dationis inspicitur. Mobilising  the  lex cum quid  (D  12.1.3)  to  adapt  monetary  obligations  to  currency  turbulences  raised  a  further  problem.  The  test  established  by  reliance upon D 12.1.3 works for financial loans, but not for monetary obligations arising out of other contracts. Monetary obligations can arise from longterm  leases,  and  indeed  from  all  obligations  to  make  recurring  payments,  originating, for example, from public law (pensions) or last wills, and, in all  of  these,  the  considerable  time-lag  between  the  initial  agreement  and  the  time  for  payment  renders  the  obligation  susceptible  to  turbulences  in  the  currency. The test established by Pillius – later adopted by other glossators  – does not work for such obligations. Perhaps Pillius was already conscious  of the problem, because he subtly shifted his language from tempus dationis inspicitur to tempora primitive servari, the latter phrase being applicable to  contracts other than those originating in a datio. Azo’s brocard was unambiguously applicable to all sorts of contract. Another lasting effect of the glossators’ approach was to place the issue of  debasement firmly within the framework of interpretation of contracts. But  for one small problem, this was a true achievement. Treating the problem as  one of interpreting the contract leaves the parties the freedom – and indeed  encourages them – to specify, in their contract, which coins should be used.  Such clauses, already known in antiquity and still employed in the time of  the  glossators,  raise,  with  the  glossators’  blessing,  the  problem  of  “metallism”  versus  nominalism.  If  the  law  lets  the  parties  decide,  the  only  issue  remaining  is  the  default  rule.  For  the  glossators,  the  default  rule  was  that  money payments must make good the amount of silver represented by the  sum owed, while a contractual provision was required to render acceptable  the  tender  of  coins  of  the  same  denomination.  Later,  when  one  turned  to  nominalism, the situation was reversed, and nominalism needed to be ousted  by coinage-clauses, like the gold-clause, or – later on – by index-linking the  debt. The small problem is that, if one sees the issue as effectively controlled  by sensible interpretation of the contract, there is no solution for obligations  arising otherwise than by contract. 

238

the creation of the ius commune

The  glossators  put  money  on  the  same  level  as  other  instruments  used  to  make  measurements  (mensura vel moneta).  The  linking  of  money  to  units  of  measurement,  typically  statutorily  defined,  is  a  key  feature  of  the  medieval  approach  to  currency  issues.  It  was  advocated,  most  notably,  by  Thomas Aquinas.70 We shall meet the money/measurement parallelism again  in the texts of canon law.71 Seeing money as a kind of yardstick has become  a  commonplace,  the  story  of  which  has  been  looked  into  elsewhere.72  It  is  sufficient  to  note  that  putting  money  on  the  same  level  as  other  means  of  measurement  is  an  approach  that  does  not  have  modern  economists’  undivided support.73 Given the blurring of the distinction between coins and bullion, currency  issues were handled on the basis of a few Roman-law texts addressing issues  of non-cash loans. Solving these issues effectively on the basis of contractual  interpretation meant that there was no incentive to develop a specific legal  concept of money and provide solutions specifically adapted to the peculiarities of money as distinct from commodities.  It seems reasonable to assume that the nominalist position, described in  Azo’s brocard by the words superveniens moneta vel mensura debetur, sive maior sive minor sit (“the reformed money or measurement is owed, whether  it be greater or lesser”), was not introduced just for the sake of argument,  but was or had actually been subscribed to by one scholar or another. We  often find that, in the textual transmission of the glossators’ writings, minority  positions have been very successfully suppressed. The formation of the ius commune, requiring  the  development  of  new  doctrines  in  great  number  –  some  altogether  alien  to  the  Roman  law  –  may  have  involved  much  more  scholarly  dissension  than  we  now  tend  to  assume,  given  that  texts  such  as  the Glossa ordinaria convey the impression of relative harmony among the  glossators. 

70 Taeuber, Geld und Kredit (n 21) 227 n 648; on Thomas’ monetary doctrine(s), see exhaustively  Wittreck, Geld als Instrument (n 35). 71  See section D. below. 72  On this issue, see W Ernst, “Money as Measure and Measure for Money”, in M Welker (ed),  Monetarization and Standardization (2010) (forthcoming). 73  See, most notably, L Mises, Human Action: A Treatise on Economics (1949; repr 1996) II 327-333.

the glossators’ monetary law

239

D. CANON LAW ON COIN DEBASEMENT (1) Introduction Around the turn of the century, a number of decretals dealt with changes in  coinage or measurements. The cases that they were meant to settle were less  clear-cut than the cases which the civilians presented, in somewhat streamlined  versions,  to  their  students  in  the  classroom.  The  decretals  are  more  ambiguous,  the  monetary  issue  often  being  intertwined  with  other  issues  such as prescription. Over time, however, a number of decretals gravitated  towards  a  rule  similar  to  that  stated  in  Azo’s  brocard.  We  shall  list,  in  an  overview, this “network” of decretals, before looking for traces of a dissenting  opinion. (2) Decretals on instabilities in the fields of money and measurement (a) 2 Comp. 3.25.3 Querelam (later X 3.39.26) 2  Comp.  3.25.3,  issued  by  Urban  III  in  1187,74  was  the  first  decretal  that  upheld  the  position  that  an  older  standard  of  coinage,  established  by  a  year-long  use  of  a  certain  coin,  fixed  the  creditor’s  position  for  the  future  as  well.  The  dispute,  which  was  ultimately  resolved  by  the  decretal,  had  arisen because the creditor demanded payment in current Paris coin which  was actually better than the local coin that had been tendered and accepted  for several previous years. One must keep this case in mind if one wants to  understand why later references relying on 2 Comp. 3.25.3 stress that the old  coin is determinative, even if the new one is better. The decretal was reissued  with substantial alterations by Gregorius IX as X 3.39.26. (b) 3 Comp. 2.15.4 Quanto personam tuam (later X 2.24.18) A  key  decretal  turning  on  the  ruler’s  power  to  debase  coins  was  issued  by  Innocent  III  on  5  April  1199.  Since  the  decretal  has  been  intensively  discussed, there is no need for further exposition here.75 (c) 3 Comp. 3.37.3 ex parte  Compostellani (later X 3.39.18) This decretal, issued by Innocent III on 23 March 1206, dealt with difficulties arising from the lack of uniform standards of measurement. The decretal  takes a fairly liberal view, and shows some reluctance to require the use of  74 Taeuber, Geld und Kredit (n 21) 104-106. 75 Cf the literature cited in n 35. 

240

the creation of the ius commune

the contemporary bushel where the transaction was gratuitous, it not having  been established in the proceedings which bushel had been used earlier.76 (d) 3 Comp. 3.37.5 olim causam (later X 3.39.20) This is the first decretal to deal head on with an issue of debasement. It is  a  fortunate  coincidence  that  the  debasement  of  the  denarius  from  Lucca,  which was so extensively dealt with by Pillius, was also the occasion of one of  Innocent III’s decretals. Issued in the year 1200, we find the decretal in the  Compilatio Tertia (3 Comp. 3.37.5), and later unchanged in the Liber Extra,  referred to as olim causam (X 3.39.20).77  The bishop of Spoleto had a twice-yearly claim for payment against the  clerics of a particular  church  in  Rupo.  For  many  decades,  payment  of  the  claim,  which  was  for  three  denarii  of  Pavia,  had  been  made  using  denarii  from Lucca, since Pavia had discontinued the minting of its own denarii. The  denarius of Lucca had lost about half its value and was held to be only half of  an original Pavian denarius. The Pope held that the clerics must either use  the Pavian denarii or make good the difference. The case also raised the issue  of prescription – ignored here – since the “wrong” kind of coins had been  accepted over quite a period of time.  As to the monetary aspect, 3 Comp. 3.37.5 olim causam worked in favour  of the creditor, who is protected against debasement by upholding the coin  standard  established  by  the  earliest  payments.  Olim causam  thus  differed  slightly from the civilians’ doctrine. The creditor can insist on payment being  continued as before. The standard is not, as the civilians assumed, set by the  contract or – in the case of loans – by the coins given to the borrower. Rather,  one cannot depart from coinage already used in the past.  It was problematic if the old coins, to which the debtor was referred, were  no longer available or, as happened not infrequently, were no longer lawful  money. In olim causam we find, perhaps for the first time, a clear-cut answer  to this problem: if the old coins are no longer in use, the aestimatio is due.  The debt expressed in old coins is to be reckoned in new coins. By this way,  the debtor has to make good in full the debasement by paying a proportionally higher number of coins. We have seen that this idea had also found its  way into Azo’s brocard.  76 See Taeuber, Geld und Kredit (n 21) 85 nn 217, 219, 228. 77 See P Landau, “Die Bedeutung des kanonischen Rechts in der Geschichte der Geldschuld”, in G  Dilcher and N Horn (eds), Sozialwissenschaften im Studium des Rechts (1978) IV 165; Taeuber,  Geld und Kredit (n 21) 107-108; W Taeuber, “Geld und Kredit im Dekret Gratians und bei den  Dekretisten” (1954) 2 SG 445.

the glossators’ monetary law

241

X 3.39.20 olim causam was to become the chief canon-law text referred to  on the issue of disturbed currency developments, and as such has been relied  upon by civilians as well. We still find it advanced, as an argument against  nominalism, in the nineteenth century by Friedrich Carl von Savigny.78 The  decretal was also relied upon when the later monetary development saw the  introduction of tariffs for various coins, an issue that we shall not look into  here. (3) A nominalist minority position? There  are  plenty  of  lengthy  glosses  to  olim causam,  by  contemporary  and  later  authors,  and  olim causam  was  also  frequently  referred  to  in  other  money-related glosses within the canon-law corpora. It is interesting to note  that several of these glosses registered an alternative position which favoured  the use of contemporary coinage (and measurements), a position that, with  regard to money, we would refer to as nominalist. The key text upon which  this dissenting position was (or could have been) based was C. 32 q. 4 c. 6. To  understand why this was so we look into an anonymous gloss that Johannes  Teutonicus used, in his Glossa ordinaria, for C. 32 q. 4 c. 6: gloss tempori nostro to C. 32 q. 4 c. 6 It is argued that if heavier coined money is first circulating and later lighter coined  money circulates by command of the prince, I shall be freed from my obligation by  paying with the lighter. Argument based on D 18.1.71. Contrary texts: X 3.39.20;  D 12.1.3; 3 Comp. 2.15.4 [= X 2.24.18] 79

Why was this monetary statement (arguing that if a ruler decrees the introduction  of  a  lighter  coin  in  lieu  of  a  heavier  one  in  circulation,  I  shall  be  discharged by using the lighter one for payment) placed at C. 32 q. 4 c. 6 in  the first place, in the midst of the law of marriage? C. 32 q. 4 c. 6 is taken  from  St  Jerome,  who  argued  that  Abraham  had  found  favour  with  God  in  marriage,  whereas  today’s  virgins  find  favour  in  chastity:  Serviamus et nos legi et tempori nostro – “let’s meet our time and law as they seek us”.80 The  reader is assumed to apply this to the monetary issue of coin debasement. We  78 F C von Savigny, Das Obligationenrecht als Theil des heutigen römischen Rechts (Berlin 1851) I  476; cf W Ernst, “Kanonisches Recht in Savigny’s System” (2010) 127 ZRG, KA 275.  79  “Argumentum, quod si primo currebat fortior moneta et modo currit vilior de mandato principis:  liberor solvendi viliorem. argumentum D. De contrahenda emptione l. imperatores [D 18.1.71].  Argumentum contra extra, de censibus, olim [X 3.39.20], D. Si certum petatur l. cum quid [D  12.1.3] et extra III de iureiurando, quanto [3 Comp. 2.15.4 = X 2.24.18].” Also found as an anonymous gloss to the Bamberg MS of Gratian’s Decretum, BamSB MS Can 13 fol 207v. See Taeuber,  Geld und Kredit (n 21) 104-106. 80  Contra Iovinianum, lib 2.

242

the creation of the ius commune

are, the association with C. 32 q. 4 c. 6 suggests, to use the coins of our time,  regardless  of  whether  or  not  they  seem  debased  when  compared  to  older  coins. To support this position, reference is made to D 18.1.71, a text that  allows contracting parties to determine freely measure and price when selling  wine. The other (majority) position is explained as well. Does the gloss side  with the nominalist position? The placing at C. 32 q. 4 c. 6 suggests that this is  so, because the key notion of C. 32 q. 4 c. 6 is the supremacy of present mores  and  laws  over  those  of  the  past.  Later  editions  (including  Bartholomaeus  Brixiensis’ revisions) made sure, however, that the reader was referred to the  “proper” exposition of the law, namely by Bartolus and Alexander de Imola,  which rephrased the classical doctrine as stated in Azo’s brocard.81 Taueber has assumed that Johannes Teutonicus himself subscribed to the  nominalist position. Johannes’ own apparatus to the Compilatio Tertia does  not support this theory: Johannes Teutonicus’ gloss vel estimationem to 3 Comp. 3.37.5: It is argued on the School question that payment should be made in old money,  although  new  coined  money  may  be  better,  and  see  above  3  Comp.  2.15.4  [=  X 2.24.18] and above 2 Comp. 3.25.3 [= X 3.39.26] and D 12.1.3. For contrary  argument see C. 32 q. 4 c. 6, D 18.1.71. Johannes82

We find both positions referred to, the canon-law mainstream now additionally supported by a reference to the Digest’s lex cum quid (D 12.1.3) upon  which the civilian glossators relied as well. The juxtaposition of both views can also be found in a gloss by Vincentius  Hispanus83 to the Compilatio Tertia (at 3 Comp. 3.37.3): Vincentius’ gloss to 3 Comp. 3.37.3 It  is  debated  whether  performance  should  be  made  according  to  old  measurement,  not  to  new,  and  in  old  coined  money,  although  the  new  may  be  better,  as above 2 Comp. 3.25.3, provided the old is still in circulation. Otherwise, the  aestimatio  will  be  paid,  according  to  this  contrary  argument  in  C.  32  q.  4  c.  6.  Vincentius.84 81 See, e.g., the edition of 1582 published at Rome. 82 Admont Stiftsbibliothek MS 22: “arg. ad questionem dominicalem quod ad ueterum monetam  sit soluendum, licet noua sit melior, et supra de iureiur. Quanto, lib. eodem [3 Comp. 2.15.4 =  X  2.24.18]  et  supra  eodem Querelam, lib.ii.  [2  Comp.  3.25.3  =  X  3.39.26]  et  ff. si certum pet. l.iii [D 12.1.3]. Arg. contra xxxii. q.iiii. Quis ignorat [C. 32 q. 4 c. 6], ff. de contrahend. empt. Imperatores [D 18.1.71] jo.” Interim internet edn by K Pennington: http://faculty.cua. edu/pennington/edit323.htm, accessed 20 June 2009. 83  On him, see K Pennington, “The Decretalists 1190-1234”, in W Hartmann and K Pennington  (eds),  The History of Medieval Canon Law in the Classical Period, 1140-1234  (2008)  211  at  230-232. 84  BamSB MS can 20, found reproduced, with variants, in Taeuber, Geld und Kredit (n 21) 227 n 652:  “Argumentum, ad veterem mensuram solvendum est, non ad novam, et ad veterem monetam, 

the glossators’ monetary law

243

Here we meet again the parallelism between measure and money, a commonplace for civilians and canonists alike.  An earlier dispute had been settled in a way that could be read as subscribing  to a nominalist notion of money. The case is documented in the archives of  the Quedlinburg monastery.85 The monastery had undertaken, in 966, to pay  a yearly sum of “libra una argenti”, a pound of silver. In Germany, the lighter  mark replaced the libra. The abbess kept sending one mark, arguing that a  “libra” was now to be taken as a mark. The conflict was settled by arbitration,  the  archbishop  of  Mainz  being  the  arbiter.  He  decided  in  favour  of  the monastery, and the pope conceded that the monastery henceforth could  pay one mark of silver. The debtor thus could use for payment the number  of  coins  initially  fixed,  although  the  coins  now  being  paid  were  a  debased  version of them.  It is not unreasonable to assume that a “more” nominalist approach had  (or had had) some followers among the canonists as well. (4) The canon law’s classical position Early in the thirteenth century a canon-law doctrine had emerged that effectively coincided with the civilians’ doctrine epitomised in Azo’s brocard. In  a classical language  the  canon-law  rule  was  rephrased  by  Hostiensis  in  his  Summa (ca 1253), taking up the dispute underlying olim causam: Next, if some coinage has been in circulation over forty years, and for individual  years, cathedaticum has been paid in that money, and after forty years have gone  by the coinage is altered recourse must be had to aestimatio of the prior coinage,  unless prescription of payment in the second type of money be absolute … Therefore, payment ought to be made in the old coinage or in an aestimatio of it. And  you ought to observe the same if a measure of wine or oil of a specified amount is  owed, whether the standard of measurement be increased or indeed decreased.86

If  there  ever  was  a  nominalist  minority  position,  it  had  been  effectively  silenced. It would take roughly four centuries for the nominalist position to  licet nova melior sit, supra, eodem titulo, querelam, libro II [2 Comp. 3.25.3] dum tamen vetus in  usu sit. alias solvatur estimatio, ut hic argumentum contra XXXII q. IIII si quis ignoret [C. 32 q. 4  c. 6]. Vin.” 85  Taeuber, Geld und Kredit (n 21) 85; Taeuber, “Geld und Kredit Kredit im Dekret Gratians” (n  77) 462, quoting Antonius Udalricus ab Erath, Codex diplomaticus Quedlinburgensis (1764) 103  (not seen). 86  Hostiensis, Summa aurea lib 3 de censibus etc no 8 (1570) fol 304v/304r: “Porro si aliqua moneta  currit per annos 40, et singulis annos fuit cathedaticum solutum ad illam monetam, et elapsis 40  annis  mutetur  moneta,  ad  estimationem  primae  redeundeum  est,  nisi  prescriptione  solutionis  monetae secundi generis sit completa. [...] Ergo ad antiquam monetam, vel aestimationem eius  solvi debet. Et idem intelligas si census vini, vel oloei ad certam mensuram debetur, sive augeatur  mensura, sive etiam minuatur.”

244

the creation of the ius commune

be revived, notably by Dumoulin.87 It was to become the position to which all  advanced legal systems now subscribe. E. CONCLUSION: THE HUMBLE BEGINNINGS OF LEARNED MONETARY DOCTRINES It  is  now  generally  accepted  that  the  various  contributions  of  the  authors  of the scholastic age cannot be attributed to any shared uniform theory of  money.  There  was  no  such  thing  as  a  uniform  “feudal”  concept  of  money,  as once advanced by Emile Bridrey.88 Walter Taeuber, who has looked into  a  huge  number  of  relevant  glosses,  found  for  the  much  narrower  field  of  learned  lawyers  that  their  money-related  deliberations  did  not  result  from  a preconceived abstract theory of money.89 The surviving glosses therefore  cannot be put together in order to arrive at “the” general theory of money  prevalent in the age of the glossators. There was no “grand design” to which  all glossators would have subscribed, nor any concept of money over-arching  law, philosophy and theology. The glossators did not turn to ideas of distributive justice, which might easily have been employed to deal with the contractlaw problems caused by unforeseen currency turbulence.  As in other fields of law, the glossators were anxious to stay close to the  Roman texts. Each single lex was read on its own merits and interpreted in  a way to make sense of the text in the light of the glossators’ own monetary  environment. Mostly, the glossators relied on texts dealing with loan, disregarding the fact that monetary obligations arise out of other contracts as well.  The loan-related texts, upon which the glossators relied, did not deal with  financial loans at all, but with non-cash loans, because only in non-cash loans  is quality an issue.  The texts relied upon and the way they were construed betray a particularly  medieval understanding of money (coins). The commodification of money is  the  most  striking  feature  of  the  time.  The  power  of  coins  to  discharge  an  obligation was acknowledged only insofar as the coins effectively contained  silver. The power to discharge an obligation was not linked to the coins being  money – a mere means of exchange which can in turn be used to purchase  other goods. With a little exaggeration, one could say that the glossators did  not have any concept of money at all. Coins themselves were demonetised.  Owing to the commodification of money, the idea that money, functioning  87 On Dumoulin’s contribution, see W Taeuber, Molinaeus’ Geldschuldlehre (1928). 88 E Bridrey, Nicole Oresme: La théorie de la monnaie au XIVe siècle (1906; repr 1978). 89 Taeuber, Geld und Kredit (n 21) 331-339.

the glossators’ monetary law

245

as a medium of exchange, could call for theories and rules of its own did not  occur. To have a general (legal) theory of money would have been at odds  with the prevailing notion that coins were a commodity and governed by the  laws applicable to commodities in general.  The commodification of money – of coins – was not the glossators’ doing.  The  commodification  of  money  was  a  fact  of  life  for  which  the  glossators  merely  provided  a  matching  legal  approach.  Given  the  state  of  monetary  affairs,  the  distinction  between  precious  metal  –  bullion  –  and  “money”,  minted  coins,  could  not  be  perceived,  from  a  legal  perspective,  as  strictly  as  within  a  fully  developed  monetary  system.  To  abstain  from  nominalist  doctrines was a sheer necessity.90 Given that the silver content of coins was  in  flux,  not  only  every  now  and  then  but  practically  all  the  time,  the  idea  that coins, simply by their identical denomination, could be interchangeable  regardless  of  their  silver  content  would  have  been  ludicrous.  The  field  of  money  proves  that  the  glossators  were  capable  of  reading  the  Roman-law  texts in a way that extracted from these texts a contribution to the pressing  legal issues of the time. The glossators thus tackled – and mastered – the task  of  adapting  the  Roman-law  texts  to  a  monetary  environment  which  posed  questions unheard of in the time of the Roman lawyers. It is to be admired,  I think, that all issues, great and small, were looked into and answers were  provided,  based  on  ingenious  interpretations  of  sometimes  unrelated  law  texts of antiquity. It would take centuries for later doctrinal lawyers to undo  the legal commodification of money and to restore an approach to money as  a specific means of exchange. The canon-law doctrine emerged rather erratically, owing to the peculiar  facts of the actual cases which the popes had to address. By and large, canonists and civilians were marching in the same direction, but not quite to the  same tune. While the civil lawyers saw the debasement issue as one of interpretation of contracts and thus employed means of interpretation to establish, as between the contracting parties, the “correct” standard of coinage, the  canon-law approach was more broadly based on a preference for old over new  coinage, the old coinage being preferred because it had already been relied  upon in the past. The idea of a parallelism between money and measure was  employed by civilians and canonists alike. Canonists occasionally advanced  90 Speaking  of  nominalism,  one  should  bear  in  mind  that  the  argumentative  juxtaposition  of  nominalistic and “metallistic” (or “valoristic”) concepts of money belongs to a much later period,  namely the eighteenth and nineteenth century, when a huge battle for and against paper money  was fought. If we speak of nominalism today, in an environment of pure token money, we have  yet another problem in mind, the problem of inflation. Debasement of commodity-money and  inflation, in times of token money, while showing similar effects, are different issues.

246

the creation of the ius commune

leges from the Corpus iuris civilis, and since these leges were the ones the  civilians relied upon for the issue of debasement, it is fair to assume that the  canonists  were  acquainted  with  the  legists’  monetary  doctrines.  Like  their  civilian counterparts, canon-law scholars abstained from grand theories. They  did not fully develop the pros and cons of nominalism or “metallism”. Since  debasement unsettles what can be seen as the just equilibrium of contracts,  one  might  expect  notions  of  distributive  justice  to  be  mobilised.  That  this  did not happen seems even more surprising for the canonist side, especially  in the light of recurring theoretical criticism of monetary nominalism voiced  later on from a specific canon-law perspective.91

91 It has been held, as recently as 1974, that canon law must take a stand against nominalism, G  Hubrecht, “La monnaie en droit canonique” (1974) 18 L’année canonique 115.

10  Citations and the Construction of Procedural Law in the Ius Commune Richard H Helmholz A. INTRODUCTION B. THE SOURCES (1) The Roman law (2) The canon law C. WORKING OUT THE DETAILS (1) The necessity for citation (2) Assignment of the day for appearance (3) The citation’s contents D. ENGLISH COURT PRACTICE E. CONCLUSION A. INTRODUCTION The subject of this volume is the creation of the ius commune. It is a large subject, even an unwieldy one. Where should a student begin? A possible starting point, the one taken in this chapter, is to select a vantage point and use it to examine in detail the ways in which the law was formed. It has long seemed to me that procedural law might provide a profitable view. So I picked it as a topic, finding in subsequent investigation that it does provide a good example of some of the ways by which the mature ius commune came into being. There were other ways, no doubt, and they may provide greater insights into the process. However, this one shines a light on the nature of the work done by the medieval jurists. This chapter concerns only one aspect of procedural law: that relating to citations, the process by which parties were summoned to appear before a court. The choice was not quite random. It was influenced by my own experiences in working through the records of the English ecclesiastical courts, some of which are the source of a brief description of court practice placed at the 247

248

the creation of the ius commune

end of this chapter. But it was more than that. Chaucer’s summoner and the  disreputable reputation of the men who delivered citations in English court  practice,1 together with the disputes about citations found in the records of  litigation, were also spurs for this choice. The subject is of intrinsic interest.  How law suits should be begun is a problem faced by every legal system.2  And as my investigation went forwards, more and more the choice began to  appear an appropriate one for studying the construction of the ius commune.  I cannot say that it was typical. Indeed some things point to the existence of  special characteristics in the law of citations. However, the subject does tell  us something worth knowing.  Three  specific  parts  of  the  law  of  citations  have  been  included  in  the  chapter’s  examination.  First  is  the  question  of  its  necessity.  Did  one  have  always to issue a formal citation before adjudication? Second is the question  of time. How long an interval had to come between service of the citation  and the required appearance in court? Third is the question of the citation’s  contents. How much information had to be placed in it to establish its validity?  Other  questions  could  have  been  selected,  of  course,  for  example,  asking  about the expenses of citation or about proof of service. There is material  aplenty. Long folio treatises devoted exclusively to many aspects of subject  came to be written.3 However, some limits were required. The task was to  investigate how the ius commune came into being. The three topics chosen  were important and are illustrative.4 B. THE SOURCES The  starting  point  for  examination,  however,  cannot  be  the  work  of  the  medieval jurists. It must be with the formal sources of the ius commune, the  Roman law and the canon law. That is where the ius commune itself began.  Doing so produces an apparently strange result. A survey shows how little  the medieval jurists had to work with. Not nothing, of course, but much less  than one might have expected for an aspect of court practice that would come  to be called the fundamentum of any valid judicial sentence.5 Less also than    1 See S L Rogers, All Things Chaucer: An Encyclopedia of Chaucer’s World (2007) II 420-421.   2 S F C Milsom, “Trespass from Henry III to Edward III: Part III” (1958) 74 LQR 561 at 572-574;  reprinted in S F C Milsom, Studies in the History of the Common Law (1985) 72-74.   3  See the treatises listed in M Lipenius, Bibliotheca realis iuridica (1757-1789; repr 1970-1971) I  212-213.    4  A  useful  description  in  English  is  contained  in  A  Bevilacqua,  Procedure in the Ecclesiastical Courts of the Church of England with its Historical Antecedents in Roman and Decretal Law  (1956) 13-40.   5  See,  e.g.,  Glos ord  ad  X  1.2.10  sv  non conventi:  “[C]itatio  est  principium  sive  fundamentum 

citations and construction of procedural law in the ius commune 249

one would have supposed by starting from the rule that the medieval church  vivit iure Romano. At least in this area of the law, one cannot fall back on  the stock notion that the canon law simply took over an established part of  Roman law. The institution was there, together with texts that could serve to  shape the law in specific ways, but for several reasons the Roman law proved  quite incomplete. Most of the detail necessary in a workable law of citations  had to be discovered or created.6  (1) The Roman law In  its  earliest  days,  Roman  law  seems  to  have  required  not  only  that  a  complainant notify the defendant of his intent to initiate proceedings against  him (the in ius vocatio); sometimes he had physically to bring an unwilling  defendant to court.7 The summons was, however, a private act, and it is not  always  easy  to  collect  its  basic  elements  from  the  Justinianic  texts  alone.8  They do, however, contain information about the initiation of litigation and  the  process  of  bringing  parties  before  a  court.  For  instance,  a  text  in  the  Digest  envisioned  three  preliminary  edicts,  separated  by  ten  days,  then  a  peremptory  edict,  after  which  the  magistrate  could  proceed  to  sentence  whether the person summoned came or not (D 5.1.72). One text from the  Code  made  a  similar  point,  speaking  of  three  denunciationes  as  effective  against a contumacious party (C 7.43.9). Other texts, almost incidentally it  seems, mentioned citation as an ordinary part of court procedure in specific  instances, one without which any judicial action might be suspect.9 Some of  them seem to have been based on a legal principle that both parties must be  heard by the judge before their interests could be adversely affected.10 Whatever the potential inherent in them, the presentation of the subject  seems  disappointing  at  first  sight.  The  Corpus iuris civilis  did  not  contain  a full or organised treatment of the process of citation. The Institutes had 

ordinis iudiciarii”; Robertus Maranta (d ca 1530), Speculum aureum (1549) pt VI tit De citatione  no 1: “[E]t dicitur fundamentum ordinis iudiciarii”.    6  K Pennington, “The Penal Process and the Protection of Rights in Canon Law”, in P Dugan (ed),  Proceedings, Gratianus Series (2005) 45 at 53: “Although founded on Roman law, the ordo was  new”.   7  Twelve Tables I:1, in M H Crawford (ed), Roman Statutes (1996) II 584-585; H F Jolowicz and  B Nicholas, Historical Introduction to the Study of Roman Law, 3rd edn (1972) 175-176; Kaser,  ZPR2 § 10.   8  E Metzger, “Lawsuits in Context”, in J W Cairns and P J du Plessis (eds), Beyond Dogmatics: Law and Society in the Roman World (2007) 187 at 187-188; J M Kelly, Roman Litigation (1966) 6-15.   9  See, e.g., D 40.5.26.9. 10  See, e.g., D 42.1.47; D 1.7.39.

250

the creation of the ius commune

virtually nothing on the subject.11 Neither the Digest nor Code had a title  called De citationibus, and their most relevant titles, both called De in ius vocando (D 2.4; C 2.2), consisted mainly of cases in which a specific class of  persons could not lawfully be cited – for instance, the Digest’s texts noted  that children could not cite their parents (D 2.4.6), no one could cite a person  who had taken refuge in his house (D 2.4.18), and a girl below puberty who  was in another’s power was exempt from service of process (D 2.4.22). Only  slightly more useful was the eleven-law title in the Code (C 7.43) in which  it was stated that a sentence could be given against an absent person if that  person had earlier been cited and deliberately absented himself.12  The likely explanation for the incomplete, indeed meagre, treatment of  the  citation  –  that  the  compilers  of  the  Digest  had  to  draw  upon  sources  describing  a  procedure  that  had  become  obsolete  by  the  sixth  century  –  would  have  been  quite  unknown  to  the  medieval  jurists.13  They  could  not  have known that Tribonian and his associates had felt obliged to leave much  out because it was no longer relevant. In fact, official citation by the authority  of the government had taken a more prominent place in practice. However,  the  men  who  formulated  the  medieval  ius commune  had  to  deal  with  the  Roman law’s texts as they found them, working without any “historical sense”  of the way the Digest had been put together, and what they found would have  seemed inadequate for the task.  Roman law thus treated the citation as a normal part of court procedure,  but it contained no coherent or adequate treatment of the citation. No law  set forth the contents required in a valid citation, although it seems to have  been  assumed  that  it  should  contain  the  name  and  authority  of  the  judge  before whom the defendant was being called to appear.14 Apparently, use of  both oral and written citations was envisioned, at least under some circumstances.15 This aspect of procedural law must have appeared to the medieval  jurists  to  have  been  an  area  of  the  law  outside  the  main  interests  of  the  Roman jurists. They found many individual texts scattered throughout the  Digest that touched upon the citation, and they made use of them. They were  quite capable of making imaginative leaps in doing so – as in reading laws  restricting rights gained by prescription to show that usage without citation  11 It was mentioned, for example, in J Inst 4.6.12. 12 Quomodo et quando iudex sententiam proferre debeat praesentibus partibus, vel una parte absente. 13 I am grateful to Professor Wolfgang Ernst for raising and discussing this point with me. 14 See P Collinet, La procédure par libelle (1932) 103-104; L Wenger, Institutes of the Roman Law of Civil Procedure, trans O H Fisk (1940) 94-99. 15 D 2.13.1.1; Collinet (n 14) 97, 99-100; see also Glos ord ad D 2.4.21 sv nemo extrahi.

citations and construction of procedural law in the ius commune 251

of the party affected was a nullity.16 Many leges thus proved relevant to the  jurists  through  the  process  of  analogy.  However,  they  did  not  define  what  information had to be conveyed to summon a party to litigation. They did  not tell readers what a citation had to be or contain. That was a product of  later choice. (2) The canon law The medieval canon law was not much more directly informative about the  citation,  although  it  too  contained  texts  relevant  to  the  subject.  Gratian’s  Decretum  (ca 1140)  had  little  to  say  about  the  mechanics  of  court  procedure, including this one.17 The Gregorian decretals (1234) contained no title  expressly  devoted  to  citations,  nor  did  the  Liber Sextus  (1298).  All  these  sources did include multiple texts devoted to stating jurisdictional rules, and  some  of  them  would  become  important  in  determining  the  validity  of  any  citation.18 The texts also made certain advances on the existing Roman law.  For example, a papal decretal of Urban III (1185-1187) used the word citatio  to  describe  the  opening  step  of  litigation,  purporting  to  make  its  issuance  determinative of a judge delegate’s continued power to deal with the matter  assigned to him after the death of the person in whose name he exercised  jurisdiction (X 1.29.20). Similarly, papal decretals used the term in holding  that a summons to court might be made in a defendant’s church or his house  if he could not otherwise be found (X 2.14.10; X 2.14.3). A decree from pope  Clement V’s council of Vienne (Pastoralis cura) stated emphatically that no  person could lawfully be cited to appear outside the jurisdiction to which he  belonged. The text was placed in the Clementines (Clem 2.11.2), and a long  explanation of its meaning was incorporated into the glossa ordinaria. These  texts stated rules of law. They could be read to emphasise the citation’s importance. However, only indirectly did they describe (or affect) the contents of  the citation itself. No part of the corpus of the Church’s emerging law focused  directly upon that question. It is to be contrasted with the libel and the litis contestatio, both of which were accorded a separate title in the Decretals (X  2.3.1-3; X 2.5.1).  This  did  not  mean,  of  course,  that  the  canon  law  was  useless  in  the  construction of a law of citations. Indeed its texts turned out to be vital in  16 C 7.22.1, cited in the marginal gloss to X 1.2.10 and Glos ord ad id. sv non conventi. 17 See E Jacobi, “Der Prozeß im Decretum Gratiani und bei den ältesten Dekretisten” (1913) 3 ZRG, KA 223 at 265-270. 18  For example, could a defendant ignore a citation from a judge who lacked jurisdiction? Or must  he appear and make a proper exception against the asserted jurisdiction? 

252

the creation of the ius commune

several respects. As noted, several of the Decretals mentioned the existence  of citations, even requiring that a written record of their issuance be made  (X 2.19.11). Evidently, the citation was to be considered a necessary part of  the church’s procedural law. Some of the texts would also prove useful by way  of analogy, and several were important for their statement of a principle of  fair treatment of defendants. For example, Gratian’s Decretum contained a  quaestio involving a bishop who was publicly accused of the commission of a  crime. A papal letter stated that if the matter could not be settled peaceably,  the bishop was to be cited to appear before the next provincial synod and be  given adequate forewarning, including a written notice of the accusations so  that he could be prepared to respond (C. 5 q. 2 c. 1). Gratian concluded that  some sort of notification was obviously necessary; without it the bishop could  not be required to respond, even if he were otherwise required to be present  at the synod.19  The Decretals contained a text stating the principle even more forcefully.  It was a papal letter from pope Gregory I refusing to decide a dispute without  having heard both parties (X 2.12.1). The text made no specific mention of  citation, but it did provide a good reason for requiring one, and it would later  be taken up and employed for that purpose. The Decretals also contained  a papal letter that would be similarly fruitful, one from pope Innocent III  requiring particularity in stating the claim in a civil action over land (X 2.3.2).  The defendant was entitled to know exactly which of the several parcels of  land he owned was at issue before he could be required to respond in a cause  being initiated against him. None of these texts was detailed or definitive, it  may justly be said. Both could either have been distinguished or said to relate  to something other than the formal citation used in court practice. However,  they raised a matter of principle that was related to the purpose any citation  served. They become important in an evolving law of citations. They both  gave one reason for requiring citations in the first place – to give anyone to  be adversely affected by civil or criminal litigation the chance to prepare a  response to the case against him.  C. WORKING OUT THE DETAILS The  resources  available  to  the  medieval  jurists  for  working  out  a  sensible  law  of  citations  in  the  years  around  1200  thus  amounted  to  more  than  a  blank page. The citation was an established part of the ordo iuris. It was in  19 d.p. C. 5 q. 2 c. 4. 

citations and construction of procedural law in the ius commune 253

regular use in European courts. Its existence was mentioned in the laws of  the Frankish peoples,20 and it was noted in some of the texts of the canon  law.  Principled  reasons  for  requiring  citation  of  a  party  to  litigation  when  his interests were likely to be adversely affected were also stated in several  places in both the canon and Roman laws. The necessity of prior notification  in analogous situations, such as diocesan synods or monastic elections, was  stated in the canon law. As a model of sorts, such texts were available to the  commentators. Still, considerable work remained to be done. The earliest writers on the law did not do that work. The first decretists,  writers on Gratian’s Decretum, were not encouraged by their text to discuss  questions  of  court  procedure,  and  for  the  most  part  they  did  not.21  The  civilians, some of whom wrote expressly about procedure, were not greatly  different. For instance, Placentinus, author of a Summa known as Cum essem Mantue  (ca  1160),  described  the  various  actions  available  under  the  civil  law  without  describing  how  the  parties  were  to  be  summoned.22  Johannes  Bassianus, writing in the 1180s, began his discussion of procedure with the  libel rather than the citation. He did take note that in principle the defendant  should not be obliged to respond to a plea in the absence of a written petition,  but he seems not to have been thinking of a citation, but rather the plaintiff’s   statement of claim. He did not comment on the citation.23 The Summa known  as  Ulpianus de edendo,  written  ca  1200,  included  a  brief  section  entitled  “De in ius vocando” and mentioned the existence of the apparitors who did  the work of citing defendants, but did not describe the citation’s contents.24  Similarly  Pillius,  thought  to  have  been  a  teacher  at  Bologna  and  to  whom  an Ordo iudiciorum from the late twelfth century has been ascribed, noted  that the judge “should warn the pars rea to appear” at the start of any cause.  However, he recorded no more about this part of the process.25 He referred  20 See, e.g., Lex Salica I:1, in E A T Laspeyres (ed), Lex Salica ex variis quae supersunt recensionibus  (1833) 15; see generally A Engelmann, Der Civilprozess: Geschichte und System II § 35 (1890;  repr 1970) 40-42. 21  See,  e.g.,  J  F  von  Schulte  (ed),  Paucapalea, Summa über das Decretum Gratiani (1890;  repr  1965). 22 Placentinus, Summa De actionum varietatibus, ed L Wahrmund [= QGRKP IV pt 3] (1914; repr  1962) 1-4. 23  Johannes Bassianus, Summa Quicumque vult, ed L Wahrmund [= QGRKP IV pt 2] (1925; repr  1962) 1-8 at 7: “Reus enim non debet respondere, nisi scriptae petitioni.” 24  G Haenel (ed), Incerti auctoris ordo iudiciorum (Ulpianus de edendo), Tit. De in ius vocando  (1838) 5-6. 25  Pillius Medicinensis, Summa de ordine iudiciorum, in F Ch Bergmann (ed), Pillius, Tancredus, Gratia, Libri de iudiciorum ordine (1842; repr 1965) 4. The same minimalist treatment is found  in Damasus Bononiae, Summa composita, de ordine judiciario, in A Wunderlich (ed), Anecdota quae processum civilem spectant (1841) 46-47.

254

the creation of the ius commune

his  readers  to  the  Novels,  rather  than  to  the  Digest’s  texts  on  procedure,  perhaps because he knew that the texts from the title De in ius vocando no  longer fitted the procedural pattern of litigation current in his times. Much  the same was true of the earliest decretalists, commentators on then current  papal decretals.26  Real progress on the subject seems to have begun only with Tancred of  Bologna,  whose  influential  Ordo iudiciarius  appeared  between  1214  and  1216. This treatise contained a separate section devoted to citations, using  that name rather than the title De in ius vocando taken from Roman law,27  and it dealt with the subject in greater detail. With it in hand, one can turn  directly to the three questions to be examined.  (1) The necessity for citation The first question is whether or not a citation was required before process  could be initiated against any person. An affirmative response seems axiomatic; it is a matter of fundamental fairness that a person be warned and given  a  chance  to  explain  himself  before  his  person  or  his  property  can  be  put  in  jeopardy.  So  it  seemed  then  too.  The  rule  was  regarded  as  part  of  the  law of nature by jurists in earlier centuries,28 and it was also stated, at least  by  implication,  in  some  of  the  basic  texts  of  the  Roman  and  canon  laws.29  What was not so clear was that it could stand as an absolute rule. Practice of  earlier centuries, including actions taken by some strong and saintly men, had  sometimes  paid  no  attention  to  any  such  requirement.  They  acted.30  They  had right on their side and they knew it. The Church’s sanction, excommunication latae sententiae, under which anyone committing certain acts in effect  excommunicated  himself,  also  seemed  to  stand  against  any  argument  that  citation was necessary.31 The law itself provided sufficient warning. Anyone  26 See, e.g., Bernard of Pavia, Summa Decretalium, ed E A T Laspeyres (1860; repr 1956) 32-35;  Bonaguida  de  Arezzo,  Summa introductoria super officio advocationis in foro ecclesiae,  in  Wunderlich (ed), Anecdota (n 25) 166-171. 27  Tancred, Ordo iudiciarius pt II tit 3, in Bergman (ed), Libri (n 25) 132. 28  See, e.g., Panormitanus (d 1445 or 1453), Commentaria super Decretalium libros (1615) ad X  1.33.8 no 1.  29  See,  e.g.,  C  7.43.7.  See  generally  M  Schmoeckel,  “‘Neminem  damnes,  antequam  inquiras  veritatem’: Die Entwicklung eines hohen Beweisstandards als Vorgeschichte der Verdachtsstrafe”,  in  M  Bellomo  and  O  Condorelli  (eds), Proceedings of the Eleventh International Congress of Medieval Canon Law (2006) 539. 30  Some pertinent evidence is collected in R H Helmholz, “Excommunication and the Angevin Leap  Forward”  (1997)  7  Haskins Society Journal  133;  A  L  Winkler,  “The  Excommunicated  Castle:  Clerical Power and the Natural World”, in A Duggan et al (eds), Medieval Studies in Memory of Leonard Boyle, O.P. (2005) 231. 31  See P Huizing, “The Earliest Development of Exommunication latae sententiae by Gratian and 

citations and construction of procedural law in the ius commune 255

who violated the law knew at once that he had incurred the church’s most  serious penalty. In addition, there were practical problems. What if citation  became impossible for practical reasons? Suppose the person to be cited hid  himself whenever the process server appeared. Was the legal system to be  powerless to do justice? The question of the citation’s necessity was therefore  a real problem. When  we  look  at  the  earliest  proceduralists  in  the  ius commune,  what  do  we  see?  As  noted  above,  the  best  starting  point  is  Tancred’s  Ordo.32  It  made an advance on most contemporary canonical and decretal collections.  His treatment also drew upon both Roman and canon law to state that no  defendant  should  be  declared  contumacious  unless  he  had  been  lawfully  called to court.33 Tancred went on to state that although it was permissible  to  proceed  against  a  defendant  who  ignored  a  single  peremptory  citation,  equitable principles should move judges to cite the party again and wait for  him  to  appear.  In  short,  he  at  least  opened  up  the  subject  to  considering  broader  questions  of  how  to  do  justice.  It  is  true  that  Tancred’s  treatment  was not comprehensive. He devoted more space to stating the Roman law’s  rules excepting certain persons and seasons from lawful citation than he did  to discussing either the requirement of citation itself or the contents necessary in any written citation. Perhaps Tancred’s decision to devote a separate  section to it as a part of statement of the ordo iuris was as important as any of  his actual conclusions, but his treatment shows signs of fresh thought, based  upon the underlying purpose of the citation. The thirteenth century witnessed a multiplication of works devoted to the  ordo iuris.34  Not  all  of  them  discussed  the  citation,  but  most  of  them  did,  generally (as it appears to me) following the lead given by Tancred. Aegedius  de Fuscarariis, for example, whose work dates from just after the mid-point  of the century, described the citation as the foundation of the ordo and went  on to discuss what it should contain. He also provided forms to be used in  preparing one.35 He asked what would happen if a person lawfully cited did 

the earliest Decretists” (1955) 3 SG 277; E Vodola, Excommunication in the Middle Ages (1986)  28-35. 32  Tancred, Ordo iudiciarius pt II tit 3 (n 27) 132. 33  Tancred, Ordo iudiciarius pt II tit 3 (n 27) 132. The two citations were to D 42.1.53 and X 2.8.1,  the latter mentioned by Tancred only as furnishing a possible argument.  34  See generally L Fowler-Magerl, Ordo iudiciorum vel ordo iudiciarius: Begriff und Literaturgattung (1984) 33-153; K W Nörr, “Die Literatur zum gemeinen Zivilprozess”, in Coing, Handbuch I 383-397; Lange, RRM § 54 I 443-445. 35  Aegidius  de  Fuscarariis,  Ordo iudiciarius,  ed  L  Wahrmund  [=  QGRKP  III  pt  1]  (1916;  repr  1962)  5.

256

the creation of the ius commune

not appear, and answered his question after a fashion,36 but Aegedius did not  confront directly the question of whether adequate citation was necessary for  all valid sentences. He only hinted that it might be. When  we  reach  the  Speculum iudiciale  of  William  Durantis,  the  most  influential proceduralist of the early ius commune, whose treatise dates from  the end of the thirteenth century, we seem to be in a different world. He  followed Tancred in devoting a separate section to the subject, but he was  anything but brief in his discussion of it. Indeed, he was prolix in the extreme.  For  example,  where  Tancred  had  simply  mentioned  peremptory  citation,  neither defining it exactly nor discussing its implications, Durantis devoted  three folio pages to it. He stated that a special peremptory citation could be  the basis of a valid sentence, though he also outlined the contrary view that  was (apparently) held by some.37 He devoted many lines to the question of  whether a citation could be peremptory in effect if it did not state in so many  words that it was. In some sense this change in emphasis was appropriate.  Single citation, presumed to be peremptory in nature, in place of the original  threefold  citation,  had  become  the  norm  in  ecclesiastical  court  practice.38  Durantis did not say this, but the attention he devoted to it may have been  more than a result of his own habitual verbosity. But what did Durantis say about the necessity of citation? Very little actually.  He raised the question of whether a person known to be contumacious had  to be cited, giving brief textual authority for both sides of the question.39 He  also opened up the difficult question of what status a citation “of all interested  persons” had in law, discussing it inconclusively but at length.40 Such citations  to appear and object to some step in litigation became a regular feature of  court practice – as in the citation of all persons who had an interest in the  estate of a decedent – and the matter deserved discussion. Durantis took it  up, but he did not really face the larger question of the citation’s necessity  and its practical effects. For all its length, his treatment of the centrality of  the citation in canonical process disappoints. It  appears  that  the  question  of  the  citation’s  necessity  only  crystallised  around the formal adoption of summary process in the canon law at the start  of the fourteenth century. In the decrees Saepe and Dispendiosam, the church  accepted and defined legal process simpliciter et de plano absque iudiciorum 36 Aegidius de Fuscarariis, Ordo iudiciarius (n 35) 11. 37 William Durantis (d 1296), Speculum iudiciale (1574; repr 1975) Lib II pt 1 § 3 no 8. 38 The commentators read a text in the Clementines to authorise use of a single citation; see Glos ord ad Clem 5.11.2 sv citationem. 39  Durantis, Speculum iudiciale (n 37) Lib II pt 1 § 4 no 19. 40  Durantis, Speculum iudiciale (n 37) Lib II pt 1 § 4 nos 20-23.

citations and construction of procedural law in the ius commune 257

strepitu et figura.41 The announced purpose of this reform measure was to  cut  down  on  unnecessary  delays  in  litigation.42  The  decrees  had  the  inevitable  if  incidental  effect  of  requiring  a  decision  about  which  parts  of  the  ordo iuris could be omitted and which could not. The citation belonged with  the  latter;  indeed  it  was  expressly  so  stated  in  Saepe.  The  initial  jurist  to  devote his attention to the Church’s law on summary procedure, Johannes  de  Lignano,  began  his  coverage  with  the  question,  “First  is  the  question  whether a citation is necessary?” answering it with a strong affirmative.43 It  was required by natural and divine law. Indeed God had himself given the  lead, as shown by the example of Adam in the Garden of Eden, whom God  cited with the words “Adam ubi es?” He surely knew of Adam’s whereabouts,  but he notified him anyway. Thereafter affirmation of the centrality and the  necessity of the citation was a regular part of treatises dealing with procedure  in the ius commune. Panormitanus, for example, reasoned that the purpose of  summary procedure was to exclude “the pernicious subtlety of actions”, not  to take away any form of defence available to defendants under divine law  and the ius gentium.44 Therefore, it followed, citation was a necessity. It could  not lawfully be removed by local statute or contrary custom. This view became a standard point of departure among procedural writers  in later centuries. It was particularly apt for the criminal law. A representative  example  comes  from  the  work  of  Jodocus  Damhouder.  He  began  his  coverage of criminal procedure with the statement that citation was “of the  substance” of any criminal cause, stating that its necessity was shown by the  divine, natural, canon, and civil laws.45 Similarly, Julius Clarus began his own  discussion of the law of citations with the statement that, “before all else, the  defendant must be legitimately cited … otherwise the process and sentence  and everything that follows against him are ipso iure void”.46 The point was  established. As an accepted rule, its force for court practice seemed unequivocal. Despite such apparently conclusive statements, however, some commentators  did  go  from  this  starting  point  to  consider  situations  where  citation  might  not  be  required  by  the  law.  Some  were  in  cases  one  might  expect.  41 Clem 2.1.2; Clem 5.11.2.  42 H K Briegleb, Einleitung in die Theorie der summarischen Processe (1859). 43 See Johannes de Lignano (d 1396), Super Clementina “Saepe”, ed L Wahrmund [= QGRKP IV  pt 6] (1928; repr 1962) 1. 44  Panormitanus, Commentaria (n 28) ad X 2.1.6 nos 1 & 6: “perniciosam subtilitatem actionis non  autem naturam actionis”. 45  J Damhouder (d 1581), Praxis rerum criminalium (1601) C. 13 no 1. 46  Julius Clarus (d 1575), Practica criminalis (1595) Lib V q. 31 no 1.

258

the creation of the ius commune

A  person  already  in  court  need  not  be  cited  by  normal  process;  it  would  have been a pointless formality.47 A pope might validly confer a benefice on  a supplicant without citing other parties who might have an interest in it, for  example. Acting from the fullness of papal power, he could be presumed to  have acted within the bounds of reason.48  If such exceptions were relatively narrow and unambiguous, others were  not. They were more open-ended and debatable. If it were absolutely certain  and notorious that a defendant had no defence to a claim made against him,  reasoned  Sebastianus  Vantius  and  some  others,  a  formal  citation  might  be  omitted.49 Such a person should not be able to hide behind a formal requirement to escape punishment. Similarly, the person who declared publicly that  he would not come to court under any circumstances might be dealt with  summarily.50 If he declared that no citation would have any effect on him and  rode away, it seemed that none should be required.  Process  against  the  person  who  hid  himself  and  could  not  be  reached  by ordinary methods also called upon juristic ingenuity to devise a remedy.  Some jurists concluded that no citation would be necessary in these circumstances.51 The man’s contumacy would be undeniable. If a citation were to be  required before he could be declared contumacious for disregarding a prior  citation, the requirement would have been an invitation to paralysis. Therefore, a citation to have him declared contumacious would be superfluous. The  general principle held that he should be given a fair warning and a chance to  prepare himself, but that would not apply to the truly contumacious. There thus came to be many limitations to the rule, and modern readers  are sometimes tempted to regard the process of elaboration of possible exceptions as a kind of a game, or worse. Adding to the list of exceptions does seem  to have been something like a sport. The Neapolitan jurist Robertus Maranta   compiled a total of thirty-one limitations to the rule,52 and it was said (by him)  that another commentator had been able to put together a list of 115 exceptions to the rule requiring citation. However, for our purposes, the lesson that  is more helpful for understanding the formation of the ius commune is that  the exceptions occurred in situations where the purpose of the law was not 

47 See, e.g., S Scaccia (fl 17th cent), De iudiciis causarum civilium, criminalium et haereticalium  (1663) Lib I, c. 88 no 23.  48  A Gabrielius (d 1555), Communes conclusiones (1593) Lib II concl 1 nos 13 & 25.  49  S Vantius (d 1570), Tractatus de nullitatibus processuum (1567) tit ex defectu citationis no 20.  50  Gabrielius, Conclusiones (n 48) Lib II concl 1 no 317. 51  See, e.g., Panormitanus, Commentaria (n 28) ad X 1.33.8 no 2. 52  Maranta, Speculum aureum (n 5) pt IV tit De citatione nos 6-38.

citations and construction of procedural law in the ius commune 259

violated.53 The reason for citation was to give the person being cited a chance  to be heard in his defence, just as God listened (briefly) to Adam’s explanation for his behaviour. Saepe and Dispendiosam, which were placed in the  Clementines, stated the rule that followed from the example. Citation was not  one of those elements that could be omitted, even in summary process. This  had an effect both on court practice and on juristic thought. It established the  basic principle. However, commentators could also look behind the texts to  the reasons for the law. Those reasons turned out to admit of exceptions – just  how many was open to the fruits of ingenuity and speculation. If the process of thinking about the reason for requiring citation permitted  the  jurists  to  make  exceptions  to  the  rule,  it  also  worked  the  other  way.  Although  excommunication  latae sententiae  seemed  to  permit,  and  even  to require, punishment without citation, in fact the law was interpreted to  require  citation  before  a  person  could  be  denounced  as  excommunicate.54  The sentence of excommunication, in theory, retained its automatic character,  but  the  denunciation  by  which  it  was  made  effective  in  practice  required  citation. The reason was that the person might have something to say in his  own defence, and that value was what the citation provided. In this case, therefore, the ius commune was formed by taking an institution  found in the Roman and canon laws and applying to it a principle ascribed to  the law of nature. That principle could be found recognised in several texts of  both laws. Its relevance to an institution of obvious and growing importance  – the citation – seemed obvious. The spur for the process was the adoption  of summary procedure in the early fourteenth century and the explicit statement  in  the  decree  permitting  abbreviation  of  process  that  citations  were  not among those parts of the ordo that could be omitted. With this start, the  jurists moved ahead. They developed the law of citations by thinking through  the implications of the natural law’s rule that every person was entitled to be  heard, and applying it to the exigencies of actual litigation. The result was to  produce a complex body of law, one full of amplifications and limitations, and  limitations of the amplifications, and amplifications of the limitations. Basic  agreement among the jurists about the requirement of initial citation in most  situations nonetheless remained intact.

53  See K W Nörr, “Von der Textrationalität zur Zweckrationalität: das Beispiel des summarischen  Prozesses” (1995) 81 ZRG, KA 1; K W Nörr, “Rechtsgeschichtliche Apostillen zur Clementine  Saepe”, in O Condorelli (ed), Panta Rei: Studi dedicati a Manlio Bellomo (2004) IV 225.  54  Antonius de Butrio (d 1408), Commentaria in Decretales (1578) ad X 5.39.9 no 3: “Tertio nota  quod excommunicatus a canone est denunciandus”. For English practice, see F D Logan, Excommunication and the Secular Arm in Medieval England (1968) 139.

260

the creation of the ius commune

(2) Assignment of the day for appearance How much  time  must  intervene  between  the  service  of  a  citation  and  the  party’s  appearance  before  the  judge?  This  is  the  chapter’s  second  subject.  Obviously, the interval could not be as short as an hour, but it need not be  as  long  as  a  year.  What  then  should  it  be?  This  was  an  obvious  question;  obvious once the church had created a system of public courts. Some of the  earliest proceduralists recognised it as a question.55 One may begin again with  Tancred. He found his starting point, not in the Corpus iuris civilis,56 but in  Gratian’s Decretum, where it was raised, albeit indirectly, in three different  places  and  with  three  different  results.  Tancred  appropriately  began  his  discussion with a qualification: “according to the opinion of some”, he wrote,  the answer is as follows.57 Then he took up the three possibilities he had found  in the Decretum. The first (C. 5 q. 2 c. 2) came within a Causa supposing that  a bishop had been slandered by a malicious writing which the writer could  or would not prove. The canon, a decree of pope Silvester, stated that the  writer should be punished, but that he must first be cited to appear and that  no action could be taken against him for seven days. Thereupon further and  more severe sanctions could be added to the first if he defaulted, but again  not  before  additional  time  had  elapsed,  first  another  seven  days,  then  two  days and then two days and then one final day. This suggested a period of  seven days might be proper in most situations. The second (C. 4 q. 5 c. 1), a  decree from the Council of Carthage, held that no bishop could be suspended  from communion without being given a warning and a chance to respond;  the hearing must be appointed to take place no less than a month from the  time of the initial warning. This suggested a month. The third (C. 24 q. 3  c. 6), again taken from a church council, set thirty days as the appropriate  period of penance for a cleric who had acted wrongfully in excommunicating  a person he should have known was innocent; but since it required a threefold  citation, the period envisioned was actually longer. On that account ninety  days might seem to be the appropriate interval between citation and appearance. Tancred buttressed the last with a citation to a text in the Authenticum,  which was commonly inserted into early editions of the Code.58 All  of  these  canons  rested  upon  a  principle:  that  some  delay  between  55 See Damasus, Summa (n 25) 46-47, giving a period of thirty days for each citation, but allowing  judges to set a lesser period, citing D 5.1.72. 56  The Novels contained texts providing for ten days and for twenty days (Nov 53.3.1-2); the Codex  Theodosianus provided for a period of forty days (CTh 2.6.1, 2.18.2) and one of twenty days (CTh  1.16.10), but none of these texts figured in Tancred’s treatment. 57  Tancred, Ordo iudiciarius (n 27) 132-135. 58  C 7.43.8; see also Nov 112.3.

citations and construction of procedural law in the ius commune 261

warning and hearing must occur, evidently to give the accused a chance to  consider  and  respond.  However,  they  were  not  in  agreement  on  just  how  long that delay had to be. The principle did not answer the question. Tancred  found  a  possible  resolution;  his  response  was  twofold.  First,  he  assigned  a  short  period  of  seven  days  to  cases  in  which  the  judge  and  the  defendant  were in the same city. Where they were “in the next city or not far away”, the  longer period of a month would apply. If they were multum remoti, however,  they must be cited by three citations separated by thirty days apiece, although  this was somewhat undercut by Tancred’s admission that there could be one  peremptory  citation  in  place  of  the  three.  In  any  case,  he  brought  some  coherence  to  the  possibilities  raised  in  the  Decretum.  It  made  sense  too  – the further the distance, the longer the appropriate period of time. The  distinction may have been something he hit upon himself.59 None of the texts  actually mentioned the distance involved as a factor in setting the appropriate  time. It was open to a commentator to look behind the words of the texts,  however,  in  order  to  discern  their  true  meaning,  and  Tancred’s  suggestion  fitted that way of thinking. So  it  might  seem.  In  fact,  it  turned  out  to  have  been  something  of  a  feint. Having made this distinction and pronounced it “reasonable enough”,  Tancred swept it all away. He continued, “[T]he general custom of the church  holds  the  contrary”.  There  must  certainly  be  some  delay  between  citation  and punishment. The judge cannot lawfully eliminate it. However, just how  long the delay had to be was “arbitrary”, something to be determined by the  judge’s discretion according to the quality of the parties, the difficulties of the  cause, and the nature of the circumstances. Tancred supported this result with  several citations from Roman law and papal decretals. None of the specific  texts he mentioned actually required the result Tancred embraced, but some  came close. For instance, a text from the Digest observed and endorsed the  judicial practice of abridging proceedings during their course, according to  the nature of the cause and the attitudes of the parties (D 42.1.2). A text from  the Code also allowed procedural latitude to judges (C 7.45.4). Neither dealt  specifically  with  citations  or  named  any  actual  intervals  of  time,  but  their  spirit certainly supported his conclusion.60  This resolution meant that Tancred’s initial attempt at harmonisation of the  relevant texts from the Decretum became unnecessary, except as they tended  59 It might be said that the second was appropriate because it involved a bishop being judged by his  metropolitan, and they would not have been in the same city, though they would also not have  been separated by long distances, at least in Italy. 60  The others were X 2.28.4; X 2.28.5; C. 5 q. 3 c. 1; C. 3 q. 2 c. 8; C. 3 q. 3 c. 2; and D 5.1.72.

262

the creation of the ius commune

in the same direction, which he may reasonably have considered they did. His  usage illustrates the possibility of making “creative” use of prior canons, but  the result he chose was based most directly upon a solution that had been  worked out in practice. Other jurists of the time took the same tack.61 In this  case, therefore, customary usages, tempered by an assessment of the values  involved, became the vehicle for introducing a solution to a problem in the  ius commune. And it was not an isolated situation. When faced with a difficult  problem in which there were many variables, very often it had seemed necessary in practice to leave the decision to the sound discretion of the judge. To this solution, the canon law also contributed something – a papal letter  of pope Alexander III which Raymond of Peñafort inserted in the Gregorian  Decretals.62 This decretal forbade judges from appointing a terminum nimis breve for the defendant’s appearance, unless there was “urgent necessity” for  doing otherwise. The decretal did not itself embrace Tancred’s solution in so  many words, but it was consistent with the principle he had advocated. It left  the meaning of the nimis breve term undefined and hence to decision by the  individual judge.  It might seem that acceptance of this text would have ended discussion  among the jurists. But it is not so. A good example is furnished by Durantis.  Oddly, he did not so much as mention Alexander III’s decretal in discussing  the length-of-time question. His approach was a slightly confusing one; he  repeated what Tancred had said, almost (but not quite) word for word, and  then he added to it.63 He began by opening the possibility of distinguishing  between different classes of persons to be cited. Thus bishops might be given  longer than ordinary clerics to appear. This would distinguish the second of  the canons in the Decretum from the first. The second had in fact concerned  a bishop, although it did not say that a shorter period would apply to others.  The  distinction  also  fitted  distinctions  between  orders  drawn  elsewhere  in  the  canon  law.64  However,  it  is  not  clear  that  Durantis  endorsed  it.  He  moved on quickly to raise two other possible distinctions – one between civil  and criminal causes, the other between arduous causes and routine causes.  He  had  some  texts  to  support  both  of  these,  but  he  again  rejected  them  as  “antiquated”  and  ended  by  repeating  what  Tancred  had  said  about  the  custom of the church and the responsibility of the judges for setting a reasonable  period.  The  most  that  can  be  said  for  this  treatment  is  that  it  added  61 William of Drogheda, Summa aurea c. 9, ed L Wahrmund [= QGRKP II pt 2] (1914; repr 1962)  15: “secundum consuetum usum litigandi in Anglia”. 62  X 2.8.1, and see esp Glos ord ad id. sv nimis.  63  Durantis, Speculum iudicialis (n 37) Lib II pt 1 § 2 (Nunc). 64  He cited X 2.1.13 and X 1.29.28.

citations and construction of procedural law in the ius commune 263

complexity  and  a  few  more  texts  to  the  subject.  It  also  made  more  widely  available the conclusions Tancred had earlier reached.  The pattern established by Durantis continued. If we look at sixteenthcentury  procedural  treatises  –  of  which  a  great  many  were  written  –  they  essentially  endorse  the  reasonability  standard  he  had  established.  “A  term  should be given and a suitable delay assigned, such as shall seem just according  to a determination made by the judge.” So concluded Sebastianius Vantius,  adding references to treatises compiled by other jurists.65 This became the  communis opinio. It is true that there was some disagreement; it would be  possible, for example, to distinguish between different kinds of citations. A  citation to determine what monetary fine should be levied could be treated  differently  from  one  to  proceed  with  the  acta inter partes.66  A  criminal  cause might be treated differently from a purely civil cause.67 However, to  most jurists it seemed safest to leave the specific term to the disposition of  local statute and custom. Where these were uncertain, it should be decided  according  to  the  sound  discretion  of  the  individual  judge.  If  he  erred,  the  error might be corrected by timely appeal. Thus,  for  this  question  the  ius commune  was  formed,  first,  by  application of a norm taken from practice rather than from a text of the Roman or  canon  laws.  The  texts  were  used  in  developing  the  law,  but  they  were  not  themselves  the  source  of  the  rule.68  They  supported  a  response  that  came  from good sense and experience. This component of the ius commune is too  often overlooked. Tancred’s choice and the endorsement of many later jurists  provide a reminder of its existence and its importance.  The law that emerged endorsed a general rule or principle that allowed  judges  considerable  discretion.  It  would  have  been  difficult  to  hit  upon  a  fixed  period,  one  that  would  be  practicable  in  the  wide  geographical  area  where the ius commune held sway. A general rule, the details of which were  to be worked out by judicial discretion in light of individual circumstances,  was probably a sensible solution. It was also a characteristic solution of the  65  S Vantius, Tractatus de nullitatibus processuum ac sententiarum (1567) tit Ex defectu citationis  no  70. 66  Scaccia, De iudiciis causarum (n 47) Lib I c. 32 no 17; the former would not require naming a  specific date, the latter would.  67  D Tuschus (1620), Practicarum conclusionum juris in omni foro frequentiorum (1661) tit C concl  230 no 7. 68  See  a  similar  development  of  the  same  period  described  by  G  Drossbach,  “Die  Entwicklung  des  Kirchenrechts  als  raumübegreifendes  Kommunikationsmodell  im  12.  Jahrhundert”,  in  G  Drossbach  and  H-J  Schmidt  (eds),  Zentrum und Netzwerk: kirchliche Kommunikation und Raumstructuren im Mittelalter (2008) 41; M G Cheney, Roger, Bishop of Worcester 1164-1179  (1980) 166-212.

264

the creation of the ius commune

ius commune,  and  in  this  case  it  was  reached  at  an  early  date.  It  did  not,  one should add, insulate a judge’s decisions from challenge. They could be  challenged under the heading of citationes indiscretae.69 Such appeals were  quite possible under the standard reached, and this made for delay. But the  alternative, setting a fixed number of days, would certainly have been worse.  It  would  have  called  for  too  many  limitations  and  variations  to  have  been  workable. If anything, the loose standard adopted provided an incentive for  judges and their officials to be circumspect, and perhaps even generous, in  assigning the date required for a party’s appearance. (3) The citation’s contents The  third  subject  concerns  the  citation’s  contents.  If  oral,  what  must  the  summoner tell the person being cited? If written, what information must the  citation contain? What information about the judge and his authority must  there be? What information about the time and place of court proceedings?  And what information about the plaintiff and the nature of his claim against  the defendant? In other words, how much detail should there be and how  much must there be?  It might be thought that principles would have applied identical to those  relevant to deciding whether or not a citation had to be used at all – that is,  the need to alert the defendant to the nature of the cause against him so that  he could prepare an answer. A decretal of Alexander III stated that it was  not necessary that a complaint should go into the “subtleties” of the civil law  of actions; it was enough that it state “purely and simply” the basis for the  plaintiff’s claim (X 2.1.6). This could be read a contrario sensu as holding that  the citation must at least give notice of what the claim was. However, another  decretal, one from Pope Celestine III, showed why this did not necessarily  follow (X 2.8.2). It stated that if a citation fully informed the defendant of the  nature of the cause being brought against him, he had no claim of right to a  procedural delay to consider his answer. Being fully informed in advance, he  should have come prepared. But this also meant that the negative effects of  an uninformative citation could be cured by granting a delay after the defendant had appeared. The plaintiff’s libel was, in any event, the proper place  for setting forth the complaint, and as often happened in fact in England, a  delay could be provided for a defendant after the libel had been introduced.70  If this course was followed, no harm to the defendant would have ensued. 69 See Tuschus, Practicarum Conclusionum (n 67) tit C concl 232.  70 See A Wunderlich (ed), Joannis Andreae Summula de processu judicii, c. 2 (1840) 8.

citations and construction of procedural law in the ius commune 265

Under a careful reading of this second decretal, the form of the claim that  actually went into the citation was a matter either for choice by the plaintiff or  for settlement by local custom. It would become important in practice later  on, when objections on principle were raised to the full exercise of ecclesiastical jurisdiction. Then defendants might object in principle to being cited  and take independent steps to quash the action. But contention on that point  was in the future in the thirteenth century. Most jurists held that defendants  must appear in court and raise jurisdictional exceptions there. Little can be  said  of  early  practice.  Examples  of  relatively  full  statements  have  survived  from the early period,71 but the glossa ordinaria to the Decretals admitted  that even in documents from the papal chancery full statements of the nature  of the claim were unusual.72  It seemed otherwise with respect to questions of jurisdiction and dates of  hearing. Tancred stated that, especially for papal judges delegate, the citation  should contain the commission under which the judge was acting, word for  word, and state by whose authority he had issued the citation.73 He did not  say the citation was invalid without such a statement; nor did he give any rule  applicable to ordinary judges. But he did raise the question of jurisdiction,  one that played a larger role in the thinking of the early decretalists than it  would later on, probably because the twelfth and thirteenth centuries were  the periods when litigation before papal judges delegate was at its height.74  For this question, statement of the source of the citation was critical. A text  from  the  Code  supported  the  conclusion  (C  9.3.3).  It  long  continued  to  furnish the standard rule.75  Tancred added that the citation should contain the name of the plaintiff and  the date and place at which the person cited should appear. There was some  textual support for the last, a decretal of pope Innocent III that mentioned,  almost casually, that the day and place had been included in letters of citation  (X  1.3.25).  There  was,  however,  a  danger  lurking  here.  A  citation  for  one  particular  matter  or  for  one  specific  day  and  place  might  not  be  valid  for  another  matter,  day  or  place.76  The  greater  the  specificity,  the  greater  the  obstacle  to  allowing  for  a  change  of  circumstance.  It  might  tie  the  judge’s  71 72 73 74

See, e.g., Gratia, Summa de iudiciorum ordine, in Bergmann (ed), Libri (n 25) 331-332. Glos ord ad X 2.8.2 sv plene. Tancred, Ordo iudiciarius pt II tit 3 § 2 (n 27) 133-134.  J  E  Sayers,  Papal Judges Delegate in the Province of Canterbury, 1198-1254  (1971);  Cheney,  Roger, Bishop of Worcester (n 68) 159. It made a continuing difference, in that fuller statement  was long required in a citation from a judge delegate than in a citation from an ordinary judge;  see, e.g., Scaccia, De iudiciis causarum (n 47) Lib I c. 32 nos 5-6. 75  See Tuschus, Practicarum conclusionum (n 67) tit C concl 230. 76  Tuschus, Practicarum conclusionum (n 67) tit C concl. 243. 

266

the creation of the ius commune

hands, or at least require a fresh citation. Perhaps it would be preferable to  state this requirement in more general terms, leaving practical details open  to local decision. There might of course have been quite practical reasons for  setting a time and place, and these would have been as obvious to individual  officials as they were to learned commentators. The matter could safely be  left in the former’s hands.  Turning to William Durantis, a student finds the much longer treatment  of the subject that he expects. In this instance, it also includes several forms  for citation – no doubt meant to be followed with appropriate modifications.77  Those he included were relatively full of detail, typically giving information  about the nature of the claim and the details about the parties, the judge, 78  and the circumstances of the proposed hearing. He did not say that the details  were necessary for validity. It may be that he was only giving an example of  good practice. Other contemporary commentators did the same. In fact, at  several points Durantis also included comments that seem to have cut against  any  requirement  that  the  citation  be  full.  What,  he  asked,  if  the  date  and  time of appearance were not specified in the citation? Is it nonetheless valid,  and was the person cited required to appear? Yes, he answered, he should  come within a reasonable time, although this requirement was itself to be  interpreted with moderation.79 The canon law called for willing obedience,  not a stubborn spirit, and this situation provided an occasion for calling this  characteristic of the canon law to mind. Some jurists, Durantis went on, had  opposed him on this point. They held that persons so cited were by no means  required to appear, though they might come if they wished to be heard. The  reader comes away from the passage uncertain of the law’s requirements on  this score. If one were a judge in the late thirteenth century, he might surely  have wished to include the projected date and place for appearance in order  to avoid controversy. But the judge would have had an argument on his side  if he neglected to do so. This characteristic of the law of citations – one may call it imprecision or  call it flexibility – continued in the statements of the commentators on the  mature ius commune. For example, was it necessary to state the name and  the source of authority of the judge before whom the defendant was to be  cited and that of the person executing the citation? One sixteenth-century  jurist wrote that it was, adding however that this was especially true where  the claim was brought before a judge delegate who came from a long distance  77 See Durantis, Speculum iudiciale (n 37) Lib II pt 1 tit De citatione § 4 nos 9-10; § 5, nos 4, 5, 6.  78 See William of Drogheda, Summa aurea c. 11 (n 61) 16-17.  79 Durantis, Speculum iudiciale (n 37) Lib II pt 1 tit De citatione § 4 no 20.

citations and construction of procedural law in the ius commune 267

from the domicile of the parties.80 Where did that leave the ordinary citation  in the many cases brought before local tribunals by the officials of a bishop  or archdeacon? The implication, at least as I see it, is that they had a choice:  take the high road by including the information, or the low road by assuming  that the fact would be obvious to the people involved. The latter would be  easier and cheaper. But in neither situation would the citation be subject to  practical challenge on this score.81 There was also the question of whether the citation must state the nature  of  the  plaintiff’s  claim,  and  if  so,  with  what  degree  of  particularity.  Sigismundo  Scaccia,  surveying  the  literature  on  the  subject  in  the  seventeenth  century, began by citing Cyno da Pistoia and Bartolus and a few others to the  effect that such inclusion was necessary, and that it must be stated in specific  terms.82  However,  Scaccia  then  took  the  opposite  tack,  conceding  that  the  opinio magis communis was to the contrary. More doctores had held that the  causa need not be included, at least in any detail. If a plaintiff were free to  modify his libel at the next stage, as the communis opinio of the jurists held  him to be, what practical advantage was gained by requiring the causa to be  stated in the citation? The defendant would know it soon enough. Scaccia  also  cited  the  lessons  of  practice,  saying  that  the  minimalist  position  was  confirmed by normal usage in the courts. He took special note of practice in  the courts of Florence, where a statute required that the plaintiff’s petition  be  included  but  where  nonetheless  ordinary  court  practice  confirmed  a  minimalist position. In the end it is not wholly clear what position he took.  The  contrarietates opinionum  of  the  jurists  can  sometimes  become  bewildering, and this treatment provides an example. One has to conclude that  Scaccia meant to leave room for “the usage received in many places”.83 Robertus Maranta, speaking of the courts in Naples, took the same position  as to the necessity for citation at the other end of a legal process, the term  for definitive sentence. It was the custom that a separate citation be issued  at that point,84 and the evident question was what it should contain. Maranta  wrote that, “[t]oday in the kingdom, by the common style of the holy council  sentences  are  given  without  statement  of  a  causa  in  the  citation,  and  they  80 Vantius, De nullitatibus processuum (n 49) no 32: “cum insertione nominis et potestatis ipsius  iudicis (maxime si foret delegatus et citatio fieret de longinquo)”. 81  See,  e.g.,  Tuschus,  Practicarum conclusionum  (n  67)  tit  C  concl  230  no  29.  Having  given  the  reasons for requiring that the authority of the person citing be fully stated, he added: “Limita  quando stylus tribunalis esset contrarius, quod immo sufficiat citatio data a parte sine commissione iudicis.”  82  Scaccia, De iudiciis causarum (n 47) Lib I c. 32 no 14.  83  Scaccia, De iudiciis causarum (n 47) Lib I c. 32 no 7. 84  See, e.g., N Boerius (d 1539), Decisiones Burdegalenses (1544) q. 285 nos 1-2. 

268

the creation of the ius commune

are valid because they are given under the name of the king”.85 He added a  supporting reference, in this case a provision from the Roman law’s Code (C  7.53.8). It left open a question of appointment of the executor of a sentence  until  after  the  merits  of  a  cause  had  been  fully  heard  and  discussed,  and  Maranta read it as adding a good reason for the Neapolitan practice. In most  circumstances, a defendant would not be allowed to introduce new material  after the conclusio in causa. That was the moment when the merits would  have been clear; therefore he would suffer no prejudice by being deprived of  specific citation after that time.86 It followed that the practice in Naples did  not offend against basic principles of justice.  Such allowance of scope to local variation in court practice and the repeated  reference by commentators to the stylus curiae on questions involving the  form and manner of citation are of interest and importance in assessing the  creation of the ius commune.87 In one sense, it is surprising. The ius commune  taught in the schools was an ideal law. It was the best law. Practice could be  another matter. What went on in the courts did not always match the views  of the jurists, as they themselves were quick to recognise. Where there was  deviation, it might show that things had gone wrong. However, it also might  call for further enquiry into the reason for the rule of the ius commune. As  in the case of the citation to hear definitive sentence, court practice might  be determinative, requiring adjustment to standards set by the law itself. If it  did, it would not necessarily change the ius commune, but it might be worthy  of note and even approbation. It might be accepted as the source of a rule,  as  it  did  with  the  second  question,  that  of  the  delay  between  citation  and  hearing. But this question was a little different. Reference to practice meant  that  differences  in  the  necessary  contents  of  a  citation  would  be  allowed.  They did not call forth objections from the commentators.  These examples show that the ius commune itself could leave room both  for difference of opinion and influence from practice. It also allowed – even  invited – local statutes to regulate some aspects of practice.88 They were in  fact enacted in many places.89 The law of Scotland, for instance, apparently  85  Maranta, Speculum aureum (n 5) pt IV tit De citatione no 10. 86  See Mattheus de Afflictis (d 1523), Decisiones sacri consilii Neapolitani (1604) Dec 283 no 6.  87  See, e.g., O Vestrius (d 1573), In Romanae aulae actionem et iudiciorum mores introductio (1547)  Lib IV tit De citatione extra curiam no 1.  88  M Wesenbecius (d 1586), Commentaria in Codicem (1665) tit De in ius vocando no 10: “citatio  porro fieri debet secundum statuta cuiusque loci”. 89  See, e.g., those noted in P Fournier, Les officialités au moyen âge (1880) 148; “Canons of the  Council of London (1268)” c 25, in F M Powicke and C R Cheney (eds), Councils and Synods with other documents relating to the English Church, (1964) II 772-773; “Synod of 1553 (Diocese  of Astorga)” c 8, in A García y García (ed), Synodicon Hispanum (1984) III 68-69; “Constitutions 

citations and construction of procedural law in the ius commune 269

required the statement of six specific matters in the citation; they were: (1)  name of the party being cited; (2) name of the person bringing the complaint;  (3)  description  of  the  matter  to  be  litigated;  (4)  day  and  (5)  place  set  for  appearance; and (6) name of the person before whom the party cited was to  appear.90 Without them, the summons was invalid. Such a mandatory inclusion of specific information was quite permissible under the ius commune,  although it was not required. The texts stated some foundational principles –  that a party was entitled to an effective warning before his interests could be  prejudiced. How that principle was applied in practice might depend upon  the circumstances and on local custom. In other words, lessons drawn from  practice, even practice on a very local level, might work their way into the  law.91 At least they would be recognised as legitimate.  D. ENGLISH COURT PRACTICE At the outset, my interest in the subject of citations had come from reading  cases  involving  citations  in  the  court  records  of  the  English  ecclesiastical  courts. This was not only a personal idiosyncrasy. There are good reasons for  linking procedural theory with court practice. One is that we understand a  legal rule if we see what it turned out to mean in practice; as the commentators often said, “Practice is the best interpreter of a law”. Another is that in  this area of the law, the commentators themselves took note of the relationship between common court practice and the ius commune. They sometimes  treated the former with respect. Of course, they sometimes also criticised it.92  At any rate, it seems that lessons apparently drawn from daily court practice  figured in the development of the ius commune.  It is possible to describe English court practice with some particularity. So far  as I know, none of the actual citations used in medieval practice has survived.93 

of  Florence  (1372)”,  in  R  C  Trexler,  Synodal Law in Florence and Fiesole, 1306-1518  (1971)  299-301;  “Statuta  provincialia”  c  3  (1349),  in  R  Zeleny  (ed),  “Councils  and  Synods  of  Prague  (1343-1361)” (1972) 45 Apollinaris 497 at 502-503. 90  Regiam Majestatem c 6, in T M Cooper (ed), Regiam Majestatem and Quoniam Attachiamenta  [= Stair Society XI] (1947) 64. I am grateful to Professor Olivia Robinson for calling my attention  to this text. 91  This approach was by no means unique; see, e.g., M Rady, “Hungarian Procedural Law and Part  Two of the Tripartitum”, in M Rady (ed), Custom and Law in Central Europe (2003) 47 esp at  48-52. 92  See, e.g., Tractatus super appellacionibus tam directis quam tuitoriis”, in F D Logan (ed), The  Medieval Court of Arches [= Canterbury & York Society XCV] (2005) 178-179. 93  There are some from a later period, however; see A Tarver, Church Court Records: An Introduction for Family and Local Historians (1995) 86.

270

the creation of the ius commune

Only copies have.94 The forms to be used in litigation were given and discussed  in formularies and other manuals of practice, however, and the court books  regularly  contain  notations  about  their  use.  It  is  hard  to  reach  a  confident  judgement, but if the records tell the truth, it seems that the law applied in  practice corresponded quite well with the ius commune.  First, citations were evidently a routine part of court practice.95 It would be  hard to prove a negative – that they were never omitted in practice – but the  positive evidence about their use is quite strong. Where the act books were  full, they recorded that letters of citation were routinely read out in court as  evidence of the fact of citation.96 The summoner (also called the apparitor)  was named and details of the process given.97 Citations were usually written  documents, because the act books record that the details, which would have  certified the delivery of the citation and have been written by the apparitor,  appeared on the dorse of the document.98 Many mandates of citation to them  were recorded.99 Cases in which the person being cited tore up the document  the summoner had put into his hand have survived.100 Multiple citations in the same cause were often issued in practice, though  surely not in most ex officio matters. It was customary, for example, to issue  a special citation in instance causes before the definitive sentence was delivered, as the commentators themselves recommended, and it was also normal  practice  to  do  so  before  any  party  was  certified  to  the  royal  Chancery  as  having stood excommunicate for more than forty days, by which process the 

  94 See “Introduction” in N Adams and C Donahue, Jr (eds), Select Cases from the Ecclesiastical Courts of the Province of Canterbury c 1200-1301 [= Selden Society XCV] (1981) 38-39; D M  Owen, Ely Records: A Handlist of the Records of the Bishop and Archdeacon of Ely (1971) 30.   95  See, e.g., M Ingram, Church Courts, Sex and Marriage in England, 1570-1640 (1987) 56-57.    96  See, e.g., Fel c Alard (Canterbury 1305), Lambeth Palace Library, MS 244, fol 31: “lecta litera  certificatoria per quam constabat de vocatione dicti viri [the defendant]”; Mannyng c atte Hunt  (London 1470), Guildhall Library, London, MS 9064/1, fol 25: “et Bogot … certificavit quod  citavit eum in domo officii”; Smythsonne c. Smythsonne (York 1561), Borthwick Institute, York,  Cons.AB.23, fol 10: the plaintiff’s proctor appeared “et dicit se partem pro eadem et introduxit  citationem”.   97  Ex officio c Johanson (Winchester 1521), Hants RO, 21 M65 C1/1 (Act book 1) fol 15: “Johannes  Smyth certificavit se peremptorie citasse eandem die dominico ultimo ad comparendum istis  die et loco.” For what is known about the exercise of this office, see B L Woodcock, Medieval Ecclesiastical Courts in the Diocese of Canterbury (1952) 45-49; and L A Haselmeyer, “The  Apparitor and Chaucer’s Summoner” (1937) 12 Speculum 43.   98  Gravest c Cordrie (Archdeaconry of Berkshire 1585), Berkshire RO, Act book D/A2/C21 fol 98,  where the citation is mentioned “ut patet per certificatorium in dorso citationis”.   99  See,  e.g.,  M  Bowker  (ed),  An Episcopal Act Book for the Diocese of Lincoln 1514-1520  [=  Lincoln Record Society XVI] (1967) 95-102. 100  See, e.g., Ex officio c Darnet (Hereford 1501), Herefordshire RO, Act book O/22 at 175: “fregit  et laceravit quandam citationem sibi directam a commissario”. 

citations and construction of procedural law in the ius commune 271

contumacious party would be imprisoned by the “secular arm” in England.101  Indeed we know that some courts issued repeated citations at several stages  in proceedings, because payment for them figured in lists of court expenses  assessed  at  the  end  of  canonical  trials.  This  could  become  an  abuse.  The  losing party in litigation had to pay court costs at the end, and this included  the cost of serving all the citations.102 The more citations, the more expensive  the case.  Second, we know that the citations used in practice normally stated a day  on which the person cited was to appear in court. At least the formularies  compiled for use in court practice so stated, and it appears that the interval  assigned  settled  upon  a  fixed  number  of  days.  The  “trend”  seems  to  have  been  towards  greater  specificity.103  In  a  Norwich  notebook  on  the  subject,  for example, the author recorded that the normal practice there had come  to be to set appearance for the third day after the citation had been served,  if the court met on that day. Otherwise it was to be the next following court  day.104  This  may  seem  a  quite  short  period,  but  most  ecclesiastical  courts  were local tribunals, and perhaps three days was a reasonable enough interval  under  such  circumstances.  Longer  delays  were  normally  set  for  litigation  before papal judges delegate, for example; in these cases the distances would  normally have been greater.105 It also would become the rule in London that  more than the normal three days would be given for defendants domiciled  outside the city itself.106 It would not be sensible to suppose that abuses never  occurred; exceptions taken against the assignment of “overly short terms” are  to be found in the court records. They were not frequent enough, however, to  have raised sustained or serious objection to the church’s jurisdiction.  Third, we have examples where judges apparently paid attention to the  advice of the commentators, framing their citations to include information  about their authority and about the substance of the litigation itself as they  found it described in the academic literature.107 The Norwich treatise stated  that the original citation was to contain: (1) the style and the name of the  101 Ex officio c John de Cornhelle (Canterbury 1305), Lambeth Palace Library, London, MS 244  fol 64.  102  See  Robertson  c  Fever  (York  1503),  in  “Registrar’s  book”,  Borthwick  Institute,  York,  D/C  R.Reg. fol 17v, listing the service of eight separate citations. 103  See, e.g., Anon, The Clerk’s Instructor in the Ecclesiastical Courts (1740) 12-13, giving the date,  place and hour. 104  Treatise on the law of citations (seventeenth century), Norfolk RO, Norwich, DN/PCD 2/5 at 1.  105  See Sayers, Papal Judges Delegate (n 74) 73-74. 106  F Clarke, Praxis in foro ecclesiastico (1684) tit 12; T Oughton, Ordo judiciorum sive Methodus Procedendi in negotiis et litibus (1728) tit 20. 107  For a Spanish example, see A García y García, “Un proceso ante juez convervador pontificio,  Santiago de Compostela, 1334” (1989) 19 (ns) BMCL 55 at 56.

272

the creation of the ius commune

judge; (2) the name of the plaintiff; (3) the name of the defendant being cited;  (4) the causa for which the litigation was being initiated; and (5) the time and  place for appearance.108 As noted above, such a full statement never became  an absolute requirement under the law, and in one cause where objection was  taken to omission of the causa in the citation, the exception was rejected.109  However, fuller statement could have been described as the best practice. It  avoided controversy. It appears to have become the norm in English practice.  Controversy on this point did erupt in England during the constitutional  conflicts of the seventeenth century. It happened in response to what seems  to  have  been  a  growing  trend  to  cite  defendants  in  ex officio  matters  “to  respond to certain matters relating to their soul’s health” or “relating to the  reformation of the manners” of the person being cited.110 Such intentionally  vague statements furnished some protection against aggressive use of writs  of prohibition on the part of the common law courts.111 It was harder to issue  a  writ  when  the  underlying  facts  were  unknown.  At  the  same  time,  it  was  objected that these citations provided little or no information to the defendant. They became the source of complaint – seemingly legitimate – against  the ecclesiastical tribunals and the Courts of High Commission.  It should also be said that even a valid citation did not guarantee that the  litigation would actually begin. It might be that there would be delay, as in  Scottish practice where a case was said to be “asleep” and to require a fresh  citation “of awakening”.112 The ordo iuris was not meant to be a strait-jacket;  the parties to litigation were not required to fit their procedural action into  an  exact  form.  The  jurists  who  developed  the  law  of  citations  themselves  recognised this. They left room for manoeuvre by judges and litigants. We  see it in the court records. 

108 See above n 104. 109 Arnsby c Bayly (Peterborough 1638), Northants RO, Northampton, Act book Instance 52, s.d.  5 July, a defamation cause. 110 Ex  officio  c  Garnett  (Durham  1604),  University  of  Durham  Library,  Act  book  IV/6,  fol  41v:  “causa morum reformationis”. See also the extract from “Bishop Alnwick’s Court Book (Lincoln  1446)”, in A H Thompson, The English Clergy and their Organization in the Later Middle Ages  (1947) 208. 111  See, e.g., William Lyndwood (d 1446), Provinciale (seu Constitutiones Angliae) (1679) 315 sv  perjurio.  112  Leah  Leneman,  Alientated Affections: The Scottish Experience of Divorce and Separation, 1684-1830 (1998) 14.

citations and construction of procedural law in the ius commune 273

E. CONCLUSION This description has barely scratched the surface of the full law of citations  found in the ius commune. For instance, it has not covered general citations,  that is the citation of groups of potentially interested parties. It has not dealt  with  the  several  moments  of  the  ordinary  trial  at  which  (some  thought)  a  repeated citation was necessary or desirable. It has not covered the contentious question of how legitimate citation was to be proved. It has not even  described how the threefold citation found in many of the texts was reduced  to  a  single  instance.  Nevertheless,  it  has  (I  hope)  done  something.  It  has  come  to  grips  with  particular  examples  of  the  ways  the  ius commune  was  formed. It has shown three ways the jurists created an important institution  of the mature law. In each instance, they began with an institution or point of law taken from  the Roman law. When they found it only sketchily described, the jurists turned  first to other texts in the Roman law that could be used to fill in the content of  the existing institution. One might fairly describe their approach as expansion  by analogy, although some of it merely required them to notice parts of the  Digest that touched on citations indirectly. They looked also to texts from the  canon law for authority and guidance. Many of the church’s decrees dealt with  the nature of notice that had to be given before any person could be called  to account for his actions. They stated principles useful in forming a law of  citations, even though the Corpus iuris canonici had no special title devoted to  the subject. The canon law became an equal partner, and the formal adoption  in the canonical system of summary procedure in the late thirteenth century  played a particularly important role in forcing and enabling the jurists to think  through what the ordo iuris must and should contain.  The mixture of principle, text, and practicality was particularly marked in  the first of the three questions dealt with above – when was it necessary to have  a valid citation, and when might it be omitted? The solution reached by the  jurists illustrates the pulls exerted by all three.113 It also underlines the utility of  natural law for the jurists. They were guided in considering the question by the  requirement that in ordinary circumstances a person should not be punished  or penalised without having had the chance to prepare a defence. The jurists  uncovered situations where this rule did not apply, but they were guided in  formulating those exceptions by the natural law’s own underlying purposes. It  was entirely characteristic of the medieval ius commune that it would begin  113 See D Johnston, “Suing the Paterfamilias: Theory and Practice”, in Cairns and du Plessis (eds),  Beyond Dogmatics (n 8) 173.

274

the creation of the ius commune

with a blanket rule, to which exceptions were added. These exceptions – it was  thought – were allowed or even encouraged by the demands of natural justice.  The law of citations provides a good example. Development of a rule on the minimum period required between citation  and appearance happened a little differently, but no less characteristically of  the ius commune. Here the jurists concluded that no specific number of days  could be set. They therefore rejected the various possibilities that Roman law  offered to them, in favour of a rule that left determination to the discretion –  although not the unfettered discretion – of judges and other officials involved  in litigation. I would describe the process as having been driven by experience  in litigation coupled with serious thought about the great variety of circumstances that might have an effect on fixing a suitable period. Some parties  lived far from where the court sat; others lived closer. Some had important  duties to attend to; others did not. Practical considerations, in other words,  played a strong role in answering this second question in the law of citations. The third aspect of the citation examined – how much information had  to  be  included  in  it  to  ensure  its  validity  –  matured  in  a  slightly  different  fashion. No text from the Roman or canon law stated a rule about the subject  in  anything  but  the  most  general  terms.  The  early  jurists,  responding  to  a  jurisdictional necessity – establishing the judge’s authority over the parties  and the proper delegation of the power to cite to the man who did the work  – paid more attention to the formal aspect of the citation’s contents than they  did to the particularity with which the claim itself was stated. The detailed  information about what an English common lawyer would call the “cause of  action” could be supplied by the libel. This stress on establishing a judge’s  jurisdiction  was  never  lost  but,  at  the  same  time,  the  preference  of  most  jurists moved towards fuller statements of the claim (or in criminal causes,  of the articles). They advised that citations should be full. However, most did  not make this a prerequisite for validity of the citation. In other words, this is  one of the areas of the ius commune in which an act’s validity did not depend  on following the law’s prescriptions. It was something like the canon law’s rule  that a marriage should be solemnised in facie ecclesie, but if it was not, the  marriage was still valid.  In  recent  years,  historians  have  devoted  a  great  deal  of  attention  to  marriage law; much less to procedural law. This is natural. Marriage law is of  immediate interest to most historians, and studying its application in practice  illuminates the social assumptions of earlier centuries.114 For understanding  114 Leneman, Alienated Affections (n 112) 1.

citations and construction of procedural law in the ius commune 275

the  formation  of  the  European  ius commune,  however,  procedural  law  is  probably a better choice. It illustrates an important way the European ius commune was created, and it was a more typical way than the history of the  law of marriage. Procedural law also provides a clearer view of jurists at work,  developing a sophisticated but practical law that grappled with the question  of what was meant by procedural due process. Some of the conclusions they  drew remain alive today. The law of marriage and divorce has changed quite  dramatically over the centuries, so much so that a divide exists between its  medieval form and the modern institution. The law regulating citations, by  contrast, has changed less. Besides helping to understand the construction of  the ius commune, it may deserve attention on that account.

11  Doctoribus bona dona danda sunt : Actions to Recover Unpaid Legal Fees 1

James A Brundage A. ROMAN LAW B. ADVOCATES’ FEES C. CLERICS AS ADVOCATES D. CONTROLLING AND NEGOTIATING FEES E. COLLECTING FEES F. TAXATION OF COSTS G. CONCLUSION A. ROMAN LAW The idea that one must pay for legal advice and representation in order to secure one’s rights has never had wide appeal, save to those equipped to provide these services – and even some of them have occasionally voiced reservations about the practice.2 “How glad I am that I have always kept clear of any contracts, presents, remunerations, or even small gifts for my conduct of cases”, wrote the younger Pliny (61/62-113). He added: “It is true that one ought to shun dishonesty as a shameful thing, not just because it is illegal; but, even so, it is a pleasure to find an official ban on a practice one would never have permitted oneself.”3   1 Marginal gloss to Decretum Gratiani C. 3 q. 7 d.p.c. 2 § 12. Glosses to Gratian’s Decretum and the other books of the Corpus iuris canonici are cited from the edition published in four volumes at Venice: Apud Iuntas, 1605. See likewise C 2.7.23 pr. The Glossa ordinaria of Accursius is cited from the five-volume edition published at Lyon: Jacobus Iuntas, 1584.   2 R H Helmholz, “Ethical Standards for Advocates and Proctors in Theory and Practice”, in S Kuttner (ed), Proceedings of the Fourth International Congress of Medieval Canon Law [= MIC, Subsidia V] (1976) 283 at 289-290; repr in R H Helmholz, Canon Law and the Law of England (1987), 41 at 47-48. Similarly, Ammianus Marcellinus, Res gestae 30.4.20, 22. References to classical Roman texts throughout are to the editions in the Loeb Classical Library, unless otherwise stated.   3 Pliny, Epistolae 5.13. 9; the translation is by Betty Radice. Cf Quintilian, Institutiones oratoriae 12.7.10-12.

277

278

the creation of the ius commune

This last remark refers, of course, to the Lex Cincia, a plebiscite of 204 bce,  which forbade advocates to accept remuneration for pleading in the courts.4  Although the measure was apparently seldom enforced, Augustus (r 27 bce-14  ce)  persuaded  the  Senate  to  re-enact  the  Lex Cincia  and  to  bolster  it  with  a  fourfold  penalty  for  violations.5  Enforcement  remained  lax,  however,  and  Claudius (r 31-54 ce) yielded to arguments put to him by a group of senators  who were also advocates. They insisted that it was only fair that they receive  compensation  for  using  their  skill  and  knowledge  on  behalf  of  others.  In  response Claudius agreed to abrogate the earlier law and to permit advocates  to  receive  remuneration,  but  only  up  to  a  maximum  of  10,000  sesterces  for  each case.6 Early in the reign of Claudius’ successor, Nero (r 54-68), the senate  reversed course and reinstated the Lex Cincia.7 Around the beginning of the  third century, Septimius Severus (r 193-211) changed policy once more. He  reinstituted a cap of 100 aurei on advocates’ fees, a sum virtually equivalent to  the 10,000 sesterces of the Senatusconsultum Claudianum.8 Advocates of course found ways around these limits by, among other things,  persuading clients to top up the maximum fee with a so-called “loan” (mutuum)  that had no repayment date, or to promise a “bonus” (palmarium) for winning  the case. The authorities soon caught on to these artifices and outlawed them,9  while bemoaning the greed and wiles of advocates.10

B. ADVOCATES’ FEES Medieval popes and councils, as well as lawyers trained in the ius commune,  were aware of much of this history, partly from classical and patristic references to it, and partly from the Roman law texts studied in the law schools.  A few medieval writers, notably Gandulphus (fl 1160-1170) and Placentinus  (ca 1135-1192),  inferred  from  a  statement  by  Ulpian  (d  223)  that  law  was  sacred  and  jurists  were  its  priests  whose  services  could  not  be  valued  in    4  Livy, Ab urbe condita 34.4.9; Tacitus, Annales 11.5-6, ed C D Fisher [Scriptorum classicorum  bibliotheca  Oxoniensis]  (1906;  repr  1966);  K  Visky,  “Retribuzioni  per  il  lavoro  giuridico  nelle  fonti del diritto romano” (1964) 15 Iura 1 at 17. The section of the Lex Cincia concerning gifts  to advocates has not survived; the fragments of the law that have been preserved appear in M H  Crawford (ed), Roman Statutes, 2 vols [= Bulletin of the Institute of Classical Studies, Supplement 64] (1996) II 741-744.   5  Cassius Dio, Roman History 54.18.2.   6  Tacitus, Annales 11.6-7; Pliny, Epistolae 5.4.2 and 5.9.4.   7  Tacitus, Annales 13.5 and 13.42; Suetonius, De vita Caesarum, Nero 17.   8  D 50.13.1.12.   9  C 2.6.3 and 10.65.2.1; D 50.13.1.10 and 12. 10  CTh 1.29.5; cf C 1.55.3.

doctoribus bona dona danda sunt

279

money terms, that lawyers had no right to demand fees for pleading in court  or providing legal advice.11 Most legal writers, canonists and civilians alike, rejected this interpretation.  To be sure, they deplored mercenary advocates and iurisperiti who were so  enslaved by avarice that they would take any case that promised to pay well. Such  men, an anonymous twelfth-century canonist wrote, were “abject degenerates  numbered among the vilest creatures” on earth.12 But mainstream authorities,  such as Azo (ca 1150-1230), Tancred of Bologna (ca 1185-1236), St Raymond  of  Peñafort  (ca 1180-1275),  and  Johannes  Teutonicus  (ca 1170-1245/46),  insisted that lawyers were not obliged to work free of charge. As a later writer  observed, “Advocates cannot live on air”.13 They maintained instead that most  lawyers under most circumstances were entitled to contract with their clients  about fees, and that the courts should enforce those contracts when they were  properly made.14 Canonists found authority for this position in a passage from  a  letter  of  St  Augustine  (354-430)  included  in  the  Decretum Gratiani  (ca 1140). While a judge must not sell his judgments, Augustine declared, and a  witness should not accept payment for his testimony, an advocate was free to  sell honest advocacy and a jurist was entitled to payment for sound counsel.15  Pope Innocent III (r 1198-1216) agreed, observing that no one has a duty to  fight for another at his own expense.16 The advocate’s capacity to contract for fees, however, was subject to limitations.  Jurists  generally  agreed  that  it  was  unethical,  immoral  and  illegal  for  11 D 1.1.1 and 50.13.1.5; Johannes Teutonicus, Glos ord to C. 11 q. 3 c. 71 v. iustum: “Gan  dixit quod advocatus vel iurisperitus non debet vendere consilium vel patrocinium, quia si causa  est  iniusta,  non  debet  iudex  eam  assumere,  immo  semper  acceptam  vel  assumptam  deserere  debet, C. de iudi. rem non novam [C 3.1.14]. Si est iusta, gratis tenetur consulere vel patrocinari,  et dicit Augustinum hic loqui secundum leges. Sed tu dicas quod tales operas bene potest locare  quis, nemo enim de suo cogitur facere beneficium 10 quaes. 2 Precariae [C 10 q. 2 c. 4].” See  also Placentinus, Sermo de legibus  lines 150-157,  in H  Kantorowicz, “The Poetical Sermon of  a Medieval Jurist” (1938) 2 Journal of the Warburg Institute 22 at 39; repr in H Kantorowicz,  Rechtshistorische Schriften, ed H Coing and G Immel [= Freiburger Rechts- und Staatswissenschaftliche Abhandlungen XXX] (1970) 111 at 132. 12  G Fransen and S Kuttner (eds), Summa “Elegantius in iure diuino” seu Coloniensis 6.65 [= MIC,  Corpus glossatorum I pts 1-4] (1969-1990) II 133: “Nam si lucro pecuniaque capiantur, ueluti  abiecti et degeneres inter uilissimos numerabuntur.” 13  J F Autenrith, De eo quod justum est circa salaria ac honoraria advocatorum (1727) 20-21 § 25. 14  Azo, Summa Codicis to C 2.6, in Summa Azonis (1508; repr 1966) 28; Tancred, Ordo iudiciarius  1.5.4, in F Ch Bergmann (ed), Pillius, Tancredus, Gratia, libri de iudiciorum ordine (1842; repr  1965) 113; Raymond of Peñafort, Summa de penitentia 2.5.39, ed X Ochoa Sanz and A Diez [=  Universa bibliotheca iuris I pt B] (1976) cols 517-518; Johannes Teutonicus, Glos ord to C. 14 q.  5 c. 15 sv patrocinium. 15  C. 14 q. 5 c. 15 § 1, taken from Augustine’s Epist. 153 § 23 Ad Macedonium (414 ce). 16  X 2.26.16 (Potthast 1778), paraphrasing 1 Cor. 9:7. Similarly C. 10 q. 2 c. 4 § 1; C. 13. q. 1 d.p.c.  1 § 4; C. 28 q. 1 c. 8; as well as Thomas Aquinas, Summa theologiae 2-2.71.4 ad 1 & 3.

280

the creation of the ius commune

an advocate to demand as his fee a share in the judgment awarded in a case  (in essence, what we now call a contingent fee). Gratian incorporated in the  vulgate  version  of  his  Decretum  the  substance  of  a  Roman  rule  against  this  practice that had been in place at least since the time of Constantine,17 and  most canonists were prepared to accept this.18 A handful of writers, however,  preferred the view of Martinus Gosia (ca 1100-after 1157), one of the Four  Doctors  of  Bologna,  who  distinguished  between  contingent  fees  that  would  seriously  injure  the  client’s  interests  and  those  that  would  not.  He  further  distinguished  between  fee  arrangements  that  entitled  the  advocate  to  some  specific piece of property – say, a horse – that he would receive if he won the  case and broader agreements that gave the advocate the right to claim a share  in  the  property  awarded  to  the  client  but  did  not  spell  out  precisely  which  items his share would include.19

C. CLERICS AS ADVOCATES Disputes abounded about whether clerics should be allowed to practise law at  all, much less to receive payment for doing so. The Byzantine emperor Justin  (r 518-527) had declared in 524 that it was “incongruous, indeed shameful,  for  clerics  to  wish  to  be  known  as  experienced  in  legal  disputes”,  a  view  that medieval writers often referred to and sometimes repeated verbatim.20  Medieval canonists, however, argued that this blanket prohibition was overly  broad. Master Rufinus, writing in round 1164,21 for example,  distinguished  17 C 2.6.5, paraphrased by Gratian at C. 3 q. 7 c. 2 § 8. 18 Thus, e.g., T P McLaughlin (ed), The Summa Parisiensis on the Decretum Gratiani to C. 3 q. 7 c.  2 v. arcentur (1952) 121; Summa “Elegantius” (n 12) 6.64 at II 132; Johannes Bassianus, Libellus de iudiciorum ordine § 330 in BIMAE II 211-248 at 234; Tancred, Ordo iudiciarius 1.5.4, ed  Bergmann (n 14) 113; Johannes Teutonicus, Glos ord to C. 3 q. 7 c. 2 § 8 sv partis; Bernard of  Parma, Glos ord to X 3.1.10 v. more saecularum; Hostiensis, Summa, tit De postulatione § 2 (1537;  repr 1962) fol 62rb. 19  Johannes Teutonicus, Glos ord to C. 3 q. 7 c. 2 § 8 sv partis: “De parte litis non licet pascisci, ut  ff. de pac. Sumptus [D 2.14.53]; sed super certa re licet, ar. C. de sententia passis, Si debitor [C  9.51.3], licet M dixerit quod licet de aliquota re pacisci sine gravi detrimento partis, ut  hic; tamen ultra centum aureos non licet pacisci, ut ff. de va. et extraor. co. l. 1 § In honorariis [D  50.13.1.10].” See also G F Haenel (ed), Dissensiones dominorum sive controversiae veterum iuris Romani interpretum qui glossatores vocantur, Hugolinus § 22 (1834; repr 1964) 267, and William  of Drogheda, Summa aurea §§ 46 and 54, ed L Wahrmund [= QGRKP II pt 2] (1914; repr 1962)  46, 55-56. 20  C 1.3.40(41): “Absurdum etenim clericis est, immo etiam opprobrium, si peritos se velint disceptationum esse forensium.” Cf Bernard of Parma, Glos ord to X 1.37.1 sv saecularibus and 1.37.2  sv ecclesiastici. 21  On Rufinus, see R L Benson, “Rufin”, in DDC VII 779-784; R Weigand, “Frühe Kanonisten und  ihre Karriere in der Kirche” (1990) 76 ZRG, KA 135 at 138; repr in R Weigand, Glossatoren des Dekrets Gratians (1997), 403* at 406*. But cf R Deutinger, “The Decretist Rufinus – A WellKnown Person?” (1999) 23 Bulletin of Medieval Canon Law 10.

doctoribus bona dona danda sunt

281

the situation of monks and priests, who must not appear as advocates before  secular judges, save on their own behalf or that of widows, orphans, and other  disadvantaged persons (miserabiles personae), from that of other clerics, who  might do so. Monks and priests could, however, plead civil cases, and perhaps  criminal  cases  as  well,  before  ecclesiastical  judges.22  The  Third  Lateran  Council  in  1179  extended  the  prohibition  on  clerical  advocacy  in  secular  courts to include beneficed deacons and subdeacons as well as priests.23 The  council of Reims (1131) had earlier banned monks and canons-regular from  the study of law or medicine, and pope Honorius III (1216-1226) renewed  it again in his bull Super specula.24 None of these measures proved terribly  effective.  Considerable  numbers  of  clerics,  including  priests,  monks  and  canons-regular,  found  ways  not  only  to  study  law,  but  also  to  practise  as  advocates, proctors and notaries in church courts and even in secular courts.25  Nor did it go unnoticed that those who did so often fared remarkably well in  the competition for high office in both church and state.26 A  long-standing  tradition  that  clerics  should  furnish  their  services  as  advocates without seeking remuneration gradually crumbled before the reality  that  clerics  usually  did  expect  that  their  advice,  counsel  and  representation  in the courts would be rewarded in this life as well as in the next. Gratian’s  Decretum included a canon from the council of Tarragona in 516 that forbade  clerical  advocates  to  demand  fees,  although  it  permitted  them  to  receive  spontaneous free-will offerings from grateful clients. Gratian noted, however,  that  this  canon  had  been  abrogated  by  contrary  custom.  27  Not  all  canonists  approved  of  this.  The  anonymous  author  of  the  Summa “Elegantius in iure divino”, for one, dourly observed in 1169 that the contrary custom that Gratian  mentioned  was  primarily  a  Roman  custom,  and  added:  “It  would  be  more  perfect to consider not what is done at Rome, but rather what ought to be done  at Rome and to accept the words of the canon as they are written.”28 22  Rufinus, Summa decretorum to C. 15 q. 2 pr., ed H Singer (1902; repr 1963) 347-348. 23  3 Lat. (1179) c. 12 in DEC I 218. 24  Council  of  Reims  (1131)  c.  6  in  G  D  Mansi  (ed),  Sacrorum conciliorum nova et amplissima collectio, 53 vols in 60 (1901-1927) XXI 459. The relevant section of Super specula appears in X  3.50.10.  25  L Charvet, “Accession des clercs aux fonctions d’avocat” (1966) 67 Bulletin de littérature ecclésiastique 287; as well as J A Brundage, Medieval Origins of the Legal Profession: Canonists, Civilians, and Courts (2008) 194-198, 270-271, 277-281 and passim. 26  Brundage, Medieval Origins (n 25) 344-349, 467-477. 27  C. 15 q. 2 d.p.c. 1: “Generali namque ecclesiae consuetudine receptum est et moribus approbatum, ut clerici more aduocatorum patrocinia inpendant, et pro inpendendis munera exigant, et  pro inpensis sucipiant.” 28  Summa “Elegantius” (n 12) 9.47-48, at III 72: “Hoc Gratianus de canonicis regularibus interpretatur, maxime pro Romane ecclesie consuetudine, que aduocatis et de percipiendo pactum facere 

282

the creation of the ius commune

Decretists  during  the  late  twelfth  century  and  the  early  decades  of  the  thirteenth  century  introduced  a  distinction  between  clerical  advocates  who  possessed an ecclesiastical benefice or other stipend that furnished them with  adequate support, and those who did not. Beneficed clerics, they maintained,  should not take fees for advocacy, but those who lacked such support could  do so, provided that they sought only moderate compensation.29 A Legatine  council convened by cardinal Robert Courson (ca 1160-1219) at Paris in 1212  formally adopted this position,30 and Bernard of Parma (d 1266) incorporated  it into the Glossa ordinaria on Gregory IX’s Decretals, so that it became the  standard position taught in the law schools.31

D. CONTROLLING AND NEGOTIATING FEES Medieval  authorities,  like  their  Roman  predecessors,  feared  (not  without  reason) that advocates, proctors and notaries might exploit their clients’ appre hensions about legal proceedings to justify demands for exorbitant fees. In  order to prevent, or at least to alleviate, this possibility, church leaders sought  to  place  a  cap  on  legal  fees.  Early  in  the  thirteenth  century,  for  example,  papal chancery regulations limited advocates in the courts of the curia to a  maximum fee of twenty librarum denariorum senatus.32 In 1274 the Second  Council of Lyon made a more sweeping effort to limit legal fees throughout  Latin  Christendom.  In  the  constitution  Properandum  the  council  decreed  et de non soluto actiones institutere concedit … Perfectius est non quid Rome fiat set quid Rome  fieri debeat considerare et uerba canonis uti sonant accipere.” Similarly R Sorice (ed), Distinctiones “Si mulier in eadem hora” seu Monacenses to C. 14 q. 5 c. 15 [= MIC, Corpus glossatorum  IV] (2002) 116, echoing C 1.3.40(41), called it “wicked” (turpe) for clerics to demand fees for  pleading cases. 29  See, e.g., Sicard of Cremona, Summa to C. 15 q. 2 (written ca 1170-1179); Ricardus Anglicus,  Apparatus  to  1  Comp.  1.38.1  sv  stipendiis  (written  in  the  1190s);  Summa “Prima primi uxor Adae”, to C. 15 q. 2 (written after 1203); Apparatus “Ecce vicit leo”, to C. 15 q. 2 sv pro expensis  (second recension ca 1210). For transcriptions of these texts, see Brundage, Medieval Origins (n  25) 196-197 nn 105-108. Likewise, Goffredus de Trano, Summa super titulis decretalium to X 1.37  De postulando (1519; repr 1968) 127; Innocent IV, Apparatus super quinque libris decretalium  to X 1.37.1 sv sustententur (1570; repr 1968) fol 165ra; Hostiensis, In quinque decretalium libri commentaria [= Lectura] to X 1.37.1 § 3 (1581; repr 1965) fol 182ra. 30  Council of Paris (1212) c. 6, in Mansi, Collectio (n 24) XXII 820-821. 31  Glos ord to X 1.37.1 sv stipendiis. On Bernard of Parma and the glos ord see P Ourliac, “Bernard  de Parme ou de Botone”, in DDC II 781-782; as well as S Kuttner and B Smalley, “The Glossa  Ordinaria to the Gregorian Decretals” (1945) 60 English Historical Review 97; repr in S Kuttner,  Studies in the History of Medieval Canon Law  (1990)  no  XIII.  On  the  authority  of  the  glos ord,  see  also  W  Engelmann,  Die Wiedergeburt der Rechtskultur in Italien durch die wissenschaftliche Lehre (1938), 189-204; and G R Dolezalek, “Les gloses des manuscrits de droit: Reflet  des méthodes d’enseignement”, in J Hamesse (ed), Manuels, programmes de cours et techniques d’enseignement dans les universités médiévales (1994) 235. 32  Institutio cancellarie super petitionibus § 6, in M Tangl (ed), Päpstlichen Kanzleiordnungen von 1200-1500 (1894; repr 1959) 55.

doctoribus bona dona danda sunt

283

that advocates could receive no more than twenty and proctors no more than  twelve livres tournois in a single case.33 Any who received more was supposed  to refund the surplus.34 This provision dismayed advocates at the curia, who  sought to suppress it.35 Although  legal  writers  took  note  of  Properandum,  Bolognese  jurists  preferred  to  centre  their  discussions  of  fee  arrangements  on  the  details  of  when and how to arrive at them.36 They generally agreed that it was best to  reach an agreement with the client about these matters early on. Most authorities advised advocates to negotiate their fee, as well as any bonus they might  expect if they won the case, preferably at the very outset, but definitely before  litis contestatio commenced.37 Raymond of Peñafort apparently considered that when an advocate actually  collected his fees was a matter of indifference; as he was a Dominican friar  sworn to poverty, perhaps collecting fees did not interest him greatly.38 Other  jurists took the matter far more seriously, but customary practices varied from  region to region and court to court. Prudence dictated that advocates should  receive payment after final arguments had been given but before the judge  announced  his  decision,  and  this  seems  to  have  been  a  common  practice.39  The rationale behind it was that clients were likely to be nervous about the  outcome  at  this  point  and  hence  might  be  willing  to  part  with  their  money  more readily.40 William Durand (1241-1296), the leading procedural authority  of the later Middle Ages, advised advocates to follow the customary pro cedure  on this matter at the court in which they practised. If local custom were lacking,  however,  he  considered  it  prudent  to  insist  that  half  the  fee  be  paid  when  the case began, with the other half due at the end, as was customary at the  33 2 Lyon (1274) c. 19 in DEC I 324-325; B Roberg, Das Zweite Konzil von Lyon [= Konziliengeschichte, ser 1] (1990) 314-316. 34  William Durand, Speculum iudiciale 1.4 De salariis § 3.3 (1574; repr 1975) I 347-348. 35  Properandum was in fact the only constitution of 2 Lyon that was omitted from the Liber Sextus.  Nonetheless  local  authorities  in  many  regions  adopted  its  rules  and  attempted  (with  mixed  success) to enforce them: Brundage, Medieval Origins (n 25) 303-304. 36  Jurists in northern Europe were often more interested in the ethical issues raised by legal fees: A  Gouron, “L’enseignement du droit civil au XIIe siècle: De la coutume à la règle”, in Hamesse (ed),  Manuels (n 31) 183-199 at 196; repr in A Gouron, Juristes et droits savants: Bologne et la France médiévale (2000) no VII. 37  Thus  Accursius,  Glos ord  to  C  2.6.6  sv  contractum ineat;  Tancred,  Ordo iudiciarius  1.5.3,  ed   Bergmann (n 14) 112; Durand, Speculum iudiciale (n 34) 1.4 De advocato § 3.8 and De salariis  §§ 3.2 and 3.5 at I 266, 347, 348. 38  Summa de penitentia 2.5.39, ed Ochoa and Diez (n 14) col 517. 39  Accursius, Glos ord to C 4.3.1 sv consequantur; Haenel (ed), Dissensiones dominorum, Cod. Chis.  § 131 and Hugolinus § 110 (n 19) 221-222, 342. 40  Thus,  e.g.,  Accursius,  Glos ord  to  C  3.1.13.9  v.  protrahantur;  Gratia  of  Arezzo,  De iudiciario ordine 2.9, ed Bergmann (n 14) 319 at 378; A Alciato, Commentarium to C 2.3.5 in A Alciato,  Opera omnia, 4 vols (1582) III 329.

284

the creation of the ius commune

Roman curia.41 Even there, however, practice seems to have varied. Thomas  of Evesham (d 1236) tells us that in a lawsuit at the papal consistory toward  the beginning of the thirteenth century he paid his four advocates daily before  hearings commenced.42 Baldus (ca 1327-1400), at the other extreme, tells us  that in his time it was customary (he failed to say where) to pay advocates and  law  professors  only  at  mid-year.43  It  seems  to  have  been  common  in  many  places for the client to entrust the advocate’s fee to the proctor in a case and  for the proctor to transmit it to the advocate.44 Lawyers  received  fees  for  their  advice  as  well  as  for  representing  clients  and  pleading  for  them  in  court.  Learned  lawyers,  especially  eminent  jurists  (iurisperiti) who spent much of their careers lecturing in the law schools, were  frequently  called  upon  to  provide  legal  counsel to prospective litigants, and  often to judges and public authorities as well, frequently as written opinions  called  consilia.45  They  could,  and  generally  did,  charge  for  their  opinions,  although Johannes Teutonicus suggested in his Glossa ordinaria on Gratian’s  Decretum that if a jurist could provide an opinion without paging through his  books or looking up references, he ought to do so free of charge.46 Johannes  Andreae (ca 1270-1348), for one, vehemently disagreed: By this reasoning (he wrote) it follows that I, who have laboured at my studies for  more than fifty years, so that nowadays I can deal with the present question with  little effort, will be in a worse situation than someone of modest experience who  has worked only a short time at his studies and now has to struggle in order to reply  to it. For this reason I firmly maintain that in these matters account must be taken  of past as well as future work.47 41 Durand, Speculum iudiciale (n 34) 1.4 De salariis § 3.14 at I 349. This apparently remained the  practice at the Roman rota in modern times: J J Hogan, Judicial Advocates and Procurators: An Historical Synopsis and Commentary [= Canon Law Studies, no 133] (1941) 174. 42  Thomas of Marlborough, History of the Abbey of Evesham § 285, ed and trans J Sayers and L  Watkiss (2003) 284-286. Thomas, who had taught Roman law at Oxford before becoming a monk,  was acting as proctor for his monastery in this case. 43  Baldus, Commentaria in corpore iuris civilis, 10 vols in 8 (1586) to C 2.7.26 §§ 1-2 at VI fol 132vb. 44  Gregory XI, Quamvis a felicis (1375) § 7 in Tangl, Päpstlichen Kanzleiordnungen (n 32) 130. 45  D  50.13.1.11.  See  also  I  Baumgärtner  (ed),  Consilia im späten Mittelalter: Zum historischen Aussagewert einer Quellengattung [= Studi: Schriftenreihe des Deutschen Studienzentrums in  Venedig XIII] (1995); M Ascheri, I Baumgärtner and J Kirshner (eds), Legal Consulting in the Civil Law Tradition (1999). 46  Glos ord to C. 11 q. 3 c. 71 sv iustum: “Tamen dico quod si sine labore possunt dare consilium non  reuoluendo libros, nec laborando in praelegendis allegationibus tunc gratis teneantur subuenire.”  Likewise, Raymond of Peñafort, Summa de penitentia, ed Ochoa and Diez (n 14) 2.5.39, col 518. 47  Johannes Andreae, Additio to Durand, Speculum iudiciale (n 34) 1.4 De salariis § 3.1 v. labor, at I  347: “Ex hac ratione sequitur quod ego qui ultra quinquaginta annos laboraui in studio, per quod  cum modico nunc labore studii praesentis quaerentum expedio, ero deterioris conditionis quam  qui  parum  laborauit  in  studio  et  modicum  prouectus  qui  habet  nunc  ut  respondeat  laborare;  quare constanter dico in his non solum praesentis, sed et praeteriti laboris rationem habendum.”

doctoribus bona dona danda sunt

285

He had good reason for concern. Giving formal legal opinions was an extremely  profitable business for prominent law professors such as he. Charges for their  consilia  were  usually  steep.  A  single  consilium  in  an  important  case  could  readily bring in as much as a whole year’s academic stipend or even more,  and Johannes, a married layman with children, had a vital economic interest  in preserving this lucrative source of income.48 Advocates  and  proctors  also  commonly  entered  into  agreements  with  corporate bodies, such as monasteries and municipalities, or great magnates,  such as bishops, counts, and kings, to furnish legal advice and assistance when  needed in return for an annual retainer payment.49 Retainer agreements were  sometimes set forth in formal written contracts that specified the amount of  the fee, as well as the obligations of the parties and sometimes the timing and  conditions of payment.50 E. COLLECTING FEES Although  Theodosius  I  (r  379-395)  toward  the  end  of  the  fourth  century  had admonished  clients  to  pay  their  lawyers  promptly,  medieval  advocates  and proctors were well aware that this did not invariably happen.51 A client  who had lost his case almost invariably blamed his lawyer and might well be  disinclined to pay for a disappointing outcome, while even those who won  might be reluctant to pay.52 Although in principle the advocate, proctor or  48 A  Gouron,  “Le  rôle  social  des  juristes  dans  les  villes  méridionales  au  moyen  âge”  (1969)  9/10  Annales de la Faculté des Lettres et Sciences Humaines de Nice  55  at  56-58;  M  Bellomo, The Common Legal Past of Europe, 1000-1800, trans Lydia G Cochrane [= Studies in Medieval and  Early Modern Canon Law IV] (1995) 213-214. 49  Proctors for corporate bodies were formally styled syndics; X 1.39.1 and glos ord sv generaliter. 50  Thus, e.g., Anthony Beek’s retainer agreements with an advocate and a proctor at the Roman curia  can be found in London, British Library, MS Harley 3720 fol 13v. An advocate’s claim for payment  of his retainer fee appears in M G Dupont (ed), Registre de l’officialité de Cerisy in (1880) 10 3rd  ser Mémoires de la Société des Antiquaires de Normandie 271-662 at 294-295, no 9h. For model  contracts of this sort, see P Herde, “Ein Formelbuch Gerhards von Parma mit Urkunden des  Auditor litterarum contradictarum aus dem Jahre 1277” (1967) 13 Archiv für Diplomatik, Schriftgeschichte, Siegel- und Wappenkunde 225 at 276 and 285, §§ 26 and 43, and P Varin (ed), Liber practicus de consuetudine Remensis 1.103 in Archives législatives de la ville de Reims (1840-1852)  I 135 at 106-107. Retainers sometimes took the form of appointment to prestigious and profitable  posts, rather than payments in cash or in kind. 51  C 4.3.1 pr. Cf the rescript of Septimius Severus and Caracalla quoted by Ulpian in D 50.13.1.10;  Quintilian, Institutiones 12.7.12; Ammianus Marcellinus, Res gestae 30.4. 22. 52  Helmholz, “Ethical Standards” (n 2) 47-48; R Delachenal, Histoire des avocats au Parlement de Paris, 1300-1600 (1885) 274-275. Durand, Speculum iudiciale (n 34) 1.4 De salariis § 3.6 at I 348  noted that some advocates actually demanded a larger fee for cases that they lost than for those  that they won, on the grounds that the increased fee was compensation for the damage to their  reputations that flowed from losing the case.

286

the creation of the ius commune

notary was entitled to whatever the client had promised, if payment failed  to materialise within a reasonable time it was up to the lawyer to take steps  to recover the unpaid fee.53 But what was a reasonable time? Cino da Pistoia  (1270-1336/37) claimed that it might run as much as two years, but practitioners no doubt considered this excessive.54 When a lawyer decided to take steps to collect from a slow-paying client,  his first recourse (almost certainly after a few oral hints) was usually to send  a  dunning  letter  to  remind  him  that  it  was  past  time  to  settle  up.  Thus,  for  example, Berengar de Açanis, a proctor at the Roman curia, wrote to the dean  and chapter of Tudela: My lords, I have received your letters filled with large promises and little profit, at  which I am not a little disturbed. Whether there is legitimate reason for this, God  knows … I served as your faithful proctor, working and arranging matters as well  as anyone on earth could have done to procure the documents you wanted. I even  fought against three great lions, among them Egidius Petri de Pertusa, who was  looking for the twenty petits livres tournois due to him for his fee. I sought assistance from Master Sancho, lord abbot of the monastery of St Sernin at Toulouse,  who was then an archdeacon. He helped me to reach a settlement with the aforesaid  Egidius  about  the  aforesaid  fee.  The  lord  abbot,  as  arbiter,  ordered  me  to  pay  four  livres tournois  immediately  on  behalf  of  the  church  of  Tudela,  saving  [Egidius’] claim to the remainder of his fee … By a decision of the judge Egidius  was to have only seven gold florins, and these I paid him out of my own purse …  Likewise I made a settlement with Master Andreas de Setia and paid him on your  behalf 240 silver grossi … And to be brief, you know perfectly well that you have  not even sent me my own fee – nothing at all in fact, save promises.55

If dunning letters of the kind that appear in professional handbooks56 failed  to  produce  the  desired  outcome,  writers  on  professional  matters  advised  advocates, proctors and notaries to escalate pressure on clients by refusing to  hand over the records and documents in the case until they had their fees in  53 Accursius, Glos ord to C 3.1.13.9 sv litigator; Durand, Speculum iudiciale (n 34) 1.4 De salariis §  3.10 at I 348; Helmholz, “Ethical Standards” (n 2) 47. 54  Cino da Pistoia, Commentaria in Codicem to C 2.6.3 sv si sub specie, ed N Cisnero, 2 vols (1578;  repr 1964), I fol 71rb, citing the Casus to C 2.6.3. 55  Text in P Linehan, “Spanish Litigants and Their Agents at the Thirteenth-Century Papal Curia”,  in S Kuttner and K Pennington (eds), Proceedings of the Fifth International Congress of Medieval Canon Law [= MIC, Subsidia VI] (1980) 487 at 500-501; repr in P Linehan, Past and Present in Medieval Spain  (1992),  no.  VIII.  See  also  P  Linehan,  Spanish Church and the Papacy in the Thirteenth Century [= Cambridge Studies in Medieval Life and Thought, 3rd ser IV] (1971)  289-290. 56  See, e.g., Durand, Speculum iudiciale (n 34) 4.4 De magistris §§ 1-2, 4-5 at II 486; Herde, “Ein  Formelbuch  Gerhards  von  Parma”  (n  50)  288,  §  48;  L  Wahrmund  (ed),  Der “Curialis”  §  206  [=  QGRKP, I pt  3]  (1905;  repr  1962),  61-62.  See  also  E  F  Jacob, “To  and  From  the  Court  of  Rome in the Early Fifteenth Century”, in E F Jacob, Essays in Later Medieval History (1968) 58  at 69-70.

doctoribus bona dona danda sunt

287

hand.57 This was probably more likely than a letter alone to secure payment, as  the experiences of Thomas of Marlborough during the action that the abbey  of Evesham brought against the bishop of Worcester show.58 Lawyers, if sufficiently provoked, could sometimes be considerably more ruthless in dealing  with a defaulting client. Pierre de Maucreux and Jacques la Vache, advocates  at the Parlement de Paris, for instance, faced with a client who refused to pay  them the thirty livres tournois he owed, seized the client’s house and sold it  to the highest bidder.59 Not many, however, seem to have gone quite this far. If these tactics failed to produce payment, the next step was to sue.60 Court  records show that lawyers were not shy about doing so. Nearly twenty lawsuits  over unpaid legal fees appear, for example, in the register of the court of the  official  of  the  bishop  of  Paris  between  December  1384  and  October  1386,  which averages out to nearly one case per month. They show up with considerable frequency in church court records elsewhere as well.61 By the thirteenth century the ius commune provided four different actions  for this purpose. Which was appropriate in a given situation depended on the  kind of agreement that lawyer and client had reached about fees, if indeed they  had made any agreement at all.62 If they had entered into a stipulation (which  by this time had long been stripped of its original formalities) that specified the  precise compensation to be paid – such as a specific sum of money or a definite  quantity and kind of property – the lawyer could sue to recover the unpaid fee  through a condictio certae pecuniae or condictio certae rei.63 If the stipulation  simply promised remuneration without specifying just how much it should be,  an alternative action ex stipulatu was available to deal with this contingency.64  57 D 3.3.25; D 3.3.26; William of Drogheda, Summa aurea § 95 (n 19) 90; Guillaume du Breuil,  Stillus superincliti Parlamenti cum scholiis Stephani Auffrerii (1525) fol. 2ra; Autenrith, De eo quod iustum est (n 13) 37 § 45. 58  Thomas of Marlborough, History of the Abbey of Evesham §§ 386, 390, 447-451, 505 (n 42) 372,  376, 428-432, 472-474. 59  Delachenal, Histoire des avocats (n 52) 274-275. 60  C 3.1.13.9; D 50.13.4.  61  J Petit (ed), Registre des causes civiles de l’officialité épiscopale de Paris, 1384-1387 [Collection  de documents inédits sur l’histoire de France VIII-4] (1919) cols 14, 22, 90, 94, 102, 153, 165,  173, 212, 220, 236, 275-276, 291, 294, 303, 376, 380; Helmholz, “Ethical Standards” (n 2) 47; C  Donahue, Jr, “Introduction”, in N Adams and C Donahue, Jr (eds), Select Cases from the Ecclesiastical Courts of the Province of Canterbury, c. 1200-1301 [= Selden Society Publications CV]  (1981) 25 n 1. 62  Accursius, Glos ord to D 17.1.7 v. si extra ordinem; Durand, Speculum iudiciale (n 34) 1.4 De salariis § 3.4 at I 348. 63  J Inst 3.15 pr; C 8.37.10; Hostiensis, Summa (n 18) tit De postulando § 2 at fol 62rb; Zimmermann, Obligations 78-91; Kaser, ZPR 2 111-113. Both Master Arnulphus, Summa minorum § 23,  ed L Wahrmund [= QGRKP, I pt 2] (1905; repr 1962) 20 and Durand, Speculum iudiciale (n 34)  4.4 De magistris § 5 at II 486 provide sample libelli for instituting such an action. 64  D 45.1.75pr; Zimmermann, Obligations 90; Kaser, ZPR 2 110-111.

288

the creation of the ius commune

Proctors ad litem invariably worked under a mandate contract (mandatum),  which empowered them to act on behalf of a client in a lawsuit (sometimes  called their principal), so that what they did and said carried the same force  and effect as if the client himself had said or done them. Mandate originated  as a gratuitous contract and accordingly a proctor was not legally entitled to  a fee as compensation for his services, although he could claim remuneration  for  expenses  incurred  on  the  client’s  behalf  through  an  actio mandati.65  An  undated rescript of Alexander Severus (r 222-235), however, provided that if  the client had promised to pay the proctor a specific salarium66 for his efforts  but had failed to do so, the proctor could sue for payment through cognitio extraordinaria.67 Although legal writers continued to assert that mandate was  gratuitous,68 this no longer reflected actual practice.69 But  what  if  the  client  simply  promised  that  he  would  reward  his  lawyer  without  entering  into  any  kind  of  contract  that  the  classical  law  recognised  and then failed to do so when the case ended? Did the lawyer have any legal  recourse? Classical Roman law held that he did not for, according to Ulpian: “A  bare agreement (nudum pactum) does not give rise to an obligation.”70 Canonists  insisted,  however,  that  the  agreement  became  legally  as  well  as  morally  binding if it was confirmed by an oath.71 Jurists long debated whether a bare agreement was enforceable at law unless  it was clothed (vestitum) by an oath or some other contractual formality, such  as a stipulation.72 Canon lawyers ultimately concluded that a simple unsworn  agreement created an obligation enforceable by the courts, even if it lacked the  formalities associated with traditional Roman contracts. They based this in part  65 D 17.1.1.4; J Inst 3.26.13; Zimmermann, Obligations 415-416. 66 A salarium by the time of Severus meant a specified sum of money paid as compensation for  services rendered; it retained that meaning through the Middle Ages; D 17.1.7; Accursius, Glos ord to C 4.35.1 sv salario; Zimmermann, Obligations 416-417. Accursius, Glos ord to C 2.6.3 also  used salarium as a synonym for honorarium. Visky, “Retribuzioni” (n 4) 18-19 maintains that the  proctor mentioned in D 17.1.7 must “indubbiamente” mean a proctor acting as advocate, but fails  to supply a rationale for his assertion. 67  C 4.35.1; D 17.1.7; Zimmermann, Obligations 416-418; agreement on an uncertain honorarium,  however,  was  not  enforceable;  C  4.35.17  and  Glos ord  v.  incerta pollicitatione.  On  cognitio extraordinaria, see Kaser, ZPR 2 438-439, 451-471; as well as W N Turpin, “Formula, cognitio  and Proceedings extra ordinem” (1999) 46 RIDA 499. 68  See, e.g., Ricardus Anglicus, Summa de ordine iudiciario § 18 ed L Wahrmund [= QGRKP, II pt  3] (1915; repr 1962) 21; William of Drogheda, Summa aurea § 54 (n 19) 56, but cf § 103 at 98-99. 69  Zimmermann, Obligations 418-420. 70  D 2.14.7.4: “Nuda pactio obligationem non parit, sed parit exceptionem.” Likewise D 2.14.7.5,  19.5.15; Paulus, Sententiae 2.14.1-2, in P Krueger (ed), Ulpiani liber singularis regularum; Pauli libri quinque sententiarum fragmenta minora (1878) 66. 71  C. 22 q. 5 c. 9 and d.p.c. 11; Rufinus, Summa to C. 22 q. 5 c. 9 (n 22) 403. 72  D 17.1.56.3-4; Haenel (ed.), Dissensiones dominorum, Cod. Chis. § 130 (n 19) 221; Azo, Summa Codicis to C 2.3 (n 14) 23-24.

doctoribus bona dona danda sunt

289

on the authority of the canon Quicumque suffragio of the Fourth Council of  Toledo (633), which Gratian had incorporated into his Decretum. Quicumque suffragio  declared  that  promised  remuneration  must  be  paid.73  In  addition,  the  decretal  collection  later  known  as  Compilatio prima,  which  Bernard  of  Pavia (d 1213) completed around 1190, included the canon Antigonus, drawn  from the First Council of Carthage (348), which ruled that failure to respect  simple agreements was punishable. Bernard also inserted in his collection the  canon Indicante, a selection from the letters of pope Gregory I (r 590-604) in  which the pope asserted that executors of decedents’ estates were obliged to  execute even simple verbal legacies.74 Raymond of Peñafort included all three  of these canons in the Liber Extra, which pope Gregory IX promulgated in  1234 as an official supplement to Gratian’s Decretum. It immediately became  a fundamental part of the curriculum of the canon law schools and remained  so until 1917.75 Bernard of Pavia, however, was cautious in his interpretation of these canons.  In the Summa decretalium, which he wrote not long after completing Compilatio prima,  Bernard  declared  that  simple  agreements  were  binding  unless  they were contrary to the law or good morals, but he failed to specify whether  they created only a moral obligation or whether their provisions were legally  enforceable.76 In contrast, Ricardus Anglicus (ca 1161-1242), another canon  law teacher at Bologna in the 1190s, declared that fees promised in a simple  agreement could be sued for by a condictio ex lege.77 This view ultimately found  its way into Bernard of Parma’s Glossa ordinaria on the Liber Extra, and was  adopted by most leading canonistic authorities, including cardinal Hostiensis  (ca 1200-1271),  cardinal  Goffredus  de  Trano  (d  1245),  pope  Innocent  IV  (r  1243-1254),  master  William  of  Drogheda  (d  1245),  and  bishop  William  Durand among others.78  73  C. 12 q. 2 c. 66 and Glos ord sv prouiderint. 74  1 Comp. 1.26.1 (= X 1.35.1) and 3.22.13 (= X 3.26.4) in E Friedberg (ed), Quinque compilationes antiquae, nec non collectio canonum Lipsiensis (1882; repr 1956). 75  Zimmermann, Obligations 542-544; Brundage, Medieval Origins (n 25) 118, 264. 76  Bernard of Pavia, Summa decretalium 1.26.4, ed E A T Laspeyres (1860; repr 1956) 21: “Effectus  pactorum est, ut serventur, nisi sint contra leges vel contra bonos mores …” 77  Ricardus Anglicus, Summa de ordine iudiciario § 26 (n 68) 34: “Salarium nuda sponsione constitutum petatur condictione ex lege, C. de suffragio, l. unica [C 4.3.1].” On the condictio ex lege, see  D 13.2.1; C 4.9.1-4; Zimmermann, Obligations 838-839; and J A C Thomas, Textbook of Roman Law (1976) 77-78, 328. 78  Bernard  of  Parma,  Glos ord  to  X  1.35.1  sv  pacta custodiantur  and  X  3.26.4  sv  nudis verbis;  Hostiensis, Summa, book 1, tit De postulando § 2 (n 18) fol 62rb, and Lectura to X 1.35.1 §§ 1-3  (n 29) fol 176vb–177ra; Goffredus de Trano, Summa super titulos 1.35 De pactis § 6 (n 29) 122;  Innocent IV, Apparatus to X 1.35.1 (n 29) fol 161va; William of Drogheda, Summa aurea § 54 (n  19) 54-55; Durand, Speculum iudiciale 1.4 De salariis § 3.10 (n 34) I 348.

290

the creation of the ius commune

Opinions  on  this  matter  among  civilians,  however,  remained  divided  through  out  the  thirteenth  century,  and  indeed  for  centuries  thereafter.79  Thus Azo at one point in his Summa Codicis adopted the canonists’ teaching  that an advocate could sue for a salarium promised in a nudum pactum by a  condictio ex lege, but he elsewhere declared that although a nudum pactum  created a natural obligation, it could not give rise to an action at law.80 Accursius (ca 1185-ca 1266) in his Glossa ordinaria on Justinian’s Code rejected this  position. He observed that although some people (quosdam) claimed that a fee  agreement cannot be binding unless it takes the form of a stipulation, in his  opinion a bare agreement was enforceable by a condictio ex lege.81

F. TAXATION OF COSTS An  advocate,  at  least  under  certain  circumstances,  could  avail  himself  of  a judicial process to coerce a reluctant client to pay his fees by securing a  taxation  of  costs  order.  This  procedure  became  available  as  a  result  of  the  adoption by canon law of a rescript of Justinian in 530 that obliged the judge  in cases where property was at issue to order the losing party to reimburse the  winner for the expenses he had incurred in the lawsuit. Judges were supposed  to include the award of costs in their decision, and a judge who failed to do so  could (at least theoretically) become liable to pay them himself.82 The council  of Tours in 1163 adopted the gist of these provisions in the canon Calumnia.  Bernard of Pavia included that canon in Compilatio prima, whence it passed  into the Liber Extra.83 A lawyer who had won a case and whose client failed  to pay might petition the court to tax costs, which would result in a judicial  decree  ordering  the  court’s  bailiffs  to  collect  the  fee  from  the  defaulting  79 Zimmermann, Obligations 542-544. 80 Azo, Summa Codicis to C 4.3.1 (n 14) 110: “Petitur autem salarium tribus modis, vel condictione  certi si certum fuerit promissum, vel ex stipulatu actione si in facto consistit quod est promissum,  vel condictione ex lege ubi nudo pacto tantum non per stipulationem est promissum, ut in hac l.  dicitur …” But cf Summa Codicis to C 2.3 (n 14) 24: “Si autem non fuerit nudum sed vestitum  actionem parit, sic enim ei nomen recte impono, quia si aliquod est nudum ex eo quia actio non  datur, ut infra eodem titulo l. Legem et l. Traditionibus [C 2.3.10 & 20], ergo aliquod est vestitum  quia actionem indulget.” 81  Glos ord to C 4.3.1 sv constrinxerint: “Stipulanti, alias non obligantur secundum quosdam. Sed  dicas hic nudo pacto et quod subiicit restituent, scilicet condictione ex hac lege [C 4.3.1], cum  stipulatio  non  interuenit  …  Sed  quidam  dixerunt  quod  ubi  pactum  nudum  interuenit,  de  eo  praestando detur actio praescriptis verbis, vel de dolo, si dicas incidere in speciem illam ‘facio ut  des’, per l. ff. de prescrip. verb. l. Naturalis § Quod si faciam [D 19.5.5.3], et hoc praestito prius  patrocinio, quae opinio non placet.” 82  C 3.1.13.6 and C 7.51.3, 5; Auth 82 coll 6.10 post C 3.1.15 = Nov 82.10. 83  Council of Tours (1163) c. 9 in Mansi, Collectio (n 24) XXI 1183; 1 Comp. 5.32.5 = X 5.37.4.

doctoribus bona dona danda sunt

291

client.84 A judge could also order a taxation of costs for other reasons, such as  to penalise a litigant who caused unreasonable delays in a lawsuit by making  him pay the expenses thus incurred.85 Costs included the fees of advocates, proctors and notaries, insofar as the  judge deemed them reasonable.86 Lawyers’ fees often accounted for about a  third, and sometimes more than half, of the costs that victorious parties claimed  in the bills they submitted. It is sometimes difficult or impossible to estimate  legal  fees  exactly  because  they  may  include  goods  of  unknown  value.  Thus,  for  instance,  Adam  Usk  (d  1430)  tells  us  that  his  legal  fees  for  successfully  representing  Lord  Morley  in  an  action  against  the  earl  of  Salisbury  totalled  £5 plus twelve lengths of scarlet cloth, whose value he does not specify. Other  advocates, as Maitland noted, “took their fees in bishoprics and deaneries”, and  these may likewise prove difficult to quantify with precision.87 Litigation expenses were often substantial, but it was a basic principle of  the  process  that  taxed  costs  should  not  exceed  the  value  of  the  property  at  issue in a case.88 Judges routinely reduced the claims that parties submitted in  their taxation petitions, often by a sizeable amount. William Durand provides  a hypothetical example of how a judge should bargain with the party submitting a claim: 84 Delachenal,  Histoire des avocats  (n  52)  452-453  provides  an  example  of  how  this  worked  in  practice;  see  also  Dupont  (ed),  Registre de l’officialité de Cerisy,  no  284  (n  50)  473-474;  J  E  Sayers,  “A  Judge  Delegate  Formulary  from  Canterbury”  (1962)  35  Bulletin of the Institute of Historical Research 198 at 210 § 27; repr in J E Sayers, Law and Records in Medieval England: Studies on the Medieval Papacy, Monasteries and Records (1988) no VIII. 85  L Wahrmund (ed), Der Ordo iudiciarius “Scientiam”, § 2 [= QGRKP, II pt 1] (1913; repr 1962)  6. 86  D 46.8.18; Bernard of Parma, Glos ord to X. 1.38.6 v. propter hoc; Decisiones Rotae Romanae,  edn Mainz, by Peter Schöffer (4 Jan 1477) no 533. 87  C  Given-Wilson  (ed  and  trans),  The Chronicle of Adam Usk, 1377-1421  [Oxford  Medieval  Texts] (1997) 96; F W Maitland, “English Law and the Renaissance”, in Select Essays in AngloAmerican Legal History, 3 vols (1907-1909) I 168 at 184. Examples of bills of costs appear in R  C Van Caenegem (ed), English Lawsuits from William I to Richard I, 2 vols [= Publications of  the Selden Society, CVI-CVII] (1990-1991) II 397-404; B L Woodcock, Medieval Ecclesiastical Courts in the Diocese of Canterbury [Oxford Historical Series] (1952) 136-137; D L Smail, The Consumption of Justice: Emotions, Publicity, and Legal Culture in Marseille, 1264-1423 (2003)  67-69; F Harrison, Life in a Medieval College: The Story of the Vicars-Choral of York Minster  (1952) 192-193; Delachenal, Histoire des avocats (n 52) 268-269; as well as P Alonso Romero  and C Garrigas Acosta, “El régimen jurídico de la abogacía en Castilla (siglos XIII–XVIII)”, in  Assistance dans la résolution des conflits [= Recueils de la Société Jean Bodin LXV] (1997) pt 4,  51 at 86 n 177.  88  H Muller, “Streitwert und Kosten in Prozessen vor dem papstliche Gericht – Eine Skizze” (2001)  87 ZRG, KA 138 at 138-142. For examples of vast litigation expenses, see, e.g., Thomas of Marlborough, History of the Abbey of Evesham § 433 (n 42) 414; J E Sayers, Papal Judges Delegate in the Province of Canterbury, 1198-1254: A Study in Ecclesiastical Jurisdiction and Administration  [Oxford Historical Monographs] (1971) 267; P A Linehan and J C de Lera Maíllo, Las postrimerías de un obispo alfonsino: Don Suero Pérez, el de Zamora §§ 26-27, 39 (2003) 104-106, 112.

292

the creation of the ius commune

The judge should say, “Friend, what did you spend on this case?” (or “for this appeal  that your adversary wished on you”). How much for advocates, for witnesses, and  the like?   And if [the party] replies that he spent more than seems probable – say, twenty  pounds – and the judge deems it likely that he spent only fifteen pounds, let the  judge say, “I tax you fifteen pounds. Now tell me under oath how much less than  fifteen pounds you [actually] spent.”   But if [the litigant] says that he spent twelve pounds, let the judge say, “And I  award you twelve pounds as costs,” seeking to minimise rather than to increase the  costs. And if [the litigant] says that he spent a penny less than fifteen pounds, let  the judge say, “And I therefore award you fifteen pounds minus a penny as costs.”     You  see,  therefore, how judicial taxation is done.  You also see how the oath  follows [taxation] and that costs are awarded for the sworn amount, not for less.89

G. CONCLUSION The  history  that  I  have  sketched  here  shows  remarkable  continuities  over  space and time. A widespread feeling, at least among non-lawyers, that lawyers  ought  to  make  their  expertise  available  to  clients  free  of  charge  stretches  deep into the history of the Roman republic. This ceased to be a pressing  issue in the West, to be sure, after professional lawyers virtually disappeared  in the aftermath of the so-called barbarian invasions, but it quickly revived  once formal legal training again became available in the twelfth century and  professional lawyers began to reappear.90 Lawyers  naturally  tend  to  see  matters  differently.  They  argue  that  they  deserve  compensation  for  the  time,  trouble  and  skill  they  invest  in  serving  their clients’ interests. Public authorities since at least the time of the emperor  Claudius  have  frequently  adopted  a  compromise  between  these  opposing  views by legitimising payment for legal services but imposing limits on how  much lawyers are allowed to charge for them.  Clients,  ancient,  medieval  and  modern,  often  seem  mystified  about  the  89 Durand, Speculum iudiciale (n 34) 2.3 De expensis § 6 pr. at I 907-908: “Debet enim iudex dicere:  ‘Amice,  quid  expendisti  in  causa,  uel  ratione  huius  appellationis  a  tuo  aduersario  interpositae;  quid expendisti in aduocatis, quid pro testibus et similibus,’ ut C. de iudi Sancimus [C 3.1.15].  Et  si  ille  responderit  ultra  quam  sit  uerisimile  eum  expendisse,  puta  dixerit  XX.  lib.,  iudex,  si  sibi uerisimiliter uidetur quod expendit tantum XV., iudex dicet, ‘Et ego taxo XV.; unde dic per  tuum sacramentum, quantum expendisti infra XV’. Quod si ille dixerit quod expendit XII., iudex  dicet, ‘Et ego tibi condemno in XII. nomine expensarum’, eas restringens potius quam laxans. Et  si ille responderit se expendisse XV. minus uno denario, iudex dicat, ‘Et ego tibi in quindecim  condemno minus uno denario nomine expensarum.’ Vides igitur quomodo praecedit iudicialis  taxatio.  Vides  etiam  quomodo  post  eam  sequitur  iuratio,  et  post  eam  condemnatio  in  id  quod  iuratum est, non enim fieri debet in minus, ut ff. de in lit. iur. In actionibus [D 12.3.5].” For a  concrete example of how the process actually worked in the case of Forde c Bunse (1498), see  Woodcock, Medieval Ecclesiastical Courts (n 87) 126. 90  On this, see Brundage, Medieval Origins (n 25) chs 6 and 7.

doctoribus bona dona danda sunt

293

justification  for  their  lawyers’  fees.  Many  of  a  lawyer’s  services  are  oral  and  fleeting – they give advice, they argue before a judge, sometimes about matters  that the client may find unintelligible or even silly. The only tangible product  of the lawyer’s work that a client can touch and feel are a few documents filled  with  seemingly  impenetrable  gibberish.  From  the  client’s  viewpoint  the  fee  that his lawyer demands can easily seem wholly disproportionate to what he  has done. For that reason, at the close of the lawsuit or other matter that led  the client to engage his lawyer’s services the client often feels reluctant (and  reasonably so in his own mind) to pay more than a nominal fee for the services,  he has received—particularly if the result has not been the one he wanted to  begin with. I  have  tried  to  show  how,  faced  with  clients  who  paid  grudgingly,  and  sometimes not at all, medieval lawyers either adapted Roman contract law to  force clients to part with the fees they owed or else, in the case of the nudum pactum, as Reinhard Zimmermann put it, they “used a few Roman bricks in  order to create a totally un-Roman edifice”.91

91 Zimmermann, Obligations 549.

Index

The index is arranged in word by word order. References to authors are those discussed in the text. Accursius bare agreement, 290 causa, 97 cumulatio, 60 discharge of smoke, 83 fraud, 100 generally, 24–5, 27, 50–1, 54 Glossa ordinaria, 225–6 heir, 151 just price, 81–2 latent defects, 195–6, 198, 200, 216 liability of statements, 85 linking of texts, 95–6 Lombard laws, 18 ownership, 131–2, 135 right to build a mill, 87 schema, 99–100 succession, 146, 147 tenancy, 170–1 two separate remedies quanti minoris, 198 usability of a coin’s metal, 227 actio dotis, 113 actio empti, 178–83, 185, 186, 190, 191, 192–4, 196–8, 208; see also liability for latent defects actio ex lege, 125 actio in factum, 180, 182–3 actio quanti minoris, 176, 177–83, 190, 192–4, 196, 208; see also liability for latent defects actio quanti minoris civilis, 196, 197, 199, 200, 205, 208–14; see also liability for latent defects actio quanti minoris praetoria, 190n, 193–4, 196, 200, 207, 208–14;

see also liability for latent defects actio redhibitoria, 176, 177, 180, 182–3, 197 action of status, 121, 122, 125 Ad decus constitution, 16, 50 Adam Usk, 291 Adriano Cappelli, 15 advocates’ fees, 278–80; see also legal fees: recovery of unpaid aediles curules, 177 Aegedius de Fuscarariis, 255–6 affectio maritalis, 114 Africanus, 63, 164 agnates, 146 agricultural tenancy, 162–4, 168–9; see also hypothec Albericus de Rosate, 29, 54, 62, 137–8, 186, 208 Alexander III, 124 Alexander Severus, 288 Andreas Bonellus de Barulo, 153 Andreas de Iserna, 52 Angelus de Ubaldis, 206, 212 Anselm of Laon, 119 apparatus, generally, 15, 23–5 arguing, art of Belleperche’s dispute with Revigny, 66–70 Revigny’s citation of texts, 62–6, 72 Aurea practica libellorum, 200–1 autenticae, 16 Authenticum, 16–17 Azo Portius Brocardica sive generalia iuris, 234–6, 238, 243 bullion and coins, 228–9

296

index

concurrence of actions, 59–60 fraud, 100 generally, 22, 23, 24, 26, 27, 29 hypothec, 167–9, 170 latent defects, 184, 190–4, 195, 199 legal fees, 279, 290 nudum pactum, 290 ownership, 133, 134, 135 price reduction, 216 Summa Codicis, 191, 192, 193 usability of a coin’s metal, 227

Baldus de Ubaldis causa, 98, 99 commentaries, 29, 84 concurrence of actions, 62 hypothec, 172–4 latent defects, 204, 210–11 legal fees, 284 Bartholomeus Brixiensis, 41, 47, 54 Bartholomeus de Saliceto, 205–6, 211–12 Bartolus de Saxoferrato causa, 97–8 citation, 267 concurrence of actions, 73–4 consensual contracts, 98 doctor de Aurelianis, 139–41 generally, 23, 29 ignorant sellers, 205 interference with land, 84 ownership, 4,127–8, 129, 131, 139–42 praetorian actio quanti minoris for slaves and animals, 207–8 price reduction, 209–10, 214, 217 public policy arguments, 90 right to build a mill, 87–8, 89–90 rules, 61 Bassianus, Johannes see Johannes Bassianus Battista da Sanbiagio, 212–14 Beckhaus, F C G, 24 Belleperche, Pierre de see Pierre de Belleperche Bellomo, M, 2, 11, 21 Beneventanus, R, 23 Berengar de Açanis, 286 Bernadus Compostellanus, 40 Bernadus Parmensis de Botone, 54, 136, 137 Bernard Balbi, 38, 40, 41, 42

Bernard of Parma, 44, 45, 282, 289, 290 Bezemer, K, 51 bibliographical surveys, 159 bibliographical tools, 9–11 Biblioteca Medicea Laurenziana, 13 Bibliothèque Nationale Paris,10 Blackstone, W, 83, 88 Blume, F H, 13 Bolognese bushel, 229–31, 233 Boniface VIII, 45 Breviarium extravagantium, 38, 40, 41, 42 Bridrey, E, 244 brocardica, 22 Brundage, J A, 11 Bulgarus apparatus, 23, 24 Bolognese bushel, 231, 233 generally, 53, 55 liability for defects, 186 questions, 22 summae, 25, 26 bullion, 221, 224, 228, 229 Burgundio of Pisa, 14 bushel, 229 Canning, J, 2 canon law, 2–3, 30 caro unus, 109 Carolo de Tocco, 18 Casus Codicis of Wilhelmus de Cabriano, 3, 55, 186–7, 188 causa doctrine, 97–8 children emancipation, 108 legitimi, 110 liability of tutor, 104 natural, 118 natural obligations, 107 status, 121, 122, 123, 125 status of liberty, 114 succession, 104, 118, 121 Cinus of Pistoia citation, 267 comparison with Dinus, 72–4 distinctiones, 71, 73 generally, 28–9 latent defect remedies, 212–13, 214, 216 liability of statements, 85, 86

index

price reduction, 203–4, 208, 209, 210, 216, 217 recovery of unpaid legal fees, 286 statement of claims, 201 succession, 148, 150, 151–2, 153, 154 citation assignment of the day for appearance, 260–4 canon law, 251–2 citationes indiscretae challenge, 264 contents, 264–9, 274 dates of hearing, 265–6 day stated, 271 Decretals, 251, 252 defendant to be fully informed, 264, 265 definitive sentence, 267–8 delay before appearance, 260–1, 274 discretion of judges, 262, 263, 264, 274 distinctions, 262–3 English court practice, 269–72 exceptions, 258 excommunication, 254–5, 259 flexibility, 266–7, 268, 274 generally, 247–8, 273–5 in ius vocatio, 249 ius commune formation, 263 jurisdiction, 265–6, 274 mandates, 270 multiple, 270–1 nature of claim, 267 necessity for, 254–9 new material, 268 Norwich treatise, 271–2 ordo iuris, 255–6, 272, 273 reason for, 259 Roman law, 249–51 sources, 248–52 unnecessary delays in litigation, 257 validity in Scotland, 268–9, 272 Claudius, 278 Clementinae, 46 Codex Florentinus, 13–15 codex rationum, 107 Codex secundus, 14 cognito extraordinaria, 288 cohabitation, 108–9, 114, 118, 121–4 coin debasement adapting lex cuum quid to monetary doctrine, 236–8 Azo’s brocard, 234–6, 238, 243

297

Bolognese bushel, 229–31, 233 canon law, 239–44, 245–6 eadem mensura rule, 234 generally, 224–5 glossators on, 229–38 nominalist minority position, 241–3, 244 olim causam, 240–1 Pillius’ quaestiones, 231–4 Quedlinburg monastery archives, 243 Coing, H, 11, 54, 159 coins circulation, 224 debasement see coin debasement dual use, 227 exchangeability, 229 hoard, 221 minting, 220 personal chattels, as, 220, 221 seignorage, 228 standards of measurement, 239–41 underweight, 228 usability, 227 see also money coitus matrimonii, 124 colonus, 163 comitia conturiata, 106 commentum, 26–7 commixtio, 221, 226 commodity money, 220, 221, 244–5 community of goods, 109 Compilatio prima, 289 Compilationes antique, 40–3 Concordat of Worms, 119 concubinage, 112 concurrence of actions, 59–62, 71, 73–4 condemnatio pecuniaria, 234, 235 condictio ex lege, 289, 290 confarreatio, 113 conflict of laws, medieval jurists, 91–2 coniugium, 111, 116, 119, 123–4 coniumctio, 111 consilia, 29 consumptio nummorum, 221, 226 contingent fees, 280 contracts, consensual, 98–9 contrahere matrimonium, 113–14 conubium, 114 Corpus iuris canonici Clementinae, 46 Compilationes antique, 40–3

298 early glosses to Decretum, 35–6 early summae, 36–7 extravagantes, 39, 43, 46 Gratian’s Decretum, 31–6 ius novum, decretals, 37–40 Liber extra, 40, 43–5 Liber sextus, 45–6 Corpus iuris civilis abbreviations, 15 Codex Justinianus, 15–16 chronological survey of printed editions, 14n Digest, 13–15 index, 15 Justinian’s novellae, 16–17 references to Roman law texts, 18–20 text, 12–20 Volumen, 16, 17–18 corpus nummorum, 221 Cortese, E, 11, 139 cumulatio, use of term, 59 customary law, 68 Damhouder, J, 257 Dayoz, 15 De actione et eius natura, 212–14 Decretales Gregorii IX, 43–5, 136, 137 decretals, generally, 37–40, 43–5 Demoulin-Auzary, F, 122 denarius, 223–4, 228 differentiae legum et canonum, 30 Dinus de Mugello comparison with Cinus, 73, 74 disclosure in public interest, 85 generally, 24 heir, 152, 153, 154 materia cumulationis, 60, 61 natura feudi, 155–6 regulae, 71 disputations, 23 dissensiones dominorum, 53, 55 distinctiones, 22 divorce, 30, 114, 115, 125 Dolezalek, G, 10, 59 domina, 108 dominium directum, 129, 137, 138 dominium utile doctrine Bartolus, 127–9, 139–42 conflict with Roman property law, 131–41

index development, 130–1 longi temporis praescriptio, 130 dominus, 108, 109 Donahue, C, 2–3, 21 double-entry book-keeping, 107 dowry, 113, 114, 133, 134, 136, 137 Duggan, C, 11 Dumoulin, C, 244 duo non possunt habere dominium eiusdem rei in solidum, 131–41 duplex dominium, 132–6 Durand, Guilliaume see William Durand duty of disclosure, 62–4 ecclesiastical courts, 120–1 Edict of Theodoric, 115 eiusdem generis et eadem bonitate, 233 Epitome Iuliani, 17 excommunication, 254–9, 259 extravagantes, 38, 39, 43, 45–6, 50 familia, 106–10 actio familiae erciscundae, 147, 154 family law, 106–10 famuli, 106, 107 Feenstra, R, 130 female succession, 146–7 feudal law, 49–52 fiducia, 160–1 First Council of Carthage, 289 Florentina, 13–15 Four Doctors of Bologna, 22, 23, 53, 55, 280 Fowlier-Magerl, L, 48 fraud, 100 Friedberg, E, 31, 33, 34, 40, 44 Fulbert of Chartres, 18, 49, 50 fundum tributarium, 189 Gaius, 104, 107, 114 Gandulphus, 278 García y García, A, 42 Gaudemet, J, 123 Gebauer-Spangenberg edition of Corpus iuris civilis, 15 gifts and donations, 114 Giovanni Crispo de Monti, 216 Glossa ordinaria, 24, 25, 32, 41 glossae, meaning, 21 glossators, meaning, 21–2

index

299

glosses, early development, 21–2, 225–6 Gombrich, E, 90 good faith, 178, 180, 181 Gordley, J, 33 Gouron, A, 25, 54 Gratian’s Decretum citations in procedure, 251, 252, 253, 260 clerics as advocates, 281 composition, 120 Corpus iuris canonici, 30, 31–6 early glosses, 35–6 early summae, 36–7 ecclesiastical courts, 120–1 editio Romano, 32, 33 feudal law, 49 legal fees, 280 matrimonium, 117–19 palea, 32, 33, 35 principal parts, 32–3 recensions, 34 Roman law added to, 35 system of citations, 33–4 Gregory I, 289 Gregory IX, 25, 40, 43, 45, 136, 137, 289 Gregory X, 45 Guibert de Bornado, 35 Guillelmus de Cabriano, 55 Guillelmus de Cuneo, 28, 148–50



implied term, 165, 167 legal remedies of creditor, 161 listing of specific objects, 164 Lo Codi, 166–7 medieval learned law, 165–74 principle of subsidiarity, 162 Roman law, 160–5 sancta rusticitas, 168–9 security, 160 Summa Trecensis, 166, 167 tenancy, 162–5, 168–9 urban tenancy, 162, 164–5 valuable movable objects, 165

habere dominium alio iure, 132–6 Habita, 16 Heimbach, G E, 17 heir, discussion as to meaning, 150–4 Helmholz, R H, 11 Henricus de Baila, 186 Honorius III, 43, 281 Hostiensis, 44–5, 105, 125, 243 Hughes of Saint-Victor, 123 Hugo, 96 Hugo à Porta, 14 Hugolinus, 24, 27, 194–5, 228, 229 Huguccio, 39 hypothec actio conducti, 167 agricultural tenancy, 162–4, 168–9 crops, over, 163 general contractual clause, 164–5 generally, 159–60 goods invecta et illata, 162, 165, 167

Jacobus Balduini, 195 Jacobus Butrigarius, 23, 28, 74 Jacobus Columbi, 51 Jacobus de Ardizone, 50, 143–4 Jacobus de Arena, 24, 27, 91–2 Jacobus de Belvisio, 156 Jacques de Boulogne, 69 Jacques de Revigny arguments by analogy, 65–6 causa, 97 duty of disclosure, 62–4 heir, 150, 151–2 lectures, 23, 28 liability of statements, 85, 87 price reduction, 201, 202, 203, 208 property, discharge onto another’s, 83, 84 regula, 60 tenancy, 172, 174 title to a fief, 147–8

incest, 111, 118, 119 inheritance rights, 118 Innocent, 40–1 Innocent III, 41, 136, 279 Innocent IV, 44, 45, 105 intestacy, 110 invecta et illata, 162, 165, 167 Investiture Controversy, 119 iura naturalia, 105 ius accrescendi, 153 ius commune feudorum, 4, 143–4; see also Libri feudorum ius gentium, 95–6 ius liberorum, 104 ius patronatus, 144

300

index

Jakobs, H H, 24, 135 Jean Chappuis, 31, 46 Jean Faure, 71–2 Jean Nicot, 139–40 Jodocus, 12 Johannes Andreae, 46, 284–5 Johannes Bassianus Arbor actionum, 48, 49 citation, 253 generally, 24, 26, 27, 29, 53 ignorant sellers, 188–90, 192, 195, 200 latent defect remedies, 214 ownership, 133, 134, 135 price reduction, 216 tree of actions, 194 vassals, 144 Johannes Bertachinus, 54 Johannes de Lignano, 257 Johannes Fasolus, 51 Johannes Teutonicus, 41, 42, 241–2, 279, 284 Julius Clarus, 257 just price arguments, 81–2 Justin, emperor, 280 Kantorowicz, H, 5 Kaser, M, 128, 129 Kelly, A, 44 kidushin, 108, 109, 116, 124 Koptev, A, 8 Kriechbaum, M, 9, 73, 159 Krüger-Mommsen edition of Corpus iuris civilis, 12, 13, 16 Kuttner Institute, 9 laesio enormis, 80, 81 land, interference with use of, 83–4 Landau, P, 11 Lanfranc du Bec, 120 Lange, H, 9, 159 Latin vocabularies, 12 law of persons, 104–5 lectura, 26–7 legal fees: recovery of unpaid action ex stipulatu, 287 advocates’ fees, 278–80 bare agreement, 288 beneficed clerics, 282 clerics as advocates, 280–2 collection, 285–90

condictio certae pecuniae, 287 condictio certae rei, 287 condictio ex lege, 289, 290 contingent, 280 controlling and negotiating, 282–5 dunning letter, 286 generally, 292–3 mandate, 288 mutuum, 278 palmarium, 278 proctors ad litem, 288 refusal to hand over documents, 286–7 retainer agreements, 285 Roman law, 277–8 salarium, 288, 290 suing for, 287 taxation of costs, 290–2 legal tender, 221 letting and hiring contract, 162, 164–5 Lex Cincia, 278 lex origo emendi, 226 Lex Romana Burgundionum, 116 liability for latent defects Accursius, 195–6, 198 actio empti, 178–83, 185, 186, 190, 191, 192–4, 196–8, 208 actio quanti minoris, 176, 177–83, 190, 192–4, 196, 208 actio quanti minoris civilis, 196, 197, 199, 200, 205, 208–14 actio quanti minoris praetoria, 190n, 193–4, 196, 200, 207, 208–14 aedilician actions, 180, 182–3 Azo, 190–4 Bassianus, 187–90 Cinus of Pistoia, 201, 203–4 commentators, 204–13 condictio ex lege rem maioris, 184 Corpus iuris, development from, 177–9 early legal scholarship, 185–7 European Directive, 176 fair price, 184 generally, 176–7, 215–17 glossators, 184–98 Hugolinus, 194–5 ignorant sellers, liability of, 205–6 iustum pretium, 184 Justinianic law, 179–84 laesio enormis, 184

index

Odofredus, 195–6 Placentinus, 187–90 praetorian remedies, 197 price reduction, 181, 206–14 remedies in Corpus iuris, 179–80 Roman law, 177–84 Romano-canonical procedure, 199–201 school of Orleans, 198–9, 201–4 seller in good faith, 178, 180, 181 statement of claims, 199–201 liability of statements, 85–6 libel, 199–201 Liber Extra, 40, 43–5, 136–7 Liber Sextus, 45–6 liberi, 106 libra, 224 Libri feudorum generally, 17, 18, 49–52, glosses, 144, 145 succession, 145–7 supplementary agreements, 154, 155 validity as source of ius commune, 156 Ligurian law, 109 Litewski, W, 48 Littera Bononiensis or Vulgata, 14, 15, 16 Littera Florentina, 13–15 Littera Pisana, 13 Lo Codi, 166, 167, 185 locatio conductio, 162 Lombarda, 18 longi temporis praescriptio, 130 Lotharius, 87 Lyon, Councils of, 44, 45, 282–3 McLaughlin, T P, 37 McManus, B, 42 mandate, 288 maritagium, 112 maritum habere,111 marriage, 110–19, 123–5 Martini sequaces, 54 Martinus (Gosia), 24, 25, 53, 184, 280 materfamilias, 108 materia forma publica percussa, 220 matrimonium, 113–19, 123–4 matrimonium legitimum, 118 matrimony early medieval legal texts, 115–19 twelfth-century case law, 119–23

301

Medicinensis, Pilius see Pillius de Medicina medieval jurists application of texts to new situations, 80–93 ascribing explanatory power to terms, 97–9 conflict of laws, 91–2 imposition of order, 93–101 juxtaposition of texts, 94–7 linking related texts, 93–4 order through a schema, 99–100, 101 medieval learned law apparatus, 23–5 commentum, 26–7 consilia, 29 disputations, 23 feudal law, 49–52 index to decretals and gloss, 44 lectura, 26–7 procedural law, 47–9 repetitio, 27–9 research strategy, 52–6 summae, 25–6 teaching and writing, 20–9 Meijers, E M, 109 Metzger, E, 8 mill, right to build, 87–8, 89–90 Mitteis, L, 115 money Bolognese bushel, 229–31, 233 bullion 221, 224, 228 coins see coins commodification , 220, 221, 244–5 denarius, 223–4, 228 different currencies, 222, 223 glossators’ environment, 223–5 Justinian’s laws, in, 220–3 libra, 224 nature of, 226–8 unit of measurement, as, 238 moveables, 68, 70 Müller, J, 9 Müller, W P, 39 Murano, G, 10 mutuum, 228, 235, 236, 278 natura feudi, 150, 154–6, 157 naturales liberi, 118 Norwich treatise, 271–2

302

index

notabilia, 22 Novel of Valentine III, 120 novellae, 16–17, 19 nudum pactum, 288, 290, 293 nummi, 220 nuptiae, 111, 112–13, 116 oath of fealty letter, 18, 49, 50 Obertina, 49 Obertus de Orto, 18, 49, 146–7 Odofredus de Denariis discharge of smoke, 83, 84 duplex dominium, 132, 133 generally, 20, 27, 29, 54–5 habere dominium alio iure, 132–6, 137 latent defects, 195, 196, 198, 199 ownership, 132–7 urban tenancy, 16–70 olim causam, 240–1, 243 order: imposition on texts, 93–100 Ordinaturus magister, 36 ordo iudiciarius, 47 Osler, D, 10 Otto Papiensis, 22, 187 ownership Bartolus, definition, 127–8, 129, 131, 141–2 different modi in canon law, 136–7 dominium utile, compatibility with duo…solidum rule, 139–42 dowry, 133, 134, 136, 137 feudal law, 128 French Code civil, 127–8 glossa ordinaria, 131–2 habere dominium alio iure, 132–6, 137 modern concept, 128 partial, 131–2 Roman law, 128 synthesis, 137–8 vertical division, 128n pacta sunt servanda, 30 pactum adiectum, 67 palmarium, 278 Panormitanus, 257 Paris, Council of, 282 paterfamilias, 107, 108, 109, 114, 115 paternal fief, 146 Patrologia latina, 32 Paucapalea, 36, 37

Paul, 1,164, 221, 226 Paulus de Castro, 205, 212, 214 pecunia numerata, 224 pecunia reproba, 225 Pennington, K, 9, 11, 42 person, use of term, 104–5 persona representata, 105 Petrus Beneventanus, 40 Petrus de Bellapertica, 85–6, 97 Petrus Jacobi, 200–1 Peter Lombard, 124 Pierre de Belleperche generally, 28 heir, 150, 151–2 latent defect remedies, 212–13, 214, 216 dispute with Revigny, 66–70 price reduction, 202–3, 204, 208, 209, 210, 216, 217 title to a fief, 148, 153 pignus, 4, 160–1, 168, 173n, 174 Pillius de Medicina citation, 253–4 generally, 144 questions, 23, 230, 231–4, 237 Quidam creditor Lucenses, 231–4 recensio Ardizona, 50 Pius V, 32 Placentinus citation, 253 claims, 48 generally, 24, 26 ignorant sellers, 187 legal fees, 278 pledge, 160, 161–2, 165 Pliny the younger, l277 Pomponius, 232, 233, 236, 237 possession, 96–7 possessiso status, 121, 122 possessorium, 70 preiudicium, 121, 122 price reduction for defects, 181, 206–14 procedural law, 47–9; see also Speculum iudiciale Properandum, 282–3 property, smoke discharge onto, 82–4 public interest disclosure, 86–7 public policy arguments, 90 public road, right to use, 88

index

303

Quaestiones Dominorum Bononiensium, 230 quanto minoris empturus esset, 181, 196, 197, 200, 206, 207 Quia iudiciorum, 47 Quicumque suffragio, 289 Quintus Neratius Priscus, 163–4

ownership, 128 persons, 104–5 references to texts, 18–20 transposition tables, 20 use of cases to explain concepts, 77–9 Römisches Recht im Mittelalter, 9, 159 Rufinus, 36, 37, 280–1

Rabanus Maurus, 117, 118 Rainier of Pomposa, 40 Raphael Fulgosius, 208 Rather of Verona, 118 ratione materiae, 120 ratione personae, 120 Raymond of Peñafort, 43, 44, 279, 283, 289 recensio antiqua, 49 recensio Ardizona, 50 rectum feudum, 150, 154–6 Regino of Prüm, 116–17 regula, use of term, 75 regulae concurrence of actions, 59–62 development, 58 examples of use of citations, 62–6 France, 70–2 freedom to argue, 66–70 generally, 57–9 Italy, 72–4 Jacobus Butrigarius, 74 Jean Faure, 71–2 Reims, Council of, 281 repetitio, 27–9 rescriptum, 69 retrait lignager, 67, 68 Revigny, Jacques de see Jacques de Revigny Ricardus Anglicus, 41–2, 289 Richardus Malumbra, 93 Robert Pullus, 123, 124 Robertus Maranta, 258, 267–8 Roffredus Beneventanus, 23, 48–9, 200, 214 Rogerius, 24, 25–6, 185–6, 187, 201, 203 Roland, 36, 37 Roman law canon law and, 30 family, 106–10 liability for latent defects, 177–84 marriage, 110–15

Sacrementa puberum, 16 Saint Augustine, 111, 279 Saint Jerome, 116, 169 saisine, 68, 70 salarium, 288, 290 sale and barter distinction, 221 sancta rusticitas, 168–9 Savigny, F C von, 1, 241 Scaccia, S, 267 schema, 99–100, 101 Schultz, F, 77–8 Scott, S P, 13 Second Lateran Council 1139, 32 security, 160–1 seignorage, meaning, 228 Septimius Severus, 278 sexual relationships, 110, 116, 117, 122 Shaw, Justice, 88 sic utere tuout neminem laedas, 83 Simon de Bisignano, 30, 38, 39 Sinibaldus Fliscus, 44, 45, 105 smoke discharge, 82–4 Speculum iudiciale, 47–8, 51, 200, 256, 266, 283, 289, 21–2 statement of claims, 199–201 status of liberty, 106 Stein, P, 2, 77–8 stipulatio, 97–8 subscriptiones, 16 succession Accursian gloss, interpretation, 147–54 children, 118 generally, 143–4 Libri feudorum, 145–7 natura feudi, 150, 154–6, 157 rectum feudum, 150, 154–6 Summa Trecensis, 25, 47, 54, 146, 166, 167 summae, 25–6 tabulae nuptiales, as evidence of marriage, 113

304

index

tacit emancipation by cohabitation, 108–9 tacit emancipation by wedding, 108 Taeuber, W, 225, 242, 244 Tancred of Bologna citation, 254, 255, 260–2, 263, 265 generally, 41, 42, 47, 48 legal fees, 279 Tanta est vis matrimonii, 124 tantum valet unus nummus quantum argenti tantundem in massa, 227 taxation of costs, 290–2 tenancy agricultural, 162–4, 168–9 urban, 164, 172–4 see also hypothec Theodosius I, 285 Third Lateran Council, 281 Thomas of Marlborough, 284, 287 Thompson, A, 33 Toledo, Council of, 289 toll collection: demonstration of arguing, 65–6 Tours, Council of, 38, 290 transposition tables, 20 Tres libri codicis, 16, 17, 24, 26, 27, 29 trial, medieval procedure, 47–9 tutela, 104, 105 Twelve Tables, 106, 107 Ulpian hypothec, 161 legal fees, 278–9 nudum pactum, 288 order, 94, 95, 96–7 ultramontani, 152, 153, 165, 172 under the huppa, 108

urban tenancy, 164, 172–4; see also hypothec usucapio, 143 uxorem ducere, 111 Vacarius, 100, 188 Valantine III, 120 Vantius, 258, 263 vassals, 143–4 Vervaart, O, 9 Viglius of Ayytta, 216–17 Vincentius Hispanus, 242 Vinogradoff, P, 1–2 vir et uxor, 111–12 Virgil, 188, 192, 195 Vismara, G, 120 Vocabularis utriusque iuris, 12 Voet, J, 6 Volumen, 16, 17–18 Vulgata, textual variation, 13–15 Vulgate of Saint Jerome, 116 Watson, A, 12, 87 Weigand, R, 11, 32, 33, 35–6 Weimar, P, 21, 51 Wieacker, F, 2 Wilhelmus de Cabriano, 3, 55, 186–7, 188 William Durand, 47–8, 51, 200, 256, 262–3, 266, 283–4, 291–2 William Naso, 44 Winroth, A, 31–2, 34 Young Research Library (UCLA), 32, 44 Zimmermann, R, 78, 293