Historical introduction to the study of Roman law 9780511995019, 9780521082532, 9780521088756

Citation preview

HISTORICAL INTRODUCTION TO THE STUDY OF

ROMAN LAW (THIRD EDITION)

BLANK PAGE

HISTORICAL INTRODUCTION TO THE STUDY OF

by

the late H. F. SOLOWICZ sometime Regius Professor of Civil Law im the University of Oxford

and

BARRY NICHOLAS Professor of Comparatwe Law in the Unwersity of Oxford Fellow of Brasenose College

THIRD EDITION

CAMBRIDGE UNIVERSITY PRESS CAMBRIDGE

LONDON? NEW YORK: MELBOURNE

CAMBRIDGE UNIVERSITY PRESS Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo, Delhi

Cambridge University Press The Edinburgh Building, Cambridge CB2 8RU, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/978052 1082532

© Cambridge University Press 1972 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First edition by H. F. Jolowicz 1932 Reprinted with corrections 1939 Second edition by H. F. Jolowicz 1952 Reprinted with corrections 1954, 1961, 1965 Third edition by Barry Nicholas 1972 Reprinted 1978 This digitally printed version 2008 A catalogue record for this publication is available from the British Library Library of Congress Catalogue Card Number: 74—164452

ISBN 978-0-521-08253-2 hardback ISBN 978-0-521-08875-6 paperback

| page Preface xi Dates xiii Contents

Abbreviations XV Periods in history 1 Periods in the history of the law Ay

1 Periods in the history of Rome and in the history of her law 1

Elements 8

2 The republican constitution 8

The struggle between the orders 9

The assemblies of the people 17

The senate , 30 The magistracies 45

Characteristics and procedure of Roman assemblies 27

Italy 58 The provinces 66

8 Italy and the provinces during the republic 58

Excursus — Double citizenship 71

during the republic 75

4, Social conditions and the different classes of the population

5 Sources of law in the republic 86

Lex and plebiscitum 86 Interpretatio 88 Edicta magistratuum 97

Custom 101 Vv

Contents

6 Law for foreigners, ius gentium and ius naturale 102

7 The XII Tables 108

Tables 114 Marriage 114 Patria potestas 118

8 The law of the family and of succession at the time of the XII

Guardianship 121 The agnatic relationship 122

Succession on death | 123 Intestate succession 124 Testamentary succession 126

9 Slavery and manumission at the time of the XII Tables 133

10 The law of property at the time of the XII Tables 137

Ownership 140 Methods of acquiring ownership 143

Res mancipi and nec mancipti 187

Restrictions on ownership 156

Introduction 159 Nexum 164

11 The law of obligations at the time of the XII Tables 159

Delict 167

Trial 176 Execution 187

12 The law of procedure at the time of the XII Tables 175

Summons (in tus vocatio) 175

procedure 191

13 Private law from the XII Tables to the fall of the republic:

Changes in the system of legis actiones — 192

The formulary system 199 vi

Contents

Marriage 233 Patria potestas and bondage 238

14 Private law from the XII Tables to the fall of the republic:

the law of the family and of succession 233

Guardianship 239 Succession 242

15 Private law from the XII Tables to the fall of the republic:

the law of slavery and manumission 256

the law of property 259 The possessory interdicts 259

16 Private law from the XII Tables to the fall of the republic:

Origin of the possessory interdicts 261

ownership 263 Ownership of provincial land 267 The actio Publiciana and the development of ‘ bonitary ’

Ownership by peregrini 268 Tura in re aliena 268

Delict 272 Contract 279

17 Private law from the XII Tables to the fall of the republic:

the law of obligations 271

18 Criminal law in the republic 305

From the XII Tables until the legislation of Sulla 305

The quaestiones perpetuae after Sulla 318

19 The constitution under the principate 321

The beginnings of the principate 8321 The development of the principate 324

under the principate 345 The extension of the citizenship 345

20 Classes of the population and the government of the cities

| vii

Contents

20 Classes of the population and the government of the cities

Local government 347 Classes of the population 350 under the principate — continued

21 Sources of law in the principate 3538

The old sources 353 The new sources 3638

22 Legal science during the principate : 374

The work of the jurists 374 The two schools 378

The chief individual jurists 380

General 395 Criminal jurisdiction 401

28 Jurisdiction and procedure in the principate 395

Innovations in civil procedure 397

Introductory 405 Influences and methods 406

24 General character of the classical law 405

| Decreasing formalism and rigidity 408 Increasing tendency towards abstract methods of thought 412

Increased use of writing 414 25 The constitution under the Dominate 421

The imperial power 421 The administrative machine 4,23 The senates and the remaining republican magistracies 431

Classes of the population 433

26 Procedure and jurisdiction in the Dominate 439

Civil procedure 439 The courts and jurisdiction 445 Vili

Contents

Justinian 451 Legal science 451

27 Sources from the end of the classical period to the accession of

Imperial legislation 460 The barbarian codes 466

28 The character of the law from the end of the classical period

to the accession of Justinian 469

‘Vulgar law ’ 473

The reception of Greek institutions , 469

29 The legislation of Justinian 478

The course of legislation 479

Legal education 498

The history of Justinian’s legislation in the East 500

Index 517

80 General character of Justinian’s law 505

1x

_ BLANK PAGE

Preface The purpose of this book remains what it was forty years ago, when Jolowicz wrote in the preface to the first edition: ‘This book is intended, in a sense, to provide a version of Hamlet with-

out the Prince of Denmark. The central subject of Roman legal studies must remain the private law of classical and later times, but an historical background has always been necessary, and is even more so now that the new problems discussed by modern authors are mainly of an historical rather than a dogmatic nature. My object has therefore been to... give the student what he needs to know about Roman law rather than the Roman law itself. The main emphasis is on the sources of law and the constitutional developments without which the nature of the sources cannot be

understood, but I have tried in addition to explain procedure and to give an idea of the judicial structure. It seemed also that for the republican period it was possible to give some account of the state of our knowledge without explaining or presupposing the institutions of the developed law in any detail, and thus to portray the foundations on which the later edifice was built. No attempt has been made to avoid controversial topics, because in these studies controversy is the breath of life, and there is often more to

be learnt from an appreciation of opposing theories than from an acknow- , ledged fact.’

But the study of Roman law and Roman constitutional history have not stood still, and, since Jolowicz was able to make only minor alterations in the second edition twenty years ago, a thoroughgoing revision seemed now to be essential. The character of the book is, I hope, unchanged, but some sections are entirely new, many have been largely rewritten, and there are few pages which have remained altogether unaltered. My chief endeavour has been to maintain the balance which I have always regarded as a great virtue of the book, and which was manifested not only in its assessment of controversial

questions but also in the relationship between text and footnotes. In introducing the second edition Jolowicz expressed the hope that it was § still pos-

sible for a reader to understand the text although he has no previous acquaintance with the subject, while the scholar is told of the direction in which he might profitably dig ’. My difficulty has been to take account of the mass of literature which has appeared in the last forty years without destroying Jolowicz’s careful balance or swamping the notes with a flood of biblio-

graphy. For, paradoxical as it may seem, at a time when Roman law at an xi

Preface

elementary level has been almost everywhere in retreat, there has been a great increase in the interest devoted to it by the learned. A simple measure of this is provided by the fact that whereas in 1932 there were only four periodicals wholly or mainly given over to Roman law, there are now eight or nine. I have therefore had to be severely selective in citation and have endeavoured to give an account only of what seemed to me the most significant developments, providing bare sign-posts to the rest. As far as the sub-

stantive law and the law of procedure are concerned I have done this the more readily because of the existence of Kaser’s Privatrecht and Zivilprozess-

recht, which provide a very thorough bibliography. With very few exceptions I have been able to take account only of literature which reached me before the end of 1970.

I am indebted to my colleague Mr. D. L. Stockton for criticising the draft

of some of the more purely historical chapters, and to Professor J. A. C. Thomas for reading the proof of the chapter on the criminal law. And I must express my appreciation of the patience and care with which the staff of the Cambridge University Press and the printer have dealt with a very tangled manuscript. Barry Nicholas. Oxford March 1972

Gi

Dates MONARCHY B.C. 753 Foundation of Rome 510 Expulsion of the Tarquins

REPUBLIC 451-450 Compilation of Twelve Tables

445 Lex Canuleia

367 Leges Liciniae Sextiae Admission of plebeians to consulate

Institution of praetorship

338 Dissolution of Latin league

304 Ius Flavianum 300 Lea Ogulnia

Admission of plebeians to pontificate

287 Lea Hortensia

Final equation of plebiscita with leges

252 Ti. Coruncanius first plebeian pontifex maximus C. 242 Institution of peregrine praetorship

241 First province (Sicily)

123-122 Tribunate of C. Gracchus

82-79 Dictatorship of Sulla

44 Assassination of Caesar

31 Battle of Actium

PRINCIPATE 27 Augustus regularises his power

A.D. 14 Death of Augustus

Earlier classical period of law

117-188 Reign of Hadrian 188-161. Reign of Antonius Pius

P 161 Institutes of Gaius

161-172 Reign of M. Aurelius and L. Verus (Divi Fratres) 172-180 Reign of M. Aurelius alone xill

Dates Later classical period of law

A.D. 198-211 Reign of Septimius Severus 211-217 Reign of Caracalla (Antoninus)

P 212 Constitutio Antoniniana

222-235 Reign of Alexander Severus

DOMINATE 284-305 Reign of Diocletian

812 Conversion of Constantine the Great 895 Death of Theodosius the Great Division of the empire

438 Codex Theodosianus 476 End of western empire 527-565 Reign of Justinian

xiv

Abbreviations” Abbott and Johnson, Municipal Administration: F. F. Abbott and A. C. Johnson, Municipal Administration in the Roman Empire (Princeton, 1926).

ACII: Acta Congressus Iuridici Internationalis, 1934 (2 vols., Rome, 1985).

AG: Archivio Giuridico (1868— ). AJA: American Journal of Archaeology (1897- ). AJPhil: American Journal of Philology (1880— _). Albertario, Introd.: E. Albertario, Introduzione storica allo studio del diritto romano giustinianeo 1 (Milan, 1935).

Albertario, Studi: E. Albertario, Studi di diritto romano (6 vols., Milan, 1983, 1941, 1936, 1946, 1937, 1958).

Ann. stor. dir.: Annali di storia del diritto (1957— ). APal.: Annali del Seminario giuridico dell? Universita di Palermo (1913—_ ). Arangio-Ruiz, Compravendita: V. Arangio-Ruiz, La compravendita in dit-

ritto romano (2nd ed., Naples, 1954). Arangio-Ruiz, [stit.: V. Arangio-Ruiz, Istituzioni di diritto romano (14th ed., Naples, 1960). Arangio-Ruiz, Rariora: V. Arangio-Ruiz, Rariora (Rome, 1946). Arangio-Ruiz, Storia: V. Arangio-Ruiz, Storia del diritto romano (7th ed., Naples, 1957).

Athenaeum: Athenaeum — Studi periodici di letteratura e storia (1913-). Atti Bologna: Atti del Congresso internazionale di diritto romano, Bologna, 1933 (Pavia, 1934).

Atti Roma: Atti del Congresso internazionale di diritto romano, Roma, 1933 (Pavia, 1934). Atti Verono: Atti del Congresso internazionale di diritto romano e di storia del diritto, Verona, 1948 (Milan, 1958).

B.: Basilica. Beseler: G. Beseler, Beitrige zur Kritik der rémischen Rechtsquellen 1-4 (Tibingen, 1910-20).

Bethmann-Hollweg: M. A. von Bethmann-Hollweg, Der rémische Civilprozess 1-8 (Bonn, 1864-66; repr. Aalen, 1959). * For details of places of publication etc. of periodicals see L. Caes, Collectio bibliographica operwm ad ius romanum pertinentium, Series 1, Suppl. I (Brussels, 1969). For general bibliography see Kunkel, Intro. 179ff.; Buckland xxiiff. XV

Abbreviations

BGU: Berliner Griechische Urkunden (Agyptische Urkunden aus den Museen zu Berlin, Griech. Urk.) (Berlin, 1895- ). BIDR: Bullettino dell’Istituto di diritto romano (1888—_). Biondi, Dir. rom. crist.: B. Biondi, Il diritto romano cristiano 1-8 (Milan, 1952-4),

Bleicken, Senatsgericht: J. Bleicken, Senatsgericht und Kaisergericht. Eine Studie zur Entwicklung des Prozessrechtes im frihen Prinzipat (G6éttingen, 1962). Bonfante, Corso: P. Bonfante, Corso di diritto romano 1, 2 (in two parts), 83, 6 (Rome, 1925, 1926/28, 19338, 1980 ; repr. Milan, 1963- ).

Bonfante, Scrittt: P. Bonfante, Scritti giuridict 1-8 (Turin, 1926, 1918, 1921) 4 (Rome, 1926).

Bonfante, Storia: P. Bonfante, Storia del diritto romano (4th ed., 2 vols., Turin /Rome, 1934; repr. with revised bibliography, Milan, 1958/9). Boyé, Denuntiatio: A.—J. Boyé, La denuntiatio introductive d’instance sous le principat (Bordeaux, 1922). Broggini, Coniectanea: G. Broggini, Coniectanea. Studi di diritto romano (Milan, 1966).

Broggini, Judex: G. Broggini, Iudex arbiterve (Cologne / Graz, 1957). Bruns: Fontes Iuris Romani Antiqui, ed. C. G. Bruns (7th ed., 2 vols. in 1, Tiibingen, 1909; repr. Aalen, 1958). Buckland: W. W. Buckland, A Teaxt-book of Roman Law from Augustus to Justinian (8rd ed. by P. Stein, Cambridge, 1968). Buckland, Main Institutions: W. W. Buckland, Main Institutions of Roman Private Law (Cambridge, 1931).

Buckland: Manual: W. W. Buckland, A Manual of Roman Private Law (2nd ed., Cambridge, 1939).

Buckland, Slavery: W. W. Buckland, The Roman Law of Slavery (Cambridge, 1908).

C.: Codex of Justinian. CAH: The Cambridge Ancient History, 1929-39. C. Cordi, C. Deo auctore, C. Haec, C. Imp. mai., C. Omnem, C. Summa, C.

Tanta, C. Aédwxev: Introductory constitutions to parts of Justinian’s compilation; see below, 479, n. 3. CIEL: Corpus Inscriptionum Latinarum (Berlin, 1863-—_).

Class. Quart.: The Classical Quarterly (1907- ). Class. Rev.: Classical Review (1887— ). CLJ: The Cambridge Law Journal (1921—_ ).

Coll.: Collatio legum Mosaicarum et Romanarum (FIRA 2.548ff.; and see p. 456, below).

xvi

Abbreviations

Collinet, Ecole de Beyrouth: P. Collinet, Histoire de l’Ecole de droit de Beyrouth (Etudes historiques sur le droit de Justinien, 2, Paris, 1925). Collinet, Le caractére oriental: P. Collinet, Le caractére oriental de lVoeuvre législative de Justinien et les destinées des institutions classiques en occident (Etudes historiques, 1, Paris, 1912).

Collinet, La procédure par libelle: P. Collinet, La procédure par libelle (Etudes historiques, 4, Paris, 1982).

Const. Sirm.: Constitutiones Sirmondianae (in vol. 1 of Mommsen-Meyer edition of CTh.; see below, 465). Consult: Consultatio veteris cuiusdam iurisconsulti (FIRA 2.598ff.; and see below, 458).

Corbett, Marriage: P. E. Corbett, The Roman Law of Marriage (Oxford, 1980).

Costa, Cicerone giureconsulto: E. Costa, Cicerone giureconsulto (2 vols., Bologna, 1927).

Crook, Consilium Principis: J. A. Crook, Consiliwm Principis (Cambridge, 1955).

Crook, Law and Life: J. A. Crook, Law and Life of Rome (London, 1967). CTh.: Codex Theodosianus.

D.: Digest of Justinian. Daube, Biblical Law: D. Daube, Studies in Biblical Law (Cambridge, 1947). Daube, Forms: D. Daube, Forms of Roman Legislation (Oxford, 1956). David-Nelson: Gai Institutionum Commentarit IV mit philologischem Kom-

mentar herausgegeben von M. David u. H. L. W. Nelson (=Studia Gaiana, vol. 8, Kommentar; Leiden, 1954—_).

Entretiens Hardt 18: Fondation Hardt pour |’étude de l’antiquité classique. Entretiens sur antiquité classique, vol. 18 (Geneva, 1967). Epit. Ulp.: Epitome Ulpiani (FIRA 2.261ff.; and see below, 458). Eranion Maridakis: Eranion in honorem G. S. Maridakis (8 vols. Athens, 1968 / 4).

Essays in Legal History, ed. Vinogradoff: Essays in Legal History read before The International Congress of Historical Studies held in London in 1913, ed. Sir P. Vinogradoff (Oxford, 1918).

Ferrini, Opere: C. Ferrini, Opere, ed. V. Arangio-Ruiz (5 vols., Milan, 1929 / 80).

FIRA: Fontes Iuris Romani Anteiustiniani, ed. S. Riccobono, J. Baviera, C. Ferrini, J. Furlani, V. Arangio-Ruiz (8 vols., 2nd ed., Naples, 1941, 1940, 1943).

xvii

Abbreviations de Francisci, Primordia: P. de Francisci, Primordia Civitatis (Rome, 1959). Fschr. Koschaker: Festschrift P. Koschaker (8 vols., Weimar, 1939). Fschr. Rabel: Festschrift fiir E. Rabel (2-vols., Tiibingen, 1954). Fschr. Schulz: Festschrift F'. Schulz (2 vols., Weimar, 1951). Fschr. Siber: Festschrift fir H. Siber (2 vols., Leipzig, 1941/8).

Fschr. Wenger: Festschrift fiir L. Wenger (=Minch Beitr. 84, 1944; 85, 1945).

FV: Fragmenta Vaticana (FIRA 2.468ff.; and see below, 456).

Gai.: Institutes of Gaius. Gai. Aut.: The Autun Gaius (FIRA 2.207ff.; and see below, 458). Gai. Epit.: The Epitome of Gaius (FIRA 2.231ff.; and see below, 466). Gai. Sympos.: Gaio nel suo tempo. Atti del symposio romanistico (Naples, 1966).

Garnsey: Social Status: P. Garnsey, Social Status and Legal Privilege in the Roman Empire (Oxford, 1970).

Gaudemet, Formation: J. Gaudemet, La formation du droit séculier et du droit de l’église aux IVe et Ve siécles (Paris, 1957).

Gaudemet, Institutions: J. Gaudemet, Institutions de. l’antiquité (Paris, 1967).

Girard: P. F. Girard, Manuel élémentaire de droit romain (8th ed. by F. Senn, Paris, 1929).

Girard, Mélanges: P. F. Girard, Mélanges de droit romain (2 vols., Paris, 1912, 1928).

Gradenwitz, Interpolationen: O. Gradenwitz, Interpolationen in den Pandekten (Berlin, 1887). Greenidge: A. H. J. Greenidge, Roman Public Life (Oxford, 1901).

Guarino, Ordinamento: A. Guarino, L’ordinamento giuridico romano (8rd ed., Naples, 1959).

H.A.: Historia Augusta. Hammond: Antonine Monarchy: M. Hammond, The Antonine Monarchy (American Acad. in Rome, Papers 19, 1959).

Heimbach: Basilicorum libri LX, ed. G. Heimbach (7 vols., Leipzig, 188397).

Hill, Roman Middle Class: H. Hill, The Roman Middle Class in the Republican Period (Oxford, 1952). Historia: Historia. Zeitschrift fiir alte Geschichte (1950—_ ). Holdsworth, HEL: W. 8S. Holdsworth, A History of English Law (16 vols., London, 1923-66). Honoré, Gaius: A. M. Honoré, Gaius. A Biography (Oxford, 1962). xviii

Abbreviations

H.-S. or Heumann-Seckel: Heumanns Handlexikon zu den Quellen des rémischen Rechts (9th ed. by E. Seckel, Jena, 1907; repr. as 10th ed., Graz, 1958).

IRMAE: Ius Romanum Medii Aevi, auspice Collegio antiqui iuris studiis provehendis (Milan, 1961— ). Iura: Iura. Rivista internazionale di diritto romano e antico (1950—_ ).

J.: Institutes of Justinian. JHA: Journal of Egyptian Archaeology (1914— ). Jhering, Geist: R. von Jhering, Geist des rémischen Rechts auf den verschie-

denen Stufen seiner Entwicklung (8 parts in 4 vols., 4th and 5th ed., 1883-94; repr. Aalen, 1968). JJP: Journal of Juristic Papyrology (1946—_ ).

Jolowicz, De Furtis: Digest XLVII.2, De Furtis, ed. with Introduction, Translation and Notes by H. F. Jolowicz (Cambridge, 1940).

Jolowicz, Roman Foundations: H. F. Jolowicz, Roman Foundations of Modern Law (Oxford, 1957). Jones: A. H. M. Jones, The Later Roman Empire (8 vols., Oxford, 1964).

Jones, Studies: A. H. M. Jones, Studies in Roman Government and Law (Oxford, 1960).

Jérs—Kunkel: W. Kunkel, Rémisches Privatrecht auf Grund des Werkes von P. Jérs (8rd ed., Berlin ete., 1949=2nd ed., 1935, with bibliographical supplement). JRS: Journal of Roman Studies (1911-_ ). Jurid. Rev.: Juridical Review (1889-_ ).

Karlowa: O. Karlowa, Rémische Rechisgeschichte (2 vols., Leipzig, 1885, 1901).

Kaser, AJ: M. Kaser, Das altrémische ius (Gottingen, 1949). Kaser, EB: M. Kaser, Eigentum und Besitz im dlteren rémischen Recht (2nd ed., Cologne / Graz, 1956 = Ist ed., 1942, with appendix).

Kaser, RPL: M. Kaser, Roman Private Law (2nd ed., transl. R. Dannenbring, London, 1968).

Kaser, RPR: M. Kaser, Das rémische Privatrecht 1 (2nd ed., Munich, 1971), 2 (1st ed., Munich, 1955)= Handbuch der Altertumswissenschaft, ed. H. Bengtson, vol. x.3.3.1, 2).

Kaser, ZPR: M. Kaser, Das rémische Zivilprozessrecht (Munich, 1966= Handbuch der Altertumswissenschajft, ed. H. Bengston, vol. x.3.4). Kelly, Roman Litigation: J. M. Kelly, Roman Litigation (Oxford, 1966).

xix

Abbreviations

Kipp: Th. Kipp, Geschichte der Quellen des rémischen Rechts (4th ed., Leipzig / Erlangen, 1919).

Kriiger: P. Kriiger, Geschichte der Quellen und Literatur des rémischen Rechts (2nd ed., Munich / Leipzig, 1912).

Kubler: B. Kibler, Geschichte des rémischen Rechts (Leipzig / Erlangen, 1925).

Kunkel, Herkunft: W. Kunkel, Herkunft und soziale Stellung der rémischen

Juristen (2nd ed., Vienna, 1967=I1st ed., 1952, with bibliographical supplement). Kunkel, Intro.: W. Kunkel, An Introduction to Roman Legal and Constitutional History, transl. J. M. Kelly (Oxford, 1966). Kunkel, Krim.: W. Kunkel, Untersuchungen zur Entwicklung des rémischen

Kriminalverfahrens in vorsullanischer Zeit (Munich, 1952=Abh. d. Bayer. Akad. d. Wissensch., phil.-hist. Kl., 56). Labeo: Labeo. Rassegna di diritto romano (1955—__).

Lawson, Negligence: F. H. Lawson, Negligence in the Civil Law (Oxford, 1950).

Lenel, Gesch.: O. Lenel, Geschichte und Quellen des rémischen Rechts (Munich/Leipzig, 1915, Holtzendorffs Enzyklopddie der Rechtswissenschaft, vol. 1, 7th ed.).

Lenel, EP: O. Lenel, Das Edictum Perpetuum (8rd ed., Leipzig, 1927; repr. Aalen, 1956). Levy, Ehescheidung: EH. Levy, Der Hergang der rémischen Ehescheidung (Weimar, 1925). Levy, Ges. Schr.: E. Levy, Gesammelte Schriftén (2 vols., Cologne / Graz, 1968).

Levy, Obligations: E. Levy, Westrémisches Vulgarrecht, Das Obligationenrecht (Weimar, 1956). Levy, Privaistrafe: E. Levy, Privatstrafe und Schadensersatz (Berlin, 1915). Levy, Property: E. Levy, West Roman Vulgar Law, The Law of Property (Philadelphia, 1951). Levy, Sponsio: E. Levy, Sponsio, fidepromissio, fidetussio (Berlin, 1907). Lévy-Bruhl, Nouv. Et.: H. Lévy-Bruhl, Nouvelles études sur le trés ancien droit romain (Paris, 1947). Lévy-Bruhl, QP: H. Lévy-Bruhl, Quelques problémes du trés ancien droit romain (Paris, 1984). Lévy-Bruhl, Recherches: H. Lévy-Bruhl, Recherches sur les actions de la loi (Paris, 1960).

Lipsius: J. H. Lipsius, Das attische Recht und Rechtsverfahren (8 vols., Leipzig, 1905, 1908, 1915). LQE: Law Quarterly Review (1885—_ ).

xx

Abbreviations L.R.Burg.: Lex Romana Burgundionum (see below, 467). L.R.Visig.: Lex Romana Visigothorum (see below, 466). Magdelain, Actions Civiles: A. Magdelain, Les actions civiles (Paris, 1954). Magdelain, Recherches sur Vimpertum: A. Magdelain, Recherches sur l’imperium, la lot curtate, et les auspices d’investiture (Paris, 1968). Maine, Ancient Law: Sir Henry Sumner Maine, Ancient Law (ed. Pollock, London, 1906).

de Martino: F. de Martino, Storia della costituzione romana 1-5 (Naples, 1951, 1954/5, 1958, 1962/5, 1967). v. Mayr: R. von Mayr, Rémische Rechtsgeschichte (7 vols., Leipzig, Berlin, 1912/18).

Mél. Cornil: Mélanges de droit romain dédiés a G. Cornil (2 vols., Ghent / Paris, 1926). Mél. Gérardin: Mélanges Gérardin (Paris, 1907). Mél. Lévy-Bruhl: Droits de Vantiquité et sociologie juridique. Mélanges

Henri Lévy-Bruhl (Paris, 1959). |

Mél. Meylan: Mélanges Philippe Meylan (2 vols., Lausanne, 19638). Mél. de Visscher: Mélanges F. de Visscher 1-4= RIDA 2-5 (1949-50).

Meyer, Rém. Staat: E. M. Meyer, Rémischer Staat und Staatsgedanke (8rd ed., Zurich / Stuttgart, 1964). Mitteis, Chrest.: L. Mitteis, Grundztige und Chrestomathie der Papyruskunde, 2.2 Chrestomathie (Leipzig, 1912). Mitteis, Grundztige: Ibid. 2.1. Mitteis, Reichsr.: L. Mitteis, Reichsrecht und Volksrecht in den Gstlichen

Provinzen des rémischen Kaiserreichs (Leipzig, 1891). ,

Mitteis, RPR: L. Mitteis, Rdmisches Privatrecht bis auf die Zeit Diokletians 1 (Leipzig, 1908).

Momigliano, Quarto Contributo: A. Momigliano, Quarto contributo alla storia degli studi classici e del mondo antico (Rome, 1969). Momigliano, Terzo Contributo: A. Momigliano, Terzo contributo etc. (Rome, 1966). Mommsen, Abriss: Th. Mommsen, Abriss des rémischen Staatsrechts (2nd ed., Leipzig, 1907).

Mommsen, Ges. Schr.: Th. Mommsen, Gesammelte Schriften (8 vols., Berlin, 1905-18; repr. Berlin/Dublin, 1965; Juristische Schriften= vols. 1-38).

Mommsen, Rém. Chronologie: Th. Mommsen, Die rém. Chronologie bis auf Caesar (Berlin, 1858). Mommsen, StR: Th. Mommsen, Rémisches Staatsrecht 1, 2 (8rd ed., Leipzig, 1887), 8 (1st ed., Leipzig, 1888); repr. Tibingen, 1952. xxi

Abbreviations Mommsen, StrR: Th. Mommsen, Rémisches Strafrecht (Leipzig, 1899; repr. Darmstadt, 1955). Monier: R. Monier, Manuel élémentaire de droit romain 1 (6th ed., Paris, 1947), 2 (Sth ed., Paris, 1954); repr. Aalen, 1970.

Miinch. Beitr.: Miinchener Beitrdge zur Papyrusforschung und antiken Rechtsgeschichte (1915- ).

Nicholas, Introduction: B. Nicholas, Introduction to Roman Law (Oxford, 1962).

1966). , }

Nicolet, L’ordre équestre: Cl. Nicolet, L’ordre équestre 4 Vépoque républicaine 1 (Bibl. des écoles frangaises d’Athénes et de Rome, fasc. 207, NNDI: Novissimo digesto italiano (Turin, 1957—_).

Noailles, Fas et Ius: P. Noailles, Fas et tus. Etudes de droit romain (Paris, 1948).

Not. Dign. Occid.: Notitia dignitatum in partibus occidentis, ed. O. Seeck

(Berlin, 1876).

Nov.: Novels of Justinian. Nov. Maioran.: Novellae Maiorani (in vol. 2 of the Mommsen—Meyer edition of CTh; see below, 465, n. 5). Nov. Marcian.: Novellae Marciani (ibid.). Nov. Th.: Novellae Theodosii (ibid.). Nov. Val.: Novellae Valentiniani (ibid.).

Ogilvie, Livy 1-5: R. M. Ogilvie, 4 Commentary on Livy Books 1-5 (Oxford, 1965).

Pais, Ricerche: E. Pais, Ricerche sulla storia e sul diritto pubblico di Roma 1 (Rome, 1915). P. Antinoop.: The Antinoopolis Papyri (London, 1950—_).

Partsch, Schriftformel: J. Partsch, Die Schriftformel im rémischen Provinzialprozesse (Breslau, 1905).

Pernice, Labeo: A. Pernice, Labeo, Rémisches Privatrecht im ersten Jahrhundert der Kaiserzeit (1, 1878; 2.1, 2, 2nd ed., 1895/1900; 8.1, 2nd ed., 1892; all Halle; repr. Aalen, 1968). Perozzi, Scritti: S. Perozzi, Scritti giuridici (8 vols., Milan, 1948). Peters, Die ostrémischen Digestenkommentare: H. Peters, Die ostrémischen

Digestenkommentare und die Entstehung der Digesten (Berichte d. ! sdchs. Gesellsch. d. Wissensch. 65, 1918; repr. Labeo. 16 (1970) 188ff., 3385ff.).

P. Giess.: Griechische Papyri im Museum des oberhessischen Geschichtsvereins zu Giessen (Leipzig, etc., 1910-12).

xxii

Abbreviations , P. Hal.: Dikaiomata: Ausztige aus alexandrinischen Gesetzen ... herausgeg. von der Graeca Halensis (Berlin, 1918). P. Lips.: Griechische Urkunden der Papyrussammlung zu Leipzig, ed. L.

Mitteis (Leipzig, 1906). . Pollock and Maitland: F. Pollock and F. W. Maitland, The History of English law before the time of Edward I (2 vols., 2nd ed., Cambridge,

1898, 1911; repr. with introduction and bibliography by S. F. C. Milsom, 1968).

P. Oxy.: The Oxyrhynchus Papyri, ed. B. P. Grenfell and A. S. Hunt e¢ al.

(London, 1898— ). Pringsheim, Ges. Abh.: F. Pringsheim, Gesammelte Abhandlungen (2 vols., Heidelberg, 1961).

Pringsheim, Greek Law of Sale: F. Pringsheim, The Greek Law of Sale (Weimar, 1950). PS: Pauli Sententiae (FIRA 2.819ff.; and see below, 457).

Pugliese, Proc.: G. Pugliese, Il processo civile romano, vol. 1: Le legis actiones (Rome, 1961/2); vol. 2: Il processo formulare 1 (Milan, 1968). PW: Paulys Realenzyklopddie der klassischen Altertumswissenschaft, neue

Bearbeitung, ed. G. Wissowa, etc. (Stuttgart, 1894— ; in two series, the second, beginning at R, being distinguished by the addition of A after the volume number; Suppl. =supplementary volume).

Rabel: E. Rabel, Grundziige des rémischen Privatrechts, in Holtzendorfjs Enzyklopddie der Rechtswissenschaft 1 (7th ed., 1915) 899ff. (separate reprint, Darmstadt, 1955). Rend. Ist. Lomb.: (Reale) Istituto lombardo di scienze e lettere. Rendiconti. Classe di lettere e scienze morali e storiche (1864—_).

RHD: Revue historique de droit francais et étranger (4th series, 1922/ );

from 1877 to 1921: Nouvelle revue historique de droit francais et étranger.

Riccobono, Scritti: S. Riccobono, Scritti di diritto romano 1, 2 (Palermo, 1957, 1964).

RIDA: Revue internationale des droits de l’antiquité (1948—_ ). Riv. it. sci. giur.: Rivista italiana per le scienze giuridiche (1886—_). Rostovtzeff, Economic History: M. Rostovtzeff, The Social and Economic History of the Roman Empire (2nd ed. by P. M. Fraser, 2 vols., Oxford, 1957).

Rotondi, Leges publicae: G. Rotondi, Leges publicae populi romani (Milan, 1912; repr. Hildesheim, 1962). Rotondi, Scr. giur.: G. Rotondi, Scritti giuridict (8 vols., Milan, 1922). XXII

Abbreviations

1963). :

Schmidlin, Rekuperatorenverfahren: B. Schmidlin, Das Rekuperatorenver-

fahren. Eine Studie zum rémischen Prozess (Freiburg, Switzerland, Sch. Sin.: Scholia Sinaitica (FIRA 2.637ff.; and see below, 459). Schulz, CRL: F. Schulz, Classical Roman Law (Oxford, 1951). Schulz, History: F. Schulz, History of Roman Legal Science (Oxford, 1946). Schulz, Principles: F. Schulz, Principles of Roman Law (Oxford, 1986). Schwind, Publikation: F. v. Schwind, Zur Frage der Publikation wm rémischen Recht (Minch. Beitr. 81, 1940).

Scr. Ferrini (Milano): Scritti in onore di Contardo Ferrini. pubblicati in occasione della sua beatificazione (4 vols., Milan, 1947-9).

Scr. Ferrint (Pavia): Scritti di diritto romano in onore di Contardo Ferrini pubblicati della R. Universita di Pavia (Milan, 1946). Scr. Giuffré: Scrittt in memoria di A. Giuffré (Milan, 1967).

Scullard, History: H. H. Seullard, A History of the Roman World 753-146 B.C. (8rd ed., London, 1961).

SDHI: Studia et documenta historiae et iuris (1985—_). | Sherwin-White: A. N. Sherwin-White, The Roman Citizenship (Oxford, 1989).

Siber: H. Siber, Rémisches Recht, vol. 2: Rémisches Privatrecht (Berlin, 1928).

Sohm: R. Sohm, Institutionen, Geschichte u. System des rém. Privatrechis (17th ed. by L. Mitteis and L. Wenger, Munich, 1928). Solazzi, Scritti: S. Solazzi, Scritti di diritto romano 1-4 (Naples, 1955, 1957, 1960, 1968).

St. Albertario: Studi in memoria di E. Albertario (2 vols., Milan, 1958). St. Arangio-Ruiz: Studi in onore di V. Arangio-Ruiz (4 vols., Naples, 1958). St. Betti: Studi in onore di E. Betti (5 vols., Milan, 1962). St. Biondi: Studi in onore di B. Biondi (4 vols., Milan, 1965). St. Bonfante: Studi in onore di P. Bonfante (4 vols., Milan, 1980). St. de Francisci: Studi in onore di P. de Francisci (4 vols., Milan, 1956). St. Grosso: Studiin onore di G. Grosso (8 vols., Turin, 1968). St. Paoli: Studi in onore di U. E. Paoli (Florence, 1955). St. Redenti: Studi in onore di E. Redenti (2 vols., Milan, 1951). St. Riccobono: Studi in onore di S. Riccobono (4 vols., Palermo, 1936). St. Solazzi: Studi in onore diS. Solazzi (Naples, 1948). St. Volterra: Studi in onore di E. Volterra (in course of publication, Milan). Stein, Histoire: Ei. Stein, Histoire du Bas-Empire (2 vols. in 8, Paris, etc., 1959, 1949).

Stein, Regulae: P. Stein, Regulae Iuris. From Juristic Rules to Legal

| XXIV

Maxims (Edinburgh, 1966).

Abbreviations Steinacker, Die antiken Grundlagen: H. Steinacker, Die antiken Grundlagen der frihmittelalterlichen Privaturkunde (Leipzig, 1927). Steinwenter, Versdumnisverfahren: A. Steinwenter, Studien zum rém. Versdumnisverfahren (Munich, 1914).

Strachan-Davidson: J. L. Strachan-Davidson, Problems of the Roman Criminal Law (2 vols., Oxford, 1912; repr. Amsterdam, 1969). Symb. David: Symbolae iuridicae et historicae M. David dedicatae (2 vols., Leiden, 1968).

Symb. Frib.: Symbolae Friburgenses in honorem Ottonis Lenel (Leipzig, 1981).

Symb. Taubenschlag: Symbolae R. Taubenschlag dedicatae, 3 vols. = Kos 48 (1956).

Synteleia Arangio-Ruiz: Synteleta V. Arangio-Ruiz (2 vols. in 1, vol. 2 beginning at p. 607, Naples, 1964).

SZ: Zeitschrift der Savigny-Stiftung fiir Rechtsgeschichte. Romanistische Abteilung (1880— ).

Tab.: XII Tables (FIRA 1.23ff.). TAPA: Transactions of the American Philological Association (1869-—_ ).

Taubenschlag, Law: R. Taubenschlag, The Law of Greco-Roman Egypt in the light of the Papyri, 332 B.C.—A.D. 640 (2nd ed., Warsaw, 1955). Taylor, Roman Voting Assemblies: L. R. Taylor, Roman Voting Assemblies (Ann Arbor, 1966).

Tenney-Frank: Tenney-Frank, An Economic History of Rome to the end of the Republic (2nd ed., London, 1927). TR: Tijdschrift voor Rechtsgeschiedenis. Revue d’histoire du droit (1918).

Tulane L.R.: Tulane Law Review (1929- ). Varia: Varia. Etudes de droit romain (Institut de droit romain de |’Université de Paris; 4 vols., 1952, 1956, 1958, 1961).

Vinogradoff, Hist. Jurisp.: P. Vinogradoff, Outlines of Historical Jurisprudence (2 vols., Oxford, 1920, 1922).

de Visscher, Etudes: F. de Visscher, Etudes de droit romain (Paris, 1931).

de Visscher, Les édits d’Auguste: F. de Visscher, Les édits d’Auguste découverts & Cyréne (Louvain, 1940; new ed. with additions, 1965).

de Visscher, Nouv. Et.: F. de Visscher, Nouvelles études de droit romain

(Milan, 1949). |

Voci, Modi di acquisto: P. Voci, Modi di acquisto della proprieta (Milan, 1952). XXV

_ Abbreviations Walbank, Polybius: F. W. Walbank, A Historical Commentary on Polybius, vol. 1 (Oxford, 1957). Watson, Obligations: A. Watson, The Law of Obligations in the later Roman Republic (Oxford, 1965).

Watson, Persons: A. Watson, The Law of Persons in the later Roman Republic (Oxford, 1967).

Watson, Property: A. Watson, The Law of Property in the later Roman Republic (Oxford, 1968). Watson, Succession: A. Watson, The Law of Succession in the later Roman

Republic (Oxford, 1971). , Wenger: L. Wenger, Institutes of the Roman Law of Civil Procedure (revised ed., transl. O. H. Fisk, New York, 1940).

Wenger, Quellen: L. Wenger, Die Quellen des rémischen Rechts (Vienna, 1958).

Westrup: C. W. Westrup, Introduction to Early Roman Law (5 vols., London / Copenhagen, 1934, 1939, 1944, 1950, 1954).

Wieacker, Textstufen: F. Wieacker, Textstufen klassischer Juristen (G6ttingen, 1960; Abh. Akad. Wiss. Géttingen, phil.-hist. Kl., 48). Wieacker, Vulgarismus: F. Wieacker, Vulgarismus u. Klassizismus im Recht der Spdtantike (Sitzungsb. Akad. Heidelberg, 1955). Willems: P. Willems, Le droit public romain (7th ed., Louvain, 1910). Willems, Sénat: P. Willems, Le Sénat de la République romaine (Louvain,

etc., vol. 1 with appendix, 2nd ed., 1885, vol 2, 1883; repr. Aalen, 1968).

Windscheid—Kipp: B. Windscheid, Lehrbuch des Pandektenrechts (8 vols.,

9th ed. by Th. Kipp, Frankfurt am Main, 1906; repr. Aalen, 1963). Wlassak, Provinzialprozess: M. Wlassak, Zum rémischen Provinzialprozess (Akad. der Wissenschaften in Wien, phil.-hist. Kl., 190.4, 1919). Wlassak, Prozessformel: M. Wlassak, Die klassische Prozessformel 1 (ibid. 202.8, 1924). Wlassak, Prozessgesetze: M. Wlassak, Rémische Prozessgesetze (2 vols., Leipzig, 1888, 1891).

Wlassak, Studien: M. Wlassak, Studien zum altrémischen Erb- und Vermichtnisrecht 1 (Akad. der Wissenschaften in Wien, phil.-hist. KL., 215.2, 1988).

de Zulueta, Gaius: F. de Zulueta, The Institutes of Gaius, Part IT, Commentary (Oxford, 1958). de Zulueta, Sale: F. de Zulueta, The Roman Law of Sale (Oxford, 1945).

®ee' CHAPTER 1

Periods in the history of Rome and in the history of her law 1 PERIODS IN HISTORY The history of Rome is commonly treated in three great divisions corresponding to the different forms of government which prevailed — the Monarchy, the Republic and the Empire. These main divisions may be subdivided

in different ways. Mommsen,’ whose great history reaches only down to the fall of the republic, divides his work into five periods. Of these, the first stretches from the earliest times down to the abolition of the kingship. The ancient historians ? have a good deal to tell us of this period: that the city was founded in 753 B.C. by Romulus, that the last three of the six kings who followed him were Etruscan, and that it was the last of these Etruscan kings, Tarquinius Superbus, whose tyrannical conduct led to the abolition of the monarchy in 509 B.C. These historians, however, all lived long after the events which they described, in an age when the sifting of evidence was not considered the historian’s chief duty, and much of the picturesque detail is probably therefore no more authentic than the story of King Alfred and the burnt cakes. But some features are demonstrably true. That there

were kings is confirmed by the echoes which survive in the offices of interrex * and rex sacrorum;* that there was a period of Etruscan domination in the sixth century is clear from the archaeological evidence. Much else is disputed, but the historians’ account represents the living tradition of the Roman people and cannot lightly be discarded.° 1 Rémische Geschichte (9th ed. Berlin, 1903) ; English translation by Dickson (3rd ed. London, 1894).

3 Below, 30. :

2 In particular Livy (59 B.C—A.D. 17) and Dionysius of Halicarnassus (c. 25 B.C.). 4 Below, 45.

5 For a valuable survey see Momigliano, Terzo Contributo 545ff. (=JRS 53 (1963) 95ff); ef. ibid. 669ff. (=Riv. storica italiana 76 (1964) 803ff.). The archaeological evidence is collected and discussed by E. Gjerstad, Harly Rome 1— (Lund, 1953— ). For a brief review by the same author see Entretiens Hardt 13.3ff., and cf. Historia 16 (1967) 257ff. See also generally H. H. Seullard, The Etruscan Cities and Rome (London, 1967)

243ff. Although the site of Rome was continuously inhabited from at least the eighth century (and traces have been found from c. 1500 B.C.), Gjerstad identifies the foundation of the city with, inter alia, the laying out of a permanent forum, which in his view marked the unification of the pre-existing villages and which he dates’ c. 575 B.C.; but this dating is not universally accepted, and the making of the forum could in any case 1

Periods in the history of Rome and in the history of her law The next of Mommsen’s periods runs from the abolition of the monarchy until the unification of Italy. When she first appears in history, Rome is a small city-state, but already occupies a pre-eminent position in the confederation of kindred city-states known as the Latin league.* The relationship between Rome and her confederates was not always that of peace, but in the wars against the Latins it was Rome who was finally successful, as she was also, but only after centuries of hard fighting, against the other

nations of Italy. The last great war of this period was one in which the Greek city of Tarentum in South Italy (where there were many Greek colonies) was allied with a non-Italian power in the person of Pyrrhus, king

of Epirus. With the final defeat of Pyrrhus in 275 B.C., the conquest of Italy was virtually complete. Roman colonies’ had been planted at most of the points of strategic importance, and of her late enemies, some were forced to accept direct Roman rule, while others were bound by treaties in which Rome was so much the predominant partner that they were, in fact, under her dominion. They could have no treaties with other foreign states and were bound to supply contingents to serve with the Roman armies.

The internal history of Rome during this period was hardly quieter than her external history, being taken up mainly by the ‘ struggle between the orders ’, i.e. between the patricians and plebeians. The patricians were the nobility or privileged citizens, the plebeians the unprivileged citizens, and

the central purpose of the plebeians was to obtain political and legal equality, a purpose which was only fully achieved in 287 B.C.°®

The third period sees Rome at grips with non-Italian powers against whom she has to fight, not only in Italy but abroad. Chief among these powers were Carthage and Macedonia, and Mommsen ends the period with the decisive victory over the Macedonian forces at Pydna in 168 B.C. The

result of Rome’s victories was that she now acquired territories outside Italy which came to be known as ‘ provinces ’. The first province was Sicily,

which fell to Rome after the first Punic war, in 241 B.C. The later wars with Carthage brought Rome the greater part of Spain and finally ‘ Africa ’, i.e. the Punic possessions in North Africa which were formed into a province after the destruction of Carthage in 146 B.C. Macedonia itself had become

a province in 148 B.C. This was but the logical consequence of Pydna; already at the beginning of the second century B.C. Roman influence both represent merely the beginning of Etruscan domination. The evidence also shows that

trade with Etruria continued until c. 450 B.C., and Gjerstad places the end of the monarchy at about this date; but the expulsion of Tarquin need not have led to an immediate end of trade between Rome and Etruria, and there are other difficulties in the way of this and other alternatives to the traditional account; see below, 8 n 2. 6 Below, 58ff.

7 The Roman, unlike the Greek colony, remained closely bound to the mother city, and served as a military garrison in conquered territory; ef. below, 59. 8 Lex Hortensia, below, 25. 2

Periods in history in the eastern and in the western Mediterranean had been supreme. But ‘captive Greece her captor captive took’, and this is the period in which Rome begins to feel the full impact of Greek culture, for better and for worse. This is the period also in which the economic transformation of Roman society begins. The city-state becomes the capital of an empire, the commercial centre of the world; the yeoman farmer is replaced by the great landowner, relying on slave-labour; the compact citizen body gives way to a society of far greater extremes of wealth and poverty; a proletariat emerges.

The fourth period Mommsen calls ‘ the revolution ’. Its history is that of the breakdown of the old republican form of government, and it is characterised by fierce political strife leading on several occasions to civil war. Mommsen ends it with the temporary triumph of the senatorial nobility under Sulla, dictator 81-79 B.C., who re-established the constitution on an entirely undemocratic basis. The fifth period, the last in Mommsen’s book, stretches from Sulla to 46 B.C., in which year the victory of C. Julius Caesar over the Pompeians at Thapsus left him master of the Roman world. This period Mommsen calls ‘the foundation of the military monarchy ’. It was one of renewed civil conflict, accompanied by expansion abroad, and in the extra-constitutional positions assumed by such leaders as Pompeius and Caesar it already foreshadows the empire. But, though Caesar was, in fact, a military autoerat for the short remainder of his life, and in a sense the founder of the Roman empire, it is not he who is usually regarded as the first emperor. There was no real revival of the republic after his assassination in 44 B.C., but there was a revival of civil war, ending only in 31 B.C. with the battle

of Actium, won by C. Julius Caesar Octavianus, the great-nephew and adopted son of C. Julius Caesar, over his rival M. Antonius. It is this Octavian, better known by the title of ‘ Augustus ’, which he received in 27 B.C., who is generally spoken of as the first Roman emperor.

The long centuries of the empire fall naturally into two periods, which are not, however, marked off from each other by any very definite dividing line. In the earlier period, usually known as the ‘ principate’, though the emperor is in fact supreme, his power is clothed in republican forms; but the disguise becomes ever thinner, and in the third century A.D. vanishes altogether. It is common to refer to the succeeding period of undisguised autocracy as the ‘ dominate’, because the emperor is now no longer even in theory merely the princeps, or ‘ first citizen ’, he is dominus — ‘ master ’ —

of his subjects. If we must choose a specific date for the beginning of the ‘ dominate ’, we can take A.D. 284, in which year, after about half a century of confusion during which emperors followed each other in rapid succession, 3

Periods in the history of Rome and in the history of her law Diocletian ascended the throne.*® This emperor introduced important reforms

amounting to a change in the constitution. From his time until its end, the Roman empire remained an absolute monarchy, but no simple answer can be given to the question, ‘ when did it end?’ Diocletian himself, among his reforms, instituted an administrative division between the western and

eastern halves of the empire. This cleavage was accentuated when the Emperor Constantine transferred his residence to Constantinople, and became final in A.D. 895. But by that time already the end of the western empire was nearing. The invasions of the barbarians could no longer be kept in check, and the last Roman emperor of the West, Romulus Augustulus, was deposed in A.D. 476.

The eastern empire, on the other hand, had a long future still before it, which was only to end with the capture of Constantinople by the Turks in A.D. 1458. Of all the emperors who reigned during its remaining thousand years of life, there is one who is of supreme importance for lawyers, the Emperor Justinian, who came to the throne in A.D. 527 and died in 565, for it was in his time, and largely through his personal interest, that the Roman law assumed the shape in which it has come down to succeeding generations. It is consequently with the year 565 that histories of Roman law usually close, and, apart from a few isolated matters, reference will not be made in this book to later events.

2 PERIODS IN THE HISTORY OF THE LAW The history of a people’s law is as long as that of the people itself, for law of _ @ sort exists in every stage of human society. In the case of Rome, the history of the law is even longer than that of the people, for even after the fall of the

empire, Roman law never quite died out, and from the eleventh century A.D. onwards, through the renewed study of Justinian’s compilations, it was revived as an active force, and deeply influenced the development of the law

throughout Europe. But even if we go no further than the death of Justinian, the stretch of time with which we are concerned is enormous, some thirteen hundred years, counting from the traditional date of the foundation of the city. If it is difficult to divide general history into periods without introducing

a false idea that a people develops by starts rather than continuously, it is still more difficult with legal history, for there are seldom any violent breaks.

In general, changes come gradually, lagging rather behind economic and social developments, and it is only after a long time has elapsed that we can see how great the transformation has been. Moreover, constitutional changes, ® But already in the early second century the ex-consul Pliny the Younger regularly addresses Trajan as dominus. 4,

Periods in the history of the law such as that from republic to empire, may have very little effect on the private law. The periods into which it is convenient to divide the history of the law will therefore not necessarily coincide with those of general history and will usually be even less precisely definable. We may here, however, distinguish five such periods.

(a) The archaic period. This includes the period of the monarchy and the first three or three and a half centuries of the republic. The dominant event is the drawing up of the famous code known as the ‘ Twelve Tables ’, which, according to tradition, took place in the years 451-450 B.C. No part of the Tables themselves has survived, but later authors provide us with a considerable number of what are generally thought to be reliable quotations and paraphrases, and these form the most substantial basis for our knowledge of

the private law of the period. (For the constitutional law of the republic, thanks to the references which the historians necessarily make to it, our information is fairly good.) Otherwise our sources are meagre. We know of some pieces of legislation affecting private law in the period after the XII Tables and we know of the existence of certain legal institutions; moreover, we can sometimes deduce an earlier state of affairs from a close inspection of the law of a later age, and comparison with other systems of primitive law may suggest how gaps in our information should be filled or may explain facts which by themselves would be unintelligible. Beyond this we are left to conjecture. (b) The formative period. The last 150 years of the republic and the first

century of the empire are the great formative period of Roman law. The economic and social developments which followed the end of the second Punic war in 201 B.C. had their counterpart and consequence in the law. Our information for the second century is still tantalisingly slight, but it is clear from later sources that it was a period of great beginnings: the old, stiff law of the XII Tables is giving way to a flexible system more in keeping with the needs of an imperial society. The great jurists who mould this system are clear historical figures, though no fragment of their writings survives. In the last century of the republic we can see more clearly. The jurists of

this period are frequently cited by later writers in Justinian’s Digest,’ and there are a few direct quotations; we have the works of Cicero,” in all of which (not merely in the speeches delivered in court) there are numerous references to legal matters; and altogether we have a good deal of literature

from which, though it is not legal, information on legal matters can be deduced. We have, too, the text of a few laws preserved in inscriptions.

By the end of the republic the main features of the law as we know it appear to be settled and the pace of development slows up. The volume of

1 Below, 480ff. 2 106-43 B.C. 5

Periods in the history of Rome and in the history of her law legal literature, however, increases. For not only are the consequences of 150 years of rapid growth being worked out, but it is probably also true that under the empire the law was an important avenue for men who aspired to a public career, since opportunities for political distinction were necessarily few.

(c) The classical period. By the reign of Hadrian the law had reached a stable maturity.® In the second century and the first half of the third century A.D. a number of great lawyers, who were as a rule both practitioners * and writers, undertook the work of elaboration and consolidation, and these 150 years are accordingly usually regarded as the classical period of the law.* It may be divided into an earlier classical period covering the reigns of Hadrian and the Antonine emperors, and a later classical period under the Severi.° Not that there is a break in the continuity of development, but the work of the earlier age was of a more creative character, while that of the later represents rather the working out of existing principles over the whole field of law. More than nine-tenths of the quotations in the Digest come from authors of the classical age, but those taken from two writers of the later period ’ alone comprise more than half the work.

(d) The post-classical period down to the reign of Justinian. In the era of confusion which succeeded the death of Alexander Severus (A.D. 285) the

line of jurists came abruptly to an end, and even the restoration of order under Diocletian did not revive legal literature. There was no place in the absolute monarchy for the creative work of the individual jurist. Lawyers there were, of course, but anonymously in the imperial chancery, and their work is to be found in the legislation and decisions issued in the emperor’s name. Apart from these, the literature consists merely of a few collections and epitomes of classical works. The quality of legal thought falls sadly away. In the fourth and fifth centuries the practising lawyer, particularly in the provinces, no longer observes the fine distinctions of the classical law. The crudely simplified ‘ vulgar law ’ which results can best be seen in the codes made for their Roman subjects by the Germanic successors of the Roman emperors in the West,® but its influence can be found even in the pronouncements of the imperial chancery. (ce) The reign of Justinian forms a period by itself, because of the work, which may be roughly described as codification, undertaken by that emperor. 8 The consolidation of the praetorian edict by Julian (below, 356ff.) puts an end to the most potent source of radical change. 4 As to the sense in which the jurists can be called practitioners, see below, 94f.

5 Since, as has been said above, the period of most rapid growth seems to end with the republic, the beginning of the classical period is sometimes taken back to that time. 6 For dates, see table, p. xiii.

7 Ulpian and Paul; see below, 392f. 8 Below, 466ff., 473ff. 6

Periods in the history of the law Already in the preceding generation there had been something of a revival in

legal matters, and Justinian made use of it for the purpose of reducing to order the mass of existing authorities. He was also himself a legislator who enacted a number of statutes, some of which were of great importance. But Justinian’s great claim to fame is not his original legislation, nor indeed was the intellectual revival of sufficient strength to enable much original work to be done. The importance of his work lies in the fact that in his ‘ Digest ’ he

gathered together, and thereby preserved, a great mass of excerpts from classical authors, that in his ‘Code’ he made a companion collection of imperial enactments, and that thus he gave to Roman law what was, in a sense, its final form.

7 J.—2

CHAPTER 2

The Republican Constitution 1.ELEMENTS According to tradition the monarchical constitution contained three elements: the king, the council of elders (senatus, connected with senez), and the assembly of the people. Of the distribution of functions between the three we know nothing, nor is it likely that the constitution was at all definite. The king was leader in war and chief priest, and exercised some judicial functions; the senate was his council. Presumably the questions which were reserved for the assembly of the whole people varied a good deal with the character of the king; a weak king would ask the people’s approval for a proposed course of action where a strong one would do without it.

In republican times two of the elements survive, the senate and the assembly, and indeed the third does not suffer as much change as would appear, for the magistrates are the successors to the royal power. When the change from monarchy to republic was made, two magistrates who were eventually called ‘ consuls ’ ! were elected as heads of the state in place of the

king, but the royal power was not cut down. It was limited, indeed, by the fact that there were now two heads instead of one and by the fact that the consuls only held office for a year, whereas the kings had held it for life, but,

apart from this, the consuls had still the great powers which the king had exercised.? In course of time these powers were restricted by statute,* but 1 Originally ‘ praetors’; Livy 3.55.12; Festus 2491; ef. Tab. X11.3.

2 Many scholars have found it difficult to accept the traditional account, Among the arguments adduced against it are (i) that it is improbable that the eventual complex system of two co-equal consuls appeared fully grown from the very beginning; (ii) that for most of the years 448-368 there were in fact more than two supreme magistrates (military tribunes, below, 14), and since a ‘restoration’ after so long an interval is improbable, the dual magistracy can date only from the reforms of 367 (below, 15); (iii) if the dual system does date from 367, the change of name from praetor to consul is easily accounted for (the ancient derivation of praetor was from prae-ire, therefore ‘leader’: Cic. Leg. 3.8, and though the etymology of consul is obscure, the con-, it is argued, suggests joint activity; see further, de Martino 1.192n, 345n; Pugliese, Proc. 1.123n) ; (iv) Livy, 7.3.5, cites a lex vetusta which spoke of qui praetor maximus sit, and this, it is said, is incompatible with a dual magistracy of co-equals. These and other arguments have led to a wide variety of hypotheses as to the course of constitutional development before 367 B.C., all involving originally either a single magistrate, such as

the dictator (below, 55) or at least one magistrate pre-eminent among others (see

Staveley, Historia 5 (1956) 90ff.; de Martino 1.185ff.; de Francisci, Primordia 743ff. ; 3 E.g. the right to put citizens to death was taken away; below, 12, 306ff. 8

The struggle between the orders there remained throughout a large residue undefined by strict law, though generally controlled by the conventions of the constitution. The name of this undefined power, at first given to the consuls alone but afterwards to a few. other magistrates, is imperium.‘ The magistracy was, we are told, confined to the patricians, plebeians being ineligible, and this was one of the points round which the struggle of the orders centred. Of this struggle something must now be said.

2 THE STRUGGLE BETWEEN THE ORDERS What was the origin of the plebeians is a question which has never been satisfactorily settled.5 According to Mommsen they were originally § clients ’, that

is people in a position of dependence on the great families (e.g. former slaves, resident foreigners), but to this it can be objected both that such a body could hardly have maintained an effective struggle, and that our sources make a distinction between plebeians and clients. There have been numerous hypotheses in terms of a difference of race (e.g. that the plebeians were the original Latin inhabitants and the patricians were the Etruscan conquerors),° but the evidence to support them is insubstantial. It is in any case clear that at all times of which we know anything for certain the plebeians were citizens

and not mere resident foreigners. The division was probably simply the Arangio-Ruiz, Storia 28ff., 407ff.). The most substantial argument is the fourth. Unless the accuracy of Livy’s story is discredited (as by Staveley, loc. cit.) praetor maximus has to be explained (by defenders of the traditional account) not as a title but merely as a descriptive reference to the senior of the two consuls (Meyer, Rém. Staat 483; see further, Momigliano, St. Grosso 1.161ff.) or as denoting the supremacy of the office of consul (Magdelain, Zura 20 (1969) 257ff.). On the other hand the traditional account finds its chief independent support in the fasti consulares, which show a succession of two annual consuls in the years before 451. (The fast¢ are lists of the eponymous magistrates — consuls, military tribunes etc.—by reference to whom the Romans counted the

years; although the surviving copies were made only in the late republic and their credibility has been attacked, they are found reliable when they can be checked.) If there were only one supreme magistrate in these years, half the names in the fasti would have to be interpolated; the obvious motive for such interpolation would be the glorification of the later leading families, but the names of such families are in a minority in these years. See Meyer, Musewm Helveticum 9 (1952) 176ff., criticising the thesis of K. Hanell, Das altrém. eponyme Amt (Lund, 1946), that the beginning of the fastt in 509 marks

not the establishment of the republic but the dedication of the temple of Jupiter Capitolinus, and that the earliest part of the fasti was originally a list merely of successive holders of a single eponymous magistracy connected with the new temple. For Hanell the end of a gradual transition to the republic would be marked by the appoint-

ment of the decemviri in 451; cf. the related views of Gjerstad, above, 1 n 5, and the survey by Momigliano cited there. 4 For a survey of the controversy as to its origin see Staveley, Historia 5 (1956) 107ff. For its relation to iurisdictio see below, 47 n 9.

5 See de Martino 1.54 ff. 6 Arangio-Ruiz, Storia 43ff., 412ff, 9

The republican constitution common social and economic one: the greater part of the land presumably fell into the hands of a minority of families, which then arrogated to themselves superior rights in political and religious matters. Moreover it is now

widely held that the division did not harden until the early years of the republic.” For all the kings except Romulus and the two Etruscan Tarquins bear names which in later times are plebeian, and plebeian names appear among the consuls until 487:B.C. Moreover all our sources seem to agree that the prohibition against intermarriage between patricians and plebeians which was contained in the XII Tables was not, as modern writers have universally understood it, a restatement of an immemorial rule, but was on the contrary a harsh innovation.* On this view the struggle between the orders originates in an attempt by an inner group within the aristocracy to isolate itself as a ruling caste, thus uniting against it the lesser aristocracy and the common people.® But in any case, even if the division was not recent, the objectives of the struggle and the course which it took make it plain that there must have been within the plebeian body an aristocracy capable of aspiring to the leadership of the state. The struggle between the plebeians and the patricians was of a twofold character, partly economic and partly political. The chief economic questions were those concerning the public land and the law of debt. (a) The public land. Not all land was in private ownership; ' some belonged

to the state. This was increased from time to time by successful wars. It might be dealt with in any of three ways, either assigned free to private individuals, as their property (ager assignatus), sold to private individuals, or left open for any citizen who wished to occupy it for agricultural purposes ”

not clear. ,

or for use as pasture. The plebeians were evidently under a disability in regard to the enjoyment of public land in any of these ways, but to what extent they were disabled and whether the disability was legal or economic is

(b) The law of debt. An even more pressing evil for the poor than their exclusion from the advantages of the public land was the fearful stringency of the law of debt. The poor man was liable to be taken from his farm to serve in the army and had not, like the rich, slaves who could carry on the 7 Last, JRS 35 (1945) 39ff.; Kunkel, SZ 77 (1960) 369f.; Momigliano, Terzo Contributo 590ff. (=JRS 53 (1963) 117f.); Magdelain in Hommages J. Bayet (Brussels,

1964) 427ff. 8 Last, JERS 35 (1945) 31f.; contra, Arangio-Ruiz, Storia 414. 9 Cf. Momigliano, Quarto Contributo 419f. (=Lntrettens Hardt 13.199f.). 1 On early communal ownership of land see below, 138. 2 Strictly, the occupation of such land gave no right except at the good pleasure of the state, which exacted a rent in kind and could reclaim the land at any moment; but in fact this right to reclaim was seldom exercised and the land, though not strictly owned by the occupant, descended from father to son, while the rent, which had never been large, was often not exacted by the patrician magistrates from the members of their own order. 10

The struggle between the orders work in his absence. He might even, for Rome was not always successful in her wars, return to find that his farm had been ravaged by the enemy. The only thing to do was to apply to some wealthy person for a loan to enable him to start again. Next year the same thing might happen and the load of debt would grow. This would be serious enough in a modern state, but in ancient Rome a man who could not pay his debts could be taken by his creditors and either killed or sold abroad as a slave. Even if the creditors did not proceed to these lengths they could, by threatening to do so, keep the debtor in a condition of abject poverty dependent entirely on their will, and this in fact seems to have been the position of a large number of the plebeians.®

In the lands which they had helped to conquer they had no share, and they ran the risk of losing not only their own land but their life or liberty. The chief political questions were those concerning the magistracy, the validity of resolutions passed by the plebeian assembly and intermarriage between the orders.

(a) The magistracy. The regular heads of the state were the consuls, and from the consulship plebeians were, at least from the second quarter of the fifth century,* excluded. Nor might they be appointed to the dictatorship, which was a temporary revival of the monarchy used in times of emergency when it was necessary to concentrate the whole power of the state in a single person. The only regular magistrates besides the consuls were the quaestors, who were originally simply assistants of the consuls and appointed by them. There was of course no likelihood that the patrician consuls would wish to appoint a plebeian quaestor. The priesthoods too, which were of considerable political importance, were entirely in patrician hands. (b) The validity of resolutions passed by the plebeian assembly. The plebs from very early times met together in an assembly (concilium plebis) and passed resolutions (plebiscita). It was one of their objects to obtain for these resolutions the force of law.

(c) Intermarriage. The XII Tables excluded conubium between the orders;° that is to say, a marriage between a patrician and a plebeian was not recognised as lawful: the children of a patrician father and a plebeian mother would be plebeians, and would not be legally related to their father, with the result, in particular, that they would have no right to his estate on his death. This prohibition of intermarriage was of political as well as social importance, because it emphasised the patrician contention that the plebeians were too basely born to be admitted to the magistracies. The struggles in connexion with the economic and the political questions

3 Below, 164. 4 Above, 10. 5 Above, 10. 11

| The republican constitution are closely interwoven. There were, of course, rich plebeians who did not suffer under the economic grievances of their poorer brethren, but they realised that it was only by making common cause with the rest of their order that they could obtain the political concessions in which they themselves were interested. In the end, the political equality which they sought was completely. achieved, but palliatives, at best, were provided for the poverty of the lower classes, which a wise use of the public land might have effectively remedied.

The details of the struggle belong for the most part to the period of tra-

ditional history, and what is said by the historians is, at least in part, unhistorical. The difficulties of reconstruction are increased by a tendency to read far back into the past reforms which, in fact, took place comparatively

late. Thus we frequently find references to a series of laws apparently all enacting the same thing, and only two explanations seem possible; either the earlier laws were not carried into effect, or they never existed, except in imagination. For instance, the rule that no Roman citizen might be put to death without appeal (provocatio) * to the assembly is referred to no fewer than three leges Valeriae of 509, 449 and 300 B.C. respectively.’ It is, to say the least of it, strange that it should in each case be a member of the same clan who secured the passing of these laws and the probability is that only the last ever really existed.® In spite of these difficulties certain facts emerge. The first step taken by

the plebeians was to band themselves together into a corporation, to hold meetings and to elect officers of their own, these officers being known as tribunt. According to tradition their institution goes back to 494 B.C. in which year is supposed to have taken place the ‘ First Secession of the Plebs ’, that is, the plebs left the city in a body and only returned on being granted certain concessions. These concessions are said to have been the recognition of the plebs as a corporation and the granting of specified powers to the tribunes (ten in number by 449 B.C.).° These officers were given (a) Ius agendi, i.e. the right of convening the assembly of the plebs (concilium plebis) and eliciting resolutions from them. These resolutions, however, were mere self-regarding ordinances and bound no-one outside the corporation. (b) Intercessio,

i.e. the right to veto any magisterial act, including such acts as bringing a bill before the assembly. (c) Auwilium, i.e. the right to protect the plebeians, especially against punishment by the magistrates.’

6 Below, 306ff. 7 To say nothing of the leges Porciae.

8 Laws were called after the name of the magistrate who proposed them; hence a lez Valeria is necessarily one passed by a man with the name ‘ Valerius ’.

® Our sources conflict as to the stages by which the number reached ten; see Ogilvie, Livy 1-5 301ff., 381f.

1 The plebs also swore to uphold the sacrosanctity or inviolability of the tribunes: anyone assaulting them was to be sacer and an outlaw, and his property was to be forfeit 12

The struggle between the orders The tribunes undoubtedly had these powers later; but it is more than unlikely that they possessed such weapons — in particular, the formidable intercessio — from the first.

Strife continued in the succeeding years which saw some plebeian suc-

cesses, such as the lex Icilia (traditional date 456 B.C.), a law passed under tribunician pressure which distributed the land on the Aventine Hill to poor citizens; but the next important movement is that in favour of codification which resulted in the enactment of the XII Tables. The story is that as early as in 462 a tribune, C. Terentilius Arsa, proposed that five men should be elected to draw up a code of law which should bind the consuls in the exercise of their judicial powers. The patricians successfully opposed the project for eight years, but then they were forced to give way. They managed, however, to delay matters by sending ambassadors to Greek cities to study their laws, and specially the code of Solon, the famous Athenian lawgiver.” In 451, after the return of the ambassadors, ten men (instead of the five originally proposed) were elected as chief magistrates instead of consuls (decemviri consulari imperio legibus scribundis), the ordinary constitution, including the appointment of tribunes, being temporarily suspended. Although plebeians were declared eligible for this position, the influence of the aristocracy was so great that only patricians were elected to the first board. The decemviri drew up their code and brought it before the assembly for ratification, after which the laws were inscribed on ten bronze tablets and set up in the market-place. As, however, the work was not considered complete another board of ten was elected for the following year (450), and two further tablets of laws were drafted. On this second board there were some plebeians, who were thus the first members of their order to hold a magistracy of the Roman people. At this point history fades more and more into legend. The new decemviri are said to have behaved tyrannically and to have refused to lay down their office although their work was done. Popular indignation was aroused especially against one of them, Appius Claudius, who being enamoured of a certain Virginia, instigated a dependent of his own to claim her as a slave, and then, in his capacity as magistrate, gave interim possession to the claimant instead of allowing, as the law required, that a person whose liberty was questioned should remain in freedom until the case was decided. The girl’s father then slew her to avoid her dishonour and led a

‘Second Secession of the Plebs’ which resulted in the overthrow of the decemviri and the restoration of the constitution. The two draft tablets were, however, put before the assembly and passed, thus bringing the number up to twelve. to the plebeian temple of Ceres. According to the tradition this sacrosanctity was given legislative sanction by one of the leges Valeriae Horatiae of 449 B.C. 2 The date of Solon’s legislation was probably 594-593 B.C. 18

pn thet ss The republican constitution

In spite of the obviously legendary character of much of this story, some points are fairly clear. The compilation of the XII Tables was an episode

in the struggle of the orders, and constituted a victory for the plebs. Obviously the law had not only been administered by patrician magistrates but had been unknown in a large measure to the general public. The plebeians wanted a code, so that, if a plebeian were wronged by a patrician magistrate, he could point definitely to the provision in the code which the magistrate had broken. It is likely enough that the code was intended to be a substi-

tute for the tribunician power. The main use of the tribunes lay in their function of auwilium, i.e. they could stop the magistrates from treating plebeians unjustly, but this method of preventing injustice is one of the clumsiest ever invented. First a magistrate is appointed and then another person — the representative of a particular class — to interfere with the magistrate in

the interests of that class. The class representative is given the power of bringing the whole of the state machinery to a standstill by his veto (intercessio), and there is thus a constant element of anarchy in the state. That the tribunes were an anarchical institution had been shown by the fifty odd years of strife since their institution. Very probably it was intended by codification to substitute for the personal protection of the tribunes the more regular protection by definite law, with which the plebeians might be expected to content themselves, as members of their own order could be on the commission which was to draw up the code. But if this was the intention, the violent end of the decemvirate frustrated it. The restored constitution included the appointment of tribunes, and there was never another attempt to abolish them. Indeed when the struggle between the orders was over, they became, as we shall see,’ an important instrument of senatorial government. In addition to the restoration of the constitution, the plebs, by their secession, obtained certain concessions which were embodied in several leges Valeriae Horatiae, passed by the new consuls for 449. Of these the most important were one de provocatione * and another which is represented as giving to resolutions of the concilium plebis the force of law.* A few years

later, in 445, by a lex Canuleia, the prohibition of marriage between the orders (which had been included in the XII Tables) was removed, and at the

same time a compromise was arranged with regard to the admission of plebeians to the magistracy. Instead of consuls there might now be elected military tribunes (generally six, sometimes four) with consular power,® it being left to the senate to decide

each year which form the highest office in the state should take.’ In the

3 Below, 54. 4 But see above, 12. 5 Below, 24.

6 Tribuni militum consulari potestate. They must not be confused with tribunt plebis. 7 Technically it was a matter for the magistrate who presided over the elections to decide, but we can hardly suppose that he would take such an important decision without reference to the senate. 14

_ The struggle between the orders event, in the years up to 3867, when the office was abolished, military tribunes were elected twice as often as consuls. Plebeians were eligible, and Livy presents the creation of the office as a political concession; but he also cites the view that more magistrates with imperium were needed because of the numerous wars of the period. Neither explanation is free from difficulty. On the one hand, only two plebeians were elected before 400 and only seven between then and 367; and on the other hand, there seems to be no relation between the incidence of wars and the election of tribunes.’

The next important moment comes with the passing of the leges Liciniae Seatiae in 367. The intervening years were a period of continuous strife during which the richer plebeians continued to use the economic grievances of

the poor as an instrument for keeping agitation for political concessions alive. But the struggle did not always go their way. One victory which the patricians won (in 448) was the establishment of a new high office confined to patricians, the censorship.® Two censors were henceforward elected (even-

tually every five years) to take over certain duties hitherto performed by the consuls, especially the drawing up of the census, the official list of the Roman people for purposes of military service and taxation. On the other hand in 421 plebeians became eligible for the quaestorship. Something too, but not much, was done to alleviate the distress of the poor by laws assigning portions of conquered territory to citizens. In 877 began the agitation by the tribunes Licinius and Sextius for the enactment of a programme of reform which included both economic and political measures. It took ten years of violent unrest before they succeeded, but in 867 their proposals became law.! The measures passed were as follows:

A. Economic

(i) No citizen to ‘ occupy ’ more than 500 jugera (approx. 820 acres) of public land or keep more than 100 oxen or 500 sheep on the common pasture. (11) Landlords must employ a certain proportion of free labourers. The object of this provision was to provide employment for the growing number of citizens who were thrown out of work by the increase of § De Martino 1.262ff.; Staveley, JRS 43 (1953) 30ff.; Beddington, Historia 8 (1959) 356ff,

® It has been contended that the censorship did not emerge as a distinct magistracy

until perhaps 367, its functions being discharged until then by additional consular tribunes; but see de Martino 1.270ff.

1 The historical aceuracy of this traditional account is disputed; see von Fritz, Historia 1 (1952) 1ff. and, on the measures concerning land, Tibiletti, Athenaewm 26 (1948) 173ff.; 27 (1949) 1ff.; 28 (1950) 245ff. The provision for the employment of free labourers looks particularly anachronistic. Slave labour can hardly have been plentiful enough to constitute a threat before the great wars of conquest of the second century B.C.

15

The republican constitution large estates which their owners found it more economical to cultivate by means of slaves. (i131) Debtors are relieved by subtracting interest already paid from the

capital and making arrangements for balances to be paid in instalments. This was, of course, a measure which could have only a temporary effect. The law of debt was not altered. B. Political (iv) The military tribunate is abolished and in future ONE OF THE CONSULS MUST BE A PLEBEIAN.” The election of patricians to the first decem-

virate and the small number of plebeian military tribunes had shown the necessity of reserving a place in the consulate for plebeians. (v) The keepers of the oracles, of whom there are now to be ten (x viri sacris faciundis), are to be half patrician and half plebeian.

This first admission of plebeians to the priestly colleges is important politi-

cally, for the interpretation of oracles might seriously influence affairs of state. The same year in which the leges Liciniae Sextiae were passed saw the establishment of two new magistracies, the praetorship and the curule aedileship.® The abolition of the military tribunate had again reduced the number of magistrates with imperium to two. The creation of the praetorship raised it to three. The new magistrate was intended to take over the judicial work of the consuls, and his office is therefore, for the study of Roman law, the most important magistracy of all. Livy represents the praetorship as originally confined to patricians and its institution therefore as a compromise, but, at any rate, it did not long remain a patrician preserve, for in 887 we know that a plebeian was elected to the office.*

With the reservation of one consulship to their order the plebeians had gained their chief point, and the opening of other offices to them followed in due course. In 351 a plebeian was first elected censor and in 3389 a lex Publilia reserved one censorship to the plebeians. In 800, by a lex Ogulnia, the ponti-

ficate, the most important of the priestly colleges from a political point of view, was thrown open to the plebs,*® and therewith the struggle as regards 2 But it is only from 320 onwards that this provision is consistently observed, and not until 172 that two plebeians are elected together. 3 Below, 48, 49. They were probably not introduced by the rogations themselves; see Magdelain, Recherches sur Vimperium; ef. id. RHD (1964) 199ff. Since the consuls themselves were originally called praetors (above, 8), the new magistrate is at first sometimes referred to as praetor minor. 4 Mommsen, StR 2.204, doubts whether the office was ever confined to patricians by statute as there is no mention of a statute opening it to plebeians later.

5 The first plebeian to become pontifex maximus was Tiberius Coruncanius in 254; below, 91.

16

The assemblies of the people offices was over. The last act of the whole struggle was the lex Hortensia of 287, whereby the resolutions of the concilium plebis were given the force of law. The details of this matter, which is of great importance, will be discussed later when we deal with the popular assemblies. ®

The result of the plebeian victory was that the political importance of the distinction between patricians and plebeians disappeared. But Rome did not become democratic. The old aristocracy was replaced by a new one, often described as an aristocracy of office, which consisted of the patrician families,

now comparatively few in number, with the addition of those plebeian families which had gained sufficient wealth and influence to be elected to the higher magistracies. A man was considered a nobleman if one of his ancestors had held a curule office.’? But there was a great difference between the new nobility and the old; the privileges of the patricians had been secured to them

by law, those of the new aristocracy were purely the result of practice. In law there was nothing to prevent the assembly from electing to office a person who could point to no curule magistrates in his family tree, and sometimes the assembly made use of this power.® Generally, however, only members of the old established families, patrician or plebeian, had sufficient influence to secure election, and the feeling of the assembly appears to have been that it

was best to continue entrusting the government to men in whose family government had become a tradition. The poor, it must be admitted, did not benefit much by the admission of rich plebeians to share in the political privileges of the patricians. The limits imposed by the leges Liciniae Sextiae on the occupation of public land were not enforced for any length of time, and the provision for free labourers (if historical at all) remained without much effect. On the other hand, independent circumstances, especially the increase of land available for colonies and the general growth of wealth which was due to Rome’s conquests, did undoubtedly tend to better the material condition of the lower classes. Of the modification of the law of debt we shall have to speak later.

38. THE ASSEMBLIES OF THE PEOPLE From the very first, Roman popular assemblies appear to have had two peculiarities which remained characteristic of them so long as they continued

to exist: (a) the voting was always by groups, and (b) the assembly could 6 Below, 24.

7 Dictatorship, consulship, praetorship, censorship, curule aedileship. For a full discussion of the concept of nobilitas see M. Gelzer, The Roman Nobility (Eng. trans., Oxford, 1969). Gelzer would confine the term nobilis to men whose ancestors had held the consulship (or its equivalent). 8 As in the case of Cicero, consul 63 B.C., who was a novus homo. 17

ee The republican constitution

meet only when summoned by a magistrate, and could transact only the business put before it by the magistrate who summoned it. The assembly did not, like the British House of Commons, vote by a simple counting of heads. First the heads were counted in each group, the majority determining the vote of the group; then the votes of the groups were counted, and the majority determined the vote of the assembly. The first group to be used for this purpose was the curia, and the earliest assembly was therefore known as the comitia curiata.°®

The exact nature of the curia is a matter of some doubt.’ The oldest division of the Roman people was evidently into the three tribes, Ramnes, Tities, and Luceres,? and there were ten curiae in each of these tribes. The curia was plainly an artificial unit, but it appears to have been based on kinship,* and therefore was probably related in some way to the gens, which corresponds to the clan and was based on a real or supposed relationship between its members.* But it was not purely a kinship unit. We are told that each curia had its own territory,® and that the curiate organisation was linked to the military, each curia providing 10 horsemen and 100 foot-soldiers.* There is no sign of any distinction between patricians and plebeians, perhaps because the division had not yet hardened.’ ® Comitium was properly the place where the assembly met, but the use of the singular is rare. Comitia (plural) means ‘ an assembly ’. 1 See Momigliano, Terzo Contributo 571ff. (=JRS 53 (1963) 109ff.).

2 The origin of these tribes is quite uncertain. They must not be confused with the ‘Servian’ tribes, a later institution, below, 20. 3 The antiquarian Laelius Felix, quoted by Aulus Gellius (15.27.5) says that the voting in the comitia curiata was by genera hominuwm. This has sometimes been taken as a direct reference to gentes, but it can hardly signify more than kinship (see de Martino 1,.120ff.). Dionysius (2.7.3) equates the curia with the Greek gparpia, a kinship unit. 4 Since gentes were natural units and would be of different sizes, the curiae probably contained varying numbers of gentes. The gens itself continued to be of importance; see de Francisci, Primordia 162ff. All members of a gens have a common name (e.g. in the case of Marcus Tullius Cicero, Tullius is the gentile name, Marcus being the praenomen

or personal name and Cicero that of the particular family within the gens), there are common customs, and there is an organisation which conducts the common cult (sacra gentilicia). It seems to have been possible for the meeting of the gens to pass resolutions binding on its members, e.g, that they must not bear a certain praenomen which had been dishonoured. 5 Dionys. 2.7.4.

6 Varro, L.L. 5.89, says that 1,000 milites were provided by each tribe, and Servius, Aen. 9.368, that ten equites came from each curia; cf. Festus, s.v. celeres, Bruns 2.4. 7 Cf. above, 10. Livy, 10.8.9, represents the patricians as maintaining that the plebeians

gentem non habere, and it has therefore been argued, on the supposition that the curiae were composed of gentes, that the comitia curiata was confined to patricians; but in later times there were certainly plebeian gentes, and the rule giving intestate succession in the last resort to the gentiles (below, 125) is attributed to the XII Tables without any suggestion that there was some other rule for plebeians. The patricians’ claim reported by Livy, above, was perhaps propaganda, to be understood in the sense that the plebeian 18

The assemblies of the people

What functions, if any, were reserved for the comitia curiata under the monarchy we do not know. The historians do indeed say that it was for the people to decide questions of peace and war, but this is probably only a deduction from the law of a later period. We also hear of leges proposed by

the king and carried by the comitia, but legislation is rare in primitive societies and it is more than doubtful whether any such laws were ever passed.® Nor do modern authorities believe that the Roman historians were right in thinking that the king was elected by the comitia. He was probably identified by an omen, the function of the comitia being merely to acclaim the new ruler, not to choose him.?®

We must indeed not imagine that there was any very definite constitution.

Probably it would depend very much on the personality of the king what questions he decided by himself or with the advice of the senate, and what he left to be decided by the comitia. In any case the comitia curiata did not retain any political importance in historical times; perhaps at about the time of the abolition of the monarchy

it was replaced, for political purposes, by another comitia, that is, by an assembly in which the citizens were summoned according to different group-

ing. But although it was politically unimportant in republican times, the comitia curiata continued to meet for certain purposes connected with private law, the making of wills ' and adrogations,? and it had also one formal function connected with public law. When a magistrate with imperium had been

elected in the new comitia his office had to be confirmed by a lex curiata passed by the curiate assembly.* In fact, however, for all these purposes, which become purely formal in historical times, the thirty curiae were represented by thirty lictors, the attendants of the magistrates. gentes did not deserve the name; see Magdelain in Hommages a J. Bayet (Paris, 1964)

464. For another view see Giuffré, Labeo 16 (1970) 329. 8 Below, 86 n 2. ® See Gaudemet, Institutions 267ff.; Kunkel in Ius et Lex (Festgabe fiir M. Gutz-

1 Below, 127. 2 Below, 120.

willer) (Basle, 1959) 3ff.

8 Cicero, Agr. 2.26, represents the lez as repeating the election and thus giving the people an opportunity for second thoughts, but this cannot have been its original function,

when the comitia cwriata was the only assembly. It is usually seen as a survival of the part played by the assembly in the creation of the king (see text above), but its nature is much disputed. It is sometimes called lex de imperio and is thus commonly supposed to have conferred, or to have been in some way connected with, imperium (above, 8). On

the other hand, it is also linked to the power of taking the auspices (below, 30 n 2), and may have conferred it (Staveley, Historia 5 (1956) 84ff.). But again, the jurist M. Valerius: Messalla seems to say (Gell. 13.15.4) that a lex curtata was necessary even

for lesser magistrates who lacked imperiwm or the power of taking the auspices. Magdelain, Recherches sur Vimperiwm 15 n 1, supposes that, while the choice lay with the

comitia centuriata or tributa, the investiture and the determining of the limits of the magistrate’s power was the function of the lez curiata; but this conflicts with what Cicero says (above), and he must have known the wording of the lex (ef. Nicholls, AJPhil. (1967) 257ff.). 19

The republican constitution The individual curiae nevertheless continued to exist; they had sacred rites and apparently communal funds, but they were of no great importance. The assembly which replaced the organisation by curiae for political purposes was the comitia centuriata, the basis of which was the centuria, literally ‘hundred ’. This comitia was the central feature of a new constitution which was, according to tradition, the work of Servius Tullius, the last but one of the seven kings (traditional date 578-585 B.C.). Some of the features of this ‘ Servian ’ constitution were later additions, but it is now widely held that

tradition is right at least in placing it at about the end of the monarchical period. From a reference in the XII Tables * to comitiatus maximus, it seems certain that the comitia centuriata was in existence by the middle of the fifth century, and the main features of the constitution can readily be accounted for in terms of what we know of the political and military position of Rome c. 500 B.C.5 Rome was evidently then economically prosperous and must have attracted numerous immigrants, but these immigrants were necessarily excluded from citizenship under the curiate organisation, based as it was on

kinship. At the same time Rome was embarking on a period of military expansionism, which demanded a larger army. The new constitution was plainly connected with the military organisation of the Roman people and very probably made possible an extension of the citizen body. The three old tribes were superseded by new ones based on place of residence, the city itself being divided into four tribes, and the land outside the walls originally perhaps into a further sixteen; the number of these ‘ rustic ’ tribes was increased as Rome expanded until in 241 B.C. the total number of tribes was fixed at thirty-five. Residence, not kinship, was thus the basis of the new organisation, and the consuls (and after them the censors) would in

these early years have had the power of enrolling immigrants in the tribes when they made the census. Residence was, moreover, sufficient; ownership of land was not required.* The new tribes existed only for political purposes and as a basis for taxation; unlike the curiae, they had no self-government and no sacra. In the constitution as we know it from the historians, the citizen body was also divided into five classes according to wealth. We are given the property classifications for the various classes in money, but as the organisation goes 4 Tab. 1x.1, 2.

5 Last, JRS 35 (1945) 30ff.; ef. de Francisci, Primordia 668, St. Arangio-Ruiz 1.1ff. (abandoning an earlier sceptical view); Staveley, Historia 5 (1956) 75ff.; de Martino 1.132ff.; Momigliano, Terzo Contributo 594ff. (=JRS 53 (1963) 119ff.). Nicolet, RHD

20 ,

(1961) 341ff., however, prefers a date in the fourth century. 6 Last, JRS 35 (1945) 30ff. Appius Claudius, censor in 312 B.C., was not, as Mommsen (StR 2.402ff.) thought, the first to admit landless citizens; he probably re-distributed the urban population over all the tribes, thus curtailing the power of the country-dwellers ; his innovation was reversed by the censors of 304 B.C.

The assemblies of the people back to a time much earlier than the first coinage of money at Rome,’ they must originally have been expressed in terms of other property.® Sons under the power of their fathers, who could not own any property themselves, were presumably placed in their fathers’ class.°

Each class was divided into military companies or centuriae (literally ‘hundreds ’). In the first (wealthiest) class there were eighty centuries, in the second, third, and fourth, thirty centuries each. All these were infantry. Above them, without any definite property qualification but drawn presumably from the wealthiest members of the first class, were eighteen centuries of equites (cavalry), and there were also four centuries of artificers and buglers, and one consisting of all those who had not even the qualification for

the lowest class (proletarit or capite censi). In all there were thus 193 centuries, and these were the voting groups of the new assembly, the comitia centuriata. The significant feature of the new comitia, it will be seen, was its timocratic basis, that is, the preponderance given in it to wealth. For the wealthiest class together with the equites could muster ninety-eight centuries and could therefore outvote the rest, though they must have constituted only a small part of the citizen body. Although as we know it the centuriate organisation serves a political purpose, it clearly was originally connected with the military organisation of the Roman people. The arrangement into cavalry, infantry and technical troops

is sufficient proof, and we hear further that in each class half the centuries consisted of iuwniores (men under the age of forty-six), who formed the active

army, half of seniores (men between forty-six and sixty), who formed the reserve. We also hear that the weapons and armour with which the members of the different classes had to provide themselves varied; the first class having to provide the panoply of a fully armed infantryman, the others a less expensive equipment. Moreover, its ceremonial was military.” On these grounds Mommsen held that the comitia centuriata was in origin and long remained essentially the Roman people in their military array,* the military organisa-

tion being used for voting purposes. There are however objections to this view. The centuries of the voting assembly cannot ever have been the tactical

unit used in the field, because military organisation requires a unit of standard strength, whereas the number of centuries in each class was settled once

and for all and their strength therefore must have varied with growth of population and the changes in wealth. Further it would be impossible in any 7 Below, 145.

8 Mommsen, StR 3.247, conjectured that only land was included; contra, Last, JES 35 (1945) 40f. Perhaps any res mancipi counted; Greenidge 69; Kaser, HB 175ff. ® Festus, s.v. duicensus (Bruns 2.7).

1 And within the eighteen there was an original nucleus of six (sex suffragia) which

2 Below, 26. $ StR 3.253.

voted first; cf. below, 23 n 8.

21

The republican constitution community for the number of men between forty-six and sixty to be as large as that of men under the age of forty-six.* It is nevertheless inconceivable

that the centuriate organisation should have acquired its many military features without having any connexion with the army in the field, and the

explanation is probably that what we have is not in detail the original organisation.°® , In 241 B.C. the number of tribes reached thirty-five, and thereafter no more were added: newly acquired land was incorporated into the existing tribes.® Then or soon afterwards’ the composition of the comitia centurtata was altered to achieve some co-ordination between tribes and centuries. Our information is slight and obscure, and the nature of the reform has been much debated.* From a statement of Cicero’s ° it seems that in his day there were

seventy, instead of the original eighty, centuries in the first class, and it is generally supposed that the new figure resulted from an allocation in this class of two centuries (one of seniores and one of iwniores) to each tribe. If the same principle had been applied to the other four classes there would have been a total of 373 centuries (5 x 70 +18 centuries of equites +5 supernumerary centuries of the unarmed), but it seems clear from what Cicero says that the total remained at 198, and in any case so drastic a shift in voting power is hardly conceivable. Mommsen ' therefore conjectured that the third century reform involved the creation of new centuriate units (two to each tribe

in each class) in addition to the old voting centuries. Only in the first class did each new centuriate unit have one voting century; the 280 (4x 70) divisions of the next four classes were distributed unequally among the remain4 For these and other arguments see G. W. Botsford, The Roman Assemblies (New York, 1909) 80ff., 203f.

5 We are told that the first class was called simply classis, and the others infra

classem (Festus 100u; Gell. 6.13). Perhaps classis was the legion and infra classem the lighter troops and non-combatants. Moreover, it seems likely that there were originally

only six centuries of equites, and that they were not drawn from the nobility (who fought on foot) but constituted the king’s bodyguard; see Momigliano, Quarto Contributo

377ff. (=JRS 56 (1966) 16ff.). The difficulty of tracing the development of the centuriate organisation is compounded by the difficulty of tracing the early development of the army; see de Martino 1.142ff.; Scullard, History 423ff. 6 The tribes therefore gradually lost their territorial significance and the tribal desig-

nation became almost part of a citizen’s name, passing normally from father to son. Thus Cicero’s official description was M(arcus) Tullius M(arci) fi(lius) trib(u) Arn(iensi) Cicero.

7 Before 218, because the surviving portion of Livy’s history resumes in that year, and in 216 he takes the revised arrangement for granted. The reform was probably made by the censors, and a newly discovered inscription may, on one interpretation, show that it was made by Q. Fabius Maximus, censor in 230 B.C.; see Cassola, Labeo 8 (1962) 307; contra, Guarino, Labeo 9 (1963) 89ff. 8 For a conspectus see Walbank, Polybius 1.683ff.; Meyer, Rém. Staat 88ff., 497ff. ;

Arangio-Ruiz, Storia 415ff. |

® Rep. 2.39. The interpretation of the passage is disputed; see Sumner, Historia 13 (1964) 125ff., with references. 1 Sth 3.270ff. 22

||

The assemblies of the people

ing 100 voting centuries, thus retaining the timocratic principle while achieving a relationship between the tribal and the centuriate organisation. This system was generally rejected by Mommsen’s critics as improbably complex,’ but an inscription (commonly now called the Tabula Hebana*) discovered in 1947 at Magliano in Tuscany (the ancient Heba) suggested that in

fact it might be near the truth. This inscription, dating from A.D. 19-20, reveals that a lex Valeria Cornelia of A.D. 5 set up a special electoral body to make the pre-election (destinatio) of candidates for the consulship and the praetorship. This body, consisting of the senators and those equites who were eligible to serve as judges, was drawn from 38 tribes, but voted by centuries (at first ten, but raised to fifteen in A.D. 19 and to twenty in A.D. 23),* two or three tribes being allocated to each century by lot on each occasion. Since this body must be in substance a part of the comitia centuriata,‘ it is tempting to suppose that the allocation by lot was derived from the procedure of the comitia, and various detailed systems have been proposed.® But the pos-

sibility remains that the third century reform involved no more than a redistribution among the other classes of the ten centuries taken from the first class.

The reform is represented by Dionysius ’ as a ‘ democratic ’ one, and it is

true that the conservative Sulla (88 B.C.) temporarily restored the old system, but the significant feature was probably not the slight reduction in the number of centuries in the first class ®* but the shifting of the voting basis to the tribes. Under the old system it must have been possible for the urban

population, who could more easily attend meetings, to outvote the country dwellers in each class, whereas now the urban tribes could control at most eight votes in each class (two for each urban tribe), regardless of how many of them voted.° The tribes were already the basis of another assembly, the comitia tributa. 2 But see Momigliano, Quarto Contributo 363ff. (=SDHTI 4 (1938) 509ff.).

8 V. Ehrenberg and A. H. M. Jones, Docwments illustrating the Reigns of Augustus and Tiberius, 2nd ed. (Oxford, 1955) 76ff. For discussion see G. Tibiletti, Principe e magistratt repubblicant (Rome, 1953); de Visscher, Etudes de droit romain, 3e série (Milan, 1966) 3ff., 51ff. (=RHD, 1951, 1ff.; St. Arangio-Ruwi2 2.419ff.), and further, below, 326 n 8.

4 This further increase was revealed by another inscription (Année Epigraphique, 1952, 80).

5 It evidently performed the function of, and presumably replaced, the centuria

6 See above, 22 n 8. 7 4.21.3.

praerogativa (below, 28).

8 It was probably at this time (certainly after 296 B.C., see Livy 10.22.1) that another change was made. The privilege of voting first which had previously been

enjoyed by the sez suffragia (above, 21 n 1) was transferred to one of the centuries of the first class, chosen by lot. For the significance of this privilege, see below, 27f. This gain by the first class may well have outweighed the reduction in its number of centuries. 9 Meyer, Rom, Staat 90f.; Staveley, AJPhil. 74 (1953) 23ff. 28

The republican constitution It had full legislative power, but its functions were otherwise parallel but inferior to those of the comitia centuriata: election of lesser magistrates, adjudication on provocatio against fines levied by the curule aediles. When it originated is uncertain, but the reference in the XII Tables to comitiatus maximus 1 may imply its existence as the lesser comitia, and there seems no reason to doubt Livy’s statement that a tax on the manumission of slaves was imposed by it in 357 B.C.? But the whole growth of the institution is obscured by confusion with the concilium plebis, the assembly of the plebs alone, which also met by tribes. As the patricians must have become in the later part of the republic a numerically insignificant body of nobles, and as the purely plebeian assembly attained in the end equal legislative capacity with the assemblies of the whole people, it was easy for anyone unacquainted with the niceties of the constitution to confuse the two bodies, and this probably accounts for the difficulties which modern historians find in interpreting the sources.°®

Of the concilium plebis we have already had occasion to speak in discussing the struggle between the orders. At first, it appears, the plebeians were in the habit of meeting by curiae for the election of tribunes and perhaps other business. At any rate according to Livy * it was enacted, by a certain lex Publilia of 471 B.C., that henceforward the arrangement should be by

tribes, and so it remained. It seems that this reform was of democratic nature because it diminished the influence of the patricians over the plebeian assembly.*° So long as the plebs had met by curiae the clients of the great patrician houses had been members of the concilium plebis and had been able sometimes to turn matters as their patrons desired; once the tribe was substituted for the curiae these landless persons were necessarily excluded and the plebs became more independent. The history of the steps by which the concilium plebis attained legislative power is more than usually obscure. We hear quite definitely of three leges,

all of which are represented as enacting, almost in identical words, that plebiscita, i.e. resolutions of the plebs alone, should have the force of law. The first is a lex Valeria Horatia, said to have been passed in 449 B.C. as a 1 Above, 20.

2 Livy 7.16; but this was anomalous as it took place in castris. According to Tacitus, Ann. 11.22, the election of quaestors (which was, at least in later times, a function of the comitia tributa) first took place in 447, and Livy, 3.72.6, says that it was the tribes that voted the so-called lex de agro Coritolano in 446. De Martino, 1.330f. regards both these as anticipations, and dates the origin of the comitia to the fourth century on the

ground that only then would the rural population (whom the tribal voting system favoured, above, 23) have become sufficiently substantial to obtain this innovation. 3 For a summary of the differences between the assemblies see Taylor, Roman Voting

4 2.58, 5 Livy 2.56.

Assemblies, table facing p. 5; and against the view that in the later republic patricians attended the conctliwm plebis, ibid. 60ff.

24

The assemblies of the people result of the fall of the decemvirs,* the second a lex Publilia Philonis of 389, the third the lex Hortensia of 287.° The relation of these laws to each other has been the subject of much conjecture. The lex Hortensia is the only one

mentioned by legal writers, and some think that it alone had any real existence.® The two earlier statutes would then be a product of the tendency of Roman historians, which we have already noticed, to read back comparatively recent events into the remote past. Here the legend might have arisen

from the fact that both 449 and 839 were really moments at which democratic reforms were introduced.’ Many scholars, however, concede historical existence to one or both of the earlier statutes, but suppose that the validity granted to plebiscita was qualified in some way, e.g. by the requirement of

patrum auctoritas? or of confirmation by the comitia centuriata.* But the lex Canuleia,* the leges Liciniae Sextiae,® and the lex Genucia™ were all plebiscita, and it may be therefore that the historians are right and that the patricians tried to disregard the two earlier statutes.”* The legal writers would

have been concerned only with the one which was finally effective. In any case there is no doubt that after the passing of the lex Hortensia, plebiscita, that is enactments of the plebs in the concilium plebis, an assembly in which the patricians had no place, were equally valid with leges, that is enactments of the whole populus, i.e. patricians and plebeians together, in the comitia centuriata or tributa. We thus have the strange result that in the later republic there were three bodies all equally capable of passing binding statutes, three sovereign legislatures, as we should call them,** the comitia centuriata and tributa, consisting of the same people, though organised differently, and the concilium plebis, consisting almost entirely of the same people (for the patricians must by 287 have become a numerically insignifi6 Ut quod tributim plebs wussisset populum teneret; Livy 3.55. 7 Ut plebiscita omnes Quirites tenerent; Livy 8.12. 8 Ut quod plebs iussisset omnes Quirites teneret; Pliny, Nat. Hist. 16.15 (10). See also Gai, 1.3; J. 1.2.4; D.1.2.2.8 (Pomponius). 9 E.g. von Fritz, Historia 1 (1952) 19. 1 For another lex Publilia Philonis of 339 see below, 33. 2 I.e. approval by the patrician members of the senate; below, 33. 3 De Martino 2.127ff.; Guarino, Ordinamento 376ff. (=F schr. Schulz 1.458ff.). Staveley, Athenaeum 33 (1955) 3ff., supposes that the lex Valeria Horatia validated both plebiscita

and the enactments of the comitia tributa, subject to the requirement of patrum auctoritas, and that the lex Publilia removed this restriction for enactments of the comitia tributa and the lex Hortensia for plebiscita; but see the criticisms of Arangio-Ruiz,

Storia 414f.14. | | 5 Above, 15. 4 Above, 6* 342 B.C., forbidding loans at interest.

7* Stuart-Jones, CAH 7.528, points to the fact that in the years 355-247 (i. after the leges Liciniae Sextiae had required that one consul should be a plebeian) on at least six occasions two patrician consuls were elected. 8* Subject, in the case of the comitia centuriata at least, to the formality of patrwm auctoritas, below, 33.

25

The republican constitution cant minority), and meeting like the comitia tributa by tribes. We must also remember that the comitia curtata, an assembly of the whole populus, organised again on a different basis, though it had lost its political importance, still maintained a formal existence.

Which of these assemblies was summoned in any particular instance depended on the magistrate who wished to put a proposal before the people. The normal presidents of the comitia centuriata were the consuls, both for legislation and for elections; the comitia tributa could only be summoned by * patrician ’® magistrates, usually the consuls or praetors, and the con-

ciium plebis could only meet under the presidency of a magistrate of the plebs. Though the legislative competence of all the assemblies was equal, the concilium plebis became the usual organ for the passing of laws in the later republic as its presidents, the tribunes, had more time for, and interest in, legislation than the consuls, who were frequently engaged in military duties.’ So far as elections were concerned the functions of the assemblies were more strictly differentiated, the magistrates with imperium and also the censors being elected in the centuriata, the lesser magistrates of the people (e.g. curule aediles and quaestors) by the comitia tributa, the tribunes and plebeian aediles by the concilium plebis.? The centuriata was thus in practice confined usually to the most important elections, one reason for the comparative rarity of its meetings being the cumbrousness of the ceremonial involved. It was (in theory) the people in arms and could only meet with full military ceremonial outside the walls, usually in the Campus Martius. When it met, a watch was placed on the Janiculan Hill (on the opposite side of the Tiber from the main part of the city and the Campus Martius) and a flag hoisted, the striking of which would be the signal that the enemy

was approaching and that the consul should break up the meeting. This practice continued until well into the empire, although the Tiber had for centuries ceased to be the boundary of Roman territory. ® This term does not imply that the magistrate is necessarily himself a patrician; it is used to differentiate the officers who were strictly magistrates (of the populus), e.g. consuls, praetors and curule aediles, from those who were not strictly magistrates at all but only officers of the plebs, the tribunes and the plebeian aediles. Justinian (J. 1.2.4), in distinguishing between leges and plebiscita, is somewhat misleading. He says: ‘A lex is an enactment passed by the populus on the motion of a senatorial magistrate (senatore magistratu interrogante), e.g. a consul; a plebiscitwm is an enactment passed by the plebs

on the motion of a plebeian magistrate, e.g. a tribune.’ This might lead the reader to suppose, wrongly, that tribunes were not members of the senate. 1 But even when a consul or a praetor proposed legislation he apparently did so before

the comitia tributa; we know of no lex passed by the comitia centuriata between 287 B.C. and Sulla; see Gaudemet, Institutions 388 n 2. 2 For the criminal jurisdiction of the assemblies see below, 305ff. 3 Dio Cassius 37.28.

26

The republican constitution

4,.\CHARACTERISTICS AND PROCEDURE OF ROMAN ASSEMBLIES As already noticed, a Roman assembly can meet only when summoned by a magistrate, and it can transact only the business put before it by the magistrate. His proposal it must either accept or reject; there is no possibility of amendment, nor, in the actual assembly itself, of any discussion. When legis-

lation was proposed, the magistrate had to draft the bill (rogatio) which he intended to put before the people, or cause it to be drafted for him, and this bill was, by constitutional practice, usually debated in the senate. By strict law, however, this was not necessary, and the consent or disapproval of the senate made no difference to the validity of a law once it was passed. It was the duty of the magistrate when the bill was drafted to publish it in an edict (proponere, promulgare) in which he also announced the date on which he would summon the assembly to vote on it. The interval between promulgation and voting had, by a lex Caecilia Didia (98 B.C.), to be at least twenty-four days (trinundinum — three Roman weeks), and this | rule seems to have been followed in practice even before it was laid down by

law. After promulgation the bill might be withdrawn but it could not be altered. During this interval the magistrate could, and usually did, hold informal public meetings for discussion (contiones) in which he spoke himself and might allow anyone else to speak for or against the bill. Sometimes the contio took place immediately before the assembly proper. Before the

voting the bill was read out again and the presiding magistrate asked the people whether they agreed to it or not, velitis iubeatis, Quirites ? The people

then arranged themselves in their divisions (tribes or centuries) and voted separately. Originally each man gave his vote orally, but with the growth of bribery in the late republic an attempt was made to check it by introducing the secret ballot.* Each man was given a small wax-covered tablet, which he marked either V (uti rogas - ‘as you propose ’) if he wished to vote for the proposal, or A (antiquo — ‘I vote for the old state of things ’) if he wished to vote against.°

In the comitia centuriata, as we have seen, the classes voted in order (though within each class the centuries apparently voted simultaneously), and since the votes of each class were announced before the next began to vote, the familiar tendency of the floating vote to follow a lead gave an 4 Introduced for legislation by a lez Papiria of 131 B.C. For elections it had been already introduced eight years previously by a lex Gabinia. 5 Or, when the assembly was sitting as a court, L (libero) or D (damno). Alternatively perhaps both letters were written on the tablet in advance, and the voter struck one out.

In elections the tablet would be larger and the voter would inscribe the initials of the candidates he preferred. See Taylor, Roman Voting Assemblies 34f., 76f., 146.

27

The republican constitution advantage to the class that voted first (the equites). This advantage was sharpened when it was transferred to a single century of the first class (cen-

turia praerogativa) * and that century was chosen by lot. For the lot was seen as expressing the will of heaven.’ In the two tribal assemblies the voting, except in elections, was likewise successive,’ and since the entire order

was decided by lot, the first tribe to vote (principium) must have had an advantage similar to that of the centuria praerogativa. Once accepted by the assembly a bill became law immediately, unless, oi course, it contained provisions postponing its coming into force. Publication was not necessary for validity, though it was common; sometimes the law itself provided that its provisions should be put up in public * where they

could be easily read from the ground’ (unde de plano recte legi possit). _ Originally wooden tablets were used for this purpose, later bronze ones. An official copy of the law was in any case kept in the aerarium (treasury), but there appears to have been no adequate provision for filing and arrangement, for Cicero complains that only the skilled assistants could discover an enactment which it was desired to consult, and that they therefore in fact decided what was to be law.? The Romans had the principle, as must all communities with legislative bodies, that a previous enactment can always be repealed by a subsequent one, and also that where a subsequent enactment conflicts with a previous one it must be taken as repealing it to the extent of the inconsistency. The

Roman constitution was also, like the British, a ‘ flexible’ one, that is to say there were not, as in the constitution of the United States of America, any fundamental constitutional rules which could not be abrogated by the ordinary legislature. Just as the British Parliament could repeal, if it thought fit, the most important constitutional laws, e.g. the Act of Settlement, and

would use for that purpose precisely the same procedure as that which it uses for passing a Licensing Act, so the Roman comitia or concilium plebis 6 Perhaps always a century of tuniores; Taylor, Roman Voting Assemblies 91f.; ef.

above, 23.

7 Cicero, Planc. 49, says that in consular elections a candidate who had the vote of the centuria praerogativa was nearly always successful; Taylor, Roman Voting Assemblies 151 n 16. He is even more emphatic in pro Murena 38.

8 Taylor, Roman Voting Assemblies 40f., 128ff., following P. Fraccaro, Opuscula 2 (Pavia, 1957) 235ff., against Mommsen, StF 3.396ff., who thought that voting was always

simultaneous. Even in elections the order in which the results were announced might affect the outcome in some circumstances, since the announcements ceased once a suffi-

cient number of candidates had obtained the necessary number of votes; see Taylor, Roman Voting Assemblies 81f. . 9 Cic. Leg. 3.46. No doubt an exaggeration. On the passage see Schwind, Publikation 30.

1 This was expressly recognised by the XII Tables (Tab. x11.5). Livy 7.17: in XII tabulis legem esse, ut quodcumque postremum populus tussisset, id ius ratumque esset.

28

Characteristics and procedure of Roman assemblies could, by using its ordinary procedure, alter the Roman constitution, and indeed frequently did so. But this fact did not prevent assemblies from attempting to bind their successors by laying down rules as to legislation. Most famous of such rules is one contained in the XII Tables.? Privilegia ne inroganto, ‘no law may be passed against an individual’, i.e. laws are to lay down general rules; the procedure of legislation is not to be used to penalise a particular individual who has not broken some general rule of the community. An example of a privilegiwum would be an English ‘ Act of

Attainder ’, i.e. an Act of Parliament ordering 4 particular person to be executed, such as was used for instance by Henry VIII to get rid of his minister Thomas Cromwell.

A later enactment which similarly attempted to lay down a rule for legislation was contained in the lex Caecilia Didia.* This forbade proposals deal-

ing with unrelated subjects to be included in the same bill (ne quid per saturam ferretur), the object of such ‘ tacking’ being, of course, to induce the people to accept an unpopular proposal because they could not reject it without at the same time rejecting the proposal which they welcomed. Now it is clear that, given the principles that a later enactment repeals a former and that there is no difference between fundamental constitutional laws and others, such limitations cannot strictly bind future assemblies. If the assembly chooses to pass a bill which does contain unrelated matters,

then the bill should, at any rate in strict logic, become law and the lex Caecilia Didia be deemed to be repealed in so far as it conflicts with the new law.

This seems indeed to have been the strict legal position: there was no external limitation on the validity of a lex, but there were strong conventions which internally restricted the freedom of the assembly. Thus the Romans themselves sometimes argued that an act of the people must necessarily be valid even if it was contrary to some previously enacted constitutional principle, because of the implied repeal,* but the assembly no doubt felt the same repugnance to the enactment, for example, of privilegia as the

British Parliament would to the repealing of the Act of Settlement. It is probably in this sense that Cicero argues * that a lex cannot deprive a citizen of his liberty or citizenship, i.e. that it is improper for a lex to impose such a deprivation on a particular citizen or class of citizens.* He does, however, also tell us ’ that a clause was added to every statute which expressly limited 2 Tab. 1X.1. 3 Above, 27.

4 An example of this occurs in Livy 7.17, where it is argued that an election of two patricians as consuls is valid, in spite of the lex Licinia Sextia, because the election itself

is an act of the sovereign people. Cf. also the arguments which Cicero puts into the

mouth of his opponents; Caec. 95; Dom. 106. 5 Caec. 96. 6 Pugliese, Atti Verona 2.61ff.; cf. Stein, Regulae 14. 7 Caec. 95. 29

The republican constitution its effect to what was lawful. Sz quid ius non esset rogarier, eius ea lege nihilum rogatum. ‘ If there be anything that it would be contrary to law to enact, then no such enactment is contained in this statute.’ Since in logic, as we have seen, express repeal was necessarily effective, this should perhaps be interpreted as excluding repeal by implication.®

5. THE SENATE The senate was in strict theory of law a purely consultative body, at first of the king, later of the consuls. It could not pass laws, as could the comitia; it

could only give advice, and that only when asked by a magistrate with whom the executive power lay.’ In the early republic it may be that the magistrates were in fact as well as in law more powerful than the senate, but

later, through the growth of a constitutional convention which made the consultation of the senate by the magistrate necessary for almost all business except routine matters and the conduct of operations in the field, it was the senate which became the actual governing body at Rome.

A. Composition. Originally, then, the senate was the king’s council, and according to tradition he freely appointed the members of it.' The existence, however, within the republican senate, of a body of patres, and their func-

tion in an interregnum, suggest that his freedom was not unlimited. For while in a broad sense the whole senate came to be known as patres, the term

was more strictly applied to an inner aristocracy, and it was the patres in this narrow sense who had the ultimate power when the supreme magistracy was vacant, as might happen if both consuls died suddenly. In such circumstances there was no-one competent to hold elections and, as the technical phrase went, the ‘ auspices ? returned to the patres ’. The patres then nominated an interrex * who after holding office for five days nominated his successor and so on until one of the interreges held the consular elections.* In later republican times the patres exercised their functions in this matter only on the initiative of the senate as a whole, which suggested that they should meet

for the purpose, but there must presumably have been a time when the 8 For the related question of leges imperfectae see below, 87. ® The name for a decision of the senate remains throughout consultwm, i.e. resolution ;

it is never iusswm as is a decree of the people. 1 Festus, s.v. praeterits (Bruns 2.26). 2 Before every public act, such as an election or the putting of a bill to the people, it must be ascertained by omens whether the gods are favourable. This is called ‘ taking the auspices’, and thence the auspices are identified, as in the phrase here quoted, with the

right to take them. 8 Livy 1.17.5. 4 Any interrex could conduct the elections except the first, probably because no election

eould be valid unless the auspices had been taken, and he alone could first take them (Magdelain in Hommages 4 J. Bayet, Brussels 1964, 427ff.).

30

The senate patres were the only senators. At any rate, the terminology shows that the procedure was originally used for the appointment of a king and it seems unlikely that the composition of a body which was in this way constitutionally prior to the king would have been simply dependent on his will.®

Who the patres were is again uncertain. They can hardly have been the heads of all the families, even in the extended Roman sense of family,® and it is commonly therefore suggested that they were the heads of the gentes.’ According to tradition the original number of senators was 100; the steps by which the republican figure of 800 was reached are variously stated,*® but we are told that when Tarquinius Priscus created 100 new senators the original

body was called patres maiorum gentium and the new members patres minorum gentium.® But this does not of course necessarily mean that each was a pater gentis, and there is nothing else to suggest that such a figure existed. Moreover, the question of membership of the senate is confused in the sources with the question of membership of the patrician order (patres in yet another sense).

To the question how and when plebeians were first admitted no clear answer can be given, and if the distinction between patricians and plebeians

did not harden until the early years of the republic,’ the question itself is misconceived. Later writers of course approach the early senate in the terms

of their own time, when the distinction, though no longer important, is nevertheless sharp, because it is seen in the perspective of the struggle between the orders. There are, however, ambiguities and variations in the accounts given by these writers which may have resulted from imposing the distinction on a period when it was not yet so clearly defined.” The existence of an inner aristocracy of patres in the narrow sense is, as we have seen, beyond question. It appears in the famous phrase, patres conscripti, which was always used by speakers addressing the senate. In later times this was taken to express a single idea and to apply to all senators, as being patres who had been selected and enrolled, but in origin it was evidently equivalent to patres et conscript, the conscripti being those who were not patres in the narrow sense but had been enrolled.* In the later republic the conscripti were

5 For patrum auctoritas see below, 33. 6 Below, 114, 118.

7 E.g. de Martino 1.117ff. 8 See Ogilvie, Livy 1-5 148. 9 Cie. Rep. 2.35; Fam. 9.21.2; Livy 1.35.6; Suet. Aug. 1, 2; Tacitus, Ann, 11.25, attributes the introduction of the patres minorum gentiwm to Brutus, one of the first two consuls. The significance of the distinction between maiores and minores gentes is not

further explained; Ogilvie, Livy 1-5 147f. 1 Above, 10. 2 See Momigliano, Quarto Contributo 424ff. (=Entretiens Hardt 13.204ff), who suggests a threefold division into patres, populus, and plebs, the last two being identified with classis and infra classem (above, 22 n 5). 3 Festus, s.vv. adlecti, conscripti, qui patres qui conscripti (Bruns 2.2, 5, 29) and other evidence in Mommsen, St 3.838ff.; contra, U. von Libtow, Das rom. Volk (Frankfurt, 1955) 144ff.; Ogilvie, Livy 1-5 236.

31

The republican constitution certainly plebeian and Livy * attributes their origin to the first consuls, who filled the gaps which the preceding revolution had left in the ranks of senators _ by drawing on the equestrian order. But this may well be a rationalisation in terms of the later class structure, and there are other versions which place

the admission of plebeians earlier.’ Similarly, it is not clear whether the patres minorum gentium are to be seen as patrician or plebeian.® In short, all that can safely be said is that on the one hand if there were non-patrician (to use a neutral word) kings and, in the early years, non-patrician consuls, there must have been non-patrician senators, and on the other hand that, in terms of the struggle between the orders, once the tribunate with consular power was opened to plebeians it must have been possible for them to be admitted to the senate. But since Livy constantly represents that body as the stronghold of the patrician party and since, as we have seen, the number of plebeians who held the consular tribunate was small, it is clear that plebeian senators must have been a minority, and probably a small minority. In their choice of senators the consuls were subject to no formal restrictions. They were no doubt more or less constrained by custom to choose men who had held high office, but so long as the magistracies were few there must have been a considerable field in which they could use their discretion. in making up the number of 800. By a lex Ovinia, however (c. 312 B.C.),’ the function of appointing senators was given to the censors,*® and they were to

‘choose the best men of every order ’.° This is not further explained, but since the lex Ovinia was a plebiscite the orders in question are most probably

the patrician and plebeian.’ Alternatively, the-reference would be to the different classes of ex-magistrates, both curule and other.? Be that as it may, ex-magistrates did acquire iri the course of time, by constitutional practice or in some cases by law, a right to be chosen unless some

good reason could be assigned for excluding them. At the time of the Punic

wars this right was still confined to the ex-curule magistrates (consuls, 4 2.1.10, 11; cf. Plut. Qu. Rom. 58, Rom. 13; Dionysius 5.13.2, says that Brutus chose the best of the plebeians, made them patricians and then enrolled them in the senate, but this is evidently because he assumed that senators must be patricians and that a personal grant of patrician status was possible; see Mommsen, StF 3.839 n 2.

5 See Mommsen, loc. cit. 6 Momigliano, op. cit. above, note 2. 7 The first known lectio senatus by censors was that of Appius Claudius and C. Plautius

in 312. Livy, 4.24, records that the censors of 319 removed the name of the consular Mamercus Aemilius from the tribes but makes no mention of his exclusion from the senate, and from this Mommsen, StF 2.418 n 2, inferred that the lex Ovinia had not yet

been passed; ef. de Martino, 1.408 n 3. 8 Below, 53. 9 Festus, s.v. praeteriti (Bruns 2.26): ... ut censores ex omni ordine optimum quemque curiati in senatum legerent. On curiati see Willems, Sénat 1.169ff.

1 O’Brien Moore, PW Suppl. 6.692f. |

2 Willems, Sénat 1.160ff., shows that it cannot have been confined to ecurule magistracies because the number of qualified candidates would have been so small in relation to the number of vacancies as to make the dex pointless. 82

The senate praetors and curule aediles), but it was extended first to plebeian aediles, then to tribunes and finally under Sulla to quaestors. As a result of these extensions there were always enough ex-magistrates to fill all the seats in the

‘senate, and in practice the censors’ power of appointment (though it was regarded as their most important function) was limited to the removal of a qualified person from the list if they considered him to have been guilty of serious misconduct. This remained true although Sulla raised the number of senators to 600, for he also increased the number of quaestors to 20. The result of the filling up of the senate with ex-magistrates was that it became in effect a body elected by the assemblies, the chief importance of election to the quaestorship being indeed that it carried with it a seat in the senate. But it must not be imagined from this that the senate became a democratic body. It was only seldom that the comitia elected men who did not belong to the aristocracy, and, once elected to office, a man held his seat in the senate for life.

B. Funcrions. (i) Functions in connexion with legislation. Patrum auctoritas. The inner aristocracy of patres, who retained a special function in an interregnum, had also another power, which must in origin have been very important. No act of the comitia (legislative, elective, or judicial) was valid without the auctoritas of the patres. For plebiscita, the enactments of a body with which the old constitution had had no concern, auctoritas was probably unnecessary.* It was originally given after the voting, but a lex Publilia Philonis of 889 B.C. required that it be given before the proposal was put to the people in the case of legislation, and this principle was extended to elections by a lex Maenia in the following century.*® The result must have been to change the character of the grant of auctoritas. Previously the patres were validating the act of the comitia; now the comitia was giving effect to a recommendation of the patres, just as it might to any resolution of the whole senate.© Whether for this reason or simply because the change in the political

balance in the senate had made the exercise of such a function by one part only of the senate seem unjustifiable, patrum auctoritas now merges in the general functions of the senate and becomes a mere form.

Consultation by magistrates before the submission of a bill to assembly. The real powers of the full senate were a matter of constitutional convention. It was a convention that the magistrates should not, except for routine 3 Above, 30.

4 But see above, 25 n 2. The evidence for the comitia tributa is so scanty that we cannot say for certain that auctoritas was needed for acts of that body, but the probability is that it was; Mommsen, StR 3.1040; contra, Willems, Sénat 2.86ff.

5 Known only from Cie. Brut. 55, which shows that it had not yet been passed at a time some years before 290; since Livy could not have failed to mention it, it must

3.1042. 6 Meyer, Rém. Staat 213. belong to the period 292-219 of which Livy’s account does not survive; Mommsen, Stk 33

The republican constitution purposes, use their powers without first consulting the senate, and it was customary therefore for the magistrate who intended proposing legislation in the assembly to bring the matter first of all before the senate. This was all the more necessary as, once the proposal had been promulgated, there was no possibility of amendment and the contiones afforded but little scope for real discussion. In the senate on the other hand the proposal could be pro-

perly debated, and the magistrate might alter his original draft in accordance with amendments suggested before he finally promulgated it. The initiative remained of course with the magistrate, but in the later republic, when the number of magistrates who had a right to summon the senate was large, it was always possible for any considerable party in the senate to find a magis-

trate who would call a meeting at their request. Similarly, if a magistrate, in defiance of constitutional practice, wished to put a proposal before the people without first consulting the senate, it was normally possible for the senate to find a tribune who would veto his action. Instances in which magistrates successfully ignored the senate belong to the revolutionary period.

Even if the proposal was not vetoed and passed the assembly, the senate had still a weapon left, for, as we have seen, it came to be considered the guardian of the constitution and it would generally not be difficult to find some flaw in the enactment, for instance that the auspices had not been properly taken, which would invalidate the proceedings. Strictly, it was for the magistrates, as executive officers, to decide whether they should ignore the questionable enactment or put it into force, but they could hardly take so serious a step as ignoring it on their own authority, and the ultimate decision rested in fact with the senate.’ Dispensing power. Strictly, as the assembly was the only body which could

make law, so it alone could exempt an individual from the operation of a law. This power was, however, gradually usurped by the senate, which would

grant a dispensation in urgent cases subject, at first, to ratification by the comitia, this ratification being subsequently reduced to a formality or omitted altogether. An attempt made in 67 B.C. to deprive the senate of the right it had usurped only resulted in its confirmation, by a lex Cornelia which provided that at least 200 senators must be present when a dispensing resolution was passed.® 7 The senate’s position as guardian of the constitution really rests on the idea that it is the great council of state and as such must be the ultimate arbiter when any religious questions arise. The fundamental principles of the constitution are not purely secular matters. They, and indeed the whole of Roman public life, were conceived as having divine sanction and any departure from established custom might call down divine wrath. The senate’s particular connection with religion can be seen from the practice that the first business laid before it by the consuls on assuming office should be a motion concerning religious matters. See e.g. Livy 36.1; 37.1. 8 Asconius, Corn. p. 58c. (Bruns 2.69). Confirmation by the assembly appears to have remained necessary in theory at least. Dio Cassius 36.39 says nothing of the quorum of 200.

34

nner The senate

One important feature of Roman rule, the government of provinces by pro-magistrates, i.e. magistrates whose term of office had expired, rested essentially on the dispensing power of the senate, which prolonged the individual’s imperium by dispensing him from the original term to which he had been restricted.?

(ii) Functions in relation to the magistracy. With regard to the actual election of magistrates the senate exercised little power. We have already mentioned the formality of the patrum auctoritas,’ and the power of the patres to appoint an interrex.? The senate (i.e. the whole senate) also exercised de facto power with reference to another extraordinary magistracy, the dictatorship. As we shall see, the power of nominating a dictator in an emergency belonged to the consuls, but, already early in the republic, it became customary that this power should only be exercised at the suggestion of the

senate, which finally went so far as to suggest, not only that a dictator should be nominated but also who he should be. This practice of course fell into abeyance with the disuse of the dictatorship itself after 202 B.C.’ On the appointment of magistrates elected in the various assemblies the senate had no influence at all beyond those indirect methods of influencing votes which are known to governments of all ages, but these were usually, though by no means always, sufficient to prevent the return of candidates who were distasteful to the majority in the senate. After their election, however, the senate, by the power which it acquired of assigning their different spheres of activity to the magistrates, exercised a very important control.

The consuls, for instance, were each fully competent to exercise all the powers inherent in imperium, but they in fact often arranged a division, each frequently taking command of one of two armies destined for different campaigns. Though strictly the division was made either by arrangement or by lot, the senate sometimes suggested that one magistrate was better fitted

than another for a certain sphere, and, above all, it came to be the undisputed prerogative of the senate to mark out the spheres themselves, i.e. decide which were to be the consular * provinces ’.* In the late republic, the assemblies came from time to time to use their unfettered legislative power to override the dispositions made by the senate;* but these were, of course, revolutionary times and usually the decision in this matter rested with the senate and gave it some measure of control over the consuls. It was perhaps ® Originally a vote of the people was needed, but the senate had become competent already by the second century B.C.; Mommsen, Stk 1.643.

1 Above, 33. 2 Above, 30. 3 Below, 56.

4 For the meanings of provincia see below, 66. 5 Thus in 108 B.C. Marius, who had been elected consul for 107, was given the province of Numidia and hence command of the war against Jugurtha by a plebiscitum, although

the incumbent commander had already been prorogued in command by the senate. Similar methods brought Pompey and Caesar their great commands, in face of strong senatorial opposition. 35

The republican constitution in order to minimise this control that Gaius Gracchus in 123 B.C. passed his

lex Sempronia de provinciis consularibus,® which required the senate to designate particular provinces as ‘ consular’ in advance of the elections at which the consuls in question would be appointed. But it must often have been not too difficult to guess in advance who was likely to win election.’ In the pre-Sullan period the consuls regularly proceeded to military commands or governorships abroad during their year of office; but after Sulla they almost always remained at Rome, proceeding to their provinces only at the end of their year as ‘ proconsuls ’.* Thus the separation of provincial governorships and military commands from the civil magistracies of Rome hardened into a settled practice, and as such was inherited by the principate. Under the heading of functions in connexion with the magistracy may also be put the power, assumed by the senate during the revolutionary period of the late republic, of passing a resolution known as senatus consultum ulti-

mum, the force of which was, at any rate according to the Optimates (the ‘ Conservative’ party), to arm the magistrates with extraordinary powers, including, so it was alleged, that of putting citizens to death without appeal to the people. In form the resolution was only advice to the consuls (sometimes with the addition of other magistrates) to ‘ see that the commonwealth did not suffer ’, but its object was to proclaim something like martial law in times of crisis, and it was in part a device intended to fill the gap in the constitution made by the disuse of the dictatorship, the original constitutional provision for creating a strong executive in times of emergency. Though the Romans would probably have agreed generally that if there was actual armed force used against the state it was permissible for the consuls to go outside

strict law in meeting it, and that the senate as the great Council of State might point out their duty to the consuls, the fact was that the senatus consultum ultimum was only used for party purposes and during the period at which the senate really represented only one party (the Optimates); its legality was consequently never recognised by the democratic party. In particular the execution of the Catilinarian conspirators without trial by Cicero when consul in 63 B.C., although the senate had passed the consultum ulti-

mum, was generally regarded as illegal and was the chief reason for his banishment in 58 B.C.® 6 This law remained in force down to 52 B.C., when it was superseded by the shortlived lex Pompeia.

7 Thus in 60 B.C., foreseeing that Caesar was certain to be elected to one of the consulships of 59, his leading opponents in the senate secured that the consuls of 59 should be assigned unimportant provinces (provinciae minimi negotii: Suet. Jul. 19). Caesar used his political domination of the assemblies to have the lex Vatinia passed giving him the governorship of Gaul for five years. 8 The older view, that they invariably remained at Rome, and that this was the consequence of a specific enactment of Sulla, has been disproved by Balsdon (JRS 29 (1939) 57ff., an important discussion of the whole nature of imperiwm in the late republic).

9 See Last, JRS 33 (1943) 93ff. 36

The senate The two spheres in which the influence of the senate made itself most strongly and most consistently felt were those of finance and foreign affairs.

Here, as throughout, the action of the senate is in form advisory, but nowhere else does the form wear so thin and the fact emerge so clearly that it is the senate itself, not the magistrates whom it advises, with whom the real

decision lies. |

(iii) Financial functions. To understand republican finance we must first rid ourselves of the modern preconception that the greatest part of a state’s income must necessarily come from taxation. At Rome the principle was rather that the state should have enough income to meet all ordinary charges without taxation.1 Nor again must we suppose that all expenditure of a public character is necessarily met by payment out of the public treasury. The Romans so far as possible were accustomed to provide for ordinary recurring expenditure by appropriating specific sources of revenue. The expenses of religion were thus largely met by appropriating the rents of certain state

lands once and for all to the service of particular temples. Even military expenditure might be dealt with in a similar way. Until a comparatively late

period the payment of money to which cavalrymen had a right for their horse and its fodder was a burden on women and orphan children who owned

property but could not be rated in the census because the census list contained only the names of those capable of bearing arms. That the money pay-

able in such cases did not go through the state treasury, but direct, is clear from Gaius, who says that if the soldier was not paid he could seize a piece of property from the person liable in order to enforce payment.” Even the

payment of the infantry appears originally to have been a matter for the tribes, if indeed they were paid at all, but was early taken over by the state (406 B.C.).

This system of permanent appropriation was one of the chief causes of senatorial control over finance; expenditure not covered by it was necessarily a non-routine matter and so, on the general constitutional principle, needed the senate’s sanction.’

The main ordinary source of income was the revenue from public lands. To this must be added some indirect taxes, especially the portoria, export and import duties, and a certain amount of fines inflicted for various offences.

But Roman indirect taxes differed from ours in that they were not reconsidered annually and so variable in amount from year to year, but laid down

once and for all. In the later republic, when Rome had already extended her conquests widely, there was an additional source of income in the taxation of the provinces, but here too the taxes were levied under permanent schemes and according to the final theory represented rent for the land which, 1 As the king of England was at one time expected to ‘ live of his own’.

2 Gal. 4.27. How it was decided which person was to pay which soldier we do not

know. Cf. below, 190. 3 Above, 30. 37

The republican constitution although left to its original owners, was held to have become the property of the Roman state. Where the ordinary revenue was not sufficient for state expenses, as generally happened during the wars of the earlier republic, it was necessary to have recourse to taxation. The tax, known as tributum, consisted of a proportion of the citizen’s property as assessed for the census list (which was also the basis of arrangement in classes for purposes of the centuriate assembly), and it was for the consuls to say when it should be levied and what the proportion was to be,® but here the consuls probably never acted without senatorial authorisation. Strictly the tributwm was not so much a tax as a forced loan exacted to meet an emergency and repayable if circumstances permitted. There are indeed instances of repayment after a successful campaign, but such cases were rare, and tributum was, in fact, almost always levied, until the victory of Pydna in 168 B.C.,°® by providing the treasury with enormous booty, made it unnecessary. Direct taxation was never again imposed on citizens during the republic.’

As to expenditure, the chief extraordinary outgoings besides those necessary for war were for public works, roads, buildings, etc. The civil service was little developed; the higher officials were all unpaid (though opportunities for making money were not always lacking) and though they had paid assistants this did not form a large item. Most of the innumerable burdens which a modern state takes on itself (e.g. education, relief of poverty, supervision of the conditions of employment) were lacking, a fact which to some extent explains the simplicity of Roman financial arrangements when compared with our own. In strict theory of law the state treasury (aerarium) was under the control of the chief magistrates, the consuls, and they alone were entitled to take money from it; but this theory does not correspond with the facts, for the senate kept a firm hold on finance, and the main financial officers were the censors and the quaestors. The duties of these officers were however quite different and must be dealt with separately. Censors.® These magistrates are sometimes said to have been the ‘ budget-

makers ’ of Rome, but, from our point of view, it was a very imperfect budget that they made. They had to compile the census-list and for this purpose they appear to have had a considerable discretion in the valuation of property. We hear, for instance, that Cato in 184 B.C. assessed articles of luxury at ten times their real value, which meant of course that their owners

would have to pay an increased amount of tribute. It was the censors too who made the contracts necessary for getting in the public revenue; they 4 Below, 267.

5 The normal amount was one thousandth part (tributum simplex), but we hear also of tributum duplex and triples. 6 Above, 2. 7 Exeept perhaps in 43 B.C. 8 Cf. below, 51ff. 38

The senate could for instance lease public land or such sources of wealth as mines or fisheries belonging to the state at a rental, and it was they who made the contracts with the publicani, the tax-farmers. Taxes such as the portoria, or the vectigal (payment for public land ‘ occupied ’ by squatters),°® and the various forms of taxation in the provinces, were not, in the republic, collected directly by state officials, but farmed out to speculators who paid or promised a lump sum for the right to collect the taxes, expecting, of course, to make a profit by collecting more than they had to pay. The system resulted in considerable oppression, as the publicani were not properly controlled, and vast profits were often made.’ Sometimes however they overreached themselves, either when the competition in bidding between different companies of publicani was particularly keen, or when a harvest turned out badly,? and were faced

with bankruptcy if held to their bid. In such cases only the senate could release them from their contract, and we know of several instances in which this was done.? Besides arranging for the collection of public revenue the censors were also

to some extent spending officers; they, for instance, often arranged contracts for public buildings, roads, etc., in which the state was the paying, not the receiving, party. In these matters, however, they had less discretion, being only allowed to draw from the aeraritwm up to a definite amount placed to their credit by order of the consuls and senate. They were thus in this respect

in a position similar to that of other magistrates, who had similar credits opened for them for their expenses, for instance in the payment of their subordinates. The registers of the censors (tabulae), showing as they did a large proportion of the income which the state might expect, were the nearest thing which the Romans had to a budget and were no doubt largely used in estimating the amount of money which might be spent, but they were necessarily incomplete, for censors were only elected every five years and held office for not longer than eighteen months,‘ and in the intervals business which could not

wait was transacted by the other magistrates. Also extraordinary income, such as that derived from booty, would find no place in the censorial accounts.

he quaestors.” The aerartum itself was under the supervision of the quaes-

tors. Originally there was but this one treasury administered by the two 9 Above, 10.

1 The contracts were commonly not given to individuals but to companies, and it was possible to take a share in such a company by investing capital, without doing any work, in much the same way as one takes a share in a joint-stock company to-day. 2 Many taxes being calculated by taking a percentage of the crops. 8 One famous case occurred in 59 B.C. when the question whether the publicani of Asia should be released or not was one of the most important issues between the political parties; Balsdon, JRS 52 (1962) 135ff.

4 Five years were thus the normal term for censorial contracts. 5 Cf. below, 50f. 39 J.——-d

The republican constitution quaestors as assistants to the consuls, the consuls having control of the treasury as of all other departments of state. In 421,° the number of quaestors was raised to four. Two (quaestores urbant) were to remain in charge of the aerartum at Rome and each consul was to have one of the others as a separate assistant when he took the field. A separate treasury was then established for him and similar separate treasuries were established later for provincial governors. These, though under the control of the governor, remained in relation with the central treasury at Rome, the provincial quaestors being given money or credits on the central treasury for the expenses of government and being bound to render account on their return. Meanwhile the governor had, however, full discretion as to the use to which he would put moneys which flowed into his treasury. By no means all moneys went through the central treasury. Besides their duty of keeping the accounts of the aerarium, the two quaestores urbani had other duties, such as sale of property (e.g. slaves and booty) which fell into the hands of the state. They also sometimes made financial statements in the senate. The key to the understanding of Roman finance lies in the principle that every holder of imperium was bound to delegate the keeping of the account

of moneys which came under his control. For the consuls and most promagistrates these delegates were provided in the elected quaestors and exquaestors, who, like their superiors, usually went to a province after their year of office; but where there was no such delegate provided, it was the duty of the holder of imperium to choose one himself. The object of this rule was clearly that there should be a record of every payment made by order of a magistrate with imperium which would make it possible to call him to account for moneys spent improperly, without limiting his discretion in the spending of them. The urban quaestors had a more independent position than any others because their superiors (the consuls) were frequently away from Rome. This meant that in fact they laid out moneys only on the authorisation of the senate. They were entitled to pay on the command of the consuls alone, but in the later republic even the consuls never gave such orders without previously consulting the senate. We thus get the position that the quaestors are the people who hold the keys of the treasury; they pay out moneys at a consul’s order either to him or to other magistrates, but owing to the convention that the consul must consult the senate, it is the senate which has the real control.’ (iv) Functions in connexion with foreign affairs. That the declaration of war was a function reserved for the people has already been mentioned,‘ 6 When the office was opened to plebeians; above, 15.

7 But in the late republic the supervision of public accounts was often only sketchy ;

ef. Cic. Verr. 11.1.36f.; Plut. Cato Minor 16. &§ Above, 19. 40

The senate though in this most vital matter the magistrate was of course bound in practice to consult the senate before bringing a proposal before the assembly. It is not so easy to answer the question where, according to the Roman constitution, lay the power of making peace, or, what is in effect the same,° the

treaty-making power. It is likely enough that the power of binding the Roman state by treaty was originally regarded as inherent in the imperium of the magistrate as the representative of the commonwealth, but in historical times this was no longer true.

Treaties cannot be deemed binding unless the magistrate’s act is either authorised beforehand by the community or subsequently ratified. On more than one occasion a deluded enemy learnt, to his cost, that the Romans would refuse to acknowledge the most solemn oath taken on their behalf by a magistrate, and consider their consciences satisfied by delivering him and all who had taken part in the ceremony naked and bound to the people with whom they had contracted. For purposes of the authorisation or ratification of treaties the community meant the ‘ Senate and People of Rome’, the style of the Roman state used particularly for international relationships, in which, be it noticed, the senate

comes first. In fact the decision lay from early republican times with the senate, and the ratification of the senate’s decision by the people was a formality. Towards the end of the republic it was occasionally omitted, though Cicero could still argue that it was legally necessary.” The treaty once authorised was usually concluded by sending a mission consisting of two or three Fetiales, members of the priestly college specially charged with international ritual, who took an oath on behalf of the senate and people.

The concentration of all international matters in the hands of the senate was the natural outcome of its permanence when contrasted with the everchanging individual magistrates, and one result of this concentration was that a general in the field, once peace was in sight, directed the ambassadors

of the other state straight to the senate at Rome before which all but the preliminary negotiations took place. Also, after an enemy had been so com® For, in theory, any state not bound to Rome by treaty is an enemy.

1 The most famous instance is that of the ‘Caudine Forks’ in 321. The Roman army under the consuls T. Veturius Calvinus and Sp. Postumius was trapped in a defile by the Samnites. As escape was impossible, the consuls concluded a treaty by which they agreed to make peace if the lives of the Romans were spared. This was done, and the army, after being stripped of its arms and made to pass under a yoke (the symbol of defeat), returned home. The senate, however, on the advice of Postumius himself, repudiated the agreement, the argument being that the people could not be bound by a treaty made without their consent and that all that law demanded was the delivery of the actual contracting parties to the enemy. Cf. Lévy-Bruhl, Nouv. Ht. 116ff. (=RHAD, 1939, 533ff.) ; Kaser, AJ 260ff.; La Rosa, Jura 1 (1950) 283. A later example is the repudiation of the Numantine treaty negotiated by Tiberius Gracchus on behalf of the consul C. Hostilius

Mancinus in 137 B.C. 2 Pro Balbo 34. 41

The republican constitution pletely conquered that there was no question of a treaty, but only of complete surrender (deditio), it was the senate which sent out commissioners to assist the commander in settling the organisation of the conquered territory. This organisation was generally laid down in a lew, a sort of charter, which however did not receive the sanction of the comitia. During the period when provinces were being added to the Roman empire the senate’s supremacy was so clearly recognised that its authority was sufficient.® In speaking of the ‘ foreign affairs ’ of Rome one must of course remember

that in the later republic Rome was already supreme in the Mediterranean and indeed over most of the known world, so that there was no international intercourse such as exists to-day between independent states, the only power which could at all measure itself with Rome being the distant kingdom of

Parthia. But it must also be remembered that the vast majority of the inhabitants of the lands ruled by Rome were not citizens, but foreigners (peregrint), so that the government of the provinces outside Italy was also in a sense part of ‘ foreign affairs ’. Here too the senate, as the permanent government, exercised what supervision was exercised at all over the provincial governors. Permanent representation either of subject communities or of independent powers at Rome, such as exists in our modern embassies, was unknown, all business being transacted by delegates appointed for the particular occasion. Roman history records many instances of these and of the

impression made on the delegates by the assembled senate, a body which must have seemed to many of them, as it certainly did to one, * an assembly of kings ’.*

(v) Functions in connexion with religion. As the great Council of State, the

senate was, of necessity, often concerned with matters affecting the state religion, and it is difficult to say that any part of Roman religion was quite independent of the senate. Not only was the first sitting of each year devoted

to the presentation of a report by the consuls on sacred matters,® but at every meeting religious affairs took precedence of all others. To some extent the most important sacred ‘ colleges ’, such as the pontiffs

and augurs, whose members would generally, though by no means necessarily, be leading members of the senate, acted as standing committees for matters falling within their province; but these bodies possessed little initiative, and senatorial authority was needed to carry out their proposals. Thus the senate might order ceremonies of purification to be undertaken if the 8 In 133 B.C, Tiberius Gracchus did challenge the senate’s prerogative in the matter of the newly acquired province of Asia; and ten years later his brother Gaius regulated the

fiscal organisation of Asia by a plebiscitum. ae

4 These words were used by Cineas, ambassador of Pyrrhus, king of Epirus, when in spite of Pyrrhus’s victory at Heraclea in 280 B.C. the senate refused to discuss terms so

long as foreign troops remained on Italian soil. 5 Above, 34 n 7. 42

The senate priestly college announced the appearance of some ‘ prodigy’ which was taken as indicating divine wrath; it might also order the consultation of the Sibylline books,* a matter sometimes of very great political importance. The senate’s approval was of course necessary, on general principles, if money was to be spent on any religious ceremonies or festivals out of the ordinary course of events, and it could decree extraordinary festivals, a power which was sometimes used for the purpose of obstructing undesirable public activities. The reception of a new deity into the public worship of Rome, though

strictly perhaps a matter for the people, was generally decreed by senate, sometimes as a result of consulting the Sibylline books.

C. ProcepureE. To understand the procedure of the senate it 1s necessary to remember throughout that the senate is a council whose business it is to give advice to the magistrate, and that it has no formal executive authority of its

own. This appears first from the fact that the senate can only meet when summoned by a magistrate, originally only a magistrate of the populus with imperium,’ who consults it previously to acting himself. Later, however, owing probably to the growth of the legislative power of the concilium plebis, and the consequent necessity that its presidents should be able to consult the

senate before proposing plebiscita, the tribunes obtained the right of convening the magistrates’ council.* This multiplication of possible conveners of the senate is important, for it meant that any considerable section of opinion could normally secure the holding of a meeting in order to express its views even if all the magistrates with imperium were desirous of preventing such expression. In this way, as well as by the use of their veto in accordance with the senate’s wishes, the tribunes became, in the later republic, an important

instrument of senatorial rule, to such an extent that they could be termed mancipia nobilum.

As the senate could only be summoned by a magistrate, so it could only discuss business put before it by a magistrate; there were no ‘ private members’ bills *, though as we shall see it was possible for the motion on which the House voted to be framed by a non-magisterial member. The normal conveners were the consuls, but once the senate was convened it was not the convener alone who could put business before the House. The ¢ A collection of oracular utterances in Greek, said to have been bought by the last king of Rome, Tarquinius Superbus, from the Sibyl at Cumae. They were under the custody of a special college, X viri, later XV viri sacris faciundis (above, 16). 7 Dictators, consuls, praetors, interrex and necessarily the military tribunes with consular power when these existed, above, 14f. 8 The date cannot be determined, and the tribunes evidently did not make much use of

the right (as opposed to the right to put business before the senate once it had been convened by the consul or the urban praetor) ; O’Brien Moore, PW Suppl. 6.700. 43

The republican constitution consul’s business came first,® then the praetor’s and then the tribune’s. The proceedings on each question began with a statement by the magistrate who brought forward the business, and it was also permissible for any magistrate

to speak, without invitation, at any moment during the debate. After the opening statement the House might proceed immediately to a vote, but more usually a debate followed. In the form of debate, however, the theory of consultation by the magistrate appears again clearly. [It was not for any member wishing to speak to ‘ catch the Speaker’s eye ’, it was for the presiding magistrate to call upon the member to deliver his opinion (e.g. Dic, Marce Tull, quid censes?). In this duty, however, the magistrate was more or less constrained by custom to follow a prescribed order of seniority. When the cen-

sorship was still a living force, it had been customary for the censors to choose some particularly distinguished man to head the list of the senate and he, as princeps senatus, would always be asked first. After Sulla’s reforms had reduced the importance of the censorship this was no longer done, and the president could choose any person of consular rank (i.e. an ex-consul) to speak first. After the consulars came the praetorians, then the ex-aediles, the ex-tribunes and finally the ex-quaestors.! Those who had not held curule office would thus seldom have the chance of speaking, but they could make their opinion clear by moving and taking a seat near a speaker with whom they agreed. Those who thus expressed their opinion ‘ with their feet ’ were called pedarii.2 Once invited to speak, a senator could express his opinion on anything he considered of public importance, regardless of its relevance to

the business before the House.* Nor was there any time limit to speeches except that the sitting had to be suspended at sunset, and obstruction by filibustering was by no means unknown. When the debate was over it was for the presiding magistrate to choose the order in which the opinions expressed by members should be put to the vote,

and the voting took place by those in favour moving to one side of the House, those against to the other. If one motion on a subject was passed, the remainder were automatically dropped. Once passed, a motion became a senatus consultum provided no tribune or

other magistrate with the right of intercessio vetoed it. It was open to veto because, in strict theory of law, it was only advice by the senate to the magis9 ‘Business ’, not ‘motion’, for, strictly, the magistrate asks for advice and does not propose anything himself. In fact of course his statement might contain a proposal. 1 Normally, for obvious reasons, the consules designati were asked for their opinion before the consulars. 2 Gell. 3.18; Festus s.h.v. (Bruns 2.23); Monro, Journal of Philology 4 (1872) 113ff.; O’Brien Moore PW Suppl. 6.680; and see Jura 18 (1967) 248ff. 3 This was known as egredi relationem. The most famous case is that of Cato (censor 184 B.C.), who tacked on to every speech the remark ‘ Carthage should be destroyed ’, a piece of advice which the Romans finally took. 4,4

| The magistracies trate who had elicited it and took its binding force from his authority, and an act of a magistrate was open to veto by a tribune or colleague of equal or higher rank. If vetoed, the motion which the senate had passed became not a senatus consultum, but merely an auctoritas, an expression of opinion

which bound nobody but might command the respect of magistrates or others concerned. All resolutions were committed to writing under the supervision of a small committee chosen for each case by the presiding magistrate, and were deposited in the aerartum under the supervision of the quaestors. To prevent forgery the plebeian aediles were, during the time of the struggle between the orders, given a share in the control — we hear of the deposit of senatus consulta under their care in the temple of Ceres — but the exact way

in which responsibility was shared is unknown. , As to publication, there was no general system by which the senate’s resolutions were made public; sometimes they were announced in a contio, some-

times put up in a public place to be read, and sometimes copies might be sent to interested parties, such as provincial governors or the ambassadors of allied states.

6. THE MAGISTRACIES A. Tue Consu.LatE. We have seen that on the downfall of the monarchy, according to the traditional account, the kingly power was limited by being transferred to two annually elected officers, known in historic times as consuls,* and that the struggle between the orders turned largely on the question of the eligibility of plebeians to this office, which continued throughout the republic to be the highest in the state. Like the royal power of which it was the successor, the consular imperium was, strictly, unlimited, extending

to all departments of government, and including leadership of the army, jurisdiction and the right of putting business before the assembly as well as, of course, before the senate, the advisory body originally of the kings, now of the consuls. The only sphere in which the consuls did not succeed fully to the kingly power appears to have been that of religion. The position of head of the priestly college of pontifices passed to a separate official, the pontifex maximus, and for certain minor religious purposes the name of ‘ king’ was retained by a priest known as the rex sacrorum. The election of the consuls took place in the comitia centuriata,® which was summoned for this purpose normally by a consul, exceptionally by a dictator, an interrex or a military tribune with consular power. For a praetor to hold the consular election was quite irregular. The date of the elections, the most important regular political event in the year, which drew numbers of citizens 4 Originally praetores; above, 8 n 1. 5 The formal confirmation of the imperiwm by the comitia curiata (above, 19) took place at the beginning of the year of office. A5

The republican constitution from all parts of Italy to Rome, varied considerably from year to year until settled by Sulla, after whose time they were normally held in July. The consular year however (since a law of 153 B.C.) always began on 1 January, so that there was a considerable interval during which the successful candidates were consules designatt before entering on their office. Once elected, a consul could not legally be removed from office, except perhaps by a dictator, but removal of a consul did occur in a few cases during the revolutionary period. If a consul died or resigned it was for his colleague to hold an election to fill his place for the remainder of the year. If both consuls died, an interrex had to be appointed.°®

The consul’s imperium is in effect the royal power, limited by the principle of collegiality and the annual tenure of office. It is thus a general authority,

military and civil, and undefined except in so far as special limitations are put upon it by specific statutes or by the assignment of special functions to other magistrates. These limitations apply however almost exclusively within the city or the first milestone therefrom (domi); outside (militiae) the holder

of imperium is untrammelled. The most important restriction based on statute was that concerning provocatio.’ The right of choosing the senate given to the censors,® and the jurisdiction in civil matters assigned to the praetor urbanus are examples of powers originally belonging to the consuls and subsequently transferred to other magistrates. But apart from these legal restrictions, and much more important, is the all-pervading power of the senate of which the consuls came in effect to be merely the chief executive officers. Indeed, in Cicero’s day the consuls are during their year of office no more than the chief executive officers in Rome and Italy. Previously, although special praetors or ex-magistrates had been appointed as occasion required to be provincial governors, the consuls themselves had frequently commanded armies of the republic outside as well as within Italy, but in the post-Sullan period governorships of the provinces, and therefore the command

of the troops stationed there, were regularly held by magistrates (consuls and praetors) whose year of office had come to an end, with the result that during their year of office the consuls and praetors did not leave Italy. The importance of this politically was that within Italy the civil and the military powers were separated; for Italy was regarded as permanently pacified (all its inhabitants were now citizens), and it became a constitutional principle that there should be no troops stationed within its boundaries, save in the event of a tumultus Italicus or a slave uprising. The inconvenience of the principle of colleagueship, which meant that there were two officials whose functions and competence were precisely the same,

was surmounted by a number of different expedients. The ultimate resort

7 Below, 306ff. 8 Above, 32.

6 Above, 30.

46

The magistracies was intercessio, 1.e. either consul could veto any act of his colleague, so that matters remained as they were. But of course government would have been impossible if it had been necessary to have recourse to this expedient often.

To some extent the consuls divided their spheres of action by agreement; more commonly they acted together. They could, for instance, convene the senate together, or the assembly, and that they did so is witnessed by a number of laws bearing the names of consular colleagues. Where the business was such that only one man could do it, the matter might be arranged either by agreement or by lot, but there was also an old practice by which the consuls took turns of such duty. At home they changed every month, the older

magistrate, or the one first elected (consul prior), generally taking the first turn; only the consul whose turn it was had the fasces borne before him. In the field the more questionable practice of taking daily turns of supreme command prevailed.

The consuls had in principle a general jurisdiction,® but in fact, when a special jurisdictional magistracy, the praetorship, had been established, the consuls seem no longer in practice to have had any concern with civil litigation.*

Of outward signs of rank the consuls had many. The magisterial toga praetexta (purple-edged toga) and the curule chair they shared with some other magistrates, but they alone were preceded by the full number of twelve lictors, bearing the fasces, or bundles of rods, the symbol of coercive authority, to which, as soon as the consul quitted the precincts of the city, the axe

was added to show that he had exchanged the limited authority which he exercised domi for the unrestricted powers of the magistrate with imperium militiae. Most notable privilege of all was that by which the names of the 9 The relationship between iurisdictio and imperiwm is a source of difficulty; see Pugliese, Proc. 1.162ff.; Kaser, ZPR 29f., 133f. They can hardly be treated as wholly distinct (as by Lauria, St. Bonfante 2.481ff.) in the face of texts in which the two powers are clearly linked (e.g. D. 1.21.3; 5.1.58), or of the rule, for example, that a magistrate has no jurisdiction over another of equal imperiwm (D. 36.1.13.4), or again of the category of iudicia imperio continentia (below, 223). But it is equally difficult to treat them as merely two aspects of one idea, since although every magistrate with imperium has iurisdictio, at least in principle, there are some magistrates who have turtsdictio but no imperiwm. These include the curule aediles (for a suggestion that they had imperium see Sargenti, BIDR 69, 1966, 201ff.; G. Impallomeni, L’editto degli edili curuli, Padua, 1955, 118f.), the duoviri and quattuorviri, the quaestors in senatorial provinces, and the censors (if indeed they were regarded as having turisdictio, see Sargenti, loc. cit.). Such magistrates were unable to perform acts which were magis imperii quam turisdictionis, such as ordering missio in possessionem or in integrum restitutio, or exacting praetorian stipulations (D. 2.1.4; 50.1.26) ; it is not clear whether interdicts fell into this category. It is probably best to regard iurisdictio as a part of imperiwm which can nevertheless be attributed independently; cf. F. de Martino, La giurisdizione nel dir. rom. (Padua, 1937).

1 Though transfers of property by in ture cessio and manumissions vindicta, which were actions in form, could take place before them; and they could of course veto a praetor’s act. As to criminal jurisdiction see below, 305ff.

AT

The republican constitution consuls formed the official designation in the calendar of the year in which they had held office.

B. Tue Prartorsuip. The events which led up to the creation of the first praetorship in 367 B.C. have already been described.” The new magistrate was to take over the duties hitherto performed by the consuls with respect to civil jurisdiction. This, it may as well be noted at once, does not mean that the praetor was a judge in our sense of the word. A Roman civil trial, until well on into the empire, always took place in two stages, the first of which alone was the concern of the jurisdictional magistrate. This was the stage in ture, in which only the preliminaries and especially the issue * between the parties was settled. It was in the second stage, apud iudicem,* that the actual trial took place and the issue raised was decided, the judge (tudex) in this stage being, not a magistrate, but a private person appointed for that purpose. It is this division of functions between the magistrate and the iudex which alone makes it possible to understand how the Romans were able to manage

for so long with a single jurisdictional magistrate, for it was not until about 242 B.C.*° that a second praetor was appointed and a division of duties made, one praetor superintending the jurisdiction between citizens (praetor urbanus) and the other that between foreigners or between citizens and foreigners and hence known as praetor qui inter peregrinos ius dicit, or more shortly praetor peregrinus (‘ foreign ’ praetor).° The number of praetors did not remain at two for long, but the subsequent creations were for a different purpose. The first provision which Rome made for the government of provinces which she acquired beyond the seas was by the appointment of new praetors, two being added about 227 B.C. for Sicily

and Sardinia, acquired by Rome as a result of the first Punic war, a further two in 197 B.C. for the two new Spanish provinces. At this number of six

the praetorship remained until Sulla, for although new provinces were created they were governed more and more under the new principle of pro-

magistracies.” |

The position of the praetor is best understood by thinking of him, in the

2 Above, 16. 3 Settling the issue means settling exactly what is the question, whether of law or fact, between the parties which the trial is to decide.

4 Below, 185. ,

5 Mommsen, StR 2.196 n 2; Jérs-Kunkel 59 n 2.

6 Daube, JRS 41 (1951) 66ff., observing that this appears to be the original title and that the form qui inter cives et peregrinos ius dicit is not found until the beginning of the principate, argues that until then it was the urban praetor who dealt with cases between citizens and foreigners; contra, F. Serrao, La ‘ iurtsdictio’ del pretore peregrino (Milan, 1954) 10ff., 139ff. Certainly either could delegate his jurisdiction to the other,

Livy 24.44.2. See also Bonifacio, Iura 6 (1955) 242ff. 7 Below, 67. 48

ees The magistracies

Roman phrase, as a minor colleague of the consul. He, like the consul, is elected by the comitia centuriata and he has full impertum, which means that he is in principle capable of all the duties which the consul performs, military and administrative as well as judicial. But he is subordinate to the consul; the consul’s intercessio prevails against his act, whereas he cannot veto an act of the consul, and his inferiority is made outwardly manifest by the lesser number of lictors, six, by whom he has a right to be accompanied.

Above all he differs from the consul in that he is always given a definite sphere (provincia) in which to exercise his imperium, while that of the consul is not so limited. But, although each praetor has a provincia, he is not elected to it in the first instance by the people. They only elect six praetors and the

provinciae are afterwards distributed among the successful candidates by lot. The normal provinciae are the two jurisdictional spheres of the city and the four oversea governorships, but it was possible for this arrangement to be

varied at times; two spheres, e.g. those of the praetor urbanus and praetor peregrinus, might be combined, in order to leave a magistrate free for some particular duty outside the usual six spheres; or the same purpose might be achieved by continuing the existing governor of a province for a year beyond his original term of office. The decision in such matters rested with the senate. After Sulla’s reconstruction of the constitution the position was somewhat different. There were eight praetors elected annually, and they all remained at Rome during their year of office, two, as before, taking the civil Jurisdiction and the remaining six being used as presidents of the quaestiones perpetuae, or standing criminal courts which had been made a permanent fea-

ture of Roman public life by Sulla.* Only after the year did the praetors, like the consuls, proceed to governorships of provinces abroad. Though primarily confined to his allotted provincia, each praetor, being endowed with imperium, could also act outside it if necessary. Thus we find

the praetors summoning the senate or the people or undertaking the levy, but only (in normal cases) in the absence of the consuls. In such cases it was

the praetor urbanus who was expected to take the lead, his office ranking highest. He would also almost necessarily be present in Rome, for we hear that he was forbidden by statute to leave the city for more than ten days during his year of office.? The special functions of the urban and peregrine praetors with respect to jurisdiction, especially their edicts, which were of the very greatest importance in the development of the law, must be left for separate discussion.*

C. Tue Aepitesuip. The aediles were originally purely plebeian officers, assis-

tants of the tribunes, as the quaestors were assistants of the consuls, and were elected, like the tribunes themselves, in the concilium plebis. A complete

8 Below, 318ff. 9 Cie. Phil. 2.31. 1 Below, 97ff. 49

The republican constitution change, however, came over the office as a result of the creation in 367 of two new officers with the same title who were true magistrates of the whole people and as such were elected by the comitia tributa. Henceforward the connexion with the tribunate is forgotten ? and the four aediles act largely together as a single ‘ college’. Some differences between the plebeian and ‘curule’ aediles remain; the former must always be plebeians, the latter are elected in alternate years from the two orders;.the curule aediles have not only the curule chair from which they derive their name but also the magisterial praetexta, whereas the plebeian aediles, like the tribunes, have no outward sign of their office, and are in general considered to hold a lower rank. The duties of the office, though rather heterogeneous, can be described as in the main municipal. The aediles were responsible for the cura urbis, which

included the supervision of cleanliness in the city, the repair of roadways, public places and public buildings, and the water supply. They were also in charge of the corn supply and of some of the public games, and had some share in the control of the state archives.* In connexion with their general cura urbis they exercised certain police functions and could inflict fines and take other enforcement measures, subject to the need for a comitial trial if the fines were above a certain maximum.‘ Their most interesting function from the point of view of legal history is the control of the marketplace, for in connexion with it the curule aediles ° exercised a certain amount of civil jurisdiction for the purpose of which they issued edicts corresponding to those of the praetors;* these edicts were of great importance in the development of the law of sale.’ D. Tue Quasestorsuip.* The quaestors were perhaps originally assistants of the consuls and, as such, chosen by the consuls themselves. In historical times, however, they are elected in the comitia tributa and count as magistrates of the Roman people. The number, originally ° two, was raised in 421 2 So much so that the aedileship ranks higher than the tribunate. 8 Above, 45. 4 Below, 307; the power to convoke a comitial trial is an unexplained anomaly, since the aediles had no imperiwm (but see above, 47 n 9) and could not summon the assembly for any other purpose. Perhaps they ‘ borrowed the auspices ’ from a praetor. 5 This jurisdiction was not shared by their plebeian colleagues, probably because at the time when it was instituted the plebeian aediles had not yet acquired the character of magistrates of the people; Mommsen, St# 2.501. Even the curule aediles had no émperium; for the consequences see above, 47 n 9; see also Kaser, Mél. Meylan 1.173ff. 6 For some remedies (e.g. actto de peculio) they made use of the praetorian edict; Kaser, op. cit.

7 Below, 293f. 8 Wesener, PW 24.1.801ff.

9 When they originated is uncertain. Our sources, in so far as they refer to the matter, attribute their creation to one or other of the kings (Ulpian, citing Junius Gracchanus, D. 1.13.1 pr.; Tac. Ann, 11.22; Plut. Popl. 12), but Latte has argued (TAPA 67, 1936, 24ff.; PW Suppl. 7.1610f.) that they date from 447 B.C. and were from the first inde50

The magistracies to four,’ and probably about 267 another four were appointed to assist the consuls in the administration of Italy. Sulla, as we have seen, raised the number to twenty.’ The two urban quaestors (quaestores aerarii) were, essentially, independent magistrates. Their financial functions, which were the most important, have

already been dealt with,® and their part in the criminal trial before the comitia will be considered below.* Those quaestors, on the other hand, who were attached to magistrates with zmperium outside Rome were in a more dependent position. Though their chief duties were financial, they were used by their superiors as delegates in all types of duty, military, administrative and even judicial.® The relationship between the higher magistrate and his quaestor was considered to be a personal one, and when the superior’s office was prolonged, that of his subordinate was regularly prolonged also. Those quaestors, however, who were concerned with the administration of Italy, though strictly subordinate to the consuls, had in fact definite posts allotted to them which they administered with considerable independence. Not all these positions are known, but there was one at Ostia, concerned with the corn supply, and another appears to have been concerned with the department of § woods and forests ’ (calles).

K.. Tue Censorsuip.® We have already spoken of the institution of the censor-

ship (in 443 B.C.) as an incident in the struggle between the orders,’ which constituted a patrician victory in so far as the new office was confined origin-

ally to patricians. When the office became open to plebeians is uncertain, but we hear of a plebeian censor for the first time in 351 B.C. and, soon after-

wards, the same rule applied as to the consulship, that one of the two colleagues must be a plebeian. The original purpose of the new magistracy was clearly to relieve the supreme magistrates of one of the duties that they had previously performed,

that of taking the census, and at first the censorship appears to have been pendent magistrates concerned with finance. Tacitus’ statement, loc. cit., that they were first elected in 447 is therefore correct, but his further statement that they were earlier nominated (which is the chief basis for Mommsen’s view that they were originally assistants of the consuls) must be a false inference. In Latte’s view, which is accepted and elaborated by Kunkel, Krim. 37ff., the quaestores parricidii, who are said by Pomponius, D. 1.2.2.28, to have been mentioned in the XII Tables and to have presided over res capitales (cf. Festus, s.v. parricidit quaestores, Bruns 2.21) were quite distinct. Mommsen, StR 2.536ff., had identified them with the quaestores aerarti and had accounted for the incongruous combination of duties by his assumption that they were assistants of the consuls, See further, below, 310, 316 n 6.

1 Above, 40. 2 Above, 33. 3 Above, 39f. 4 Below, 306ff.

5 Gaius, 1.6, says they had the jurisdiction of the curule aediles.

6 F. Cancelli, Studi swi censores (Milan, 1957); J. Suolahti, The Roman Censors

(Helsinki, 1963) is mainly prosopographical. 7 Above, 15. 51

The republican constitution (Sn eenr NEE INE TINT TN TT Ta eT comparatively unimportant. But the control which it gave over the whole moral life of the people, and especially the power of appointing senators,* made it, in the end, an office of even greater dignity than the consulship itself, and it was, with hardly an exception, held only by men of consular rank. The election was held in the comitia centuriata under the presidency of a consul, but the censor had no imperium,® and, though he sat on the curule chair, he did not have the fasces borne before him. On the other hand the eminence of his office became apparent when he died, for he alone of all magistrates was buried in the full purple toga which had been the emblem of royalty. The censors also differed from other magistrates in that they were not elected annually; they were only elected for the purpose of taking the census, a duty which ended with the lustrum, or ceremonial purification of the people, and they had to lay down their office when this was completed, or at the latest eighteen months after their election. Then no others were appointed until a new census was to be taken, the usual interval being four or, later, five years. Re-election was forbidden. The census was far from being a mere numbering; it was the registration of the whole people in their divisions of tribes and centuries, and its purpose was chiefly military and financial. Incidentally it was also political, for the centuries and the tribes were both voting units. For the purpose of registration the censors summoned the members of the centuriate assembly, that is the arm-bearing citizens, to meet them in the Campus Martius, and each citizen, or at least each head of a family, had to make a return of the members of his family and his property; this latter return was needed partly for the purposes of assessing the tributum, and partly because the divisions in the centuriate assembly depended upon wealth. In addition the censors held a separate review of the eighteen centuries of equites in the forum. The super-

vision over the morals of the community, the famous regimen morum, was an outcome of the census. In the lists which they made up the censors could affix a nota or mark of censure against a man’s name if they disapproved of his conduct, and there was no restriction as to the reasons for which they might express this disapproval. It might be some action in his private life, such as luxurious living, or divorcing a wife without taking the opinion of a family council; it might be the following of a disgraceful trade; it might be the commission of some disgraceful wrong, such as theft, against another person, or again some act in connection with public life, desertion or cowar-

dice in the army, or what the censors considered improper conduct as a magistrate. The censors appear to have published the principles on which they would act in an edict and, no doubt, the views of different individuals 8 Above, 32.

9 Cancelli, Studi sui censores, above (ef. Labeo 6 (1960) 225ff.), and Magdelain, Recherches sur Vimperiwm, argue for a limited imperium.

52

The magistracies

varied, although the general attitude was that the censors were there to uphold the ancient simplicity and sternness of Roman manners, The only safeguards against abuse of these great powers were that the reason for the nota must always be given and that the two colleagues must concur in affixing it. It was also usual to summon the person affected to defend himself, but there was no formal trial.

The effect of the nota was to remove the censured person from the tribe to which he belonged, usually to remove him altogether from the tribes, so that he became an aerarius, or citizen without the right of voting and sub-

ject to a poll-tax, but apparently the result might sometimes only be to degrade the citizen from a ‘ rustic’ to one of the four ‘ urban ’ tribes, which were less honourable, partly because they contained all the freedmen. In the case of equites the censure was expressed by the command § sell your horse ’, which implied dismissal.

The revision of the list of senators was no part of the census, but it was perhaps the most important power of all, and here the censorial nota meant the exclusion of a person otherwise entitled to a seat. In addition to affixing the nota the censors might punish conduct they considered reprehensible by

increasing the amount of tributum a man had to pay, either directly by multiplying the usual thousandth or indirectly by valuing certain property, especially objects of luxury, at a much higher figure than their real value. Of the financial duties of the censors, especially the giving of public contracts, we have already spoken.’ In connexion with these functions they had, no doubt, incidentally to decide disputes, for the Roman principle was that where the state was concerned the magistrate decided, not the courts. But it appears that, in the late republic at any rate, if a dispute occurred between individuals, the censors could appoint a itudex or recuperatores.”

As to criminal matters, the censors did, it seems, sometimes levy fines in connexion with their various duties — the idea that a magistrate should not be able to do so was repugnant to the Roman conceptions of magisterial dignity — but there is no evidence that this was an exercise of any regular jurisdiction. The regular interval between elections to the censorship was, as we have said, five years, but there were many cases in which longer periods were allowed to elapse, and appointment became very irregular after Sulla. After 22 B.C. censors were never again appointed, though the emperors continued the supervision of morals and occasionally took the title of censor. The fact

was that the census had become unwieldy in the late republic when the number of citizens had greatly increased, and, in effect, it was only a compilation taken from individual lists made in the various municipalities. It 1 Above, 38f, 2 Lex agraria, line 35 (Bruns 1.80) ; cf. Cancelli, Studi sui censores (above) 59ff.

53

The republican constitution was also unnecessary either for taxation, as tribute was no longer imposed on citizens,* or for military purposes, now that the old citizen army had been replaced by a professional force voluntarily recruited. The powers of the censorship were also disliked by the senate as a check on the power of the nobility and were no longer necessary to fill up the senate as, under Sulla’s scheme, there were always sufficient ex-quaestors to take the vacant places.

F. Tue Tripunate. The institution of the tribunate as a measure of defence by the plebeians, the raising of the numbers to ten and the chief powers of the office have already been described. By the time that the struggle between the orders was over the tribunes had been for so long part of Roman public life that there was no attempt to abolish them and they continued to exist throughout the republic and well into the empire. Strictly, they remained magistrates not of the whole populus, but of the plebs only; they continued to be elected in the concilium plebis; they must be plebeians, and they bore no outward sign of rank whatever; in effect however they ranked as magis-

trates and the office was one regularly held by men of leading plebeian families * as part of the cursus honorum® or magisterial career. They also obtained the right not only of sitting in but also of convening the senate.° The three powers which gave to the tribunate its greatest political importance were the intercessio by which. they continued to be able to bring the whole business of the state to a standstill, their legislative power as presidents of the concilium plebis and their general power of coercitio. The relationship of this last to their conduct of political trials, especially those of magistrates after the expiration of their term of office, and the connexion of the powers of the tribunes with the right of provocatio are discussed below.’ Although it had become an integral part of the Roman constitution, the tribunate retained its anarchical possibilities, and it was constantly used as an instrument of party strife, now by one side and now by the other. Frequently its great negative powers*® made it an instrument by which the senatorial party kept control over magistrates who might wish to use their theoretically enormous powers without sufficient respect for the senate’s wishes, but the tribunate was also the office chosen by reformers, such as the Gracchi, and demagogues, such as Saturninus, and its powers remained a 8 Above, 38. When Augustus created the aerariwm militare to provide for the pensioning of discharged soldiers, it was in part funded by an inheritance tax (vicestma here-

5 Below, 79. 6 Above, 43.

ditatwm) imposed on citizens. 4 Kunkel, SZ 77 (1960) 381. 7 Ch, 18.

8 It is generally said that the tribunes had no positive powers, or only very few, such

as those in connexion with guardianship under the lez Atilia (Gai. 1.185). But see

Karlowa 1.224f.

54

The magistracies danger to senatorial supremacy. For this reason they were drastically curtailed by Sulla when he re-established the constitution on an aristocratic basis, but the original position was shortly afterwards restored, and the tribunate remained important down to the end of the republic. Its powers and also the sacrosanctitas, or inviolability, which still attached to it, made it one of the two chief bases on which Augustus constructed the legal justification of his autocracy.

G. Tue Dicratorsuip. The republican constitution made provision from the first for a temporary return to monarchy in times of crisis when the principle of colleagueship was likely to cripple the efficiency of the state.? Nor was it, as we might have expected, for the people to decide when this was necessary, but for the consuls themselves. In an emergency either consul could nomin-

ate a dictator who then became superior to himself, his colleague and all other magistrates. Which consul was to exercise the power was decided either

by the possession of the fasces, or (in default of agreement) by lot, and, in any case, during the long period of senatorial government the power was only

exercised on the authority of the senate. Though he was appointed without reference to the people, the dictator’s imperium, like that of the consuls and praetors, had to be confirmed by a lex curiata.? He could not hold office for longer than six months, and apparently his powers also came to an end when the term of office of the consul who had appointed him expired. The persons appointed were nearly always men who had held the consulate, and the office, originally confined to patricians, probably became open to the plebs at the same time as the consulate. During his tenure of the office the dictator was supreme; his superiority over the consuls was shown by his right to twenty-

four lictors, a greater number than even the kings had had, and the axes appeared in the fasces even within the city. He was not, until comparatively late, subject to provocatio,® and originally he was apparently also free from tribunician intercessio; even after this had been changed (perhaps about the beginning of the third century B.C.) the tribunes appear to have used their rights very rarely, if ever. As bearer of imperium the dictator was entitled to do all the acts that a consul could do and he could further do one thing that no other magistrate could do, namely, give imperium to his own delegate, for every dictator nominated a magister equitum, who, though primarily, as the name implies, appointed to command the cavalry, was a general assistant and might have various duties delegated to him, e.g. full command at

Rome while his superior was in command of the army in the field. The ® Some writers who doubt the traditional account of an immediate transition from monarchy to co-equal consulship see the dictatorship as an intermediate form: above, 8n 2; contra, Staveley, Historia 5 (1956) 101ff. 1 Above, 47.

2 Above, 19. 3 Perhaps not till the lex Valeria of 300 B.C. 55

The republican constitution

magister equitum had the right to six lictors and ranked equally with a praetor.*

The dictatorship as part of the regular constitutional machinery did not last as long as the republic. Already in 217 B.C. there was a departure from the established practice when the dictatorship was conferred by election on

Q. Fabius Maximus, and the original character of the office was further weakened when, in the same year, the comitia conferred equal power on the

magister equitum. After 202 B.C. no dictator in the old sense was ever appointed. The title, it is true, was taken both by Sulla and by Caesar as the

basis of their arbitrary powers, but in neither case was the nomination regular, and, in fact, both were successful party leaders able to impose their wills on the community in the revolutionary period because of their command of military force. The old dictatorship had disappeared because it was incompatible with complete senatorial supremacy.

H. Tue Minor Macistrrates. First among these must be reckoned the military tribunes. There were normally six tribunes to each legion and hence for the eventual army of four legions there were twenty-four tribunes, but the stages of growth are uncertain. Until 862 B.C. they were appointed by the consuls, but in that year the election of the six tribunes ® was entrusted to the comitia tributa. Minor civil magistracies were: (a) the tresviri capitales who exercised subordinate functions in connexion with criminal jurisdiction, providing for the imprisonment of accused persons pending trial, arranging for executions and probably dealing on their own authority with slaves and foreigners. From their responsibility for the prevention of nocturnal disorder they were commonly known as tresviri nocturni.® (b) The tresviri monetales, or masters of

the mint, whose creation in 289 B.C. is thought to mark the beginning of 4 A dictator could also be appointed for a limited and momentary purpose, such as convoking the comitia, or even for a purely ceremonial function, such as holding certain games; but there seems to have been no difference in principle between these dictators imminuto ture and those created optima lege; Luzzatto, St. de Francisci 3.41 0ff. 5 Or the election of six out of a larger number? See Mommsen, StR 2.187 n 1; Livy 7.5.9. The practice of popular election had, however, fallen into disuse by the time of the Gracchi; Gaius Gracchus seems to have revived it. 6 See further, below, 314. They also had some functions in connexion with the jurisdiction of the urban praetor, but the nature of these functions is obscure. Festus, s.v. sacramentum (Bruns 2.34), records that a lex Papiria (after 242 and probably before 124) provided that sacramenta exigunto iudicantoque. This need mean no more than that they

exacted the penalty which was forfeited by the loser in the legis actto sacramento (below, 180) and adjudicated on any attendant disputes, but iuwdicare has been taken to refer to some independent jurisdiction. They also had some function in connexion with manus iniectio (below, 188), perhaps to supervise its exercise on behalf of the praetor. It was apparently the lex Papiria which provided for their election by the comitia. For discussion see Pugliese, Proc. 1.211ff., with references. 56

The magistracies coinage at Rome.’ (c) Six commissioners for the upkeep of roads, of whom

four were responsible for the streets of Rome itself, two for those in the immediate neighbourhood. They worked under the control of the aediles. More important for the lawyer are the judicial magistrates, (d) four praefectt Capuam Cumas, praetorian delegates for jurisdiction in these cities,*®

and (e) xvii stlitibus iudicandis.’ These were associated with lawsuits involving freedom, and though they acted, during the republic, only as jurymen, were elected, and counted as magistrates. The reason is perhaps to be found precisely in the great importance, especially to the lower classes whose family tree might be hazy, of these trials for liberty. The electing assembly was in all cases the comitia tributa, and by the end of the republic, all the minor civil officers were spoken of collectively as the

vigintisexviri; it was usual to have served in one of these offices before holding the quaestorship.

7 Below, 145 n 6. 8 Cf. below, 330. ® Below, 198.

57

CHAPTER 3

Italy and the provinces during the republic By the end of the republic the extent of Rome’s dominions was very nearly

as great as it ever became, and included already the greater part of the known world. This great empire was not, like that of Alexander, the work of a single conqueror, but the result of centuries of warfare and expansion, and its government was therefore not a coherent scheme imposed by a single mind, but an intricate structure embodying the expedients of many successive generations of rulers. Nor can it be said that the political institutions of Rome, though no doubt they helped her to acquire her empire, proved a success as a means of governing it. The city-state constitution broke under the strain imposed on it and had to be replaced by a military monarchy, for the constitution of Rome at the end of the republic, it must be remembered, was still that of a city-state. If the territories which we call the ‘ Roman empire ’ can be said to have a constitution at all, it must be described as a kind of association of city-states of which the sole bond is their common subjection, though in varying degrees, to the supreme city, whose governing organs have to serve the purpose of governing also the whole federation. In tracing the development of this federation it is best to distinguish three

stages, those of the Latin league, the conquest of Italy, and the acquisition of the provinces beyond the seas and beyond the Alps. Not that these stages represent distinct periods; Rome had already conquered several Italian states

before she finally suppressed the last revolt of the Latins, and she had acquired provinces overseas before she had finished with Italy, but it was

only through her leadership in Latium that she was able to achieve her supremacy over Italy, and at the end of the republic, as well as during the early empire, Italy represented a privileged part of the Roman dominions contrasted with the subject territories beyond her borders.

1 ITALY A. Tue Latin LEAGUE. Rome’s relations with her Latin neighbours emerge into the light of history with the making of a treaty, the foedus Cassianum (traditionally dated 493 B.C.) between Rome and a league of Latin cities.} This treaty, in which Rome appears not as a member of the league but as an equal, provided that each party should give assistance in war to the other, 1 For the reliability of the tradition see Last, CAH 7.487ff. 58

Italy that booty should be shared equally (also perhaps that the command of the combined forces should alternate), and that suits arising out of private transactions should be tried in the city in which they were made. What was more important for subsequent legal development, however, than any of these was

that this relationship between Rome and the Latin cities was evidently accompanied by the sharing of certain private and public rights. Seen from the Roman point of view this meant that the Latins had some, but not all, of the rights of Roman citizens. Moreover, this ‘ Latinity ’ came in time to be an abstract status capable of being conferred on communities which had no connexion with the original Latin cities. This resulted from one of the important joint activities of Rome and the league, the foundation of colonies. A colony in the Roman sense is primarily a military settlement planted on conquered land with the object of holding down the surrounding territory, though of course it served also the purpose of providing for surplus population, for the colonists were always given parcels of land in the territory of the new cities. The Roman or citizen of any Latin state who became a member of a Latin colony lost his original citizenship, for the colony was a separate

state, and the rule was that no one could be a citizen of two states at the same time.” At first the colonies became members of the league on the same footing as the original cities, but this appears to have changed even before the dissolution of the league, the newer colonies being allies, not members. Their |

citizens, however, had the same Latin rights as the citizens of the older colonies, and, as we shall see, after the dissolution of the Latin league Rome continued to found colonies with these rights elsewhere in Italy and even in

the provinces, and also to concede the Latin status to existing peregrine communities.

Chief of the private rights was commercium, which the Latin cities had originally not only with Rome but with each other. The general rule of antiquity was that the law of a community was for the members of that com-

munity only, and that the stranger was without rights. If there was no treaty to the contrary with his state the foreigner could be seized as a slave and his property taken by the first comer as goods without an owner. Where there was a treaty, protection of the citizens of the contracting states could be arranged, and the development of the ius gentium secured protection even for those without a treaty.* But the bond of commercium meant more than this.* It entitled the Latin to use the Roman means of acquiring property by mancipation,® and perhaps also the other acts per aes et libram.® Latins could also, unlike other aliens, benefit under a Roman will, but this was probably not as a result of their commercium. Similarly, no doubt, though we hear

2 See further, below, 71ff. 3 Below, 102ff.

4 Kaser, St. Arangio-Rwiz 2.130ff. 5 Below, 143ff.

6 Below, 161, 164ff. Other transactions would become open to them with the development of the ius gentiwm. 59

Italy and the provinces during the republic little about it from the other side, the Romans enjoyed the same privileges in the Latin cities. The right to use mancipation for the acquisition of property did not mean that the Latin, when he acquired property, owned it ex ture Quiritium (by Roman right), for he was not one of the Quirites.” Consequently he could not use the old Roman procedure by legis aciio to assert his ownership in a Roman court ® and could not acquire property by in ture cessio, which in-

volved the use of the legis actio in a collusive manner. The difficulty that mancipation, as described by Gaius, also involves the assertion of ownership ex ture Quiritium is best met by assuming that there were two forms, one, that described by Gaius, for use between citizens, another, in which the words ex ture Quiritium did not occur, which was used when one of the parties was not a Roman.

There was also probably conubtum between Latins and Romans.’ This meant that if a Latin married a Roman the union was recognised by Roman law, with the result that the children took the status of the father, and, if he was a Roman, were in his power (patria potestas), whereas in unrecog-

nised unions they took the status of the mother. :

The Latins’ most important public right (until perhaps the second century B.C.) ! was that of becoming a Roman citizen by taking up permanent residence in Rome and abjuring their citizenship of origin (tus migrand?).? Other Latins resident in Rome were permitted to vote in a tribe chosen by lot on each occasion. The Romans no doubt had corresponding rights in Latin cities. The Latins, however, became dissatisfied, and in 340 B.C., after Rome had rejected demands for greater political rights, they revolted. The revolt was suppressed in 338 and a radical change followed. The league was dissolved and the federal government was abolished, but the settlement was otherwise generous. Some cities received full Roman citizenship while retain7 The origin of this word (denoting Roman citizens) is obscure; see Labruna, Labeo 8 (1962) 340ff. Ius Quiritiwm can mean either the law applicable to Roman citizens (J. 1.2.2) or the Roman citizenship; Gai. 1.35 has both meanings. The latter is probably the original meaning. De Visscher (Etudes 3.225ff.=Mél. Lévy-Bruhl 317ff.), following Mitteis, RPR 67 n 15, notes that in the ancient formulae (vindicatio, in ture cessio, mancipatio) the term is confined to rights in rem, and suggests that such absolute rights were seen as deriving from citizenship, whereas the derivation of obligations was only indirect, via the act creating them (sponsio, etc.). 8 Pugliese, Proc. 1.233ff.; Mitteis, RPR 124 n 45; contra, Sautel, Varia 1 (1952) 75ff. 9 Epit. Ulp. 5.4 says that Latins did not have it unless it was specially granted, and Gaius, 1.79, implies that a marriage between a Roman and a Latin was one without conubiwm, but these are Latini coloniarii, i.e. the members of the later Latin colonies and the other communities to which Latin rights were given. The Latiné prisci, i.e. the original members of the league and of the early colonies, clearly had some advantages over the coloniarii, and conubiwm was very probably one of them. 1 Below, 61. 2 Castello, BIDE 61 (1958) 210ff. The term is not Roman.

60

Italy ing their local government, but were required to surrender control of their relations with other states, whether Latin or not, and to provide contingents to serve with the Roman armies. Others were given civitas sine suffragio, which meant that they had the same rights and burdens (including military service and tributum*) as Romans, but could not vote or be elected to a magistracy and were themselves subject to the jurisdiction of Roman magistrates or praefecti iure dicundo. They had the ius migrandi, however, and their status was not then, as it later came to be, regarded as penal, but rather as a transitional condition. Others, again, remained as allies, deprived of commercium or conubium with each other, but retaining commercium, and perhaps in some cases conubium, with Rome herself, and probably also the ius migrandt. The subsequent history of Latin rights is that of the Latin colonies. Settlements with rights similar to those of Latin cities continued to be made, and

were among the chief methods adopted for securing Rome’s hold on Italy, but it was, of course, now Rome alone which decided when and where they were to be sent, and who chose the colonists. These might be Romans or be taken from any of the allied states, but every colonist who was enrolled, the Roman included, lost his original citizenship on receiving that of the new city.*

The ius migrandi was curtailed, perhaps about 265 B.C.,° by the requirement that a Latin wishing to acquire citizenship in this way must leave a son in his own city, the object being no doubt to maintain the fighting effectiveness of the colonies, and in the course of the second century the right seems to have disappeared altogether, being replaced by what is subsequently called minus Latium,® the first unequivocal reference to which is found in 89 B.C.’

Latinity thus became a status, privileged in comparison with that of the other allies by reason of the private rights and the possibility, even though reduced, of obtaining Roman citizenship. It could, as has been said already, be granted not only to colonies but also to already existing peregrine cities. It was thus granted in 89 B.C. to the inhabitants of Italy north of the Po (Gallia Transpadana) who were not yet considered ripe for the citizenship which was then extended to the rest of Italy,® and it continued to be granted in the remainder of the republican period, especially by Caesar, and in the early empire to communities in the provinces as a kind of half-way house to citizenship. Latin rights were even, under the empire, taken as a model for

3 Above, 38. 4 Gai. 1.131.

5 See Sherwin-White 96ff. 6 Below, 345 n 4.

7 Asconius in Pis. p. 3 (Clark), but clearly the institution was already established then. If Mommsen’s supplement is accepted there is a reference to it in the lez Acilia of 123

B.C. (Bruns 1.72; FIRA 1.101), line 78. 8 Below, 66. 61

Italy and the provinces during the republic the status granted.to certain freedmen who did not become citizens on manu- |

mission, the ‘ Junian’ Latins; but this ‘ Latinity ’, which did not imply membership of any community with Latin rights, is a mere anomaly from the point of view of republican law.°®

B. THE REMAINDER OF ITaLy. Towards the rest of Italy Rome pursues a policy similar to that adopted towards the Latins after the dissolution of the league. Some of the states over which she gains supremacy she dissolves and incorporates, others she binds to herself by treaties which, though leaving them technically autonomous, embody such restrictions on that autonomy that they become in fact dependent. Confederations are everywhere broken

up and the allies of Rome are forbidden to conclude treaties with other powers.

If we look at the position in 266 B.C., when her victory over Pyrrhus in 275 and subsequent victories over her Italian enemies have made Rome supreme in Italy south of the Apennines, we find that the inhabitants fall already into four classes :

(i) Roman citizens with full rights. These include not only the inhabitants of Rome itself, but also those of communities incorporated with full citizen-

ship as well as members of the Roman colonies. For Rome used settlements of citizens with full rights as well as those with Latin rights for securing her hold on conquered territory, the former being in the earlier period smaller and more definitely military in character.1 As the colonists in these cases remained Romans, the foundation was not, like that of a Latin colony, the creation of a new state. The land given to them became a res mancipi, fully theirs ex iure Quiritium, and was incorporated in one of the Roman tribes,? just as the owners themselves remained on the census lists, lived under Roman law and were subject to the jurisdiction of the praetor. A minor degree of self-government perhaps originally of a military nature * was, however, given to the colony, and the colonists were commonly dispensed from military service with the Roman armies because they were in fact intended to act as a standing garrison where they lived.

(ii) Cives sine suffragio. Citizenship without the right of voting, such as had been given to some of the Latin cities, was given also to other conquered

communities which were too dangerous to be left independent even under a treaty of alliance, and the penal nature of this grant was usually made clearer by the confiscation of a considerable proportion of the conquered land. Now if Roman citizenship was, as the Romans certainly sometimes 9 See e.g. Gai. 1.22. 1 Sherwin-White 72ff. He holds that there was a change of policy about 180 B.C.

2 Above, 22. 8 Sherwin-White 78. 62

Italy said themselves, incompatible with any other citizenship, such incorporation, no less than incorporation with full rights, would have implied the utter extinction of the conquered city as a separate state, and this view was in fact taken by Mommsen. It is, however, now very generally held that Mommsen

applied the rule of incompatibility of citizenship more rigidly than the Romans themselves, and it seems that in fact considerable powers of selfgovernment were exercised by the local authorities, not merely by delegation from Rome, but as a relic of their former sovereignty.* According to Roman ideas, however, Roman citizenship implied the use of the Roman law (though some local rules managed to survive) and subjection to the jurisdiction of the praetor urbanus, but with the increase in the number of citizens it was obviously impossible to send every case to Rome for trial; nor could the praetor go on circuit, for he might not be absent from Rome for more than ten days during his year of office. The device of delegation was therefore adopted and praefecti ture dicundo were appointed by the praetor to take the civil cases in outlying communities. It is probable that these delegates also exercised general functions of government in those cities which were allowed no local magistrates. In cities with local self-government they exercised some supervision over the local administration, but their main task was jurisdiction, and for this therefore they were sent to cities whose inhabitants had the full citizenship as well as to those sine suffragio, at any rate where the city was so far from Rome that to send every case before the praetor would have been an intolerable hardship. Originally all prefects were nominated by the praetor whose delegates they were, but, as we have already seen, four were in later times elected in the comitia tributa and counted as minor magistrates of the Roman people.°® The recognition of local self-government among communities of citizens implies a departure from the general principle, which applied in Italy as well as in Greece, that state and city are convertible terms, for it means that a man may be a citizen of Rome and also have rights as a member of a subordinate community. This double citizenship is at the bottom of the Roman conception of municipium *® in the later sense of the term, i.e. a township of citizens or Latins with minor rights of self-government. In the period of which we are speaking the word had not yet this technical meaning, but the fact was there and was to be of the greatest importance in the history of Roman, and indeed European, governmental institutions. 4 See Excursus to this chapter. 5 Above, 57.

6 The word municipiwm was derived by the ancients (Festus, s.v. municipium, Bruns 2.15) from munus capere and thought to refer to the position of cives sine suffragio who ‘took the burdens’ of Roman citizenship but did not share in its privileges. It has been suggested that munus signified not ‘burden’ but ‘ gift’, municipes being thus originally ‘gift takers’, ie. members of states allied to Rome by a treaty of friendship which was 63

Italy and the provinces during the republic (iii) Latins, This term now means almost exclusively the inhabitants of the Latin colonies, for the original Latin cities had nearly all obtained Roman

citizenship. The Latins were free from tributum, sent separate contingents to the army which served under their own officers, and owned their own land according to their own law, but their position was more favourable than that of the other allies on account of their privileged position in private

law and their tie with Rome was closer in other ways also. The newer colonies had been planted by Rome alone; their legal’ and social system was modelled on the Roman, and they looked upon Rome as their mother city and their support against the frequently hostile populations of alien race among whom they were settled.

(iv) Soci (allies). The position of the ‘ allied ’ states, though less favourable than that of the Latins, was in this period better than that of cives sine suffragio. The provisions of the treaty which bound them to Rome no doubt varied in the different cases, but in general the allied state gave up her right of concluding treaties or making war separately from Rome and agreed to send troops, the number of which was probably fixed by the treaty, to serve

with the Roman forces. The coinage of silver (first undertaken at Rome about 269 B.C.) was in effect reserved to Rome,* but for the rest the states

were independent. They retained their own constitutions (though these might be remodelled to suit Rome), their own systems of law, their own finance and administration (except that Romans were dispensed from paying import and export duties), and their ownership of their territory was recognised by Rome. They need not admit Roman garrisons ® and were free from

Roman taxation except in so far as the provision and payment: of troops

necessarily involved expense. |

A century after the defeat of Pyrrhus, when the third period into which Mommsen divides his History comes to an end with the battle of Pydna’ (168 B.C.), the condition of the Italians has changed considerably for the worse. This is due, in the main, to the great struggle with Carthage in which Rome has been forced to employ ruthlessly all the resources at her disposal

and particularly to the measures of punishment meted out to those of her allies who had made common cause with the enemy. Many of the Campanians and others in the south of Italy were denied even the citizenship sine suffragio, and remained dediticii. This meant that the condition of sursymbolised by the interchange of gifts; but see, in support of Festus, Pinsent, Class.

leges. 8 CAH 7.663. Quart. (1954) 158ff. 7 Cicero speaks (Balb. 21) of the voluntary acceptance by the Latins of several Roman

® The right to place. garrisons in allied towns is occasionally found in exceptional foedera (e.g. that with Tarentum), and was naturally combined with the right to take

hostages; CAH 7.655f. 1 Above, 2. 64

Italy render (deditio) into which they had fallen when subdued by the Romans, instead of being exchanged for incorporation or alliance, was perpetuated. They were not rightless in private law, for they remained free, and all free persons were regarded as having rights under the tus gentium,? but they had

no status in public law at all, and were simply subjects with whom the Roman government could do what it liked.*

But even apart from these cases of exceptionally harsh treatment the Roman yoke began to bear more heavily on the Italian ‘ allies ’ and even on

the Latins. Many cities had to submit to an unfavourable revision of the treaties which bound them to Rome; military burdens were imposed on Latins and allies more heavily than on Romans, and they were not given a fair share of the spoils of war. Roman legislation * and Roman administrative measures ° were sometimes extended without legal justification to both classes, and Roman magistrates, knowing that there could be no effective resistance, did not scruple to behave like foreign conquerors in ‘ federated ’ cities.

One symptom of the changed attitude of the Romans was that they became much less generous with the grant of their citizenship. They were now

definitely the masters in Italy, not merely the leading state, and they did not want to diminish the value of their privileges by sharing them with too many others.* The citizenship thus came to be the chief object of ambition to the Italian peoples who could now no longer hope for any real independence.’

The claim of the allies to the citizenship was favourably viewed by the leaders of the democratic party at the beginning of the revolutionary period (in particular C. Gracchus) and from 125 B.C. onwards proposals were brought forward for granting it. All, however, failed because of the united opposition of the senatorial party and the city mob who were jealous of their privileges and in this matter refused to follow their usual leaders. The climax was reached in 95 B.C. when the consuls of the year carried a proposal (lex Licinia et Mucia) which, so far from granting the citizenship

to the Italians, instituted a severe scrutiny of the citizen-lists to weed out those whose claims were at all doubtful. This measure and the failure of a 2 Below, 102ff.

3 Strictly deditio is the agreement by which the enemy surrenders completely on the promise of mercy (fides), and the word does not apply to the conquest of a people whose city is taken by storm, but the position of the conquered is similar in the two cases. For dedttictt under the empire see below, 346f.

4 E.g. in 193 B.C. when usury laws were extended both to Latins and to other

Italians. 5 The Sc. de Bacchanalibus (below, 313) was applied all over Italy. 6 Cf. the restrictions on the acquisition of citizenship by Latins, above, 61. 7 One of the chief reasons was the protection which the right of provocatio gave to a citizen, but to no one else, against arbitrary punishment by a Roman magistrate. 65

Italy and the provinces during the republic proposal of Drusus in 91 to extend the citizenship to the allies helped to precipitate the Social war (91-88 B.C.). Though the Romans succeeded, after more than two years’ hard fighting, in suppressing this revolt, in which the

greater part of Italy took up arms against them, it was only at the cost of relinquishing the principle for which the struggle had been begun. TI'wo laws were proposed and passed before the war was over, one a lex

Iulia by the consul L. Caesar in 90 B.C. which gave the citizenship to all allies who had not yet revolted (and perhaps to those who laid down their arms), the other, a lex Plautia Papiria in 89 B.C. by two tribunes allowing two months to any person domiciled in Italy during which he could obtain the citizenship by giving in his name to the peregrine praetor in Rome.® These measures clearly did not include Italians still in revolt, but in fact citizenship was extended shortly after their submission to all cities in Italy proper,® and this extension remained the solid achievement of the Social war. As a result a kind of double citizenship, Roman and local, became the rule in Italy. The Latin or Italian city which had previously been, at least in theory, a sovereign state in alliance with Rome, became what municipia in the strict sense had long been, merely a community for the purposes of local government, and the difference between coloniae, municipia and praefecturae became virtually one of name only. The acquisition of the Roman citizenship by the Italians also brought with it, in principle, the application of Roman law, and the new citizens, like the old, were subject to the jurisdiction of the praetor or his delegates. But the powers of local self-government remaining to the cities, or newly bestowed on them, are considerable. Local assemblies elect magistrates, local senates deal independently with their

own finances; criminal jurisdiction is largely local, and there comes into existence a civil jurisdiction under Roman law exercised by local magistrates

in subordination to the praetor at Rome. ,

2THE PROVINCES Provincia literally means just ‘ sphere of authority of a magistrate’, and the sense from which our word province is derived is simply the result of the territorial application of the same idea. A ‘ province ’ is a sphere of authority ‘with territorial limits assigned to a magistrate with 7mperium, and, as such, 8 More restrictively, Sherwin-White 132. Our knowledge of the content of the lex derives from one passage of Cicero (pro Archia 7), who cites only the chapter of the lez ‘which is relevant to his client’s case. How many other chapters there were we cannot know.

, “Within northern Italy (the ‘ Cisalpine Gaul’ of the Romans) a distinction was made between the regions south and north of the Po. The former received full citizenship, the latter only Latinitas, the further step to citizenship being taken under the supremacy otf Caesar in 49 B.C.

66

The provinces is outside Italy, for within Italy there were no such geographical limits on the zmperium. It is the unity of command given to the governor within these limits which really constitutes the unity of the province within which may live people actually governed in different ways and standing in varying relationships to Rome. For all of them the governor represents the authority of the Roman state, whether they are Roman citizens resident in the province, members of Roman or Latin colonies or other inhabitants living in their original communities which might or might not have a considerable measure of self-government. It is this fact which explains how the Romans managed to govern their great empire with so few officials. When they acquired territory they did not. incorporate it into a coherent administrative system of their own; they left it as a rule in much the same condition as that in which they found it, but subjected great areas of it to the general

supervision of a magistrate holding imperium and thus vested with the supreme authority which imperium gave outside the city. The governorship was, in fact, the military command of the Roman general perpetuated as a system of control in times of peace, and the chief limit on the governor’s power was not any legal restriction but the physical impossibility of attention to detail with the very small staff which the state put at his disposal.

The first method by which the Romans provided for magistrates with | imperium to govern their provinces was an increase in the number of praetorships, but with the growth of the empire this measure proved insufficient, and resort was had to the prolongation of the power of magistrates after their year of office had elapsed. This was never deemed permissible within the city, but elsewhere the senate frequently used its dispensing power for

this purpose. It was also the rule that a military commander outside the city retained his command until relieved by his successor. Prolongation however was of magisterial power, not of the office itself, and the ex-magistrate exercised his «mperium pro consule or pro praetore, ‘in place of the consul or praetor ’, according to the rank he had held. It must be remembered that, in the period before Sulla, consuls frequently undertook commands overseas

and these commands were often prolonged in this way, as were those of praetors whose successors were needed for other purposes. It was thus already common for provinces to be administered by pro-magistrates before the time

of Sulla, and after Sulla it became the norm that both consuls and all the praetors should remain in Italy during their year of office and proceed to a province only afterwards. By a law of Pompey’s, passed in 52 B.C., this was again altered, and five years were to elapse between the tenure of office at Rome and the provincial governorship, a rule which, though it fell into abeyance during the civil wars, was revived by Augustus.

The governor’s staff included only one actual magistrate, the quaestor, of whose duties we have already spoken; there were legati, i.e. senators 67

Italy and the provinces during the republic without office (one usually in a praetorian, three in a consular province), whom he could employ as his delegates in any type of duty, military, admini-

strative or judicial. These men were originally appointed by the senate, but in later times chosen by the governor himself, and they were commonly men of considerable seniority and experience, sometimes themselves exgovernors or former generals. The governor had also a number of comites (literally ‘ companions ’) with him, i.e. usually younger men who were being initiated into public life, whom he could use for any purpose he thought fit. Both the legati and the comites have an important later history under the

empire. | Rome’s first provinces came to her as a result of her struggle with Carth-

age; Sicily, the first, was created in 241 B.C. after the end of the first Punic war, and Sardinia ten years later; the two Spanish provinces were kept at the end of the second war in 201, and Africa was added after the destruction of Carthage had ended the third in 146. Meanwhile, Rome had become involved in the affairs of the eastern Mediterranean. Macedonia became @ province in 148, Asia in 129, and the expansion did not cease during the period of the revolution. By the end of the republic there were in all fifteen provinces: Sicily, Sardinia, Hither and Further Spain, [llyricum, Macedonia, Achaea, Africa, Asia, Gallia Narbonensis, Gallia Cisalpina, Bithynia, Cyrene with Crete, Cilicia and Syria.

In these extra-Italian dominions Rome’s policy was partly similar to that adopted in Italy, partly different. Incorporation of provincials as citizens either with or without voting rights, such as had been common in the earlier phases of the conquest of Italy, was unknown, but the policy of alliance by treaties which, in fact, left the other party dependent on Rome was widely adopted. In some cases, as in that of Marseilles, these treaties represented older agreements which had been concluded when the parties were really on an equal footing, and many states retained real self-government in so far as all internal affairs were concerned, including even criminal jurisdiction over Roman citizens within their territory.! External relations were, of course, strictly controlled by Rome, or, rather, no external relations were, as a rule, allowed except with Rome herself.

Next in rank to these ‘ free and federated ’ states come those which are ‘free’ but not ‘ federated ’, i.e. those whose freedom is guaranteed, not by

a treaty but by a unilateral act of Rome herself, which, unlike a treaty sworn to by both parties, could legally be revoked at any moment. One , document of this nature, or rather a considerable fragment of it, has sur1 Treaties with kings, though unknown in Italy, occur frequently elsewhere. Rome’s sphere of influence, like that of imperial states in modern times, was not bounded by the territory which she ruled directly, and a common preliminary to incorporation in the empire was the recognition and support of a native king, who in return took his orders from Rome and had to protect her frontiers. 68

The provinces

vived in an inscription.? It is a plebiscite of about 70 B.C.* and enacts, among other things, that the citizens of Termessus (in Pisidia) are to be ‘free friends and allies of the Roman people and enjoy their own laws in so

far as they are not incompatible with this statute’. There is a significant absence of any statement that the Termessians have consented to this arrangement. The rest of the provincials who did not belong to either class of favoured

state were mere subjects, and, although they are included in the general term soci (allies), it is difficult to find any legal formula which will distinguish them from the dediticii.4 There were, however, important variations in the actual treatment of different classes. Where, as in Sicily and the Hellenised East, there were city-states and a high standard of civilisation in existence before the coming of the Romans, these city-states were left with a considerable amount of local self-government and the use of their own law, but it must be remembered that, even in these countries, there were large numbers of people who were not citizens of the city-states but subject to them, as well as others who had no connexion with any city-

state at all but lived in villages and cultivated the soil. In the East, especially in the province of Asia, these last had in many cases been serfs attached to royal estates, and though Roman law itself did not at this time ° recognise serfdom (as opposed to slavery) their position does not appear to _ have been bettered by the Roman occupation. Nor, indeed, was that of the subjects of the communities who continued, in the Roman phrase, to be ‘ attributed ’ to those cities. It is difficult however, with the small amount of evidence we have, to know how far the depressed condition of these classes

was due to burdens imposed on them by general rules of law, and how far it was due simply to their poverty in comparison with the wealthier citydwellers and Roman landowners whose tenants they became.

The organisation of a province, once it was acquired, was usually laid down in a lew data® or charter, commonly the work of the conquering general himself with the assistance of a senatorial commission of ten legati, or at least submitted to the senate for ratification. Of this nature was the lex Rupilia governing conditions in Sicily, of which we know a good deal from

Cicero’s Verrine orations. Such a lex data would, among other things, define the status of the different classes of inhabitants, establish principles of taxation, regulate local government and lay down rules for the adminis2 Lex Antonia de Termessibus (Bruns 1.92; FIBA 1.135). 3 Last, CAH 9.896.

4 Above, 64f. 5 For the ‘ colonate’ of the later empire see below, 435ff.

6 Lex data is not, as Mommsen thought, antithetical to lex rogata, as emanating from a magistrate rather than from the legislature. It denotes simply an enactment regulating the constitution of a province, colony or municipium, regardless of the enactment’s constitutional form. It may be itself a lex rogata (e.g. the lex Antonia de Termessibus, above) ; Tibiletti, St. de Francisci 4.595ff.; Dizionario epigrafico 4 (1956) s.h.v.

69

Italy and the provinces during the republic tration of justice. Of the lew Rupilia, for instance, we know that it provided that where one member of a state had a dispute with another member the matter was to be decided in the court of that state and by the law of that state; only where members of different states were concerned did the case come into the governor’s court.’ Similar rules no doubt applied elsewhere.® Much, however, was left undetermined by the lex of the province and had to be supplied by the edict of the governor. These edicts which the governor, like every other Roman magistrate, was empowered to make at the beginning of his term of office, laid down the principles which he intended to apply during his tenure of power. As happened with other magistrates, especially the two urban praetors, each governor usually took over the bulk of his predecessor’s edict and added only a comparatively small amount of his own,

with the result that these provincial edicts, like the urban edicts, grew into an important and permanent body of law. The power of issuing edicts, which no court or other body could disallow, is in itself sufficient to show how little

the powers of the governor were fettered even by the law of the province. It was, in fact, the great weakness of Roman provincial government under the republic that there was no-one to control the governor, the only check provided being under the laws against extortion and treason (maziestas), which enabled him to be prosecuted after his term of office had come to an end. This was a safeguard which the corruption of the criminal courts at the

end of the republic often made illusory.° |

The fact is that the Romans of republican times regarded the provinces from a purely selfish point of view and chiefly as sources of income. The ‘ federated ’ cities and some others were immune from taxation, but on the

. bulk of the provincials the burden was very heavy. Not only had they to pay taxes to the Roman state; but these taxes were generally farmed out. Roman companies or individual bidders paid an agreed sum for the right to collect them during a specified period, with the result that a great number of Roman private citizens were able to enrich themselves at the expense of the provincials. The governor too and his staff, it must be remembered, received. no salary and expected to find opportunities of making money out of their position. So far as military service was concerned the position of the provincials differed radically from that of the Italians. The treaties with the ‘ federated ” states stipulated for military support and the maritime Greeks had from time to time to furnish ships, but the mass of the population was not called upon, and this policy shows clearly the difference between Italy, which the Romans 7 Cie. Verr. 11.2.39.

8 Cf. the fourth edict from Cyrene (FIRA 1.409); de Visscher, Les édits d’ Auguste 123ff.; RIDA (1964) 324ff.; and, generally, Pugliese, Proc. 2.1.245ff. 9 The first law against extortion by provincial governors was the lex Calpurnia repetundarwm of 149 B.C.; below, 308. 70

Excursus — Double citizenship

regarded as a military confederacy under their leadership, and the subject world whose business it was to pay taxes.

The leading ideas of Roman provincial government in the republic are thus: (1) The subjection of great areas to the autocratic authority of a single magistrate whose duties are military, administrative and judicial. Individual states of the ‘ free and federated ’ or ‘ free ’ class, where geographically within the province, are not strictly subject to the governor, but his influence in fact extends to them. Where, as often happens, such a city is the chief one in a province he may make it his residence and the local authorities will have to let him have his way. (2) Retention in a large measure of the existing territorial organisation, especially when this is of the city-state type or readily adaptable to it. (8) Exclusion of the provincials from the citizenship (and from Latinity) as well as from the military burdens which fall on the Italian allies, in place of which they have to pay heavy taxation. The vast majority of Roman subjects are, so far as her law is concerned, peregrini, ‘foreigners ’, outside the pale of the strict Roman law and only entitled to such rights as all free persons have under the ius genttum. The word peregrinus in fact comes to mean, not a foreigner who belongs to an independent state, for of such, by the end of the republic, there were few in the known world, but a subject of Rome who is not a Roman because he is not a citizen.

Eacursus — Double citizenship See above, p. 63

Mommsen’s view ' that Roman citizenship was incompatible with any other was pressed to its full logical consequences by H. Rudolph,? who held that after the Social war, when the whole of Italy south of the Po obtained the citizenship, all local independence ceased, and all Italians remained under the exclusive jurisdiction of the Roman officials until a new system of local

self-covernment was devised by Julius Caesar. But Sherwin-White* has shown that in detail his argument will not hold, and de Visscher,* among other arguments, has pointed out that Gellius* cannot have been simply talking nonsense, as Mommsen ® holds, when he described municipes as ‘Roman citizens from municipia, using their own leges and ius’, for his authority was the Emperor Hadrian, who having been born in a municipium and having himself reorganised the system of government, must certainly

1 Sth 3.1.773ff£. 2 Stadt u. Staat im rom. Italien (Leipzig, 1935). 3 Roman Citizenship 136ff. 4 Nouv. Et. 51ff., 109ff.; and see his Les édits d’ Auguste 108ff.

5 16.13.6. 6 StH 3.1.796 n 3. 71

J.—4

Italy and the provinces during the republic have understood the subject thoroughly. De Visscher is also no doubt right

when he attributes the strictness with which Mommsen applies the rule against duality of citizenship to the influence of modern theories of state sovereignty. But there must be more doubt about De Visscher’s own interpretation of the rule. Briefly, he holds that the rule was a unilateral one, in the sense that her political superiority enabled Rome to maintain the position that none of her own citizens might accept the citizenship of any other community without forfeiting hers, but that there was nothing to prevent other communities, if they so chose, from allowing their members to become Roman citizens while retaining their citizenship of origin. And though, if they came to Rome, they forfeited their local citizenship, this was not because of any principle of incompatibility but because loss of local domicile entailed loss of local citizenship (mutatio civitatis).’ In support of this thesis he is able to quote some passages from Cicero in which a contrast between Romans and others appears to be made,® and other evidence. But in answer to the obvious logical objection that if A is incompatible with B, B is incompatible with A, he has to explain not only that the rule is one of Roman, not international, law, but also that it applies only to persons who are Roman citizens by origin. And here lies the weakest part of his theory, for ‘ Roman

citizen by origin’ cannot be taken to mean simply one who has had the citizenship from birth, because in that case the problem of incompatibility in the municipia would merely arise one generation later than on the ordinary view. He.has therefore to explain that the category of people he refers

to as ‘ citizens by origin’ changed its character in the course of history. Originally, and still in Cicero’s time, it included only citizens of Roman origin

‘in the strict sense ’—a very small number of people in the last century of the republic, and not including, for instance, Cicero himself, who was born in

the municipium of Arpinum. After Caesar, however, when the Romanisation of Italy was complete, it embraced all Italians as opposed to the members of provincial cities.

But if we accept this view, we must suppose without any direct evidence that when Cicero said ® ‘ we cannot belong to this state and another as well’,

he had forgotten that the rule of which he was speaking did not apply to himself, and further, that this rule, evidently moribund in his day, was revived shortly after his death, though with a highly significant change of meaning. Even Mommsen’s view, that the original strict rule of incompatibility was suddenly reversed under Augustus,’ is more easily credible. In truth, it would seem that we cannot attribute a completely logical system to the Romans at any time. No doubt it was generally felt in the republic 7 St. de Franciset 1.39ff.

8 Pro Balbo 29: Sed nos non posswmus et huius esse civitatis et cuusvis praeterea; ceteris concesswm est. Cf. ibid. 30; pro Caec. 100.

® See preceding 72note. 1 Sth 3.1.699.

Excursus — Double citizenship that a man could not have two patriae,” but the political exigencies of various

periods compelled the acceptance of situations that could logically lead to no other result. Closely connected with his treatment of citizenship, and of greater impor-

tance for the private lawyer, are de Visscher’s views on the subject of the law to be applied to Roman citizens who were members of municipalities. A rigid interpretation of the personality principle would mean that it could only be Roman law, and this again was Mommsen’s opinion, but the matter is not so simple. Gellius not only speaks of municipes as ‘ using their own law ’ but goes on to say that they are not bound by a Roman lez unless they have ratified it, though he admits that in fact it would be impossible in his own day to apply the local law because it had been forgotten.°

De Visscher therefore holds that while Roman citizens by origin were bound by Roman law only, new citizens continued to live by their original law, though it is true that this became gradually Romanised as the political preponderance of Rome and the technical superiority of her law made itself

felt. For this view he also relies, in large part, on the inscription from Rhosos,* and to a lesser extent on the Cyrenean edicts.* In the former the eitizenship is given by Octavian under a hitherto unknown lex Munatia Aemilia (42 B.C.) to a certain Seleukos for services rendered, and it is stated that he, and his family, are to have the choice of being judged in their own city according to their own laws, or before the Roman magistrates, with perhaps the further possibility of a free city of their own choosing.® They are thus given a choice of jurisdictions, but de Visscher points out that the law to be applied is only mentioned in connexion with their own city, and takes it to be the general rule that, so far from citizenship implying that Roman law and nothing else applied, the normal course was that new citizens not only continued to be members of their old city but were still subject to its laws. They were also subject to its jurisdiction, unless specially exempted as | in this case. In no case however, he argues, could they be given a choice of law, for that would turn them into Proteus-like beings, with whom nobody could safely do business.

De Visscher draws a somewhat similar argument from some words in the third Cyrenean edict’ in which Augustus orders that the inhabitants of the province of Cyrene who have been honoured with the (Roman) citizenship are, unless specially exempted, to Actrovpyetv otdév EXawov én péper TH TOV “HAAQVwv copaTe.

Most scholars have taken this to mean something like ‘ be subject to litur2 It is evidently with the idea of producing a paradox that Cicero says, Leg. 2.5:

8 Gell. 16.13.6-9. 4 FIRA 1.308.

omnibus municipibus duas esse censeo patrias, unum naturae, alterum civitatis.

line 55. 7 Lines 57-8. 5 FIRA 1.403; ef. de Visscher, Les édits d’ Auguste.

6 §8, lines 53-9; the alternative of a free city rests on a very doubtful restoration in 73

Italy and the provinces during the republic gies in their turn to the body of the Greeks’, but de Visscher adopts * a rendering of Wilhelm’s,® which would take éu pépe: r@ rHv “HAAHvwv together

and separate this phrase from owpaz7t, so that the whole would mean ‘ be subject to personal (i.e. as opposed to financial) burdens in their capacity as

Greeks ’. If this were right, there would of course be no doubt that the original citizenship was maintained, and that, in one maiter at least, the local law applied, but a considerable amount of special pleading by a master of the Greek language is required to give the desired meaning.

This idea that new citizens remained subject to their local law has farreaching implications. Its. acceptance would mean not only upsetting the general belief that, normally, the principle of personality prevailed, but in particular a revision of our views on the effect of the Constitutio Antoniniana' extending the citizenship to the whole, or almost the whole, of the free population of the empire. Mitteis ? took it to have implied necessarily an

equally wide extension of Roman law, and an attempt therefore to alter the law of millions by a stroke of the pen, and though this view has been attacked, it appears, as we shall see,* to be in its essentials correct. To hold otherwise is to fly in the face of too much evidence.

But if we may still believe fundamentally in Mitteis’ view, and in the general principle of the personality of law, all the riddles are certainly not yet solved,* and we probably err in our understanding of the ancient position

by attempting to express it in modern terminology and modern concepts.° There is evidence ® that some local rules survived in Italian municipia in spite

of the extension of Roman citizenship, and in the principate it seems that Roman citizenship ‘ remained compatible with local citizenship whose obligations were unaffected by it ’.’ We can only conclude that in the principate, as in the republic, anomalies existed, but in spite of them, the general idea that law went with citizenship was maintained, and the Constitutio Antoniniana thus provided a theoretical basis for the uniformity of law throughout Roman territory which was, in the late empire, to become a reality. 8 Ann. del sem. giur. di Catania 3 (1949) 14f. For his arguments against previous interpretations see Les édits d’Auguste 89ff. 9 Anzeiger Akad. Wissensch. Wien 80 (1943) 8. He thinks that the stonemason intended to write éu uépes 7@ rov “HAAHvwy cwpyarr. Approved by Wenger, Mél. de

Visscher 2.523. 1 Below, 346. 2 Reichsr. 159ff. 3 Below, 346 and 470.

4 A passage which does not seem to have been quoted elsewhere in support of the personality principle for the Republican period is Diodorus Siculus, Bk. 13. ¢.35, roAAai yobv rav Kara Thy virov méAewv xpwdpevat SveréAeorav rois rotrov [i.e. Diocles’] vépous péxpr Srov mévreg of SuxeArwrat 77g “Pwpaiwy modtretas 7£1b6noav, which seems to show that he

at least thought that attainment of the Roman citizenship would involve a change of law for the members of the Sicilian cities. 5 De Visscher, Osterreich. Zettschr. fiir Offentl. Recht 12 (1962) 71ff.; Norr, TR 31

(1963) 525ff., especially 577ff. 6 Above, 72f. 7 A. H. M. Jones, The Greek City from Alexander to Justinian (Oxford, 1940) 172.

And see Arangio-Ruiz, Scritti F. Carnelutti 4 (Padua, 1950) 53ff. TA

CHAPTER 4

Social conditions and the different classes of the population during the republic Rome owed her rise to the fighting qualities of her peasant-proprietors. The traditional picture of the earliest republic and the struggle between the orders

clearly show that then already the normal Roman was a landowner. It is true that the distress of the plebeians indicates that their plots must often have been very small, and it is also true that, according to modern archaeological theories, there had probably existed, before recorded history begins, in Latium generally, as in Etruria, a system of villeinage, i.e. that the bulk of the population had been semi-free tenants working on large estates owned by great landlords.’ But what we know of the economic side of the struggle between the orders is sufficient to show that this state of affairs was over when that struggle took place. One of the chief complaints of the plebeians is that they are forced to serve as soldiers and that on their return from a campaign they find their land ruined through lack of attention, or worse still actually devastated by the enemy, and that they are thus forced to borrow from the larger landowners in order to begin again. This is not the sort of complaint which would be made by a semi-free tenant, whose lord would

have the responsibility for setting him up again. The cry too is that the patricians monopolise the advantages of the public lands instead of dividing them up among the poor plebeians, which again shows that the plebeian is capable of holding land and regards such ownership as his normal condition. If the normal citizen was a small landowner we must not imagine that the estates of the aristocracy were very great either. The tradition of the patricians who had to be called from the plough to lead the armies of the republic

probably represents the truth. Rome was a poor and weak state from the expulsion of the kings until at least the time of the capture of Veii (traditional date 896 B.C.), her first great conquest of a foreign people. The earlier part of the struggle between the orders took place during this period of weak-

ness and the main plebeian victory (the Licinio-Sextian laws of 367) was won very early in the period of expansion. The fact that it was necessary ° to limit the amount of public land which could be occupied by any one individual to the relatively large figure of 500 tugera and to require landlords to employ a certain proportion of free jabourers shows that Rome was already beginning to go along the disastrous path of large estates and slave labour. But these measures themselves, though they were not strictly enforced, at

1 Tenney Frank 12. 2 If it is a fact; sce above, 15 n 1. 75

a Social conditions and different classes during the republic

any rate after the first few years, must have done something towards remedying the evils at which they were aimed, and still more was done to help the

peasants by the use which Rome made of her conquests. After the capture of Veii the city was razed to the ground and her territory divided among the citizens of Rome in equal lots, the plebeians having their fair share. And subsequently Rome made many ‘ assignations’ of conquered territory to individuals in addition to sending out colonies, which, besides their military function, served the purpose of providing land (in very small lots generally)

for her increasing body of citizens. }

The result of this policy was a population consisting chiefly of a homogeneous race of peasant-proprietors, men ‘ with a stake in the country ’, who were patriotic enough, hardy enough and numerous enough to provide the relatively very large armies which enabled Rome to continue her career of conquest once it was begun. These peasants tilled their land themselves with the help of their families and in some cases of one or two slaves or persons in mancipio.* Commerce and industry seem to have played a comparatively small part in the early development of Rome, and the government was, as we have seen, aristocratic. The success of the plebeians in their contest with the patricians did not change matters; it merely replaced the ancient aristocracy, which had a legal monopoly of office, by a new one in which, though every citizen had equal rights before the law, a comparatively few families, plebeian now as well as patrician, in fact furnished nearly all the magistrates and retained its power through the senate. The aristocracy thus evolved was probably the most successful the world has ever seen. It was not brilliant, and great statesmen were rare, but, like the rest of the Roman people, it was characterised by deep devotion to the state, a steadiness which never failed in the face of the most desperate situations, and a readiness to accept respon-

sibility which made it possible to entrust the great powers of the curule magistracies to a succession of ordinary aristocrats, and thus build up a governing body of men with wide experience of public functions. The Romans had the same fondness for entrusting great powers to individuals which is shown by

the English and, like the English, they took a sort of pride in the eccentric use of power; M. Livius, who when censor in 204 B.C. disfranchised thirtyfour out of the thirty-five tribes because they had first condemned him in spite of his innocence and then elected him consul and censor, would have precipitated a revolution in most countries. A great change, both in the external conditions of life and in the spirit of the country, began with the ending of the second Punic war. In seventeen years of continuous warfare the fairest districts had been ravaged, enormous numbers of men had fallen and the whole manhood of the country had been demoralised by continued military life. Agriculture suffered, not only from

devastation and lack of the cultivators who were called away on military 3 Below, 114 n 3. 76

Social conditions and different classes during the republic service, but from competition with the corn which was now imported in great masses from Sicily. Worst of all, the peasant-proprietor was gradually giv-

ing way to the great landlord who cultivated his estate mainly by slave labour. The wars enriched the aristocracy who were able to make great sums out of booty and out of the government of the conquered peoples and they also enriched tax-farmers and business men who found new fields for their enterprise. In Italy, as a result of war and confiscation, large tracts of country were acquired by the Roman state, and now that the number of citizens no longer sufficed to cover them with peasant settlements, the only profitable

method of exploitation was to let the land in large blocks to people with money, or to continue the old Roman custom of allowing them to ‘ occupy ’ it, 1.e. simply take it informally and thus come under an obligation to pay

the state a proportion of the produce. Everything favoured the capitalist. The successful wars made it possible for him to get as many slaves as he liked, and slaves were not only cheaper than free labour but there was no danger of their being called away for military service. Much land was suitable only for use as pasture, and pasture does not pay unless it is managed on a large scale, for a few herdsmen can look after a great number of cattle. On the other hand, competition with the corn-growing provinces was making the fertile parts of Italy turn more and more to wine and olive growing, both of which need capital, for the small man cannot afford to wait the five years

before vines or the fifteen years before olive trees will bear. The capitalist therefore tended to absorb the new land and to buy up the peasant-proprietor,

who might of course settle in a colony, or go abroad to the provinces as a business man, but who might also drift to Rome and become a more or less idle proletarian living largely on the corn imported from the provinces which the government now already began to distribute, not yet for nothing, but at very low prices. In the second century B.C. there were, no doubt, still large numbers of peasant-proprietors, and it was these men who conquered the Kast as they had conquered Hannibal, but the process of concentration of the land in fewer hands was going on all the time and was to be largely instrumental in bringing about the revolution. Latifundia perdidere Italiam — ‘ The great estates ruined Italy ’. That the wealth which flowed to Rome from her conquest of the East had the effect of corrupting her population is an old and true story. Contact with the East meant contact with decadent Hellenism, and wealth gave the oppor-

tunity of copying Hellenistic vices. But with wealth and vice came more desirable things — literature, art and philosophy. The state, the family and the farm had been the only things that mattered to the old-fashioned Roman;

now he saw that there might be more in life. In the third and second centuries B.C. the upper classes became permeated by Greek culture; Greek literature became the basis of education and Greek rhetoric already began to exercise its dangerous fascination over Roman minds. For speculative 17

Social conditions and different classes during the republic philosophy the practical Roman had little taste, but the great days of Greek speculation were over and emphasis was already laid rather on the ethical than on the metaphysical side of philosophical teaching. This was true of all the three chief systems, the ‘new Academy’ and Epicureanism as well as. Stoicism, but it was Stoicism which found the most favour at Rome. In the first place its speculative system left a place for the gods and their worship, whereas the E;picureans taught that the gods, if they existed at all, had no concern with the lives of men, and it thus did not come into conflict with that religion of the state which the Romans cherished from patriotic motives, even when they did not believe in it. But above all the Stoic ethic, with its ideal of the perfectly wise man who masters his passions in order to live the ‘life in accordance with nature ’, appealed to the Roman sense of duty and gave a theoretical justification for that service of the state which had been the guiding principle in the city’s life. The decay of agriculture in Italy and the consequent growth of a landless proletariate, together with the increasing differences in wealth brought about by the new opportunities of acquiring a fortune, of which naturally only a minority were able to avail themselves, were the causes which led to the attempted reforms of the Gracchi (1338-121 B.C.) and to the revolutionary period which was to last a century, and then only end in the establishment of the empire. This period presents a number of the most astonishing contrasts. From the point of view of the governmental system it was a time of breakdown, occasionally anarchy, and from that of the governing class of

moral decay, and yet it was the most brilliant epoch of Roman history. _ Foreign conquest did not cease, literature reached a standard never approached before and equalled only by the immediately succeeding period under Augustus, a series of brilliant personalities from Gaius Gracchus to Augustus himself passes across the stage of history, and in the latter years, when civil war was imminent or actually raging, the whole scene is lit up for us, not merely by professional historians, but by the literary genius of two men who were themselves among the chief actors in the drama — Caesar and

Cicero. Though the lawyer finds his ‘ classical age’ in the more orderly, if duller, times of the early empire, it is the last century of the republic which is § classical ’ for the student of language and literature.

It is, of course, impossible to classify exhaustively all the different orders of men who went to make up what had become a vast and highly civilised state, almost as complex in its social and economic structure as those of modern Europe. We can, however, roughly classify as follows:

(a) The senatorial nobility. Office was very largely confined to the high nobility, that is to say to members of families who could count curule magistrates among their ancestors,* though from time to time a novus homo might, 4 For the more restrictive view of Gelzer see above, 17 n 7. 78

Social conditions and different classes during the republic by his ability, force his way into the privileged circle. A young man of this class who, as was nearly always the case, was destined for a public career, would as a rule spend some years in military service, though this was apparently no longer legally necessary at the end of the republic. He would then start on his cursus honorum (career of office) by holding one of the magistracies which formed the XXVI virate and, on election to the quaestorship, would become a member oi the senate for life. The offices had to be held in order; the praetorship might not be held before the quaestorship, nor the consulship before the praetorship, and a minimum age was laid down for each. For the quaestorship this had been twenty-eight; in Cicero’s time it was thirty; for the consulship it was forty-three. Between these two offices there were not only the other offices — aedileship, praetorship, and, for plebelans, tribunate — but service in the provinces, and the rule was that two years must elapse between the tenure of two ‘ patrician ’ offices. Not before the lapse of ten years might the same office be held again. The exercise of any trade or profession was a disqualification for office and the senators had thus necessarily to be men of wealth. They were, in fact, all large landowners, often owning several estates in Italy and in the provinces. From direct participation in business, senators had long been excluded by a lex Claudia (?218 B.C.), which prevented them and their sons from owning a sea-going ship beyond a modest size (sufficient to carry the produce of their own lands) and probably also forbade them to participate in contracts for the farming of state revenues. In any case, all types of speculative business were regarded as unbecoming to a senator and could only be indulged in secretly, if at all. By the end of the republic the senatorial nobility was unworthy of the position it held in the state. Cicero, champion of the ‘ order’ though he is, inveighs against the degenerates who loll at ease in their villas and care more for their fishponds than for their public duties. The craze for luxury caused many of the aristocracy to run into debt, and the unfortunate provincials had to find enough money, not only to repair the shattered fortunes of their governors, but also to bribe the juries when the governor was prosecuted for extortion on his return. At the height of her power Rome was, in fact, the

prey of a degenerate governing class and it needed the strong hand of a monarch to reduce the turbulent nobles to order.

(b) Equites.* We have seen ° that according to the ‘ Servian ’ constitution there were eighteen centuries of equites who formed the cavalry of the army and who were evidently (since they took precedence in voting over the other 5 Hill, Roman Middle Class; Henderson, JES 53 (1963) 61ff.; Nicolet, L’ordre équestre.

It is difficult to find any appropriate English rendering for equites. ‘Knights’ suggests medieval chivalry, and ‘middle class’ is probably now too extensive and suggests too clear a separation from the senatorial order; Scullard, JRS 45 (1955) 181f. 6 Above, 20f. 79

Social conditions and different classes during the republic centuries) drawn from the well-to-do members of the first class, though without any fixed property qualification above the minimum for that class. They were known as equites equo publico, because the money for the purchase and upkeep of their horses was provided out of public funds, and they presumably reverted to the first class on reaching the age-lhmit for active service of 46.” These eighteen centuries were not, however, sufficient in times of crisis, and the need was met by other members of the first class who volunteered to serve as supernumerary cavalry, providing their horses at their own expense.* One can thus distinguish between the elghteen centuries, who have a special place in the centuriate organisation, and the supernumerary cavalry, who do not, but there is as yet no question of a distinction between senators and equites. For possession of a ‘ public horse ’ was not incompatible with

a seat in the senate, and senators were no doubt included in the eighteen centuries until they reached the age-limit (and sometimes evidently even longer).°

Three steps, probably all connected with the reform movement of C. Gracchus, contribute to the emergence of the equites as a separate order. A lower limit is established by the fixing of a minimum property qualification, which

at least in 67 B.C. was 400,000 sesterces,’ and an upper limit by a plebiscitum reddendorum equorum (c. 129 B.C.) which excluded senators from the eighteen centuries. The third step was taken by the Gracchan lex iudiciaria, which gave to equites the right, hitherto confined to senators, of sitting as jurymen. This right was to be a source of contention between the two orders until the lex Aurelia (70 B.C.) effected a settlement.? This much is 7 Hill, Roman Middle Class 9f.; contra Nicolet, L’ordre équestre 75ff., but see below, 81.

8 They are often called by modern writers equites equo privato, but this usage has no ancient warrant; Hill, Roman Middle Class 18 n 3; Nicolet, L’ordre équestre 66ff. Livy, 5.7, says that they first served at the siege of Veii (captured 396). The historicity of this

ean here be left aside; see Hill, op. cit. 17££. 9 See n 7, above.

1 This at least was the qualification stated in the lex Roscia theatralis, which restored to the equites the privilege (perhaps originally conferred by C. Gracchus and taken away by Sulla) of sitting in the fourteen rows of seats in the theatre immediately behind the senators. The earliest mention of a census equester is in 76 B.C, (Cic. Rosc. Com. 42), and it is commonly thought to have originated in the Gracchan lez iudiciaria (next note), but it may well go back to the second Punic war (Nicolet, L’ordre équestre 47ff.). 2 It is not clear whether the reform of Gracchus excluded senators altogether or simply added the equites, or whether it applied only to the quaestio de repetundis (below, 308) or to all criminal and civil courts. Perhaps the conflicting evidence can best be reconciled

by assuming that senators were excluded from the quaestio de repetundis (where the accused would be of their own order) and that for other courts the tudices were drawn from a mixed list; Brunt, Crisis of the Roman Republic, ed. R. Seager (Cambridge, 1969) 83ff. Fragments survive of a lex (Bruns 1.55; FIRA 1.84) dealing with trials for repetundae and containing provisions evidently aimed at excluding both senators and their immediate circle. This is usually identified as the lex Acilia mentioned by Cicero, Verr. 1.51, which is taken to be the Gracchan lex iudiciaria, but the crucial clauses defin80

Social conditions and different classes during the republic tolerably clear, but there remains a central ambiguity. There were obviously many who satisfied the property requirement of 400,000 sesterces (which was not a vast sum) * but who could find no place in the eighteen centuries, even if, as seems likely, these now contained more than 1,800 men. Were these ‘ supernumerary cavalry ’ included in the equestrian order, and in particular were they included in the category of those who were now entitled to sit as jurymen ? Cicero, writing in 76 B.C., contrasts those who are equites Romani from the standpoint of the census with another group, presumably

the eighteen centuries,* and the better view seems to be that, at least by this time, the order could be treated as including all those who satisfied the property qualification, and that the eighteen centuries constituted an elite of younger men within the order.* And it seems likely that the Gracchan tudices were also drawn from this larger group,® defined perhaps then for the first time in terms of the property qualification.

The exclusion of senators from business,’ though not always strictly ob-

served, meant that the equestrian order was identified with the class of wealthy business men, and in particular, since the chief activity of such men was the undertaking of state contracts, with the publican (Cicero ® calls them

‘the flower of the order’). The interests of this class would often diverge from those of the senatorial nobility, but the divergence must not be exaggerated: land was still the main capital, and many equites were landing positively those who may serve are missing; see Last, CAH 9.75ff., 892ff. For the next

fifty years the composition of the jury list varied according to which order was predominant. The lez Aurelia divided it equally between senators, equites, and tribunt aerartt. The identity of these last is obscure (see e.g. Henderson, JR&S 53 (1963) 63f.; Hill, Roman Middle Class 212ff.), but they were evidently socially close to, if not a part of, the equestrian order, and the settlement was no doubt unfavourable to the senate. The tribuni aerarit were excluded (or absorbed into the equites) by Caesar (46 B.C.), and further changes were made by Augustus and Caligula, but the albwm remained confined to senators and equites. See Kunkel, PW 24.1.749ff.; Mommsen, StR 3.527ff.

3 Cicero, Parad. 49, regards 100,000 a year as enabling a gentleman to live no more than a frugal life; to live in a great luxury could cost 600,000 (Brunt, op. cit. above, note 2, 83 n 1). And compare the pay of high officials in the principate (below, 337 n 2). 4 Rosc. com. 42: quem si tu ex censu spectas, eques Romanus est. There is also a clear distinction between the eighteen centuries of young men and the equester ordo itself in the Commentariolum petitionis, 33, but the validity of the attribution of this work to Cicero’s

brother, Quintus, and therefore its date, have been questioned (as to this see Balsdon, Class. Quart., 1968, 2424f.).

5 Nicolet, L’ordre équestre 163ff., argues that ordo equester has only a strictly juridical meaning, and that it includes only equites equo publico, who alone are equites Romani, but this is very difficult to reconcile with the literary evidence, particularly that of Cicero, which reveals a much more fluid usage; see Henderson, JRS 53 (1963) 61ff. 6 Mommsen, Stk 3.530, StrR 209, confined them to the eighteen centuries; contra,

Strachan-Davidson, 2.85ff. 7 Above, 79. 8 Plane. 23.

9 In the provinces especially the two might come into collision; if a governor was strict in his supervision of the tax-farmers he would find a jury of equites unsympathetic if he had to stand his trial for extortion afterwards at Rome. 81

Social conditions and different classes during the republic owners. Nor was there a sharp social distinction. Apart from the fact that the sons of senators would normally be recruited to the eighteen centuries, there were often close links between senatorial and equestrian families.’

The equites, like the senators, had certain outward signs of rank; they wore the angustus clavus, i.e. narrow purple stripes on the tunic (whereas senators wore broad stripes); they obtained the right to the gold ring, which in earlier times had been a senatorial privilege, and, as we have seen, they also had special seats in the theatre.

Under the empire the equestrian nobility became even more important; admission to it was a necessary first step to any public career, civil or military, for others than the sons of senators, and many very important posts were confined to men of equestrian rank, to the exclusion of senators. (c) The middle and lower classes. Among these we must reckon all free persons who did not belong either to the senatorial or equestrian orders. In the last centuries of the republic, in spite of the anarchical state of the government, Italy appears to have been prosperous. In the growing cities there were business men wealthy enough to build themselves elegant houses, and many others living on the income brought them by estates of moderate size, worked, like the great ones of the senators, mainly by slave labour. Among this well-

to-do municipal aristocracy, from which the local. magistracies were filled, must also be reckoned some of the ex-soldiers who were given plots of land on their discharge. We know, for instance, that a number of Sulla’s veterans were settled at Pompeii and became the leading element in the population. In the country the free peasantry was certainly not extinct, and though many suffered from the loss of their holdings as a result of confiscation in the civil wars, their places were taken by the veterans settled on the land by the military leaders. There must also have been considerable numbers of free tenants of the landlords, though in the main cultivation was done by slaves. These tenants were not independent people like the English tenant-farmer, working lands in much the same way as an owner and finding their own markets for their produce, but quite small men forming part of the larger economic unit constituted by the estate as a whole, and very closely dependent on the landlord. _ In the cities there was no great class to correspond with the mass of working-men in a modern industrial state, first because most of the corresponding work was done by slaves, and secondly because industry in the ancient world never developed to the extent to which it has grown in the modern. But there were, of course, numbers of artisans, some working in small shops, which made goods to order for customers, and others making standard articles for 1 Brunt, op. cit. above, 80 n 2; R. Syme, The Roman Revolution (Oxford, 1939) 13£.; and, very fully, Nicolet, L’ordre équestre 247ff. 82

Social conditions and different classes during the republic an indefinite market. The factory system, however, did not prevail to any great extent; most of the articles appear to have been made in small shops and distributed through business men who bought them from the makers; Roman industry never played such a large part in producing great fortunes as did commerce.

The greater part of the working-classes were, however, not free men but slaves, of whom great masses worked on the estates, great and small, and in the workshops of the cities. The status of all these men was legally identical;

they were owned like any other piece of property by their masters, but in practice their positions varied very greatly from the farm-labourer who was often forced to work in chains, who slept in a sort of barracks and was excluded from all possibility of family life, to the slave-bailiff in charge of a great estate owned by an absentee landlord, or the confidential secretary of a Roman of high rank who might, like Cicero’s slave, Tiro, become a ‘ humble friend ’ of the family. Often a man would appoint his slave manager of a business, in which case he acquired the profits but became liable in full on contracts made in connexion with the business. Otherwise, apart from authorisation, he could not be sued on his slave’s contract for an amount greater than the peculium, i.e. the property which, though it remained legally the master’s, the slave was permitted to administer for himself. As the master could appropriate any gains that accrued, he stood to profit without restriction, though his liability was limited. These better-placed slaves were often allowed by their masters to accumulate money in their peculia and purchase their own freedom with their savings, that is to say the master would agree (though such an agreement would not be enforceable at law) that if the slave saved a certain sum he would take that sum and manumit the slave. Such an arrangement naturally acted as a powerful inducement to the slave to work hard. Many slaves were also manumitted from motives of liberality, or indeed of ostentation. This last motive applied particularly to manumissions by will, for the Romans were much addicted to funereal pomp, and it made a good impression if a large number of grateful freedmen followed a man in his last procession. The social and political importance of this practice of manumission was very great indeed, especially as Roman law, more generous in this respect than the Greek systems, gave to the manumitted slave (provided the proper formalities were fulfilled) not only complete freedom but also citizenship.? The result was that from the time when slaves began to be numerous a great part of the citizen population was of servile birth, and, what is still more important, of foreign race, for most of the slaves were prisoners captured in war. Greeks and other races of the eastern Mediterranean especially mingled their stock with that of Italy in this way, for they had the civilisation which 2 See below, 136.

83

Social conditions and different classes during the republic fitted them for the better positions, while the Gauls, for instance, were mostly used for hard physical labour in the fields or mines and had little chance of manumission. The freedman (libertinus),? although a citizen, was not on an entire equal-

ity with the free-born man as regards political rights. He had a vote, but as all freedmen were confined (except for short periods when democratic leaders succeeded in passing laws removing the restriction) to the four ‘ city ’ tribes, there were thirty-one out of the thirty-five tribes which he could not influence. From voting in the centuriate assembly he was no doubt originally excluded, for, throughout the republic, freedmen might not serve in the legions (though

regularly used for the less honourable service in the fleet), but when the comitia centuriata ceased to have any close connexion with the army and was based to some extent on tribal divisions * freedmen were probably included. For magistracies and the senate freedmen were not eligible.°

In spite of their political disabilities freedmen formed a very important class of the population. Most of them came of quick-witted races and it was naturally the most intelligent that secured their freedom, though a youth passed in the abominable condition of slavery did not tend to make them too scrupulous.* They congregated mostly in the towns, where they probably

made up the larger part of the free working population. Some, no doubt, became rich, though there is little sign of the colossal fortunes which became proverbial in the early empire.’ Particular importance attaches to the proletariate in Rome itself. Now that

the citizen body was spread all over Italy it was, of course, impossible for the great majority of voters to attend the assembly except very occasionally. Numbers of out-voters came for the consular elections in the summer, but legislation was practically in the hands of those who happened to reside in Rome. It must be remembered that the comitia remained throughout a primary assembly; the device of representative government was never adopted. Anyone therefore who could keep the populace in the city in a good temper had in his hands the legislative organ of the Roman state. The chief means adopted for this purpose was the distribution of corn by the state at very low prices, a practice which had begun under C. Gracchus. In Cicero’s time every citizen who applied received a ration equal to that of the soldier at a stable and heavily subsidised price. These distributions were made at the expense of the state, but there were other advantages which came out 8 Libertinus is the general word, including ‘Latin’ freedmen; libertus implies citizen-

ship; Buckland, RHD (1923) 293ff. 4 Above, 22.

5 Libertinus originally included the children of freed slaves and ingenwus consequently meant a person whose grandfather had been free, but by the end of the republic anyone who was born in a state of freedom was ingenuus; Gai. 1.11: Ingenwi sunt qui liberi nati

sunt. 6 Nec turpe est quod dominus iubet; Petronius, Cena Trimalchtionis 75. 7 See generally S. Treggiari, Roman Freedmen in the late Republic (Oxford, 1969).

84

Social conditions and different classes during the republic of private pockets, especially the costly games which an aspirant to the higher

magistracies was almost forced to provide if he did not want to receive an unpleasant check to his career.

The result was naturally that the plebs urbana degenerated and that the comitia became merely a machine for registering the wishes of the man who could obtain the greatest popularity at the moment or could bring an army to overawe the population. This last proceeding would, of course, have been quite inconceivable in the earlier republic when the army was almost identical with the population, but a@ momentous change had taken place. Under the older system military service was compulsory and based on a property qualification: the capite censt ® had not been liable. In the third and second centuries B.C. owing to wars, and emigration to the provinces, the number of citizens and especially (owing

to the growth of latifundia) of citizens with the requisite amount of property decreased, and military service also became unpopular with the wellto-do classes. On the other hand there were a number of unpropertied citizens who were only too glad to have a livelihood provided for them in the army and resented their discharge at the end of a campaign. Successive lowering of the census required for enrolment did not meet the case and the final step had to be taken in 107 B.C. after a crushing defeat by the Gauls had annihilated a whole army. Marius, then consul, and one of the greatest of Roman soldiers, opened the ranks to all citizens who cared to enlist, irrespective of any property qualification. The Roman army became as a result a largely mercenary one.® The soldier was a professional; his interests lay with the army and not with the state as a whole, and his loyalty was for the general who led him and on whose good offices he depended for an allotment of land when his time of service was over. It thus became possible for Roman generals to use their armies as instruments for working their will upon the state, and this power, once realised, led through the long agony of the civil war to the establishment of the military autocracy that we call the empire. 8 Above, 21. % But conscription was still common when very large numbers of soldiers were needed ; P, A. Brunt, Italian Manpower 225 B.C.-A.D. 14 (Oxford, 1971) ch. xxXII.

85

CHAPTER 5

Sources of law in the republic The phrase ‘ source of law ’ is used in a number of different senses which will

be found discussed in works on jurisprudence.’ In the sense in which it is used here it means a method by which new rules of law can come into existence. In England to-day there are two chief sources of this nature, Statute and Precedent; if a rule is laid down in an Act of Parliament it becomes a

rule of law; if in deciding a case which raises a new point of law a judge applies a certain principle the decision is a precedent and the principle becomes (subject to certain limitations) a rule of law. The corresponding sources at Rome during the republic were:

1DLEX AND PLEBISCITUM Lex was strictly an enactment of the comitia,? * either the centuriata or the tributa,* while plebiscitum was one passed by the concilium plebis. As soon, however, as its enactments had been put on a level with those of the comitia,° 1 E.g. H. L. A. Hart, The Concept of Law (Oxford, 1961) 246f.

2 This refers only to republican times. The historians indeed represent the kings as authors of various leges (see collection, Bruns 1.1-15; FIRA 1.1-18), and Pomponius even supposes that the kings, like the republican magistrates, put proposals for laws before the assembly (D. 1.2.2.2). It is, however, unlikely that there was any real legislation at so early a date. The rules mentioned may be old, but their ascription to leges is a result of the common error that there can be no law without legislation. A book purporting to contain a collection of leges regiae is said by Pomponius (loc. cit.) to have existed in his day and to have been called ius Papirianwm because it was compiled by a certain Sextus Papirius in the time of the last king, Tarquinius Superbus. Further on, however (§ 36), he speaks of Publius Papirius as the man who made a collection of royal laws, and Dionysius (3.36) says that Gaius Papirius, the first pontifex maximus, restored a collection made by the fourth king, Ancus Martius, of a number of ordinances of the second king, Numa, which had been put up in the forum on tablets and had gradually become illegible. These stories are mere legends, and the collection was certainly made much later than they suggest. In D. 50.16.144 there is a reference to a commentary on the tus Papirianum by a certain Granius Flaceus, and as the only man of that name known is one to whom Caesar dedicated a book on religious formulae (de indigitamentis), the work has been attributed to him. It must be at least not later than 46 B.C., the probable

date of a letter (ad Fam. 9.21) in which Cicero is evidently ignorant of its existence. Careopino, in Mélanges d’archéologie et d’histoire, Ecole francaise de Rome, 54 (1937) 3 In its widest sense lex embraced much more than statute; see above, 69 n 6 with reff.,

and below, 244 n 8. The root meaning of the word is probably declaration; Stein,

19. 5 Above, 24f.

Regulae 9ff.

4 The curiata had ceased to have any political importance in republican times; above, 86

Lex and plebiscitum the concilium plebis became the usual organ of legislation, and laws passed by it were frequently called leges. This practice indeed was so common, even among lawyers, that it can hardly be called incorrect.°®

Apart from the XII Tables themselves, which were of course a lex (indeed in the mouth of a jurist lex without qualification commonly means the XII

Tables), statute was not a very fruitful source of law during the republic. Roman legislation, like our own, was usually of a political character and the

development of private law was, in the main, left to the other sources, especially interpretation and magisterial edict. There were, however, a number of leges (or plebiscita) of constitutional importance which affected private law and some which, so far as we know, had no political significance. Chief among the latter class is the lex Aquilia,’ which recast the whole law of damage to property, and was the basis of all subsequent law on this subject. Among laws which were the result of political struggles we can men-

tion the lex Canuleia of 445 B.C., which removed the prohibition of inter- : marriage between patricians and plebeians,* and the lex Poetelia (? 826 B.C.), which mitigated the severity of the law of debt.® A political character also probably attached to the various statutes dealing with suretyship discussed by Gaius 3.121ff.' A distinction is drawn in a post-classical compilation ? between three different degrees of ‘ perfection ’ which may attach to a law. A lex perfecta for-

bids an act and invalidates it if done; a lex minus quam perfecta does not invalidate the forbidden act but imposes a penalty on the person doing it; a lex imperfecta forbids the act but neither invalidates it nor imposes a penalty. An example of the last class is the lex Cincia (204 B.C.), which forbade gifts above a certain amount except among near relatives, but neither invalidated them nor punished the offender.‘ The lex Furia testamentaria, on the other hand, is a lex minus quam per344ff., held it to be a Pythagorean forgery, made between 46 and 37 B.C. Whatever its. date there can be little doubt that the work really represented, in part at least, genuine pontifical tradition. Its actual title appears to have been de ritu sacrorwm (Servius, Aen. 12.8386, Bruns 2.78), and this fits well with the nature of most of the rules attributed to regal laws of which we hear, for they all belong to the borderland between law and religion. Cf. Schulz, History 89; Kaser, AJ 65f. 6 E.g. Ulpian in D. 9.2.1.1 says Quae lex Aquilia plebiscitwm est, cum eam Aquilius

tribunus plebis a plebe rogaverit. 7 Below, 275.

8 Above, 14. 9 Below, 164.

1 Below, 300. Other important leges are the lex Aebutia (below, 218), the lex Plaetoria (below, 241), leges Atilta and Iulia et Titia (below, 239), leges Furia, Voconia, Falcidia

(below, 247f.). 2 Epit. Ulp. 1.2.

3’ Perhaps even a lex perfecta did not make the act automatically void but merely

voidable; di Paola, Synteleia Arangio-Rui2z 1075ff.

4 But the law was not without effect: the praetor enforced it, within limits, by exceptio and replicatio; ef. below, 207; Buckland 254f.; Kaser, RPR 1.603f.

87

Sources of law in the republic fecta. It forbade legacies of more than 1,000 asses except to certain persons, but did not invalidate them; instead it gave an action for a fourfold penalty

against the person who had received such a legacy.°® y¥

What reason there can have been for passing leges imperfectae is unknown;

it has been suggested that it expresses the idea that lex had no power to alter ius, but more probably it reflects the formalistic conception of legal transactions in early Roman law: once a transaction had been completed in due form, only another appropriate form could undo it. At the end of some leges there was a paragraph called sanctio, which laid down penalties for contravention of the law, but whether we are to imagine that all laws normally contained such a paragraph and that leges imperfectae were characterised by its absence is unknown.’

2INTERPRETATIO The history of Roman law, so far as we really know anything about it, begins with a code, the XIT Tables: No doubt much of the contents of the code was

taken from existing customary law, and much more must have remained unwritten. The development of this amalgam of statute and custom was achieved (apart from the edict) mainly by what was known as ‘ interpretation ’.

According to a unanimous and entirely credible tradition this function was, in early times, exercised by the pontifices, a small ‘ college’ of men who, though they may be described as priests of a sort, did not have to belong to any special caste, except that until 300 B.C.* they had to be patricians. Membership of the college was in no way incompatible with other offices; on the contrary, it was an added distinction to the public career of a member of the aristocracy. Nor must we imagine, on the other hand, that the pontifices were judges; ® their business was, almost certainly, like that of the later prudentes, advisory. They advised the magistrate as to the law, and they also apparently advised individuals; at least, this is very probably what is meant by Pomponius’ enigmatic statement that one of the pontifices 5 Gai. 2.225; 4.23, 6 See Stein, Regulae 14 ff.

7 Cf. J. 2.1.10: Legum eas partes quibus poenas constituimus adversus eos qui contra leges fecerint sanctiones vocamus. The last paragraph of the lex de imperio Vespasiant (Bruns. 1.202; FIBA 1.154) is also headed sanctio. It does not threaten any penalties,

but on the contrary relieves any person acting in accordance with the law from any penalty he might incur under previous enactments. In late Latin, sanctio sometimes means ‘law’ in general, e.g. Const. Deo auctore 2, where Justinian orders the compilers of the Digest to ‘ collect and amend the whole law of Rome "—-omnem Komanam sancttonem. 8 Above, 16. But the first plebeian pontifex maximus was not created until 254 B.C. 9 Broggini, Judex 90ff. And see next note.

88

Interpretatio was appointed every year ‘ to be in charge of private matters ’.’ It is not difficult to understand why it should have been the pontiffs who were the earliest legal authorities. It is true that Roman law took on a secular charac-

ter at a comparatively early stage in its history, but with the Romans, as with all peoples, law and religion were not originally differentiated, and there were many spheres, even after the XII Tables, and in later times, where the tus sacrum, the religious law strictly so called, touched the ordinary civil

law. The pontiffs were the guardians of religious tradition, and, as such, would naturally be the authorities to be consulted in purely legal matters as well. Thus, for instance, the calendar was primarily a religious matter and, as such, regulated to a large extent by the pontiffs; but it was also of great legal importance, as on very many days there were religious reasons _why the magistrate might not sit in his court.” Again, much of the law relating to the family was of a religious character. Adrogations and, originally, will-making could only take place in the comitia curiata meeting under the

presidency of the pontifex maximus, who was especially interested in the preservation of sacra, i.e. the family religious rites, which might be adversely

affected by the changes in the natural order of descent which these acts involved.

By their different kinds of advice, the pontiffs were able to influence the development of the law very considerably. They might even, under the cover of ‘ interpretation ’, create an entirely new institution, as they did in emancipation. The XII Tables apparently provided no method by which a father could voluntarily set his son free from his power, but there was a clause, intended apparently to punish cruel misuse of his rights by the father, which enacted that if the father sold the son three times the son was to be free from his power.’ A triple sale of this sort was possible because if the buyer of the son manumitted him (which he might do just as he could manumit 1 D. 1.2.2.6; qui praeesset privatis, i.e. probably rebus — private law matters as opposed to sacred law matters; Wlassak, Prozessformel 103. Paoli, Mél de Visscher 4.281ff., con-

nects it with (verba) praeire, in the sense of leading or prompting the recital of the appropriate procedural forms by the magistrate and the parties; on this see Bonifacio, Iuwra 2 (1951) 335ff. F. de Martino, La giurisdizione nel dir. rom. (Padua, 1937), and Storia 1.172, goes further and holds that until the institution of the praetorship in 367 B.C. the supervision of litigation was in the hands of the pontifex maximus; but see Kaser, AJ 348ff.; Pugliese, Proc. 1.120.

2 It must be remembered that the Roman calendar was a complicated matter, being an adaptation of an original lunar year, which was brought into harmony with the solar year by the occasional intercalation of an additional month, until Caesar’s reform introduced the system of leap-years.

3 Tab. Iv.2: St pater filium ter venwm duit, filius a patre liber esto. This is the traditional explanation, but other misuses of patria potestas are always visited with sacral penalties, and it has been suggested (Kaser, SZ 67, 1950, 474ff.; ef. Lévy-Bruhl, Now. Et. 80ff.) that the clause was always intended to provide a way of ending patria potestas, the requirement of three successive sales being imposed to mark the seriousness of what was being done.

89

Sources of law in the republic a slave), the son fell back into the power of the father. By ‘ interpretation ’ this clause was used for the purpose of emancipating the son from the father’s power. The father made a pretended sale * of the son to a friend three times;

after each sale the son was manumitted by the friend; after the first two manumissions he reverted to the power of his father, but after the third he Was Sui iuris.

Here we find, as in several other cases in ancient Roman law, the use of a recognised legal proceeding for purposes for which it was never intended; a

pretence of a sale is made in order to achieve objects which have nothing to do with ordinary real sales. But when we come to the emancipation of daughters or grandchildren, there is something further, a definite twisting of the clause in the XII Tables to mean what it almost certainly did not mean. The clause mentioned only sons; it was probably only with reference to them that there was this limit on the father’s power; it was intended that the head of the family should be able to sell the less important members of the family as often as they were manumitted by a buyer. But once the clause

had come to be used to permit of the emancipation of sons, it was interpreted to mean that only in their case were three sales necessary; in the case of daughters or grandchildren one was allowed to suffice; the daughter

manumitted. ,

or grandchild was therefore emancipated by being ‘ sold’ once, and once Equally important with this interpretation was the work of the pontiffs

in shaping the legis actiones, or forms of words used to bring a claim before

the court. Such claims had to follow closely the text of the law on which they were based,* and they had to be exactly correct in every word. Gaius tells us that anyone who made the slightest mistake lost his case,* and he gives an instance’ of a person who was non-suited because, wishing to bring an action under the clause of the XII Tables dealing with ‘ cutting down of trees’ against someone who had cut down his vines, he spoke of ‘ vines ’ instead of ‘ trees’ in his claim. It is clear that the people who could ultimately decide what forms were correct had very great power and that litigation was a very perilous thing for the layman who thought that the justice of his claim was sufficient guarantee of success. The pontifical monopoly of law, for as such it is represented, was one of the great strongholds of the patricians, from whose ranks the pontifices were originally exclusively drawn. One successful attack on it had, of course, been 4 Cf. below, 119. ‘Emancipation ’, the freeing of a child from the power of his father, must be carefully distinguished from ‘manumission’ the freeing of a slave or person in mancipio by his master. 5 This at least is one of the explanations given by Gaius (4.11), but see Stein, Reguiae

7 Gai. 4.11. .

lif.

6 Gai. 4.30.

90

Interpretatio made when the XII Tables laid down a written text, but much remained to be done.

Tradition ascribes the breakdown of the monopoly to the action of a certain Cn. Flavius,* secretary to Appius Claudius Caecus (censor 312 B.C.), and son of one of his freedmen, who is said to have stolen and published a collection of legis actiones made by his master. It became known as the

tus Flavianum; and Flavius’ act was so much appreciated by the people that they elected him tribune and then curule aedile (304 B.C.), in which latter capacity he put up a copy of the calendar in the forum, so that anyone could see for himself on what days an action might be brought. As Appius Claudius, though belonging to a great patrician house,’ was himself

a democratic innovator, and is mentioned as a jurist of note, it is likely enough that Flavius’ publications were really made at his instigation. It is also highly probable that we should connect the popularisation of law with the opening of the pontificate to the plebeians by the lex Ogulnia of 800 B.C. ; at any rate it was the first plebeian pontifex maximus, Tiberius Coruncanius

(254 B.C.), who, according to Pomponius, primus profitert coepit.' Exactly what these words mean we do not know, but they clearly refer to some sort of public activity, perhaps to the admission of members of the public. generally, and especially of students desirous of learning the law, to his consultations. At any rate, there now came into existence a class of men known as iuris consulti or iuris prudentes, persons ‘ learned in the law ’, who made law their speciality. For the jurists of the earlier republic almost our only source of information is the long fragment from Pomponius’ liber singularis enchiridti? (single volume handbook) preserved in Digest 1.2.2. The fragment is probably the whole of the ‘ historical introduction ’ to the handbook and falls into three parts, the first ® dealing with the origin of the law, the second * with the different magistracies, and the third ° with the prudentes. The text is unfortunately very bad; some of the authorities that Pomponius used, especially

for the most ancient period, were unreliable, and, like other Roman historians, he is fonder of picturesque anecdote than accurate detail, but the value of the fragment is still very great. For the later republic we have, in addition to Pomponius, a number of references in Cicero’s philosophical works as well as allusions to his contemporaries in the letters and speeches, and there are references in the legal writers of the imperial age. 8 On Flavius see Pais, Ricerche 1.217ff., summarised with reference to other literature by Kreller, SZ 45 (1925) 600ff.; Schulz, History 9f., is highly sceptical about the whole

tradition. | 9 He was a great-grandson of the decemvir. 1 PD. 1.2.2.38. Cf. ibid. 35: Ha omnibus qui scientiam nancts sunt ante Ti. Coruncanium publice professum neminem traditur. Schulz, loc. cit., again is sceptical.

History| 168f. 3 Pr. —12. 42 Schulz, 13-34, 5 35-53. 91

eerie Sources of law in the republic

After Papirius ®° Pomponius first mentions Appius Claudius the decemvir and then passes immediately to his great-grandson Appius Claudius Caecus,’

who, he says, was the author of a lost treatise de usurpationibus (the inter-

ruption of prescription).* Next come P. Sempronius (consul 304), called * the wise ’;° P. Scipio Nasica,' who received the surname Optimus and was given

a house in the via sacra at the public expense so that it should be easier to

consult him; and Q. Maaimus, of whose legal work nothing is said. Of Tiberius Coruncanius ? Pomponius says that none of his writings had survived, but that his responsa were ‘ numerous and memorable ’. Sextus Aelzus Paetus * (consul 198) is next mentioned. He was the author of the tripertita,

so called because ‘ the law of the XII Tables came first, then followed the interpretatio and finally the legis actio ’. This may mean that each clause of the XII Tables was given separately and followed immediately by its juristic developments and the appropriate legis actio, or that the three parts were separate; we have no means of judging. Sextus Aelius is also said * to have compiled the ius Aelianum, a collection of legis actiones which superseded the ius Flavianum, but whether this was identical with the tripertita or not is unknown.® The tripertita still existed in Pomponius’ time and were, he says, called ‘ the cradle of the law’. With Sextus Aelius are mentioned his brother, P. Aelius (consul 201), and P. Atilius, who was surnamed sapiens. After the Aelii come M. Porcius Cato (censor 184) and his son who bore the same name and died before his father, in 152. The elder Cato is the famous

statesman, but the younger was the more important as a lawyer, and more books of his than of his father’s survived. There follow three men of whom Pomponius says that they ‘laid the foundations of the civil law’, which probably means that they were the first to write books which were not mere collections of forms, but contained independent discussions. The three were P. Mucius Scaevola (consul 138 and later pontifex maximus), M. Iunius Brutus (praetor in 142, but not consul) and M’. Manilzus (consul 149). Brutus’ work, which was called de ture civili,** was in part in the form 6 Above, 86 n 2. For biographical information on the jurists see Kunkel, Herkunft;

for their work see Schulz, History. 7 Above, 91. 8 Apocryphal according to Schulz, History 9; but see Mayer-Maly, Mnemosyne P.

Bizoukides (Salonica, 1960/3) 221ff., who suggests that the usurpatio in question was the trinoctis absentia (below, 116), which would have acquired social and political significance once marriage between the orders became possible and thereby a plebeian might acquire manus over a patrician wife. ® Pomponius says that the people called him cogds, but the people would hardly choose a Greek name. Pomponius is simply copying a Greek source; Kipp 98.

1 Pomponius confuses him with his father, whose praenomen was Gaius; Wenger, Quellen 487 n 140.

2 Above, 91. 3 Surnamed catus (sagacious).

4D. 1.2.2.7. 5 If the three parts were separate it might be identical with the third. 6* Cie. Cluent. 141; de Or. 2.142, 224; Schulz, History 92f.

92

Interpretatio of a dialogue with his son, but also contained all his responsa. Manilius’ monumenta are said by Pomponius to have survived in his time and it was probably the same man who was also author of a collection of forms for contracts of sale (Manilianae venalium vendendorum leges) mentioned by Cicero.’

After the three ‘ founders of the civil law’ came P. Rutilius Rufus (consul 105), a disciple of the Stoic philosopher Panaetius, and probably the originator of the Rutilian type of action;* A. Verginius, of whom nothing is known;*® Q. Aelius Tubero, like Rutilius a disciple of Panaetius; Sextus Pompeius, uncle of Pompey ‘ the Great ’, and Caelius Antipater, who, Pomponius says, wrote histories and gave more attention to oratory than to law. The next name is that of Q. Mucius Scaevola (son of P. Mucius), an active

statesman, consul in 95 and, like his father, pontifex maximus; he met his

death at the hands of the Marian party in 82. With him we enter a new period, and one of which more is known. The writings of the earlier jurists (though copies of some of their works survived) were known in the classical age almost exclusively through quotations, but those of Q. Mucius were clearly read in the original, and even in the Digest there were some excerpts taken directly from his works, while citations from him in the excerpts taken from other writers are innumerable. He was, moreover, the first jurist to be plainly influenced by Greek dialectic. Pomponius says of him that he was the first to * arrange the tus civile in genera ’, and we can obtain some notion of what this means from Gaius’ statement that he distinguished five kinds of tutela.? His chief work, eighteen books iuris civilis, was the first systematic legal treatise ever produced and became the basis of most of the later works on the 2zus civile.* In addition he wrote a liber singularis épwv, i.e. of definitions,* but containing rules as well, which is used in the Digest. Some modern scholars would make him the founder of a school of jurists, subsequently continued by Labeo and the Proculians.* In any case he had many pupils, chief of whom was C. Aquilius Gallus, a colleague of Cicero’s in the praetorship (66 B.C.) * and originator of the formulae de dolo™ as well as of the Aquilian stipulation.** A pupil of Gallus was Servius Sulpicius Rufus * (consul 51), whose influence on the development of the law was as great as 7 De Or. 1.246. Some are preserved in Varro, #.R. 2.5.11 (Bruns 2.63). They may have formed part of the monwmenta.

8 Gai. 4.35. Below, 209 n 1. ® See Schulz, History 47.

1 Schulz, History 62ff.; Stein, Regulae 36ff. 2 1.188.

3 For its plan see Schulz, History 94f.; cf. M. Lauria, Ius Romanwm I.1 (Naples,

1967). 5 Below, 380. 7 Cic. Off. 3.60; de nat. deor. 3.74. 8° J. 3.29.2. 1963), reviewed by Wieacker, Jura 16 (1965) 269. And see below, 376.

4 Stein, Regulae 36ff. Or perhaps distinctions; Albanese, Studi G. Scaduto (Palermo, ¢ He was president of the quaestio de ambitu (bribery court) ; Cic. Cluent. 147.

-% Stein, Regulae 41ff. For a biographical study see P. Meloni, Servio Sulpicio Rufo e ¢ suot tempt (Cagliari, 1946).

93

Sources of law in the republic that of Q. Mucius.! He was a friend and rival of Cicero’s in the courts, and, it is said, only took up the study of the law on account of a reproach levelled at him for his ignorance by Q. Mucius.? According to Pomponius, he left ‘nearly 180 books ’, several of which were still extant. His works included reprehensa Scaevolae capita, i.e. corrections of Q. Mucius’ views, a book on dowries, and the first commentary on the edict in ‘ two very short books ’. His pupils were many, among them A. Ofilius, who was a friend of Caesar and survived into Augustus’ time. He remained a member of the equestrian

order and is especially noteworthy as having written the first full commentary on the edict. Another pupil of Servius’ was P. Alfenus Varus, said to have begun life as a bootmaker, who became consul in 389 and attained the honour of a funeral at the public expense. His works included Digesia

in forty books, two epitomes of which are used in Justinian’s Digest. A number of other pupils of Servius are mentioned by Pomponius, of whom scarcely anything is known except that the works of some of them were collected by a certain Aufidius Namusa. More important were A. Cascellius, who was still alive in the time of Augustus, and may have been the creator of the zudicium Cascellianum,’ C. Trebatius Testa, a friend of Cicero’s who enjoyed a great reputation as a jurist and was consulted by Augustus as to the advisability of enforcing codicils,* and Q. Aelius Tubero, who is said to have become a jurist only after prosecuting Q. Ligarius before Caesar and failing to secure his conviction. He achieved a reputation for learning in

both private and public law and left books in both branches. That these turisprudentes were not professional lawyers in our sense is clear; not only did they not receive any remuneration for their services, but they were public men who devoted only some of their time to law, and indeed

did so as part of their public career. As we have seen, many of them were consuls, which means that they had gone through the whole cursus honorum, and some were distinguished as generals and as provincial governors. Tiberius Coruncanius was the leader of an army which withstood Pyrrhus in 280; Sextus Aelius was in command of an army in the first Macedonian war, and Q. Mucius’ tenure of the proconsulate. of Asia was regarded as a model of what a governor should do.* In some cases, no doubt, the chief stimulus to

learning was the hope that, by giving legal aid to citizens, the jurist might gain the popularity needed for success at the polls.* Towards the end of the

1 Cf. below, 380. 2 D. 1.2.2.43.

3 Gai. 4.166, 169.

4 J. 2.25 pr. 5 He was also one of the proposers of the unfortunate lex Licinia et Mucia which precipitated the Social war; above, 65.

6 Witness the story of C. Figulus, who, being disappointed at a consular election, afterwards refused to give legal advice, sending clients away with the pun, An vos consulere scitis, consulem facere nescitis? (Val. Max. 9.3.2). 94

Interpretatio republic the legal career was apparently becoming slightly more specialised, though the chief lawyers were still men who held high office, for we hear that Aquilius Gallus refused to stand for the consulship in order to devote

himself more intensely to law,’ and that he retired at times to the island of Cercina where he wrote several of his books.* Cicero,’ discussing what qualities are needed for an ideal jurisconsult, says that he must be one who is skilled ‘ ad respondendum et ad agendum et ad cavendum in all matters of law and custom that can be needed by private people in the state’. Of these activities,’ respondere is the most important. It means giving advice in the sense in which lawyers use that word, i.e. especially advice as to what the law is, and such advice might be given either to a private individual, as with our ‘ opinions’ of counsel, or to a judge who was trying a case, for it must be remembered that the judges (iudices) at Rome were not, like our judges, professional lawyers, but laymen, more like our jurymen, except that they generally sat singly and belonged to the wealthier classes of the community. Such a judge, if in doubt as to a point of law, might very well wish to ask the opinion of a jurisconsult, and would be almost certain to follow the opinion when he got it, though, during the republic, there was no compulsion to do so. It might also happen that a litigant had taken the opinion of a jurisconsult and gave evidence of this opinion before the judge who was trying his case.” The result was that these opinions were, in effect, very much like decisions, and, though Roman law did not attach any binding force to precedent, the opinions of the jurists helped to mould the law

in a manner not entirely different from that in which judgments mould English law. From the beginning of the empire the importance of responsa as a source of law was to become much greater, but they were already among the sources during the republic. As a rule, no doubt, responsa were given for

an actual case, whether one which led to litigation or not, but this was not necessary. A purely hypothetical case might be raised, for instance, by a pupil, discussed and decided by the jurist, and, since there was in any event no formality, the influence of the decision might be equal to that of one given

on real facts. That discussions in the circle of a jurist did take place is evidenced by Cicero,? but whether the curious phrase disputatio fort used as a synonym for interpretatio,* i.e. for the law created in this way, has any connexion with this practice is uncertain.

7 Cie. ad Att. 1.1.1. , 8 Pomponius, D. 1.2.2.43.

9 De Or. 1.212. 1 See also Schulz, History 49ff.

2 Pomponius, D. 1.2.2.49: Ante tempora Augusti...(prudentes) ... plerumque iudictbus tpsi scribebant, aut testabantur qui illos consulebant. But see also below, 359f. 3 Top. 56: Vestras in respondendo disputationes. 4 Pomponius, D. 1.2.2.5, where it is also said that ius civile was sometimes used in this restricted sense, just as we speak of ‘common law’ sometimes as opposed to statute law, and sometimes including it as opposed to equity. 95

Sources of law in the republic Under respondere in its widest sense can also be included advice given to magistrates in connexion with their legal duties, for the magistrates, though as public men they would have some knowledge of law, were only exceptionally- experts. Especially in the highly important work of drawing up their edicts they must have been assisted by jurists, so that the edictal law, as well as the civil law proper, is in fact largely the work of the prudentes. Of the other jurisprudential activities mentioned by Cicero, cavere means the drafting of legal forms for contracts, wills and other transactions where expert help was needed.* Perhaps we should include under it what Cicero elsewhere ° calls scribere, the formulation of written documents, but it must be remembered that many important transactions, which would with us be embodied in written documents and merely signed by the parties, had in Roman law to be concluded by the spoken word (stipulation), even if a document was also prepared as evidence of what had been said.

Agere refers to assistance in litigation — help on points of procedure, the drafting of forms to be used by the parties to a lawsuit especially, if modern

authorities are right, the drafting of the formula (in the technical sense), which ‘under the ‘ formulary ’ system of procedure formed the basis for the trial of the action by the tude.” Advocacy proper was not the business of the jurisconsult but of the orator, who, though of course he might also be a jurist, was much more often not legally trained. According to ancient ideas, the training needed was one in oratory itself, and many young Romans went to the Greek schools for the purpose of receiving this sort of education; the strictly legal knowledge necessary for arguing points of law involved in his case the orator was expected to

law. ,

get from an expert on each occasion, though in course of time he would

naturally pick up a fair amount for himself. Cicero, for instance, though he did not in the least consider himself a lawyer, obviously knew a great deal of In addition to his immediate practical activities of respondere, cavere and agere, the jurist, as we have seen, sometimes spent part of his time in writing

books on legal subjects. This was the only type of literature in which the Romans were independent of Greek models. In Greece there was nothing between the generalities of philosophers who were not interested in the detailed system of any one state and mechanical collections made by attorneys.*® 5 Cavere literally means ‘to take precautions’, hence to draft a form of words which seeks to take precautions against various eventualities, as legal documents do. N.B. especially the use of the word as a synonym for stipulari, because stipulation, being (unlike the informal contracts) capable of giving binding force to any sort of engagement into which the parties might wish to enter, was always used where elaborate provisions had to be made. The noun cautio very frequently means the written document which is evidence of the stipulation; cf. below, 418, 509f.

6 Mur. 19. , 7 Below, 200. 8 P. Hal. 1 is believed to be such a collection; see p. 26 of the edition. 96

Edicta magistratuum The Romans, as we have seen, were influenced by Greek dialectical method, and they borrowed some of the philosophical generalities, but these remained

without much effect in practice, and Roman legal literature was above all practical; the jurist never loses sight of the facts of Roman life and the way in which they had to be treated in the forum. Of republican literature, indeed, nothing has survived except a few isolated fragments from Q. Mucius, but it is clear from the great period which begins with the empire that a solid foundation must have been laid during the republic. The days of mere collections of forms like the tus Flavianum and the ius Aelianum were over long before it ended. Detailed and systematic works of an original character were written, though these too, no doubt, were constructed largely on the ‘ casuist ’ principle, i.e. on the discussion of individual cases (actual or hypothetical) which continued to characterise Roman legal literature throughout, except in introductory works for students. Of such works there were very few in this period.° That some jurists were also active as teachers has already been mentioned incidentally. Generally this teaching was confined, in republican times, to the admission of young men to consultations and to the discussions which accompanied them,’ and some preparation was provided by a knowledge of the XII Tables, which in Cicero’s boyhood were still regularly taught at school.? It

is presumably to the presence at consultations that Pomponius is referring when he speaks of one jurist as having been the auditor * of another. Regular lecturing by professional teachers did not exist until the empire.*

8 EDICTA MAGISTRATUUM All the higher magistrates had the right to issue edicts, i.e. proclamations in which they notified the people of their orders and of their intentions, each naturally within his own sphere.® From the edicts of those whose duty included jurisdiction, especially from that of the praetor urbanus,® there arose ® Brutus’ dialogue with his son (above, 93) was presumably one; Schulz, History 93f. 1 Above, 95; Schulz, History 55ff, 2 Leg. 2.59: Discebamus enim puert XII ut carmen necessariwm, quas tam nemo discit.

3 E.g. D.1.2.2.42: Mucti auditores fuerunt complures, sed praecipuae auctoritatis Aquilius Gallus, Balbus Lucilius etc.

4 Of Ser. Sulpicius Pomponius says (§ 43) that he ‘heard’ several jurists but was institutus a Balbo Lucilio instructus autem maxime a Gallo Aquilio —instituere referring presumably to elementary instruction, instruere to more advanced, but this is insufficient ground for supposing that there was a systematic division of legal education into stages; Kunkel, Herkunft 337 n 714. 5 We know, e.g., of an edict issued by the censors of 92 B.C. in which they denounced

the growing practice of attending schools of rhetoric as contrary to mos matorwm (ancestral custom). 6 But also that of the praetor peregrinus, the provincial governors, the curule aediles (cf. above, 49) and the quaestors, whose position in the senatorial provinces was analogous to that of the aediles at Rome; Gai. 1.6.

97

ee ee eel Sources of law in the republic

the ius honorarium ’ or magisterial law, which came to be placed side by side

with the ius civile arising from statute and interpretation, and was interwoven with it in a way which, in spite of important differences, may be compared with the manner in which the common law and equity have combined to make up the English legal system. In the case of the praetor and the other

jurisdictional magistrates, it was the practice that the edict should be published each year when they entered upon their office and put up in a conspicuous place in the forum.’ As it was intended that it should be valid throughout the year, it was called perpetuwm (continuous).® Each praetor had, in theory, a perfectly free hand in the matter of his edict, but it became customary for him to take over and republish as his own the bulk of his predecessor’s edict, making only such erasures or additions as he or his technical

advisers saw fit, and there thus grew up a document of considerable size, known as the edictum tralaticium, because it was thus § carried on’ from year to year; it was this document on which the jurists wrote commentaries.’ There was originally no compulsion on the magistrate to adhere to the intentions he had expressed in his edict;? presumably the pressure of public opinion

was enough, but towards the end of the republic, when, as we know from Cicero’s account of Verres’ misdeeds, unscrupulous magistrates did not hesi-

tate to misuse their powers in their own or their friends’ interest, a lex Cornelia of 67 B.C. was passed forbidding praetors to depart from their edicta perpetua.®

The praetor was entitled to issue edicts and, in fact, these edicts were a very

important source of law, but the praetor was not a legislator; he could not alter the law directly and openly * as could the sovereign assembly by a lex or a plebiscitum, and his edict consequently did not take the same form as a statute. It consisted, on the contrary, chiefly of statements by the praetor of what he would do in certain circumstances, of the way in which he would carry out his duty of jurisdiction, and it was the great freedom he had in this 7 From honor=magistracy; the term does not occur before Hadrian’s time: A. MagdeJain, Les actions civiles (Paris, 1954) 63ff. Volterra, Scritti U. Borst (Padua, 1955) 17ff,, Iura 7 (1956) 141ff., argues that the aedilician edict did not create ius honorarium; but see Guarino, Ordinamento 385ff. (=Labeo 1 (1955) 295ff.; 2 (1956) 352ff.). 8 It was written on white boards, hence albwm.

9 It was also open to the praetor to issue further edicts during the year if occasion arose. Cicero once speaks of such an edict as repentinum (Verr. 11.3.36) but this was

not a technical term. } 1 Above, 94. 2 Except the intercessio of a colleague or a tribune, which was always possible. Cicero relates that it was constantly used when Verres was praetor (Verr. 11.1.119). 3 Ascon. in Cornel. (Bruns 2.69). This cannot be taken strictly, since the praetor evidently continued to grant actions on the merits of an individual case (so-called decretal

actions) and also to refuse actions which were set out in the edict (see immediately below). See B. Vonglis, La lettre et Vesprit de la loi (Paris, 1968) 187ff., who suggests that such grants were permissible if made on the auctoritas of jurists. 4 As to this see Buckland, Tulane L.R. 13 (1939) 163ff. 98

anit teh Edicta magistratuum

respect that made it possible for him to influence the law to such an enormous extent. He would thus say that in such and such a case he would give an action (iudicium dabo), i.e. if a man came to him with a complaint against another which did not, at civil law, give him any claim against that other for redress, the praetor might nevertheless allow him an action. Or the praetor might say that in certain circumstances he would put a man into possession of property (possidere iubebo, bonorum possessionem dabo), or that he would put a man back in his original position (in integrum restituam), i.e. account

some transaction, e.g. a contract into which the complainant had been induced to enter by fraud, as never having taken place, and so on. The praetor could also refuse to allow a plaintiff to proceed with his claim (denegare actionem) if he thought that justice required this.° The essence of the praetor’s power lies in fact in his control over remedies. He does not give

a right (as a law can), he promises a remedy, and once there is a remedy there is, by implication, a right also. This is perhaps most clearly seen from an example taken from the law of inheritance. If a man died intestate leaving

no children and no relations,® then there was, at civil law, no heir to his estate and it was open to anyone to seize the property without any fear, so far as the civil law was concerned, that any other person would be able to bring an action to get it away from him. The praetor was of opinion that in these circumstances the widow of the deceased’ should have a claim. He could not however assert that she was heir; at civil law she was not, and he could not alter the civil law, but he could, and did, say in his edict that he would ‘ give her possession of the goods’, i.e. that he would give her a remedy ® by which she could get the property of the deceased from. anyone who had taken possession of it, and since there was no-one who could show a better title she would be able to keep what she had got. We can thus say that by the ius honorarium the widow had a right, although she had none at civil law, and though the praetor did not, in so many words, say that she had one

at all. This is, of course, but one example of a whole complex system of bonorum possessiones, all of which taken together form the praetorian law of succession which became engrafted on the civil law rules of hereditas. In a

similar way, as we shall see, it becomes possible to speak of praetorian (‘ bonitary ’) ownership, i.e. ownership protected by praetorian remedies in opposition to ownership strictly according to the civil law (dominium ex ture Quiritium), and the parallelism extended throughout the whole legal system. How the civil and praetorian rules worked in with each other in practice is of 5 Kaser, ZPR 178.

6 Apart from the gens; not everyone belonged to a gens and in any case the right of the gens to inheritance on intestacy disappeared at the end of the republic or beginning of the empire; see Gai. 3.17 and below, 125. 7 The civil law took no account of the widow unless she had been married with manus;

ef. below, 124. , 8 Below, 252f. 99

Sources of law in the republic course a matter for detailed study in each instance, but one famous remark on their relationship may be explained here. Papinian says that the function of the ius honorarium is to ‘ aid, supplement or correct’ the civil law.°® Again the system of inheritance, of which Papinian is probably thinking, helps us to see best what is meant. ‘ Aiding ’ refers to the provision of praetorian remedies in addition to those of the civil law for the use of a person who has a civil law right; thus the interdict quorum bonorum (like all interdicts a praetorian remedy) was in many cases available for the person who was civil law heir. ‘ Supplementing ’, though it is not possible to draw a hard and fast line between it and ‘ aiding’, refers especially to the granting of remedies to persons chosen according to the praetorian system in default of any who had rights at civil law, as in the case of the widow mentioned above. ‘ Correcting ’ occurs when the praetor gives remedies to a person who is not entitled at civil law although there does exist someone who is so entitled, e.g. @ person nominated heir in a will which satisfies praetorian but not civil law requirements will be preferred to the intestate heir, who, since the will is invalid at civil law, is by the civil law entitled to succeed.? The right to issue edicts is no doubt in a sense immemorial, but it does not

follow that the praetor had from the beginning of things the great powers which we have just outlined.? It may indeed be regarded as certain that in very early times the primitive system of legis actzones had to suffice for all purposes without modification; in other words that the civil law alone prevailed. It is also certain that the praetor’s power must have received a con- . siderable addition when the lex Aebutia* extended the formulary system of procedure, for that system substituted for the rigid and unalterable legis actiones a method of trying cases by a form of words which depended on the

praetor. What scholars are not agreed about is the amount of power which the praetor had in the later stages of the pre-Aebutian period, some believing that he was then still something of an automaton, so that if the proper words were spoken by the plaintiff he had to let the case go on and could neither stop it nor influence it, others that even before the lex Aebutia the praetor had a very free hand. This question however must be deferred for later consideration.

Of the edicts other than that of the praetor urbanus and the curule aediles we know little. The praetor peregrinus, seeing that he dealt with foreigners to whom the ius civile and the legis actio system did not in any case apply, 9 D. 1.1.7.1.: Ius praetoriwm est quod praetores introduzerunt adiwvandi vel supplendt vel corrigendi iuris civilis gratia. 1 But the praetorian successor will not always keep what he has got by these remedies ; below, 253. Kelly, Irish Jurist 1 (1966) 341ff., conjectures that it was only after the lex

‘correct ’ the civil law. 2 Gai. 4.11.

Aebutia, and probably by virtue of a provision of that lex, that the praetor could 3 Date unknown, but almost certainly second century B.C. Cf. below, 219ff.

100

Custom

must have had a very free hand from the beginning, and it is probable that many innovations were first made in his edict and only subsequently taken over into that of the urban praetor. At any rate the two seem to have corresponded to a great extent by the time of Cicero, who says that he intends to model a part of his own edict as proconsul of Cilicia on those of the praetors at Rome,* which would only be possible if they were similar to each other. Most of our knowledge about the provincial edicts comes from what Cicero says in this connexion and from references to the Sicilian edict in the speeches against Verres. His remarks on his own intentions with regard to Cilicia show

us what an enormous amount was left to the discretion of the individual governor. He says, for instance, that he will follow the example set by Q. Mucius Scaevola in the edict he promulgated as proconsul of Asia, and allow the Greek cities to live according to their own law — which shows that, if he had wished, he might have substituted some entirely different system during his term of office.

4. CUSTOM Roman law, like that of other nations, of course originated in custom, but the part played by custom as a source of law in historical times is comparatively small. Nevertheless, it was responsible for the introduction of some rules — Gaius, for instance, tells us that the legis actio per pignoris capionem was used for the enforcement of some claims in accordance with statute, in others by custom,* and Cicero enumerated mos * among the sources of law. But the republic had no definite theory in the matter, and the discussion of imperial views must be postponed.’ 4 Ad Att. 6.1.15. On this text see Pugliese, Synteleia Arangio-Rwiz 972ff.; Marshall, AJPhil. 85 28. (1964) 5 Gai. 4.26, 6 Top. 7 185ff. Below, 353ff.

101

CHAPTER 6

Law for foreigners, 7us gentium and wus naturale The strict theory of Roman law which remained throughout its history was that the ius civile was only for citizens, and, as there was originally no other law than the tus civile, the foreigner was both rightless and dutiless. It was

open to any Roman to seize him and his property as things without an owner, and, on the other hand, there was no court in which he could be sued. Whether there was ever a time at which practice was entirely in consonance with this theory may be doubted, but at any rate so soon as intercourse with

other states became at all common and civilisation advanced, such a barbarous system could no longer be maintained. We have seen that an exception was made in the case of members of the Latin league, whose admission to commercium means that their rights were protected at Rome, and very

early history already provides an example of a treaty with a foreign state which guarantees mutual protection of legal rights, at least so far as they arise out of commerce. This is the treaty said to have been concluded in the first year of the republic’ (509 B.C.) between Rome and Carthage. Unfortunately we have no record of the nature of the protection granted in this treaty, but other treaties appear to have made provision for rectperatio,* i.e. the appointment of a court of several jurymen (recuperatores), perhaps taken from nationals of both the states concerned. Finally, quite apart from special treaties, the foreigner was no longer in fact treated as rightless;' to treat him so would have been to put a stop to the possibility of commerce, and the commercial interests of Rome were growing. We have seen * that about 242 B.C. a special praetor was appointed to deal with disputes in which foreigners were concerned, and from our accounts there can be no doubt that the single praetor had been dealing with such cases, as well as those in which citizens alone were involved, for some time previously. About the same time, it must be remembered, Rome also acquired her first provinces,* which meant 1 This is the date given by Polybius (3.22). Reasons for accepting it are given by Last, CAH 7.859ff.; Walbank, Polybius 1.337£. 2 Reciperatio est...cwm inter populum et reges nationesque et civitates peregrinas lex convenit, quomodo per reciperatores reddantur res reciperenturque, resque privatas inter

se persequantur (Festus, s.v. reciperatio, Bruns 2.30). Their competence, as this text suggests, was probably restricted to the recovery of property seized in war; Schmidlin, Rekuperatorenverfahren 3ff.; Pugliese, Proc. 2.1.41n. Cf. below, 103 n 7. 3 Above, 48. 4 Above, 68. 102

Law for foreigners, ius gentium and ius naturale that her governors would have to concern themselves with jurisdiction abroad in which foreigners would necessarily be involved. Two questions arise in connexion with the courts in which a Roman magistrate exercised jurisdiction over foreigners; first as to the procedure used and secondly as to the law applied. The strictly Roman legis actio procedure was almost certainly not available,* and the magistrate would have to find some method for himself. This may be one of the roots of the formulary procedure.°

The law of course could not be the tus civile pure and simple because that applied to citizens only, and probably the foreigners concerned were far from

eager that it should be extended to them in its entirety; its cumbrous formalities, in which a single slip might mean disaster, would be particularly unattractive to those who came from Greek states and were used to a more developed and freer system.:The problem might have been solved by applying the principle of personality, i.e. of judging a man according to the law of his

own state,’ and to some extent this was done. Foreigners belonging to the same state were, we know,® generally allowed to settle their disputes according to the law of that state when they came before the local courts, and it is probable that the same principle applied if the matter was litigated before the

Roman governor’s tribunal.® But if this principle had been consistently carried through it would have been necessary to develop rules of § conflict of laws ’? in order to decide in doubtful cases, where a Roman and a peregrine

or peregrines of different states were concerned, which system was to be applied. Such rules did not come into existence,’ partly perhaps because of the multitude of states whose systems the Roman courts would have had to notice, but more especially because, as the Roman power grew, so the ‘foreigners ’ concerned were more and more generally, in fact, subjects of Rome, with the niceties of whose law the Roman magistrate might get a little

impatient. In any case there did grow up, through the edicts of the praetor peregrinus and the provincial governors, a system which was neither the Roman ius civile nor a code of ‘ private international law’, but a general system of rules governing relations between free men as such, without reference to their nationality. Much of this system of law, seeing that it was based on the edicts of Roman magistrates, was Roman in origin, but it was Roman law stripped to a great extent of its formal elements, and influenced by other, especially Greek, ideas. Thus the Roman contract of stipulation was one of

the institutions extended in this way to foreigners, i.e. a foreigner can be bound and entitled under it,? and it is not difficult to see why, for although

5 See Kaser, ZPR 45. 6 See below, 219ff.

7 Above, 63, 71ff. 8 Above, 70 and 100f. 9 Mitteis, Reichsr. 125.

1 Mitteis, Reichsr. 123; Wesenberg, Labeo 3 (1957) 227ff. finds occasional steps in this direction. 2 Provided he did not use the word spondeo, which was reserved for citizens; Gai. 3.93. 103 J.—od

Law for foreigners, 1us gentiwm and ius naturale the stipulation is what we call a formal contract,*® the forms required are the simplest imaginable, and the contract is useful for all manner of purposes. Mancipation,* on the other hand, with its elaborate ceremony, involving the use of scales and bronze and the speaking of set words, remains exclusively a transaction of the ius civile. It must also be noticed that the rules of which we are speaking refer almost exclusively to transactions inter vivos; matters of family law and of succession remain under the personal law of each particular man.° That the phrase tus gentiwm was ever applied to these rules developed in the peregrine and provincial edicts cannot be shown, but it is clear that they were of great importance in leading up to the conception of the tus gentium, for, once established, they in their turn influenced the development of the law

as applied between citizens, especially in the direction of making it less formal, and thus there came into existence the ius gentium, in its practical sense, i.e. ‘ that part of the law which we apply both to ourselves and to

property. :

foreigners ’. In this sense there is a great deal of law which is iuris gentium, e.g. not only the stipulation among contracts, but all the informal contracts, both ‘ real’ and ‘ consensual ’,* and the informal methods of acquisition of

In the sources this ‘ practical’ sense of the phrase tus gentium is not

always clear, because the Roman writers themselves do not distinguish it from a rather different sense which is not practical but derived from Greek philosophical theories. Aristotle, speaking of law in general, had divided it into two parts, that which was ‘ natural’ (@vo.xdv) and that which was manmade (voutxév),” and he asserted that natural law was the same everywhere

and had equal validity everywhere; as well as being ‘ natural’ it was ‘common ’ (xowdv).2 This idea became a commonplace, especially among the Stoics, with whose ideal of a ‘ life according to Nature ’ it of course fitted

admirably. Cicero repeats lofty sentiments about the law of Nature in a similar strain; he does not, any more than the Greeks, get beyond elementary

moral rules when he gives instances of the precepts of this law,’ but it is clear that for him, as for Aristotle, the universality of a principle is a proof of its naturalness and hence of its validity, for the law of Nature is no mere ideal, it is a binding law and no enactment of the people or senatus-consult 3 I.e. a particular form, in this case an oral question and answer concluded in corresponding terms, is necessary for its validity. 4 Below, 148.

5 E.g. the question of patria potestas can only arise if both father and child ure Roman citizens. No foreigner can take under a Roman will; whether a Roman can take under a foreign will is a question for the foreign law. 6 Below, 285ff., 288ff. 7 Eth. Nicom. 5.7.1.

$ Rhet. 1.13.2. 9 E.g. de inv. 2.65ff.; 2.161. 104

Law for foreigners, ius gentiwm and ius naturale

can prevail against it. The argument, though not put in these words, is obvious: if all races of mankind acknowledge a practice it must be because

it has been taught them by their universal mother, Nature. Cicero thus identifies the law of Nature with the ius gentium,? in the sense of law common

to all peoples, and draws the inference that what is part of the tus gentium should also be part of the zus civile,* i.e. of the law of each particular state, although what is zus civile is not necessarily zus gentium, for, as in Aristotle’s view, there are matters on which Nature is indifferent and each community can lay down rules for itself. This theoretical view of the ius gentium as law

common to all mankind became current coin among the jurists, though it does not occur in the surviving sources until Gaius,* who indeed begins his Institutes almost in the words of Aristotle, ‘ All peoples who are governed by laws and customs apply partly their own law, partly law which is common to all mankind; for the law which each people has made for itself is peculiar to

that people and is called its ius civile, the special law of the state; but that which natural reason has appointed for all men is in force equally among all peoples, and is called ius gentium, being the law applied by all races. Thus the Roman people applies partly its own law, partly that common to all men.’

The difference between this ‘ theoretical? meaning and the $ practical ’ meaning of which we spoke above, is best seen when we consider the correla-

tive term to ius gentium. In each case it is ius civile, but in translating this phrase into English we have to differentiate. Where the ‘ practical ? meaning is In question, we say ‘ civil law ’, meaning ‘ Roman law ’, e.g. the answer to the question whether mancipatio is an institution of the ius gentium is ‘ No; it is an institution of the civil law.’ But where the theoretical meaning is in

question, as in the passage from Gaius above, we have to refer to the particular state in question, e.g. if we ask, ‘Is the rule that a husband or son must authorise a woman’s contract ® part of the ius gentium?’ the answer would have to be: ‘ No; it is a rule of Bithynian civil law.’ Justinian makes this quite clear. ‘ If,’ he says, ‘a man wishes to call the laws of Solon or Draco Athenian civil law, he will not be wrong.’ °

It is true that this distinction is not made by the Romans themselves * — 1 Rep. 3.33. Cf. Blackstone, Commentaries 1.41: ‘ This law of nature, being coeval with mankind, and dictated by God himself, is of course superior in obligation to any other... no human laws are of any validity if contrary to this.’

2 Off. 3.23; Har. resp. 32. 3 Off. 3.69.

+ It occurs in a text of Pomponius, D. 50.7.18, in what has become in modern times its commonest sense, i.e. the law governing relations between states (public international law).

Ca 1.193. 6 J. 1.2.2.

7 G. Lombardi, in a valuable survey (Sul concetto di ‘ius gentiwm’, Rome, 1947; ef. icherche in tema di ‘ius gentiwm’, Milan, 1946) has maintained that the ‘ theoretical ’ use of the term ius gentiwm is prior to the ‘ practical’, and indeed that this latter sense hardly existed as a separable thing at all. But though the term may have originated in 105

Law for foreigners, wus gentium and ius naturale they did not in fact succeed in distinguishing morals from law in theory — but

it is there all the same and must be grasped if we are to understand the different senses in which an institution can be said to be part of the ius gentium. The sense in which the stipulation, for instance, can be so classified

has been explained; it is part of the ius gentiwm because an Athenian or a Gaul or a Junian Latin, who has no state, may be bound or entitled under it. But when the Romans say that slavery and manumission ° or the right of capturing things in war (occupatio bellica)® is part of the tus gentium they mean something different. They mean that these institutions are common to all systems of law which they know, not that they will be recognised in a

Roman court. A person who alleges that he has been manumitted by a foreigner cannot claim to be free by Roman law;: whether he is free or not will depend on the law of his manumitter’s state. It is true that all states know manumission, but the rules on the subject differ. greatly from state to state; whereas when a stipulation is mentioned the Roman institution of that name is meant, and any rules there may be in the systems of other states

with regard to some similar contract do not matter in the least. It is this latter sort of tus gentzum alone which really concerns the practical lawyer;

the rest is philosophical ornament. ,

As appears from what has already been said the law of Nature is only

another name for the (theoretical) cus gentium. Not only does Cicero identify

them,” but the lawyers generally use the two phrases indiscriminately.’ Indeed, when they invoke the idea of ‘ nature’, it is usually in the much less precise sense of the intrinsic character of a thing or institution. In this sense traditio is a ‘ natural’ method of transferring ownership.* Only one refinement which occasionally appears needs mention. If there was one institution which was really common to all peoples of antiquity it was slavery, which consequently is always reckoned as turis gentium. According to Aris-

totle it was also natural, for some men were slaves by nature,*® but other Greek philosophers had different views; man was by nature free, and we thus find slavery defined occasionally as an institution of the ius gentiwm contrary to Nature and resulting from war.® But this is the only case in which a disGreek theory, one can hardly deny, in face of such passages as Gai. 3.93 and 154, that Roman jurists sometimes used it with practical implications (unless one assumes that the

passages are interpolated or that Gaius is ‘ post-classical’ before his time; ef. below, 390 n 4). See also Pugliese, Riv. it. sci. giur (1948) 453ff.; Grosso, Mél. de Visscher

1.395ff. 8 DP. 1.1.4. 9 D. 41.1.5.7. 1 Qui enim potest iure Quirttium liber esse is qui in numero Quiritium non est? Cic.

Caec. 96. 2 Above, 105. 3 Compare e.g. Gai. 1.86 with 1.89. 4 See Nicholas, Introduction 56f., and in greater detail, Levy, Ges. Schr. 1.3ff. (=SDHI 15, 1949, 1ff:); C.A. Maschi, La concezione naturalistica del diritto e degli

istitutt giuridics romani (Milan, 1937). 5 Pol. 1.2.18. 6 E.g. J. 1.2.2. See generally M. Villey, Lecons d@’histoire de la philosophie dw droit (Paris, 1957) 121ff. (=RAD, 1953, 475ff.). 106

Law for foreigners, ius gentium and ius naturale crepancy between the two systems can be found. Normally, it seems, the jurists thought in terms of a simple dichotomy between tus civile on the one

hand and ius gentium (identified with tus naturale) on the other. Ulpian, however, seems to have adopted a trichotomy.’ For he equates the law of nature with the instincts which men share with animals.* Justinian in his Institutes begins by adopting the same trichotomy, but as he then goes on to incorporate the passage of Gaius which states the dichotomy, his treatment is confusing.° 7D. 1.1.1.2, 3=J. 1.1.4; 1.2 pr. It has been contended that the trichotomy and the idea of ius naturale which it introduces are interpolated; see e.g. Lombardi, Concetto (above) 194 ff.

8 The idea, seems to have been developed in the Greek schools of rhetoric; see Pollock,

Note E in his edition of Maine’s Ancient Law, and Castelli, St. S. Perozzi (Palermo, 1925) 53ff. Its basis was Pythagorean; see Cic. Rep. 3.19. It occurs in no other legal text and is an unfortunate confusion of thought (see e.g. Nicholas, Introduction 55f.),

but its prominent position in the Institutes and the Digest gave it an undeserved influence in later thought; see e.g. Aquinas, Qu. 94, art. 2 (fin.). 9 Gai. L1=J. 1.2.1; ef. J. 1.2.11.

107

CHAPTER 7

The XII Tables We have already described how, according to the story, the XII Tables came to be compiled and enacted;! now something must be said of their contents, so far as they can be ascertained from the surviving fragments, and of the stage in legal development to which these fragments point. The greater part of the text has perished completely; the original bronze ? tablets are said to have been destroyed when the Gauls burnt Rome in 890 B.C. It may be that

substitutes were afterwards made and put up in the forum, but no such authoritative text was in existence at the end of the republic.* On the other hand private copies must have been very numerous, as is shown by frequent quotations and by Cicero’s statement that in his youth the XII Tables had been learnt by boys at school. The lack of an official text however had as a result that the language gradually became modernised. Though the fragments that we have look archaic it is known that their language, except in a few instances, is nearer to the Latin of classical times than to that of the fifth century B.C.* The surviving fragments come from quotations in authors of the last century of the republic or later times, who sometimes give the text of the law in what purports to be the original words, sometimes merely state its provisions in their own language. A good deal is to be got from Cicero,

something from the jurists, especially Gaius, a fair amount from grammarians and antiquarians who were interested especially in curiosities and obsolete words, and the remainder is gathered at large from Roman literature. Of the arrangement very little is ascertainable; in a few cases the number of the tablet on which a provision appeared is known; Cicero, e.g., says ° * we learnt the sz zn tus vocat when we were children ’, which seems to

show that these words (concerning the summons of the defendant by the plaintiff) stood at the beginning of the whole text, and consequently that the 1 Above, 13.

2 That this was the material used is stated by the historians. Some modern writers think that at so early a date wood would have been used. The text of Pomponius (D. 1.2.2.4) gives eboreas (ivory), but that has been conjectured to be a MS error for rovoreas (wooden).

3 St. Cyprian, writing about A.D. 245 (ad Donat. 10), and deploring the evil behaviour to be seen in the forum at Carthage, says incisae sint licet leges duodecim tabulis ... inter leges tpsas delinquitur, but even if this refers to the XII Tables it is probably no more than a rhetorical flourish. That no text dating from 450 or 390 survived is evident from the modernised form in which fragments are quoted even by antiquarian writcrs such as Varro; see further below, 111 and ef. Lenel, SZ 26 (1905) 500f. 4 See further below, 111. 5 De leq. 2.9. 108

The XII Tables code began with the provisions relating to precedure, which is likely enough

in any case, as these were probably considered the most important of all. The prohibition of conubium between patricians and plebeians was, we know, in one of the last two tablets (those supposed to have been drawn up by the second ‘ college ’ of decemviri) which, Cicero says,® contained ‘ unfair ’ laws,

and it is interesting to hear that testamentary succession was dealt with before intestacy.” Modern editions all follow more or less the order adopted by Dirksen (1824) and Sché6ll (1866), but rather for convenience of citation than for any other reason.’ What portion of the original bulk appears in these

works it is of course impossible to say, but, seeing that the Romans were fond of referring to their great code where possible, it is likely enough that we have at least a reference to most of the more important provisions.

If we speak of the XII Tables as a code, we must not misconceive their scope and scale. Livy does indeed describe them as ‘ the source of all public

and private law ’,° and Ausonius,’ in the fourth century A.D., goes even further and adds ‘ sacred law’, but these statements are exaggerations. In the first place, almost the whole of the surviving fragments are concerned with private law. A few rules which belong to the ius sacrum are incorporated,? but the compilation as a whole is secular in character, and it is an indication of the legal genius of the Romans that they were able, at so early a stage in their development, to separate law so completely from religion. Again, public law, in the sense of constitutional law, is represented in our fragments only by two provisions, that forbidding privilegia* and that forbidding the trial of a citizen on a capital charge by any assembly except the comitia centuriata.* Both of these are of great importance from the point of view of ‘ the liberty of the subject ’, and, if genuine, were no doubt included for that reason. Moreover even within the field of private law the scale of the

fragments suggests that only the leading rules were set out, though the amount of detail varies. Procedure seems to have been dealt with carefully, presumably because most of the difficulties of which the plebeians complained 6 De Rep. 2.63; cf. above, 13.

7 D. 38.6.1 pr. For these and other materials for a possible reconstruction (including the order of treatment in Gaius’ commentary on the XII Tables), see M. Lauria, op. cit. (above, 93 n 3) 21ff. For earlier literature, see Kriger, 13 n 31. 8 See Bruns 1.15ff.; FIRA 1.21ff.; English translation, A.C. Johnson, P.It. ColemanNorton, F.C. Bourne, Ancient Roman Statutes (Austin, Texas, 1961) Off. 9 3.34. 1 Idyll. 11.61-2. 2 Especially the regulations for burial collected in Tab. x, though even these are chiefly dictated by a desire to avoid lavish expenditure and unseemly expressions of grief. See also VIII.21: Patronus si clientit fraudem fecerit sacer esto. 3 Tab. 1x.1: Privilegia ne inroganto; ef. above, 29.

4 Tab. 1x.2: De capite civis nisi per maximum comitiatum ne ferunto; see further, helow, 306ff.

109

||

The XII Tables

had arisen in connexion with it,> whereas mancipation, which was no doubt wel] known, is dismissed with a simple confirmation of its validity.®

The broad picture which the fragments give is of a community of peasantproprietors, in which commerce is as yet unimportant (though archaeological evidence suggests that there was in fact a certain amount of ‘ foreign ’ trade) and writing was still uncommon. But is the picture authentic ? Of course the story of the decemviri plainly contains, as we have seen, some elements of

legend, but in the years around 1900 and thereafter criticism went much further: the whole central story of a codification dating from the fifth century was an invention. Either‘ it was an example of the Roman habit of reading recent events far back into the past,® the Appius Claudius who figures among

the decemviri® being a shadow of the real Appius Claudius, the censor of 312 B.C.,' or, more radically still,? the compilation was not legislative at all but was a collection, partly indeed of very old materials, made in the first half of the second century B.C., very probably by Sextus Aelius,* from whose

tripertita all subsequent writers derived their knowledge of the so-called code. But it is really inconceivable that the Romans of the first century B.C.

should have been so entirely mistaken about an event which, on these theories, was then comparatively recent. The time of Sextus Aelius lies in the full light of history, and even that of Appius Claudius the censor is, and was, quite well known. Moreover the fasti * record the names of the decemviri and there is nothing to justify a supposition of such extensive falsification. Above all, the state of the Jaw shown by the surviving fragments is much too archaic for 200 B.C. or even for 300, and neither date would leave enough time for

the development of the law as shown by the XII Tables into the already complex system of Cicero’s time. In spite of all possible scepticism as to the details of the story of their compilation, it is nowadays generally accepted * that the XII Tables are really an enacted code of law and that tradition 1s not far wrong in assigning them to the middle of the fifth century B.C. 5 It is noteworthy that the next step in the popularisation of the law, the tus Flavianum (above, 91) was also concerned with procedure. 6 Tab. Vi.l: Cum nezwm faciet mancipiumaque, uti lingua nuncupassit, ita tus esto.

7 BE. Pais; Storia di Roma (Turin, 1898, 1899) .1.1.550ff., 1.2.546ff., 631ff.; Storia critica di Roma 2 (Rome, 1915) 217ff.; Ricerche sulla storia e sul diritto pubblico dt Roma 1 (Rome, 1915) 1-240. An account and review of Pais’ later work is given by

Kreller, SZ 45 (1925) 8 Above,91. 12. ® Above, 13. 589ff. 1 Above, 2 Lambert, RHD (1902) 149ff.; Revue générale dw droit 26 (1902) 385ff., 480ff.; 27 (1903) 15ff.; Mélanges Ch. Appleton (Paris, 1903) 503ff.; La fonction du droit civil

comparé (Paris, 1903) 1.398ff. 3 Above, 92. 4 Above, 8n 2.

5* But it is consistent with what we know of the fifth ecntury; Wieacker, Entretiens Hardt 13.300ff.; ef. Gjerstad, ibid. 357. 6* Against Pais and Lambert see especially Girard, Mélanges 1.1ff. (=RHD (1902) 381ff.) ; Lenel, SZ 26 (1905) 498ff.; Greenidge, Hnglish Historical Review 20 (1905) Iff.; Wenger, Quellen 360ff.

110

The XII Tables This is not to say that the precise verbal form in which the provisions are

quoted (where they are not merely paraphrased or summarised) can be attributed to the fifth century. Our earliest example of Latin is provided by an inscription found in the forum’ and dating probably from c. 500 B.C. Very little can be deciphered, but there is enough to show that the forms of the words in our fragments of the XII Tables are not from this period.* Thus the word iouwmenta occurs in the inscription and is taken to be the archaic form of iumenta, but in what purports to be a verbatim quotation from the XII Tables Aulus Gellius has simply iumentum.® Moreover it is generally agreed ' that there have been some additions and alterations deriving from misunderstanding or attempted rationalisation or from the incorporation of explanatory comments.” Changes such as these are only to be expected where no authoritative text survives. On the other hand, some primitive forms are preserved unaltered (e.g. sam for eam, escit for est), and even where the forms are modernised the words themselves are often obviously old, some being incomprehensible to the writers who preserve them.® Again, the syntax * has probably been little affected by the hazards of transmission, and the style is uniform in its extreme brevity and simplicity. The whole code, so far as we can tell, was a series of staccato imperatives, but the language is always careful and exact.° A question which has always been much discussed is that of Greek influence.° The traditional story includes an embassy to Athens to study the laws 7 The so-called lapis niger, FIRA 1.19; Bruns 1.14; see Wenger, Quellen 3492. 8 Wieacker, Entretiens Hardt 13.300f.

® Tab. 1.38 (Gellius 20.1.25): Si morbus aevitasve vitium escit, qui in tus vocabit vumentum dato; si nolet, arceram ne sternito ; see also Tab. VII.7. 1 Cf. Wieacker, RIDA (1956) 462ff.

2 Thus, in Tab. 1.3, quoted above, n 9, qui in ius vocabit, which conflicts with the usual staccato style, is generally taken to be a later addition; and Fraenkel, Hermes 60 (1925) 440ff., has shown that vitiwm must be an unintelligent gloss on the text of Aulus Gellius, perhaps echoing the morbus vitiwmve of the aedilician edict (below, 294). For an example of rationalisation see below, 171 n 9. 3 E.g. lessus in Tab. x.4 (mulieres genas ne radunto, neve lesswm funeris ergo habento),

the meaning of which, according to Cicero, de leg. 2.59, was obscure even to Sextus Aelius; and cf. anfractus (Tab. v11.6), adsiduwus (meaning wealthy: Tab. 1.4), pedem struere (Tab. 1.2, quoted below, n 5). 4 F. Sbordone, Synteleia Arangio-Ruiz 334ff.

5 In some fragments there is what appears to be a frequent and abrupt change of subject; e.g. Tab. 1.1, 2: Si in ius vocat (i.e. the plaintiff), ni it (the defendant), antestamino (the plaintiff). Igitur em (the defendant) capito (the plaintiff). Si calviiur pedemve strut (the defendant) manum endo iacito (the plaintiff). But Daube, Forms 57ff., has suggested that these are examples of a primitive use of the third person singular in an impersonal sense. Thus, si in ius vocat means ‘if there is summoning’. On the text of Tab. 1.1 see Daube, Forms 28f. 6 For recent discussions see Wieacker, Entretiens Hardt 13.330ff., St. Volterra 3.757ff. ; Delz, Musewm Helveticum 23 (1966) 69ff.

lil

The XII Tables of Solon and also to other cities to study their laws;’ and an Ephesian named

Hermodorus is said to have assisted the decemvirs.* No more authority attaches to these tales than to the other picturesque details which gathered round the XII Tables, and modern opinion has generally been sceptical. But

in so far as this scepticism has been founded on the supposition that the horizons of the Roman world did not yet extend so far, it has been shown by recent archaeological discoveries to be unjustified: there is good evidence from the period before the XII Tables of contacts between Rome and both Magna Graecia and Greece proper.® This is borne out by the existence of Greek loan-words, one of which, poena, actually occurs in the XII Tables.’ It is thus entirely possible that there should have’ been Greek influence on the law, even though the story of the embassy to study the laws of Solon may well have arisen from a desire to link the Roman code with its greatest Greek counterpart.” What is surprising is rather the paucity of internal evidence of such influence.* Two passages from Gaius’ commentary preserved in the Digest allege identity between rules of the XII Tables and those of Solon, but the identity is not very striking and suggests at most indirect borrowing via Magna Graecia.* A more persuasive instance is given by Cicero, who speaks of restrictions on display at funerals as having been taken from Solon’s legisJation; here the details correspond so closely that borrowing seems certain,

though it might still be only indirect.* Other instances of similarity with Greek rules have been collected by modern authors,* but when all has been ‘ Livy 3.31.8, 32.6. Dionys. Hal. 10.51.5, 52.4, however, says that the other cities were in Italy (Magna Graecia). 8 Pomponius (D. 1.2.2.4) says that Hermodorus suggested (auctor fuit) some laws to

the x virt, Pliny, H.N. 34.21, that he was ‘ interpreter ’. :

9 E. Gjerstad, Harly Rome 4 (Lund, 1966) 582, 586f., 597f£. (summary). 1 Tab. v111.3; cf. Mitteis, RPE 15. 2 Ruschenbusch, Historia 12 (1963) 250ff., attributes the origin of the story to 8. Sulpicius Rufus. Momigliano, Entretiens Hardt 13.357, remarks that an embassy to the Athens of Pericles would surely have been given something more modern than the laws of Solon.

3 Unless clarity of style and arrangement (Schulz, Principles 7) or the absence of any

evidence. .

attempt to attribute the code to divine influence (Momigliano, loc. cit.) be seen as 4D, 47.22.4 (=Tab. v111.27), allowing sodales (members of societies) to make their own rules provided they do not contravene ‘ public law’, and D. 10.1.13 (=Tab. v11.2, concerning the space to be left between buildings erected by adjoining landowners. The

Solonian rules on the latter point (cf. Paoli, RHD, 1949, 505ff.) were in force at Alexandria in the third century B.C.; see P. Hal. cols. 4 and 5. Bonfante, Storia 1.114, points out that Gaius himself is really dealing with a different matter —the principles which are to guide a iudex in the actio finium regundorwm. 5 Cic. de leg. 2.59, 64 (Tab. x.2, 3, 4); ef. Wieacker, Entretiens Hardt 13.345ff.

6 Pais, Ricerche 1.147ff.; Mitteis, RPE 12f. Most interesting is the correspondence of privilegia ne inroganto (Tab. 1x.1) with the Athenian rule quoted by Demosthenes, c. Aristocr. 86.649: pydé vopnov éx’ avipt eEeivar Ocivat, éav un tov abrév éri zac ’AOnvaiots.

, 112

The XII Tables said, the evidence covers but a small part of the code and almost exclusively

matters of detail. The great majority of the surviving fragments appear to have a purely native origin and this view of them fits well with the tradition that what the plebeians wanted was not so much reform as certainty. No doubt some disputed points were settled and some innovations were introduced, of which a few may well have been copied from Greek originals, but as a whole the XII Tables are based on the customary law of Rome herself.

1138

CHAPTER 8

; i“

The law of the family and of succession at the time of the XIT Tables The Roman family is purely patriarchal, and Roman law is characterised by

the exceptionally great power which it allows to the father as head of the family. Originally, no doubt, this power was a general, rather vaguely conceived, supremacy, but by the time of the XIT Tables it has crystallised and become differentiated, so that its name varies according to the persons over whom it is exercised.! Over the wife and the sons’ wives it is manus, over the children and slaves potestas, patria potestas in the former, dominica potestas in the latter case, while the bondsmen are, at least in classical terms,’ said to

bein mancipio.® :

1. MARRIAGE Marriage was to the Romans, as to the other peoples of antiquity, a de facto

rather than a de jure matter, in the sense that two people were held to be married, not because they had gone through any particular ceremony, but because they in fact lived together as man and wife.* This state of affairs begins ordinarily with the bringing of the bride to the house of the bridegroom (domum deductio), but no special legal significance attaches to this 1 Gai. 1.49. The word originally used for the undifferentiated power seems to have been

manus (cf. Old English mund), which still appears in compounds, e.g. manwmissio, emancipare, without reference to power over a wife. Cf. Jhering, Geist 2.162; Mitteis, RPE 75. 2 Mancipium occurs in three senses: (a) the power described in the next note; (0) the

act per aes et libram (ef. below, 151); (c) the slave as object of that act. Sense (b) is probably the original, and (a) a relatively late derivative; see L. Capogrossi Colognesi, La struttura della proprietaé (Milan, 1969) 221ff., 288ff., who argues that in the phrases mancipio dare or accipere (e.g. Gai. 1.119, 121, 140, 162) the word is in the ablative (‘ by mancipation ’ and not in the dative (‘into the power or status of mancipiwm’). See also Kaser, HB 180ff. On causa mancipii see also David and Nelson, Th 19 (1951) 439ff.; id.,

Gat. Inst. Komm. 162f. ) 3 This status is the result of the power possessed by the Roman father to sell his

children. If the child was sold abroad he became a slave among the foreign people, but if he was sold to a Roman he was held in mancipio. He was subject to the buyer and, so far as his private life was concerned, no doubt in much the same position as a slave, but he remained legally a free man and a citizen, though his political rights were perhaps in abeyance. On manumission he fell back into the potestas of his father, unless there had already been three sales, in which case, by a rule of the XII Tables, the patria potestas was finally broken. Cf. above, 89f. 4 Jolowicz, Roman Foundations 141ff.

114

Marriage

proceeding, except in so far as it is in fact the beginning of conjugal life. Marriage may, however, be accompanied by manus, i.e. the woman passes out of her father’s family and becomes a member of her husband’s family and subject to his power: she is in the position of a daughter to him.* Since the importance of the family organisation in early Rome would make it very inconvenient for the wife to belong to a different family from her husband, and since manus would, as we shall see, normally result from the continu-

ance of marriage for a year, it may properly be said that marriage with manus was the rule at the time of the XII Tables, but the two institutions are nevertheless distinct.®

Manus may arise in any one of three ways: by confarreatio, coemptio, or usus.” Confarreatio was a religious ceremony in which the essential point was

the transfer of the woman from the domestic cult of her father’s family to that of her husband’s. It was almost certainly confined to patricians. Gaius says that the name comes from the use of a cake of spelt (far, a kind of coarse wheat) in the sacrifice made to Jupiter Farreus, that there must be ten witnesses and that certain solemn words had to be spoken. A late source ® adds that the pontifex maximus and the flamen Dialis (priest of Jupiter) were also present. In Gaius’ time it was still necessary that the rex sacrorum and the greater priests should be born of parents who had gone through the ceremony, and that they themselves should do so.

Coemptio, as opposed to confarreatio, is purely secular and is open to plebeians. It is an application of the act per aes et libram to the creation of manus. A man can acquire ownership over certain kinds of property by man-

cipation ® (i.e. by a formal act of purchase with scales and bronze in the presence of five witnesses), and in the same way (save that the words used are different ') he can acquire manus over a woman. In classical law coemp5 Gai. 1.111. According to Gai. 1.148, 2.159, 3.41, if the husband is in patria potestate, the wife is nevertheless in his manus and presumably in the potestas of the paterfamilias. 6 E. Volterra, La conception du mariage d’aprés les juristes romains (Padua, 1940) ; St. Solazet 675ff. The texts never speak of two types of marriage, and they refer to con-

farreatio, coemptio, and wsus as modes of creating manus, not as forms of marriage. Moreover usus adds manus to an existing marriage (Gai. 1.111; see below, 116). For a different emphasis see Kaser, Iura 1 (1950) 64ff.

7 Gai. 1.110-13. As to the relative antiquity of these three forms there is much difference of opinion, but that they all existed already by the time of the XII Tables is generally agreed. That usus did is clear from the provision that it might be prevented by the usurpatio trinoctit; below, 116. 8 Servius, in Georg. 1.31 (Bruns 2.78); a few more details in tn Aen. 4.374 (Bruns 2.76). For an account see Corbett, Marriage 71ff.; and for the view that confarreatio

contains elements from different periods, see Noailles, Fas et Ius 29ff. (=F schr. Koschaker 1.386ff.). 9 Below, 143.

1 Gai. 1.123 (but we do not know what the words were). Plutarch (Qu. Rom. 30; cf. Cic. Mur. 27) tells us that the bride said to the bridegroom ‘ubi tu Gaius, ego Gaia’, and Boethius (in Cic. Top. 3.14; Bruns 2.73) that the bridegroom asked the bride whether 115

a Family and succession at the time of the XII Tables

tio, like mancipation, is an imaginaria venditio, and no price is paid, but no doubt in origin both were genuine acquisitions by purchase.’ Usus. Just as coemptio was parallel to conveyance by mancipatio, so also usus was parallel to acquisition of ownership by prescription.* Gaius* says that if a woman remained married to a man for an unbroken year she came under his manus, and the XII Tables provided that if she wished to prevent this happening she could do so by absenting herself for three nights in each year (usurpatio trinoctii). It should be noted that Gaius says ‘ remained married ’ (nupta persevera-

bat), thus clearly showing that manus and marriage are distinct. On the

other hand, the very institution of usus reflects a period in which manus was, as has been said above, the normal accompaniment of marriage. The usur-

patio trinoctti was no doubt not introduced for the first time by the XII Tables but represented the degree of interruption which had come to be accepted as enough to break the year, since plainly not every casual absence could suffice and yet the possibility of interruption was implicit in the idea of USUS.

Difficulties however remain. (i) Why did usus and its corollary usurpatio

trinoctit evolve at all? We probably place the emphasis wrongly if, with Gaius, we regard it as in origin a means of acquiring manus. One does ‘not enter into marriage casually, and one must therefore have an intention one way or the other as to manus. One would hardly decide simply to let it grow — or not — by usus.° The original function of usus, as of its parallel usucapio, she would be his materfamilias and she asked him whether he would be her paterfamilias, but these were no doubt customary parts of the accompanying marriage ceremony rather than requisites of the act of coemptio. The ridiculous idea found in some later authors (Servius in Georg. 1.31, Bruns 2.78; Isidor. Or. 5.24—6, Bruns 2.81) that the coemptto was @ reciprocal sale, the wife buying the husband and the husband buying the wife, 1s due to their misunderstanding something that Ulpian evidently said in a passage now lost. Boethius, who refers to Ulpian, no doubt reproduces him correctly when he speaks of reciprocal questions. 2 In early law it is possible that women were never sui turis, but on the death of the paterfamilias fell into the power of the nearest agnates (Kaser, Iura 1, 1950, 89ff.; see below, 122); the purchase would therefore always be from the person in whose power the woman was. But once women can become sw turis there is the difficulty that the woman

seems to sell herself; see Corbett, Marriage 80f.; Watson, Persons 24 n 6. Some scholars (e.g. Lévy-Bruhl, Now. Et. 74ff.; Karlowa 2.166) have held that coemptio never was a genuine sale, but was a conscious adaptation of mancipation, the woman and not

her father having always been the party; but there is no branch of the law in which deliberate innovation is less likely than that of the family. 3 Below, 151ff.

41.111. 5 Explanations have been offered in terms of a kind of marriage on approval comparable to the ‘handfast’ marriages found at one time in the north of England and in Scotland, which became permanent if the woman bore a child or became pregnant within

a year and a day, but might be dissolved if she did not (Vinogradoff, Hist. Jurisp. 1.246; Lévy-Bruhl, TR 14, 1936, 452ff.) but there are no traces of such a marriage in 116

Marriage was probably to simplify proof. After a year the husband no longer needed to

prove a coemptio or confarreatio.® (ii) What would be the reason for preventing manus from arising by resort to usurpatio trinoctii? There may be some broad social explanation hidden in Roman pre-history, but more probably the separation of marriage and manus which, as we have seen, is implicit

in Gaius’ account of the XII Tables rule, was the result of evolution, the impetus to the making of the separation coming, it has been suggested,’ from the development of the capacity of women to own property. It may be that in the very early law a woman could never become sui iuris * and could therefore never own property: either she was in the potestas of her father, or, on his death, she fell, as his property did, to the nearest agnates.® The creation of manus in her husband would therefore not per se effect the transfer of any property. But once it was accepted (as it already was in the earliest period of which we have any knowledge) that a woman became sui iuris on the death of the paterfamilias, and was then equally entitled with her brothers to share in the family property (or, more importantly perhaps, to take it all if she had no brothers) there would be a strong motive for her agnates to wish to prevent the family property from passing with her on her marriage. Divorce. Since the existence of marriage was a matter of fact, so also was its termination. All that was required was the actual ending of conjugal life

~the parting of husband and wife with no intention of coming together again.’ On the other hand, manus could only be broken by a legal act, a remancipation of the wife to her father, or to some other person who could then set her free, with the result that she became suz iuris. Where manus had been created by confarreatio, there was apparently a corresponding diffareatio for its termination,” but we know very little about it.

Although no form was necessary for divorce, we hear of a number of formulae the use of which was presumably a matter of custom.*® Again, Rome, and it is hardly consistent with Gaius’ nupta perseverabat (Kaser, Iura 1 (1950) (0ff., also against a later variant by Lévy-Bruhl, Nowv. Et. 63ff.).

6 Kaser, Jura 1 (1950) 75ff. But the need for such simplification of proof is more obvious in the case of ownership of property than of manus over a wife, the creation of which would be a momentous and memorable event. Mitteis, RPR 252, thought that the purpose of usus was simply to cure defects in the ceremony of coemptio. 7 Kaser, Iwra 1 (1950) 86ff.

8 Above, 116 n 2. 9 Below, 125.

1 The marriage of the fiamen Dialis was, however, indissoluble; Festus, s.v. flammeo, Bruns 2.9; Gell. 10.15.23.

2 Festus, s.hv. (Bruns 2.7); Plutarch, Qu. Rom. 50, refers, apparently in this connexion, to ‘many strange and terrible rites ’.

3 Corbett, Marriage, 224f.; Kaser, RPR 1.82. There is a difficulty that Cicero says (Phil. 2.69) that the formula res tuas tibé habeto was laid down in the XII Tables (cf. D. 48.5.43, where the reference may, however, be to the lex Iulia de adulteriis; Volterra, St. Biondi 2.123ff.). No satisfactory explanation has been found: if the words were legally required, they would hardly have gone out of use in this most traditional part of 117

Family and succession at the time of the XII Tables although no grounds were necessary for a valid divorce, a Roman husband was not able to use his powers lightly. Custom required that, before he sent his wife away, he should summon a family council (on which his wife’s relatives must be represented) and show a good ground for his intended action.* Plutarch * gives a list of such grounds, which he says was established by Romulus: adultery, wine drinking, tampering with keys, and witchcraft. Divorce on insufficient grounds might involve punishment by the censors ° or possibly something more severe in very early times,’ but it would be contrary to the whole trend of Roman thought to suppose that it would have been

invalid.

We have so far spoken only of divorce by the husband. In the law as we know it, however, either party could, as the logic of the institution demanded,

take the initiative in terminating the marriage. Even if the wife were in manu, the legal power of her husband could not prevail over the fact of her repudiation of him, and since as a result she would be still in his manus but no longer his wife, she was able to compel him to release her.’ This cannot, however, have existed in early law, and presumably the power of the husband would then have been sufficiently effective to make divorce by a woman in manu unthinkable.

2 PATRIA POTESTAS The complete power of the Roman father over his children has become pro-

verbial, and the Romans knew that it was an institution peculiar to themselves. It extended not only over all sons and daughters (so long as they had not passed into the manus of a husband), but also over the children of the the law; if they were merely a matter of custom (Kaser, RPR 1.82) it is difficult to see why they should have been included in the XII Tables. Against Yaron, TR 28 (1960) lff., see Watson, TH 33 (1965) 43f. 4 Val. Max. 2.9.2; ef. below, 237.

5 Rom. 22. Contrast Dionys. 2.25. Discussion, Noailles, Fas et Ius 1ff.; Westrup 1.1.77, 198.

6 Val. Max., loc. cit., relates that in 307 B.C. the censors expelled L. Antonius from the senate for putting away his wife without the consent of the family council.

7 According to Plutarch, Rom. 22, a law of Romulus laid down that a man who divorced his wife, except on one of the permitted grounds, should forfeit half his fortune to the wife, and half to Ceres; whereas one who sold his wife should be sacrificed to the gods of the underworld. On this see Noailles, Fas et Jus 1ff.; Perrin, St. Albertario 2.405f.

8 Mitteis, RPR 252. There is a remarkable concurrence of tradition (references in Corbett, Marriage 218) that the first divorce ever to take place was that of Spurius Carvilius Ruga, which is most commonly dated about 230 B.C. But both considerations of general probability and the amount of contrary evidence (see above) make this incredible.

Either the date must be wrong (Corbett, Marriage 227f.) or it was the first divorce without fault of the wife (childlessness), or the first in which the dos was returned; see

Kaser, RPR 1.82n.; Watson, TR 33 (1965) 38ff. $ Gai. 1.137a; below, 236. 118

Patria potestas sons and more remote descendants through males, without any limit other than that imposed by the span of human life.

The oldest male ancestor (paterfamilias) has complete control over the persons of his descendants, even to the extent of inflicting the death penalty

on them. This vitae nectsque potestas! was not, however, an arbitrary power, but was subject to the requirement, certainly by custom and perhaps by law, that a consilium be convened to hear the case; and it seems that the paterfamilias was bound by the verdict which this consilium passed.” Again, only the paterfamilias has any rights in private law: no subordinate member of the family can own any property, and any acquisitions that they make go straight to the pater, just as the acquisitions of a slave become the property of his master. No person, male or female, in potestate can marry without the consent of the pater, and if the sons marry with manus it is the pater who obtains authority over their wives.? One limitation there was, presumably from the earliest times: patria potestas has no concern with public law, and a son under power could vote and hold a magistracy just as freely as a paterfamilias.* In private law the only limitation, if it can be called one, which we

know to have existed at the time of the XII Tables, was the rule that if a father sold his son three times the son was to be free from the father. This rule, as we have seen,*® was used to make emancipation possible.

If the tie between father and child can be artificially broken by emancipation it can also be artificially created by adoption. Desire for continuity of the family was always strong at Rome, and it was of especial importance that there should be a son to carry on the sacra or religious cult of the family. In republican law as we know it there were two entirely different forms of adop-

tion, according as the person to be adopted was one under potestas (i.e. alient iuris), who was merely to be transferred to another potestas, or a paterfamilias (i.e. a male sui iuris), who was to give up his independence and come under the patria potestas of someone else.* In the former case the pro1 Yaron, TR 30 (1962) 243ff.

2 The existence of this domestic tribunal has been disputed, but see Kunkel, SZ 83 (1966) 219ff., who adduces instances from the late republic and early principate of trials of sons (Sen. de clem. 1.15.2; Val. Max. 5.9.1; Jos. A.J. 16.356). For trials of wives see above, 118. Even women swi iuris might be tried. Kunkel conjectures that the defective

passage in Gai. Aug. 85f. refers to a XII Tables rule forbidding the putting to death of sons sine tusta causa (which would be established by the consiliwm). For Kunkel’s related views on the consiliwm in ordinary criminal trials see below, 311ff., 317. 3 Above, 115 n 5.

4 Pomponius, D. 1.6.9: Filius familias in publicis causis loco patris familias habetur,

velutéi ut magistratum gerat, ut tutor detur. 5 Above, 90.

6 A person is alient iwris if he or she is in manu, in mancipio, or in potestate (patria or dominica) to another. All other persons are sw turis, and if male are patresfamilias, which is simply another way of saying the same thing in their case, for the word has nothing to do with being married or having children; the new-born infant, if he is not

in potestas (e.g. if he is illegitimate), is a paterfamilias. The word materfamilias is 119

Family and succession at the time of the XII Tables cess was adoptio (in the strict sense) and the rule about the three sales was again pressed into service to break the potestas of the real father. The child (or rather * person to be adopted ’—for he might be of any age) is twice sold to a third party and after each sale manumitted; then he is sold a third time, so that the patria potestas has gone irretrievably. The next step is that the adopting father should bring an action against the third party who now holds the child in mancipio, claiming that he is his son. The action is of course

collusive, so that the third party makes no defence and the magistrate adjudges the child to the adopter, into whose patria potestas he thus passes. This is one of a number of cases where the Roman jurists used a collusive action to bring about results for which no direct method was provided by law.’

Adrogatio was a more important affair; first there was an investigation by the pontiffs into the desirability of the transaction and then, if they had no objection, the comitia curiata, meeting probably under the presidency of a pontiff, had to give their approval, the institution taking its name from the rogatio or bill which was submitted to them.® The reason why there was need

here of approval by the religious authorities and by the assembly, whereas none was needed -for adoptio, is not difficult to see. Adrogatio, being the adoption of a person su? zuris, meant that a family was extinguished and merged into another. The family might consist only of the person adrogated himself, or of him together with those whom he held in his potestas (who would follow him into the potestas of the adrogator), but in either case it was extinguished, and provision might have to be made for the continuation of its domestic cult. This no doubt was the special interest of the pontiffs.°

That adrogatio existed already at the time of the XII Tables is generally assumed, though there is no actual proof; the institution has an archaic flavour about it, and, above all, the use of the comitia curiata, the earliest form of assembly, which ceased to function as a political institution from early republican times,’ points to the origin of adrogatio at a period before that of the XII Tables. In historical times the comitia curiata was for the purposes of adrogations represented by thirty lictors, one for each of the thirty curiae, so that its approval was a mere matter of form. The preliminary investigation of the pontiffs however remained a reality. occasionally used in a corresponding sense as the opposite of filiafamilas (e.g. D. 1.7.25),

but its older meaning appears to be ‘wife’ and indeed especially one married with manus; cf. the formula used in coemptio, above, 115 n 1. 7 For in ture cessio generally see below, 149. 8 Gai. 1.99.

® It is probable that there was a second vote of the assembly declaring that the adrogatus had no longer any part in his family sacra (detestatio sacrorum). 1 Above, 19.

120

Family and succession at the time of the XII Tables

38. GUARDIANSHIP Throughout the history of Roman law a child under the age of puberty ” needed a guardian and so also at the time of the XII Tables (and for long afterwards) did a woman of any age who was sui iuris. In these cases the guardian was called tutor. Beyond incapacity due to youth or sex Roman law also recognised two other cases, that of the lunatic (furtosus) and the spendthrift (prodigus), but in these cases the guardian was called by a different name — curator. It must however be understood that there is never any question of guardianship of either sort except where the person concerned is sui iuris; a person who is in potestate, in manu or in mancipio can have no guardian, because the chief function (and in later law the only function) of

the guardian is to administer the property of the incapable person, and a person alieni iuris can have no property to be administered.

At the time of the XII Tables the rules for the appointment of tutores were simple in the extreme. A man might in his will appoint a tutor for any person in his manus or potestas who would become sui iuris at his death.* * Failing a tutor appointed by will, the tutela went to the person who would,

according to the rules of intestacy,’ succeed to the property of the ward (pupillus) on the latter’s death,® i.e. to the nearest agnate’ or the nearest agnates jointly; failing agnates it went to the gentiles.* In the case of freed persons (who could have no agnates) the tutela went to the patron (i.e. the person who had manumitted them) or his children. This was not definitely laid down by the XII Tables, but was inferred from the rule that the patron

or his children were to take the succession to freed persons if they died intestate.° The cura of lunatics went first to the agnates, failing them to the gentiles;+ there could be no appointment by will. The XII Tables also made provision for the management of the property of spendthrifts (cura prodigi). If a man wasted property which he had acquired by succession on intestacy as suus 2 Fixed at twelve for girls and later at fourteen for boys; see Gai. 1.196; J. 1.22 pr. 3 Provided, of course, that the person in question would need a tutor. Thus he could appoint for his wife (in manu), and for his daughter (unless she were married with manus), at any age, and for his son or grandson by a deceased son in case they were under puberty at his death. He could not, on the other hand, appoint a tutor for e.g. a grandson by a living son, for that grandson would come under the potestas of his own father. Vestal virgins were also exempt from tutela. See Gai. 1.144ff.

4 The view that testamentary tutela goes back to the XII Tables depends on those versions of Tab. v.7 which mention tutela (see below, 139 n 7). But there are versions which omit it, and Wlassak (Studien 4, 19), refusing to believe that the XIT Tables gave more than the power of disposing of individual property, prefers them.

122. 81Below, 125. 97 Below, Gai. 1.165. Tab. v.7. 5 Except that a woman could succeed but could not be a tutor. 6 Below, 124. 121

ie Family and succession at the time of the XII Tables

' heres? he could be prohibited by magisterial interdiction from dealing with the property, and the management of it would then pass to the agnates.* In later times the interdiction could extend to all property.* Guardianship (both tutela and cura) became in the developed law an institution intended, as it is with us, to shield the incapable person from the con-

sequences of his own inexperience, disease or folly, but at the time of the XII Tables this was clearly not its main purpose. It was then assigned ° to the successors on intestacy, the very people who would benefit if the ward died without attaining testamentary capacity,® for their own, or at any rate the family’s, advantage. This is particularly clear with respect to the tutela of women; the agnates were given power over them to prevent their losing

the family property or taking it with them in an undesirable marriage. Marriage with manus needed the consent of the guardians, who could thus prevent a match of which they disapproved.’ The whole institution is in fact one intended to keep the property in the agnatic family.

4 THE AGNATIC RELATIONSHIP Outside the immediate circle of persons subject to the same paterfamulias relationship was still traced exclusively on the agnatic principle, i.e. only relationship through males was recognised. Two people are related agnatically if they are in the patria potestas of the same man, or if there is some common ancestor in whose power they would both be if he were alive.* Two brothers or a brother and a sister are thus agnates, and so are a man and his brother’s son or daughter, because they are both.in the potestas of the same man, i.e. the father of the one and the grandfather of the other (or would be 2 Below, 124. In other words, property which he has received from a direct ascendant. 3 Emit. Ulp. 12.2. No doubt if there were no agnates cura went here also to the gentiles, but no text says so.

4 The question whether interdictio existed at the time of the XII Tables is debated. For the view adopted here see Solazzi, Scritti 3.245ff. (=St. Bonfante 1.47ff.) ; Kaser, RPK 1.85; ef. Hpit. Ulp. 20.13; D. 27.10.1 pr.; 28.1.18 pr. The form of the interdiction is preserved in PS 3.4a.7: quando tibi bona paterna avitaque nequitia tua disperdis liberosque tuos ad egestatem perducis, ob eam rem tibi ea re [? aere; Kaser, St. ArangtoRuiz 2.152ff.] commercioque interdico. De Visscher (Htudes 57ff.= Mél. Cornil 2.570ff.; ef. Collinet, Mél. Cornil 1.147ff.) takes bona paterna avitaque to refer to property left to the

‘spendthrift in his father’s will, and since we know that the XII Tables applied only to property acquired on intestacy, he supposes that the interdiction originated only later, when the magistrate found it advisable to extend the restraint to other property. On this view it was the agnates themselves who, under the XII Tables, had power to make the

interdiction. |

5 Apart from testamentary appointment of tutors. 6 By reaching puberty in the case of a boy, by recovering his senses or having the interdiction removed in the cases of the lunatic and the prodigus respectively.

7 Cf. above, 116 n 2. 8 Or, of course, if they are father and child. 122

Succession on death if he were alive), but a man and his sister’s son are not related agnatically because they could never be in the same potestas—they are mere cognates, cognates being all relatives whether the relationship be traced through males or females. The bond of patria potestas can, as we have seen, be created artificially by adoption, which gives to the adopted child all the rights he would have had if he had been born in potestas; he becomes by the adoption the agnate of all his adopting father’s agnates. On the other hand he loses all agnatic connexion with his previous family, as does also the person who is emancipated. Manus has the same effect as patria potestas; on marriage with manus the woman passes out of the agnatic family of her father into that of her husband, in which she occupies a position similar to that of her husband’s daughter. The working of the agnatic principle has already been seen in relation to

guardianship; still more important is its operation in relation to intestate succession, to which we shall shortly come.

5. SUCCESSION ON DEATH Few aspects of early Roman law have given rise to as much controversy as the law of succession on death, and some account must be given of the main issues in this controversy. But the issues will be clearer if we sketch first the most characteristic features of the law as we know it in historical times.° The essential of the Roman system is the heres or universal successor. When a man dies, his heres, or several heredes jointly, take over his legal position and succeed to all his rights and duties (so far as they are transmissible at all) as a single whole. The heres becomes owner of the things which the dead man owned, he becomes creditor where the dead man was creditor and debtor where the dead man was debtor. Where there are several heredes, the inheritance is shared between them in the sense, not that each will get certain particular things which belonged to the dead man, but that they will

be co-owners of everything that he owned; they will each be entitled to receive a fraction of every debt owed to the dead man and will each be liable

for a fraction of every debt owed by him; and their liability for his debts will not be limited to the value of his assets —if he did not leave enough property to pay his debts, they must pay them out of their own pockets.

The heres or heredes may either be appointed (‘ instituted ’) by will or designated by the rules of intestacy, but without a heres no succession can take place. Further, there is an important distinction between sui heredes and extranet heredes. A suus heres is any person in the potestas or manus of the dead man who becomes sui iuris by his death. All other heredes are ® See also Nicholas, Introduction 235ff. 1238

Family and succession at the time of the XII Tables extranei.1 It is, as we shall see, the sui who are primarily entitled on the death of the paterfamilias. For example, if he leaves a widow in manu, a son who has himself a son, a grandson by a predeceased son, a married daughter and an unmarried daughter, his estate will be divided into four parts of which the widow, the son, the grandson by the predeceased son ? and the unmarried daughter will each get one. The grandson by the living son gets nothing; he is not a suus heres because he does not become sui iuris by the death of his

grandfather but comes under the potestas of his own father. The married daughter also gets nothing, because, provided she was married with manus, she has passed out of her father’s power.

The sui differed from the extranei in three important respects: (a) they became heirs automatically on the death of the paterfamilias, whereas extranet had to make an act of ‘ entry ’ (aditio) signifying acceptance of the succession; (b) as a corollary of this they became heirs even without their knowledge or against their will—they were sui et necessarit heredes; (c) equally, if the paterfamilias made a will, they could not be passed over in silence; they must either be instituted or expressly and solemnly disinherited

— otherwise the will must fail. |

A will may also make gifts out of the inheritance (legacies), appoint tutors, or manumit slaves, but none of these provisions can take effect unless there

has been an institution of a heres or heredes. If there is no institution, or if the institution fails, the whole will is void. For there is a sharp line drawn between testamentary succession and succession on intestacy, which is expressed in the maxim nemo pro parte testatus pro parte intestatus decedere potest. Legacies etc. cannot be charged on a heres ab intestato; and a will cannot dispose of only part of the inheritance, leaving the rest to go as on intestacy.

6. INTESTATE SUCCESSION By the XII Tables, if a man dies intestate the first people entitled to succeed are his sui heredes. The form in which their rights are mentioned is worthy of notice.? They are not so much given the inheritance as assumed to take it,

for the statute simply says ‘if a man dies intestate without having a suus heres, the nearest agnate is to take ’. And Gaius* says that they are called sui because ‘ they are household heirs and even in their father’s lifetime are considered in a manner owners ’. The explanation of this and of the special 1 We are not here concerned with the necessarius heres; Gai. 2.153ff.

2 If there are several children by a predeceased son they share the part which would have gone to their father —i.e. the succession is per stirpes, not per capita; Gai. 3.8. 3 Tab. v.4: St intestato moritur cui swus heres nec escit, adgnatus prozimus familiam habeto. See also below, 131 n 4 and n 8.

4 2.157: domestict heredes sunt et vivo quoque parente quodammodo domint existtmantur. Cf. D. 28.2.11. 124

Intestate succession features of the position of the suus which have been mentioned above is presumably to be found in an early stage of legal development when the paterfamilias was not so much the owner as the administrator of the family pro-

perty, and on his death the sui merely entered into the full enjoyment of what already in a sense belonged to them. Hence they succeeded automatically, and hence they could not be excluded except solemnly in a will; and a will, as we shall see,* in early times had to be approved by the pontifices and

made before the comitia. ,

Failing sui heredes the succession goes to the nearest agnate or agnates, if there are several in the same degree.*” In the case of women and children this is necessarily the first class because, since a woman cannot have patria

potestas, she cannot have any sui heredes, and a child cannot have any children.

It is to be noted that the XII Tables do not say of the agnates that they are to be heredes, but that they are ‘ to take the familia’. From this it has been conjectured that they did not originally count fully as heredes, as did the sué and perhaps also the heirs appointed in a will, but in historical times the agnate certainly does count as an heir in the full sense.*®

In default of agnates the succession went to the gentiles.° We do not know

whether they took as a corporation or whether, as the use of the word gentiles, rather than gens, suggests, they took as individuals. It seems that, at any rate in the latest period in which gentile rights survived, the succession was taken by individuals, for we read that Caesar was ‘ deprived of his gentilician successions ’,' but this may well have been different at the time of the XIT Tables.’ In all this there is no trace of either primogeniture or preference for males

over females. Daughters take equally with their brothers in the class of suz 5 Below, 127ff.

6 Degrees are reckoned by the number of steps to be taken in the family tree from one person to the other; thus a grandfather and grandchild are related in the second degree, and an uncle and nephew in the third, two steps up from the nephew to the common ancestor and one down again to the uncle. 7 In the classical law proximus was interpreted narrowly, and if the nearest agnate refused, the next nearest was not entitled (i.e. in the language of the commentators there was no successio graduum) ; Gai. 3.12, 22; Hpit. Ulp. 26.5. But the rule could lead to strange results, and Yaron (TR 25 (1957) 385ff.) has suggested that originally proximus

indicated merely that the agnates did not take as a class, and that the limiting interpretation was adopted only when the praetor had admitted cognates to bonorum possessio (below, 250), when the effect was to place all but the nearest agnate on an equality with

the cognates. 8 Cf. Buckland, Text-book 368; Manual 228. 9 Tab. v.5: Si agnatus nec escit, gentiles familiam habento. We do not know whether this applied also when the nearest agnate refused; contrast Buckland 369 and Schulz, CRL 224.

1 Suetonius, Caesar 1.2. The latest evidence for survival is the Laudatio Turiae (B.C. 8-2, Bruns 1.321) which presupposes gentilician tutela. Gaius 3.17 says totum genttlicium

ius in desuetudinem abisse. 2 See, however, below 138 n 1. 125

sect ee SS is rarest Family and succession at the time of the XII Tables

heredes, and similarly, in the class of proximi agnati, a sister of the dead man (by the same father) will share equally with a brother and will exclude an uncle. The absence of primogeniture and the peculiarities of the succession of suz heredes * have been adduced in support of the hypothesis, associated in this connexion above all with the name of Bonfante,* that the fundamental unit in pre-historic times was the patriarchal joint family. This institution, for which analogies are claimed in other Indo-European peoples, would not have been confined, as the classical Roman family was, to the direct descendants of a single living paterfamilias, but would have embraced a number of such families, living together and jointly exploiting the family property. This joint family would have had an administrative head or sovereign (appointed, as we shall see, by will), but the property would have remained in the undivided ownership of all the members of the family from generation to generation (subject, obviously, to economic pressures which would lead to ‘ emigration ’ etc.). At first sight this hypothesis might seem to be supported by the fragments of Gaius discovered in 1933,° which showed that in early law there existed between sui heredes what Gaius calls a form of partnership, by which

they remained in undivided ownership of the property which they had inherited, each having power to dispose of the whole. But this partnership was evidently very different from the supposed joint family. For there is no trace of any administrative headship,°® and this is essential if the joint family

is to survive as a stable unit. On the contrary, indeed, this is a partnership of independent equals; it survives only so long as none of the suz chooses to dissolve it, and so long as it does survive each of them has power to dispose of the whole property. For these and other reasons Bonfante’s theory has lost favour, but since it is also central to the controversy about the origins of the Roman will, further consideration of it may be postponed.

7 THESTAMENTARY SUCCESSION When we pass from intestate to testamentary succession our difficulties in ascertaining the state of the law at the time of the XII Tables become much greater. We know that a will of some sort existed,’ and this is in itself a note8 Above, 124f.

4 Scritts 1 (Turin, 1926); Corso 6.37ff. The patriarchal joint family is the central theme of C. W. Westrup’s Introduction to Early Roman Law, vols. 1-3 (London and Copenhagen, 1944, 1934, 1939). 5 Below, 389.

6 De Zulueta, Gaius 177; and for this and other arguments see Crook, Class. Quart. (1967) 1134.

7 The rules of intestate succession presuppose the possibility of a will: st intestato moritur...And Tab. v.3 (uti legassit ...swae ret) presumably refers to some disposition 126

Testamentary succession worthy fact, for most peoples have not yet invented a will at a parallel stage in their legal development. The existence of a will means that a man’s power over his property extends beyond his lifetime, that he can take it away from those to whom the law would give it at his death and say that someone else is to have it. We are so accustomed to this extreme development of the right of property that it seems natural to us, but it is in fact a development which comes as a rule only late in a people’s legal history.* We know also that the XII Tables dealt with testamentary before intestate succession,® and that throughout their history the Romans continued to regard testamentary succession as normal.' Beyond these facts however all is more or less plausible conjecture. We can tabulate the chief points on which there is difference of

opinion as follows: .

(i) What type of will existed in the time of the XII Tables? We know from Gaius ” that ‘ in the beginning there were two kinds of will, one made comittis calatis, the other in procinctu ’. The former was made in the comitia curiata, summoned twice a year for the purpose * probably under

the presidency of the pontifex maximus and known as the comitia calata.* The testamentum in procinctu was made when the citizens were met, not in the assembly, but in military array, ‘for’, says Gaius, ‘ procinctus is the host ready and armed ’. ‘ Thus’, he continues, ‘ they made the one sort of will in peace and ease and the other when they were about to proceed to battle.’ It appears from other sources * that the moment for making the will

in procinctu was between the two takings of the auspices, and that the citizen-soldier made it in the hearing of three or four comrades.°®

In addition to these two public forms of will’ there was, at least in later times, a private form, that per aes et libram (mancipatory will). This is in appearance the mancipation of the whole of the testator’s fortune to a friend (here called familiae emptor) who was asked to dispose of the property after the testator’s death in accordance with his instructions. As we know it, this mortis causa, but its exact meaning is disputed; see below, 139 n 7. For the early origin of the word testamentwm see Goldmann, SZ 51 (1931) 223ff.

8 Cf. Maine, Ancient Law 207ff. 8 Above, 109.

1 Indeed, Bonfante held that the will was aboriginal, but only for the purpose of designating the head of the joint family (above, 126); property remained joint. See

further, below, 131f. 2 2.101.

3 The dates were, according to Mommsen (StR 2.38; Rém. Chronologie 242f.), 24 March and 24 May.

4 The word was apparently applied to the comitia only when meeting under religious

authority. 5 Schol. ad Verg. Aen. 10.241, Bruns 2.77; Plut. Coriolanus 9. 6 Scherillo, Scr. Giuffré 1.781ff., conjectures that it was made, not when the people

were really in battle array, but before the comitia centuriata (above, 20). It would therefore have been open to plebeians, who were, in his view, excluded from the comitia curiata because they had no gentes; but see above, 18 n 7. 7 Both were obsolete before the end of the republic.

127 }

Family and succession at the time of the XII Tables is a true will, resulting in the institution of a heres (who is someone other than the familiae emptor), and by the later republic it is the normal form, remaining so even in the classical law.’ In origin, however, it is obviously not so much a will as a device for circumventing the impossibility of making a will. But since we know from Gaius that the two other forms already existed in early times, we have to ask in what sense the making of a will was impos-

sible. It is commonly suggested that the mancipatory will arose to provide for emergencies when the making of a will was impossible in the sense that a testator could not wait until the next meeting of the comitia; more probably perhaps, as is suggested below, it achieved a purpose which was beyond the scope of the comitial will. It is disputed whether the mancipatory will (in its primitive form) existed at the time of the XII Tables. On the one hand Gaius says, as we have seen, that the other forms existed ‘ in the beginning ’, and he says of the mancipa-

tory will that ‘it was added to the earlier forms’. As the Romans always regarded the XII Tables as fundamental, this suggests that Gaius thought of the mancipatory will as a subsequent introduction. On the other hand, in the formula of the mancipatory will there occur the words secundum legem publicam, and this is most probably a reference to a provision of the XII Tables. But even if the beginnings of the mancipatory will are to be found in the XII Tables, there can be no doubt that the form in which we know it is a later

development, and we shall therefore postpone discussion of it to a later chapter.? Here we shall consider only the questicns which arise in connexion with the comitial will.

| (ii) Was the part played by the people in the comitial will that of legislators or merely that of witnesses ?

At first sight it would seem obvious that their part was that of legislators, i.e. that they had to ratify by their votes the will which the testator proposed to make. A will generally means an interference with the legal rights of a man’s family, or gens, to his property after his death, and it seems natural enough that such an interference should first of all have been allowed only if sanctioned by the sovereign people itself in a sort of private Act of Parliament, just as was an adrogation, where voting certainly took place. Indeed, the will may simply be a form of adrogation.' Many scholars however hold the opposite opinion and think that there was no voting and that the people were summoned only to witness the testator’s declaration. The chief arguments for this view are: (a) The word testamentum itself is connected with testis, a witness, and the formula of the later mancipatory will, which is in all probability taken from the comitial will, includes an appeal to witnesses

8 Below, 242. ® Below, 242ff.

1 See below, 130. Koschaker, SZ 63 (1943) 449ff., finds a parallel to the comitial will as a legislative act in the will of a Hittite King at least 1,000 years before the XII Tahles. 128

setae ee SSS nn nl Testamentary succession

—ita do ita lego ita testor, itaque vos, Quirites, testimonium mihi perhibe-

tote.” (b) According to Labeo * the comitial will was made in populi contione, and there was never any voting in a contio.*

(ili) Was the comitial will one in which, as in the will of the developed law, a universal successor (heres) was appointed, or did it merely contain legacies ? ®

The former of these two views is now probably universally accepted, but

the latter was vigorously maintained by Lenel.* At the time of the XII Tables, he thought, the only true heres was the successor on intestacy; and the possibility of creating a similar universal successor by will did not come

until a late stage in the development of the mancipatory will, when the familiae emptor ceased to have any real function and the mancipation to him became a mere piece of ceremonial necessary for the passing of the rights and

duties under the will to the heir mentioned in the written tablets.’ This change, Lenel thought, had not yet taken place in the time of Plautus,’ in whose plays there was, in his view, no trace of the appointment of an heir by will. Lenel’s thesis rests in part on the use of the word legare in the fragment of the XII Tables dealing with wills (uti legassit . . . suae rei, ita ius esto), but it is likely that this refers to the early mancipatory will, which probably arose, as we shall see,°® to get round the impossibility of making legacies by the comitial will. The insuperable difficulty in Lenel’s account is, however, that of explaining how the Roman will ever developed the features which characterised it in the mature law, and in particular how the validity of the whole will came to depend on the nomination of the heir. So fundamental a rule cannot have been merely a late by-product of the mancipatory will. It is 2 “Thus I give, bequeath and make my will, and so do you, citizens, bear me witness.’ See Gai. 2.104, and below, 243. 3 As reported by Laelius Felix, quoted by Gell. 15.27.1-3.

4 Contio is an informal gathering of the people on the summons of a magistrate to listen to speeches, such as preceded the voting on any legislative proposal. But there seems to be no reason why the proposed will should not similarly have been read first in a contio before the people proceeded to vote on it. See Paoli, St. Betti 3.529ff. An intermediate opinion is possible: that what was originally a legislative function degenerated into that merely of witnessing. Thus Girard 853 supposes that the change was definitely made by the XII Tables. He points out, 855, that the will in procinctu must have been introduced subsequently to this change, for the citizen-soldiers can never have had any

function but that of witnessing in this form of will, and it is improbable that the

iuniores (men under forty-six), who alone formed the active army, should have had a freedom of testation denied to their elders. (On Scherillo’s view, above, 127 n 6, both wills were legislative acts.)

5 And the appointment of tutors, if the version of Tab. v.7 given in Hpit. Ulp. 11.14 is correct: wutt legassit super pecunia tutelave suae rei, ita ius esto; see below, 139 n 7. 6 Essays in Legal History, ed. Vinogradoff 120ff., repr. Labeo 12 (1966) 358ff.; also SZ 87 (1916) 129ff. For criticism see Buckland, LQR 32 (1916) 97¢f.

7 Below, 243. 8 Died in 184 B.C. (if Cicero, Brut. 60, is reliable).

9 Below, 244.

129

Family and succession at the time of the XII Tables also very difficult to believe that the peremptory heres esto (the words by which the heir is appointed) originated, as Lenel contended, in the mancipa-

tory will, which was an adaptation of a conveyance, rather than in the original comitial form itself, where in a quasi-legislative form the imperative

appears quite natural. |

To some extent, however, Lenel was concerned to deny that the primitive will could appoint an heir from outside the family. Roman law, he argued, would, if this had been so, have been unique. No other early system of law knows of a universal succession independent of blood relationship either real or fictitious (i.e. by adoption). What is found in both Greek and early Germanic law is that testamentary dispositions begin by being of two kinds. On

the one hand there is adoption, which enables a childless man to provide himself with an heir who will take the position of a natural son, and on the other hand there are individual gifts of particular things, made during a man’s lifetime but taking effect only at his death. In Lenel’s view the Roman equivalents were adrogatio and the comitial will. Although here too majority opinion has not followed him, the divergence is not in all cases very sharp, for the question which we are asking has changed. We are no longer asking simply whether the comitial will was confined to the making of legacies, but are concerned more widely with the extent to which the will could interfere with the order of succession on intestacy. This wider question needs separate

examination. .

(iv) Who could be appointed heir by will?

In the mature law, as we have seen,’ anyone may be appointed heir, whether he is simply one of several sui heredes or comes from altogether out-

side the family, but this freedom of testation is subject to the requirement that if a suus heres is not appointed he must be expressly and formally disinherited. In the view dominant among modern scholars,” this entrenchment of the rights of, the suz reflects an original state of the law in which they could not be disinherited at all. The paterfamilias was not free to dispose of the family property on his death. His powers were those of a trustee or administrator of the property, and, together with his functions in connexion with the family sacra, they devolved automatically on to the swus (or the sui jointly). At this stage there would be no place for a will. A need would ‘first be felt in the case where there were no sui and where in addition it was for one reason or another undesirable that the nearest agnate should take.’

This need could be met (as it was, as we have just seen, in other early systems) by the device of adoption, and this would account for the apparent 1 Above, 123f.

2 Kaser, RPR 1.92ff., 95ff.; Wieacker, Fschr. Siber 1ff.; Arangio-Ruiz, Istit. 510. This view does not differ essentially from that of Maine, Ancient Law, ch. 6, and other

nineteenth-century writers. 3 Cf, above, 125. 130

Testamentary succession similarity between the comitial will and adrogatio, the will being perhaps in origin an adrogation with effect delayed until the death of the adrogator. If and so long as it was regarded as an adrogation, it would of course have been

confined to the appointment of an extraneus in the absence of sui, and in particular could not have been used to appoint one suus to the exclusion of the others. Whether this was ever possible we have no means of knowing, because the whole will was early obsolete and our information is accordingly very scanty.*

This account is in its broad outlines widely accepted. Against it must be set the view of Bonfante, which has already been mentioned.*® For him the will was not in origin merely an exceptional device to provide for the devolution of property in the absence of suz heredes, but was from the earliest times a normal institution, serving to designate one son out of several as the successor to the administrative headship or sovereignty of the joint family (the property remaining in the undivided ownership of all members). It thus took the place which in some other systems is taken by the principle of primogeniture. Designation by a predecessor always played an important part in the succession to public office at Rome,® and since the chieftainship was analogous to a public office, it is natural to suppose that the Romans, among whom primogeniture was quite unknown, used designation to provide for this case too. Only thus, in Bonfante’s opinion, can we account for the predominance of testamentary over intestate succession even in the earliest times, and only on the supposition that the heir succeeds to a type of sovereignty can we

explain the rule nemo pro parte testatus, pro parte intestatus decedere potest.’ This maxim has no justification in a system which looks at succession merely from the point of view of property, but it is perfectly intelligible that a family cannot have two heads, one appointed by will and the other by

direct operation of law. Legacies are quite a different matter; they are a development of donationes mortis causa and it is to them that the words of the XII Tables, uti legassit etc., are to be applied. Not until long afterwards

did it become possible to appoint an heir who was not a member of the family.®

4 At first sight it would seem that the provision of the XII Tables, si intestato moritur cut suus heres nec escit, implies that a man who had sui heredes might nevertheless appoint other heirs (see e.g. Franciosi, Labeo 10 (1964) 352ff.), but it can equally well envisage merely the case of the man who, having no sui heredes, is in a position to make a will and fails to do so. For a third interpretation see below, n 8.

5 Above, 126, with reff. 6 Cf. above, 30 and 55, and below, 341.

7 *No one can die partly testate, partly intestate’, i.e. if a man leaves a valid will at all, it must regulate the whole of his succession. Hence, if X was appointed heir to an individual thing, even though the testator intended the rest of his property to devolve as an intestacy, X took the whole; Buckland 296. 8 Lévy-Bruhl, Now. Et. 33ff. (ef. id., Mél. de Visscher 2.137ff.; Stojéevié, Mél LévyBruhl 273ff.) makes an extreme application of Bonfante’s theory to the provision of the 181

Family and succession at the time of the XII Tables Bonfante’s theory has the great merit of accounting for the will as an immemorial institution without the assumption that there could be, even in primitive times, a ‘ universal successor ’ unconnected with the deceased by blood or adoption, but (even apart from the objections, set out above, to the thesis of a primitive Roman joint family) it raises more difficulties than it solves. It cannot, for instance, explain the rule that suz must be disinherited expressly if they are not to succeed. This rule must in fact rest on an original right to succession vested in all suz, which could only be defeated by a direct exercise of despotic power by the head of the family. On Bonfante’s theory that succession of one son under a will was normal, there can have been no

such right, and the further rule that the birth of a suus (postumus)? after the making of a will, upset it, becomes quite incomprehensible.* If a man had made a will appointing one son, it would have been nonsensical to invalidate

that will because a child was born whom no one would certainly think of appointing to ‘ sovereignty ” at least if others were available, until he had grown up. The successoral rights of woman are also difficult, if not impossible, to explain on Bonfante’s theory.*

Without more evidence the riddle must remain unsolved, but the more probable conjecture is that with the Romans, as with other peoples, a will appointing a ‘ universal successor ’ could originally be made only where there was no natural descendant to succeed as of course. It may well be that it was

only with the development of the mancipatory will that it became possible both to achieve what Bonfante saw as the prime function of the will—to choose one suus heres to the exclusion of the others * — and, as a logical con-

sequence, to exclude all the suz and appoint an extraneus in preference to them. It is likely also that the comitial will was confined to the appointment of an heir, and that the earliest purpose of the mancipatory will was the making of legacies, but this question is considered more fully below.® XII Tables: si intestato moritur cui suus heres nec escit ... (above, n 4), according to which the words mean ‘If a man dies intestate and therefore no swus becomes his heir ..-’, the assumption being that heres applies only to a son succeeding under a will to the sovereignty of the family; if not appointed in a will, he would acquire the property as

1 Gai. 2.123ff, 2 Gai. 2.130ff.

nearest agnate, but not the headship (D. 38.16.12). 9 126.

3 Rabel, SZ 50 (1930) 295ff., at 330. The whole article is a searching criticism of

Bonfante’s theory. 4 Rabel, «bid. 321. 5 Wieacker, Fschr. Siber l1ff., argues that this development did not occur before the latter half of the fourth century B.C. Solazzi, Jura 4 (1953) 149ff., even contended that as late as the lex agraria of 111 B.C. the mancipatory will could not institute a heres. 6 242ff.

132

CHAPTER 9

Slavery and manumission at the time of the XII Tables Slavery is an institution common to all the races, whether civilised or barbarian, of the ancient world and it was certainly recognised from the earliest times at Rome. Its chief source was then, as at all times, capture in war, and there is no doubt that the rule which makes a child the slave of the person who owns his mother goes back to remote antiquity. Such minor modifica-

tions of this rule as exist in classical times are due to a late principle of favouring liberty wherever possible and were no doubt absent from the earliest law. The slave was always at Rome, so far as the law was concerned, a piece of property and any limitations on the power of the master to do as

he liked with him were, at the times of which we are speaking, still in the distant future. In two respects only can we say that the civil law recognised the humanity of the slave, (a) in the possibility that the master might acquire rights through him and (b) in his capacity to become a free man, incapable of being owned, on manumission.

(a) The civil law rule was that an act-in-the-law carried out by the slave enured to the benefit of the master, but could not place the master under any

liability. A slave could thus, in the later law, make a stipulation for his master, because stipulation is a unilateral contract out of which duties arise only on the side of the promissor. If then A’s slave says to B: ‘ Will you pay my master 10?’ and B replies ‘I will’, B thereupon owes A 10. Not so with a sale, because a sale is bilateral, and the seller only acquires a right to the payment of the price in return for placing himself under the duty of delivering the thing sold. If therefore A’s slave purports to sell A’s property to B, B cannot force A to do anything, but A can enforce the contract against B

if he likes, provided, of course, that he is willing to do his part. These examples belong to a period later than the XII Tables, but there is no reason

to doubt that the principle is ancient. A slave could, in classical times, receive a thing by mancipation on behalf of his master,’ and so presumably could he as soon as mancipation had become a mere conveyance, not necessitating the real payment of any price.” In iure cessio to a slave was however not possible because, as Gaius says, he could not claim anything as his own. The only way in which a man’s legal rights could be adversely affected by

2 Cf. below, 143f. 3 2.96.

1 For discussion see Buckland, Slavery 712ff.

133

Slavery and manumission at the time of the XII Tables his slave’s act was by delict on the part of the slave, for if the slave committed a delict the master must either pay damages, or lose his slave. But this cannot be regarded as an admission of the slave’s humanity because very similar rules applied in the case of damage inflicted by animals.*

(b) Manumission. The developed civil law knew of three methods of manumission, all of which had the effect, provided they were correctly carried out, of making the slave not only free but a citizen.

(i) Censu. This consisted in the enrolment of the slave with his master’s consent in the list of citizens drawn up at the census.® It was the slave who actually gave his name to the magistrate charged with the census, but the authorisation of the master was of course necessary, and if it were not given the magistrate would refuse to accept the name, as, no doubt, he might also refuse if he himself considered the man unfit for citizenship.

It is disputed whether this is a declaratory or a constitutive act, 1.e. whether the magistrate fictitiously records that the man already is a citizen, or whether it is the entry in the list which itself changes the man’s status (just as enrolment in the list of Latin colonists made a Roman filiusfamilias a Latin).® Since plainly the magistrate had power to change a man’s position within the list,” the more probable view ° is that at least for the early period

the constitutive interpretation is correct. In the later republic, when the avenues to citizenship were more numerous and the possibility of errors greater, the other view may well have prevailed.

(ii) Vindicta. A more common method was that which consisted in the use for the purpose of manumission of the form appropriate to an action in which liberty was claimed (causa liberalis).® It might, of course, happen that a person held as a slave wished to claim that he was really free. In such a case it

was not open to him to bring the necessary action himself; he had to get some citizen to act as claimant (adsertor libertatis) and the proceedings took

the form (so long as the legis actiones were in use) of a legis actio sacramento.’ The parties came before the magistrate, and the adsertor, touching the man with the wand (vindicta or festuca), asserted his freedom; then the defendant also touched the man and asserted that he was his owner, and finally after further formalities,? the matter was sent for trial by a zudex in the ordinary way.

46 Below, 171. 75 Above, Above, 51f. Gai. 1.131. 52.

§ Last, JRS 35 (1945) 35ff.; Daube, JAS 36 (1946) 6O0ff.; doubted by Watson, Persons 186f.; for a restatement of the declaratory view see Frezza, St. de Francisci 1.201ff.

9 But see below, 135 n 6. 1 Below, 180.

2 As in any other action tried by sacramentwm, but there was a special provision that the amount of the stake must never exceed fifty asses, so that people should not be discouraged from acting as adsertores; see Gai. 4.14. 184

Slavery and manumission at the time of the XII Tables When a master wants to use this process for the purpose of freeing his slave, he gets some citizen (in fact one of the magistrate’s lictors was commonly used) to act as adsertor and make a claim that the man is free; then, instead of contesting the claim (as he would do if involved in a real causa liberalis), he remains silent, and so admits the claim. He does, however, here too place his wand on the man, and it appears indeed that it is from this act

of his rather than from that of the adsertor that the proceeding takes its name.*® The process then ends with the declaration (addictio) by the magistrate that the man is free.* Manumission vindicta is thus in form a collusive action, but it cannot, any

more than in wre cessio, be explained simply as such. Not only does the magistrate ° know the purpose of the transaction, but the effects are not the same as they would be in the case of an ordinary action. For the manumission is not effective merely between the parties: the man becomes free for all purposes and he becomes free only from the moment of manumission, whereas the decision of a causa liberalis in favour of liberty would necessarily imply that the man had been free before the action was begun. He becomes, too, the freedman of his former master, whereas if he had been free all along he might not be a freedman at all but free born. Manumission vindicta is in fact a form of cessio in ture, and here too we have to deal with a

hybrid institution, beginning no doubt as a collusive action, but coming to be recognised as something quite different.°

(iii) Testamento. Manumission by will was the most common of the three forms, for a man who dislikes parting with his property during his lifetime, may have no objection to being generous at the expense of his heirs. The gift of liberty must, like a legacy, be in authoritative words, the usual form being

Stichus liber esto or Stichum liberum esse iubeo; like a legacy, too, it is dependent on the validity of the will as a whole, and, where the heir is an extraneus, takes effect only from the moment of his entry. 3 Karlowa 2.133.

4 It appears from certain non-legal texts (Karlowa 2.1383 n 4; ef. Wlassak, SZ 28, 1907, 2f.) that the master also gave the slave a box on the ear and turned him round, but it is probable that these acts (for which it is easy to find symbolic meanings: Nisbet,

JES 8, 1918, 8ff.) had no legal significance, and they may not even be primitive (Charvet, SDHI 19, 1953, 334ff.).

5 Not necessarily the praetor or proconsul; any magistrate with imperium could exercise this ‘ voluntary jurisdiction ’.

6 Below, 149f. It is noticeable that the Romans themselves never speak of either manumissio vindicta or in ture cessio as fictitious, whereas Gaius (1.113, 119) does call mancipation imaginaria venditio; Wlassak, SZ 28 (1907) 76. Lévy-Bruhl, QP 56ff., accordingly argues that it has no connexion with the causa liberalis but is a straightforward act of manumission by the master, followed by the praetor’s addictio. See also Meylan, St. Arangio-Ruiz 4.469ff., and ef. below, 150. But the features on which LévyBruhl relies are probably to be explained as later developments; Kaser, SDHI 16 (1950) Tor,

135 J.—6

Slavery and manumission at the time of the XII Tables Of the relative antiquity of the three forms we can say little. Manumission by will certainly existed as early as the XII Tables, for we find a reference to conditional gifts of liberty,’ and a condition could be attached only to this form. Whether the other two forms also existed at this time we do not know.® It has been thought surprising that the manumitted slave became not only free but also a citizen, the private individual having thereby a power of conferring the citizenship, and it has therefore been conjectured that the rule was not primitive. But this involves a sharp separation of liberty and citizenship which is foreign to ancient ideas.® In a system of personal laws, if a man is free, he must necessarily belong to some community by whose laws he lives, and since the slave is made free according to Roman law, the community to which he belongs must necessarily be that of Rome. In the same way, a slave manumitted at Rome by a peregrine took the status of his master.” All this is no more surprising than the common principle of the: modern world that a foreign wife takes, or is entitled to, the nationality of her husband. In both cases a private citizen can be seen as exercising a power to confer the citizen-

ship. Indeed it would seem that in the Roman case it is in origin not the citizen by himself who confers the citizenship. For in manumissions vindicta and censu the public authority is clearly involved in the shape of the magistrate, and if manumission testamento goes back to the comitial will, it must originally have required the approval of the assembly. Eventually, of course, as the private will per aes et libram developed, and as, with the increase in population, it became impossible for the magistrate to know anything about the people brought before him for manumission, this control became illusory, but its existence in the early stages is presumably to be explained precisely by the fact that manumission did confer citizenship. In the Greek cities, it is true, the freed slave was given the status only of metic or resident foreigner, but the metic was a member of the community in

which he resided, albeit not a full citizen member. In a similar way, the Romans eventually, by the lex Iunia (?17 B.C.) conferred on the imperfectly

manumitted slave the status of Latinity.* In the primitive Roman society there was no such category of second-class member. If anything is surprising it is that none was created (or adapted) until the lew Iunia. 7 Tab. vir.12. 8 Cf. Daube, Forms 72ff.; Buckland, Slavery 443.

® For what follows see Volterra, St. Paoli 695ff. 1 Above, 103. 2 Volterra, St. de Francisci 4.73ff. 3 Below, 345. But the metic was better placed than the Junian Latin.

136

CHAPTER 10

The law of property at the time of the XII Tables 1 RES MANCIPI AND NEC MANCIPI Most, if not all, systems of law are compelled by the very nature of things to distinguish landed from all other types of property, for land is necessary for the production of food and the erection of dwelling-places, and, unlike other

things, it cannot be moved. In English law, in the guise of the distinction between ‘ real’ and ‘ personal’ property, this contrast is particularly marked ; in Roman law, though it existed as early as the XII Tables,’ its importance is overshadowed by a different and purely Roman distinction, that between res mancipi and res nec mancipt. Res mancipi were land subject to Roman

ownership,” slaves, beasts of draft and burden, including cattle, and rustic servitudes * belonging to land subject to Roman ownership; res nec mancipi were all other things. In the developed law the point of the distinction is that full Quiritarian ownership in res mancipi can only be transferred by the solemn method of conveyance known as mancipatio or the equivalent ceremony of in iure cessio, whereas the ownership of res nec mancipi can be transferred by mere delivery (traditio). Thus if A wants to make B the gift of a sheep and delivers it to him for that purpose the sheep becomes immediately the full Quiritarian property of B, but if he wants to give him an ox and does the same the ox remains the property ex ture Quiritium of A, because an Ox iS a Tes Mancipi whereas a sheep is not.

It is not difficult to see some reason why the particular classes of things known as res mancipi should have been specially treated; they are the things which are most important in a settled community which is both agricultural and warlike: the land, the slaves and beasts with which it is worked, and the rights of way and water without which, if it is away from the public road or has no water on it, it cannot well be farmed, and the horses used in battle.' 1 The times for usucapion are different according as the thing in question is mobilis or immobilis; see below, 152.

2 After the extension of citizenship to the whole of Italy (above, 66) this meant ail land in Italy, but when the citizenship was further extended the deduction that the land of the new citizens was capable of Quiritarian ownership was not drawn. Occasionally however land in the provinces might be assimilated legally to land in Italy by grant of

the ius Italicum; below, 267. 2 Below, 157. 4 The list is not intelligible if seen purely in terms of agricultural importance (de

Visscher, SDHI 2, 1936, 263ff.). Indeed, it can be argued that in an agricultural society 137

The law of property at the time of the XII Tables Such things must not pass from hand to hand as less important things may; if the ownership in them is to change there must be a public act of transfer of which witness can afterwards be given in case of any dispute.® Neverthe-

less it is probable that the category of res mancipi as it appears in the developed law is not exactly the same as it had been in some earlier period. The ceremonial of mancipation is not at all appropriate to a conveyance of land, because it includes the grasping of the thing to be acquired by the transferee ° and the grasping of land is an impossible, or at least an undignified, gesture.” And it may well be that the reason why land was not originally included among the res mancipi was that it was not capable of private ownership, or if capable of ownership, was not alienable but had to descend from

father to son. We know that very early law does not as a rule recognise private property in land; among the Germanic tribes, for instance, it was unknown in Caesar’s day,® and even in the time of Tacitus it existed only for the homestead, the remainder of the land being owned by the village com-

munity as a whole and redistributed to individuals for cultivation every year.” Traces of original clan or tribal ownership are also to be found in Greek law,’ and indeed, so long as there is plenty of land for everyone who can till it, and a great part of the available land is necessarily left fallow each sheep and goats would certainly count as an important element of wealth, and should therefore have been classed as res mancipt. But the point is perhaps that a res mancipt must not only be important in bulk, but individually identifiable. The smaller beasts are not so easily identifiable as the larger, and a person who has received one of them as one receives consumable things, without a ceremonious transfer before witnesses, is not so likely to be suspected of theft.

5 Below, 145. For de Visscher, op. cit., the distinction is not an economic one at all. The key is the word mancipium, of which the root meaning, in his view, is the power exercised by the paterfamilias over those human beings and animals which help him in his work and his battles (ef. above, 114 n 2). With the development of the idea of property,

the free subordinate members of the family came to be excluded from the category of res mancipt, though continuing to be objects of mancipation (it is indeed in relation to them that Gaius, 1.116ff., explains the ceremony), and eventually the idea of mancipiwm is swallowed up in dominiwm. Against this, however, the root meaning of mancipiwm seems rather to be the act of taking with the hand which occurs in mancipatio, the power which results from the act being only a derivative sense; and the inclusion of land as the object of a power of command is difficult to explain (de Visscher has to justify it on the ground that it is the area in which the power is exercised).

6 Below, 144. :

7 There is no reason to think that the land was symbolised by a elod of earth in the case of mancipation (though it was in a legis actio sacramento; Gai. 4.17). For Gaius could not have omitted to refer to such a practice, had it existed, in his description of the

8 B.G. 6.22. |

mancipation of land in 1.121. In the case of servitudes, grasping was of course completely impossible.

9 Germ. c. 26. See however Dopsch, Appendix 111 to Reeb’s edition (1930), 152ff.

1 Vinogradoff, Hist. Jurisp. 2.206. Bonfante, Scritti 2.260ff., in accordance with his general conception (above, 126) saw res mancipt as in origin those things which were owned by the joint-family or gens. 188

Res mancipi and nec mancipi year because the proper use of manure is unknown, the desire for individual

ownership will hardly make itself felt. It is then likely enough on a priori grounds that there was a time at Rome also when land, or the greater part

of it, was not owned privately, and this probability is supported by the tradition that at the foundation of the city Romulus distributed a $ lot’ (heredium) of two iugera to each citizen.” As this would not be nearly enough to support a family the distribution would imply that there existed also some form of communal cultivation, the heredium being the homestead merely, and very probably, as its name seems to imply, inalienable. But this tradition is not particularly trustworthy,® and in any case we do not know what form the communal ownership of the remaining land took, whether it was distributed among the different gentes, as is most probable, or among some other subdivisions of the state. At the time of the XII Tables at any rate, it is clear

that private property in land already existed and was not confined to the heredium; the definite rule for the usucapion of land * is sufficient evidence, and indeed the early history of the struggle between the orders * is incomprehensible without this assumption. On the other hand, it has been conjectured that ownership, in the sense of a right protected by the vindicatio, was

at first confined to res mancipi, a man’s interest in res nec mancipi being enforced only to the extent of an action for theft.° Another difficult question is as to the connexion, if any, between the divi-

sion of things into res mancipi and res nec mancipi and the perhaps even older division into familia and pecunia. These two words occur, sometimes together, sometimes separately in a number of ancient legal phrases, and the confusion of the versions in which they are preserved both makes it difficult to know what the original distinction was and shows that in historical times it was no longer understood.’ Familia could by then include a man’s entire 2 Varro, #.R. 1.10.2, Bruns 2.62; Pliny, H.N. 18.2.7; see Kaser, BPR 1.122; MayerMaly, SZ 77 (1960) 40ff.

3 It may be a dating back of what occurred in quite different circumstances when colonies were founded; Lenel, Gesch. 1.311. 4 Below, 151f.

5 The struggle was to a certain extent one between rich and poor (above, 10f.) and in early times this can only mean those who are rich or poor in land. It is however possible that land remained inalienable even though privately owned. 6 Kaser, SZ 68 (1951) 147£f., pointing to the similarity between the words used in the mancipatio (below, 143) and in the legis actio sacramento in rem (below, 180), and to the cumbersomeness of the ritual of the latter for the less important types of property. See further, below, 141.

7 For a thorough examination of the texts see Albanese, 4Pal. 20 (1949) 258ff. The words appear together in the formula spoken by the familiae emptor in the mancipatory will (Gai. 2.104; below, 242); familia appears alone in the XII Tables rules on intestate succession (above, 124) ard in the phrases familiae emptor and actio familiae erciscundae; pecunia appears alone in the words quoted by Cicero from the XII Tables concerning the curator of a lunatic (Tab. v.7): st furtosus escit adgnatum gentiliumque in €0 139

The law of property at the time of the XII Tables property, but more narrowly it denoted his slaves (famuli). Many * therefore have held that familia was the early equivalent of res mancipi (before the inclusion of land), and pecunia (from pecus) the equivalent of res nec mancipt. Familia (probably etymologically connected with domus) would thus originally have been the ‘ household property ’, especially slaves, and pecunia

the remainder, especially the flocks and herds. An attractive alternative view ° is that the distinction is not the same as that between res mancipi and

res nec mancipi, but that familia is that property (including res mancipi) which is necessary for the existence of the family as an agricultural unit and

which cannot therefore be freely alienated by the paterfamilias, whereas pecunia is his ‘ personal’ property. In view, however, of the inconsistent and confused usage in the surviving texts, the more cautious conclusion ' is that there was no sharp dichotomy, but rather that the two words represent the two main items of early property (apart from land), i.e. slaves and animals, and therefore together stand for the entire family property.

2 OWNERSHIP We have spoken so far of ownership as something which explains itself, but in order to understand the next point it is necessary to enquire a little more closely into its nature. The Roman law of classical times is dominated by what is commonly called the absolute conception of ownership which it has

evolved and by the action through which this right is asserted, the vindicatio. Ownership, in the developed law, may be defined as the unrestricted right of control over a physical thing, and whosoever has this right can claim the thing he owns wherever it is and no matter who possesses it.” If I possess a thing and you own it, then all you have to do is to prove your ownership and I must give it up; it is not necessary for you to allege that I have done you any wrong. I, on the other hand, the possessor, have nothing to do but

to sit tight and wait for you to prove your right; if you do not succeed in proving that you are owner, I remain in possession. Now this very clear-cut - conception is not to be found in all, or even most, systems of law. English pecwmaque eius potestas esto. What the precise form of the rule in the XII Tables concerning testamentary succession (above, 129) was is uncertain as our authorities give different accounts. Gaius (2.224) and Pomponius (D. 50.16.120) give wit legassit suae rei,

ita wus esto, pit. Ulp. 11.14 gives uti legassit super pecunia tutelave suae rei, and Paul (D. 50.16.53 pr.) similarly speaks of pecunia and tutela (though the passage is corrupt). Cicero on the other hand (Inv. 2.148) gives uti super familia pecuniaque sua. 8 E.g. Mitteis, RPR 81; Karlowa 2.73; Bonfante, Scritti 2.206f. ® Wlassak, Studien 42ff.; Wieacker, F'schr. Stber 1ff.; cf. Festus (Bruns 2.22): Peculawm servorum a pecore item dictum est, ut est pecunia patrum familiae. 1 Kaser, APR 1.45; AJ 159ff.

2 Subject, of course, to any special right the possessor may have and subject to the rules of ‘ bonitary ’ ownership, below, 264f. 140

Ownership law, for instance, has never known an action corresponding to the vindicatio, at any rate with respect to movables. The actions by which an owner recovers his thing which has got out of his possession all allege that the defendant has done some wrong to the plaintiff, that, for instance, the defendant ‘* unjustly detains ’ the thing * or that the plaintiff lost it, that it came into the hands of the defendant and that the defendant has converted it to his own use;* to use technical language, the actions for the recovery of movables all ‘ sound in tort ’. Further, the wrong alleged is, strictly, not one to ownership, but to possession, or the immediate right to possession, for, if the thing be ‘ bailed ’, e.g. lent, by the owner to another,’ it is the bailee who can bring the action against the third party in whose hands the thing is found, the bailor being confined to his rights against the bailee. Hence it can be said that ‘ although the bailor was the owner, the sum of his rights as owner was criginally his better right as against the bailee to get possession; for this better right to get possession was the only form of ownership which the mediaeval common law recognised ’.® This idea of a relative right to possession as contrasted with the absolute Roman dominium, is common to other Germanic systems of law besides the English. In origin, however, it is said, Roman law was less clear-cut, both in the sense that full ownership was confined to res mancipi, and in the sense that even this ownership was not absolute but relative. These contentions must be briefly considered.

In the developed law, as we have seen, ownership applies to all sorts of things, res nec mancipi as well as res mancipi, the only difference being that res mancipt need mancipation (or in iure cessio) if they are to be transferred, whereas res nec mancipi do not, but Kaser ’ conjectures that at one time the

distinction went further, and that res nec mancipi originally did not need mancipation and indeed could not be mancipated, for the simple reason that they were not capable of being fully owned, and that therefore the assertion of full ownership ex ture Quiritwum which is a necessary part of mancipation,*® as of in ture cessio ®° would not be applicable to them. The right which a man

acquired in them would no doubt be protected by the law of theft, but not by

an actio in rem, with the result that the owner could assert his right only against a person who had come into possession of the thing in bad faith. If this was ever the law, it had certainly ceased to be so before the end cf the legis actio period. For Gaius’ in his account of the legis actio sacramento

8 “ Detinue ’. 4 ‘Conversion ’.

5 For a fixed period; the ‘bailor at will’ has the immediate right to possession; see generally Winfield and Jolowicz on Torts, 9th ed. 431ff. 6 Holdsworth, HEL 3.337. 7 See above, 139 n 6; in addition to the reasons there given, he adduces the stipulatio habere licere, which in his view protected the purchaser only from interference hy the seller or his successors; see below, 292. See also Kaser ZPR 68 n 14.

8 Below, 144. 9 Below, 149. 1 4,17. 141

The law of property at the time of the XII Tables in rem,” of which an assertion of Quiritary ownership formed an essential part, mentions a column, a ship, and a flock of sheep as things which could be claimed (but which, since they could not conveniently be brought into court, were represented by parts only). It is not even beyond doubt that res nec mancipi could not be mancipated * (they could certainly be conveyed by in ture cessio), though the question could only in practice arise if the mancipatio were not accompanied by a handing over of the thing (traditio). | The second point is more fundamental. We have seen that one of the characteristic features of the classical law is the position of the defendant in a vindicatio; he himself need not show how he came to be in possession, need not, in technical language, show title; so long as the plaintiff does not prove title he can keep the thing. But in the earliest form of vindicatio known to us this appears not yet to be the case.* If we look at the form of the legis actio sacramento in rem as recorded by Gaius ° we see that not only the plaintiff

asserts that the thing in question is his, but the defendant makes a similar assertion; then the plaintiff asks the defendant on what grounds he makes the claim. It would seem therefore that the defendant could not simply rely on his possession, but had to show some title. The issue, it is said, is simply that of which party has the better right, as it is in Germanic systems, whereas in classical law the plaintiff has to show absolute ownership. This is superficially correct, but one must beware of being misled by merely verbal distinctions. On the one hand, the contrast between absolute and relative owner-

ship cannot be as clear in practice as it is in theory, simply because it is rarely possible to prove ownership. Even in the classical vindicatio the plaintiff must in practice have shown merely a better right to the thing.’ And on the other hand, we do not know what happened in the legis actio sacramento in rem if the judge was satisfied by neither party’s claim.’ Without this piece of information the distinction between absolute and relative ownership hangs in the air. 2 Below, 180.

3 See Bush, Juridical Review 52 (1940) 66ff., citing in particular Pliny, N.H. 9.35.58 and Tac. Ann. 1.73, as against Cic. Top. 45 and Lpit. Ulp. 19.3.7-9.

5 4.16. ;

4 Kaser, HB 6ff.; convenient summary in English in Tydskrif vir Hedendaagse Romeins-Hollandse Keg (1964) 5ff.; criticised by Voci, Modi di acquisto 279ff.; ef. also Watson, Property 94ff. (but see below, 181 n 2); and see Kaser, RPR 1.124f.

6 Cf. below, 154, 266, and see further, Nicholas, Introduction 154ff. Kaser’s view has

been widely criticised by Italian scholars, but not, as here, on the ground that the contrast with a supposedly ‘absolute’ classical concept is illusory, but on the contrary on the ground that the primitive as well as the classical concept is ‘ absolute’; see e.g. Talamanca, TR 26 (1958) 243ff.; L. Capogrossi Colognesi, La struttura della proprieta (Milan, 1969) 121ff., 396ff.

7 Gaius says that the defendant did not necessarily retain interim possession of the thing, ie. until the action was decided, for the praetor gave this to ‘one or other of the parties’, We do not know what factors the praetor took into account in making his 142

The law of property at the time of the XII Tables

8. METHODS OF ACQUIRING OWNERSHIP The Romans themselves did not devote much discussion to the abstract questions considered in the preceding paragraph; the main topic which they con-

sider under the heading of ownership is that of the methods by which it is acquired. These methods they divide into civil law methods and natural law methods.® The civil law methods discussed are mancipatio, in iure cessio and usucapio, those of the natural law include occupatio (i.e. the right of a man who has been the first to take possession of an ownerless thing such as a wild beast), spectficatio (the right obtained in some cases by the maker of a new thing out of existing materials), traditio (informal delivery of possession) and others. The civil law methods are open only to citizens (or those with commercium), the natural law methods, being part of the ius gentium, are open to foreigners as well. In this connexion Justinian ® ventures an historical conjecture. ‘ It is more convenient ’, he says, ‘ to begin with the older law, and

it is clear that the natural law is the older, seeing that it is the product of Nature herself and so coeval with the human race; for civil rights only came into existence when states were first founded, magistrates appointed and laws written down.’ Now, in a sense, as applied to methods of acquiring ownership this is true; no doubt grabbing came before mancipation, but in another

sense it is the reverse of what we know now to be the general truth with regard to the historical evolution of law, namely that formality (and the civil law methods are characterised by formality) comes before informality. The recognition that mancipation gave title and the working out of legal rules on the subject certainly preceded the recognition in legal theory of the rules of occupatio. Traditio similarly escaped notice originally as a method of acquiring ownership, because, though a man who received a res nec mancipi in this way was protected, the informality of the method disguised its legal importance. In speaking of the law of the XII Tables we can thus confine ourselves to the civil law methods.

A. Mancrpatio. * Mancipatio’, says Gaius,’ ‘is a sort of symbolical sale ’ (imaginaria venditio). For its accomplishment are needed the two parties to the transaction (the transferor and transferee), at least five witnesses, who must be Roman citizens above the age of puberty, a pair of scales and another decision. Of course not even the classical vindicatio established the successful plaintiff as the absolute owner in the sense that the judgment was valid against a third party who might subsequently seek to vindicate the thing from him, (The contrary has, however, been maintained for the legis actio procedure — reff. in Kaser, ZPK 93 n 36; but see

Pugliese, RIDA, 1959, 347ff.) 8 Gai. 2.65; J.2.1.11. 8 The words are attributed in the Digest (41.1.1 pr.) to Gaius, Res cottidianae (on

which see below, 388 n 3*). 1 1.119. For textual questions see David-Nelson ad h.l.

143

The law of property at the time of the XII Tables citizen of full age to hold them (libripens), and a piece of bronze (aes, also called raudusculum).? The ceremony consists in the transferee’s grasping the thing to be transferred, if it is a movable,’ and saying (literally,* and assuming that it is a slave that is being transferred) ‘ I assert that this man is mine by Quiritary right, and let him have been acquired to me with this piece of bronze and a bronze scale’ (Hunc ego hominem ex iure Quiritium meum esse aio, isque mihi emptus esto hoc aere aeneaque libra). Then he strikes the scales with a piece of bronze and gives it to the transferor * by way of price ’ (quasi preti loco).° Such was the ceremony in the classical age, and at that time mancipation was a mere conveyance, i.e. whatever the reason for which one man wanted to transfer the ownership of a res mancipi to another, mancipation was open to him as a means of so doing. It might be that the reason was a sale, i.e. _that he wanted to convey the ownership in the thing because he was under an obligation to do so arising out of a contract of sale, but it might also be any other reason, because e.g. he wanted to make a gift or to fulfil a promise he had made by stipulation, and even where the reason was a sale, the sale is quite distinct from the mancipation; the sale is the contract out of which arises the obligation to convey, the mancipation is the conveyance by which that obligation is fulfilled. The piece of bronze given by the transferee to the transferor has no connexion with the real price even if there has been a preceding sale and it is given equally when there is no sale at all. But at one time, at least according to the traditional view, mancipation had

not been a symbolical sale but a real one. Until coined money came into existence its place as a medium of exchange was, to a certain extent, taken by uncoined bronze, which naturally had to be weighed in order that the recipient might know that he was getting the amount for which he had bargained. The primitive ‘sale’ by mancipation was however quite different from the sale (emptio venditio) of the developed law. Whereas emptio venditio was a mere contract, i.e. an agreement giving rise to obligations (on the

part of the seller to transfer the thing sold and on the part of the buyer to pay the price for it) and quite separate from the subsequent conveyance and payment by which these obligations were fulfilled, the primitive mancipation was a sale and a conveyance in one, for there is no previous contract and the payment of the price (i.e. the weighing out of the metal) is part of the form

necessary for the conveyance. This ‘ sale for ready money’ (or rather for ready bronze) is in fact the only sort of sale known to primitive law. It used to be generally assumed that the occasion of the change from sale (of a sort) 2 Cf. Festus (Bruns 2.33): In mancipando cwm dicitur ‘ rudusculo libram ferito’ asse tangitur libra.

3 Above, 138. 4 See further, immediately below. 5 The formula may have included a statement of the price; see Paul, FV 50:emptus

mths esto pretio.

144

Methods of acquiring ownership to conveyance pure and simple by the substitution of a pretended weighing out of bronze for a real one must have been the introduction of coined money. For, of course, once this existed it could be counted instead of weighed, and the payment of the price necessarily became something separate from the mancipation itself. We know now, however, that money was only coined at Rome at a surprisingly late date. Recent authorities © put it in the early third century, more than 150 years after the XII Tables, and it is likely that the symbolical form had already evolved at the time of the XII Tables. In this connexion it is as well to distinguish three senses in which mancipatio can be described as a symbolical sale. Either there is a genuine conveyance on sale, but the price is not paid at the time, i.e. there is a sale on credit; or there is in substance no sale, but the ceremony is used to effect a conveyance on gift (e.g. a dowry for a daughter); or, finally, there is no sale and only notionally a conveyance, as when manczpatio is used for the making of a will,’ or for emancipation,® or for the creation of manus by coemptio.® It is likely, as we have seen,’ that at least the mancipatory will existed at the time of the XII

Tables, and it is commonly thought that the credit sale also was recognised.” * It is at any rate inconceivable that none of these uses of mancipatio had come into existence before the third century. The witnesses, who were the chief element in the ceremony besides the weighing out of the bronze, are presumably there to ensure the publicity of the act * and to prevent future disputes. The transfer of a res mancipi was,

no doubt, an important matter in early Rome and the man found to be in 6 H. Mattingley, Roman Coins (2nd ed., London, 1960) 3ff.; id., JRS 35 (1945) 65ff. ; R. Thonisen, Early Roman Coinage, vol. 3 (Copenhagen, 1961). Silver coins were first struck in 269 B.C., but there were cast bronze coins (aes grave) and bronze currency bars (commonly now called aes signatwm) a little earlier, very probably beginning in 289 B.C.,

when the tresviri monetales were first appointed (above, 56). 7 Above, 127. 8 Above, 90. 9 Above, 115. See also nexwm, below, 164. 1 Above, 128. 2 Justinian (Inst. 2.1.41) says that the XII Tables provided that the property in

things sold should not pass until the price was paid (or security given). He states the rule for traditio (the only conveyance surviving in his day), but it is unlikely that the XII Tables would have dealt with traditio at all (contra, Watson, Obligations 61ff.). There has been much controversy (the classical law also is uncertain); see de Zulueta, Sale 37f£.; Arangio-Ruiz, Compravendita 276ff.; Kaser, RPR 1.46, 418ff. with reff. It has often, however, been conjectured that in the XII Tables the rule applied to manctpatio, either in the sense in which Justinian applies it to traditio, or in the sense that the actio auctoritatis (below, 146) did not lie until the price had been paid; Buckland, 240 n 2; Arangio-Ruiz, loc. cit. Kaser, EB 111ff. See also below, 146 n 1.

3 Paul, FV 50, says that the XII Tables ‘confirmed’ both mancipatio and in iure cessio (but see Watson, Property 21 n 2). This has been linked (Girard 310) with the provision (Tab. v1.1), cwm nexum faciet mancipiumque, uti lingua nuncupassit, ita ius esto, but it is more likely (Buckland 238; Kaser, RP# 1.47) that this provision refers to terms added to the formula of the mancipatio (leges mancipio dictae) ; e.g. if A wanted to transfer ownership in a piece of land to B, but to reserve a right of way for himself, he could get B to say ‘TI assert that this thing, subject to a right of way, is mine’ ete. 4 Contra, Gallo, SDHI 23 (1957) 174ff. 145

The law of property at the time of the XII Tables possession of a slave or of cattle which the whole district probably knew to have belonged previously to someone else must be able to justify the fact. The obvious thing therefore, at a time when writing was little used, was to summon witnesses who would be able to testify afterwards that all was in order. The presence of the witnesses, who would of course normally be neigh-

bours, is also a certain guarantee to the transferee that the transferor has the right to convey the thing in question, for they would usually know what res mancipt the people in the district owned.° The effect of mancipation, when it is completed,‘ is, of course, normally to transfer ownership; but ownership can only be transferred by one who has it;

if the transferor was not the owner the transferee could not become owner either. In such case however the transferor was responsible if eviction took place, that is to say in case the real owner brought an action against the transferee and was able, by proving his title, to get the thing away from the transferee. When such an action was threatened it was the business of the transferee to inform the transferor and call upon him to support the title he had transferred (laudare auctorem, i.e. to call upon him as guarantor of the

title), and if the transferor failed to defend the action so that eviction followed, the transferee could bring another action against him for double the purchase price. This action was almost certainly called the actio auctoritatis,’ and is of great importance as the starting-point for the liability of the seller for title as developed in the later law.* Apparently the liability to the action arose necessarily, without any special agreement, from the mancipation itself, and could not even be avoided by a special agreement excluding it, for when it was desired to exclude it the practice was, at. any rate in the later law, to mention a single sestertzus (nummus unus) as the purchase price, so that the actio auctoritatis would be only for the nominal sum of two sestertizt and not be worth bringing. This was done, for instance, when the mancipation was used to transfer property by way of gift.® The liability however only lasted until the period of usucapion had elapsed. If two years in the case of land or one year in the case of movables had passed and the transferee had remained in undisturbed possession, then there would be no need for any actio auctoritatis because by that time the position of the transferee would have become unassailable,’ even if the thing had not originally belonged to the transferor. 5 For similar provisions in Anglo-Saxon laws see Holdsworth, HHL 2.81: ‘A person who bought secretly ran great risks. Not only might the real owner of the property claim his own, but he might also make a charge of theft against the purchaser. A purchaser who had bought secretly ... might find it hard to disprove this charge.’ 6 Provided (in case of sale) that the price has been paid, if there really was a rule to

this effect. 7 PS 2.17.3. 8 Arangio-Ruiz, Compravendita 310ff.; Girard, Mélanges 2.5ff.; Kaser, RPR 1.132;

Lenel, BP 542ff. ® Bruns 1.335; FIRA 3.241; ef. de Zulueta, Sale 43 n 4. 1 On the nature of usucapion see below, 151. The meaning of auctoritas and its relationship to usucapion give rise to much difficulty (for literature see Kaser, RPR# 1.134f.; 146

Methods of acquiring ownership Another action which might arise by reason of the mancipation was that de modo agri. Where a piece of land which was mancipated was stated to be of a certain area and was found afterwards, on measurement, to be smaller, this action lay for double the proportionate part of the price. This liability of the transferor by mancipation is certainly very old and probably as old as the XII Tables, for there are some words of Cicero quoting the XII Tables which almost certainly refer to it.” esp. Kaser, SZ 68 (1952) 155ff., Mayer-Maly, SZ 78 (1961) 221ff., Yaron, TR 35 (1967)

191ff.). In this context (there are other uses, e.g. auctoritas of a tutor), texts from Plautus onwards clearly give it the meaning of ‘ guarantee’, as above (see Mayer-Maly, op. cit.), but this meaning cannot easily be applied to the three earliest legislative uses of the word. (a) Tab. v1.3 probably ran: usus auctoritas fundi bienniwm, ceterarwm rerum omniwmn annus esto. Usus is probably here nominative (Cicero, Caec. 19.54, has lex usum ET auctoritatem fundi iubet esse bienniwm) and the text is usually taken to mean: ‘possession (for usucapion) and the duty to guarantee lasts two years...’ Mayer-Maly, op. ctt., prefers to take usus as a genitive and to confine the provision to imposing a limitation on the actio auctoritatis: ‘the guarantee of possession lasts two years...’ The

objection to these explanations (especially the latter) is that a limitation on the duty to guarantee only makes sense if it coincides with the period necessary for the acquirer to perfect his title by usucapion, and that period might last more than two years from the date of mancipation if possession were interrupted; Kaser, EB 26, quoted by Yaron, op. cit. 201. Kaser, SZ 68 (1952) 155ff., RPL 105, takes usus as a nominative and says that the sense of the provision is that the possession which depends on the alienor’s auctoritas lasts two years (se. after which time his position becomes unimpeachable) ; but to arrive at this meaning he has to assume that the provision was badly drafted, and in any event it seems to leave the main point (the perfection of the acquirer’s title) to be implied from an incidental consequence (the lapsing of the duty to guarantee). Of course, in accordance with Kaser’s thesis that ownership in the strict sense was at this time confined to res mancipi (above, 141) this derivation of usucapion from auctoritas confines usucapion also to res mancipi; see below, 154. (b) The same objection can he made against the interpretations usually offered of Tab. 111.7 (v1.4 in FIRA 1.44): adversus hostem (i.e. a foreigner) aeterna auctoritas esto. This is taken to refer to the exclusion of foreigners from usucapion (Gai. 2.65), and the meaning most commonly offered is ‘in favour of a foreigner the duty to guarantee lasts for ever’ (sc. because he cannot usucapt). To this there are indeed the further objections both that it gives an unusual meaning to adversus, and that an express provision in favour of foreigners is unlikely in the XII Tables. Kaser therefore suggests (RPL 106) ‘the guarantee (sc. and the position of the acquirer which depends on 1t) lasts eternally to the detriment of an alien’, ie. he cannot usucapt; but, apart from the initial objection already made, this gives a very forced meaning. See also Mayer-Maly, op. cit. 270ff. (c) The same objections apply against giving the same meaning to auctoritas in the provision of the lex Atinia (Gell. 17.7.1ff., below, 153): quod subruptum erit, eius ret aeterna auctoritas esto. Yaron, however, op. cit., following Lévy-Bruhl, Now. Et. 15ff., argues that in all three texts auctoritas denotes the power to recover a thing. This gives a straightforward meaning for Tab. 111.7 (the owner can always recover a thing from a foreigner, i.e. there is no usucapion) and for the lex Atinia (no usucapion of stolen things), but Tab. v1.3 has to mean that the possession necessary for usucapion lasts two years for land etc., and so does the owner’s power to recover it; and it can be objected that the second proposition is superfluous, since it is a necessary consequence of the first. Moreover, it is not easy to explain the transition from the supposed early meaning to the well-evidenced later mean-

ing of guarantee. 2 Cie. Off. 3.65 (Tab. v1.2). 147

The law of property at the time of the XII Tables The foregoing account of the primitive mancipatio as a genuine sale was until recent decades generally accepted, and it still seems in essentials to be

probably correct, but out of a great deal of controversy an alternative version, formulated by Kaser,? has emerged with a good deal of support. According to this version, mancipatio did not originate as a sale but as a unilateral taking, the purpose of the payment of the bronze before witnesses being to secure the acquiescence of the loser. If the loser failed to protect the

acquirer against a third party who claimed the thing, he was regarded as being guilty of a kind of theft of what the acquirer had paid and was therefore liable for twice that amount (actio auctoritatis).+ Three main pieces of evidence are adduced in support of this hypothesis.

First, mancipatio is clearly in form unilateral: it is only the acquirer who says or does anything, the loser (or seller) merely acquiescing in silence. But mancipatio is not alone in this. Buckland * has pointed out that it is a common feature of Roman formal acts in private law that the ritual actions and

words proceed only from the person benefiting or claiming, and has suggested that the exceptions in which there is a dialogue were sacral in origin. Secondly, the first part of the formula of the mancipatio (hunc ego hominem ex iure Quirittum meum esse aio) is identical with that of the vindicatio.® The two, Kaser therefore argues, are related. Just as the vindicatio is a conflict of two acts of force, which has to be settled by a trial, so also manci-

patio is a single act of force which does not go to trial because the other party accepts a composition.’ But the use of the same form of words need show no more than that in both cases there is a public declaration of ownership, In one case by agreement, in the other case not.® Finally, the second part of the formula (isque mihi emptus esto hoc aere aeneaque libra) seems illogical as it is usually translated (‘ and let him be bought to me with this piece of bronze and this scale ’). For the first part has just asserted that the 3 EB (esp. 136ff.); RPR 1.43ff., 131ff.; SZ 68 (1951) 174ff. De Visscher, Now. Et. 141ff. (=RHD, 1933, 603ff.) first advanced the idea that mancipatio was unilateral, but associated it with the contention that the main purpose of the act was the obtaining of the auctoritas of the loser, this being the only basis of the acquirer’s title until usucapio gave him ownership. It would, however, be strange if the main purpose of the act were

not mentioned in it. For discussion of this and intervening contributions to the controversy, see Kaser, loc. cit. It must be emphasised that here, even more than in most discussions of primitive Roman law, elaborate structures are built on very little direct evidence, eked out with evidence drawn from other primitive systems, the relevance of

which is itself conjectural. EB180. 122f. 5 Fschr. Koschaker 1.16ff.4 6Kaser, Below,

7 Kaser, EB 140, assumes that mancipatio goes back to a primitive stage of self-help between groups. 8 The absence from mancipatio of the laying on of the vindicta or festuca, which is a feature of the vindicatio, may also be significant. But it is not mentioned in the account of in ture cessio, which almost certainly is related to the vindicatio (below, 150), though the omission may there be merely a chance defect of the evidence, since the vindicta is mentioned in our account of the closely related manwmissio vindicta (above, 134). 148

Methods of acquiring ownership thing already belongs to the speaker. It has accordingly been said that the first part appears to state an untruth and the second to confess the untruth, and one or other part has been rejected as a later addition.*® Kaser, however,

argues that the original meaning of emere is not ‘ to buy’ but ‘ to take’, and there is then, on his view of mancipatio as a unilateral seizure followed by a composition, no illogicality. Against this Prichard? has pointed out

that the supposed illogicality derives from a mis-reading of the tense of emptus esto, which is not a present-future imperative, as both the traditional translation and Kaser’s variant assume, but a past imperative;* and, further,

that the original meaning of emere is rather ‘ to acquire’ than ‘ to take’. On Prichard’s view, therefore, emptus esto means literally ‘ let it have been

acquired to me’ and refers back to the symbolical grasping of the thing (mancipium) with which the ceremony begins. The whole act is a conveyance

on sale, the substantial meaning of the formula being ‘I declare that this man belongs to me by the rules of the civil law, and let my receiving of him have been legally effective by virtue of this piece of bronze properly weighed out as the price ’.°

B. In rure cessio. This, like mancipatio (to which it was frequently an alternative), was a form of conveyance, and, as its name implies, took place before the magistrate. In classical times, according to Gaius’ description,*

the parties go before a magistrate of the people, the urban or peregrine praetor at Rome, or the governor in the provinces, and the intended transferee, grasping the thing to be conveyed, says (if e.g. it is a slave) ‘I assert

that this man is mine by Quiritarian right’ (Hunc ego hominem ex ire Quiritium meum esse aio). The magistrate then asks the transferor whether he makes a similar claim (an contra vindicet); the transferor either remains silent or says ‘no’; and the magistrate then ‘ adjudges ’ (addicit) the thing to the transferee. This process is usually, and probably rightly, described as a sort of collusive action; in order to achieve the desired result of transferring the owner® De Zulueta, Gaius 59. Thormann, Miinch. Beitr. 33 (1943) 115, argues that the two parts derive from two different racial elements which made up the central Italian popu-

lation, Unilateral acquisition, in his view, was characteristic of the invading IndoGermans, the ‘ people of the spear’, whereas the ceremony of weighing belonged to the indigenous Mediterranean world of peaceable commerce. 1 LOR 76 (1960) 4124. 2 Cf. Tab. v1II1I.12: Si nox furtum faxsit, si im occisit, ture caesus esto.

3 Kaser also contends that the right which the taker acquires is, until usucapio is complete, only a relatively better one than the loser’s; and since the title is acquired by taking, it is a new one and not one derived from the loser. But the acquirer speaks of the thing’s being his ex wre Quiritium; and there is no suggestion that he could not vindicate at once. Moreover, his title is not new in the sense that it is free of the encumbrances or flaws which affected the title of the loser; cf. Voci, Modi di acquisto 37ff.;

Prichard, LQR 76 (1960) 418f. 4 2.24, 149

The law of property at the time of the XII Tables ship in a thing, the parties pretend that it already belongs to the transferee, who claims that it is his; his claim is then admitted by the transferor and confirmed by a judgment of the court. The process begins in exactly the same way as does a vindicatio under the old system of procedure by legis actio,° and the language of Gaius corresponds to this idea, for he speaks of the claim as a vindicatio and says that the process was called legis actio. There are however some important features which distinguish in iure cessio from an ordinary action.® In the first place the addictio did not, it seems, take place in an ordinary action; if the defendant made no answer the plaintiff would simply take the thing with him.’ Secondly the addictio cannot merely be construed as a judgment by consent, for a judgment is only effective between the parties to the action,® whereas in iure cessio has an effect as regards third parties also. Suppose for instance that A is in possession of a thing and B claims that it is his, brings a vindicatio and gets judgment. That judgment settles the matter between A and B, so that if A claims the thing again B can

simply answer that the matter has been decided; but if a third party, C, claims the thing from B, the judgment in the action B v. A is no answer to him. Similarly if C has got possession of the thing and B wants to claim it, he cannot allege the judgment in support of his claim. But if A had ceded the thing im iure to B and C got possession of it, then B can use this to support his claim and he will succeed if he can show that A owned it and then ceded it to him just as he would if he could show that A owned it and then mancipated it to him. For these and other reasons it has been held that in ture cessio was not a collusive or ‘ fictitious ’ action at all but from the beginning a conveyance in which the machinery of the court was used to give expression and effect to the will of the parties.? The chief objection to this view is that the Roman law never seems to have known any such principle as that the magistrate can give effect by his authority to the will of the parties to a transaction, and in fact in iure cessio is especially used to bring about results which the will of the parties was originally incapable of producing.’ Further, there are some rules which can only be explained on the view that in iure cessio is a form of litigation, especially the rule that persons in potestate cannot take part in one any more than in a legis actio.? The truth is that neither explanation alone suffices to account for all the facts and the institution must therefore be assumed to have a hybrid character. It began, in all probability, as a collu5 Below, 180. And see immediately above, on the formula of mancipatio.

6 Buckland 233f. 7 See however below, 180 n 8. 8. And those deriving title through them. See however above, 142 n 7.

9 Wlassak, SZ 25 (1904) 102; Lévy-Bruhl, QP 114ff.; Nouv. Ht. 141ff.; ef. Schulz, CRL 348f. Contra, Kaser, AJ 104 ff. 1 Cf. below, 151 n 6; Mitteis, RPR 278, See also Wolff, Tulane L.R. 33 (1959) 525ff.

2 Gai. 2.96; Schol. Sin. 49. 150

Methods of acquiring ownership sive action, an expedient invented by practitioners, but it became subsequently so common that it was recognised as a conveyance and emancipated

from some of the rules which would have attached to an action pure and simple.

Whether in wure cessio is of earlier or later origin than mancipation is not certainly known, for both go back beyond our records, but the probability is that mancipation is the older.* Mancipation belongs to a group of transactions ‘ by bronze and scales’ (per aes et ltbram) including a form of loan * and a method of discharging debts,® and both of these, like mancipation itself in its original form of sale, are institutions such as even the most primitive legal system needs. Those transactions, on the other hand, for which in iure cessio is necessary (and not merely alternative to mancipation) belong in no case, so far as we can see, to the oldest stratum; it is, for instance, the only

way of creating urban and personal servitudes inter vivos, and both these types are known to be of later origin than the rustic servitudes, which can be mancipated.* But in any case it is probable that in iure cessio existed already

at the time of the XII Tables, for Paul is recorded as saying that the XII Tables ‘ confirmed it ’, as well as mancipation.’ In the later law it was an alternative to mancipatio as a conveyance of res mancipi and to traditio as a

conveyance of res nec mancipi, but it was little used owing to the greater convenience of the other methods, and its chief sphere of application was to those ‘ incorporeal things ’ to which mancipation was inapplicable. As it was a civil law method of acquiring ownership it could only be used by citizens

and it was not applicable to provincial land, because, like mancipation, it involved the assertion of Quiritarian ownership of which provincial land was not capable. Its effect was also, naturally, to transfer Quiritarian property, provided the transferor was Quiritarian owner, but there was nothing to correspond with the actio auctoritatis, because, seeing that the process was in form a claim that the property already belonged to the transferee, the transferor could not logically be made liable for defective title. Nor, for similar reasons, could there be any question of making the passing of ownership dependent on the payment of the price where the ground of the conveyance was a sale.

C. Usucario. It will be best to begin with an account of the essential elements of usucapio as we know it in classical law. It then meant the acquisition of ownership by continued possession for a certain time, the requisite

3 Cf. above, 148 n 7. 4 Below, 164. 5 Below, 161.

6 Gai. 2.29, 30. Other cases are the transfer of an inheritance before acceptance by the intestate heir (Gai. 2.35), the transfer of the tutela of women by their tutor legitimus (Gai. 1.168ff.), and the claim made by the adopting father with which the ceremony of adoptio (above, 120) ends. Manumission vindicta is also a kind of in iure cessto, though the action which was brought collusively was not a vindicatio but a causa liberalis. 7 Above, 145 n 3. 151

The law of property at the time of the XII Tables periods being two years for land, one year for all other things. There was much elaboration of the conditions in which possession would lead to owner-

ship in this way; in particular it was necessary that the possession should have begun in good faith and that there should have been a tusta causa for its inception. This is not the place for a detailed discussion of these require-

ments, but their nature can be understood to some extent by taking the simplest and commonest case of usucapion, that is when a man buys and has delivered to him a thing which does not belong to the seller. In such case he will not become owner of it immediately, but provided he was in good faith (and provided the thing was not a stolen thing) he will become owner after his possession has continued for the requisite year or two years. Here the ‘ good faith’ (bona fides) required is the belief that the seller was qualified to transfer ownership in the thing; the tusta causa is the sale, ‘ an event or dealing which is normally the basis of acquisition ’.? Other such tustae causae are gift, legacy and dowry. In addition to the requirements of bona fides and iusta causa, there was a rule that stolen things could not be usucaped.® This, as Gaius is careful to point out,’ does not merely mean that the thief cannot usucape the thing that he has stolen, but that no one else can either, not for instance a man who has

bought it from a thief, however innocent the man may be. The combined effect of the rule forbidding the usucapion of stolen things and the wide definition of theft in Roman law ? was to make the usucapion of movables com-

paratively rare. Generally, when a man got into possession of something belonging to another, theft would have taken place at some time. Thus if A has obtained possession, even quite innocently, of a thing belonging to B and sells it to C, knowing that he has no right to do so, then C can never usucape

because in the very act of selling a thing which he had no right to sell A committed the theft and so made the thing a res furtiva. With immovables the case was different for, in the opinion which prevailed, land could not be stolen, so that in the case supposed above, if land were in question, C would be able to usucape.* The purpose of usucapion, as is explained by Gaius,* was that ‘ ownership of things should not remain uncertain for too long a time’, in other words that where a man has failed to take steps to recover his property he should, after a certain time, be deprived of it in favour of the possessor; people who have been in possession of a thing without disturbance for some time must not suddenly be confronted with claims founded on titles which the claimant

has not hitherto troubled to enforce. It must however be noticed that the periods in Roman law are short, and, though no doubt suitable at a time

8 Buckland, Manual 129. 9 Gai. 2.45; cf. above, 146 n 1.

1 Gai. 2.49. 2 Below, 170.

3 But a lex Plautia (between 78 and 63 B.C.) and a lex Iulia (of Caesar or Augustus)

excluded the usucapion of things taken by force; Gai. 2.45, 49ff. 4 2.44. 152

Methods of acquiring ownership when the city was small and people could find their property with compara-

tive ease, they were found inconvenient in later times when the empire expanded.* Another use of the institution was that it cured defective convey-

ances. If a res mancipi were transferred by delivery (traditio) only, or if there had been an attempt at mancipation but some mistake had occurred in the ceremony, ownership ex iure Quiritium did not pass, but, provided that the transferee remained in possession for the requisite period, he would subsequently become full owner in spite of the defect in the original conveyance. Usucapio, as foreign writers sometimes put it, cured defects both of substance and of form; when an attempted conveyance has been ineffective, either because the transferor had no title to convey (acquisition a non domino) or

because the form of conveyance adopted was inadequate (acquisition a domino), the transferee nevertheless becomes owner after a certain period has elapsed. Such, in outline, was the classical usucapio. In primitive law, however, it was almost certainly very different, though there is considerable controversy as to the precise extent of the differences.* That the exclusion of stolen goods

was laid down in the XII Tables we know from Gaius,’ but the rule is also attributed to the lex Atinia (probably of the mid-second century B.C.).°® Since we hear of a debate as to whether this lex was retroactive,’ it cannot have merely restated existing law, and the most probable explanation of the double attribution is that the rule was applied by the XII Tables only to

usucapion by the thief himself and was extended to usucapion by third parties by the lex Atinia.?

It is, moreover, unlikely that the requirements of good faith and iusta causa existed in early law.? If they had, the much clumsier requirement that the thing should not have been stolen would hardly have come into existence, whereas the reverse process is easily conceivable with the advance

of legal science, and would explain the co-existence of both rules in the developed law. Furthermore, this supposition derives support from the existence, even in the developed law, of several cases in which good faith was not required.* These are most easily explained as survivals from an older state of the law. 5 Longer periods were adopted for the provincial institution of longi temporis praescriptio, and Justinian prolonged the period of usucapion for movables to three years (Inst. 2.6 pr.). 6 See especially Mayer-Maly, SZ 77 (1960) 16ff., 78 (1961) 221ff., 79 (1962) 86ff.; St. Betts 3.451ff.; Yaron, TR 35 (1967) 191ff.

7 2.45; ef. above, 146 n 1. 8 See generally Jolowicez, De Furtis LXXXIX.

® Gell. 17.7.3; cf. Stein, Regulae 22ff.

1 Daube, CLJ 6 (1937) 217ff.; but see Yaron, TH 35 (1967) 215ff., and ef. Watson, Property 26f8. 2 Contra, Mayer-Maly, SZ 79 (1962) 97f.; but see Yaron, TR 35 (1967) 215.

3 Gai. 2.52ff. See de Zulueta, Gaius 71ff. The chief case is usucapio pro herede. If a person took possession of a piece of property belonging to a vacant inheritance (i.e. 153

The law of property at the time of the XII Tables Discussion has recently, however, centred principally around the question of the original function and effect of usucapio. It is usually supposed that usucapio is referred to in the provision of the XII Tables: * usus auctoritas fundi biennium esto, which is taken to contain two separate, though related, rules: that the time for usucapion (and during which the actio auctoritatis may be brought) is two years in the case of land. Kaser,*> however, takes the two more closely together and argues that the kind of prescription involved

is different from the classical wsucapio: the XII Tables do not recognise usucapio as a method of acquiring ownership, but are concerned only with what has to be proved in a vindicatio. The transferee in a mancipatio, if he needed to show title in a vindicatio before the one or two years had elapsed,

had to call upon the auctoritas of the transferor, whereas afterwards his position became unimpeachable. He then needed only to show his possession

for the requisite period, any better right being extinguished by its holder’s failure to assert it. Kaser’s view here is of course in accord with his thesis that the early vindicatio was concerned only with a relatively better title.° It is indeed a logical corollary of it. For a system which recognises only ‘relative ’ rights cannot admit an ‘ acquisitive ’ form of prescription which confers ‘ absolute * ownership.’ And the converse also is true: a system which thinks in terms of ‘ absolute’ ownership cannot admit a system of merely ‘ extinctive’ prescription. Hence Kaser argues that the transition from primitive to classical usucapio occurs when the character of the vindibetween the death of a deceased pefson and the acceptance of the inheritance by a heres __ extraneus) he could usucape it even though he knew well that he had no right to take it. The rule about theft did not prevent him, for it was a principle that there could be no theft of a thing forming part of a vacant inheritance, seeing that such a thing had no owner. 4 Discussed above, 146 n 1. The exact meaning of usus here is uncertain. As soon as the lawyers give it a precise sense it means ‘ possession ’, and the possession required for usucapion is, in general, the same as that required for the possessory interdicts (below, 259). But when the meaning of ‘ possession’ is established, the word itself gives way to possessio, and Yaron (TR 35, 1967, 209ff.) argues that it must therefore originally have meant something different, and conjectures that the primitive requirement was the actual use of the thing as opposed to the abstract possibility of control which was sufficient for

pOssessio. . 5 See further, above, 146 n 1; and for a clear restatement of his position see Deutsche Landesreferate zwm VI. Internationalen Kongress fiir Rechtsvergleichung, 1962, 19ff. Mayer-Maly argues (but see above, 146 n 1) that the XII Tables provision is con-

cerned only with the actio auctoritatis, but that the other surviving references are to usucapio, the classical institution being derived from both these sources. 6 Above, 142.

7 See Nicholas, Introduction 120f. Indeed, if the early usucapio was based on the failure of the person with the better right than the present possessor to assert that right, one would expect the relevant time to be that which had elapsed since he was in a position to make the assertion, even if no one person had in the meantime been in continuous possession. But Kaser’s interpretation of usus auctoritas ... bienniwm makes the criterion the possession of the acquirer (to use the later terminology) and not the dispossession of the loser. 154

Methods of acquiring ownership catio changes and ‘ absolute ’ ownership is recognised. However, if the validity of the sharp distinction between relative and absolute ownership may be doubted,® so also may that of the distinction here between the early and the mature usucapio. For the conception of classical usucapio as enabling the plaintiff in a vindicatio positively to prove ownership must in practice have been illusory. Good faith and the absence of theft are inherently unprovable, and usucapio must therefore in classical law have been not so much a method of acquiring ownership as a method of shifting the burden of proof.’ The

plaintiff who showed that he had had possession, and that it had been acquired ex iusta causa (and even this would often be difficult in the case of movables) could reasonably say that it was for the defendant to show that there had not been usucapion. The position of Kaser’s primitive litigant does not seem significantly different in this respect: once he had shown possession for the requisite period it was for the other party to show that the thing had been stolen ' or was otherwise incapable of being usucaped.? D. Tue Remarinine Civit Law MErTHopS oF ACQUISITION. These can be dealt

with shortly. First come the methods of acquisition from the state. It was the custom, no doubt from the earliest times, for booty taken in war to be sold

by the magistrates, usually the quaestors, at a public auction. At such auctions a spear was set up, perhaps as a symbol that the right acquired would be under the protection of the state forces,*® and the sale consequently

known as venditio sub hasta. The ownership in property bought in this way passed to the purchaser, apparently even before delivery ;* at any rate there was no need for any mancipation in the case of res mancipi (and slaves would often be amongst the things sold). This is indeed according to the general principle of Roman law that the state is not bound by the rules that apply to individuals. Not only movable booty but land which belonged to the state (whether by reason of capture from an enemy or for any other reason) might be sold or, as we have seen,°* assigned to individuals without payment, as 8 Above, 142; and see below, 266f.

8 See Nicholas, Introduction 124. An exception to this is the function of wsucapio in

curing the formal defect in the title of a recipient of a res mancipi by traditio. (De Visscher thought that this was the original function; SDHI 22 (1956) 98f., 23 (1957) 28; but see Yaron, TR 35 (1967) 193f.) 1 By the first party? Above, 153.

2 The XII Tables forbade the usucapion of other things besides res furtivae. These were (a) the five feet between estates belonging to different owners (Tab. Vi1I.1); (0) the forum sepulchri, explained as the entrance to a tomb, and the bustwm, the place where

a corpse was burnt (Tab. x.10); (ce) a res mancipi belonging to a woman in the tutela of her agnates, unless it has been alienated with auctoritas (Tab.v.2; Gai. 2.47) ; the rule was subsequently altered by a constitutio Rutiliana of uncertain date (Karlowa 2.404). 3 Below, 198 n 6.

4 At least this is so in classical times for fiscal property sold by an agent; D. 49.14.5.1. 5 Above, 10.

155

The law of property at the time of the XII Tables was frequently done on the foundation of a colony. Here too the property passed without any form such as would have been needed in private law. The only other method of acquisition which needs mention here is adiudi-

catio. In partition actions the iudex of classical times can not only, as in other actions, condemn one party to pay the other a sum of money, but can also assign sole ownership in the whole or part of the property concerned. If, for instance, A is co-owner of a piece of land with B and wishes to end the co-ownership, he can bring an action against B for division. The iudex can then either assign part to A and part to B or he can assign the whole to one

and condemn that one to pay the other compensation. In either case the judgment acts as a method of vesting ownership, dominium ex iure Quiritium if the trial is a zudicium legitimum,® otherwise only praetorian ownership

which will ripen into dominium ex i. Q. after the period for usucapion has elapsed. In classical law there were three actions in which the iudex had this peculiar power, the actzo familiae erciscundae for the division of property which was held jointly because it had been jointly inherited, the actio communi dividundo where it was held jointly for any other reason, and the actio fintum regundorum for regulating boundary disputes.” Of these actions the first * and the last ° certainly go back as far as the XII Tables.'

4, RESTRICTIONS ON OWNERSHIP Restrictions on the general right which an owner has of doing what he likes with his thing are of two kinds. They may be such as are imposed on all owners of a particular kind of property in the interest of the public generally or of the neighbours, or they may be merely imposed in a particular case by

reason of a right in the thing which happens to be vested in a particular person other than the owner. Restrictions of both sorts existed already at

the time of the XIT Tables. |

A. RESTRICTIONS IMPOSED ON ALL OWNERS OF A PARTICULAR SORT OF Pro-

PERTY.” In practice this means restrictions on owners of land. (i) As we have seen already, an owner may not cultivate or build up to the extreme limit of his property.’ (ii) He must allow the branches of his neighbour’s tree to over6 Te. tried within a mile of Rome before a single tudez, all parties being citizens; Gai. 4.104.

7 The nature of this action is disputed; see especially Buckland, RHD (1936) 741ff.;

Arangio-Ruiz, BIDR 32 (1922) 5ff. 8 Tab. v.10; D. 10.2.1 pr. ® Tab. v11.5. For the second, see below, 183.

1 It must be remembered that the ‘natural law’ methods of acquiring ownership are omitted here, not because they did not exist at the time of the XII Tables, but because there is no evidence of their regulation by law. Anything that could be said about them would be merely a statement of the rules as worked out by the later jurists.

2 Kaser, RPR 1.125f. 3 Above, 112 n 4. 156

ree ene) Restrictions on ownership

hang his land provided they are at least fifteen feet from the ground;

branches lower than fifteen feet he can insist on having cut off. Where the tree itself overhangs there is a similar right, or perhaps even the whole tree has to be cut down.‘ (iii) Where fruit from his neighbour’s tree has fallen on his land he must allow the neighbour to come upon the land to collect it.° (iv) He must not make such alterations on his land as will interfere with the natural flow of water from his neighbour’s land to his.® B. REstRictions BY REASON OF RIGHTS VESTED IN A PERSON OTHER THAN

THE Owner. In the developed law there were several kinds of these zura in re aliena, rights in property of which the ownership is with someone else,’ but

at the time of the XII Tables there existed only the most ancient of what came to be known as ‘ rustic praedial servitudes ’. A praedial servitude is a right vested in the owner of one piece of land (the ‘ dominant estate ’) to do something on a neighbouring piece of land (the ‘ servient estate ’) or to prevent the owner of that estate from doing something which he would otherwise be at liberty to do; the right must ‘ run with the land ’, that is to say it must be such that if the dominant estate passes into other hands, the right goes with it, and similarly, if the servient estate passes into other hands, the

liability goes with it. Praedial servitudes in the developed law are either ‘rustic ’ or ‘ urban’ according, generally, as they serve the needs of agricul-

tural land or of buildings, and thus we find that rights of way, rights of drawing water or taking it across the lands of another and rights of pasturage are classed as rustic, while the right of preventing a neighbour from increasing the height of his house or obstructing the access of light, and the right to support, are said to be urban.® As a class the urban servitudes are unquestionably later in origin than the rustic, because the need for them only arose 4 Tab. v11.9. In the later law these rights were sanctioned by interdicts (below, 230) ; the XII Tables apparently gave actions; D.43.27.1.8; 2.

§ Tab. v11.10. In the later law there was an interdict de glande legenda for the enforcement of this right. ‘Glans’ properly means ‘acorn’, but was used to include the fruit of all trees.

6 Tab. v11.8. The action was aquae pluviae arcendae. 7 Below, 268.

8 There is great difficulty in establishing the exact principle on which the distinction between urban and rustic servitudes was drawn. Some texts (e.g. J. 2.3; D. 8.2.2; 8.3.1) give lists of each kind of servitude from which it would appear that the same servitude was always either rustic or urban, but sometimes a servitude generally appearing as urban is classed as rustic (D. 8.3.2 pr.) and vice versa (6.2.11.1), reflecting perhaps the principle that the same servitude might be urban or rustic according to the nature of the dominant tenement, i.e. according as it was built upon or not; Buckland 262. Kaser has suggested (SZ 70 (1953) 144ff.) that the distinction originated relatively late and in the schools, and that the chief classical jurists adhered to the second view. The practical results would in most cases be the same, for a right of support (e.g.) cannot be wanted unless there is a building on the dominant tenement, and rights of way and water are usually only wanted in the country, not in the town where everyone has access as a rule to a, public way and a public water supply.

157

The law of property at the time of the XII Tables when building space in the city became valuable and the old-fashioned Roman house, built round a courtyard with its windows looking inwards, began to be superseded by structures more like our blocks of flats, and when,

in spite of the rule in the XII Tables providing for passage ways, the detached house was superseded by the ‘ semi-detached ’ or ‘ terrace’ house.°®

All rustic servitudes however are not equally old; the only one which is actually mentioned in the extant fragments of the XII Tables ' is that of via (the right to use a prepared roadway across another’s land *), though it is highly probable that the other rights of way (iter, right to walk or ride, and actus, right to drive cattle) and the right to take water across another’s land (aquaeductus) are equally old,® for they are all necessary to agriculture as soon as private property in land exists. With regard to via, the XII Tables laid down a fixed width for the roadway (eight feet, sixteen at the bends), and also provided that if the road were not kept in repair the person entitled ‘might drive his beast where he would ’,° i.e. go off the road. It must be remembered that only the rustic servitudes (and perhaps only the oldest of them) are res mancipi and can be created by mancipation; the reason for this is not difficult to understand if we consider that a primitive mind would see in a ‘ right of way ’ not an abstract limitation on an abstract right of ownership but a very concrete path, part of a concrete field and so transferable by the same methods as was the field. When servitudes of the type of * ancient lights? (ne luminibus officiatur) come to be recognised it is impossible to regard them as a part of the neighbour’s land, and the later servitudes are therefore not res mancipi. There is however no reason to suppose that in iure cessio could not be used as an alternative to mancipation for those servitudes which were res mancipi as well as being the only method applicable to the others. The form of an in wre cessio in the case of a servitude would be a

claim by the owner of the estate which was to become dominant that he already had the right and an admission by the owner of the intended servient estate that this was so.’ 9 The oldest urban servitude is very probably that of drainage (cloacae immittendae), which may have arisen when Rome was hastily rebuilt after its destruction by the Gauls in 390 B.C. If, as Livy says (5.55), it was built in such a haphazard way that the public drains, originally made under the public streets, passed under private houses, it may also have happened that the drains of one man’s house passed under that of another man and

that the right to keep them there had to be secured. 1 Tab. vii.7. 2 This is the usual explanation, but see Arangio-Ruiz, Istit. 236f.

3 The other important water servitude, aquae haustus, the right of drawing water,

4 Tab. v11.6. 5 Tab. v1I.7.

appears to be of later origin.

6 This is sometimes expressed by modern writers as a division of ownership in terms of function, each party owning the land for different purposes; Kaser, RPR 1.38, 143f.

7 No mention is made above of the ‘ personal’ servitudes of usufruct and usus, because, in spite of their great importance in the developed law, they certainly did not exist at the time of the XII Tables. 158

CHAPTER 11

The law of obligations at the time

of the XII Tables lL INTRODUCTION Justinian defines an obligation as ‘ a legal bond by which we are placed under

the necessity of performing some duty in accordance with the laws of our state’,! and this definition, though imperfect,” sufficiently describes the developed Roman and the modern conception of an obligation. But it must be realised that this apparently simple conception is by no means primitive. When we say to-day that A is under an obligation to B, we imply that, if A does not voluntarily perform his duty in accordance with that obligation,

B can bring an action? in the courts against him and, after getting judgment, have A forced by state authorities either to do what he should have done originally, or, if that is impossible or undesirable, to do something else instead, generally to pay him money. Thus if A is under an obligation to pay B £100 or to convey Blackacre to B, the state will, in the end, sell sufficient of A’s goods to bring in the necessary £100 or take away Blackacre from A and give it to B. If A’s obligation be to act as secretary to B the state will

not force him to act in that capacity, but it will make him pay money instead. The duty to do something for B is always coupled with the liability

to be forced to do that thing or to have the next best thing done at one’s expense. In the early stages of law, on the other hand, these two things, duty and liability, are by no means necessarily combined. At a time when the state has not yet undertaken generally the enforcement of duties between private persons, it by no means follows that because a duty is recognised, therefore it will be enforced, and the primitive creditor consequently asks for something tangible which will serve as security for the performance of the duty by the debtor; he wants to have in his power some person or some thing which will be liable; in other words, he wants a hostage or a pledge with whom or with which he can do what he likes if he does not get his rights. It must be noted especially that it is not a question (originally) of getting the hostage to 1 J. 3.13 pr. 2 Buckland, Main Institutions 235f. In particular, the word could denote not only the debtor’s duty but also the ereditor’s right. 3 There are, even in modern law, some cases in which an obligation, though recognised by law, is not enforceable by action; see e.g. Salmond, Jurisprudence, 12th ed. (1966) §43, and cf. the ‘ natural obligations’ of the later Roman law. 159

The law of obligations at the time of the XII Tables

him.

do what the debtor ought to do, or of selling the pledge and paying oneself out of the proceeds; it is a question of having the hostage or pledge in one’s power, and, in the case of the hostage, of being able to wreak vengeance on A further difference between developed and primitive law is that primitive systems have not yet made the distinction, common to advanced systems, between delict and contract as sources of obligation. In the developed Roman law and in modern systems derived from it we can say that if A is under an

obligation to B it is usually (by no means always) either because A has agreed to perform the act which forms the content of the obligation (contract), or because he has committed some wrong (delict) against B for which he must pay B damages. The distinction itself is evidently not primitive, but there has been some debate as to whether obligation originated in what is later called contract or in delict. It can be argued * that what comes first is the undifferentiated idea of a wrong, and that it is only very gradually that the particular sort of wrong which consists in not doing what one has agreed to do is separated from other wrongs. On the other hand it can be said * that the primitive consequence of a wrong is vengeance, and that this does not

entail the idea of binding with a view to compelling satisfaction which is essential to obligation. For obligation is a means to securing satisfaction, whereas vengeance is itself the satisfaction sought. The idea of obligation emerges therefore only when the stage is reached at which the person wronged may agree to forgo his vengeance if the wrongdoer will pay him composition, and it is precisely in these agreements for composition that we find the begin-

nings of contract. Later still the optional composition is replaced by a fixed

compulsory composition, i.e. by a fixed penalty, and finally by variable damages, but meanwhile contract has developed on its own.

It is also in the agreement for composition that we find the clearest example of the distinction between duty and liability. Suppose that A has committed a delict against B—has stolen his property—B is justified in avenging himself on A, and if he is strong enough, will capture A for the purpose. A may then promise to pay composition and B may agree to accept it, but it may easily happen that A will be unable to collect the money so long as he remains in custody, whereas B, having once got hold of him, does not want to let him go. A therefore gets a child or kinsman to be hostage for him in the meanwhile. The result of this is that although A owes the money, 4 Perozzi, Scritti 2.443ff.; Bonfante, Storia 1.208ff.

5 B. Betti, La struttura dell’obbligazione romana, 2nd ed. (Milan, 1955) 109ff., who also emphasises the early importance of the contract of guarantee. De Visscher, Hiudes 257ff., argues that delict was not brought within the concept of the classical obligatio until the early second century A.D. It is certainly striking that in Gaius’ Institutes delict

appears as an appendage to a treatment of obligation in terms of contract; cf. de Zulueta, Gaius 141f. 160

Introduction he is no longer liable, whereas the hostage, who does not owe the money, is liable, because liability is something quite physical; it means being in the

power of the creditor. Then, in some systems, at any rate, comes a stage when A can become hostage for himself, or, rather, as one can now say, surety. That is to say B lets him go on his undertaking to be liable, which means that B shall be allowed to take him again if he does not pay. The physical binding is replaced, for the time being, by a metaphorical bond, which may in certain circumstances be converted into a real one again. The distinction here drawn between duty and liability has been studied by modern scholars especially in their investigations into ancient Germanic law,

and it 1s in Germanic law only that its existence can be proved to demonstration.° None the less it is now widely held that somewhat similar ideas formed the foundation of the Roman law,’ and the chief evidence for this view comes from the Roman terminology itself. Obligatio means literally a * binding ’, and there can be little doubt that originally it was no mere metaphor; the same idea is to be seen in the word nexum, which, as we shall see,*® refers to the earliest known contract. More significant still is the word solutio, which in the developed law means payment, or rather fulfilment of an obliga-

tion. Literally it means ‘ loosening ’, and that this was originally no mere metaphorical loosening of a metaphorical ‘ bond’ appears clearly enough in the formula of the solemn payment per aes et libram, where the debtor who is to be discharged says ‘I free and release myself from you ’— me a te solvo liberoque.°®

Now, at the time of the XII Tables the law was no longer in its most primitive condition. The state had already restricted self-help very consider-

ably, but the time when it would take upon itself the task of seeing that people carried out their duties to each other was still in the far distant future; the most it could do was to say when one man was liable to another, to decide in some, indeed in most, cases how that liability could be redeemed, and to provide methods for preventing unauthorised enforcement of liability. Thus, so far as what is later called delict is concerned, it already provides generally that a certain penalty is to be paid by the wrongdoer to the person wronged, in other words, that the person wronged must forgo his right to vengeance if he receives a certain sum by way of composition, but in other cases the right to take vengeance (though in a restricted form) is expressly preserved, and it is for the person wronged to consider whether he will accept composition instead, and, if so, at what rate. Contract is more developed § Brunner, in Holtzendorfi’s Enzyklopddie der Rechtswissenschaft, 7th ed. by J. Kohler, 1 (Munich etc., 1915) 137. For similar ideas cf. Pollock and Maitland 2.185ff. The words

‘duty’ and ‘liability’ are here used as (not very exact) equivalents of the German words ‘Schuld’ and ‘ Haftung ’. For Greek law see J. Partsch, Griechisches Biirgschafts-

recht (Leipzig, 7 Kaser, AJ 189ff. 8 Below, 164.1909) 9 Gal.13ff. 3.174. 1 Below, 172. 161

The law of obligations at the time of the XII Tables than might perhaps be expected. The agreement to pay and accept composition for delict is of course recognised, and so are forms of suretyship: of two old types of surety, vas and praes, the former was certainly mentioned in the

XII Tables.? But since the discovery of the new fragments of Gaius* we know that stipulation also goes back to this period,‘ i.e. that an oral promise preceded by a question could create an actionable obligation. It is probable, however, that stipulations were as yet confined to promises to pay a definite sum of money (as in the example Gaius gives), and the survival of nexum,*®

a transaction by which the debtor was ‘ bound’ to the creditor, seems to show that the mere ‘ obligation ’ (in the later sense) produced by a stipulation was by no means always considered sufficient. This bare statement is, moreover, misleading. How, it may be asked, could

men do without at least sale and hire, loan of chattels and deposit? The answer is twofold. First one must distinguish the question of the existence of the economic relationship covered by each of these contracts of the later law

from the question whether there arose as yet any liability from promises made or implied in such relationships, and secondly one must realise that many of the results achieved later by the idea of contract can be, and were, achieved by that of delict. Thus it is quite clear that the economic relationship of sale or barter existed, that people in fact exchanged goods for the primitive equivalent of money (especially uncoined metal), or for other goods, but this proceeding did. not result in any obligations and was not a contract because the two parties almost always carried out their respective parts of the bargain simultaneously. Thus, from the economic point of view the primitive mancipation, before the weighing out of the bronze became purely symbolical, was, as it is ordinarily understood, clearly a sale — the exchange of a thing for money of a sort — but it was no sale in the sense of the later contract, for in that sense a sale means an agreement to transfer a thing in return for a money payment, and when the agreement is once reached the

seller is under an obligation to do what he has promised to do, as the buyer on the other hand is under an obligation to pay the price that he has promised. In the mancipation there was no scope for these obligations because the transfer of the thing and the weighing out of the bronze were both parts

of the same transaction; the property in the thing passed only when the bronze was given, and the bronze was only given when the property in the

thing passed. Once the mancipation is complete there is nothing left for either party to do, and before it is complete no legal effect is produced.°

2 Tab. 1.10. Below, 184. 3 Below, 389.

4 Below, 182. 5 Below, 164.

¢ If after mancipation the buyer is evicted, the seller is, it is true, under an obligation to pay double the price (above, 146), but the double liability appears to show that this is delictal, a punishment for ‘ stealing ’ the price. 162

Introduction When the weighing out became symbolical it did become possible to manci-

pate property without receiving an equivalent at once; in other words it became possible to give credit, but there is no reason to think that, at the time of the XII Tables, a seller who had thus given credit had an action for the price. It may well be that he could secure himself by making a nexum whereby the price was treated as a loan from him to the buyer, though there is no direct evidence of this practice. Further, if there was a rule that property did not pass until the price was paid, the seller was protected by his ability to vindicate the property.’ The converse case — that the buyer should pay the price before the transfer of the thing — would presumably be very

rare indeed. If a mancipation had taken place, but the thing had remained in the possession of the seller, it would none the less belong to the buyer, who would thus be protected by his power to vindicate. Where it was a res nec mancipi that was sold we cannot say what rules there were, but there can be no doubt that such sales were almost invariably for ready money and that there was no general principle that an unpaid seller was entitled to recover the price or a buyer who had paid to sue for the thing.*®

Of the relationships covered by the later contract of ‘ hire’ (locatio conductio), which included the renting of land and chattels, contracts of service and contracts to do jobs, we know even less. That they all existed cannot be

doubted, though none of them can have been as common as they subsequently became. The renting of land and houses was necessarily rare in a community which consisted chiefly of peasant-proprietors; the occasions for

the hiring of movables would be few; the trades which undertake ‘ jobs ’

were hardly developed, and, though there were, no doubt, some free labourers, the work of tilling the soil must have been done mainly by the owners, their families and their slaves. There was certainly no general principle that money promised in payment for the use of a thing or services was recoverable,® but the law of property and the law of delict would prevent some kinds of injustice. The man who refused to return a thing he had hired would have to give it up on vindication by the owner and could very possibly be treated as a thief;! and the same reasoning applies to the case of the man who had received a thing in order to do some work on it —a piece of cloth, for instance, to make up into a garment. 7 Above, 145 n 2.

8 That the seller who did not insist on immediate payment had no action as a rule to enforce it afterwards is best shown by the particular case mentioned in Gai. 4.28. The XII Tables, he says, allowed pignoris capio (one of the forms of action) against a man who had bought an animal for sacrificial purposes and did not pay for it. Had there been any general rule allowing an action for the price such special provision would not have been necessary. 9 Again we have one special case where pignoris capio is allowed; Gai. 4.28. 1 Cf. above, 141, on the question whether the early vindicatio was available for res nec mancipt.

163

The law of obligations at the time of the XII Tables So too with the relations later covered by the ‘ real ’ contracts of loan (of

chattels) and deposit. The borrower and the depositee who attempted to retain things which did not belong to them would be in the same position, and in the case of deposit there is the definite evidence that the XII Tables allowed an action in duplum—an action that is for a penalty exactly the same as that for theft.? That some form of pledge too had at least a de facto existence we cannot doubt, but the early law on this subject is wrapped in mystery.*

2 NEXUM* In the accounts of early Roman history considerable prominence is given to the sufferings of certain people called newi (literally ‘ bound ’), who, it is

clear, were debtors (generally plebeians) who, being unable to pay their debts, became the bondsmen of their creditors, liable to work for them, to be chastised by them and in a very literal sense ‘ bound ’. It is further clear that the transaction on which the position of these nevi rested was, like mancipation, one ‘ by the scales and bronze ’. The possibility of this bondage resulting from debt came to an end by the passing of a certain lex Poetelia, prob-

ably of 826 B.C.° This is about all that can be said with certainty, for naturally enough no classical legal writer deals with an institution which was obsolete some four centuries before the classical period, and the historians are not concerned to give an exact legal account of the position which gave rise to the sufferings they describe. The best evidence we have consists of a few passages from the lexicographers (the chief one of which is however badly

mutilated) ® and a passage of Gaius which describes, not nexum, but the method of releasing a debt contracted ‘ by the scales and bronze ’.’ The result is that scholars differ very considerably in their interpretation of the

2 Cf. below 172 n 2. 8 Cf. below, 301ff.

4 The literature is immense. See de Zulueta, LQR 29 (1913) 137ff.; Kaser, RPR 1.166; also Tomulescu, Iwra 17 (1966) 39ff. (with further reff.) ; MacCormack, SZ 84 (1967)

850ff.; Ogilvie, Livy 1-5 296ff. 5 See below, 189.

6 Varro, L.L. 7.105 (Bruns 2.60): Nexwm Manilius scribit omne quod per libram et aes geritur, in quo sint mancipia; Mucius quae per aes et libram fiant ut obligentur, praeter quae mancipio dentur. Hoc verius esse ipsum verbum ostendit, de quo quaerit;

nam id est, quod obligatur per libram ‘neque suum’ fit, inde ‘nexwm’ dictum. Liber, qui suas operas in servitutem pro pecunia quam [debebat] dum solveret, nexus vocatur, ut ab ‘aere’ ‘obaeratus’. Hoc C. Poetelio Libone Visolo dictatore sublatwm ne fieret; et omnes qui bonam copiam iurarunt, ne essent next, dissolutt. For textual questions see Tomulescu, Iura 17 (1966) 49 n 16. On bonam copiam turare see Berger, St. Arangio-Rwiz 2.117ff. The passage is particularly important as it cites the opinions of two lawyers, Manilius and Mucius (above, 92), but even they lived a long time after the lex Poetelia. If the statement in Varro’s text that the passing of the law was due to the dictator Poetelius is right, then its date would be 313 B.C., but Livy (8.28) refers it to the consuls of 326, one of whom was a Poetelius. 7 3.17348.

164

Nexum evidence, and the opinions which have been held as to the nature of nexum are innumerable. The theory which was dominant in the second half of the nineteenth cen-

tury, and of which there have been many subsequent variants, is that of Huschke.* According to his view nexwum was a form of loan, contracted with the ceremonial of the scales and bronze, the libripens and the five witnesses,

and thus was the counterpart, in the law of obligations, to mancipation in the law of property. The lender originally weighed out the bronze to the borrower; later, by a development analogous to that which took place for mancipation,® the weighing became symbolical, and the actual payment of the money lent was separated from the ceremony. The XII Tables confirmed this transaction in the same way (on this view) as mancipation itself, by the words cum nexum faciet mancipiumque, etc.' So far the transaction is precisely similar to mancipation, but the words spoken in the case of a loan necessarily differed from those used in a conveyance, and Huschke conjectured that they were spoken by the lender (as the person acquiring rights under the transaction), and consisted of a form in which he declared that the borrower was damnas (condemned) to repay him the money, the importance of this word lying in the fact that if the debtor did not repay at the due date, the obligation could be enforced directly against his person, without first

getting judgment, by the process known as manus iniectio.? The debtor might not of course proceed to these extremities at once but might keep the debtor in dependence on him by threatening to exercise his rights. The first to propose a radically different explanation was Mitteis.* He argued that Huschke’s theory was inconsistent with the picture of nexum

given by the historians, from which it appears that the entering into the state of bondage was not the result of the loan itself, but was the last desperate act of insolvent borrowers when faced with an action which would inevitably lead to judgment, manus iniectio, and either death or slavery.* Mitteis

conjectured that there were two transactions, both called nexum. The first was the making of the loan, which would have been enforceable by legis actio per sacramentum, and the second a mancipation of himself by the

8 Ueber das Recht des Nexwm (Leipzig, 1846). ® Above, 144. 1 But see above, 145 n 3. We lack direct evidence that they come from the XII Tables, but the way in which Festus quotes them (s.v. Nuncupata pecunia, Bruns 2.18) leaves little doubt about the matter. If they do come from the XII Tables, then they prove that nexzum existed at that time, but in any case the institution is so primitive that there ean be no question as to its existence at the time of the codification. 2 Below, 188; de Zulueta, Gaius 245f. Against attributing this consequence to the word damnas, Kaser, AJ 124ff. Kaser himself conjectures that the creditor said guod tu

mihi nexus es... 3 SZ 22 (1901) 96ff.; 25 (1904) 282; RPR 136f.

4 E.g. Livy 7.19.5; cf. MacCormack, op. cit. Kaser, AJ 232ff., supposes that nexum did

indeed involve a loan, but that in the cases referred to by the historians the loan was

made to enable the debtor to pay off the existing debts. 165

The law of obligations at the time of the XII Tables debtor to the creditor, whose bondsman he thus became by a sort of self-sale — or rather self-pledge, for he would be released if the debt were subsequently paid. This self-mancipation was made in order to escape, by voluntary entry

into bondage, the execution which would otherwise have followed on judgment. This theory explains the constant use, especially by Cicero, of nexum as the equivalent of mancipation; it explains Varro’s statement that a nexus was a ‘free man who gave his services into slavery, for money which he owed until he should pay it ’,° and it is in accordance with developments in other systems in which self-sale or self-pledge is frequently found, but the difficulties are at least as great as those which stand in the way of Huschke’s conception. If we assume that bronze was weighed out on the first newum, there would be none left to weigh out on the second, the self-mancipation,*® and further, however common self-sale may be in other systems, it is otherwise unknown in Roman law and appears to be contrary to principle. In fact, though much of the argument on which Huschke based his opinion has been shown to be erroneous, his hypothesis is still, in the main, probably the most acceptable. Newum, so we can still believe, was a transaction carried out by

the scales and bronze whereby the debtor made himself liable to manus intectio without judgment if he failed to pay on the appointed date. It had thus a ‘ real’ element because it gave the lender a right (eventually) against the body of the borrower,’ but that is not the same thing as saying that it

was a form of self-sale. It gave rise to manus iniectio because that was originally the only form of liability known; it was in fact a form of legalised self-help. Only subsequently, when other legis actiones necessitating a trial were introduced, did the treatment of debtors so liable come to be regarded as particularly harsh. Indeed as a positive argument in favour of this view there still remains the passage in Gaius ® where we are told that a judgment

debt (which certainly did give rise to manus iniectio) and one contracted per aes et libram both had to be released by the form per aes et libram. If nexum had been a self-sale the release would surely have involved a re-sale or a manumissio vindicta, since the debtor would have been in mancipio to the creditor.* !

5 Above, 164 n 6. 6 Cf. Mommsen, SZ 23 (1902) 350. 7 It was thus a transaction directed towards the creation of ‘ liability’, in the sense explained above, 159.

8 3.173-5; cf. de Zulueta, Gaius 144. § Gai. 1.138. 1 Livy 2.27 causes a difficulty. The historian there pictures one of the consuls as com-

pellmg debtors to enter the condition of nexi. MacCormack, op. cit., suggests that the absence of any reference to such a proceeding after the XII Tables is attributable to its having been superseded by manus iniectio iudicati (Gai. 4.21). Tomulescu, op. cit., takes the text as evidence that nexum was not an act per aes et libram at all, but a form of in tiwre cessio, the reference in Livy being to the addictio. It is true that there is no clear evidence of the existence of a transaction of loan per aes et libram, but Tomulescu’s thesis involves the assumption that by the time of Manilius (above, 92) 166

The law of obligations at the time of the XII Tables

38 DELICT Of the four * chief delicts of the developed law two were already recognised

under the names which still attached to them later, furtum (theft) and iniuria (later including all offences against a person’s dignity and a good deal

more, but at this early date going, in all probability, no further than actual assault).

A. Furtum.’ A distinction was drawn between furtum manifestum,* when the thief was caught in the act, and furtum nec manifestum, when he was not. In the former case the thief, if a free man, was scourged and ‘ adjudged ’

{addictus) to the person from whom he had stolen. It was doubtful, says Gaius,° whether he thereby became a slave or was in the position of an adtu-

dicatus (i.e. one who was seized by his creditor after judgment by manus iniectio).° If the thief was a slave he was scourged and then thrown from the Tarpeian rock. In the case of furtum nec manifestum the penalty was already

pecuniary, double the value of the thing stolen having to be paid.’ It was further provided that a man on whose premises a stolen thing was found after a search with witnesses should be liable, in the actio furti concepti, to a threefold penalty.* Gaius also says that if the searcher went naked except nezum had aequired another and quite different meaning: omne quod per aes et libram geritur.

2 Furtum, rapina (robbery), damnum iniuria datum (unlawful damage to property) and inturia. 3 Tab. vi11.12-16; Gai. 3.183-208. On all that follows see Jolowicz, De Furtis; de Zulueta, Gaius 198ff.

* In later times there were differences of opinion as to the exact definition of furtum mantfestum; Gai. 3.184; J. 4.1.3. On the origin of the distinction see Jolowicz, De Furtis LXVIII-LXXIV.

* 3.189. 6 And so eventually liable to death or slavery abroad; below, 189. 7 Gai. 3.190. Tab. v111.16 (St adorat furto quod nec mantfestum erit —the rest of the sentence is not preserved) is usually taken to have laid this down, adorat referring to the accusation of the plaintiff in the actio furti nec manifesti. But the XII Tables elsewhere go straight to the wrongful act (e.g. st membrum rupsit, Tab. VIL1.2) without wasting words on the need for accusation, and Yaron (7'R 34 (1966) 510ff.) suggests, with support from other primitive systems, that the provision related to the penalty for a false accusation of theft. § Gai. 3.186, 191. A man was liable to this action even if he did not know that the

thing was on his premises or that it was stolen, but the injustice of this could be remedied in a rough and ready way by the actio furti oblati, also for threefold value, which the householder could bring against anyone who had put the stolen thing in his house. Gaius (3.187) says that the defendant is lable only if he acted with the intention of passing off the goods, but this qualification is probably later than the XII Tables (Daube, Biblical Law 262, 269). Kaser indeed thinks (BIDR 65 (1962) 84) that Gaius did not intend to exclude the liability of the innocent defendant but was merely giving the typical case, but this is difficult to reconcile with the words he uses: utique si ea mente tibi data fuerit ut... 167

JT

The law of obligations at the time of the XII Tables for a lictum, and holding a lanz * in his hand, then the offence was treated as furtum manifestum. In Gaius’ time this search had long been obsolete * and was evidently no longer understood. There has in consequence been much modern discussion.” Ritual searches are widespread in primitive societies,

especially those of Indo-Germanic origin,? and nakedness is a common feature, its purpose being probably to prevent the searcher from ‘ planting ’ stolen property in the house. But the significance of the lanx and liciwm, for which there are no close parallels, is obscure. Gaius has rationalistic explanations (the licitum is a loincloth, for modesty’s sake, and the lanza is a platter, either to put the stolen thing on, if found, or to keep the searcher’s hands occupied and thus prevent him from smuggling anything in) but their significance is more likely to be magic or religious. Licitwm elsewhere means a thread, and it may here indicate the wearing of an amulet or a ritual headband, and the lanz may be a dish to carry an offering to placate the gods of the house which is searched.‘ There is equal difficulty in determining the relationship between the search

lance et licio and the informal search with witnesses. That they should both have existed side by side seems to be contrary to the ‘ economy of means ’ which is usually attributed to early law, and yet the simple explanation ° that the formal search alone existed at the time of the XII Tables and that the actio furtt concepti was a later praetorian substitute for the old capital penalty, is hardly acceptable. For it involves denying Gaius’ plain statement

_ that the actio furti concepti went back to the XII Tables, and Gaius, who wrote a commentary on the XII Tables and would also have known the formula (which would show whether the action was civil or praétorian), is not. likely to have made a mistake about so straightforward a matter. The explanation which best tallies with all that Gaius says is that of Daube.* For

him both searches did exist at the time of the XII Tables, but the search lance et licto was invoked only if the accused declined to allow the informal search. This is the natural meaning of Gaius’ statement 7 that the XII Tables

provided no penalty for refusal of search but simply laid down that the searcher should proceed lance et licto. Gaius, from his rationalistic point of view, finds this ridiculous, since a man who has refused the one will also refuse the other, the more so as it involves a higher penalty, but in primitive times a refusal of the ritual search would no doubt have entailed a religious 9 Gai. 3.192-3a. For the meaning of these words see immediately below. 1 Gellius (16.10.8) includes it among the antiquities which disappeared as a result of the lez Aebutia (below, 218). 2 For a comprehensive summary and criticism see Horak, PW 24.1.788ff. (1963). 3 C, von Schwerin, Die Formen der Haussuchung in indogermanischen Rechten (Leipzig, 1924) ; review by Goldmann, SZ (German. Abt.) 45 (1925) 457ff. 4 Horak, PW 24.1.793ff. Wolf, Sympotica F. Wieacker (Gottingen, 1970) 59ff.

5 Mommsen, Sirk 748. 6 Biblical Law 259ff. 7 3.192f. 168

Delict sanction.*® In the classical law, and long before, only the informal search leading to the actio furti concepti survived, and refusal of this search now

led to an actio furti prohibiti, introduced by the praetor, for a fourfold penalty.

If the thief came by night or if in daytime he defended himself with a weapon the XII Tables allowed him to be killed without a trial, provided that, at least in the latter case, the killer shouted aloud.° It is likely, as the etymology suggests,’ that furtum implied in early times an actual taking away of the thing. At any rate it seems clear that originally

theft was possible only in the case of things that were capable of being removed, and not therefore in the case of land.? This was indeed still the 8 Daube, Biblical Law 284ff., seeks to meet the argument from ‘economy of means’ by supposing that in a period before the XII Tables the ritual search alone existed and that it took place only as the culmination of a hot pursuit (whence the assimilation of finding after such a search to furtwm manifestum). The informal search would have been first invoked when there was no hot pursuit. De Visscher (Htudes 1.217ff.=TR 6 (1925) 249ff.) also invokes the presence or absence of hot pursuit. He supposes that at the time of the XII Tables there was only one search (lance et licio) but that the penalty of furtum manifestum applied only if there had been hot pursuit or an express denial of possession; in other cases there would have been only the actio furti conceptt. But see Daube, Biblical Law 278ff. ® Tab. vr11.12, 138. The requirement of shouting (endoplorare) is mentioned by Cicero

only for the latter case (Tull. 50), and though Gaius mentions it for both cases (D. 9.2. 4.1) the text is with reason suspect. Nevertheless, since endoplorare seems originally to

have denoted a formal call to neighbours to attest the guilt of the culprit before the injured party resorted to self-help, it probably was in early law required in both cases. In later times the justification of the killing was seen as self-defence (Cicero, loc. cit.) and the shout was therefore probably taken as evidence that the killer was prepared to justify his act. In such a context the absence of a shout by night could be excused. See generally Wieacker, F'schr. Wenger 1.129ff. Kaser, HB 38ff., conjectures that endoplorare

was required also before the killing of a manifest thief or one treated as such after search lance et licio. 1 Niederlander, SZ 67 (1950) 185ff. The word fur (Greek gop) is connected with ferre and so means ‘one who carries away’. This derivation is already given by some ancient writers, e.g. J. 4.1.2, where it is mentioned as an alternative to suggestions that the word is connected with furvwm (black) or with fraus. These, though etymologically fanciful, no

doubt reflect the wider idea of theft that prevailed in classical times; see immediately below.

2 The suggestion that ‘ carrying away’ formed part of the original conception of theft may seem at variance with the opinion expressed above (163) that the bailee who refused

to return the thing bailed to him was treated as a thief. Both views may however be correct. Primitive law has ‘ objective’ standards, i.e. it lets certain consequences follow from certain outward facts because it regards those facts as proof of the other facts with

which it seeks to deal. If a man refuses to give up a thing lent to him it is generally because he has taken the thing away or at least hidden it with the intention of keeping it to himself; that is enough to make the law say that everyone who acts similarly is a thief. In any case it is not suggested that there was any statutory rule needing strict construction. The fact that the XII Tables dealt with the cases of deposit and of the dishonest tutor (below, 172) by punishing them in the same way as they punished thieves, ean be used as an argument either way, for, as Mommsen says (Str& 738 n 2), it is 169

The law of obligations at the time of the XIT Tables snmnseseeiesenreeineeegineanan ene a EAA NANT LTLTTLTTeaaL classical law, but the exclusion of land had been disputed by some of the earlier jurists,* and with reason, since the scope of theft was early widened to include cases where there was no taking away. Thus in the second century B.C. the jurist Brutus held that a man was a thief if he borrowed a horse for

one purpose and used it for another;* and it was evidently theft to sell a thing belonging to someone else, knowing that one had no right to do so, even though one had come into possession of it quite innocently.° The act which constitutes theft is accordingly in the classical law commonly described as a contrectatio, meaning literally a touching or handling.°

B. DaMaGE To PROPERTY. The later law on this subject is based so com-

pletely on a statute (the lex Aquilia, of doubtful date, but certainly later than the XII Tables) that it is impossible to discover what were the earlier provisions which were superseded.’ There are a number of fragments of the

XII Tables which mention particular cases of damage to immovable property, but none concerning movables, and there is no reliable evidence of any general provision. The offences which are dealt with are those which would

specially concern an agricultural population; the nocturnal cutting of another’s crops or letting animals graze on them was punished capitally,° unless the offender was under the age of puberty, in which case he was flogged or had to make good the damage twice over.’ The person who set fire to a dwelling-house, or a heap of corn near such a house, was burnt if he had done so intentionally; if by accident, he had to make good the damage and,

if too poor, was ‘ lightly chastised ’.' For the unlawful cutting down of another’s trees a penalty of 25 asses for each tree was laid down,” and there uncertain whether these cases are to be regarded as particular cases of theft, or as analogues of theft (cf. Watson, Obligations 157f.). : 3 Gai. 2.51; D. 41.3.38; Gell. 11.18.13. 4 Gell. 6.15.1; ef. Gai. 3.196. Of course if he thinks that the owner will not mind he is not liable; Gai. 3.197. 5 Cf. Gai. 2.50. 6 A strict interpretation of this requirement of physical contact would result in anomalies (e.g. where a man sold a thing belonging to someone else but only the buyer touched it), and certain texts can be read as admitting theft without handling. Buckland accordingly argued that contrectatio came to mean meddling (LQE 57 (1941) 467ff.). Against this, inconclusively, and with reff. to further controversy, Watson, Obligations 220ff.

7 Cf. below, 275. Ulpian says (D. 9.2.1) that the lex Aquilia ‘derogated from’ previous rules on the subject contained in the XII Tables and other statutes. For discussion see Lawson, Negligence 4ff. 8 Tab. vi11.9. The punishment (crucifixion as a sacrifice to Ceres, the goddess of corn) reveals the primitive affinities of this offence with sacrilege; Perrin, Annales Universitatis Saravienstis 2 (1953) 173ff. 9 The words are obscure; possibly, as Mommsen suggests (see Bruns 1.31), the flogging

was inflicted on an impubes sui turis, while in the case of one in potestate the father had either to pay double damages or surrender the boy (noxae deditio).

1 Tab. vt1r,10. 2 Tab. vitt.11; cf. below, 174. 170

Delict were also provisions, details of which are lost, against the use of magical incantations to destroy or get for oneself another’s crops.* Damage done by quadrupeds (pauperies) was specially regulated, the rule

being that the owner had either to make compensation (nowiam sarcire) or give up the animal,* a rule which, as we shall see,’ also applied in the case of delicts by slaves and children under power.°

C. Inruria.’ The XII Tables punished the ‘ breaking of a limb’ (membrum ruptum) * by talion, that is, by allowing the infliction of a similar injury on the wrongdoer, but only ‘if no agreement for composition be made’. The fracture of a bone (os fractum) was dealt with by a fixed pecuniary penalty of 800 asses where the victim was a free man, 150 asses where he was a slave.

‘For all other iniuriae ’, says Gaius, ‘ the penalty laid down was 25 asses.’ In all probability this refers merely to blows such as do not result in serious injury. It is unlikely that the Romans of the fifth century B.C. were very susceptible to insult, and, if we imagine that iniuria can here refer to the innumerable different kinds of attack on a man’s personality which it covered in the later law, it is difficult to explain how they could all be punished by the same fine. The XII Tables did indeed punish the singing of defamatory

sorcery.” * . 3 Tab. VIIrs. + Tab. vi0i.6. * Below, 173. verses, and that capitally, but the evil here is almost certainly that of

6 See generally Daube, CLJ 7 (1939) 23ff.; Kaser, AJ 219ff. * Gai, 3.225ff.

8 This can hardly be confined to the severance of a limb, since a serious wound which involved neither such amputation nor the breaking of a bone (os fractum) would entail the same penalty (25 asses) as a minor blow. Binding, SZ 40 (1919) 106ff., supposed that a clause dealing with such wounds had been forgotten, but this is unlikely. It is more commonly assumed that membrum ruptum meant any permanent loss or disablement of some part of the body, including for example an eye. To the objection that such injurv might be less severe than os fractum and yet would still carry the grave penalty of talion, it can be said (Appleton, Mél. Cornil 1.51ff.; G. Pugliese, Studi sull’ inturia, Milan, 1941, 29ff.) that the XIT Tables allowed talion in these cases precisely because it was impossible to fix a tariff which would meet the widely varying degrees of seriousness, whereas the fracture of a bone was a relatively easily ascertainable and stable injury for which a price could be laid down (Birks, TR 37, 1969, 163ff., goes much further). It must, moreover, be remembered that composition was no doubt usual and may even have been indirectly enforced by the magistrate’s refusing to grant addictio of « defendant who made a reasonable offer of composition (Simon, SZ 82 (1965) 166). Alternatively os fractum may have been confined to such fractures as produced no permanent disablement (di Paola, Ann. Cat. 1 (1947) 268ff.). De Zulueta, Gatus 217, suggests that the distinetion lay in the weapon used: ‘membrum ruptum was wounding with a cutting weapon — a sword, dagger, or knife — which would be an act of private war deserving of talio.’ For a much more radical and conjectural suggestion see Birks, op. cit. 9 Tab. VIII.1; Ronconi, Synteleia Arangio-Ruiz 958ff. Pliny and Cicero give different versions and different interpretations. Pliny, H.N. 28.4.18, quotes qui malwm carmen incantasstt, which he regards as directed against sorcery, while Cicero, in a fragment of the de republica (4.12): preserved by St. Augustine (de civ. det 2.9) has si quis occent171

nt mn een) The law of obligations at the time of the XII Tables

D. OTHER DELICTS AND THE GENERAL CHARACTER OF THE LAW OF DELICTS. In

addition to these main rules for the chief offences against the person and against property, the fragments of the XII Tables contain a number of scattered provisions for penalties, some of which would not, in a developed system, come under the heading of delict at all, but must be so classed here precisely because the cases they concern are dealt with by the imposition of

penalties, that is, are treated as delicts. Thus if a depositee fails to return the thing deposited with him he is liable, like the thief, to an action for ‘ double ’, and not, as in the later law, regarded as having merely broken his contract.” Similarly the tutor who embezzles the property of his ward has to pay ‘ double ’,*? whereas in later law, though this action survived,* the rela-

tions between a tutor and his ward were regulated generally by an action (actio tutelae) which has no delictal character and is placed by Justinian ° under the heading of quasi-contract.°®

The picture given by this description of almost all that has survived of the decemviral legislation with respect to delict is an interesting one, for it shows

that Roman law was at the time of the XII Tables in a state of transition from voluntary to compulsory composition for private wrong. In nearly all cases, so far as we can see, the law already insists that the party wronged shall accept a pecuniary penalty in lieu of exercising his primitive right of vengeance, and fixes an amount at which the penalty is to be assessed, but in the cases of membrum ruptum and manifest theft ’ it is still open to him to insist on vengeance of a sort, though this too is now regulated. But if the simple system of vengeance or composition at the option of the party wronged has almost disappeared already at the time of the XII Tables, the law of delict not only then, but long afterwards, retained many traces of

its origin. Throughout the history of Roman law it remains true that a delictal action dies with the wrongdoer, and the reason is no doubt that ven-

geance was only permissible against the wrongdoer, not against his heir. Originally it was also probably true that the action died with the victim of avisset sive carmen condidisset, and this is expressed in terms of defamation: quod infamiam faceret flagitiumve aiteri. It has been argued that there were indeed two separate provisions, but it is improbable that the XII Tables concerned themselves with defamation, in the modern sense, or, if they did, that they punished it capitally. It is more likely that the two versions reflect a single original, Pliny’s giving the primitive meaning and Cicero’s a rationalisation; cf. Ronconi, op. cit. 962f.; Wieacker, RIDA

(1956) 462f., 466. Pliny, loc. cit. (=Tab. vi11.8), also preserves part of a provision against ‘singing away’ crops (qui fruges excantassit). 1 In the developed law only intentional acts could count as iniuriae, but it is unlikely that at the time of the XII Tables the law considered the state of mind of the individual

defendant; see below, 173f. 2 Tab. vi1I1.19; ef. above, 163, 169 n 2.

3 Tab. vr11.20. 4 As actio rationibus distrahendis. 5 J. 3.27.2.

6 Cf. also the actiones auctoritatis and de modo agri, above, 146f. (obsolete for Justinian). 7 Here too the party wronged might accept compensation if he liked; D. 2.14.7.14: Nam et de furto pacisct lex permittit, where lex no doubt refers to the XII Tables. 172

Delict the wrong because his heir would not feel the same thirst for vengeance, and this rule remains even in the developed law for some actions, especially that for iniuria, which was held to be of a particularly vengeful character. Again, the whole system of ‘ noxal’ actions which remains in existence throughout the history of the law bears, on its face, the traces of its origin in the idea of vengeance. It was the rule that, if a slave or a child under power

committed a delict, an action known as ‘ noxal’ lay against the master or father, in which he had the choice either to surrender the wrongdoer (nowae

deditio) or to pay what a free person would have had to pay had he committed the same delict. Originally this meant that the person wronged had the right to take vengeance on the wrongdoer, but that the father or master could buy off that vengeance; obviously the slave or child could not do so himself, for he had no property with which to pay. If before the action was

brought the slave changed hands or the child was adopted into another family, then it was the new master or father who was liable, again clearly because the primary right was to vengeance, and the payment of damages merely a possibility open to the domestic superior for the time being to ransom his inferior.* If the child was emancipated or the slave manumitted, the liability was transferred to him himself, and he had to pay just as if he had not been under power when he committed the delict; for the liability to vengeance is not affected by any change in status or family relationships. In the XII Tables ’ noxal liability extended to furtum and nowia (the latter probably denoting offences against property other than furtum) but not to offences against the person (iniuria) for which a noxal action was provided only by the praetor.’ The idea of guilt also is in a state of transition in the XII Tables.? The primitive principle may have been that a man was responsible for his acts 8 Similarly, death before the action was brought extinguished the liability. On surrender of a dead body after litis contestatio see de Zulueta, Gaius 274, with references. ® Tab. XI1.2.

1 On the details of the early law there is conjecture and controversy; see especially F. de Visscher, Le régime romain de la noxalité (Brussels, 1947), also incorporating the results of a number of earlier studies; de Visscher summarised his main ideas in Jura 11 (1960) 1ff.; see also Kaser, RPR 1.163f., AJ 225ff. One difficulty is to find a juridical basis for the condemnation of the paterfamilias. This should on ordinary principles be found in an obligation existing before litis contestatio, and it is commonly held that the obligation here derives from the act of the paterfamilias in protecting the culprit. De Visscher, however, observing that this is incompatible with the rule (see text above) that the paterfamilias is liable only if the culprit is still in his potestas at litis contestatio, contends that the obligation arises only at that point. He believes that before the noxal actions were introduced the injured party could, by a formal but extra-judicial summons, require the paterfamilias either to make surrender or to offer appropriate compensation. The defect of this proceeding — that the paterfamilias could not contest the validity of the claim — was, de Visscher supposes, removed by allowing him to make his offer of composi-

tion conditional upon proof of guilt. But the evidence for all this is very scanty, and it is difficult to see why, if the offer of compensation is the basis of the action, the option of

surrender should also survive. 2 Cf. Perrin, RHD (1951) 383ff. 173

The law of obligations at the time of the XII Tables irrespective of his state of mind, but this principle is no longer universally applied. One of the very few things that we know about the law of murder is that a distinction was already made between intentional and unintentional killing, for a fragment tells us that ‘ if the weapon sped from his hand rather than was thrown by him’, then a ram was ‘ substituted ’ * — probably given to the agnates ~a relic of the time when the agnates of the slain man were entitled to take vengeance on the slayer. A similar distinction between an intentional and an unintentional act was also made, as we have seen, in the case of arson,* and a rough and ready concern with the need for a guilty mind is reflected in the fact that some penalties at least were mitigated for those not of full age.® Indeed the approach of the XII Tables to wrongs other

than murder and arson is perhaps better expressed ® in terms, not of the primitive liability for the act alone, but of hability for an act which the law characterises as necessarily involving a wrongful intention.’ As yet no practical consequence flows from this idea of guilt, since proof of the absence of a wrongful intent is not admitted, but the possibility of admitting such proof will allow the law later to develop a principle of liability only for fault.® If we turn from the question of guilt to the nature of the penalties imposed,

we see that the XII Tables, like other primitive systems, are detailed and rigid; the code tries to lay down exactly what is to happen in every case, so that when the question of guilt is decided, nothing is left to the discretion of the judge. In particular the fixed money penalties must be noticed, the 25 asses, for instance, which is always the penalty for simple iniuria, and for the cutting down of each tree,’ although, no doubt, the severity of the blow and the value of the tree might vary considerably from case to case. In the later law, as we shall see, the praetor found it necessary to improve on this very rough and ready method by introducing an action for iniuria in which

it was left to the judge to estimate the damages payable in any particular case according to his view of the seriousness of the offence,’ and similarly in

the case of the trees he substituted an action for double the amount of the loss inflicted.? The early penalty on the other hand is always a fixed sum or a simple multiple of a definite quantity —the value of an article —not, as sometimes in the later law, of something which is harder to calculate, the loss inflicted on the person wronged by the delict (id quod interest).*”

3 Tab. Viii24. 4 Above, 170. 5 Tab. viri.9, 14. 6 Cf. Kaser, BIDR 65 (1962) 79ff.

Cf. furtum conceptum, above, 167. ’ Below, 275. ”27Actto Above, 170, 171. 1 Gai. 3.224. arborum furtim cacsarum: Lenel, EP 337. The action provided by the XII Tables has usually been distinguished as actio de arboribus succisis; contra, Albanese, APal, 23 (1953) 15ff. ** The multiplication was originally perhaps not of the value at all, but in kind, so that

the man who stole a cow, for instance, had to give two cows: Jolowicz in Cambridge Legal Essays (Cambridge, 1926) 203ff. 174

CHAPTER 12

The law of procedure at the time of the XII Tables’ In the history of Roman law there are to be found, apart from differences of detail, three systems of procedure — that of the legis actiones, the formulary system, and cognitio extraordinaria. The periods during which these systems were in use overlapped each other, but broadly it may be said that the legis actio system prevailed until the passing of the lex Aebutia, probably in the second half of the second century B.C.,? that the formulary system was that chiefly used from the last century of the republic until the end of the classical period ana that cognitio extraordinaria was the system in use in post-classical times. At any rate, for the purposes of the XII Tables we have only to consider the first of these types of procedure — the legis uctiones.

In any discussion of procedure three main questions have to be asked: (i) How does a man who wants to set the law in motion against another begin; how, that is, does he get the other into court? (ii) How is the trial conducted when the parties are before the court? (iii) Supposing that the judgment is in favour of the plaintiff, how is it enforced against the defendant ? We have therefore to consider (i) Summons, (ii) Trial, (iii) Execution.

1 SUMMONS (UN IUS VOCATIO) This process was the simplest that can well be imagined; the man who wished to begin legal proceedings summoned his opponent orally,? wherever he might find him, to follow him to court (in tus — before the magistrate), and

it was the duty of the opponent to obey the summons. If he refused, the summoner called the bystanders to witness and then proceeded to use force,

for the state as yet, and for a long time afterwards, provided him with no help.* If the defendant was sick or infirm with age he had to be provided with a beast to carry him but he could not insist on a cushioned carriage.° 1 See generally Kaser, ZPR; Pugliese, Proc. 1. 2 Below, 218.

3 Whether formal words were necessary is not clear; Kaser, ZPhk 48; Kelly, Roman Litigation 6 n 3. 4 The powerful man, or the man with powerful friends, must therefore, it would seem, have often been able to snap his fingers at a weaker adversary; see discussion by Kelly, Roman Litigation 6ff., and Garnsey, Social Status 189ff.

5 Tab. 1.1: St in ius vocat [ito]. Nt it antestamino: igitur em capito. 2: Si calvitur 175

The law of procedure at the time of the XII Tables The only way in which a defendant could escape from the duty of obeying the plaintiff’s summons was by finding a vindea, i.e. one who would guarantee his appearance before the magistrate when wanted, and, as the plaintiff could not be expected to let a substantial opponent go merely on the guarantee of some impecunious bystander, the XII Tables laid down that where the defendant was a member of the wealthier class (assiduus) the vindex must be one also.°®

2, TRIAL The trial of an action under the legis actio procedure (and also later under the formulary system) was characterised by a remarkable division of the proceedings into two stages, the first of which took place before the magistrate (2m ture), under whose supervision all the preliminaries were arranged, while the second, in which the issue was actually decided, was held before a iudez,’ who was neither a magistrate nor a professional lawyer, but a layman agreed on by the parties and appointed by the magistrate. He was more than a mere private arbitrator, however, because the decision which he subsequently gave

was a judgment which was recognised by the state and which gave rise to execution proceedings (though in the last resort, as with in ius vocatio, it was the successful plaintiff who had to put these into effect).® This, in outline, is the Roman system as it emerges into the light of history,

but if we seek for origins there are a number of points of uncertainty. Was there a time when the magistrate (or the king) conducted the entire proceed-

ings himself? Later tradition said that there was,® and certainly if one pedemve struit, manum endo iacito. 3: St morbus aevitasve vitium escit, iwumentum dato. Si nolet, arceram ne sternito. We know that these provisions stood at the beginning of the

XII Tables, for Cicero (Leg. 2.9) refers to the whole code by the initial words si in tus vocat. On [ito] see Daube, Forms 28f. Whether in ius vocatio was necessary in a proceeding in rem is disputed; Kaser, 2PR 48; Lévy-Bruhl, Recherches 159. 6 Tab. 1.4: Assiduo vindex assidwus esto; proletario iam civi quis volet vindex esto. The exact position of the vindex, who also appears in execution by manus iniectio (below, 188) is uncertain. In manus iniectio he was probably a substitute who took over the whole

liability of the defendant, i.e. became the actual party to the action, but it may be that here he was a mere guarantor of the defendant’s appearance; see Kaser, ZPR 49f. with reff.

7 The phrase in iudicio sometimes used in modern books for the proceedings before the iudex, as opposed to those before the magistrate, does not appear to be warranted by the sources, and it is better to say apud tudicem. According to Wlassak, Prozessgesetze 2.26ff., followed by Wenger 190.n 1, iwdictwm meant the proceedings from litis contestatio until judgment, but there are some passages, especially Cic. Part. Or. 99, which go to show that it might also-have the narrower meaning; Buckland, Class. Rev. 40 (1926) 83ff.; Pugliese,

Proc. 2.1.74. 8 Cf. above, 175 n 4.

9 Dionys. 4.25, 36; 10.1; Cie. Rep. 5.3; Pomponius, D. 1.2.2.1. The question has been much debated; see Broggini, Iudex 59ff.; Pugliese, Proc. 1.77ff., 97ff.; Kaser, TR 32 (1964) 336ff., all with reff. The addictio by the magistrate in in iure cessio (above, 149), if that can be regarded as a reliable reflection of an early action, also supports the hypo176

Trial assumes that there must have been in Rome as in other societies a period in which questions of proof and therefore of the decision of a case were left to irrational or supernatural methods, such as ordeal or the taking of auspices,’ the division of proceedings would then have had no point. Again, what was the reason for the division, whenever it originated P * The

view which for a long time dominated the approach of many writers to this subject was that of Wlassak,* who saw Roman proceedings as essentially a voluntary submission by the parties to arbitration, the state merely giving its approval. In support of this view were cited the lack of state intervention in the proceedings, which has already been noticed, the freedom to choose their own iudex which the parties certainly enjoyed in classical law, and the supposedly contractual character of the litis contestatio which terminated the proceedings in iure.* The second and third of these arguments are now, however, commonly thought to be ill-founded,° and the first need be no more than a reflection of the paucity of means of detailed law enforcement in primitive society. More fundamentally the objection to Wlassak’s theory ° is that it paints too idyllic a picture of that society, for it assumes that voluntary submission to a rational settlement must have preceded any state organisation. It is more likely, as has been said above, that the arbitrament to which primitive man submitted himself was an irrational or supernatural one, and that the function of the king, who was also the chief priest, was to determine the method which should be used (i.e. to deliver a so-called medial judgment).” Here perhaps was the origin of the division of the proceedings, and it may well be that the function. of the tudex when he was first appointed was not to weigh the evidence (such a transition from the irrational to the rational would be too abrupt), but to determine the verdict from his own knowledge, in somewhat the same way as did the early English jury. thesis. So also, on one view, does Tab. 1.8, which provides for the non-appearance of one party: Post meridiem litem praesenti addicito. But though addico is normally appropriate to the praetor (Kaser, HB 78n), it may well here be used of the twdex (Jolowicz, Mél. de

Visscher 1.488n). For another explanation see below, 178 n 2. Broggini, Iudex 87ff., argues that the division of proceedings did not become obligatory until the introduction of the legis actio per iudicis arbitrive postulationem (below, 182). 1 Lévy-Bruhl, Recherches 73ff.; Broggini, Coniectanea 133ff. (=Recuetls de la Soctété Jean Bodin 16, 1965, 223ff.); J. Ph. Lévy, Mél. Lévy-Bruhl 133ff.; ef. Pollock and Maitland 2.598ff.

2 For discussion see Kaser, Irish Jurist 2 (1967) 129ff.; TR 32 (1964) 329ff.; Pugliese, Proc. 1.77ff. Kaser, while accepting the likelihood of an initial stage in which the king made the decision himself, thinks that the reason for the division was the recognition of the desirability of recourse to an impartial outsider, and the sense that the mere trial of a case was not the function of the bearer of the royal or magisterial powers of command. But this explanation is surely too rational. 3 In many places, but see especially Der Judikationsbefehl der rém. Prozesse in Sit-

eungsb. Ak. Wiss. Wien 197.4 (1921). 4 See below, 184. 5 See below, 178f. and 184 n 8. 6 Kaser, opp. citt. (above, n 2). 7 See Jolowicz, Atti Bologna 2.59ff.; ef. Mél. de Visscher 1.477ff. L77

The law of procedure at the time of the XII Tables We may, moreover, be distorting our view of the primitive proceedings if we think in terms of the single tudex who was normal in historical times. On

the one hand we hear also of the appointment of one or more arbitri.* In classical law iudex and arbiter are not clearly distinguished, but in origin the difference was probably that the twdex was appointed when the question was one of liability or not, while an arbiter or arbitrt acted when the matter was more complex ° (e.g. the determination of boundaries in the a° finium regundorum or the division of the common property in the a° familiae erciscundae). More radically, on the other hand, we have to reckon with the possi-

bility ' that important cases were decided, not by a single wudew or by arbitri, but by a jury court presided over by a magistrate. For it has been conjectured ? that the centumviral and decemviral courts, which were of this

kind, go back at any rate tothe early republic. There remains the question of how the zudeax or arbitri were appointed.’ Clearly at all times they were normally drawn from an official list (album 1udicum) of authorised persons.* Until the Gracchan reform * only senators could serve, and the list was therefore co-extensive with that of the senate. Thereafter the categories of persons qualified changed more than once. In the post-Gracchan period it seems that the parties could, if they wished, choose someone from outside the album, but it is hardly likely that this freedom existed in the early law,® both because it would fit ill with the rigidity of

early society and because the storm aroused by the Gracchan proposals would then be difficult to explain. A more important question is whether the person who was to serve was chosen by the parties (as on Wlassak’s theory ‘ of the consensual origin of Roman procedure he waquld have to have been) or by the magistrate. On the one hand there is the language of the legis actio per tudicis arbitrive postulationem, as it is revealed by the new fragments of 8 Three are mentioned in Tab. x11.3; ef. Cic. Ley. 1.55. For recuperatores see below, 203 n 7,

9 Broggini, Iudex; Pugliese, Proc. 1.169ff. The blurring of the distinction perhaps oceurred when bonae fidet iudicia appeared, posing questions both of liability and of evaluation. 1 Kaser, 7k 32 (1964) 338ff.

2 By Kunkel; see below, 199. If Kunkel’s arguments as to the nature of Roman jury courts can be accepted, it is likely that the presiding magistrate was bound by the verdict

of the jury. Kaser, TR 32 (1964) 347ff., conjectures that even the relationship of the single tudez to the magistrate may have been seen in this way. This, he thinks, would account for Tab. 1.8 (above, 176 n 9). The tudez is acting as the jury or consilium of the magistrate, and therefore if one party does not appear, the matter is taken back to the magistrate, who makes addictio. This involves the assumption that, in the normal case, at the end of the hearing before the iudexz there was originally what would amount to a third stage, in which the magistrate promulgated the decision of the tudea. Kaser finds an echo of this in Censorinus on the lex Plaetoria (Bruns 1.45), and Servius, in den. 12. 727. 3 Kaser, TR 32 (1964) 355ff.; Pugliese, Proc. 1.100ff.; 2.1.215ff.

+ Cf. above, 80f. : 5 Below, 315. 6 Contra, Broggini, Zudex 18f. 7 Above, 177. 178

Trial Gaius.* The plaintiff (and apparently he alone) asks the magistrate to ‘ give ’ a tudex, which is hardly the language appropriate to a request for approval of a tudex already agreed by the parties. But on the other hand Cicero says ° that ‘ our ancestors ’ wished that no-one should be a tudew in any case unless

he had been agreed to by the parties; and other evidence suggests that the parties had some say.' The explanation is probably that the appointment was indeed made by the magistrate, but that he would in practice take account of the wishes of the parties, and more especially would not force any particular iudex on an unwilling party. A. PROCEEDINGS IN IuRE. It is in this stage that the highly formal character of the system makes itself evident. Once before the magistrate * the plaintiff

had to set the proceedings in motion by making his claim in a set form of words appropriate to his cause of action. The defendant then (if he disputed the matter) replied similarly in set words* and the magistrate intervened, again in a prescribed form, so that the case might be sent for trial before the tudex. It was these forms of spoken words (for the procedure was entirely oral) which constituted the actual legis actiones, and the forms laid down had

to be followed so exactly that if a plaintiff made the slightest mistake he failed in his action.* Thus Gaius records * that a man who wished to sue for the destruction of his vines and used the word ‘ vines ’ in his claim lost his case because the clause in the XII Tables * under which he was suing spoke only of ‘ trees ’. Had he used the word ‘ trees ’ all would have been well, for

vines are trees, but his failure to use the right word was fatal. From this example we can see that where the claim was based on a statute it had to follow exactly the wording of the statute,’ but there must also have been forms of claim not directly based on any statute but the product of custo8 Below, 182. On addicere iudicem see Kaser, AJ 108 n 21. 9 Clu. 120. 1 Festus, s.v. Procum (Bruns 2.28): ...Hst enim ‘ procare’ poscere, ut cum dicitur vi iwdice conlocando ‘si aliwm procas’, ‘nive ewm procas’, hoc est ‘ poscis’..... 2 After the institution of the praetorship the magistrate concerned was normally the praetor, but originally the only magistrates with tmperium and hence the only ones with jurisdiction were the consuls, or, in the period between the XII Tables and the institution of the praetorship the military tribunes with consular power (above, 14). It must be remembered tvo that the magistrate could not sit on all days but only on those which were marked F (fasti) in the calendar or on those marked C (comitiales) if no comitia

5 4,11. 6 Above, 170.

3 But see below, 182. 4 Gai. 4.30.

were in fact held on them. Days on which he might not sit were marked NV (nefastt).

7 Gaius also says (4.11) that the forms were called legis actiones ‘ either because they were provided by statute ... or because they were adapted to the words of the statutes themselves and so were adhered to as unchangingly as statutes ’. For another explanation see Stein, Regulae 11f. Pugliese, Proc. 1.11f., suggests that the qualification of acttones as legis would only be necessary if there were other actions, and therefore that it may be no older than the first appearance of procedure per formulas. 179

The law of procedure at the time of the XII Tables mary law,® and in either case it was presumably the pontiffs, in their capacity as advisers to magistrate or iwdex, who finally decided whether a form was

admissible or not. When Gaius came to describe the legis actiones (which were almost completely obsolete in his day)' he said that there were five kinds (modi) of legis actio,? but it is clear that these modi were only general moulds in which the action might be cast, and that within these moulds each cause of action had its own appropriate form. In any case, so far as the procedure for beginning an action at the time of the XII Tables is concerned, we need only discuss two of the five modi —sacramentum and iudicis arbitrive postulatio, for, of the remainder, two (manus iniectio and pignoris capio) are primarily methods of execution,* and one (condictio) is of later origin.* (i) Sacramentum. The form of proceedings by sacramentum differed according as the claim was in rem (a vindicatio or claim of ownership especially)

or in personam (a claim to enforce an obligation). As to the former we are well informed, for Gaius’ description is fairly full;*> unfortunately the manuscript is defective where he deals with the latter.°®

(a) Legis actio sacramento in rem (vindicatio). Where the thing claimed was movable it had to be present in court and the plaintiff began by grasping the thing and saying (e.g. if it was a slave) ‘ I assert that this man (? according to a proper title) is mine by Quiritarian right; see, as I have said, I have put my wand upon him’ (Hunc ego hominem ex ture Quiritium meum esse aio secundum suam causam; sicut dizi, ecce tibi, vindictam imposut).” He then laid a wand (vindicta) * on the slave. The defendant in his turn made a claim in the same words and with the same gestures as the plaintiff, and the praetor then called on both of the parties to loose their hold (Mittite ambo hominem). This is normally seen as a formalisation of the self-help which was supposedly the forerunner of litigation,®* but it may equally have originated 8 De Zulueta, Gaius 231; Magdelain, Actions Civiles 8ff., 22ff.; Pugliese, Proc. 1.11ff.; contra, Kaser, Z2PR 25 n 6.

® Note that in the story about the ‘ vines’ Gaius says ‘ there was a responsum that he had lost his case ’. He does not say whether the responswm was given to the magistrate, or later when the case had reached a tudez.

1 Gai. 4.31. -2 Gai. 4.12. the of a right ’. 4 Below, 193. 5 enforcement Gai. 4.16f. 6 Gai, 4.14f. 3 Cf. Buckland 609: ‘ actio did not necessarily imply litigation; it was a process for

7 Noailles, Fas et Ius 45ff. (=RHD, 1940/1, 1ff.) points out that Probus 4.6 (FIRA

2.456) gives among the abbreviations which he expands ‘8.8.C.8S.D.E.T.V.’, showing that he took secundum suam causam with what follows, not with what precedes. For discussion see de Zulueta, Gaius 234; Pugliese, Proc. 1.276ff.

8* The meaning of vindicta is doubtful. Gaius treats it as the equivalent of festuca (wand), but the original meaning may have been different; de Zulueta, Gaius 234; Staszkow, Labeo 8 (1962) 317ff.; SZ 80 (1963) 83ff. 9* De Zulueta, Gaius 233; Kelly, Roman Litigation 2; Pugliese, Proc. 1.27ff., 46f. 180

a Trial as a ritual assertion with magical properties.’ After the praetor’s intervention the parties take up the dialogue again. The plaintiff ? says: ‘I demand this;

will you say on what ground you have made your claim?’ (Postulo anne dicas qua ex causa vindicaveris), and the defendant answers: ‘ I have done right and thus I have laid my wand on him’ (Ius fect sicut vindictam imposui). Plaintiff: ‘ Seeing that you have claimed unrighteously I challenge you to a sacramentum of five hundred asses’ (Quando tu iniuria vindicavisti, D

aeris te sacramento provoco). Defendant: ‘ And I you’ (Ht ego te). The sacramentum in historical times was a sum of money ® which had originally to be deposited, later promised with security,‘ as a sort of stake by each of the parties and was forfeited to the public use by the one that eventually lost the case. As however sacramentum literally means ‘ oath ’ it is supposed that in an earlier form of the proceeding * the parties each made an oath as to the justice of their claims, and the issue then was which oath was justified,® the loser forfeiting his stake as a penalty for his false oath. If one accepts that a judge or judges originally played no part in the proceeding,’ the issue would presumably have been settled by some invocation of the supernatural, such as ordeal or the taking of auspices.’

Once the ‘ bet’ was made, the magistrate would proceed to assign the interim possession of the thing claimed to one or other of the parties, who had to give security that if he turned out not to be entitled he would return it together with any fruit that had accrued meanwhile.* Where the thing claimed was an immovable it appears that originally the parties went and transacted part of the ceremony at least on the disputed land, and later there 1 Kaser, AJ 321ff.; Lévy-Bruhl, Recherches 45ff., 80ff. (but see Pugliese, T& 30, 1962, 518) ; Broggini, SZ 76 (1959) 113ff.

2 Gaius (4.16) says with greater accuracy ‘he who had made the first vindication ’, for strictly, as each has to claim that the thing is his, neither is more of a plaintiff than the other; ef. above, 142. Watson, RIDA (1967) 455ff., pointing to the oddity that the question is asked by only one party, argues that it is the defendant, and therefore that the proceedings were begun by the defendant and not (as above) by the plaintiff. But Gaius describes this party as ‘ qui vindicabat’, and though Gaius later uses vindicare of both parties, he must, when he uses it to identify only one of them, mean the plaintiff. 3 Fifty asses if the matter at issue was worth under 1000 asses, 500 if it was worth 1000 or more, except that a question of liberty could always be raised for a stake of 50. Originally the amounts were probably assessed in oxen and sheep (Cic. Rep. 2.60). The stakes appear to have been deposited at one time with the pontiffs (Varro, L.L. 5.180, Bruns 2.54) and the loser’s no doubt forfeited to the gods whom his perjury had outraged. 4 Praedes, Gai. 4.16. Cf. below, 187f£. and 299f. 5 According to Kunkel’s view (below, 311f.) the oath survived in that form of the legis actto which was used for capital proceedings.

6 Cic. Caec. 97; Dom. 78. 7 Above, 176f.

8 Lévy-Bruhl, Recherches 73ff.; Broggini, Coniectanea 133ff. (=Recueils de la Société Jean Bodin 16, 1965, 223ff.). ® The assignment of interim possession, i.e. until the action has been tried, was called

vindicias dicere, and the sureties praedes litis et vindiciarum; Gai. 4.16; Tab. x11.3; Santoro, 4 Pal. 30 (1967) 5ff. 181

The law of procedure at the time of the XII Tables was a pretence of doing so.' A clod of earth, Gaius-says, was used to represent the land and the parties spoke their words and made their gestures over it * as if it were the whole thing that was present ’.?

(b) Legis actio sacramento in personam. Here we have fewer details as to the form of words used, but the proceedings must clearly have been simpler, as there was no thing of which both parties claimed the ownership, and no need therefore for any touching or wands or interim possession. The plaintiff asserted whatever it was that he claimed as owing to him from the defendant, saying ‘I assert that you owe me’ (Aio te mihi dare oportere),’® and perhaps also adding the ground for the claim, and the defendant denied the debt (if he wished to dispute the matter). Then no doubt there followed the

challenge to the sacramentum, as where the action was in rem, and the appointment of the iudex. If the defendant did not wish to dispute the plaintiff’s claim, he must admit it; he could not remain silent, as he could in sacramentum in rem, for there was here no thing that the plaintiff could take away with him; the plaintiff wanted the defendant to pay him something or

do something for him, and the defendant must therefore either admit the claim or dispute it in such a way that it could be tried. If he admitted it, the effect of such admission before the magistrate was already at the time of the XII Tables * equivalent to that of a judgment, so that execution could proceed just as if there had been a trial followed by a judgment for the plaintiff.

(21) Iudicis arbitrive postulatio. Before the new fragments of Gaius were discovered very little was known of this action, and there was no evidence even that it dated back to the XII Tables. Now we learn that, in contradistinction to sacramentum (described as ‘ general ’),> it could be used only in those cases for which it had been specifically authorised by statute, and that the plaintiff had to state the ground on which he was suing.° If, for instance, he were suing on a stipulation he said Ex sponsione te mihi X milia sestertiorum dare oportere aio: id postulo aias an neges (I affirm that by a sponsio

you are under a duty to pay me 10,000 sesterces; this I ask whether you affirm or deny). The defendant denied the debt (no formal words being, it seems, needed for his denial), and the plaintiff proceeded to ask the praetor for a tudex or arbiter. Quando tu negas, te praetor iudicem sive arbitrum postulo uti des. This means probably that in some cases he said tudicem, in others arbitrum, i.e. that a tudex was asked for when the claim permitted of a simple answer ‘ yes’ or ‘no’, an arbiter when it required the exercise of 1 Festus, s.v. superstites and vindiciae, Bruns 2.42, 46; Cie. Mur. 26; Gell. 20.10.7. 2 Gal. 4.17; ef. Gell. loc. cit. 3 Probus 4.1 (FIRA 2.456). 4 This is shown by the treatment of confession and judgment together in Tab. 111.1: Aeris confesst rebusque ture iudicatis xxx dies iusti sunto; cf. below, 188.

5 Gai. 4.13. 8 Gai. 4.17a. 182

Trial

some discretion,’ as in the partition actions. For after mentioning: these actions Gaius says itaque nominata causa ex qua agebatur statim arbiter petebatur, without adding sive iudez.® It is also likely that the words quando

tu negas were omitted in these cases, for there would be nothing to deny. Gaius speaks only of actions on a stipulation (authorised by the XII Tables),

and of the two partition actions, familiae erciscundae (also under the XII Tables), and communi dividundo, under a lex Licinnia.® Even before the new

discovery it had been conjectured that the partition actions were among those for which 7.a.ve p. had been used, for they would not lend themselves

to sacramental procedure involving an oath that could be declared true or false; but the inclusion of stipulation was a great surprise, for the question whether A owes B 10,000 sesterces or not is as definite as can be,! and it was not even known that stipulation was as old as the XII Tables. The cases mentioned appear to be instances only, but we know of no others. It is likely enough that the a? finium regundorum, in view of its resemblance

to the partition actions, was included,” and possibly also the a° pluviae arcendae and the a° damni infecti.®

Though actions could not be brought by i.a.ve p. without statutory authorisation, Gaius does not say that authorised actions could be brought in that way only, and we know that in one case at least in the later law a stipulation was enforced by sacramentum.* Probably this was always an alternative in the case of stipulation, but it is doubtful whether the partition actions could ever have been brought within the sacramental procedure. It is not clear, however, who had the choice where there was an alternative. Prima facie one would imagine that it was for the plaintiff to begin proceedings in whichever way he chose, but it is noticeable that in the only other case in which we know of alternative penal and non-penal proceedings * the choice lay with the defendant. In any case the great advantage which i.a.ve p. had over sacramentum was that it was non-penal, i.e. there was no risk of forfeiture to the state if the action was lost. Gaius stresses this point particularly —itaque sine poena quisque negabat — and, though these words refer literally only to the defendant, the position of the plaintiff must have been similar. Furthermore there is a great simplification;® all the antiquated preliminaries are gone, 7 Cf. above, 178.

8 Arangio-Ruiz, BIDE 42 (1934) 614; contra, de Zulueta, JRS 26 (1936) 182, especi-

ally because Probus 4.8 gives te praetor iudicem arbitrumve postulo uti des as a set form; ef. de Zulueta, Gaius 239f. ® Cf. D. 4.7.12. Date unknown; de Zulueta, JAS 26 (1936) 178f. 1 For conjecture as to why stipulation was included see Kaser, AJ 253ff.; Broggini, Iudexz 169ff.

2 Contra, Buckland, RHD (1936) 741ff. 2 Kaser, ZPR 79; Broggini, Iudex 168f. 4 Gai. 4.95; below, 196 n 6. 5 Gai. 4.163; below, 231; cf. de Zulueta, JRS 26 (1936) 184. ¢ Lévy-Bruhl, Recherches 95ff. 183

The law of procedure at the time of the XIT Tables

there is no trace of an appeal to the supernatural, and the parties come straight to the issue between them. We cannot tell whether this simplified legis actio was an innovation made by the XII Tables themselves, but it is not

impossible when we consider that the Tables were in part a concession to plebeian pressure, and that the penalties involved in sacramentum would press particularly hardly on the poorer classes. In any case 2.a.ve p. is certainly of later origin than sacramentum. _ The last proceeding in ture in both these legis actiones was the appointment of the itudex, and the arrangement of the trial before him for the next day but one (in diem tertium sive perendinum). Gaius says this expressly for sacramentum in personam, but it is no doubt true of sacramentum in rem and i.a.ve p. also.” The critical moment however was that at which the parties ‘ joined issue’ by the interchange of the solemn forms of assertion and counter-assertion or denial, and this was called litis contestatio, probably because it was originally preceded by a solemn calling upon the bystanders to witness (testar?) what took place.* The chief reason why this moment was

so important was that from then onwards the plaintiff’s right was held to have been ‘ consumed ’, i.e. even if judgment was not obtained, no fresh action could be brought on the same claim. _ It sometimes happened that the proceedings in ture could not be terminated on a single day; in this case, in order to avoid the necessity for a fresh in tus vocatio, sureties of a special sort (vades) had to ‘ go bail’ for the reappearance of the defendant; this sort of suretyship (vadimonium) was replaced at a later date by an ordinary stipulation made by the defendant himself and guaranteed by later forms of suretyship, but apparently not until

after the lex Aebutia, for Gellius® mentions vades among the antiquities abolished by that statute. For the adjournment at the end of the proceedings in iure for the purpose of meeting again before the tudew no bail was needed, because, as we shall see, the zudex could condemn a defendant who failed to 7 Originally the appointment was made at once, and from Gaius’ newly discovered paragraph 4.17a it appears that this remained the rule in #.a.ve p. But for sacramentum a lex Pinaria (date unknown) prescribed an adjournment of thirty days before the meeting for the appointment (Gai. 4.15), presumably to give the parties time to come to terms. Vadtmoniwm would be used to secure the presence of the defendant after the adjournment. 8 Festus, s.v. Contestari (Bruns 2.5): Contestari tem dicuntur duo aut plures adversarii, quod ordinato iudicio utraque pars dicere solet: ‘testes estote’. There has been immense controversy as to the juridical nature of litis contestatio. Wlassak, in accordance with his view of Roman procedure as originally a voluntary submission to arbitration (above, 177), saw it as a contract. As opinion has moved away from Wlassak’s general position, the emphasis has shifted to the formal and unilateral character of the parties’ declarations. Even for the classical law to see it as a contract is no doubt anachronistic, but the element of agreement is clearly more significant. For discussion and survey of the considerable recent literature see Kaser, SZ 84 (1967) 1ff., 521ff.; Labeo 15 (1969) 190ff. 9 16.10.8; cf. below, 219.

184 |

sneer Trial

appear even in his absence, and this was considered sufficient guarantee that he would not fail. Such, in outline, was the nature of the proceedings in iure in the legis actio period. There are many points, not only of detail but of substance, on which

our information fails us. We do not know, for instance, exactly how proceedings were begun. In the subsequent formulary system the plaintiff had to let the defendant know what the case was about, probably by showing him the draft of the formula which he proposed to use (editio actionis), and he had to ask the magistrate to allow the action (postulatio actionis). Something of the sort must presumably have been necessary under the legis actio system as well, but we have no details. The chief point of controversy is the position of the magistrate himself, but this question, being intimately bound up with the formulary procedure, must be left for discussion in connexion with that system.' B. PROCEEDINGS APUD IUDICEM. The proceedings before the zudex * appear to

have been from the earliest times free from restrictions of form. The tudex sat in some public place in the city ;* he could hear cases even on days when

no proceedings in ture could take place, and, like the magistrate, he was assisted by a consilium of advisers. Proceedings began with a brief statement of the case by the parties.* Then came the actual trial. At least in the later republic and afterwards this was conducted as a rule by advocates who spoke on behalf of their clients and produced the evidence on which each side relied. This evidence might be either that of witnesses or of documents, though until well into the empire the testimony of witnesses was preferred. No doubt in

very early times, as has been said above,° the irrational methods of proof _ commonly found in primitive law were found also in Rome.® In the developed

law, however, although there were some rules of evidence (for instance, the general principle that the burden of proof rests on the plaintiff) the judge had a very wide freedom in weighing the evidence that the parties put before him.’ He had, as is already laid down in the XII Tables, to hear both sides,* unless one of them did not appear by midday and had no valid excuse, for the law laid down that in that case judgment could be given for the party 1 Below, 218ff.

2 Or other person or group to which the decision of the ease is remitted; see above, 178.

3 Tab. 1.6, 7: Rem whit pacunt, orato. Ni pacunt, in comitio aut in foro ante meridiem caussam coiciunto. Com peroranto ambo praesentes; Kaser, TK 32 (1964) 349ff. 4 Causae coniectio, Gai. 4.15. Under the formulary system the formula itself would inform the tudez what the case was about. 5 177.

6 Broggini, Coniectanea 133ff. (= Recueils de la Société Jean Bodin 16 (1965) 223ff.) ; J. Ph, Lévy, Mél. Lévy-Bruhl 193ff.; ef. Pollock and Maitland 2.598ff. 7 On questions of proof in classical law see Pugliese, Recuetls de la Société Jean Bodin

16 (1965) 277ff. 8 Ambo praesentes; cf. above, n 3. 185

The law of procedure at the time of the XII Tables who was present.? Serious disease and an appointment made for a trial of some matter with a foreigner ' are the only excuses of which we know. The

hearing had to come to an end at sunset,” but, at any rate in later times when advocates had learnt to make long speeches, adjournments were common in important cases. When both parties have finished, the zudex withdraws to consider his judgment with the help of his consilcum. If he is unable to come to a decision he can swear that ‘ the case is not clear to him ’ (sibi non liquere) and the magistrate can then release him from his duty.* The parties will have to go back to the magistrate and get another zudex appointed.* In the ordinary way, when he could make up his mind, he delivered his judgment (sententia) orally at

once, but he must do so in the presence of all parties, for a party is not bound by a judgment given in his absence. In sacramentum the form of judg-

ment appears to have consisted in a declaration that the oath of one of the parties was zustum,* and if the proceedings had been in rem, or in personam for a definite sum of money, execution could follow directly, as described in the next paragraph. This would also be possible in proceedings by 1.a.ve p. where the claim had been for a liquidated amount. But it is not clear what happened if the claim was for certa res or an incertum.® Under the formulary system we know that judgment could only be for a definite sum of money,’ and it may be that this rule goes back to the legis actiones,

but there are difficulties. It can be argued that in i.a.ve p. an arbiter, at least, could make a valuation in money himself,® but even if this is accepted,

there is nothing to justify a similar assumption in regard to the l.a. sacramento. The explanation may be found in an arbitrium liti aestimandae, of which we know little but the name, which is preserved in Valerius Probus’ collection of abbreviations.” This was perhaps a proceeding for arriving at a

valuation in money after judgment had been given in an ordinary action. There is, however, a further difficulty in a defective passage of Gaius ™ deal-

ing with actions in rem, in which he appears to say that while judgment in the formulary system had to be for a definite sum, in earlier law it could be ® Tab. 1.8: Post meridiem praesenti litem addicito; cf. above, 176 n 9, on the question who made the addictio. When it was the defendant who was absent the plaintiff had probably to make out a prima facie case; ef. Buckland 638. 1 Tab. 11.2: Morbus sonticus and status dies cwm hoste.

. 2 Tab. 1.9: Solis occasus suprema tempestas esto. 3 Gell. 14.2.25, 4 For the difficulties involved in any change of iudex see Buckland 715; Broggini, Coniectanea 227ff. (= TR 27 (1959) 313ff.). 5 Cie. Caec. 97; Dom. 78. 6 Below, 193. The problem would arise, for example, in the case of a thief who had to be condemned in a multiple of the value of the thing stolen.

7 Condemnatio pecuniaria; below, 204. 8 But see Wenger 144 n 19.

9* 4.10 (FIRA 2.456) ; ef. Lex Acilia repetundarum 58 (Bruns 1.68; FIRA 1.96). 1* 4.48. For discussion see de Zulueta, Gaius 264. See also Broggini, Coniectanea 187ff. (=St. Betti 2.119ff.). 186

Execution in tpsam rem. This clearly cannot refer to any enforcement by state action,

but it may be that since in the l.a. sacramento in rem the verdict of the 1udex that the oath of one party was tustum amounted in effect to a declaration of ownership, this could be regarded as a judgment in ipsam rem, the evaluation being left to the arbitrium litt aestimandae. But since in an action in rem there is no relationship between plaintiff and defendant on which the liability of the defendant could be based, it is probably better to assume that failure to hand over the thing in dispute constituted a delict only.’

38. EXECUTION There remains the question how a judgment is enforced if the defendant against whom it js pronounced does not voluntarily comply with it. In modern law we think of execution as directed towards the fulfilment of the actual judgment; if the defendant will not do himself whatever it is that he is ordered to do — hand over the possession of a thing, pay a sum of money, or whatever it is — then the state will, so far as possible, take steps to secure the desired result itself; it will, for instance, seize the thing in question and hand it over to the successful plaintiff or seize and sell sufficient of the defendant’s property to enable it to pay the requisite sum of money to him. This

is not the standpoint of early law; execution is there regarded rather as a method of putting pressures on a defendant who is obstinate in order to break

his will and make him do whatever it is he ought to do. It is therefore directed almost exclusively against the person of the defendant, for it is thus that he can be made to suffer most effectively; only in exceptional cases is execution against the property employed. It must also be noticed that in this early period it is the business of the plaintiff to carry out the execution proceedings; for this he needs the authorisation of the magistrate but the magistrate does not act for him. For the discussion of details it is again necessary to distinguish between actions in rem and in personam. A. ACTIONS IN REM. Here the need for execution can arise only if the judgment goes against the person to whom interim possession has been awarded,°

for if it is in his favour then there is no more to be done. In most cases, at any rate, where the defeated interim possessor did not give up the thing of his own accord, execution proceedings would be against the praedes litis et vindiciarum, who had gone surety for him at the beginning of the trial, for 2 This is the interpretation put by Kaser (Jura 13, 1962, 22ff.; cf. BIDE 65, 1962, S8f., 94ff.) on a provision of the XII Tables (x11.3): St vindiciam falsam tulit, st velit is, tor arbitros tres dato; eorwm arbitrio fructus duplione damnum decidito. For a full discussion of this provision and of other interpretations see Santoro, APal. 30 (1967) 5ff., who himself argues for the more commonly accepted view that the provision relates to the measure of the liability of the praedes litis et vindiciarum; see

immediately below. 3 Above, 181. 187

The law of procedure at the time of the XIT Tables the purpose of their existence was precisely to facilitate execution.* The possibility of proceeding against the party himself has already been considered.°* B. ACTIONS IN PERSONAM. Under this heading we have to consider in strict-

ness only the legis actio per manus iniectionem, which is the only method of execution for judgment debts and goes against the person of the debtor. It will be convenient however to discuss also the legis actio per pignoris capionem,® which is a method of execution against the property, permissible in exceptional cases for the execution of certain debts without judgment. (i) Manus iniectio.’ Thirty days of grace must be allowed after admission in court ® or judgment; if not paid within this period the creditors could take

the debtor before the magistrate and make a ceremonial seizure (manus iniectio) of him, reciting in a set form of words ® the fact of the debt and its non-payment. If the debtor still could not or did not pay, and if no vindea * came forward to dispute the plaintiff’s right of seizure, the creditor had the right to take him away to his private prison. It is usually assumed that there was an actual addictio of the debtor to the creditor by the magistrate, such as occurred in the case of the manifest thief,” and it is clear that there must have been an official authorisation of some sort, for otherwise there would

have been no point in requiring the magistrate’s presence, but our chief authorities do not mention addictio in this connexion.® While in captivity the debtor might be bound ‘ with cord or with fetters of fifteen pounds weight, not less ’,*“ but he remained a free man, owner of his property and capable of

contracting, at least to the extent of coming to some arrangement with his creditor. If he still did not pay and did not manage to come to any arrange4 For the question of the extent of their liability see Santoro, APal. 30 (1967) 63ff. with reff, From Varro, L.L. 6.74 and Festus, s.v. praes (Bruns 2.26) it appears that they originally undertook their liability by saying praes swm in answer to a question put by the magistrate. The extent of their liability must either have been fixed by Tab. x11.3, as Santoro and others hold, or been left to assessment by arbitri (Kaser, Iura 13, 1962,

31f.). 5 Above, 186f. 6 Cf. above, 180. 7 See de Zulueta, Gaius 242ff. 8 Confessio in iure; cf. above, 182 n 4. 9 Gai. 4.21: Quod tu mihi iudicatus (sive damnatus) es sestertium X milia, quandoc non solvisti, ob eam rem ego tibt sestertium X milium iudicatt manuwm inicto. The alternative sive damnatus was probably used in cases where execution could be taken without

judgment (such as legatum per damnationem and confessio: in ture); see de Zulueta,

Gaius 246. 1 Below, 189. 2 Above, 167. 3 Gai. 4.21 and the Lex coloniae Genetivae 61 (Bruns 1.123; FIRA 1.179), which follows the Roman rules of manus iniectio fairly closely. And it is difficult to see how the dispute. about the position of thieves who were addictt (above, 167) could have arisen if tudicatt were also addictté. On the other hand, Gellius, 20.1.44, says Nist dissolverant ad praetorem vocabantur: et ab eo quibus erant iudicaté addicebantur; cf. Kaser, AJ 110f. 4* Vincito aut nervo aut compedibus xv pondo, ne minore: aut si volet maiore vincito.

Gell. 20.1.45. Many editors change the text so as to give a maximum, not minimum, weight, but see Wenger, SZ 61 (1941) 372ff., with literature. 188

Execution ment he might be kept thus for sixty days, but he must be produced on three consecutive market-days ° before the magistrate in the comitium, and the amount of the debt must be publicly announced, in case someone should take

pity on him and pay on his behalf or perhaps to give other creditors an opportunity of asserting their rights.® If nothing was done by the end of the sixty days, the debtor could be killed or sold as a slave abroad; if there were several creditors they could cut the body up into pieces corresponding to the

amount of their debt, but the XII Tables added that no responsibility attached to them if they cut too much or too little.’ The frightful severity of this process of execution, which is vouched for by

the fragments of the Tables,* shows clearly that the law of debt was still regarded as part of the law of delict; the creditor who is not paid what is owed him has suffered a wrong; he desires to take vengeance on his debtor and the law permits him to have his way. The only possibility of escape is the appearance of a vindex; by this is meant a person who prevents the removal of the debtor into captivity by himself offering to dispute the matter, not only at his own risk, but, almost certainly, as in the parallel case of the actio iudicati in the formulary system, at the risk of having to pay double the amount owing if he fails to dispute it successfully.*® It is almost

certain, also, that even the vindex could not dispute the merits of a case which had once been decided by judgment, but could only dispute the validity of the judgment itself,’ or its sufficiency as a basis for manus iniectio on the ground, for instance, that the debt had already been paid, or that the debtor had already made some arrangement with the creditor.’ The extreme severity of the process was mitigated in some way by the lex Poetelia (? 826 B.C.) already referred to in connexion with nexum.*® This

land 619 n 10. 6 Beseler 4.104. 5 Probably the last three market days (nundinae) of the period of sixty days; Buck-

7 Gell. 20.1.49. Tertiis nundinis partis secanto: si plus minusve secuerunt se fraude esto. It has been held that this famous clause is not to be taken literally, but means that the debtor’s property was divided. Gellius however takes it literally, though he says ($ 52)

that he has never heard of anyone being dissected, and so does Quintilian (Inst. Or. 8.6.84). Most modern authorities also regard a literal interpretation as more in accordance with the spirit of primitive law (e.g. Wenger 224f.; Kaser, ZPR 102), but Radin (AJPhil. 43 (1922) 40ff.) thinks that secare refers to a sale of the debtor’s goods (cf. interdictum sectorium, Gai. 4.146), and da Nobrega (SZ 76 (1959) 499ff.) that it refers to the decision to be made as to the sharing of the price if the debtor is sold trans Tiberim. Lévy-Bruhl (QP 152ff.) sees a reference to magical rites devoting different parts of the debtor’s living body to different divinities, but unaccompanied by actual mutilation. 8 Tab. 111.1-6.

9 Cf. below, 197; and for a discussion of the position of the vindex see de Zulueta, Gaius 243; Kaser, ZPR 99. The Lez col. Gen. 61 (above, 188 n 3) says definitely that he must pay double.

1 K.g. on some formal ground, or on the ground that one of the parties had not been

2 Cf. Buckland 619. 3 Above, 164.

present when ‘it was delivered.

189

The law of procedure at the time of the XII Tables law, apparently, made killing and selling into slavery abroad illegal; presum-

ably it also authorised detention beyond the sixty days allowed by the XII Tables in order that the creditor might use the labour of the debtor,* for otherwise, if he had had to let him go free, there would have been no effective process for execution at all. Gaius ° in describing the result of failure to provide a vindex says simply that the debtor was ‘ led home and bound ’. How far execution against property accompanied manus iniectio we do not

know. It is clear that the debtor remained owner of his property during his sixty days’ captivity, for the XII Tables * say expressly that if he wishes he is * to live on his own ’, but whether, after the sixty days, his property went with his person to the creditor is uncertain. In any case the question would not be of great practical importance, for a man would almost always part with his last shred of property before allowing manus tniectio to take place. (ii) Pignoris capio.” This ‘ taking of a pledge ’ was the seizure of a piece of property belonging to a debtor as a means of putting pressure on him to pay

the debt. The seizure had to be accompanied by the speaking of a solemn form of words, but it did not have to be in the presence of a magistrate or of the other party and so some authorities, says Gaius, doubted whether it should be classed as a legis actio at all. It does not appear that the pledge could be sold; probably, if the debtor proved obdurate, it was destroyed. In any case pignoris capio was never a general method of execution, nor was it used for judgment debts. It was allowed only in a very limited number of cases, in some by custom, in others by the XII Tables themselves. All these cases appear to be such as concern the state or religion, and the probability is, therefore, that it was primarily a state privilege, allowed to individuals only by a kind of delegation when it was recognised that their claims were of peculiar public importance.® 4 Kaser, AJ 247ff., thinks that this must long since have been the practice, and that the lex only mitigated the conditions in which debtors could be held (ef. Livy 8.28.8). 5 4,21; ef. Lex col. Gen. 61 (above, 188 n 3): Secwm ducito. Iure civilt vinctwm habeto.

6 {II.4.

7 Gai. 4.26-29; de Zulueta, Gaius 248ff.

8 Buckland 624. None of them could have been enforced by an ordinary legis actio; Pugliese, Mél. Meylan 1.279ff. The two cases concerned with religion, and said to have been introduced by the XII Tables, have already been mentioned incidentally (above, 163 n 8 and n 9). The customary cases mentioned by Gaius are all those where an individual has been placed under a duty to provide a soldier with money, either for his pay (aes milttare) or, if he were a knight, for the purchase or upkeep of his horse. This last burden (aes hordeariwm) is said (Livy 1.43) to have been placed on unmarried women, presumably because, though they might be possessed of property, they could not be rated in the census (Greenidge 74). Pay is said to have been taken over by the state in 406 B.C. The remaining case of pignoris capio, mentioned by Gaius as having been introduced lege censoria (i.e. presumably in the terms of a contract laid down by the censors), is that of the tax-farmers (puwblicani) who were allowed to use this method to enforce the payment of taxes. Here the idea of delegation by the state appears most clearly. 190

CHAPTER 18

Private law from the XII Tables to the fall of the republic: procedure It was in this period that the foundations were laid for the great edifice of the classical law. The four centuries which intervene between the compilation of the XII Tables and the beginning of the principate saw the growth of Rome from a small city-state to be the mistress of a great empire, and the increasing complexity of life which this expansion brought about was necessarily

accompanied by an increasing complexity of the law. In 450 B.C. the Romans were still a community of peasant-proprietors, organised largely on the basis of the gens or clan, whose life could be regulated by a few comparatively simple institutions; in 31 B.C. they were a diversified population, in which the gentile organisation had ceased to have any importance, and their Jaw, as a result of foreign conquest and the spread of commercialism, was already marked by that individualism which was to become one of its most

distinctive features. : There is great difficulty in dealing satisfactorily with the law of this period,

for the direct sources from which we can obtain any information are very scanty, and such as exist belong mainly to the latest period.’ In general we have to take the law as we know it to have existed in the time of the great 1 There are a few direct extracts from the late republican jurists in the Digest (sce below, 482), and a larger number of citations of their opinions by their successors of the classical period. Chief among the lay sources must be reckoned the works of Cicero (106— 43 B.C.). Many of the speeches were actually delivered before the courts, though mostly in criminal eases, and thus necessarily contain much legal material, but the philosophical writings are also of great importance. The Republic and the Laws, though professedly descriptive of the ideal state, are to a considerable extent based on idealisations of actual Roman practice and they, as well as the other works, contain many legal anecdotes. Even the letters often refer to legal matters, especially, of course, to Cicero’s private affairs. i. Costa, Cicerone Giureconsulto (2nd ed., Bologna, 1927) is a useful guide; for discussion see Pugliese, Raccolta di scrittt in onore di A. C. Jemolo (Milan, 1963) 4.561ff.; on problems in some speeches, H. J. Roby, Roman Private Law (Cambridge, 1902) 2.451ff. Other authors of importance are the elder Cato (above, 92) and Varro (116-27 B.C.), whose de lingua Latina contains many explanations of legal technical terms and was the chief source from which was drawn indirectly Festus’ work On the Significance of Words. The comedies of Plautus (254-184 B.C.) are full of legal allusions, but have to be used with great care, for the plots are taken from Greek originals, and even where the dramatist uses Roman technical terms we can be by no means sure that he has made the necessary alterations in the facts to make them fit. Terence (185-159 B.C.) is of less value than Plautus as his works are much closer copies of the Greek originals. Watson’s series of volumes on the law in the later republic (see Abbreviations, above, Xxv1I) makes considerable use of these lay sources. 191

Private law to the fall of the republic: procedure classical jurists, and, since we cannot suppose it to have sprung into existence fully developed, deduce the earlier state of affairs as well as we can from

the internal evidence of the institutions themselves, helped out by the little that the Roman jurists themselves tell us of the history of the law, by stray references in non-legal writers, especially the historians, and, to some extent, inscriptions. Even so it is not only details which remain doubtful; there is plenty of room for differences of opinion even on such fundamental matters, for instance, as the development of the contractual system which we find existing in the early empire. Before proceeding to deal with the substantive law it will be best to begin by continuing the history of procedure at the point where it was left at the end of the last chapter, because the procedural changes were not only of great importance in themselves, but a condition precedent to much of the substantive development. The chief creative instrument was, in the later part of our period, as in the early empire, the praetorian edict, and it was to a great extent through the formula that the praetor. worked. The system of procedure, under which the question for decision between the parties is put to a iudex in a form of words (formula) authorised by the praetor, is indeed the system with reference to which the main body of Roman law was worked out. It is thus impossible to understand the development of the law without some knowledge of this procedure.

1. CHANGES IN THE SYSTEM OF LEGIS ACTIONES Even before the introduction of the formulary procedure (and the date and manner of its introduction are much disputed) considerable changes must have been made by way of suiting the old system to altered conditions. In the first place the number of actual forms of words to be cast in one or other of the ‘ moulds ’ of legis actiones was presumably increasing continuously as new grounds of action came to be recognised, either by legislation or by cus-

tom, and we have some evidence of this process in the statement of Pomponius that Sextus Aelius composed a book of actions to supplement that of Flavius ‘ because, with the growth of the state, some forms of action were lacking ’.2 This can only mean that since Flavius’ time new forms had in fact

been recognised and that Aelius put them in his ‘ book of precedents ’; he cannot possibly have created them on his own authority.® 2 2D. 1.2.2.7: Augescente civitate quia deerant quaedam genera agendt, non post multwm

temporis spatiwm Sextus Aelius alias actiones composuit et librum populo dedit, qui

appellatur ius Aelianwm; cf. above, 92. |

_ 3 Cf. Lenel, SZ 30 (1909) 343. We can compare in a general way the growth in the number of writs in the twelfth and early thirteenth centuries; see Holdsworth, HEL 1.398; id., Sources and Literature of English Law (Oxford, 1925) 21. 192

Changes in the system of legis actiones Apart from this gradual development the only changes of which we have any definite knowledge are those introduced by legislation and mentioned by Gaius in his account of the legis actto procedure. Though there is great diffi-

culty in dating the individual laws, they appear to belong to the period stretching from about the middle of the third to about the end of the sécond century B.C., and they tend, not only to a certain simplification of procedure, but more particularly towards ameliorating the position of the poor litigant who had no kin able and willing to help him. The system of sacramentum

meant that each plaintiff or defendant had to find a considerable sum of money for the stake, or, in the later period, find friends who would go surety for him in the requisite amount,* and in the case of manus iniectio the defendant’s position was still worse, for he could not defend himself at all, but had to find a vindex who would undertake the cause at the risk of having to pay double if he failed in the defence. It is true that these disadvantages did not attach to zudicis arbitrive-postulatio, but the field of this simplified legis actio was restricted. The harsh use of the judicial machine by the rich against the poor was a question of political importance throughout the republic, and it must have become increasingly acute with the breakdown of the old gentile system.° The influx of new citizens made it more difficult for a poor man to find the friends whose help he needed in litigation.® A. INTRODUCTION OF LEGIS ACTIO PER CONDICTIONEM. One of the most im-

portant changes made in the old system was the creation of a new legis actio

~that per .condictionem. Of this Gaius’ says that it was introduced by a certain lex Silia for the recovery of. definite sums of money (certa pecunia) and by a lex Calpurnia for all ‘ definite things ’ (certa res).°

The dates of these laws are unknown, but they are commonly assigned to 4 Cf. above, 181. That 500 asses was a large sum is clear from the fact that a penalty of 300 was inflicted for the serious offence of breaking another’s bone; above, 171. 5 Including the protection of ‘ clients’ by the gens to which they were attached.

6 Cf. E. Jobbé-Duval, Etudes sur Vhistoire de la procédure civile chez les Romains (Paris, 1896) 28. In early law the man without family is a man without rights. It is worth noticing that the period'of these reforms (slight as they are) seems to correspond with the appearance of democratic tendencies at Rome about the end of the third and beginning of the second centuries B.C. Cf. also the laws relating to suretyship (below, 300f.) which were probably the result of agitation on the part of the popular party. 7 4.19,

8 ‘A claim for certa res is'a claim for the conveyance either of a specific thing already

identified (‘the slave Stichus’, ‘the estate at Tusculum’) or of a definite quantity of fungible things of a definite quality (‘a thousand sesterces’, ‘a hundred bushels of the best African corn’, ‘a hundred jars of the best Campanian wine’). Any other claim is one for an incertwm, in particular where the claimant alleges that the other party is

under an obligation to perform any act other than a conveyance of property (D. 45.1.68; 74; 75). The question is of special importance with respect to stipulations. Under the formulary system the proper action on a stipulation for certwm was a condictio, on one for an incertwm an actio ex stipulatu; Gai. 4.136, 187; D. 12.1.24. Cf. below, 215 n 6. 193

Private law to the fall of the republic: procedure the third century.® The proceedings are described shortly by Gaius in a passage of which the beginning was not known until the recent discoveries.’ The plaintiff said Aio te mihi sestertiorum X milia dare oportere: id postulo aias

an neges. The defendant denied his hability and the plaintiff proceeded : Quando tu negas, in diem tricesimum tibi rudicis capiendt causa condico (Since you deny, I summon you to appear on the thirtieth day for the purpose of appointing a iudex). The parties appeared again for this purpose on the thirtieth day, the legis actio taking its name, as Gaius says, from the summons, ‘for condicere in the ancient language is equivalent to denuntiare ’.? It will be seen that there are great similarities with wudicis arbitrive postulatio;* there is in both cases a direct assertion and denial of obligation, free from the troublesome preliminaries and from the danger of forfeiture involved in sacramentum. But there are also important differences. In condictio the plaintiff does not have to state the ground of his action, as he does in i.a.ve p., and the thirty days’ adjournment, introduced for sacramentum

by the lea Pinaria, applies to condictio but not to t.a.ve p. It may be that there was another difference. I.a.ve p. was completely without risk to either

party beyond the matter in dispute, but in the condictio of the formulary period, which took the place of the legis actio of that name,* when the action

was for a definite sum of money there was a risk, for the loser had to pay one-third of the sum at issue to the winner. This was secured by sponsio et restipulatio tertiae partis, i.e. the plaintiff said to the defendant: ‘Do you promise to pay me one-third of the amount claimed if judgment goes in my favour? ’* The defendant promised and then proceeded in his turn to stipulate for a similar payment if judgment went against the plaintiff. There was thus a bet on the result of the trial, and the loser’s stake went to the winner, not as in sacramentum to the state. The predominant opinion ° is that this practice goes back to the legis actio period. It is at first sight surprising, if this is correct, that Gaius makes no mention of it, even though his account is very short, but the explanation may be’ that the purpose of the thirty days’ adjournment was to encourage the parties to come to a compromise, and therefore that the sponsio et restipulatio took place only at the second stage of the proceedings, when the zudex was appointed; and of this stage Gaius

says nothing. | There remains the question of the scope of the new action. Gaius says *

that he does not know why it was introduced, seeing that actions de eo quod

9 Kaser, ZPR 81, with reff. 1 4,17b.

2 Cf. Festus, s.vv. condicere and condictio, Bruns 2.5. * Above, 182f. 4 Below, 214 ff.

5 The exact form is uncertain, but see Gai. 4.180; Lenel, HP 238. 6 See especially Arangio-Ruiz, BIDRH 42 (1934) 622ff.

is in issue. 8 4,20.

7 In any case there was almost certainly no sponsio et restipulatio in actions under the lex Calpurnia, for even in the formulary period there is no trace of them when certa res 194

Changes in the system of legis actiones nobis dari oportet could be brought by sacramentum or iudicis postulatio. This shows that he at any rate did not think that condzctio made any claims actionable which had not been enforceable before by at least one of the older methods. But it is possible that he was mistaken. It may be that reform was needed because there were new grounds of claim pressing for recognition — the literal contract,® mutuum,' and possibly some forms of what was later called quasi-contract.? [.a.ve p. would not be available for them without statutory authorisation, and perhaps at the time the grounds on which sacramentum in personam could be used had become so stereotyped that legal conservatism would not readily permit its extension to others. Subsequently, when they had been made enforceable by condictio, it may have been held that sacramentum as a ‘ general action ’ must have been available also, and this might account for Gaius’ mistake.* In any case it is notable that the innovation did not take the form of prescribing specific grounds of claim for which the new procedure could be used. Instead it prescribed the object; on whatever ground a man claimed that certa pecunia or certa res was owing, he could sue by condictio, and there can be no doubt that the ‘ abstract ’ nature of the new legis actto must have facilitated the recognition of fresh grounds of obligation.*: °

B. PRocEDURE PER SPONSIONEM. Another simplification of proeedure, this

time for the decision of questions of property, which also almost certainly goes back to the time of the legis actiones, is the device of trying questions of ownership directly, by means of a stipulation. The plaintiff says to the defendant, ‘Do you promise to pay me twenty-five sesterces if the slave 9 Below, 282.

1 Prichard, Synteleia Arangio-Rwiz 260ff., argues that it was introduced to fill the gap in the ereditor’s recourse against the debtor which had resulted from the emasculation of nexum by the lez Poetelia (but the effect of this lex is uncertain; above, 189). Tomulescu,

Irish Jurist 4 (1969) 180ff., suggests that the new action, by virtue of the sponsio et

time of inflation. 2 Below, 284. 3 De Zulueta, JRS 26 (1936) 183ff. 4 Cf. below, 214f. restipulatio, the oath, and the abstract formulation, strengthened the creditor’s hand in a

5 It is uncertain also whether there might be a tusturandum in iure delatum (necessariwm). Under the formulary system where there was a condictio for certa pecunia, and in some other cases (Buckland 633), the plaintiff might offer the defendant the chance of deciding the matter by his oath. If the defendant accepted the offer and swore that the claim was unfounded the plaintiff lost the action. The defendant might, however, refuse

to swear and ‘refer’ the oath to the plaintiff, and then, if the plaintiff swore that the claim was justified, the result was as if judgment had been given. If the defendant refused either to swear or to ‘ refer’ he was treated as one who would not defend himself properly and liable to have pressure put upon him by missio in bona, i.e. by the praetor’s putting

the plaintiff in possession of his property. If the plaintiff to whom an oath had been ‘referred’ refused to take it, then the matter was also settled, for the praetor would refuse him his action if he tried to begin again (denegattio actionis). It may be that there is a reference to this oath in Plaut. Pers. 478, Curc. 496, Rud. 14. 195

Private law to the fall of the republic: procedure (e.g.) in question is mine by Quiritarian right ?’ and the defendant promises. Action is then brought for the twenty-five sesterces, and the zudew in deciding

whether these are owed or not has necessarily to decide the question of ownership, which is the object of the proceeding, for the twenty-five sesterces

are a purely nominal sum and are not even in fact paid.* The advantage of this process is that even if the trial is by sacramentum it is sacramentum in personam, where the formalities are simpler than in the case of sacramentum in rem, and as the amount formally at issue is only very small, the stake is only 50 asses, whereas if the object whose ownership was disputed was valuable it would have to be 500 asses in sacramentum in rem.” Once legis actio

per condictionem had been introduced, the nominal amount of the stipulation could be claimed by this form and so no stake forfeited to the state at all. Of course the mere decision on the stipulation in his favour did not secure to the plaintiff the surrender of his thing. In order that he should be able to

enforce this the defendant had to make another promise by stipulation by which he undertook, in effect, 1f the decision on the first stipulation went against him, either to restore the property with interim profits or pay the value.*®

For this promise he had to find sureties, whose function was consequently

similar to that of the praedes litis et vindiciarum in the legis actio sacramento in rem, and the promise was therefore called stipulatio pro praede litis et vindiciarum.® C. INTRODUCTION OF MANUS INIECTIO PURA. The ancient process of manus

iniectio has been described above as a method of proceeding to execution when judgment had been obtained,’ and as applicable, very probably, even without judgment in the case of debts contracted by nexum.? Gaius * after describing its nature goes on to explain that its application was extended by a number of statutes which permitted its use also in the case of all other debts; one of the statutes was the lex Publilia, which permitted the sponsor (surety) who had paid a debt on behalf of the principal debtor to use manus iniectio against the principal debtor if he were not reimbursed. within six 6 In Gaius’ time when the indirect method per sponsionem was used the summa sponsionis was usually claimed by an ordinary formulary condictio, but that the procedure dated from before the formulary system is made highly probable by the fact that it was claimed by sacramentum if the case came before the centumviral court, which used the old procedure (Gai. 4.95). In any case the process could. hardly have been invented at a time

when the more effective and simpler formula petitoria (below, 211) was already in existence.

7 Kaser (see ZPR 77, with reff.) finds support here for his thesis (above, 154f.) of a transition from a relative to an absolute conception of ownership. In the older procedure

both parties assert ‘mewm esse’, whereas here only the claim of the plaintiff is

(indirectly) in issue. 8 Lenel, HP 516ff. 9 Gai. 4.94. 1 :Above, 188.

2 Above, 164. 3 4,22, 196

Changes in the system of legis actiones months ;* another was the lex Furia de sponsu,* which gave similar rights to one of several sponsores or fidepromissores who had been forced, contrary to

the provisions. of that law, to pay the creditor more than his share of the debt. In both these cases the manus iniectio was pro iudicato, i.e. the words ‘as if on a judgment ’ were added to the words spoken when the seizure was

made, and the position of the defendant was similar to that of a judgment debtor, in particular with regard to the necessity for a vindex; if none such appeared he had either to pay or be led away to captivity, and the vindex could only prevent this result by taking over the defence himself at the risk of having to pay double the amount due if it were found to be really owing. We cannot, however, imagine that in this case the burden of proof was on the vindex, for that would mean that anyone by merely alleging e.g. that he had paid as sponsor for the defendant could force the defendant to find a vindex who would prove the negative.‘

Other statutes, Gaius then goes on to say, introduced a different kind of manus iniectio for various debts, namely manus iniectio pura; in these cases there was no need for a vindex, the defendant being allowed to ‘ ward off the hand of the plaintiff himself and himself defend the action ’.’ Finally, by a certain lew Vallia, all manus iniectio except that on a judgment and that under the lex Publilia was made pura,® and the process therefore became merely a method of beginning an action available in particular cases and characterised by the liability of the defendant to condemnation for double if his defence were unsuccessful. Some writers, it is true, hold that there was no such double liability, and there is in fact no direct evidence of it, but without it the process would have no advantage over the ordinary form of procedure

by sacramentum or condictio, and it is difficult to see why it should ever have been invented.? D. INSTITUTION OF THE CENTUMVIRAL AND DECEMVIRAL Courts. Among the

changes in the legis actio system we can include the setting up of two new courts before which proceedings ‘ in tudicio ’ might take place — the centum4 Perhaps this was rather a restriction of a previously existing right to proceed to manus iniectto immediately (if the payment of the debt had itself been made by solutio per aes et libram, above, 161). See Kaser, RPR 1.172, with reff.; RPL 217f. The date of

the lex Publilia is unknown. 5 Below,.300. 6 Cf. Buckland 621.

7 Gai. 4.23. The instances are (a) the lex Furia testamentaria giving manus iniectio for a fourfold penalty (Hpit. Ulp. 1.2) against anyone not coming under one of the classes excepted from the provisions of the law, who had received more than 1,000 asses by way of legacy or donatio mortis causa from the same person; and (b) the lex Marcia providing for the recovery of interest illegally exacted. The date of the lex Furia appears to lie between 204 and 169 (Girard, Mélanges 1.101; contra, Mitteis, RPR 52 n 30); that of the lez Marcia is unknown (104 B.C. according to Rotondi, Leges publicae; 342 B.C., Seckel-

Kubler’s edn. of Gaius, ad 4.23). 8 Gai. 4.25. Date unknown.

® Buckland 622, Mitteis, SZ 22 (1901) 116, held that the doubling was characteristic of m.t. pro tudicato only, but see Kaser, AJ 121ff. 197

Private law to the fall of the republic: procedure viri and the decemviri stlitibus iudicandis.1 The centumviri were a panel, numbering at least in the later republic 105 persons, from which the actual court (consilium) had to be selected for each particular case. We do not know the method of selection, nor the usual number forming a consilium, but some

indication is given by the fact that under the empire, when the number on the panel was 180, the centumviri usually sat in four divisions.” From the time of Augustus they sat under the presidency of the decemviri, and before then of ex-quaestors.* Their jurisdiction plainly covered claims concerning inheritances: hereditatis petitio* and the related querela inoffictos: testamenti* (which probably originated at about the end of the republic). It may have extended more widely than this to include assertions by vindicatio of ownership of land, of tutela, and of liberty, though the evidence of this wider scope is not compelling.* Whatever its exact extent, the jurisdiction seems, in

the republic at least, to have been exclusive. For the principate a difficulty is presented by the fact that the preliminary proceedings in iure’ were still by legis actio sacramento, even after the leges Iuliae*® had otherwise completed the replacement of the legis actiones by the formulary system, and yet there plainly also existed a formulary hereditatis petitio.° The explanation may be that the matter in issue before the centwmviri had to be of a certain minimum value.” Certainly it was a court where causes celébres were heard and the greatest orators appeared;** its reputation, at any rate from the 1 JT.e. literally, ten men for the trial of cases. 2 Pliny, Yp. 6.33.3; Quint. Inst. Orat. 12.5.6. 3 Pomponius, D. 1.2.2.29, gives the need for presidents of the centumvtri as the reason

for the first creation of the decemviri, but both Suetonius (dug. 36) and Dio Cassius (54.26.6) say that this function was given to them by Augustus, and (unless there had been an intervening change from decemviri to ex-quaestors) Pomponius is probably mistaken; but see Brassloff, SZ 29 (1908) 179.

4 Below, 252. 5 Buckland 327.

6 See Pugliese, Proc. 1.202ff.; contra, Kaser, ZPR 39, with reff.; Kunkel, Krim. 119 n 437. There is no sign of any jurisdiction in actions in personam. Gaius (4.16) says that the spear which was set up in the court was the origin of the festuca used im the legis actio sacramento in rem, and though this explanation is very probably wrong (the spear being an ancient symbol of public authority, quite distinct from the festuca; Kunkel, Krim. 117, following Alféldi, AJA 63 (1959) 1ff.), it presumably reflects Gaius’ view of

the court’s functions. 7 Before the praetor urbanus or peregrinus; Gaius 4.31.

8 17 B.C.; below, 218. ® See also Quint. Inst. orat. 5.10.15.

strong. ,

'1* An obscure text (PS 5.9.1) may show that the minimum was 100,000 sesterces; see Lenel, HP 525ff. and Kaser, ZPR 39. If the court’s jurisdiction was exclusive there must have been a like limitation on the querela, and this Kaser, loc. cit., supposes to have been so; but this is conjectural, and the evidence linking the querela with the centumviri is 2* Most famous of all cases was the cause Curiana in which the great jurist Q. Mucius Scaevola and the great orator L. Licinius Crassus appeared on opposite sides in a question of inheritance, the former arguing for a strict, the latter for a more equitable interpretation of the law; Cic. de Or. 1.242ff.; Brut. 144£. See Wieacker, Irish Jurist 2 (1967) 151 ff.

198

The formulary system orator’s point of view, fell in the latest days of the republic,* but rose again in the early empire when the absence of openings for political oratory made forensic opportunities of addressing considerable audiences more valuable. An alternative explanation of the concurrence of jurisdiction may be that after the lex Iulia the unus iudex was allowed to try the ordinary hereditatis petitio, while the querela remained the exclusive province of the centumviri.* The earliest known trial before the centumvzri dates only from c. 145 B.C.,°®

but the court is probably much older.* Festus’ says that the 105 members were made up of three from each tribe, and since the number of tribes did not reach 35 until 241 B.C., it cannot in the form we know it be older than that, but the name centumviri strongly suggests that there must have been an earlier period when the members did in fact number 100.* Moreover the court has features (its tribal composition, the use of the old legis actio per sacramentum, the spear which stood as its symbol)’ which are more likely to have originated before than after 241 B.C. Of the decemviri! even less is known, partly because they ceased to have

any separate existence when Augustus made them into presidents of the centumviral courts.. Pomponius ? speaks of them as being instituted after the

creation of the praetor peregrinus, but, like the centumviri, they may be very much older. The earliest certain evidence we have, however, is an inscription * referring to a certain M. Cornelius Scipio Hispanus, praetor peregrinus in 189 B.C. and decemvir a few years earlier. At least at the end of the republic they were elected in the comitia tributa and counted as magistrates. Of their jurisdiction we know only that they heard claims thai a man held as a slave was free, and vice versa (vindicatio in libertatem, in servitutem).

2 THE FORMULARY SYSTEM A. NaTURE OF THE SystTEM. The chief difference between the legis actiones

and the formulary system is summed up by Gaius when he says* that the result of the lex Aebutia and the leges Iuliae * was to introduce litigation per

35 ThatTae, Dial. 38. 4 Pugliese, Proc. 1.205ff. of Cn. Hostilius Mancinus, probably shortly after 146 B.C.; Cic. de Or. 1.181;

see Girard 1038 n 5. 6 Kunkel, Krim. 115ff. 7 S.v. centumvtralia, Bruns 2.5.

8 La Rosa, Labeo 4 (1958) 30ff. (who, however, goes much further and identifies the original court with the senate of the regal period, which is supposed to have numbered 100). Festus himself rather weakly says that the centumviri were so called for the sake

of simplicity. 9 Above, 198 n 6. 1 See Franciosi, Labeo 9 (1963) 163ff. 2 D. 1.2.2.29. They are often identified with the otherwise unknown iudicés decemviri, who, according to Livy (3.55.7), were included with the tribunes and aediles of the plebs

in the grant of ‘sacrosanctity’ by one of the leges Valeriae Horatiae (449); but these were evidently plebeian officials, whereas both patricians and plebeians could be zviri stlitibus tudicandis; so there would have to have been a development (which Franciosi,

3 CIL 1.38. 4 4.30. 5 Below, 21S.

op. cit., assumes). Against the identification, Pugliese, Proc. 1.189ff. 199

J.——8

Private law to the fall of the republic: procedure concepta verba, i.e. by words adapted in each case to the particular matter in dispute between the parties, the phrase being used in contradistinction to the certa verba—the unalterable forms— which had been characteristic of the legis actiones.* Under the new system the question at issue is submitted to the iudex in a form of words making plain to him that if he finds certain assertions of the plaintiff to be true it is his duty to condemn the defendant, and that if he does not find them true he is to absolve him. Thus in a claim for a definite sum of money (to take the simplest formula as an example) the form is L. Titius iudex esto. Si paret Numerium Negidium Aulo Agerio’ sestertium decem milia dare oportere, iudex Numerium Negidium Aulo Agerio sestertium decem milia condemnato,® si non paret absolvito (If it appear that Numerius Negidius ought to pay ten thousand sesterces to Aulus Agerius, the judge is to condemn Numerius Negidius to pay Aulus Agerius ten thousand sesterces, if it does not appear he is to absolve). What is now done ?n iure is to lay down the formula; apud iudicem, as before, the actual trial takes place and the tude gives his decision. The actual stages in the process were as follows. Summons could still take place in the old form of in tus vocatio, but this might be replaced by a vadimonium,® i.e. the defendant instead of accompanying the plaintiff immediately into the presence of the magistrate promised (by stipulation) to appear on such and such a day. When summoning his opponent the plaintiff had to make it clear to him what the claim was and this notification was known as

editio actionis;' when, on the appointed day, the parties came before the magistrate, a second editio actionis took place, in-which the plaintiff placed

6 Gai. 4.29. |

7 In the pattern formulae (below, 203) the name of the plaintiff (he who brings the action —agit) is always given as Aulus Agerius, that of the defendant as Nwmerius Negidius (he who pays—numerat—and denies—negat). The parallel sometimes drawn between these names and the Richard Roe and John Doe of English law is misleading, for these latter persons are actually feigned to exist and act for certain purposes, whereas the Roman names mean no more than the A.B.’s and C.D.’s of our books of precedents. $ There has been much discussion as to whether the formula was cast in the third person of the imperative or the second. See G. Jahr, Litis contestatio (Cologne/Graz, 1960) 91ff. with reff. The third person would be more in accord with Wlassak’s view (above, 178) of the formula as the work of the parties and not of the praetor. The usage in the sources, however, is inconsistent. The appointment of the iudez is found only in the third person, but the condemmatio occurs in both forms (Gai. 4.43, 46f., 50f.). 9 Vadimonium took two forms: (a) if the hearing could not be completed in one day the defendant was required to give (by stipulation) an undertaking to re-appear (Gai. 4.184; Lenel, HP 80ff.) ; this replaced the old vades (above, 184); (0) in order to avoid the need for in tus vocatio (with its requirement of immediate compliance under threat of missio in bona) the parties might agree on a convenient date, to which the defendant

would bind himself by a similar stipulation. The latter is evidenced only in literary sources (e.g. Cic. Quinct. 61) and in the tablets found at Herculaneum (Arangio-Ruiz, BIDR 62 (1959) 226ff.).

1 Lenel, EP 59ff. It is not clear how precise the notification has to be; see Kaser, ZPR 162; Pugliese, Proc. 2.1.359ff.

200

The formulary system before his opponent and the magistrate the draft formula, drawn no doubt usually with the help of his legal adviser, on which he proposed that the case should be tried.

At the same time occurred the postulatio actionis—the request by the plaintiff to the magistrate that he would grant the action, i.e. order a zudex to undertake the trial on the formula indicated. The defendant, also no doubt usually acting on professional advice, might declare himself satisfied with the

draft or he might demand alterations, in particular the insertion of an exceptio, and the magistrate, who also had his legal advisers, would take part in the proceedings by indicating what form of words he would allow.’ When once the form of words was arranged there remained the question of the itudex. It would usually be after he had been chosen* and his name inserted in the draft that litzs contestatio* took place and the magistrate made his decree ® granting a trial. It is this decree which gives the zudezx his authority, and it is the magistrate’s power of refusing it which preserves in the last resort his complete control over litigation. The decree would be made

orally, perhaps in formal words (tudicitum do), and the formula was probably in principle ora] too,® though in practice it must have been reduced to writing. B. THe Epicr aND THE FoRMULA. The new system gave a new importance to the position of the magistrate. As no formulae were laid down by law and the

particular formula to be used in each case needed his authority to make it effective, he could, if so minded, assent to the use of a formula even if such formula had no basis in the civil law; and, on the other hand, where a party sought to-enforce a civil law right he could render such right nugatory by refusing his concurrence to the formula proposed. The general principles by which he intended to be guided in this matter were, like his other rules, set out in the edictum perpetuum. The praetor however did more than merely announce that in such and such cases he would § grant an action ’.’ He also 2 He could threaten a plaintiff who would not accept certain modifications with denial of action, and a defendant who refused to accept a formula in a certain form with the use of his powers of coercion. 3 See above, 178. A further re-appearance in ture might be necessary, e.g. if the avail-

ability of a particular iadex was in doubt, but litis contestatio would have to take place immediately; Kaser, ZPR 215f.; ef. Lenel, SZ 43 (1922) 570. 4 See above, 184. It was probably now a formless act, requiring an expression by both parties of their willingness to accept the magistrate’s decree (but the expressions were

probably directed to the magistrate rather than to the other party, and an analysis in terms of contract therefore strictly misplaced) ; Kaser, ZPR 215ff., with reff. 3 Kaser, ZPE 217ff., 237.

6 Arangio-Ruiz, Iura 1 (1950) 15ff.; Kaser, ZPR 237 n 10; contra, most recently, Biseardi, St. Biondi 1.647ff. 7 And occasionally would not: D. 12.2.9.5; 25.4.1.10 sub fin. (Lenel, FP 430, 313).

201

Private law to the fall of the republic: procedure set out in his album patterns of the formulae which he would permit to be used,® and this he did, not only where the action itself was one of his own creation, but also where a formula was needed for causes of action already existing at civil law. Thus where the cause of action was, for instance, a loan

of money, there was no need for the praetor to say that he would grant an action, because such a loan created a civil law obligation, and all that was needed was that the plaintiff should have indicated to him a formula for claiming a definite amount of money due at civil law.* Where, on the other hand, the plaintiff’s cause of action was the fraud of the defendant, a fact which, by itself, gave no claim at civil law, the edict said: ‘ If it be alleged that fraud has been committed, and there be no other remedy available on the facts, and the cause appear to me to be a just one, I will grant an action, pro-

vided that not more than a year has passed since proceedings might have been begun ’.? Then followed the pattern formula, which was probably much in the following form: ‘ If it appears that, as a result of fraud on the part of Numerius Negidius, Aulus Agerius ? mancipated * the estate with which this action 1s concerned to Numerius Negidius, and provided that not more than a year has passed since proceedings might have been begun, then, unless restoration be made in accordance with the directions of the iwdez,* the iudex is to condemn Numerius Negidius to pay to Aulus Agerius as much money as shall be the value thereof; if it does not appear, he is to absolve.’ ° In addition there are set out the forms of words to be used for the different exceptiones © which may be introduced, at the instance of the defendant, into the formula in particular cases.

It is obvious that the exact formulation of the issue to be tried in these forms of words was a matter of great importance, and that the praetor, who had the last word in the formulation, exercised a vast influence on the growth of the law in this way. It is also obvious that there is a great deal to be learnt

as to the substance of the law from the study of the formula, because it 8 Perhaps originally in an appendix; below, 357. ® For the formula see below, 214 n 1.

1 Quae dolo malo facta esse dicentur, si de his rebus alia actio non erit et vusta causa esse videbitur, intra annum, cum primum experiundi potestas fuerit, iudicium dabo; D. 4.3.1.1; Lenel, HP 1144f. 2 For these names of the plaintiff and defendant respectively sce above, 200 n 7.

3 This is merely an example of an act involving loss which might be induced by fraud. In the particular case the particular method by which the fraud had caused the plaintiff loss would be set out. 4 For explanation of this clausula arbitraria see below, 213. 5 St paret dolo malo Numerti Negidit factum esse, ut Aulus Agerius Numerio Negidio fundum quo de agitur mancipio daret, neque plus quam annus est, cum experiundt potes-

tas fuit, neque ea res arbitrio iudicis restituetur, quanti ea rcs erit, tantam pecuntam iudex Numeriuwm Negidium Aulo Agerio condemnato; si non paret absolvito. It must be noticed how the formula is made to fit exactly the conditions for granting the action laid

down in the edict. * Below, 206. 202

The formulary system expresses in precise language exactly what are the powers and duties of the iudex in each case. It may even be said that the invention of this instrument of procedure, at once flexible and precise, was not only a sign of the Roman genius for law, but also, to some extent, a cause of the success which the

Roman jurists achieved. We must therefore say something here in more detail of the structure of the formula — the § parts ’ of which it might be made up, and of the different classes of formulae which were in use.

C. Tue ‘ Parts’ oF THE ForMULA. (i) Intentio. Beyond the appointment of the zudex or recuperatores’ (Titius iudex esto; Tittus Maevius . .. recuperatores sunto), which appears at the head of each formula, the most important part, appearing in almost ® all formulae, is the intentio. It is here, as Gaius says,*® that the plaintiff formulates what it is that he claims as his right, and so it is here that is crystallised the question at issue between the parties. Thus in the actio certae creditae pecuniae,’ the intentio is the clause ‘ if it appear that the defendant owes the plaintiff ten thousand sesterces ’, that being precisely what the plaintiff claims, and the question whether he is entitled or not being precisely what the iudea has to decide. In this case the claim is for a certum? and hence the intentio is certa, i.e., it describes the claim exactly, but there are also many cases where the intentio is incerta, and there we find not ‘ if it appear .. .’, but ‘ whatever it appear...” — not si paret, but quidquid paret. Thus e.g. in an action on a stipulation for an incertum the intentio reads ‘ whatever on that account the defendant ought to pay to or do for the plaintiff ’.* 7 A bench of several jurors. They apparently originated in the redress provided for foreigners under early treaties (above, 102) but they appear in certain ordinary actions under the formulary system, including suits for repetundae (see below, 308; originally a civil rather than a criminal action), certain suits involving violence (Cic. Zull. 7; Caec. 23; Gai. 4.141), and others which seem to have as a common feature a marked public interest. Proceedings before them were speedier because inter alta they could sit on dies nefasti (ef. above, 179 n 2). See further, B. Schmidlin, Das Rekuperatorenverfahren (Fribourg, 1963) ; Pugliese, Proc. 2.1.194ff.; Kaser, ZP2 142ff.

8 In some cases (especially actio iniuriarum) the formula consists of 1 demonstratio followed by quantum bonum aequum videbitur condemnato or similar words (below, 213) so that there is no intentio which can be separated from the condemnatio. For the problems connected with these ‘ formulae without intentio’, Buckland 652; Wenger 141 n 3, and literature there quoted, especially Audibert, Mélanges P. F. Girard. Etudes de drott romain dédiés 6 M. P. F. Girard (Paris, 1912) 1.35ff. De Visscher, tudes 359ff. (=RHD, 1925, 193ff.), holds that intentio is confined to formulae in tus and that the st paret clause

of formulae in factum has a quite different nature. But sce Lenel, SZ 48 (1928) 1ff.; Buckland, Jurid. Rev. 48 (1936) 357ff. 9 4,41: Intentio est ea pars formulae qua actor desiderium suum concludit. This definition, which really only fits intentiones in tus, is probably taken over from the legis actio period; Lenel, SZ 48 (1928) 13; de Zulucta, Gaius 260; ef. Schulz, Principles 46.

1 Cf. below, 215 n 6. 2 Cf. ahove, 193 n 8.

3 Quidquid paret Nm. Nm. Ao. Ao. dare facere oportere; Gai, 4.4], 131.

203

Private law to the fall of the republic: procedure In a few cases the intentio stands alone, without anything further; this happens jn what we should call ‘ actions for a declaration ’, i.e. where no remedy is sought, but the court is only asked to decide a certain question, whether e.g. A is the freedman of B or not. These formulae are called praetudiciales.*

(11) Condemnatio. Much more commonly, of course, the plaintiff wants not

only a declaration but a condemnation, and condemnatio is the name given to the clause ‘ in which the iudez is given the power of condemning or absolving ’.° Thus in the actio certae creditae pecuniae this clause runs ‘ condemn

the defendant to pay the plaintiff (ten thousand) sesterces; if it does not appear absolve him ’.® In this case, as there is a definite sum of money mentioned in the intentio, so there is the same definite amount mentioned in the condemnatto, which is consequently certa also. The iudezx has only the choice of condemning in that sum or absolving. But this is not usually the position. Generally the zudex has himself to fix the amount of money for which he will condemn (if he finds for the plaintiff), and the condemnatio clause only indicates to him how he is to arrive at the sum. He may e.g. be told to take the

value of something,’ or a multiple of some value,® or simply ‘ whatever amount of money seems right and just to him’.® Where the intentio is

incerta, the condemnatio simply tells the iudex to condemn for $ that amount ’, i.e. the value of ‘ whatever it appears that the defendant ought to pay to or do for the plaintiff ’.1 In some cases the judge’s discretion is further limited by a clause (taxatio) which fixes a maximum above which he cannot go.” But in any case it is for a sum of money only that the judge can condemn; he cannot condemn the defendant to hand over a house or a slave to the plaintiff, or to perform any service for him, or do anything else whatsoever except pay him a sum of money.* The explanation of this apparently 4 “Pre-judicial’, because the decision is generally needed as a preliminary to further litigation. It was e.g. forbidden that a freedman should summon his patron in tus without special authorisation from the magistrate. If A wanted to bring an action against B and B alleged that he was A’s patron this question would have to be decided before the

ease could proceed. 5 Gai. 4.43. 6 iudex Nm. Nm. Ao. Ao. sestertium x milia condemnato ; si non paret absolvito. 7 Hither quantt ea res est (i.e. at the time of litis contestatio) or quanti ea res erit (i.e.

at the time of condemnatio), e.g. in a condictio for a certa res the whole formula reads as follows: Iudex esto: s.p. Nm. Nm. Ao. Ao. tritici Africt optimi modios centum dare oportere, quanti ea res cst, tantam pecuniam condemnato; s.n.p.a. For q.e.r. erit see e.g. below, 214 n 3.

8 E.g. in actio furti nec manifesti ‘ quanti ca res fuit cum furtum factum est, tantae pecuniae duplum iudex Nm. Nm. Ao. Ao. c.s.n.p.a.’ (Lenel, HP 328). 9 See below, 213, on actiones in bonum et aequum conceptae. 1 See e.g. the formula of bonae fidei actions, below, 205, 211f.

2 The tazatio may mention a specific sum (Gai. 4.51) or the judge may be restricted in some other way, e.g. to a certain fund, as in the actio de peculio et in rem verso.

3 (aj. 4.48,

204

The formulary system inconvenient rule of condemnatio pecuniaria is not known, nor is it certain that it goes back to the legis actio period.* It remained in force throughout the formulary system, however, and, although, as we shall see,® pressure might be put upon an obstinate defendant to fulfil his obligations, in the last resort all that could be done was to order him to pay a sum of money.®

(iii) Adiudicatio. This name is given to the clause appearing in the formulae of partition actions 7 which entitles the zudex to assign the whole or part of the subject matter of the action to one party as his sole property: ‘ let the tudex adjudge what should be adjudged him (? to him to whom it should be adjudged)’.®

(iv) Demonstratio. In certain cases the formula begins (after the nomina-

tion of the judge) with a quod (‘ whereas’, ‘in so far as’)°* clause, the object of which is to define the issue to which the intentio relates.’ If e.g. the seller brings an action on the contract of sale (actio venditt) the formula reads as follows: ‘ Whereas the plaintiff sold to the defendant the slave who is the object of this action, which sale is the matter involved in this case, whatever

on that account the defendant ought in good faith to pay to or do for the plaintiff, that (i.e. the value thereof) the zudex is to condemn the defendant

to pay to the plaintiff; if it does not appear he is to absolve ’.* Here the words ‘ whereas . . . case’ form the demonstratio, ‘ whatever .. . plaintiff ’ the intentio, and the remainder the condemnatio. A demonstratio is only found in actions in personam where the intentio is incerta, thus in all bonae fidei actions and in the actio ex stipulatu, not in condictiones or in actions im rem.* 4 See above, 186. Kascr, ZPR 287, conjectures that it is rooted in the need in primitive delictal liability to buy off the wronged party’s vengeance and therefore to have a fixed ransom. See also Santoro, 4Pal. 30 (1967) 81ff. Kelly, Roman Litigation 69ff., links it with a shortage of money in the late republic and sees it as perhaps an instance of the way in which the law was weighted against the poor and weak; but see Garnsey, Social

Status 198ff. 5 Below, 213f. 6 That the English conmon law courts could also only give damages and not specific performances appears to be not a real parallel but a coincidence, for the rule is due only

to the early disappearance of ‘ real’ actions. * Above, 156. 8 Gai. 4.42 gives quantum adiudicari oportel, iudexz Titio adiudicato, but Titio can hardly be right; Lenel, #P 208.

9 See below, n 3. 1 Gai. 4.40. For criticism see Schulz, Principles, 46.

2 Quod As. As. No. No. hominem quo de agitur vendidit, qua de re agitur, quidquid ob cam rem Nm. Nm. Ao. Ao. dare facere oportet ex fide bona, eius tudex Nm. Nm. Ao. Ao. ¢.s.n.p.a.; Lenel, HP 299. All b.f. iudicia (cf. below, 211) have exactly parallel formulae.

3 The difficulty with formulae containing a demonstratio is that they appear to he illogically constructed. The quod-clause apparently states a fact and yet the truth or falsehood of this fact is one of the matters which the iudez has to decide; one would have expected ‘ whereas the plaintiff alleges that he sold .. .’ or something of the sort. Koschaker, SZ 41 (1920) 339, takes quod to mean ‘in so far as’, as elsewhere in official style (cf. Schulz, History 258), but even so there is no paret to balance si non paret at 205

Private law to the fall of the republic: procedure (v) Praescriptio. There were originally two sorts of praescriptio: (a) pro actore (on behalf of the plaintiff), (b) pro reo (on behalf of the defendant). (a) The praescriptio in this case was a clause limiting the scope which the action would otherwise have. Thus if a stipulation had been made for a num-

ber of payments at different dates and one or more, but not all, were overdue, the promisee, if he wished to sue on these, inserted a clause ‘ let only those payments which are due be the object of the action ’.* If he did not insert this clause, litis contestatio would ‘ consume’ his right and he could never bring an action for the remaining instalments when they, in their turn,

became payable. | (b) Certain defences, subsequently raised by exceptio, were originally

raised by praescriptio, e.g. in some cases an action might not be brought if

its decision would prejudice that of another and more important issue. Originally if the defendant wished to make use of this rule as a defence he inserted a praescriptio in which the judge was told to deal with the matter only if it would not prejudice the more important case.° Demonstratio, intentio, adiudicatio and condemnatio are the only ‘ parts ’ of the formula enumerated as such by Gaius,° but, as we have seen, they do not by any means all of them appear in every formula. Nor, on the other hand, do they exhaust the possible contents of the formula. In any concrete

case it might include additions by which the attention of the judge was directed to special defences put forward by the defendant, answers by the plaintiff to these defences, and so on. To these we must now turn.

(vi) Exceptio. An exceptio is a clause, the effect of which is to direct the judge not to condemn, even though he finds the intentio proved, if he also finds a further set of facts to be true. It is thus always a conditional clause with a negative, i.e. beginning ‘ if not’ or ‘ unless ’, and it is used where the defence is not a denial of the right asserted by the plaintiff, but an allegation, that even though that right may exist, it is unjust that the defendant should be condemned.’ If, for instance, in a claim for money lent the defence

is that the money was never received, or that it has already been returned, the end. Arangio-Ruiz, Rariora 25ff. (ef. Turpin, CLJ 23, 1965, 262f.) suggested that the demonstratio originated in cases where the parties were agreed on liability (or else it had heen already established in some other way) and were only concerned to obtain an assessment of what was due by the defendant; the illogicality came in when the formula was

required to embrace the question of liability also. This involves assuming that the formula originally ended not with a condemnatio, but with an order to make an assessment (e.g. iudex aestimato). It may be better, however, simply to accept that the formula does, illogically, state the case from the plaintiff’s point of view; see Daube, Forms 35; ef. Turpin, op. cit. 270. Against Ashton-Cross, CLJ 18 (1960) 81ff., see Turpin, op. cit. 268ff. And see further, below, 221.

4 Ka res agatur cuius ret dies fuit, Gai, 4.131. 5 Gai. 4.133.

¢ 4.39; Buckland 649. Praescriptio has been mentioned together with them above for the sake of convenienee. 7 Gai. 4.116. 206

The formulary system the defendant can allow the simple formula ‘ if it appear that the defendant ought to pay the plaintiff ten thousand sesterces ’ to stand, because, if the judge believes him, then he must find that the money is not owed; but if the defence is that the plaintiff agreed (informally) not to sue for the money, then, as an informal pact of this sort has no effect on the existence of the obligation at civil law, if the formula is left as it is the judge will be bound to condemn. The defendant can therefore insist on the insertion of an exceptio pacti, so that the instruction to the zudex reads: ‘ If it appear that the defendant ought to pay the plaintiff ten thousand sesterces and if there has not been an agreement between the plaintiff and the defendant that that money should not be sued for, the judge is to condemn, etc.’ * The result of this form is that if the judge finds that such a pact was in fact made he must absolve the defendant. The power of authorising such exceptiones, which, if proved, will result in the plaintiff’s losing his case although his right is perfectly good at civil law, is, equally with the power of granting actions, a source of the praetor’s influence on the growth of the law, and the rules concerning these exceptiones are

just as much part of the tus honorarium as are those concerning praetorian actions, but it must not be thought that all exceptiones are due to praetorian initiative. In a considerable number of cases rules introduced by a lex or a

senatus consultum are made effective by means of an exceptio. If, for instance, it was desired to plead in answer to a claim for the repayment of a money loan, that the loan was contrary to the provisions of the Sc. Macedonianum, which forbade such loans to sons under power, an exceptio had to

be inserted, and similarly where the defence was that the transaction on which the plaintiff founded his claim was contrary to the lex Cincia.® The form of the exceptio was usually in these cases ‘if in this matter there has

been no contravention of any statute or senatus-consult ’,’ it being left apparently to the defendant to make clear apud iudicem, if necessary, what was the lex or senatus consultum on which he was relying.” The reason why the rules under these statutes were enforced by means of exceptiones and not by treating the forbidden transaction as void were not the same in all the cases,° but the use of this praetorian method is a good example of the way in 8’ Sp. Nm. Nm. Ao. Ao. sestertium decem milia dare oportere et si inter Am. Am, et Nm. Nm. non convenit ne ea pecunia peteretur, iudex Nm. Nm. Ao. Ao. sestertiwm decem mMilie ¢.8.r.p.a. (see Gai. 4.119).

® Forbidding gifts above a certain maximum except between near relatives; see above, 87.

1 St in ea re nthil contra legem senatusve consultum factum est; Lenel, #P 513. The case might e.g. be that of A who is suing on a gratuitous promise which B has made by stipulation to pay him a sum of money larger than that allowed by the lex Cincia. 2 In some cases an exceptio in factum specifying the statute was used; Lenel, loc. cit. 3 See Buckland 653, and cf. above, 87. 207

Private law to the fall of the republic: procedure which the ius honorarium and the civil law worked in together; nothing could be further from the truth than to conceive of them as antagonistic systems working in opposition to each other.*

(vii) Replicatio. It may happen, as Gaius says, ‘ that an exceptio which prima facie appears just, really acts unjustly against the plaintiff, in which case an addition is needed to assist the plaintiff ’.° This addition is called a replicatio, and is in the form of a clause which tells the zudex to condemn even though the facts alleged in the exceptio are true, if a further set of facts is also true. If, for instance, a pact not to sue has been made and after that another pact whereby the debtor releases the creditor from that pact and permits him to sue again, in order that he may have the advantage of this second pact it is necessary to allow him to put in the formula, in answer to the debtor’s exceptio pacti, a replicatio pacti, so that the whole formula reads: ‘If it appear that the defendant ought to pay the plaintiff ten thou-

sand sesterces, and there has not been an agreement... that the money should not be sued for, or if afterwards there was an agreement that it might be sued for, the judge is to condemn, etc.’ °

The replicatio was not. necessarily the last word; the formula might be extended by a duplicatio, a triplicatio, or even further.” ® D. CLassiIFICATION OF FoRMULAE. (i) Formulae of civil and of praetorian

actions. All formulae fall first of all into two classes according as they are used for civil or praetorian actions, for, as already observed, there are some cases where the praetor merely authorises a formula for the enforcement of a right already existing at civil law, and others where the right only exists because the praetor ‘ gives the action ’. But the praetor does not always act in exactly the same way when he allows an action which falls outside the

Buckland, Tulane L.R. 13 (1939) 163ff. 5 4,126. 64 Cf.Gai. 4.126. 7 Gai. 4.128f.

8 The Roman exceptiones, replicationes, etc. have often been compared with the pleas, replications, rejoinders, etc. of the old English system of pleading, and there is indeed this central point of resemblance, that the object of both systems was the formulation, by the allegations of the parties, of the issues to be tried between them. In the English, as in the Roman procedure agreement is necessary before the case can proceed (Stephen, Pleading, 5th ed., 137f., quoted by Holdsworth, HEL 3.627f.). But the differences are as great as the resemblance. In form the English pleadings were separate allegations of fact put into the mouths of the parties, whereas the Roman allegations are in the form of clauses conditioning a condemnation. Secondly, in the English system there had always to be a plea in answer to the declaration, even if it were only a direct traverse, whereas, if the Roman defendant’s answer is a traverse he need only accept the formula as proposed by the plaintiff. Also, although an exceptio is often called a ‘ plea in confession and avoidance ’, the defendant is not, as in the English system, necessarily taken to admit the truth of the plaintiff’s first statement. In some ways the English system, though infinitely more developed, is more closely comparable with the system of the legis actiones, where the issue is also formulated by assertion and counter-assertion of the parties. 208

The formulary system civil law, and we can distinguish three different sorts of formulae according to the different types of actiones honorariae. (a) Formulae with a fiction. In some cases where one definite requirement for a civil law action is absent and it is desired that an action should never-

theless be permitted, the praetor exercises his powers by authorising a formula in which the iudezx is simply told to assume that that requirement is present and decide accordingly. Thus the actio furtz proper lies only between citizens, but if the thief is a foreigner an action may nevertheless be brought against him and the formula will be similar to that of the civil law action,

but contain a fiction; it will say not ‘if he ought (i.e. at civil law) to pay damages as a thief’, but ‘if he ought, were he a Roman citizen, to pay damages as a thief ’.°

(b) Formulae in which a different name appears in the condemnatio from that which appears in the intentio. This occurs where a right which, at civil law, belongs to A is to enure, at praetorian law, for the benefit of B, or vice versa, a right which, at civil law, avails only against A is, at praetorian law, to be made the ground of an action against B. For instance if A has autho-

rised his son B to purchase a thing from C, then, at civil law, B alone is liable on the contract, but this is one of the cases in which the praetor allows an action against the father on the son’s contract, and the formula will run something like this: ‘ Whereas B, by authorisation of A, when he was in A’s potestas, bought such and such a thing from C, which thing is the subject matter of the present action, whatever on that account B ought in good faith to convey to or do for C, thereto the judge is to condemn A, etc.’ ! _ (c) Formulae in factum conceptae. Where there is no analogy in the civil law sufficiently close for the use of a fiction, and the case cannot be met in the manner described under (b), the formula used is one in factum concepta,

i.e. there is no reference to a civil law conception such as ‘ owing’ (dare oportere) or ‘owning’ (alicuius esse ex iure Quiritium), but the judge is simply told to condemn, if he finds certain facts described in greater or less detail in the intentio to be true, if not to absolve. Thus, in the example given by Gaius of an action for a penalty against a freedman who has begun pro-

ceedings against his patron without special permission, the formula is as follows: ‘ If it appear that such and such a patron has been summoned to a court by such and such a freedman contrary to the edict of such and such a 9 Quam ob rem ewm, si civis Romanus esset, pro fure damnum decidere oporteret. A fiction was similarly allowed where it was the victim of the theft who was a foreigner; Gai. 4.37. Among the most important ‘ fictitious’ actions is the actio Publiciana, ibid. 36; ef. below, 263.

1 Lenel, EP 278. The formulae of the other actiones adiecticiae qualitatis were similar in the respect here considered. Equally important examples of the same device are the formulae used by and against representatives (Gai. 4.86), and purchasers of bankrupt estates (Kutiliana species actionis, Gai. 4.35). See also below, 257f.

209

Private law to the fall of the republic: procedure praetor, the recuperatores are to condemn that freedman to pay that patron ten thousand sesterces; if it do not appear they are to absolve.’ ?

Opposed to such formulae in factum conceptae, are those in ius conceptae,* i.e. where the intentio refers not merely to the existence of certain facts, but to the existence of certain civil law relationships, in particular those of

‘owning’ and ‘ owing’. For instance in a condictio for a definite sum of money the intentio is ‘if it appears that the defendant owes .. .’ (dare oportere); in an action on sale it is ‘ whatever the defendant ought to convey

or do in good faith .. .” (quidquid dare facere oportet ex fide bona); in a vindicatio it is ‘if it appear that the thing is the property of the plaintiff at Quiritarian law’ (Ai. Ai. esse ex iure Quiritium). Included among formulae im ius conceptae are thus not only those of civil law actions, but also those of praetorian actions with fictions or change of persons, because here too the intentio refers to the civil law conceptions of ‘ owning’ and ‘ owing ’, the formula of the actio Publiciana being, for instance, ‘ if the thing in question would have been the property of the plaintiff at Quiritarian law, had he possessed it for a year’ (si... anno possedisset tum si eius ew 7. Q. esse opor-

teret...).*° |

In addition to the main distinction between the formulae of civil and praetorian actions and those between different classes of praetorian formulae, 2 Gai. 4.46: LRecuperatores sunto. St paret illum patronum ab illo liberto contra edictwm illius praétoris in tus vocatum esse, recuperatores illum libertum illo patrono sestertium decem milia condemnate; st non paret absolvite. The amount of the penalty is not certain; see Lenel, EP 69. 3 Gai. 4.45. 4 The difference between formulue in tus and in factum conceptae cannot be expressed simply by saying that in the one case it is a question of law and in the other one of fact

which is submitted to the judge. In either case the judge has to decide all questions whether of law or of fact which arise; whether e.g..4 owes B money depends not only on

law but on fact, and similarly questions of law might, no doubt, arise in the action against a freedman for beginning proceedings against his patron without leave, whether for instance the act complained of amounted to an in ius vocatio or not. 5 The exact meaning of aclio in factwm (as distinguished from action with a formula in factwm concepta) is disputed. One view (Lenel, EP 203) is that it is simply equivalent to actio praetoria, i.e. would include actions with a fiction and actions with change of persons. Wenger (162 n 12, 169 n 26) believes that actio in factwm can never mean anything but one with formula in factum concepta. Such formulae might be granted for a special case (so-called decretal action) and then forgotten, but they might also form precedents and result in the inclusion of a new pattern formula in the edict; when this happened the action would nevertheless continue to be called in factum. Against this it can be argued that some actions which are called in factwm seem likely to have been formulated with a fiction, rather than in factum, e.g. D. 9.2.11.8, 17, where the actio in factum on the lex Aquilia which is given to the 0.f. possessor is likely to have been formulated with a fiction of ownership. The expression actio utilis, which is also frequently found, apparently covers all praetorian actions which are based on civil law analogies, whether they have formulae in factum or with a fiction or of any other sort. The terms actio in factum and actio utilis thus overlap to some extent. Cf. Wesener, SZ 75 (1958) 220ff. 210

The formulary system there are also a large number of variations in type corresponding to differences in the nature of the right asserted by the plaintiff and the nature of the relief to which he is entitled either by civil or praetorian law. The formula is not, of course, a complete instruction to the tudex as to the law which he is to apply, but it does indicate briefly what it is that the plaintiff demands, and what the duty of the zudex is, and its form therefore varies with the variations in the nature of the claim, and the nature of the relief which it is possible to grant. Among the more important distinctions and categories are the following:

(ii) Formulae of actions in rem and actions in personam. Whether an action is in rem or in personam is immediately clear from the structure of the formula, for when a man is claiming in rem, the defendant’s name does not (apart from exceptional cases)*® appear in the intentio at all, in a claim in personam it necessarily does. Thus in the typical case of a vindicatio,’ because the plaintiff is only asserting a relationship between himself and the

thing he claims, the intentio runs ‘if it appear that the thing belongs at Quiritarian law to Aulus Agerius’, whereas if he is claiming in personam, i.e. is alleging that some other person is under a duty towards him, then to make the extent of his assertion clear it is necessary that the name of the person from whom he claims, i.e. the defendant, should be mentioned. In an actio certae creditae pecuniae therefore the intentio reads ‘ if it appear that Numerius Negidius ought to pay Aulus Agerius.. .’ (iii) Formulae of actiones bonae fidei® and of actiones not bonae fidei. In a number of actions 7n personam where the intentio is incerta (quidquid dare facere oportet), there are added the words ex fide bona, the judge being thus definitely instructed to take ‘ good faith ’ into account, and to condemn for an amount representing what the defendant ought to do in accordance with good faith. Of this nature are for example all the actions on consensual contracts.” Thus if A has sold B a slave and wishes to claim the price or enforce 6 The intentio of the actio negatoria does mention the defendant’s name, but this is necessary to define the extent of the real right that the plaintiff is claiming; Buckland 677. So also with the actio prohibitoria; Lenel, HP 190. 7 In the condemnatio the name of the defendant necessarily always appears, as it is he who, failing restitution, will be condemned to pay money to the plaintiff. 8 The classical phrase is iudicia bonae fidet. ® The list given by Gaius (4.62) comprises the actions on emptio venditio, locatio conductio, negotiorum gestio, mandatum, depositum, fiducia, pro socio, tutela and the actio ret uxoriae, but the last of these seems strictly to have been in bonum et aequum concepta

(below, 213); see Lenel, HP 302ff.; Kaser, RPR 1.337. (B. Biondi, Iudicia bonae fidei (Palermo, 1920) 178ff., doubts the MS.) On the other hand we must add to Gaius’ catalogue the actto commodati (Lenel, EP 253) and possibly the actio pigneraticia (ibid. 255, but see Kaser, RPR 1.537). In the case of depositwm and commodatum the existence of an alternative (and earlier) formula in factwm is proved by Gai. 4.47, and it may also be regarded as certain for pignus (Lenel, HP 254) and negotiorwm gestio (ibid. 102). For an 211

sn nineteen

, Private law to the fall of the republic: procedure

any other claim arising out of the contract the formula will read as follows: ‘Whereas A sold to B the slave in question, which matter is the subject of this action, whatever on that account B ought to convey to or do for A in accordance with good faith, that sum the judge is to condemn B to pay A; if it do not appear he is to absolve.’ ! The practical importance of the insertion of the words ex fide bona is considerable. In particular two points should be noticed.’

(a) Inherence of exceptio doli. If the defence to an actio venditi, for example for the price of goods sold, were that the defendant had been induced to enter into the contract by the fraud of the plaintiff, there is no need (as there would be if the contract were one of stipulation) for him to insist on the insertion of an exceptio doli in the formula. He can simply, apud iudicem,

raise the point that it is unfair for him to be condemned in such circumstances, and the judge, if he finds the facts to be as the defendant alleges, will have to absolve him. The same is probably true of the exceptio pacti.* (b) Possibility of set-off (compensatio). In the ordinary way, if A brought an action against B, B could not originally plead in answer that A was under

an obligation to him (B); B must, if he wished to enforce that obligation, bring a separate action against A. In the case of b./. cudicia however it was already in Gaius’ time possible for the judge to take account of such counterclaims, provided that they arose out of the same transaction. Thus e.g. if A and B are partners and A is suing B for a share of profit made by B out of partnership business amounting to 10,000 sesterces and B wishes to claim from A 8,000 sesterces, A’s share of expenses incurred by B on partnership business, it is open to the judge to take this into account and condemn B only for the difference, 7,000 sesterces.*

In the later law it became possible to plead set-off to actions which were not bonae fidei, but the subject is too complicated and disputable for discussion here.®

In the law of Justinian’s time the antithesis to actio bonae fidet was actio stricti iuris (or strictum iudicium),*® but it is improbable that the classical law had any inclusive expression of this sort;’ the variety of actions which were not bonae fidei was too great for them all to be included under one earlier list see Cic. Off. 3.70. In Justinian’s time, when the phrase iudiciwm bonae fidei has lost the very definite signification that it had in the formulary period, the list of b./. actions is considerably enlarged; see J. 4.6.28. 1 Quod As. As. No. No. hominem q.d.a. vendidit, q.d.r.a., quidquid ob eam rem Nm. Nm. Ao. Ao. dare facere oportet ex fide bona eius iudex Nm. Nm. Ao. Ao. condemnato, s.n.p.a. 2 For other points see Buckland 680; Manual 364. 3 D. 18.5.3. The text is suspect (see e.g. Schulz, CRL 53) but is probably in substance genuine. Biondi, Iudicia bonae fidet (above) 37ff., argues that the exceptiones metus and ret tudicatae were also ‘ inherent ’.

4 Gai. 4.61ff. 5 J. 4.6.30; Buckland 703ff. é J. 4.6.28, 30. 7 Pringsheim, SZ 42 (1921) 649ff.; Biondi, BIDR 32 (1922) 61ff. 212

The formulary system heading, and it has long been recognised that in any case the classification could not be exhaustive.® Instead, therefore, of attempting to formulate a definition of a * strict action ’, it will be better to indicate some of the principal classes of action which are not bonae fidei. (i) Actiones in bonum et aequum conceptae. Most closely allied to iudicia bonae fidei are these actions, where a phrase referring the iudex to equitable

considerations is also found in the formula, but not in the same position. Instead of the words ex fide bona in the intentio there is here a clause (following a demonstratio) referring the zudex to his feeling of justice for the amount for which he is to condemn. Thus the formula of the actio iniuriarum (aestimatoria) ran something as follows: ‘ Whereas . . . Aulus Agerius was

(e.g.) struck in the face by Numerius Negidius . . . whatever amount of money the recuperatores think right and fair that N. N. should be condemned

to pay A. A. on that account that amount of money . . . they are to condemn N. N. to pay A. A. ete.’ *»?

(ii) Actiones arbitrariae. The rule of condemnatio pecuniaria? is a clumsy one; there are many cases in which damages are an insufficient remedy and the plaintiff should be given, not a sum of money, but that to which he has

a right. The formulary system, though it retained the rule that in the last resort the judge could only condemn for a sum of money, knew, in some cases, of a roundabout way in which pressure might be put on the defendant, if unsuccessful, to induce him to fulfil his primary duty instead of waiting to be condemned in damages. This was achieved by the insertion in the condem-

natio of a clause which had the effect of making the judge’s duty to condemn the defendant in damages dependent on the defendant’s not fulfilling his original duty. Thus to take the most important case as an example, if A. A. brings a vindicatio against N. N. to recover his property, there will appear after the intentio (‘ if it appear that the property in question belongs to A. A. at Quiritarian law ’) a condemnatio in the following form: ‘ and N. N. do not make restitution to A. A. in accordance with the directions of 8 Wenger 165 n 13; Buckland 679. In any case the distinction only fits non-penal actions in personam with a formula in ius. Thus condictiones (below, 214), the actio ex stipulatu and the actio ex testamento are counted as ‘ strict’ because the intentio runs si paret dare oportere or quidquid dare facere oportet, but the words ex fide bona are not inserted. 9 Quod... Ao. Ao. pugno mala percussa est ... qg.d.7r.a., quantam pecuniam recuperatoribus bonwm aequum videbitur ob eam rem Nm. Nm. Ao. Ao. condemnari, tantam pecuniam ... recuperatores Nm. Nm. Ao. Ao. c.3.n.p.a. The manner in which the defendant’s name appeared in the demonstratio is uncertain; Lenel, HP 399. 1 List of actions having similar formulae, Buckland 686 n 8. Formulae of this kind were perhaps used especially where sentimental as well as pecuniary damage had to be considered; E. Costa, Profilo storico del processo civile romano (Rome, 1918) 57. On the whole subject see Pringsheim, SZ 52 (1932) 78ff. 2 Above, 204.

213

Private law to the fall of the republic: procedure the tudex, the iudex is to condemn him in the value thereof, etc.’ * If therefore the judge finds for the plaintiff he must announce the fact and give the defendant an opportunity of complying with his findings ;* only if the defendant fails to comply will he proceed to a condemnation; otherwise he must absolve.

The defendant, it must be noticed, is under pressure to comply, because, if he does not do so, the judge will allow the plaintiff to assess the value of what he claims on oath (iusiurandum in litem) himself, and the plaintiff is not likely to be too modest in his assessment.°

The list of actiones arbitrariae (i.e. of actions with the clause referring to the discretion — arbitrium ~ of the judge) includes, in addition to the rez vindi-

catio, also all other actions in rem ° and some actions in personam, e.g. the actio de dolo and the actio quod metus causa.‘ (iii) Condictiones. Under the rubric sz certum petetur in the edict there were two ® pattern formulae, one for use when it was a definite sum of money

that was claimed, the other for the claim of any other certa res.’ The former (already quoted several times as an example) ran: ‘If it appear that N. N. ought to pay A. A. 10,000 sesterces, the judge is to condemn N. N. to pay

A. A. 10,000 sesterces; if it do not appear he is to absolve.”’ : The latter was: ‘If it appear that N. N. ought to convey to A. A. 100 bushels of the best African corn the judge is to condemn N. N. to pay A. A. the value thereof; if it do not appear he is to absolve.’ ? It is characteristic of both these formulae that they allege a civil law debt (dare oportere) without mentioning in any way the cause of action.” They were used, in fact, in some cases where the cause of action was a contract — if A had lent B a sum of money or a sack of corn (mutuum), if B had pro3 neque ea res arbitrio iudicis Ao. Ao. restituetur, quanti ea res erit tantam pecuntam iudex Nm. Nm. Ao. Ao. ¢.8.0.p.0.

4 The word commonly used for this finding of the judge is pronuntiare; see H.-S. s.h.v.

5 The judge is not bound to accept the plaintiff’s figure but, at any rate where the defendant’s refusal to restore appeared to be inexcusable, he would presumably do so. This is indeed given as the rule in D. 12.3.2, but other texts conflict; D. Medicus, Id quod interest (Cologne/Graz, 1962) 205ff., 248f. 6 Except probably those whereby praedial servitudes were claimed; Lenel, EP 193. But see Broggini, Iudex 74 n. 7 There is considerable doubt as to the list; Buckland 659; Kaser, ZPR 257. 8 Lenel, HP 232. Perhaps a third, with a slave instead of the hundred bushels of corn, served as a model for claims of a ‘species’ as opposed to those of a ‘ genus’; Lenel, HP

240. 9 Above, 193 n 8,

1 §.p. Nm. Nm. Ao. Ao. sestertiwm decem milia dare oportere iudex Num. Nm. Ao. Ao. sestertium decem milia c.s.n.p.a. and S.p. Nm. Nm, Ao. Ao. tritici Africt optim modios centum dare oportere quanti ea res est tantam pecuniam Nm. Nm. Ao. Ao. 6.8.N.p.0. Though the intentio is certa in both cases the condemnatio of the second form is incerta because the judge has to arrive at a money value of the corn for himself. 2 Compare the legis actio per condictionem, above, 193.

214

The formulary system mised A a sum of money or any other certa res by stipulation — but they were also used in a number of non-contractual cases. If, for instance, A had paid B 10,000 sesterces in the mistaken belief that he owed B that sum, he could get the money back by an action of this sort as money paid when it was not owed (indebitum). Indeed a number of causes of action came to be recognised: because these formulae were available for cases where it was difficult to assign a definite cause of action for the plaintiff while it was felt to be unjust that the defendant should retain the money (or other thing) that he had got,° e.g. if A had given money to B as a dowry because B was going to marry his

daughter, and the marriage in fact never took place. The word used for bringing an action of this sort was condicere,* and hence condictio may be said to be an action alleging a civil law debt, but not mentioning any cause of action.* In this sense it is opposed, not only to praetorian actions, but also

to civil law actions with a demonstratio, whether ‘strict’ (e.g. actio ex stipulatu) or bonae fidei.*® 7: ®

EK. Triau aND ExecuTIon. The formulary system appears to have brought with it no important change in the proceedings apud iudicem. All we can say with any certainty is that the brief statement of the case with which the trial had previously opened ® had now become unnecessary, as the formula itself was sufficient to inform the judge what the case was about. Judgment too 3 Cf. below, 284f. 4 Thus Gaius (3.91), in explaining the possibility of claiming return of indebitum, says proinde et condict potest ‘si paret ewm dare oportere’ ac si mutuum accepisset. The word

is used presumably because actions with such formulae took the place of the old legis actio per condictionem. 5 Originally only elaims for a certwm could be made in this way. The condictio incertt

is very doubtfully classical; Buckland 683; Lenel, HP 156ff; Schulz, CRL 614; contra, Kaser, EPR 1.599. 6 We cannot say that this usage was at all strict in the classical period, for the following reasons: (i) the name for the action claiming certa pecunia was actio certae creditae

pecuniae, not condictio (Lenel, HP 234); (ii) Gaius says (4.5) that actions in personam with an intentio referring to dari fiert oportere are called condictiones (unless fierive is a gloss, de Zulueta, Gaius 229; but the contrast in the text is then unbalanced) ; and again (4.18) that a condictio is an action in personam with an intentio claiming dari oportere. The definition would include actio ex stipulatu (although that has a demonstratio) and even the actio certt ex testamento although that certainly contained a reference to the eause of action (Lenel, HP 367). This however is a minor matter of terminology — the important thing is to realise in what cases these particularly simple formulae were used. 7 The difficulties in understanding condictio come in part from the close connexion here

between the substantive law and the law of procedure. If condictio is the name of a certain type of action it is also the name under which a very important class of ‘ quasicontractual’ causes of action have always been known to the Roman lawyer; Schulz, CRL 610ff.; Kaser, RPR 1.592ff.; 2.304ff.

8 The three categories here given do not by any means exhaust the types of action which are not b.f. Actions not b.f. include all those in rem, all in factum, all condictiones and a few actions which, though ‘ strict ’, state the ground on which they are brought. 9 Above, 185.

215

Private law to the fall of the republic: procedure was delivered as before.’ We do however find great changes when we come to

execution, the most important innovation being the introduction of execution against the property of the judgment debtor. Not that execution against the person was abolished; it remained, on the contrary, normal throughout the classical period,’ and common even in the late empire, but already in the last century of the republic it was no longer the only possibility.* Apart from this change the most important reform is that it is no longer permissible to proceed at once to execution after the days of grace; instead we find the curious system that the judgment creditor must first bring another action, this time an action on the judgment — actio tudicati. As in the case of every other action, there must be summons and editio actionis, but the cause of action is now the judgment itself. In the normal case, however, there will be no litis contestatio and no trial before a iudex, because it is usually hopeless for the defendant to dispute the judgment. He will generally either pay if he can, or, if he cannot, will admit his liability and then execution will begin. It may, however, happen that he does wish to dispute the matter. He cannot, of course, dispute the judgment on its merits,* but he may wish to plead that it is invalid, e.g. for want of jurisdiction or want of form, or that he has already satisfied it. If he does this there will be litis contestatio and trial in the ordinary way, but there are two rules which must have effectively discouraged frivolous defences; first that the defence will not be admitted at all unless the defendant furnishes security,’ and secondly that if he loses the case

he will be condemned in double the amount of the original judgment. The result is thus in effect much the same as it was under the system of manus iniectio 1udicati where no trial could take place unless the defendant found a vindex, and condemnation was similarly for double if the defence was unsuccessful.®

If the defendant neither pays nor defends’ the actio iudicati then the magistrate issues his authorisation to the plaintiff to take him away into cus-

tody (duci twbet), that is to say, his position is the same as it would have 1 Ibid.

2 The lez Rubria (FIRA 1.169, Bruns 1.97) allows the local magistrates to order ductio (xx1.15), but reserves missto for the praetor (Kx11.47). 3 Missio in possessionem followed by bonorum venditio existed in 81 B.C., the date of Cicero’s pro Quinctio, though judgment is not actually mentioned as one of the grounds on which it might be granted (5.60). How much further back it goes is uncertain; Kaser, ZPR 301. 4 Appeal is an innovation of the empire; see below, 400. 5 Gai. 4.102. This must no doubt sometimes have been a hardship to poor and friend-

less persons who really had an honest defence. 6 Above, 189. 7 The possibilities are payment, defence, admission of liability (confessio) and refusal to defend, i.e. failure to concur in the steps necessary for litis contestatio. In either of the last two cases the praetor could order ductio, for he could always do this on confession or failure to defend in an action for certa pecunia (Lenel, EP 410), and an action on a judgment is necessarily for certa pecwnia. 216

The formulary system

been under the system of manus iniectio after the disappearance of the creditor’s right to kill or sell his debtor.®

But the magistrate may also now authorise execution against the goods of the debtor. In this case he issues a decree putting the judgment creditor in possession of all the debtor’s property (missio in bona); the creditor then advertises the seizure in order to give other creditors the chance to come in

and claim also; at the end of thirty days the creditors meet and elect a magister from among their number who is to proceed to the sale. This magis-

ter, after the lapse of a further period of days, during which he prepares a list of the property and of the debts, then sells the goods to the highest bidder (bonorum venditio),® that is to say to the person offering the creditors

the highest percentage on their debts. If, for instance, the buyer offers a quarter, then he is given a right to the debtor’s assets, in return for which he has to pay each creditor a quarter of what the debtor owed that creditor.

The process, it will be seen, is in effect bankruptcy; at this period in Roman law a creditor must make his debtor bankrupt in order to enforce the

payment of the smallest sum that the debtor will not pay voluntarily; he cannot just take one piece of property sufficiently valuable to satisfy his debt and sell that. The method is clumsy, for it often means imposing much greater

hardship on the debtor than is necessary to secure the creditor his rights. But from the ancient point of view there was no objection to this; the object was not that the state should do for the creditor what the debtor would not do, but that the state should help the creditor to put pressure on the debtor and punish him if he did not pay his debts, a result normally accomplished by locking him up, but one which could also be achieved by taking away all his property; that the creditor was also paid was a secondary, not a primary, consideration.

The relationship between the two forms of execution is to some extent doubtful. We do not know whether missio in bona always accompanied the authorisation to imprison the debtor, or whether personal execution was pos-

sible without execution against the goods; normally at any rate the two would go together. It seems clear, however, that the creditor could waive his right to imprison and rely on missio alone.” In any case there existed, prob-

ably only from the time of Augustus* however, a method by which the debtor could in many cases escape execution against his person. This was by making a voluntary surrender of his property (cessio bonorum) to his credi-

8 Above, 191ff. 9 Gai. 3.77ff.

1 He has an interdict to recover property in anyone else’s possession (Gai. 4.145) and ean bring a ‘ Rutilian ’ action against the judgment debtor’s debtors, in which the intentio will contain the judgment debtor’s name, the condemnatio his own (Gai. 4.35). Cf. above,

209. 2 Wenger 232. 3 It is referred to as cessio e lege Iulia (e.g. Gai. 3.78), i.e. probably Augustus’ law on procedure of 17 B.C., though Mommsen (R6m. Gesch. 3.536) refers it to Caesar. 217

Private law to the fall of the republic: procedure (coeteammetnanennateappadeninnneeigeienenicareeneneapeanemiaeananeantnaaenenaceneteneaa iam seananeneicatenaangireiaiata anh eaaa aa tia tamed aaaiane areata eaten ea nO ae nN AN AeA EARN ATL NTN

tor or creditors. This surrender took the place of the forcible putting in possession by the magistrate and led similarly to a sale of the property, but it had great advantages for the debtor. He escaped the infamia which resulted from an enforced sale and he remained for ever free from danger of imprisonment for his debt.* Not all persons, however, could avail themselves of this means of escape; it was closed probably not only to those whose insolvency was due to their own fault, but also to those who had no property worth the mention to hand over to their creditors.° F. ORIGIN OF THE FORMULARY SYSTEM; DATE AND EFFECT OF THE LEX

ArsBuTia. According to Gaius ® the abolition of the legis actiones (except in cases before the centumviral court ’ and in the proceeding for damnum infectum) * and their supersession by the formulary system was the result of a certain lex Aebutia and two leges Iuliae. This brief statement leaves us in doubt on two fundamental questions. We do not know the date of the lex Aebutia,

though the leges Iuliae are plainly the Augustan procedural legislation of 17/16 B.C.;° and we do not know what the relationship was between the statutes, or to what extent the introduction of the formulary system was due to legislation at all. For although Gaius makes it clear that the abolition of the legis actiones was expressly enacted, he says no more of the introduction of the formulary procedure than that its use was a result of the legislation.’ 4 If the creditors were not paid in full and the debtor afterwards acquired enough pro-

perty to make it worth while (J. 4.6.40; D. 42.3.4; 6; 7), the creditors might bring another action against him and proceed to another sale, but in such action he had the socalled beneficium competentiae, i.e. in the condemnatio of the formula was a clause limit-

ing the condemnation to id quod facere potest, i.e. what the defendant had. He could therefore always pay the amount of the judgment and need not suffer personal execution. The debtor whose goods had been forcibly taken had similarly the beneficiwm competentiae, but only for a year, whereas the man who had made cessio had it for ever (Lenel,

EP 432). In the law of Justinian’s day the beneficitwm meant that the debtor was allowed to retain the necessaries of life; Buckland 694. 5 Woess, SZ 48 (1922) 485ff. There is very little direct evidence for either of these

exceptions but there certainly must have been exceptions, for otherwise all insolvent debtors would have made cessio, and yet we know that personal execution survived. 6 4,30f.

7 Above, 197. Before a case went to the court the formalities of the legis actto sacramento were gone through before the praetor urbanus or peregrinus. 8 Cf. below, 227. It also remained true that where collusive actions were used for creating or transferring rights (manwmissio vindicta, cessio in iwre) the form was always that of the legis actio. The explanation of the exceptions is that litis contestatio by agreement on a formula is intimately bound up with the appointment of the single tudex; where this is not needed, either because, as in in iwre dessio, there is no question to try, or in the case of centumviral matters because the court is already constituted, then the only procedure

possible is the old one. ® Girard, SZ 34 (1913). 295ff.; see further, below, 225. 1 4.30: Per legem Aebutiam et duas Iulias sublatae sunt istae legis actiones, effectwmque est ut per concepta verba, id est per formulas, litigaremus. For a parallel to this use of effectum est to indicate the indirect result of an enactment see Gai. 2.254. 218

The formulary system The only other reference to the lex Aebutia is in a passage of the Noctes Atticae ? in which Aulus Gellius recalls that he put to a lawyer friend of his the question of the meaning of proletarius in the XII Tables.* The friend did not know the answer, and justified his ignorance by the plea that as a lawyer he was concerned only with what was still in force, whereas ‘ proletarii and adsidui and sanates and vades and subvades and twenty-five asses and talions

and searches lance et licio have vanished, and all those antiquated things from the XII Tables (except for legis actiones in centumviral cases) have with the passing of the lex Aebutia, fallen asleep ’.* Here there is no mention

of the formulary system at all, and the use of the words ‘ vanished’ and * fallen asleep ’, suggest that the effect of the lex on the ‘ antiquated things ’

was indirect rather than direct. In regard to the legis actiones themselves this is indeed obviously true, both because there would otherwise have been nothing left for the leges Iuliae to do, and because there is ample evidence that legis actiones were still in use in the time of Cicero.® It is therefore safe

to assume, as a minimum, that the leges Iuliae finally abolished the legis actiones (subject to the exceptions noted). There remain the questions of the date and scope of the lex Aebutia. The two are obviously inter-related — if we could determine the date we could with greater confidence infer the probable scope, and vice versa — and there is a danger of arguing in a vicious circle ° from a conjectural answer to one to an even more conjectural answer to the other. It is important therefore to establish the limits beyond which conjecture begins. The date of the lex Aebutia must obviously be before the leges Iuliae, and probably before the beginning

of the first century B.C. For if the lex had been passed only shortly before the leges Iuliae it could hardly have escaped mention along with them in the later sources; and if it had been passed in the time of Cicero he would surely have referred to it. That much can be said with reasonable confidence, but when one attempts to determine the terminus post quem one is dependent on one’s sense of historical probability. Granted that the lex was in some way concerned with procedural reform, what is the earliest date by which the need for such a reform is likely to have been felt and to have been met by legislation ? Inevitably opinions differ, but there is general agreement that it cannot have been earlier than:the first part of the second century.’

2 16.10.8. 3 Tab. 1.4; above, 176 n 6.

4 Cum ‘proletarii’ et ‘adsidui’ et ‘sanates’ et ‘vades’ et ‘subvades’ et ‘ viginti

quinque asses’ et ‘taliones’ furtorumque quaestio ‘ cwm lance et licio’ evanuerint omnisque tila Duodecim Tabularum antiquitas, nisi in legis actionibus centumviralium causarum

lege Aebutia lata consopita sit. ...

5 For the evidence see Jobbé-Duval, Mél. Cornil 1.542ff.; Marrone, APal. 24 (1955) o39f. On the question of the l.a. per condictionem see below, 223f. 6 See Pugliese, Proc. 2.1.20ff., 58. 7 Girard, Mélanges 1.99f., argued that the tripertita of Sextus Aelius (consul 198; see

above, 92) provided a terminus post quem, since the three parts of which it was com219

Private law to the fall of the.republic: procedure That the lex Aebutia created the formulary‘system is very unlikely. Such changes are seldom made at one stroke,* least of all in a period as relatively primitive as that with which we are concerned. The formula is more likely to have evolved within the jurisdiction of the praetor, and there are two areas in particular in which the need for it would have been felt — that of the consensual contracts and other relationships actionable by bonae fidei iudicia, and that of commercial relations with peregrines. The two areas are plainly not mutually exclusive, since the most obvious form that commercial relations with foreigners would take would be that of sale. It is true that if the development is taken to have occurred after 242 B.C. the two areas would have been within the jurisdiction of different praetors, and there has been debate on the question whether the origin of the formula should be attributed to the urban or the peregrine praetor.® But on the one hand it is unlikely that the need to do justice in relations with peregrines had not arisen before 242 B.C., and on the other hand there is no need to assume that the formulary system sprang from only one root. We may therefore consider the possibility of a praetorian origin of the formulary system in each of the two areas without embarking on this further question (for the resolution of which there is

in any case no sufficient evidence).

It has been suggested ? that the formula might have originated within the legis actio as a convenient method of recording the issue between the parties at the end of the proceedings in ture. Certainly some aide-mémoire would have been useful especially in the |. a. per condictionem and (after the lex Pinaria) the l. a. sacramento, since there was a thirty days’ delay between the determining of the issue and its presentation to the zudex, but it has been objected ? that the style of the legis actio would lead one to expect a formulation in the first person rather than the third. Thus, Gaius * says that when the l. a. per tudicis arbitrive postulattionem was used for the partition of an

inheritance (when of course the issue was not one of hability, but of how much each party should have) parties did not use the normal dialogue, in which, as we have seen, there is an assertion of a claim by the plaintiff and posed were the XII Tables, interpretatio, and legis actiones, without any mention of the formula; but just as ‘XII Tables’ must have included what subsequent legislation there was, so also ‘legis actiones’ could have included what innovations there had been by formula; see Kaser, St. Albertario 2.46ff. Kaser himself inclines on general grounds to a date in the first half of the second century, Pugliese (Proc. 2.1.58) to one in the last decades of the century. 8 The great reform of English procedure by the Judicature Act of 1873 was preceded by piecemeal legislation, the most important of which was included in the Common Law Procedure Acts of 1852-1860. ® See Pugliese, Proc. 2.1.36ff. 1 E.g. Luzzatto, Procedura civile romana 3 (Bologna, 1950) 150ff.; G. Jahr, Litis contestatio (Cologne/Graz, 1960) 24ff., 84ff. For earlier variants of this view see Pugliese,

2 Pugliese, Proc. 2.1.29. 3 4.17a.

Proce. 2.1.23ff.

220

The formulary system its denial by the defendant, but that there was simply a statement of the cause of action (nominata causa ex qua agebatur) and a request for an arbiter. Since the usual dialogue is cast in the first person,* one would expect this statement of the cause (which evidently has some affinities with the sup-

posed aide-mémoire) to take the same form. On the other hand, ArangioRuiz * long ago propounded in this connexion an attractive explanation of the apparently odd wording of the formulary demonstratio. As we have seen,®

the word quod (if taken in its most obvious sense of ‘ whereas ’) appears to admit the plaintiff’s claim and therefore to leave nothing for the tudex to settle except the question of how much the defendant shall pay. And this, suggested ‘Arangio-Ruiz, was exactly what in its earliest origin the demonstratio did mean. For in the case of a partition action under the legis actio, or where the defendant had admitted liability by confessio in ture on other claims, the assessment of the amount payable was remitted to a tudea or

arbiter’ in a primitive formula containing simply a demonstratio and an order to make an adiudicatio or an aestimatio. From this it would be a short step to the use of the same device for settling the quantum due on any other agreed claim, even though it were not recognised by the zus civile (e.g. an informal sale). And in course of time, as the possibility of litigating such ‘ extra-legal’ relationships became familiar, the formula would be extended to allow consideration of the question not only of the quantum but also of liability itself (i.e. whether the sale or other relationship out of which the claim grew had in fact been made). This would have been the origin of the bonae fidei iudicia.

An apparent difficulty in the way of any attribution to the praetor of the origin of the bonae fidez wudicia is that in classical law they plainly belong to the ius civile and not to the tus honorarium. The intentio asserts an oportere,

and this, at least within the formula, is taken as the mark of the civil law obligation. How then could it be a praetorian creation? The explanation probably lies ® in the qualification which follows the oportere and gives the actions their name. The intentio claims quidquid dare facere oportet ex fide bona — whatever it is proper on the basis of good faith that the defendant 4 Above, 182.

5 St. econ.-giur. della Facolta di Giurisprudenza di Cagliari 4 (1912) 75ff. (= Rariora 25ff.) ; and see Turpin, CLJ 23 (1965) 260ff. 6 Above, 205 n 3.

7 Or recuperatores? One argument adduced in favour of an origin of the formulary system in the jurisdiction over peregrines is that recuperatores are not found in the legis actio system but do appear in some formulary actions, and that they themselves originate in international dealings; cf. above, 203 n 7. 8 See above, 209f. Cicero, Off. 3.61, speaks of the bonae jfidei iudicia as being sine lege. ® Kunkel, F'schr. Koschaker 2.1ff.; Kaser, SZ 59 (1939) 68ff.; 61 (1941) 179ff. Against the variant version proposed by Wieacker, SZ 80 (1963) 1ff. see Turpin, CLJ 23 (1965) 264ff.; Kaser, ZPR 110 n 30. 221

Private law to the fall of the republic : procedure should convey or do. In the ordinary (so-called strictt turis) intentio ? the unqualified oportet refers in origin to a basis in lex or interpretatio, but here the standard is the extra-legal one of good faith. No doubt the practice had grown up of entering into informal transactions outside the civil law in reliance on recognised standards of good faith. As has been said above, we need not inquire whether the practice first grew up in dealings between citizens or in those involving peregrines;* what matters is that the praetor would now be directing the iudex to determine what the obligations of the parties were in the light of those same recognised but extra-legal standards. This would in the classical law have been categorised as tus honorarium, but in early times when the law-making powers of the praetor are relatively undeveloped and

systematic thought in its infancy, the lines have not yet been sharply drawn.* * In the classical law, moreover, when the oportere is seen as referring to the civil law, the ew fide bona clause came to qualify also the condemnatio and to give the tudex a wide discretion in assessing the extent of liability.®

If it be accepted that the formulary system was in existence in some form before the lex Aebutia, what was the purpose of the lex ? The view which held

the field until quite recently was that of Wlassak,’ who maintained that until then there had been no formulary actions within the area of the zus

1 Daube, Forms 16ff. 2 Above, 203; ef. 211.

3 We do not know how early the bonae fidet tudicia were recognised, but in D. 19.1.38.1

an opinion of Sextus Aelius on an arbitriwm concerning sale is quoted, and this must refer to something outside the legis actio system. The evidence of Plautus is inconclusive (Ferrini, Opere 3.49ff.). 4 Schulz, History 83. 5 Magdelain, Le consensualisme dans Védit du préteur (Paris, 1958) 109ff. (ef. éd., Actions Civiles), while accepting the broad thesis outlined above, seeks the origin of the

bonae fidet iudicia in an edict and in particular in the edict, de pactis, in which the praetor says pacta conventa servabo. The contracts would later come to be assimilated into the civil law, and the scope of the edict confined to its classical scope, i.e. to those agreements which are valid only ture exceptionis. But this involves the assumption of a degree of forgetfulness which it is difficult to credit. It is easier to believe that the bonae fides tudicia originated before the practice of issuing edicts in such matters had arisen, and therefore that their origin was unrecorded. 6 Another possible root of the formula is provincial practice, but the evidence of Cicero’s time (e.g. Cic. Verr. 11.2.37-42) is too late to be significant. Partsch, Schriftformel, cites also inscriptions of the second century (the earliest being from Magnesia, probably c. 190 B.C.; FIRA 3.501) in which the senate is shown intervening to settle a dispute between independent cities and following a method which echoes the formulary procedure, but it is more likely that the senate would borrow procedure from Rome itself than from provincial practice; Pugliese, Proc. 2.1.30ff. 7 Prozessgesetze 1.104, 128ff., 153f.; 2.17ff., 362ff. He was followed especially by Girard (Manuel 1057; Mélanges 1.67f.), who, however, went further and maintained that until the lez there were no formulary actions at all between citizens, and that the great powers of the praetor, especially that of refusing actions — denegatio actionis — did not exist until

that time. He found support for this in what he saw as a sudden multiplication of 222

The formulary system civile, i.e. that the praetor had given actions only when no legis actio was available. The innovation made by the lex was to allow the giving of formulary actions even for civil law claims. In such cases therefore there was a choice between the old and the new. The leges Iuliae simply abolished this choice (subject to the exceptions already mentioned). In this way Wlassak could account for the evident fact that in the time of Cicero both formulae and legis actiones did exist side by side,’ and also to give significance to Aulus

Gellius’ statement that the legis actiones ‘ fell asleep’. For the leges Iuliae would have done no more than recognise the accomplished fact that the old had been unable to hold its own in competition with the new. In recent years, however, opinion has moved away from Wlassak in favour of a less sweeping theory propounded by Kaser.® Kaser’s principal criticism of Wlassak is that he sees the lex Aebutia through modern eyes as a general reforming statute, whereas everything that we know of the second century and of the style and

content of its legislation would lead us to expect something limited and narrow. Moreover, at this time substance cannot be separated from procedure, and a reform of the kind envisaged by Wlassak would therefore have been much more sweeping in its effect than a corresponding reform in a modern system. Kaser himself accordingly suggests a more limited purpose for the lex. Starting from the fact that although we hear of all the other legis actiones in the time of Cicero, there is no mention of the l.a. per condictionem, he suggests that what the lex Aebutia did was to allow a formulary action only for claims covered by the condictio. In other words, it was only in this area that the choice postulated by Wlassak was made available, and since the formulary action was obviously preferable, it drove out the l.a. per condictionem. It is evident, however, as we have seen, that in the late republic there were other formulary actions for civil law claims, and Kaser therefore supposes that the praetor, encouraged by the success of the condictio, must in the course of time have allowed other formulary actions as alternatives to legis actiones. But these, having no authority in lex, would have been only iudicia imperio continentia and not, like the condictio, iudicia legitima.'

The most important practical consequence of this would have been that examples of praetorian activity in the years after c. 127 B.C. But so late an extension of

the benefits of formulary actions for claims not recognised by the legis actiones is scarcely conceivable, and it is equally difficult to believe that the praetor was transformed overnight from an automaton. into a source of rapid changes in the law (see Lenel, SZ 30 (1909) 329ff., esp. 333). And the evidence for Girard’s date is unconvincing (Mitteis,

RPE 52 n 30). Kelly, Irish Jurist 1 (1966) 344f., conjectures that until the lex the praetor had no power to override the civil law. 8 For legis actiones see above, 198 n 2*; and Cicero, Rose. Com. 24, says sunt formulae

de omnibus rebus constitutae.

9 St. Albertario 1.27ff., restating a thesis put forward earlier by Eisele and others (citations on p. 29). 1 For the distinction and its consequences see Gai. 4.103ff.; Buckland 688; cf. Bonifacio, St. Arangio-Ruiz 2.207£. 223

Private law to the fall of the republic: procedure bringing the formulary action would not bar a subsequent proceeding on the same cause by legis actio,? and there would therefore be a reason for the survival of the other legis actiones, by contrast with the assumed withering away of the l.a. per condictionem, and the leges Iuliae.

Kaser’s view is now widely accepted, but there is room for doubt. The central supposition that the l.a. per condictionem did not survive the lex Aebutia rests on an argument from silence, and the references to the other legis actiones are not so numerous as to make the argument compelling. Again, it is not obvious why the lex should have abolished the most recent and least formal of the legis actiones.* And it is difficult to account satisfactorily for the width of language used by Aulus Gellius’ lawyer friend (even though he does not present the disappearance of the things he mentions as the direct consequence of the lex). To these and other reasons for hesitation Birks ¢ has added another. Both Wlassak and Kaser assume that where two modes of procedure are in competition, the good will necessarily drive out the bad —that because objectively, from the point of view of doing justice between the parties, the formula is better than the legis actio (provided both are iudicia legitima), the latter will necessarily wither away if the choice is

free. But the history of English law and a consideration of human nature shows that this is not so. The parties to an action are not concerned with doing abstract justice by the best method. They want to win the case, and they, or their lawyers, will seize any tactical advantage which procedural devices offer. In the Roman context this will sometimes lead to the disappearance of the legis actio, as in the case of Aulus Gellius’ twenty-five asses; the plaintiff will prefer the actio aestimatoria iniuriarum, because it will give him more. But it will not always be so. One advantage of the formulary procedure from the objective point of view is that by means of excepitones etc. it permits a far more exact and equitable definition of the issue than Is possible in the rigid forms of the legis actio. A plaintiff might therefore see an advantage in proceeding by legis actio instead of by the alternative formulary action, if he could thereby deprive the defendant of the ability to plead

an exceptio. (How far he could in practice do this we cannot, it is true, be sure;° Gaius ° says that there were no exceptiones in the legis actzo, and this is inherently likely, but the praetor could perhaps induce the plaintiff to modify his position by the threat of denegatio actionis; but even if this is so, the plaintiff would still be better off than in the formulary action in which the 2 Except presumably by denegatio actionis; ef. text immediately below.

3 Kaser, St. Albertario 1.43ff., holds that the sponsto et restipulatio terttae partis (above, 194) was compulsory in the !. a., but optional in the formulary action. It therefore made possible litigation without penalty for the loser and favoured poorer persons in a time of tight credit. But since the option is in any case the plaintiff’s, whereas the poorer party is likely to be the defendant, this does not seem very persuasive.

4 Trish Jurist 4 (1969) 356ff. 5 See Kaser, ZPR 53f. de Zulueta, Gaius 231. 6 4,108. 22-4

The formulary system defendant had only to ask for an exceptio to be sure of getting it.) More often, no doubt, a plaintiff would find that by suing under the formula he could obtain an advantage not available in the legis actio, such as the ability to sue by cognitor, or to insert a praescriptio pro actore,’ but so long as there were advantages to be derived from the legis actio there could be no question of its withering away. Birks therefore returns to a modification of Wlassak’s thesis. Before the lex Aebutia was passed the praetor had created formulary

actions for civil law claims, but these were of course imperio continentia iudicia. Litigants would exploit the advantages to be derived from the choice between formula and legis actio (most obviously the possibility of bringing a

legis actio after failure in a formulary action), and this would lead to the passing of the lex Aebutia to give to the formulary actions the status of iudicia legitima. The result would be a growth in the use of formulary actions at the expense of the legis actio, but plaintiffs would continue to find sufficient advantage in the latter to ensure its survival — and to call eventually for legislation to ensure that the objectively better system alone should survive. This would have been the reason for the intervention of the leges Tuliae.

There remains the question why the reform should have been contained in two leges Iuliae. We do indeed hear of two leges Iuliae de iudiciis, but only one was concerned with iudicia privata, the other dealing on the contrary with tudicia publica (i.e. criminal trials). Since there seemed to be no reason

why the latter should’ have anything to say about the abolition of legis actiones, Wlassak * conjectured that there had been altogether three leges Iuliae, one dealing with iudicia publica and two with the iudicia privata, of which one concerned Rome itself and the other municipia and colonies. There is, however, no trace of any such lex dealing comprehensively with trials outside Rome, and it is normally assumed ® that the jurists were in the habit of referring to the two leges dealing with tudicia publica and privata as if they were a single enactment, although here in fact only one was in question. Kunkel! has recently proposed another explanation. As we shall see,? he argues that for murder and other wrongs there was a proceeding by legis actio sacramento leading to the addictio of the wrongdoer, and that this private capital proceeding, though it would have been in practice superseded by the growth of the more normal criminal jurisdiction of the quaestiones in the later republie, would have been formally abolished only by Augustus. And since it had more affinities with criminal proceedings than with a private action, its abolition would have been enacted in the lex Iulia de iudiciis publicis. 7 Cie. Lose. Com. 32; de orat. 168. In the eases in which we know that the formula was used there was a particular advantage to be gained; Kaser, St. Albertario 2.52f. 8 Prozessgesetze 1.191ff. ® Girard, Manuel 1059; SZ 34 (1913) 341ff.; Pugliese, Proc. 2.1.65f.

1 Krim, 120, 2 Below, 311. 225

Private law to the fall of the republic: procedure G. PRAETORIAN REMEDIES OTHER THAN ACTIONS. The history of procedure

during the republic must not be left without some mention of those praetorian remedies which lie outside the system of actions, but contributed hardly less than the actions and formulac to the growth of the tus honorarium. The activity of the magistrate in the ordinary procedure leading to an action is a result of that part of his function known as iurisdictio, but with the remedies we have now to discuss the position is different. Here we

have to deal with orders issued by the praetor as a holder of imperium.* These orders are however issued for the purpose of the administration of justice; the rules which the praetor adopts with respect to their issue, or at least some indications of them, appear in the edict, and, as we shall see, the praetor generally avoids using direct means of enforcing obedience, so that any dispute concerning them will very often lead to an action which has to be tried in the ordinary way. The remedies which we have to consider under this head are of four kinds: (i) Praetorian stipulations, (ii) Misstones in possessionem, (iii) In integrum

restitutiones, (iv) Interdicts.* : (i) Praetorian stipulations. In some cases the praetor * supplements what he considers to be the deficiencies of the civil law by ordering one party to 3 See above, 47 n 9.

4 That all four types of remedy existed before the end of the republic is clear, but otherwise their history raises many difficult questions, the answers to which depend in part on the view taken of the magistrate’s powers before the lex Aebutia. Some interdicts, it

is agreed, must have existed before that law, for though the reference to wtrubé in Plautus’ Stichus (acted 200 B.C.) 696, 750, is very doubtful (cf. Watson, Property 86f.), we find a form of words obviously based on that of the interdict uti possidetis in use for the settlement of a dispute between two Greek cities about the middle of the second century (above, 263 n 4). From the latter example it would also seem likely that the procedure was by this time out of the stage in which the magistrate settled everything himself (below, 232), for the decision is referred to the assembly of a third city just as in a private matter it is referred to a judge. That misstones were known to Q. Mucius Scaevola (consul 95 B.C.) is clear from D. 41.2.3.23 and, since he is there quoted as making a distinction between different kinds, they can hardly have been quite new in his time. A stipulation damni infecté is reported (Pliny, HN 36.2.2) to have been exacted as early as 123 B.C. (ef. Karlowa 2.1241), and the rather archaic form given in D. 46.8.18 also points to an early date (Lenel, HP 552 n 1). In integrum restitutio for metus certainly existed im 59 B.C. (Cie. Flacc. 49), and other cases are pretty certainly older. The rubric of the edict promising restitutio on account of absence (Hx quibus causis maiores XXV annis in integrum restitwumtur) points to a time when this was the only reason for which persons above the age of twenty-five could get restitutio, and this is indeed probably the earliest case, except that of minors, which very probably goes back to a date not long after the lex Plaetoria (below, 241). 5 The aediles, although not invested with imperium, also used their powers of coercion to insist on the making of stipulations with relation to matters within their jurisdiction. These stipulations were of great importance in the history of sale; see below, 293f. 226

The formulary system make a promise to the other, such promise being in the form of a stipulation, and giving to the promisee a right, or at least a remedy, which he would not otherwise have. Thus if A’s house is in a dangerous condition and likely to fall and damage B’s land, B can insist on A’s promising to pay damages for any harm that may be done (cautio damni infecti —i.e. damage threatened — ‘not done ’).® If without such promise having been exacted the house fell and did damage, B would have no claim against A unless he could show that A had been guilty of some wilful or negligent act such as would make him

liable under the lex Aquilia, and this would, of course, not always be the case. Again, suppose that A has been left the ownership and B the usufruct (life-interest) in an estate, A must, it is true, already at civil law, refrain from actively injuring the property, as for instance by cutting down all the timber; but he is, at civil law, under no obligation to do any positive acts to keep it in repair; also at civil law he (or his heir) is under no active obligation to hand over the estate to the owner on the cessation of the usufruct, though of course he has no right to remain on the land. The praetor, on the other hand, will insist that the usufructuary promise the owner that he will perform active duties, will look after the property as a careful man should and return it when the usufruct is at an end (cautio usufructuaria).” ®

The methods by which the praetor enforces these stipulations are not always the same. In the case of usufruct the praetor would simply refuse the usufructuary failing to make the promise the action for claiming the estate

from the owner.® In the case of damnum infectum, if the stipulation was refused, the praetor made an order entitling the person whose land was threatened to enter into possession of the dangerous building (missio in possesstionem),' and if opposition was offered to the entry and the damage was done, there was an actio ficticia, i.e. one in which the judge was told to con-

demn for as much as would have been payable had the stipulation been made.? Direct methods of constraint are not evidenced in classical times, though they may have been used.° 6 That there was a liability at civil law is clear from Gaius’ statement (4.51) that procedure by legis actio was still in principle possible. What the legis actio was we do not know (see de Zulueta, Gaius 255; Pugliese, Proc. 1.108 n 185). In any case the praetorian law here completely superseded the civil.

7 Usurum se boni viri arbitratu et, cum usus fructus ad eum pertinere desinet, restituturum quod inde exstabit ; D. 7.9.1 pr. 8’ Sometimes a2 mere promise was enough, sometimes there had to be security as well,

e.g. in the case of damnum infectum if the occupier was not the owner; D. 3.2.7 pr.; 13.1.

® D. 7.1.13 pr. If the usufructuary were in possession of the thing the owner could claim it (vindicatio) and if the usufructuary pleaded his usufruct in an exceptio the

pr. 1 Below, 228. 2 Lenel, EP 372f. 3 Wenger 242. Pignora are mentioned in J. 1.24.3, but this passage is probably not owner would have a replicatio alleging that the stipulation had not been made; D. 7.9.7

classical.

227

Private law to the fall of the republic: procedure Whatever the method used it is clear that the praetor’s insistence on the making of the stipulation in fact altered and amplified the law, and the rules of this part of the tus honorarium were, like the rest of it, to be discovered from the edict. An appendix gave the forms to be used in each case, and scattered among the earlier provisions were to be found individual edicts con-

cerning their application. Whether in any particular case a stipulation had,

according to these rules, to be made, was a question which the praetor decided himself, if necessary after hearing argument; there is no place here for reference by means of formula to a iudex.*

(ii) Missiones in possessionem. Missio is an authorisation by the praetor to

enter into possession, either of a particular thing (as with damnum infectum), or of the whole of a person’s property (as in the case of execution).° Sometimes it merely entitled the person authorised to enter and hold the property concurrently with the person to whom it belonged, sometimes it gave

full possessory rights, with authority to eject the owner and eventually become owner by usucapion. The primary object of such missiones was to put pressure on the person to whom the goods belonged, as in the case of the person who failed to ‘ defend himself properly ’ in an action,® or failed to carry

out a judgment, but, as we have seen, it might have the secondary purpose of enabling the creditor to get paid by having the goods sold if the pressure proved ineffective. It was also used where there was no one on whom pressure could be put, as where the creditors of a deceased person who leaves no heir

are put in possession of his property and allowed to proceed to bonorum venditio.? Or again it might be a purely provisional measure, as where the mother of an unborn child is put in possession of the property which the child will inherit if it is born alive, but which will go to someone else if it is not.® The rules differ very considerably in the different cases, but again we can take damnum infectum as an example. If the stipulation ordered by the praetor were refused, there was first a missio in rem which entitled the complainant to enter into the dangerous building, but not to eject the owner. If this proved of no avail a second decree would issue giving full possession,°

which, if the other party remained obdurate, would ripen into ownership after the time for usucapion had elapsed.’ Again here, as with stipulations, the praetor did not as a rule use direct force if his commands were disobeyed. In the particular case of damnum infectum, if opposition was offered to the entry of the person authorised to take possession there was, as we have seen,” a fictitious action, at which it was assumed that the desired stipulation had 4 For technical distinctions between different classes of stipulations see Buckland 437

and 728.

5 Above, 217. In some cases it might be a hereditas, not a single thing nor yet the

whole of a person’s property. 6 Above, 216f, | 7 Gai. 3.79. 8 D. 37.9; Lenel, /P 347. It was only available where the child would be a swus heres.

® D. 39.2.7 pr. 1 D.39.2.5 pr.; 12. 2 Above, 227. 228

The formulary system been made.® In other cases other means were used. Thus where the missio was the result of a judgment, and in connected cases,* there was an action for damages against anyone who prevented the persons authorised from taking possession of anything forming part of the estate.» When bonorum venditio had taken place, the emptor had an interdict to enable him to recover anything belonging to the estate which was not voluntarily surrendered to him.* An action of course involves reference to a zudex, and so, as we shall see, does an interdict, so that, where opposition was offered to the magistrate’s order, there would in fact usually be a trial apud tudicem before the matter was decided. The importance of missiones can hardly be exaggerated ; on them depended, in the last resort, the working of a system in which no trial could take place without the concurrence of the defendant, and no judgment be executed except by the sale of the whole of the judgment debtor’s estate. (iii) In integrum restitutiones. In some cases the praetor’s way of dealing with the possibility that general rules of law may have inequitable results is

to annul the result which he considers inequitable by restoring the party injured to his original position (in integrum). Thus if a man has been prevented by absence abroad in the public service from taking steps to interrupt another’s possession of his land, with the result that that other has been able to complete usucapion, the praetor will restore him to his original position by

decreeing that the usucapion is to be held not to have taken place; or if a man has been induced by threats (metus) to make a mancipation of his property, the praetor may decree that the conveyance is to be held not to have

taken place. Particularly important too is the rule that a minor (i.e. one under twenty-five years of age) may get in integrum restitutio if his inexperience has led him to enter into a transaction which turns out to be disadvantageous, even though he cannot show that the other party actually took advantage of his youth.’ The praetor does not, of course, act arbitrarily in granting this relief, and like his other remedies it is provided for by clauses in the edict. As, however, it is the praetor himself who decides in each case

whether restitutio is to be granted or not, he does, in some of the edicts, leave himself a considerable liberty; that concerning minors, in particular, reads simply: ‘ If a transaction with a person under twenty-five years of age be alleged, I will take such steps as each particular case shall call for.’ ° The actual grant by the praetor is only a necessary preliminary to which effect is given by other proceedings. These consist most commonly of a tudicium rescissorium, i.e. the person benefiting by the restitutio is allowed 3 There was also an interdict forbidding interference with the entry of one authorised to take possession; Lenel, EP 469.

6 Gai. 4.145. 7 Cf. below, 241f.

4 T.e. those called rei servandae causa; see Buckland 724. 5 Lenel, HP 424.

8 Quod cum minore quam vigints quinque annis natu gestum esse dicetur, uti quaeque res erit, animadvertam; Lenel, EP 116. 229

Private law to the fall of the republic: procedure to bring the action which would have been open to him but for the event whose effects have been rescinded by the praetor.* Thus, where land has been usucaped owing to absence, the original owner brings a vindicatio with the fiction that the usucapion has not taken place, and the formula reads something like this: ‘ If the land in question had not been usucaped by Numerius Negidius, then if that land ought, by Quiritarian law, to be the property of Aulus Agerius, etc.’.?

(iv) Interdicts. Literally interdictum means a prohibition, but the word came to be used for all praetorian orders of a certain class, whether positive or negative in form.” These orders are issued by the praetor, not on his own initiative, but on the application of some person who either considers himself aggrieved or thinks that some public interest is in danger :* they are in a stereotyped form which is set out for each sort of case in the edict,* and lead, where there is any opposition, to a trial before a tudex or recuperatores. Suppose, for instance, that A has let B occupy a farm precario, i.e. on condition that B returns it at any moment that A chooses to ask for it back,* and that when A does so ask for it, B refuses to give it up. A goes to the praetor, and the praetor issues an order to B in the following form: ‘ What you hold precario from A, or by your own wilful wrongdoing have ceased so to hold, that you are to return to him.’ ® Now this order is carefully framed so as to leave open the question whether B really does hold the farm precario from A or

not, and is equivalent to an order ‘if you hold precario . . . you are to return *. Of course if B recognises that the praetor’s order does apply to him and gives up the farm, the matter is at an end, but if he does not, the question has now to be decided whether, in not returning the farm, B has been guilty of disobeying the praetor’s order or not. Only if he does so hold precario from A has he been guilty, and this is what, in effect, the tudex will have to decide. Two methods of procedure are possible, one, no doubt the older, per sponsionem, and the other per formulam arbitrariam. ®° For other possibilities see Buckland 723.

1 The exact form of the fiction is uncertain; see Lenel, #P 123. For another case of iudicium rescissorium ef, Gai. 3.84.

2 Gai. 4.140 says that decretum was the proper word when something was ordered to be done, interdictum when something was forbidden. 3 Certain interdicts were popularia, i.e. could be brought by anyone whether personally affected or not, e.g. ne quid in loco publico vel itinere fiat for the prevention of unauthorised building on public land; see D. 43.8.2.2.

4 In cases where the form set out did not exactly fit, an interdictum utile might be granted; cf. actio utilis, above, 210 n 5. 5 Precarium is distinguished from loan (commodatum) by the intentional absence of

any contractual relation between the parties; neither has any action against the other arising out of the relationship. It probably originated in grants at will of land by great

land-owners to their ‘ clients’.

¢€ Quod. precario ab illo habes aut dolo malo fecisti ut desineres habere, qua de re agitur, id illi restituas; Lenel, HP 486. 230

The formulary system (a) Per sponsionem. The parties by means of stipulations (sponsiones) make a sort of bet. B promises to pay A so-and-so much ’ if he has disobeyed

the interdict, and A promises to pay B a similar amount if B has not disobeyed it. A and B then bring actions against each other in the ordinary way ® for these sums; the question to be decided in the two actions is of course exactly the same and they are sent to the same tudex to try. The iudex, in order to decide whether there has been disobedience or not, must go into all the facts of the case, and in particular must find out whether the land claimed was held precario by B from A or not, for, if it was, B has disobeyed the interdict, if it was not, although he has disregarded it, he has not disobeyed it.°® If the twdex finds that B has disobeyed, then he must condemn B in the first action, and absolve A in the second; if he finds that B has not disobeyed he must absolve B in the first action and condemn A in the second. The loser thus pays a penalty in any event, but, if B is the loser, the mere decision on the stipulations does not give A his land back. There is therefore in this case a third action, also sent to the same zudex (but necessitating no further trial because the point at issue has already been decided), in which the iudex is instructed to condemn B to pay the value of the land unless he restores it to A (iudicium secutorium).! Here again, as we have seen with actions in rem,* there is no specific restitution; in the last resort the iudex can only award damages, but the defendant is given the chance of escaping condemnation by making restoration. (b) Per formulam arbitrariam. In certain classes of interdicts it is possible

for the defendant to avoid the risk involved in the sponsiones by asking, before he leaves the presence of the praetor, for the appointment of an arbiter. A formula arbitraria* will then issue at once and the case will be tried on that; the result will then be exactly the same as it would be under the other procedure except that the defendant will not, if he loses, have to 7 How great the amount was in classical times is unknown. Originally it was probably the full value of the plaintiff’s interest in the matter estimated by him on oath. At a time when the iudiciwm secutoriwm (see below, n 1) did not yet exist this would have been the only way in which the plaintiff could get full satisfaction. With the invention of the tudicium secutorium, this reason no longer applied, and the full value appears to have been merely a maximum beyond which the praetor would not allow the parties to go. Cf. Lenel, EP 471; Buckland 739 n 6. 8 In classical times these would be actiones certae creditae pecuniae; under the legis

actio system condictio or sacramentum would have been available. 8 Cf. Buckland 737.

1 Gai. 4.165, 167. The exact wording of the formula is not known, but at any rate it contained a clausula arbitraria (cf. above, 213f.) where the interdict was exhibitory or restitutory. In the case of prohibitory interdict there is more difficulty; see Lenel, EP 450f. The tudicium secutorium is presumably later than the sponsiones, for it could not exist under the legis actio system. Cf. above, n 7, and Lenel in Festgabe fiir R. Sohm

(Munich/Leipzig, 1914) 207. 2 Above, 214. 5 The formula is here quite uncertain (see Lenel, EP 449; Kaser, ZPR 329 n 42), but

the effect was that the defendant either had to put matters right or pay damages. 231

J—9

sn Ee Private law to the fall of the republic: procedure

pay the penalty, and that the plaintiff will similarly be free from penalty if the case is decided in favour of the defendant.

Such, in the barest outline, is the procedure on this very important and

common type of praetorian remedy. Interdicts served all manner of purposes. Some were no more than cogs in the procedural wheel,* some were devised to safeguard public rights;*° others were the vehicles of immensely important

praetorian innovations and amplifications of the civil law, as in particular

the interdict quorum bonorum, by means of which the praetor carried through a great part of his reform in the law of succession.® Of even greater significance, if we consider not merely Roman law but the history of law in general, were the interdicts by which the praetor protected possession. This was a matter outside the civil law, but, as we shall see,’ the praetor made it a rule that an existing possession, whether rightful or wrongful, must not be disturbed except by making a proper claim in a court of law. If I am in pos-

session of land belonging to you and you take it from me, the praetor, by means of an interdict, will force you to return it to me quite regardless of your ownership, and you will not be able to get it again by the proper action, the vindicatio, which is open to owners who are out of possession of their things. The interdicts are, in fact, the most far-reaching of magisterial remedies and a very considerable appendix to the edict was needed to include all of them; they are also that form of procedure which most clearly shows the magistrate not merely as an intermediary who helps the parties to come to an

issue, but as a superior who can command.* It is true that this command leads in historical times to a trial, but it is highly probable that there was originally a personal investigation of the facts by the praetor before he issued

the command,’ and that this system only had to be given up because of an intolerable pressure of business on the single judicial magistrate. It should be noticed in this connexion that interdicts have to a certain extent a ‘ police ’ character; many are concerned with public ways and rivers; a number, such as those regulating possession, are provisional — the peace must be kept until the question at issue is settled in a lawful way; one protects the tenant who

is prevented by his landlord from moving out with his goods,’ and others regulate disputes between neighbours.” All are further characterised by comparative rapidity of procedure; the praetor can issue them on dies nefasti and they can be tried outside term time.*

4 Ii.g. interdicta secundaria, Gai. 4.170. 5 Above, 230 n 3.

6 Below, 253. 7 Below, 259ff. 8 Gaius begins his account of interdicts (4.139) by saying: ‘In certain cases the

praetor or proconsul uses his authority directly for the settlement of controversies’ (principaliter auctoritatem suam ...interponit). 9 Which would then naturally not be in the hypothetical form subsequently used. 1 De migrando; Lenel, HP 490.

2 E.g. De glande legenda; Lenel, EP 487. 3 Cf. Wenger 75, 246, 254. 232

CHAPTER 14

Private law from the XII Tables to the fall of the republic: the law of the family and of succession 1. MARRIAGE A. Berrorua..’ Marriage at Rome, as elsewhere, was preceded normally by engagement. This, in early times, took the form of reciprocal promises ” (sponsiones) between the intending husband or his father and the father or guardian of the woman. The consent of the woman herself was almost certainly not needed * if she was in her father’s power, and perhaps not even that of a son under power. The contract appears to have been enforceable originally in the sense that if either party broke off the engagement an action lay against him for damages, but it was certainly not enforceable in classical times, when all restrictions on the freedom of the parties were regarded as

improper, and it had evidently ceased to be so long before the end of the republic.* This change is presumably reflected in the replacement of the sponstones by the mere informal agreement which we find in classical times.* For breaking off the engagement a mere formless notice (renuntiatio) was suffi-

cient. Even in classical times this could be sent by a father without his daughter’s assent, if she was in his power,® and probably at one time the same had been true in the case of sons. 1 Kupiszewski, SZ 77 (1960) 125ff. (early law) and 84 (1967) 7Off. (classical law) ; Watson, Persons 11ff. 2 D. 23.1.2.

3 As it was in the classical law; D. 23.1.7.1, 11-13. Contra, Solazzi, Seritti 3.1ff. (=BIDR 34 (1925) 1ff.). 4 Gellius, 4.4, quoting Veratius’ account of a statement made by Servius Sulpicius, says that sponsalia had been enforceable in the law of Latium until the lex Iulia (90 B.C.), when all the Latins received the citizenship. This shows pretty clearly that they were no longer enforceable in Roman law at this period, but original Roman law can hardly have differed from the allied systems of Latium in so important a particular, though it is easily comprehensible that it should have developed the more liberal principle earlier. That the agreement was not enforceable in classical times does not mean that it.was without legal importance; the engagement created an affinity in some respects similar to that resulting from marriage. Thus a prospective father-in-law could not be forced to give evidence against his daughter’s fiancé (D. 22.5.5) and gifts which would be invalid under the lex Cincia between unrelated persons were permissible between betrothed persons (FV 302). On the whole question see Corbett, Marriage 16f.

5 D. 23.1.4. 6 D. 23.1.10. 233

Private law to the fall of the republic: family and succession B. MarriacGe. We have seen that at the time of the XII Tables marriage must ordinarily have been accompanied by manus.’ By the end of the republic manus, though it still existed, was the exception. The ordinary married

woman therefore was not in the power of her husband and did not even become a member of his (agnatic) family. If her father was still alive she remained under his potestas; if he was dead she was sui iuris and had a tutor appointed in the same way as if she had remained single. To her children she was not related according to the civil law, for they were in the patria potestas of her husband. As we have seen, no ceremony was required for the contracting of a marriage; all that was necessary was the consent of the parties, and, if they were under power, of their patres,* together with the de facto beginning of conjugal life. It was usual for this to take place by a ceremonial bringing of the bride

to the bridegroom’s house, where she was lifted over the threshold and received from the bridegroom the symbolical gifts of fire and water.® But this was not legally necessary; any other actual beginning of life as husband and

wife was sufficient. Thus marriage might be contracted by the bringing of the bride to the house of the bridegroom in his absence, provided he signified his consent by letter or message,’ or the couple might take up their residence together in the bride’s house.” The setting up of a common household is, of course, compatible not only with marriage but with concubinage,’ but, as already noticed,* it would usually be quite clear from the attendant circumstances when marriage was intended,® and there was further a presumption in favour of marriage when the parties were of equal social standing.° 7 Above, 115.

8 In the case of 4 woman sw iuris the auctoritas of the tutor was not necessary, as it was for coemptio, but it appears from later legislation that he and the relatives had some say in the matter; Corbett, Marriage 60ff. Solazzi, op. cit. above, n 3, tries to show that even in classical times the consent of a filia familias herself is not needed, but it is difficult to see how consent (of a sort) could be dispensed with in a system which regards marriage as a de facto matter, unless the girl is carried struggling into the husband’s house. The question is quite different from that of betrothal. Cf. Watson, Persons 41ff.

9 Symbolical apparently of her share in the most necessary elements of human life; Festus, s.v. aqua et igni, Bruns 2.3. 1 DQ, 23.2.5.

2 Levy, Ehescheidung 69. If, as is said in D. 23.2.5, the woman cannot, in the same way as the man, be married in her absence, it is because her entry into the husband’s house is the normal thing, whereas his entry into hers in her absence would not be equally unequivocal evidence of the beginning of matrimonial life. 3 Nicholas, Introduction 83f. 4 Above, 114.

5 Preceding betrothal, festivities, and, in particular, the drawing up of instrumenta

marriage. 6 D. 23.2.24; 39.5.31 pr. dotalia, documents concerning the dowry, which could only exist where there was lawful

234

Marriage There were, of course, certain fundamental conditions for the validity of the marriage; the parties must have conubium, be above the age of puberty, of sound mind and not related within the prohibited degrees. With respect to this last point the rules became progressively less strict; thus marriages between first cousins, originally forbidden, were known already towards the end of the third century B.C.’ Marriage as such (i.e. apart from manus) can scarcely be said to create a legal relationship between husband and wife. Where the marriage is dissoluble at will by either party it is idle to impose a duty for instance on the husband to support the wife, and such a duty did not in fact exist. There were, however, some mutual disabilities.* For example, so long as the mar-

riage lasted, neither could bring against the other a penal action, or one which entailed infamia for the unsuccessful defendant; most importantly perhaps, gifts between husband and wife were invalid.°

C. Divorce. Where marriage was without manus divorce was entirely free to either party. As marriage began by the de facto beginning of life as husband and wife, so it ended if such community of life was broken by the action of either spouse. This does not, of course, mean that mere separation constituted a divorce any more than mere living together constituted marriage; it must be the intention of one of the parties to treat the separation as putting an end to the marriage and such intention must, normally at any rate, be made clear to the other spouse.’ The husband might send his wife from the house, using perhaps the traditional formulae,” or the wife might leave it, declaring that she considered the marriage at an end. No particular form was laid down, but as such declarations are usually only made when personal intercourse between the spouses has become painful, it was usual for a written or oral message to 7 Livy 20 (fragments published in Hermes 4 (1870) 372), quoted by Karlowa 2.175; cf. Tac. Ann. 12.6. 8 Corbett, Marriage 125f. ® The object of this rule is said to be that the spouses should not be led by affection to deprive themselves of their property (D. 24.1.1), and that no material considerations

should enter into married life (D. 24.1.3 pr.). The text of D. 24.1.1 says that the rule is one introduced by custom. On the other hand, it appears to be subsequent to the lex Cincia (204 B.C.; above, 87) since husband and wife were ‘ excepted persons’ under that statute (FV 302), and it is difficult to see how such a rule could be introduced at so late a date merely by custom (cf. below, 353). Some have thought that D. 24.1.1 is interpolated and that the rule was introduced by the marriage legislation of Augustus, but this is difficult to maintain; Watson, Property 229f. For another suggestion see Kaser, EPR 1.331; A. Sdllner, Actio ret uxoriae (Cologne, 1969) 127ff. 1 Levy, Ehescheidung 84. Where one spouse has been missing for so long that there is

no reasonable likelihood of being able to re-establish communication with him, the marriage appears to be ipso facto at an end, so that the other spouse can marry again without sending any repudium; see Levy, Ges. Schr. 2.47ff. (=Geddchtnisschr. E. Seckel, Berlin, 1927, 147ff.). | 2 Above, 117. 235

Private law to the fall of the republic: family and succession be sent, and repudium or nuntium mittere became a common expression for divorcing. ) In the less usual case of marriage with manus, the manus could still only

be broken by a re-mancipation undertaken by the husband,* and this he could no doubt undertake when it pleased him ;* in the classical law a woman married with manus could apparently also insist on being freed; she repudiated her husband in the same way as a woman married without manus, and could then insist on re-mancipation ;° but how long this had been the rule is unknown. By the end of the republic divorce had become extremely frequent, at any rate in the high society of the capital, of which alone we have any real knowledge, and this was a state of affairs that Augustus set himself, without much success, to remedy.®

D. Dowry. Dowry (dos) is always a gift to the husband coming from the side of the wife and intended as a contribution to the expenses of married life, which fall upon the husband. So long as marriage with manus was the rule, the property of a woman sui iuris necessarily passed to her husband on her marriage, and in the case of a daughter under power it was usual for the father to provide for his daughter by giving a dowry to her husband; to the daughter herself he could, of course, not give it. It was also possible for persons other than the father who were interested in the wife to provide for her

in the same way. When the marriage was without manus it was perfectly possible (if the woman was sui iuris) for her to own property and to have property given her, but it nevertheless remained usual for a dowry to be given to her husband on the marriage. If this was given by her father it was called dos profecticia, if by the woman herself or any other person dos adven-

ticia.” Neither where the marriage was with manus, nor where it was not, does there appear to have been originally any legal duty on the husband or his heirs to return the dowry when the marriage came to an end. Where there was manus the wife, if she survived her husband, shared with the children on 3 Ie. where the manus resulted from coemptio or usus. For divorce in confarreate marriage see above, 117.

4 For the difficulties which would arise if the wife refused to be present for the re-mancipation ceremony see Levy, Ehescheidung 41. 5 Gai. 1.137a. The text is defective, but see Levy, Ehescheidung 40 n 1.

6 For the tradition that Sp. Carvilius Ruga was the first man ever to divorce his wife see above, 118 n 8.

7 It was not necessary that property should be transferred at once; it might be promised. Anyone could bind himself to give a dowry by the ordinary contract of stipulation, but certain persons could bind themselves by a formal oral promise (dictio dotis) even though it was not (as was necessary in a stipulation) preceded by a question, and consequently without necessarily meeting the promisee. The persons to whom this rule applied were the woman herself, her debtor acting with her authorisation and her father or other male ascendant related solely through males (Epit. Ulp. 6.2); see Daube, Jurid. Rev. 51 (1939) 11ff., and below, 279 n 6.

236

Marriage intestacy, or might be provided for by will; if the wife predeceased the husband the dowry remained with him as part of the family property which, in the normal course of events, he would transmit to his children. Where there was no manus the position was still satisfactory; the wife could be provided for by will if her husband predeceased her, and to give her a claim to the return of the dowry would have meant depriving the children of the marriage of all hope of succession since, at this period, a woman could make no will, and her property would go on her death, not to her children (who were in the agnatic family of their father), but to her own agnates. In the rare cases of divorce (always at the instance of the husband) it was customary, as we have seen,® for the husband to sit in judgment on the wife and either give her back part of her dowry, if the fault were a minor one, or refuse to return any of it if she had been guilty of serious misconduct, but this was a matter of custom, not of law.® The change must have come when divorce became more frequent, especially divorce where no fault was imputed to the wife, or at her instance. If in such cases the husband did not choose to return the dowry a woman still of marriageable age might be left without provision and thus find

difficulty in contracting another marriage. We find consequently, in the developed law, quite a different state of affairs. In many cases the person giving the dowry would actually make a stipulation for its return on the dissolution of the marriage,’ but even where this was not done, the wife had a special action (actio ret uxoriae) * for the return of her dowry or part thereof to herself, when the marriage was dissolved by divorce or predecease of the husband; only if the marriage was dissolved by her death did the husband retain the dowry.*® This action was almost certainly in bonum et aequum concepta, and the husband could make deductions in some cases, as, in particular, in case of divorce occasioned by the wife, for her misconduct (propter mores), and for children, one-sixth being allowed for each child up to three. The origin of this action is obscure, but the most probable view appears to be that it originated in agreements for the return of an equitable portion of the dowry in case of divorce, which it became the custom to make at the time of the marriage when the increasing frequency of divorce made this a con-

tingency which had to be taken into account. Later an action would be granted even if the parties had not taken this precaution. The agreement 8 Above, 118.

® Speech of Cato reported in Gell. 10.23: Vir, cwm divortium fecit, multeri iudex pro

censore est; imperium quod videtur habet. Si quid perverse taetreque factum est a multere, multatur; si vinum bibit, si cwm alieno viro probri quid fecit, condemnatur. For discussion see Volterra, Riv. it. sci. giur. (1948) 114 ff. 1 In such case the dowry was said to be recepticia; Epit. Ulp. 6.5. 2 Watson, Persons 67ff.; Sdllner, op. cit. (above, 235 n 9). 3 Even then he did not retain it if it was profecticia and the giver himself was still alive; he had however the right to retain one-fifth for each child; Mpit. Up. 6.4. 237

Private law to the fall of the republic: family and succession cannot have been by stipulation for it was probably not possible before the late republic to make a stipulation for an indefinite sum such as an ‘ equitable portion ’,* which is precisely what, according to our only authority,° these agreements did provide for.® It is thought, however, that the action may have originated in the praetor’s enforcing an informal pact made with this object.” In any case the liability of the husband to return the dowry goes back well into the republic. It is related * that in 160 B.C. the heirs of L. Aemilius Paulus had to pay back the dowry to his wife and were hardly able to find the money by selling the whole of his movable property, while in 121 B.C. the claim of C. Gracchus’ widow was held justified by Q. Mucius on the ground that the riot in which Gracchus was killed had been caused by his own fault.° Both these cases refer to claims where the marriage was dissolved by death, and the rule requiring return of the dowry on divorce was probably earlier.’

2 PATRIA POTESTAS AND BONDAGE The milder manners of the later republic did not bring about any change in the legally despotic power of the father over his children; in 63 B.C. it was still possible for A. Fulvius to be executed by his father’s orders for having been concerned in the Catilinarian conspiracy.” Mancipation of children still existed, not only as a device for achieving the objects of emancipation and adoption,® but as a reality in the case of noxal surrender,** and was probably still occasionally used as a way of setting children to work where the father could not usefully employ them himself.** But the status of the bondsman had probably already changed considerably for the better, for Gaius records that

such a person could insist on obtaining his freedom censu, even against the

46 AlsoBelow, 280. 5 Boethius, below, n 1. where marriage was with manus a stipulation would be useless to the wife. 7 Mitteis, BPR 53. 8 Polyb. 18.35; 31.22; Plut. dem. Paul. 4; Livy, Epit. 46. 9 D. 24.3.66 pr.; ef. Daube, St. Biondi 1.199ff.

1 Another view as to the actio ret uxoriae is that of Esmein (RHD (1893) 145ff.), who thinks that it originated as a penal action against the husband who divorced without good cause. The chief objection to this view would appear to be that the words quod melius aequius erit, which were certainly part of the formula of the actto rei uxoriae (Lenel, HP 304), are given as part of the cautio rei uxoriae by Boethius (ad Cic. Top.

65), which would be strange if the action had no connexion with the cautio. In any case some features of the action, in particular the fixed fractions which may be retained

in certain cases, appear to be due to specific legislation probably under Augustus; Lenel, HP 304 n 5. For Kaser, RP# 1.337, the action originated in the need to provide

2 Sallust Cat. 39.5. 3 Above, 119f.

for the wife in manu, who on divorce would be without means. 4* Above, 173. It still existed in classical times; Gai. 1.141.

5* It is maintained by S. Cugia, Profilo del tirocinio industriale (Naples, 1921) that mancipation was used in order to apprentice a son to a trade; see SZ 47 (1927) 531. 238

||

Guardianship will of his master,® and this is a rule which can hardly have arisen during the

empire, when the census was very rarely taken. So far as proprietary relationships are concerned the old rule that all acquisitions go to the father also remained untouched, though it became common for sons, like slaves, to have peculia. These would be all the more necessary as civilisation became more urban and the son no longer usually worked with his father on the land. In the plays of Plautus they appear already as a well-recognised institution, so that they were certainly not new at that time. Action against the father on the son’s contract could be brought in the same cases as against the master on the contract of his slave.’

8 GUARDIANSHIP A. Tuteta. The XII Tables had dealt only with tutors appointed by will or right of kinship. At some time previous to 186 B.C. was passed a lex Atilia,*® which laid down that where an incapable person sui zuris was without a tutor the praetor urbanus should appoint one with the co-operation of the majority

of the tribunes of the plebs. It may well be that the praetor had had the right of nomination before, and that the lex Atilia really restricted his power by bringing the tribunes into the matter,® but even so the increasing attention paid to the magisterial or ‘ dative ’ guardianship shows how the idea of tutela as a public duty was superseding the early conception of it as an advantage to the guardian. More important however in this respect was the introduction of the actio tutelae, a bonae fidei action, in which the tutor could be called upon to account for any loss that the ward had suffered by his fault (intentional or, at least in the later classical law,} negligent) in administering the estate. This action existed at any rate at the time of Q. Mucius Scaevola,’ and shows a great advance on the earlier law according to which the tutor had only been liable if he actually embezzled the property of the ward.* This 6 Except where he had been noxally surrendered, and in the cases where there was a trust (fiducia) for re-mancipation to the father; Gai. 1.140. In 1.141 Gaius says that an actio iniwriarum will lie for dishonouring treatment of a person in mancipio, but it is

impossible to say how far back this rule goes. 7 Below, 256ff.

8 Gai. 1.185. The date must be earlier than 186 B.C., for the woman whose informa-

tion led to the discovery of the Bacchanalian conspiracy of that year had a tutor appointed by the praetor and tribunes; Livy 39.9. Gaius, loc. cit., mentions that a similar

power of appointment was given in the provinces to the governors by a lex Iulia et Titia; if this was a single law it must have been passed in 31 B.C., but there were more probably two separate laws; see Kaser, RPF 1.357 n 47. 9 Mitteis, RPR 41 n 4. The need for nomination would not often arise until the tutela of the gentiles became obsolete. 1 For liability for negligence in the republic see Watson, Persons 141f. 2 It appears in the list of b.f. tudicia quoted by Cicero, Off. 3.70, from Scaevola. 3 Te. by the actio rationibus distrahendis; above, 172. 289

Private law to the fall of the republic: family and succession action however only lay in the case of guardianship over boys and girls; in that of adult women no such action was ever developed, because the position of the tutor was here quite different. He did not administer the woman’s fortune at all; she administered it herself, and only needed the tutor’s authori-

sation for certain classes of transactions.* In fact, although the lifelong guardianship of women remained, it was only as a burdensome technicality left over from an earlier stage of civilisation. Tutors, other than those holding tutela legitioma as parents or patrons,’ could be forced, if necessary, to give their authorisation to transactions which the woman desired to carry out, and a method was devised whereby a woman could get a different guardian in place of cone whom she disliked. For this purpose a coemptio (called coemptio fiduciae causa ® to distinguish it from one entered into on marriage) was used. The woman made a coemptio with some man — it did not matter with whom — and thus fell under his manus; this man then mancipated her to the person whom she wished to have as her tutor, and he (who now held her in mancipio) manumitted her, thereby becoming her tutor in the same way as a man who manumitted a female slave became her tutor.’ The original coemptio needed, it is true, the authorisation of the tutor, but this could, as we have seen, generally be enforced. It was also possible for a husband when appointing by will a tutor for his wife in manu to leave her to choose the individual, and he might even give her the right to choose a new one as often as she liked.* In such a case it is clear that the burden was not a serious one at all. B. Cura. (i) Furiosi and prodigiz. Here too there was a development, in some respects obscure, of magisterial appointment, in addition to the agnatic and

gentilician rights recognised by the XII Tables. Both in the case of the lunatic and in that of the spendthrift the praetor came to have the right to appoint in default of agnates, and though there was never any strictly legal right to appoint by will, it became the rule that the magistrate should confirm an appointment so made by the father of the incapable person.°® It seems also

clear that the interdiction of the spendthrift, originally confined to inherited property and probably only to that inherited on intestacy,’ could, in later law, extend to all property however acquired, but in this matter, as in many

4 Gai. 1.1908. 5 Gai. 1.192.

6 J.e. ‘for the purpose of a trust’, the trust being one for manumission imposed by the coemptionator on the person to whom he mancipated; Buckland 120. 7 The tutela was however fiduciaria whereas that of the patron was legitima; Gai. 1.166a. The use of coemptio for the purpose of changing tutors is not actually attested

before Gaius (1.115), but the mention of coemptio for the purpose of getting rid of sacra by Cicero (Mur. 27) immediately after that of the powerless tutor suggests that it was already known in his time. 8 Gai. 1.150ff. 9 Not attested before classical times; D. 27.10.16 pr. 1 Above, 122.

240

Guardianship others, it is impossible to give even approximate dates.? At no time does there appear to have been any lex granting the power of appointment to the magistrate; it seems to have been regarded as inherent in the imperium. (11) Minoris. More significant of the change in the conception of guardianship was the rise of an entirely new kind of cura, that over young men above the age of puberty but under twenty-five (minores). This sort of guardianship was never anything but an institution for the benefit of the ward, and curators of this sort were never appointed except by the magistrate. The first step away from the old system whereby a youth was considered fully adult as soon

as he had reached the age of puberty was taken apparently by a lea Plaetoria,* probably passed about 200 B.C.* This statute permitted an action to be brought against anyone taking advantage of the inexperience of a young man under twenty-five (circumscriptio adulescentium). The action was probably penal and popularis*® (i.e. it could be brought by any member of the public), and the law evidently had the effect of making people chary of dealing with minors, for in a play of Plautus we find a young man complaining that the law about being twenty-five years old will be the ruin of him, since everyone is afraid to lend him money.® Transactions with minors became still more dangerous when the praetor began to grant them in integrum restttutio,’ not only if advantage had been taken of them, but even if, without any fraud on the side of the other party, they had been led by their inexperience into an unprofitable transaction — had, for instance, bought something for more than it was worth. The dangers which persons transacting business with minors thus ran could however be minimised if, before concluding the

transaction, the minor received the consent of some experienced adult appointed by the praetor to consider whether it would be advantageous. Perhaps the consent of such a curator did not exclude an action under the lex Plaetoria, and it certainly did not exclude an application for in integrum restitutio,® but it would make the success of either much less probable, and thus anyone doing business of importance with a minor would generally insist on a curator’s consent. Whether the lex Plaetoria itself provided for the nomination of curators is uncertain,® but at any rate the practice became common. 2 The interdiction of a prodigal who had received property under the will of: his father is attested for the year 92 B.C. by Valerius Maximus, 3.5.2; see Collinet and

de Visscher, opp. citt. above, 122 n 4. 3 Or Laetoria; Schulz, CRL 191. 4 There is a clear reference to it in Plautus’ Pseudolus (acted 192 B.C.) 303, and

perhaps in his Rudens (acted about the same time) 1376; ef. Watson, Persons 157f.

5 It is sometimes said, mainly on the strength of Tab. Heracl. 112 (Bruns 1.108; FIRA 1.149) that there were two actions, one criminal (iudiciwm publicum), and the other private, for recovery of the amount lost or a multiple thereof; but see Duquesne,

7 Above, 229. 8 C, 2.24.2.

Mél. Cornil 1.217ff.; Schulz, CRL 192; Kunkel, Krim. 53. 6 Pseud. loc. cit. ® The only text which speaks of them in connexion with the lez is H.A., Vita Marct 10.12: de curatoribus vero, cwm ante non nisi ex lege Laetoria, vel propter lasciviam 241

Private law to the fall of the republic: family and succession

4, SUCCESSION? A. TEsTAMENTARY SuccESSION. The chief feature here is the development of

the mancipatory will and the final disappearance of the earlier forms, but something must also be said of the so-called * praetorian will’, that is to say of the recognition by praetorian law for certain purposes of documents which might not fulfil the civil law requirements for a valid will. (i) Disappearance of the early forms of will. That the wills comitzis calatis and in procinctu had become obsolete by the time of Cicero is clear, but no precise date can be given. Cicero states definitely that wills in procinctu were no longer made,? and they are last recorded as occurring in 148 B.C.* For the comitial will we have no evidence.‘

(ii) The mancipatory will (testamentum per aes et libram).> We have a clear description of the manner in which wills of this sort were made in classical times from Gaius,® and as the previous history of the institution is almost completely a matter of conjecture it will be as well to begin with this description, and discuss the theories of its original nature afterwards. According to Gaius the testator first writes out his will, which must, at this period, contain the nomination of an heir,’ on wax tablets, then he gets together the usual personnel of a mancipation, the five witnesses and the balance-holder,

and a man to act as familiae emptor (* purchaser of the estate’). He then goes through a formal ceremony of mancipating his estate to the familiae emptor, such ceremony being exactly like any other mancipation, except that the transferee (the familiae emptor), instead of speaking the ordinary words,°®

makes use of a more complicated formula: ‘I say that your familia and pecunia are at your orders and in my custody (endo mandatela tua custodelaque mea), and be they bought to me with this piece of bronze so that you may make a will in accordance with lex publica.’ ° Then follows somevel propter dementiam darentur, ita statuit ut omnes adulti curatores acciperent non redditis causis. This may mean that before Marcus curators were only appointed for special reasons. Schulz, CRL 193, argues that they were until then only appointed in cases of prodigality or insanity. Since the question whether the minor had been overreached was one of fact and the approval of the curator therefore only evidentiary, an official appointment was not logically necessary; Buckland 170.

1 See now Watson, Succession. 2 Nat. deor. 2.9.

8 Vell. Pat. 2.5. The mention in Cicero, de Or. 1.228, hardly helps as to date. 4 Labeo (quoted by Gell. 15.27) is the earliest authority to mention it at all, and he speaks of it in the past tense, but we can be fairly sure that if it had existed in Cicero’s

time some mention would have survived.

7 Above, 123, 8 Above, 144.

5 Cf. above, 127f., and see especially de Zulueta, Gaius 87ff. 6 2.104,

8 Familiam pecuniamque tuam endo mandatela tua (MS tuam) custodelaque mea esse aio, eaque (the words esse aio, eaque are not in the MS) quo tu iure testamentum facere 242

Succession

thing quite different from the proceedings in an ordinary mancipation. Whereas in an ordinary mancipation the transferor is silent,’ here the testator takes an active part in the proceedings. He takes the prepared tablets in

his hand and says: ‘ As it is written on these tablets and wax, so I give, bequeath and declare, and so do you, Quirites, bear me witness.’? This speech, says Gaius, is called the nuncupatio, and by it the testator confirms generally the specific directions given in the tablets.*

Now in classical times the whole ceremony is merely a legally necessary form, not a real conveyance. It does not matter who the familiae emptor is, for he takes no rights and comes under no duties by the transaction.* No effect whatever is produced until the testator dies, and then the person who succeeds to the rights and duties is the heir nominated in the will, who is also lable for the legacies, if any. Further the testator, as long as he lives, can always entirely invalidate the whole proceeding by making another will. Clearly this had not always been the position. As the choice of the form of mancipation and the whole nature of the proceeding shows, the transaction was originally a conveyance and provides an example of the rule that testamentary forms commonly originate in bilateral transactions inter vivos.° Gaius himself clearly distinguishes an earlier state of affairs in which the familiae emptor ‘ took the place of an heir and for that reason the testator gave him instructions as to what things he wished given to each person ’.° Gaius further” represents the mancipatory will as originating in a device adopted by persons who were in danger of death and could not wait for the next opportunity of making a will in the normal way before the comitia. In such an emergency the ‘ testator ’ conveys his property to a friend, who will presumably outlive him, and asks the friend to make such dispositions as he himself would have made had he been able to make a comitial will. This friend (the familiae emptor) has then, like an English executor, to dispose of the property in accordance with the ‘ testator’s ’ instructions. The instructions were no doubt originally oral, and there was no mention of any heir, the nearest thing to an heir being the familiae emptor himself. possis secundum legem publicam hoc aere esto mihi empta. Some people add, says Gaius, after hoc aere the words aeneaque libra. And see below, 244 n 6. 1 Cf. above, 148.

2 Haec ita ut in his tabulis cerisque scripta sunt, ita do, ita lego, ita testor (MS

testator), itaque vos, Quirites, testimontium miht perhibetote. 3 Sealing by witnesses, the libripens and the familiae emptor for the purpose of subsequent identification of the document was usual, and may have been legally necessary, for the edict concerning bonorwm possessio secundum tabulas in Cicero’s time spoke of seals

‘not less in number than the law requires’ (cf. below, 248 n 2), but neither Gaius nor Justinian (Inst. 2.10.2) mentions this requirement. 4 ‘He is brought in’, says Gaius, 2.103, ‘for form’s sake (dicis gratia) in imitation of the ancient law.’ 5 See Pollock’s notes in his edition of Maine, Ancient Law, ch. 6.

6 2.103; ef. also 105. 7 2.102. 243

Private law to the fall of the republic: family and succession This much is clear, but there are many points of difficulty and dispute, among them the following: (a) What was the significance of the mention of lex publica in the formula spoken by the familiae emptor? If, as is usually supposed, the reference is to the XII Tables,* the only clause known to us which could possibly be in point is one already mentioned,’ the essential words being those given by Gaius:*

uti legassit suae rei, ita wus esto. Other versions ? are fuller and differ as to what may be ‘ legated ’ — familia, pecunia, tutela — but the primary difficulty

is to see how this clause could justify the testamentum per aes et libram as we know it, i.e. an act involving the appointment of a heres. If legare has its classical meaning of ‘ to make a legacy ’, i.e. to charge a gift on a testamentary heres, the clause can only refer to the comitial will, and since the legacy would then derive its validity from the act of the comitia, it is difficult to see the need for a special provision in the XII Tables. It is accordingly a tempting conjecture * that the word should be given the wider meaning of ‘ to

appoint to a duty’ or ‘to commission’. For if one further makes the not uncommon assumption * that what the XII Tables allowed to be the subjectmatter of this commission was only pecunia, in the sense of that part of the

property of a paterfamilias which he was free to alienate (as opposed to familia which he was not), then what the clause originally envisaged was a conveyance (mortis causa and subject to the commission) of this limited

class of property. Only later, when the distinction between familia and pecunia had gone, would an abusive interpretation have allowed the clause to be used to justify the making of a full testamentum. Whether or not this conjecture is right, some historical development must be postulated if we are to account at all for the transformation of a mere conveyance into a testamentum, in which the appointment of a heres was not only possible but indispensable, and in which the familiae emptor was no longer the active executant of the ‘ testator’s ’ wishes but a mere cipher.® (6b) What was the purpose of the careful phrase endo mandatela tua custodelaque mea? ® It must presumably have been meant to qualify the effect of 8 Stein, Regulae 12f., suggests that lea here has the sense of ‘ declaration’ and that the reference is to the nuncupatio (see text immediately below), made before members

of9 the public asnwitnesses. | Above, 126 7. 1 2,224. 2 See FIRA 1.37£. 3 See de Zulueta, Gaius 87ff. 4 Kaser, RPE 1.107£.; cf. above, 139 n 7.

5 But technicalities which testify to his original position remained even in classical law. Thus, no person in his power or in whose power he was or in the same power with him could be a witness; Gai. 2.105-8. Another trace probably in D. 48.18.1.6; Mitteis, RPR 287. 6 Weiss, SZ 42 (1921) 102ff. David and Nelson, 315ff., omit tua[m] (see above, 242 n 9) because que suggests a close link between the two nouns; the meaning would then be, by hendiadys, ‘in my care for the purpose of fulfilling a commission ’. 24:4:

Succession

the mancipation. If one applies the strict categories of later law, ownership

has necessarily passed and the ‘ testator’, in any interval between the making of the mancipation and his death, is dependent on the goodwill of the familiae emptor.’ How the early law understood the matter, and what precisely was the relationship between familiae emptor and ‘ testator’ we cannot tell, but the words which we are considering must in substance have had the effect of postponing the effect of the conveyance until the ‘ testator’s ’ death.* They would otherwise have been mere verbiage.

Similarly, we do not know whether the beneficiaries of the ‘ legacy ’ had

any way of forcing the familiae emptor to carry out the instructions, or whether there was simply an unenforceable trust.* There is indeed no reference to a fiducia as in some other cases where mancipation is used to give legal effect to purposes other than those of a mere conveyance,’ but the careful words suggest some method of enforcement. (c) Was the familiae emptor (before he became a mere cipher) a universal successor in the sense in which a heres was? It may well be that if the testator’s instructions did not cover the whole of his property the familiae emptor

could retain the beneficial ownership of what was left over, and this fact would account for Gaius’ reference to him as being heredis loco, for the heir also retained what remained after he had distributed the legacies, but that mancipation can in early times have been deemed capable of transferring the abstract ‘ inheritance’ seems unlikely, for the simple reason that this abstract conception is not primitive at all. On the crucial] question whether or not the familiae emptor was liable for the debts of the deceased we have no clear evidence.” Nor do we know what happened to the sacra. (d) When the change in the position of the familiae emptor took place we

cannot tell. Lenel put it as late as about the beginning of the second century B.C.,? but there can be no doubt that by Cicero’s day, when the old forms 7 Maine, Ancient Law 217: ‘Still we must believe that, if the Testator did recover, he could only continue to govern his household by the sufferance of his Heir.’ 8 Kaser, AJ 151, sees it in terms of divided ownership. 9 Weiss, SZ 42 (1921) 102ff. (especially 112) argues that the custodela of the familiae emptor meant simply that he had to keep the things at the disposal of the legatees. In his view the legacies were all originally per vindicationem (cf. below, 246f.) and the

legatees could consequently fetch the things as soon as they became theirs at the

testator’s death. This theory has the advantage of explaining how the legatees enforced their rights, which is a difficulty if we dismiss the theory of a mere unenforceable trust. 1 As in fiducia cum amico and cum creditore; Gai. 2.59f.; below, 286.

2 Of course if the familiae emptor was not liable for the debts it is difficult to say exactly what did happen. It may be that the rules were not as definite as could be wished, and that the need for someone who would be liable for them was precisely one of the reasons which led the later jurists to insist that an heir not only might but must

be appointed. But it may well be that in the early law obligations were in any case intransmissible (Kaser, RIDA, 1952, 507ff.). 3 Hssays in Legal History, ed. Vinogradoff 141 (=Labeo 12 (1966) 374).

245

Private law to the fall of the republic: family and succession had disappeared, the evolution was complete. By then too the testator’s instructions were already written down in the normal case and remained secret until his death, but it seems certain that there existed also the possibility of an oral will, the ceremony being exactly the same except that the

witnesses.* .

dispositions were spoken by the testator and hence necessarily known to the

(iii) Restrictions on institution. The testator, whether he has sui or not, is free to choose his heir or heirs as he wishes,*® but if he has suz whom he wishes to exclude he cannot simply pass them over in silence. In the case of a son he must put in a clause disinheriting him expressly (nominatim), otherwise the will will be entirely void.® If he has other sui (grandchildren or daughters) he must at least put in a clause saying that all not instituted heirs are to be disinherited; otherwise the omitted persons will be able to come in and share with those instituted.’ The reason for these rules is, no doubt, that the sui are regarded as having a sort of dormant ownership in the estate of their father even during his lifetime, and, at his death, if they are heirs they succeed without the need for any express acceptance. The power of the father is great enough to exclude them, but this power must be exercised expressly and not by implication.® The rules of disherison are thus restrictions of form; the testator can do as he likes, provided he takes the proper steps. The only restriction of substance introduced during the republic with respect to the institution of heirs is that of the lex Voconia (169 B.C.), which forbade the institution of women as heirs by persons rated in the highest (i.e. wealthiest)

class at the census.°® |

(iv) Legacies. A legacy is defined in the Digest ' as a lessening (delibatio) of the inheritance, a disposition which decreases the amount that the heir or heirs would otherwise take. The ‘ lessening’ might take either of two forms — per vindicationem or per damnationem.? A legacy per vindicationem 4 Horace is said by Suetonius (Vita Horatii, sub fin.) to have made a will orally in favour of Augustus because he was too ill to observe the usual form; see Girard 860 n 1. The form of the nuncupatio must, of course, have been slightly different im these

cases from that given by Gaius.

5 Provided the person chosen has testamenti factio. 6 Gai. 2.123.

7 Gai. 2.124. The rules given in the text are those stated by Gaius. There was still some doubt about details when Cicero was a child (see de Or. 1.175), but, in spite of some controversy, it seems clear that the general rule is old; Girard 905 n 1; Kaser, RPR 1.703.

8 Cf. above, 124.

9 Gai. 2.274. The scope of the statute was perhaps originally wider; Buckland 290 n 10. 1 30.116 pr.

2 Two less important forms, per praeceptionem and sinendt modo, are known to the classical law, Gai. 2.192, but the former is a variation of per vindicationem and the latter of per damnationem, and the antiquity of both is disputed; Buckland 334 n 7. 246

Succession

is one in which the testator has used the form do lego (‘I give, bequeath ’) or an equivalent, and the result is that, as soon as the will becomes operative by the entry of the heir, the ownership in the thing bequeathed passes to the legatee, who can then bring a vindicatio to obtain it from the heir or anyone else who happens to possess it. A legacy per damnationem, on the other hand, is given by the words heres meus damnas esto dare * (‘ Let my heir be damnas to convey ’) or an equivalent; it gives no right in rem to the legatee, but imposes an obligation on the heir. The thing bequeathed (if it was the property of the testator) passes, like the rest of the property of the deceased, to the heir, but the legatee can bring an action (in personam) against the heir to force him to transfer it. It is even possible by this method to bequeath things which do not belong to the testator; the heir is then under an obligation to get the thing if he can and transfer it to the legatee, or, if he cannot, to pay its value. This form was particularly useful for giving legacies of sums of money. A pecuniary legacy per vindicationem was only good if the testator actually owned coins to the requisite amount when he died,* for only so

] could he pass ownership in the coins, but however little actual cash a man owned he could impose a duty on his heir to pay a sum of money, which the heir could then raise in any way he thought fit. Of these two main types of legacy that per vindicationem is probably the older, for the words do lego which are used in giving it are precisely those which the testator uses in his nuncupatio.®

(v) Restrictions on legacies, Even after the institution of an heir had become a definite requirement of the mancipatory will, there seems to have been a tendency to comply only in form with the demands of this rule by nominating an heir, while neglecting the substance of it by giving away so much in legacies that little or nothing was left to the heir. One result of this

practice was that heirs often found that it was not worth their while to accept the inheritance, and refused to enter, thus causing the whole will to fail, and a series of statutes had to be passed to remedy matters.® The first of

these was the lex Furia (testamentaria), of unknown date,’ which forbade legacies of more than 1,000 asses to be given except to certain persons closely related to the testator. It did not have the desired effect because by giving a sufficient number of legacies, none of which exceeded 1,000 asses, the largest 3 For the word damnas see above, 165, and Liebs, SZ 85 (1963) 242ff., with reff.

4 A legacy of non-fungible things could not be given per vindicationem unless the testator owned them ex ture Quiritiwm both when he made the will and when he died. For fungible things ownership ex 74. Q. at death sufficed; Gai. 2.196.

5 Cf. above, 243. On the whole question see Wlassak, SZ 31 (1910) 196ff.; Kaser, AJ

150ff.; SZ 67 (1950) 320ff.; RPR 1.110 n 8. 6 Gai. 2.22448.

7 It is probably later than the lez Cincia (204 B.C.) and certainly before the lex Voconia; Girard 975 n 7; Mitteis, RPR 52 n 30. For the machinery of the law cf. above, 88.

247

Private law to the fall of the republic: family and succession estate might be exhausted and nothing left for the heir. No greater success attended the provision of the lex Voconia which forbade any legacies to be greater than the amount which remained for the heir,® for again, provided there were a large enough number of legacies, although no one individually exceeded the heir’s share, that share might still be too small to make it worth his while to accept. The matter was not finally settled until the passing of the

lex Falcidia (40 B.C.), which enacted that if the amount of the legacies exceeded three-quarters of the estate the heir might cut them all down proportionately and so retain a quarter for himself.°

(vi) The ‘ praetorian will’, The praetor could not make an heir but he could grant bonorum possessio, which meant that he could grant to persons of his choice the right to use certain remedies which would enable them to get possession of the goods of the deceased. Whether they would be able to keep what they had got is another question, which has to be answered differently according to the circumstances and the period, but at any rate they could get possession in the first instance.’ The rules governing the praetor’s choice of persons to whom he will grant bonorum possessio are set out in the edict and extend both to testamentary and intestate succession. For the moment, however, we are concerned only with testamentary succession. So far as this is concerned the praetor based his action on the view that the whole formality of mancipation and nuncupation was so much useless ceremonial. What was important was that the testator’s dispositions should be committed to writing and that the document in which they were contained should be identi-

fiable by the seals of persons who could testify that it was genuine. The praetor therefore promised bonorum possessio .secundum tabulas (i.e. in accordance with the tablets) to any person nominated heir in tablets sealed with the requisite number of seals, i.e. seven,? for, so far as the praetor is concerned, the familiae emptor and the libripens are just two witnesses in addition to the five required by civil law for mancipation. This bonorum 8 The combined result of this with the other provision of the lez Voconia (above, 246) was that no person rated in the first class of the census could leave a woman more than

half his property. Some scholars hold that there was an exception in the case of daughters, but the evidence is not convincing. For literature see Girard 870; Kibler 196. 9 Wesel, SZ 81 (1964) 309ff., doubts Gaius’ account of legislative incompetence (which certainly almost passes belief) and supposes that only the lex Falcidia was concerned to secure a sufficiency for the heir. The lez Furia would have aimed merely at excessive

legacies, just as the lez Cincia was aimed at excessive gifts; and the purpose of the lex Voconia, in this provision as in the other (above, 246), would have been to prevent the accumulation of wealth in women’s hands. 1 Cf. below, 253f.

2 The edict in its final form seems to have said that there must be ‘not less than

seven seals’, In Cicero’s time we know however (Verr. 11.1.117) that the words were ‘not fewer than the law requires’ (cf. above, 243 n 3), but whether that meant five or seven is not clear; Lenel, HP 349; Kaser, F'schr. Schulz 2.26f.; Watson, Succession 13ff.

248

Succession possessio was given whether the civil law ceremonies had been gone through

or not, so that if they had been gone through the person nominated in the tablets was both heir at civil law and entitled to bonorum possessio; if they had not been gone through he was entitled to bonorum possessio only, and the document may be called a ‘ praetorian will ’.*

At what date bonorum possessio secundum tabulas was first granted is unknown, but it was already in existence in the time of Cicero.*

(vii) Praetorian rules of disherison. As the civil law will is subject to cer-

tain restrictions with respect to sui,> so the praetorian will is subject to analogous restrictions with respect to the praetorian class corresponding to sui —the liberi. In Gaius’ time all male liberi must be disinherited nominatim if they are not to succeed; only females can be excluded inter ceteros. If these rules are not complied with the omitted person can claim b.p. contra tabulas, i.e. contrary to the dispositions in the tablets. These rules, or at any rate b.p. contra tabulas, go back as far as Labeo,’ but cannot be much older, as the class of liberi had not yet been defined in Cicero’s time.’

B. Inrestate Succession. (i) Civil law. The civil law rules of succession as laid down by the XII Tables remained unchanged throughout the republic, except in one ® particular, the restriction of the agnatic rights of females. By the law of the later republic only sisters, not more remote female agnates, could inherit on intestacy, whereas there had been no such limitation originally.1 This change was apparently not due to legislation, but to the interpretatio of the jurists, and is said in a late text ? to have been made ‘ on the principle of the lex Voconia’. This probably means that it took place after the passing of that statute, but it may only mean that it was produced by the same tendency which produced the lex Voconia, a tendency to restrict the power and luxury of women by restricting their chances of sharing the great fortunes which were becoming common at Rome.”

(ii) Praetorian law. More important and destined to have much more farreaching effects were the reforms brought about by the praetor’s edict, for it was the praetor who first recognised the claims of the cognatic, as opposed to ’ This phrase was not used by the Romans, and in some respects documents of this sort did not attain the efficacy of a will; Buckland 285f.; Voci, St. Grosso 1.99ff.

47 D. Above,n2. - 5 Above, 246. 6 Below, 250. 37.4.8.11; Girard 910 n 4. 8 Below, 250 n 8. ® There may also have been the change from corporate to individual succession by the

gentiles; above, 125. 1 Above, 125. 2 PS 4.8.20 (22).

s* Some modern scholars have thought that the restriction must be older on the ground that the complete equality of males and females is inconsistent with a primitive system, but the evidence brought forward is insufficient to upset the uniform Roman tradition; Kiibler, SZ 41 (1920) 15ff. Against Kiibler: S. Brassloff, Studien zur rom, Rechtsgesch. 1 (Vienna/Leipzig, 1925), reviewed SZ 47 (1927) 440ff. 249

Private law to the fall of the republic: family and succession the agnatic, system of relationship, i.e. of blood-relationship traced through females as well as through males and independent of potestas. Here again it will be well to mention briefly the classical system, which is well known, before indicating how far the reforms had got by the end of the republic. In classical times, so far as ordinary cases of succession to freeborn persons was concerned,* the praetor had four classes of persons to whom he offered bonorum possessio ab intestato in turn: (a) Liberi (‘ children ’). This is an extended class of sui, i.e. it includes all who are sui together with those who would have been suz, but for capitis deminutio minima,’ in particular emancipated children. (b) Legitimi, i.e. all who have a civil law claim, including sui and agnatz, but not gentiles, whose rights are obsolete by classical times. (c) Cognati, i.e. blood-relations up to the sixth degree,® without distinction of male and female lines, the nearer relation excluding the more remote, as was the case with agnates under the civil law system. (d) Vir et uxor. Failing any of the earlier classes the husband might get bonorum possessio to his deceased wife or the wife to her deceased husband. This refers only to marriage without manus, because where there was manus the woman could own nothing, and if her husband predeceased her would be a sua heres to him. It will be seen that this system breaks with agnation in favour of cognation in two respects, first in the category of liberi, for here it disregards the element of potestas, and secondly by giving rights to cognati as such, but the breach must not be exaggerated. The category of liberi, like that of suz, exists only on the death of a paterfamilias; the praetor gave children no right of succession to their mother except as cognates; and cognatz come in only

after all agnati, so that the remotest agnate will exclude, for instance, an uncle or a first cousin on the mother’s side.

In republican times the praetorian categories had not yet assumed their classical form. After the words promising b.p. secundum tabulas there followed immediately words promising b.p. to those persons who had a civil law claim,’ i.e. the legitimi; the class of liberi had not yet apparently come to be recognised, and the legitimi still included the gentiles.* The rights of cognates

also do not appear to have been as yet recognised expressly in the edict. Probably the edict merely said that if there was no claim under a will or under the civil law rules of intestacy the praetor would grant bonorum possessio to the person whose claim appeared most just.® But there are instances 4 Special rules applied where the deceased had been emancipated; Buckland 384. 5 Except that children given in adoption were not included so long as they remained

in their adoptive family, and adopted children were not included if they had been emancipated.

in one case, that of second cousins once removed. 7 Cie. Verr. 11.1.114. 86 Seventh Verr, 11.1.115. 9 Cic. Part. or. 98. 250

Succession

of grants to cognates who of course may well have been the persons whom such an edict was primarily intended to benefit.’ C. ACQUISITION OF HEREDITAS AND OF BONORUM POSSESSIO. Sui heredes

acquire the rights and duties of an heir immediately on the death of the deceased, but an extraneus upon whom an inheritance devolves does not become heres until he has manifested his acceptance of the position.” In the classical law such manifestation might be made in either of two ways.* Nor-

mally it was by pro herede gestio (acting as heir), i.e. by doing some act, such, for instance, as selling the goods or providing the slaves with food, which showed that the extraneus regarded himself as heir.* But sometimes a

much more formal method called cretto was used, which consisted in the speaking of a solemn formula in which the chief words were hereditatem adeo cernoque (‘I enter and decide upon the inheritance ’).° Acceptance in this way was only necessary in classical times where it had been prescribed in a special clause of a will,® and therefore only took place in the case of testa-

mentary succession, but it is held by many scholars that originally it had a much wider scope and was in fact the only method by which an ewtraneus could acquire an inheritance immediately, whether it came to him under a will or on intestacy. It would be strange indeed if the highly formal system

of early Rome had allowed so important a matter as the entry upon an inheritance to take place informally, and there is no other ceremony of which we have any trace.’ It is also probable that cretio included originally more

than the mere speaking of the words, that, as the word adeo suggests, it necessitated physical entry upon the land of the deceased.’ Unlike hereditas, bonorum possessio is never acquired simply by the death of the person whose estate is in question. It has always to be asked for from

the praetor, and one of the chief characteristics of the praetorian system is that it has to be asked for within a limited time. For each class there is a period (usually 100 days, but a year in the case of children and ascendants) 1 See Cic. Cluent. 165; ef. Kubler 190 n 4. 2 Cf. above, 124.

3 The text of Gaius, 2.167, mentions also nuda voluntas (informal declaration of intention) as a way of acquiring an inheritance, but there is considerable doubt whether this correctly represents the classical law; see de Zulueta, Gaius 103. Possibly it represents provincial practice (Arangio-Ruiz, Istit. 555). At any rate Labeo seems clearly not

to have known of it; D. 29.2.62 pr. 4 Enit. Ulp. 22.26,

5 Gai. 2.166. Though it was no doubt usual to have witnesses they were probably not necessary; Buckland 313; TR 3 (1922) 241. 6 The object of the clause, which always contained a time-limit, was to ensure that the heir made up his mind whether to accept or not within a reasonable time. There was no limit at civil law or otherwise; Buckland 313. 7 Girard 923f.; Karlowa 2.896; Buckland, TR 3 (1922) 249. Contra, Lenel, Essays in Legal History, ed. Vinogradoff 124 (=Labeo 12 (1966) 361). 8 Sohm 567 n 7. 251

Private law to the fall of the republic: family and succession during which it is open to them to claim, and then the next class has its chance. Thus, if the deceased died intestate leaving an agnatic first cousin and a maternal uncle as his only relatives, the cousin has first claim; if he fails to ask for b.p. during 100 days the maternal uncle can claim as cognate.

It is true that the cousin is also a cognate (he is a blood-relation), but as such he will be postponed to the uncle, for an uncle is in the third degree of relationship whereas a first cousin is in the fourth. Only if the uncle fails to claim can the cousin now come in. This system of ‘ succession of orders and of grades’ (i.e. of one class to another and of a more remote relative within the same class on the failure of a nearer) is purely praetorian, and is among the praetor’s most important innovations. At civil law if there was e.g. a suus, and he did not want the inheritance, no right passed to the agnates; if there was an agnate and he failed to accept, no right passed to the gentiles. Similarly, if there was a nearer agnate who failed to accept, a remoter agnate had no right. The object of the praetor, on the other hand, was to multiply successors, to see, as Gaius says, that no.one should die without a successor.° D. REMEDIES OF THE HEIR AND OF THE BONORUM POSSESSOR. The civil law

heir, no matter what his title to the inheritance, succeeds to the rights and duties of the deceased, so far as they are transmissible.! Having thus become owner of the goods which the deceased owned, he can bring a vindicatio to recover any of such goods if they are not in his possession; being now creditor where the deceased was creditor, he can use the same action (whatever it might be in the particular case) for recovering the debt as the deceased might

have done. But, in addition, he has an action in his capacity as heir — the hereditatis petitio — which is the proper one for him to bring where the question to be raised is whether he or the defendant is entitled to the succession. Thus, if he finds a horse which he believes to have belonged to the deceased.

in the possession of a third party, X, and X on being asked to give it up refuses on the ground that the horse never belonged to the deceased or that he bought it from the deceased before he died, the proper action for settling

the matter is an ordinary vindicatio; but if X’s reply is, ‘I agree that the horse belonged to the deceased when he died, but in fact I am his successor, not you ’, the proper action is the hereditatis petitio.? So also if a third party, on being asked to pay a debt which he owed to the deceased, refuses to do so on the ground that he himself is heir and the debt has thus been cancelled, hereditatis petitio is the proper action. The bonorum possessor as such has neither the particular actions nor the hereditatis petitio; the praetorian grant simply means that he can use certain

® Gai, 3.33. 1 Cf. above, 123.

2 If the plaintiff insisted on bringing the vindicatio he would be defeated, originally by a praescriptio, later by an exceptio praeiudicialis; see Gai. 4.133.

252

, Succession ornate mmtr aa ee eend

praetorian remedies, of which the earliest and probably the only one existing

in republican times was the interdict quorum bonorum.® This interdict enabled him to get possession of any of the goods of the deceased which were in the hands of any person who held pro herede or pro possessore, i.e. who believed himself (rightly or wrongly) to be heir, or knew that he had no title,

and did not put forward any ground of possession independent of inheritance.* It was thus possible for the bonorum possessor to get possession even from the heir himself, if there was one. True, the heir could sometimes in classical law, always originally,® get back possession from the bonorum pos-

sessor by means of the hereditatis petitio, but that does not mean that the interdict quorum bonorum was useless. In the first place it enabled the bonorum possessor to be defendant in any action that was brought — he could get the goods and force the person who alleged himself to be heir to prove his

case; secondly, if he retained possession long enough, he would become owner of the property by usucapion, and then any claim by the heir would be too late; thirdly, there might be no heir at all, in which case his possession would be unchallengeable and necessarily ripen into ownership by usucapion.

Further, it must be remembered that the person entitled to bonorum possessio would very often be the heir himself; if, in such case, he availed him-

self of the praetor’s offer, he had the advantage of being able to use the praetorian remedy (quorum bonorum), which was more expeditious than the hereditatis petitio.® K. ORIGIN OF BONORUM POSsSESSIO. The chief historical importance of bono-

Tum possessio lies in its use as an agency for the reform of the old law of succession — supplendi or corrigendi iuris civilis gratia — but it is unlikely that in its origin it was at all revolutionary. Even in Cicero’s time, as we have seen, the praetor’s scheme differed but very slightly from that of the civil law, the only class of intestate successors who were recognised being in fact the legitami — those entitled at civil law. It was not until later that the class of liberi, in which the praetor definitely broke with the civil law, was recognised, and, though cognates might be admitted to bonorum possessio, they do not appear

to have been specifically mentioned in the edict. In any case, if bonorum possessio did come to be granted to someone who was not civil law heir, it is 3 Cf. above, 232. It was already known to Cicero; see ad fam. 7.21; Top. 18. For the

4 Gai. 4.144. 5 Below, 254.

words of the interdict see D. 43.2.1 pr.

6 In the classical law (and probably since Labeo’s time, D. 37.1.3.1; ef. Sen. benef. 6.5.3) the bonorum possessor could also sue the debtors of the deceased and be sued by his ereditors by actions in which there was a fiction that he was heir (Gai. 4.34). Until these were invented he could apparently do nothing more than get the res corporales by the interdict quorwm bonorwm. Where he succeeded in usucaping them, according to the older theory (Gai. 2.54), he was regarded as usucaping the inheritance, and was, conse-

quently, it is to be presumed, liable for the debts in proportion to the share he had usucaped.

253

Private law to the fall of the republic: family and succession clear that the civil law heir, if there was one, would always be able, at this period, to succeed in a hereditatis petitio against him, provided, of course, that he brought his action before the bonorum possessor had had time to usucape. In fact, whereas in the classical law the praetor would sometimes uphold his own nominee against the civil law heir,’ in Cicero’s time bonorum

possessio, if given to someone other than the heir, was always sine re, i.e. ineffective against the heir. In its origin then bonorum possessio was almost certainly given adiuvandi iuris civilis gratia, i.e. to provide persons who were

already civil law heirs with additional praetorian remedies, but what the motive was which determined the praetor to act in this way remains an open question. Leist * maintained that the central idea of bonorum possessio was to be found in the praetorian system of ‘ succession of orders and degrees ’. The civil law gave no right to a more remote agnate if the nearest did not accept

the inheritance open to him, or to the gentiles if there were agnates who failed to accept; further, at civil law, if there was an heir nominated under a will, intestate succession did not open until it was certain that he would not accept, either by reason of his death or his repudiation, and there was nothing to force him to make up his mind within a reasonable time. It might thus happen very easily that an inheritance was vacant, without anyone to

protect it, a state of affairs necessarily dangerous and particularly so at Rome, where the rule was that anyone might seize on any portion of a vacant inheritance and usucape it in one year even though he was not in good faith.® The praetor therefore, so Leist supposed, gave to all those who would

be entitled at civil law, in turn, the chance of getting possession of the inheritance, i.e. in Leist’s view, not merely the corporeal assets, but a possessory right to the position of heir, so that they would be able to bring all necessary actions and be liable to be sued by the creditors. This procedure was in the interest of all parties — the heir himself, other possible claimants, the creditors and the pontiffs, who would thus have someone on whom they could fix the responsibility for the sacra. If it so happened that possession was granted to a non-heir, there was always the possibility that the real heir would make it sine re by claiming in time, but meanwhile the inheritance was protected.' 7 This was done by giving the bonorwm possessor an exceptio doli against the heredi-

tatis petitio of the heir. See eg. Gai. 2.120, which describes how bonorwm possessto secundwm tabulas became cum re as against agnates by reason of a rescript of Antoninus Pius. 8 In Glick’s Erlaéuterung der Pandekten, Serie der Biicher 37~38.1. Also published as Der rémische Erbrechtsbesitz in seiner urspriinglichen Gestalt, 1870. ® Above, 153 n 3.

1 The idea of this provisional regulation of the inheritance was, in Leist’s view, taken by the praetor from Athenian law. Sohm 566ff. went even further in this direction. 254

Succession The principal objection to Leist’s theory is that it involves the assumption that the grant of bonorum possessio was a grant of the ‘ inheritance ’ as such,

and not merely of the corporeal assets. This means that the bonorum possessor (even if not heir) must have been able to sue and be sued in the same way as an heir, and Leist indeed says that the actions with a fiction date from the time when bonorum possessio was first given to a non-heir.? This is

a very difficult assumption to make, for the interdict quorum bonorum is probably older than the fictitious actions. The interdict quorum bonorum was

clearly known to Cicero,’ the first trace of the fictitious actions dates from the time of Labeo.* It is thus probably better to follow the opinion of those scholars who regard bonorum possessio in its origin as simply a scheme for the regulation of the possession of the goods of a deceased person pending the settlement of the question of inheritance by hereditatis petitio.® If two people were about to dispute the title to a thing by hereditatis petitio under the old system of legis actio sacramento, the interim possession of it would have to

be assigned to one or other of them, just as was the case with an ordinary vindicatio, and the praetor might very well settle this question by assigning it to the one who appeared prima facie to be entitled according to the civil law; originally this would have been done after investigation by the praetor

himself: later, the machinery of an interdict would be used in this as in other cases. Often, no doubt, the parties would content themselves with this procedure, and the hereditatis petitio would never need to be tried. According to this theory, then, the interdict quorum bonorum would have had an origin very similar to that suggested for utz possidetis,® in the desire of the praetor to grant the advantage of possession to the person who appeared to be prima facie entitled; it is also pointed out that the words si de hereditate ambigitur may well have referred to the proceedings by hereditatis petitio for which the interim assignment of possession was first needed.

2 Gluck, Erlduterung der Pandekten (above) 102. 3 Top. 18; ad ‘fam. 7.21. 4 Above, 253 n 6. 5 Girard 844ff.; Bonfante, Corso 6.420; Buckland, Manual 235. 6 Below, 262.

255

CHAPTER 15

Private law from the XII Tables to the fall of the republic: the law of slavery and manumMission The old civil law had recognised the humanity of the slave only by admitting that he could acquire rights for his master and by allowing him to become a

citizen on manumission.' By the end of the republic praetorian rules had modified this position only in two directions: (a) by admitting that in certain circumstances the slave might not only acquire rights for the master but also

place him under a liability; (b) by allowing that certain acts, not falling under any of the forms of civil law manumission and not resulting in citizen-

ship, might give a slave the right to protection against the exercise by the master of his rights of ownership. (a) Here the praetor carried out his innovation by granting actions against the master on transactions entered into by the slave.” These actions fall into

two classes, according as they are based on the idea that a master who has authorised his slave to contract must be responsible for the contract, or on the idea that the master who has allowed the slave to administer a peculium for himself must be prepared to lose the peculium if the slave administers it badly. To the former class belong the actiones quod iussu, exercitoria and institoria, to the latter the actio de peculio et in rem verso and the actio tributoria. The actio quod iussu enabled a person with whom a slave had made a contract which was authorised by his master to sue the master on the contract for the full amount owing. The actio institoria concerns the general authorisation implied in the appointment of a slave as manager of a business; if any person contracted with the slave in connexion with the business which he was appointed to manage he could sue the master for the full amount by

this action. Almost exactly the same rules applied where the slave was appointed master of a trading ship, but here the action was called emer1 Above, 135.

2 Or filius familias; what follows applies equally to both. The name actiones adtecticiae qualitatis usually given to these actions, though not Roman, is useful. For the

obligation is that of the slave or son (even though it is unenforceable—a ‘natural’ obligation in later terminology; Buckland 552) ; the liability of the master is additional or accessory. This is made clear in the formulae; see the example below, 257f. 256

Private law to the fall of the republic: slavery and manumission citoria.* These actions, it must be noticed, greatly increased the usefulness of slaves, for they made it possible for masters to use their slaves as agents, a thing which had been impossible under the old law when slaves could only

acquire rights for their masters and not put them under liabilities. The actio de peculio et in rem verso also enables the master to be sued on his slave’s contract, but there are two important features which distinguish this case from those that precede. There is no need for any authorisation — indeed the action lies even if the master forbade the transaction; all that is necessary

is that the slave should have a peculium.* But, on the other hand, the master’s liability is limited; he is only hable up to the amount of the peculium and for any proceeds of the transaction by which he has profited, if, for instance, the slave has borrowed money and used it to pay a creditor of his master’s. In estimating the amount of the peculium the master could deduct any debts due from the slave to himself,® but, on the other hand, must include any debts due from himself to the slave —‘ debts ’ meaning here not actionable debts (for there could be no action between a master and his slave), but unenforceable debts arising out of transactions between the slave and the master, because so far as practice went the peculium was treated as the slave’s property. He could, for instance, ‘ sell’ a thing to his master and

the result would be that the price of the thing had to be reckoned as part of the peculium even before the master gave it to the slave, because he already ‘ owed ’ it. The actio tributoria shows a difference in this respect. If the slave traded with his peculium or with part of it to the master’s know-

ledge and the fund so created was at any time insufficient to pay all the creditors, it was the master’s business (if required to do so) to divide what remained up fairly in proportion to the debts; in such a case he might not deduct first what was ‘ owing ’ to himself, but had to treat himself just as he would any other creditor. If any creditor thought he was not being treated fairly in this respect he could sue the master by this action. The whole field of these actions provides a good example of the techniques of purely praetorian law-making right outside the province of the civil law.

Several edicts laid down the circumstances in which the different actions would be allowed, and in the formulae the praetorian devices of fiction and change of person between the intentio and the condemnatio were employed.° K.g. the formula, in the case of a purchase made by a slave on the authori-

sation of his master, ran something as follows: ‘ Whereas A.A. with the authorisation of N. N. sold a toga to Stichus at a time when Stichus was in 3 From exercitor, t.e. the person who takes the profit of a ship, her owner or charterer. Institoria is derived from institor, a business manager. These actions were also available if the agent was sui iuris; Gai. 4.71. 4 See above, 83.

5 Or, what comes to the same thing, to any other member of the familia. 6 Above, 209. And see F. H. Lawson, The Roman Law Reader (New York, 1969) 1344f.

257

Private law to the fall of the republic: slavery and manumission the power of N. N., whatever on that account Stichus, if he were free, would have in good faith to pay A. A., thereto the judge is to condemn N. N. ete.’ * A great deal of the elaboration of the law in connexion with these actions is of course due to the work of the classical jurists, but that the actions themselves were known by the end of the republic or very early in the principate seems clear. The actiones institoria * and de peculio ® are mentioned by Ser. Sulpicius,: the exercitoria? by Ofilius,* and the actiones quod iussu* and tributoria * by Labeo.*®

(b) The ‘ informal ’ methods of manumission. It happened sometimes that a master who failed for one reason or another ™ to use recognised methods of manumission, nevertheless declared that he wished a slave of his to be free.

This had no effect at civil law, but, provided certain conditions had been complied with, the praetor would not allow the master to go back on his word to the extent of forcing the slave to work for him again; he would, in fact, refuse the master his vindicatio in servitutem. For all other purposes | however the slave remained a slave, and the children of a woman were slaves of her owner. It was not every informal declaration that had this effect; the praetor only acted if the master had provided the slave with some evidence which he could subsequently adduce if the declaration were called in question, either by writing him a letter or by making the declaration before wit_ nesses. Hence there are said to be two kinds of informal manumission, that per epistolam and that inter amicos, for the ‘ friends’ are here present as witnesses.

7 Lenel, EP 278. 8 D. 14.3.5.1. 9D. 15.1.17; ef. 15.3.16. 1 Above, 93. 2D. 14.1.1.9. 3 Above, 94. 4 D. 15.4.1.9, 5 D. 14.4.7.4; 9.2. 6 Below, 380. 7* E.g. because being only ‘ bonitary’ owner he was not competent to use one of the civil law methods, or because being deaf or dumb he was similarly incompetent; see PS 4,12.2,

258

CHAPTER 16

Private law from the XII Tables to the fall of the republic: the law of property 1 THE POSSESSORY INTERDICTS In this period there emerges the clear distinction between ownership and possession which is one of the characteristic features of classical Roman law.'

And to understand the Roman idea of possession it is necessary to understand the interdicts by which it was protected. The characteristic of these interdicts is that they protected possession as an existing state of fact without reference to its rightfulness or wrongfulness,? and the result of their existence is that if anyone wishes to disturb an existing possession he can only do so by means of a regular judicial proceeding in which he himself proves a title. The Romans themselves divided possessory interdicts into three classes,

according as they served the purpose of acquiring, retaining or recovering possession (vel adipiscendae vel retinendae vel reciperandae possessionis causa comparata).* An example of the first class is the interdict quorum bonorum, which enables the bonorum possessor to get possession of the goods of the deceased.* This is possessory in the sense that it deals with possession

only; no question of superior title may be raised by the defendant; if, for instance, he thinks that he is heir and has a better right to the goods than the bonorum possessor, he must nevertheless give them up and will have to make good his title, if he can, in a separate hereditatis petitio. But these interdicts of the first class are not relevant to the present discussion, because,

as their name shows, they do not protect an existing possession, but are a means of obtaining a new one. The remaining two classes we must consider.

(i) Interdicta retinendae possessionis causa comparata. Of these there were two, one, ut2 possidetis, for use where the property concerned was immovable, the other, utrwbi, for use where it was movable. (a) Uti possidetis. The simplest way of understanding this interdict is to take the case where it is used as a preliminary to a vindicatio. If two people are in dispute as to the ownership of a piece of land, it is usually quite clear 1 D. 41.2.12.1: Nihil commune habet proprietas cum possessione.

2 Except that it must not have been obtained vi, clam or precario from the other party; below, 260. 3 Gai. 4.143. The classification is misleading since, as will be seen below, those which were ostensibly for retaining possession might, in their developed form, operate to recover

possession. 4 Cf. above, 253. 259

Private law to the fall of the republic: property that one possesses and the other does not, but it may happen that each claims to be actually in possession; one may say that he has mown the grass,

the other that he has pastured his beasts on the stubble, and each may claim that his act amounts to the exercise of possession. Which is right is decided by the interdict proceedings. The praetor issues an order, addressed to both parties, forbidding any interference with the existing state of possession;° each party then does some act (for form’s sake) which, if the other possesses, amounts to such interference, and the question which of them has thus disobeyed the praetor’s order is tried on sponsiones and restipulationes.®

Interim possession is granted to the party who makes the higher bid for it, i.e. Offers to pay the higher sum should he get it and be found subsequently not to have been entitled.’ If it turns out that the interim possessor is the person entitled, i.e. really did possess at the beginning of the proceedings, he of course remains in possession and will be defendant in the vindicatio; if the other party turns out to be entitled, possession will have to be handed over to him, and he will have the desirable position of defendant. The interesting

thing to notice is that the whole proceeding is ‘ double’, as the legis actio sacramento in rem was; there is no plaintiff and no defendant, and interim possession pending the settlement of the question at issue has thus to be given to one or other, but the question at issue is now simply that of the actual possession, and in no way prejudices further proceedings to settle the right of ownership. This use of ut possidetis as a preliminary to vindicatio is typical, but it might, of course, happen that the party defeated in the interdict rested content with that decision and never brought a vindicatio. It is also generally agreed that uti possidetis might be used in certain circum_ stances to recover a possession which had been lost.* The interdict, from a very early date, contained words which limited the praetor’s protection to such possession as had not been obtained vi, clam or precario (violently, clandestinely or as tenant-at-will) from the other party to the proceedings. If, therefore, A was found to be in possession but it was also found that he had obtained that possession by violently ejecting B, then it is clear that B in exercising the formal force did not disobey the praetor’s order. It is not quite so clear that A disobeyed it, but it appears that he was held to have done so and consequently had to give up possession to B.

(6) Utrubi.? Where a movable object was concerned, the praetor’s order 5 Uti eas aedes quibus de agitur nec vi nec clam nec precario alter ab altero possidetis, quo minus ita possideatis vim fiert veto (Ulpian, D, 43.17.1 pr.; older version, Festus, 8.v. possessio; see Lenel, EP 470).

é Cf. above, 230f. The interdict being ‘ double’ there is need for two bets, (i) that A has disobeyed, (ii) that B has disobeyed, and consequently for four stipulations.

7 Gai. 4.167. 8 Buckland 734; Schulz, CRL 450f.

9 Utrubi hic homo quo de agitur matiore parte huiusce anni nec vi nec clam nec precario ab altero fwit, quo minus is ewm ducat vim fieri veto (Lenel’s reconstruction, EP 489, amended by Fraenkel, SZ 54 (1934) 312f.); ef. Gai. 4.150ff., 160. 260

The origin of the possessory interdicts was framed so as to give possession not to the person who was found to have it at the moment when proceedings commenced, but to the one who had had it for the greater part of the last year. It is obvious therefore that this interdict might be used to regain possession — if e.g. A had had possession for the last five months, but B for the seven months preceding, then B could recover from A. Otherwise the effect was similar to that of uti possidetis.

(ii) Interdicta reciperandae possessionis causa comparata. Of these there were two, or perhaps at one time three. (a) Unde vi.1 This enabled a man to regain possession of land from which

he had been ousted violently. It had two forms according as the violence complained of had been ‘ armed’ or ‘ ordinary ’. The chief difference was this: if A complained that B had ousted him by ordinary violence, it was sufficient answer for B to show that A himself had previously got possession vt, clam or precario from B; if A complained of armed violence such defence was not open to B. (b) De precario.? This enabled the man who had let another have a thing precario to get it back. (c) Perhaps there existed at one time an interdict de clandestina possessione enabling a person who had lost possession by the clandestine entry of another to get it back. As the two other recuperatory interdicts correspond

to two of the limbs of the clause nec vi nec clam nec precario found as a defence in other interdicts, so it is thought that there must have been one corresponding to the third limb.°

2 ORIGIN OF THE POSSESSORY INTERDICTS Of the origin of the possessory interdicts nothing is known for certain. Niebuhr’s* theory, which is still dominant, is that they were originally developed for the protection of the interest in public land which settlers had acquired by occupation; such interest was, of course, not ownership, and consequently an occupier deprived of his enjoyment could not avail himself

of the vindicatio; the praetor therefore, according to this view, provided protection by interdict. It is true that persons who occupy public lands are frequently referred to as possessores, but nevertheless there are serious objections to the theory. As |

is pointed out by Jhering,® a dispute about ager publicus would not be a 1 Gai. 4.154; Lenel, HP 461ff. 2 Cf. above, 230. 8 Cf. Kaser, HB 252. 4 B. G. Niebuhr, History of Rome (tr. J. C. Hare and C. Thirlwall, London, 1855) 2.142ff. Followed e.g. by Savigny, Possession (tr. Sir E. Perry, London, 1848) § 13, and more recently Kaser, HB 243ff. 5 Der Besitzwille (Jena, 1889) 124 n 1. He also points to the existence of the interdict de loco publico fruendo (D. 43.9), which would have been superfluous if the praetor had already protected the tenants of public land by the possessory interdicts, and is only 261

Private law to the fall of the republic: property matter for the jurisdiction of the praetor, but for administrative action on the part of the magistrates, in particular the censors, whose duty it was to manage state property, and, above all, it is difficult to see how a remedy, originally intended for the protection of legitimate interests in public land, could have subsequently been used for the totally different purpose of protecting mere de facto possession of private property. Jhering’s own view ° is that the original purpose of the possessory interdicts was to form a substitute

for the procedure of vindicias dicere in actions in rem, so that instead of granting interim possession according to his own discretion, the praetor now caused the question who was actually possessor to be investigated by means of the interdict procedure, and saw that the party who was successful in the interdict remained in possession until the action was decided. As regards ut? possidetis and utrubi this corresponds closely with what Gaius says about the use (though not expressly about the original purpose) of these interdicts,’ and to the origin suggested for quorum bonorum.®

But the same arguments do not apply equally to the interdicts for the recovery of possession. So far at least as unde vi is concerned, the obvious suggestion is that it began as a police measure, and, as such, it might even when first introduced have been available against persons who disturbed existing enjoyment of Jand without regard to the question whether such land was or was not capable of ownership ex iure Quiritium. The de facto nature of possession is often stressed by the jurists,’ and legal niceties would have no place in what was originally a matter of public order.? explicable on the assumption that he intervened late in what had been a censorial preserve, and therefore spoke not of possidere but of fruwi. To this type of argument Kaser (ZB 271) answers that there is nothing unusual in the praetor’s concerning himself with such matters, as is shown by his interdicts concerning public roads and rivers (D. 43.10.12), and that the definite lease of public land protected by the interdict de loco publico fruendo is quite different from the holding of ager occupatorius. But to this one can reply that there is a yet greater difference between preventing interference with roads and rivers used by people ‘ promiscuously’ and protecting individuals who hold a part of state-owned land in their private interest; the praetor’s intervention in the former type of case would still leave his concern with the other anomalous. 6 Uber den Grund des Besitzschutzes (2nd ed., Jena, 1869) 76ff. 7 4.148; ef. J. 4.15.4; D. 41.2.85. 8 Above, 255. Karlowa (2.318) objects that the assignment of vindiciae came right at

the end of the old procedure in iwre as described by Gaius (4.16) and that it would not be possible to insert at this point a possibly lengthy investigation of the facts of possession. Somewhat similar arguments are used by Ubbelohde, in Gliick’s Hrlauterung der Pandekten, Serie der Biicher, 43—44.5.630ff. Cf. Cuq, RHD (1894) 12ff. Probably, however, possession would only be disputed in exceptional cases, to which Gaius was in no way bound to refer in his summary account of an obsolete procedure. 9 E.g. Labeo’s famous derivation, D. 41.2.1 pr.; cf. h.t. 1.3, 4; 29. 1 Another view is that it was simply a question of the need for a more summary remedy than vindicatio where a person’s possession was interfered with, and that this need became noticeable when the extension of Roman territory made disputes concerning 262

The actio Publiciana and the development of ‘ bonitary ’ ownership For the dates at which the different interdicts were introduced we have little evidence. Terence, writing in 161 B.C.,? uses vi, clam and precario together, evidently supposing that his audience would recognise the legal connexion, though we do not know which interdict or interdicts he had in mind, and before the end of the second century probably all those known later had been introduced. At any rate unde vi is attested in the lex agraria of 111 B.C.,° and uti possidetis evidently served as a model for three arbitra-

tions between Greek cities in the second half of the second century. We cannot doubt that utrubi was known too by then, but a supposed reference which would date it back to Plautus ° (c. 200 B.C.) is quite uncertain.

38. THE ACTIO PUBLICIANA AND THE DEVELOPMENT OF ‘BONITARY’ OWNERSHIP The actio Publiciana is the action which is allowed under the praetorian edict to a man who, after possessing a thing in such circumstances that he would

in time usucape it, loses possession of it before the time for usucapion has elapsed. Such a man can clearly not bring a vindicatio because he has not yet become owner, but the praetor desires that he should be able to get the thing back from subsequent possessors who have no better right than himself and therefore permits him to bring this action, the formula of which contains a fiction ® that the time for usucapion has already run, i.e. the judge is told to condemn if the thing in question would have become the property of the plaintiff ex ture Quiritium had his possession lasted long enough. It must be noticed that the fiction only extends to the lapse of time; whether the plaintiff’s possession was such as to lead to usucapion the judge must of course investigate. Now an action of this sort would benefit two classes of possessors who had not got good title to the thing they possessed,’ (a) the possessor who was not

owner because the person from whom he had acquired the thing was not owner, and (b) the possessor who was not owner because of some defect in the method by which the thing had been acquired, in particular, where a res mancipi had been received by mere traditio. The position, however, of these land at once commoner and more difficult to settle quickly; Karlowa 2.318. If the need for more summary process was the driving force, the case is closely parallel to that of the English possessory assizes; Holdsworth, HEL 3.8.

2 Eunuchus 319. 3 § 18. FIRA 1.106.

4 Dittenberger, Sylloge Inscriptionwm Graecarwm (3rd ed.) 679, 683, 685 (with Supplementum Epigraphicwm Graecuwm 2.511; see Cary, JRS 16 (1926) 194ff.). 5 Stichus 696, 750; ef. Watson, Property 86f. 6 Gai. 4.36. For formulae with a fiction cf. above, 209. * Of. above, 154f., on the functions of usucapio.

263 J.—10

Private law to the fall of the republic: property two classes is not the same, because possessors of the first class (whom we will call * bona fide possessors ’) will be protected only against persons who

have no better title than themselves, whereas those of the second class (whom we will call ‘ bonitary owners ’) will be protected even against. the Quiritarian owner. The way in which the praetor brings this result about is best seen by examples.

(i) Bona fide possessor. If A sells and delivers to B a thing which in fact belongs to X, and B receives it in good faith, B (provided there has been no

theft) will become owner in one or two years (according as the thing is a movable or immovable). If now before the time has elapsed the thing gets into the hands of C, B can get it back by the actio Publiciana;® but if it gets

back into the hands of X, its real owner, the actio Publiciana will not be successful, because, although it is true that B would have been owner had his possession continued long enough, X is allowed to plead his ownership as a defence. X puts into the formula an exceptio iusti dominii, i.e. a clause telling the judge not to condemn even if B would have been owner had his possession lasted long enough, if in fact X is owner.® It is, of course, also open to X at any time so long as B possesses and the time for usucapion has not elapsed to bring a vindicatio against B.

(ii) Bonitary owner. If A sells and delivers (tradit) to B a slave (res manctpt) which is his (A’s) property, B does not become owner at once ex iure Quiritium, but will do so in time by usucapion. If he loses possession to C he can therefore use the actio Publiciana in the same way as the mere b./. possessor. But unlike the mere b./. possessor he can use the actio Publiciana successfully even against A (who has remained owner ex ture Quiritium) if the slave happens to get back into A’s possession. It is true that A can put in the exceptio wustt dominii, but B will meet this by a replicatio rei venditae et traditae,' i.e. the judge will be told to condemn even if the thing is the defen-

dant’s (A’s) property if it be further shown that the defendant sold and delivered it to the plaintiff (B). If A tries to take advantage of his Quiritary ownership and bring a vindicatio against B then B will use the exceptio rei venditae et traditae,’ i.e. the judge will be told not to condemn even if the 8 In some cases of course a possessory interdict would be available, but by no means in all, 9 D. 6.2.16; 17.

1 The existence of this replicatio is not directly attested, but can be inferred with certainty from the exceptio r. v. et t. The exceptio iusti dominii is only given causa cognita, according to D. 17.1.57.

2D. 21.3. As its name shows, the exceptio was originally only intended for the commonest case, i.e. where the delivery was on account of a sale, but protection was also given, apparently by allowing exceptio dolt, where there was some different ground, e.g.

an exchange; D. 44.4.4.31. The exceptio was available to successors in title of the transferee and against successors in title of the transferor; D. 21.3.3. 264

The actio Publiciana and the development of ‘ bonitary ’ ownership property belongs to the plaintiff (A) if the plaintiff sold and delivered the thing to the defendant. In fact the praetor treats failure.to mancipate a res mancipi as a mere technicality and regards anyone who seeks to rely on it as unworthy of consideration; the recipient he supplies with such remedies that

he in practice has all the security and very nearly all the advantages of ownership. In such a case, where the advantage of ownership was with B and

the mere nominal right ex wure Quiritium with A, B was said to have the thing among his goods — in bonis — from which phrase the Greek writers subsequently coined the word ‘ bonitary ’ to express this type of ownership.’

Like the ‘ praetorian will’, this ‘ bonitary’ or ‘ praetorian’ ownership exemplifies the dislike of the praetorian system for the old formal requirements of the civil law, requirements whose importance it greatly diminished by providing remedies for those whose rights were defective because some piece of ceremonial had been omitted. Obviously once the position described above had been reached there was little danger if a buyer did not insist, when buying a res mancipt, on having it conveyed to him by mancipation or in ure CeSst0.

By the end of the republic this development was certainly not yet complete, though it is generally supposed that it had at least begun. But who the Publicius was who introduced the Publician action we do not know,* and the earliest surviving text ° to mention it is from Neratius, who lived in the years around A.D. 100. There has been controversy as to which of the two pur8 The word is first used by Theophilus, Paraphr. 1.5.4. Gaius is alone among classical lawyers in calling it ownership (Gai. 1.54: duplex dominiwm; cf 1.167: ex ture Quiritiwm tua, in bonis mea); other jurists say in bonis habere or esse (as indeed Gaius also does: 2.40, 41; hence some scholars have argued that the other passages are not Gaian; David—Nelson ad 1.54). Gaius mentions only the case of res mancipi transferred by traditio, but it is clear that other cases were included in in bonis habere (e.g. where a man has been put in possession on the ground of damnum infectum; above, 227); see Buckland 195. No doubt the formula of the actio Publiciana would in such cases differ from that given by Gai. 4.36, which refers to the common case of delivery on sale; similarly the exceptio rei venditae et traditae would have to be altered or replaced by the exceptio dol. Although Modestinus (D. 41.1.52) says generally, ‘We are deemed to have a thing in bonis nostris whenever we have an exceptio if we possess it and an action to recover it if we lose possession’, the term was probably restricted to cases where the

protection was available against everyone, including the true owner; i.e. it did not extend to the case of the bona fide possessor on the way to usucapion; Kaser, SZ 78 (1961) 177ff.

4 J. 4.6.4 says that the action was called ‘ Publician’ because it was first put in the cdiet by a praetor named Publicius, and it is commonly taken as probable that this was the Quintus Publicius who was one of the praetors in 67 B.C. But even if he is not ruled out by his having apparently been praetor peregrinus (Cicero, Cluent. 126, mentions him second), there is nothing to connect him with the action and there may have heen many other Publicii who held the praetorship; see Watson, Property 104ff. It is

also possible that Justinian is mistaken and that the name is that of a jurist who suggested the formula as adviser to a party. 5 D. 6.2.17; ef. ht. 9.3. 265

Private law to the fall of the republic: property poses of the action mentioned above was the original, but the evidence is insufficient for more than conjecture.* It can indeed be argued that the first step was the granting not of the action but of the exceptio rei venditae et traditae. To give a transferee a defence against the transferor who attempts to rely on his purely technical title seems a more obvious piece of equity than

to provide the transferee with means of attacking third parties, but such a priori reasoning cannot carry us very far.

More recently it has been argued by Wubbe’ that the earliest purpose of the action was neither of the two mentioned so far. We have seen * that according to the view maintained principally by Kaser the legis actio sacra-

mento in rem asserted only a relative ownership, whereas the formulary vindicatio required the plaintiff to prove absolute ownership. Wubbe contends that the actio Publiciana was introduced to play the part in the formulary system which had been played in the legis actio system by the sacramentum in rem. Every plaintiff who had not yet had possession of the thing

for the period of usucapion would bring the actio Publictana, even if he believed himself to be Quiritary owner, because he needed only to show possession ex iusta causa in order to put on the defendant the burden of proving that he (the defendant) was owner.’ Now it may readily be conceded that a bona fide possessor cannot know that he is one, and therefore that the plain-

tiff who had not acquired the thing by an original mode and had not had possession for the period of usucapion would have been well advised to bring the actio Publiciana rather than the vindicatio, and thus to guard against the possibility that his belief in the soundness of his title was misplaced. No text says that this was in fact done, but the tactic is so obvious that it can hardly have failed to occur to the lawyers of the day. Wubbe, however, maintains

that this was the prime purpose of the praetor in introducing the action in the first place, whereas plainly for Gaius the paradigm case is that of the man

who does need to usucapt, but has not yet done so.' Moreover, Wubbe’s thesis that the defendant must show a better title than the plaintiff is hardly 6 De Visscher, SDHI 22 (1956) 87ff., 23 (1957) 26ff.; RIDA (1958) 469ff., has most recently maintained the thesis that the original purpose was to protect the acquirer of res mancipi by traditio (ef. his views on the original function of usucapion, above, 155 n 9); see also reff. in next note.

7 Res aliena pignori data (Leyden, 1960; in Dutch, but with summary in German; review by Beinart, Iura 12 (1961) 387ff.); RIDA (1961) 417ff. Of course justice cannot here be done to the elaboration of Wubbe’s argument. See Kaser, SZ (1961) 173ff. 8 Above, 142. ® Wubbe, SZ 80 (1963) 175ff.

1 Wubbe contends that Gaius, when he says (4.36) that the actio Publiciana is given to one who has lost possession of a thing which he received ex tusta causa but has not yet usucapted (nondum usucepit), means merely that he has not yet had possession for the period of usucapion, not that his title is necessarily in any need of perfection by usucapion.

266

Ownership of provincial land compatible with what we know of the formula of the action.? Gaius* says that the intentio ran: ‘ Supposing that the plaintiff had possessed for a year

the slave who was bought by and delivered to him, if it would then be correct in law for that slave, who is the subject of this action, to be his by Quiritary right .. .’ The fiction relates only to the lapse of time, the other requirements of usucapion being left, it would seem, to be proved in the usual way. No doubt, as we have seen,* the plaintiff could not be expected to

prove that he was in good faith or that the thing had not been stolen, but it is difficult to believe that the defendant could not seek to show the absence of these requirements unless he was also prepared to show that he had a better title.

4, OWNERSHIP OF PROVINCIAL LAND In classical law it was settled doctrine that land in the provinces belonged to the state and could not be owned ex ture Quiritiwm by any private individual, Roman or peregrine.* Exception was only made in the case of a few favoured localities, where the soil was assimilated for legal purposes to that of Italy, by special imperial grant of the ius Italicum.° In spite of this rule, however, rights heritable and alienable were in fact enjoyed by individuals in provincial land, and differed in practice from full Roman ownership mainly by reason of the liability to one form or another of land-tax which attached to the soil. For the protection of this quasi-ownership the ordinary vindicatio was of course not available, but an action analogous to the vindicatio was certainly given under the provincial edicts.’ The doctrine of state ownership seems indeed to have no substantial consequence, and Jones ® has argued that it is merely a lawyer’s explanaticn of a change in practice. In the early empire the government ceased to draw the

conclusion that if citizenship were given to a community its soil became 2 Kaser, SZ 78 (1961) 190ff.

3 4.36: St quem hominem Aulus Agerius emit is e1 traditus est anno possedisset, tum si ewm hominem de quo agitur etus ex ture Quiritium esse oporteret. For the translation of oporteret see Daube, Forms 15, 19. 4 Above, 155. Of course, questions of the burden of proof were little regulated in the classical period and the tudez had a very wide discretion (including the discretion to wash his hands of the suit and swear sibi non liquere; cf. Gellius 14.2.25) ; see Pugliese, RIDA (1956) 349ff. Against Sturm’s contention (RIDA (1962) 357ff.) that proof of ownership was simple see Kiefner, SZ 81 (1964) 212ff.

5 Gai. 2.7: In provinciali solo... dominiwm populi Romani est vel Caesarts, nos autem possessionem tantum vel usumfructum habere videmur; ef. 2.21. 6 Below, 345 n 2. 7 The formula is uncertain. Lenel, EP 189, conjectures Si paret Ao. Ao. fundwm quo de agitur habere possidere frui licere, mainly on the ground of Gai 2.7 (above), and the occurrence of similar words in the lez agraria of 111 B.C. (Bruns 1.73-89), lines 32, 40,

50, 52 and 82. ’ Studies 143ff. (=JRS 31 (1941) 26ff.). 267

Private law to the fall of the republic: property Roman. People then transferred possession in such land by traditio, but this could never ripen into ownership by usucapio. The missing dominium was therefore ascribed to the state.

5 OWNERSHIP BY PEREGRINI Peregrint (unless they had commercium) could not own property ex iure Quiritium, and the ordinary forms of vindicatio could therefore not be used by them; nor could the Publician action, for it involved the proposition that the plaintiff would have acquired ownership ex iure Quiritium had he possessed long enough. There can be no doubt however that their property was in fact protected, perhaps by actions with a fiction, and their relation to such property must therefore be classed as one of the inferior modes of ownership known to Roman law.”

6. TURA IN RE ALIENA A. PRaEDIAL SERVITUDES. By the end of the republic the number of servi-

tudes recognised by law had grown considerably, but it was not, and never became, the rule at Rome that parties could attach the characteristics of a servitude to any restriction on the ownership of land that they chose.’ Of the

rustic servitudes which appear on Justinian’s list? (beyond the original rights of way and aquae ductus), aquae haustus was already known to Cicero as well established,* pecoris ad aquam adpulsus to Trebatius,* and others may be equally old. At no time was there, it appears, a special pattern formula in the edict for any rustic servitude except the rights of way and aquae ductus ;* the formula would have to be drafted specially in each case. Urban servitudes, too, had long been in existence. The need for these came, as has been said,® with the growth of urban life and the raising of buildings above the ground floor, a practice which goes back at least to the end of the third century B.C.’ A number of the servitudes known to the later law are actually attested by republican writers; thus stillicidiwm (the right to let rain-water drop onto a neighbour’s land) and flumen (where the water flows in a continuous stream) are mentioned by Varro,® and Cicero treats of ura parietum, luminum, stillicidiorum as three definite groups.” ® Buckland 190; Girard 380; Monier 1.366. 1 Schulz, CRE 383; Watson, Property 176ff.; and see Buckland 259ff. for the general principles to which any servitude must conform.

2 J. 43.20.1.18. 2.3 pr. and 52.Lenel, 3 Caec. 74. 4D. EP 192. 6 Above, 157f. 7 Livy 5.55.

8 LL. 5.27. 9* De Or. 1.173. 268

Iura in re aliena B. Ususrructus anp usus. Usufruct ' is the right of using and taking the fruits of property belonging to another, salva rerum substantia,’ i.e. without the right of destroying or changing the character of the thing, and lasting only so long as the character remains unchanged.’ It is usually for the life of the person entitled and cannot be for a longer period, and it thus corresponds most closely to what we call a ‘ life interest ’. Usus is a similar right, but to the use of the thing only, not including the taking of fruits. Rights of this nature are classed in the law of Justinian’s day as ‘ personal servitudes ’,‘ because they exist for the benefit of individuals as such, without reference to the ownership of any particular piece of property, whereas praedial servitudes enure to the advantage of the owner for the time being of a particular piece of land, but this classification we know now to date from the end of the

classical period at the earliest.*° They can exist over either movable or immovable property.

We do not know precisely when either of these rights came to be recognised, but usufruct was well known to Cicero,* and we hear of a question concerning it that was debated by the jurists of the middle of the second century B.C.’ There can be little doubt that it occurred first as the result of legacies, and legacy remained always the main source of the ‘ personal servitudes ’. Usus is probably of later origin, for the jurists treat it as a variation of the more normal usufruct and apply the rules worked out for usufruct to

it by analogy.* The normal, and no doubt the earliest, method of creating these ‘ personal servitudes * was legacy. C. CONDUCTIO AGRI VECTIGALIS AND SUPERFICIES. The classical law knew of

two further rights in relation to land, both arising from a sort of lease, which had already through praetorian legislation almost completely attained to the status of rights in rem, a process which was completed in Justinian’s time. Conductio agri vectigalis (which became emphyteusis under Justinian) arose from the grant of a heritable but probably not alienable lease of land, by the state or municipalities in return for a comparatively small rent, payable by 1 Pugliese, Tulane L.R. 40 (1966) 523ff. 2 J. 2.4 pr. 3 Buckland, Manual 162, but see also Main Institutions 145ff. 4D. 8.1.1. The Institutes, following Gaius, distinguish ‘servitudes’ from usufruct ; J. 2.2.2 and 3; Gai. 2.14. 5 Buckland 268; St. Riccobono 1.277ff. It is not very apt; Nicholas, Introduction 1444, 6 E.g. Caec. 19.

* Cic. Fin. 1.12; D. 7.1.68. For possible references in Plautus see Watson, Proper! y 203ff. Against the common suggestion that it arose as a method of providing for a widow where marriage was without manus, Watson, TF 31 (1963) 614ff. 8 D. 7.8.1.1; Karlowa 2.540. 269

Private law to the fall of the republic: property the tenant for the time being (and thus constituting a burden on the land). Superficies was a somewhat similar grant for building purposes. The practice of granting leases of this sort had apparently begun in republican times and was apparently extended to private land in the course of the classical period.’ 9 See Mitteis, Zur geschichte der Erbpacht (Abh. Sdchs. Ges. 20, 1901); F. Lanfranchi, Studi sull’? ager vectigalis 1 (1938); 2 (Ann. Univ. Camerino 13, 1939) ; 3 (Ann. Triestint 11, 1940); Kaser, SZ 62 (1942) 34ff.; L. Bove, Ricerche sugli agri vectigales (Naples, 1960) ; and on later developments Levy, Property 43ff., 774.

270

CHAPTER 17

Private law from the XII Tables to the fall of the republic: the law of obligations In tracing the history of obligations it is necessary, more even than in other

parts of the law, to distinguish the growth of theoretical conceptions, of general rules and of classifications, from the development of the actual procedural possibilities which underlay the generalisations of theory. So far as the former are concerned the end of the republic still finds the Romans comparatively backward. The word obligatio itself is not yet a technical term,’ and indeed even in the classical period, when it had become technical, it did

not correspond to the Byzantine and modern conception, for there still remained about it traces of its original literal meaning. Nexum, it is true, had long fallen into disuse, but a man who was ‘ under an obligation ’ might find himself quite literally bound, if he did not discharge it, so long as execution against the person remained a normal thing. Further, in the dominant classical usage obligatio was a purely civil law word, and did not include those cases where a man was liable because a praetorian action lay against him.? With other familiar words the position is similar. ‘ Contract ’ seems to

us a fundamental conception, denoting the formation of an obligation by agreement, and this was the Byzantine meaning also; but the Latin for an agreement is conventio or pactum,* and it was by no means every agreement which gave rise to an obligation. Contractus is simply the verbal noun formed from contrahere (literally, ‘ to tie’), and it is used elliptically for contractus negotu or obligationis, whether what we call a ‘ contract’ was involved or not. Logically there was no reason why it should not be used of obligations contracted by delict, but usage seems to have confined the word to negotia (transactions), in opposition to delict.* At the end of the republic indeed ter1 The verb obligare is old in juristic use, e.g. Varro (Z.L. 7.105), quoting Mucius: quae per aes et libram fiant ut obligentur (above, 164), but it is not confined to ‘ obligations’. Plautus uses it of pledge (Truc. 214: aedes obligatae sunt ob amoris praediwm), as indeed the classical lawyers still do (e.g. D. 20.4.21). The noun obligatio is rare before Gaius; Cicero uses it, ad Brut. 1.18.3, but without any definite legal implication; see Schulz, CBRL 455.

2 Buckland 409; cf. 8. Perozzi, Le obligazioni romane (Bologna, 1903) 135ff. (also in his Scritti 2.422ff.). The usage was not rigid; G. Segré, Ser. vari di dir. rom. (Turin, s.d.) 345ff. (=St. Bonfante 3.571ff.). In particular, Gaius’ treatment of delictal obligations includes not only praetorian developments of the civil law, but also a wholly praetorian creation, rapina (below, 277) ; ef. Schulz, CRL 457ff. 3 Bonfante, Scritti 3.136ff. 4 Ibid. 3.107ff.; Mitteis, RPR 147; Schulz, CRL 465ff. The payment of an indebitum

would be as much a negotium as the payment of money by way of loan, and it is clear 271

Private law to the fall of the republic: obligations minology had not got even as far as this, for, though Cicero uses the word contrahere, he does not use the noun.” Still less was there any classification of obligations into those arising from contract and delict, or of contracts according to their different methods of formation. In the discussion which follows, the usual classification under contract and

delict will be preserved for the sake of convenience, but it must be remembered that the terminology is in advance of the time of which we are speaking, and that contract often grows up out of delict. It will also be necessary to discuss some obligations usually classed under quasi-contract in close connexion with contract.

1 DELICT A. Tue C1vit Law DELIcTS AND THEIR EXTENSIONS. (i) Iniuria. (a) Praetorian

reforms. These are here of a twofold nature: (1) the praetor substitutes for the old actions, involving talion or fixed penalties, actions of his own in which the penalty is always pecuniary and is always variable at the discretion of the court according to the gravity of the offence (actio aestimatoria); (2) he makes actionable a number of offences which were unpunished by the XII Tables, thus carrying the conception of iniuria far beyond that of

physical assault. |

(1) First under the heading de iniuriis in the praetor’s album stood a

‘ general edict ’ promising a court of recuperatores to assess damages where iniuria was alleged.® It is probable that the original intention of this edict,

traces of which are found as early as Plautus,’ was not to extend redress beyond those cases already covered by the XII Tables,® but merely to modfrom Gai, 3.91 that the two cases were traditionally treated together as ‘ contracts made re, though Gaius himself criticises the arrangement. In his Institutes (3.88) Gaius only mentions two categories of obligations, ex contractu and ex delicto, and it is easy to see that, so long as contractus retained its original significance, this classification was an exhaustive one. In the Aurea according to the Digest (44.7.1 pr.) Gaius added a further miscellaneous category of obligations arising ex variis causarum figuris, but this passage is suspected of interpolation. Even stronger suspicion rests on D. 44.7.5, where quasi-contractual and quasi-delictal obligations are mentioned. 5 Costa, Cicerone giureconsulto 1.202. How different even Labeo’s terminology was from that to which we are accustomed can be seen from D. 50.16.19. 6 The plaintiff assessed the damages himself and the court could condemn for the sum claimed or less; where the iniwria was atrox the praetor in effect fixed the sum (Gai. 3.224), There is considerable difficulty about the composition of the court. Gellius

(20.1.13) says clearly that the edict promised recuperatores, but Gaius speaks of a single tudex and so do others, e.g. Rhet. ad Herenn. 2.13.19; see Lenel, HP 397; Girard, Mélanges 2.385ff. (=Mél. Gérardin 255ff.). Sehmidlin, Rekuperatorenverfahren 29ff., points out that the context of Gellius’ statement is the anecdote of L. Veratius (see text, below) and argues that recuperatores were confined to physical iniwuriae.

7 Asinar. 371, where pugno malam si tibi percussero seems to be a humorous allusion to the pattern formula; Lenel, BP 398. 8 Daube, CLJ 7 (1939) at 45ff. argues that the edict did not originally cover mem272

Delict ernise the procedure and the penalty. No doubt the fall in the value of money had helped to make the old fixed penalties ridiculous, as is illustrated in the

anecdote of L. Veratius, who is said to have amused himself by slapping people in the face and then ordering a slave who followed him with a bag full of money to pay each of them the twenty-five asses fixed by the XII Tables.’ (2) After the general edict *' came two special edicts dealing with particular cases of non-physical affront — convicitum ? (public insult), adtemptata pudi-

citia (e.g. following a respectable woman) and a wider provision (ne quid infamandi causa fiat *) which was aimed at defamation but which was appar-

ently limited to such conduct as was calculated to bring infamia* upon the plaintiff.* There was also a clause dealing with iniuriae to slaves and persons under power. The penalty was, of course, always assessable. That these special edicts were later than the general one is probable, but well before the end of the republic the meaning of iniuria itself (i.e. under the general edict) had been extended far beyond that of physical assault,® and in the classical law it is a very wide conception indeed, covering not only our law of assault and defamation but including any act which could be construed as an attack on a man’s honour or dignity, even e.g. a refusal to allow him to use the public baths.’ brum ruptum or os fractum but only the lesser assaults covered by the old jtniuria; see Watson, Obligations 248ff.

® Quoted from Labeo by Gellius, 20.1.138. Birks, TH 37 (1969) 175, questions the verisimilitude of the story on the ground that 25 asses was too large a sum for even 4 rich man to throw away; but the calculations of F. R. Cowell, Cicero and the Roman Republic (London, 1948) 104ff. (which Birks adduces in support) seem to show that 25 asses in the early second century was about ;35 of the annual cost of living of a free labourer. H. F. Hitzig, Iniwria (Munich, 1899) 60ff., finds Greek models for the praetorian action, but they are probably no more than parallels: G. Pugliese, Stadi sulViniuria (Milan, 1941) 39ff.; Schmidlin, Rekuperatorenverfahren 32 n 3. 1 See Girard, Mélanges 2.385ff.; Watson, Obligations 250ff. 2 D. 47.10.15.2. 3D. 47.10.15.25.

4 Those who were found guilty of certain offences, or who in actions arising out of certain relationships were found to have fallen short of the standard of good faith required, or who were found to have acted dishonourably in certain other ways were infames and suffered certain disabilities in legal proceedings (Gai. 3.182), were disabled from holding certain offices, and, no doubt, suffered social disabilities also; see Kaser, SZ 73 (1956) 2208.

5 So Daube, Atti Verona 2.413ff. He shows that the pattern formula in the edict (Coll. 2.6.5) must have taken the case of capillum inmittere infamandi causa, i.e. going

about with hair dishevelled (a sign of mourning) to suggest that the plaintiff had behaved dishonourably, and relates this to Seneca, Controv. 10.1.30, where the defendant

to an action for iniuria has acted similarly and the plaintiff has been refused public office.

6 Rhet. ad Herenn. 2.26.41 already (early first century B.C.) rejects the view that it did not extend beyond blows and convicitum, and Labeo, D. 47.10.15.26, held the special edict ne quid infamandi causa fiat to be superfluous. 7 D. 47.10.13.7. ©

273

Private law to the fall of the republic: obligations (6) Lex Cornelia. This statute of Sulla’s (about 81 B.C.) provided specially for three cases of forcible tnturta — pulsare (blows), verberare (explained as blows which hurt) * and vi domum introire (forcible entry into a house). The action under the statute was of a criminal nature in so far as it was tried by a quaestio,® but counted as private ? and could be brought only by the per-

son wronged, whereas anyone could prosecute in a criminal action. The penalty is not known for certain, but was probably pecuniary.” It has been

held that, until late in the classical period, no ordinary praetorian actio aestimatoria could be brought in the cases covered by the statute,? but the better view appears to be that the statute merely provided an alternative procedure for those cases.

(ii) Furtum. How long the system of the XII Tables remained: untouched we do not know, but it may have been until towards the end of our period, for there is no trace of the praetorian edicts on the subject before Labeo.* The praetor, according to Gaius,® retained the civil action for double value in case of furtum nec manifestum and the actions concepti and oblati for triple value; for furtum manifestum he gave an action for fourfold. The old search lance et licio evidently became obsolete and only the simple form survived, refusal to permit it being sanctioned by the praetorian actio furti pro-

hibitt for fourfold.* Justinian mentions also a praetorian actio furti non exhibiti’ which lay against a person who refused to ‘ exhibit ’ a stolen thing

found after search. This action is not mentioned in any other text and the penalty is not known. The right to kill the thief who came by night or who defended himself with a weapon by day no doubt continued to exist; indeed 8 D. 47.10.5.1.

® This appears from the remains of the rules concerning the composition of the court preserved in D. 47.10.5 pr.

1 D. 3.3.42.1; Pugliese, Studi sull’iniuria (above) 117ff. If Kunkel’s account of the early history of quaestiones (below, 310ff.) is right, this mixed character is a survival. 2 Mommsen, Strk 804, relying on D. 47.10.37.1, holds that the plaintiff fixed the sum, the court being bound either to condemn for this sum or acquit. 3 Girard, Mélanges 2.406ff. (=Mél. Gérardin 278ff.). But this view seems inconsistent

with Gaius’ treatment (3.220) of verberatio as an ordinary case of iniuria without any mention of the lex Cornelia; see Lenel, EP 397. 4 P. Huvelin, Etudes sur le furtuwm (Lyons/Paris, 1915) 567.

5 3.190, 191. 6 Above, 169.

7 J. 4.1.4. Theophilus (Paraphr. ad h. 1.) relates the action to a refusal to hand the thing over on the spot, but the word exhibere (cf. actio ad exhibendum) suggests rather a refusal to produce it in ture (Kaser, 2B 43f.). In neither case, however, is it easy to see a compelling need for such an action in addition to the actiones furti concepti and ad exhibendum. For discussion see Yaron, TK 34 (1966) 519ff., who ingeniously suggests that the action lay for the searcher’s failure to find the thing, i.e. for a false accusation. This involves, however, the supposition that the original text was different and clumsily ambiguous (qui furtivam rem [apud se] quaesitam [et] inventam non exhibuit) and it requires a forced meaning for exhibere. 274

Delict

self-defence.’ :

it was probably not until post-classical times that it sank to a mere right of (iii) Damage to property.® The date of the lew Aquilia, which became the

foundation of the law on this subject, is not known. It was certainly later than the XII Tables, for Ulpian says‘ that it partially repealed their provisions, and it would appear to have been subsequent to 287 B.C., for it was strictly not a lex but a plebiscitum,? and a plebiscite would probably not have had the force of law before the passing of the lex Hortensia in that year. On the other hand it was evidently earlier than the middle of the second century B.C., for it was commented on by M. [unius Brutus.*

We know of three chapters of the law, but only the first and third concerned damage to property.* The first dealt with the unlawful killing of slaves or four-footed pecudes * belonging to others, the third with the causing of loss (damnum *) by * burning, breaking or smashing’ any other property.

The third chapter, at any rate as interpreted later, included also minor injuries to slaves and pecudes. Under both chapters the act must have been done iniuria.” This meant originally non ture, in the sense that once it was proved that the defendant had caused the damage (in the appropriate way) he was liable unless he could show a recognised justification, such as selfdefence. At the end of the republic, however, the meaning is shifting to a requirement that the defendant should have acted dolo aut culpa (maliciously

or negligently). The measure of damages under the first chapter was the highest value the slave or animal had borne during the preceding year; under

the third chapter the period of time was different —thirty days — but the 8 Cf. above, 169. That D. 9.2.5 pr. has been interpolated is shown by comparison with Coll. 7.3.2-3; see Lawson, Negligence 82.

8 Lawson, Negligence. 1 D. 9.2.1 pr.

2 D. 9.2.1.1.

3 D. 9.2.27.22. For Brutus see above, 92. The lex is commonly dated to 287 B.C. on the basis of two remarks by Byzantine authors. Theophilus (Paraphr. 4.3.15) says ‘ at the

time of the secession’ and the Scholiast on B. 60.3.1 (Heimbach 5.263) says ‘ when the plebs was in revolt from the senate and seceded from it’. As the latest secession was

in 287 it is concluded that this was the date. But (a) the law is of a non-political

character (contra, Beinart, Butterworth’s S. African Law Rev. (1956) 7O0ff., who sees a number of its provisions as redressing grievances of the plebs), (b) Theophilus and the Scholiast probably had no independent evidence, but, not knowing much early history, jumped to the conclusion that the lex was connected with political strife because it was a plebiscite.

4 For the second, see Gai. 3.215; ef. Lévy-Bruhl, RIDA (1958) 507ff.; Daube,

St. Solazet 154ff, 5 Explained, J. 4.3.1, as ineluding all beasts which graze.

6 For damnum as referring to pecuniary loss to the owner and not to the physical damage to the thing (which might not result in any loss: D. 9.2.27.17) see Daube, St. Solazzt 93ff., contra, Liebs, SZ 85 (1968) 195ff. * Beinart, St. Arangio-Ruiz 1.279ff.; Lawson, Negligence 36ff.

275

eres sites Private law to the fall of the republic: obligations

measure was still apparently the highest value. ‘ Highest ’ was not, however,

in the text of ch. m1, though it had been read into it by the middle of the

first century A.D.° The lex Aquilia is peculiarly interesting historically as it affords examples of development both by interpretatio and by praetorian action.

(a) Interpretation. According to the text of the statute only the actual value of the thing could be recovered by way of damages, but interpretation, going back at least as far as Labeo,° allowed the plaintiff to include consequential damages as well, if e.g. the thing destroyed was one of a pair, so 8 The fact that the third chapter apparently gave the plaintiff the full value of the thing, even in cases of minor damage, together with certain oddities in the text of the chapter as it is preserved in D. 9.2.27.5, led Jolowicz (LQR 38 (1922) 220ff.) to conjecture that both chapters were originally confined to total destruction, ch. I being concerned with animate property and ch. 111 with imanimate. Partial damage to all forms of property was included in ch. 111 by interpretatio later (but still very early, since M. Iunius Brutus (above, 92) allows an action for striking a slavewoman and causing her to miscarry: D. 9.2.27.22), and this extension led to a corresponding shift in the measure of damages—from value to actual loss suffered. The main obstacles to this explanation (the other objections of Lenel, SZ 43 (1922) 577, are less substantial) are (a) the lack of evidence for so extensive a shift of meaning (though evidence for the law of the third and second centuries is very slight), and (b) the improbability that the lex would have omitted damage falling short of killing or total destruction (unless it be assumed that such damage was covered by the XII Tables; ef. above, 170). Daube (LQR 52 (1936) 253ff.; cf. Tliffe, RIDA (1958) 493ff.) conjectured that ch. III never

gave the plaintiff full value; quanti ea res erit in the text of ch. 111 referred (as commonly in the formulae of actions) to the plaintiff’s actual loss, and the thirty days were those following, and not, as was plainly the classical interpretation, those preceding, the wrongful act. Moreover, ch. 111 was originally confined, like ch. 1, to animate

property, but dealt with damage less than killing, and the delay of thirty days had therefore a purpose —to give time for the extent of the injuries to become clear. Later (but necessarily before the late first century B.C., see Labeo in D. 9.2.29.3) the chapter was extended to inanimate things, and since the extent of the damage to such property was immediately apparent, the thirty day period came to be treated, by analogy with

eh. 1, as relating back. Only in this new context would the question of reading in ‘highest’ arise. The only objection to this very ingenious argument is, again, the improbability of all these developments being forgotten in so strongly traditional a system

as the Roman. The dominant view (Kaser, FPR 1.620; D. Medicus, Id quod interest (Cologne, 1962) 238ff.) remains that the plaintiff recovered the full value regardless of the extent of the damage. But see Daube, Roman Law, Linguistic, Social and Philosophical Aspects (Edinburgh, 1969) 66ff. There is no text (with the possible exception of, by implication, D. 9.2.24) which positively excludes this interpretation, and the actto legis Aquiliae was, it is true, like other delictal actions (above, 172), in part at least penal (Gai. 4.9), but it is difficult to believe that so grossly inequitable a system existed throughout the classical law period. Kelly, LQE 80 (1964) 73ff., also makes a radical reconstruction of the original text of the lez, and Pringsheim, Ges. Abh. 2.410ff. (=Mél. Lévy-Bruhl 233ff.), finds successive layers within it. In so far as all these conjectures are based on a minute examination of the surviving text, it should be remembered that the Roman jurists evidently did not have the same concern for scrupulous accuracy in quoting legislative texts as a modern lawyer has. ® D. 9.2.23.4,

276

Delict that the loss suffered was greater than the value of the individual thing.’

Further, the text of the third chapter ran ‘ sz quis... . usserit, fregerit, ruperit ...’, and strictly rumpere only means ‘ to break ’, but the republican jurists already interpreted it as equivalent to corrumpere, ‘ to spoil’, thereby bringing all other sorts of injury within the statute.’ (b) Praetorian actions. Apart from this extensive interpretation of rum-

pere, the words of the third chapter and the word ‘ kill’ of the first were strictly construed at civil law. Hence ‘ causing death ’, as opposed to $ killing ’, did not give grounds for an action at civil law, nor did damage done to a thing without physical contact. Already, however, in Labeo’s time a person who caused a slave’s death by giving him poison to drink was lable to a praetorian actio in factum,® and similar actions were given where the damage was not considered direct enough for a civil action under the third chapter.* Equally important was the granting of praetorian actions to persons who, although not owners, had an interest in the property (e.g. to the usufructuary), whereas the words of the statute mentioned the owner only.’ But there seems to be no evidence that this step was taken before the empire. Again, since a man could not be said to be the owner of his own limbs, he could not sue for personal injury ° (unless it were inflicted intentionally, when it would constitute the delict of znzuria).

(iv) Violent damage to property and rapina. Among the four main delicts treated by Gaius in his Institutes is robbery (rapina), which is simply theft accompanied by violence.’ For this an action lay under the edict for fourfold damages provided it was brought within the year, for simple damages if brought later. The history of this action is curious and, in some respects, obscure.* We know that in 76 B.C. M. Terentius Lucullus, when praetor, found it desirable, no doubt on account of the disturbed state of the country, to promise in his edict an action for damage (damnum) inflicted violently by armed bands (vi hominibus armatis coactisve). It is probable that this edict was intended to deal with violent damage to property and not with violent theft, which seems to have been sufficiently covered by a previous edict, that of Octavius, dating from about 80 B.C.® But before Labeo’s time a second clause was inserted which expressly dealt with violent theft, whether committed by a band or not, the reason probably being that the edict of Octavius had in the meanwhile been restricted, so as not to cover this case any more. 1 Gai. 3.212. 2 Ulpian, D. 9.2.27.13, refers the extension to the veteres.

4 Gai. 3.219. 5D. 9.2.11.6, 12.

3. D. 9.2.9 pr.

6 One text (D. 9.2.13 pr., attributed to Ulpian) allows him a praetorian action, but even this is probably interpolated. 7 Gai. 3.209.

8 For the following description see Schulz, SZ 43 (1922) 216ff.; cf. Rouvier, RHD

(1963) 443ff.; Balzarini, St. Grosso 1.323ff. 9 Cf. below, 278. 277

Private law to the fall of the republic : obligations In classical times both the clause about damage by armed bands and that dealing with violent theft were in the edict, but under more settled conditions the latter was the more important in practice and alone thought worthy of a place in Gaius’ elementary work. B. DELICTS RESTING PURELY ON THE IUS HONORARIUM. So far we have spoken

only of such praetorian innovations as were in extension or alteration of provisions already known to the civil law; something must now be said of those which gave actions where the civil law saw no wrong at all. Some of these really belong to the law of procedure and were required for perfecting the machinery of justice. Such e.g. were the provisions giving actions for damages against anyone who refused to obey the decree of a municipal magistrate,' and against anyone who forcibly removed a person who had been zn

ius vocatus.” Other actions, unconnected with procedure, but of minor importance, were that for corrupting a slave,® and that which lay against a surveyor who fraudulently made a false report on the acreage of a field.* But the most important praetorian delicts were metus and dolus, for here the praetor introduced new general principles which were to influence the law

through all its branches. .

(i) Metus. In the classical law there were three possible remedies available to a person who had been induced by threats to do some act which was detri-

mental to him—in integrum restitutio,® exceptio and action. Of these the action appears to be the oldest, and to date back to the praetorship of an Octavius about 80 B.C.* In its original form it lay, apparently, only for actual deprivation of property by violence or through menaces,’ and was for four times the value of the property. Subsequently the reference to violence was dropped, but the action was made available in all cases where a man was induced by threats to perform any act by which he suffered loss, whether this was a legal transaction such as the conveyance of property or a physical act such as pulling down a house.®

(ii) Dolus. Originally, in accordance with formalist principles, there was no redress for mere deceit any more than for threats. If a man had been cheated e.g. into making a conveyance of his property he had made it none the less, and perhaps in very early times the deceit was not even regarded as morally reprehensible.® No remedy was found until Aquilius Gallus, in Cicero’s words,

1 Lenel, EP 51. 2 EP 73.

8 EP 175; Gai. 3.198; Watson, Obligations 264f. 4 EP 219. 5 Above, 229.

6 Cicero, Verr. 11.3.152, refers to it as formula Octaviana, but which Octavius was its

author is uncertain; see Schulz, SZ 43 (1922) 217. On the date of the exceptio see Watson, Obligations 257f. 7 Quae per vim et metum abstulerant, Cie. loc. cit.; cf. ad Quintum fr. 1.1.7.21.

8 Schulz, SZ 43 (1922) 220. 9 Cf. Maine, Ancient Law 321f. 278

Contract ‘ produced the formulae de dolo’.' We cannot say for certain whether this refers to the actio or the exceptio, but the probability is that here, as in the case of metus, the action came first. It was, like the actio metus, arbitraria, but it gave only simple damages for the loss which had been inflicted by the fraud, and was available only if the plaintiff had no other remedy open to him.?

2 CONTRACT A. Format Conrracts. (i) The disuse of nexum. The lex Poetelia ®* does not appear actually to have abolished nexum, but to have led to its disuse.* As

we have seen, it in some way eased the debtor’s lot; and if, as is likely enough, the practice of enforcing the nexum without judgment had become obsolete, no advantage would remain over the newer forms of contracting an obligation which were coming into use.’

(ii) Stepulatio.® In the developed law stipulation is a form of contract by which any agreement can be made actionable by the simple expedient of reducing it to the form of an oral question and answer. The person to become promisee says e.g. ‘ Do you promise to pay me 10,000 sesterces ?’ or ‘ Do you promise that you will build a house according to such and such specifications ?? and the other answers ‘I promise.’ No witnesses and no writing are required. Whether in the classical law there was entire freedom of choice as to the words which were used is debatable,’ but originally it seems clear that 1 Off. 3.60; in de nat. deor. 3.74 Cicero speaks of iudiciwm de dolo, which points to the action rather than the exception. Aquilius was praetor with Cicero in 66 B.C., and hence the action is usually assigned to that date. The difficulty however is that he is

known (Cic. Cluent. 147) to have acted when praetor as president of the quaestio ambitus. Though he may have held the peregrine praetorship at the same time (Kubler, SZ 14 (1893) 80f.), it is more likely that it was in his capacity as legal adviser to a magistrate and not as a magistrate himself that he was the author of the formulae, in which case the date may be any time during his activity as a lawyer; see v. Lubtow, Eranion G. S. Maridakis (Athens, 1963) 185ff., with references. 2 For the formula see above, 202 n 5. 3 Cf. above, 189f.

8.28.8. 5 Kaser, AJ 247ff. 4 Cic. Rep. 2.59: Omnia nexa civiwm liberata nectierque postea desitwm; ef. Livy

6 There were two very much less important forms of verbal contract —dictio dotis, the promise of a dowry, and iusiurandwm liberti, the oath by which a slave on manu-

mission promised certain services to his former master. Both of these differ from stipulation in that only the promisor need speak —there is no preceding question by the promisee; Gai. 3.95f. It is commonly supposed that dictio dotis was originally enforceable as an accessory to the enforceable promise to give in marriage, but see Watson, Persons 60f. The tusiurandwm liberti is almost certainly primitive, but its enforcement may have been left to self-help until the later republic; Kaser, RPR# 1.300; C. Cosentini,

Studi sui liberti 1 (Catania, 1948) 181ff. .

7 For the view that only certain words might be used see Nicholas, LQR 69 (1953) 63ff.; contra, Winkler, RIDA (1958) 603ff.; van Oven, TR 26 (1958) 409ff.; RIDA 279

Private law to the fall of the republic: obligations they had to be ‘ Spondesne ?’ ‘ Spondeo ’, for the use of this particular word was always confined to citizens, whereas the stipulation, if other words were used, was iuris gentium.®

We speak of stipulation as a formal contract because it requires for its validity that the parties should meet and formulate their agreement in an oral question and answer, but the form is about the simplest which can be imagined. It used to be thought that so flexible an institution could not be primitive. We know now, however, that it existed at the time of the XII Tables,® though we do not know how far the development had then gone. The example given by Gaius is of a sponsio for a definite sum of money, but what this really represents for the time of the XII Tables when money did not exist is uncertain. It may be that stipulations were possible for any certa res, but not yet for an incertum.' The origins of stipulatio have been extensively debated, but no consensus has emerged. It is often said? that the word spondeo has religious associations and that it originally denoted a promise under oath, breach of which would entail the vengeance of the god who had been invoked, but the evidence for this * is far from compelling. On the other hand, the noun sponsor is used only of a guarantor or surety and not in the wider sense of a promisor generally, and this has been invoked as support for the view* that stipulation originated as a contract of suretyship. The development of the Roman

contract would on this view have been similar to that found in other systems.° To begin with, when A commits a wrong, B goes surety for him, B alone and not A being liable; then A is allowed to go surety for himself, so that duty and liability fall on the same person; finally the arrangement of (1961) 391ff. Since Latin has no word for ‘ Yes’, the occurrence of e.g. ‘ Dabis?’ ‘ Dabo’

in Plautus is (pace Watson, Obligations 1) no evidence for the admissibility of

8 Gai. 3.93. ® Above, 182.

stipulations in this form at this time. See further below, 509f.

1 Stipulations for incertum were known before the end of the republic, Varro, L.&. 2.2.5; 2.5.10 (Bruns 2.63), but even in Labeo’s time it was usual to fix a money penalty in case of breach, D. 45.1.137.7. For the meaning of certa res and incertum see above, 193 n 8.

2 See especially Kaser, AJ 256ff.; A. Magdelain, Hssai sur les origines de la sponsto (Paris, 1943). 3 Principally: (1) it is usually thought that spondeo is etymologically connected with

onéviw (I bring a drink-offering); (2) Festus 59 (Bruns 2.5) gives coniurator as the meaning of consponsor; (3) Festus 329 (Bruns 2.40) quotes Verrius Flaceus as deriving sponsus and sponsa from the Greek, ‘quod ii ozovdas interpositis rebus divinis faciant’ (but this need be no more than an inference from the ceremony of betrothal, especially as Festus also says that Flaccus elsewhere connected spondere with what was promised

sponte sua). |

4 Mitteis, Festschr. fiir E. I. Bekker (Weimar, 1907) 107ff.; id., RPR 27, 266ff. 5 See Brunner, in Holtzendorff’s Enzyklopddie der Rechtswissenschaft, 2nd ed. by J. Kohler, 1 (Munich etc., 1915) 1.139; Pollock and Maitland 2.184; Holdsworth, HEL

2.83ff.

280

Contract going surety for oneself turns into a method of making binding promises in a more general] way.® It has been pointed out, however,’ that in the law as we know it the sponsor could guarantee only obligations which were themselves

contracted by stipulation, and this rule hardly tallies with an origin in a guarantee of a non-contractual obligation. And Daube has shown * that one cannot argue from the meaning of a verbal noun to that of the verb itself. Not everyone who undertakes is an undertaker.

Kaser ® finds truth in both the theories discussed in the preceding paragraph. Stipulation, he argues, had a twofold origin, religious and secular, corresponding to the two terms sponsio and stipulatio. The religious institution, which is of course the older, survives into historical times in the use of sponsio for making treaties with foreign states.! Stipulation, on the other hand, is a purely secular institution and is found in its earliest form in the guarantees given in court procedure by praedes and vades.? Eventually, when sponsio has lost its religious associations, the two institutions merge to produce the contract as we know it. Leaving aside the uncertainty of the religious origins of sponsio, this and other theories which start from a dichotomy between sponsio and stipulatio * are open to a serious objection made by Arangio-Ruiz.* As far back as the evidence of our sources goes, the terminology used gives no support to this dichotomy. Gaius never uses sponsio of the entire verbal contract of question and answer, and very rarely uses stipulatio at all; his normal term is verborum obligatio. It is only in the late law that stipulatio is habitually used in the sense which we now take for granted. Normally stipulatio and stipulari are used only of the question, the

answer being indicated by spondere or promittere, as in the traditional formulary put at the end of documents: stipulatus est L. Titius, spopondit C’. Seius, or in the description of the betrothal contract (sponsalia) by Servius Sulpicius * (d. 43 B.C.) as contractus stipulationum sponsionumque. If stipulatio and sponsio were once two separate contracts, it is difficult to see how

this terminology could have become established.

6 Cf. above, 159ff. 7 Monier 2.22; ef. Gai. 3.119.

8 LOR 62 (1946) 266ff.; Roman Law (above, 276 n 8) 2ff. 9 AJ 256ff. and RPR 1.168f. 1 As in the affair of the Caudine forks, above, 41 n 1; ef. Gai. 3.94. 2 Above, 162.

3 F,. Pastori, Appuntt in tema di sponsio e stipulatio (Milan, 1961) supposes that stipulatio (as opposed to the religious sponsio) originated in fidepromissio in international dealings; it was turis gentium and based on fides, the words being merely a declaration of agreement; it only became a formal verbal contract on its reception into the ius civile. Against this see Talamanca, Labeo 9 (1963) 96ff. For another view see

Biondi, BIDE 65 (1962) 105ff. 4 BIDE 65 (1962) 193ff. 5 Quoted by Aulus Gellius, 4.4. Cf. above, 233.

6* Whereas in legal texts sponsor is used of a surety (see above), sponsto seems never to refer to the contract of suretyship; in the earliest known use (in the formula of the

legis actio per iudicis postulationem given in the new fragment of Gaius, 4.17a) it 281

Private law to the fall of the republic: obligations (iii) The literal contract.’ Of this contract little is known for certain, as Gaius,® almost our only authority, describes rather the use to which it was put than the way in which it was formed. It appears, however, that A might become the debtor of B for a sum of money by a fictitious entry made in the account book (codex accepti et expensi) of B (the creditor), alleging that B had paid out that sum of money by way of loan to A. Gaius tells us of two uses to which the contract might be put: (a) Transscriptio a persona in personam, If A owed B 10,000 sesterces and it was desired to substitute C for A as B’s debtor (e.g. because C owed A a similar sum and the position could be simplified by cutting him out altogether), then this might be achieved by entering in B’s book a fictitious loan

of 10,000 to C. :

(b) Transscriptio a re in personam. If A owed B 10,000 sesterces because e.g. B had sold him goods, this debt could be turned into a debt owed under

a literal contract by entering fictitiously in B’s book that B had lent A 10,000 sesterces. The object of such a transaction might be to give the creditor an easily provable debt enforceable by the simple and profitable ° remedy of condictio instead of the bonae fidei action under the original contract, and the debtor might have to agree to such a course, because e.g. the creditor made it a condition of granting him further credit. There is no evidence that in either case it was necessary, or even customary, for B to make a balancing entry recording the discharge of the debt which was being replaced by the literal contract. Indeed, in thinking of the account book we have to bear in mind that Roman accounts were kept in simple narrative form, and in particular that there were no separate debit and credit columns, and therefore that the idea of a counter-entry has no place.’ Gaius says nothing of any participation’ by the debtor in the making of the contract. Plainly he must have agreed to the making of the creditor’s entry, apparently refers to the promise of the debtor in the verborwm obligatio in a general sense, whereas in texts from Plautus to the end of the second century A.D. it refers uniformly to procedural promises (sponsiones et restipulationes). The terminology of Cicero (in legal contexts) is consistent with this: spondeo is used in a wide sense of any promise verbis, whereas sponsor means a surety and sponsio a procedural promise. See Arangio-Ruiz, op. cit.

7 See de Zulueta, Gaius 163ff., qualified as to the form of the account book by de Ste

Croix in Studies in the History of Accounting, ed. A. C. Littleton and B. S. Yamey (London, 1956), 14ff. at 72ff.

8 Gai. 3.128~33. In addition we have an account, probably of little authority, in Theophilus, Paraphr. 3.21, a fragmentary inscription dating from the middle of the first eentury A.D. (Bruns 1.353), two fragmentary tablets recently recovered from Herculaneum (see below), and a number of references in legal and lay literature (discussed by Watson, Obligations 18ff.). ® Above, 214f., 194.

1 De Ste Croix, op cit. above (n 7) 74; Joannique, RHD (1968) 5ff.

282

Contract and it used to be conjectured that he made a corresponding fictitious entry in his own account book, but some fragmentary tablets recovered from Herculaneum have prompted other conjectures.” The first ‘ page’* of one of these tablets (evidently a list of contents) bears the words chir(ographum) L. Comint Primi ex nomine facto, followed, after a gap, by the amount of the debt. Another tablet is similar, except for the name. A few scraps of papyrus are stuck to the first, and from this Arangio-Ruiz conjectured that the creditor kept the debtor’s acknowledgment (chirographum)* together with the note of the corresponding entry in his account book. This suggests that, as in the case of stipulation,® the form which constituted the contract was distinct from the evidence for it; in other words, the entry (though a necessary form) was not in itself enough to prove that it had justifiably been made.®

Gaius carefully distinguishes from these fictitious entries (nomina transscripticia) what he calls nomina arcaria, i.e. entries which recorded payments

actually made by way of loan, where the binding element lay, not in the written entry (which was merely evidence), but in the actual payment of money (mutuum)."

How old the institution of nomina transcripticia is we cannot say; bookkeeping with entries of loans and repayments certainly existed at the beginning of the second century B.C.,*° but this does not necessarily mean that

RIDA (1948) 15ff. 3 Below, 415 42 Arangio-Ruiz, Below, 415. 5 Above, 280.

6 Contra, Watson, Obligations 23f. Arangio-Ruiz later (Atti Accad. Modena (1948) 131ff.=St. Redentt 1.113ff.) conjectured that what was stuck to the chirographwm was the nomen itself, and since, if nomina had had to be made in the account book, this would have necessitated cutting up the book, he suggested that there was no necessary connexion between the contract and the account book. Any statement of the debt by the creditor in writing, even on a single sheet, would be sufficient form, and the transaction would be another example of the Roman principle that the acquirer of rights formulates them. It is important in this connexion not to make too sharp a distinction between account books and other memoranda. It is generally accepted that Cicero’s speech pro Koscto Comoedo connects the literal contract with accounts, but there is disagreement as to whether the accounts referred to are the account books proper or the adversaria, or day-sheets (see discussion in Watson, Obligations 24ff., 38f., and add Tumulescu, Labeo 15 (1969) 285ff.). But since it is likely that adversaria might take any form, including that of loose sheets (de Ste Croix, op. cit. above, n 7, 41), the distinction between them and other memoranda is probably illusory. On the interpretation of the tablets from Herculaneum proposed by Stojeevic, Iwra 13 (1962) 53ff., see Watson,

Obligations 23. It must be emphasised that no explanation of the piece of papyrus stuck to the chirograph can be more than conjecture.

7 He does not say how one could distinguish one type of entry from the other. We

appear to have the paradox that the difference between them is that proof of the falsity of the entry will validate the one and destroy the other. Arangio-Ruiz, Mélanges

KE. Tisserant (= Studi e Testi, Vatican, 1964) 9ff.,-calls attention to certain of the tablets from Herculaneum which specify that the payment was expensum ex arca, and conjectures that these are references to nomina arcaria; ef. Tumulescu, Labeo 15 (1969)

285ff. 8 Plautus, Truc. 70ff. 283

Private law to the fall of the republic :- obligations fictitious entries were already known. In Cicero’s day, on the other hand, they were clearly in common use,’ as they presumably must have been still at the time of the destruction of Herculaneum (A.D. 79), but the brevity of Gaius’ treatment suggests that in the middle of the second century they were no longer of much importance. In Cicero’s day, on the other hand, they were clearly in common use,’ and by the classical era the institution had already become practically obsolete.

B. Inrormau Contracts. (i) Mutuum and the quasi-contractual obligations giving rise to condictio. In the classical law if A handed B money with the intention that it should be a loan and B received it with the same intention, the transaction was called mutuum and A could bring a condictio to recover the amount.” The obligation arising in this way was said to be one contracted ré, i.e. the transfer of the money was necessary in order to create the obligation to return an equal sum; if no money had been in fact transferred, then

there could be no mutuum. When formless loans of this sort first became actionable we do not know, but it was presumably well back in republican times.* The question is perhaps not of such great practical importance as would appear at first sight, because the formless loan could never give the right to recover a greater sum than was actually lent, and gratuitous loans were comparatively rare. If interest was to be charged, it was necessary, even in classical times, to make a special stipulation to this effect. In all probability mutuum first became actionable not because it was recognised specifically as a contract worthy of enforcement, but as an application of an as yet incompletely articulated principle that if A had been enriched at the

expense of B without any justification, then B should have an action to recover the amount of the enrichment.* In the classical law a number of such cases are recognised, and the remedy

is always condictio. Thus, if B has paid A a sum of money under the mis® Rosc. Com. 4.13; 5.14; Off. 3.59. 1 Off. 3.58f£. (Watson, Obligations 29ff.).

2 The contract is also possible with other ‘ fungible’ things, e.g. corn; Gai. 3.90. 3 Phrases such as Si cut mutuom quid dederis (Trin. 1051) in Plautus do not actually

prove more than the existence of loan as an economic fact, but in Cicero’s day we have proof that formless loan was actionable: pro Rose. com. 5.14, where Cicero, enumerating the possible grounds of action for a elaim of pecunia certa, says: haec pecunia necesse est aut data aut expensa lata aut stipulata sit. Data must here include loan, though it may include other things as well. Cf. Watson, Obligations 15 n 1. 4 More precisely, the classical principle was probably that if B had conveyed (cf. data in the preceding note) to A a sum of money or a thing, and there was no justification for A’s retaining it (or, in the case of an overdue mutwwm for example, no longer any justification), a condictio would lie. For an outline see Nicholas, Tulane L.R. 36 (1962) 611ff., and for a full examination F. Schwarz, Die Grundlage der Condictio im klass. rom. Recht (Minster, 1952), and Donatuti, Studi Parmensi 1 (1951) 35ff. Earlier still the claim may have been delictal, the failure to return being akin to theft (e.g. Kaser, AJ 2844f.).

284

Contract

taken belief that he owes it, he has a condictio for recovery (condictio indebiti). Similarly, if B has given A money to manumit a slave within a certain time and A fails to manumit, B can get the money back by condictto (condictio ob rem dati).> Mutuum is really a similar case; if B has lent A

money and A does not return it, A has been enriched unjustifiably at the expense of B. The principle that unjustifiable enrichment should be recoverable by the person at whose expense it has occurred was actually expressed

early, for it is attributed by Sabinus ° to the veteres, i.e. the republican jurists, but there is no need to suppose that it did not exist until it was expressed, or that it first came into existence with the introduction of the legis actio per condictionem. Debts recoverable by condictio seem, at least in some cases, to have been recoverable previously by legis actto sacramento,’ and there is no reason to suppose that the range of facts which would support the assertion aio te mihi dare oportere had not been extended so as to include the debts we are discussing before the introduction of condictio. The history of our own law teaches us that such processes are inevitably gradual. There

ean be little doubt, however, that the abstract nature of condictio (at least under the formulary system) * helped to extend the number of cases where a debt was assumed, for it was obviously easier for a iudex (or his professional

adviser) to recognise that what was claimed ‘ ought to be paid’ than it would have been for a pleader to define exactly the cause of action if the formula had not absolved him from his duty in such cases. As Rabel put it,° ‘ condictio, seeing that it was a formula which did not mention the cause of action, was excellently adapted to cover those cases in which it was desired to allow a claim for recovery, not because the plaintiff had a special justification for his claim, but because the defendant had no sufficient justification for retaining what he had got ’. When ‘ contract’? came to connote the idea of agreement,’ mutuum of course could be classed as a contract, whereas the obligation arising from the payment of indebitum had, since there was no agreement, to be called quasicontractual,” but the origin of the two lies in the same idea, as does also that of the other non-contractual condictiones which modern authors place with indebitum solutum under the heading of quasi-contract.

(ii) Fiducia and the ‘ real’ contracts other than mutuum. Fiducia,* or pactum fiduciae, is not an independent contract but an agreement subsidiary 5 Also given by Justinian (D. 12.4, rubric) the strictly meaningless name (by which it was nevertheless subsequently known) of condictio causa data causa non secuta. 6 PD. 12.5.6 (Ulpian 18 ad Sab.): Perpetuo Sabinus probavit veterwm opinionem existimantium id quod ex iniusta causa apud aliquem sit posse condict. Cf. Watson,

Obligations 12f. 7 Cf. above, 194f. 8 Above, 214f.

® Rabel 470. 1 Cf. above, 271. 2 J. 3.27.7.

3 See W. Erbe, Die Fiduzia im rém. Recht (Weimar, 1940).

285

Private law to the fall of the republic: obligations eaten ent enn ene ate ane mnemneeaemertanneen areas pinmnentaarnanaateameenaataaacnaneaerr annem anmaaneeinam neater anees uaneeaireae eet nent tne an naar eerie eee nected

to a conveyance by mancipation or in iure cessio imposing a trust. upon the transferee with reference to the thing conveyed, most frequently a trust for reconveyance of the thing in certain circumstances.* Thus, if a man thought for some reason that his property would be safer in the hands of a friend than in his own, he could convey it (by mancipation or in ture cessio) to the friend with a fiducia for reconveyance on demand or when the particular danger which he had in mind had passed (fiducia cum amico).° The friend would in the meanwhile be technically owner and able to exercise all the rights of an

owner, for instance to vindicate the thing if it got out of his possession. More important however was fiducia cum creditore —a form of pledge. Here a debtor, in order to secure his debt, conveyed some piece of his property to his creditor with a fiducia for reconveyance if and when the debt was paid. Something more will have to be said later about this arrangement under the

heading of ‘ real security ’,* but for the moment the important point is the fiducia. This was at all times a purely informal agreement, and its informality as well as the word fiducza (fides) points to a time when it was unenforce-

able and the transferor had to rely on the faith of his friend or creditor.’ Already some way back in the republic, however, this was no longer the case, and an actio fiduciae lay against a transferee who was guilty of a breach of

trust. The first mention of this action occurs in a list of bonae fidei iudicia quoted by Cicero from Q. Mucius,® but the probability is that the action is older than Mucius, for some words quoted by Cicero from the formula are of a peculiar and archaic character.’ Possibly it was the first departure from the old principle of giving effect only to formal transactions and was the prototype of the actions of good faith. Depositum, commodatum and pignus. These three contracts (deposit, loan for use, and pledge) are classed by Justinian with mutwum under ‘ obligations contracted re ’.? They are all ‘ real’ in the sense that there is needed for their formation (in addition to the agreement of the parties) only the handing over by one party to the other of a thing which is the object of the contract, and that the main duty which arises under the contract is that of 4 The trust might also be for the manumission of a person either free (see e.g. above,

240 n 6) or slave who was mancipated, or for the further transfer of property to a

third party; see e.g. D. 39.6.42. 5 Gai. 2.60. 6 Below, 301. 7 It is sometimes argued, e.g. by Watson, Obligations 172ff., that fiducia was action-

able already by legis actio under the words of the XII Tables (v1.1): cwm nexum faciet mancipiumque uti lingua nuncupassit, ita ws esto (above, 145 n 3), but see Kaser TK 34 (1966) 416; the provision is referred to as pactwm conventwm in the formula Baetica (Bruns 1.332; FIRA 3.296). 8 Off. 3.70. Plautus uses the word, but (pace Watson, Obligations 172ff.). nothing certain can be inferred as to the existence of an action. ® Ut inter bonos bene agier oportet, Top. 66; Off. 3.61, 70; ad fam. 7.12. 1 On the character of the formula and the question whether there were two actions

or one, see Kaser, TR 34 (1966) 417f., and RPR 1.461f. . 2 J. 3.14.

286

Contract

the recipient to return the thing.* Late‘ analysis thus places them with mutuum, but Justinian is careful to point out the differences; whereas the borrower in mutuum becomes owner of what he borrows and has only to restore an equal quantity of similar things, the borrower in commodatum, the depositee and the pledgee do not become owners and have to return the identical thing received. Now it is obvious that, from the economic point of view, pignus and depositum cover the same ground as fiducia cum creditore and fiducia cum amico, and it may be that fiducia cum amico was also used to effect the purposes of a commodatum. The technical legal difference

is great, for the recipient under fiducia becomes owner, whereas in the other cases he does not, but from the point of view of the layman both pignus and fiducia cum creditore are ways of raising money on security, and both depositum and fiducia cum amico are ways of getting someone to look after property. It is also clear that the ‘ real’ contracts are the simpler and, when recognised, would tend to supersede fiducia. It is consequently some-

times said that they are simplifications of fiducia, but it is very doubtful whether this accurately expresses the historical process. Fiducia could only be attached to mancipatio and in iure cessio, and would hence be inappropriate to res nec mancipt, and yet people must, in fact, have delivered res nec mancipi by way of loan, deposit or pledge, without transferring the owner-

ship therein, before they were protected by the development of special actions. Even then they would not be entirely without protection, for the owner who had lent or deposited his thing could vindicate it, and was perhaps also protected by the law of theft.*® The position of the pledgor would be

similar if the debt had been paid, but what happened if he attempted to vindicate before then we cannot say. As has been indicated already, the first specific recognition of these ‘ real contracts ’ was by praetorian actiones in factum, dating from before the end of the republic. It was only later that actiones in ius also became available. Any more precise dating is a matter of conjecture.° 5 The pledgee only if and when the debt is paid.

4 How late is uncertain. In his Institutes Gaius does not mention any of the three under the heading ‘ obligations contracted re’ (3.90f.), which for him is evidently confined to debts which are recoverable by condictio (above, 284f.) and which therefore entail the passing of property. On the other hand the quotation from his Res cottidianae in D. 44.7.1 is closely parallel to Justinian’s Institutes and mentions them all. How far this latter passage is genuine and, if not genuine, how old it is, is disputed. On Gaius’ treatment of these contracts see de Zulueta, Gaius 150f.; Nicholas, Introduction 168f.; Honoré, Gaius 63ff., 97£.; Wubbe TR 35 (1967) 500ff.

5 Girard 558ff. For this kind of protection ef. Pollock and Maitland 2.186: ‘If the gage was not restored the claim for it would take the form “you unjustly detain what is mine ”’.’ For depositum there was also the old action under the XII Tables; above, 164, 172.

6 For discussion see Watson, Obligations 157ff., 167ff.; Burillo, SDHI 28 (1962) 233ff.

287

Private law to the fall of the republic: obligations (iii) The consensual contracts and the quast-contractual relationships of negotiorum gestio and tutela. In the classical law there are four contracts said to be concluded consensu, i.e. nothing more is necessary for their formation than the agreement, no matter how expressed, of the parties.” These are emptio venditio (sale), locatio conductio (hire), societas (partnership) and mandatum (agency). All were in existence well before the end of the republic, and all appear in the list of bonae fidet tudicia quoted by Cicero from Q. Mucius,® though this does not necessarily prove that they were already

recognised as consensual. So long, however, as the history of procedure before the lex Aebutia remains as obscure as it is at present, we can have no detailed knowledge of the stages by which the consensual contracts came to be recognised, and their history before the time of Q. Mucius must remain almost wholly conjectural.

The grouping of the four contracts together is, no doubt, like the other institutional classifications, an achievement of theoretical jurisprudence dating from the early empire, and we can see that historically the four fall into two distinct groups, the one consisting of emptio venditio and locatio conductio, which are commercial relationships, the other of societas and mandatum, which appear to have had their origin in the family. In sale and hire the parties are men whose interests are opposed, and each seeks to make a bargain which will be profitable to himself,® whereas partnership and mandate are both of a fiduciary nature and require that the parties should have special trust in each other,’ a point which is clearly illustrated by the rule that condemnation in an actio pro socio or mandati leads to infamia.’ It is

more dishonourable for one partner to try to overreach another than for a seller to try to overreach a buyer. - (a) Emptio venditio and locatio conductio. (1) Emptio venditio. (a) Theories of origin.* There can be no doubt that at Rome, as elsewhere, sale began by being an exchange in which the parties carried out their respective parts

of the bargain simultaneously and left no obligations outstanding, but beyond this point there is no agreement as to the origin of the contract or the date at which it first became consensual. Many solutions have been offered, and it may be that there is an element of truth in all of them, but the difficulty has not been solved because the story was probably a long and complex one which can never be reconstructed without some evidence of the actual decisions of the courts, any more than it would have been possible to 7 Gai, 3.135ff. 8 Off. 3.70.

9 In the English phrase the contract is ‘at arm’s length’. 1 Cf. Cie. Rose. Amer. 112: Neque mandat quisquam fere nist aniico. 2 Gai. 4.182; D. 3.2.1. But only condemnation in the ‘direct’, not in the ‘ contrary ’, action. On infamia see above, 273 n 4. 3 See for discussion Arangio-Ruiz, Compravendita 45ff. 288

Contract reconstruct the history of contract in early English law without the evidence of the Year Books. Among the chief hypotheses are the following: Hypothesis of development from state contracts.* According to this view both sale and hire originated in the imitation, by private persons, of arrangements which had existed from time immemorial in the practice of the state.

The typical public sale was that of booty by the quaestor on behalf of the state. The magistrate in these cases published the conditions of the sale orally or in public notices and the bargain was concluded when a bid made by the private purchaser was accepted by the magistrate without any further formality. If difficulties arose, the magistrate himself decided them on equitable grounds, and this procedure became a pattern for the subsequent D./. judicia on private contracts. It may, of course, well be that these public contracts did have an influence on the development of informality in private contracts, but the real difficulty of knowing what actions the private parties used is not touched, and, with-

out any evidence, it is difficult to accept the view that the administrative process used by the magistrate in dealing with public contracts could have formed the model for the totally different private procedure before a iudea. Hypothesis of double stipulation.® According to Bekker and a number of other writers, the earliest method of concluding a contract of sale which was not to be executed at once was by recourse to the universal method by which any undertaking could be rendered actionable — the stipulation. The buyer,

it is supposed, stipulated for the delivery of the thing, the seller for the payment of the price.

That stipulation played some part in the development of sale is certain,‘ but there are great difficulties in supposing that these double stipulations were the origin of the contract; if they had been we should have expected the actions, like other actions on stipulations, to have been strict and not bonae fidei, and, above all, it is strange, on this hypothesis, that in the forms

of sale given by Cato’ there is no suggestion that the whole contract was confirmed by stipulation though he recommends that there should be stipula-

tions for certain special points.® |

4 Mommsen, Ges. Schr. 3.132ff. (=SZ 6 (1885) 260ff.); F. Caneelli, L’origine del contratto consensuale di compravendita nel dir. rom. (Milan, 1963), criticised by Mayer-

Maly, Iura 15 (1964) 357ff.; Gallo, SDHI 30 (1964) 299ff., supposes that the early credit sale was by stipulation (cf. immediately below), but that the transition to consensual sale was made by way of the private auction.

5 E. I. Bekker, Die Aktionen des rém. Privatrechts 1 (Berlin, 1871) 156ff.; Girard 570; Jhering, Geist 3.202; Arangio-Ruiz, Compravendita 57ff. 6 Below, 292ff.

7 See especially de agr. cult. 146 (Bruns 2.49); von Libtow, Symb. Taubenschlag 3.006ff.

8 Watson, TR 32 (1964) 245ff., conjectures that the consensual contract at first simply filed the gaps left hy these stipulations, i.e. that although the stipulations were 289

Private law to the fall of the republic: obligations Hypothesis of a ‘ real’ stage in the evolution of sale. Another view, advocated especially by Pernice,*® supposes that before sale became consensual

there was a stage during which it was ‘ real’, in the sense that the seller could bring an action for the price if he had delivered, but only then, and that there was no action by which the buyer could enforce delivery. Comparison with other systems would appear to support this view,! but the positive arguments in its favour are not as strong as was thought at one time.? Hypothesis of original formation by payment of arra.* In the classical law

the function of arra, or earnest, was purely evidentiary,‘ i.e. it served to actionable independently, the bonae fidet action could (at first) only be brought if one

or more of the stipulations existed. He contends that this would account for two features of the Roman contract: the persistence of the stipulations after the consensual

contract was established, and the apparent failure to recognise the advantages of developing a contract which could be made inter absentes. But the reason for the first

feature could be that the stipulation had the advantage of imposing strict liability, while liability under the consensual contract would have to be based on lack of bona fides; and the second feature is no easier to explain on this hypothesis than on any other (it: should, moreover, be remembered that we know next to nothing about how long-distance commerce was conducted). Further, one would expect there to be some

stipulation. 9 Labeo 1,456ff.

trace of juristic discussion of the legal effect of a consensual contract without any 1 For English law see Glanvil x.14, quoted by Holdsworth, HEL 3.414 n 2.

2 (a) The rules that risk passed on the conclusion of the contract and that for usucapion pro empto bona fides is needed both at the time of the contract and at the time of delivery, quoted by Pernice in support of his view, may equally well be relics of the time when delivery and payment were simultancous.

(b) The exceptio mercis non traditae (Gai. 4.126a) does not prove that the seller had to deliver before he could sue for the price, for Gaius is referring to actions brought by bankers, not ex vendito, but on stipulations for the price of goods sold by auction; Girard 567 n 3; Lenel, HP 503.

(c) Varro, REF 2.2.6 (Bruns 2.63), after giving the form of stipulation which the buyer of sheep should exact as to their soundness, says: Cwm id factum est, tamen grex dominum non mutavit, nist st est adnumeratum; nec non emptor pote ex empto vendito illum damnare, si non tradet, quamvis non solverit nummos; ut ille emptorem simili iudicio, si non reddit pretium. From this Pernice argues that the seller had to deliver first and that this is a relic of the time when there was no obligation until he had delivered. But the words can equally well mean that the buyer can insist on delivery even before he has paid, provided he is ready and willing to pay, and that the seller can similarly insist on payment provided he is ready and willing to deliver. It is true that

Varro does not actually say of the seller that he can secure payment quamvis non tradiderit, but this seems to be implied in the comparison he makes between the positions of seller and buyer; had he meant to contrast their positions he would have said at ille

non potest emptorem damnare simili iudicto nisi ipse tradidit. ,

The argument from the rule that the ownership in the thing only passes if payment has been made falls to the ground if the view be accepted (cf. above, 145 n 2) that this rule is post-classical. Thus, in the passage quoted from Varro Pringsheim (Der Kauf mit fremdem Geld (Leipzig, 1916) 73) takes adnuwmeratwm to refer, not to payment of the price, but to the counting of the sheep on delivery. 3 v. Mayr 11.2.11.59ff.; Pringsheim, Greek Law of Sale 333ff., 415ff.; ef. Watson,

Obligations 46ff. 4 Gal. 3.139. 290

Contract mark the moment when the parties had ceased chaffering and had definitely concluded a bargain. But in other systems, and particularly in Greek law, it served a different purpose, that of a kind of pledge that the bargain would be carried out; the rule in some Greek systems, at any rate, being that if the party who had given the earnest failed to complete the contract he lost his earnest, and if the other party failed he had to give the earnest back with as much again.’ The position was then that although there was no possibility of action to enforce fulfilment, the party who failed to fulfil lost the amount of the earnest. It may be that some similar rule existed in early Rome and that the practical binding secured in such cases resulted finally in the rule that the mere consent, even without arra, would be sufficient to constitute obligations on either side. This suggestion, though not unlikely in itself, is without sufficient evidence

to warrant its acceptance. In any case, the key to the problem is probably to be found in the actions by which the parties enforced their rights. As we. have seen, at least one of the roots of the bonae fidei actions must have been in the need to provide for cases in which foreigners were concerned, and they may have been thence extended, together with the formulary system, into the law as administered between cives. The reference to b.f. was probably at first regarded as controlling not so much the measure of damages as the question of actionableness itself. In sale therefore, when the iudex was told to condemn for quidquid dare facere oportet ex {.b., his functions would have included that of deciding whether (even if there had been what later amounted to a

consensual sale) good faith demanded that any payment at all should be made. Not until later was it recognised that once such a sale had been proved, the only question to be assessed according to good faith was that of the damages payable. On this view the crucial step was the recognition of bona fides as a basis of liability outside the old civil law.‘ } (@) Liability of seller for defective title. The primary duty of the seller is to deliver possession of the thing sold (vacuam possessionem tradere); he is not bound, as he is usually in modern systems, to convey ownership therein,’ i.e. the buyer cannot complain merely because he discovers that he has not become owner on delivery.® If, however, there has been eviction,’ or if, without actual eviction, the buyer has lost the benefit of his bargain, then he can

make a claim against the seller. Such loss without eviction may happen if, 5 Considerable literature; Kaser RPR 1.547f. 6 Cf. Pringsheim, Ges. Abh. 2.179ff. (=RHD (1954) 475ff.); Meylan, TR 21 (1953) 129ff.; id. in Aequitas u. Bona Fides — Festgabe fiir A. Simonius (Basle, 1955) 247ff. 7 Sale of Goods Act 1893, ss. 1, 12 (1); German Civil Code, § 433; Swiss Law of Obligations, § 184. § Unless the seller was in bad faith; D. 19.1.30.1. ® For the meaning of ‘ eviction ’ see above, 146. 291

Private law to the fall of the republic: obligations for instance, the buyer keeps the thing, but only because he buys off the claim of the real owner, and thus has to pay twice for the same thing.

This liability of the seller is inherent in the contract in the classical and later law, but it develops out of the stipulations which ordinarily accompanied sales.2 We have seen that where a buyer had received a thing by mancipation he could, if subsequently evicted, bring an actio auctoritatis for double the purchase price, provided he had warned the seller of the threatened action. This was only possible in the case of mancipation, but where, for one reason or another, the thing was not mancipated,' it was customary to provide for the seller’s liability by exacting from him a promise by

stipulation that in the event of eviction he would pay double the purchase price (stzpulatio duplae),* i.e. to copy the liability arising ipso facto from mancipation. There was also another stipulation in which the seller promised

that the buyer ‘ should be allowed to hold’ the thing (habere licere). Here the seller’s liability is not for a fixed sum but for whatever damage the buyer suffers. In classical times the custom seems to have been to take the stipulatio duplae only in sales of res mancipi (as above) and of res nec mancipt of considerable value.*®

With the growth of the conception of bona fides, it came to be regarded as inconsistent with good faith not to make the promises which were usual, and, if in any particular instance they had not been made, an actio empti could be brought by the buyer to insist on the seller’s making them. This stage had been reached at any rate by the time of Neratius*® (late first century and early second century A.D.). Finally it was admitted that even if the stipulations had not been made at all the actio empti could be used by the buyer to get what he would have got if they had been made, and that was the classical

position.’ |

The rule that the seller is not bound to convey ownership cannot have been of great practical importance, since failure to do so would normally manifest 1 DP. 19.1.13.15; 21.2.9; 21.2.56.2; PS 2.17.8; Girard 593ff. 2 Girard 589; Monier 2.153; de Zulueta, Sale 43. 3 E.g. because it was not res mancipi or because one of the parties was a peregrine. Even where there was mancipation it would be necessary for there to be a stipulation if the obligation were to be guaranteed by sponsors or fide-promissors, and it is perhaps

to this guarantee that the phrase satisdatto secundum mancipiwm (Cie. ad Att. 5.1.2; tab. Baetica, Bruns 1.334) refers; Lenel, EP 546f.; Kaser, RPR 1.130. For another explanation see Sargenti, BIDE 65 (1962) 151ff.

4 Or, if the parties preferred, simply for the price (stipulatio simplae); Varro, Lh 2.10.5 (Bruns 2.64; de Zulueta, Sale 63). 5 —D. 21.2.37.1. Cf. h.t. 6 (local custom). The requirement of eviction was less strictly construed in the case of the stipulatio habere licere; D. 19.1.13.15; 21.2.9; 21.2.56.2;

PS 2.17.8; Girard 593ff. It has been argued that the stipulatio habere licere covered only eviction by the seller or his heirs, but see Lévy, RHD (1954) 345f.; Watson, Obligations 86.

6 D. 19.1.11.8. | 7 De Zulueta, Sale 44. 292

Contract itself in the dispossession of the buyer, who could then sue on the warranty

against eviction, but in principle the rule is at first sight surprising. The explanation is perhaps * that it would have been difficult to define precisely what ownership had to be transferred. Mancipatio is necessary for the conveyance of Quiritary ownership, but even if there was a duty on the seller to mancipate (and this is far from certain),® this would be impossible if either party were a peregrine. Again, provincial land was incapable of ownership in the ordinary sense. In these circumstances it may well have seemed better to leave the question to be expressly provided for, and that what emerged was @ warranty against eviction rather than a warranty of title is hardly surprising.’ (y) Liability of seller for latent defects.? Roman law, like other systems, began with the principle that no lability attached to the seller for defects in the thing sold, whether or not they were noticeable at the time of the sale. In classical law it was indeed recognised that the seller was liable (in the actio empti) if he knew of the defect and failed to disclose it, but it seems that in the time of Cicero this liability was confined to sales of land and to legal, as opposed to physical, defects (e.g. the existence of servitudes).* It was, of course, possible to provide by stipulation that the seller should be liable in case the thing turned out to have some particular defect or defects, and it is clear from the forms given by Varro that this was frequently done. Thus he says, for instance, that when oxen are sold the buyer stipulates that they are ‘ healthy and free from noxal liability ’.° Further development was achieved through the action of the curule aediles, who, having charge of the market-place, were able to make regulations concerning the sales which took place there.* We know of two edicts dealing with the matter, one of. which concerned the sale of slaves, the other the sale of beasts.” The slave edict cer-

tainly dates from republican times, and indeed appears to have existed 8 See de Zulueta, Sale 36f. % Arangio-Ruiz, Compravendita 167ff., denies it.

1 Alternatively, if it is accepted that early Roman law had no concept of ownership hut only one of relative title, then the seller did not promise ownership because a better possession was all that the law could envisage. Cf. above, 142. See also Meylan, Festschr. O. Riese (Karlsruhe, 1964) 423ff. 2 De Zulueta, Sale 46ff.; R. Monier, Za garantie contre les vices cachés (Paris, 1930) ; ef. Crook, Law and Life 180ff. 3 Off. 3.65ff.

4 For discussion see P. Stein, Fault in the Formation of Contract (Edinburgh, 1958) off.; Watson, Obligations S86ff. Stein thinks that even in classical law there was nu liability for non-disclosure of physical defects in land; cf. Daube, LQR 73 (1957) 379ff. D. 21.1.38.7 seems to show that for Ofilius, a younger contemporary of Cicero’s, the hability extended to physical defects in slaves.

5 R.R. 2.5.10, 11 (Bruns 2.63). 6 Above, 49f.

7 Iumenta, literally beasts of burden, but a special clause declared that all pecora were included; D. 21.1.38.5.

293

nn inet ee teen) Private law to the fall of the republic: obligations

already in the first half of the second century B.C.;° the edict concerning beasts is later, but also appears to have existed by the end of the republic.° The object of the aediles was, no doubt, to punish fraudulent sellers,’ and the two classes of dealers with which their edict was concerned were clearly

particularly suspect, but the result of their action was to import into sales which fell within their jurisdiction entirely new rules of law; what were ostensibly police regulations became a significant source of private law. In particular, the seller was made responsible for knowing the defects of a slave

or beast that he sold. In the form in which we know it? the edict required the seller to declare any physical defect (morbus or vitium),® and in addition,

in the case of a slave, whether he was noxally lable, a runaway, or a vagabond. If a defect appeared which had not been declared, the buyer could claim either rescission of the contract (by actio redhibitoria, which had to be brought within six months) or a rebate on the purchase price corresponding to the difference which the defect made to the thing’s value (by actio quanti minoris * within a year).° In the classical law it seems clear that the principles of the aedilician edict were applied to sales of slaves and beasts even outside the market-place, but

that the extension had gone any further is very unlikely. There is indeed a text in the Digest which attributes to Labeo the statement that the rules applied indiscriminately to sales of all things, movable or immovable, but this. is undoubtedly interpolated and so evidence for the law of Justinian’s day only.® ’ (2) Locatio conductio.** Modern expositions of Roman law have customarily

distinguished between three different types of agreement: I.c. rez, the hire of 8 Cato (d. 152 B.C.) is quoted on the meaning of morbus in D. 21.1.10.1. Daube, Forms 91ff., conjectures that the edict was first made at the prompting of Cato as (plebeian) aedile in 199 B.C., but see Nicholas, Class. Rev. 71 (1957) 251. Cicero mentions the edict, Off. 3.71. ® Labeo is quoted on sows by Gellius, 4.2.8, and Ofilius on mules in D. 21.1.38.7. 1 D. 21.1.1.2; in D. 21.1.23.4 the aedilician actions are actually called ‘ penal ’. 2 D. 21.1.1.1, 38 pr. Gellius, 4.2.1, preserves part of an earlier version. 3 But this was interpreted to include, for example, a tendency in a mule to shy. 4 This action is not mentioned in our text of the edict, and for Monier, op. cit. above,

293 n 2, 170ff., it is not classical; but see Arangio-Ruiz, Compravendita 383ff. | 5 On the provision (sanctioned by the same actions, but with limitation periods of two and six months) requiring the seller to give the stipulatio duplae against eviction and (in the formulary) to promise the absence of the edictal defects, see de Zulueta, Sale 50f.; Lenel, HP 562. 6 D. 21.1.1 pr.; ef. h.t. 63. The Digest preserves elaborate discussion by the jurists of

what constitutes a morbus or a vitiwm in a slave or a beast, but no word of anything else.

7: In giving effect to their edict the aediles applied some parts of the Praetorian edict (e.g. the actiones de peculio and exercitoria); Kaser, Mél. Meylan 1.173ff. 8* T. Mayer-Maly, Locatio conductio (Vienna, Munich, 1956); Watson, Obligations 100ff.; Crook, Law and Life 152ff., 192ff., 221ff.

294

Contract a thing, movable or immovable, l.c. operarum, the hire of services, and l.c.

operis, the giving out of a job of work to be done, e.g. the building of a house, the repair of a garment, the transport of merchandise.* This terminology, however, is not Roman, and even the trichotomy itself 1s no more than implicit in the texts.1 The Romans seem never to have devoted much thought either to the analysis of the contract or to the social needs which it regulated. For us, however, it is important to realise that the renting of land

and dwellings was not treated on different principles from the hire of a chattel, so that what corresponds to the English law of landlord and tenant

is to be found under the heading of l.c. rei. | There is even less evidence about the origin of locatio conductio than about

that of emptio venditio, but the two are no doubt closely linked, and here also the key must lie in the development of the bonae fidet actions. Mommsen, as we have seen,” finds the origin of both contracts in the imitation by private persons of contracts concluded informally by magistrates on behalf

of the state, the typical l.c. operarum of public law being the contract by which the magistrate hires the services of lictors and other free attendants, while the contracts of the censors formed the model both of l.c. rei and l.c. operis, for they were the magistrates chiefly concerned both with the leases of public land, where the state received money, and with the placing of con-

tracts for public works, where the state paid the money. Some think that, here also, reciprocal stipulations were first used, while others assume a stage

during which the contract was ‘ real’, in the sense that it did not become binding until executed on one side, in particular in the case of l.c. rez.? (b) Societas and Mandatum. (1) Societas.t The Roman societas has a wider

meaning than the English ‘ partnership ’, for it includes any agreement for the common exploitation of capital or labour for & common purpose, whether

that purpose be profit or not. Its scope in the developed law may vary from societas omnium bonorum, in which the parties agree to own all their present property in common and share future acquisitions, to societas unius ret, 1.€. for a single transaction. Between these two extremes lies s. unius negotzationis (partnership in a single branch of business) and s. universorum quae ® The origin of this rather confusing terminology is probably to be found in the literal meaning of the words. Locare is to ‘ place’ a thing with another, whether that other is to enjoy the use of it or to do work on it; the workman similarly ‘ places’ himself with the employer who ‘ takes him with him’. Cf. Nicholas, Introduction 182f.

1 For Schulz, CRL 542ff. it is not even implicit. 2 Above, 289.

3 Pernice, Labeo 1.466ff. One root may be in the relations between patrons and their clientes (see e.g. Kaser, RPL 183; RPR 1.564f.: contra, Mayer-Maly, SZ 82 (1965) 408). On H. Kaufmann, Die altroém. Miete (Cologne, 1964) see Mayer-Maly, op. cit. 406ff.; on Amirante, BIDE 62 (1959) 9ff. see Kaser, Zura 11 (1960) 229ff. Also Alzon,

RHD (1963) 558ff.

4 V. Arangio-Ruiz, La societa in dir. rom. (Naples, 1950), on which see Wieacker, SZ 69 (1952) 302ff.; Watson, Obligations 125ff.; Crook, Law and Life 229ff. 295

J.—I]

Private law to the fall of the republic: obligations ex quaestu veniunt (in which the parties share all acquisitions made by any type of business transaction). The discovery of the new fragments of Gaius * has added to our knowledge of the history of this contract. S. omnium bonorum, we can now say with some confidence, developed out of a more primitive arrangement sometimes (though not by Gaius) called consortium,® whereby children succeeding to their father, instead of dividing their inheritance, continued to enjoy it in common. This would often be the most practical course to adopt in the case for instance of a small farm. We also learn that a similar arrangement for pooling property could be artificially created between persons who were not coheirs by some form of legis actio, and that there was thus at one time a civil law form of partnership, open only to Roman citizens and not created by mere consent, as opposed to the consensual contract of classical times, which was iuris gentium. This form had the further peculiarity that a single member could validly dispose of common property by manumitting a slave or alienating a thing by mancipation, though it is probable that the others could prevent such disposition if they vetoed it in time. We do not know

what sort of legis actio was used for creating artificial consorttwm, nor whether anything was needed to constitute a ‘ natural ’ one beyond the mere

omission of the steps required for dividing the inheritance.’ It is more

doubtful whether forms of societas other than omnium bonorum have the same origin; they are, in classical law, treated as species belonging to the same genus as s. omnium bonorum; they are equally characterised by fraternitas;* condemnation results in ignominia,® while, on the other hand, the socius cannot be condemned for an amount greater than his resources permit (beneficium competentiae).! On the other hand, it is urged that the objects of a commercial partnership are so different from those of s. omnium bonorum that its origin must have been different, though the s. omnium bonorum may subsequently have influenced its rules.” The truth is, presumably, that there existed in early law many sorts of arrangement, each having its own rules, 5 3.154-154b; see de Zulueta, Gaius, ad h.l.; id., JRS 24 (1934) 168ff., 25 (1935) 19ff.; Bretone, Labeo 6 (1960) 163ff. 6 Gellius, 1.9.12, speaks of anticum illud consortwwm quod ture atque verbo Romano appellabatur ‘ ercto non cito’. Gaius also uses the last phrase and explains it as meaning dominio non diviso. See Festus s.vv. erctwm citumque and sors, Bruns 2.8 and 40. The etymology is still an unsolved puzzle; de Zulueta, JRS 25 (1935) 21.

7 Gaius treats consortiwm as obsolete, but Kunkel, dnnales de la Facwlté de Droit @Istanbul 4-5 (1955) 56ff., argues from Pliny, Ep. 8.18, and other texts that the original ‘natural’ consortiwm retained some life, not indeed with the peculiar incidents mentioned above, but in the sense that the relationship of children succeeding their father was not seen as a consensual one and was not therefore actionable by the actio pro socio. 8 D, 17.2.63 pr. (interpolated in form but not in substance, Lenel, ZP 298). 9 Gai. 4.182.

1 D. 42.1.22.1; Lenel, HP 298. 2 Pernice, Labeo 1.443ff. 296

Contract and that it was juristic abstraction which brought them under one head, just as it brought all the different arrangements for providing goods and services

in return for money under the two carefully distinguished categories of emptio venditio and locatio conductio.

One form of societas unius negotiationis needs special mention, that entered into by publicani, i.e. tax-farmers and other persons who paid the state money in return for permission to exploit public resources, such e.g. as the state salt-works.*® These publicant formed a powerful capitalist class in the later republic and it is clear that their contracts differed from ordinary societates not only by reason of their complicated organisation, the magnitude of their undertakings, and their relationship to the state, but also in the legal relationship of the partners inter se.* It also appears that in addition to the socii proper there were persons who contributed capital and shared in the profits without taking any part in the management and without risking more than the amount of their contributions. There is no clear evidence that these ‘ shares ’ (partes) were alienable,* but they obviously provided opportunities for investment. It is sometimes asserted on the strength of one text ® that

societates publicanorum could, unlike ordinary partnerships, have a corporate personality, but it is difficult to see why an association formed to undertake a temporary contract (usually for five years) should have needed incorporation.’

(2) Mandatum.? Mandatum is a contract by which one party undertakes gratuitously to perform some service for the other. The principal (mandator) can, by an actio mandati directa, obtain damages if the undertaking is not carried out properly and can insist on the agent’s handing over to him any advantage accruing through its performance; the agent, on the other hand, has an actio contraria to secure reimbursement of his expenses.

The date at which it first became possible to bring an action on mandate is not known; it must have been after the passing of the ler Aquilia® and

before 128 B.C., for in that year it is said that Sextus Iulius as praetor refused to grant an action against the heirs of the agent, whereas a few years later another praetor, Livius Drusus, allowed one in such cases.!

3 D. 39.4.12.3; 13 pr. 4 See e.g. Buckland 513.

5 Mitteis, RPR 413. 6 D. 3.4.1 pr.: vectigalium publicorum socits permissum est corpus habere.

* For discussion see P. W. Duff, Personality in Roman Private Law (Cambridge, 1938) 141ff.

8 V. Arangio-Ruiz, Zl mandato in dir. rom. (Naples, 1949); A. Watson, Contract of Mandate in Roman Law (Oxford, 1961) ; id. Obligations 147ff.

9 Because the second chapter gave the principal creditor an action against the

297 |

adstipulator who fraudulently released the debtor; such action would have been unnecessary had the actio mandati already existed, for the adstipulator is necessarily the agent

of the principal creditor and can be sued on the mandate if he causes loss to his

principal; Gai. 3.215f. 1 Auct. ad Herenn, 2.13.19.

Private law to the fall of the republic: obligations (c) Negotiorum gestio ? and tutela (quasi-contract). Closely allied to man-

datum are the quasi-contractual relationships of negotiorum gestio and tutela. In both these cases there is an agent who acts in the interests of a principal, but there is no contract (according to later ideas) * because there is no agreement, for the negotiorum gestor is precisely a person who * carries on affairs ’ which concern another without any instructions, and the tutor, though he acts on behalf of his ward, is appointed, not by agreement with

the ward, but by the law. These agents are however responsible for the proper management of their principals’ affairs, and can, within limits, secure

reimbursement of their expenses. As in the case of mandate, the action brought by the principal is the actio directa, that brought by the agent the actio contraria.* It is now, however, very generally held that the respective spheres of man-

date and negotiorum gestio were not originally, or even in the earlier classical law,* the same as those assigned to them in the system found in the Corpus Iuris. There an agent who has any authority at all is a mandatary, while a negotiorum gestor is a person who intrudes unasked into another’s affairs. In the earlier law, on the other hand, it seems that the actio mandati lay only when there was special authorisation of the act done, and that anyone acting under a general authority (in particular a procurator in the sense of ‘ general agent ’)® was a negotorium gestor. At any rate a procurator is commonly regarded, even by the classical lawyers, as one who negotia gerit, and the actio n.g. is the only one available to or against a curator,’ who is certainly not a person who pushes himself into other people’s affairs.

C. Security. The law of obligations is completed by the law of security, i.e. : of the methods by which a creditor seeks to guard himself against the possible insolvency of his debtor. Such security may be either personal or real according as the creditor obtains a right against a third person who will be liable for the debt in addition to the debtor himself, or a right over a piece of property which he can use in some way to satisfy himself if the debtor does not pay the debt. The two institutions take widely different paths, but in their origin they are not so different. The primitive (personal) surety is, as

2 The abstract noun is not Roman. 3 Above, 271f. 4 There appears to have been an actio in factum as well as a bonae fidet tudicium;

on the relation between them see Arangio-Ruiz, Ii mandato in dir. rom. (above) 34ff.

5 The dating of the transition is disputed; sce A. Watson, Contract of Mandate in Roman Law (Oxford, 1961) 36ff., with reff.

6 It also embraced procedural representation by cognitor and procurator ad litem (though the date of origin of the latter is uncertain), and since in the edict it comes immediately after the provisions dealing with such representation, this may be one of its roots; Kaser, RPE 1.587. But see Watson, Obligations 193ff. 7 Some texts describe it as utilis, but they are probably interpolated; Buckland 538 n 20; Lenel, SZ 35 (1914) 203ff. 298

Contract we have seen,® a hostage, i.e. a person who is § liable’, not in addition to, but instead of the person whom we call the ‘ principal debtor ’; if the prin-

cipal pays, the surety is released. So, too, in the earliest times, with real security; the debtor has the right to release the thing pledged by paying his debt, but if he does not the thing is ‘ liable ’, i.e. the creditor keeps it.® Per-

sonal security appears to have developed at an earlier stage in Roman history than did real security; it was a social duty to become surety for one’s

relations and friends, and we have already noticed the importance of this practice in connexion with procedure.'! Even in the law of classical times and

later, when security is needed for procedural purposes, it is almost exclusively in the form of suretyship that it is required.’ (i) Personal security. The original nature of suretyship is visible, in historical times, only in the praedes, who, as we have seen, probably engaged them-

selves instead of, not in addition to, the ‘ principal debtor ’.® In historical times, however, praedes only occur where security has to be given to the state or in connexion with legis actio sacramento,* and for private as well as most procedural matters the form of suretyship used is quite a different one, in which the surety makes a promise by stipulation, by which he becomes indebted to the creditor in addition to the person who may now be described quite properly as the ‘ principal debtor ’. Three forms of such stipulation are described by Gaius,* sponsio, fidepromissio and fideiussio, which last is of considerably later origin than the other two, and may not have come into use until after the end of the republic. Sponsio is probably the oldest form; here the surety is asked idem dari spondes? and replies, spondeo, the use of this

word being confined, as in the case of ordinary stipulations, to Roman citizens. The fidepromissor is asked idem fidepromittis ? and in this form the

contract is open to peregrines as well. Otherwise there is no difference between the effects of the two forms;® both enable the creditor to take his choice of suing either the principal debtor or the surety;‘ both can be used

8 Above, 159. » Cf. beluw, 801£. 1 Above, 193.

2 Cf. Nicholas, Introduction 149ff. 3 Above, 187f.

4 It is held by some authorities, e.g. Kaser, ZPR 302, Girard 797, that praedes as well as vades were never used except for contracts with the state and for procedural purposes. This view is prima facie supported by Festus, s.v. praes (Bruns 2.26): Praes est is qui populo se obligat interrogatusque a magistratu si praes sit, lle respondel: praes. But see Mitteis, Festschr. fir E. I. Bekker (Weimar, 1907) 121.

5 3.115ff. 6 Except that only the sponsor has the actio depensi; above, 196.

7 But litts contestatio with one will make it impossible for an action to be brought against the other, and the creditor will thus always sue the surety if there is any doubt

about the solvency of the principal debtor. To this extent the surety is in the same position as a correal debtor, and it has been contended (Levy, Sponsio 79ff.) that there was originally no legal distinction between the two, though de facto a correal debtor who became party to a debt which did not concern him would be a surety; legal differen299

Private law to the fall of the republic: obligations only to guarantee obligations which have themselves been created by stipulation; in neither. case is the heir of the original surety liable, and both forms were subjected to restrictive legislation which began some two centuries

before the end of the republic. The first statute was a lex Apuleia which enacted that where there were several sponsores or fidepromissores any one

of them who had paid more than a proportionate share of the debt might recover the excess from his co-sureties. A little later the lex Furia ® went further by providing, so far as sureties in Italy were concerned, that the debt should be divided by the number of sponsores or fidepromissores living at

the due date and that anyone who had been made to pay more than his share should have an action, exercisable by manus iniectio, against the creditor for recovery. It also enacted that sponsores and fidepromissores should be

released from their lability two years after the due date of the debt. The restriction of liability to a proportionate share where there were several sureties made it important for each surety to know how many others there were, and a lex Cicereia completed the system by requiring that the creditor should state openly the amount of the debt and the number of sureties, and that 1f he did not the surety could secure his release. A further restriction was introduced by a lex Cornelia, which, with some exceptions, forbade any

one man to be surety for any one other for a greater amount than 20,000 sesterces in the same year.°

The object of all this restrictive legislation was presumably to lighten the burden cast upon sureties and so make it easier for people in need of credit

to obtain the necessary security, but apparently it overshot its mark by reducing too greatly the value of the surety to the creditor, and this led to the introduction of the new type of surety, the fidezussor, to whom none of the restrictions (except that of the lew Cornelia) applied. The fidetussor, like the older sureties, entered into his obligation by stipulation, being asked id + fide tua esse iubes? but this method could be used to guarantee any sort tiation, Levy thinks, came with the lex Cicereia. General opinion is, however, against him. See W. Flume, Studien zur Akzessorietéit der Biirgschaftsstipulationen (Weimar, 1932) 1Liff.; Kaser, APH 1.662 n 21.

’ This statute was confined to suretyships in Italy, whereas the lex Apuleia applied in the provinces as well.

® There is no direct evidence for the dates of any of these statutes, but it is clear from Gaius that the order was 1. Apuleia, |. Furia, 1. Cicereia, l. Cornelia. The author of the l. Cicereia may have been C. Cicereius who was praetor in 173 B.C. (Livy 42.1) and the use of a praeiudiciwm (Gai. 3.123) where one would have expected an exceptio

points to a time when the formulary procedure was at any rate still undeveloped.

prove this. .

The 1, Apuletia, on the other hand, since its provisions extended to the provinces, cannot well have been passed earlier than 241 B.C., the date of the first province (above, 68).

The 1. Cornelia may well be due to Sulla (circa 81 B.C.), but there is no evidence to 1 The ms of Gaius reads, somewhat doubtfully,. idem, but other authorities give id,

eg. D. 45.1.75.6; CIZ 3.934 (Bruns 1.352): id fide sua esse iussit Titius Primitius. 300

Contract of obligation whether it had been created by stipulation or in any other way. There is, however, no trace of fideiussio in Cicero and it is probable that it did not exist until the time of Labeo.? In addition to the three stipulatory types of suretyship there were in the classical law two others which were free from rules of form, mandatum pecuniae credendae® and constitutum. The origins of both go back to the repub-

lic, but neither was developed until imperial times. Mandatum pecuniae credendae is the use of mandate to achieve the purposes of suretyship. If A gives B a mandate to lend: money to C, A has in effect become surety for C, because if C fails to repay, B will be able to bring an actio mandaii contraria for reimbursement against A on the ground that he has suffered loss by carrying out A’s instructions. This method had evidently been tried before the end of the republic, because Servius Sulpicius denied its validity on the

ground that the mandator himself had no interest in the fulfilment of the mandate; Sabinus however thought differently, and in classical times it was extremely common.*

Constitutum was an informal promise to pay an already existing debt, enforceable by a praetorian action.® It had two advantages over the actio certae creditae pecuniae, by which the existing debt would have been enforced. It gave the plaintiff not just the amount of the debt, but damages for the delay in payment (i.e. interest); and it had a sponsio et restipulatio for half the sum in issue instead of a third.® In classical times the debt might be one owed by the promisor or by a third party, and in the latter case there

was clearly a kind of suretyship. Only constitutum of a debt owed by the promisor, however, seems to be evidenced for republican times.’

(ii) Real security. We have already considered * the contractual aspects of the two basic forms of real security. In fiducia ownership in a thing was conveyed to the creditor by mancipation or in iure cessio, and the debtor conse-

quently lost all rights in rem with respect to the thing, retaining only his Whether the distinction between the fideiussor’s id and the idem of the older sureties points to a fundamental difference in the newer institution, as Pernice holds (SZ 19 (1898) 182; see the exposition by Schulz, CRL 499ff.) is doubtful. Gaius’ complete silence is against Pernice’s view; Levy, Sponsio 79ff. Buckland, Jurid. Rev. 53 (1941)

281ff., contended that litis contestatio with a fideiussor (and perhaps with other adpromissores) did not release the principal debtor; but see Levy, Seminar 2 (1944) Off. 2 Levy, Sponsio 123; Flume, Studien (above) 41 n 2.

3 The term mandatum qualificatum is not Roman. 4 Gai. 3.156.

5 See Schulz, CRL 560ff. ; 6 Gai. 4.171; cf. above, 194.

7 Cic. pro Quinct. 5.18. The receptum argentariu, whereby a banker promised to pay the debt of his client, was closely similar, and was merged by Justinian in constitutwm. There appears to be no reliable trace of its existence in republican times. 8 Above, 285ff. The word fiducia occurs in Plautus, e.g. Trin. 117, 142, but the institution is probably older; see Manigk, PW 6.2290. For comparison with the English law of mortgage see Hazeltine, Preface to R. W. Turner, The Equity of Redemption (Cambridge, 1930). 301

Private law to the fall of the republic: obligations personal action against the creditor if the creditor alienated the thing before the debt fell due or in any other way broke the agreement. In pignus the debtor parted only with possession of the thing ® and was therefore protected by vindicatio against unlawful alienation, as well as having a personal action against the creditor for this or other breaches of the agreement. In classical times it was evidently usual for there to be a special term in the agreement allowing sale,’ and if the thing fetched more than the amount of the debt the excess had to be paid to the debtor,’ also, the creditor could probably not simply keep the thing as his own if the debt were not paid unless there were a special clause (lex commissoria) allowing him to do so, just as there had to be such a clause if he were to retain a pignus. But these rules are no doubt comparatively late developments; even in Cicero’s time the rule was probably simple: if the debt remained unpaid after it had become due the creditor was freed from his duty to re-convey.* Indeed, it is probable that there was a still earlier stage at which the pledge, like the personal surety, was primarily given in satisfaction of the debt and not as an accessory security, and there-

fore that all that the debtor retained was a right of redemption up to a certain date for the amount of the debt; after that date the creditor simply kept the thing.* This, at any rate, is what the analogy of other early systems, and in particular of both Greek and Germanic law, would lead us to suppose,° but the early history of Roman pledge is the subject of debate.°® The last form of pledge to develop was that known as conventio pignoris

or hypotheca,’ i.e. the pledging of a thing by mere agreement, without the ® According to the usual view therefore the creditor was at first protected against third parties only by a possessory interdict, and before the introduction of the interdicts would have had no protection at all. He would, in other words, have had only a de facto security. Kaser, RPA 1.457f., 471ff., consistently with his view (above, 142) that primitive ownership was no more than a relative right, holds that the creditor was protected by the legis actio sacramento tn rem, and that the actio Serviana (below, 303) merely

took the place of this remedy (the need for a special action being attributable to the fact that the formulary vindicatio required proof of absolute ownership).

1 Formula Baetica (Bruns 1.334; FIRA 3.295). 2 PS 2.13.1. Obligations 179 n 2). 4 Manigk, PW 9.295, 355; Rabel 494. 3 Manigk, PW 6.2299; Cie. Flacc. 51 (but for a different interpretation see Watson,

5 For early Greek law see Manigk, PW 17.355; for English law, Pollock and Maitland 2.184: ‘In the case of the gage there probably was at first no outstanding duty on the side of the debtor when once the gage had been given ... he handed over

something of sufficient value to cover and more than cover the debt; the debt was satisfied; the only outstanding duty was that of the recipient of the gage, who was bound to hand it back if within due time its giver came to redeem it.’

6 See literature cited by Watson, Obligations 180. The old conception seems to be reflected in the practice still found in classical times of inserting a special clause which reserved the creditor’s right to exact any deficiency from the debtor; as Pomponius says, D. 20.5.9.1, the creditor had (according to classical principles) exactly the same right in any case. 7 The argument of M. Fehr, Beitrdége zur Lehre vom r. Pfandrecht (Upsala, 1910) that hupotheca is everywhere interpolated has been generally rejected. 302

Contract transfer of either ownership or possession. Clearly there were grave inconveniences about both fiducia and pignus. In the first place these both meant that the debtor was deprived of the use of his thing, unless, as sometimes happened, it were given back to him either precario or on hire, and secondly the credit-raising value of the thing was necessarily exhausted by a single pledging, for a thing cannot be conveyed or delivered successively to two creditors even though the amount of the first debt be very much smaller than the value of the thing. The first case in which it became possible (among private persons) for a thing to be pledged by mere agreement was apparently that of the agricultural tenant who desired that the effects which he brought on to the land (invecta et illata) should be available as security to the landlord for the payment of the rent. Obviously he could not be deprived of the possession of them for they were probably all that he had, and the land could not be worked without his slaves, beasts and agricultural implements. If then he had agreed that they were to be pledged it was allowed that the landlord

should have an interdict (interdictum Salvianum) which enabled him to obtain possession of them if the rent were not paid when it fell due.* This may have been available only against the tenant ® but a praetorian action (actio Serviana) was later allowed whereby the landlord could get possession of the things even from third parties into whose hands they had come, e.g. by purchase from the tenant. This action was subsequently generalised, so that it lay when any debtor had agreed that a thing belonging to him should be pledged for his debt, and the creditor could thus claim the thing wherever it was if the debt were not paid when it fell due.’ Once he had got the thing he was in exactly the same position as if it had been pledged to him from the first by delivery of possession. It was at one time commonly held, owing to the Greek name, hypotheca, under which the institution commonly appears in the Corpus Juris, that its introduction was due to Greek influence, but it is now generally agreed to be of native growth. Pignus and hypotheca are treated throughout as one and the same thing;? in some cases possession is transferred at once, in others it is not, but that is all. The word pignus is freely used for both cases and indeed occurs in the formula of the very action which made pledge without possession possible.* It is clear too that the particular case from which hypotheca arose was one which was but a slight 8 Gai 4.147.

® C. 8.9.1, but perhaps interpolated; see Kreller, SZ 64 (1944) 306ff.; Kaser, RPE 1.472.

1 Justinian (Inst. 4.6.7) distinguishes actio Serviana available to the landlord from the actto quast Serviana or hypothecaria used by ordinary hypothecary creditors, but this distinction is doubtfully classical; Lenel, HP 493; for another explanation see Watson, Obligations 180.

2 D. 13.7.1 pr. (Ulpian): Pignus contrahitur non sola traditione, sed etiam nuda conventione, etst non traditum est; D. 20.1.5.1 (Marecian): Inter pignus autem et

hypothecam tantum nominis sonus differt. 8 Actio Serviana; Lenel, HP 494. 808

Private law to the fall of the republic: obligations extension of the original principle. The land onto which the tenant’s goods are brought is in the possession of the landlord, though they themselves are not, and it was doubtless the feeling that they were to some extent within

the landlord’s control which prompted this departure from the original necessity of possession.

As to date—the existence of pledge without possession is attested for Cato’s time (284-149 B.C.) by a clause in his form of contract for the sale of olives on the tree, which provides that everything which the purchaser brings into the olive-yard is to serve as security for payment,‘ and a similar

clause with respect to slaves and cattle in the form for a sale of pasture, there being in the latter case a further clause which provides that any litigation concerning the matter is to take place at Rome.® But there is nothing to show against whom the pledgee could proceed, or by what remedy.* One thing, however, is certain: though the a° Serviana was definitely a praetorian action there is no trace of any edict that promised it. In the Edict as revised by Julian, the pattern formula followed immediately after the interdictum Salvianum.’ Some authors * have held this to be evidence of an origin so early that it preceded the practice of issuing edicts, and so, probably, the lew Aebutia. Others ® hold, on the contrary, that the action was an innovation of Julian’s, who had no need, since he was settling the whole Edict, to insert a special clause promising the new formula that he created. This view would be difficult to support if the interdictum Salvianum was already in early times subject to the restriction that it could not be used against third parties, for creditors would have insisted on better protection. But if the restriction is interpolated, it may well be that the actio Serviana dates only from Julian’s time, and was introduced by him rather as a technical improve-

ment in procedure than as an important substantive alteration. It is more likely,? however, that the action is late republican (its originator being perhaps Servius Sulpicius Rufus),? and that since it served essentially the same purpose as the interdictum Salvianum, one edict covered both formulae.

4 146.5 (Bruns 2.49). 5 149.7-8 (Bruns 2.50).

6 For discussion see Watson, Obligations 180ff., with reff. Kaser, RPR 1.457f., 471f., in accordance with his thesis that early ownership was relative (above, 142, 154), argues that the pledgee would have had an ordinary vindicatio, the need for a special remedy

emerging only as ownership came to be absolute. 7 Lenel, HP 493.

8 E.g. Manigk, PW 9.353. 9 E.g. Siber 2.124.

1 Kreller. SZ 64 (1944) 334ff.; Kaser, RPR 1.472. 2 Above, 93.

304

CHAPTER 18

Criminal law in the republic 1. FROM THE XII TABLES UNTIL THE LEGISLATION OF SULLA We have seen that in early Roman law, as in other early systems, the field covered by modern criminal law falls largely within the province of the ordinary civil law of delict. Thus, theft and assault, typical crimes according

to modern law, led only to private actions at the instance of the party injured, though the result of such action might be the infliction of a physical penalty on the wrongdoer. Moreover, even where the wrong appears not to be regarded as private, it is often a religious matter. Crime and sacrilege are not originally distinct categories. This is probably true even of treason (perduellio), but at least by the early republic this seems clearly to be a secular crime. Beyond this, for the pericd down to the appearance of the quaestiones in the later second century B.C. and more particularly down to the reforms of Sulla, there is little agreement as to how the scanty evidence should be

interpreted. It should be noted, however, that the question at issue is primarily one of procedure, not of substantive law. That murder, to take the most obvious case, was legally punishable is not in doubt. What is disputed is whether the proceeding by which the penalty was enforced was a private

one, as in the case of theft, or was in some sense an exercise of public authority. A. MommMseEn’s Account. Until nearly the middle of this century Mommsen’s

brilliant reconstruction ? went unquestioned. For him the root of the early criminal law lay in coercitio, the higher magistrates’ general power of enforcing order and exacting obedience to their commands. So long as this police power remained unlimited and unregulated, crimes were indeed punished but there was nothing which could be called criminal law. The crucial step from coercitio to zudicatio * was the result of the introduction. first attributed * to 1 E.g. Tab. vri11.9, above, 170 n 8; ef. 118 n 7. 2 It is summed up in his last great work, Rdém. Strafrecht (StrR), published when he was 81 (1899), but it had been worked out long before and is embodied in his R6m. Staatsrecht (Stk). 3 But the distinction between the two is not always clearly maintained by Mommsen; Strachan-Davidson 1.103ff,

4 Hardly reliably (see above, 12ff., and below, 309); but it is usually held that provocatio must have existed by the time of the XII Tables, because they enacted (Tab. 1x.2) that only the comitia centuriata might decide de capite civis. But see below, 309.

305

Criminal law in the republic

the lex Valeria of 509 B.C., of provocatio, the right of every citizen to appeal to the people against a sentence of death passed on him domi (i.e. in Rome or within the first milestone from the city).° Out of this there developed the trial before the comitia centurtata, which took place in two stages, a trial proper and an appeal.® The first stage is the investigation which the magistrate would in any case normally make before proceeding to coercitio; but if there is a possibility of provocatio, this investigation (anquisitio) is held (after a summons to the accused to appear on a certain date — diez dictio) before an informal gathering (contio)‘ of the people, so that they may acquire the information which they will need in the event of an appeal.

The anquisitio must be adjourned at least twice, with at least one day between hearings. The magistrate then gives his decision. If it is an acquittal, the matter is at an end, but if it is a condemnation and the accused makes a

provocatio, the second stage then follows. This is a sitting of the comitia, held after the twenty-four days’ notice required for resolutions. There is probably no discussion or evidence; the people can merely, as with legislation, either accept or reject the magistrate’s proposal. Mommsen saw in this

power to quash the magistrate’s sentence the explanation of a fact which was apparently quite incompatible with his thesis, namely, that none of the comitial trials of which we have any knowledge was conducted by a magistrate who had the power of coercitio. It was not, he maintained, consistent with the dignity of the consuls (or later of the praetors) that their sentences should be quashed by the comitia, and in capital cases therefore they must have delegated their functions, and in particular their power to convoke the comitia, to their assistants the quaestors * or, in the case of treason, to the duoviri perduellionis.°

The same jurisdiction came to be exercised also by the tribunes, but only for political offences. For their competence in this respect derived originally

from their position as heads of the plebeian community: they used what were in effect revolutionary powers to punish any person, in particular any 5 For a long time provocatio did not prevail against the act of a dictator, and exceptional magistracies during the revolutionary period, such as that of Sulla, were expressly exempt from it. 6 Most of our information comes from Cicero (Dom. 45) and from the commentarius vetus anguisitionis of M. Sergius (below, 310). 7 Little is known of the procedure, but the magistrate no doubt had a very free hand. As at other contiones, he could allow selected persons to speak, and this meant that the accused usually had the assistance of counsel; Mommsen, Sirk 163ff. 8 See above, 50f., and below, 310. 9 The name should probably be duoviri perduellioni iudicandae (Mommsen, StR 2.617).

We hear of them on only three occasions: the legendary trial of Horatius (Livy 1.26; Bleicken, SZ 76 (1959) 333ff.), the trial of M. Manlius (384 B.C.), which according to some, says Livy, 6.20.12, was conducted by them, and the trial of Rabirius, who was defended by Cicero, in 63 B.C.; but this last was an artificial revival, the functions of the duoviri having long since been taken over by the tribunes.

306

From the XII Tables until the legislation of Sulla patrician magistrate, who infringed the rights of the plebs. After the end of the struggle between the orders, the tribunes, now in fact though not in law! magistrates of the whole state, continued to exercise similar powers, especially against magistrates and others guilty of abusing the public authority

with which they had been entrusted. In the later republic they entirely superseded in this the [Tviri perduellionis.

Provocatio, and with it the comitial trial, was extended in several ways. According to tradition magisterial fines were, already before the XII Tables, subjected to a limit (eventually expressed as 8,020 asses),? and sentences

above this limit were therefore open to provocatio. This was the basis of the criminal jurisdiction of the aediles (both curule and plebeian) before

the comitia tributa, a jurisdiction which was especially concerned with offences in the markets (including usury and profiteering in grain) and the streets. The pontifex maximus dealt in the same way with the misconduct of the priests subject to him. Magisterial sentences of scourging were also subjected to provocatio, apparently by alex Porcia proposed by the elder Cato

in 198 or 195 B.C. It was perhaps another lex Porcia, of about the same time, which took the important step of allowing provocatio beyond the limits of the city (militiae).* The citizen was thus protected anywhere in the Roman world. It was a remarkable feature of Roman practice that the death penalty was hardly ever in fact inflicted. No doubt the magistrate who initiated the proceedings could use his coercitio for the purpose of keeping the accused in

custody,* so that if the people did not pardon him the sentence might be executed; but any tribune might then use his auzxilium to free the prisoner, and it is clear that persons of consideration could always count on being freed in this way. From the second century B.C. it seems to have become the practice that the accused should remain at liberty until the people had actually decided against him and should thus be able to go into exile.* After he had left it was usual for the comitia to pass a decree of ‘ interdiction of

fire and water’ against him. The exact effect of this is disputed,* but it 1 Hence a tribune wishing to bring a citizen before the people on a capital charge had

to ‘ask for a day for the comitia’ from a magistrate with imperiwm; see e.g. Livy 26.3.9. 2 Originally thirty oxen and two sheep; see Bleicken, PW 23.2.2450f. 3 See further, Brunt, TR 32 (1964) 447£. Mommsen held (Str 143) that women had no right of provocatio against capital sentences (as opposed to fines); for arguments against this inherently improbable view see Strachan-Davidson 1.141ff.

4 Imprisonment was never regarded by the Romans as a normal form of punishment in itself; ef. D. 48.19.8.9: Carcer enim ad continendos homines, non ad puniendos habert debet.

5 This is attributed to one of the leges Porciae and aliae leges by Sall. Cat. 51.22, 40; ef. Cie. Dom. 78; but the significance of this is disputed; Bleicken, PW 23.2.2449. G. Crifé, Ricerche sull’ exiliwm 1 (Milan, 1961) argues that exile always was a right,

: 807

but see Fuhrmann, SZ 80 (1963) 451. 6 Strachan-Davidson 2.23ff.

Criminal law in the republic certainly had the result that he could not return without making himself liable to be put to death. In Mommsen’s account, the next stage in the development of the criminal trial is the appearance of the permanent courts, known as quaestiones perpetuae, before which most cases came in the late republic. These quaestiones were established by a number of different statutes, the first of which we have any knowledge being a lex Calpurnia of 149 B.C., which provided for the trial of provincial governors who were accused (after the end of their term of office) of extortion (res repetundae). Other laws followed, and Sulla, by a whole series of leges Corneliae, reduced the practice to a system, afterwards completed by further statutes. In their Sullan form these quaestiones are jury courts presided over by a magistrate or his deputy. Technically perhaps they produce no infraction of the principle that it is for a magistrate to investi-

gate crimes and to pronounce the fitting penalties, but in substance they differ radically from the comitial court. The proceedings now are accusatory, not inquisitory: the prosecution is no longer brought by the magistrate, but

by a private individual, and it is he who adduces the evidence, though he may have official assistance in collecting it. Again, though the jury is called

a consilium and is, in form, similar to the consilium which a magistrate chooses to advise him in making any important decision, there are fundamental differences. The ordinary consilium is freely chosen by the magistrate

and he is legally at liberty to disregard its opinion, whereas in a criminal quaestio the composition of the consilium and the way in which it votes are laid down by lex, and its vote is decisive. The presiding magistrate or officer

does not even sum up, and altogether he does not play nearly such an important part in the trial as does an English judge. B. Kunxeu’s ALTERNATIVE. A systematic alternative to Mommsen’s recon-

struction was not propounded until 1962, by Kunkel.’ He started from certain weaknesses in Mommsen’s account which earlier critics had pointed out. (i) Provocatio and the comitial trial. Brecht * had shown that in the comitial trials of which there is any record, other than those concerning fines by the pontifex maximus, there is no trace of Mommsen’s division into trial and appeal. The first stage was not a trial culminating in a judgment, but a preparatory hearing leading up to a proposal by the magistrate, the second stage being a decision on that proposal. On this view there was no need to resort to Mommsen’s unsupported hypothesis of a delegation of authority in

order to explain away the fact that the magistrates in question had no 7 W. Kunkel, Untersuchungen zur Entwicklung des rom. Kriminalverfahrens in vorsullanischer Zeit (Bayer. Akad. d. Wissenschaften, phil.-hist. Kl. 56), referred to hereafter

as Krim. Modifications or elaborations of the thesis in SZ 83 (1966). 219ff.; SZ 84 (1967) 218ff.; SZ 85 (1968) 253ff.; Symbolae M. David (Leiden, 1968) .1.111ff.; PW

24.1.720ff. | 8 SZ 59 (1939) 261ff. 808

From the XII Tables until the legislation of Sulla coercitio: they were not purporting to exercise this power, but were merely proposing a sentence to be imposed by the comitza. Brecht conceded, however, that Mommsen’s two-stage trial did occur in the case of fines imposed by the pontifex maximus, and it was left to Bleicken ° to argue that even here Mommsen’s reconstruction was unjustified. He pointed out that the pontifex maximus was not properly a magistrate, that there was no evidence that he could convoke the comitia tributa, and that in any case the decisions of which we know did not, as on Mommsen’s theory they should have done,

simply confirm or quash the sentence imposed by the pontifex: they both | quashed the fine and required the accused to obey the pontifex. This compromise resolution was more probably proposed, suggested Bleicken (and the proceedings therefore conducted), by a mediating third party who did have the power to convoke the comitia, perhaps a curule aedile.

Again, although we hear of many leges which introduce or confirm the right of provocatio, there is no reliable evidence of its use against a capital sentence; and although the Romans of the late republic plainly regarded it as one of the foundations of their civic liberties,’ there is no sign of any political reaction on this score to the introduction of the Sullan quaestiones,

to which it was clearly inapplicable.’ | Heuss and Bleicken had therefore argued that provocatio was a political institution arising, like the tribunate, out of the struggle between the orders and obtaining legal sanction only by the last lex Valeria, in 300 B.C.* Until then it was essentially a revolutionary device backed only by the political power of the plebs and directed against abuse by patrician magistrates of their legally unlimited coercitio. (On this view the famous provision of the XII Tables,* de capite civis nist per maximum comitiatum ... ne ferunto, was concerned not with limiting the magistrate’s power by provocatio, but with debarring the concilium plebis from exercising lynch justice.) The politi-

eal origin of provocatio would also suggest that it was not concerned with ordinary crime, which was in any case not within the province of coercitio: magistrates would only be concerned to take repressive action against crimes

of a public nature. : 9 §Z 76 (1959) 333.

1 Cie. de Or. 2.199: patronam illam civitatis ac vindicem libertatis; Livy, 3.45.8, describes provocatio and the auailium tribunicium as duas arces ltbertatis tuendae; ef, Cic. Rep. 2.53ff.; Rab. Perd. 11ff.; Livy 3.55.4.

2 Special courts immune from provocatio had also been created from time to time in the hundred years before Sulla, either by lex or by senatusconsultwm. Where they were ereated by lex, Mommsen could argue that what the people had given the people could take away; where they were set up by senatusconsultwm, especially the Sc. de Bacchanalibus (below, 313), he conjectured that if provocatio was indeed excluded, it was done

unconstitutionally in an emergency; StH 2.110ff.; StrR 153 n 1; 172 n 2; 257 n 2; Strachan-Davidson 1.229ff.

factum. 4 Cf. above, 305 n 4. 3 And even then, Livy says (10.9.5), contravention was only declared to be improbe 309

Criminal law in the republic (ii) The ‘ political? character of the comitial trial. That the cases brought before the comitia by the tribunes were political in character was recognised

by Mommsen. Kunkel argues that the same was true of all such trials. In the case of trials conducted by the aediles this involves a considerable exten-

sion of what can be regarded as political, so as to include acts which are detrimental to the public interest and which concern the police functions of the aediles. But serious crime was excluded, since the aediles could inflict only fines, and the crucial question is therefore that of the jurisdiction over those serious offences which are not political even in this extended sense, especially murder. Mommsen, as we have seen, attributed this jurisdiction to the quaestors, as the assistants of the consuls. The principal reason for this

attribution was the existence at the time of the XII Tables of quaestores parricidit whom he identified with the ordinary quaestores aerarii. The evidence of actual trials is inconclusive. Our sources mention only four,® of which two come from before the XII Tables and are therefore hardly to be relied on for legal details, and the third, the prosecution of Camillus in 891, even if

taken as reliable, was for the crime of embezzlement of public funds (pecu- , latus), which can be classed in this context as political. The only clearly dependable evidence is to be found in quotations by Varro ® from a commentarius anquisitionis written by a quaestor, M. Sergius, apparently for a particular trial on a capital charge, but there is nothing to show what the offence was. With such scanty evidence, the absence of any instance of a trial for ordinary crime can obviously prove nothing, especially as the historians

were not interested in such things, but Kunkel argues that more positive support can be found in the time-consuming procedure of the comitial trial. This would be justifiable for state trials, but intolerably cumbersome for the

prosecution of ordinary crimes. :

(iii) The treatment of ordinary crimes. If the foregoing criticisms are accepted, there is obviously a gap in the system propounded by Mommsen: there is no provision for the punishment of non-political crimes, except to the small extent that they are included in the aedilician jurisdiction. Moreover, Mommsen’s distinction between a magistrate’s coercitio and his iudicatio " rested on his view of provocatio; if that view is rejected, there is left only his coercitio. It was Kunkel’s purpose to fill these gaps, and in so doing to remedy two further and related inadequacies which he saw in Mommsen’s

system: that by its exclusive emphasis on the comitial trial it omits altogether the element of private vengeance which is commonly found in 5 Kunkel, Krim. 344,

§ LL 6.90 (Bruns 2.59). The commentarius dates perhaps from the early second

century B.C. and is certainly later than 243. 7 Neither is a technical term; Kunkel, Krim. 140 n 479.

810

From the XII Tables to the legislation of Sulla early law,® and that it is unable to explain satisfactorily how the system of quaestiones which does contain a private element and which, as we have seen, makes a total break with the comitial procedure in other ways, could have emerged. Kunkel argues (a) that non-political capital offences were not in the early law the subject of a public prosecution at all, but were left to be

punished by a private suit by legis actio sacramento; (b) that the Sullan quaestiones had been preceded both by extraordinary quaestiones in the second century, and also even earlier, by the exercise of a magisterial power, conferred perhaps by lez,*® to try common crimes with a consilium, or jury,

by whose verdict the magistrate was bound; and that against such acts of iurisdictio as opposed to coercitio, provocatio was never available. (a) The legis actio sacramento in capital matters. Starting from the premise

that there is no evidence of a criminal proceeding in early law for the ordinary crimes and that the element of vengeance was then still strong, Kunkel looks for evidence of a private capital suit. Since we know that in the earliest quaestio perpetua (the quaestio de repetundis established in 149 B.C.) ? the proceedings were by legis actio sacramento, there is some likelihood that this form of action would have been used, and Kunkel finds support for this in a fragment of Probus and another of Festus.’ Among the abbreviations concerning legis actiones which are preserved by Probus ° is one of which no satisfactory explanation has hitherto been found : S.N.S.Q. si negat sacramento quaerito. Kunkel conjectures that this comes

from a commentary on the legis actiones and that quaerere means here to conduct a quaestio. The sentence would then mean: ‘if he (sc. the defendant) denies, let him (se. the praetor) conduct a quaestio ’, the sense of the preceding sentence (which is not preserved) being probably ‘ if he admits, let the praetor make addictio’. This would then be a reference to a legis actio sacramento. But it would differ from the ordinary legis actio in that there would be no division into two stages; for it is evidently the praetor who is to conduct the quaestio.* Kunkel further invokes the later use of quaerere and quaestio to support the conjecture that the praetor sat with a consilium.-This would have developed out of the familiar Roman practice of

8 Kunkel, Krim. 37ff., 124f. 9 Kunkel, Krim. 68.

1 Above, 308. The inscription which is usually identified as the lex Acilia (above,

80 n 2) and attributed to 123/2 B.C. provides that the legis actio sacramento be replaced by the procedure by nominis delatio, which was to become standard in all quaestiones (below, 318f.); and the defendant’s liability was raised to double the sum extorted.

2 He also argues in a later note (SZ 84 (1967) 382ff.) that the evidence of Livy 3.13.8, and more especially of Dionysius 10.7.2, 5, suggests that the trial of Caeso Quinctius for murder was the result of 4 private suit. 3 Notae iuris 4.5 (FIRA 2.456).

4 It cannot be a tudez because at the time of the sacramentum none has yet been

appointed; and quaerere is not elsewhere used of the functions of a tudez.

311

Criminal law in the republic consulting a consilium before making any important decision, but here there

would be the peculiarity that the verdict was binding on the magistrate. This feature, in Kunkel’s view, is characteristic of an exercise of jurisdiction

by a magistrate as opposed to an exercise of coercitio. The centumviral court, which shared this feature and which also tried legis actiones, would provide a close parallel, and the juries of the Sullan quaestiones would be in this respect merely maintaining an ancient practice.°

From the si negat of Probus it is evident that the plaintiff must have asserted something, and Kunkel supposes that it was a claim to the plaintiff’s person (i.e. a vindicatio, as in the legis actio sacramento in rem, but without a contravindicatio) and he finds support for this in the fragment of Festus,° which comes from a speech of the elder Cato. It is imperfectly preserved, but Kunkel proposes the restoration: scelera nefaria fie[ bant, de quibus ut sacra-

me|nto traderetur lege est cautum. This could show that the unsuccessful defendant in our legis actio was handed over to the plaintiff for punishment, the addictio of the fur manifestus providing a close parallel, save that in that case there was probably no trial.” ° Kunkel therefore concludes that ordinary crime was from early times the subject not of a public prosecution, as Mommsen had supposed, but only of a private suit by sacramentum, and that the other capital wrongs of which we hear at the time of the XII Tables ® were probably dealt with in the same

way. Other wrongs may also have been included, such as adultery and stuprum, which one would expect to be actionable in a system based on private vengeance but which are usually held to have been made punishable only by Augustus’ legislation.’ 5 The contrast with the ordinary procedure would be less marked if, as is possible

6 §.v. sacramento (Bruns 2.34).

(above, p. 178), the place of the unus iudex was in early law taken by a jury.

7 Kunkel argues that in our case too, if the guilt were manifest, or if the accused ‘pleaded guilty’ (cf. the Common law system), there would be no trial (confessus pro iudicato est); Symbolae David (above) 118ff.

8 Festus is explaining the statement sacramento dicitur quod [turis turandt sacratio|ne interpostia actum est, i.e. sacramentum is derived from the sacratio by oath which it involves. Elsewhere, however, (e.g. Gai. 4.13f.) sacramentum is used of the stake in a wager (above, 181). The contrast is usually explained by assuming that Festus is here going back to the primitive meaning, but since he is commenting on words occurring in authors such as Cato, he is more likely to be concerned with a living institution. There must then have been some context in which the oath still survived alongside the usual wager, and Kunkel, Krim. 106ff., suggests that it was that of the capital legis actio. In the ordinary case the amount of the stake varies with the value in issue, but since a man would not be valued he would take an oath of self-damnation. It can be objected, however, that there was a stake in the case of a (possibly) free man in the vindicatio in libertatem or in servitutem (Gai. 4.14), and that it is likely enough that even the ordinary proceeding by sacramentwm included an oath, since the stakes were deposited with the pontiffs and the loser’s used for a sacrifice; Pugliese,

BIDR 66 (1963) 173¢. 9 Above, 305. 1 Below, 355, 401. 812

From the XII Tables to the legislation of Sulla As we shall see, Kunkel also supposes that the private action was in prac-

tice largely superseded in the period from the end of the third century onwards, but he conjectures that it maintained a formal existence until the legislation of Augustus. This would explain why Gaius says that the legis actiones were finally abolished by two leges Iuliae.2 We know that one was de iudiciis privatis and the other was therefore presumably de iudiciis publicis, and yet it is difficult to see what the latter could have had to do with legis actiones. On Kunkel’s view it was so called because, although the legis actio which it abolished was private, the quaestiones perpetuae which took its place were public. _ (0) The antecedents of the Sullan quaestiones. As we have seen, one of the

weaknesses which Kunkel finds in Mommsen’s account is that it does not satisfactorily explain the growth of the quaestiones. He himself argues that they had antecedents at the beginning of the second century and probably even earlier, both in the field of political crimes appropriate to the comitial trial and in that of ordinary crimes justiciable by legis actio. As political questions became more complicated, the comitial trial would become increasingly unsatisfactory, and the more vulnerable the comitia became to demagogic influence the less willing would the senatorial aristocracy be to allow

its members to be exposed to the chances of a popular trial. This would account for the fact that from the beginning of the second century quaestiones extraordinariae * were set up from time to time by senatusconsultum.* Nor were these special courts confined to political crimes. Ordinary offences also were dealt with in this way, but only, it would seem, if they were sensa-

tional and involved numbers of people. The best known of these causes célébres is the ‘ Bacchanalian conspiracy ’ * in 186 B.C. The senate, shocked by the excesses of crime and vice which the new cult of Bacchus had introduced, directed the consuls to make inquiry, and decreed that participation in the cult should be a capital offence. Many persons of both sexes, citizens and others, were put to death. For Mommsen these trials, when authorised only by senatusconsulta, were 2 Above, 218.

3 But it is then surprising that Gaius, 4.30, says only that they replaced the legis actiones by the formulary system; Pugliese, BIDE 66 (1963) 170. 4 There are earlier examples of such quaestiones in affairs outside Rome and involv-

29,36.10ff. (204 B.C.) oO |

ing allies: Livy 9.26.6 (314 B.C.); 10.1.13, 10.1.3 (302 B.C.); 28.10.4f. (206 B.C.); 5 And later (perhaps as early as 187 BC.; Livy 38.54f.; cf. Kunkel, Krim. 58 n 217) by plebiscite; probably exclusively by plebiscite after the lex Sempronia de capite civis, below, 315; in these cases for Mommsen also there was a jury court. 6 Strachan-Davidson 1.232ff.; Kunkel, Krim. 68 n 256. The chief authorities are Livy 39.8-19 and the inscription known as Se. de Bacchanalibus (FIRA 1.240; Bruns 1.164).

Cf. the poison trials.in 184 B.C. (Livy 39.41.5); 180 B.C. (Livy 40.37.4, 43.2f.); 152 B.C. (Livy per. 48).

313

Criminal law in the republic merely an exercise of the magistrate’s ordinary judicial function, but with an exceptional exclusion of provocatio. For Kunkel, however, as we have seen, the magistrate could act judicially only with a consilium whose verdict bound him and whose existence justified the absence of provocatio from the judicial sphere. If he acted on his own or against the verdict of the consium, he was exercising only coercitio; the accused was indemnatus and

provocatio was available.’ A specially constituted court was therefore necessary. For everyday crimes the machinery of these special quaestiones is obviously

inappropriate, but Kunkel finds traces of a regular jurisdiction. In the late republic and thereafter, the law of murder was treated as being based on Sulla’s lex Cornelia de sicariis et veneficis, and yet the lex itself was concerned directly only with the carrying of weapons for a criminal purpose and with possessing and dealing in poisons. Kunkel argues that the substance of the lex is much older than Sulla, and that the reason for the omission from it of murder itself is that this was already provided for by the capital legis actio. For evidence of the existence early in the second century B.C. of an offence of carrying weapons Kunkel points to a passage of Plautus * in which

one character threatens to lay an information (deferre nomen) against another, a cook, for going about with a knife. The information is to be laid with the tresviri capitales.®* These magistrates had no imperium and are normally credited with jurisdiction only over slaves and foreigners, but Kunkel conjectures that in spite of their lack of tmperium they may in fact have tried citizens of the humbler sort. However this may be (and so great a breach with constitutional principle is difficult to accept) it is reasonable to assume that such carrying of arms was at this time criminal, and since we hear of a praetor presiding over a quaestio inter sicarios in 142 B.C.,° it may be that, at least in the case of citizens of higher standing, jurisdiction lay with him.! A crime in relation to poison was also certainly dealt with by a quaestio before the lex Cornelia, since an inscription records the existence (? 98 B.C.) of a tudex (i.e. president) quaestionis veneficis.” 7 Cf. Cic. Verr. 11.5.12, 18; Gell. 13.25.12; Festus, s.v. parricidit quaestores (Bruns 2.21).

8 mn 415ff. 8* Above, 56. 9 Cie. Fin. 2.54.

1 We hear (Ascon. Mil. 32) that L. Cassius Longinus (cos. 127 B.C.) was quaesitor in a number of murder trials. Perhaps, as in the Sullan system (below, 318), a nonmagisterial president could be appointed. The lex Acilia (above, 311 n 1) accords the presidency of the quaestio repetundarwm to the praetor peregrinus. Cicero, on the other hand (Leg. 3.8), gives as the functions of the praetor only privata iudicare iudicarive tubere, but since in his time praetors presided over the quaestiones, this presents difficulties on any account; Kunkel, Krim. 50. 2 Inscriptiones Latinae Selectae, ed. H. Dessau (Berlin, 1882-1916) 45; and there are numerous references in the Khetorica ad Herenniwm (i.e. before the Sullan reforms) to 4udices in what are fairly clearly criminal contexts; Kunkel, Krim. 46.

814

From the XII Tables to the legislation of Sulla It seems reasonable therefore to suppose that criminal quaestiones had developed well before Sulla, and that he merely systematised and extended what already existed. It is likely enough also, if a capital legis actio did exist, that it was in practice superseded by the quaestiones as these developed. For the information would most naturally be laid, in murder trials at least, by those same relatives who would be the plaintiffs in the legis actio, and the new procedure would have the advantage that witnesses could be subpoena’d and that the successful layer of an information would receive a reward.°®

(c) The legislation of C. Gracchus. The pre-Sullan growth of quaestiones

provides a convincing context for two leges put through by C. Gracchus. We ) are told that one provided ne de capite civium Romanorum iniussu popult tudicaretur,* and this has hitherto been regarded as merely re-inforcing the existing right of provocatio (as understood by Mommsen). Kunkel, however, argues that its purpose was to forbid the creation of capital quaestiones by the authority of the senate alone (as had been done most recently in 182 in the case of the supporters of his brother Tiberius Gracchus). This interpretation is borne out by the fact that we hear of no subsequent quaestio set up by senatusconsultum. The other lex was that by which Gracchus gave to the equites the right to sit as jurors.° This had previously been the prerogative of senators, which

meant both that the verdict in political trials was in the hands of a close group, and also that the establishment of a wider system of quaestiones was hardly possible for lack of jurors. The substantial growth of the system of quaestiones therefore comes after Gracchus, and although Sulla restored to

the senators their right to serve on juries, he was able to do so only by adopting a measure which Gracchus himself (and his brother before him) had proposed in vain — the increase in the size of the senate to 600. C. OBJECTIONS TO KUNKEL’S ALTERNATIVE.” The evidence is so fragmentary

that one cannot look for more than probability, but it is perhaps fair to say (i) that Kunkel is most convincing in his treatment of the antecedents of the Sullan quaestiones; (ii) that direct evidence for the exclusion of ordinary crimes from the comitial trial is quite inconclusive, but that his argument from the unsuitability to such crimes of so cumbersome a procedure must have some weight; and (iii) that his greatest difficulty lies in accounting for 3 On the other hand, the quaestiones as we know them in the Sullan system are

perhaps in one respect nearer to the legis actio than they had been earlier. For though, as we have seen, the Sullan quaestiones are accusatory, the accuser is said nomen deferre (and so also in Plautus, above, 314 n 8). Kunkel argues from this that the procedure

was in origin inquisitorial, but that it was inevitable, since there..was no official investigating staff, that the delator would do most of the work; Krim. 92, 121, 134.

4 Cic. Rab. perd. 12 (intiussu vestro); it is often referred to as lex Sempronia de

provocatione. 5 Above, 80. 5* See Addendum, below, 320. 815

Criminal law in the republic all the evidence on the vexed question of provocatio. On this question and on Kunkel’s conjectural legis acttio something more must be said, and: we must also consider the crucial matter of the nature of the verdict of the consilium.

(i) Provocatio. The principal obstacle to Kunkel’s account is the evidence of Cicero.* One may readily concede that when he attributes the origin of

provocatio to the regal period and says that under the XII Tables it was allowed ab omni iudicio poenaque,’ he is exaggerating, but if there was never provocatio from any tudictum at all we have to convict him of fundamental

ignorance. Again, he plainly does see provocatio as connected with the comitial trial, in the sense that an attempt by a magistrate to impose the death penalty would lead to an appeal to the people.* For Kunkel this must be an appeal only against coercitio, and the citizen has no legal right to a hearing by the people—his provocatio ad populum is in effect merely an appeal to a tribune to ‘ intercede ’.? Moreover, since the distinction between coercitio and an act of jurisdiction lies in the presence of a consiliwm, and since the consilium is chosen by the magistrate, it would be easy to exclude provocatio altogether by collecting a compliant consilium. Would a right of | such a Platonic character (as Kunkel describes it)! have been regarded as a palladium of civic liberty ? There remains, of course, the difficulty that the quaestiones of the second

century B.C. and later were evidently not subject to provocatio. Here Mommsen’s explanations may still be right.? In the case of quaestiones set up by senatusconsultum, as in the case of the Bacchanalian conspiracy, the specious argument of a national emergency may well have prevailed.* In the case of those established by plebiscite the exclusion of provocatio may have been express, and if the quaestiones perpetuae before Sulla were created by unrecorded leges (a conjecture which Kunkel rejects,* but which does not seem improbable), the same explanation could apply to them, an explanation which would be consistent with Kunkel’s interpretation of the lex Sempronia de capite civis. 6 Pugliese, BIDF 66 (1963) 159. And it is not easy wholly to discount the cases of pontifical fines (above, 309). For they do contain the three elements of condemnation, provocatio, and a judgment (though not one which simply quashes or confirms). Moreover, Pomponius, D. 1.2.2.23, does assert the consuls’ inability to pronounce a death sentence and does connect with this the creation of the quaestores qui capitalibus rebus pracessent, who were called quaestores parricidit and were mentioned in the XII Tables;

ef. above, 50 n 9. 7 Rep. 2.54. 8 Leg. 3.6, 10, 27.

9 And yet Livy treats provocatio and auxilium as not one but two bastions of liberty; above, 309 n 1. The magistrate who ignored provocatio would also risk impeachment at

the end of his term of office. | t Krim. 131. 2 Sth 2.110ff.; Strachan-Davidson 1.225ff.; ef. Brunt, TR 32 (1964) 447. 8 A similarly fundamental principle of the British constitution was allowed to be set aside in the war-time case of Liversidge v. Anderson [1942] A.C. 206. 4 Krim. 46; cf. 25.

316

From the XII Tables to the legislation of Sulla (ii) The capital legis actio. There are difficulties in the way of Kunkel’s use of both Probus and Festus.*° His explanation of S.N.S.Q., st negat sacramento quaerito, requires it to be taken from a commentary on the conduct of proceedings in the legis actio, whereas all the other abbreviations preserved

by Probus in this context are of formulae used by the parties;* and even if such a commentary existed, would it have used the imperative in referring to

the praetor? Again, Kunkel’s restoration of the fragment from Festus involves the requirement that the convicted defendant be handed over to the

plaintiff by the praetor (ut... traderetur), but since the proceeding is private, the praetor would surely not have custody of the defendant. One would expect an addictio, not a traditio.

(iii) The verdict of the consilium. We have seen that a distinction between coercitio and jurisdiction is fundamental to Kunkel’s view of provocatio, and

that the distinction turns on the existence of a consilium whose verdict is binding on the magistrate in the same way as in Common law systems the verdict of the jury is binding on the judge. Since the Sullan quaestiones observe this principle and there is the early analogy of the centumviral court, this cannot be said to be improbable, but there is little direct evidence.

That the magistrate normally would accept the verdict can hardly be doubted: pressure of custom and the opinion of his class would see to that. The question is whether he was bound in law to accept it. Certainly Verres, when trying peregrini seems to have behaved as if he were in some way bound,’ and the first Cyrene Edict of Augustus reveals that in a similar con-

text the consilium was not a loose group of advisers, but was formally defined and constituted. But the most compelling evidence is that of the quaestio conducted by the consuls in 138 B.C. into the murder of prominent

men in the Sila forest.® Cicero records that the consuls twice ordered an adjournment (ampliatio) de consili sententia. This must be the same procedure as that laid down for the quaestio perpetua repetundarum by the lex Acilia fifteen years later, whereby if a third of the jury cannot arrive at a verdict, an ampliatio must be ordered, and there is then an entire re-hearing. If the consuls could act without the consilium, asks Kunkel, why should they

go to such lengths to obtain a verdict from it? Here, then, the balance of probability seems to be in favour of Kunkel.

5 Pugliese, BIDE 66 (1963) 170ff. 6 Sacconi, SDHI 29 (1963) 310ff. 7 Cie. Verr. 11.1.71ff.; 2.68ff.; 5.114; Kunkel, Krim. 79ff.; SZ 84 (1967) 233ff. 8 Krim. 83. Kunkel explains the distinction in the fourth Edict between the governor’s himself trying the case and his leaving it to the consiliwm as being between his sitting in person with the consiliwm and his leaving it wholly to them; SZ 84 (1967) 231. ® Cic. Brut. 85; Kunkel, Krim. 84; SZ 84 (1967) 224ff. 1 And see below, 340. Kunkel even contends (SZ 83 (1966) 219ff.) that the paterfamilias was bound by the verdict of the consiliwm in his iudiciwm domesticum, but this is more difficult to credit; Guarino, Labeo 13 (1967) 124.

317

Criminal law in the republic

2 THK QUAESTIONES PERPETUAE AFTER SULLA? Although, as we have seen, the main features of the quaestiones perpetuae are recognisable before Sulla, it was he who, in a whole series of leges Corneliae, rounded out the system, so that subsequent legislation, including

that of Augustus, effected only modifications. ,

The essentials of the system were that for each offence or group of offences

there was a court, consisting of a magistrate and a number of jurors, and that the statute which set up the court defined with greater or less particularity the law to be applied by it and the punishment to be inflicted. The presiding magistrate of each quaestio was in principle a praetor, and the raising of the number of praetors to eight by Sulla was intended to provide six such presidents in addition to the urban and peregrine praetors who were

mainly concerned with civil matters. Since there were probably only six quaestiones,® this provision would have sufficed if there had not been, at least for some quaestiones, more business than one court could cope with. Thus, in 66 B.C., when Cluentius was tried, there were evidently three courts inter sicarios sitting simultaneously.* The need for additional presidents was

met by the assignment of additional officers of magisterial status (iudices quaestionum).*°. The president held office of course for a year. The jurors were chosen by lot from a list the composition of which, after C. Gracchus’ initial displacement of the senators, varied with the ebb and flow of political power between the senate and the equites.® The size of the jury, where chance records it in individual cases, varies between thirty-two and seventyfive, and this may reflect variations not only from time to time, according to the size of the list from which they were drawn, but also from court to court.’ Any reputable citizen might bring the accusation * and the conduct of the

case was left to him, though he (but not the accused) could subpoena wit-

2 Cf. Kunkel, Intro. 63ff. 3 Below, 320. 4 Cie. Cluent. 147. 5 This was an annual office, held usually between the aedileship and the praetorship (Kunkel, Krim. 48). The presidency could even in particular cases (perhaps because the usual president was unable or unwilling to act) be assigned to an individual chosen ad hoc, probably from the members of the jury. 6 Above, 80 n 2. For details of the probable methods of selection by lot at different times see Kunkel, PW 24.1.751ff.

7 Cie. Cluent. 74; Pis. 96; Flacc. 4. Kunkel, PW 24.1.755, suggests that after the lex Aurelia (see above, 80) the quaestiones inter sicarios and de vi had 51 jurors and the quaestiones repetundarwm and maiestatis 75.

8 Such a system is obviously open to abuse, and the léx Réemmia (not later than 80 B.C., and perhaps contemporary with the Sullan reforms) provided that a person convicted of knowingly making a false accusation (calwmnia) before a quaestio should be declared infamis (above, 273 n 4) and branded K on his forehead; Mommsen, Sth 491ff.

318

The quaestiones perpetuae after Sulla nesses and seize documentary evidence. To the extent that the procedure was thus accusatory, and not, as in civil law countries today, inquisitorial, there is an obvious parallel with the jury trial of the Common law.°® But there are marked differences. The president is not (or not necessarily) a lawyer, but a man of affairs making his way up the political ladder, and his role, even if he

is not entirely passive,’ is far less important than that of a Common law judge. There was virtually no law of evidence, and there was therefore no place for the continuous control exercised by our judges: the court listened to whatever the parties or their advocates chose to put before it (and they in their choice were hampered by no strict code of conduct such as that which governs the English bar). Again, the president did not sum up to the jury, and there was therefore no room for our distinction between law and fact: each juror made up his mind by himself on the whole case as he had heard it. Such a system offers no scope for the development of exact legal rules and definitions, and it is therefore not surprising that the republican jurists seem

to have taken no interest in criminal law or procedure. It is not until the time of Hadrian and later that we hear of any substantial literature on the subject and even then the attention given to it is superficial compared with that lavished on the civil law.? Our dependence on the later classical jurists makes it difficult to reconstruct the Sullan legislation in any detail.* For the jurists were interested in it only in so far as it had survived the reforms of Augustus, and even then they were not careful to distinguish between the original law and subsequent accretions and alterations.* The best known of the leges Corneliae in criminal matters is that de sicariis et venefictis.5 We have seen ® that quaestiones for these two offences already existed; the effect of the lex was apparently to amalgamate the two courts into one and to add a few ill-assorted crimes to its jurisdiction: The lex testamentaria nummaria’ established a court to deal with falsification of testaments and coins and was eventually at least made to include a variety of other falsifications, so that Justinian * calls it lex de falsis. The lex maiestatis established a court for treason and sedition (crimen imminutae marestatis, i.e. any act calculated to diminish the greatness or ® Strachan-Davidson 2.112ff,

1 As Mommsen thought (Str# 421f.; ef. Strachan-Davidson 2.125); contra, Kunkel, PW 24.1.761, citing Rhet. ad Herenn. 4.35.7; Ascon. in Mil. 32.45; Cie. Verr, 1.29,32. 2 Schulz, History 140, 256.

3 For a survey see Kunkel, PW 24.1.741ff. There is much to be gleaned from the speeches of Cicero, but it is necessarily fragmentary and unsystematic, and it throws more light on the conduct of proceedings than on the exact content of Sulla’s legislation. 4 Even when they appear to quote the ipsissima verba, there are often divergent versions, suggesting that they set no great store on the precise wording. 3 Coll. 1.3 (Ulpian) ; D. 48.8.1, 3; PS 5.23.1; Cic. Cluent. 148; ef. D. 29.5.25. 6 Above, 314.

7 Cie. Verr, 11.1.108. 8 J, 4.18.7 and the rubric of D. 48.10; ef. PS 5.25. 319

Criminal law in the republic authority of the Roman people).’ Definitions of treason are seldom precise, but the Romans do not seem to have attained even a moderate degree of pre-

cision in the matter, and trials for maiestas were decided (as indeed the nature of the court would lead one to expect) mainly on political considerations. The lex repetundarum presumably regulated the existing quaestio, and as

there were, at least after Sulla, quaestiones perpetuae also for ambitus (bribery at elections) and peculatus (embezzlement of public funds), these too were probably the subject of special leges. Since the lex Cornelia de inturus } appears not to have established a standing court, there were thus probably six Sullan quaestiones perpetuae, and though there was a certain amount of legislation in criminal matters before the end of the republic? it did not make any great alteration in their scope or number. The jurisdiction of the comitia was not abolished, but it fell into disuse as

the sphere of the quaestzones extended. Non-citizens were, of course, still subject to unregulated coercitio and to the summary punishments of the III viri capitales.' ® The concept of maiestas was not new; a quaestio to deal with at least some aspects was set up by the lex Appuleia (7103 B.C.), but it is not clear that this was a-quaestio perpetua. On the crime see Kubler, PW 14.1.544ff.; Bleicken, Senatsgericht 27ff. The older crime of perduellio was eventually absorbed in maiestas. 1 Above, 274. 2 Especially the lex Pompeia parricidté (on the murder of a near relation or a patron),

the lez Plautia de vi (on violence against public authorities), and, almost certainly from this period, the lex Fabia de plagiariis (on kidnapping). Mommsen, StrR 203, thought that the last two established quaestiones perpetuae, but see Kunkel, PW

24.1.746£. : 3 Above, 314.

Addendum. When these pages were in proof, the posthumous work of A. H. M. Jones, The Criminal Courts of the Roman Republic and Principate (Oxford, 1972), was published. Chapter 1 is concerned with the comitial trial. Although Jones evidently intended

this as a counter to Kunkel’s thesis (above, 308ff.), he proceeds, as was his custom, simply by way of an evaluation of the evidence which is inconsistent with Kunkel’s (on what are essentially Mommsen’s lines). Space does not permit of a detailed examination, but we

may note that he finds two comitial trials for ordinary crime even in Cicero’s lifetime (pp. 5f.), but to identify them as such he relies on the use (by Orosius and Valerius Maximus) of diem dicere; but see Kunkel, Krim. 47 n 179. In general, the point in Kunkel’s argument which emerges as weakest is his treatment of the evidence of Cicero (above, 316), and of the records of pontifical trials (above, 316 n 6). With Kunkel’s eriticism of Mommsen Jones does not of course deal, except for the argument that the comitial trial would have been too cumbersome for the repression of ordinary crime (above, 310). To this Jones replies (p. 19) (i) that there is no evidence that it was effectively repressed (and he remarks that England tolerated for long a judicial system which failed to suppress violent crime in London; but this failure was not due to a defect in the trial procedure, but to inadequate police); (ii) that the commonest offences (theft and assault) were civil delicts; (iii) that we are concerned only with crimes by citizens. But even if one thinks only of murder trials, Kunkel’s argument still has force.

320

CHAPTER 19

The constitution under the principate 1 THE BEGINNINGS OF THE PRINCIPATE! For a hundred years before Augustus finally became supreme, republican institutions had failed to function with any degree of smoothness, and Rome had become familiarised with arbitrary powers placed in the hands of individuals. The sovereignty of the comitia made it possible, as a rule, to give a legal basis to such powers by securing a vote of the people, but in fact such a vote was little more than a formality which presented no difficulty to the man or men who commanded a sufficient army. Most far-reaching had been the powers conferred on Sulla and on Julius Caesar, both of whom bore the title of ‘ dictator ’, though their office had little more than the name in common with the temporary and constitutional dictatorships of the republic. After the assassination of Caesar (44 B.C.) Mark Antony conciliated republican sentiment by securing an enactment which abolished the dictatorship for ever, and the title was never again borne by any Roman. Nevertheless, as Cicero put it, Caesar’s death had done away with the ‘ king’ only, not the ‘ kingship ’.? The rival leaders could not be restrained within the bounds of the republican constitution, and when, in 43 B.C., the three most powerful, Mark Antony, Octavian (then but twenty years old) and Lepidus, had come to an agreement between themselves, it was easy for them to secure a resolution of the people appointing them ‘ triumvirs for settling the republic’ for

five years. Their power in fact continued for longer than the five years originally laid down, the continuation being probably again legalised by a popular vote. After the end of 83 B.C., when, so it appears, the second term of office allotted to the triumvirs expired, the legal position is obscure. But the law mattered very little; Octavian was master in the West and his victory over Antony and Cleopatra at Actium in 31 B.C. made him master in the East as well. In that year he was indeed consul, but this office alone _ would not account legally for the position he held; actually, though not in name, he was an unconstitutional dictator. Once his authority was unchallenged and peace restored throughout the empire, Octavian desired to put an end to the unconstitutional position and to begin afresh. In 28 B.C. all the previous illegal acts of the triumvirate were cancelled, and at the beginning of 27 B.C., in his own words, Octavian 1 The literature is immense. Bibliography, M. Hammond, The dAugustan Principate (Cambridge, Mass., 1933); PW 22.1998ff. (1955). 2 Ad fam. 12.1.1. 321

The constitution under the principate ‘ transferred the state from his power to the management of the Senate and People of Rome ’.* In a speech to the senate he renounced his extraordinary powers and the ordinary republican constitution ipso facto revived. This purported restoration of the republic has however always been regarded as the beginning of the empire.‘ In fact it would have been impossible for Octavian to retire even if he had wished. The tranquillity of the Roman world rested upon his enormous prestige, and the old republican system had really broken down before he was born. Some system had to be put in its place, and that system could only survive if it left room for him as supreme power in the state. The solution adopted involved as little break as possible with republican tradition. The authority of the comitza, of the senate and of the magistrates was not touched, but there were concentrated in the hands of Octavian sufficient powers to make it possible for him to control the whole administration of the state. These powers individually were such as had been known to the republican system or at any rate involved, each in itself, no very great departure from republican practice. What was new was their concentration in the hands of a single individual and the absence de facto of any time limit.

For there is no doubt that the arrangement was meant to be permanent. One cannot tell what Octavian thought at that time about the succession to the position he had created for himself, but his subsequent dynastic plans

show that he did not contemplate the possibility of a real return to the republic. Outwardly perhaps the most striking sign of the change was the assumption by Octavian, at the request of the senate, of the name ‘ Augustus ’, by which he has ever since been known. The use of honorific cognomina by distinguished persons and even their conferment by the senate was not unknown in republican times, but the title itself was unprecedented. It was a rare word and had a strong religious flavour about it, its meaning being almost equiva-

lent to our ‘holy’; in Greek it is rendered by ocBaords, * worshipful ’, a word which perhaps conveyed more meaning to the bulk of the population than did the Latin term. His real power Augustus apparently intended at first to base mainly upon the consulship, which he continued to hold until 23 B.C. But the consulship as it had existed in the later republic did not give him all that he needed, for since the time of Sulla it had been an office con8 Res Gestae Divi Augusti 34: In consulatu sexto et septimo, postquam bella civilia extinxeram, per consensum universorum potitus rerum omnium, rem publicam ex mea potestate in senatus populique Romani arbitriwm transtult. The Res Gestae, an account of his achievements written by Augustus himself and published after his death, is one of the most important sources of our knowledge of his career. The Latin text and a

Greek version are preserved (imperfectly) in an inscription set up in what is now Ankara (Monwmentum Ancyranwm) and in two other fragments; critical edition by J. Gagé (2nd ed., Paris, 1950); edition of Latin text with translation and commentary by P. A. Brunt and J. M. Moore (Oxford, 1967). 4 Dio Cassius 52.1 says that from this moment the Romans ‘ began again to be ruled

by a monarch ’. 5 Mommsen, Stk 2.771. 822

The beginnings of the principate fined to Rome and Italy and giving no military power, the armies being all under the command of provincial governors. This lack of control of the ultimate source of power was remedied by his being assigned (for ten years in the first instance) those frontier provinces in which the bulk of the army was stationed,® the actual administration of the provinces being in the hands of legates appointed by him. The other provinces were left to the senate and were henceforth governed by proconsuls of the republican type.

In 23 B.C. Augustus made a change in his position so far as the basis of his power in Italy was concerned, by laying down the consulate, to which he thereafter permitted himself to be elected only twice, in 5 B.C. and 2 B.C. One objection to the earlier system may have been the scarcity of consulars (i.e. men who had held the consulship), which necessarily resulted if the princeps filled one of the two available places every year,’ but there was probably a political objection to so flagrant a breach with the republican constitution as an apparently unending succession of consulships. His position in the provinces was now securely based on his imperium as proconsul,® but in Rome itself he needed some other foundation for his authority. This was found in the grant of the ‘ power of tribune ’ for life.® This tribunicia potestas evidently suited Augustus’ purposes admirably. It was free from the inconvenience of colleagueship, for the ordinary tribunes were not regarded as in

any real sense the colleagues of the princeps, who indeed, as a patrician, would have been ineligible for the tribunate proper; it provided the necessary powers of summoning the senate and people, and power of veto enabled the princeps to invalidate the act of any other magistrate. Moreover, the ancient 6 This has commonly been treated as a grant of imperium, and there has been much debate as to whether the imperium was consular or proconsular. But Augustus had imperium as consul and probably therefore all that was granted to him was a provincia consisting of these frontier provinces; Jones, Studies 5f.; de Martino 4.133ff.

7 Even after 23 there would not have been sufficient to fill posts reserved for consulars if the practice of holding the consulate for less than the full period of a year had not been adopted, though this did not become regular practice until 5 B.C. The increasing need for consulars in the imperial service led to increasing interference by Augustus in the elections and thus to interference with the senatorial government generally; F. B. Marsh, The Founding of the Roman Empire (2nd ed., Oxford, 1927) ch. 9; see especially the lists (246) of consular holders of imperial posts in the periods 22-13 B.C. and 12 B.C.-A.D. 3 respectively.

8 It was, however, extended in three ways (Cassius Dio 53.32.5): it was for life; it did not lapse (and therefore did not need subsequent renewal) whenever he returned to the city; and it was superior to that of all other provincial governors. The extent of this imperium maius (if that is what it was) is much debated. ® He had had some aspects of the tribunician power before. Indeed Appian, BC 5.132, and Orosius, 6.18.34, say that he was granted the power itself in 36 B.C. Cassius Dio, 49.15.5-6, however, speaks only of a grant of sacrosanctitas, and since Augustus in his Res Gestae, 4.4, dates the years of his tribuwnicia potestas from 23 B.C., this account is to be preferred. Cassius Dio, 51.19.6, 20.4, also says that Augustus accepted the tribunicta potestas when it was voted to him in 30 B.C., but probably he accepted only the wus auxilit (Last, Rend. Ist. Lomb. 84 (1951) 93ff.).

323

The constitution under the principate sacrosanctitas or inviolability attached to the person of the tribune meant that any indignity in act or word offered to him might be treated as a crime.!

Less tangible, but probably politically equally important, were the sentimental associations of the tribunate in the minds of the common people.’ The proconsulare imperium * and the tribunicia potestas were the two main bases of Augustus’ power. They were hardly by republican standards consti-

tutional, since they were separated from the magistracies from which they derived, but they provided at least a constitutional colour. They were supplemented by a number of minor powers.* Thus Augustus had the right, formerly reserved for the people, of declaring war and making peace,’ as well

as that of concluding treaties. He had special rights with respect to the senate beyond those implied in his tribunician power, including the right to put a proposal before it by letter, and he also exercised considerable control

over the composition of that body, partly through his power of ‘ recommending ’ candidates for the magistracies which carried with them a right to a seat, and partly in other ways.® In these last powers, however, we are passing into the sphere of auctoritas, that element of influence or charisma which

can have no legal basis or definition, but which was a vital element in Augustus’ regime.’

2 THE DEVELOPMENT OF THE PRINCIPATE A. THE POWERS OF THE PRINCEPS. The legal basis of the princeps’ powers

as it had once been established by Augustus remained almost unaltered throughout the duration of the principate. In spite of the preservation of republican forms Augustus was himself supreme, and the most autocratic of 1 The power of an ordinary tribune was confined to the city, but this limitation did not apply to the emperor. Exactly why is not clear; see Mommsen, Stf 2.880. 2 Jones, Studies 11. 3 Jones, Studies 12ff., argues that from 19 B.C. the imperiwm was consular, since this alone can justify many of his actions, such as the command of the troops stationed in Italy and the exercise of jurisdiction. 4 The chief evidence for the minor powers of the princeps comes from the so-called Lez de imperio Vespasiani (Bruns 1.202, FIRA 1.154), a bronze fragment containing

part of a senatusconsultwm passed on the accession of Vespasian in A.D. 69 and apparently ratified by the people. In several places the document says that it shall be lawful for Vespasian to exercise some power as it had been for certain named predecessors, e.g. uti licuit divo Augusto, Ti. Iulio Caesari Augusto, Tiberioque Claudio Caesart Augusto Germanico. These statements need not necessarily refer to specific grants of the powers; the earlier emperors may have assumed some of them under their general auctoritas, whereas Vespasian, the founder of a new dynasty, preferred to have them specifically conferred; Last, CAH 11.407.

57 Below, Mommsen, Stk 2.954. | 6 Below, 327. 343f. For the general power ‘to do all things which he should deem to be for

the benefit of the state’ (Lex de imp. Vesp. 16-20), see below, 365. On the imperial title see Syme, Historia 7 (1958) 172ff. 824

The development of the principate his successors needed for the enforcement of their will but little that he had not already had. On strict republican theory the most significant change was perhaps the assumption by Domitian in A.D. 84 or 85 of a permanent censorship,® the chief advantage of which was that it gave him complete control over the senate. Augustus had three times refused the offer of an absolute cura legum et morum,® which would have included censorial powers, no doubt because he considered that such an office, amounting in effect to a

dictatorship, would make the independence of the senate too obviously unreal, but these considerations did not weigh with Domitian, who detested the senate. After his assassination the title of censor was never assumed, even temporarily, by any emperor, but the necessary functions of the office continued to be carried out by the emperor as such.’ Far more important than any technical change was the gradual penetration of imperial authority, without the assumption of fresh legal powers, into all departments of government and the recognition of the new system as normal and necessary, so that even on the death of an emperor only a few visionaries

could ever regard the restoration of the republic as a real possibility. This process, in its turn, was not without influence on the view which the lawyers took of the imperial attributes. Originally there is no doubt that the princeps was regarded as a citizen, subject like all other citizens to the laws, though

the senate might, and occasionally did, exempt him from the operation of particular rules, such, for instance, as those imposing disabilities on unmarried and childless persons. Domitian and his successors frequently usurped this senatorial privilege of dispensation and it came finally to be recognised as an imperial right. Henceforward if the emperor acted in contra-

vention of any rule from which dispensation was possible, he was held to have given himself the necessary dispensation. It was in this sense, and in this sense alone, that the lawyers of the later classical period spoke of the emperor as legibus solutus,? though the phrase meant much more under the dominate, and played an important part in the development of autocracy on the basis of Roman law in European history.® Still more significant than the dispensing power was the legislative power of the emperor. This, as we shall see, was also a growth of customary law, and was recognised already in the

time of Hadrian.* | 8 Dio Cassius 67.4.

9 Res Gestae 6.1. Suetonius, Aug. 27, and Dio Cassius, 54.10.5, 30.1, appear to contradiet this. The explanation lies probably in Augustus’ statement that he would not accept any office inconsistent with mos maiorwm: he probably accepted temporary grants of censorial powers but refused the cura legum et morwm summa potestate;

Jones, Studies 21ff. 1 Mommsen, StF 2.945. 2D. 1.3.31; ef. Lex de imp. Vesp. 20ff. For discussion see Mommsen, StR 2.750ff.;

de Martino 4.444ff. .

(1947) 62ff. 4 Below, 365. 3 Esmein, Essays in Legal History, ed. Vinogradoff 201ff.; Jolowicz, Tulane L.R. 22 325

The constitution under the principate B. THE ANCIENT ELEMENTS OF THE CONSTITUTION. Of the three elements which

went to make up the republican constitution, the people, the senate and the magistrates, it was the first which suffered the most obvious curtailment of its power on the establishment of the principate. In form indeed no change was made even here by Augustus, and the dogma of popular sovereignty continued to play a part in legal theory even in the late empire,’ but from the first, legislation by the people was merely the ratification of the emperor’s wishes, and, though not uncommon under Augustus, became progressively rarer, until it ceased altogether at the end of the first century.® In the same

period the elections of magistrates (in so far as they were not in effect nominated by the emperor) ’ also passed to the senate,* and the only popular function which survived the first century was that of conferring the tribuniclan power upon a new emperor.’ Augustus, as we have seen, deprived the

people of one of its most ancient prerogatives by accepting the power to make peace and war, and the judicial functions of the assemblies, which had fallen into disuse before the end of the republic, were not revived. Unlike the people, the senate received a considerable accession of dignity

with the inception of the principate. In 29-28 B.C. and again in 18 B.C. Augustus purged it of some undesirable elements which had entered under Caesar and the triumvirate, and fixed its number permanently at 600. Membership, as before, was normally obtained after holding the quaestorship, but there were considerable changes in the qualifications for a senatorial career. In the republic already the senate had become de facto largely hereditary, because members of the old families were generally elected to the magistracies,' and in the principate this hereditary element was emphasised by the rule that no-one except the son of a senator might stand for the vigintivirate, 5 Below, 423.

6 The last known instance is an unimportant lex agraria under Nerva; D. 47.21.3.1. * This might take the form of recommendation of individual candidates (commendatio being probably written and suffragatio oral) or pre-selection of the list in various ways

(loosely called nominatio); cf. Tacitus, dnn. 1.81. By the time of Vespasian (lex de imp. Vesp. 10ff.) such nominations are recognised as binding; they would earlier have rested on auctoritas; Levick, Historia 16 (1967) 207ff. 8 The procedure of destinatio revealed by the Tabula Hebana (above, 23) is evidently a transitional device, which gives a preponderating influence in elections (of consuls and praetors) to a mixed body of senators and equites (the latter in a large majority, however: Jones, Studies 27ff.). Tacitus, Ann. 1.15, says that the elections were transferred to the senate already in A.D. 14, whereas the Tabula Hebana shows destinatio in operation in A.D. 19. On the conflict, and on the political purpose of the introduction of destinatio, see Jones, loc. cit.; R. Syme, Tacitus (Oxford, 1958) 756ff.; Brunt, JRS 51 (1961) 71; de Martino 4.1.535ff. ® But this was only a formality confirming a previous decree of the senate; Mommsen, Sth 2.875f. Formal meetings were also still held, even in the early third century,

to hear the announcement (renuntiatio) of the results of the elections made by the

senate; Dio Cassius 37.28.3, 58.20.4. 1 Above, 17. 326

The development of the principate normally ? a necessary preliminary to the quaestorship, unless he had special dispensation.* But there was a steady influx of new blood as the old families

died out.* It seems that by the end of Hadrian’s reign no member of the patrician families of the republic survived, and the new creations of the early principate died out equally quickly. The gaps were filled increasingly by men from the provinces, and by the early third century such men were probably in a majority. The resulting fall in the numbers resident in Rome emphasised the decline in the importance of the senate as a political body, leaving only its identity as a class. The emperor could exercise control over the composition of the senate not only by his influence on elections to magistracies but also by his censorial or quasi-censorial powers. By virtue of these he could not only remove members, but could grant membership directly, by adlectio, according such rank

(e.g. inter tribunicios, inter praetorios) as he pleased, though adlectio to consular rank did not take place until the third century.’ Augustus also instituted a property qualification, probably of 1,000,000 sesterces and a yearly revision of the list of members, at which the names of the deceased were removed, as well as of those who no longer had the necessary property (unless the emperor chose to make it up by gifts), or who had been guilty of disgraceful conduct.

With these changes in composition there came an entire change in the

position which the senate held in the state, and this change cannot be described merely by pointing to the overwhelming power of the princeps. What happened was that the senate became, at the expense of the comitia, the representative of the republican element in the constitution. In fact the comitia had become an absurdity before the end of the republic, and the principate drew the necessary inference by transferring those powers not exercised by the emperor to the senate, including eventually that of actual legislation. In addition, so far as imperial administration permitted, the senate continued to exercise those general powers of supervision which it had had during the republic, but its rdle was now definitely secondary. Its control of foreign policy it lost entirely, though the emperor might ask or allow it to receive embassies from foreign nations; the provinces which remained under its jurisdiction were those where no great questions of policy were likely to arise, and it lost such control of the army as it had ever had. The old treasury of the republic, the aerarium, remained under senatorial control, though the

emperor had a share in the nomination of the officials in charge.* But the 2 See McAlindon, JRS 47 (1957) 191ff.

3’ Such dispensation was signified (at least from the later first century A.D.; Jones,

Studies 30ff.) by permission to wear the latus clavus, or broad stripe on the tunic, which thus became the badge of the senatorial order. 4 For what follows see Hammond, Antonine Monarchy 249ff. 5 Mommsen, StR 2.942. 6 These varied from time to time; Mommsen, StF 2.557f., 1012f. 3827

J.—12

The constitution under the principate importance of the aerarium was progressively diminished by the emergence of a new imperial treasury, the fiscus, which provided the bulk of the money needed for public purposes throughout the empire. The fiscus received the taxes from the imperial provinces and eventually a considerable amount of other revenue besides, but it is impossible to delimit exactly or for any particular period the spheres of the two treasuries or that of the patrimonium, the fortune of the emperor himself.” When Diocletian abolished the distinc-

tion between senatorial and imperial provinces, the distinction between aerarium and fiscus lost what justification it still retained, and though the aerartum survives into the Dominate, it is little more than the municipal treasury of Rome. The republican hierarchy of magistrates was retained almost intact by the principate, but, like the rest of the republican machinery, it now functioned mainly on the non-political side of government and was, in the most important spheres of administration, superseded by the new imperial civil service. Nevertheless the consulship was still the highest titular office in the state

open to a subject and a great object of ambition; and still, if there were a vacancy in the principate, the responsibility of government rested for the moment with the consuls. Even in normal times, too, they must have had a sufficiency of real business to transact, for beyond the presidency of the senate in its ordinary and its judicial capacities they had also some important judicial functions of their own. Moreover, the highest offices in the imperial

administration, such as army commands and the City Prefecture, were almost exclusively confined to ex-consuls; thus, what the consulship might 7 For a summary attempt see de Martino 4.810, with reff. The course of development

of fiscus and patrimoniwm and even the validity of a distinction between them is debated; see Last, JRS 34 (1944) 51ff.; Jones, Studies 99ff.; Millar, JRS 53 (1963) 29ff.; Brunt, JRS 56 (1966) 75. The reason is in large measure that in non-legal texts (covering in time the first two centuries A.D.) the word fiscus has several senses (Jones, op. cit.) including at the widest the entire finances of the emperor, but being often synonymous with patrimonium (Millar, op. cit.). No doubt the distinction, which seems not to have crystallised until the reign of Claudius (Last, op. cit.), depended ultimately on the needs of the individual emperor and on his sense of the proper. In legal texts

(none earlier than Antoninus Pius) the fiscus is clearly a public treasury. Ulpian, D. 43.8.2.4, when he wishes to justify-the exclusion of fiscal lands from the interdict ne quid in loco publico, goes no further than to argue that they are quasi propriae et prwatae principis, and this at a time when the absolute power of the emperor is openly acknowledged. At the same period a distinction is clearly made between the fiscal and other property of the emperor (D. 49.14.3.10, 6.1). It would be anachronistic to attribute legal personality to the fiscus (P. W. Duff, Personality in Roman Law (Cambridge, 1938) 51ff.) but our distinction between the property of the Crown and the property of the monarch in person is analogous. The origin of the fiscus is, however, reflected in the

fact that it is in principle subject to private law (whereas the aerariwm is subject to public law), though this principle is increasingly eroded as the fiscus absorbs the aerartum and as special officials are created to deal with claims by and against it (below, 337). And see further below, 427.

828

The development of the principate lead on to, and the status it conferred, was in a sense more important than the consulship itself. This is reflected in the fact that each pair now held office

only for a few months, those who were elected at the beginning of the year (consules ordinarit as opposed to their successors who were consules suffectt) being considered most highly honoured and giving their names to the whole year.

Praetors were appointed in numbers varying from ten to eighteen, and were now, as under the late republic, confined almost entirely to judicial or quasi-judicial work. The urban praetor continued to be the chief magistrate for civil suits between citizens so long as the formulary system lasted, and the peregrine praetorship probably lasted until the constitutzo Antoniniana conferred the citizenship on nearly all the inhabitants of the empire.* Praetors continued to preside over the criminal quaestiones perpetuae, until these too

were superseded; one praetor presided over the work of the centumviral court,’ and others had special jurisdictions. The number of quaestors which had been raised by Caesar to forty was

reduced to twenty by Augustus. As before, two bore the title urbani, but, except for a short period, they had no longer the control of the aerartum and their exact functions are uncertain. Two of the remainder were in the direct service of the emperor (quaestores Augusti), who probably was entitled to them as proconsul, and used them mainly for communications to the senate. The remainder were, as under the republic, assistants to the consuls or to the proconsuls of the senatorial provinces, those in the provinces now bearing the title pro praetore.? Among their duties in the provinces was that of exercising a jurisdiction similar to that of the curule aediles at Rome and issuing the corresponding edict. There were no quaestors in the imperial provinces (their financial duties there were carried out by imperial procurators,*® normally of equestrian rank) and Gaius tells us that the aedilician edict was not issued.* What steps, if any, were taken to supply its place is not stated. Even in the

early principate the importance of the aediles diminished considerably. Caesar had raised their number to six by adding two plebeian aediles Ceriales for the purpose of supervising the corn supply. But this duty had to be taken over already in Augustus’ reign by a special imperial prefect,’ and most of the real work in connexion with the police of Rome was done by the prefect of the city.® Trials before the people under the presidency of the aediles had

become very rare before the end of the republic’ and disappeared entirely under the empire, but a minor criminal jurisdiction remained, and in Nero’s

8 Below, 345f.; Mommsen, St& 2.226. 9 Above, 197ff. 1 Below, 395f. From c. 23 B.C. the charge of the aerariwm was given to two praetores

aerartt. Claudius replaced them by quaestors appointed by himself, but Nero created praefectt aerarti, who were also imperial officials and who were chosen from men of

praetorian rank. 2 Mommsen, Sth 2.247. 3 Below, 336. 4 Gai. 1.6. 5 Below, 335. 6 Below, 334. 7 Mommsen, Sth 2.496. 829

The constitution under the principate time it was still worth while to issue a new regulation limiting the amount of the fines they could inflict.* The civil jurisdiction of the curule aediles continued to exist and it would seem from Gaius’ language ° that they still issued their edict in the middle of the second century A.D., but this, like the praetorian edict, had then been standardised and no longer gave to the promulgating magistrate any opportunity of developing the law.’ For tribunes the empire had even less use than it had for the aediles, since all their real power was submerged in the superior tribunicia potestas of the

emperor himself. Nevertheless they continued to be elected and it was a necessary step for a plebelan in his cursus honorum to become either aedile or tribune between holding the quaestorship and the praetorship.? Intercessio was still legally possible but only very occasionally used in quite minor matters of administration.* A minor criminal jurisdiction also survived.+ Augustus abolished the 11 virt vis extra urbem purgandis and the four praefectt Capuam Cumas, whose jurisdiction had become superfluous with the reorganisation of municipal government.> The xxvi virate of the republic thus became a xx virate. The x viri stlitabus tudicandis lost their jurisdiction in matters concerning freedom and became presidents of divisions of the centumviral court,® but the remaining minor magistrates continued to exer-

cise some at any rate of their earlier functions. The tenure of an office included in the vigintivirate was, as we have seen, the normal preliminary to the quaestorship and a senatorial career. At the end of the principate the only republican offices which still survive with any real functions at all are those of consul, praetor and quaestor, and the functions even of these have been constantly diminished by encroachments of the emperors themselves or of their officers. The practical consequence of election which bulked most largely in the minds of the holders was probably the expense which they would be

bound to incur in the provision of public games and other festivities. Tribunes and aediles appear de facto to have ceased to exist, though the titles may still have been conferred and a corresponding rank in the senate granted. The vigintivirate can be traced at the beginning of the third century, but whether it fell into disuse before Diocletian or was only abolished by his or Constantine’s reforms is unknown.

8 Tac. Ann. 13.28; Mommsen, StR 2.496. 9 1.6. 1 Below, 358. 2 Dio Cassius 52.20. 3 Tae. Hist. 4.9. The schoolmaster in Juvenal, 7.228, can rarely get his fee without recourse to a tribune (rara tamen merces quae cognitione tribuni non egeat). No ordinary action for such fees lay, and this perhaps shows that the tribunes had an extraordinaria cognitio in such matters; see below, 396 n. 5*. If the reference is to tribunician intercessio (Mommsen, Stk 2.291 n 1; ef. Kaser, ZPR 357 n 42 and 365) Juvenal must mean that the schoolmaster gets judgment against his pupil and is then deprived of it by the pupil’s invoking the tribune’s help; but this is forced. 4 And was limited, like that of the aediles under Nero, Tac. Ann. 13.28.3.

5 Mommsen, StR 2.610. 6 See above, 199. 330

The development of the principate C. THE IMPERIAL CIVIL SERVICE. One reason for the failure of the republic to

govern satisfactorily the vast territories over which it held sway was the inadequacy of such permanent staff as existed to assist and guide the constantly changing magistrates. This defect was remedied by the principate, which introduced for the first time in Roman history a trained, paid and permanent service, responsible to the princeps himself, which was able to reduce chaos to order, but finally developed all the worst vices of a bureaucracy and was among the contributing causes of the downfall of the empire. It is important to have some idea of this service,’ not only in order to understand how the empire was governed, but also from the legal point of view, because many of the officials came to have powers of jurisdiction, and because with the growth of the emperor’s own powers of jurisdiction it became necessary for him to have a staff of officials and advisers to assist

him in this as in his other tasks.

The civil service derived its character from the princeps’ position as a magistrate. He was no monarch with ministers who were necessarily also ministers of the state; he was a magistrate, and expected, like other magistrates, to carry out his duties himself. But for his enormous sphere of activity he needed more help than other magistrates. In part this help was provided by delegates who were themselves in a quasi-magisterial position, analogous to that held by some magisterial delegates under the republic; such were his prefects, who may be compared with the praefecti Capuam Cumas, who exercised a jurisdiction delegated by the praetor,® and the legates through whom he governed the provinces under his control.® Under the republic, magis-

trates and pro-magistrates had relied a very great deal for clerical and administrative assistance on members of their own personal households. Thus

the bulk of the princeps’ assistants came into existence simply as members of his household. Only gradually did it come to be recognised that, since the princeps’ business was so largely state business, and since his tenure of public

responsibilities was in practice permanent, these assistants were in fact government officials. The free choice of these officials by the princeps was never restricted by law, and the service remained quite separate from the hierarchy of republican offices which, as we have seen, continued to exist; but rules concerning the classes of persons who might be appointed to the different posts were laid down by the emperor or grew up by custom, and were rigidly observed. In this connexion we may note three points particularly. 7 For the ‘ clerical and sub-clerical grades’ see Jones, Studies 153ff. 8 Above, 57. But the emperor’s prefects are always chosen by him himself, whereas the praefectt Capuam Cwmas were elected. 9 The position of the curatores below, 333 n 2, is similar.

331

The constitution under the principate (i) The use of freedmen. In the household of a wealthy Roman, freedmen and even slaves were commonly used in responsible positions as business agents, estate managers, accountants and secretaries, and at first the emperors continued this practice in their own households. The result was that men of this sort sometimes occupied positions of the very greatest importance, and were given the opportunity of amassing enormous fortunes. The climax was reached under Claudius, in whose reign, or at least in the latter part of it, the imperial freedmen came close to being the real governors of the empire. With the gradual recognition that the emperor’s service was the state service this system could not continue, and the more influential posts were reserved for members of the equestrian order by the reforms of several emperors and in particular of Hadrian. (ii) The distinction between senatorial and equestrian posts. For some purposes, even in the imperial service, senators were always employed, so that the holding of these posts might be part of a senatorial career fitting in with

the tenure of the ordinary republican magistracies. But it was felt that the loyalty of a senator might be divided, for he was a member of the corporation which was to some extent a rival of the emperor in the government of the state and hence from the beginning there were some very important posts from which senators were excluded, and which together with the great bulk of responsible posts throughout the service were in the hands of equites. This order therefore played a central part in the government of the empire, and Augustus reorganised and enlarged it, and emphasised its honorific and ceremonial aspects.? As in the republic,® the terminology seems fluid, but a distinction should probably still be made between the broad category of equites Romani, who are those who merely have the property qualification (which is still 400,000 sesterces), and the élite of equites equo publico.* These now number over 5,000 and are organised (but only, it would seem, for ceremonial purposes) into six turmae or squadrons.* Admission to the equestrian order in this narrow sense is now in the hands of the emperor (but 1 H.A. Hadr. 22.8.

2 Especially the angustus clavus and the gold ring (above, 82). Augustus revived the annual parade on 15 July, which served as a kind of festival. In the later principate the gold ring was freely granted and came to mean no more than free birth, either real

or fictitiously granted; Kubler, PW 6.1.286. 3 Above, 79ff.

4 See Henderson, JES 53 (1963) 61ff. Nicolet, L’ordre équestre 177ff., concludes that on the contrary eques Romanus is the really prestigious title, but on the rather paradoxical ground that it is much affected by little men. There is also the category of iudices. This must in fact have existed since the lex Aurelia (above, 80 n 2), which included only that minority of the equites who were on the jury list. The existence of this category under Augustus is shown by the lex Valeria Cornelia (above, 23), which draws its ‘ destining ’ centuries from those equites who are wudices.

5 Dionysius of Halicarnassus, 6.13.4, says that 5,000 might take part in the 15 July parade.

332

The development of the principate sons of senators have the public horse as of right).* Control is exercised by a special bureau a censibus equitum Romanorum, and admission is in effect permission to enter on a public career. (iu) The relation between the civil and military services. In the public life of the republic there had never been any clear separation between civil and military office. The imperium of the highest magistrates included the power of command in the field as elsewhere; the tribunate of a legion counted as one of the minor magistracies, and during a public career a man would in most cases see some military service on the staff of a general, even if he did not ever hold an independent command of his own. This system was continued by the early principate; the young man of senatorial rank frequently

served as a military officer before being elected to a magistracy, and the equites who held posts under the emperor were usually also men who had

served in the army, and might again have military duties to perform. Hadrian however introduced innovations in this as in other respects, by separating the civil from the military career. In his time and later it was possible to enter the emperor’s service without being a soldier, the post first

held being often that of an advocatus fisci.? Septimius Severus did, in a sense, bring about a renewed militarisation of the service by increasing the number of offices open to equites as opposed to senators, and by the possibilities of promotion which he opened to the rank and file of the army. Under him indeed a procuratorship was often the reward of long military service. Nevertheless his officials were by no means all soldiers, as the tenure of the praetorian prefecture by Papinian in his reign shows.° D. CLASSES OF IMPERIAL OFFICIALS.® One important class of imperial officers

has already been mentioned, the legati Caesaris pro praetore, who were governors of the principal imperial provinces. These legates were in the earlier principate uniformly of senatorial rank, but in the second half of the third century they were in many cases replaced by equestrian governors.’ Apart from the legates, the chief imperial officials were the prefects, the procurators, the secretaries and the members of the imperial council.” 6 It is generally held that equestrian rank was not itself hereditary; Mommsen, Sth 3.500; contra, Kubler, PW 6.1.286 and SZ 48 (1928) 651ff. 7 O. Hirschfeld, Die kaiserliche Verwaltungsbeamten (Berlin, 1877) 51.

8 Below, 391. ® Hammond, Antonine Monarchy 444ff. 1 Petersen, JRS 45 (1955) 47ff. We are told (Aur.Vict., Caes. 33.33-4) that

Gallienus excluded senators from all military commands, but some provinces in which legions were stationed were still under senatorial governors after his death. Perhaps the rule was not strictly enforced, or else the governors were confined to civil matters ~— an anticipation of Diocletian’s reform; see below, 425.

2 Of rather less importance were senatorial curatores appointed, generally, by cooperation between the emperor and the senate, e.g. for public works in Italy and the conservancy of the Tiber; Greenidge 413; de Martino 4.1.585.

333

The constitution under the principate (i) The Prefects. (a) Praefectus urbi.* The prefect of the city was originally a mere delegate appointed in case of the temporary absence of the emperor, but the office became a permanency owing to Tiberius’ continued residence

away from Rome in the latter part of his reign, and under subsequent emperors the prefect remained in office even when the emperor was present. His duties included generally the maintenance of order in the city, and he had under his command the four urban cohorts, in effect a police force of 2,000 men. He early assumed criminal jurisdiction,’ and in the end became the chief criminal court for Rome and for the district within 100 miles.* He also took civil cases in matters connected with his criminal jurisdiction. The office was one which made most obvious to the Roman populace that they had now a master and appears to have been unpopular at first.® Its offensiveness was however to some extent mitigated by its ancient name’ and by the fact that senators, almost without exception of consular rank, were always appointed. (b) Praefectus praetorio.® The famous praetorian guard, which was later to exercise such great influence on the choice of the emperors and was responsible for the fall of several of them, came into existence as the development of a republican institution. It had long been the practice for every general to have a body-guard, cohors praetoria, consisting of soldiers with higher pay

- and other privileges, and Augustus, as permanent imperator, turned this body into a standing force of nine cohorts of 500 men each.® At first he retained the command himself, but in 2 B.C. he delegated it to two prefects, and the office thus created came to be the highest in the empire next to the

principate itself. Following the Augustan precedent it was often, but not always, held by two men at the same time, and then the usual principle of collegiality evidently applied. The praefectus praetorio was no mere soldier; he was chief of staff to the emperor, and since the emperor’s powers as proconsul were both civil and military, he became the chief adviser and executive officer for both spheres. The actual influence of the prefect depended very largely, as is always the case with a chief of staff, upon his own and his commander’s personality, but

it might be, and frequently was, very great, simply because he was the natural person to do anything that the emperor did not do himself. In the 3 Sachers, PW 22.2.2502ff.; Schiller, RIDA (1949) 319ff. 4 Above, 329.

5 Beyond 100 miles the praefectus praetorio had jurisdiction; Ulpian, D. 1.12.1.4. Until perhaps M. Aurelius the jurisdiction of the praefectus urbi covered all Italy. 6 Mommsen, St#& 2.1063.

7 It had existed under the monarchy and maintained a shadowy existence during the republic; Pomponius, D. 1.2.2.33. 8 Ensslin, PW 22.2.2391ff.

® There were ten cohorts from Domitian’s time, and Severus raised the number of men in each to 1,000.

334

The development of the principate course of the second century he acquired a mass of judicial work, originally as the emperor’s delegate, and three of the great Severan jurists, Papinian, Paul, and Ulpian,’ held the office. The great position held by the praetorian prefect made him a possible rival to the emperor himself, and the emperors were, not without reason, jealous of their powerful subordinates. It is this fact which accounts for the rule, observed with but few exceptions until the time of Alexander Severus, that the prefect should not be a senator but an eques. (c) Praefectus annonae.? The provisioning of the city had always been one of the cares of the Roman government and difficulties had led at times to the creation of extraordinary magistracies, notably those held by Pompey in 67 B.C., when he was entrusted with the command against the pirates, who intercepted the sea-borne supplies on which Rome depended, and again in 57 B.C., when he was appointed curator annonae. Such was really also the cura annonae undertaken by Augustus at a time of shortage in 22 B.C. At first he appointed curatores of senatorial rank as his deputies in the matter,

but towards the end of his reign a prefect was appointed and the office became a permanency.® The chief duty attached to it was that of seeing to the provision of sufficient corn at reasonable prices for the Roman market and for this purpose the prefect had a number of subordinates in Italy and in the provinces; with the actual distribution of the corn doles to the city populace, he apparently had no connexion. The prefecture was an equestrian office, and like the other imperial posts acquired both a criminal and a civil jurisdiction in matters. which arose out of the functions with which it was concerned.

(d) Praefectus vigilum.* According to the republican system the officials chiefly concerned with the duty of dealing with fires at Rome were the 11 virt capitales, but they were not alone, and Mommsen °* reckons that at any fire there were twenty-four people entitled to give orders, a circumstance which did not make for efficiency. Augustus took the matter in hand in A.D. 6 and created a fire-brigade, consisting of seven cohorts, each numbering 1,000 to 1,200 men, organised under the command of a prefect of equestrian rank. The post became one of great importance and its holder a high police officer

second only to the prefect of the city. He tried criminal cases of minor importance and appears eventually to have acquired a civil jurisdiction in matters arising out of these cases. (ce) Prefects of provinces.® The principal imperial provinces were. governed, 1 Below, 391ff.; and cf. 400. The extent to which the prefect had jurisdiction at first

instance is unclear.

2 Ensslin, PW 22,.2.1262ff. 3 Mommesen, StF 4.1041.

4 Ensslin, PW 22.2.1340ff. 5 St& 2.1055.

6 Jones, Studies 115ff.; A. N. Sherwin-White, Roman Society and Roman Law in the New Testament (Oxford, 1963) 5ff.; Brunt, Latomus 25 (1966) 461ff. 335

The constitution under the principate. as we have seen, by legati of senatorial rank, but for some small provinces, such as Judaea, Augustus appointed equestrian governors with the title of praefectt. In this he was echoing the republican practice whereby provincial governors might delegate some of their functions to members of their staffs who likewise bore this title. These provincial prefects evidently exercised all the normal functions of provincial governors, and they must therefore have had imperium and iurisdictio; the difference in title merely marked their lower social status and the minor importance of their provinces. To the rule that only minor imperial provinces were assigned to prefects

there was one large exception. Egypt had come under the dominion of Augustus after Actium and was henceforward part of the Roman empire,

but, by reason of its great strategic and economic importance and the absence of local government, it was not treated as an ordinary province. The emperor himself succeeded to the position previously held by the kings

and, instead of appointing a senatorial legatus, kept a closer hold on the country by means of a governor of equestrian rank, the praefectus Aegypti. Senators were not even allowed to set foot on Egyptian soil without special permission.’ The prefecture became one of the greatest prizes of an equestrian career, yielding in rank only to that of the praetorian guard. From the reign of Claudius these provincial prefects, with the exception of the prefect of

Egypt, are called procurators * and are thus assimilated in title to other imperial agents, who must now be considered.°

(ii) Procuratores. Procurator is a word taken from private law, and means

an agent entrusted by his principal with the management of property.' A wealthy Roman might have several such agents, one e.g. for each of a number of estates, and they would nearly always be freedmen, i.e. persons who though dependants were free, and thus had the capacities, such as that for

representation in a lawsuit, which were necessary for their duties. The imperial procurator was originally just such an agent appointed to manage some part of the emperor’s property, to collect moneys due to him and for

similar purposes, but the vast extension of the wealth belonging to the emperor and its use in reality as state property, quite apart from the fact that these men also handled large sums of public money on the emperor’s behalf, made these procuratorships into offices of such importance that already in the early principate the higher ranks were filled by men of the equestrian class. In time there came to be a whole hierarchy of such officials 7 Tae. Hist. 1.11; Ann, 2.59.

8 Tacitus, Ann, 15.44.4, using the terminology of his own time, calls Pontius Pilate procurator, but in a contemporary inscription he is praefectus (Année Epigraphique (1963) 104). On the powers of the praefectus Aegypti see Last, JEA 40 (1954) 68ff. 9 No reference is here made to the purely military prefectures or to a number of less important civil prefectures; see Ensslin, PW 22.2.1257ff.

1 It has also the special meaning of ‘ representative in litigation’; Gai. 4:82ff. 836

The development of the principate graded according to the pay they received,? and promoted from one post to another on a fairly uniform principle. Some were employed at Rome where

the central office was presided over by a procurator a rationibus,? who became in effect the Finance Minister of the emperor. More, naturally, were to be found in the provinces, where they collected taxes and looked after the emperor’s property. The wasteful system of tax-farming which had been in

use under the republic was not given up at once, but gradually replaced either by direct collection or by collection which, if not strictly direct, was closely supervised by the procurators.* Even in those cases where taxes, such

e.g. as the portoria (customs duties), were still farmed to publicant, the machinery for checking fraud and extortion was much improved. Procurators were sent not only to imperial provinces, but also to those under senatorial government. In the former the chief procurator held a position analogous to

that of a quaestor in a senatorial province, but in the latter, where the proconsul had his own quaestor and legates, the imperial procurator stood apart from the main work of government and only represented the interests of the fiscus. There were also, in both types of province, procurators with special functions, in particular for the administration of the great estates which were

acquired in increasing numbers by the emperors and formed part of the patrimonium.® Later, under Septimus Severus,® perhaps because the patri-

monium had acquired the character of crown property rather than private property of the emperor, there was created a further department, the res or ratio privata, made up chiefly of confiscated lands. The res privata eventu-

ally absorbs the patrimonium.’ |

Following the usual Roman principle that administration brings with it jurisdiction, Claudius gave jurisdiction to some of these procurators in matters connected with imperial finances.* The policy of subsequent emperors

varied. Thus, Nerva entrusted such matters to a special praetor,® but this reform evidently did not last. The procedure used was of course that known as cognitio extra ordinem,' for the procurators were not magistrates, and thus had no power to authorise a trial by a zudex according to the formulary system. And an appeal lay, directly or indirectly, to the emperor.* 2 There were four classes, trecenarii (HS 300,000), of whom there were very few, ducenarit (HS 200,000), centenaritt (HS 100,000) and sexagenarit (HS 60,000). 3 Later rationalis. 4 Last, CAH 11.428. 5 Above, 328.

6 Or perhaps under Antonius Pius; Nesselhauf, Historia Augusta Colloqmum 1963, ed. J. Straub (Bonn, 1964) 73ff. 7 Below, 427; see now A. Masi, Ricerche sulla res privata (Milan, 1971).

8 Tac. Ann. 12.60; Stockton, Historia 10 (1961) 116ff.; Brunt, Latomus 25 (1966) 461 ff.

1 Below, 397. 2* Cf. below, 400.

9 D. 1.2.2.32.

337

The constitution under the principate (iii) The Secretaries.* As the emperor’s chief accountant, a rationibus, had developed into a Finance Minister with a great body of officials under him, so the imperial secretaries also developed into Secretaries of State at the head of departments, and, from the time of Hadrian onwards, were, almost without exception, members of the equestrian order. But it must be remembered

that the Roman emperors, for the most part, ruled as well as reigned, and transacted a surprising amount of business of all sorts themselves, so that the secretaries were, except for the size of their staffs, more like those of a Tudor monarch than like the Secretaries of State of to-day. The two chief departments were those ab epistulis and a libellis.* Of these the former dealt with the great mass of imperial correspondence, and it was

also the department chiefly concerned with the appointment of imperial officials. In the second century the work was so heavy that there was one secretary ab epistulis Latinis and another ab epistulis Graecis, the former being the more important.° The office a libellis was concerned with petitions, of which great numbers

were naturally addressed to the emperor. These petitions included those requests for advice on points of law out of which there grew in the second century the regular practice of settling cases by imperial rescript.® It was consequently necessary that the head of the department (and, no doubt, some of his subordinates), should have legal knowledge, and the post was in fact held by several distinguished lawyers, including Papinian and Ulpian,’ both of whom subsequently became praetorian prefects.*®

Another officer who was concerned with the judicial work of the emperor was the a cognitionibus. It is not clear how his competence differed from that of the a ltbellis,® and in the fourth century only the latter survived. A memoria was the title of an official, at first of subordinate rank, whose original duty was perhaps to reduce to writing speeches and oral decisions made by the emperor, and to prepare the necessary material for them. The post however grew in importance and its holder’s duties came to include the preparation of patents for the appointment of imperial officers as well as the 3 Millar, JRS 57 (1967) Off.

4 The evidence for the functions of .these officials is scanty; Millar, JRS 57 (1967) 14ff,

5 See Townend, Historia 10 (1961) 375ff., who dates the effective division of the office to c. 166. 6 Below, 369f. 7 Below, 391, 393.

8 Honoré, SDHI 29 (1962) 162ff., tries, by stylistic analysis of the writings of individual jurists and of imperial constitutions of the time, to identify in greater detail the jurists who held the post. ® Perhaps he dealt with personal as opposed to written petitions (Bleicken, Senatsgericht 110), or perhaps he was concerned with organisation of the court, preparation of the cause list, ete. (Premerstein, PW 4.221). 838

The development of the principate issue of permits for the use of the imperial post, a privilege which was rigidly

restricted. In the dominate it became equal if not superior to the offices ab epistulis and a libellis.

(iv) The imperial council. The ancient Roman custom of consulting a consiium before making any important decision was followed by the emperors, who might, of course, summon for this purpose any persons they chose, in much the same way as did the Angevin kings of England.’ This much is

clear, but Mommsen? maintained that, apart from this informal body of amici, there was a more stable legal council, deriving from republican pre-

cedents, which became a permanent and salaried body under Hadrian. Crook,* however, has argued that this rigid distinction is unmaintainable, and that though the emperor might choose to summon different persons for judicial purposes from those whom he consulted on political matters, there is no clear distinction and no institutional form, even under Hadrian.* To some extent this is a matter of emphasis, for there can in the nature of things be no sharp division between a fluctuating group and a formal body, when the composition of both depends simply on the will of one man, but some degree of permanency and some degree of distinction of function seems to be implied in the statement in the Historia Augusta *® that when Hadrian ¢ sat in judgment he had in his consilium not only his friends or companions, but

jurists and especially Iuventius Celsus, Salvius Iulianus, Neratius Priscus and others, all of whom the whole senate had approved ’. The senate’s approval (whatever precisely it amounted to) seems to imply that the composition of the consilium for judicial purposes was not wholly a matter of the whim

of the emperor at the moment.* Moreover, there is obviously a difference between a judicial sitting (which even in the second century was normally in public, and which entails a hearing) and a political discussion. Kunkel therefore argues’ that whereas the amici whom the emperor consults on political

1 Stockton, Jwra 7 (1956) 210. 2 StR 2.902ff., 988ff.

3 Consiliwm Principis; cf. Hammond, Antonine Monarchy 370ff. 4 Augustus did indeed create a consilium semestre (Dio 53.21.4; cf. Sc. Calvisianum in the fifth edict of Cyrene, FJRA 1.410) consisting of the consuls, one praetor, one aedile, one quaestor, possibly one tribune, and fifteen senators chosen by lot every six

months. This body not only prepared legislation, but also occasionally, we are told, acted as a court (possibly for cases which were not important enough for the whole senate; Kunkel, SZ 85 (1968) 271ff.). In A.D. 13 it underwent a change the nature of which is disputed, and in A.D. 26, when Tiberius finally left Rome it ceased to function

and probably never revived; contra, Kunkel, SZ 85 (1968) 270, who thinks it was revived by Claudius (Dio 60.4.3) and continued to exist as the judicial constliwm. 5 Vita Had. 18.1: cwm iudicaret in consilio habuit non amicos suos aut comites solum sed tiuris consultos et praecipue Iwentium Celsum Salvium Iulianum Neratiwm Priscum aliosque, quos tamen senatus omnis probasset. See also D. 37.14.17 pr.; 28.4.3. 6 Save that when the emperor was travelling in the provinces his constliwm would necessarily consist of amict et comites. 7 SZ 85 (1968) 253ff.; ef. PW 24.7834. 839

The constitution under the principate matters are indeed a fluctuating group of, as it were, courtiers, the consilium in its judicial form was, as Mommsen saw, a descendant of the judicial consilia of the republic, consisting of leading men in public life,* and that just as in the republic.a trial without a consilium was (in Kunkel’s view) * no trial, so also the emperor felt himself bound to consult a properly constituted consilium..Certainly Philo,* who led the Jewish embassy from Alexandria to Caligula, expresses his disgust at having been received by the emperor alone, and takes it as self-evident that the three requirements for the constitution of a proper court are that there should be a consiliwm chosen on the basis of distinction,? that both sides should be heard, and that the consiliwm should be consulted when the decision is reached. Kunkel goes further and maintains that just as (in his view) a republican magistrate in a criminal trial was

bound by the verdict of his consilium, so also was the emperor when he judged a criminal case.* Here the evidence falls far short of proof. It is true that there is evidence of the taking of votes,* but more than this must be an extrapolation from whatever conclusion is reached for the republican

practice. :

If we turn to the part played in the development of the consilzum by

Hadrian,* we must beware of attributing to him all the developments which had occurred by the end of the second century. For we know that by then the praefecti and the heads of departments were regular members, and that some of those who sat were paid, being divided into classes according to the

amount of their salaries. (These last were perhaps the ‘ civil service ’ element, preparing the work of the council; they cannot have sat on an equality with the great figures whose names we know.) Certainly Hadrian sat often in a judicial capacity, and equites were included in his consilium,*® but we do not know that this was an innovation. The significance of Hadrian’s reign is probably that he gave new impetus to the process of attracting able

men into the imperial service as equites rather than via the old republican cursus honorum, and that he was thus the source of the new professionalism which marks the government of the later principate in general and the legal system and the consilium principis in particular. 8 Cic. Quinct. 5; Val. Max. 8.2.3. 9 Above, 311f., 317. 1 Leg. ad Gas. 349f. 2 apiozivdny.

3 Dio, who was himself a member of the consilium of Septimus Severus, makes Maecenas advise Augustus (52.33.3) only to give judgment with the most distinguished senators and equites. This is anachronistic for the time of Augustus, but Kunkel, SZ 85 (1968) 286ff., thinks it reflects the practice of Dio’s own time, the equttes being the praefectus praetorio and other heads of departments; Crook, Consiliwm Principis 88,

regards it as pure invention. 4 Below, 402 n 7. 5 Kunkel, Herkunft 296ff.; Crook, Consiliwm Principis 56ff., 135ff. 6 H.A., Vita Had. 8.9.

340

The development of the principate E. SUCCESSION TO THE PRINCIPATE.’ NATURE OF THE CONSTITUTION.® The

principate began by being in part, at any rate, an extraordinary magistracy, and it never entirely lost its magisterial character. There was consequently at no time any acknowledged legal system of succession even when it had become obvious that the institution was a permanency. What we have to ask therefore is this: ‘ How did people come to occupy such a position that the usual powers of the princeps were conferred upon them ?’ and this is, in the main, not a legal question at all. We may say that there were four chief factors at work, heredity, choice by a predecessor, choice by the army, and choice by the senate. All of these four were closely connected and their relative importance varied greatly from one case of succession to another. Especially closely connected were the first two, for the choice of an emperor

generally fell upon one of his sons if he had any, and the expression of his

choice, if he had no sons, was normally by adoption. But this is not the whole story. Choice was frequently made clear by granting to the person chosen a share in the imperial power, and here we can distinguish two separate periods. In the earlier the position of the co-regent is, even in form, definitely subordinate to that of the princeps himself, in the later there are two emperors, and, though the senior may take by far the greater share in the government, they are legally on a footing of equality. The change came when Marcus Aurelius, immediately on his accession, insisted that the senate

should confer on his adoptive brother, L. Verus, exactly the same honours and titles as he himself received, and thereafter there were a number of cases in which two Augusti ruled over the empire. When this happened legislative enactments, for instance, bore the names of both emperors,’ whereas previously the single emperor’s ‘ minor colleague’, as he may be called, had taken no part in legislation. The existence of two Augusti of course simplified greatly the question of succession; on the death of one, the other continued

to hold office and might or might not himself cause a co-regent to be appointed. An interaction of choice and heredity can be seen also in the interventions

of the army. For the troops, especially the praetorian guard, played a considerable part in the whole troubled history of the succession, and on more than one occasion, even before the long horrors of the third century, armies from different parts of the empire, each wishing to raise its leader to the throne, had fought the question out by force of arms. Where armed forces are willing to fight for such a cause they can naturally impose their will upon 7 Wickert, PW 22.2137ff. 8 De Martino 4.1.234ff.; Kunkel, SZ 75 (1958) 302ff.

9 Thus a number of enactments bear the names of Severus and Antoninus (i.e.

Caracalla). 341

The constitution under the principate an unarmed and unorganised population. But this element of force was itself connected with the element of heredity which, for the soldier as for the mass of the population, constituted a title to legitimacy.! The loyalty which made the soldier willing to fight for his general when the succession was in doubt would extend to quite unsoldierly members of that general’s family once the general had become emperor, and some of the worst of Rome’s emperors were really imposed on her in this way. But even though the army could, and often did, force its nominee upon an

unwilling senate, there were occasions, as in the case of Nerva, and later (A.D. 275) of Tacitus, when the senate was able to exercise a real choice. And here the principle was different. The senate was the repository of republican legitimacy, and the princeps, as the first citizen, should be chosen on his merits, and even if he were chosen not by them but by the ruling emperor, should be taken from among their number.?

Given these facts as to the succession and bearing in mind the imperial powers already discussed, how are we to construe, from a legal point of view, the nature of the constitution P Mommsen’s * view was that, from the point of view of law, the principate

was not conferred at all, but assumed by the individual emperor, either at the invitation of the senate, or, more frequently, at that of the army. The invitation by the army was, in his opinion, identical.in law with the acclamation of a general as imperator after a victory, a well-known republican

custom, which gave to the general concerned the right to use that title officially thereafter. Mommsen even goes so far as to say ‘ any armed man had the right to make anyone else, if not himself, emperor ’.* Of course the number and strength of the ‘ armed men’ might make all the difference to the possibility of their carrying their nominee to real power, but it made no difference to the legal quality of their act, and, if there were several nominees, each was legitimate until overthrown by another. On this view it is not diffi-

cult to see why Mommsen went on to describe the principate as ‘ autocracy tempered by legally permanent revolution ’,> and to say that there ‘ never was a system of government which had lost so completely the conception of legitimacy as the Augustan principate ’.® This startling conclusion drawn by the greatest master of Roman constitutional history has not remained unchallenged, and many attempts have been made to fit the principate into one clear constitutional category or another — 1 Jones 1.1ff.

2 B. Parsi, Désignation et investiture de Vempereur romain (Paris, 1963) 78ff., lays emphasis on the lez de imperio (ef. below, 365), which was looked upon as the constitutional basis of the emperor’s power (Gai. 1.5; J. 1.2.6). 3 Sth 2.842. 4 Ibid. 2.844.

5 Ibid. 2.1133. 6 Ibid. 2.844. 342

The development of the principate that it was a continuation of the republic, since the sovereignty remained with the senate as representing the people of Rome; that the republican forms concealed the true legal position that it was a monarchy; or, following

another famous description of Mommsen’s,’ that it was a ‘dyarchy’, in which the sovereignty was shared between the princeps and the senate. No one of these explanations can really tell the whole truth of the matter,®

for none of them can satisfactorily draw the distinction between law and fact. Like the British constitution, that of Rome was unwritten, and changed

imperceptibly as the de facto methods of exercising governmental power hardened into conventions which, whether called law or not, became insepar-

able from the legal elements. It was thus that the senate had become the governing body in the republic, and it was thus that the public law of the empire was formed. The powers attributed to the emperor were defined by reference to offices well known to the republic; a lawyer might have said that their concentration in the hands of a single man was accidental, but in fact it was vital, and became a convention of the constitution from which further deductions were constantly being made. The property of this multiple magis-

trate becomes, as we have seen, state property and his servants obtain judicial powers, both developments being a logical inference not from his magisterial position, but from his personal supremacy. Even a right to legislate comes to be recognised, but so gradually that no one can point to a definite date. In the formation of this conventional constitution, the most varied factors play a part, traditional reverence for republican institutions, the brute force of the army, dynastic sentiment, oriental ruler-worship, and many others. The final result was the undisguised autocracy of Diocletian, but the process by which it was reached was a continuous one, and the principate was never capable of being fitted into the ready-made categories, of ‘monarchy ’, ‘ aristocracy ’ or even ‘ dyarchy ’. Nor indeed is it sufficient to think of the princeps merely as a magistrate, even in the earliest period.°® The principate as Augustus shaped it defies definition because of an essential

duality. The republic survives, but there appears beside it a monarchical element which gives it impetus and direction and at need controls it. Because it is for the most part clothed in republican forms, this monarchical element evades precise identification, but that it exists and that the supremacy of the princeps goes beyond his republican powers is most clearly shown by the title ‘ Augustus ’, which connotes no magisterial powers at all, and is yet the

highest that the princeps bears. The same extra-constitutional aspect is expressed, though with more republican echoes,” in the term auctoritas. In

7 Ibid. 2.748. 8 See especially Schonbauer, SZ 47 (1927) 2644f. 9 Cf. R. Syme, The Roman Revolution (Oxford, 1939) 313ff. 1 Arangio-Ruiz, Storia 215ff.; Kunkel, Intro. 47f., Gymnasium 68 (1961) 353ff.

2 In the’ language of the republic auctoritas belonged especially to the senate, and also to individual senior senators or principes viri.

343

The constitution under the principate the Res Gestae* Augustus says that although he had no more legal power (potestas) than his colleagues in the magistracies, he excelled them all in auctoritas. This is no doubt how Augustus wished matters to appear, but it ignores the peculiarity of his position—that this intangible authority was backed by a greater combination of lasting powers (and a greater control of financial resources) than lay even remotely within the reach of any of his transient colleagues. It was in this combination of legal and extra-legal forces

that the strength of the principate lay. |

3 34.3. See P. A. Brunt and J. M. Moore, Res Gestae Divi Augusti (Oxford, 1967) 78ff., 832.

844

CHAPTER 20

Classes of the population and the government of the cities under the principate 1. THE EXTENSION OF THE CITIZENSHIP! The extension of citizenship to the whole of Italy was complete before the end of the republic, and with the principate began the policy of extending it also gradually to the provinces, both by the foundation of colonies and, more frequently, by direct grant to already existing communities.” The grant, whether to a community or to individuals, was now always made by the emperor, who also frequently conferred Latin rights on Western communities which were not yet considered fit for full citizenship. In the East, where city life was already developed, this was not done. Nor was Latinity ever given to individuals, except in the anomalous case of ‘ Junian ’ Latins, i.e. defectively manumitted slaves, who were given a modified form of the status by

the lex Iunia (?17 B.C.).° Subsequent legislation used the citizenship itself as an inducement to these and other Latins to engage in a wide variety of activities in the public interest‘ (e.g. service in the fire brigade at Rome, putting up buildings in the city, shipping grain, bearing children). Citizenship was also the normal reward of those who served their time in the auxiliary forces and received an honourable discharge; and it was frequently given to individuals as an imperial favour. In A.D. ?212° a great extension of the 1 Sherwin-White 167ff.

2 The extension of the citizenship to Italy had always been taken to imply that the new citizens owned their land according to Roman law, and, of course, that any other Roman could do so similarly. This consequence was not deemed to follow when citizenship was granted to communities outside Italy. Here (above, 267) the principle prevailed that the soil was the property of the Roman people or of the emperor and as such liable to, taxation. Freedom from taxes was granted by conferring the ius Italicwm, which assimilated the soil to that of Italy and thus also made it a res mancipi. For a list of communities known to have had tus Italicwm see Mommsen, St# 3.807f., and cf. Sherwin-

White 216f, For a possible personal grant of tus Italicwm see the inscription discussed by Triantaphyllopoulos, Iwra 14 (1963) 109ff. 3 Gai. 1.16, 22, 23; 3.56. See Buckland 78, 93f.; and cf. above, 61f. 4 Gai. 1.28ff. Many Latins must also have become citizens by holding office or becom-

ing decuriones in their own towns. Gaius, 1.96, tells us that in his day there were two sorts of Latin right in this respect, the maius Latiwm, whereby all decuriones became Roman citizens, and the minus Latiwm, whereby only magistrates had this privilege. 5 The date has always been taken to be 212, but Millar, JEA 48 (1962) 124ff., has shown that the assumptions on which this is based (that Dio Cassius is adopting an annalistic structure and that the decrees in P. Giess. 40 (below) are in chronological 845

Population and government of cities under the principate citizenship was made by the Emperor Caracalla. The exact scope of this famous constitutio Antoniniana is a matter of doubt.® Ulpian, Dio Cassius and others’ speak as if it had turned all the inhabitants of the empire into citizens, but since it is clear that even after the constitution there were some who did not possess the citizenship, it is normally assumed that there was some limitation. It may well be, however, that the grant was indeed universal and that the non-citizens after the constitution were those slaves manumitted after that date who failed to obtain citizenship because of either the lex Iunia or the lew Aelia Sentia,® or else were immigrant barbarians.® A papyrus which is believed to have contained a copy of the decree ' is so fragmentary that its discovery has increased rather than diminished the uncer-

tainty.” In the crucial part all that is clear is that there is a grant of citizenship, subject to a reservation (the details of which are lost), and that dediticu are excepted.* There are three questions. Who are the dediticii? Are they excepted from the grant of citizenship or from the reservation? What is the reservation ? Gaius * defines dediticii as those who had taken up arms against Rome and

had been forced to surrender. Unless Rome gave them citizenship or set them up as an allied community they remained free but without any further status. In this sense the term covers all those who were not organised in cities, 1.e. the bulk of the agricultural population. If these were the deditici referred to, then it cannot have been from the grant of citizenship that they order) are inadequate. The papyrus appears to connect the constitution with a thanksgiving by the emperor Caracalla; Millar conjectures that this was for his escape from shipwreck in the Hellespont in 214. There is some support for this in papyri which show the taking of the name Aurelius at this time. (The new. citizens had to take a Roman gentile name, and they took that of the emperor, whose official style was M. Aurelius Antoninus, Caracalla being merely a nickname.) Contra, Gilliam, Historia 14 (1965) 90ff. Seston, Mélanges J. Carcopino (Paris, 1966) 877ff., prefers to connect it with the emperor’s deliverance from an illness in 213 (H.4A. Car. 5.2, 3).

6 The literature is immense. See C. Sasse, Die Constitutio Antoniniana (Wiesbaden, 1958), with bibliography. For later writings (of which see especially Jones, Studies 127ff.) see id. JIP 14 (1962) 109ff.; 15 (1965) 3204f. 7 Ulpian in D. 1.5.17; Dio Cassius 77.9; H.A. Sept. Sev. 1; St. Augustine, de civ. dei 5.17; Nov. 78.5. 8 Diplomas granting citizenship to ex-soldiers were still given; Mommsen, Ges. Schr. 5.402ff., 419ff.; Meyer, SZ 46 (1926) 264ff. For grants of conubium to peregrine women see Volterra, Jura 3 (1952) 217.

® Below. Junian Latinity and dediticia libertas were definitely abolished only by

Justinian; J. 1.5.3; C. 7.5 and 6. 1 Jones, Studies 133f.

2 P. Gtess. 40.1; Mitteis, Chrest. No. 377; FIRA 1.445. Sherwin-White, 226f., doubts if the papyrus contains the actual text of the constitution.

8 The text runs: didwpe rou[v] vv ara[c. 28 letters]y olxoupévyy m[Lodir]eiav ‘Pwyaiwv [p]ev-

ovros [e. 28 letters])arwv yup[is] rav [Se]deurixiwv. It is generally agreed that the first lacuna contained a grant of citizenship to all the peregrine inhabitants of the Roman world; the controversy is as to how the second lacuna (the reservation) should be filled

and as to its relationship to ywpis. .. . 4 1.14; and cf. above, 64f. 346

Local government were excepted, since the native Egyptian peasants, for example, evidently did become citizens.° In a narrower sense, however, dediticit denotes either barbarians from outside the empire who surrender themselves or manumitted

slaves who were debarred from the citizenship by the lex Aelia Sentia because they had been submitted to degrading punishments.® If these were the dediticit of the papyrus, the exception could be from the grant of citizenship,

and the failure of our sources to mention it would not be surprising in view of the small numbers involved. On the other hand, the exception is grammatically more likely to be related to the reservation, and dediticii would then most naturally be understood in the wider sense. Many conjectures have

been made as to the content of the reservation, the most attractive being those which suppose it to have concerned the local citizenship of those enfranchised: only the dediticti were excluded from local citizenship.’

From the political point of view the constitution was not as important as might appear. Citizenship no longer implied either real political power or immunity from taxation. Indeed Dio is probably right when he says that Caracalla’s true motive was to make more persons liable to those taxes which were paid only by citizens.* Even in the matter of appeal to the emperor the distinction was now not between citizen and non-citizen but between hones-

tior and humilior.’ From the point of view of private law, however, it was more important, though rather for its indirect effect on the character of the law itself than for its direct impact on the new citizens. For though, on the generally accepted view,’ large numbers of provincials who previously had lived by their own law, were now expected to adapt themselves to Roman law, the immediate result, as we shall see, was often no more than the adoption of the forms of Roman law as clothing for the unchanged substance of local law. The more important consequence was the eventual reception into the imperial Roman law of a number of institutions of Greek origin and, more broadly but more disputably, a spreading influence of Greek ideas.’

2 LOCAL GOVERNMENT? When all the inhabitants of Italy had become Roman citizens it became a logical necessity that the local constitutions, previously those of independent

city-states living under their own law and merely in alliance with Rome,

1926) 32. 6 Gai. 4.13. 5 E. Bickermann, Das Edikt des Kaisers Caracalla in P. Giess. 40 (diss., Berlin, 7 See especially the tabula Banasitana (Seston and Euzennat, C. R. Acad. Inscr. et

Belles Lettres (1961) 317ff., full publication expected ibid., 1971; cf. Schénbauer, Zura 14

(1963) 71ff.). This records a grant of citizenship in Mauretania in the later second

century A.D. salvo ire gentis. 8 Dio Cassius 77.9. 9 Below, 351. 1 Cf. below, 470. 2 Below, 414ff., 469ff., 509.

3 Abbott and Johnson, Municipal Administration; G. H. Stevenson, Roman Provincial Administration (Oxford, 1949) ; de Martino, vol. 4.2.

347

Population and government of cities under ‘the principate should be reorganised as instruments for the local government of communities which now merely formed part of a larger whole. This reorganisation was

indeed gradually completed, but the process by which it was achieved is obscure. Something was done by individual leges datae,* and it may be that Julius Caesar either passed or intended to pass a statute laying down a model

constitution for the municipalities. At any rate in the early empire a considerable degree of uniformity had been attained and the form of constitution adopted was also used for those provincial communities which received the citizenship. There were indeed always a number of local variations, due either to local conditions or to the ancient traditions of the community concerned — sometimes as old as those of Rome herself — but the general scheme was the same. The difference between colonies and municipia, formerly of importance, came by the second century to be one of name only,® and even the ‘ Latin ’ cities of the West were organised on lines sufficiently similar to those of the citizen communities for all to be treated together in this very summary treatment of an obscure subject.” The constitutions were, as might be expected, models on a small scale of the constitution of Rome, with a popular assembly, a council of decuriones,

corresponding to the senate at Rome, and annual magistracies. The magisstrates were usually four in number, 1 viri ture dicundo, who corresponded

next note.

to the Roman consuls, and 11 virt aediles, generally ranking below the former, 4 Cf. above, 69. An example is the lex municipt Tarentint; Bruns 1.120; FIBA 1.166; Abbott and Johnson, Municipal Administration no. 20. But see Frederiksen, cited in 5 The question turns on the character of the Tabula Heracleensis (Bruns 1.102; FIRA 1.140; Abbott and Johnson, Municipal Administration no. 24; E. G. Hardy, Siz Roman Laws (Oxford, 1911) 136ff.). Savigny, Verm. Schr. 3.279ff., held this to be a law of Caesar’s passed in 45 B.C. (cf. Cie. ad fam. 6.18.1), and identified it with a lex Iulia

municipalis, the name (only) of which is preserved in CIL 5.1.2864. The disparate character of its contents, however, creates difficulties, and Premerstein (SZ 43 (1923) 45ff.) thought it consisted of drafts prepared by Caesar and put into force by Antony with the rest of Caesar’s acta after his assassination, without receiving the necessary revision. For other conjectures and bibliography see Wenger, Quellen 375ff.; de Martino

3.306ff. Schénbauer (RIDA (1954) 373ff.) dates it a quarter of a century earlier and denies it any connexion with Caesar or any general significance. The latter opinion may well be right. It seems likely that such leges datae were compiled locally from Roman laws of various kinds, adapted more or less freely to local needs. The tabula Heracleensis

would then be such a compilation, not yet adapted; see Frederiksen, JES 55 (1965) 183 ff.

6 Cf. Norr, TR 31 (1963) 566ff. Aulus Gellius, 16.13, says that even the inhabitants would rarely know the difference. C. Saumagne, Le droit latin et les cités romaines sous Vempire (Paris, 1965), maintains that municipia outside Italy were confined to Latin rights; but see Sherwin-White, JES 58 (1968) 269f. 7 Municipal constitutions are best illustrated by two inscriptions containing parts of the leges datae of two Spanish towns which became municipia when Vespasian gave Latiwm minus to the whole of Spain (lex Salpensana, lez Malacitana, Bruns 1.142ff.; FIRA 1.202ff.). 348

Local government who carried out duties similar to those of the Roman aediles. Sometimes, however, six or eight magistrates are found. They were elected, in the earlier period, by popular vote, but this practice seems to have fallen into disuse in nearly all cities before the end of the principate. By then municipal office was usually regarded as a burden which those eligible tried to avoid if possible, and Ulpian® speaks as if election by the ordo decurionum were the usual thing. Thus the development in the municipalities followed, after an interval, what had taken place at Rome. Every five years the 1 viri ture dicundo had to undertake a census of the city and they were then called quinquennales. Like the censors at Rome, they had to make up a list of the senate, which usually numbered 100. They had to choose ex-magistrates in the first instance and fill up the remaining places from other qualified persons. Private law legislation, though a theoretical possibility,® did not take place except in matters of quite minor importance, but administrative decrees were possible, and there was considerable independence in matters of finance. The power to impose local taxation appears indeed not to have existed generally, but

the cities derived income from the lands and other property which they possessed, from monopolies, from fines and other sources.’ Administration was the work of the decuriones, not of the assembly. The 11 viri ture dicundo

presided over the decuriones, but, as the name implies, their chief duties were judicial. They, or courts under their presidency, had exercised criminal

jurisdiction of some importance in republican times,? but this was taken from them under the principate and transferred to imperial authorities, the municipal magistrates being left with no more than minor powers of police.* Their civil jurisdiction was more important, though limited to cases

involving small sums,* and further by the principle that the municipal magistrates could not exercise rights, such as that of missio in bona or in integrum restitutio, which depended on imperium.* Acts of voluntary jurisdiction — manumissions vindicta, adoptions and in iure cessiones — since they

all required the use of the legis actzo, were not normally within their power, but the right of using the legis actio might be given in exceptional cases,‘ and the magistrates of the Latin cities could certainly allow manumissions, presumably by a process of local law parallel to the Roman legis actio, which resulted in the slave’s becoming a Latin.’ 8 D. 49.4.1.3. In the fourth century some towns in Africa apparently still elected their magistrates in the assembly; CTh. 12.5.1.

® Mommsen, who in principle excluded local legislation affecting Roman citizens, admitted that minor variations might sometimes have been allowed by the origiuul charters; St# 3.811f. 1 For a summary of municipal finance see Abbott and Johnson, Municipal Administration 138ff.

2 Lex municipt Tarentini 1.4ff. $ Mommsen, StrR 225ff.

46 Kaser, ZPR 7436f. D. 50.1.26. 28. PS 2.25.4. Lex5Salpensana 849

Population and government of cities under the principate Apart from the colonies and the municipalities with Roman or Latin rights there were of course great numbers of cities, particularly in the Kast,‘ where urban life was old, which remained purely peregrine until the constitutio Antoniniana. The Greek cities had constitutions of widely differing types, and Roman influence was at first limited to minor changes, or changes

in the direction of oligarchical as opposed to democratic institutions, but from the early second century onwards ° imperial interference increases, pro-

voked to a great extent by the incompetence of the local authorities. Curatores are sent to deal with the disorganised finances of the cities and the appointment of these officers then becomes general. At first they are of sena-

torial or equestrian rank, but in the third century the local senates appoint members of their own order, and the office of curator retpublicae or civitatis thus becomes a municipal one, overshadowing the older magistracies.!

With the extension of the citizenship by Caracalla the raison d’étre of local sovereignty ceased to exist, and well before the end of the principate it had become clear that everywhere the city institutions were for loca] govern-

ment only and the local authorities entirely dependent on the central government.

8. CLASSES OF THE POPULATION? We have seen that the old distinction between citizens, Latins and peregrines, though it continued to exist, became of diminishing importance. On - the other hand distinctions, originally purely social, acquired legal significance, so that even when all men, or nearly all, were citizens there was still no equality. The senatorial and equestrian classes formed a higher and a lower imperial nobility,? though both were recruited in increasing proportions from provincials. In this respect, as in others, the principate saw the levelling of the provinces with Italy. Next to these imperial orders came the local governing classes in the municipalities, for, as we have seen, the tendency in local government was towards aristocracy. Not only was a property qualification required for membership of the local senates,* but before the 8 A. H. M. Jones, The Greek City from Alexander to Justinian (Oxford, 1940) ; D. Norr, Imperiwm und Polis (Minch. Beitr. 50, 1966). . ® The earliest example is the sending of an official by Trajan to reorganise the free cities of Greece. Pliny, Fp. 8.24, urges him to deal tenderly with them in view of their

glorious past. See A. N. Sherwin-White, Letters of Pliny (Oxford, 1966); and, on the subsequent curatores, Abbott and Johnson, Municipal Administration 90ff., 189ff.

1 Cf. below, 430. |

2 J. Gagé, Les classes soctales dans Vempire romain (Paris, 1964); Rostovtzeff, Economic History, esp. ch. 5.

3 The nobilitas, i.e. the descendants of families which had held the consulship in the republic, died out in the course of the second century; M. Gelzer, The Roman Nobility

(Oxford, 1969) 141ff. 4 At Comum 100,000 sesterces; Pliny, Ep. 1.19. 850

sich iheennenennessee Classes of the population

end of the principate office had become very largely a matter of hereditary right and duty. The cwriales, i.e. the members of official families, in fact took a place in the municipality similar to that held by the senatorial class in the empire at large, and both were privileged persons, honestiores, in the eye of the law.® The rest of the population were humiliores or plebeii, which latter

word had thus assumed quite a different meaning from that which it bore

under the republic. ‘ Plebeit’, says Paulus, ‘cannot hold magistracies reserved for decuriones.’ ©

Among the upper classes in the provinces — before Caracalla’s constitution —the Roman citizens formed a peculiarly privileged class, but in the East it is clear that the members of the Greek cities were also regarded as privileged.’ If they attained the citizenship they did not forfeit their membership of their own city,*® and indeed in Egypt a man had to become a citizen of Alexandria before he could become one of Rome.*® Gradations also existed among the

lower classes, but the great cleavage was in reality between the well-to-do city-dwellers and the rest of the population. Fostering city life seems to have exhausted the efforts of the government for the well-being of its subjects and the condition of the agricultural population was often miserable enough. In Kigypt at any rate, where, owing to the papyri, our knowledge is most extensive, forced labour and heavy taxation were the lot of the fellaheen in Roman times as during the rest of their history. But it was not only in Egypt that this was true, and with the decline of wealth and prosperity the note of compulsion becomes dominant. Before the principate comes to an end we can see the beginning of the colonate, the institution which in the later empire was to bind great numbers of peasants to the soil in the manner of the mediaeval villein.*

If the lower classes were bound to labour at their hereditary tasks, the lot of the curiales was sometimes scarcely more desirable. Wealthy citizens had to contribute large sums to the public expenditure of their cities, and office

involved such great financial liabilities that in the later principate those eligible had to be forced to serve as magistrates. Should they attempt to 5 The distinction between honestiores and humiliores is a social rather than a legal one and is therefore never precisely defined, but in the course of the principate it acquired legal consequences, especially in criminal matters; cf. below, 403, and see Cardaseia, RHD (1950) 305ff., 461ff.; Garnsey, Social Status.

6 DP. 50.2.7.2. Freedmen were excluded by a definite enactment, the lex Visellia of A.D. 24, but they, as well as actors, criers and others whose professions excluded them, eould (if sufficiently wealthy) become priests of the imperial cult and enter the ranks of the Augustales, a sort of minor municipal order. 7 See e.g. Schénbauer, SZ 49 (1929) 396ff. 8 This rule was admitted for the principate even by Mommsen, StR 3.699; cf. above, 71ff.

9 Pliny, Hp. 10.6 (22). 1 Below, 435ff.

351

Population and government of cities under the principate escape they could be forced to serve,” though conditions had not yet become as bad as they did later, when office itself was used as a punishment.® Like municipal office a military career was also gradually becoming hereditary, as was the membership of certain guilds, those, for instance, of transport workers, which were bound to labour in the service of the state. The iron

system of the later empire, in which all vital occupations became compulsory and hereditary, was in fact prepared already in the century preceding Diocletian.

2 D. 50.4.9. 3 CTh 12.1.16; below, 434.

852

CHAPTER 21

Sources of law in the principate 1 THE OLD SOURCES A. Custom.’ In a famous text Justinian, following Ulpian, divides the law into 2us scriptum and tus non scriptum.? By ius non scriptum only custom is meant; all other law, that derived from responsa prudentium and magisterial edicts, as well as that enacted by leges, plebiscita, senatus consulta and imperial constitutions, is classed as written.* It is thus clear that the term ‘ written law’ is being used in a perfectly literal sense, not in the modern sense in which it is equivalent to enacted or statute law.* Responsa prudentiuum are not statutes but they exist in writing and the law which they lay down is thus classed as written. Customary law on the other hand is unwritten because it can be discovered, not by reading any document, but only

by observing what people actually do.

That much of earlier Roman law is of customary origin is obvious, though this is expressly stated of relatively few institutions.* But it is widely held that custom was not acknowledged to be a source of law until after the end of the classical period. According to the classical jurists, on this view, custom made law only indirectly, through the medium of interpretatio. And certainly it is not included in the lists of sources of law given by Gaius and Papinian,® and appears for the first time in what is probably a post-classical compilation.” On the other hand, a text attributed to Julian * says that it is observed pro lege, and provides a theoretical justification in terms of the tacit consent

of the people: ‘ For what does it matter whether the people declares its 1 Much has been written on this in the last thirty years. See e.g. Thomas, TR 31 (1963) 39ff. and RIDA (1965) 469ff.; Schiller, Virginia Law Review 24 (1938) 268ff.; Schulz, Principles 14f.; Gaudemet, Formation 106ff.; G. Stihff, Vulgarrecht im Kaiserrecht (Weimar, 1966) 37ff.; B. Schmiedel, Consuetudo im klass. u. nachklass. rom. Recht (Cologne, 1966), on which see Norr, SZ 84 (1967) 4544f.

2 J. 1.2.3; D. 1.1.6.1 (from Ulpian’s Institutiones). Widely held to be interpolated (Schulz, History 73f.; Kaser, RPR 2.35), but the Aristotelian distinction between vépos Eyypagos and vouog &ypagos, from which (though not identical) it derives, must have been familiar to the jurists, and might well have been used in an elementary work.

8 Pomponius, D. 1.2.2.5, classes the law arising from disputatio fori as unwritten;

this also is commonly held interpolated. 4 Blackstone, Commentaries 1.63.

5 E.g. the interdiction of prodigals (PS 3.4.7; D. 27.10.1 pr.); pupillary substitution (D. 28.6.2 pr.) ; prohibited degrees (D. 23.2.8; 39.1); patria potestas (D. 1.6.8 pr.) ; for the prohibition of gifts between husband and wife (D. 24.1.1) see above, 235.

6 Gai. 1.2; D. 1.1.7. G. 3.82 is corrupt, but in the normal restoration (derived from the parallel passage in J. 3.10 pr.) it speaks of wus quod consensu receptum est. This is, however, regarded by some as a gloss; ef. Lombardi, SDHI 18 (1952) 38.

7 Epit. Ulp. 1.4; ef. below, 458. 8 D. 1.3.32.1. 353

Sources of law in the principate wishes by vote or by its actual conduct?’ This text is widely held to be interpolated,® but not for any compelling internal reason, and since the idea occurs in Cicero and other non-legal writers,’ Julian was certainly familiar with it. It appears no doubt disingenuous in the mouth of a jurist who lived at a time when supreme power had passed from the people to the emperor,

but we must remember that the theory of popular sovereignty remained alive long after his day, and even beyond the end of the principate. Even so, of course, the explanation is very artificial, since the number of persons who

actually have occasion to follow any particular custom can scarcely ever amount to a majority of the citizens entitled to vote, but artificiality is compatible with authenticity.

The truth is probably that our careful distinction between custom as a direct and an indirect source, and indeed the whole question of the theoretical basis of custom was something which held little interest for the jurists. For, except for local custom, which was certainly recognised as a subsidiary

source (and to which most of our texts in their original context probably referred),? custom must have been, in the sphere of private law, of negligible

importance as an active source. (And in the constitutional sphere, where much of the emperor’s power, and indeed partly that of the senate, cannot be

attributed strictly either to any enactment or to the application of republican rules, the Romans do not invoke the idea of custom.) Moreover too many different things are included by the different expressions which we translate ‘custom ’ (mos (maiorum), mores, usus, consuetudo), and too little distinc-

tion is drawn between them for us to formulate rules with any certainty. Sometimes what is meant is that an institution forms part of the primeval Roman law (here mos or mores is generally used);* sometimes the reference is to the practice of the courts, or to a series of precedents which the Romans do not distinguish from what a modern lawyer calls custom.* Again, custom is not clearly distinguished from usage,* nor general from particular custom. 9 See Steinwenter, St. Bonfante 2.419ff., and the references in n 1, above.

1 Cic. de inv. 2.67; Varro, quoted by Servius, ad Aen. 7.601; Quint. Inst. Orat.

5.10.13; Gell. 11.18.4, 12.13.5. 2 Lombardi, SDHI 18 (1952) 21ff. 3 E.g. patria potestas, ef. above, n 5; Allen, Law in the Making, 7th ed. 82; for a

survey see Kaser, SZ 59 (1939) 52ff.

4D. 1.3.38: Nam imperator noster Severus rescripsit in ambiguitatibus quae ex legibus proficiscuntur consuetudinem aut rerum perpetuo similiter iudicatarwm auctoritatem vim legis obtinere debere. D. 47.11.9 refers to the custom of the proconsul’s court in punishing crimes peculiar to the province. Cf. D. 50.1.25. Precedent, though unrecognised in the lawyers’ lists of sources, is well enough known as exemplum or res tudicata to the rhetoricians (e.g. Quint. Inst. orat. 5.2.1; cf. Pugliese, Recueils de la société Jean Bodin 16 (1965) 340ff.) and undoubtedly played a part in the development of the law; Jolowicz, BIDR 46 (1939) 394ff.; id., Journal of the Society of Public Teachers of Law (1937) 1ff. Justinian definitely forbade the use of precedents in C. 7.45.13 — non exemplis sed legibus iudicandum est — but see below, 461.

5 I.e. usages which have no binding authority in themselves but are deemed to be implied in a contract unless the parties expressly exclude them; see Allen, Law in the 354

The old sources

Whatever the attitude of the classical lawyers was, custom is clearly accepted as a source in the law of Justinian,® and the rules which modern systems have evolved for testing the validity of an alleged custom are taken

partly from the Roman texts.’ But the Roman rules themselves never received any exact formulation. At no time was it laid down how old a custom must be, and though a constitution of Constantine® says that a custom cannot prevail against reason, this is not further explained. The greatest difficulty has always been found in the relationship between custom and statute. If, as the text attributed to Julian has it, custom is itself a sort of statute, then it follows of necessity —and the text expressly says ' — that

a statute can be repealed as easily by the growth of a customary contrary custom as by another statute. On the other hand, the same constitution which denies the validity of custom against reason says that it cannot prevail against statute, and, in spite of the appearance of the contrary doctrine in Justinian’s Institutes,” this is more likely to express the practical attitude of the Dominate. Whether the practice was different in the classical period it

is difficult to say. We are told of a number of statutory rules which had fallen into disuse,* but the language used is compatible with their being merely in fact not observed, without such non-observance affecting their formal validity. B. LEGISLATION BY THE POPULAR ASSEMBLIES. A number of leges of funda-

mental importance for the subsequent history of the law were passed through the comitia or the concilium plebis in the reign of Augustus. These included laws concerning marriage and divorce,* manumission,’ and legal procedure.™ Although in some cases they bear the names of the consuls who actually put

them before the assembly, they were all part of the emperor’s policy and passed at his wish. The lex Papia Poppaea indeed of A.D. 9, which imposed disabilities on unmarried and childless persons, can hardly have been welcome to the proposing magistrates, who were both unmarried. A few leges were passed in the succeeding reigns, especially under Claudius,” who in this as in other respects showed his love for ancient republican forms, but, although Gaius still speaks of leges and plebiscita as existing sources of law, Making, 7th ed. 130. Salmond, Jurisprudence, 12th ed. 193ff., uses ‘ conventional custom ’ in this sense. For usage used to explain provisions of a will see D. 32.65.7. 6 J. 1.2.9; D. 1.8; C. 8.52. 7 H. F. Jolowiez, Roman Foundations of Modern Law (Oxford, 1957) 21ff. 8 C. 8.52.2: Consuetudinis ususque longaevi non vilis auctoritas est, verwm non usque adeéo sut valitura momento, ut aut rationem vincat aut legem (A.D. 319). 9 See Thomas, RIDA (1965) 469ff., with reff.

1 See also J. 1.2.11; Gell. 11.18.4, 12.13.5. | 2°1.2.11.

3 Ch. 2 of the ler Aquilia (D. 9.2.27.4); the penalties laid down by the XII Tables for inwwria (J. 4.4.7) ; lex Genucia on usury (Appian, Bell. Civ. 1.54; Val. Max. 9.7.4). 4 L. Iulia de maritandis ordinibus ; l. Iulia de adulteriis ; 1. Papia Poppaea. 5 L. Aelia Sentia; l. Fufia Caninia.

6* Leges Iuliae, above, 218, 313. 7m E.g. Gai. 1.157. 355

||

Sources of law in the principate

the practice of popular legislation had in fact disappeared by the end of the first century A.D.° C. THE EDICTS OF THE MAGISTRATES. The empire brought with it no immediate change in the legal position of those magistrates whose edicts created

the ius honorarium, but their independence suffered, like that of all the other organs of the state, from the new all-embracing authority of the princeps. Changes were still occasionally made in the edicts, but additional clauses

were seldom inserted except in order to carry out the provisions of other law-making agencies, especially senatus consulta,® but sometimes even leges.’

Even the theoretical power to alter the ius honorarium was, however, ultimately incompatible with the emperor’s supremacy, and it seems that, here as elsewhere,’ it was Hadrian who took the decisive step towards bringing the republican form into line with the new realities. Our accounts of this reform are, however, meagre and late,® and even if they are to be trusted * they leave much that is obscure. It seems that

8 Above, 326. |

® Below, 363. Independent action of Cassius as praetor is mentioned, but it does not appear to have resulted in permanent changes in the text. In D. 4.6.26.7 he is said to have issued an edict promising in integrum restitutio if the court were prevented from sitting by the proclamation of a public holiday, but this does not seem to have been incorporated in the permanent edictum, for the passage goes on to say that Celsus approved of the principle, which he would not have been likely to do if there had been a clause embodying it specifically. D. 29.2.99 refers to a special case. On D. 42.8.11 see Lenel, HP 500, n 2, and on D. 44.4.4.33 see Watson, Obligations 257f. Pliny, Ep. 5.9.3, refers to a breve edictwm issued by a praetor who presided over some special court in which he announced his intention of enforcing a sc. concerning gifts to advocates. It was perhaps an old sc. which had not been observed, for the excitement of which Pliny speaks clearly shows that the praetor need not have issued it had he not wished, and the ecentumviral praetor adjourned his court to consider whether he should follow the example or not.

1 The l. Papia Poppaea gave in some cases bonorwm possessio (not hereditas), e.g. Gai. 3.50, and the praetor put a clause in his edict granting b.p. wherever a statute required it (Lenel, HP 360); see Buckland 9. There is no definite instance of a clause put in at the emperor’s wish, except perhaps D. 43.4.3.3 (Lenel 348), but the edict in several places (e.g. D. 3.1.1.8; 4.6.1.1) puts decreta principwm on a level with leges, ete. 2 Pringsheim, Ges. Abh. 1.91ff. (=JRS 24, (1934) 141ff.). 3 C. Tanta § 18; C. 4.5.10.1 (A.D. 530); Eutrop. 8.17 and Aurelius Victor, de Caes. 19.2 (both of the late fourth century, but probably deriving from a common source early in the century); Epitome Leguwm (Zachariae, Ius Graeco-Rom. 2.280; tenth century). The date of the revision is probably between 130 and Hadrian’s death in 138; Kriiger

94 n 9; Girard, Mélanges 1.214ff., Manuel 57 n 3. St. Jerome (Chron. a Abr. 2147) gives 131, but we do not know from where he derived this date. —

4 Their credibility is forcefully questioned by Guarino (Ordinamento 396ff., 425ff. =St. Albertario 1.623ff. and Attt Verona 2.167ff.). It is certainly surprising that Gaius makes no mention of the reform, and the other classical legal writers were presumably equally silent, since Justinian’s compilers would surely have excerpted a reference if there had been one; ef. Honoré, Gaius 46ff., 54f. Moreover, the Latin and Greek versions of C. Tanta do not give identical accounts; see immediately below. But it is difficult to believe that what we are told has no foundation at all. See also below, 384f.

856

The old sources Hadrian commissioned the great jurist Salvius Iulianus to revise or consolidate the praetorian edict in some way, and that the finished work was given some confirmation by a senatus consultum passed at the emperor’s wish.°

It is difficult to tell what Julian in fact did. We know that he introduced one new substantive rule ® (concerning a particular case of bonorum posses-

sio), but the fact that this is referred to as nova clausula Iuliani suggests that there were at any rate not many others.’ He evidently made no attempt to rationalise the arrangement of the edict. For it is possible to reconstruct the order of the titles from the fragments of the great classical commentaries ad edictum which survive in the Digest, and it is plainly an arrangement which is the result of gradual accretion.* The same conclusion can be drawn even more clearly from the lack of any uniformity of style.® The conjecture has been made ' that, whereas previously the pattern formulae had appeared

all together in an appendix, Julian placed each one after the appropriate clause in the body of the edict, or in some other suitable position, but the evidence for this is flimsy. Nor is it clear what powers the praetor retained. Gaius * speaks of the ius edicendi in the present tense, and the edict continued to be published at the beginning of each year of office, so that the distinction between ius civile and ius honorarium persisted, but as a source of new law the edict apparently ceases to count.? For though Gaius,‘ for example, speaks of a rescript altering the rules of bonorum possessio (an institution of the zus honorarium), he says nothing of any change in the text of the edict, nor is there anything to show that the numerous actiones utiles allowed on the authority of the jurists found any mention. The instruments

of praetorian law reform therefore still exist, but they are in the hands of the emperor and the jurists. 5 C. Tanta § 18. 6 D. 37.8.3. Cosentini, St. Solazzi 219, declares the text interpolated, but unconvincingly.

7 He probably made a correction in the edict de eo quod certo loco dart oportet; Lenel, EP 243.

8 The universally accepted reconstruction is that of Lenel, HP. See also Schulz, History 148ff.

9 On the style of the edict see Kaser, Fschr. Schulz 2.21ff.; Daube, Forms; Dernburg attempted unsuccessfully to use stylistic differences to trace the development of the edict (Festgaben A. W. Heffter (Berlin, 1878) 91ff.). 1 By Wlassak, Hdict u. Klageform (Jena, 1882). Girard, Mélanges 1.300ff., adduced

in support a fragment of the Notae Iuris of Valerius Probus (who wrote in the second half of the first century A.D.), which gives a list of abbreviations used in the early part of the edict, but gives no abbreviations for formulae (FIRA 2.456). Probus may

well, however, have devoted a separate section of the book to actions. 2 1.6.

3 In its permanent form it is thus perpetuwm in a new sense, though according to Pringsheim, Ges. Abh. 1.102ff. (=Symb. Frib. 1ff.), the word did not acquire the

meaning of ‘unchangeable’ until about A.D. 300. 4 2.120. 5* C. Tanta § 18 appears to say that any gaps are to be filled by the emperor (and attributes this to a sc.), but the Greek version in C. AdSwxev leaves this task with the magistrates (and says that Hadrian laid this down in a speech to the people). 357

Sources of law in the principate Our accounts speak only of one edict, but if what has been said of the pur-

pose of the revision is correct, it must have embraced not only the urban edict but those of the peregrine praetor and the curule aediles as well, for these magistrates cannot have been allowed to retain powers which were taken from the urban praetor. It has been suggested that the urban and pere-

grine edicts were fused,* but Gaius speaks of them as separate,’ and the silence of our few authorities is easily explained by the fact that they all lived long after the extension of the citizenship by Caracalla. Gaius also speaks of the aedilician edict as a separate document, and commentaries were written on it. In Justinian’s time those commentaries were evidently treated simply as appendices to their authors’ commentaries on the praetorian edict,* but there is nothing to justify our saying that Julian himself treated one edict as an appendix to the other.? The provincial edicts give rise to a difficulty. Gaius speaks as if each governor still promulgated his own, and one would imagine that local cireumstances would require the preservation of differences. On the other hand Gaius wrote a commentary ad edictum provinciale, which suggests that there had

been only one, and the fragments of this work in the Digest show that the text on which he commented was largely identical with that of the urban edict. The explanation is probably that the document in question did not concern provincials generally but only Roman citizens living in the provinces, to whom the governor in each case extended the protection of the ius honorarium. Had this not been done they would have remained dependent solely on the archaic civil law.? In addition to the edicts issued at the beginning of their term of office and

consisting almost wholly of matter taken over from their predecessors, the provincial governors could publish edicts dealing with special matters. A 6 Rudorff, Zeitschr. f. Rechtsgesch. 3 (1864) 21, cited by Kriiger 95. But they were probably largely identical; Lenel, HP 3. 7 1.6.

8 Theophilus, 1.2.8, speaks of the aedilician edict as an appendix to the (single) edict of the two praetors, and the Florentine Index auctorwm adopts the same attitude to the commentaries, since it ascribes 80 books instead of 78 to Paul’s and 83 instead of 81 to Ulpian’s, and in each case makes no mention of the two books on the aedilician edict; but the inscriptions of the individual fragments preserve the original distinction. 9 As Lenel does, EP 48; contra, Kriger 95 n 15; Volterra, Scrittt U. Borst (Padua, 1955) ; Iwra 7 (1956) 14148.

1 Lenel argues (£P 5) that the subsequent unification of law throughout the empire would necessarily mean that the compilers would have no use for those parts of the commentary which dealt with rules peculiar to the provincial edict. But see next note. The existence of a general provincial edict cannot be established from the evidence of the papyri; Katzoff, TR 37 (1969) 415ff. (against Wilcken, SZ 42 (1921) 135, and Wenger, Quellien 411).

2 Buckland, RHD (1934) 81ff. Julian could easily revise such an edict. With other edicts peculiar to particular provinces he would have had great difficulty, and they had presumably nothing to do with his special task.

358

The old sources number of such edicts issued by the prefects of Egypt have been preserved,’

some, especially that of Ti. Iulius Alexander of A.D. 68* (which forbade imprisonment for debt), of great interest. The Egyptian edicts at any rate appear to have remained in force even after their authors had ceased to be governors.°

Apart from the governors of imperial provinces, the new imperial officials did not follow the republican practice of issuing general edicts when they entered upon their offices,® but they could naturally give general instructions within their sphere when they thought it necessary, and a constitution of A.D. 235" recognises a kind of subordinate legislative power in the prefects of the praetorian guard, by commanding that general instructions given by them were to be observed, provided they were not contrary to leges or imperial constitutions. The practice may be older than the constitution. D. RESPONSA PRUDENTIUM AND THE IUS RESPONDENDI. Responsa prudentium

continued in the principate to be an increasingly important source of law, but in this case, as in that of edicta, it was evidently felt that complete independence would be incompatible with the new regime. For Augustus already made a move towards bringing the jurists under imperial influence, and

Hadrian may have taken some further step. Much, however, is obscure because our information derives from a very defective text of Pomponius and a brief reference in Gaius’ Institutes, and there has in consequence been a great deal of controversy. Pomponius * says that whereas previously anyone who had sufficient confidence in his own learning gave responsa, and gave them without any particular form, Augustus decided, in order to give greater authority to the law, that responsa should be given ‘ with his authority ’ and under seal. Thenceforth the right to give such responsa — the ius (publice) respondendi — was sought as a privilege (beneficitum). On the other

3 See Wilcken, SZ 42 (1921) 187-9. 4 Bruns 1.243; FIRA 1.318. 5 Wilcken, SZ 42 (1921) 139ff.

6 It is also noticeable that the consuls and praetors concerned with jurisdiction extra

ordinem (e.g. fideicommissa, below, 395f.) did nothing of the sort. 7 C, 1.26.2. 8 Pomponius, D. 1.2.2.48-50: Massurius Sabinus in equestri ordine fuit et publice primus respondit: posteaque hoc coepit beneficiwm dari, a Tiberio Caesare hoc tamen ill concessum erat. Et, ut obiter sciamus, ante tempora Augusti publice respondendi ius non a principibus dabatur, sed qui fiduciam studiorum suorum habebant, consulentibus respondebant: neque responsa utique signata dabant, sed plerumque iudicibus ipsi scribebant, aut testabantur qui illos consulebant. primus divus Augustus, ut maior turis auctoritas haberetur, constituit, ut ex auctoritate etus responderent: et ex illo tempore peti hoc pro beneficio coepit. et ideo optimus princeps Hadrianus, cwm ab eo viri praetorti peterent, ut sibt liceret respondere, rescripsit ets hoc non peti, sed praestari solere et ideo, si quis fiduciam sui haberet, delectari se