Professional Documents
Culture Documents
ULPIAN
Pioneer of Human Rights
TONY HONORÉ
Second Edition
1
3
Great Clarendon Street, Oxford
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide in
Oxford New York
Athens Auckland Bangkok Bogotá Buenos Aires Cape Town
Chennai Dar es Salaam Delhi Florence Hong Kong Istanbul Karachi
Kolkata Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi
Paris São Paulo Shanghai Singapore Taipei Tokyo Toronto Warsaw
with associated companies in Berlin Ibadan
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
© Tony Honoré 2002
The moral rights of the author have been asserted
Database right Oxford University Press (maker)
First edition published 1982
Second edition published 2002
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose this same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
Library of Congress Cataloging in Publication Data
Data Avaliable
ISBN 0–19–924424–3
1 3 5 7 9 10 8 6 4 2
Typeset by Hope Services (Abingdon) Ltd.
Printed in Great Britain
on acid-free paper by
T. J. International Ltd., Padstow, Cornwall
To Detlef Liebs
studiosus studioso
Preface
T H E first edition of this work on Ulpian, the early third-century lawyer from
Syria, came out in 1982. Though there had been substantial articles on Ulpian
by Pernice (1885), Jörs (1903), and Crifò (1976), mine was the first modern
study of his life and works at book length. Texts from Ulpian take up two-
fifths of Justinian’s sixth-century Digest, for many centuries the staple of legal
education. For most topics Justinian’s compilers chose Ulpian’s account as
their core text. Through the Digest his exposition of the law has influenced the
western legal tradition at least as much as that of any later writer, including
Irnerius, Grotius, Blackstone, Pothier, and Savigny.
I set out to study him as a writer and imperial office-holder in the Severan
age (AD 193–235) and, as Millar put it, a figure in the complex cultural land-
scape of the Roman empire. Assessments of his role differ. In Syme’s eyes
(1972) Ulpian, a lawyer in politics, deserted his books to ingratiate himself
with the ruling dynasty and paid the penalty when he was murdered by rebel-
lious troops. To me, in contrast, his career was that of an intellectual in
government, concerned to give effect in law and administration to the cos-
mopolitan ideas of his time.
The book had a mixed reception. It was welcomed by Rodger (1983), Liebs
(1984), Birks (1983), and Millar (1986), a welcome tempered by admonitions
against pushing the evidence too far. Gordon (1984) and Crifò (1985) had
reservations about the analysis of Ulpian’s style. The book was sharply criti-
cised by Watson (1983) and, more soberly, by Frier (1984), whose point of
view was close to that of Syme. Birks showed that Watson’s criticisms were
largely mistaken. But some of the points made by scholars, favourably dis-
posed or sceptical, need to be taken seriously and some arguments in the first
edition strengthened. As Frezza (1983) pointed out, there was an imbalance
between my depiction of Ulpian as a writer and as a lawyer in government.
One problem in writing about Ulpian is that hardly anyone has read the
bulk of his surviving work, which runs to some 300,000 words. Most students
of the ancient world read some Ulpian, and his work has been available in a
consolidated form since Lenel’s Palingenesia of 1889. But most scholars read
only as much as is needed for the purpose in hand. The Digest, and with it the
western legal tradition, is in thrall to Ulpian, but he comes in bits and pieces.
The sporadic reader can easily miss the quality and flavour of his writing.
The first edition sought to fix the main features of Ulpian’s style and, on
this basis, to distinguish between the works that are genuinely his and the five
attributed to him that are in my view spurious. There is no need to change the
lists of genuine and spurious works arrived at (Chapters 5 and 10), though
viii Preface
some scholars still defend the authenticity of Rules in one book (Liber singu-
laris regularum) and possibly Replies (Responsa). Some argue that Opinions,
though not a work of Ulpian, is classical. The debate continues.
The list of mature works that seem certainly or probably genuine comprises
216 or 217 books in all. By a book is meant a liber of on average 10,000 to
12,000 words, corresponding to a modern chapter. Most of Ulpian’s work
can be dated to Caracalla’s sole rule (December 211–April 217). My hypoth-
esis was that Ulpian’s massive survey was composed in the five years 213–17,
after and in the light of the enactment of Caracalla’s Antonine constitution,
which is traditionally (and I think rightly) dated to 212. The author may have
planned to continue writing after 217 and, in particular, to compete his
commentary On Sabinus. If so, he did not or could not carry out his plan. The
five-year period 213–17, Quinquennium Ulpiani, was devoted to a large-scale
exposition of law and public administration for the benefit of rulers and
citizens of the new cosmopolis. Undertaken privately but with official encour-
agement, it was, as Liebs (2000) points out, a forerunner of the later Roman
codifications by Gregorius, Hermogenianus, Theodosius II, and Justinian.
If Ulpian composed his survey in the five years 213–17 he must have com-
posed at speed and worked to a plan. I suggested in the first edition that his
plan was to complete a chapter (liber) a week, apart from a holiday period of
eight weeks. To some this seemed absurd. But, as emerges in Chapter 8, the
argument for my suggestion is stronger now that it was in 1982.
In this and other respects I am tempted to say that most of the detailed
conclusions reached in the first edition were correct or at least defensible but
that the book was not reader-friendly. It did not sufficiently stress certain
themes that mark off Ulpian’s contribution to Roman law from that of oth-
ers. It did not do justice to the egalitarian philosophy that inspired his work,
to his use of dictation, his easy shifting from Latin to Greek, and his empir-
ical approach to the solution of legal problems. The work lent itself to cari-
cature as ‘Law on the Installment Plan.’ Caricatures distort but also
illuminate. So in this edition the emphasis has changed. Three new chapters
have been added and one discarded. Certain themes, previously muted, are
now stressed. I now think (Chapter 2) that Ulpian dictated the whole of his
survey of Roman law. His survey reflects speech, not writing. In the ancient
world dictation was important to the Roman government and to Christian
writers but was not confined to them. Dictation accounts for the linking of
sentences, the conversational flow, the sense of debate, and the superficial
egocentricity of an author who, though self-confident, often expressed his
opinion in a tentative way. Oral composition fits the many examples he gives
by way of illustration and his recourse to argument by analogy (Chapter 4).
In the upshot Ulpian reasons rather as do common lawyers, whose style
derives from a practice of oral exchange between judges and counsel and of
argument from example.
Preface ix
Moreover, Ulpian is the Roman lawyer whose thinking most clearly has a
basis in theory. His philosophy, implied in his Teaching Manual (Institutiones)
and elsewhere, is cosmopolitan and egalitarian (Chapter 3). It fitted the con-
temporary extension of citizenship to all free people in the Roman empire.
Ulpian here reflects, but also promotes, the Zeitgeist. He can therefore justly
be accounted a pioneer of the human rights movement. Drawing mainly on
the Stoic conception of natural law, his philosophy allows slaves the share in
human dignity that is required by the belief that human beings are born free
and equal. At present, when human rights are widely debated, it may promote
a balanced view to see them in a perspective that goes back to antiquity.
Human rights are not a product of the Enlightenment, still less of the twenti-
eth century, as some otherwise well-educated people suppose. The values of
equality, freedom, and dignity, to which human rights give effect, formed the
basis of Ulpian’s exposition of Roman law as the law of a cosmopolis. The
citizens of the empire did not enjoy political freedom. But they possessed civil
rights, and Roman lawyers were concerned to see that these were respected.
To ensure this respect was the emperor’s duty. No one made a greater contri-
bution to this end than Ulpian, whose mission was to expound Roman law to
the rulers and citizens of the cosmopolis as a system of law based on reason,
utility, and equity. That is one reason why the study of Roman law can never
be out of date.
The book has been reshaped in, I hope, a more reader-friendly style. This
has meant, among other changes, translating into English the titles of
Ulpian’s works. Latin cannot be avoided in a work that turns in good part
on the style and outlook of a Latin author. As much Latin as possible has,
however, been translated or relegated to the footnotes. If the result strikes
some students of the ancient world as odd, this is the price to be paid for
introducing a great writer to the wider audience he deserves.
I am grateful to Dr Dennis Flynn for his help with computing problems,
and to Dr Leofranc Holford-Strevens, to whose scholarly acumen as copy-
editor both editions of this work owe a debt.
T.H.
Contents
Spurious
Encyclopaedia Pandectae 10 (pand.)
Replies Responsa 2 (resp.)
List of Ulpian’s Works with Method of Citation xiii
1 Lenel (1889) 2.379–1200, 1264 with Sierl (1960) 12–7; Pernice (1885/1962); Kalb (1890)
126–35; Schulze (1891); Jörs (1903); Fitting (1908) 99–120; P. Krüger (1912) 239–50; Volterra
(1937); Berger (1952) 750; Leclercq (1952); Wenger (1953) 519f., 526f.; Pflaum (1960–1) no.294;
Mayer-Maly (1961); Honoré (1962a) 207–12 (unreliable); Frezza (1968); Syme (1970/1979);
Carcaterra (1972); Orestano (1973a); Nörr (1973); Crifò (1976); Nörr (1978) 131–6; Honoré
(1982); Rodger (1983); Birks (1983); Frezza (1983); De Marini Avonzo (1982); Liebs (1984);
Mantello (1984); Lannata (1984) 214–9; Liebs (1987a); Yaron (1987); Honoré (1988); Waldstein
(1985); Crifò (1985); Millar (1986); Schermaier (1993); Winkel (1993); Marotta (2000); Honoré
(2001). Literature on his career n.49
2 Syme (1980) 102 = (1984) 1412 3 Liebs (1997) §424 p. 187
2 1. Background and Career
fifteen, Marcus promoted him joint Augustus and so ensured that he would
in due course succeed his father as emperor. The reign of Commodus (180–92)
ended in misgovernment and murder. Of the various contenders who emerged
in 193, the ultimate victor was Septimius Severus (193–211).4 But it took two
civil wars and nearly four years for him to become sole master of the Roman
world. The effort left its mark both on him and on the empire.
The rule of the Severans was harsher but more cosmopolitan than that of the
Antonines. Severus needed to establish a link with Marcus, since Marcus was
then and later considered a model ruler. So in 195, after defeating his rival
Pescennius Niger, he arranged to be adopted by Marcus. The adoption was a
fiction, since Marcus had been dead fifteen years. Nevertheless it was
put through in detail. Severus’ elder son, Bassianus (generally known as
Caracalla), then aged eight or nine, received the dynastic name ‘Antoninus’.5
Severus tried, however, to avoid repeating Marcus’ mistake over Commodus,
who had been allowed too much freedom. He gave Caracalla a good education
and saw to it that he had military experience. He made him consul three times
during the next sixteen years, and from 197 or 198 introduced him to civil and
legal business as joint Augustus. Caracalla was able, though not in a literary
way, quick-witted, a good judge of character. He had plebeian tastes, liked sol-
diering, and was not averse to menial tasks. With the soldiers he was and
remained popular.
At times one could think of him as a future emperor who would justify the
name Antoninus. At others he appeared unhinged. Then his violent temper
and bitter feuding with Geta, his younger brother, left no room for illusion.
Geta, a year younger, was thought incompetent.6 After Caracalla was pro-
moted Augustus, the younger son was kept in the subordinate position of
Caesar for another eleven years. Ultimately Severus, perhaps on the pretext of
administrative convenience, made Geta a third Augustus during the British
campaign of 209, so that there was someone left behind at York to attend to
civil business while he and Caracalla went on campaign in Scotland. He
thereby committed the empire to the joint succession of the two brothers. Once
again, dynastic considerations overrode the claims of good government.
Septimius followed the traditional Roman policy of extending the limits of
empire. Not specially gifted as a general, he repeatedly won victories in both
civil and external wars. In 197, when the civil wars were over, he invaded
Parthia and pushed the frontier of the empire eastwards. At the end of his
4
Hasebroek (1921) has a good account, with dates. See also Platnauer (1918), Hammond
(1940), Murphy (1945), Hannestad (1944), Barnes (1967), Alföldy (1968), Herzig (1972), and
Birley (1988) with bibliography 258–73 and prosopography 212–29
5
Dio 78.1–10 (hostile); Herodian 4.7.4–7; RE 2.2434 Aurelius (46); Schulze (1909). For
Caracalla as a judge Nörr (1972b) 25; Kunkel (1953/1974) 255. On young emperors Hartke
(1951)
6
Dio 77.7.1–2; Herodian 3.10.3–4; 4.3.2–4.4.3; Fluss (1923); Alföldy (1972) 19
1. Background and Career 3
reign, in 209–11, he tried to advance the British frontier north into Scotland,
perhaps with a view to occupying the whole island. When he died in February
211 the attempt was anyhow abandoned and the army, with the two emper-
ors Caracalla and Geta, returned to Rome. They were accompanied by two
eminent lawyers: Papinian, a supporter of the joint succession, and Ulpian,
who turned out to be a supporter of Caracalla.
Severus had debts to the army. It was the troops of Pannonia that had
brought him to power, and his campaigns required loyal troops. The balance
between civil and military power shifted in favour of the soldiers. They
obtained concessions: higher pay, permission to live with their wives.7 The
events of 193–7 revived the lesson of AD 68, that the army legions made and
unmade emperors. Severus took steps to ensure that in future the praetorians
in Rome, who had killed Pertinax and made Didius Iulianus emperor, would
not do this again. In 193 he tricked and disarmed them.8 But this tour de force
was not a long-term solution to the problems of military indiscipline. An
emperor could not survive if the troops and their generals were disloyal to
him.
Severus was well equipped to manage civil as well as military affairs.
Though not as well educated, says Dio, as he would have liked to be, he had
an inquiring mind. Trained in ‘philosophy’, he was a man of many ideas but
few words.9 Whether he had received any legal education is obscure, but he
was a good administrator and assiduous as a judge.10 He claimed, though not
bound by law, to conform to it.11 He appointed lawyers to important posts.
Paul’s reports of decisions in the imperial council12 show that he took an active
part in legal argument and was prepared to differ from the opinions of his
lawyer councillors, particularly Paul.13 He was keen to improve the working
of the legal system. The cumbrous operation of the permanent criminal com-
missions (quaestiones perpetuae), which from the republic onwards had exer-
cised criminal jurisdiction in Rome, now ceased.14 The urban prefect in Rome
was given unlimited jurisdiction at first instance over crimes committed in the
7
Herodian 3.8.5. Whether soldiers’ marriages were previously void in certain cases is debat-
able: see Kaser (1971) 1.317. Herodian 3.8.4–5 also mentions the right to wear gold rings (the
mark of equestrian rank, in practice confined to centurions and other middle-ranking officers).
See also Whittaker (1969–70) 2.308–9
8 Herodian 3.13.2–12
9 Dio 77.16.1–2; Eutropius 8.9.1 cf. HA Severus 18.5–6. He composed an autobiography:
attamen legibus vivimus. The formulation is probably Ulpian’s in view of the parallel licet . . . atta-
men in CJ 5.18.2 (4 Apr 207) and the Institutes text (saepissime rescripserunt) comes from his
work: ch.2 n.769
12 Lenel (1889) 1.959–65, 1111–2 nos. 57, 58, 62, 63, 64, 65, 67, 68.2, 69, 70.1, 2, 73, 5, 76, 7,
79, 878
13 Coriat (1997) 561–3 14 Garnsey (1967)
4 1. Background and Career
city or within a hundred miles of it. The praetorian prefect15 had similar juris-
diction in Italy beyond the hundred-mile limit,16 and unlimited appellate juris-
diction in both civil and criminal cases.17 In this he acted, theoretically, as the
emperor’s delegate.18
Delegation was inevitable, especially with an emperor as ambitious as
Severus. The office of praetorian prefect, now generally held by two prefects
at a time, had grown piecemeal into the most important in the empire next to
that of the emperor himself. Originally a military post,19 its duties spread
from the command of the praetorian guard, stationed in Rome, to the super-
intendence of the armies in the provinces. Its civil business, especially legal
appeals, now increased to the point at which one of the two prefects was
sometimes a lawyer. It was thus that, following Tarrutienus Paternus
(177–82)20 in the reign of Commodus, Papinian (205–12), Macrinus (212–7),
Ulpian (222–3/4), and conceivably Paul (219–20), men whose careers were
purely civilian, came to hold this, the highest equestrian office.
Another aspect of the administration of justice was the rescript system.21
The emperor provided what was in effect a free legal advice service, for which
the secretary for petitions (a libellis) was responsible. One could petition
about anything: privileges, pardon for crimes, grants of land, honours. But
when a petition involved a point of law it was referred to the secretary for
petitions who drafted a reply for the emperor’s consideration. From the
Severan age far more rescripts on points of law survive than from previous
reigns, and there is no doubt that their number really increased. In this
domain, too, the emperor took his duties seriously.
It was the concern of the government for civil administration that made the
age a great one for law and lawyers. The Severan jurists, Papinian, Paul,
Tryphoninus, Messius, Menander, Ulpian, Modestinus, all equestrians, had an
interest in seeing that justice was freely available and that law prevailed.22
Whether they were advising the emperor as members of his council, composing
rescripts for him as secretaries for petitions, or writing treatises for the use of
governors, judges, officials, and private citizens, they were conscious of being
members of an élite circle of lawyers. It is a mistake to think of advice given to
the emperor as ‘bureaucratic’, private writing or practice as ‘free’.23 Both rested
on professional discipline, and the test of their worth was the opinion of fellow
members of the élite. There is nothing to suggest that lawyers felt under pres-
sure to slant their opinions in a sense favourable to the government.
15
Howe (1942) 42; Strachan-Davidson (1912) 1.158
16
Collatio 14.3.2 (Ulp. 9 off. proc. referring to the Lex Fabia); Passerini (1939) 236
17 Mommsen (1887–8) 2.1113f.; Ensslin (1954) 2391; Kaser (1966) 365; Howe (1942) 29f.
18 Howe (1942) 40 19 Howe (1942) 7; Palanque (1933); Durry (1938); Passerini (1939)
20 AE 1971.534; CIL VI.4.1.27118; Lenel (1889) 2.335; Pflaum (1960–1) 1.420–2; Kunkel
(1967) 219–22 (‘Tarruntenus’); Giuffrè (1974b) 61–5; Liebs (1976) 341–4; HLL 4 §419.6
21 Honoré (1994) ch.2 22 A. Stein (1927); Howe (1942) 43; Schiller (1953)
23 As does Syme (1972) 406 = (1984) 863
1. Background and Career 5
Severus was the only emperor apart from his rival Albinus to come from
Africa. He was also the first emperor not to be of purely Italian stock.24 That
does not make him an ‘African emperor’,25 but it helps to explain the
cosmopolitan outlook of his regime. He came from an area where Punic was
spoken,26 and was himself fluent in the language. Punic is the language
that Ulpian mentions as an example, after Latin and Greek, as one that
the parties to certain transactions may choose.27 Severus’ second wife, Julia
Domna, came from a part of Syria where, though Punic had not been
spoken for two centuries or more, the vernacular was Aramaic. Ulpian also
mentions this language (‘Assyrian’) as a language that can be used for certain
legal transactions.28 This does not imply that these languages were commonly
used to make contracts or create trusts. Ulpian’s point is that for informal or
semi-formal transactions any language may be used provided it is understood
by the parties. The new citizens are reassured that they can participate in the
legal system to which they are subject without mastering Latin or Greek.
Though he wrote in Latin, the language of Roman law, the readership for
which he was writing included Greeks, many of them newly enfranchised as
citizens, as we shall see in chapter 3.29
It was an opportunistic act on the part of Severus’ son Caracalla in AD 212
to make all free inhabitants of the empire, with obscure but unimportant
exceptions, Roman citizens. But it fitted the spirit of the dynasty and the age.
By the Antonine constitution (constitutio Antoniniana)30 the different
provinces, east, west, north, and south, were put on a level. Other distinctions
too were blurred. With their new privileges, soldiers had a status closer to that
of civilians. Women were more prominent than before, both as property own-
ers and in politics. In the imperial circle the Syrian princesses Julia Domna,
her sister Julia Maesa, and Maesa’s two daughters Soaemias and Mamaea
had influence behind the scenes. But they also figure prominently, with
official titles, on coins,31 and Caracalla entrusted his correspondence to his
mother Julia Domna.32 Rank and class still depend on wealth,33 but other
boundaries, social and conceptual, are blurred.
Gilliam (1965); Sasse (1958), (1965); De Visscher (1961); Seston (1966); Gundel (1966); Segrè
(1966); Saumagne (1966); d’Ors (1966); Gaudemet (1967) 528; A. H. M. Jones (1968); Talamanca
(1971); Herrmann (1972); Wieling (1974); Wolff (1976); Buraselis (1989); Honoré (1996) 383
31
RE 11.916, 926, 948; Mattingly 52 (1975) 156–70 and passim; Robertson (1977) 98–102,
127–33, 163–8
32 33
Dio 79.18.2–3 Garnsey (1970) 221f.
6 1. Background and Career
The levelling that occurred caused strains and resentments. The wars of
Marcus had brought to a close the days when the resources of government
amply met military and civil requirements. Even without civil war, the end of
the second century would not have been comfortable. Civil war set the emperor
and senate at variance. At the behest of Didius Julianus, the senate in 193
declared Severus a public enemy. Later it changed sides, but lost prestige.34
When in 196–7 Severus quarrelled with his Caesar, Clodius Albinus, many sen-
ators sided with Albinus, a man who prided himself on his clemency.35 Severus,
whose character fitted his name, was shocked. When he defeated Albinus he
had some dissident senators executed on his own authority. This was to violate
the model of a good ruler, and mistrust festered.36
Like many a ruler Severus came over time to rely increasingly on his own
family and associates. He trusted a kinsman and friend of his youth, Fulvius
Plautianus, to whom he was bound by emotional ties, and treated him almost
as a partner.37 A man of ability and, like Severus, from Africa, in the end
Plautianus became sole praetorian prefect and accumulated great power. Had
it not been for Severus’ two sons, the situation would have been manageable.
Severus would have made his friend a Caesar or junior Augustus, and
Plautianus in due course might have proved himself a good emperor. But
Caracalla had been given the name Antoninus in 195 and made Augustus in
197 or 198, so that this course was ruled out. The clash was resolved in
another way. By an ingenious plot Caracalla and his mother Julia Domna
trapped Plautianus and had him killed in 205.38 The strain was temporarily
reduced, but effective rule had once again been sacrificed to dynastic interests.
Two praetorian prefects were then appointed, one military, Laetus, and one
civilian, the lawyer Papinian. Severus, conscious of the tension between his
sons, tried to improve their ways by moving them from Rome to the Italian
countryside (205–7),39 and later took them with him to Britain on his last mil-
itary expedition (208–11).40 When he died Julia Domna tried to keep the
peace between them.41 She failed. At the end of 211 Caracalla had Geta mur-
dered in her presence.42 She still tried to compensate for her elder son’s lack
of balance. Later her niece Julia Soaemias tried in turn to restrain her son
Elagabal (218–22),43 but with no more success. So intractable were the prob-
lems of an empire conducted by, or in the name of, young men spoiled by
34 35
Dio 74.16–7 Herodian 3.5.2; Dio 75.7–8
36
Dio 76.8.4; Herodian 3.8.1–3
37
A. Stein (1912) 270–8; PIR2 F 554; Howe (1942) 69 no.17; Grosso (1968a) 7; Birley (1988)
128, 162–3, 207. Their quarrels and reconciliations and Severus’ remorse after Plautianus’ death
point to the truth of Herodian 3.10.6 cf. Dio 77.5.1–2
38
Dio 77.3–4; Herodian 3.11.1–3; 3.12.12 (giving an official version); Hohl (1956) 33; Birley
(1988) 161–3
39 40 41
Herodian 3.13.1 Herodian 3.14–5; Dio 77.11–3 Herodian 4.3.8–9
42
Dio 77.18.2–3
43
Herodian 5.5.5–6; 5.7.1–3. On Elagabal see Hay (1911), Barnes (1972), Pflaum (1978)
1. Background and Career 7
early adulation that some Romans became disillusioned with the tradition
that saw in public service the true end of human endeavour.
In this age of strain the minds of many thoughtful people turned inwards.
There were powerful Christians. The ex-slave Callistus took the see of Rome
(c.217–23) 44 against the challenge of his opponent and critic Hippolytus. Julia
Mammaea, the mother of Alexander Severus, ‘a most religious woman, if ever
there was one’ and a supporter of Ulpian in his later years, was inquiring
enough to bring the Christian intellectual Origen from Antioch by military
escort for a meeting.45 Other religions, hovering on the border of monotheism,
attracted devotees. The cult of Elagabal, the god of Emesa, whence Julia
Domna and her family came, was one, a cult unknown to Greeks and
Romans.46 Elagabal, the last Antoninus, nicknamed from the god to whom he
was devoted, saw the deity as a jealous god. His demands, hardly less exclusive
than those of Jehovah, led the emperor-priest to challenge the traditional
Roman state religion.47 Elagabal wanted to invert the relations of politics and
religion. Far from being the handmaid of state policy, as hitherto, religion was
to be the prime concern of the emperor and people.48 Elagabal’s reign warns
us that, in sacred affairs also, the easy-going days are over. It foreshadows the
Christian revolution that is to come a century later.
The main source of information about Ulpian’s career 49 is his own writings.
These survive to the extent of about 300,000 words, roughly an eighth of the
original, and tell us, directly or by inference, a good deal. They were in the
main composed under Caracalla’s sole rule (211–7) as the references to joint
constitutions of him (imperator noster, imperator Antoninus) and his dead but
deified father (divus pater, divus Severus) make clear.50
Apart from this, two rescripts of Alexander Severus51 mention the two pre-
fectures that Ulpian held in 222 under that emperor. There is an inscription
in his honour from Tyre that records these prefectures.52 A papyrus from
44
Gianelli–Mazzarino (1956) 2.291f.
45
Eusebius, Church History (Hist. Eccl.) 6.21 cf. 6.28.1: ‘the household of Alexander consisted
mostly of believers’
46 Herodian 5.3.5 47 Herodian 5.5.3–10; 5.6.3–10; 6.1.3 48 Herodian 5.7
49 Bremer (1883); Balog (1913) 339–421; A. Stein (1943); Howe (1942) 44–52, 75f., 100–5;
Pflaum (1948) 41–5; Kunkel (1967) 245–54; Pflaum (1960–1) 2.762–5 no.294;
Modrzejewski–Zawadski (1967); Grosso (1968b); Syme (1970/1979), (1972/1984), (1980/1984);
Honoré (1982) 1–46; Liebs (1987b) 76f., 90–2, 110, 113f.; Salway (2000) 137f.
50 Ch.7 n.21–70; ch.8 n.2–12, 23–41; ch.9 n.11–2, 50–1, 93–6, 104–7, 130–1
51 CJ 8.37.4 (31 March 222), 4.65.4.1 (1 Dec 222)
52 AE 1988.1051 pp.285f: Domitio Ulpiano praefecto/praetorio eminentissimo viro/iurisconsulto
Egypt shows that he was killed before May/June 224.53 Of the historians
Dio,54 as summarized by Xiphilinus, mentions his prominent role early in
Alexander’s reign, about which Herodian is silent.55 Aurelius Victor (c.369)56
mentions him briefly, as do the epitomes of Festus (363–70)57 and Eutropius
(364–78).58 There is an important passage in Zosimus (c.500)59 and briefer
ones in Zonaras (1118–43)60 and Syncellus.61 The ‘mythistorical’ Augustan
History of the late fourth century, drawing on Marius Maximus, Herodian,
Victor, and other sources, along with a talent for invention, gives him a
prominent role in the lives of Niger,62 Elagabal,63 and especially Alexander
Severus.64 He is there depicted as guiding the young emperor Alexander’s
footsteps in the paths of justice. This shows that towards the end of the fourth
century Ulpian was highly regarded in certain pagan circles in Rome. It can
be used with caution to fill out some aspects of his career. Following Straub
(1978), I have assumed that, at least in certain instances, the author had access
to reliable information about the law and lawyers.65 In dealing with them his
method is often distortion of the truth rather than pure invention. At times it
seems that the distortions have a serious purpose. It is not impossible that the
author was himself a lawyer.
Ulpian’s fellow lawyers call him Ulpianus or, in Greek, Ολπιανς.66 But a
few texts, including two rescripts of Alexander, speak of Domitius Ulpianus.67
So does a text of Ulpian’s contemporary Paul, in which he reproduces a letter
from a friend or client that refers to an opinion of Ulpian.68 Lactantius calls
him Domitius69 and both Modestinus and Dio have ∆οµτιος Ολπιανς.70
The gentile name Domitius is confirmed by the inscription found in Tyre.71
53
P.Oxy. 2565: Barns–Parsons–Rea–Turner (1966) 102 54 Dio 80.1.1; 2.2–4
55
Herodian 6.1.4 says in general terms that legal and civil business was entrusted under
Alexander to men of eloquence and legal experience
56 The Caesars (De Caesaribus) 24
57 Summary of Deeds of the Roman People (Breviarium rerum gestarum populi Romani) 22 pace
Chron. 1.673; Zonaras 12.15 cf. 26.6.2.5 (Mod. 1 excus.); 27.1.4.1 (2 excus: Ολπιανς
κρτιστος); 27.1.13.2 (4 excus: ∆οµτιος Ολπιανς)
67 CJ 8.37.4 (31 March 222); 4.65.4.1 (1 Dec. 222); HA Severus Alexander 68.1
68 D 19.1.43 (Paul 5 quaest.)
69 Church Teaching (Div. inst.) 5.11.19 70 Dio 80.1.1; D 27.1.13.2 (Mod. 1 excus.)
71 AE 1988.1051; above n.52
1. Background and Career 9
though the dot between the two Ns is awkward. If this reconstruction is cor-
rect, the owner of the large country house near Rome was Gnaeus Domitius
Annius Ulpianus. Is he the lawyer Ulpian? There is a possible point of con-
nection apart from the name. The villa belonged to a wealthy man and a
statue of Meleager has been found in the area. Meleager,74 poet and Cynic
philosopher of the first century BC, came from Gadara but lived in Tyre.
Ulpian also came from Tyre. The statue seems appropriate to the residence of
a wealthy scholar proud of his connection with Tyre. He might be the lawyer
or some other member of a well-educated family from that area.
The lawyer took pride in his connection with Tyre. He combined working
in Rome or as part of the emperor’s entourage with strong provincial links.75
In his work on Taxation (De censibus), written under Caracalla and probably
datable to 213,76 Ulpian describes Tyre as his town of origin, splendid, famed
for its various quarters, possessing an ancient history, strong in arms, faithful
to its treaty with the Romans:77
est in Syria Phoenice splendidissima Tyriorum colonia, unde mihi origo est, nobilis
regionibus, serie saeculorum antiquissima, armipotens, foederis quo cum Romanis
percussit tenacissima.
72
CIL XI.3587 = XV.7773
73
Kunkel (1967) 252; Crifò (1976) 738; A. Stein (1903); PIR 21 19, 25; 32 39; Passerini (1939) 324
74 75
Geffcken (1931); Radinger (1895); Wifstrand (1926) Liebs (2001) v–xv
76
Ch.9 n.11–25
77
D 50.15.1 pr (Ulp. 1 cens.). On Tyre see Krall (1888), Fleming (1915) and on municipal
administration Liebenam (1900), Vittinghoff (1951), Nörr (1969)
78 79
Nörr (1963) 525 = (1965) 433. D 50.1.1 pr (2 ed.)
80
CJ 10.40.7 pr (Dio. et Max. AA et CC: cives quidem origo manumissio adlectio adoptio facit)
10 1. Background and Career
why Tyre supported Severus. Tyre and Sidon were the founding cities of
Carthage, and Severus, apart from his Caesar, Clodius Albinus, was the first
and only emperor from Africa.
To prevent a repetition of the revolt that Niger had mustered in the east,
the victorious Septimius divided Syria into two. He deprived Antioch of its
status as capital of the much-truncated Syria Coele, and made Tyre capital of
the newly founded Syria Phoenice. Many of his supporters came from these
provinces.90 Phoenice extended to the north to include Emesa and Palmyra.
It became the seat of the council (κοινν) of Syria and of the High Priest of
Phoenicia (φοινικρχης).91 Some years later, perhaps after 198, Severus made
Tyre a Roman colony and settled in it veterans of Legio III Gallica.92 It was
now called the metropolitan colony of Septimius Severus (Colonia Septimia
Severa metropolis). Ulpian mentions, after the lyrical passage quoted,93 that
Severus and Caracalla conferred Italian rights (ius Italicum) on Tyre, for its
‘signal and outstanding loyalty to the state and the Roman empire’.94
Caracalla probably became Augustus in the autumn of 197,95 so that this
grant cannot have occurred earlier. It may have been made along with the
grant of colonial status, or perhaps later. Italian right96 carried with it notable
privileges, including a degree of autonomy and exemption from capitation
and land tax (tributum capitis, tributum soli).
It has been said that Elagabal reduced Tyre in rank because of a military
revolt, and that Alexander Severus restored it to its previous position.97 In the
civil war between Macrinus and Elagabal in 218 Sidon, Tyre’s twin town,
sided with Elagabal. While it is not proved that Tyre sided with Macrinus, it
evidently lost some privileges, and is not called a colony or metropolis on
coins of Elagabal apart from the early ones. It seems, then, to have suffered
disgrace, but not immediately on the fall of Macrinus. Whether the disgrace
was connected with a revolt of Legio III Gallica, which, having raised
Elagabal to the throne, later rebelled, is uncertain, since the legion was not
stationed in Tyre itself but in nearby Raphanae. At any rate Alexander
restored Tyre’s privileges, as one would expect given his dependence on
Ulpian in the early part of his reign.98
Ulpian’s connection with Tyre has elicited puzzlement on another point.
Is he identical with the Oulpianos who presides as toastmaster in the
Sophists at Dinner (Deipnosphistae) of Athenaeus?99 The relation between
them has been canvassed in writings about Ulpian and his career. It merits
a brief discussion.100
The Sophists at Dinner is a work in fifteen books in the form of a sympo-
sium, or discussion at dinner. The talk, in turns entertaining and boring, is
about the meaning and derivation of words and the passages in which they
feature in Greek literature. Oulpianos, the toast-master, refuses to eat a dish
before he is satisfied about the etymology of the word used to name or
describe it.101 He is depicted as a learned but bad-tempered pedant, con-
cerned to preserve the purity of Greek from Roman intrusion.102 He pretends
not to understand Latin words or their derivatives.
None of this would point to a connection with the lawyer were it not that
Oulpianos is described as coming from Tyre.103 The date of the Sophists at
Dinner must be later than 192, since in book 12 the author refers to
Commodus, who died in 192, in a disrespectful way. Commodus ‘in our own
times’ sat in his chariot, with Hercules’ club beside him and a lion skin spread
out below, ‘trying unsuccessfully to imitate the God’.104 Some of the particip-
ants in the symposium are drawn from real life. The host Larensis, a Roman
knight, who gives the party in Rome, seems to be P. Livius Larensis, known
from a touching but undated inscription on a tombstone dedicated to him by
his wife.105 Galen the doctor, also a guest, can hardly be other than the famous
anatomist and physiologist who, like the guest, came from Pergamum.106 The
real Galen, born in 129, is now thought to have lived on into his eighties, per-
haps as late as 216.107 At the end of the work Oulpianos dies,108 while nothing
is said of Galen, who is presumably still alive. So the dramatic date can hardly
be later than 216.
Some scholars, beginning with Kaibel, have identified Oulpianos with
Ulpian.109 Others, such as Kunkel, have for good reason dissented.110
Athenaeus could hardly depict Ulpian, who died in 223 or 224, as dying
during Galen’s lifetime. In any case they differ intellectually. Oulpianos,
assuming he is a real person, is a partisan of Greek culture, an Atticist in mat-
ters of language. He is a word-fancier,111 hypercritical,112 and is nicknamed
100 Kaibel (1887) intro: Dittenberger (1903) 1–28; P. Krüger (1912) 239143; Kunkel (1967)
249–51; Crifò (1976) 715–34. The references to Oulpianos in Athenaeus’ text are collected in
Kaibel (1887) 3.564 cf. Jörs (1903) 5.1435. Background in Bowersock (1969); Swain (1996) 43–64.
Editions by Kaibel (1887) and Gulick (1927–41) cf. for technique Mengis (1920)
101 Athenaeus 9.401 cf. Swain (1996) 49–51 102 Athenaeus 3.12f, 3.98c, 8.361
103 Athenaeus 1.1d (Τριος) cf. 8.346c (Σρων) 9.368c (Συραττικ)
104 Athenaeus 12.537; Wenger (1953) 232112 105 CIL VI 2126
106 Mewaldt (1912); PIR2 G 24. He was born in 129 and according to the Suda attained 70 but
in reality lived to some date between 204 and 216: Swain (1996) 368–9, 430–2
107 Nutton (1995) 25–39 108 Athenaeus 15.686c
109 Kaibel (1887) 1.vif.; Mengis (1930) 23 f.; Crifò (1976) 721–2, 734 contra Wenger (1933)
23214
110 Wenger (1933) 23214 ; Kunkel (1967) 251
111 Athenaeus 9.401b: ροντιστ κα λογιστ 112 Athenaeus 14.613c: ιλεπιτιµητ#ς
1. Background and Career 13
tas: ut puta si quis forte chartas sic reliquerit ‘chartas meas universas’ qui nihil aliud quam libros
habebat, studiosus studioso: nemo enim dubitabit libros deberi. namque et in usu plerique libros
chartas appellant) cf. 50.16.60 pr (69 ed: ‘locus’ and ‘fundus’: it is not size but intention—nostra
affectio—that determines whether something is one or the other)
117 Athenaeus 9.368c 118 Syme (1980) 101 = (1984) 1411123
119 Kunkel (1967) 248 120 Kunkel (1967) 253 121 Above n.71–4
14 1. Background and Career
Corbulo, the only Domitius known to have governed the province before
Ulpian’s time. Corbulo had a stepson Annius Vinicianus, so that this second
gentile name, found on the Centumcellae pipe, could go back to the same
period.122 If these clues are reliable, Ulpian’s family had probably been estab-
lished in Tyre for a considerable time. He certainly knew Greek as well as
Latin, and may have known Aramaic. But the attempt to show that his writ-
ing betrays strong Semitic or Greek influence has so far carried little convic-
tion.123 He composes in Latin but has a good command of Greek, as is shown
in many texts where he gives the Greek equivalent of some Latin term or
inserts a brief word or phrase, but never a whole sentence, in Greek. His Latin
is at times colloquial, because he is dictating, but it is clear and correct, closer
to some first-century Latin, for example that of Quintilian, than to third-
century Latin. Cicero and Quintilian seem to have had a strong influence on
his style and thought.
Ulpian must have been born about 170.124 His career can be divided into
three periods. Comparison of his private writings with the rescripts of the
early third century shows that he was drafting rescripts as secretary for peti-
tions (procurator a libellis) or his assistant on behalf of Severus and Caracalla
from at latest August 202 to at earliest May 209. From 213 to 217 under
Caracalla and Macrinus he composed his major restatement of Roman law in
the light of the Antonine constitution of 212, according to a plan explained in
chapters 7–9. In March 222 he was prefect of supply (praefectus annonae) and
in December of that year praetorian prefect (praefectus praetorio), a post that
he held under Alexander Severus until his assassination in 223 or 224. There
are gaps between the first and second period, and between the second and
third. These can be filled to some extent from a combination of evidence and
speculation. But we know nothing about his career before 202.
Frezza supposes him to have had a period practising as an advocate before
202/3.125 This is possible but not proved. It has also been suggested that he
began as assessor to a praetor, or to Papinian, or with a procuratorship in
Gaul. None of these ideas withstands careful scrutiny, if ‘assessor’126 is taken
to refer to a junior paid post, often held by a trained lawyer, of the sort about
which Paul wrote a monograph.127 But if by ‘assessor’ is meant a legal expert
122 Kunkel (1967) 254
123 Heineccius, ‘De Ulpiani Hebraismis’, (1765–71) 2.249; Kunkel (1967) 251; Kalb (1888)
127; Volterra (1937) 3.158–63; Wenger (1953) 519311; Yaron (1987) 3–17
124 Liebs (1997) §424
125 Frezza (1983) 415, citing the reference to popularis meus in D 45.1.70 (11 ed.)
126 Hitzig (1893); Behrends (1969) 192–226; Coriat (1997) 243
127 D 1.22.1 (Paul 1 de off. adsessorum)
1. Background and Career 15
128
e.g. Celsus senior in D 31.29 pr (Cels. 36 dig.). Kunkel (1967) 331697 contra Wenger (1953)
519311; NNDI (1964) 355
129 130 131
D 4.2.9.3 (11 ed.) Ch.8 n.15–21 Jörs (1903) 1438
132
Seeck (1894) 423–6; Kaser (1966) 367; Kunkel (1967) 331 but contra Behrends (1969) 192,
216 citing Seneca, Peace of Mind (De tranq. animi) 3.4
133 134
D 1.22.6 (Pap. 1 resp.) D 34.1.14.3 (2 fid.)
16 1. Background and Career
praetor ( %&τινι συν#δρευσα) but in the context this can only mean ‘advising’
139 D 40.2.5 (Iul. 42 dig.) cf. 22.1.3.3 (Pap. 20 quaest.) 140 Ch.8 n.15–21
141
HA Pescennius Niger 7.3–4 (qui Papiniano in consilio fuerunt); Severus Alexander 26.5 (qui
tamen ambo assessores Papiniani fuisse dicuntur)
142
D 12.1.40 (Paul 3 quaest.)
143
CIL XI 6337: Ti. Claudius Zeno Ulpianus ex sacra iussione adhibitus in consilium praef.
praet. item urbi; Pflaum (1960–1) no.294; Howe (1942) 37
144
Ch. 6 n.18–24
1. Background and Career 17
A wife’s personal effects, not part of her dowry, are called παρερνα by the
Greeks but peculium by the Gauls. Calder points out that peculium is used in
Galatia as well as Gaul.151 None of these passages requires that Ulpian held
a post in Gaul, though as an equestrian he probably held one or more provin-
cial procuratorships early in his career.152 He travelled with Severus in 208
through Gaul on the way to Britain, and came back without him in 211. So
he knew something of Gaul. But he may not have had a close connection with
it. Unlike Paul,153 he does not list anywhere in Gaul as having Italian rights
(ius Italicum).154 The examples of languages other than Latin and Greek,
which in his Trusts (213 or 214) are Punic and Gallic,155 become Punic and
Assyrian (Aramaic) in the later books On Sabinus (216).156 In 214 a testator
speaks of his Gallic estate157 while in 216 a tutor is appointed to a ward’s
‘African or Syrian estate’.158 Caracalla and Julia Domna had moved east in
the interval, and Ulpian may have done the same.159 On the other hand he
may have remained in Rome with its convenient access to libraries.
We do not know how Ulpian’s career began, nor where he studied law, nor
who his teacher or teachers were. At some stage he must have taught law pri-
vately, because he refers to Modestinus as his pupil, studiosus meus.160
Moreover, Marcianus, writing after Caracalla’s death in 217, was clearly his
pupil.161 Ulpian’s Teaching Manual (Institutiones) also points to time spent
teaching. So do his ten books of Disputations (Publicae disputationes), which
involved the public discussion of moot points of law. He will have engaged in
both public and private teaching. The references he makes to special condi-
tions in Arabia,162 Egypt,163 Asia,164 and Africa165 show that he was abreast
of conditions in a number of provinces.166 But he need not have held posts in
those provinces. His upstage way of referring to magistrates and high officials
suggests a man who has not spent much time in humble provincial posts and
has long been associated with the central government. My guess is that
Ulpian assisted Papinian at least for part of the time when the latter was
drafting rescripts for Severus as secretary for petitions (194–202). The secre-
tary for petitions ( procurator a libellis, later magister libellorum) was the
official who dealt with petitions to the emperor that raised points of law, at
any rate those of any difficulty, and suggested a draft reply. The drafts were
often, though of course not always, accepted without modification and in due
course issued as imperial rescripts.167 If Ulpian assisted Papinian in this way
he could have been with Severus for much of his reign, including his visit to
Egypt in 199–200 and to Syria in 201.168 Though the post of assistant to the
secretary for petitions is not attested its existence cannot, given the increased
volume of work in Severus’ reign, be ruled out.
Secretary/assistant for petitions (202–9). That Ulpian held the office of secre-
tary for petitions ( procurator a libellis) is asserted in a muddled passage of the
Augustan History.169 Eutropius170 and Festus171 both say that he held this
office under Alexander. That is impossible,172 since within three weeks of
Alexander’s becoming Augustus Ulpian is already a prefect.173 Nor is the vari-
ant suggested by Syme that Ulpian was secretary to Alexander as Caesar
viable,174 though a Caesar could have a secretary for petitions.175 The HA
account is more plausible if only because by implication it puts Ulpian’s tenure
of the office of petitions in the reign of Severus.176 The passage says that
Pescennius Niger proposed that assessors should administer the departments
in which they had been assessors. Severus and others adopted this policy, as is
162
D 47.11.9 (9 off. proc.)
163
D 47.11.10 (9 off. proc.), 32.55.5 (25 Sab.): see Bonneau (1969)
164 165
D 1.16.4.5 (1 off. proc.) D 34.1.14.3 (2 fid.) contra Bremer (1868)
166
D 3.2.4.2 (6 ed.), 32.55.6 (25 Sab.)
167 Honoré (1994) 43–8. Millar (1992) 648–52 thinks this model too bureaucratic for the
period
168 Coriat (1997) 386–8. Birley (1988) 133 thinks that by 198 both Papinian and Ulpian were
influential. Contrary to the accepted view which treats him as African (Liebs 1997, § 416) Birley
makes Papinian Syrian
169 HA Pescennius Niger 7.3–4: intimavit, ut assessores, in quibus provinciis adsedissent, in
his administrarent. Quod postea Severus et deinceps multi tenuerunt, ut probant Pauli et Ulpiani
praefecturae, qui Papiniano in consilio fuerunt ac postea, cum unus ad memoriam alter ad libellos
paruisset, statim praefecti facti sunt
170 Eutropius 9.33: adsessorem habuit vel scrinii magistrum Ulpianum, iuris conditorem
171 Festus 22
172 Despite Syme (1980) 81 = (1984) 1395. For differing views see Coriat (1997) 261, 559, 570
173 CJ 8.37.4 (31 March 222) 174 cf. Syme (1980) 93 = (1984) 1405
175 CIL XIV.5347 (L. Volusius Maecianus) 176 Jörs (1903) 1437
1. Background and Career 19
177 Liebs (1987b) 111f.; Syme (1980) 86f. = (1984) 1399f. is highly sceptical
178 CJ 2.1.3–3.9.1 (30 Aug 202) 179 CJ 2.3.2 (12 Feb 202)
180 CJ 2.3.3 (25 March 202)
181 In the first edition of Emperors and Lawyers I dated Ulpian’s tenure of office from this
rescript (Honoré 1981, 59–64). In the second edition I postponed it out of caution to CJ 5.66.1,
5 April 203 (Honoré 1994, 81)
182 Honoré (1994) 80 183 Honoré (1994) 62–3
184 CJ 2.1.3–3.9.1 (30 Aug 202): ‘permultum interest’ points strongly to Ulpian: ch.2 n.688–93.
The parallel of authority and equity (auctoritas, aequitas) in 2.1.3 is also telling
185
Syme (1980) 95 = (1984) 1406–7
186
Honoré (1981) 59–64; (1994) 81–91; Birks (1983) 153; Liebs (1997) §424 p.176
187
Honoré (1994) 81 n. 65–6; 131 n.789–90. He may also have drafted Caracalla’s speech to
the senate about donations between husband and wife: below n.240; Birley (1988) 166–7
188 189
Honoré (1994) 86 n.168–9 Liebs (1983a) 496; (1987) 179; (1997) §424 p.176
20 1. Background and Career
Syme,190 has suggested that Ulpian, though drafting imperial rescripts for
Severus, did not hold the office of secretary for petitions for the whole period.
In the first two years, up to the fall of Plautianus in 205, Aelius Coeranus191
was secretary, paid at the 300,000 sesterces rate, and Ulpian was his assistant,
paid at the lower 60,000 sesterces rate. It is, however, possible that Coeranus
was secretary for petitions to Caracalla while Ulpian was secretary to
Severus, since each Augustus will have had his own secretary, though only the
rescripts issued by the senior Augustus were collected as having general
authority.
The important point is that the rescripts from August 202 onwards are in
Ulpian’s style and reflect his vocabulary. Allowance must, however, be made
for the fact that in composing rescripts Ulpian is composing in writing
whereas his works in his own name are dictated and come some years later.192
His literary style, as it emerges from texts in Justinian’s Digest, is discussed in
chapter 2. The parallels between what he wrote as a private author and the
rescripts of April 203 to May 209 were set out in the first edition of this
book193 and in Emperors and Lawyers.194 They will not be repeated here. But
it is important for the understanding of Ulpian’s public role to illustrate one
feature of the rescripts for which he was responsible: the vigour and directness
of their approach, unparalleled in the later principate. This vigour contrasts
with the understatement that Papinian favoured both in his private writing
and in the rescripts he drafted for Severus between September 194 and
February 202.195
A rescript of Papinian’s time can modestly end ‘you will not raise the
defence in vain’ (non inutiliter defenderis).196 Ulpian is forthright. He often
begins a rescript by advising or scolding the petitioner. ‘Approach the gover-
nor’ (Adite praesidem);197 ‘You wrongly believe’ (Falsa suasione credis),198
‘You are asking for something unjust and unusual’ (Neque aequam neque usi-
tatam rem desideras);199 ‘It is unusual and a bad example’ (Insolitum est et
grave exemplo);200 ‘You are asking for something contrary to law’ (Incivilem
rem desideras), 201 ‘You are wrong to be afraid that’ (Frustra times);202 ‘Your
thinking accords with legal principle’ (Secundum rationem iuris existimas);203
‘We do not see why you want’ (Non animadvertimus cur velis);204 ‘You should
know’ (Scire debes).205 He can also criticize judges: ‘The procurators were
wrong in refusing’ (Non recte procuratores noluerunt);206 ‘The judge was too
207: secundum rationem iuris); 7.45.1 (29 May 208: non videtur rationem habere); 8.13.4 (30 May
208: non habet rationem)
222
CJ 6.3.1 (30 Dec 204: nisi necessitas suaserit)
223
CJ 6.3.1 (30 Dec 204); 4.2.1 (1 July 204); 5.15.1 (20 July 204); 5.69.1.2 (12 Oct 205); 8.15.2
(14 Oct 205); 5.62.3 (15 March 206); 8.40.3.1 (16 Aug 208); 7.74.1 (1 May 209) cf. CIL 3.14203.8
224
CJ 6.26.2 (27 July 204: et ratio suadet et divus Marcus pater constituit)
225 226
Millar (1986) 277 CJ 7.62.1 (13 Jan 209), 7.7.41 (1 May)
227 228
CJ 8.18.1 (13 July 209) CJ 7.8.3 (31 Dec 209)
22 1. Background and Career
Why did Ulpian’s term of office end in the summer or autumn of 209?
Presumably because he accompanied Severus and Caracalla on their cam-
paign in Scotland, while Julia Domna and Geta remained behind in York.229
Geta was left in charge of the civil business of the empire, including the issu-
ing of rescripts.230 During the course of the year Geta was belatedly made an
Augustus, eleven years after his brother. His promotion was known in Athens
in October or November 209.231 In 210 Severus, ill with gout, remained in
York while Caracalla again campaigned in Scotland. There were prepara-
tions for another campaign, but Severus died at York on 4 February 211.
On Severus’ death the imperial party returned to Rome. Severus had wanted
both sons to succeed him, and both had their supporters. Papinian, a close
friend of Severus, supported the idea of joint rule, and Ulpian’s successor as
secretary, whose identity is not known, remained in office at least until 26
November 211.232 This was shortly before Geta was lured to a meeting with
their mother, supposedly of reconciliation, and treacherously murdered on 19
or 26 December of that year.233 This secretary (no.3) 234 was presumably a pro-
tégé of Papinian and a supporter of Geta. A rescript of 28 Dec. 211235 perhaps
belongs to a period of overlap with the next secretary. No.3 was replaced at the
beginning of January at the latest,236 probably by Arrius Menander,237 a
lawyer and imperial councillor during the reign of Severus who had written a
treatise on military law and who was clearly Caracalla’s choice. Papinian was
dismissed as praetorian prefect after Severus’ death238 and executed in 212 on
the basis of charges brought by the praetorian guards after Geta’s death.239
Ulpian escaped Papinian’s fate. He sided with Caracalla in good time.
What he wrote in Caracalla’s reign, especially in the early years, treats him in
flattering terms and points to a good relationship between them.240 In 211
229 230
Dio 76.13; Hasebroek (1921) 142–3, 148–9; Gilliam–Mann (1976) Herodian 3.14.9
231
Kienast (1995) 166; Birley (1988) would postpone his promotion to 210, but the change of
secretary for petitions in 209 suggests that 209 is correct
232 Honoré (1994) 86–8 233 Kienast (1995) 166
234 Secretary no.3 in Honoré (1994). The secretaries are listed at pp.190–1
235 CJ 6.44.1 (28 Dec 211) 236 CJ 5.75.1 (5 Jan 212)
237 Lenel (1889) 1.695–700; D 4.4.11.2 (Ulp. 11 ed.); Giuffrè (1974a); Honoré (1994) 88–91;
241
Herodian 4.3.2–3 cf. Alföldy (1972)
242
The principle went back to the XII Tables: Kaser (1971) 695
243
Millar (1986) 274; Liebs (1997) §424 A; below ch.7 n.8–92
244
Mommsen (1904–13) 2.158–9, revived by Frier (1984) 863
245
Peters (1913) cf. Rodger (1983) 393–4
246 247
Frag. Vat. 125, 147, 149; Lenel (1889) 2.899–903: ch. 9 n.13 Honoré (1994) 85–91
248
Lenel (1889) 1.595–620 nos.37, 47, 107, 118, 200: Wolf (1959) 524f.; Liebs (1997) §424 p.179
249
Lenel (1889) 1.881–926 nos.396, 414, 526, 528, 529, 578, 624, 625, 626; Schulz (1961) 277–9;
Wieacker (1960) 341f.; Liebs (1997) §424 p.179
24 1. Background and Career
court moved with its books and files, but is no more than a guess. It is not
clear what office, if any, Ulpian held after ceasing to be secretary for petitions
in 209.
The scenario that best fits his subsequent career, and especially his literary
activity in 213–7, is that he remained close to the imperial family.250 His con-
cern for problems of public administration, on which he wrote more exten-
sively than anyone else, points in the same direction. Like Julia Domna he
was from Syria. A link with the imperial family is indicated by Ulpian’s state-
ment in Taxation (De censibus), probably to be dated 213,251 that he knows
that Caracalla allowed his cousin Julia Mammaea to retain the consular dig-
nity acquired though her first marriage, though she later married a husband
not of consular rank.252 Her second husband was in fact Gessius Marcianus,
father of Alexianus, the future emperor Alexander Severus. Ulpian is not cit-
ing a published rescript. He is reporting an item of inside knowledge. Julia
Mammaea dominated her son Alexianus and came to rely on Ulpian when
her son became emperor in 222. So Ulpian was in some way close to the
Syrian branch of the imperial family. But he was also close to Severus, and
knows that Arrius Menander, was, exceptionally, excused from a guardian-
ship already assumed.253 Menander was one of Severus’ legal counsellors,
later promoted by Caracalla to be secretary for petitions.
Given the outlook that emerges from his writing Ulpian will on theoretical
as well as practical grounds have favoured and, I suspect, advocated the
extension of citizenship that Caracalla effected by the Antonine constitution
(constitutio Antoniniana) of 212.254 That step, though presaged by earlier
developments, was revolutionary.255 As the Tabula Banasitana of 177 shows,
a generation earlier it was a complicated matter, involving the emperor’s
council, to obtain Roman citizenship.256 Up to 212 the customary law of var-
ious ethnic groups, the ius gentium, had been an element on which Roman law
drew for certain purposes, which might include criticism of existing civil law
institutions. Now Roman law became the ius gentium, the law of peoples.
Once that step had been taken, it was crucial to provide for the citizens of the
empire, new and old, a clear exposition of the common law to which they were
subject and in doing so to bring out its universal and rational character. This,
Ulpian, with his deep understanding of the problems of provincial adminis-
tration,257 was uniquely well qualified to do.
250 Syme (1971) 154 251 Ch.9 n.11–26 252 D 1.9.1 pr (2 cens.)
253 D 4.4.11.2 (Ulp. 11 ed.). On Menander see Jörs (1895); Liebs (1997) §419.7; Honoré (1994)
88–91
254 Literature above n.30; Birley (1988) 190: Paul and Ulpian may have recommended this
enactment
255 For the parallel between the constitutio Antoniniana and the extension of citizenship to
non-white South Africans, along with legal implications, see Honoré (1990); Mandela (1993).
For a more conservative interpretation see Sherwin-White (1972).
256 257
Seston–Euzennat (1971) 468–90 Millar (1986) 275
1. Background and Career 25
continued composing it for the five years 213–7. The ideological grounds for
this interpretation will be set out in chapter 3.
There was some evolution in the attitude of Ulpian and other lawyers to
Caracalla over the years when writing his survey of Roman law. From some
time about the middle of 213 he gives Caracalla priority over Severus when he
mentions the two together as authors of a law. It is now ‘our emperor
Antoninus with his father’ rather than ‘the deified Severus and our
emperor’.264 Perhaps the German campaign of 213, as a result of which
Caracalla was acclaimed Magnus Antoninus, points to his increasingly auto-
cratic frame of mind. There is some evidence that as the reign progressed
Ulpian came to feel some constraint. Texts that rest on the author’s personal
experience (with ‘I know’ or the like) belong to 213 or 214 and do not reap-
pear until Caracalla is dead.265 Passages in which Ulpian adopts a con-
descending tone towards emperors, including Caracalla, also come to an end
after 214.266 A text fairly early in 217 points out that, contrary to Cicero’s
view, a defendant to a lawsuit may fail to appear through fear of a tyrant’s
cruelty, enemy attack, or domestic sedition.267 This, which may reflect the
atmosphere of the time, seems to be the only surviving reference in Roman
legal literature to the effect of tyranny on legal obligations.
In the reign of Macrinus, which began on 11 April 217, Ulpian speaks of
Caracalla simply as ‘Antoninus’ or ‘the emperor Antoninus’, and gives
Severus back his priority.268 Despite the assassination of Caracalla, Ulpian
completed his programme of writing for the year 217, either during 217, as I
think most likely, or a few months later. Liebs has argued that he was at some
stage banished by Macrinus, who was defeated by Elagabal on 8 June 218 and
later killed. Ulpian need not have been with the imperial court when he was
writing. Perhaps he moved eastward in Caracalla’s last years,269 but one
could take a text of 216 about a wife’s listing property that does not form part
of her dowry ‘as we commonly see happen in Rome’270 to imply that he stayed
in Rome, where there were good library facilities.
If Ulpian carried through his plan, set out in chapters 7 to 9, by the end of
217, it is not clear that he published anything afterwards. In 1962 I supposed
that Ulpian revised his Taxation (De censibus) 271 under Elagabal. This was
because both Ulpian272 and Paul in his parallel work273 attribute the grant of
Italian right to Emesa to ‘our emperor’ (imperator noster), who in Paul’s case
264 265 266
Ch.7 n.33–70 Ch.9 n.11–23, 63, 108 Ch.9 n.63–4
267 268 269
D 42.4.7.4 (59 ed.) Ch.7 n.88–92 Above n.125–9
270 271
D 23.3.9.3 (31 Sab: ut Romae vulgo fiero videmus) Honoré (1962a) 212
272 273
D 50.15.1.4 (1 cens.) D 50.15.8.6 (Paul 2 cens.)
1. Background and Career 27
274 Gualandi (1963) 2.200–1; d’Ors (1942–3) 42–3; contra Mommsen (1870, 1904–13) 2.16853;
According to the Augustan History, ‘some say that Paul and Ulpian were
made prefects by Elagabal, others by Alexander.’300 The author does not
specify that they were made praetorian prefects. We know, however, that
Alexander made Ulpian praetorian prefect in 222.301 Elagabal may have con-
ferred the same honour on Paul if we suppose that the Julia Paula whom he
married in 219 was the lawyer’s daughter.302 In that case Paul will have been
banished when Elagabal divorced Paula in 220 and Victor’s statement that he
was restored by Alexander becomes plausible. Elagabal made Paul praetorian
prefect and later banished him. He appointed Ulpian, who was junior to Paul,
to a lesser prefecture. Alexander restored Paul, who was certainly alive in his
reign,303 and made him one of his councillors.304 He initially kept or rein-
stated Ulpian as prefect of supply but soon promoted him to the praetorian
prefecture. The career structure secretary (for petitions)–prefect (for supply)
was not unusual.305
On 31 March 222 a rescript states the law ‘according to a reply (responsum)
of Domitius Ulpianus prefect of supply, lawyer, and my friend.’306 ‘Friend’
emphasizes that the person mentioned enjoys the emperor’s confidence. Its
technical meaning is that of a person admitted, or who would if present be
admitted, to greet the emperor at his morning salutation.307 The prefecture of
supply demanded administrative competence.308 Just over eight months later
Ulpian has been promoted praetorian prefect. A rescript of 1 December 222
tells the owner of a warehouse, a victim of breaking and entering, to approach
the provincial governor, who, if he thinks a heavy punishment is called for,
will send the culprit to ‘Domitius Ulpianus, praetorian prefect and my par-
ent.’309 Parens can be used in rescripts both of real parents and of earlier
emperors. Indeed Alexander does just this in a rescript of 3 December 222.310
The term as applied to Ulpian is highly honorific, and supports the idea that
his relation to Alexander was close.
Both offices are also attested by a column in Ulpian’s honour found in
Tyre.311 The promotion from prefect of supply to praetorian prefect was not
unusual.312 Praetorian prefects, especially on the civil side, were mainly
300 HA Severus Alexander 26.5 301 CJ 4.65.4 (1 Dec 222)
302 ILS 477: Iuliae Corneliae Paulae Aug(ustae). There is a marble bust of her in the museum
at Ephesus (Inv. no.1566). Mentioned as possible by Syme (1980) 99 = (1984) 1410 and accepted
by Liebs (1997) §423
303 D 31.87.3 (Paul 14 resp.) 304 HA Severus Alexander 68.1 305 Coriat (1997) 271
306 CJ 8.37.4 (31 March 222: secundum responsum Domitii Ulpiani praefecti annonae iuris con-
(1967).
30 1. Background and Career
313 Jörs (1894); PIR2 A 388; Pflaum (1960–1) no.220; Kunkel (1967) 224; Honoré (1994)
Alexandrum primis diebus. The author has a story along similar lines about the evil ‘Turinus’ at
35.5–36.3
32 1. Background and Career
in power? He may have been murdered as early as 223.326 The argument for
this date depends partly on Dio’s statement that he was killed ‘not long after’
the removal of Flavianus and Chrestus327 and partly, it was supposed, 328 on
the album of Canusium, dated October 223.329 The album mentions among
senatorial patrons of the town T. Lorenius Celsus, M. Aedinius Iulianus,
L. Didius Marinus, and L. Domitius Honoratus. It was argued by Pflaum
that these were the current and recent ex-praetorian prefects. Since Ulpian
was not mentioned, he must have been dead when the album was inscribed.
Salway has, however, shown that the persons concerned were chosen as
patrons because of their ability to advance the interests of Canusium. They
are listed in an order that probably reflects their precedence as members of
Alexander’s council. There is no implication, and in three of the cases no evid-
ence, that they were at the time or had been praetorian prefects.330 The album
has therefore no bearing on whether Ulpian was still praetorian prefect in
October 223.
Another sliver of evidence is that Dio says that during Ulpian’s lifetime a
quarrel arose between the people of Rome and the praetorian guards, who
fought for three days with loss of life on both sides. When the soldiers, who
were getting the worst of it, threatened to set fire to buildings the civilians
reluctantly came to terms with them. These disturbances seem to be those
recorded by the Chronicon Paschale for three successive days and nights in
223.331 Ulpian’s murder must have occurred after these disturbances, and, if
he took steps to discipline the soldiers, may in some way have been connected
with them. The murder must in any case have taken place before the middle
of 224. A papyrus published in 1966 shows that Epagathus, who was largely
responsible for Ulpian’s downfall, and was then sent to Egypt as prefect, was
in office there in May or June 224.332 Consistently with this, the lawyer may
have been assassinated in late 223 or early 224.
It was earlier supposed that Ulpian survived until 228.333 I shared this
view,334 which Xiphilinus’ summary of Dio seemed to support. Dio, excusing
himself for his thin account of the early years of Alexander, explains that, after
an illness, he went to govern his province of Africa, then, on returning to Italy,
was sent as governor first to Dalmatia, then to upper Pannonia. The praetori-
ans complained to Ulpian about Dio, since he ruled the Pannonian troops
with a firm hand and the praetorians feared that they would be subjected to
326
Grosso (1968b); Syme (1970/1984); Coriat (1997) 233; Liebs (1987b) 110–8: Liebs (1997)
§424 p.177
327 Dio 80.2.2
328 Pflaum (1948) 37f.; Modrzejewski–Zawadski (1967) 565–611; Mercogliano (1993) 400–7
329 CIL IX 338; Salway (2000) 330 Salway (2000) 168
331 Chron. Pasch. ann. 223: διανυκτρευσις
332 P.Oxy. 2565: Barns–Parsons–Rea–Turner (1966) 102; Grosso (1968b)
333 Still supported by Bauman (1995)
334 Honoré (1962a) 194, 207. So did virtually all scholars except Howe (1942) 100f.
1. Background and Career 33
rehabilitated.344 A great many rescripts were issued. For the rest, there are
parallels between what Herodian, who does not mention Ulpian, reports and
an Augustan History version that attributes much of what Herodian says to
Ulpian but adapts it to the administrative practice of the late fourth century.
According to Herodian, who presents Alexander’s reign in a pro-senatorial
light, civil business and administration was now put in the hands of men of elo-
quence and those skilled in law. There was a council of sixteen senators.345 The
docile Alexander was encouraged to spend most of the day on judicial
duties.346 In the Augustan History version the emperor always devoted the
afternoon to signing and reading letters in the presence of the heads of the
imperial bureaus for letters, petitions, and records (scrinia).347 He gave
instructions that the heads of the bureaux and learned lawyers who were loyal
to him, of whom Ulpian was then the chief, should examine and order state
business and litigation before the business or lawsuit was submitted to himself.
He issued no constitution without the advice of twenty learned lawyers and at
least fifty men of eloquence, so that the quorum should not be less than that
for a resolution of the senate.348 Herodian says that Julia Mammaea put the
palace under strict guard so that Alexander was isolated from potentially cor-
rupting friends and bad influences. The Augustan History version takes the
form that Alexander insisted on Ulpian being present at private audiences and
when he was sitting as a judge.349 He granted no one a private audience except
Ulpian.350 According to Herodian, Alexander ruled without putting people to
death, even those guilty of serious crimes, and executed no one without
trial.351 A rescript of 11 April 223, which falls in Ulpian’s period of ascend-
ancy, announces a policy of restricting prosecutions for treason and, in par-
ticular, refusing to treat a judge’s disregard of a constitution as treason.352
Appointments were made according to precedent, and on the advice of those
best qualified to judge,353 decency prevailed.354 The Augustan History, citing
Herodian, says that he never put a senator to death.355 The various embellish-
ments that the History adds—for instance that Alexander freed himself from
the influence of eunuchs356 and enjoyed Ulpian’s stimulating conversation at
dinner357—are inventions that can safely be disregarded. But that Ulpian con-
tributed in various ways, mainly but not wholly by his writings, to the ideal of
358 Marotta (2000) 76f. makes a more ambitious attempt to detail Ulpian’s programme of
reform. See also Champlin (1978). Against a leading role for Ulpian see Cleve (1988). On
Alexander see Jardé (1925)
359 Syme (1972) 408f. = (1984) 867f. 360 Syme (1972) 408 = (1984) 868
361 D 1.1.1 pr (Ulp. 1 inst.); below ch.3
36 1. Background and Career
Chrestus. He had Flavianus and Chrestus executed about October 222 and
became sole praetorian prefect. In that capacity he remedied some of the
abuses of Elagabal’s reign and improved the civil administration. Some time
between October 223 and May/June 224 the praetorian troops mutinied, with
Epagathus playing a leading role, and Ulpian was murdered. When attacked
he fled to the palace but Alexander and Julia Mammaea were unable to pro-
tect him.
2
1 Earlier efforts by Pernice (1885), Volterra (1937). For Ulpian’s use of Greek see ch.3
n.119–46
2 Mostly collected in Lenel (1889) 2.379–1200, 1230 nos 5–8. Rescripts drafted by him in
Ulpian (1885) assumed the contrary and Kalb (1890) 135 pointed out that, while Ulpian likes cer-
tain unusual expressions, Paul avoids anything of the sort, so that it is virtually impossible to
assign any text to Paul on grounds of style alone
6 Rodger (1983) 386–8
38 2. The Oral Style
Many of the features that make his style distinctive stem from the fact that,
apart from the rescripts he composed on behalf of the emperor Severus, his
works were in my view virtually all dictated. It is no news that, as Kalb
noticed,7 legal authorship is often closer to speech than writing, but Ulpian is
in this respect exceptional. This may seem surprising, since only an author
with a very clear mind can dictate on the scale on which Ulpian seems to have
done without devoting a great deal of time to correcting the dictated version.
But to compose some 217 books in the short period of five, or at most six,
years, dictation was called for. It would also, given the rush, have been impos-
sible to spend much time revising the text. So what we have is, in my view,
largely what Ulpian dictated. As will be seen, the distinctively Ulpianic turns
of phrase are often those one would expect to come across in speech rather
than writing. Moreover the fact that the same phrases recur again and again
points to oral composition. They make his work unexciting, as Pernice
pointed out,8 but at the same time reassuring. They convey a sense of orderly
progression from topic to topic. If Ulpian had composed in writing or thor-
oughly revised the text dictated his work would be more varied.
One drawback of dictation9 is that it can lead to slips as a result of mis-
hearing. For instance the scribe hears alicui as alii.10 But Ulpian had a clear
idea of what he wanted to say and his train of thought is usually easy to
follow. He has of course some idiosyncrasies: a liking for superlatives and a
preference for the future over the present tense, which is also a mark of the
praetor’s edict.
I. METHOD
7
Kalb (1890) 126; Crifò (1985) 605, who draws attention to Vico’s parallel between the lan-
guage of the comedians and the lawyers
8 9
Pernice (1885) Yaron (1987) 16
10
D 8.5.8.5 (17 ed: alii hactenus facere licet should be alicui hactenus facere licet) cf. 15.2.1.10
(29 ed: recte actum est for recte actum esset); 9.1.1.7 (18 ed: dolore concitatus for dolone concita-
tus; 3.1.1.5 (6 ed: a Carfania for a C. Afrania). All seem to stem from mishearings and there must
be others. I am grateful to Detlef Liebs for these examples. For dictation error in the context of
the Vindolanda Tablets see Bowman–Thomas (1994) no.234: feramus tempestates et hiem (for
etiam) si molestae sint
11
For the view that I pay insufficient attention to interpolations see Crifò (1985) 606–7. But
when suspect expressions are concentrated in particular authors they are unlikely to have been
interpolated by Justinian’s compilers. Suppose these turns of phrase are due to the interpolators.
Then either they will appear at random throughout those texts on which the committee in
2. The Oral Style 39
but their authority was confined to shortening the texts and ensuring that they
read smoothly and agreed with other legal texts. We must be on the lookout
for changes, but not for the licence that a modern editor would take for
granted in preparing a new edition of a law book. The changes detected, for
instance by Kalb,12 in texts attributed to Ulpian, do not distort the picture of
style and outlook that emerges if we take what is attributed to him largely at
face value. When interpolation (interp.) has been suspected I have referred to
a 1927 collection of suspect words and phrases by Citati, though that now
needs updating.13 Texts of Ulpian from outside the Digest, in particular the
rescripts that in my view he composed for Severus and Caracalla in AD 202–9,
are initially disregarded for purposes of the analysis of his style. The argu-
ment for attributing them to Ulpian14 rests in part on how far they resemble
Ulpian’s writings in his own name. It would be question-begging, when fixing
the elements of his style, to assume that he composed them. Moreover they
are instances of written, not oral composition.
For purposes of this study there has to be a way of deciding when a specific
word or phrase has been used. The method adopted will emerge from one or
two examples. Take the phrase parvi refert (it makes little difference). The
phrase is taken to have been used whether the formulation runs parvi refert,
parvi autem refert, parvi enim refert, or parvique refert. These all count as
instances of parvi refert. But the forms parvi referre or parvique referre are not
counted as instances of parvi refert, because they report an opinion in indirect
speech rather than stating it directly. Again, Ulpian’s inversion of the normal
word-order is important. So the phrase et ait (Pomponius etc.) that Ulpian
often uses in giving the view of another author is not taken to extend to et
(Pomponius) ait. Nor does erit dicendum extend to dicendum erit. But the
adjective aequissimus (‘fairest, most equitable’) is taken to include the femin-
ine and neuter forms aequissima and aequissimum, as in aequissimum est (‘it is
most equitable’), since the point being made concerns the author’s fondness
for the superlative form of aequus rather than for any particular inflection of
the word.
Anther point concerns words and phrases that Ulpian attributes to other
authors.15 Are these to count as Ulpian’s or those of the other author? On the
whole Ulpian’s habit is to rewrite in his own words what the other author
question worked, which they do not, or they will represent a deliberate decision to use them in
Ulpian. For that the only half-reasonable explanation would be that they had established that
these were typical Ulpianisms that would make their interpolations seem more authentic. Schulz
(1951) is perversely radical about a text of Ulpian the genuineness of which was confirmed by a
papyrus. Cf. Stroux (1950)
12 Kalb (1888) 68–84. As to other supposed changes he was sometimes wrong, partly through
his belief that all classical authors lived in Rome and partly because accurate means for compar-
ing them were not then available: Kalb (1888) 68 n.1
13 Citati (1927). Fuller references to indivdual texts in Ind. Interp. (1929–35) which also needs
updating
14 Above ch.1 n.192–225; Honoré (1982) ch.8; (1994) 81–6 15 Crifò (1985) 606
40 2. The Oral Style
says.16 Take as an example parvi refert. Ulpian uses parvi refert in 28 Digest
texts against at most two in other authors.17 But in seven texts Ulpian says of
another author such as Pomponius that the other author thinks that so-and-
so makes little difference (e.g. parvi referre putat . . .).18 We should not assume
that in these seven cases the other author really used the phrase parvi refert
rather than something equivalent that Ulpian converts into parvi referre. So
these seven texts, quite apart from the fact that they are in the infinitive form
parvi referre, are not counted as instances where either Ulpian or the other
author has used the phrase parvi refert. On the other hand when Ulpian
attributes the use of the phrase to another author in direct speech, as he does
of Julian in one text, this use can properly be assigned to the author he cites.19
Sometimes Ulpian, instead of rewriting his source, copies another author
without mentioning him.20 In that case the words and phrases are attributed
to both authors, and are not counted as distinctively Ulpianic.
Take, again the superlative aequissimus (‘most equitable’). Ulpian accounts
for 80 Digest mentions of it out of 86, the others coming from Gaius and Paul.
There is also one mention outside the Digest, in Gaius’ Institutes. But in eight
of Ulpian’s 80 texts he attributes the use of aequissimus in indirect speech to
Labeo, Pomponius, Papinian, or Julian. Even if these eight texts were treated
as not being Ulpian’s the use of aequissimus would be overwhelmingly
Ulpianic. But they need not in my view be so treated, since if these four
authors really used aequissimus the word should occur at least once or twice
in texts taken from their works, which it does not. It is rather that Ulpian has
turned what they said (e.g aequum est) into, say, aequissimum est. With nisi
forte (‘unless perhaps’), another favourite, the position is similar. Here
Ulpian has 95 Digest mentions out of 117, but four Ulpian texts read as
though nisi forte were a direct quotation from another author.21 If these are
excluded from the Ulpian count, which is then 91, and added to the citations
directly taken from other authors, which then come to 26, Ulpian still has
more than three-quarters of the surviving texts with nisi forte. This is in gen-
eral enough for the expression to be counted as Ulpianic. But for a reason that
will be explained nisi forte is not taken as a distinctive mark of our author’s
style.
16
There is an example of his doing this with a rescript of Pius in D 49.1.14 pr compared with
D 49.1.5.1 (Marci. 1 appell.); Yaron (1987) 7
17
Below n.201
18
D 4.8.7 pr (13 ed.); 4.9.1.7 (14 ed.); 14.4.5.7 (29 ed.); 7.1.12.4 (17 Sab.); 38.1.7.1 (28 Sab.);
47.2.7.3 (41 Sab.)
19
D 34.3.5.2 (23 Sab: et parvi, inquit, refert <Iulianus>)
20
e.g. D 37.8.1 pr (Ulp. 40 ed.) = D 38.6.5 pr (Pomp. 4 Sab.); D 50.16.177 (Ulp. 47 Sab.) = D
50.17.65 (Iul. 54 dig.); Crifò (1985) 606
21
D 6.1.1.2 (16 ed: Pomponius); 17.2.63.3 (31 ed: Marcellus); 19.1.11.18 (32 ed: Iulianus);
37.10.1.8 (41 ed: Pomponius)
2. The Oral Style 41
the main part of the sentence. The compilers did not take the easy course of rewriting the main
text because they were not entitled to do so
44 2. The Oral Style
is that interpolation by Justinian’s compilers could not account for the con-
centration of particular suspect words or phrases in the work of a particular
author.35 All nine Digest texts with si mihi proponas (‘if you put it to me’) are
attributed to Ulpian. Eight of them have been suspected of interpolation.36
Surely the phrase, though suspect in another author, cannot be thought alien
to Ulpian? It could only have been introduced into the Ulpian texts alone by
a post-classical editor of his work. But there is no solid evidence that his work
was re-edited in post-classical times,37 though sporadic glosses or comments
on the text were incorporated in it when copied. Even if it was re-edited, why
should the editor introduce this phrase? I have occasionally drawn attention
to a text that is clearly interpolated or attributed to the wrong author, but
there are not enough instances of this sort to cast doubt on the overall picture
of Ulpian’s style that the Digest texts present.
What, then, are the main features of the writings of the scholar from Tyre
who has left so strong a mark on the law of Europe and on countries influ-
enced by European trade, colonization, or settlement? His style is clear, sim-
ple, consistent, and personal. It is an oral style, of which the only other clear
example in Roman legal literature is Gaius’ Institutes. His writing betrays the
fact that he is dictating. He links sentences and phrases so as to make the
argument flow, even if this means, as it often does, beginning a sentence with
‘and’ (et).38 If the very ordinary Latin phrases proinde, proinde si, and proinde
et si occur 214 times in his work and at most six in that of other authors it is
because the linking phrases ‘and so’, ‘and so if’, ‘and so even if’ come natur-
ally when one is talking but are much less natural in written composition. The
Latin magis est means ‘the better view is’, a phrase much used by lawyers,
including Roman lawyers. But et magis est (‘and the better view is’) is almost
monopolized by Ulpian, who has 88 texts out of 102 with this phrase. The
reason, again, is that in speech it is natural to add the et, but in writing one
inserts a pause, a real or notional full stop, before telling the reader which
view is better. Again, quid tamen si (‘but what if’), a natural phrase in speak-
ing, but less at home in a text composed in writing, is found in Ulpian 42 times
out of 43 instances.39
Consistently with this oral method, Ulpian prefers short to long words and
short clauses to complex syntax. His prose is forceful, direct, and forward-
looking. As in the praetor’s edict, the future tense is favoured. A contract ‘will
be’ (rather than ‘is’) valid. He takes pain to avoid clusters of verbs at the end
of a sentence, especially verbs drawn from different clauses. So he sometimes
puts the direct object after the main verb, and often the participle after the
auxiliary. His prime aim is to be clear, and the reader is seldom in doubt what
35 36
Kalb (1890) 133–4 Below n.223; Crifò (1985) 606
37
Daube’s argument for re-editing (1973) is based on the unfounded prejudice that Paul and
Ulpian could not have cared about keeping slave families together
38 39
Below n.48–102 Below n.151
2. The Oral Style 45
he means. Lucidity comes at a cost to elegance and variety. His is a plain Latin
prose in which familiar phrases recur. His frequent references to himself, as in
ego puto (‘I think’), also stem in part from dictating, since in speaking it is
more difficult to avoid referring to oneself than in writing. As Birks pointed
out,40 it was a mistake on my part to accuse him in the first edition of ego-
centricity. But in any case Ulpian, though respecting the opinions of other
authors, is self-confident. From this point of view his tone echoes that of the
lawyers of the age of Trajan and Hadrian, for example Iavolenus and Celsus
the younger. Not for him the muted reflections of his elder contemporary
Papinian, who never uses the word ego. A candid writer, Ulpian makes a
point of stating his opinion, even if hesitantly (‘I should think’, putem; ‘I
believe’, credo). Much of his writing, not only in Disputations (Publicae
Disputationes), which record moot points, reads like a debate between him-
self, his colleagues, and by implication the reader: ‘you may say’ (dicas); ‘sup-
pose’ (pone, finge), ‘you should take it’ (accipias). Ulpian regularly gives
reasons, often rather summary ones, for his conclusions.
In listing the words and phrases that satisfy the tests described and count
as marks of Ulpian’s style I have mentioned or referred to all the instances of
the word or phrase to be found in the Digest in Mommsen’s reading of the
text. Ulpian’s texts are listed first. They are classified according to the work
they come from. First come the major treatises On the Edict (Ad Edictum; ed.)
and On Sabinus (Ad Sabinum; Sab.), then the medium-scale works The
Proconsul (De Officio Proconsulis; off. proc.), Disputations (Publicae
Disputationes; disp.), Tribunals (De Omnibus Tribunalibus; omn. trib.), and
The Julio-Papian Law (Ad legem Iuliam et Papiam; leg. Iul. Pap.). Lastly the
minor works are listed in alphabetical order. Then may come ‘cf.’ followed by
instances of the expression or related expressions from legal sources outside
the Digest attributed to Ulpian. When there are Digest texts from other
authors there follows the word ‘but’ and instances of the word or phrase from
texts of the other authors. The reader can in this way count and, if necessary,
look up the legal texts deriving from Ulpian and the other writers. When there
are thirty or more texts of Ulpian, so that to cite them would be to overload
the footnotes, I have usually referred the reader to the relevant fiche in the
Concordance to the Digest Jurists (1980),41 published on microfiche, instead
of setting them out in a footnote. The texts in the fiches are in alphabetical
order, so that, for example, to find instances of Ulpian’s use of proinde ac si
one looks at fiche 67. This contains Ulpian’s Digest texts from POSSESSIONEM
to PROPRIUS. Within the fiche one looks for the texts listed under PROINDE.
These are set out alphabetically according to the word following PROINDE, so
that PROINDE AC SI comes near the beginning of the PROINDE texts. Those schol-
ars who do not have access to Honoré–Menner can find most of the relevant
(i) Conjunctions and other introductory phrases. Ulpian likes to link sen-
tences. He begins each sentence but the first with a conjunction or transitional
phrase. He does not like unlinked sentences, anacoloutha. A conjunction that
is virtually confined to Ulpian is proinde, which means ‘hence’. It is used either
on its own or in combinations such as proinde si, proinde et si, and proinde
tamquam si.42 All sixty-four instances of proinde et si in the Digest are from
Ulpian,43 as are eighty-five out of eighty-eight with proinde si 44 and sixty-five
of sixty-eight in which proinde is a conjunction on its own.45 In all, then, the
Digest yields 214 texts of Ulpian in which proinde means ‘hence’ and only six
from other authors. A remarkable concentration. Nor are all the other six
texts real exceptions. One is probably due to Justinian’s compilers,46 and a
second may also be compilatorial.47 The use of this word in legal writing, used
to draw a conclusion from what precedes, is virtually an Ulpian fingerprint.
He often uses et (‘and’) to link sentences or clauses. This, a notable feature
of his prose, points to the habit of dictating. Combinations of et and putare
(‘to think’) furnish many examples. All thirty-one Digest texts with et putem
(‘and I should think’) are attributed to Ulpian.48 So are eight of nine texts
42
Proinde can also mean ‘just (as)’ or ‘just (as if)’, as in proinde ac or atque. This is a use of
which Gaius, Inst. has many examples. It is not Ulpianic: VIR 4 (1) 1222.23–31
43
Honoré–Menner (1980) fiche 67 PROINDE ET SI; VIR 4.1222.31–44
44
Honoré–Menner (1980) fiche 67 PROINDE SI; but D 34.5.13.4 (Iul. 1 ambig.); 9.2.22 pr (Paul
22 ed.); 13.7.16.1 (Paul 29 ed.) cf. Gaius, Inst. 2.79. D 47.2.21.8 (‘Paul’ 40 Sab.) is an Ulpian text
45
Honoré–Menner (1980) fiche 67 PROINDE; VIR 4.1222.8–19 but D 12.1.8 (Pomp. 6 Plaut);
33.1.21.4 (Scae. 22 dig.); 48.19.28.6 (6 cogn.) cf. Gaius, Inst. 3.14 but interp. Citati (1927) 70:
Vassalli, Schulz
46 D 9.2.22 pr (Paul 22 ed.), intended to link the Paul text to the preceding Ulpian text
47
D 13.7.16.1 (Paul 29 ed.), intended to link what follows to the correct but compilatorial
statement ‘contrariam pigneraticiam creditori actionem competere certum est’
48
D 19.1.13.3 (32 ed.); 25.4.1.9 (34 ed.); 27.9.5.4; 27.9.5.12 (35 ed.); 47.2.52.6 (37 ed.); 39.3.6.2
(53 ed.); 40.12.18 (55 ed.); 42.3.6; 42.6.1.8, 16 (64 ed.); 43.5.3.15 (68 ed.); 42.8.10.5 (73 ed.); 44.2.9
pr (75 ed.); 44.4.4.24; 44.6.1.1 (76 ed.); 36.3.1.11 (79 ed.); 38.4.3.5 (14 Sab.); 23.3.9.3 (31 Sab);
24.1.32.28 (33 Sab.); 23.3.38 (48 Sab.); 18.4.2.10; 45.1.38.12 (49 Sab.); 48.2.7.2 (7 off. proc.);
2. The Oral Style 47
with et magis puto (‘and I rather think’),49 nineteen of twenty with et non
puto,50 all eleven with et ego puto51 and all three with et verum/verius puto.52
Even plain et puto occurs in his work 112 times out of 126.53 Fourteen
instances come from other lawyers,54 and one of these, attributed to Paul, is
really a text of Ulpian.55 Nor is this all: et sunt qui putent56 and et putamus57
are also confined to Ulpian.
So are many other phrases in which et is the first element in the phrase con-
necting a sentence with the previous one. One of the criticisms of the first edi-
tion was that sometimes there are only one or two instances of the phrases
introduced by et that I cite.58 Often, of course, there are more. But the crucial
point is that among Roman lawyers the overwhelming bulk of these et phrases
come from Ulpian, as VIR confirms.59 In some of these et introduces a verb: et
(magis) arbitror,60 et constat,61 et credo,62 et dicendum erit,63 et dico,64 et non
dubito,65 et eveniet,66 et exstat/extat,67 et finge,68 et intererit,69 et magis est,70 et
50.13.1.4; 50.13.1.15 (8 omn. trib.); 48.5.2.5 (8 disp.); 40.5.24.16, 17; 40.5.26.6 (5 fid.); 48.5.26.4
(2 adult.); 49.4.1 pr (1 appell.)
49
D 25.4.1.8 (34 ed.); 26.7.5.5 (35 ed.); 27.4.1.5 (36 ed.); 1.5.10 (1 Sab.); 24.1.21 pr (32 Sab.);
24.1.32.14 (33 Sab.); 48.19.9.14 (10 off. proc.); 25.3.5.1 (2 off. cons.) but 32.64 (Afr. 6 quaest.) cf.
19.1.13.9 (Ulp. 32 ed: et puto magis)
50
D 2.2.3.4 (3 ed.); 4.4.3.4 (11 ed.); 5.3.25.6 (15 ed.); 50.14.2 (31/1 ed.); 49.17.8 (45 ed.);
29.4.1.5 (50 ed.); 47.9.3.7 (56 ed.); 42.6.1.1 (64 ed.); 28.3.6.10 (10 Sab.); 32.52.5 (24 Sab.); 23.3.9.3
(31 Sab.); 23.2.29 (36 Sab.); 26.1.6.4 (38 Sab.); 47.2.39; 47.2.43.8 (41 Sab.); 42.2.6.4 (5 omn. trib.);
5.1.52.4 (6 fid.); 40.9.14 pr (4 adult.); 42.1.15.7 (3 off. cons.) but 43.24.21.1 (Pomp. 29 Sab.)
51
D 3.5.3.11 (10 ed.); 9.2.11.10 (18 ed.); 29.5.5.1 (50 ed.); 47.8.2.27 (56 ed.); 43.19.3.16 (70 ed.);
44.4.4.23, 29 (76 ed.); 7.1.13.3, 5 (18 Sab.); 34.2.27.1 (44 Sab.); 18.4.2.3 (49 Sab.)
52
D 35.1.10 pr (23 Sab.); 1.14.3 (38 Sab.); 41.1.33 pr (4 disp.)
53
Honoré–Menner (1980) fiche 56 ET cf. Kalb (1890) 131 who points out that one of the texts
attributed to Gaius (D 18.6.2.1) is really from Ulpian
54
D 16.1.32.2 (Pomp. 1 SCC); 29.4.3 (Pomp. 3 Sab.); 18.6.2.1(Gaius 2 rer. cott.); 17.1.49
(Marc. 6 dig.); 3.4.6.2 (Paul 9 ed.); 4.8.32.16 (Paul 13 ed.); 18.5.3 (Paul 33 ed.); 22.1.38.14 (Paul
6 Plaut.); 42.1.51.1 (Paul 2 man.); 34.9.5.5 (Paul 1 fisc.); 46.1.71 pr (Paul 4 quaest.); 31.69.4 (Pap.
19 quaest.); 37.8.7 (Tryph. 16 disp.); 48.18.10.2 (Arc. Char. 1 test.)
55
D 3.4.6.2 (Paul 9 ed. = Ulp. 9 ed.), below n.827–8
56 57
D 41.2.13.2 (72 ed.); 36.1.17.3 (4 fid.) cf. 40.5.24.19 (5 fid.) D 33.8.6.2 (25 Sab.)
58 59
Watson (1983) but see Birks (1983) 180 VIR 2.530–536 et (praecedit controvesia)
60
D 13.7.24.2 (30 ed.); 18.3.4.4; 19.1.11.18 (32 ed.); 27.3.1.20 (36 ed.); 37.4.3.9 (39 ed.);
38.2.12.4 (44 ed.); 33.7.12.14 (20 Sab.); 32.70.12 (22 Sab.); 29.1.19.2 (4 disp.); 48.5.4.2 (8 disp.);
36.1.18.7 (2 fid.) cf. Kalb (1890) 131
61
D 13.7.24.1; 16.3.7.3 (30 ed.); 24.3.24 pr (33 ed.); 28.2.1 (1 Sab.); 45.1.29.1 (46 Sab.)
62 63
D 28.3.6.6 (10 Sab.); 32.11.21 (2 fid.); 17.1.29.3 (7 disp) D 5.3.25.9 (15 ed.)
64 65
D 38.17.1.10 (12 Sab.) D 25.1.5.1 (36 Sab.); 47.3.2 (42 Sab.); 39.5.7.2 (44 Sab.)
66
D 37.6.1.19 (40 ed.)
67
D 27.7.4 pr; 27.8.1.2 (36 ed.); 39.2.15.12 (53 ed.); 43.20.1.27 (70 ed.); 43.24.11.1, 8 (71 ed.);
48.18.1.27 (8 off. proc.); 40.5.30.6 (5 fid.); 21.1.8 (1 ed. cur.)
68 D 5.1.18.1 (23 ed.) 69 D 18.6.4 pr (28 Sab.)
70 Eighty-eight of 104 texts: Honoré–Menner (1980) fiche 56 ET MAGIS EST but D 30.24.3
(Pomp. 5 Sab.); 33.7 (Pomp. 8 Q. Muc.); 39.3.22 (Pomp. 10 var. lect.); 45.1.137.2 (Ven. 1 stip.);
2.14.30.2 (Gaius 1 ed. prov.); 46.3.72.4 (Marc. 20 dig.); 31.28 (Marc. 29 dig.); 6.1.6 (Paul 6 ed.);
4.2.21.4 (Paul 11 ed.); 11.3.14.2 (Paul 19 ed.); 16.2.9.1 (Paul 32 ed.); 43.24.6 (Paul 67 ed.);
23.3.56.3 (Paul 6 Plaut.); 32.89 (Paul 6 Iul. Pap.); 20.1.16.8 (Marci. 1 form. hyp.); 38.4.9 (Mod. 9
pand.). D 47.2.21.5 (‘Paul’ 40 Sab.) is an Ulpian text
48 2. The Oral Style
71
D 7.8.6 (17 Sab.)
72
D 4.2.9 pr (11 ed.); 12.4.3.7 (26 ed.); 15.3.1.2 (29 ed.); 29.4.1.1 (50 ed.); 40.12.8.2 (55 ed.);
43.20.1.21 (70 ed.); 43.24.1.2 (71 ed.); 44.4.4.1 (76 ed.); 7.4.1 pr (17 Sab.); 34.3.5.2 (23 Sab.);
40.7.3.1 (27 Sab.); 47.2.3.1 (41 Sab.); 45.2.3 pr; 46.1.6.1 (47 Sab.); 28.5.35.1 (4 disp.); 36.1.3.2 (3
fid.) cf. 26.7.54 (Tryph. 2 disp: et multum refert)
73 74
D 35.1.92 (5 fid.) D 48.13.7 (7 off. proc.)
75
D 3.1.1.10 (6 ed.); 3.5.5.8 (10 ed.); 14.5.4.4; 15.2.1.7 (29 ed.); 25.6.1.10 (34 ed.); 37.4.1.7 (39
ed.); 38.5.1.22 (44 ed.); 43.24.1.5 (71 ed.); 10.2.49 (2 disp.); 36.1.1.17 (3 fid.)
76 77 78
D 18.4.2.6 (49 Sab.) D 36.1.1.13 (3 fid.) D 13.6.5.11 (28 ed.)
79 D 11.5.1.3 (23 ed.); 21.2.4 pr (32 ed.); 43.24.13.5 (71 ed.)
80 81
D 47.10.13.7 (57 ed.); 40.5.24.5 (5 fid.) D 18.4.2.7 (49 Sab.)
82 83 84
D 47.5.1.2 (38 ed.) D 19.2.19.2 (32 ed.) D 2.14.1.2 (4 ed.)
85 86
D 7.9.9 pr (51 ed.) D 42.4.7.2 (49 ed.)
87 88
D 43.29.3.12 (71 ed.) D 43.26.1.3 (1 inst.)
89
D 43.26.6.4 (71 ed.); 43.8.2.37 (76 ed.); 49.4.1.5 (1 appell.)
90
D 2.13.6.9 (4 ed.) but interp. Citati (1927) 1: Gradenwitz and others
91
D 21.1.38.4 (2 ed. cur.) 92 D 21.1.23.9 (1 ed.cur.)
93
D 14.1.1.5 (28 ed.); 42.4.3.1 (49 ed.)
94
D 32.55.7 = 50.16.167 (25 Sab.); 40.5.24.2, 17 (5 fid.)
95
D 7.9.3.4 (79 ed.); 26.2.10.2 (36 Sab.)
96
Honoré–Menner (1980) fiche 56 (forty-four texts) but D 9.4.2.8 (Afr. 6 quaest.); 21.2.24
(Gaius 1 ed. cur.); 24.2.6 (Iul. 62 dig.); 21.1.41 (Paul 2 ed. cur.); 28.2.28.3 (Tryph. 20 disp.)
97 D 15.3.3.2 (29 ed.); 30.71.1 (51 ed.); 7.1.25.5 (18 Sab.)
98 D 4.8.21.11 (13 ed); 10.2.8.1 (19 ed.); 39.9.1.1 (41 ed.); 29.4.4.1 (50 ed.); 47.8.4.3 (56 ed.);
42.4.7.7 (59 ed.); 43.24.13.5 (71 ed.); 36.3.1.13 (79 ed.); 7.6.1.3 (18 Sab.); 40.4.13.2 (5 disp.)
99
D 43.29.3 pr (71 ed.); 44.5.1.12 (76 ed.); 45.1.1.3 (48 Sab.)
100
Honoré–Menner (1980) fiche 56 ET AIT
101
D 39.6.31.3 (Gaius 8 ed. prov.); 8.3.35 (Paul 15 Plaut.); 23.2.60.4 (Paul 1 orat. Ant. et
Comm.); 30.114.3 (Marci. 8 inst.); 29.5.15 pr (Marci. 1 del.)
102
D 23.2.60.4 (Paul. 1 orat. Ant. et Comm: below n.832)
2. The Oral Style 49
To return to linking phrases without et: nec non (‘as well as’)103 and nec non
et,104 long ago picked out by Kalb as Ulpianic,105 are linking phrases meaning
‘and’. Per contrarium (‘on the other hand’), which occurs twenty-nine times in
the Digest in texts of Ulpian and twice in Marcianus,106 an author whose style
was influenced by Ulpian’s, is used to make a contrast.107 There are other link-
ing phrases confined or almost confined to Ulpian’s writings. Some, drawing
a conclusion from what has gone before, have much the same meaning as
proinde: idcircoque,108 eapropter,109 inde,110 sic deinde.111 Others introduce
provisos or limitations on what has preceded: dummodo non,112 dummodo
sciat/sciamus.113 There is adversative force in verum tamen or verumtamen,114
103
Nineteen of 22 texts: D 4.6.26.4 (12 ed.); 11.6.7.2 (24 ed.); 14.3.5.17 (28 ed.); 15.3.3.6 (29
ed.); 27.9.3.5; 27.9.5.10 (35 ed.); 36.4.3.1 (52 ed.); 28.8.3 (60 ed.); 43.3.1.3 (67 ed.); 36.3.1.8 (79
ed.); 29.2.5 pr (1 Sab.); 7.8.10 pr (17 Sab.); 7.1.13.2 (18 Sab.); 48.19.9.1 (10 off. proc.); 24.3.22.6
(3 disp.); 26.5.6 (8 omn. trib.); 50.16.141 (8 Iul. Pap.); 49.14.16 (18 Iul. Pap.); 25.3.5.4 (2 off.
cons.) but 45.1.58 (Iul. 54 dig.); 37.1.6.1 (Paul 41 ed.); 49.1.16 (Mod. 6 diff.)
104
Nine of 11 texts: D 3.2.23 (8 ed.); 37.15.7.4 (10 ed.); 4.6.1 pr; 4.6.28 pr (12 ed.); 43.16.1.2;
50.16.60 pr (69 ed.); 36.3.1.8; 43.16.1.2 (79 ed.); 47.2.12.2 (29 Sab.) but 2.14.9 pr (Paul 72 ed.);
49.16.3.17 (Mod. 4 poen.)
105
Kalb (1890) 132–4
106
D 23.4.6 (4 ed.); 2.4.8.1 (5 ed.); 4.4.22 (11 ed.); 8.5.8.5 (17 ed.); 17.2.52.18 (31/2 ed.);
19.1.13.5, 25 (32 ed.); 25.3.1.11, 13 (34 ed.); 26.7.9.9; 27.3.1.2 (36 ed.); 37.9.1.7 (41 ed.); 39.3.1.21
(53 ed.); 28.1.20.2 (1 Sab.); 29.2.30.5 (8 Sab.); 7.2.8; 7.8.4.1 (17 Sab.); 33.3.1.10 (19 Sab.); 33.8.6.4
(25 Sab.); 23.2.12.2 (26 Sab.); 8.2.17.1 (29 Sab.); 24.1.11.5 (32 Sab.); 47.2.46.8 (42 Sab.); 45.1.1.6
(48 Sab.); 48.22.7.12 (10 off. proc.); 3.3.28 (1 disp.); 48.19.1.2 (8 disp.); 2.15.8.25 (5 omn. trib.);
21.1.23.9 (1 ed. cur.) but 36.1.34 (Marci. 8 inst.); 34.9.2.2 (Marci. 11 inst.) cf. Kalb (1890) 130
107
Ch.6 n.71–2, 80–1
108
Seventeen of 19 texts: D 3.2.13.7 (6 ed.); 17.1.14 pr (31/1 ed.); 26.7.9.6 (36 ed.); 37.4.3.4 (39
ed.); 37.10.5.1; 37.10.3.7 (41 ed., twice); 38.5.3 pr (44 ed.); 39.3.4.2 (53 ed.); 42.1.4.1 (58 ed.);
43.16.1.24 (69 ed.); 43.20.1.25 (70 ed.); 43.24.15 pr (71 ed.); 47.11.7 (9 off. proc.); 28.3.12 pr (4
disp.); 40.9.12. pr (4 adult.); 25.3.5.8 (2 off. cons.), but 47.2.71 (Marc. 8 dig.); 23.2.57 (Marci., cit-
ing divi fratres)
109
D 3.1.1.1 (6 ed.); 10.3.4.4 (19 ed.); 28.7.8.2 (50 ed.); 47.10.17.9 (57 ed.); 40.7.9 pr (28 Sab.);
39.4.14 (8 disp.)
110
Honoré–Menner (1980) fiche 60 INDE (thirty texts where inde means ‘hence’) but D 47.2.1.1
(Paul 39 ed.); 26.7.37.2 (Pap. 11 quaest.); 20.1.16.6 (Marci. 1 hyp.); 35.1.52 (Mod. 7 diff.); cf.
Gaius, Inst. 2.218; Frag. Vat. 90 (Ven 1 interd.?) cf. VIR 3.698
111
D 45.1.72.2 (20 ed.); 11.7.31 pr (25 ed.); 15.1.1 pr (29 ed.); 48.22.6.1 (9 off. proc.);
21.1.25.10 (1 ed cur.); 44.3.5.1 (3 disp.); 17.1.29.3 (7 disp.); 36.1.6.2 (4 fid.); 35.1.92 (5 fid.)
112
D 50.1.25 (1 ed.); 16.3.7.2 (30 ed.); 29.4.6 pr (50 ed.); 43.22.1.8, 9 (70 ed.); 7.4.5 pr (17 Sab.);
47.20.3.2 (8 off. proc.); 48.18.1.13 (8 off. proc.); 50.13.1.10 (8 trib.); 22.1.33 pr (1 off. cur. reip.)
but 17.1.60.2 (Scae. 1 resp.)
113
D 9.3.5.5 (23 ed.); 16.3.1.10 (30 ed.); 27.9.7.4 (35 ed.); 42.3.6 (64 ed.); 43.16.3.9 (69 ed.);
46.2.2 (4 Sab.); 46.4.6 (47 Sab.); 1.16.4.2 (1 off. proc.); 47.17.1; Collatio 7.4.1 (8 off. proc.);
21.1.1.2 (1 ed. cur.) but interp. Citati (1927) 32: Beseler
114
Twenty-nine of 33 texts: D 17.2.63.8 (31/2 ed.); 19.1.11.5, 18; 19.1.17.7 (32 ed.); 4.4.49 (35
ed.); 27.2.2.2 (36 ed.); 28.1.22.4 (39 ed.); 37.5.8.2; 37.5.10.2 (40 ed.); 38.5.1.15 (44 ed.); 38.2.14.4
(45 ed.); 38.6.1.3 (46 ed.); 38.11.1.1 (47 ed.); 42.8.10.3 (73 ed.); 44.4.4.33; 44.5.1.10 (76 ed.);
35.3.1.12; 36.3.1.19, 20 (79 ed.); 33.4.1.10 (19 Sab.); 41.1.20.1 (29 Sab.); 25.1.7 (36 Sab.); 29.1.28
(36 Sab.); 9.2.41 pr (41 Sab.); 47.2.46.8 (42 Sab.); 28.5.35 pr; 28.5.35.3 (4 disp.); 37.14.16 pr (10
Iul. Pap.); 48.5.10.2 (4 adult.) but 48.10.6 pr (Afr. 3 quaest: verum testamentum, Mommsen);
33.1.19.1 (Scae. 17 dig: velim, Mommsen) 22.5.13 (Pap. 1 adult.); 20.6.8.7 (Marci. 1 hyp.); interp.
Citati (1927) 91: Beseler
50 2. The Oral Style
115 116
D 14.6.7.11 (29 ed.) D 17.2.33 (31/2 ed.); 21.1.1.10 (1 ed. cur.)
117
D 24.3.24.5 (33 ed.); 25.4.1.11 (34 ed.); 27.2.1.2, 26.7.3.3 (35 ed.); 11.1.16 pr, 1 (37 ed.);
38.5.1.6, 21 (44 ed.); 37.9.7.2 (47 ed.); 29.5.1.13 (50 ed.); 30.71.5 (51 ed.); 47.2.17.2 (39 Sab.);
47.1.1 pr (41 Sab.); 39.5.7.4 (44 Sab.); 26.5.12.1 (3 off. proc.); 26.5.8.1 (8 omn. trib.); 23.2.27(3
Iul. Pap.) but 22.3.9 (Cels. 1 dig.); 47.2.21.10 (Paul 40 Sab. = Ulp., Mommsen); interp. Citati
(1927) 12: Beseler and others
118
D 17.2.63 pr (31/2 ed.); 29.3.8 (50 ed.); 39.3.4 pr (53 ed.); 42.6.1.3 (64 ed.); 47.2.7.1 (41
Sab.); 47.2.46 pr (42 Sab.)
119
D 47.2.46 pr (42 Sab.)
120
D 46.7.5.4 (77 ed.); 30.53.2 (25 Sab.); 8.4.6.2 (28 Sab.); 26.8.5.4, 6 (40 Sab.) but 40.5.26.7
(Ulp. 5 fid. citing Trajan); 45.2.12.1 (Ven. 2 stip.)
121 122
D 28.5.35.3 (4 disp.); 49.14.29 pr (8 disp.) D 10.3.7.13 (20 ed.); 9.4.35 (41 Sab.)
123
D 26.7.3.2 (35 ed.); 43.20.1.13 (70 ed.); 28.6.2.4 (6 Sab.); 33.7.12.2 (20 Sab.)
124
D 26.4.5.3 (35 ed.); 43.20.1.13 (70 ed.); 48.18.1.27 (8 off. proc.)
125
D 26.7.3.6 (35 ed.); 26.7.7.12 (35 ed.); 39.3.1.23 (53 ed.); 32.49.4 (22 Sab.); 38.1.9.1 (34
Sab.) cf. Collatio 12.7.8 (18 ed.)
126
Twenty-four of 26 texts: D 5.3.11 pr (15 ed.); 6.1.1.3 (16 ed.); 6.2.7.13 (16 ed.); 37.9.7.1 (47
ed.); 40.12.7.2, 3 (54 ed.); 47.9.3.3 (56 ed.); 42.5.24.2 (63 ed.); 43.19.3.4 (70 ed.); 36.3.1.19 (79 ed.);
46.6.4.3; 35.2.47 pr (79 ed.); 35.1.10 pr (23 Sab.); 26.8.5.2 (40 Sab.); 47.1.1 pr (41 Sab.); 39.5.7.5
(44 Sab.); 50.4.6.5 (4 off. proc.); 47.18.1.2; 48.18.1.3 (8 off. proc.); 48.19.9.11 (10 off. proc.);
21.1.38.7 (1 ed. cur.); 39.5.12 (3 disp: sed etenim, Mommsen); 36.1.23.3 (5 disp.); 49.14.29 pr (8
disp.) cf. Collatio 14.3.2 (9 off. proc.) but D 34.3.20.1 (Mod. 10 resp.); 45.1.63 (Afr. 6 qu: etenim,
Mommsen)
127
D 6.1.9 (16 ed.); 38.9.1.6 (49 ed.); 40.5.4.5 (60 ed.); 43.16.1.35 (69 ed.); 24.3.2.2, 37.4.17 (35
Sab.); 21.1.37 (1 ed. cur.) but 28.6.23 (Pap. 6 resp.)
128
D 44.4.4.23 (76 ed.)
129
D 25.6.1.11 (34 ed.); 37.6.1.2 (40 ed.); 47.8.2.10 (56 ed.); 40.5.4.3, 14 (60 ed.)
130
Thirteen of 14 texts: D 13.7.9.1 (28 ed.); 26.10.3.2 (35 ed.); 26.7.9.7 (36 ed.); 47.4.1.14 (38
ed.); 47.9.3.3 (56/1 ed.); 43.24.3.3 (71 ed.); 24.1.3.9 (32 Sab.); 26.4.3.9 (38 Sab.); 16.3.11 (41 Sab.);
45.1.1.2 (48 Sab.); 18.4.2.4 (49 Sab.); 21.1.38.10 (2 ed. cur.); 36.1.18.2 (2 fid.) but 15.1.47.4 (Paul
4 Plaut.)
131
Twenty-six of 29 texts: D 26.7.9.7; 27.4.1.5 (36 ed.); 37.4.3 pr (39 ed.); 37.5.5 pr (40 ed.);
37.9.1.15; 37.10.1.5 (41 ed.); 38.2.16.8 (45 ed.); 38.7.2.1 (46 ed.); 36.4.5.28 (52 ed.); 8.5.10.1 (53
ed.); 39.3.8 (53 ed.); 47.10.15.19 (57 ed.); 28.8.7.2 (60 ed.); 43.22.1.1 (70 ed.); 43.24.15.8;
50.17.157.2 (71 ed.); 42.8.10.20 (73 ed.); 35.3.1.11 (79 ed.); 7.4.10.1 (17 Sab.); 26.4.3.9 (38 Sab.);
48.18.1.23 (8 off. proc.); 21.1.38.10 (2 ed. cur.); 36.1.18.2 (2 fid.); 26.5.7 (1 omn. trib.); 23.2.43 pr
(1 Iul. Pap.); 25.3.5.12 (2 off. cons.); but 35.1.50 (Ulp. 1 off. cons. citing Pius); 3.5.18.4 (Paul 2
Ner.); 7.1.3.1 (Gaius 2 rer. cott.)
132
D 17.1.12.9 (31 ed.); 1.9.10 (34 ed.); 47.10.17.19 (57 ed.); 43.5.1.2 (68 ed.); 44.4.4.33 (76
ed.); 26.8.5 pr. (40 Sab.); 13.7.4; 47.2.27 pr (41 Sab.); 48.19.9.13 (10 off. proc.); 10.2.49 (2 disp.);
50.16.131.1 (4 Iul. Pap.); 49.1.6 (2 appell.) but 47.2.21 pr. (Paul 40 Sab. = Ulp. Mommsen)
133 D 4.3.1.8; 4.3.3 (11 ed.); 15.1.11.9 (29 ed.); 37.5.6.4; 37.5.8.6 (40 ed.); 37.9.1.14 (49 ed.);
quoque (without etiam),135 non solum . . . verum omnino,136 and non solummodo
. . . sed et.137 Questions are put with an ergo,138 an vero et,139 num forte,140
utrum autem,141 and utrum . . . an vero.142 Sive autem,143 solet autem,144 quia
autem,145 and interdum autem146 foreshadow a minor modification of what has
gone before. Interdum tamen,147 interdum . . . licet,148 bring in more radical
variations. Si quidem . . . si autem149 points to an antithesis.
Ulpian likes to argue by analogy, to point to examples that are sufficiently
like the one just discussed to be decided in the same way. Some of the linking
phrases fit this trait. The reasoning often proceeds from case to case in the
manner of the common lawyer trained in the Anglo-American tradition, with
135 D 9.4.5.1 (3 ed.); 3.3.39 pr. (9 ed.); 4.4.18.5 (11 ed.); 4.1.6; 4.7.4.2 (13 ed.); 10.4.3.9 (24 ed.);
13.6.5.8 (28 ed.); 14.4.1.5; 14.6.9.3; 15.1.1.6 (29 ed.); 25.5.1.1 (34 ed.); 37.11.2.4 (41 ed.); 29.5.1.7
(50 ed.); 36.4.5.12 (52 ed.); 47.10.11 pr (57 ed.); 11.8.1.8 (68 ed.); 50.16.60.1 (69 ed.); 43.19.5.2 (70
ed.); 29.2.30.1 (8 Sab.); 7.8.2.1 (17 Sab.); 34.3.3.4 (23 Sab.); 46.8.20 (1 disp.); 15.1.3.6 (2 disp.);
23.2.43.4 (1 Iul. Pap.) but 15.1.49 (Pomp. 4 Q. Muc.)
136
D 50.16.178 pr (49 Sab.) 137 D 50.1.8 (1 ed.)
138
D 11.1.9.4 (22 ed.); 15.2.1.7 (29 ed.); 26.4.5.3 (35 ed.); 26.4.5.4 (35 ed.); 27.9.3.5 (35 ed.);
46.1.8.3 (47 Sab.); 45.3.11 (48 Sab.) but 37.8.7 (Tryph. 16 disp.)
139
D 17.1.8.4 (31/1 ed.); 37.5.8.1 (40 ed.); 7.8.10.2 (17 Sab.); 7.6.1.3 (18 Sab.); 26.1.3.2 (37
Sab.)
140
D 4.4.16 pr (11 ed.); 10.3.7.13 (20 ed.); 26.2.17.2 (35 ed.); 43.18.1.2 (70 ed.); 1.7.15.2 (25
Sab.) but 49.16.3.7 (Mod. 4 poen.)
141
Fifteen of 16 texts: D 2.14.7.8 (4 ed.); 7.6.5.1 (17 ed.); 38.5.1.6 (44 ed.); 29.3.2.7; 29.5.3.14
(50 ed.); 39.2.30.1 (81 ed.); 28.1.5 (6 Sab.); 38.17.1.11 (12 Sab.); 7.8.10.3 (17 Sab.); 7.6.1.3 (18
Sab.); 33.8.8.8 (25 Sab.); 29.1.19.1 (4 disp.); 36.1.11.2 (4 fid.); 25.3.5.2, 20 (2 off. cons., twice); but
10.2.54 (Ner. 3 membr.) cf. Gaius, Inst. 3.189
142
Twenty-one of 25 texts: D 2.8.2.5 (5 ed.); 4.6.28.3 (12 ed.); 5.3.23 pr (15 ed.); 10.3.23 (32
ed.); 19.2.11 pr. (32ed.); 37.6.1.13 (40 ed.); 38.5.1.14.(44 ed.); 29.4.10.2 (50 ed.); 39.2.13.2 (53ed.);
42.4.5.3 (59 ed.); 40.5.4.3 (60 ed.); 46.7.5.7 (77 ed.); 38.4.1.8 (14 Sab.); 7.8.10.2 (17 Sab.); 41.9.1.2
(31 Sab.); 41.1.23.3 (43 Sab.); 48.19.8.7 (9 off. proc.); 48.19.9.14 (10 off. proc.); 23.3.5.1, 2 (2 off.
cons., twice); 40.9.30.4 (4 Ael. Sent.) but 24.1.31.4 (Pomp. 14 Sab); 18.6.2.1 (Gaius 2 rer. cott.);
44.7.44.6 (Paul 74 ed.); 35.1.82 (Call. 2 quaest.)
143
Honoré–Menner (1980) fiche 72 SIVE AUTEM (33 texts) but D 7.2.4 (Iul. 35 dig.); 13.6.18.1;
16.3.14.1 (Gaius 9 ed. prov.); 24.2.2.3 (Gaius 11 ed. prov.); 39.4.13.2 (Gaius 13 ed. prov.); 18.1.58
(Pap. 10 quaest., interp.); 50.4.18.29 (Arc. Char. 1 mun. civ.)
144
D 42.6.1.1 (64 ed.); 43.5.3.10 (68 ed.); 47.20.3.2 (8 off. proc.); 48.19.9.4 (10 off. proc.)
145
D 16.3.1.47 (30 ed.); 43.3.1.4 (67ed.); 43.18.1.6; 43.20.1.31 (70 ed.); 43.23.1.7;
43.24.15.12 (71 ed.); 26.1.3.1 (37 Sab.) but 49.14.42.1 (Val. 5 fid.)
146
D 4.4.13.1 (11 ed.); 9.1.1.15 (18 ed.); 14.5.4.1 (29 ed.); 7.9.1.7 (79 ed.); 29.2.21.1 (7 Sab.);
26.4.1.3 (14 Sab.) cf. Gaius, Inst. 3.199; 4.127; D 34.5.13.6 (Iul. 1 ambig: autem interdum) but
interp. Citati (1927) 48: Pernice and others
147
Eleven of 13 texts: D 27.6.5 (12 ed.); 4.4.19 (13 ed.); 6.2.11.3 (16 ed.); 11.1.11.3 (22 ed.);
9.3.5.2 (23 ed.); 10.4.11.1 (24 ed.); 47.10.17.13 (57 ed.); 7.2.1.3 (17 Sab.); 7.1.25.1 (18 Sab.);
46.3.12.1 (30 Sab.); 21.1.1.9 (1 ed. cur.) but 30.12.3 (Pomp. 3 Sab.); 11.3.14.9 (Paul 19 ed.) cf.
Gaius, Inst. 4.155
148
D 3.3.39.1 (9 ed.); 4.4.13.1 (11 ed.); 6.2.11.3 (16 ed.); 9.3.5.2 (23 ed.); 12.6.26.12 (26 ed.);
38.16.1.7 (12 Sab.) but 36.4.15 (Val. 7 act. = Venuleius: Krüger) cf. Gaius, Inst. 3.176: licet . . .
interdum; D. 4.8.15 (Ulp. 13 ed.); 28.3.3.6 (Ulp. 3 Sab.)
149
Thirteen of 15 texts: D 14.3.1 (28 ed.); 15.3.3.10 (29 ed.); 29.4.10.2 (50 ed.); 49.17.2 (67 ed.);
43.8.2.28 (68 ed.); 43.24.13.7 (71 ed.); 42.8.10.1 (73 ed.); 44.4.4.17 (76 ed.); 46.1.33; 46.7.5.6 (77
ed.); 35.3.1.4 (79 ed.); 7.4.10.7 (17 Sab.); 21.1.4.4 (1 ed. cur.) but 25.4.1 pr. (Ulp. 34 ed. citing divi
fratres); 36.1.67.3 (Maec 5 fid.)
52 2. The Oral Style
expressions such as ‘but suppose’ ( finge autem)150 ‘but what if?’ (quid tamen
si),151 ‘but if . . . equally’ (sed et si . . . aeque),152 and ‘unless perhaps’ (nisi
forte)153 also revealing his interest in whether the law would be the same if the
facts were slightly different.
(ii) Expository phrases. These introduce the point of law to be discussed, the
opinions of a jurist, or the implications of an argument. Quaestionis est154
and quaestio in eo est155 is an example. The opinions of other lawyers are
reported with alioquin . . . inquit,156 forte . . . inquit,157 nisi forte . . . inquit,158
plane . . . inquit,159 or extat sententia.160 Relevant points are called to
mind with meminisse oportet,161 meminisse oportebit,162 meminisse autem
150 Fifteen of 17 texts: D 4.4.3.6; 4.4.7.11 (11 ed.); 4.6.23.4 (12 ed.); 4.8.3.1 (13 ed.); 5.3.31pr
(15 ed.); 6.1.13 (16 ed.); 11.7.4 (25 ed.); 14.1.1.5 (28 ed.); 14.4.7 pr. (29 ed.); 17.2.52.10 (31/2 ed.);
7.1.17 pr (18 Sab.); 36.2.14.2 (24 Sab.); 2.15.8.20 (5 omn. trib.), 36.1.13.3 (4 fid.); 48.5.24.1 (1
adult.) but 13.7.8 pr. (Pomp. 35 Sab.); 37.8.7 (Tryph. 16 disp.); interp. Citati (1927) 33, 40:
Schulz, Appleton, etc.
151
Forty-two of 43 texts: D 26.7.23 (9 ed.); 3.5.5.3 (10 ed.); 4.8.7.1; 4.8.21 pr, 10; 26.7.25 (13 ed.);
5.3.25.17; 5.3.31.1 (15 ed.); 11.7.20 pr; 47.12.3.5 ( 25 ed.); 12.2.34 pr (26 ed.); 13.5.14.2 (27 ed.);
14.1.1.5 (28 ed.); 14.4.5.8; 14.4.7.1; 15.3.3.9 (29 ed.); 17.1.8.8 (31 ed.); 19.2.13.3; 19.2.9.1;19.2.15.4
(32 ed.); 27.4.1.6 (37 ed.); 47.4.1.3 (38 ed.); 37.10.5.5 (41 ed.); 46.7.5.6 (77 ed.); 28.2.12.1 (9 Sab.);
28.3.6.9 (10 Sab.); 38.17.2.34 (13 Sab.); 7.1.12.2; 7.1.68 pr; 7.4.10 pr; 29.2.21.3 (17 Sab.); 7.1.25.1
(18 Sab.); 34.3.5 pr (23 Sab.); 30.50.2 (24 Sab.); 33.8.6.4 (25 Sab.); 40.7.6.1 (27 Sab.); 18.2.9 (28
Sab.); 17.2.14 (30 Sab.); 24.1.21.1 (32 Sab.); 24.1.32.5 (33 Sab.); 5.1.50.1 (6 fid.); 48.5.18.3 (2 adult)
but 39.4.13.3 (Gaius, 13 ed. prov.) cf. 35.2.11.6 (Pap. 29 quaest: quid tamen dicemus?)
152 Twenty-five of 27 texts: D 10.3.7.13 (20 ed.); 11.6.1.1 (24 ed.); 15.1.9.6; 15.1.30.2 ; 15.3.3.1
(29 ed.); 27.5.1.5 (36 ed.); 9.4.38.3 (37 ed.); 12.1.12 pr; 39.4.12.2 (38 ed.); 37.6.1.11; 37.6.1.17 (40
ed.); 36.4.5.14 (52 ed.); 28.5.9.20 (5 Sab.); 30.37.1 (21 Sab.); 40.7.3.4 (27 Sab.); 18.2.4.5 (28 Sab.);
26.1.14.5 (37 Sab.); 26.8.5.5 (40 Sab.); 45.1.41 pr (50 Sab.); 47.20.3.1 (8 off. proc.); 24.3.64.4 (7
Iul. Pap.); 36.1.3.5 (3 fid.); 36.1.9 pr (4 fid.); 40.9.12.2 (4 adult.); 42.1.15.11 (3 off. cons.) but
40.5.49 (Afr. 9 quaest.); 7.1.60 pr (Paul 5 sent.) cf. Gaius, Inst. 2.193 (uncertain)
153
Above n.20–4
154
Eleven of 12 texts: D 5.1.2.5 (3 ed.); 11.3.9.3 (23 ed.); 11.7.8 pr (25 ed.); 12.1.11 pr (26 ed.);
27.2.1.2 (34 ed.); 29.4.10.2 (50 ed.); 44.2.7.1 (75 ed: magnae quaestionis est); 7.1.25.3 (18 Sab.);
18.2.2 pr (28 Sab.); 15.1.3.6 (2 disp.); 40.5.26.6 (5 fid.) but 32.89 (Paul 6 Iul. Pap.) cf. est quaes-
tionis: 15.1.9.6; 15.1.11.3 (Ulp. 29 ed.); 47.10.9 pr (Ulp. 57 ed.); fuit quaestionis: 15.2.1.7 (Ulp. 29
ed.); 29.4.1.12 (50 ed.); 29.2.24 (7 Sab.); 7.8.2.1 (17 Sab.); 36.2.12.1 (23 Sab.) but 37.7.9 (Tryph.
6 disp.); quaestionis non est: 50.16.164 pr (Ulp. 15 Sab.)
155
D 37.6.1.21 (40 ed.); 43.26.6.4 (71 ed.); 47.2.43.11 (41 Sab.); 41.1.23.1 (43 Sab.) but
23.2.60.4 (Paul 1 orat. Ant. et Comm.), the latter perhaps derived from a work of Ulpian
156
D 4.7.4.2 (13 ed.); 39.2.15.12 (53 ed.); 47.10.17.19 (57 ed.); 21.1.1.9 (1 ed. cur.); 29.2.42 pr
(4 disp.) cf. Frag. Vat. 87 (Ulp. 17 Sab.)
157
D 4.2.3.1 (11 ed.); 43.23.1.8 (71 ed.)
158
D 3.5.5.14 (10 ed.); 4.9.3.1 (14 ed.); 6.1.1.2 (16 ed.); 17.2.63.3 (31/2 ed.); 19.1.11.18 (32 ed.);
37.10.1.8 (41 ed.); 7.1.25.5 (18 Sab.).
159
D 10.2.20.3 (19 ed.); 16.1.6 (29 ed.); 17.1.12.2 (31/1 ed.); 33.9.3.2 (22 Sab.); 21.2.21.3 (29
Sab.) cf. Frag. Vat. 80 (Ulp. 17 Sab.)
160
D 27.7.4 pr.(36 ed.); 39.2.15.12 (53 ed.); 42.4.7.16 (59 ed.); 43.20.1.17 (70 ed.); 43.24.11.1,
8 (71 ed.); 44.4.4.8 (76 ed.)
161
D 25.4.1.9 (34 ed.); 43.26.4.1 (71 ed.); 30.49.8 (23 Sab.); 24.2.11.2 (3 Iul. Pap.); 25.3.5.9 (2
off. cons.); 40.2.16 (2 Ael. Sent.) but 49.1.4.5 (Macer 1 appell.)
162
D 5.2.8.14 (14 ed.); 14.3.13.1 (28 ed.); 47.2.93 (38 ed.); 39.1.5.2, 10 (52 ed.); 47.10.7.2;
47.10.15.23 (57 ed.); 39.5.7.5 (44 Sab.); 1.16.10 pr (10 off. proc.); 39.6.37 pr (15 Iul. Pap.); 36.1.6.1
(4 fid.)
2. The Oral Style 53
ed.); 43.19.3.10 (70 ed.); 35.3.3.8 (79 ed.); 47.2.27 pr (41 Sab.); 41.1.23.2 (43 Sab.) cf. generaliter
dicendum erit and related expressions 4.6.26.9 (12 ed.); 13.7.9.3 (28 ed.); 47.4.1.6 (38 ed.);
37.11.1.9 (39 ed.); 37.10.1.3 (41 ed.); 29.1.13.2 (45 ed.); 43.26.8.6 (71 ed.); 48.19.8.7 (9 off. proc.);
36.1.17.2 (4 fid.); 40.5.24 pr (5 fid.); 48.5.16.4 (2 adult.) cf. Frag. Vat. 210 (off. pr. tut.) but interp.
Citati 41: Eisele and others
168 D 5.3.31.2 (15 ed.); 24.3.22.7 (33 ed.); 43.12.1.5 (68 ed.); 36.1.23.5 (5 disp.); 40.9.30.2 (4 Ael.
Sent.)
169 D 43.16.1.22 (69 ed.); 21.2.17 (29 Sab.); 23.3.5.11 (31 Sab.); 47.2.48.7 (42 Sab.); 46.1.8.6 (47
Sab.); 48.18.1.20 (8 off. proc.); 32.11.6 (2 fid.) but interp. Citati (1927) 32: Beseler
170 D 19.1.13.7 (32 ed.); 29.5.3.11 (50 ed.); 45.2.3.1; 46.1.8.12 (47 Sab)
171 D 3.3.33.1 (9 ed.); 11.1.11.8 (22 ed.); 37.5.5.8 (40 ed.); 37.9.1.15 (41 ed.); 23.2.45.3 (3 Iul.
Pap.); 4.6.38.1 (6 Iul. Pap.) cf. 14.4.5 pr. (19 ed: nec nos dubitamus)
172 Twenty texts: D 27.5.1.3 (36 ed.); 28.1.22.6; 37.1.3.6 (39 ed.); 37.5.8.4 (40 ed.); 37.13.1.1 (45
ed., twice); 48.19.2.1 (48 ed.); 47.10.11.9 (57 ed.); 24.1.5.15 (32 Sab.); 26.4.2 (37 Sab.); 18.1.28;
47.2.25.2 (41 Sab.); 50.2.3.2 (3 off. proc.); 48.19.8.12 (9 off.proc.); 27.9.8 pr. (2 omn. trib.); 1.10.1.2
(2 off. cons.); 49.15.9 (4 Iul. Pap.); 48.2.5 (3 adult.); 48.5.30.9 (4 adult.); 1.10.1.2 (2 off. cons.)
173 D 39.1.5.12 (52 ed.); 39.2.24 pr (81 ed.); 46.4.8 pr. (18 Sab.); 47.10.30 pr (42 Sab.); 48.19.6
oportet; 30.43.1 (21 Sab: dici oportebit); 10.4.7.4 (24 ed.); 12.4.5.4 (2 disp.); 33.4.2 pr. (5 disp.) all
oportere dici; 47.2.46.5 (42 Sab: dici oporteat) but 34.9.22 (Tryph. 5 disp: non oportere dici)
179
D 4.8.21.10 (13 ed.); 29.4.10.2 (50 ed.); 39.2.15.13 (53 ed.); 43.16.1.38 (69 ed.); 30.41.4 (21
Sab.); 49.4.1.9 (1 appell.) but 24.1.31.3 (Pomp. 14 Sab.)
180
D 42.4.3.3 (59 ed.); 40.5.4.15, 16, 22 (60 ed., thrice); 28.8.8 (61 ed.); 29.2.71.1, 9 (61 ed.,
twice); 44.2.7.3 (75 ed.); 46.7.3.3, 8 (77 ed.); 45.3.7 pr. (48 Sab.); 37.14.16.1 (10 Iul. Pap.); 49.4.1.9
(1 appell.) but interp. Citati (1927) 54
181 182
D 27.9.11 (3 off. proc.) D 42.8.6.7 (66 ed.); 47.15.2 (9 off. proc.)
54 2. The Oral Style
more expansively, illud sciendum est.183 Sed est verius184 and est tamen
verius185 also serve this purpose and, finally, palam est186 followed by the
accusative and infinitive or the infinitive alone. Credo (‘I think’)187 is more
tentative. The author endorses the opinions of other jurists with placet sen-
tentia,188 sententiam puto veram/veriorem189 or with sententia mihi . . .
vera/verior videtur.190 He underlines the virtue of the solution espoused with
benignum est,191 (sententia) habet aequitatem,192 habet rationem,193 non est
sine ratione,194 and est tamen tutius.195 In rejecting wrong views or conduct,
183
Eighteen of 19 texts: D. 29.5.1.24 (50 ed.); 7.9.9.1 (51 ed.); 42.4.7.13 (59 ed.); 5.1.19.4 (60
ed.); 42.6.1.10 (64 ed.); 42.8.6.9 (66 ed.); 43.20.1.23 (70 ed.); 43.24.3.1 (71 ed.); 21.1.4.1 (74 ed.);
44.4.4.34 (76 ed.); 46.7.5.8 (77 ed.); 46.6.4.2; 7.9.1.2 (79 ed.); 24.1.7.9 (32 Sab.); 21.1.1.3;
21.1.19.4; 21.1.29 pr. (1 ed. cur.); 49.1.8 (4 appell.) but 18.6.18 (Pomp. 31 Q. Muc.) cf. 14.3.15
(Ulp. 28 ed: novissime sciendum est); 28.8.3 (60 ed: illud sciendum)
184
All twelve texts: D 4.2.9.8 (11 ed.); 4.3.7.3 (11 ed.); 6.1.15.3 (16 ed.); 27.3.1.4 (36 ed.);
28.7.8.5 (50 ed.); 44.4.4.8 (76 ed.); 7.4.29.2 (17 Sab.); 7.5.3 (18 Sab.); 34.3.5.2 (23 Sab.); 24.1.32.27
(33 Sab.); 26.2.10.3 (36 Sab.); 36.1.6.1 (4 fid.)
185
D 11.7.2.1 (25 ed.); 40.5.4.1 (60 ed.)
186
D 42.8.3 pr (66 ed.); 43.16.1.24 (69 ed.); 43.21.1.7 (70 ed.); 44.4.2 pr. (76 ed.); 46.7.5.3 (77
ed.); 19.1.10 (47 Sab.); 18.4.2.3; 45.1.38.3 (49 Sab.); 49.4.1.11 (1 appell.) cf. 43.29.3.13 (71 ed:
palam erit); 35.3.3.3 (79 ed: palam sit).
187
Nineteen of 21 texts: D 4.4.7.2 (11 ed.); 11.7.14.2 (25 ed.); 18.2.16 (32 ed.); 25.6.1.11 (34
ed.); 37.5.3 pr (40 ed.); 43.19.1.11 (70 ed.); 28.3.6.6 (10 Sab.); 30.34.4 (21 Sab.); 32.70.9, 11 (22
Sab.); 40.7.3.17 (27 Sab.); 24.1.33.1 (36 Sab.); 48.19.9.3 (10 off. proc.); 29.2.42 pr. (4 disp.);
17.1.29.3 (7 disp.); 32.11.21 (2 fid.); 36.1.13.3 (4 fid.); 48.5.18 pr. (2 adult.); 29.1.6 (2 appell.) but
40.4.55 pr. (Maec. 2 fid.); 30.115 (Ulp. 2 inst. giving form of fideicommissum)
188
D 5.3.13.1 (nobis); 5.3.18 pr. (15 ed.); 10.4.19.1 (24 ed: nobis); 19.1.13.14 (32 ed: mihi);
42.8.10.16 (73 ed.); 1.6.6 (9 Sab: nobis); 33.6.13 (23 Sab: mihi); 18.4.2.7 (49 Sab: mihi) cf. 47.2.43.5
(41 Sab: placeat)
189
With variants in the word-order: D 14.1.1.8 (28 ed.); 15.3.13 (29 ed.); 25.4.1.13 (34 ed.);
42.7.2.5 (65 ed.); 28.2.3.4 (1 Sab.); 7.1.7.1; 7.1.9.2; 7.1.12 pr; 7.4.10.7; 28.5.17.1 (17 Sab); 7.1.13.3
(18 Sab.); 30.44.2 (22 Sab.) but 38.1.4 (Pomp. 4 Sab.) cf. 5.3.13. 8 (Ulp. 15 ed.); 11.3.11 pr (23
ed.); 15.1.11.2 (29 ed.); 39.2.15.34 (53 ed.); 43.19.3.16 (70 ed.); 44.4.4.6 (76 ed.); 21.1.33 pr. (1 ed.
cur.) cf. 40.7.3.2 (27 Sab: veram putamus sententiam)
190
D 2.7.1.2 (5 ed.); 15.3.7.5 (29 ed.); 16.3.1.33 (Ulp. 30 ed.); 43.8.2.42 (68 ed.); 21.1.6.1 (1 ed.
cur.); 28.5.9.14 (5 Sab.); 7.8.12.1 (17 Sab.); 24.2.4 (26 Sab.); 40.7.9.2 (28 Sab.) but cf. 18.1.35.2
(Gaius 10 ed. prov.)
191
D 21.1.49 (8 disp.); 32.5.1 (1 fid.) but interp. Citati (1927) 14: Gradenwitz etc.
192
Nine of 10 texts: D 2.2.1 pr (3 ed.); 3.5.3.9 (10 ed.); 14.4.7.1 (29 ed.); 17.2.63.5 (31/2 ed.);
47.4.1.1 (38 ed.); 37.6.1 pr (40 ed.); 37.10.3.13 (41 ed.); 43.26.2.2 (71 ed.); 23.3.16 (34 Sab.) but
43.26.15 pr (Pomp. 29 Sab.)
193
Twenty-six of 27 texts: D 19.1 32 (11 ed.); 4.8.21.4 (13 ed.); 9.2.27.11; 9.4.2.1 (18 ed.);
10.3.7.8 (20 ed.); 12.2.9.6 (22 ed.); 11.7.2.8. (25 ed.); 14.5.4.5 (29 ed.); 17.1.12.5 (31/1 ed.); 18.3.4.1
(32 ed.); 37.10.3.13 (41 ed.); 42.4.3 pr. (49 ed.); 39.2.13 pr. (53 ed.); 5.1.19.2 (60 ed.); 20.1.21.1 (73
ed.); 7.9.7 pr; 35.3.1.6 (79 ed.); 28.6.10.6; 28.5.6.4 (4 Sab.); 7.1.12.4; 7.4.3.2 (17 Sab.); 34.2.19.3
(20 Sab.); 41.9.1.4 (31 Sab.); 24.1.5.15 (32 Sab.); 27.3.5 (43 Sab.); 45.1.3.1 (49 Sab.) but 49.14.14
(Gaius 11 Iul. Pap.) cf. 13.6.7.8 (Ulp. 28 ed: videtur habere rationem); 27.3.17 (Ulp. 3 off. cons:
non habet rationem, citing Sev. et Ant.)
194
D 13.7.11.4 (28 ed.); 14.4.9.2 (29 ed.); 42.4.7.11 (59 ed.); 16.2.13 (66 ed.); 28.5.9.4 (5 Sab.);
7.1.9.4 (17 Sab.); 37.4.17 (35 Sab.)
195
D 39.2.4.6 (1 ed.)
2. The Oral Style 55
he resorts to nec ferendus est,196 grave est,197 and improbum est.198 A currently
accepted solution is endorsed by et ita utimur199 or et hodie hoc iure utimur.200
‘It makes little difference’ ( parvi refert) 201 denies a distinction, while ‘unless
perhaps’ (nisi forte) draws attention to one, both with typical understatement.
For nisi forte Ulpian has 91 texts out of 117, but as explained the phrase does
not count as Ulpianic in view of its frequent use by Marcellus.202 Both
authors like to argue from case to case and to qualify what other authors say
by pointing out that the law might be different if the facts were slightly dif-
ferent.
In expressing his view on points of law Ulpian makes free use of the first
person singular, unusually for the Severan age. His uses of ego, me, and mihi
to refer to himself, rather than to a character in a legal example (‘if I sell you
a horse’), amount to 175 out of a total of 270, or 65 per cent. This is a good
deal higher than the 41 per cent that Ulpian’s excerpts form of the whole
Digest, but short of the 75 per cent that would qualify for use as a mark of
Ulpian’s style. Iavolenus203 and Celsus,204 in the age of Trajan, used the first
person singular relatively more freely than Ulpian was to do later. But among
his contemporaries Ulpian stands out for his frequent resort to the first per-
son. He uses it at about three times the rate of Papinian205 and Paul.206 What
is more, his uses tend to be emphatic. Thus, ego puto, a more categorical
expression than ego autem puto, ego non puto, etc., is Ulpianic in the context
of the Digest as a whole. He has 42 instances of it out of 51.207 So are ego
196
Ten of 11 texts: D 3.3.25 (9 ed.); 12.2.34.7 (26 ed.); 24.3.24.2 (33 ed.); 9.4.8 (37 ed.);
42.6.1.12, 15 (64 ed.); 23.3.33 (36 Sab.); 26.10.5 (3 disp.); 2.1.15 (2 omn. trib.); 22.6.6 (18 Iul. Pap.)
cf. Frag. Vat. 207 (off. pr. tut.) but 45.1.99 pr (Cels. 38 dig.).
197 D 13.5.1 pr (27 ed.) cf. 12.3.4 pr (36 ed: grave videbatur).
198 D 28.7.8.6 (50 ed.); 21.4.49 (36 Sab.)
199 D 42.7.2.5 (65 ed.); 18.2.9 (28 Sab.); 50.17.23 (29 Sab.); 23.3.5.9 (31 Sab.); 47.2.41.3 (41
ed.); 14.1.1.6; 14.3.7.1 (28 ed.); 14.4.5.3; 15.1.3.2 (29 ed.); 26.7.7.1 (31 ed.); 27.8.1 pr (36 ed.);
9.4.38 pr (37 ed.); 37.10.1.4; 37.10.3.9 (41 ed.); 29.4.1.1; 29.5.3.24 (50 ed.); 40.12.8.2 (55 ed.);
47.10.9.2 (57 ed.); 44.5.1.8 (66 ed.); 43.24.1.2 (71 ed.); 38.16.2.3 (13 Sab.); 7.4.1 pr (17 Sab.);
32.49.4 (22 Sab.); 40.7.3.1 (27 Sab.); 8.4.6 pr; 38.1.7.7 (28 Sab.); 47.2.3.1 (41 Sab.); 45.2.3 pr;
46.1.6.1 (47 Sab.) but D 13.1.17 (Pap. 10 quaest.); 3.4.6.1 (Paul 9 ed. = Ulp.) cf. 34.3.5.2 (Ulp. 23
Sab: et parvi, inquit, refert <Iulianus>). Noted by Kalb (1890) 130
202 Honoré–Menner (1980) fiche 64 NISI; VIR 4.149–150; above n.21–2, 24
203 Five texts, plus eight in comments on Labeo 204 Nine texts
205 Seven texts
206 Twenty-two texts
207 Above n.51 (11 texts) plus Honoré–Menner (1980) fiche 54 EGO PUTO (31 other texts) but
D.29.2.62 (Iav. 1 post.Lab.); 35.1.40.3 (Iav. 2 post. Lab.); 40.7.39.2 (Iav. 4 post. Lab.); 8.1.20
(Iav. 5 post. Lab.); 46.3.73 (Marc. 31 dig.); 4.2.4 (Paul 11 ed.); 17.1.58 (Paul 4 quaest.); 46.8.15
(Paul 14 Plaut.); 45.1.91.6 (Paul 17 Plaut.) cf. VIR 4.1349–50
56 2. The Oral Style
adquiesco,208 ego adsentio,209 ego arbitror,210 ego credo,211 ego moveor,212 ego
opinor,213 ego quaero,214 and ego scio.215
Papinian, a restrained writer, does not once use ego. Ulpian does so on 102
occasions. But although self-confident his use of ego etc. need not point to
egotism. In speaking as opposed to writing it is hardly possible to avoid a
fairly frequent resort to ‘I’. Other phrases that make use of the first person,
though without ego, include mea fert opinio ‘my opinion is’,216 invenio ‘I
find’,217 memini ‘I remember’,218 ausim dicere ‘I dare say’,219 retineo 220 ‘I
remember’, quaero ‘I ask’ (where it is the jurist, not the person consulting him,
who inquires),221 and volo tractare,222 ‘I want to discuss’. One phrase deserves
special mention. Si mihi proponas, ‘if you put it to me’, comes in nine texts of
Ulpian and in no one else.223 It shows that he conceives himself as engaged in
a debate with the reader. These illuminating texts are drawn not merely from
Disputations, where they might be expected, but from the major commen-
taries on the praetor’s edict and on Sabinus. Other phrases, finge autem,
which we have already met,224 accipe,225 accipies,226 and dicas,227 point the
same way.
It is a little more difficult to assess Ulpian’s use of the plural nos or nobis
(‘we, us’) of himself. Of about 58 such texts in the Digest 31 come from our
author, which is, again, at 53 per cent, more than proportionate to the volume
of his surviving texts, but lower than for the first person singular. It is some-
times difficult to know whether nos means ‘lawyers’ or ‘those interested in the
question’, or ‘the author’. The range of cases runs from quis nos sacerdotes
appellet,228 ‘one could call us lawyers priests’ through aliud erit nobis dicen-
dum,229 ‘I/ we lawyers shall have to say something different’, to et a nobis et a
Papiniano probatum est,230 ‘I approve, and so does Papinian’.
The use of nos by a legal author to refer to himself does not seem to be
recorded before Pomponius231 and Gaius.232 There is no such use, for exam-
ple, in Labeo, Iavolenus, or Celsus, and the one instance attributed to Julian
in the Digest is palpably interpolated.233 It is tempting to think that the habit
spread to legal writing from an academic context. It expresses the relation of
the teacher and author to those who, by reading his work, hope to learn and
so are, in a sense, his pupils. In the Antonine and Severan ages most legal
authors make occasional use of the idiom.234 What marks Ulpian, once again,
is the categorical use of nos with the present indicative to underline his own
opinion. This manner of writing does not appear, so far as I can judge, in
other legal writers. Thus we have in Ulpian’s texts nos consentimus,235 nec nos
dubitamus,236 nos opinamur,237 nos probamus,238 and nos putamus,239 expres-
sions which would be taken as compilatorial were they not confined to
Ulpian. Closely related to these are uses of verbs in the first person plural
without nos, in which the subject is not specified, but the author probably
intends himself to be understood: non dubitamus,240 invenimus,241 novimus,242
opinamur,243 ostendimus,244 (supra) probavimus,245 putavimus,246 specta-
mus,247 subsistimus,248 vetamus.249
As stated, it is at times unclear whether the subject is ‘we’ or ‘lawyers’.
The author may not himself have been sure. This is true of addimus,250
adhibemus,251 admittimus (a favourite),252 aestimamus,253 animadvertimus,254
231 D 1.2.2 pr (Pomp. 1 enchir: necessarium itaque nobis videtur . . . demonstrare)
232 D 44.7.1.15 (Gaius 2 rer. cott: non de eo nos loqui). This may not convince those who think
Res cottidianae late or postclassical, but cf. 39.4.5.1 (Gaius ed. pr. urb: quaerentibus nobis), 1.2.1
(1 XII tab: libentius nos ad lectionem producunt), Inst. 1.188 (nosque . . . hunc tractatum executi
sumus), 4.60 (sed nos apud quosdam scriptum invenimus)
233 D 9.2.51.2 (Iul. 86 dig), embedded in a long Tribonianic passage
234 Three texts in Africanus, 2 in Scaevola, 2 in Callistratus, 3 in Tryphoninus, 6 in Papinian,
(44 ed.); 50.16.195.3 (46 ed.); 28.5.4.2 (4 Sab.); 7.1.25.1; 7.1.25.6 (18 Sab.); 18.6.4.2 (28 Sab.);
28.5.35.2 (4 disp.); 36.1.13.1 (4 fid.) but 28.2.29.5 (Scae. 6 quaest.) cf. ostendi 14.4.3.2 (Ulp. 29 ed.)
245 D 14.3.13.2 (28 ed.); 30.71.4 (51 ed.); 40.5.24.18 (5 fid.) 246 D 4.4.3.2 (11 ed.)
247 Fourteen texts: D 3.5.9.1 (10 ed.); 4.4.3.1 (11 ed.); 9.3.5.11 (23 ed: expectamus?); 15.1.11.2;
5.3.3.6 (29 ed.); 17.2.63.6 (31/2 ed.); 38.2.3.20 (41 ed.); 49.17.8 (45 ed.); 47.8.2.22 (56/1 ed.);
42.4.7.15 (59 ed.); 43.29.3.1 (71 ed.); 34.2.19.13, 20 (20 Sab.), 24.1.32.14 (33 Sab.)
248 D 3.3.33 pr. (9 ed) 249 D 3.3.33 pr. (9 ed.)
250 D 41.2.13.1 (72 ed.); 1.1.6 pr. (1 inst.) 251 D 25.4.1.3 (34 ed.); 28.1.22.5 (39 ed.)
252 D 3.3.33 pr. (9 ed.); 14.1.1.5 (28 ed.); 37.5.1.2 (40 ed.); 37.9.1.5; 37.11.2.4 (41 ed.); 38.6.1.6
(46 ed.); 47.8.4.3 (56 ed.); 43.24.9 pr; 43.24.11.13 (71 ed.); 38.16.2.5; 38.17.2.17 (13 Sab.); 38.4.3.2
(14 Sab.); 24.1.32.18 (33 Sab.); 26.1.3.1 (37 Sab.); 48.5.28.2 (3 adult.)
253 D 9.2.21.2 (18 ed.) 254 FV 156 (1 excus.)
58 2. The Oral Style
255 256
D 33.8.8.8 (25 Sab.) D 1.5.10 (1 Sab.); 50.17.209 (4 Iul. Pap.)
257
D 41.3.6 (11 ed.); 42.1.4.5 (56 ed.); 42.8.16.4 (66 ed.); 32.52.1 (24 Sab.); 24.3.7.8 (31 Sab.);
49.4.1.5 (1 appell.)
258
D 28.5.1.5 (1 Sab.); 24.1.5.15 (32 Sab.)
259
Eleven texts: actionem D 3.6.5.1 (10 ed.); 47.12.3.9 (25 ed.); 14.1.1.20 (28 ed.); 43.18.1.4 (70
ed. = 50.17.156.1); 47.2.12.2; 47.2.14.17 (29 Sab.) cf. 14.4.9.2 (29 ed. = 50.17.44 actionem under-
stood) cf. 4.6.28.3 (12 ed.); 26.10.1.3 (35 ed.); 13.4.2.3 (37 ed.); 38.2.3.20 (42 ed.)
260 261
D 24.1.32.14 (33 Sab.); 40.5.26.1 (5 fid.) D 50.16.195.2 (46 ed.); 15.1.41 (43 Sab.)
262
D 50.16.199 pr (8 omn. trib.)
263
D 15.3.10.8 (29 ed.); 33.8.8.8 (25 Sab.); 1.1.6 pr. (1 inst.)
264 265
D 28.6.10.6 (4 Sab.); 7.1.22 (18 Sab.) D 23.3.23 (35 Sab.)
266 267
D 1.1.6 pr. (1 inst.) D 4.4.3.4 (11 ed. nec eo movemur quod . . .)
268 269
D 17.2.63.8 (31/2 ed.); 36.4.5.3 (52 ed.) D 38.17.2.34 (13 Sab.)
270 271
D 9.2.5.2 (18 ed.); 38.5.1.1 (42 Sab.) D 15.1.41 (43 Sab.)
272 273
D 38.5.1.1 (42 Sab.) D 37.14.16.1 (10 Iul. Pap.)
274
D 36.4.5.21 (52 ed.); 39.3.1.23 (53 ed.); 50.17.9 (15 Sab.); 26.2.10.1 (36 Sab.); 50.17.34 (45
Sab.); 45.1.41 pr (50 Sab)
275
D 2.14.10.2 (4 ed.); 16.3.1.18 (30 ed.); 19.1.13.31 (32 ed.); 37.1.3 pr (39 ed.); 47.9.1.2 (56
ed.); 47.10.7.1 (57 ed.); 50.16.46 pr. (59 ed.); 43.12.1.14 (68 ed.); 44.1.2.4 (74 ed.); 28.5.6.2 (4 Sab.);
35.1.9 (20 Sab.); 34.3.5 pr (23 Sab.); 45.1.38.9; 50.16.178.2 (49 Sab.) cf. 2.14.7.5 (4 ed.); 28.5.3.4
(3 Sab.); 5.1.61 pr (26 ed.); 29.3.2.1 (50 ed.). Dicere solemus: D. 50.16.111 (Iav. 6 Cass.); 47.10.1
pr. (Ulp. 56 ed.)
276 277
D 15.3.10.8 (29 ed.) D 13.5.16.4 (27 ed.); 24.1.3.3 (32 Sab.)
278 279
D 33.7.12.27 (20 Sab.); 23.3.9.3 (31 Sab.); 1.1.1.3 (1 inst.) D 40.5.30.4 (5 fid.)
280 281
D 45.1.41 pr (50 Sab.) D 42.3.6 (64 ed.)
282 283
D 39.3.1.23 (53 ed.); 25.3.5.4 (2 off. cons.) D 29.4.6.2 (50 ed.)
284 285 286
D 29.4.2.1 (7 Sab.) D 3.1.1.5 (6 ed.) D 9.4.2.1 (18 ed.)
287 288
D 14.1.1.16 (28 ed.); 37.11.1.4 (39 ed.); 37.5.10.1 (40 ed.) D 47.10.17.7 (57 ed.)
289 290 291
D 3.1.1.10 (6 ed.) D 9.2.21.1 (18 ed.) D 38.5.1.13 (44 ed.)
292
D 42.2.6.6 (5 omn. trib.)
293
D 16.3.1.1 (30 ed.); 33.1.3.3 (24 Sab.); 45.1.41 pr (50 Sab.)
294 295 296
D 44.4.4.31 (76 ed.) D 42.6.1.6 (64 ed.) D 32.1.1 (1 fid.)
297
D 6.2.7.17 (16 ed.) cf. 9.2.27.9 (18 ed.); 11.7.4 (25 ed.); 26.7.3.2 (35 ed.); 26.7.9.1 (36 ed.)
2. The Oral Style 59
‘I remember saying that . . . ’ (et retineo me dixisse . . . ;298 ut in iunctura argentea scio
me dixisse).299
‘The deified Marcus replied in accordance with my view’ (secundum nostram senten-
tiam etiam divus Marcus rescripsit)300
‘I noted in a comment on Marcellus that I did not think this should be said to apply
to all wrongs’ (apud eum notavi non de omni iniuria hoc esse dicendum me putare)301
Ulpian uses nos and me about equally in these texts. It would be hard to find any
parallel to them in Roman legal literature. But, as will appear in chapter 7,302 an
assured sense of his standing is a feature of the earlier rather than the later years
of Caracalla’s sole reign.
(iv) Uses of the future instead of the present tense. A feature of writing about
law not shared by other branches of learning is that statements of law can be
in either the present or the future tense. ‘The contract is binding’ and ‘the con-
tract will be binding’ are equivalent. The tense has no temporal connotation,
but the future draws attention to the hypothetical character of statements
of law in relation to assumed facts. Perhaps, too, the use of the future fits a
forward-looking, optimistic, outlook, the optimism of a lawyer who is
confident that the correct view will prevail.
Ulpian is a partisan of the future tense. Whether he is so to a greater extent
than any other Digest lawyer could not be settled without excessive labour.
But it is clear that Ulpian often uses, at least as a variant, the future form of
a verb when others confine themselves to the present. An example is dicendum
erit,303 ‘it will have to be said’, viz. that the law is so and so. Whereas dicen-
dum est is commonly found in legal writing, dicendum erit304 and erit dicen-
dum305 come 179 times in texts of Ulpian and only 15 times in other writers.
Confined to Ulpian are the forms et dicendum erit,306 et ita erit dicendum,307
et generaliter dicendum erit,308 et generaliter erit dicendum,309 and et ideo
298 D 35.1.92 (5 fid.) 299 D 34.2.19.8 (20 Sab.); Pernice (1885) 382
300 D 29.1.3 (2 Sab.) 301 D 47.10.11.7 (57 ed.) 302 Below ch.9 n.10–23
303 Honoré–Menner (1980) fiche 53 DICENDUM. cf. VIR 2.217–9
304 Honoré–Menner (1980) fiche 53 DICENDUM ERIT (77 texts), but D 36.1.67.1 (Maec. 5 fid.);
23.2.48 pr (Ter. Clem. 8 Iul. Pap.); 4.2.17 (Paul 1 quaest.); 31.82 pr (Paul 10 quaest.); 18.1.15.2
(Paul 5 Sab.); 41.3.44.4 (Pap. 23 quaest.); 2.8.15.5 (Marci. 1 appell.); 48. 19.10.2 (Macer 2 pub.
iud.). The non-Ulpian texts mostly have idem dicendum erit or the like
305 Honoré–Menner (1980) fiche 55 ERIT DICENDUM (102 texts) but D 13.6.13.2 (Pomp. 11
Sab.); 39.2.29 (Gaius 28 ed. prov.); 28.5.48 pr (Afr. 4 quaest.); 33.9.4.5 (Paul 4 Sab.); 38.2.42.2
(Pap. 13 quaest.); 34.5.15 (Marci. 2 reg.); 20.1.16.8 (Marci. 1 hyp.)
306 D 5.3.25.9 (15 ed.) cf. 43.21.3.6 (70 ed: et erit dicendum)
307 D 4.4.3.3 (11 ed.) 308 D 13.7.9.3 (28 ed.) 309 D 43.26.8.6 (71 ed.)
60 2. The Oral Style
47.10.5.3 (56/2 ed.); 47.10.13.5 (57ed., twice); 42.1.5.1 (59 ed.); 40.5.4.5, 17 (60 ed.); 42.6.1.10
(64 ed.); 41.2.4; 43.3.1.12 (67 ed.); 11.8.1.3 (68 ed.); 43.17.4 (70 ed.); 43.24.3.4 (71 ed.);
43.24.15.11; 43.24.11.8, 13 (71 ed.); 20.6.4.1; 42.8.2; 42.8.10.20 (73 ed.); 44.2.7 pr; 44.2.7.5 (75
ed.); 44.4.4.21; 44.5.1.6 (76 ed.); 29.2.6.2 (6 Sab.); 18.2.4.6 (28 Sab.); 45.3.11 (48 Sab.);
48.22.7.11 (10 off. proc.); 24.2.11.2 (3 Iul. Pap.); 49.9.1 (4 appell.) but 28.5.85.1 (Paul 23
quaest.)
316 Honoré–Menner (1980) fiche 51 CESSAT but interp. Citati (1927) 16 Beseler etc cf. VIR
1.725–6 17
317 Honoré–Menner (1980) fiche 51 CESSAT
318 Honoré–Menner (1980) fiche 51 CESSABIT
319 D 2.1.7.2 (3 ed.); 25.3.5.19 (2 off. cons.); 42.1.15.6 (3 off. cons.)
320 D 39.2.15.35 (53 ed.); 33.9.3.6 (22 Sab.); 36.1.18.4 (2 fid.)
321 D 4.4.9.5 (11 ed.); 12.2.7 (22 ed.); 21.2.37.1 (32 ed.); 27.3.1.15 (36 ed.); 38.8.1.8 (46 ed.);
42.6.1.13 (64 ed.); 28.6.2 pr. (6 Sab.); 38.16.2 pr (13 Sab.); 38.17.2.43 (13 Sab.); 30.34.4 (21 Sab.);
18.6.4.2 (28 Sab.); 25.1.5 pr (36 Sab.); 30.74 (4 disp.); 9.2.49.1 (9 disp.); 39.6.37 pr (15 Iul. Pap.);
36.1.18.3 (2 fid., twice ); 48.5.24.4 (1 adult.) cf. 4.6.26.5 (12 ed.); 6.2.9.1 (16 ed: both erit accipi-
enda); 40.4.13.3 (5 disp: erunt accipienda)
322 D 2.8.2.3 (5 ed.); 3.2.2.1 (6 ed.); 4.9.1.5 (14 ed.); 5.3.20.15 (15 ed.); 9.2.5.1; 9.2.27.17(18
ed.); 14.1.1.21; 50.16.185 (28 ed.); 37.11.1.6 (39 ed.); 38.8.1.4 (46 ed.); 48.19.2.2 (48 ed.);
40.12.22.1 (55 ed.); 47.8.2.11; 47.8.4.3 (56 ed.); 50.16.45 (58 ed.); 40.5.4.11 (60 ed.); 46.7.3.6 (77
ed.); 7.9.3.1 (79 ed.); 23.3.33 (6 Sab.); 21.1.23.2; 21.1.27; 21.1.31.12 (1 ed. cur.); 40.16.2.1 (2 off.
cons.); 1.12.1.8 (off. pr. urb.); 23.2.43.8 (1 Iul. Pap.) cf. Collatio 2.4.1 (18 ed.), Frag.Vat. 188
(off.pr. tut.) but D 13.6.18 pr. (Gaius 9 ed. prov.); 24.2.9 (Paul 2 adult.); 48.16.13 pr (Paul 3
adult.)
323 D 4.6.21.3 (12 ed.); 5.2.8.9 (14 ed.); 38.8.1.5 (46 ed.); 47.9.1.2 (56/1 ed.); 33.6.11 (23 Sab.);
consequenter dicemus 326 and 26 of 27 with consequens erit dicere 327 come from
him. So do all three with definiemus,328 all five with dicet quis,329 six of seven
with melius dicetur,330 all seven with officium erit (of the praetor etc.).331 He
has the only passages with accedendum,332 recedendum,333 and distinguendum
erit.334 He alone writes in ea erit causa, as opposed to in ea causa est.335
Aequissimum est is found in other writers, but aequissimum erit (‘it will be
most equitable’) only in Ulpian, 26 times.336 Nullius momenti est occurs in
others, nullius erit momenti only in our author.337 Meminisse oportet is found
in others, meminisse oportebit is Ulpianic.338 Locus est, ‘there is room for . . .’,
is fairly common in legal writing, erit locus and locus erit with 22 mentions
between them, are confined to Ulpian.339 The list includes erit consequens,340
difficile erit,341 dubium non erit,342 fortasse or fortassis with the future indica-
tive,343 and gravabitur.344 We have already met et eveniet,345 et intererit,346
and et erit procedendum.347 Ulpian’s liking for future tenses is not a constant.
326
D 12.2.11.1 (22 ed.); 17.1.10.3 (31/1 ed.); 25.2.17 pr (34 ed.); 27.9.5.14 (35.ed.); 37.5.3.5 (40
ed.); 43.16.3.15 (69 ed.); 30.49.1 (23 Sab.); 18.6.1 pr (28 Sab.); 24.1.7.3 (32 Sab.); 48.22.7.13 (10
off. proc.); 10.2.49 (2 disp.); 28.5.35.1 (4 disp.); 4.6.38 pr (6 Iul. Pap.) cf. 37.6.5 pr. (79 ed: conse-
quenter erit dicendum)
327 D 11.1.9.4 (22 ed.); 26.10.1.4; 26.10.3.6; 27.9.5.3 (35 ed.); 29.1.11 pr (45 ed.); 29.6.1.1 (48
ed.); 47.10.3.1 (56/2 ed.); 47.10.17.1 (57 ed.); 42.1.4.4 (58 ed.); 42.4.7.3 (59 ed.); 40.5.4.5 (60 ed.);
43.4.3 pr (68 ed.); 43.24.15.12; 46.7.3.7 (71 ed.); 43.32.1.4 (73 ed.); 44.2.7.3 (75 ed.); 39.5.19.3;
44.6.1.1 (76 ed.); 46.7.5.3 (77 ed.); 46.6.4.7 (79 ed.); 39.2.30.1 (81 ed.); 18.4.2.11, 16 (49 Sab.);
46.4.13.1 (50 Sab.); 2.12.1.1 (4 omn. trib.); 36.1.17.9 (4 fid.) cf. Frag. Vat. 269 (46 Sab.) but
29.1.41.5 (Tryph. 18 disp.)
328
D 37.1.3.2 (39 ed.); 25.1.3.1 (36 Sab.); 26.1.3.1 (37 Sab.)
329
D 27.9.7 pr (35 ed.); 38.17.2.41 (13 Sab.); 32.55.7; 50.16.167 (25 Sab.); 41.1.33.1 (4 disp.)
330
Above n.179
331
D 39.2.4 pr. (1 ed.); 6.1.9 (16 ed.); 7.6.5.6 (17 ed.); 10.3.6.9 (19 ed.); 26.7.5.8 (35 ed.);
36.4.5.22 (52 ed.); 39.3.6.6 (53 ed.)
332 333 334
D 42.1.4.3 (58 ed.) D 43.29.3.7 (71 ed.) D 44.2.11.10 (75 ed.)
335 D 42.4.7.5 (59 ed); 2.11.4.1 (74 ed.); 45.1.38.22 (49 Sab.)
336 D 3.5.13 (10 ed.); 4.8.13.4 (13 ed.); 5.3.13.9 (15 ed.); 5.3.37 (15 ed.); 4.9.7.3 (18 ed.); 14.3.11
pr (18 ed.); 14.3.11.5 (28 ed.); 16.3.1.27 (30 ed.); 17.1.12.9 (31/1 ed); 25.4.1.8 (34 ed.); 26.10.3.7
(35 ed.); 27.3.1.11 (36 ed.); 37.4.3.4 (39 ed.); 37.5.3.6 (40 ed.); 36.4.5.30; 39.1.5.10 (52 ed.);
47.10.7.2. (57 ed.); 43.14.1.7 (68 ed.); 43.18.1.5 (70 ed); 30.53.8 (25 Sab.); 21.1.17.19 (1 ed. cur.);
36.1.23.3 (5 disp.); 2.15.8.22 (5 omn. trib.); 36.1.6.3 (4 fid.); 40.5.24.14 (5 fid.); 42.1.15.9 (3 off.
cons.) cf. Kalb (1890) 132
337 D 29.1.15.1 (45 ed.); 26.8.5.2 (40 Sab.); 46.4.13.1 (50 Sab., twice); 2.15.8.17 (5 omn. trib.);
26.5.8.1 (8 omn. trib.) cf. 50.9.4.2 (1 off. cur. reip: ullius erit momenti)
338 Above n.162
339 D 4.2.9.1 (11 ed.); 9.1.1.7; 9.2.7.4 (18 ed.); 14.6.7.12 (29 ed.); 26.7.5.3 (35 ed.); 27.4.1.3 (36
ed.); 13.1.10.2; 47.4.1.11 (38 ed.); 38.5.1.17 (44 ed.); 39.2.15.33 (53 ed.); 43.12.1.15 (68 ed.);
43.20.1.14 (70 ed.); 39.1.20.2, 7; 43.24.9.3 (71 ed.); 43.22.1.5 (73 ed.); 18.4.2.11 (49 Sab.); 21.1.1.5
(1 ed. cur.); 10.2.49 (2 disp.); 36.1.1.12 (3 fid.); 36.1.17.9 (4 fid.); 37.14.16.1 (10 Iul. Pap.) cf. locus
non erit 40.4.12 (50 ed.); 39.1.20.13 (71 ed.)
340 D39.5.19.3 (76 ed.); 50.17.34 pr (45 Sab.) 341 D 50.16.99.2 (1 off. cons.)
342 D 12.2.11.3 (22 ed.)
343 D 6.1.15.1 (16 ed.); 15.1.30.6 (29 ed.); 37.10.5.3 (41 ed.); 50.16.167 (25 Sab.); 32.55.7 (25
It curves upward to a peak and then declines,348 a variation that helps to plot
the order of his works.
(v) Inversions of the traditional Latin word-order. The reader will already
have noticed some inversions of the classical literary Latin word-order. The
auxiliary often comes before the main verb, erit admittendum for admittendum
erit, est decretum for decretum est. The object sometimes follows the verb:
habet aequitatem 349 or habet rationem 350 instead of aequitatem/rationem
habet. There are other instances with habere and other verbs followed by a
noun.351 This verb–object order in Ulpian’s writing rests, no doubt, on his
habit of dictating that carries with it a tendency to slip into the order that pre-
vailed in ordinary speech. It does not predominate in his writing, but it occurs
more frequently than in other legal authors of his and earlier ages. In later
Latin inversions of this sort become standard.352
If we begin with a more striking feature, the inversions of erit and the
gerundive, we find, taking the -um ending to include endings in -a or -us, a
long list. Apart from erit accipiendum, admittendum,353 dicendum,354 notan-
dum,355 probandum,356 and procedendum,357 there are uses, confined to
Ulpian among lawyers, of erit agendum,358 audiendum,359 cavendum,360 cogen-
dum,361 concedendum,362 conveniendum,363 dandum (actio etc.),364 decurr-
endum,365 defendendum,366 descendendum,367 detrahendum,368 excipiendum,369
348 Honoré (1982) 189–90
349 D 3.5.3.9 (10 ed.); 14.4.7.1 (29 ed.); 17.2.63.5 (31/2 ed.); 37.10.3.13. (41 ed.)
350 Above n.193
351 D 37.15.5.1 (10 ed.); 4.8.31 (13 ed.); 12.1.9.3 (26 ed.); 14.4.1.3 (29 ed.); 47.6.1.1 (38 ed.);
47.10.7.1 (57 ed.); 44.4.4.10; 44.4.7.1; 44.5.1.6 (76 ed.); 35.3.3.1 (79 ed.); 32.50.4 (23 Sab.); 33.8.6.1
(25 Sab.); 25.1.1.1, 2; 26.2.10.2 (36 Sab.); 46.3.7 (43 Sab.); 2.15.8.24 (5 omn. trib.) all with habere,
cf. 4.4.25 (Gaius 4 ed. prov: nullam habet dubitationem); CJ 2.11.8 (20 Feb. 205: non laesit exist-
mationem tuam); 5.37.1 (20 Sept 206: times administrare res adulescentis); D 3.3.39.1 (9 ed: agit
quamcumque actionem); 4.4.9.4 (11 ed: capit restitutionem); 27.3.1.19 (36 ed: gerunt tutelam);
9.2.27.29 (18 ed: tollit actionem); 37.11.1.8 (39 ed: facit testamentum); 29.4.1.10 (50 ed: si possideat
hereditatem); 39.3.1.23 (53 ed: non cogemus vicinum aggeres munire); 38.17.2.34 (13 Sab: matris
punimus consilium); 7.6.1.1 (18 Sab: sequitur usumfructum); 47.2.17.3 (40 Sab: movet quaes-
tionem); 27.1.7 (1 excus: dat excusationem); 48.5.14.10 (2 adult: accusat mores). There are plenty
of examples but the older object–verb form predominates
352 Pompeius in Keil (1855–80) 5.139–50; Kaster (1988) 139–60 353 Above n.321, 324
354 Above n.305 355 Above n.325 356 Above n.315 357 Above n.71
358 D 2.7.3 pr (5 ed.); 9.2.27.17, 28 (18 ed.); 13.6.3.4 (28 ed.); 47.10.15.29 (57 ed.); 43.5.3.5 (68
370 D 36.4.3.2 (52 ed.) 371 D 4.4.7.8 (11 ed.); 37.6.1.17 (40 ed.)
372 D 38.17.2.44. (13 Sab.) 373 D 49.7.1.1 (4 appell.)
374 D 46.3.24 (47 ed.); 43.13.1.8. (68 ed.)
375 D 37.6.1.11 (40 ed.); 7.5.10.1 (79 ed.)
376 D 43.3.1.11 (67 ed.) 377 D 43.13.1.8 (68 ed)
378 D 7.8.4 pr. (17 Sab.) 379 D 34.3.5.4 (23 Sab)
380 D 43.29.3.13 (71 ed.) 381 D 21.1.59.1 (74 ed.)
382 D 26.7.3.5 (35 ed.); 43.15.1.6 (68 ed.) 383 D 42.5.15 pr (62 ed.)
384 D 11.7.8.2 (25 ed); 48.19.9.15 (10 off. proc.); 48.5.30.2 (4 adult.)
385 D 19.1.13 pr (32 ed.); 30.47.1 (22 Sab.); 37.11.5.1 (4 disp.)
386 D 37.10.3.12 (41 ed.) 387 D 49.3.1 pr, 1; 49.4.1.4 (1 appell.).
388 D 26.10.3.7 (35 ed.) 389 D 12.2.5 pr. (22 ed.) 390 D 21.1.12.1 (1 ed. cur.)
391 D 13.3.3 (27 ed.) 392 D 37.9.7.1 (47 ed.); 33.4.1.4 (19 Sab.)
393 D 38.2.14.11 (45 ed.); 36.2.14.2 (24 Sab.) 394 D 2.15.8.9 (5 omn. trib.)
395 D 4.4.3.4 (11 ed.); 24.1.7.3 (32 Sab.) 396 D 42.8.10.11 (73 ed.)
397 D 36.4.3.2 (52 ed.) 398 D 48.19.32 (6 ed.); 29.4.6.1 (50 ed.); 30.39.1 (21 Sab.)
399 D 3.2.19 (8 ed.) 400 D 6.2.7.13 (16 ed.); 13.3.3 (27 ed.)
401 D 2.11.2.8 (74 ed.) 402 D 38.2.8.1 (43 ed.); 2.11.2.2 (74 ed.)
403 D 39.1.1.1 (52 ed.) 404 D 3.3.25 (9 ed.); 21.1.38.3 (2 ed. cur.)
405 D 14.4.5.9 (29 ed.) 406 D 43.13.1.12 (68 ed.)
407 Texts in n.305, 309, 311, 313, 315, 321, 324, 325, 352–406
408 Above n.305 and D 41.1.65.3 (Lab. 6 pith.); 35.1.68 (Iav. 2 Cass.); 41.2.51 (Iav. 5 post.
Lab.); 28.6.16 pr (Pomp. 3 Sab.); 21.2.16 pr (Pomp. 9 Sab.); 34.2.34.2 (Pomp. 9 Sab.); 31.43.1
(Pomp. 3 Q. Muc.); 36.1.72.1 (Pomp. 2 fid.); 32.13 (Maec. 2 fid.); 40.5.32 pr (Maec. 15 fid.); 38.5.6
(Iul. 26 dig.); 48.5.6 pr (Pap. 1 adult.); 36.3.5.3; 24.3.40 (Pap. 28 quaest.); 46.1.51.2 (Pap. 3 resp.);
31.76 pr. (Pap. 7 resp.); 31.77.16 (Pap. 8 resp.); 3.4.6 pr (Paul 9 ed. = Ulp. 9 ed.); 12.2.30.2 (Paul
18 ed.); 10.1.4.1 (Paul 23 ed.); 10.2.29 (Paul 23 ed.); 15.1.26 (Paul 30 ed.); 17.2.65.3 (Paul 32 ed.);
3.3.61 (Paul 1 Plaut.); 48.20.7.1 (Paul 1 port. lib. damn.); 27.1.45.4 (Tryph. 13 disp.); 3.5.28 (Call.
3 ed. mon.); 20.6.8.14 (Marci. 1 hyp.)
409
D 15.1.36 (2 disp.)
410
D 2.13.4.5 (4 ed.); 3.6.5 pr; 22.1.37 (10 ed.); 4.4.3.1 (11 ed.); 4.1.6 (13 ed.); 13.6.5.2 (28 ed.);
17.1.12.9 (31/1 ed.); 42.8.10.13 (73 ed.); 28.3.6.8, 10 (10 Sab.); 50.12.3 pr (4 disp., twice); 48.1.5.1
(8 disp.); 40.5.24.21 (5 fid.) but 50.16.244 (Lab. 4 pith. interp.); 16.3.24 (Pap. 9 quaest.); 3.5.27
(Tryph. 2 disp.); 47.22.1.2 (Marci. 3 inst., derivative)
411
Above n.77
64 2. The Oral Style
(vi) Verbs. A verb which VIR notes as typical of Ulpian is accipere in the sense
‘take’, ‘construe’, ‘interpret’.420 We have come across erit accipiendum, accip-
iemus, and accipietur.421 A number of other forms of accipere count as marks
of his style. One is accipe, of which all 14 instances come from Ulpian.422
Others are accipias,423 accipies,424 accipere debes,425 accipere debemus, for
which Ulpian has 82 texts of 87,426 accipere debeamus,427 accipere nos debere,428
and accipere nos oportet.429 For accipimus he has 60 texts out of 62,430 and for
412
D 4.4.19 (13 ed.); 28.3.6.9 (10 Sab.)
413
D 3.3.37.1 (9 ed.); 14.4.5.7 (29 ed.); 42.5.9.5 (62 ed.); 43.3.1.8 (67 ed.); 43.24.1.3 (71 ed.);
28.2.6 pr. (3 Sab.); 7.1.13.5 (18 Sab.); 34.2.19.2 (20 Sab.); 47.2.17.1 (40 Sab.) but 36.1.46.1 (Marc.
15 dig.) cf. 41.9.1.2 (31 Sab: est quaestio volgata)
414
D 3.2.2.2 (6 ed.); 4.2.16.2; 4.4.20.1; 13.7.36 pr. (11 ed.); 4.6.26.9 (12 ed.); 13.7.13 pr (28 ed.);
47.10.13.7 (57 ed.); 29.2.25.2 (8 Sab.); 29.1.9.1 (9 Sab.); 38.16.1.1 (12 Sab.); 36.1.19.1 (15 Sab.);
30.41.7 (21 Sab.); 34.3.9 (24 Sab.); 26.8.5.3 (40 Sab.); 48.18.4 (3 disp.); 49.1.10.4 (8 disp.);
40.5.24.5 (5 fid.); 42.1.59.1 (4 omn. trib.); 50.13.1.12 (8 omn. trib.)
415 D 34.3.7 pr. (23 Sab.) 416 D 1.1.4 (1 inst.)
417 D 3.2.2.2 (6 ed.); 3.3.39.6 (9 ed.); 37.12.1.4 (45 ed.); 43.24.7.4 (71 ed.); 49.14.25 (19 Sab.);
8.2.3 (29 Sab.); 12.7.1 pr (43 Sab.); 50.15.1.2, 3 bis, 5, 10, 11 (1 cens)
418 D 1.19.1.2 (16 ed.); 47.10.7.1 (57 ed.) 419 Above n.310–2 420 VIR 1.94.34
421 Above n.321–3
422 D 11.4.1.5; 39.2.4.1 (1 ed.); 2.4.4.2 (5 ed.); 3.1.1.9 (6 ed.); 3.2.6 pr. (6 ed.); 3.5.3.2 (10 ed.);
4.8.21.3 (13 ed.. = iudex); 39.2.15.5 (53 ed.); 43.20.1.39 (70 ed.); 43.30.3.6 (71 ed.); 29.2.30.1 (8
Sab.); 28.2.12 pr. (9 Sab.); 7.1.13.8 (18 Sab.); 21.1.25.2 (1 ed. cur.) cf. Frag. Vat. 321 (8 ed.)
423 D 15.1.3.7 (29 ed.); 28.1.20.3 (1 Sab.); 29.2.30.1 (8 Sab.); 21.1.19.4 (1 ed. cur.)
424 D 5.3.25.8 (15 ed.) 425 D 43.23.1.8 (71 ed.)
426 Honoré–Menner (1980) fiche 49 ACCIPERE DEBEMUS but D. 35.2.74 (Gaius 3 leg.ed.urb.);
6.1.78 (Lab. 4 pith. = Paul); 20.6.8.11 (Marci. 1 hyp.); 46.3.98.3 (Paul 15 quaest.); 27.1.45.3
(Tryph. 13 disp.) cf. VIR 1.95–6
427 D 14.6.7.3 (29 ed.); 38.17.1.12 (12 Sab: ? debemus)
428 D 4.8.11.2 (13 ed.); 11.5.1.2 (23 ed.)
429 D 26.2.3.1 (35 ed.); 38.5.1.4 (44 ed.); 38.8.1.6 (46 ed.); 50.16.99.1 (1 off. cons.)
430 Honoré–Menner (1980) fiche 49 ACCIPIMUS but D. 9.2.45 pr (Paul 10 Sab.); 43.3.2.1 (Paul
sic accipiendum est and variants (accipienda sunt, est accipiendum, sunt accipi-
enda), 20 instances out of 21.431 It is a remarkable concentration.
Putare is another verb that elicits Ulpian’s favour. The forms et puto, et
putem, et magis puto, et ego puto, et non puto, and et verum/verius puto have
been mentioned already.432 To these can be added ut puta, which occurs in the
Tyrian’s writings 298 times out of 317.433 Of the remaining 19 texts Paul has
13. Despite this distribution, several scholars have had the temerity to list it
as unclassical.434 Other Ulpianic forms of putare include puto autem,435 puto
tamen,436 putavimus tamen,437 and et (recte) putat followed by the name of the
author.438
No other verbs are as Ulpianic as accipere and putare. But there are a num-
ber that occur once or more in his texts and not at all, or much more rarely,
in other Digest writers. I list them alphabetically:
adaequare,439 addere,440 artare,441 capessere,442 causari,443 commonere,444 commo-
vere,445 commundare,446 concumbere,447 confringere,448 conivere,449 conqueri,450
431
D 50.16.3 pr (2 ed.); 9.4.3; 50.16.6.1 (3 ed.); 2.14.7.5 (4 ed.); 3.5.3.1 (10 ed.); 11.1.4.1. (22
ed.); 13.5.1.1 (27 ed.); 19.2.19.2 (32 ed.); 27.4.3 pr (36 ed.); 37.5.8 pr; 37.6.1.23 (40 ed); 39.4.3.1
(55 ed.); 40.12.10 (55 ed.); 42.1.5.1 (59 ed.); 43.8.2.32 (68 ed.); 39.2.24 pr (81 ed.); 28.1.21.2 (2
Sab.); 29.2.30.3 (8 Sab.); 27.1.3 (1 off. pr.tut.); 49.3.1 pr (1 appell.) cf. Frag. Vat.186 (1 off. pr.
tut.) but 47.2.4 (Paul 9 Sab.)
432
Above n.48–53
433
Honoré–Menner (1980) fiche 74 UT PUTA and cf. Frag. Vat. 177 (17 Sab.) but D 11.7.4.3,
2.7.4 pr (Paul 4 ed.); 4.8.19.2; 4.8.28 (Paul 13 ed.); 10.1.4.8 (Paul 23 ed.); 2.14.9 pr (Paul 62 ed.);
19.1.4.1 (Paul 5 Sab.); 12.6.15 pr. (Paul 10 Sab.); 9.4.31 (Paul 7 Plaut: delete, Mommsen); 31.8.3
(Paul 9 Plaut.); 45.1.91.1 (Paul 17 Plaut.); 12.6.21 (Paul 3 quaest.); 28.6.38.1 (Paul 1 sec. tab.);
11.7.4.3 (Pap. 8 quaest.), 22.1.3.2 (Pap. 20 quaest.), 45.2.9 pr (Pap. 27 quaest.); 48.5.39 pr (Pap.
36 quaest.); 5.1.45 pr (Pap. 3 resp., Mommsen); 14.2.4.2 (Call. 2 quaest.) cf. ut . . . puta 47.2.22.1
(Paul 9 Sab.); puta ut 19.5.5.2 (Paul 5 quaest.). Noted by Kalb (1890) 130
434
Citati (1927) 72: Eisele, Bonfante, Pringsheim, Donatuti, and others
435
D 50.1.27.2 (2 ed.); 3.1.6 (6 ed.); 6.1.37 (17 ed.); 15.3.3.10 (29 ed.); 19.5.20.1 (32 ed.);
44.4.4.22 (76 ed.); 38.17.2.37 (13 Sab.); 7.1.70.4 (17 Sab.)
436 Eight of 9 texts: D 5.3.25.5, 15 (15 ed.); 11.7.14.11 (25 ed.); 16.3.1.12 (30 ed.); 49.14.6 pr (63
ed.); 33.9.1 (24 Sab.); 48.3.4 (9 off. proc.); 49.1.3.3 (1 appell.) cf. Frag.Vat. 71 (17 Sab.); 198 (1
off. pr. tut.) but D. 4.6.13.1 (Paul 12 ed.)
437 D 4.4.3.2 (11 ed.)
438
D 2.2.3.1 (3 ed.); 3.1.1.10 (6 ed.); 4.8.21.11 (13 ed.); 5.3.13.4 (15 ed.); 12.6.26.13 (26 ed.);
16.3.1.11 (30 ed.); 39.1.3.2 (52 ed.); 28.5.17.4 (7 Sab.); 38.17.2.44 (13 Sab.); 38.4.5.1 (14 Sab.);
7.1.12.5; 7.2.1.1 (17 Sab.); 7.6.1.3 (18 Sab.); 47.2.12.2 (28 Sab.)
439 D 30.41.13 (21 Sab.)
440 D 3.2.6.6 (6 ed.); 19.2.13.5; 19.2.19 pr (32 ed.); 47.8.2.23 (56/1 ed.); 41.2.13.1 (72 ed.);
2.11.2.8 (74 ed.); 30.50.1 (24 Sab.); 32.11.8 (2 fid.); 36.1.13 pr (4 fid.); 40.5.24.14 (5 fid.) but
2.15.12 (Cels. 3 dig.); 40.7.34.1 (Pap. 21 quaest.)
444 445
D 13.6.12.1 (29 Sab.) D 9.1.1.4 (18 ed.); 11.7.6 pr (25 ed.); 39.6.2 (32 Sab.)
446 447
D 34.2.25.10 (44 Sab.) D 1.6.6 (9 Sab.)
448 449
D 9.2.27.31 (18 ed.); 10.4.9 pr (24 ed.); 39.2.24.9 (81 ed.) D 40.1.4.1 (6 disp.)
450
D 29.5.1.3 (50 ed.); 47.10.7.2 (57 ed.)
66 2. The Oral Style
(vii) Nouns and nominal phrases. There is nothing special to note apart from a
certain liking for diminutives: alicula,516 domuncula,517 frivusculum,518 locu-
lus,519 operula,520 sarcinula,521 tabernula,522 viaticulum,523 vulnusculum.524 (Sig-
naculum,525 though confined to Ulpian, is not a diminutive but a deverbative.)
The remaining nouns may be listed alphabetically:
abusus,526 adfirmator,527 adgressus,528 adiutorium,529 adparitio,530 adpendix,531
adpulsus,532 adsessorium,533 alternatio,534 apertura,535 apex,536 audacia,537 calcula-
tor,538 calliditas,539 colloquium,540 commendatio,541 comminatio,542 concubitus,543
contaminatio,544 corruptela,545 corruptor,546 curiositas,547 decus,548 dedecus,549 vera
511
D 18.7.1 (32 ed.); 26.7.7 (35 ed.); 29.1.13.3 (45 ed.); 38.8.1.8; 38.8.1.9 (46 ed.); 47.9.3 pr
(56/1 ed.); 43.19.1.2 (70 ed.); 44.2.7.2 (75 ed.); 40.7.2 pr (4 Sab.); 26.2.16 pr (39 Sab.); 40.5.45.2 (5
disp.); 40.1.4.12 (6 disp.); 23.2.27 (3 Iul. Pap.)
512
D 42.1.4.1 (58 ed.); 42.8.6.1 (66 ed.); 39.5.7.6 (44 Sab.); 50.4.6.1 (4 off. proc.); 10.2.49 (2
disp.); 28.5.35.3 (4 disp.); 33.4.2.1 (5 disp.)
513
D 11.1.11.4 (22 ed.); 37.4.1.5 (39 ed.); 37.12.1 pr (45 ed.); 29.4.1.1 (50 ed.); 5.1.19.2 (60 ed.);
29.2.20.1 (61 ed.); 12.6.9 (66 ed.); 41.2.6 pr (70 ed.); 39.1.20.5; 43.29.3.3 (71 ed.); 21.1.59 pr (74
ed., twice); 26.8.5.3 (40 Sab.); 45.1.1.2 (48 Sab.); 45.1.38.22 (49 Sab.); 49.1.1.3 (1 appell.)
514
D 5.1.5 (5 ed.); 4.7.4.3 (13 ed.); 14.5.4.5 (29 ed.); 26.10.3.18 (35 ed.); 37.4.8.11 (40 ed.);
29.5.3.12 (50 ed.); 40.12.12.3 (55 ed.); 47.10.3.4 (56/2 ed.); 47.10.13.2 (57 ed.); 42.4.7.5 (59 ed.);
42.8.6.2 (66 ed.); 2.11.4.1 (74 ed.); 28.3.6.7 (10 Sab.); 26.1.3 pr (37 Sab.); 47.2.17.2 (40 Sab.);
45.1.38.22 (49 Sab.); 40.5.26.10 (5 fid.); 26.5.12 pr; 50.2.3.1 (3 off. proc.); 48.5.14.7 (2 adult.)
515
D 19.1.7 (Pomp. 10 Sab.); 38.2.25 (Iul. 1 Urs. Fer.); 14.6.14 (Iul. 12 dig.); 41.4.7.4 (Iul. 44
dig.); 20.4.9 pr (Afr. 8 quaest.); 16.1.13.2 (Gaius 9 ed. prov.); 27.10.5 (Gaius 9 ed. prov.); 26.2.1.1
(Gaius 12 ed. prov.); 26.7.57.1 (Scae. 10 dig.); 40.5.29 (Paul 3 fid.); 46.5.5 (Paul 48 ed.)
516 517 518
D 34.2.23.2 (44 Sab.) D 47.12.3.11 (25 ed.) D 24.1.32.12 (33 Sab.)
519 520 521
D 32.52.9 (24 Sab.) D 50.14.3 (8 omn. trib.) D 4.6.15.3 (72 ed.)
522 523
D 5.1.19.2 (60 ed.) D 5.1.18.1 (23 ed.) but interp. Citati (1927) 91: Solazzi
524 525
D 21.1.1.8 (1 ed. cur.) D 16.3.1.36 (30 ed.)
526
D 12.2.11.2 (22 ed.); 7.8.12.1 (17 Sab.); 7.5.5.1, 2 (18 Sab.)
527 528
D 4.4.13 pr (11 ed.); 27.7.4.3 (36 ed.) D 36.1.18.7 (2 fid.)
529 530 531
D 47.2.50.3 (37 ed.) D 21.2.50 (25 ed.) D 29.2.35 pr (9 Sab.)
532
D 8.3.5.1 (17 ed.); 43.20.1.18 (70 ed., twice); 34.1.14.3 (2 fid.); 8.3.1.1 (2 inst.)
533 D 2.14.12 (4 ed.); 47.10.5.8 (56/2 ed.)
534 D 11.3.9 pr (23 ed.); 13.4.2.3 (27 ed., twice); 47.10.7.4 (57 ed.)
535 536
D 28.5.3.4 (3 Sab.) D 17.1.29.4 (7 disp.)
537 538
D 39.4.12 pr (38 ed., twice) D 38.1.7.5 (28 Sab.); 50.13.1.6 (8 omn. trib.)
539 540
Below n.618 D 48.5.10.2 (4 adult.)
541 542
D 4.1.1 (11 ed.); 1.16.4.3 (1 off. proc.) D 26.7.7.7 (35 ed.); 37.14.1 (9 off. proc.)
543 544
D 35.1.15 (35 Sab.) D 48.5.2.3 (8 disp.)
545
D 11.3.9.1 (23 ed.); 49.14.29 pr (8 disp.)
546 547
D 11.3.9.3 (23 ed.); 11.3.11.1 (23 ed.); 49.14.20 pr (8 disp.) D 22.6.6 (18 Iul. Pap.)
548 549
D 3.1.1 pr (6 ed.); 39.1.20.10 (71 ed.) D 4.6.10 (12 ed.); 37.4.3.5 (39 ed.)
68 2. The Oral Style
550 D 2.14.7.14 (4 ed.); 4.3.9.3 (11 ed.); 12.1.11 pr (26 ed.) 551 D 42.8.10.10 (73 ed.)
552 D 22.1.33 pr. (1 off. cur. reip.) 553 D 11.7.14.3 (25 ed.)
554 555
D 22.5.12 (37 ed.); 46.8.12.2 (80 ed.) D 24.1.32 pr (33 Sab.)
556 D 1.1.1.1 (1 inst.) 557 D 19.2.15.5 (32 ed.) 558 D 19.5.20 pr, 1 (32 ed.)
559 D 19.2.15.2 (32 ed.); 48.5.16.6 (2 adult.) 560 D 50.16.195.4 (46 ed.)
561 562 563
D 21.1.38.7 (2 ed. cur.) D 13.7.11.5 (28 ed.) D 33.7.12.19 (20 Sab.)
564
D 26.7.23 (9 ed.); 3.5.5.13 (10 ed.); 26.10.3.9; 26.7.5.2; 26.10.3.10 (35 ed.); 27.3.1.13;
27.5.1.5 (36 ed., interp.); 46.3.14.1 (30 Sab.)
565 566
D 28.5.35.2 (4 disp.) D 43.23.1.2 (71 ed.)
567
D 50.13.1.3 (8 omn. trib.); 21.1.4.2 (1 ed. cur.)
568
D 47.20.3.1 (8 off. proc.) cf. Collatio 15.2.1 (2 off. proc.)
569 570
Collatio 15.2.1 (7 off. proc.) Frag. Vat. 123 (1 excus.)
571 572 573
D 48.5.30 pr (4 adult.) D 33.9.1 (24 Sab.) D 19.1.13.3 (32 ed.)
574 D 22.1.33 pr (1 off. cur. reip.) 575 D 9.1.1.6 (18 ed.)
576 D 47.11.5 (5 off. proc.) cf. Collatio 15.2.5 (7 off. proc.) 577 D 28.1.20.9 (1 Sab.)
578 D 11.1.11.7 (22 ed.) 579 D 4.4.7.3 (11 ed.)
580 D 37.15.7.5 (10 ed.); 15.1.3.9 (29 ed.) 581 D 48.18.1.20 (8 off. proc.)
582 D 1.12.1.8 (1 off. pr. urb.)
583 D 22.1.3.4 (15 ed.); 14.1.1.15 (28 ed.); 27.9.5.9 (35 ed.); 7.1.7.1 (17 Sab.)
584 D 42.4.7.4 (59 ed.) 585 D 42.4.7.13 (59 ed.) 586 D 13.6.3.6 (28 ed.)
587 588 589
D 30.39.7 (21 Sab.) D 43.12.1.14 (68 ed.) D 33.9.3.11 (22 Sab.)
590 591 592
D 47.4.1.1 (38 ed.) D 14.4.5.16 (29 ed.) D 28.5.35.1 (4 disp.)
593 594
D 16.1.8 pr (29 ed.); 21.1.21 pr. (1 ed. cur.) D 14.3.5.2 (28 ed.)
595 596
D 50.16.10 (6 ed.); 26.10.4.2 (1 omn. trib.) D 26.7.7 pr (35 ed.)
597 D 34.2.19.10 (20 Sab.)
598 Eleven of 12 texts: D 38.6.1.3 (44 ed.); 38.9.1.7, 11 (49 ed.); 29.2.13 pr, 2; 29.2.17.1;
29.2.21.3 (7 Sab.); 38.16.2.7 (13 Sab.); 24.1.5.13 (32 Sab.); 49.17.9 (4 disp.); 36.1.15.5 (4 fid.) but
19.2.25.2 (Gaius 10 ed. prov.)
599 600 601
D 41.2.13.2 (72 ed.) D 5.1.2.3 (3 ed.) D 50.12.2.2 (1 disp.)
602 603
D 37.11.2.4 (41 ed.); 50.15.1 pr (1 cens.) D 1.7.17.4 (26 Sab.)
604 605
D 2.12.1.1 (4 omn. trib.) D 47.10.7.8 (57 ed.); 47.14.1.1 (8 off. proc.)
606
D 26.8.1.1 (1 Sab.); 34.2.25.10 (44 Sab.); 21.1.14.4 (1 ed. cur.)
607
D 23.2.43.1 (1 Iul. Pap.)
608
D 4.4.13 pr (11 ed.) but interp. Citati (1927) 83: Pringsheim
609
D 9.2.9.1 (18 ed.)
610 D 24.3.22.8 (33 ed.) 611 D 19.2.13.11 (32 ed.) but interp. Citati (1927) 86: Kalb etc.
612 D 26.7.7.3 (35 ed.) but interp. Citati (1927) 86: Beseler, Kunkel
613 D 47.10.15.41 (57 ed.) 614 D 9.2.27.17 (18 ed.); 21.1.14.8 (1 ed. cur.)
615 D 14.3.11.5 (28 ed.) 616 Frag. Vat. 148 (1 excus.)
617 D 34.2.23.2 (44 Sab.)
2. The Oral Style 69
All twenty-five texts with calliditas are attributed to Ulpian.618 Even if two or
three are compilatorial, or are cited from another author or a rescript,619 the
concentration is striking. He clearly disliked craftiness, and thought of him-
self as a candid person.620
618 D 2.14.7.9 (4 ed.); 4.1.1; 4.3.1 pr, 2; 4.3.7.10; 4.4.3.1 (11 ed.); 11.3.3 pr (23 ed.); 10.4.11.1
(24 ed.); 15.3.3.9; 16.1.2.3 (29 ed.); 50.17.47 pr (33 ed.); 17.1.6.7; 50.14.2 (31/1 ed.); 24.3.22.8 (33
ed.); 47.4.1.1 (38 ed.); 29.4.1 pr (50 ed.); 40.12.12.3 (55 ed.); 40.12.14 pr (55 ed.); 42.6.1.5 (64 ed.);
43.24.1.1; 43.29.3.5 (71 ed.); 47.20.3.1 (8 off. proc.); 21.1.1.2 (1 ed. cur.); 17.1.29.5 (7 disp.); 49.4.1
pr (1 appell.)
619 D 4.3.1.2 (11 ed.); 17.1.6.7 (31 ed.); Crifò (1985) 607
620 Cf. n.640 (dolosus), 736 (callide), 740 (dolose). Frier (1984) 859 thinks that this inference
‘descends to bathos’. But a person who is continually on the lookout for a particular vice must
feel strongly about it
621 D 48.5.18.1 (2 adult.); 48.5.30.8 (4 adult.) 622 D 22.1.33 pr (1 off. cur. reip.)
623 D 19.2.15.2 (32 ed.)
624 D 4.4.3 pr (11 ed.); 22.1.33 pr (1 off. cur. reip.); 50.9.4 pr (1 off. cur. reip.) cf. 36. 1.67.2
5.12.11
628 D 9.1.1.4 (18 ed.); 21.1.4.3 (1 ed. cur.)
629
Twenty-seven of 28 texts: D 4.4.3.3, 4, 5, 7, 10; 4.4.5; 4.4.7.1, 4, 5, 7; 4.4.9.2; 4.4.11.2, 3, 4,
5, 6; 4.4.13 pr ; 4.18.1; 13.7.36.1 (11 ed.); 4.6.28.1 (12 ed.); 4.1.6 (13 ed.); 12.2.9.4 (22 ed., twice);
15.1.5 pr (29 ed.); 16.3.1.42 (30 ed.); 15.1.36 (2 disp., twice) but 4.4.29 pr (Mod. 2 resp.). Perhaps
this word is too contextual to count
630 D 43.23.1.4 (71 ed.) 631 D 43.8.5.32 (68 ed.) 632 D 34.2.23.2 (44 Sab.)
633 D 36.1.23.3 (5 disp.) 634 D 21.2.37.1 (32 ed.); 1.16.9.2 (2 off. cons.)
635 Frag. Vat. 156 (1 excus.)
636 D 47.10.7.7 (57 ed.); 1.12.1.10; 22.1.33 pr (1 off. cur. reip.)
637 D 22.6.6 (18 Iul. Pap.) 638 D 22.6.6 (18 Iul. Pap.)
639 D 21.1.37 (1 ed. cur.) 640 D 4.3.1 pr (11 ed.)
641 D 44.1.2.2 (74 ed.) 642 D 34.2.25.11 (44 Sab.)
643 D 2.1.3 (2 off. quaest.) 644 D 21.1.1.9 (1 ed. cur.)
645 D 2.12.2 (5 ed.) 646 D 39.6.5 (2 inst.)
647 D 35.2.62.1 (1 Iul. Pap.) 648 D 14.1.1.6 (28 ed.) 649 D 21.1.4.2 (1 ed. cur.)
650 D 21.1.12.2 (1 ed. cur.) 651 D 40.7.9.1 (28 Sab.) 652 D 21.1.6.1 (1 ed. cur.)
653 D 43.20.1.43 (70 ed.) 654 D 33.1.3.2 (24 Sab.) 655 D 7.5.5.1 (18 Sab.)
656 Collatio 12.5.2 (8 off. proc)
657 Frag. Vat. 75.5 (17 Sab.) but interp. Citati (1927) 46: Beseler
658 D 28.8.7.3 (60 ed.) 659 D 22.1.33 pr. (1 off. cur. reip.)
660 D. 28.1.9 (45 ed.) 661 D 21.1.19.4 (1 ed. cur.)
70 2. The Oral Style
662 663
D 5.1.50.1 (6 fid.) D 34.9.9.1 (14 Iul. Pap.)
664 D 9.2.27.30 (18 ed.); 23.3.34 (43 ed.); 43.24.15 pr (71 ed.) but 46.3.78 (Iav. 11 Cass.) cf.
Frag. Vat. 269 (Ulp. 46 Sab.)
665 D. 4.8.3.1 (13 ed.); 7.1.13.8 (18 Sab.)
666 D 45.1.1.5 (twice); 46.4.8 pr (48 Sab.); 37.11.6 (8 disp.)
667 D 3.1.1.5 (6 ed.) 668 D 21.1.17.14 (1 ed. cur.)
669 D 1.16.9.3 (2 off. proc.)
670 D 11.1.11.8 (22 ed.); 14.4.7 pr (29 ed.); 37.6.1.11 (40 ed.); 18.6.4 pr (28 Sab.); 29.3.10.2 (13
Iul. Pap.)
671 672
D 7.1.13.4 (18 Sab.) D 50.13.1.11 (8 omn. trib.); 48.5.16.1 (2 adult.)
673 674
D 43.8.2.42 (68 ed. = 50.17.150) D 21.1.4.3 (1 ed. cur.)
675 676
D 38.2.6 pr (43 ed.) D 13.7.24.1 (1 ed., thrice)
677 678
D 39.2.15.13 (53 ed.); 7.8.4 pr (17 Sab.) D 47.10.15.27 (57 ed.)
679 680
D 27.8.1 pr, 4 (36 ed.) D 35.3.1.5 (79 ed.)
681 682
D 26.7.3.8 (35 ed.); 27.8.4 (3 disp.) D 28.8.5.1 (60 ed.)
683 684
D 21.1.4.9 (1 ed. cur.) D 26.7.9.6 (36 ed.)
685 686
D 33.7.12.31 (20 Sab.) D 3.2.11.3 (6 ed.)
687 688
D 36.1.13.4 (4 fid.) D 48.19.8.11, 12 (9 off. proc.)
689 690
D 3.3.78 pr (20 Sab.) D 5.3.13.5 (15 ed.)
691 692
D 34.2.19.17 (20 Sab.) D 49.1.14 pr (14 ed.)
693 D 11.7.20.1 (25 ed.) 694 D 4.4.11.4 (11 ed.); 26.10.1 pr (35 ed.)
695 D 46.5.1.5 (77 ed.) 696 D 47.4.1.1 (38 ed.)
697 D 46.5.1.4 (70 ed.) 698 D 47.10.7.7 (57 ed.)
699 D 33.7.8.1 (20 Sab.) 700 D 3.2.13.7 (6 ed.)
701 D 43.13.1.8 (68 ed.) 702 Below n.719–721 703 D 40.9.12.6 (4 adult.)
704 D 50.15.1 pr (1 cens.); 50.3.1 pr (3 off.proc.); 1.13.1 pr (1 off. quaest.);
705 706
D 37.10.1.5 (41 ed.) D 7.4.5.2 (17 Sab.); 40.5.24.9 (5 fid.)
707 708
D 5.1.19.2 (60 ed.); 38.2.1 (42 ed.) D 39.1.5.4 (52 ed.)
709 710
D 12.2.3 pr (22 ed.) D 42.4.7.2 (59 ed.)
711 712
D 3.1.1.5 (6 ed.) D 23.4.11 (34 ed.)
713 714
D 24.3.22.8 (33 ed.) D 3.1.1.5 (6 ed.)
715 716 717
D 50.15.1 pr (1 cens.) D 50.15.1 pr (1 cens.) D 27.1.19 (35 ed.)
718
D 19.1.11.16 (32 ed.); 32.11.20 (2 fid.); 36.1.15.3 (4 fid.). Est verissimum: 9.2.5.2 (18 ed.);
9.2.27.3 (18 ed.); verissimum videtur: 11.1.11.12 (22 ed.)
2. The Oral Style 71
(ix) Adverbs and adverbial phrases. The adverbial phrase eo loci723 ‘at that
place’ and similar phrases (alio loci,724 aliquo loci,725 quo loci,726 eodem loci,727
certo loci728) are notably Ulpianic.729 Many Ulpianic adverbs are related to
the verb or noun forms already mentioned. Thus, abusive730 is related to
abusus,731 acerbe732 to acerbus,733 benigne dicere734 to benignum est,735 cal-
lide736 to calliditas,737 contumeliose738 to contumeliosus,739 dolose740 to dolo-
sus,741 minus dubitanter742 and indubitanter743 to a number of expressions
rejecting doubt,744 incaute745 to cautior,746 notabiliter747 to notabilis,748
sobrie749 to sobrietas,750 and sumptuose751 to sumptuosus.752 There is a group
of negative adverbs: (non) difficile,753 gravate,754 idonee,755 mediocriter,756
719
Above n.336 (26 texts)
720
Fifteen texts: D 4.6.26.9 (12 ed.); 11.1.11.8 (22 ed.); 10.4.11 pr (24 ed.); 13.6.7.1 (28 ed.);
14.5.2.1 (29 ed.); 19.1.13.20 (32 ed.); 47.6.1.2 (38 ed.); 37.4.3.4 (39 ed.); 37.8.1.1 (40 ed.); 38.2.8 pr
(43 ed.); 43.19.3.7 (70 ed.); 2.11.2.1 (74 ed.); 36.3.1.19 (79 ed.); 42.1.57 (2 disp.); 17.1.29.6 (7 disp.)
cf. est aequissimum 3.1.1.6 (6 ed.); 14.4.5.13 (29 ed); 19.2.9.1 (32 ed.); 29.5.3.17 (50 ed.)
721
Honoré–Menner (1980) fiche 49 AEQUISSIMA, AEQUISSIMUM; VIR 1.296 but D 16.2.5 (Gaius
9 ed. prov.); 17.1.22.4 (Paul 32 ed.); 14.2.2 pr (Paul 1 SC Tert.); 48.20.7 pr (Paul 1 port. lib.
damn.); 38.17.5 pr (Paul 1 SC Tert.); 36.1.1.2 (Ulp. 3 fid. citing SC Trebellianum) cf. Gaius, Inst.
4.71. Interp. Citati (1927) 8: Bonfante, Wylie
722
D 47.6.1 pr (38 ed.); 43.15.1.1 (68 ed.); 43.21.1.1 (70 ed.); 43.26.6.4 (71 ed.); 10.3.21 (30
Sab.)
723
Nineteen of 20 texts: D 50.1.27.3 (2 ed.); 4.8.21.10 (13 ed.); 9.2.29 pr (18 ed.); 10.4.11.1 (24
ed. twice); 13.4.2.8; 13.5.16.1 (27 ed.); 29.5.1.27 (50 ed.); 39.1.5.2, 3 (52 ed.); 5.1.19.2 (60 ed.);
11.8.1.8 (68 ed.); 43.20.1.17 (70 ed.); 8.2.17 pr (29 Sab.); 47.2.3.2 (41 Sab.); 48.19.9.6 (10 off.
proc.); 50.16.199 pr (8 omn. trib.); 5.1.52.3, 4 (6 fid.); 48.5.10.2 (4 adult) but 14.2.10.1 (Labeo 1
pith. Paul epit.)
724 725
D 4.8.21.10 (13 ed.) D 18.7.1 (32 ed.)
726
D 16.3.5.1 (30 ed.) but 6.1.6 (Paul 6 ed.)
727 728
D 42.4.7.13 (59 ed.) but 22.2.1 (Mod. 10 pand.) D 5.1.19.1, 2 (60 ed.)
729 Kalb (1890) 134–5
730 D 50.1.1.1 (2 ed.); 50.16.15 (10 ed.); 29.3.2.1 (50 ed.); 47.10.15.40 (57 ed.)
731 Above n.526 732 D 43.30.1.5 (71 ed.) but interp. Citati (1927) 1: Eisele
733 Above n.622
734 D 38.2.14.2, 8 (45 ed.); 36.4.1 pr (52 ed.); 42.5.24.3 (63 ed.); 7.4.1.2 (17 Sab.); 35.1.10 pr (23
Sab.); 24.1.34 (43 Sab.); 50.2.2.3 (1 disp.) but interp. Citati (1927) 14: Beseler etc
735 Above n.191
736 D 4.3.9.1 (11 ed.); 4.8.31 (13 ed.); 16.1.2.3 (29 ed.); 47.8.2.8 (56/1 ed.); 26.8.5.4 (40 Sab.)
737 Above n.539, 618 738 D 7.1.27.1 (18 Sab.) but Gaius, Inst 1.141
739
Above n.636
740
D 4.4.3.1 (11 ed.); 4.8.31 (13 ed.); 5.3.25.5 (15 ed.); 10.4.9.2 (24 ed.); 11.6.3 pr (24 ed.);
44.4.2.5 (76 ed.); 42.5.31.2 (2 omn. trib.)
741 Above n.640 742 D 15.1.9.5 (28 ed.) 743 D 37.11.2.7 (41 ed.)
744 Above n.65, 78, 168–173
745 D 28.4.1.1 (15 Sab.) cf. Collatio 12.5.2 (8 off. proc: incautus?)
746 Above n.697; incautus, above n.656, 745 747 D 47.10.15.26 (56/2 ed.)
748 Above n.667 749 D 4.4.11.4 (11 ed.); 26.5.12.1 (3 off. proc.)
750 Above n.603 751 D 39.2.37 (42 Sab.) 752 Above n.682
753 D 1.18.13 pr (7 off. proc.) 754 D 1.16.7 pr (2 off. proc.)
755 D 27.8.1.17 (36 ed.) 756 D 47.10.7.2 (57 ed.) but interp. Citati (1927) 56: Beseler
72 2. The Oral Style
(14 ed.); 11.7.6 pr (25 ed.); 27.2.1.3 (34 ed.); 12.3.4 pr; 26.7.9 pr; 27.3.1.15 (36 ed.); 47.4.1.7 (38
ed.); 37.5.5.6 (40 ed.); 47.10.13.7 (57 ed.); 43.24.13.5 (71 ed.); 42.8.10.13, 14 (73 ed.); 44.4.4.26 (76
ed.); 28.3.6.8 (10 Sab.); 34.2.9 (24 Sab.); 47.2.12.2 (29 Sab.); 48.18.1.7 (8 off. proc.); 50.15.4.10 (3
cens.); 13.7.26 pr (3 disp.); 29.7.1 (4 disp.); 40.5.24.21 (5 fid.); 37.14.16 pr (10 Iul. Pap.); 49.2.1.4
(1 appell.) but 39.4.16.4 (Marci. 1 del.); 48.15.3 pr (Marci 1 pub. iud.); 48.24.2 (Marci 2 pub. iud.)
cf. Kalb (1890) 132, who points out that two or three instances seem to have been treated as
enough to justify saepissime
770 D 4.4.7.8 (11 ed.); 12.1.9.8 (26 ed.); 26.10.1 pr (35 ed.); 43.26.6.4 (71 ed.); 23.1.4.1 (35 Sab.)
but 20.1.15.1 (Gaius 1 hyp.); interp. Citati (1927) 23: Pernice etc.
771 D 5.1.2.5 (3 ed.); 2.14.1.3; 2.14.7.2, 10; 2.14.10 pr; 36.1.17.6 (4 ed.); 3.5.9.1 (10 ed.); 4.2.9.1;
4.3.7 pr (11 ed.); 4.8.21.11 (13 ed.); 10.2.18.4 (19 ed.); 10.3.7.13 (20 ed.); 10.4.3.11 (24 ed.);
15.1.9.4 (29 ed.); 13.7.24 pr; 16.3.1.33 (30 ed.); 18.3.4.2 (32 ed.); 25.3.1.10 (34 ed.); 13.1.12 pr, 1
(38 ed.); 29.5.1.12; 29.5.3.30; 37.8.1.16 (50 ed.); 39.2.15.28 (53 ed.); 12.6.23 pr (3 Sab.); 7.2.1.3 (17
Sab.); 32.52.7 (24 Sab.); 18.2.4.5 (28 Sab.); 21.2.21.1 (29 Sab.); 17.2.14 (30 Sab.); 24.1.7.4 (32
Sab.); 24.3.14.1 (36 Sab.); 9.2.41.1; 47.2.71 (41 Sab.); 12.6.43; 17.1.19 (43 Sab.); 29.2.40 (4 disp.);
35.2.82 (8 disp.); 40.5.30.14 (5 fid.); 1.1.1. pr (1 inst.) but D 22.1.19 (Gaius 6 XII tab.); 2.2.4
(Gaius 1 ed. prov.); 15.1.40 pr (Marci. 5 reg.) cf. 46.3.103 (Maec. 2 fid: elegantissime); 26.7.61
(Pomp. 20 epist: elegantius); VIR 2.456
772 773 774
D 1.16.6.3 (1 off. proc.) D 4.4.7.8 (11 ed.) D 30.50.2 (24 Sab.)
775 776
D 45.1.1.6 (41 Sab.) D 47.2.93 (38 ed.) but interp. Citati (1927) 24: Albertario
777
D 3.3.13 (8 ed.); 4.4.7.8 (11 ed.); 4.8.15 (13 ed.); 48.18.1.26 (8 off. proc.)
778 779
D 33.6.9.3 (23 Sab.) D 27.2.3.3 (36 ed.)
780
Above n.96 (et generaliter), n.167.308–9 (generaliter dicendum est etc.)
781
D 11.7.14.7 (25 ed., twice); 29.2.20.1 (61 ed.); 42.6.1.10 (64 ed.); 43.16.3.6 (69 ed.); 38.4.1.6
(14 Sab.); 24.1.32.5 (33 Sab.); 33.4.2 pr (5 disp.)
782
D 24.1.22 (2 Sab.)
2. The Oral Style 73
783
D 12.4.3.8 (26 ed.); 29.2.20.1 (61 ed.); 42.6.1.10 (64 ed.); 43.16.3.6 (69 ed.); 38.4.1.6 (14
Sab.); 24.1.32.5 (33 Sab.); 33.4.2 pr (5 disp.)
784 785
D 21.1.17.3 (1 ed. cur.) D 9.2.7.8 (18 ed.); 11.6.1.1 (24 ed.)
786 787 788
D 1.16.9.5 (2 off. proc.) D 4.4.18.1 (11 ed.) D 26.10.3.17 (35 ed.)
789
D 40.2.20 pr (2 off. cons.)
790
D 1.13.1.3 (1 off. quaest.) but interp. Citati (1927) 47: Pringsheim
791
D 16.2.10.2 (63 ed.)
792
Thirteen of 14 texts: D 3.3.1.2 (9 ed.); 10.4.1 (24 ed.); 27.3.1.4 (36 ed.); 37.8.1.10 (40 ed.);
28.7.8 pr (50 ed.); 42.4.3 pr (59 ed.); 20.1.6 (73 ed.); 36.3.14.1 (79 ed.); 1.16.4.6 (1 off. proc.);
48.20.6; 48.22.7.22 (10 off. proc.); 50.16.99.2 (1 off. cons.); 40.9.12.1 (4 adult.) but 1.2.2.44
(Pomp. 1 enchir.) cf. Kalb (1890) 132
793
Six of 7 texts: D 42.1.2 (6 ed.); 4.4.3 pr; 4.4.18.1 (11 ed.); 42.6.1.1 (64 ed.); 1.9.12 pr (2 cens.);
26.10.7.3 (1 omn. trib.) but 1.3.5 (Cels. 17 dig.)
794
D 26.10.3.16 (35 ed.)
795
All eleven texts: D 4.4.7 pr (11 ed.); 9.2.7.1 (18 ed.); 26.7.5.10 (35 ed.); 47.4.1.2 (38 ed.);
42.6.1.15 (64 ed.); 43.16.3.17 (69 ed.); 43.24.11.12; 50.16.63 (71 ed.); 46.8.12.3 (80 ed.); 47.2.46 pr
(42 Sab.); 48.21.1 (8 disp.)
796
D 5.3.9 (15 ed.); 15.3.3.2 (29 ed.); 30.7.1.1 (51 ed.); 7.1.25.5 (18 Sab.) but interp. Citati
(1927) 77: De Francisci
797 798
D 5.2.8.6 (14 ed.) D 26.7.7.2; 26.10.3.5 (35 ed.)
799 800
D 7.1.15.2 (18 Sab.); 24.3.14 pr (36 Sab.) D 13.1.10.2 (38 ed.); 34.1.14 pr (2 fid.)
801 802 803
D 42.4.5.1 (59 ed.) D 48.19.6 pr (9 off. proc.) D 1.9.1.1 (62 ed.)
804
D 39.2.4.5 (1 ed.); 24.3.21.6 (3 disp.); 1.12.1.8 (1 off. pr. urb.) but interp. Citati (1927) 90:
Albertario
805
D 39.3.6.3 (53 ed.); 40.12.1.1 (54 ed.); 43.29.3 pr (71 ed.); 44.5.1.12 (76 ed.); 36.3.1.11 (79
ed.); 24.1.33.1 (36 Sab.); 45.1.1.3 (48 Sab.)
806 807 808
D 27.9.5.10 (20 Sab.) D 37.9.1.18 (41 ed.) D 25.5.1.2 (34 ed.)
809 D 15.1.19.1 (29 ed.) 810 D 11.7.14.14 (25 ed.)
811 D 12.4.3.8 (26 ed.); 49.15.15 (12 Sab.) 812 D 29.2.42 pr (4 disp.)
813 D 50.5.13.3 (23 ed.) 814 D 9.2.29.7 (18 ed.) 815 D 1.6.2 (8 off. proc.)
816 817
D 28.7.8 pr (50 ed.) D 37.10.3.5 (41 ed.) cf. Collatio 11.7.5 (8 off. proc.)
818
D 10.2.18.1 (19 ed.); 17.1.12.14 (31/1 ed.); 29.4.10 pr (50 ed.); 47.8.4.3 (56/1 ed.); 23.3.5.9
(31 Sab.); 40.4.13.2 (5 disp.); 23.2.43.3 (1 Iul. Pap.)
819
D 17.1.6.7 (31/1 ed)
820
Eight of 10 texts: D 13.7.36 pr (11 ed.); 9.2.23.4 (18 ed.); 19.2.19.6 (32 ed.); 37.9.1.11;
37.10.3.13 (41 ed.); 39.2.15.10 (53 ed.); 42.4.7.7 (59 ed.); 39.2.40.1 (43 Sab.) but 35.1.62.1 (Ter.
Clem. 4 Iul. Pap.); 34.2.34.2 (Pomp. 9 Q. Muc.)
821 822
D 4.2.14.5 (11 ed.) D 43.24.15.4 (71 ed.)
74 2. The Oral Style
Chapters 5 and 10 use it to set up two lists, one of genuine, the other of spu-
rious compositions. It may be of interest here to sketch some other ways in
which the study of Ulpian’s style can be fruitful.
One is that a grasp of the elements of style can help us to recognize as
Ulpian’s a passage that the Digest attributes to another author. There is a
chapter of the Digest that deals with actions by and against corporate bodies
(universitates).823 A two-thirds majority was needed to authorize someone to
bring or defend an action in the name of a corporate body. How was the two-
thirds reckoned? A text from Ulpian On the Edict says that, according to
Pomponius, a father can vote for a son and a son for his father if they are both
members of the local council.824 The sentence continues with an excerpt
attributed to Paul, which says that persons in the same power (e.g. two sons
both in their father’s power) can vote for one another, since their act is done
as councillors, not in a domestic capacity.825 The same voting rule applies to
appointments to local office (honores):
D. 3.4.5 ULPIANUS libro octavo ad edictum Illud notandum Pomponius ait, quod et
patris suffragium filio proderit et filii patri,
D 3.4.6 pr PAULUS libro nono ad edictum item eorum, qui in eiusdem potestate sunt:
quasi decurio enim hoc dedit, non quasi domestica persona. quod et in honorum peti-
tione erit servandum.
The last sentence in reality continues Ulpian’s text, not Paul’s. The construc-
tion erit followed by the gerundive is common in Ulpian but very rare in
Paul.826 The passage in question continues with other Ulpianic phrases. For
instance, we find, a little later (3.4.6.1) parvi refert, which occurs in thirty-
seven texts of Ulpian and otherwise only once in Papinian and once (this text)
in Paul.827 Then there is a sentence (3.4.6.2) beginning et puto828 and another
short sentence (et constitui ei potest: 3.4.6.3) beginning with et. Indeed the
whole passage, from quod et in honorum petitione to the end of D 3.4.6, though
attributed to Paul, is really a passage from Ulpian that continues D 3.4.5.
Justinian’s compilers are responsible for the mistake. They normally took
Ulpian’s Edict as the fundamental text and inserted in it shorter passages
from Paul and Gaius when that seemed appropriate.829 At the end of the Paul
insertion they or the person who copied the text forgot to repeat the previous
text heading (Ulpianus libro octavo ad edictum) before going back to the text
of Ulpian. 830
This example throws light on how Justinian’s commissioners worked. The
next tells us something unsuspected about Paul. A resolution of the senate at
the instance of Marcus and Commodus laid down that a guardian could not
marry off his female ward to his son. Paul wrote a commentary on this Speech
823 D 3.4 824 D 3.4.5 (Ulp. 8 ed. = 9 ed., Lenel) 825 D 3.4.6 pr (Paul 9 ed.)
826 Above n.352–408 827 Above n.17 828 Above n.53
829 Honoré (1970) 250, (1978) 151–70 830 Honoré (1978) 166–70
2. The Oral Style 75
Is this genuine Ulpian? The proposition stated is obviously fair. If the buyer
is entitled to sue he should also be liable to be sued. But it is bold, and hence
suspect. Yet per contrarium is found twenty-nine times in Ulpian and only
twice in any other lawyer (both times in Marcianus, who is very close to
Ulpian).837 Quite apart from Ulpian’s known concern with equity, this sup-
ports the genuineness of the text.
Unlike other legal authors, the Tyrian embraces a distinct theoretical point of
view. To understand it, one must put on one side the famous ‘precepts of law’
attributed to him,4 including ‘giving each his due’. This humdrum endorse-
ment of existing law does not come from a genuine work of Ulpian. Both
Frezza5 and Waldstein6 wrestle nobly to reconcile ‘giving each his due’ with
Ulpian’s concern to show what, from the standpoint of true philosophy,
really is due to each of us. Their instincts are sound, and they downplay the
precept as an element in Ulpian’s outlook. His genuine point of view emerges
from a well-known passage in his Teaching Manual (Institutiones). There he
derives the term ‘law’ (ius) from justice and relates it to four interconnected
elements: art, religion, ethics and philosophy.7 Law is the art or technique of
1 On the importance of ideology to at least some Roman lawyers see Nörr (1972a),(1976),
the good and equitable. Lawyers can rightly be called priests of the law, for
they mark off the equitable from the inequitable, the lawful from the unlaw-
ful, and try to make people good by providing threats and incentives. They
cultivate true, not sham philosophy. A later passage from Ulpian’s Tribunals
repeats the view that there is a close connection between law and religion,
since knowledge of law is something sacred. He draws a parallel between
teachers of philosophy and law, neither of whom should press for monetary
rewards, though it is not wrong to receive money from pupils if they offer it.8
These remarks are not merely decorative, but neither are they self-
explanatory. The ‘elegant’ view that law is the art of the good and equitable,
consisting of the range of skills needed to see that justice is done, comes from
Iuventius Celsus, who wrote a century earlier.9 Ulpian’s older contemporary
Paul says that there is a type of law, natural law, that is always good and
equitable.10 Ulpian does not dissent, but he goes further. The lawyer is con-
cerned with the techniques by which what is good and equitable can be given
practical effect. His remarks can be understood in the light of what Galen, the
other leading intellectual of the age, says about medicine. Galen is now
thought to have died between 20411 and 216.12 He remained in Rome from
about 169 and Ulpian must have known him. Galen says that medicine is an
art13 and that arts must be useful.14 His list of arts includes law. But though
medicine is an art the true doctor must, Galen says, also be a philosopher.15
He regarded himself as the one person who successfully combined philosophy
with medicine.16 The study and practice of medicine should, in other words,
have a theoretical background. A good doctor embraces logic and reason. He
renounces superstition and despises money.17
Ulpian’s thought follows on the same lines as Galen’s.18 He combines the
belief, stated by Celsus, that law is an art with the view, put earlier by Cicero
ab iniquo separantes, licitum ab illicito discernentes, bonos non solum metu poenarum, verum etiam
praemiorum quoque exhortatione efficere cupientes, veram nisi fallor philosophiam, non simulatam
affectantes).
8 D 50.13.1.5 (Ulp. 8 omn. trib: Proinde ne iuris quidem civilis professoribus ius dicent
<provincial governors dealing with claims for fees>: est quidem res sanctissima civilis sapientia,
sed quae pretio nummario non sit aestimanda nec dehonestanda, dum in iudicio honor petitur, qui in
ingressu sacramenti offerri debuit. quaedam enim tametsi honeste accipiantur, inhoneste tamen
petuntur)
9 10
Frezza (1983) 417 D 1.1.11 (14 Sab.)
11
His work On Theriac to Piso (De theriaca ad Pisonem) must be later than 204 (Swain 1996,
368–9, 430–2) and has flattering references to Severus and Caracalla which could be taken as
praise of their egalitarian stance (Buraselis 1989, 25f.)
12 Nutton (1995) 25–39 showing that the Suda date (199/200) is mistaken; Swain (1996) 357–79
13 Galen, Advice on Arts (Προτρεπτικ ς λγος ες τς τχνας, ed. Kühn 1821) 1 ch.14:
νοµικ τχνη
14 Galen, Advice ch. 9: βιωελς
15 Galen, The Best Doctor a Philosopher ( Οτι ριστος ατρ ς κα ιλσοος), ed. Kühn
1.53–63; Lannata (1984) 214–9
16 Swain (1996) 366–7 17 Galen, Advice p.60 cf. D 50.13.1.15 (Ulp. 8 omn. trib.)
18 For Galen’s influence on Ulpian see Lannata (1984) 214–9
78 3. The Cosmopolis and Human Rights
in the mouths of Atticus and Marcus,19 that the discipline of law must be
based on philosophy. What does he understand by art? It is worth looking at
what he says about medicine. When discussing claims for fees by members of
various professions, including doctors, he says that those who perform incan-
tations, imprecations, and exorcisms cannot claim fees in the way that doc-
tors can, because these techniques are not part of medicine.20 They are the
work of people commonly called impostors, though there are some who
affirm that these techniques have done them good. In much the same way law
is in Ulpian’s view an art with a specific end to be attained by rational means.
As medicine is the art of healing, so law is the art of seeing that justice is done.
The lawyer learns the rules and techniques that serve this end. He must, like
the doctor, embrace reason and renounce superstition. That a certain opinion
about the law ‘has reason’21 (habet rationem) is way of saying that it is right.
And law, like medicine, must be pursued for its own sake and not for
money.22
Ulpian’s search for a philosophy underpinning law is not surprising.
Around AD 200 intellectuals were becoming dissatisfied with the view that
whatever is traditional or customary in a society is automatically right. They
were looking in both politics and religion for something more universal,
rational, and philosophical. Origen, for example, another contemporary,
speaks of the true philosophy of Christ.23 Indeed any discipline that was sys-
tematically concerned with right and wrong, even rhetoric, could lay claim
to a philosophy. Law is concerned with the difference between right and
wrong, lawful and unlawful, and so is in this broad sense a philosophical dis-
cipline. But to understand Ulpian’s ‘true philosophy’ more precisely we
must pinpoint the sham philosophy with which he contrasts it.
His rejection of sham philosophy24 lends itself to several interpretations.25
One, congenial to Plato, is that sham philosophy is sophistry, which he
regarded as subversive of morality. An alternative view is that sham philoso-
phy consists in embracing principles that have no practical application. Both
law and medicine are applied disciplines, concerned to see that what is correct
in theory is put to use in order to heal or do justice. Sham philosophy, in con-
trast, lacks utility, a notion to which Ulpian often appeals. Sham philosophy
may also be hypocritical. Here the standard example is the philosopher
who is after money, when a philosopher’s aim should be to pursue truth and
19
Cicero, Laws (De legibus) 1.5.17
20
D 50.13.1.3 (8 omn. trib: non tamen si incantavit, si imprecatus est, si, ut vulgari verbo impos-
torum utar, si exorcizavit: non sunt ista medicinae genera, tametsi sint, qui hos sibi profuisse cum
praedictione adfirment)
21 22
Ch.2. n.167 (26 of 27 texts) D 50.13.1.5 (8 omn. trib.)
23
Origen, Homily on Genesis 9.2
24
Possible sources: Aristotle, Metaph. 4.2.1004b; Cicero, De fin. 2.21; Quintilian, Inst. 12.3.12
25
Waldstein (1994),(1995a),(1996) 59–71
3. The Cosmopolis and Human Rights 79
wisdom.26 Not to apply in practice what one in theory believes is for Ulpian,
I think, a sign that one’s philosophy is a sham. To quote Euphrates as
reported by Pliny the Younger, the person who practises what philosophy
teaches, for instance judging cases, participates in philosophy.27 A fourth
possible interpretation of sham philosophy is that it is of the sort that lends
itself to irrational techniques, of which incantation and exorcism are exam-
ples. This objection does not apply directly to law, though it may imply criti-
cism of lawyers who do not give reasons for their opinions, as Ulpian is on the
whole careful to do. The true philosophy, then, will be useful and based on
reason, and the true philosopher will be honest, not hypocritical.
But Ulpian’s view that lawyers aim at the true philosophy28 goes, I believe,
beyond a concern with right and wrong and a rejection of philosophy that is
unpractical, dishonest, or irrational. His defence of law as the true philosophy
is in part a reply to Cicero’s reproach in Pro Murena, where the orator is con-
cerned to show that rhetoric is superior to law as a qualification for public
office. According to Cicero, lawyers make no more than a wordy pretence of
wisdom.29 Stripped of the cloak of secrecy with which the pontiffs surrounded
it, law was revealed as a trivial pursuit. Ulpian in response lays claim on
behalf of lawyers to genuine wisdom. His philosophy has accordingly a posi-
tive content.
One way in which this is shown is in the emphasis in his writing on what is
natural,30 including natural equity,31 rather than artificial or, like the Roman
ius civile, confined to a given society rather than universal. The contrast
between natural and civil justice and law was a commonplace in both philos-
ophy and law. Ulpian, however, interprets it differently from his predecessors
such as Gaius. To them the natural is morally superior to the civil in that the
common custom of various peoples, the ius gentium, is founded on natural
reason and hence has a higher status than the law adopted by a particular
state, say Rome or Athens.32 ‘Nature’ can be used to criticize the narrowness
or technicality of the law of a particular state and as an instrument to broaden
and simplify the law. Ulpian, however, takes the law of nature not to be
26 D 50.13.1.4 (Ulp. 8 omn. trib.); 50.5.8.4 (Pap. 1 resp: etenim vere philosophantes pecuniam
quidem pulcherrimam partem, agere negotium publicum, cognoscere iudicare, promere et exercere
iustitiam, quaeque ipsi doceant in usus habere; Mantello (1984)
28
Possible source: Cicero, Tusc. 4.6. There is a good discussion by Birks (1983) 171–3
29
Cicero, Pro Murena 14.30 (ista vestra verbosa simulatio prudentiae) cf. 11.25 (trivial pre-
occupation with letters and divisions between words) and Quintilian, Inst. 12.3.11. Of course
Cicero’s attitude to lawyers varied according to mood and forensic need
30 D 15.1.7.7 (29 ed.); 50.16.42 (57 ed.); 43.16.1.27 (69 ed.); 1.5.24 (27 Sab.); 19.5.4 (30 Sab.);
37.5.1 pr (40 ed.); 37.10.3.13 (41 ed.); 43.26.2.2 (71 ed.); 38.16.1.4 (12 Sab.)
32 Gaius, Inst. 1.1: quod vero naturalis ratio inter omnes homines constituit, id apud omnes popu-
los peraeque custoditur vocaturque ius gentium, quasi quo iure omnes gentes utuntur
80 3. The Cosmopolis and Human Rights
embodied in common custom, the ius gentium, but to be morally superior to it.
In the conditions of his time this was a rational point of view. Roman law had
been extended to all the free peoples of the empire, so that the contrast
between the civil law and common custom was now of purely historical inter-
est. Roman law had in effect become the ius gentium. The role of natural law
in justifying some institutions and criticizing others persisted but was wider
than the role formerly performed by the ius gentium. Slavery is the prime
example of an institution contrary to nature but recognized by the ius gentium,
which ‘encroached on’ (invasit) the law of nature.33 Although most criticism of
law in the Roman world was criticism of its administration rather than its sub-
stance,34 in this instance an important institution is open to moral criticism.
Where does this line of thought come from? Philosophically minded
lawyers are not members of this or that school of philosophy. It is a mistake
to attribute to a lawyer a system of philosophy rather than a set of values.35
The nature of the discipline requires lawyers to be eclectic, to compromise
between different aims. They must be faithful to the wording of authoritative
texts, take account of the purposes they embody, and try to reach conclusions
that are morally acceptable in the particular case. Ulpian is conscious of all
three aims but inclines to give priority to achieving an equitable solution
in the particular instance. Nevertheless, it seems that Ulpian’s outlook was
predominantly Stoic. He shares with the Stoics the view that we are born free
and equal and should live according to nature.36 The precepts of nature are
accessible to reason, in which all share. Humans form a community and relate
to one another and to the gods by virtue of their power of reason.37 As
Marcus Aurelius says, if the mind is common to all, then so is reason and the
reasoning that enjoins what is to be done or avoided. It follows that law is also
common to all, and that we are all citizens in a polity of a sort (πολτευµα). It
follows further that the universe is a kind of city, in which the whole race of
men participate. From where else than from this common city is our mind,
reason, and sense of law derived?38 Marcus is not describing the Roman con-
stitution of his time, in which citizenship was restricted and eagerly sought,
but his universal city represents the ideal to which the Antonine constitution
later tried to give effect.
The Stoic thinkers and those influenced by Stoicism whom Ulpian is most
likely to have absorbed are Chrysippus, the emperor Marcus, and in ethical
matters Cicero.39 Marcianus, almost certainly Ulpian’s pupil and the lawyer
33 34
D 1.1.4. (1 inst.) Nörr (1974) 148–51
35
Sokolowski (1902); Voggensperger (1952)
36
e.g. Cicero, Ends (De finibus) 4.14: convenienter naturae vivere
37
Schmerkel (1982) 2 ch.5
38
Marcus, Meditations 4.4. Of Ulpian’s likely sources Cicero, On Laws 1.23.33 said much the
same cf. Zecchini (1995)
39
D 42.4.7.4 (59 ed.), citing and criticizing an unknown passage of Cicero, speaks to famil-
iarity with his works
3. The Cosmopolis and Human Rights 81
who most resembles him in style and thought,40 cites the definition of law by
‘the leading Stoic philosopher Chrysippus’.41 Law is sovereign over all things
divine and human. It should preside over the good and the bad alike and be
the criterion of what is just and unjust. It should be the leader and ruler of
those called by nature to live in society, prescribing to them what is to be done
and what avoided. The passage cited from Chrysippus agrees closely with
Ulpian’s exposition of law in relation to philosophy, religion and ethics. It
supports the idea that the philosophy he has in mind is mainly Stoic.42
One indication of this is that to Ulpian all humans are by the law of nature
born equal.43 It was the Stoa that first recognized this, beginning with the
equality of husband and wife. Incidentally Ulpian may have been the first per-
son to express in writing the idea that spouses take one another for better or
worse.44 To the Stoics marital equality constituted a first step towards respect
for children, who must be allowed to educate themselves even against the
wishes of their father.45 That duty of respect extends to slaves, for to
Chrysippus a slave is a person hired for life, subordinate to the person who
hired him but not possessed or owned by him. As human beings slaves and
free men are equal,46 for by nature no one is a slave. Nor is anyone noble by
birth. Only that person is free and noble who acts rightly.47 But, despite these
radical beliefs, Stoicism did not have a political programme.48
Crifò attaches importance to the circle of Julia Domna as an element in
Ulpian’s thinking.49 According to Philostratus of Lemnos Julia was the centre
of a group who shared intellectual or philosophical interests.50 The sophist
Philiscus of Thessaly won her patronage by attaching himself in about 212 to
the ‘geometers (i.e. mathematicians) and philosophers who surrounded her’.51
She persuaded Philostratus to write the life of Apollonius of Tyana,52 which he
did, but only after her death in 217. Apollonius was a miracle-worker and saint
(or, on a hostile view, a charlatan) of the time of Domitian, before whom he is
said to have been strikingly outspoken in his own defence. Philostratus’ life was
itum vel uxorem viri participem esse?) cf. Swain (1996) 119f.
45 Edelstein (1966) 73f.
46 Arnim (1903–24) 3 frag. 351 cf. Seneca, Letters (Epist.) 47.11; Favours (De benef.) 3.21.2;
3.22.3
47 48
Edelstein (1966) 83–4 Sandbach (1975) 141.
49
Crifò (1976) 734–6; Nörr (1972a) 20; RE 11.923, 926: Benario (1958); M. Gilmore Williams
(1902); Mundle (1961); Birley (1988) 168
50
Sophists 2.30.1, describing her as a ιλσοος; Dio 75.15.7; 77.18.3; Bowersock (1969) 101f.;
Swain (1999) 176f.
51
Philostratus, Sophists 622.
52
Philostratus, Apollonius (Vit. Apoll.) 1.3; Nörr (1972a) 21; MacMullen (1967). Caracalla
also honoured him in 214/5: Dio 77.18.4
82 3. The Cosmopolis and Human Rights
a defence of the ‘Pythagorean’ way of life which was now giving traditional
Greek philosophy a religious angle. Some Christians took the life of Apollonius
as a pagan counterpart to the Gospels.53 Though Ulpian must have known
Julia Domna, it is a question how far he sympathized with the partly mystical
values that the life of Apollonius seeks to propagate. Himself outspoken and
self-confident, he will have admired the freedom of speech, παρρησα, that is
attributed to Apollonius of Tyana.54 Whether we can go further is unsure.
Frezza has urged that Ulpian’s outlook is drawn from Neoplatonic rather
than Stoic sources.55 Neoplatonism was popular in his time, especially in the
circle surrounding Julia Domna. Frezza56 cites in support the view that
Ulpian takes of animals. Ulpian asserts that natural law applies to all
animals, whether domesticated or wild. They are considered, he says, to know
from experience (peritia), not merely by instinct, about what we call marriage,
the procreation of children and their education.57 As regards damage done by
animals Ulpian makes the owner’s liability depend on whether the animal
acted contrary to nature.58 Owners are liable for the actions of badly behaved
animals, who act contrary to nature, but not for well-behaved ones. Here,
too, he attributes a limited rationality to animals. The Stoic view was differ-
ent. Chrysippus asserts that other animals serve man and the gods, and that
humans do no wrong in using animals for their own purposes.59 The
Christians too, if Origen is representative, thought of man as the sole rational
creature. Lower animals, he says, are governed wholly by instinct.60 But the
Neoplatonic philosopher Porphyry of Tyre, born about 234, a decade after
Ulpian’s death, affirmed, as against Chrysippus and the Stoics, the rational
character of animals.61 Ulpian could not have read his work, but it may be
that his view that the law of nature applied to animals was influenced by
Neoplatonic philosophy. That view implies respect for animals so far as they
have reason, but is not the same thing as attributing rights to animals. There
remains a difference between human beings, who are members of a commun-
ity of rights and obligations embodied in law, and animals, who participate to
some extent in reason and are to that extent subject to natural law. Lawyers
are used to drawing on a variety of sources, and the influence on Ulpian of
Neoplatonic thinking, if it existed, does not detract from the conclusion that
the mainstream of his thought was Stoic.
53 Eusebius, Against the Life of Apollonius (Contra vitam Apoll.); Lactantius, Divine Teaching
This is probably true also of Ulpian’s view of the relation between law and
religion. Gellius, in the course of a passage citing Chrysippus, represents the
judge as the priest of justice.62 Demosthenes, cited by Marcianus, described
law as an invention and gift of the gods.63 Stoics respected the gods and the
numinous, but Stoicism was not a religion. Nörr has offered a perceptive
insight into Ulpian’s view of the relation of law to religion.64 Now that think-
ing people were no longer content with the traditional way of life, serving the
state and conforming to received laws and customs, some found solace in new
religions, such as Christianity or sun-worship. At a time when Tertullian had
abandoned law for religion65 and Gregory the Wonder-Worker was soon to
do the same, a lawyer who was alive to the new currents needed to defend his
calling by showing that it possessed a sacred element. This did not mean that
he had to embrace any of the competing religions. All, says Ulpian, may take
an oath ‘according to their own superstition’, though probably not if it is a
‘religion publicly disapproved’.66 Ulpian speaks of Jewish superstition67 and
we need not suppose that he thought much more highly of Christianity,
though it had adherents in the imperial household. Since Julia Mammaea met
Origen, Ulpian may have known something of Christianity at first hand. It is
true that his work The Proconsul contained a list of imperial rescripts directed
against Christians.68 But he could hardly fail to list these, since his treatise
was meant to help provincial governors perform their duties according to law
and current administrative practice. The list does not in itself demonstrate
hostility to Christianity. Nevertheless his attitude to it is likely to have been
ambiguous. He will have welcomed the universalism and care for the weak
that marked Christianity. A tolerant attitude to religious diversity, such as the
Augustan History attributed to Alexander Severus,69 is not ruled out. But he
will have been hostile to any tendency to seek private salvation rather than to
shoulder public responsibility. Public law combines religion with service to
the state. It is concerned with sacred things, priests, and magistrates.70
62
Gellius, Attic Nights 14.4.3 (. . . iudicem, qui iustitiae antistes est)
63
Demosthenes, Against Aristogeiton 1, 25.16; D 1.3.2 (Marci. 1 inst.)
64
Nörr (1973) 555; Frezza (1968) 368
65
Q. Septimius Tertullianus (c.160–240) expressed hostility to the Roman empire after his
conversion to Christianity (c.195) and to Montanism (c.207). He is probably the same as the con-
temporary lawyer Tertullianus: Lenel, 1889, 2.342–4; Steinwenter (1934) 844–5; Kunkel (1967)
236; Barnes2 (1985) 22–9; Orestano (1973b); Nörr (1974) 55f.; Martini (1975) 79–124; Liebs
(1976) 294 n.37a; Lehmann (1982) 194–202, 279–84; Liebs (1997) §417.2
66
D 12.2.5.1, 3 (Ulp. 22 ed: quod propria superstitione iuratum est, standum rescripsit <divus
Pius> . . . sed si quis illicitum iusiurandum detulerit, scilicet improbatae publice religionis, videamus
an pro eo habeatur atque si iuratum non esset: quod magis existimo dicendum)
67
D 50.2.3.3 (Ulp. 3 off. cons.) but recording a tolerant ruling by Severus and Caracalla
68
Lactantius, Divine Teaching 5.11
69
HA Severus Alexander 22.4 (toleration of Jews and Christians), 29.2 (statues of Apollonius,
Christ, Abraham, Orpheus, and others). The report may not be entirely the product of late
fourth-century wish-fulfilment: Swain (1999) 176
70
D 1.1.1.2 (Ulp. 1 inst.)
84 3. The Cosmopolis and Human Rights
Ulpian’s philosophy, then, was in tune with the movement that led to the
Antonine constitution and he may well have advocated the extension of citi-
zenship that it brought about. The motives behind the Antonine constitution
were perhaps three. Least important is the desire to increase tax revenue,78
though Dio, a senator, makes it the principal aim.79 Given the hostility of the
senate to Caracalla and the dilution of status that the Antonine constitution
involved, that is not unexpected. In practice the amount raised by extending
inheritance tax could not be great given that the richer provincials were often
citizens already. Non-citizens already paid capitation and land taxes. To
point to tax revenue as the explanation of the Antonine constitution is as silly
as to attribute Scottish devolution to the desire of the British Labour party to
win extra seats in Scotland in 1997. In both cases the occasion may have been
relatively trivial,80 but the root causes of constitutional change lay deep. And
Dio himself does not reject the idea of the cosmopolis. On the contrary, he
puts the case for universal citizenship in the mouth of Maecenas.81
principles require in a modern society. Because they form the framework and
underpinning of Ulpian’s writing, he is properly to be regarded as the first
human rights lawyer. He did not originate either the philosophical thinking
or the law on the subject, but he played a major role in elaborating, dis-
seminating and justifying it.
Something must be said about the Roman constitutional background. The
cult of liberty was an ancient Roman theme, resolving in a legal context into a
preference for liberty ( favor libertatis)88 in interpreting laws or private docu-
ments. ‘It is well known that many decisions are reached against the rigour of
the law and in favour of freedom.’89 This explains why everyone has the right
to demand that a free person be produced in order to ensure that he or she is
not being treated as a slave. ‘No one is barred from taking action in favour of
freedom.’90 It served to explain and justify the principle that gave citizenship
to freed slaves irrespective of their sex or ethnic origin. This ‘great benefit’,91
the policy of equal citizenship (isopoliteia), was attributed to the king Servius
Tullius, and attested by a third-century-BC inscription.92 It was inconsistent
with the widely held view that slaves owe their status to natural deficiencies or
to their own fault rather than to bad luck.93 And if in principle slaves could
became citizens why should not the free people of the empire all be citizens?
This point of view triumphed in 212 with the Antonine constitution.
The constitution did of course not apply to slaves, who did not then or later
acquire the right to be freed except in special circumstances, for example
when a price had been laid down for their freedom and they had paid it. By
Ulpian’s time, however, and in his work, the personality and dignity of a slave
(persona servi) has come to be recognized, at least to a limited extent. Hadrian
and Pius deprived the slaveowner who had treated his or her slave brutally of
the ownership of the slave.94 This was to recognize the slave’s right to a min-
imum of decent treatment. Speaking of the urban prefect’s office Ulpian says
that he should hear complaints from slaves of their owners’ cruelty or harsh-
ness, of being starved or treated obscenely or being forced into prostitution.95
In the Severan period the law, as Ulpian presents it, goes further. A sexual
88
e.g. D 43.29.3.9 (Ulp. 7 ed.); 40.5.24.10; 40.5.45.1 (5 disp: a legatee given a small legacy can
be forced to free an expensive slave); 50.17.122 (Gai. 5 ed. prov.); CJ 7.2.15.1a (531–2)
89 90
D 40.5.24.10 (5 fid.) D 43.29.3.9 (71 ed.)
91 D 38.2.1 pr (Ulp. 42 ed: tam grande beneficium, quod in libertos confertur, cum ex servitute
fectus audiat, sic accipiemus non accusantes dominos (hoc enim nequaquam servo permittendum est
nisi ex causis receptis) sed si verecunde expostulent, si saevitiam, si duritiam, si famem, qua eos pre-
mant, si obscenitatem, in qua eos compulerint vel compellant, apud praefectum urbi exponant. hoc
quoque officium praefecto urbi a divo Severo datum est, ut mancipia tueantur ne prostituantur)
3. The Cosmopolis and Human Rights 87
assault on a slave by someone other than the owner gives the slaveowner an
action for wrong (actio iniuriarum) against the other person.96 The reference
to obscene treatment shows that the law aimed indirectly at protecting the
slave against the owner’s sexual advances.
But sexual exploitation is merely one example of treatment that a slave
should not be obliged to put up with. Ulpian discusses and rationalizes the
bringing of actions for assault on or insult to a slave in the slave’s name, but
by the slaveowner,97 ‘If someone wrongs a slave, but not with a view to insult-
ing the slaveowner, the magistrate must not let the wrong go unpunished,
especially if the wrongdoer has beaten or tortured the slave. Obviously even
a slave resents that.’98 Moreover, the action for wrong to a slave is not
confined to physical assaults. If the third person has mildly hit or insulted the
slave the magistrate will not give an action against him. But if he has seriously
defamed the slave, the magistrate must inform himself about the slave’s qual-
ity, character, and occupation. The magistrate will then take account not only
of the wrong done but of the personality of the slave (persona servi) to whom
it has been done, and will allow or deny an action accordingly.99
Despite the fact that slaves, like others under disability, needed someone
else to sue on their behalf, this legal remedy views the slave as a person
endowed with dignity. In another context Ulpian says that slaves have the
right to injure themselves. ‘Even a slave has by nature the right to do violence
to himself.’100 Slaves are not bound to keep themselves in good condition
merely in the owner’s interest. Human dignity gives them the right to disable
themselves from leading a life that appears intolerable. So the owner cannot
deduct anything from a slave’s private fund (peculium) on the ground that the
slave has purposely injured him- or herself. Moreover slave couples living
together should not be split up, so that a buyer who finds that one of the cou-
ple bought is diseased will have to return both.101
96 97
D 47.10.9.4 (Ulp 57 ed.) D 47.10.15.34, 44, 48 (Ulp. 57 ed.)
98
D 47.10.15.35 (Ulp. 57 ed: si vero non ad suggilationem domini id fecit, ipsi servo facta ini-
uria inulta a praetore relinqui non debuit, maxime si verberibus vel quaestione fieret: hanc enim et
servum sentire palam est) cf. 47.10.15.43
99
D 47.10.15.44 (Ulp. 57 ed: Itaque praetor non ex omni causa iniuriarum iudicium servi nomine
promittit: nam et si leviter percussus sit vel maledictum ei leviter, non dabit actionem: at si infamatus
sit vel facto aliquo vel carmine scripto, puto causae cognitionem praetoris porrigendam et ad servi
qualitatem: etenim multum interest, qualis servus sit, bonae frugi ordinarius dispenator, an vero vul-
garis vel mediastinus an qualisqualis: et quid si compeditus vel malae notus vel notae extremae?
Habebit igitur praetor rationem tam iniuriae quae admissa dicitur quam personae servi, in quem
admissa dicitur, et sic aut permittet aut denegabit actionem). ‘These provisions on wrongs done to
the slave cannot be explained away as redressing damage to the owner’s property, for which a dif-
ferent remedy was available, or as defending his reputation. They are available when the wrong to
the slave is not an insult to the owner’ (non ad suggilationem domini id fecit: D 47.10.15.35).
100 D 15.1.9.7 (Ulp. 29 ed: licet enim etiam servis naturaliter in suum corpus saevire)
101 D 21.1.35 (1 ed. cur.) cf. 33.7.12.33 (20 Sab.). Daube (1973) went to the length of postu-
lating a post-Constantinian edition of Ulpian to explain away this text. But, though the text reads
ungrammatically because the compilers have run together several cases, the discussion is entirely
in line with the Stoic view that slaves are members of the family
88 3. The Cosmopolis and Human Rights
It is nature that gives the slave this right, and nature is specially relevant
here because debts between owner and slave are a matter of natural, not civil
law.102 The natural freedom and equality of mankind is, we have seen, a
theme of certain Stoic philosophers.103 What is of interest here is that among
lawyers only Ulpian and three of his lesser contemporaries take account of
this theory in order to justify certain legal rules and principles. Ulpian says
that slavery and the freeing of slaves depend on common customary law (ius
gentium) because by the law of nature all are born free. The grant of freedom
was originally unknown because slavery was unknown.104 Florentinus, whose
exact dates are unknown105 but who is more or less contemporary with
Ulpian since he copies a text of Paul,106 says that slavery is an institution by
which we are subjected to the dominion of others contrary to nature.107
Marcianus, a pupil of Ulpian writing after Caracalla’s death,108 says that a
freed slave can be restored on the emperor’s initiative to the status of being
freeborn. That means, says Marcianus, the state in which all men were origin-
ally born, not that in which the person concerned was born, since he or she
was born a slave.109
These three texts come from works addressed to students. Some may dis-
miss them as ornamental. That cannot be said of a text where Ulpian says that,
though in civil law a slave neither owes nor is owed money, slaveowners can
sue third persons for debts due to their slaves, and can be sued by a third per-
son for debts due from their slaves up to the amount of the slave’s private fund
(peculium). For though in civil law slaves are considered not to exist, it is dif-
ferent in natural law. So far as natural law is concerned, all men are equal.110
This is said in passing, as self-evident. In dealing with a similar problem
Tryphoninus, who was alive in 213, hence at the time of the Antonine consti-
tution, appeals to natural freedom. A slaveowner ‘owes’ his slave money and,
after the slave is freed, pays the freed slave because he wrongly thinks he is
legally bound (i.e. by civil law) to do so. Can the ex-owner recover the
payment? No, because the debt he acknowledged was owing by nature. Since
freedom, says Tryphoninus, forms part of the law of nature and domination
102
D 28.1.20.7 (Ulp. 1 Sab.)
103
Above n.25–30. The natural freedom of all to worship as they think best is asserted by
Tertullian, Ad Scapulam 2.1.2, written in the same year as the Antonine constitution: humani iuris
et naturalis potestatis est unicuique quod putaverit colere. The idea was abandoned by Theodosius I
104
D 1.1.4 (Ulp. 1 inst: cum iure naturali omnes liberi nascerentur nec esset nota manumissio,
cum servitus esset incognita)
105
Probably early third century: Liebs (1976) 348f.; (1997) §428.3; Querzoli (1996) 33f.;
Marotta (2000) 156185
106
D 38.2.28 (Flor. 10 inst.) copying D 48.20.7.1 (Paul 1 port. lib. damn.)
107 108
D 1.5.4.1 (Flor. 9 inst.) Liebs (1997) §428.1; Honoré (1982) 216; (1994), 94–5
109
D 40.11.2 (Marcian. 1 inst.)
110
D 50.17.32 (Ulp. 43 Sab: quod attinet ad ius civile, servi pro nullis habentur: non tamen et
iure naturali, quia, quod ad ius naturale attinet, omnes homines aequales sunt) cf. 28.1.20.7 (1 Sab:
slaves cannot witness wills because they do not wholly share in the civil law)
3. The Cosmopolis and Human Rights 89
111
D 12.6.64 (Tryph. 7 disp: ut enim libertas naturali iure continetur et dominatio ex gentium
iure introducta est, ita debiti vel non debiti ratio in condictione naturaliter intellegenda est)
112 113
HLL 4 § 424, 428.3, 428.1, 417.3 The Theriac (above n.4)
114
D 14.3.11.3 (28 ed: litteris utrum Graecis an Latinis?)
115 116
D 32.11 pr (Ulp. 2 fid.); 45.1.1.6 (48 Sab.) D 1.16.9.5 (Ulp. 2 off. proc.)
117
D 47.10.13.7 (Ulp. 57 ed.)
90 3. The Cosmopolis and Human Rights
to whom the emperor has granted the right to draw water from a public
source may in certain cases sue one who prevents him exercising that right,
again by the action for wrongs.118 The action for wrongs is therefore the legal
technique for vindicating human rights so far as they were recognized in
Ulpian’s time. After the extension of citizenship by the Antonine constitution
this action will have been the remedy available to anyone who was denied the
new citizen rights such as the right to hold public office.
What potential readership was Ulpian addressing? There are two interest-
ing clues to this. One consists in his use of Greek in a corpus of work basically
written in Latin, the language of Roman law. There are a number of sound
reasons for a legal author writing in Latin to include on occasion a passage in
Greek. Some presuppose that an excuse is needed for departing from the
Latin norm. One example is when the law to be discussed, for example, a
rescript, was issued in Greek119 or when the will or other document to be
interpreted is in Greek.120 Another is when the author wants to embellish his
text by citing a passage from a Greek author such as Homer,121 Plato,122
Xenophon,123 Demosthenes,124 Theophrastus,125 or Chrysippus.126 A third is
when he wishes to inform the reader of the Greek equivalent of a Latin
term,127 or the formula to use in Greek for some purpose, for example when
formally acknowledging the payment of a debt.128 A fourth is when the
author purports to derive a Latin term from a Greek root.129 None of these
involve the writer in identifying himself with Greek-speaking people or
adopting their culture. On the contrary, to say ‘the Greeks call this a so-and-
so’ implies a certain distance.
Ulpian does sometimes tell us what the Greek term is for so-and-so,130 but
a more typical use of the language is for him to insert a Greek word or two in
a Latin sentence as being more expressive or conveying nuances that the Latin
equivalent would not convey. This is a form of codeswitching.131 The contrast
can be illustrated from the phrase 'ν πλ(τει which means ‘in a broad sense’.
Paul says ‘this (proof of the date of building) is to be taken, as the Greek
expression goes, 'ν πλ(τει.’132 Ulpian in contrast says ‘this (time when a pay-
ment is ratified) is to be taken 'ν πλ(τει,’133 or ‘this (value of a slave at time of
118 D 43.20.1.43 (Ulp. 70 ed.) 119 e.g. D 8.3.16 (Call. 3 cogn.); 5.1.37 (5 cogn.)
120 e.g. D 34.1.6 (Scae. 18 dig.); 40.4.60 (24 dig.); 40.5.46.3 (Ulp. 6 disp)
121 122
D 18.1.1.1 (Paul 33 ed.) D 50.11.2 (Call. 3 cogn.)
123 D 47.22.4 (Gai. 4 XII tab.) 124 D 48.19.16.6 (Saturninus 1 poen. pag.)
125 D 1.3.3 (Pomp. 25 Sab.); 5.4.3 (Paul 17 Plaut.) 126 D 1.3.2. (Marci. 1 inst.)
127 e.g. D 19.2.25.6 (Gai. 10 ed. prov.) 128 e.g. D 46.4.8.4 (Ulp. 48 Sab.)
129 e.g. D 50.16.233.2 (Gai. 1 xii tab.)
130 D 47.9.3.6 (56 ed.); 43.8.2.22 (68 ed.); 23.3.9.3 (31 Sab.); 47.2.3 pr; 50.16.177; 46.1.8 pr (47
death) will have to be looked at 'ν πλ(τει.’.134 To use Greek without apology
implies that the reader is bilingual and is attuned to the nuances of Greek.
Sometimes the Greek phrase is one current among educated Latin-speaking
people, or is one for which there is no obvious Latin equivalent. At others,
words such as %σα)ς (‘unclearly’) are chosen because, though there is a Latin
equivalent (obscure), the Greek is more vivid and makes a better contrast, in
the context of a slave’s ability to speak properly, with %σµως (‘unintelligi-
bly’). In general the author feels free to address a readership that is in part
Greek or has learnt Greek. Of the 48 instances that I have come across of this
direct use of Greek in the legal literature 38 come from Ulpian135 and three
from his pupil Marcianus.136 Of the earlier writers there is one text of the
Augustan Labeo,137 one of Celsus,138 and three of Julian139 from the time of
Hadrian or a little later, and perhaps one each of Ulpian’s younger contempor-
aries Papinian140 and Paul.141 Many of the code-switches are reminiscent of
134 D 13.3.3 (27 ed: si vero desierit esse in rebus humanis, mortis temporis: sed 'ν πλ(τει secun-
2.14.7.2 (4 ed: hoc συν(λλαγµα et hic nasci civilem obligationem . . . contractum, quod Aristo
συν(λλαγµα dicit); 42.4.2.4 (5 ed: haec verba παρατατικ)ς scripta sunt); 4.8.29 (13 ed: nam τ"#
δυν(µει a reo petit); 4.9.1.3 (14 ed: ut ναυ$λακες et diaetarii. . .quod χειρµβολον appellant);
10.2.20.7 (19 ed: unum ex filiis %γωνοθεσαν suscepturum); 11.1.11.5 (22 ed: non respondit, id est πρ ς
-πος); 12.2.13.2 (26 ed: et quasi µονοµερς condictio ei dari debet); 13.3.3 (27 ed: sed 'ν πλ(τει
secundum Celsum erit spectandum); 14.1.1.12 (28 ed: quaedam naves onerariae, quaedam (ut ipsi
dicunt) 'πιβατηγο sunt); 27.9.3.4 (35 ed: Si ius 'µυτευτικ ν vel 'µβατευτικ ν habeat pupillus);
47.2.52.20 (37 ed: τ#ς γον#ς dumtaxat χ(ριν); 28.1.22.5 (39 ed: dum tamen habeat χαρακτ#ρα);
39.3.1.9 (53 ed: sulcos aquarios qui .λικες appellantur); 47.10.5.10 (56 ed: is qui 'πιγρ(µµατα pro-
duxerit); 5.1.19.2 (60 ed: quasi ab eo qui παραπλε/ emit); 42.5.15 pr (62 ed: %ναγραν facere);
43.12.1.2 (68 ed: perenne est, quod semper fluat, %ναος torrens 0 χειµ(ρρους); 43.17.3.7 (69 ed: non
ab eo qui κρυπτς possideret, sed ab eo cuius aedes super κρυπτς essent); 46.8.12.2 (80 ed: hoc autem
'ν πλ(τει accipiendum); 7.8.10.1 (17 Sab: sed si χρ#σις sit relicta); 32.70 pr (22 Sab: quod tinctum
non est sed α&τους); 30.53.5 (non ut carnem praestet vel cetera λεψανα); 32.55.5 (25 Sab: cum
ξ$λον hoc et naves ξυληγς appellant, quae haec %π τ)ν 3λ)ν deducunt); 18.1.9.2 (28 Sab: quia
eadem prope ο&σα est); 8.2.17.2 (29 Sab: si forte κατ %νταν(κλασιν lumen devolvatur); 47.2.43.9
(41 Sab: quid ergo, si ε4ρετρα quae dicunt petat?); 21.1.9 (44 Sab: nec qui %σα)ς loquitur: plane qui
%σµως loquitur, hic utique morbosus est); 45.3.13 (48 Sab: hic τ σ(ζον corrumpit stipulationem et
solutionem); 47.11.9 (9 off. proc: plerique inimicorum solent praedium σκοπελζειν); 48.19.9.8 (9 off.
proc: quibus sententia praecipitur δηµοσων %πχεσθαι); 32.1.2 (1 fid: nec testamenti ius faciendi
habent, cum sint %πλιδες); 36.1.17 pr (4 fid: plerique 6ποκοριστικ)ς patrimonium suum peculium
dicunt); 21.1.1.7 (puta θσις febris; aediles bis κατ το7 α&το7 idem dixisse); 21.1.1.9 (veluti con-
tingeret ρενητικ8)); 21.1.10.4 (sed et νυκτ%λωπα morbosum esse constat); 21.1.14.8 (1 ed. aed. cur:
si quis %ντι%δας habeat . . . et si %ντι%δες hae sunt quas existimo . . . qui %ντι%δας habet vitiosus est)
136
D 20.1.11.1 (1 form. hypoth: Si %ντχρησις facta sit); 48.3.6.1 (2 iud. pub: et ideo cum quis
%ν(κρισιν faceret); 48.19.17.1 (1 inst: quidam %πλιδες sunt, hoc est sine civitate)
137
D 28.7.20 pr (Iav. 2 post Lab: %δ$νατος condicio)
138
D 33.10.7.2 (Cels. 19 dig: sono quodam κα τ"# %ν(ρθρ8ω ων"# dicere)
139
D 38.7.1 (Iul. 27 dig: haec verba παρατατικ)ς accipiuntur); 38.16.8 pr (59 dig: per abu-
sionem vel potius %ναορικ)ς accidit); 30.104.1 (1 Urs. Ferox: %δ$νατος condicio). Julian seems
to have known the philosopher Euaretus (ILS 7776) and to have spent time at the temple of
Asclepius at Pergamum about 145–7 with Aelius Aristides (or. 48.9: Behr, 1986, 2)
140
D 48.5.6.1 (Pap. 1 adult: lex stuprum et adulterium promiscue et καταχρηστικ9τερον appellat)
141
D 36.1.83 (Paul 2 decr: cum solus Foebus egisset µονοµερ)ς)
92 3. The Cosmopolis and Human Rights
Cicero’s practice, especially in his letters, or may be drawn from a pool of stock
terms used in rhetorical instruction.142
This sudden shift after 212 in the way in which certain legal writers use
Greek points to a change in the intended readership of legal works. It counts
as part of the cultural change of which the Antonine constitution is the most
prominent legal expression. It does not imply that Ulpian sheds his Roman
identity. He does not. The Greeks call public roads βασιλικ(ς, our people
(nostri) call them praetorias, others consulares vias.143 He wishes, however, to
emphasize his dual identity, Tyrian and Roman.144 In this he is, to judge from
surviving legal texts, unique, though many lawyers were from the provinces
and retained links with their origins, as inscriptions prove.145 The other
reason for Ulpian’s switches into Greek is in my view his desire, as an act of
solidarity, to accommodate the greatly increased body of Greek-speaking
citizens.146 He wants openly to speak to them as participants in the cos-
mopolitan legal system.
The second clue to the intended readership concerns Ulpian’s effort to per-
suade the reader of the justice of the items in the praetor’s edict. Lawyers of
the middle and later principate appealed to values to justify their decisions
more freely than their predecessors,147 and this was pre-eminently true of
Ulpian. He was conscious of the fact that the new citizens needed to be per-
suaded that in being made subject to Roman law they were not being treated
arbitrarily. Readers of Edict cannot fail to notice the pains he takes to show
that the edict is suited not only to the traditional Roman ways but is useful,
equitable, and adapted to the needs of all. Often he points to the utility148 or
necessity149 of the provision in the praetor’s edict, a utility that can be either
public or private150 or both.151 It might consist, for instance, in the need to
protect travellers by sea or land who are forced to entrust goods to a ship’s
captain or innkeeper,152 or in the interest the holder of a right of way over
neighbouring land has in seeing that the road is kept in good repair.153 Other
practical aims are mentioned: the difficulty of proof,154 the need to protect
minors, to uphold the family, and to provide incentives for burying corpses
142 143
Adams (forthcoming) ch.3 D 43.8.2.22 (68 ed.)
144
On code-switching see Myers-Scotton (1993, 1998) and on its relation to identity Adams
(forthcoming) ch.3
145 146 147
Liebs (2001) xi–xv Adams (forthcoming) ch.3 Wieacker (1977) 29–38
148
D 3.5.1 (10 ed.); 4.1.1 (11 ed.); 4.9.1.1 (14 ed.); 9.3.1.1 (23 ed.); 14.1.1 pr (28 ed.); 14.4.1 pr
(29 ed.); 47.6.1 pr (38 ed.); 47.9.11 (56 ed.); 43.15.1.1 (68 ed.); 43.18.1.1; 43.19.3.12; 43.21.1.1;
43.22.1.7 (70 ed.)
149
D 10.4.1 (24 ed.); 16.3.1.2, 4 (30 ed.); 26.10.1 pr (35 ed.); 47.9.1.1 (56 ed.); 42.8.1.1 (66 ed.);
43.20.1.39 (70 ed.)
150
D 43.8.2.1 (68 ed.); 43.9.1.1 (69 ed.); 1.14.3 (38 Sab.); Steinwenter (1939) 84–102; Longo
(1972) 7–71; Marotta (2000) 93–6
151
D 47.8.2.1 (56 ed.); Pliny the Younger, Panegyric 67.8; Marotta (2000) 147–63
152 153 154
D 4.9.1.1 (14 ed.) D 43.19.2.12 (70 ed.) D 11.1.2 (22 ed.)
3. The Cosmopolis and Human Rights 93
and for guardians to spend their own money on behalf of their wards.155
Respect for magistrates156 and the need to see that their orders are
enforced157 are also stressed. Moreover new legislation must possess ‘clear
utility’.158
A value that Ulpian stresses at least as much as utility is fairness or equity
(aequitas),159 sometimes natural equity.160 Equity is related in his thinking to
equality, not directly, but in the sense that it requires the interests of each per-
son to be taken into account and given equal weight. Equity requires that the
parties should be on a level, so that for example the right to sue goes with lia-
bility to be sued,161 and benefit and liability go hand in hand.162 Those who
are weak or have been deceived must be protected against the strong and the
deceivers. In this connection Ulpian often denounces craftiness (calliditas)
and stresses provisions in the edict directed against the crafty,163 or those who
exploit others,164 or which serve to protect the ill-informed.165 The special
feature of natural equity is that it operates even when the civil law does not
cater for the problem. For example, it takes account of agreements that are
not enforceable by civil law,166 protects persons who have technically come of
age but are immature,167 and prevents slaves due to be freed from taking
advantage of legal technicalities.168 Natural equity is not fundamentally dif-
ferent from civil equity, but the equitable solution to a problem may or may
not already have been embodied in the civil law.
Some justifications offered are therefore practical, and consist in the good
consequences, represented by utility, of having certain provisions in the
praetor’s edict. Others are moral, and consist in giving due weight to the
interests of all, regarded as of equal worth, and represented by the notion of
equity. Whether Ulpian has provided a good justification for a given provi-
sion of the edict is sometimes questionable. What is not in question is his
determination to persuade the citizens of the new cosmopolitan society
created by the Antonine constitution that the laws governing their lives are
not arbitrary. There are good reasons for these laws, reasons that apply uni-
versally.
155 D 11.7.12.3 (25 ed.); 27.4.1 pr (36 ed.); 28.8.7.1 (60 ed.); 38.6.1.5 (46 ed.) cf. 43.8.2.10 (64
(40 ed.); 37.11.2 pr (41 ed.); 37.12.1 pr (45 ed.); 43.3.1.1, 2 (62 ed.); 43.16.1.1 (69 ed.)
160 D 2.14.1 pr (4 ed.); 4.4.4 pr (11 ed.); 13.5.1 pr (27 ed.); 47.4.1.1 (38 ed.); 37.5.1 pr (40 ed.);
ed. cur.)
164 D 2.7.1 pr (5 ed.); 38.2.1 pr (42 ed: former owners of freed slaves); 39.4 12 pr (38 ed: tax-
T H E last chapter was about the theoretical basis of Ulpian’s work. This one
is about the method he adopted in solving legal problems.1 His method was
empirical in two ways that mark him off to some extent from other Roman
lawyers. But only to some extent. Whereas his oral style and egalitarian phi-
losophy contrast rather sharply with those of other leading Roman lawyers,
his method of argument differs from some of theirs only in degree.
One way in which his method is empirical is that he likes to give and argue
from examples, to proceed by trial and error. It emerged from the study of
his style in chapter 2 that over 90 per cent of the uses of the phrase ‘for
instance’ (ut puta) in Digest texts are his.2 He is keen on arguing from exam-
ples: ‘it makes little difference if . . . ( parvi refert)’3 and of proceeding from
one example to another ‘but what if . . .’ (quid tamen si?)4 or ‘unless perhaps
. . .’ (nisi forte).5 To give and argue from examples is even more a feature of
the Anglo-American common law, both in its origin and in its modern form
of reliance on precedent. It involves arguing from analogy or lack of anal-
ogy.6 It involves refusing to commit oneself to a broad principle until one can
see how it will apply to a particular set of facts.
A second way in which Ulpian’s method is empirical is in his practice of
balancing a number of considerations before reaching a conclusion: it is plu-
ralistic and shies away from commitment to a single overriding principle. The
judge, magistrate, or legal consultant, accepting that law serves more than
one value, must see to what conclusions the different values would lead
before making a choice between them, if that should prove necessary. The
considerations that mainly concern Ulpian are (i) the wording of laws, con-
tracts, wills, and other documents (verba); (ii) the purpose or intention that
lies behind them (mens); (iii) public policy in the guise of utility (utilitas); and
(iv) the moral acceptability of the conclusion to which these considerations
point. In particular, would it be equitable (aequum)? This last value prefigures
a feature of Anglo-American law, namely the use of equity to outflank
morally unacceptable features of the strict common law. But in using it
1
On various aspects of this see Mantello (1995) (nominalism and linguistic theory); Falcone
(1997), Tondo (1998)
2 3
Ch.2 n.433 (298 out of 315) Ch.2 n.17 (28 out of 30)
4 5
Ch.2 n.151 (43 out of 44) Ch.2 n.21–2, 24, 153 (91 out of 117)
6
Bretone (1982) 311
4. The Empirical Method 95
These examples merely point to a wish on the author’s part to make plain
what it is he is talking about. Once we have an example or a range of exam-
ples the next question is how the law should handle them. As mentioned, the
second way in which Ulpian’s method is empirical is that he seeks a balance
between various considerations before reaching an opinion on a point of law.
Pernice complained that he came to conclusions not on strictly legal grounds
but for reasons of convenience or policy. There is substance in Pernice’s
complaint, and it is to Ulpian’s credit that there is. He wants to find equitable
solutions to legal problems and to fit the law to the society in which it oper-
ates, when this can be done with due regard to the constraints set by the
system as a whole. In finding the solution to legal problems some lawyers give
precedence to following settled rules, others to adherence to principle, others
again to the search for a morally acceptable solution in the particular case.
Ulpian inclines more to the last category, but always has regard to binding
texts and to the opinions of other lawyers of standing, whether he agrees with
them or not.
His discussions often appeal to the interpretation of texts of authority, to
utility, and/or to equity. The interpretation of a text involves two elements,
the wording of a law or private document (verba) and its purpose or the inten-
tion behind it (mens, sententia). A leading text, near the beginning of On the
Edict, says that ‘according to the laws’ means according to both the purpose
and the wording of the laws.14 Ulpian does not say that, if the two clash, the
purpose must prevail over the wording. In the process of interpretation both
must be taken into account and one has sometimes to choose between them.
The same is true of agreements. Whether an agreement not to sue means an
agreement not to sue a particular person or not to sue at all depends both on
the words used and the intention of the parties.15 In deciding whether a trust
to free a slave has been created one must pay attention not only to the word-
ing of the will but to the intention of the testator.16
Here are some more examples. Papinian says that a condition on which a
gift depends, say a gift of freedom, cannot be ‘taken back’ (adimi) because it
was not given in the first place. Conditions are ‘attached’ to gifts (adscriptae),
not ‘given’. Ulpian comments that it is better to give effect to the intention of
the person who purports to ‘take back’ the condition than to keep to the strict
meaning of the words used. So someone who has attached a condition to a
gift can take it back,17 whereupon the gift, as he or she intended, becomes
unconditional. Again, a rescript required a surviving mother to petition for
14 D 50.16.6.1 (3 ed: verbum ‘ex legibus’ sic accipiendum est: tam ex legum sententia quam ex
verbis)
15 D 2.14.7.8 (4 ed: non minus ex verbis quam ex mente convenientium aestimandum est)
16 D 40.5.24.8 (5 fid.)
17 D 34.4.3.9 (24 Sab: sed melius est sensum quam verba amplecti et condicione sicut adscribi, ita
et adimi posse)
4. The Empirical Method 97
sums up his view by saying that the just judge should not simply act on the
analogy of cases where one person intervenes in another’s affairs, but act in a
broadly equitable way, as the nature of the action allows him to.28 The pas-
sage in which he discusses this action for funeral expenses gives a good idea
of how, in his view, considerations of equity depend on the facts of the case.
Again, the praetor’s edict provided that an action could be brought against
a debtor in one place, say Alexandria, for a debt due in another, say Ephesus.
On occasion, says Ulpian, the right course in trying such an action is to decide
that the debtor is not liable to pay at the place where he is sued provided he
gives security for payment at the place agreed. For instance, the debtor might
have money readily available on deposit in Ephesus. The action for a debt due
in a particular place had a discretionary element, and the judge trying it
should keep his eye on the equity of the matter.29
Sometimes the right course from the point of view of equity is to resort
to a supplementary action (actio in factum). Of the Digest texts that give a
supplementary action Ulpian has 71 out of 118, which comes to three-fifths,
a disproportionately great but not overwhelming use of the notion.30 In one
text Ulpian says that if a co-heir incurs liability through mistakenly selling a
slave belonging to the deceased who was given his freedom in the will, it is
equitable to give him a supplementary action against his co-heir to recoup a
share of the loss. In this connection he remarks that when an action or defence
is not strictly available a supplementary action or defence can be given (i.e. if
it would be equitable to do so).31 Fruits from your tree fall on my land. I let
my cattle eat them. Aristo, who discusses three possible actions, thinks that
none of them will, for technical reasons, fit the facts. Well then, says Ulpian,
a supplementary action must be made available.32 One man pushes another,
who in turn pushes a slave and thereby kills him. Can the first man be sued
for wrongfully killing the slave? Proculus said not. The first man did not kill
the slave. The second was not at fault. In that case, says Ulpian, a supple-
mentary action must be given against the first man.33 A testator makes a slave
heir if no child is born to him. When he dies his wife is said to be pregnant. In
the case of a free person appointed heir there was a procedure under the edict
for ensuring that the wife really was pregnant and that a spurious child was
not smuggled in as the testator’s. Ulpian, following an opinion of Aristo, says
that the slave, though not strictly within the terms of the edict, has an inter-
est in the matter, which is also a public interest in the authenticity of family
succession. He should therefore be allowed to petition for the procedure to be
28
D 11.7.14.13 (Ulp. 25 ed: et generaliter puto iudicem iustum non meram negotiorum gestorum
actionem imitari, sed solutius aequitatem sequi, cum hoc ei et natura actionis indulget)
29
D 13.4.4.1 (Ulp. 17 ed: in summa aequitatem quoque ante oculos habere debet iudex, qui huic
actioni addictus est)
30 31
VIR 2.786–7. D 10.2.49; 19.5.21 (2 disp)
32
D 19.5.14.3 (Ulp. 41 Sab: erit agendum points to the genuineness of the phrase)
33
D 9.2.7.3 (Ulp. 18 ed: again, actio erit danda points to genuineness)
100 4. The Empirical Method
set in motion. Suppose, again, I give you some money intending a gift, but
you think it is a loan. Can I recover the amount? Julian held that there was no
gift and possibly no loan either. Ulpian agrees that there is no contract, and
in principle an action for recovery (condictio) lies. But you have spent the
money, as I intended you should, and recovery would be unjust. Hence you
are entitled, says Ulpian, if I bring an action to recover the money, to plead
my dishonesty as a defence.34
Sometimes there is a straightforward stand-off between strict law and
equity. Supposing a slave contracted a debt and his owner then died leaving
more than one heir. If the other party to the contract sues one of the heirs, the
others are in law free of liability, even though the heir sued was held liable
only to the extent of his share of the slave’s fund (peculium). But, says Ulpian,
though that is the position in strict law, an action should be given against the
other heirs so that they will be freed from liability only to the extent that the
creditor actually receives payment.35 A father who made his son heir could
provide by a secret instrument that if the son died under age the inheritance
should go to someone else. In the case in issue the son died under age and his
mother claimed the inheritance on the ground that the secret instrument had
been revoked. It was held that it had not been revoked. Then the mother dis-
covered that the father had not executed a secret instrument in the first place.
Could she reopen the case? Neratius points out that the final judgment (res
iudicata) bars her from doing so. Ulpian agrees that this is a case of res
iudicata, but says that the mother must be given a remedy because the earlier
proceedings related only to the revocation of the secret instrument, not its
existence.36 A testator leaves a legacy and asks the legatee by way of trust to
emancipate his (the legatee’s) own child. If the request had been to emanci-
pate a slave there was a legal procedure for ensuring that it was given effect.
But Papinian points out that no such procedure exists in the case of a child,
and Ulpian admits that he had formerly taken the view that the law of trusts
provides no remedy in such a case. The legatee should, however, says Ulpian,
be compelled to emancipate the children outside the normal course of law
(extra ordinem). Otherwise the intentions of testators will be thwarted.37 He
cites a ruling of Severus along the same lines.
Humanity is another value to which Ulpian appeals, especially when liberty
is in question. Anyone may appeal against a sentence in a criminal trial, not
only the accused but anyone else, and may do so even if the person sentenced
does not want an appeal.38 Anyone may bring proceedings for the production
of a freeman who is thought to be wrongfully detained (the equivalent of
34
D 12.1.18 pr (Ulp. 7 disp.)
35
D 15.1.32 pr (2 disp: sed licet hoc iure contingat, tamen aequitas dictat <rescissorium> iudi-
cium in eos dari, qui occasione iuris liberantur, ut magis eos perceptio quam intentio liberet)
36 37
D 44.2.11 pr (Ulp. 75 ed.) D 35.1.92 (5 fid.)
38
D 49.1.6 (2 appell: credo enim humanitatis ratione omnem provocantem audiri debere)
4. The Empirical Method 101
itum vel uxorem viri participem esse?) cf. Swain (1996) 119f.
42 D 17.2.55 (30 Sab: aequum est enim, ut cuius participavit lucrum, participet et damnum)
43 D 30.71.3–4 (Ulp. 51 ed: alia enim condicio est hominum, alia ceterarum rerum)
44 D 33.7.12.7 (Ulp. 20 Sab: neque enim duram separationem iniunxisse credendus est)
45 Ch.2 n.539, 618
102 4. The Empirical Method
46 D 2.13.8 pr (4 ed.), 11.1.11.11 (Ulp. 22 ed.), 16.3.7 pr (30 ed.), 47.4.1.2 (38 ed.), 43.26.8.3 (71
ed.), 30.47.5 (22 Sab,), 27.8.4 (3 disp.), 36.1.23.3 (5 disp.). The number of texts and the repetition
of the formula culpa dolo proxima tells against interpolation. There is one such text in Marcianus:
D 47.9.11 (Marci. 14 inst.) cf. 26.7.7.2 (Ulp. 35 ed: lata neglegentia)
47 D 1.16.6.3 (Ulp. 1 off. proc: nam valde inhumanum est a nemine accipere, sed passim vilissi-
Once a conclusion has been reached on the way in which the law applies to
a given set of facts, it is a question whether the law would be the same if the
facts were slightly different. Indeed Ulpian quite often qualifies a conclusion
with a possible exception ‘unless perhaps’ (nisi forte)54 to ensure that what he
says is not taken too broadly.55 The implication is not that the result would
certainly be different if the facts were different but that it might be. At other
times Ulpian commits himself to the view that the result would be the same if
the facts were different: ‘it matters little that . . .’ ( parvi refert . . .)56 or ‘but
even if so-and-so were the case . . . equally’ (sed et si . . . aeque).57 At others
again he raises the question ‘but what if?’ (quid tamen si?)58 and embarks on
a further discussion that may end either way. All these lines of discussion rest
on the assumption that the limits of a principle are to arrived at by trial and
error: by trying the rule out in a range of fact-situations.
Here are some examples of the first sort. If a testator who by will leaves tan-
gible property in trust sells the property during his lifetime without intending
to extinguish the trust, the trust is not extinguished. If he leaves a debt in trust
but exacts the debt during his lifetime the law is the same. Unless perhaps,
says Ulpian, there is a difference between the two cases. The tangible prop-
erty continues to exist but the debt does not.59 A father cannot dissolve the
engagement of a daughter whom he has emancipated. So if he has given prop-
erty by way of dowry to the daughter’s fiancé, and then changes his mind
about the proposed marriage, she can defy him by marrying her fiancé. He
cannot recover the money on the ground that the object for which it was
handed over has failed. Unless perhaps, says Ulpian, one argues that he
handed over the property on the basis that, whether or not the marriage took
place, he could recover it if (in the end) he did not agree to the marriage.60 The
‘perhaps’ is to be taken seriously, as we see from another example. An execu-
tion creditor can seize a debt owed to his debtor, provided it is admittedly
due. If the debt is said not to be due, the fairest course is not to seize it, unless
perhaps one says that, as in the case of pledged corporeal property, the mag-
istrate should decide whether the debt really is due. But there is a rescript to
the contrary.61 So, by implication, the two cases are in law the same.
To say that it matters little (parvi refert) if the facts are somewhat different
is a way of avoiding too narrow a statement of the law. A couple of examples
will suffice. If the legitimacy of two children under puberty is in issue and one
of them reaches puberty, the decision must be deferred until the other also
reaches puberty. It matters little whether the one still under puberty is plain-
54
Ch.2 n. 21–4, 153
55
Many nisi forte texts have been suspected of interpolation, some perhaps rightly. But since
more than three-quarters of the nisi forte texts come from Ulpian (Ch.2 n.20–4) the great major-
ity must be genuine
56 57 58 59
Ch.2 n.17 Ch.2 n.152 Ch.2 n.151 D 32.11.12–13 (2 fid.)
60 61
D 23.1.10 (3 disp.) D 42.1.15.9 (3 off. cons.)
104 4. The Empirical Method
tiff or defendant.62 A shipper is the person who pays the expenses and takes
the profits from operating a ship, whether the owner of the ship or someone
who has chartered the ship for a term or permanently. It matters little whether
the shipper is male or female, the head of a family, a child in power, or a slave.
But a child under age requires his or her guardian’s authority.63 As these
examples indicate ‘it matters little’ is in practice an euphemism for ‘it does not
matter’. The same is true of ‘but even if . . . equally’ (sed et si . . . aeque). If a
slave has by will been given his freedom if he climbs the Capitol, and the heir
prevents him he becomes free. But even if he is to be free if he pays one co-heir
and another prevents him, he will equally be free.64
The ‘but what if’ cases include some in which Ulpian’s train of thought
comes through with special clarity. The seller of a slave who was a thief was
liable in the action of sale if he knew of the slave’s fault, but not otherwise.
What if he did not know but told the buyer that the slave was honest and so
got a better price? Ulpian tentatively asserts that the seller is liable in the
action on sale (et putem teneri). True, he did not know. But he should not
rashly have asserted something that he did not know to be true.65 Hadrian by
rescript imposed a penalty on those who bury a corpse in a town. But what if
the statute of the local authority allows burial in the town? The answer is that
imperial rescripts possess a general force and should prevail everywhere.66 A
suit to enforce a testamentary trust must be brought where the greater part of
the inheritance lies. In deciding where this is, the rule to apply is that debts do
not diminish the assets in any particular area, because they burden the estate
as a whole. But what if certain assets are earmarked, say, for maintenance to
be provided in Rome? In that case, says Ulpian, I should think that this
reduces the Rome assets, so that the bulk of the inheritance may lie else-
where.67
These few examples, which could be multiplied many times over, give an
idea of the open and empirical character of Ulpian’s work. He shares these
features with several other Roman lawyers but in my view surpasses them in
candour and clarity. Although there are important differences between the
development of the law by lawyers, like Ulpian, who hold or have held
prominent positions in government and its development in the common law
world by a specialized group of judges, the parallels are a good deal closer
than is commonly supposed.
62 63 64
D 37.10.3.8–9 (41 ed.) D 14.1.1.15–16 (28 ed.) D 40.7.3.3–4 (27 Sab.).
65
D 19.1.13.3 (32 ed.). This highly Ulpianic text (quid tamen si . . .et putem) has often been sus-
pected of interpolation
66 67
D 47.12.3.5 (25 ed.) D 5.1.50. pr.–1 (6 fid.)
5
Genuine Works
T H I S chapter makes use of the analysis of style in chapter 2 to identify the
works attributed to Ulpian that are genuinely his. Chapter 11 deals with five
works that have been falsely attributed to him. The genuine works are
identified by the same technique as in the first edition but the evidence is set
out in a different way. There the method consisted in taking each book of
Ulpian’s works and listing the footnotes in the chapter on Ulpian’s style that
referred to a word or expression to be found in that book. That method was
accurate but not reader-friendly. To discover why the work or book was said
to be genuine Ulpian the reader had to look up the footnotes.
In this edition I have instead set out some but by no means all of the
Ulpianic words and phrases that appear in each book directly. The ones cho-
sen are those that are most typical of Ulpian’s manner and occur fairly often
in his writings. Provided the words and phrases chosen occur seldom in other
Digest authors it is relatively easy to judge the cogency of the argument for
authenticity that they provide. The reader will notice how important word-
order is in bringing out what is distinctive about Ulpian’s writing. Between 40
and 42 per cent of the material in Justinian’s Digest is attributed to Ulpian.
But a number of common words and phrases occur only or predominantly in
his Digest texts.1 This makes it fairly easy to recognize genuine Ulpian.
The argument that texts attributed to Ulpian are really by him depends on
the assumption that his style was reasonably coherent. The evidence to be pre-
sented shows that, if the major commentaries On the Edict and On Sabinus are
genuine, a number of other works attributed to our author are written in a
similar style and can also be taken as genuine. This argument could satisfy a
reader who was worried about extraneous elements in texts attributed to
Ulpian, for example marginal glosses made on manuscript copies of his work
or interpolations due to Justinian’s compilers. The argument goes to show
that, despite these intrusions in the texts and despite some mistakes in copy-
ing, the style of many works attributed to Ulpian remains distinctive. The
same is true of individual texts and sentences, though these have to be con-
sidered on their merits, one at a time. If a text contains a phrase such as
proinde, ut puta, or erit dicendum, Ulpian is likely to be its original author,
even if it has been partly altered in editing. Pastiche by post-classical editors
1
e.g. proinde, ut puta, et ait . . ., erit dicendum
106 5. Genuine Works
VIR. Examples of Ulpianic words and phrases were multiplied in the first edi-
tion. This edition is more selective.
The meaning of the word or phrase relied on as an indication of Ulpian’s
authorship is explained in chapter 2. For example, inde refers to the conjunc-
tion meaning ‘hence’, not the adverb meaning ‘thence’. Again proinde refers
to the word in the meaning ‘and so’, not to its use in a phrase such as proinde
ac si where it means ‘just (as if)’. The reader should bear in mind that some
expressions such as proinde (‘and so’) and quid tamen si (‘but what if?’) are so
often found in Ulpian and so rarely in other legal authors, that even standing
alone they point to genuineness.
14 (393) proinde; inde; et puto; ego puto; parvi refert; meminisse oportebit; cessabit;
erit dicendum; accipiemus; accipere debemus; saepissime; ut puta
15 (631) proinde; quid tamen si; et non puto; finge enim; et dicendum erit; sed enim;
parvi refert; aequissimum est/erit; dubio procul; accipiemus; accipies; erit dicendum;
ut puta
16 (369) proinde; et puto; et ait; sed enim; finge enim; erit accipienda; erit dicendum;
ut puta
17 (343) et puto; per contrarium; et ait; accipere debemus; erit dicendum
18 (660) proinde; inde; et ego puto; et generaliter; et ait; habet rationem; parvi refert;
aequissimum erit; erit danda; erit dicendum; cessabit; accipiemus; accipere debemus;
eo loci; ut puta
19 (377) et generaliter; et puto; ego arbitror; et ait; eleganter; ut puta
20 (100) habet rationem; cessabit; erit dicendum; eleganter; ut puta
21 (35) accipere debemus
22 (396) proinde; ego puto; et ait; habet rationem; consequenter dicemus; consequens
erit dicere; aequissimum est; erit accipiendum, accipere debemus; in ea causa est; ut
puta
23 (402) proinde; ego puto; quaestionis est; et finge; et est relatum; et ait; accipere
debemus; accipimus; erit dicendum
24 (315) proinde; inde; ego arbitror; et ait; nec non; erit dicendum; aequissimum est;
accipere debemus; accipimus; eleganter; eo loci; ut puta
25 (393) quid tamen si; et generaliter; ego puto; credo; finge enim; quaestionis est; et
ait; habet rationem; erit dandum; aequissimum visum est; cessabit; accipere debemus;
saepissime; ut puta
26 (404) proinde; quaestionis est; quid tamen si; et puto; et refert; quid tamen si; accip-
imus; non est ferendus; ut puta
27 (283) proinde; quid tamen si; et puto; et magis est; et generaliter; cessabit; eo loci;
ut puta
28 (678) proinde; quid tamen si; finge enim; et puto; ego puto; et est dubitatum; et
generaliter; et ait; nec non; parvi refert; admittimus; est constitutum; est rescriptum;
meminisse oportebit; aequissimum erit; accipiemus; accipere debemus; accipimus; erit
dicendum; cessabit; aequissimum est; ut puta
29 (1296) proinde; est quaestionis; quid tamen si; finge enim; si mihi proponas; et mihi
videtur; et puto; ego puto; et ait; nec non; habet rationem; parvi refert; spectamus; erit
probandum; erit dicendum; cessabit; est aequissimum; accipere debemus; accipimus;
in ea causa est; est constitutum; si mihi proponas; eleganter; ut puta
30 (400) proinde; inde; et arbitror; ego arbitror; et magis est; et constat; et ait; sole-
mus dicere; aequissimum erit; erit dicendum; eleganter; ut puta
31 (633) proinde; inde; quid tamen si; finge enim; et non puto; per contrarium; et ait;
idcircoque; verumtamen; quid tamen si; spectamus; aequissimum erit; est constitu-
tum; erit probandum; habet rationem; cessabit; erit dicendum; parvi refert; con-
sequenter dicemus; ut puta
5. Genuine Works 109
32 (772) proinde; quid tamen si; et putem; et puto; ego puto; credo; et arbitror; per
contrarium; et est relatum; et ait; verumtamen; habet rationem; ostendimus; solemus
dicere; erit dicendum; est aequissimum; erit accipiendum; accipere debemus; in ea
condicione est; eleganter; ut puta
33 (222) et puto; et constat; nec ferenda est; est rescriptum; dubio procul
34 (447) quaestionis est; et magis est; et putem; et magis puto; credo; et mihi videtur;
per contrarium; et ait; meminisse oportet; consequenter dicemus; aequissimum erit;
accipere debemus; saepissime; eleganter; ut puta
35 (779) proinde; et putem; et puto; et magis est; et generaliter; nec non; erit dubita-
tio; et ait; verumtamen; ostendimus; parvi refert; aequissimum erit; accipere debemus;
accipimus; cessabit; consequenter dicemus; consequens erit dicere; erit dicendum; in ea
condicione est; ut puta
36 (490) proinde; quid tamen si; et magis est; ego puto; et puto; et arbitror; per con-
trarium; et exstat; et generaliter; et ait; idcircoque; verumtamen; non tantum . . . verum
etiam; nulla dubitatio est; parvi refert; ostendimus; aequissimum erit; erit accipien-
dum; saepissime; ut puta
37 (233) proinde; et putem; et ego puto; parvi refert; nec ferendus est; ut puta
38 (345) proinde; quid tamen si; et magis est; et generaliter; sed et si . . . aeque; mem-
inisse oportebit; aequissimum est; accipere debemus; accipimus; cessabit; erit dicen-
dum; saepissime; eleganter
39 (288) proinde; et arbitror; et mihi videtur; credo; et magis est; idcircoque; verum-
tamen; nulla dubitatio est; solemus dicere; non tantum . . . verum etiam; aequissimum
est/erit; accipiemus; accipere debemus; erit dicendum; in ea causa est
40 (626) proinde; et ego puto; credo; et magis est; et ait; verumtamen; non tantum
. . . verum etiam; sed et si . . . aeque; nulla dubitatio est; admittimus; aequissimum
est/erit; erit dicendum; cessabit; consequenter dicemus; non est in ea causa; saepissime;
ut puta
41 (491) proinde; quid tamen si; et magis est; per contrarium; et generaliter; idcir-
coque; habet rationem; parvi refert; spectamus; admittimus; non tantum . . .verum
etiam; est aequissimum; accipere debemus; erit dicendum; ut puta
42 (84) et magis est; erit dicendum; spectamus
43 (72) aequissimum est; ut puta
44 (239) proinde; et magis est; et arbitror; et puto; et mihi videtur; et ait; idcircoque;
verumtamen; accipere debemus; accipimus; erit audiendus; erit probandum; ut puta
45 (267) proinde; et non puto; et generaliter; verumtamen; non tantum . . . verum
etiam; nulla dubitatio est; spectamus; aequissimum visum est; erit dicendum; con-
sequens erit dicere; in ea causa est; in ea condicione est; ut puta
46 (195) verumtamen; ostendimus; admittimus/admittemus; non tantum . . . verum
etiam; erit accipiendum; accipere debemus; accipiemus; in ea condicione est; ut puta
47 (40) verumtamen; sed enim
48 (36) nulla dubitatio est; accipiemus; accipere debemus; consequens erit dicere; ut
puta
110 5. Genuine Works
49 (138) proinde; et magis est; et generaliter; nulla dubitatio est; habet rationem; ut
puta
50 (652) proinde; et ego puto; et non puto; et magis est; et ait; quaestionis est/fuit; illud
sciendum est; parvi refert; est aequissimum; accipere debemus; accipimus; erit dicen-
dum; in ea causa est; eleganter; perquam; eo loci; ut puta
51 (71) illud sciendum est; aequissimum est; ut puta
52 (422) proinde; inde; et puto; et magis est; et ait; nec non; sed et si . . . aeque; mem-
inisse oportebit; et generaliter; non tantum . . . verum etiam; aequissimum erit;
accipere debemus; accipimus; eo loci
53 (622) proinde; inde; per contrarium; ego puto; et puto; et putem; et magis est; et
extat; et generaliter; et ait; idcircoque; non tantum . . . verum etiam; habet rationem;
aequissimum erit; accipe; accipiendum erit, erit probandum; erit dicendum;
spectabimus; cessabit; eleganter; ut puta
54 (87) sed enim; erit dicendum; ut puta
55 (175) proinde; et putem; et puto; et generaliter; parvi refert; accipiemus; accipere
debemus; accipimus; cessabit; non est in ea causa; ut puta
56 (424) proinde; et non puto; et ego puto; et magis est; et generaliter; sed enim; sed
et si . . . aeque; spectamus; admittimus; dicere solemus; accipiemus; accipere debemus;
accipimus; consequens erit dicere; erit probandum; cessabit; in ea causa est; in ea
condicione est; ut puta
57 (517) proinde; est quaestionis; et est rescriptum; et ait; nulla dubitatio est; parvi
refert; solemus dicere; meminisse oportebit; non tantum . . . verum etiam; aequissi-
mum erit; accipere debemus; accipmus; erit probandum; consequens erit dicere; non
est in ea causa; saepissime; ut puta
58 (72) et ait; idcircoque; accipiemus; accipere debemus; consequens erit dicere; in ea
condicione sunt
59 (209) proinde; et generaliter; ego puto; et puto; et ait; illud sciendum est; specta-
mus; solemus dicere; habet rationem; accipere debemus; erit probandum; erit dicen-
dum; cessabit; consequens erit dicere; perquam; in ea erit causa
60 (239) proinde; et puto; et magis est; sed et si . . . aeque; nec non; illud sciendum est;
habet rationem; non tantum . . . verum etiam; accipiemus; erit probandum; erit dicen-
dum; cessabit; consequens erit dicere; in ea causa est; eo loci
61 (158) proinde; et generaliter; accipimus; erit dicendum; in ea causa est; ut puta
62 (158) proinde; est quaesitum; ego puto; et ait; debemus accipere; erit dicendum
63 (117) sed enim; aequissimum est; accipimus; cessabit; ut puta
64 (166) et putem; et non puto; illud sciendum est; nec ferendus est; est aequissimum;
erit accipiendum; erit probandum; perraro; ut puta
65 (45) —
66 (125) proinde; si mihi proponas; et ait; illud sciendum est; accipere debemus; erit
danda; non est in ea causa; in ea condicione est; ut puta
67 (141) nec non; est quaesitum; aequissimum visum est; accipere debemus; accip-
imus; erit probandum; erit dicendum; cessabit
5. Genuine Works 111
68 (666) proinde; inde; et puto; et putem; et generaliter; dubio procul; solemus dicere;
aequissimum erit; accipimus; erit probandum; cessabit; consequens erit dicere; eo loci;
ut puta
69 (459) proinde; est quaesitum; ego puto; et ait; et generaliter; nec non; idcircoque;
nulla dubitatio est; fuit aequissimum; accipimus; consequenter dicemus; ut puta
70 (635) proinde; inde; et ego puto; credo; et magis est; et extat; et refert; et ait; et gen-
eraliter; idcircoque; sed enim; illud sciendum est; non tantum . . . verum etiam; aequi-
ssimum erit/est; accipe; accipimus; erit dandum; erit dicendum; erit probandum; in ea
causa est; in ea condicione est; eo loci; ut puta
71 (924) proinde; est quaesitum; ego arbitror; ego puto; ego putem; et mihi videtur; et
magis est; et exstat; et est relatum; et generaliter; et ait; idcircoque; illud sciendum est;
parvi refert; spectamus; admittimus; meminisse oportet; non tantum . . . verum etiam;
accipe; accipimus; consequens erit dicere; erit probandum; erit dicendum; in ea causa
est; saepissime; ut puta
72 (75) et puto; ego puto
73 (273) proinde; et putem; et puto; et magis est; verumtamen; habet rationem; est
constitutum; non tantum . . . verum etiam; accipimus; erit probandum; consequens
erit dicere; saepissime; perquam; ut puta
74 (168) proinde; ego puto; et ait; illud sciendum est; solemus dicere; aequissimum est;
accipere debemus; in ea erit causa; ut puta
75 (148) proinde; quaestionis est; si proponas mihi; et putem; et ait; et generaliter; erit
probandum; consequens erit dicere; in ea condicione est; ut puta
76 (483) proinde; et putem; et ego puto; si mihi proponas; et magis est; et refert; et gen-
eraliter; et ait; verumtamen; illud sciendum est; parvi refert; aequissimum est; erit
probandum; erit dicendum; consequens erit dicere; saepissime; ut puta
77 (201) quid tamen si; et magis est; illud sciendum est; accipiemus; accipere debemus;
consequens erit dicere; ut puta
78 (13) —
79 (467) proinde; et putem; et puto; ego puto; et generaliter; et ait; nec non; verumta-
men; sed enim; illud sciendum est; habet rationem; non tantum . . . verum etiam; aeq-
uissimum est; accipiemus; accipimus; erunt dicenda; consequens erit dicere; cessabit;
perquam; ut puta
80 (86) ego puto; et ait; ut puta
81 (181) et ait; ego puto; accipimus; consequens erit dicere; ut puta
All the books except 65 and 78 contain some typical word or expression. On
average one can expect one of these chosen words or expressions every thirty or
so lines, though there are naturally variations: the forty-five lines from book 65
do not have any. Despite its general consistency over the whole eighty-one
books, there are points at which Ulpian’s phrasing and vocabulary varies in the
course of the edictal commentary. These points, which provide clues to his cho-
sen method and rate of composition, are discussed in chapter 7.6 The abiding
6
Ch.7 n.101–19
112 5. Genuine Works
Much the same is true of Ulpian’s other major commentary, which concerns
civil, as opposed to praetorian law. Here, too, the coherence of style is striking.
22 (256) proinde; credo; et ego arbitror; parvi refert; accipiendum erit; ut puta
23 (324) proinde; inde; fuit quaestionis; quid tamen si; et puto; et refert; et ait; sed
enim; parvi refert; solemus dicere; meminisse oportet; consequenter dicemus; erit
dicendum
24 (294) proinde; inde; quid tamen si; finge enim; si mihi proponas; et non puto; et ait;
est rescriptum; erit dicendum; aequissimum; saepissime; eleganter; ut puta
25 (202) proinde; quid tamen si; et putamus; et puto; ego arbitror; et magis est; per
contrarium; et ait; aequissimum erit; accipimus; cessabit; ut puta
26 (139) proinde; per contrarium; et puto
27 (208) proinde; inde; quid tamen si; et puto; credo; sed et si . . .aeque; et parvi refert;
ut puta
28 (426) proinde; inde; quaestionis est; quid tamen si; et puto; ego arbitror; et gener-
aliter; et ait; sed et si . . .aeque; parvi refert; ostendimus; accipimus; erit accipiendum;
erit probandum; consequenter dicemus; eleganter; ut puta
29 (323) inde; per contrarium; et puto; nec non; verumtamen; non est in ea causa;
saepissime; eleganter; eo loci
30 (231) quid tamen si; si mihi proponas; et puto; et ait; accipere debemus; cessabit;
erit dicendum; eleganter; ut puta
31 (315) proinde; et putem; et non puto; ego puto; ego arbitror; habet rationem;
accipere debemus; cessabit
32 (443) proinde; quid tamen si; si mihi proponas; et magis est; et magis puto; per con-
trarium; et ait; nulla dubitatio est; illud sciendum est; habet rationem; consequenter
dicemus; eleganter; ut puta
33 (226) proinde; quid tamen si; et putem; et magis puto; et generaliter; spectamus/
spectabimus; admittimus; accipere debemus; erit dicendum; ut puta
34 (128) inde; et puto; et generaliter; ut puta
35 (68) et puto; est rescriptum; accipimus
36 (208) proinde; et magis est; et non puto; et ego non dubito; credo; verumtamen;
admittimus; ferendum non est; accipiemus; erit accipiendum; eleganter; ut puta
37 (64) et ait; nulla dubitatio est; admittimus; in ea causa est
38 (97) proinde; et non puto; non tantum . . . verum etiam
39 (41) proinde; in ea condicione est; erit dicendum
40 (150) proinde; sed enim; sed et si . . . aeque; est rescriptum; in ea causa est; ut puta
41 (320) proinde; inde; et magis est; et non puto; per contrarium; et generaliter; et ait;
verumtamen; sed enim; nulla dubitatio est; et parvi refert; eleganter; eo loci; ut puta
42 (278) proinde; per contrarium; ego puto; et non dubito; et ait; verumtamen;
accipere debemus; accipimus; ut puta
43 (229) proinde; ego non dubito; et magis est; et ait; spectamus; habet rationem; aeq-
uissimum visum est; erit dicendum; consequens erit dicere; eleganter
44 (157) et ego puto; et non dubito; sed enim; meminisse oportebit; in ea condicione
sunt
114 5. Genuine Works
(45) —
46 (151) et ait; erit dicendum; ut puta
47 (165) et parvi refert; et generaliter; cessabit
48 (203) proinde; per contrarium; et putem; et puto; erit probandum; erit dicendum;
accipimus; in ea causa est; ut puta
49 (309) et putem; et ego puto; et ego arbitror; solemus dicere; et magis est; et ait;
habet rationem; aequissimum videtur; erit dicendum; consequens erit dicere; in ea erit
causa; ut puta
50 (115) et puto; et generaliter; sed et si . . .aeque; consequens erit dicere; ut puta
51 (45) ut puta
The authenticity of Ulpian’s On Sabinus is not in doubt. The style is again
consistent. Thus, ut puta occurs in thirty-four books, proinde in twenty-seven,
and et ait . . . in seventeen. The typical expressions occur a bit less frequently
than in On the Edict, on average one in every thirty-four lines. Books 11, 16,
and 45 with seven, thirty-seven, and thirty-five lines respectively, lack the
chosen marks of the author’s style. But in general terms the excerpts from
books 16 (four uses of the future: locum habebit; repetetur; erit . . . necessar-
ius; valebit) and 45 (erit consequens; quid ergo si; et verum est) conform to his
manner.
There is good evidence for the authenticity of the treatises in ten and twenty
books: The Proconsul (De officio proconsulis), Disputations (Disputationes),
Tribunals (De omnibus tribunalibus), and The Julio-Papian Law (Ad legem
Iuliam et Papiam).
10 (195) per contrarium; et magis puto; nec non; credo; sed enim; accipere debemus;
meminisse oportebit; erit probandum; consequenter dicemus; perquam; eo loci; ut
puta
Books 5 and 6, with thirty-seven and twenty-four lines respectively, lack the
chosen marks of Ulpian’s style, but the other books contain enough to show
that Ulpian is the author. There is on average a typical word or expression
every thirty-three lines. This innovative treatise displays great political wis-
dom. Though Schulz says that the style departs from the classical,7 the words
and phrases he lists do not bear him out.
in style to the rest of Ulpian’s work, given that oral debate on points of law
conducted in public has its own conventions. Indeed we find on average a typ-
ical word or expression once in every twenty-eight lines, which is above aver-
age. Any passages and phrases that Justinian’s compilers have altered must
be winkled out one by one.
The two major commentaries and the four medium-scale works attributed to
Ulpian are, we have seen, coherent in style and can be accepted as substantially
genuine. The same is not true of all the minor works that stand in his name in
Justinian’s Digest or the Florentine Index. Some are demonstrably genuine.
Others are consistent in style with Ulpian’s authorship and, given that they are
attributed to him, are likely to be genuine. For some the evidence is so scanty
that no conclusion can be drawn from the texts on their own. Any inference of
genuineness rests on the inscriptions in the Digest or the attribution in the
Florentine Index alone. Others, to judge from the textual evidence, are spurious.
Ulpian’s work on trusts falls in the genuine class.
7. Trusts (6 books)
5 (404) proinde; quaestionis est; et putem; et exstat; et est rescriptum; ostendimus; est
constitutum; est rescriptum; aequissimum erit; erit dicendum; in ea causa est/sunt;
saepissime; eleganter; ut puta
6 (112) quid tamen si; et magis est; et non puto; eo loci; ut puta
There is evidence of authenticity from all six books. On average a typical
phrase or word comes every twenty-eight lines.
8. Taxation (6 books)
9. Adultery (5 books)
Books and lines:
1 (46) finge autem; erit accipiendum
2 (200) proinde; quid tamen si; ego arbitror; ego credo; et putem; et puto; et gener-
aliter; accipimus; non est in ea causa; ut puta
3 (79) quid tamen si; nulla dubitatio est; admittimus; ut puta
4 (120) et non puto; verumtamen; sed et si . . . aeque; nulla dubitatio est; accipere
debemus; eo loci
5 (26) proinde; accipere debemus; accipimus
All five books speak to authenticity. A typical word or phrase occurs on aver-
age every twenty lines.
11 D 50.15.1.2, 3, 5, 10, 11 (1 cens.) 12 D 50.15.1 pr (1 cens.); ch.2 n.654–66
13 Ch.2 n.470 14 D 50.15.1 pr (Ulp. 1 cens.) 15 AE 1988.1051
5. Genuine Works 119
et est simile; est autem manumissio; aequitate motus) though the number of
typical words and phrases is small.
This brings to an end the list of works that can be attributed to Ulpian on
grounds of style alone.
To the eight lines from Justinian’s Digest must be added some 206 from the
Vatican Fragments, some as fragmentary as the traditional name suggests.
The linguistic arguments for the genuineness of this work are inconclusive.
The general impression created by the texts is, however, consistent with
Ulpian’s manner. Thus quia est iniquum followed by the accusative and
infinitive is a striking inversion.17 Habent excusationem,18 habent immuni-
tatem,19 and habent vacationem 20 instead of excusationem/immunitatem/
vacationem habent conform to the word-order he favours.21 Non com-
putabitur 22 is a use of the future also found in Ulpian’s On Sabinus.23 Scio
dubitatum . . . invenio tamen24 illustrates a form of reasoning that has paral-
lels in his other works.25 Hence Excuses falls in the second class of works
mentioned above. It is consistent with Ulpian’s authorship but does not
point unequivocally to it. Schulz argued that it was not a work of Ulpian.26
But three texts in it reappear virtually unchanged in The Praetor for
Guardianship (no.16 above),27 so that its genuineness does not depend on
internal evidence alone. When the whole evidence is taken into account, it
can be classed as genuine. But it is an early work, composed before the death
of Septimius Severus,28 and before the author’s style had fully matured. The
Praetor for Guardianship is a second edition of Excuses, composed about
215.29
17 Frag. Vat. 153 cf. non est iniquum followed by accusative and infinitive (D 43.29.3.13, Ulp.
71 ed.)
18 Frag. Vat. 124; D 27.1.8.9 (Mod. 3 excus.) cf. Frag. Vat. 142 19 Frag. Vat. 134, 138
20 Frag. Vat. 137 21 e.g. non habet excusationem (D 50.5.13.2, Ulp. 23 ed.)
22 Frag. Vat. 161 cf. D 24.3.7.12 (Ulp. 31 Sab.) 23 D 24.3.7.12 (Ulp. 31 Sab.)
24 D 27.1.15.16 (Mod. 6 excus.)
25 Scio quaesitum (D 49.1.3 pr, 1 appell.; 28.5.35.1, 4 disp.); scio tractatum (43.8.2.33, 68 ed.);
invenio rescriptum (36.1.15.4, 4 fid.), invenio relatum (39.3.1.20, 53 ed.); invenimus rescriptum
(38.17.2.47, 1 Sab.)
26 Schulz (1961) 318–20
27 D 27.1.7 (1 excus.) = Frag. Vat. 185, 240 (1 off. pr. tut.). Frag. Vat. 145 (1 excus.) = part of
Frag. Vat. 222 ( 1 off. pr. tut.). D 27.1.15.16 (1 excus.) is a slightly shortened version of Frag. Vat.
189 (1 off. pr. tut.)
28 Lenel (1889) 2.8991; Liebs (1997) §424.10 29 Ch.7 n.104–12
122 5. Genuine Works
The three lines, though the author does not say so, are taken from a constitu-
tion of Marcus.34 The construction is non-Ulpianic, and, if the source were
not known, would even suggest that the work was spurious. As it is, no con-
clusion can be drawn.
1(3) —
1(?) —
The Digest contains two excerpts from a one-book work by Paul on this sub-
ject.35 The second of them contains a passage strongly reminiscent of Ulpian
(quaestio in eo est . . . et ait Papinianus).36 The first phrase occurs, apart from
this text, only in four texts of Ulpian.37 The et ait . . . formula is found, apart
from this text,38 117 times in Ulpian and only once in Paul.39 The suspicion is
that Paul drew on an earlier work by Ulpian on the same subject. He could
have done so, since he was still active in the reign of Alexander.40
In the upshot, the works that, on the basis of style, are clearly genuine, com-
prise:
On the Edict (81 books)
On Sabinus (51 books)
The Proconsul (10 books)
Disputations (10 books)
Tribunals (10 books)
The Julio-Papian Law (20 books)
Trusts (6 books)
Taxation (6 books)
Adultery (5 books)
Appeals (4 books)
The Aelio-Sentian Law (4 books)
The Consul (3 books)
Teaching Manual (2 books)
The Market-Masters’ Edict (2 books)
The Community Treasurer (1 book)
The Urban Prefect (1 book)
The Praetor for Guardianship (1 book)
The works which are on linguistic grounds consistent with authenticity com-
prise:
Excuses (1 book)
The Quaestor (1 book)
(hypothetically) The Speech of Marcus and Commodus (1 book?)
Those which are neutral from a stylistic point of view, and which can be
assigned to Ulpian only on the basis that the sources make this attribution
and the evidence of the texts does not contradict it, comprise:
35
D 23.2.20; 23.2.60 (Paul 1 or. d. Ant. et Comm.)
36 37 38
D 23.2.60.4 (Paul 1 or. d. Ant. et Comm.) Ch.2 n.155 Ch.2 n.99–102
39 40
Ch.2 n.101–2 D 31.87.3 (14 resp.), 49.1.25 (20 resp.)
124 5. Genuine Works
41 Lenel (1889) 1.595–620 nos.37, 47, 107, 118, 200; Reggi (1954) 21–86; Wolf (1959) 524–7,
451–509; H. Krüger (1930) 303–12; Wieacker (1960) 341f.; Santalucia (1965) 99–146; Schulz
(1961) 277–9; Liebs (1997) §424.5
43 D 9.2.41 pr (41 Sab.). 47.10.11.7 (57 ed.)
44 D 20.1.27 (Marc. 5 dig.), 26.7.28.1 (8 dig.), 29.7.9 (9 dig.) 45 D 29.7.19 (Marc. 14 dig.)
46
D 38.5.12 (Iav. 3 epist.); Eckardt (1978) 132f.
47
D 27.8.1.2 (36 ed.) cf. ch.2 n.171 (non dubitamus) in a typically future version
48
D 20.1.27 (Marc. 5 dig.)
49
D 50.8.4 (Pap. 1 resp.), 3.5.30.2 (2 resp.), 37.6.9 = P. Krüger (1881) 86, 88, 89 (all 5 resp.);
Frag. Vat. 66 (7 resp.)?; P. Krüger (1884) 171, 176, 177; D 40.4.50 = P. Krüger (1884) 173, 179
(all 9 resp.)
50 51
CTh 1.4.1 (28 Sept. 321/4) C Deo Auctore 6 (15 Dec 530)
52 53
D 16.3.24 (Pap. 9 quaest.) Ch.2 n.77–81
5. Genuine Works 125
belong to his maturity, not his youth, since Papinian’s Replies are dated to the
period 206 to 21254 and his Questions to the early or middle 190s.55
Finally we must take note of the 95 rescripts that Ulpian composed for
Septimius Severus between April 203 and May 209.56 These forceful and self-
confident rulings stand comparison with any body of third-century rescripts.
From Ulpian’s point of view, however, they count as an early work.
Moreover, rescripts were written out, not dictated, so that they strike one as
rather more formal than Ulpian’s later works.
H O W good a scholar was Ulpian? What first-hand sources did he use? Is his
reputation for lucid and accurate exposition based on the accident that
more of his work has survived than that of other Roman lawyers? Views have
differed.
In 1885 Pernice wrote an article, ‘Ulpian as a Writer’.1 In it he criticized
Ulpian’s works, in particular On the Edict, as unoriginal and unsystematic.
Each bit of the commentary was written in isolation, so that, over the whole
work, the same subject recurred twice or more. The author used arguments of
policy and convenience to support doctrinally wrong solutions. His work was
aimed at judges with little legal expertise. He often cited the views of earlier
lawyers from secondary sources without disclosing that he was doing so. ‘The
aim of this contribution will be achieved’, Pernice concluded, ‘if it succeeds in
putting people on their guard against Ulpian’s writings from various points
of view. When he speaks in his own name he may have copied from others.
When he cites others he does not usually go back to the original source. When
he asserts positive, historical facts without authority his statements must be
taken with the greatest reserve. When he attempts to develop the law on his
own initiative, he lacks for the most part incisiveness and creative power.’2
In 1905 Ulpian found a partial defender in Jörs.3 In his article for
Pauly–Wissowa Jörs came to the rescue so far as the use of sources was con-
cerned. Ulpian had used a wider range of original legal writings than Pernice
allowed for. When he used secondary sources he was following current prac-
tice. But Jörs was not wholehearted. He doubted that the Tyrian had com-
posed the bulk of his works for the first time in the reign of Caracalla. That
would have been impossible in the time available. In particular, adapting a
point made by Mommsen,4 Jörs argued that Edict up to book 52 was drafted
in the reign of Severus and then revised under Caracalla, when the later books
were added. If Ulpian’s repute as a scholar rose from the ashes, his claim to
speed of composition sank.
The views of Pernice and Jörs were influenced by notions drawn from mod-
ern scholarship and, in part, from the intellectual climate of the time. Pernice
failed to see that in the ancient world to cite from secondary sources was often
a necessity. The author has read the original and made notes on it, but does
1 Pernice (1885) 443 2 Pernice (1885) 484 3 Jörs (1903) 5.1435, 1455
4 Ch.7 n.71–86
6. Sources and Scholarship 127
themselves to their work day after day, week after week. They were expected to
be regular and assiduous. In seeking, then, to assess Jörs’s opinion that Ulpian
could not have written for the first time what he appears to have written under
Caracalla, we must disabuse ourselves of the atmosphere of a modern law
faculty or government office. If Ulpian’s career has been correctly dated in
chapter 1,8 he held the office of secretary for petitions (a libellis) from 202 and
began writing systematically only from 213. He had at least ten years to read
and make notes in the intervals of official duty. He had access to past imperial
rescripts and to recent constitutions. He could arrange excerpts from what he
read under title headings according to the scheme of the praetor’s edict and the
Digests (Digesta) of authors such as Celsus, Julian, and Marcellus. When the
time came to launch the ambitious programme of 213–7 Ulpian needed to do lit-
tle original research. It was enough to return to the main sources and to supple-
ment these from notes, weaving the whole together in a way that was coherent
from the point of view of topic rather overall structure. The sources having been
assembled, he saved time and kept the flow by dictating.9
Apart from Labeo and Paul, Ulpian’s main sources are not in doubt, at least
for the major commentaries On the Edict and On Sabinus. He used the
Questions (Quaestiones) and Replies (Responsa) of Papinian, the Digests
(Digesta) of Celsus, Julian, and Marcellus, and the commentaries of
Pomponius On the Edict and On Sabinus. Pernice also listed among main
sources the Questions and Replies of Scaevola.10 There is no evidence, how-
ever, that Ulpian used Scaevola’s Replies, and references to his Questions are
sporadic. Pernice accuses Ulpian of using Scaevola’s Questions without
acknowledgement. But Ulpian tends to generalize what Scaevola presents as
the solution to a problem presented by a particular set of facts.11
Pemice was right, however, to stress that in the major commentaries Ulpian
often took a single source as the main basis of his exposition of a topic. Thus
on false guardians12 Ulpian relies mainly on Pomponius. For possession
against the terms of a will13 he follows Marcellus. On the Lex Aquilia14 Celsus
is the leading source. On legacy of usufruct or use15 Julian is the model, with
supplements from Celsus and Marcellus. Deposit16 draws on Pomponius,
Julian, and Marcellus. In many titles On the Edict and On Sabinus the author
seems to use a selection of these sources and passes from one to the other as
occasion prompts.
9 10
Hagendahl (1971), Schlumberger (1976) Pernice (1885) 459
11
Compare D 46.3.27 (Ulp. 28 ed.) with 45.1.131.1 (Scaev. 13 quaest.)
12
Quod falso tutore auctore gestum esse dicatur
13 14
De bonorum possessione contra tabulas Ad legem Aquiliam
15 16
De usufructu et usu legato Depositi vel contra
6. Sources and Scholarship 129
17
D 33.9.3.2 (Ulp. 22 Sab., citing Aristo, who cites Labeo libro nono posteriorum)
18
Honoré–Menner (1980) intro. para. 4 for the percentages and fiches 1, 5, 6, 11, 15, 19, 25,
34, 41, 45 and 62 for the libro/libris references
130 6. Sources and Scholarship
than they. Which authors does he cite? Five stand out, as will be seen from the
table that follows, which gives the number of Ulpian’s book references to
other authors.
Legal authors cited by Ulpian by book
Julian 172
Pomponius 84
Papinian 65
Marcellus 63
Celsus 62
Neratius 17
Labeo 11
Pedius 9
Cassius 7
Scaevola 7
Ofilius 6
Sabinus 5
Maecianus 3
Mucius 3
Africanus 1
Arrianus 1
Puteolanus 1
Servius 1
Tertullian 1
Venuleius Saturninus 1
The book citations for five authors, Julian, Pomponius, Papinian, Marcellus,
and Celsus are by far the most numerous. Why did Ulpian choose them?
Papinian is the only contemporary to appear in the list. Clearly Ulpian
was a close associate of Papinian, the only contemporary lawyer whom he
cites in the imperfect, which points to oral discussion.19 A hint that he learnt
something from Papinian personally is worth noting. Ulpian is reticent about
his teachers. Apart from himself,20 he uses noster of no one except the
emperor,21 meus only of a fellow citizen of Tyre.22 In Edict book 69 he dis-
cusses the common view (quod volgo dicitur) that we can keep possession of
winter and summer pastures during the intervening months when we are away
by intention alone (animo). Ulpian says that he learnt (didici) that Proculus,
who said that we can, meant this as an example of a general principle about
retaining possession.23 From whom did he learn this? A difficult passage
of Papinian,24 which has been altered by the compilers, seems to treat the
19 D 24.1.23 (6 Sab.) 20 Ch. 2 n.300 21 e.g. ch.7 n.21–4, 30–6, 50–6, 88–90
22 D 45.1.70 (11 ed: populari meo Glabrioni Isidoro)
23 D 43.16.1.25 (69 ed.) cf. 41.2.27 (Proc. 5 epist.)
24 D 41.2.44.2; 41.2.46 (Pap. 23 quaest).
6. Sources and Scholarship 131
The references to Pomponius36 are fewer than half those to Julian, but in
long stretches of major commentary Ulpian clearly followed Pomponius’ On
the Edict (83 books)37 and On Sabinus (36 books).38 There are also references
to Pomponius’ Letters (Epistulae),39 to his Readings (Variae lectiones),40 and
his work on stipulations,41 the latter not available to Justinian’s compilers.
Ulpian probably read these lesser works, but used them only marginally.
Against fifty-one citations from Pomponius On the Edict and twenty-four
from On Sabinus we have nine from lesser works. Pernice pointed out that
Ulpian used Pomponius in fits and starts. Some titles seem to be derived
almost exclusively from him; in others he is absent. The reason for this per-
haps lies in the scale on which Pomponius wrote. His On the Edict outran all
other edictal commentaries. Hence for many topics Ulpian, working to a plan
that confined his edictal commentary to a fixed number of books, could not
afford to reproduce Pomponius in detail. In others, where Pomponius was
relatively economical, he could.
The last major source is Iuventius Celsus the younger.42 Ulpian cites by
book his Digest (Digesta), in 39 books,43 his Letters (Epistolae),44 and his
Questions (Quaestiones).45 Only the first of these was available to Tribonian,
and all but three of Ulpian’s citations appear to come from it.
The sources on which Ulpian principally relied in his major commentaries
were therefore seven:
Julian’s Digest 171 book citations from 90 books
Marcellus’ Digest 63 book citations from 31 books
Celsus’ Digest 59 book citations from 39 books
Pomponius On the Edict 51 book citations from 83+ books
Pomponius On Sabinus 24 book citations from 36 books
Papinian’s Questions 36 book citations from 37 books
Papinian’s Replies 28 book citations from 19 books
Relatively speaking Papinian’s Replies (if we assume that Ulpian had only the
first fourteen books available), Marcellus’ Digest, and Julian’s Digest were
the most thoroughly excerpted. Together the seven sources provided material
for almost every title of Ulpian’s major commentaries but not for some of the
more specialized works. But is the list of major sources complete? In particu-
lar ought Labeo and Paul to feature in it?
36
Lenel (1889) 2.15–159; Kunkel (1967) 170–1; Liebs (1997) §422
37 38
Lenel (1889) 2.15–44 cf. D 38.5.1.14, 27 (Ulp. 44 ed.) Lenel (1889) 2.86–148
39 40
Lenel (1889) 2.52–3 Lenel (1889) 2.53–8
41
Lenel (1889) 2.151; D 7.5.5.2 (18 Sab.)
42 43
Lenel (1889) 1.127–170; Kunkel (1967) 146–7 Lenel (1889) 1.127–69
44
Lenel (1889) 1.169; D 4.4.3.1 (11 ed.)
45
Lenel (1889) 1.169; D 12.1.1.1 (26 ed.); 28.5.9.2 (5 Sab.); 34.2.19.3 (20 Sab.)
6. Sources and Scholarship 133
In the Digest Ulpian cites the innovative Augustan lawyer Antistius Labeo
by book only eleven times.46 Three come from Labeo’s commentary on the
urban edict,47 one from his work on the edict of the peregrine praetor,48 six
from his Posthumous Works (Posteriora),49 one from his Plausible Views
(Pithana).50 One would expect more. Ulpian’s references to Labeo total 350
overall, which puts Labeo second only to Julian, who is cited 601 times. Is the
reason for the small number of book citations that Ulpian nearly always cited
Labeo at second hand? There is evidence that Labeo may be underrepre-
sented as regards book citations.
Ulpian cites earlier authors by book throughout the Sabinian commentary
and sporadically in certain other works: Trusts, Disputations, Tribunals, Office
of Proconsul, Office of Quaestor, Appeals. In Edict, however, book citations,
though numerous, continue only till the end of book 52.51 Is this because, as
Jörs thought,52 they were put in during Ulpian’s supposed first draft, made in
the reign of Severus, which stopped at book 52? Or is this feature due to
Justinian’s compilers, who stopped excerpting references to book-numbers at
this point?
According to the theory set out in chapter 8 Ulpian reached book 56 by the
end of 215. He continued with Sabinus books 26 to 51 in 216.53 Even if Ulpian
changed his way of citing at the end of 216 one might expect book citations in
books 53 to 56 On the Edict. There are none. Perhaps this was an accident of
excerpting. Moreover Appeals (De appellationibus), probably to be assigned
to 217,54 does contain a book citation.55 Even more telling, passages where
Ulpian cites his source with inquit, ‘his words are’, introducing an exact quo-
tation, usually run parallel to those in which he cites by book. Inquit passages
are found in the later books of Edict.56 The aim of these two devices is simi-
lar, to give the reader the exact words of the source or point the reader to the
place where he can find the exact words. So Ulpian is not likely to have kept
inquit but jettisoned citations by book in the last twenty-nine books of Edict.
More probably the compilers struck out book citations but kept inquit. To
strike out book citations did not change the sense of the text, but striking out
inquit would.
In my view the edictal committee, which progressively reduced the amount
of material it excerpted, followed this policy.57 Its second stint in excerpting
46 Lenel (1889) 1.502–58; Kunkel (1967) 32–4
47 D 11.4.1.5 (1 ed.); 50.16.19 (11 ed.); 4.8.7 pr (13 ed.) 48 D 4.3.9 4a (11ed.)
49 D 4.3.9.3 (11 ed.); 28.5.13.5, 6; 28.5.17.5 (7 Sab.); 7.8.2.1 (17 Sab.); 33.9.3.2 (22 Sab.)
50
D 46.4.8.2 (48 Sab.)
51
D 39.1.1.10; 39.1.5 pr; 39.1.5.16 (all 52 ed. citing Cels. 10 dig, Iul. 12 dig and Iul 49 dig.
respectively)
52 53 54
Jörs (1903) 1439–40 Ch. 8 n.15–6 Ch. 9 n.115–21
55
D 49.4.1.14 (1 appell., citing Iul. 40 or 45 dig.)
56
D 47.10.13.4 (57 ed.); 43.8.2.40 (68 ed.); 43.16.1.35; 43.17.3.4 (69 ed.); 41.2.6 pr; 43.19.1.11
(70 ed.); 43.24.3.4; 43.23.1.8 (71 ed.); 39.1.21.7 (81 ed.)
57
Honoré–Rodger (1970) 246, 296–306
134 6. Sources and Scholarship
Ulpian’s Edict began about the middle of book 52.58 If the stint really began
at book 53 this would explain the absence of book citations from then on
better than a supposed change of practice by Ulpian himself. The other com-
mittees need not have followed suit, so that a book citation, as mentioned, is
found in Appeals, probably excepted by the Sabinian committee in 217.59 It is
to be noted that book citations are absent from Ulpian’s Market-Masters’
Edict, which was composed about 213, but which Justinian’s compilers
excerpted after book 81 On the Edict. The change of policy stemmed from the
compilers, not Ulpian.
Despite Jörs, then, Ulpian probably continued to cite sources by book
throughout Edict. This bears on the paucity of book citations of Labeo in the
last block of edictal commentary, when Labeo is mentioned no fewer than 105
times. Indeed in this part of On the Edict he outstrips all other sources, includ-
ing Julian. The compilers have obscured the extent to which Ulpian, with the
original text of Labeo before him, cited Labeo by book.
I mentioned Ulpian’s use of inquit, which introduces an exact quotation.
Ulpian refers to named authors with inquit in the Digest 165 times. There are
only thirty uses of inquit by other Digest authors: Labeo 4, Celsus 1, Gaius 3,
Pomponius 1, Africanus 3, Paul 14, Marcianus 4. Ulpian purports, therefore,
to cite another author verbatim far more often than his predecessors. The
authors whom he cites in this way largely correspond to those in the list of
book citations set out earlier.60 Here is the list for inquit:
Julian 57, Pomponius 20, Celsus 16, Labeo 13, Marcellus 12, Papinian 12, Pedius 7,
Vivianus 5, Masurius Sabinus 3, Neratius 3, Caelius Sabinus 3, Scaevola 2, Africanus
2, Aristo 2, Cassius 1, Maecianus 1, Nerva, 1, Octavenus 1, Fulcinius 1, Quintus
Mucius 1, Ofilius 1.
The main difference between this list and the earlier list of book citations is
that Labeo appears about as often as Celsus, Marcellus, and Papinian. So it
seems that Labeo should be added to the list of Ulpian’s main original
sources. It is not likely that the verbatim citations of Labeo are taken from
Celsus or Pomponius, since these authors hardly ever cite with inquit or by
book. Ulpian cites Labeo with inquit in Edict books 31, 53, 57, 62, 68, 71, 76,
and 81 and Sabinus books 17 and 49.61 More than half the inquit references
come from the last segment of Ulpian’s edictal commentary. This supports
the view that Justinian’s edictal committee struck book references from this
block of edictal commentary and so led to an underestimate of the extent to
which Ulpian used Labeo as an original source. Jörs saw more clearly than
43.8.2.40 (68 ed.); 43.23.1.8; 43.24.5.12; 43.24.7.3; 43.30.3.4 (71 ed.); 44.4.4.15 (76 ed.); 39.1.12 pr
(81 ed.); 7.1.12 pr (17 Sab.); 18.4.2.17 (49 Sab.)
6. Sources and Scholarship 135
Pernice that, while many citations of Labeo are taken from secondary
sources, quite a lot are not.62
The list of sources referred to by book or with inquit excites puzzlement
because, of contemporary authors, only Papinian is thus cited. Why does
Ulpian not cite Julius Paulus (Paul),63 a younger contemporary who had pub-
lished works On the Edict and On Sabinus shortly before, and had written sev-
eral monographs on topics that were to concern Ulpian also? Ulpian does not
cite Tryphoninus,64 an eminent contemporary, or Callistratus,65 the latter
perhaps because he was a provincial writer. He cannot have been ignorant of
Paul’s published works. Yet he seems to overlook them.
Was he, as has often been suggested, motivated by personal rivalry?66 The
suggestion has some apparent merit. Ulpian wrote on many of the topics that
Paul had covered not long since, but adopts different methods. For instance
he cites by book or with inquit far more often than Paul.67 He seems to be
writing for a more scholarly readership. He is more concerned than Paul to
justify the provisions of the praetor’s edict.68 But this last feature of Ulpian’s
work can be explained by the need to convince the new citizens, subject since
the Antonine constitution to Roman law, of the merits of the edict. He may
have thought that Paul’s works needed to be recast to meet the new dispen-
sation. Even if there was some rivalry, this does not imply that he had a low
opinion of Paul as a lawyer.
There is some reason to think that Ulpian, in composing his major com-
mentaries, made use of Paul’s work. It would, in the first place, have been
strange not to profit from recent publications by a lawyer of standing and a
member of the imperial council. Secondly, as an earlier study suggested,69 it
is easier to explain the textual chains in the Digest if we suppose that for long
stretches the main commentaries used by Justinian’s compilers ran parallel.
This made possible a technique of excerpting by which, for example, Ulpian,
Paul, and Gaius On the Edict could be read together.70 Ulpian’s was taken
as the main text. Short excerpts from the others were added when their
commentaries diverged. This parallelism was only possible, if, to take the
same example, Ulpian used Paul On the Edict and Paul used Gaius On the
Provincial Edict. The argument is no doubt suggestive rather than com-
pelling. There are also a few passages in which it seems likely that Paul used
Ulpian as a source, again without naming him.71
62
Jörs (1903) 1477f, noting D 39.3.1.20 (53 ed.: apud Labeonem autem invenio relatum); 7.8.2
(17 Sab: apud Labeonem memini tractatum) against Pernice (1885) 476f.
63
Lenel (1889) 1.951–1308; Kunkel (1967) 244–5; Liebs (1997) §423
64
Lenel (1889) 2.351–378; Kunkel (1967) 231–3; Liebs (1997) §417.3
65
Lenel (1889) 1.82–106; Kunkel (1967) 235; Liebs (1976) 310f., 323f.; (1997) §430.1
66 67 68
e.g. H. Krüger (1912) 298 Above n.18–9 Ch.3 n.143–63
69 70
Honoré (1963) 362 Honoré–Rodger (1970) 250f.
71
D 23.2.60.4 (1 or. d. Ant et Comm.), ch.5 n.35–40; and perhaps 27.9.13.1 (1 or. d. Severi)
136 6. Sources and Scholarship
72
Honoré–Menner (1980) fiches 19, 20
73
D 20.4.12.5, 6, 9 (Marci. 1 hyp.); 20.1.11.2 (1 hyp.); 48.21.3; 49.14.18.10 (1 del.); 48.17.1.4
(2 pub. iud.); 30.113.5 (7 inst.)
6. Sources and Scholarship 137
74 D 12.4.13 (Marci. 3 reg.); 40.5.55.1 (4 reg.); 34.9.3 (5 reg.); 40.15.1.4; 49.14.18.10 (1 del.);
Tribunals, and Trusts. But all these works by Ulpian, apart from the first five
books of Edict, were written in 213 or later,92 when Papinian, who was killed in
later 212, was dead.93 In these first five books, written when Papinian was
alive,94 Ulpian refers to him only once in the words ‘I know Papinian gave a
reply,’ which is a reminiscence, not a quotation from a published work.95
It is true that in On Sabinus book 6 Ulpian appears to cite his younger con-
temporary Marcianus.96 But Sabinus went into a second edition,97 and this
isolated reference to Marcianus could be a later addition to the text. There is
no parallel elsewhere in Ulpian to the sentence ending ut . . . videtur,98 an
untypically weak ending. This phrase may therefore belong to the second edi-
tion of On Sabinus, the date of which we do not know. Marcianus himself,
writing after the death of Caracalla, cites no one later than Papinian,99 not
even Ulpian, which is remarkable given Ulpian’s evident influence over him.
Paul once refers to a reply of Ulpian cited to Paul by someone consulting
him.100 He disagreed with it. This is not a citation from a published work,
and, though Ulpian gave replies (responsa), it is not clear that he published
any.101 In five texts Paul refers to Papinian. In one he speaks of Papinian
hearing a case as praetorian prefect.102 Two others mention views expressed
by Papinian as a member of the Severus’ council.103 The other two passages
cite opinions from book 5 of Papinian’s Replies.104 The first is from a work
composed by Paul in Caracalla’s sole reign.105 The other cannot be dated, but
the passage in question is probably copied from a work of Ulpian.106 In that
case it was written in 213 or later.107 There is nothing to show, then, that Paul,
any more than the other Severan writers, cited living authors.
Given this, the problem is not to explain, by reference to rivalry, why Paul
and Ulpian do not overtly cite one another but rather why the Severans, unlike
their predecessors, do not refer to living authors. They may have been afraid of
giving offence, of iniuria. If it is in order to cite a living author, it may be insult-
ing not to cite what he has written, or to disagree with it. It is safe, on the other
hand, to express one’s opinion of a dead author or to pass over his view in
silence. Until recently English judges had a convention that the works of living
authors should not be cited in court.108 The point was to save both judge and
author embarrassment. The Severan lawyers may have had a similar sensitivity.
92 Ch.9 n.65–6, 11–34, 140 93 Dio 78.4.1a; Liebs (1997) §416
94 Perhaps in early 211: ch.1 n.249–250
95 D 2.14.7.5 (4 ed: responsum scio a Papiniano)
96 D 28.1.5 (6 Sab: iam enim complesse videtur annum quintum decimum, ut Marciano videtur)
97 C. Cordi (16 Nov 534) 3 98 Honoré–Menner (1980) fiche 74
99 Fourteen times e.g. D 48.16.14.5, 10, 13 (1 SC Turpillianum)
100 D 19.1.43 (5 quaest.) 101 Ch.10 n.4–5, 219–57 102 D 12.1.40 (3 quaest.)
103 D 29.2.97; 49.14.50 (3 decr.)
104 D 27.9.13.1 (Paul 1 or. d. Sev.); 23.2.60.4 (1 or. d. Ant. et Comm.)
105 D 27.9.13 pr (1 or. d. Sev: imperator Antoninus et divus pater eius)
106 Ch.2 n.101–2, 155, 832 107 Ch. 9 n.137–40
108 Nicholls v. Ely Beet Sugar Factory Ltd. [1936] Ch. 343, 349
6. Sources and Scholarship 139
The list of authors cited by book or with inquit affords evidence of other first-
hand sources used by Ulpian. I begin with those nearest to him in time.
(a) Cervidius Scaevola109 cannot be accounted a major source but Ulpian
cites him seven times by book110 and twice with inquit.111 All the explicit ref-
erences are to Scaevola’s Questions (Quaestiones). In all he mentions Scaevola
31 times.112 It is not clear that Ulpian used any other work of the Antonine
lawyer who taught Paul and Tryphoninus. He refers to him in his major com-
mentaries and four times in his Disputations, which correspond to Scaevola’s
Questions.113 There is one citation in Ulpian’s Adultery,114 which may how-
ever be taken from Paul’s monograph on the same subject.115
(b) Sextus Caecilius Africanus,116 a pupil of Julian who was active under
Pius, is cited once by book117 and twice with inquit.118 There is no evidence
that Ulpian used the nine books of his Questions, which were available to
Justinian’s compilers.119 The work Ulpian refers to in On Sabinus is
Africanus’ otherwise unknown Letters (Epistolae) in at least twenty books.120
Perhaps the work consisted of Africanus’ correspondence with Julian, which
Ulpian had acquired. The inquit passages come in Ulpian’s Adultery.121 In all
Ulpian cites Africanus only eight times.122 His relative neglect of a well-
known, though unoriginal, author may show that he preferred to rely on the
text of Julian rather than Africanus’ comments on his teacher’s work.
(c) Ulpian cites Volusius Maecianus,123 a member of Marcus’ judicial
council and prefect of Egypt, thrice by book124 and once with inquit.125 All
four passages come from Maecianus’ sixteen books on Trusts. Ulpian
mentions Maecianus sixteen times in all,126 ten in his Trusts127 and thrice in On
Sabinus.128 One citation in On Sabinus is to Maecianus’ Trusts.129 Ulpian must
109 Lenel (1889) 2.215–322; Kunkel (1967) 217–9; Liebs (1997) §415.6
110 D 4.4.11.1 (11 ed.); 41.3.10.2; 50.16.26 (16 ed.); 13.4.2.3 (27 ed.); 28.6.10.6 (1 Sab.); 7.1.25.6
(18 Sab.); 41.1.23.3 (43 Sab.)
111 D 13.4.2.3 (27 ed.); 41.3.10.2 (16 Sab.) 112 Honoré–Menner (1980) fiche 71
113 D 12.1.7 (Ulp. 1 disp.); 23.3.43 pr (3 disp.); 36.1.23 pr (5 disp.); 35.2.35 (6 disp.)
114 D 23.5.13.4 (5 adult)
115 Paul was a pupil of Scaevola: D 2.14.27.2 (Paul 3 ed.); 23.3.56.3 (Paul 6 Plaut.); 28.2.19
have used this work directly. In The Proconsul Ulpian cites Maecianus when
dealing with a law about murder, the Lex Pompeia de parricidiis.130 This prob-
ably refers to Maecianus’ fourteen-book work on public prosecutions (De
iudiciis publicis), which was available to Justinian’s compilers,131 and possibly
to Ulpian. It is not clear from what intermediate source he could have cited it.
(d) Neratius Priscus,132 a leading lawyer of the reign of Trajan, is cited sev-
enteen times by book133 and three times with inquit.134 The citations show that
Ulpian used Neratius’ Parchments (Membrana),135 Replies (Responsa),136
Letters (Epistulae),137 and his edition of Plautius.138 Only the first two reached
Justinian’s compilers. The citations come in Ulpian’s On the Edict, his On
Sabinus, and once in Disputations. In all Ulpian names Neratius as a source
seventy-six times.139 A good many of these citations are probably first-hand.
They include references in books on the praetor’s and the market-masters’
edicts from which the edictal committee seem to have struck out book cita-
tions.140
(e) Ulpian cites Vivianus, a late first- or early second-century writer, five
times with inquit but never by book.141 The reason for the lack of book cita-
tions is probably, again, that many of the citations come in the later books On
the Edict or in The Marker-Masters’ Edict. Scaevola142 and Ulpian used
Vivianus’ collection of material, which included excerpts from republican
authors. They refer to it with phrases such as ‘it is stated in Vivianus’ or ‘I find
in Vivianus’. Ulpian cites him 19 times in all, always in On the Edict or The
Market-Masters’ Edict.143 Justinian’s compilers did not have his work to
hand.
( f ) Sextus Pedius, a writer of the second half of the first century AD,144 is
cited nine times by book145 and six times with inquit.146 Ulpian mentions him
forty times in all.147 His works were not available to Tribonian, but he com-
posed a commentary on the praetor’s edict in at least twenty-five books. This
130 D 48.9.6 (8 off. proc.) 131 Lenel (1889) 1.587–8
132 Lenel (1889) 1.763–88; Kunkel (1967) 144–5
133 D 5.3.13.3 (15 ed.); 8.3.3 pr, 2, 3 twice; 8.3.5.1 (17 ed.); 12.4.3.5 (26 ed.); 14.6.7 pr ., twice);
15.1.9.1 (29 ed.); 19.1.11.12 (32 ed.); 13.1.12.2 (38 ed.); 7.1.7.3, 7.2.3 pr (17 Sab.); 33.7.12.35, 43
(20 Sab.); 36.1.23.3 (1 disp.)
134 D 10.4.9.8 (24 ed.); 33.7.12.43 (20 Sab., twice)
135 D 5.3.13.3 (15 ed.); 8.3.3 pr, 2, 3 (17 ed.); 12.4.3.5; 13.1.12.2 (26 ed.); 7.1.7.3 (17 Sab.)
136 D 14.6.7 pr; 15.1.9.1 (29 ed.); 19.1.11.12 (32 ed.); 7.2.3 pr (17 Sab.); 36.1.23.3 (5 disp.)
137 D 33.7.12.35, 43 (20 Sab.) 138 D 8.3.5.1 (17 ed.)
139 Honoré–Menner (1980) fiche 64
140 D 47.10.1.8, 9 (56 ed.); 47.10.7.5 (57 ed.); 42.4.7.16 (59 ed.); 43.24.7.1 (71 ed.); 20.2.3 (73
ed.); 44.2.9.1, 11 (75 ed.); 44.4.4.18 (76 ed.); 21.1.15.3 (1 ed. aed. cur.)
141 D 43.16.1.47 (69 ed.); 21.1.1.9 (1 ed. cur. aed., thrice); 21.1.17.4 (1 ed. cur.)
142 D 29.7.14 pr (Scae. 8 quaest.) 143 Honoré–Menner (1980) fiche 75
144 Lenel (1889) 2.1–10; Kunkel (1967) 168–9; Liebs (1997) §421.2
145 D 3.5.5.11 (10 ed.); 4.2.7 pr; 4.2.14.5; 4.3.1.4 (11 ed.); 4.7.4.2; 4.8.7 pr; 4.8.13.2 (13 ed.);
Ulpian cites by book in Edict books 10 to 29.148 The citations in the later
books of Edict, which lack book references, are either direct or come via Paul,
who also used Pedius as a source.149 Pedius also wrote on the market-masters’
edict.150 This work was used by Ulpian, perhaps directly, though in that case
the edictal committee has left out Ulpian’s book references to Pedius.151
(g) Probably for the same reason Caelius Sabinus,152 who was consul in AD
69 and succeeded Cassius as head of the Sabinian school, is cited three times
with inquit153 but never by book. The inquits come from Ulpian’s Market-
Masters’ Edict, which dealt with a topic on which Caelius wrote. Ulpian men-
tions him thirteen times in this commentary, fourteen if a text with ‘Caecilius’
is amended.154 Ulpian used Caelius as a main source for this work, but not in
other works. Caelius’ works were not available to Justinian’s compilers.
(h) Cassius Longinus,155 conservative co-founder of the Sabinian school,
restored from exile by Vespasian, is cited seven times by book,156 once with
inquit.157 No work of Cassius was available to Justinian’s compilers. Ulpian
cites from his Civil Law (Libri iuris civilis) up to book 10.158 All the citations
come in Ulpian’s On Sabinus. Presumably Cassius’ work was based on
Sabinus’ Civil Law. The inquit passage, on the other hand, refers to a note of
Cassius in his work dedicated to Vitellius.159 It looks as if Ulpian consulted
these notes and Cassius’ Civil Law at first hand. He mentions Cassius eighty-
five times in all.160 Many of these texts must have been taken at second hand
from Ulpian’s main sources, such as Celsus, Pomponius, and Julian.
(i) Ulpian cites Masurius Sabinus161 five times by book 162 and three times
with inquit.163 The book citations refer not to the well-known Civil Law (Ius
civile) of Sabinus, the text on which Ulpian’s On Sabinus was a commentary,
but to his books dedicated to Vitellius (Libri ad Vitellium). One of the inquit
passages, however, seems to refer to Sabinus’ Civil Law.164 Vitellius was not
a legal writer, as assumed by Lenel165 and others, but someone to whom
Sabinus dedicated one of his works.166 This work must have been available to
Ulpian, who could have used it as a direct source. In all Ulpian refers to
148 Above n.145 149 D 12.1.6 (Paul 28 ed.); 37.1.6.2 (41 ed.) 150 Lenel (1889) 2.6–7
151 Excerpted by Justinian’s compilers after On the Edict book 81, though written by Ulpian
in 213 or 214 some years beforehand: ch.9 n.26–33
152 Lenel (1889) 1.78–82: Kunkel (1967) 131–3
153 D 21.1.17.1, 16 (1 ed. cur.); 21.1.38.11 (2 ed. cur.)
154 Honoré–Menner (1980) fiche 51 and D 21.1.14.10 (1 ed. cur.)
155
Lenel (1889) 1.110–26: Kunkel (1967) 130–1
156
D 29.2.25.4 (Ulp. 8 Sab.); 7.1.7.3; 7.1.9.5; 7.1.23.1; 7.1.70 pr, 2 (17 Sab.); 26.1.3.2 (37 Sab.)
157 158
D 33.7.12.27 (20 Sab.) D 7.1.70 pr (17 Sab.)
159
D 33.7.12.27 (20 Sab.). Vitellius was a dedicatee, not a legal author: Liebs (1980) 138–9
160 161
Honoré–Menner (1980) fiche 51 Lenel (1889) 2.187–216; Kunkel (1967) 119–20
162
D 33.7.8 pr ; 33.7.12.27; 34.2.19.17 (Ulp. 20 Sab.); 32.45; 33.9.3 pr (22 Sab)
163 164
D 33.9.3 pr (22 Sab.); 18.2.11 pr (28 Sab.) D 18.2.11 pr (28 Sab.)
165 166
Lenel (1889) 1.1301–8 Liebs (1980) 138–9
142 6. Sources and Scholarship
Sabinus 119 times,167 but many of these citations must be taken from later
writers such as Pomponius, Celsus, and Julian.
(j) Working back, we now come to the republican authors. Ofilius168 is
cited six times by book169 and once with inquit,170 Quintus Mucius171 three
times by book172 and once with inquit.173 Pernice denied that any republican
texts were available to Ulpian.174 Jörs argued that these two at least were read
by him.175 In my view Ulpian excerpted these works from Sabinus’ work ded-
icated to Vitellius.176 The citations from Ofilius’ Actions (Libri actionum) and
Divisions of the Law (Libri iuris partiti) come in Ulpian’s Sabinus books 22
and 25.177 Those from Q. Mucius’ Civil Law (De iure civili) seem to come in
Ulpian’s Sabinus books 22, 25, and 44.178 But, as Lenel points out, the text in
book 44 is out of place, and we should read book 24 instead of 44.179 The text
is concerned not with the edict of the curule aediles, the subject of book 44,
but with a legacy of silverware. Finally, the books of Sabinus dedicated to
Vitellius are mentioned only in books 20 and 22 On Sabinus.180 Books 20–5
of this work are on legacies.181 Sabinus in this work, I suggest, collected texts
on legacies, which included excerpts from Q. Mucius and Ofilius, and Ulpian
drew on this collection. It is interesting that the text mentioning the two
republican authors is, uniquely, in the plural.182 A joint citation of this sort is
more natural if the source is a collection of readings. In all Ulpian cites Ofilius
thirty-two times183 and Q. Mucius twenty-one,184 but many of these citations
are likely to come at second hand from Labeo’s Posthumous Works
(Posteriora), Celsus, or Pomponius.
So far all the authors listed have been cited both by book and with inquit,
so that we can be fairly sure that Ulpian cited them from an original source
or collection of material. The rest he cites in only one mode, and for the most
part once only. So the citation is more likely to have been taken from a sec-
ondary source. By book Ulpian cites Tertullianus’ Questions (Quaestiones),185
Puteolanus’ Assessorship (Adsessoriorum libri),186 Arrianus’ On Interdicts (De
interdictis),187 Quintus (Venuleius) Saturninus’ On the Edict (Ad edictum),188
and Servius’ work dedicated to Brutus (Ad Brutum).189 Tertullianus
(Tertullian), plausibly to be identified with the Christian apologist, wrote
167 Honoré–Menner (1980) fiche 71 168 Lenel (1889) 1.795–804; Kunkel (1967) 29–30
169 D 33.9.3.5, 8 (Ulp. 22 Sab.); 32.55.1, 2, 4, 7 (25 Sab.) 170 D 33.9.3.9 (22 Sab.)
171 Lenel (1889) 1.757–64
172 D 33.9.3 pr (Ulp. 22 Sab.); 32.55 pr (25 Sab.); 33.2.27 (44 Sab.)
173 D 33.9.3.9 (22 Sab.) 174 Pernice (1885) 475f. 175 Jörs (1903) 1475–6
176 177 178
Above n.159, 163–4 Above n.169–70 Above n.172
179
Lenel (1889) 2.11791 180
Above n.162 181
Lenel (1889) 2.1079–109
182
D 33.9.3.9 (22 Sab: Quintus Mucius et Ofilius inquiunt, of which the natural translation is
‘Q. Mucius, cited with approval by Ofilius, says’.)
183 184
Honoré–Menner (1980) fiche 65 Honoré–Menner (1980) fiche 69
185 186 187
D 29.2.30.6 (Ulp. 8 Sab.) D 2.14.12 (Ulp. 4 ed.) D 5.3.11 (Ulp. 15 ed.)
188
D 34.2.19.7 (Ulp. 20 Sab.). On his citations of (Venuleius) Saturninus see Liebs (1984) 449
189 101
D 14.3.5.1 (Ulp. 28 ed.); Liebs (1980) 138
6. Sources and Scholarship 143
later than Pomponius but before Ulpian.190 His Questions in eight books were
read by Justinian’s commission. Ulpian, a near contemporary who kept to
legal studies, would naturally have read them. In his Sabinus he cites
Tertullian three times.191 Puteolanus seems to have written in the late second
century and Ulpian presumably cites him at first hand.192 The Tyrian also
refers to Arrianus,193 a first- or more probably second-century author, who
wrote On Interdicts and other works, four times.194 He probably read On
Interdicts, but the other citations may be second-hand. Servius is the famous
republican author Servius Sulpicius Rufus, consul in BC 51.195 Ulpian men-
tions Servius thirty-eight times in all.196 Most of these citations will be sec-
ond-hand, from Labeo’s Posthumous Works or Pomponius, from whom
Ulpian may have taken the book citation.197 The citation of Quintus
Saturninus On the Edict,198 whom Ulpian cites in a text about a legacy of
gold, may rest on a mistake as to author or title. The definition of gold is not
likely to have been discussed in a work On the Edict.199 Saturninus, if Ulpian
cited him, is perhaps to be identified with Venuleius Saturninus200 rather than
Claudius Saturninus, writers of the early and late second century respec-
tively.201 They were mistakenly identified with one another in the Florentine
Index and by Lenel,202 but distinguished by Kunkel and Liebs.203
One last jurist whom Ulpian cites by book204 is Urseius.205 He was a first-
century AD author whose work Julian edited.206 Ulpian, who does not tell us
the title of Urseius’ work, cites him four times, always for an opinion of
Sabinus,207 Cassius,208 or Proculus.209 Ulpian used his work, it seems, at first
hand but only as a source of the views of more eminent writers.
Some authors appear only in one or two inquit texts and are not cited by
book. Ulpian twice uses inquit of Titius Aristo,210 a distinguished lawyer of
the late first century of humble origin, a freed slave or the descendant of a
freed slave.211 He has forty-five references to Aristo, but most of them will
have been taken from Neratius, Pomponius, or Paul, who used Aristo as a
190 Lenel (1889) 2.341–4; Steinwenter (1934); Kunkel (1967) 236–40; Barnes (1985) 22–9;
source. It is not certain that Ulpian knew Aristo’s work, which was not avail-
able to Justinian’s compilers, at first hand. Octavenus, an author of the same
period,212 is cited once with inquit.213 Pomponius and Paul both cite
Octavenus, Pomponius no doubt directly. Ulpian mentions him eight
times,214 once with ut apud Pomponium scriptum est,215 which shows that on
that occasion he took Octavenus’ opinion from Pomponius. In other texts
Ulpian cites Octavenus in the imperfect, which again indicates dependence on
Pomponius.216 Probably Ulpian did not read Octavenus at first hand but
took the references from Pomponius and Paul. Octavenus’ work was not
available to Justinian’s compilers. Ulpian once cites Nerva the elder with
inquit217 in a text from book 71 On the Edict.218 The author in question is the
emperor Tiberius’ friend M. Cocceius Nerva, consul suffect in AD 22. Ulpian
cites him fifteen times.219 It is not clear that Ulpian read his work at first hand:
the references may come from collections of material by Atilicinus220 or
Vivianus.221 Nerva’s writings did not survive to the age of Justinian. Lastly
there is one inquit text222 for Fulcinius Priscus,223 an author of the first century
AD whom Pomponius and Paul also cite, Paul twice in the imperfect.224
Ulpian’s inquit, which is one of his only two references to Fulcinius, may be
derived from Pomponius or Paul.
In sum, Ulpian was widely read. He had a well-stocked library, especially
of more recent authors. The sources he read first-hand were probably more
extensive than appears from the Digest. Thus Vatican Fragments 59–64,
70–1, 74–89, which contain excerpts on usufruct from Ulpian’s Sabinus book
17, have a wider range of citations and more book references than the corre-
sponding passages that appear in the Digest.225 Justinian’s compilers were
obliged to cut down on references in order to keep the Digest within the pre-
scribed limits, and Ulpian, who likes to refer to original sources when he can,
will have suffered disproportionately from this policy. That is not to say that
Ulpian eschews citation at second hand when the original or a verbatim quo-
tation is not available. On the contrary, he freely resorts to it, as do other
Roman lawyers. To these second-hand citations we now turn.
materials by Fulcinius that included opinions of Labeo and Mela: below n.269–70
225 Lenel (1889) 2.1056–65
6. Sources and Scholarship 145
226 Above n.163–6, 176 227 Above n.35, 141–3 228 Above n.205–9
229 D 4.8.21.1 (13 ed.) 230 D 19.1.13.22 (32 ed.) 231 D 39.3.1.17 (53 ed.)
232 D 43.24.1.7 (71 ed.) 233 D 30.30.7 (Ulp. 19 Sab.) 234 D 33.9.3.9 (22 Sab.)
235
D 18.1.79 (Labeo et Trebatius); 24.3.66.1 (Labeoni Trebatio); 24.3.66.2 (Labeo Trebatius);
28.6.39 pr (Labeonis, Ofilii, Cascelii, Trebatii); 28.6.39.2 (Ofilius Cascellius); 32.29 pr (Cascellius
Trebatius); 32.29.1 (Quintus Mucius et Gallus); 32.100.1 (Ofilius Trebatius); 32.100.4 (Labeo
Trebatius); 33.4.6.1 (Ofilius Cascellius); 33.6.7 pr (Ofilius Cascellius Tubero); 33.7.4 (Labeo
Trebatius); 33.7.25.1 (Labeo Trebatius); 33.7.26.1 (Labeo Cascellius Trebatius); 33.10.10 (Labeo
Ofilius Cascellius); 33.10.11 (Labeo Trebatius); 34.2.39 pr (Ofilius Labeo); 35.1.40.4 (Labeo Ofilius
Trebatius); 40.7.39 pr (Quintus Mucius Gallus et ipse Labeo . . . Servius Ofilius); 40.7.39.1 (Labeo
Ofilius); 40.7.39.4 (Labeo Ofilius . . . Labeonis et Ofilii); 49.15.27 (Labeo Ofilius Trebatius)
236
The only exception is D 34.2.39 pr (Iav. 3 post. Lab.)
237
D 10.3.6.6 (Ulp. 19 ed.); 40.7.3.11 (27 Sab.)
238
D 4.3.9.3 (11 ed.); 28.5.13.5, 6; 28.5.17.5 (7 Sab.); 7.8.2.1 (17 Sab.); 33.9.3.2 (22 Sab.)
146 6. Sources and Scholarship
Ulpian gives Iavolenus’ opinion immediately after that of Labeo, which also
points to his using Iavolenus’ edition or epitome of Labeo.239 Whether Labeo
made the collection of opinions which form the basis of the strings of citation
himself, or used someone else’s collection, is not clear. Certainly his collection
was distinct from that apparently to be found in Sabinus’ work dedicated to
Vitellius,240 which was confined to legacies.
Atilicinus241 is an author whose name often appears as a member of a
string, for example: Proculus Atilicinus;242 Nerva Atilicinus;243 Sabinus
Atilicinus.244 Sometimes the order is reversed, as in: Atilicinus Sabinus
Cassius;245 Atilicinus Nerva Sabinus.246 Ulpian sometimes inserts et between
the names.247 Julian248 and Paul do not.249 But clearly all three are directly or
indirectly using the same collective source, probably a collection made by
Atilicinus himself. Atilicinus was a mid-first-century contemporary of
Proculus;250 so he was either junior to or coeval with the other authors listed.
His name, however, appears either first or last in the lists. When it comes last
it is in the natural chronological position of the editor of a collection of opin-
ions by others. When it comes first this is a sign that a later author is consult-
ing the edited work (and so names the editor first) and then reverts to
chronological order when he comes to name the authors cited by the editor.
This is why, in the lists Atilicinus Sabinus Cassius251 and Atilicinus Nerva
Sabinus,252 the latest writer is named first and the earliest second. If, on the
other hand we find Atilicinus in the middle, as in Neratius Atilicinus
Proculus,253 the source is not the collection of juristic opinions by Atilicinus
but some work of Neratius or (at two removes) Pomponius in which use was
made of Atilicinus’ collection. Ulpian cites Atilicinus eleven times,254 but may
not have used his collection directly. Though he does say of one opinion of
Atilicinus that it exists, he does not claim to find any information ‘in
Atilicinus’ (apud Atilicinum). His sources for the Atilicinus citations are likely
to have been one or more of Neratius, Pomponius, Julian, and Paul, all of
whom he used directly.
Atilicinus, then, collected opinions of Nerva, Proculus, Sabinus, and
Cassius and added his own. Ulpian probably cited him at second hand and
the authors whose opinions he collected at third hand. The form of citation
depends on the intermediate source. Thus, Pomponius usually mentions his
source first. Taking Aristo’s opinion from a work of Neratius, he cites
Neratius before Aristo,255 though Aristo was senior.256 So when Ulpian men-
tions Atilicinus first in a list, his source is probably Neratius or Pomponius
rather than Celsus or Julian, who prefer chronological order.
Titius Aristo annotated Labeo,257 Sabinus,258 and Cassius259 and may also,
Lenel thinks, have published a collection of legal sources, no doubt with com-
ments of his own.260 Ulpian cites Aristo’s notes on Cassius261 and cites Aristo
45 times in all.262 But he does not seem to have used Aristo’s collection of
material, if it existed, directly. He does not cite texts ‘in Aristo’ (apud
Aristonem). His source is rather Neratius, Pomponius, or Paul. Pomponius
made free use of Aristo’s work which, to go by a Paul text, he seems to have
edited.263
Ulpian seems, then, to have used the collections of material by Sabinus,
Labeo, Urseius, and Vivianus directly but those of Atilicinus and perhaps
Aristo only indirectly. Does this conclusion adequately account for his use of
ex(s)tat and apud? He uses ‘there exists’ (ex(s)tat) of opinions of Ofilius,264
Servius,265 Sabinus,266 Cassius,267 Nerva,268 Atilicinus,269 and Neratius.270
All these might be found in the works of Sabinus ad Vitellium, Labeo,
Urseius, Vivianus, Neratius, or Pomponius, if we assume that the opinion of
Nerva and Atilicinus was reproduced by Neratius or Pomponius. The list of
authors in whose work (apud quos) Ulpian finds information is longer. They
comprise Mucius, Servius, Servii auditores, Ofilius, Trebatius, Labeo, Mela,
Vitellius, old writers (veteres), Sabinus, Cassius, Minicius, Pedius, Plautius,
Urseius, Celsus, Vivianus, Pomponius, Julian, Marcellus, Scaevola, and
Tertullian.271 Most of these are authors whom Ulpian consulted at first hand
or whose works were excerpted or edited by others whom he consulted at first
hand. The two who do not fit this scheme are Mela and Plautius.
Fabius Mela272 was a lawyer of the early empire, probably a contemporary
of Labeo. Ulpian cites him thirty times,273 and also cites opinions of Aquilius
Gallus and Servius ‘in Mela’ (apud Melam).274 Another Ulpian text couples
the opinions of Labeo and Mela.275 Paul couples Mela and Fulcinius.276
Africanus cites the tenth book of some work of Mela.277 It is not easy to see
from what secondary source Ulpian derived his citations of Mela. One possi-
bility is a collection of material by Fulcinius Priscus.278 Texts of Paul lead one
to think that he used such a collection, which included opinions of Labeo and
Mela.279 Ulpian may have had access to the same source, or may have read
Mela at first hand.
Plautius is discussed in the next section.
When Ulpian cites from collections of lawyers’ opinions he cites second hand,
but the editor of the collection generally gives the text of the original author
verbatim. In the instances now to be considered Ulpian takes from his pri-
mary source, especially his main sources, opinions that he finds mentioned in
them, but not necessarily verbatim. That he often did this is obvious. Ulpian
makes no attempt to hide it. For example he says ‘I read that Cato writes
. . .’,280 ‘It is said that Mauricianus thought . . .’,281 ‘Tubero defines, as Celsus
reports . . .’.282
Pernice reproached Ulpian for citing at second or third hand.283 He was
particularly critical of his habit of citing at second hand without saying so.
That Ulpian did this is, again, clear. Discussing the vesting of annuities, he
favours vesting at the beginning of the year: ‘Labeo, Sabinus, Celsus, Cassius,
and Julian approved this’.284 The Digests of Celsus and Julian are among
Ulpian’s main sources.285 For Celsus, Labeo and Sabinus are important
sources.286 Julian often cites Cassius.287 On the other hand Celsus cites
Cassius only once 288 and Julian cites Labeo only twice.289 We can infer that
in Ulpian’s text the citations of Labeo and Sabinus are taken from Celsus,
that of Cassius from Julian. Hence the text, if expanded, would read ‘Celsus
argues for the vesting of an annuity at the beginning of the year and cites in
support Labeo and Sabinus. Julian also favours the beginning of the year and
cites in support Cassius.’ There are a number of parallels, in which the indir-
ect source precedes the main source, as in: ‘and so Sabinus and Celsus
write’290 or ‘Labeo and Marcellus write’.291 The citation of Sabinus is taken
from Celsus, and that of Labeo from Marcellus. Another way of presenting
an indirect citation is illustrated by: ‘Celsus writes, and Tubero approves that
278 279
Above n.60–1, 222–4 D 31.49.2 (Paul 5 Iul. Pap.); 25.2.3.4 (7 Sab.)
280 D 21.1.10.1 (1 ed. cur: Catonem quoque scribere lego)
281 D 41.10.1.1 (15 ed: Mauricianus dicitur existimasse)
282 D 15.1.5.4 (29 ed: Tubero definit, ut Celsus refert) 283 Pernice (1885) 473f.
284 D 36.2.12.1 (23 Sab: et Labeo Sabinus et Celsus et Cassius et Iulianus hoc probaverunt)
285 Above n.9–10, 18–9 286 Honoré (1962b) 137–40 287 Honoré (1962b) 155–60
288 D 33.7.12.20 (Ulp. 20 Sab.) 289 D 13.4.2.8 (Ulp. 27 ed.); 44.4.4.1 (76 ed.)
290 D 43.26.8.1 (71 ed: et ita Sabinus et Cassius scribunt)
291 D 12.5.4.3 (26 ed: Labeo et Marcellus scribunt)
6. Sources and Scholarship 149
opinion.’292 The late-republican writer Tubero could not approve the opinion
of his second-century AD successor Celsus. So Ulpian here mentions first the
source, Celsus, then the author cited in the source, Tubero. We can be sure of
this because Celsus cites Tubero as often as he cites any author except
Proculus.293 It says something for Ulpian’s scholarly integrity that in three of
the five texts in which he cites Tubero he makes it clear that Celsus is the
source of the citation. One could argue from this for his direct use of, say,
Mela as a source.294
When Ulpian uses a primary source A, who cites his own primary source B
first, then a third source C, cited by B, he generally follows the order BCA.
Thus ‘Neratius and Aristo and Pomponius approve . . .’.295 Ulpian is here
using Pomponius as a primary source, Pomponius has used Neratius, and
Neratius has cited Aristo. Pomponius tends to cite his source in reverse
chronological order in such cases,296 and Ulpian does the same. There can be
slightly more complicated cases in which Ulpian’s primary source has himself
used two primary sources. An example is ‘Labeo, Neratius, and Aristo think
. . .’.297 Here it is likely that Ulpian has taken the citations from Pomponius,
who himself used two sources, Labeo and Neratius, and found Aristo’s opin-
ion in Neratius. On the other hand when Ulpian puts Aristo in the right
chronological order as in ‘and that was the view of Aristo, Neratius, and
Julian’298 it is likely that Ulpian is using Julian alone as a primary source,
since Julian when he cites normally keeps to the correct chronological
order.299 Sometimes it is difficult to be sure of the primary source, for exam-
ple ‘Labeo and Sabinus think and we approve . . .’300 or ‘Labeo and Cassius
write.’301 Here it is not likely that Ulpian is using Labeo at first hand. The pri-
mary source might be Celsus, Julian, Neratius, Pomponius, or collections of
material by Urseius, Vivianus, or Fulcinius. There are many possibilities.
To go further along these lines would be to analyse the use of sources by
Ulpian’s own main sources. This lies outside the scope of the present work.
But there remain some authors cited by Ulpian who have not yet been men-
tioned. Aufidius Chius,302 a lawyer of the Flavian period, is once cited for an
opinion of Atilicinus.303 Ulpian once cites Publicius, who wrote under
Hadrian or the early Antonines,304 Paconius,305 cited by Paul306 as well as
292 D 7.8.2.1 (17 Sab: Celsus scripsit, quam sententiam et Tubero probat)
293 Honoré (1962b) 137–40 294 Above n.272–9
295 D 28.5.9.14 (5 Sab: et ita Neratio et Aristoni videtur et Pomponius probat); 7.2.3.2 (17 Sab.)
296 Above n.255–6 297 D 28.5.9.14 (5 Sab: Labeo Neratius et Aristo opinantur)
298 D 35.1.7 pr (18 Sab.) 299 D 9.1.24.1 (15 dig.); 12.4.7 pr (16 dig.); 40.12.30 (5 Minic.)
300 D 19.1.11.3 (32 ed: et Labeo et Sabinus putant et nos probamus)
301 D 28.2.6 pr (3 Sab: et Labeo et Cassius scribunt)
302 Kunkel (1967) 135–6; Martial 5.61.10; apparently a freedman
303 Lenel (1889) 1.75–6; Frag. Vat. 77 (17 Sab.)
304 D 38.17.2.8 (13 Sab: Africanus et Publicius temptant dicere); Lenel (1889) 2.186; Hanslik
Ulpian but of unknown age and Cartilius,307 a lawyer of the Augustan age or
early empire. Ulpian cites Plautius,308 a writer of the Flavian dynasty whose
work was edited by Iavolenus, Neratius, Pomponius, and Paul, only once.309
Even then he cites it not for Plautius’ opinion but for that of the authors
whom he cites.310 Presumably Ulpian knew his work but did not think highly
of him. Marcus Vindius Verus,311 consul in AD 138, is mentioned thrice.312
One text suggests derivation from Julian,313 another from Pomponius.314
Iunius Mauricianus,315 active under Pius, comes in twice, once with
‘Mauricianus is said to have thought’,316 which is clearly second hand. The
citation may have come to Ulpian through Paul, who cites him.317
There are a number of members of the ‘Proculian’ school, deriving from
Labeo and Proculus, whom Ulpian cites but does not seem to have known at
first hand. The earliest is M. Cocceius Nerva the elder.318 The famous
Proculus,319 active under Nero, is cited forty-five times320 but never by book
or with inquit. Ulpian does not seem to have read his epistulae, though these
were available to Justinian’s compilers. Ulpian cites a note by Proculus on
Labeo, for which Celsus is his source.321 Proculus’ notes on Labeo’s
Posthumous Works may have come to Ulpian via Iavolenus.322 Virtually all
Ulpian’s main sources cite Proculus. Hence there is no need to suppose that
Ulpian read him in the original. Nor need we assume that he had read Nerva
the younger,323 father of the emperor of AD 96–8, who was a contemporary of
Proculus. Ulpian cites him five times, once along with Pegasus.324 The
chronological order is upset, so that this citation may come via Pomponius.
The next head of the Proculian school is Pegasus.325 Like Aufidius and Aristo
he was a man of humble origin whose career illustrates how law was then as
later a channel of social mobility. Pegasus rose to be urban prefect under
Vespasian. Ulpian cites him twenty-three times, but never by book or with
inquit.326 There is evidence that Pomponius327 and Julian328 were the sources
from which Ulpian derived these citations. Pegasus is often cited in combina-
tion: Trebatius et Pegasus,329 Labeo et Pegasus,330 Sabinus et Pegasus,331
307 Lenel (1889) 1.106; D 13.6.5.13 (28 ed.); Kunkel (1967) 122–3
308 Lenel (1889) 2.13–4; Kunkel (1967) 134 309 D 7.2.1.3 (Ulp. 17 Sab.)
310 Ibid: omnes enim auctores apud Plautium de hoc consenserunt
311 Lenel (1889) 2.1223–4; Kunkel (1967) 167–8 312 Honoré–Menner (1980) fiche 75
313 Frag. Vat. 77 (17 Sab: Vindius tamen, dum consulit Iulianum)
314 D 5.1.5 (5 ed: Pomponius et Vindius scripserunt) cf. 2.9.2.1 (Paul 6 ed.)
315 Lenel (1889) 1.690–2; Kunkel (1967) 176–7 316 D 41.10.1.1 (15 ed.)
317 D 6.1.35.1 (Paul 21 ed.) 318 Above n.217–21
319 Lenel (1889) 2.159–84; Kunkel (1967) 123–9; Honoré (1962b) 472–93
320 Honoré–Menner (1980) fiche 67 321 D 3.5.9.1 (10 ed.)
322 D 7.8.1.4 (17 Sab.) 323 Lenel (1889) 1.791–2; Kunkel (1967) 130
324 D 3.2.2.5 (6 ed: Pegasus et Nerva filius responderunt)
325 Lenel (1899) 2.9–12; Kunkel (1967) 133–4
326 Honoré–Menner (1980) fiche 66 327 D 7.1.12.2 (17 Sab.)
328 D 7.1.25.7 (18 Sab.) 329 D 41.1.41 (9 ed.)
330 D 33.7.12.3 (20 Sab.) 331 D 12.5.4 pr (26 ed.)
6. Sources and Scholarship 151
V. IMPERIAL SOURCES
Private jurisprudence was not the whole story, nor even, in Ulpian’s time, the
main theme. Imperial constitutions were taking centre stage. These were still
primarily rescripts,345 documents addressed to private individuals or officials
answering questions put to the emperor, sometimes about law.346 In answer-
ing them the emperor’s practice was to state the existing law and a lawyer
drafted his reply. If the emperor wished to legislate, he had recourse to an
edict, such as the Antonine constitution, or a resolution of the senate (sena-
tusconsultum) passed at the emperor’s instance, for which the imperial address
(oratio) was now treated as the authoritative text.
We need to be alive to a certain irony. Ulpian cites many constitutions of
Severus, Severus and Caracalla, or Caracalla alone.347 When he cites one with
approval, he is not infrequently citing a text in the preparation of which he or
Papinian or Menander or some other professional colleague played a role.
His situation is not so different from that of Tribonian, praising himself in
texts issued by Justinian but composed by himself as quaestor.348
The table that follows349 presents a picture of the gradually changing
weight of imperial constitutions in relation to private jurisprudence. To
ensure that frequency of citation is a reliable index of the author’s practice,
the authors listed in rough chronological order are confined to those for
whom we have at least 3,000 words in Justinian’s Digest. So far as citing pri-
vate authors is concerned, jurists vary a great deal. Those most given to cita-
tion, Labeo, Pomponius, and Ulpian, come from three different centuries.
What they have in common is a lively interest in scholarship. Those authors
who are mainly oriented towards practice, like Scaevola and Papinian, cite
earlier authors sparingly. So, for a different reason, does Callistratus, a
provincial writer mainly concerned with criminal and public law,350 an impe-
rial domain. Yet when all necessary allowances have been made, Ulpian
stands out in the Severan age for his concern with the scholarly literature, a
concern stronger than Paul’s and much stronger than that of Marcianus, the
two contemporaries who mainly expound private law.
The extent to which the various authors refer to imperial constitutions
varies with the topic but, in contrast with their citing private authors, the fre-
quency changes over time. In the first century imperial law is a negligible ele-
ment in legal writing. In the second its weight relative to private scholarly
opinion grows. Callistratus writes in an eastern environment in the reign of
Severus but cites him with or without Caracalla only four times in eighty-nine
345 346
Coriat (1997) 647–8 Honoré (1994) 33–70 with literature
347
There are carefully compiled tables in Coriat (1997) 639–41, 647–8
348 349
Honoré (1978) 204 Honoré–Menner (1980) relevant fiches and intro. § 4
350
Lenel (1889) 1.82–106; Kunkel (1967) 235; Liebs (1997) §430.1
6. Sources and Scholarship 153
references to emperors. His centre of gravity lies in the emperors of the late
Antonine age. Even before Severus, therefore, the emperor’s rescripts had
become the dominant source of law in the provinces, especially criminal and
public law. Papinian reflects a further development. He had a wide private
practice, not confined to public law, yet he too, though sparing in his mention
of either, cites emperors more than private authors. In Paul and Ulpian, on
the other hand, the balance is different. They put in five or six references to
private authors for each citation of imperial law. The main reason is that in
their major commentaries they summarise the whole of the relevant law
rather than the points of growth. Marcianus, who follows Ulpian in so much,
tilts the balance the other way. He has more than two constitutions to each
citation of a private author. Modestinus has four times as many. This pre-
dominance of imperial sources heralds the decline of private legal writing, the
so-called ‘end of the classical age’.
Ulpian lives in the middle of this shift in the relative importance of imper-
ial and private sources of law. He chooses between juristic and imperial
sources according to the context. In private law he cites rather few constitu-
tions; in public law, to which he made an original and creative contribution,
many. In Justinian’s Digest he is recorded as citing the text, or part of the text,
of 69 constitutions.351 Of these twenty-nine come from areas of public law:
thirteen from The Proconsul,352 eight from The Consul,353 two from The
Community Treasurer,354 two from The Urban Prefect,355 two from
Tribunals,356 and two from Appeals.357 Seven of these come from Severus and
Caracalla, two from Severus alone, two from Marcus alone, five from
Marcus, and Verus, eight from Pius, and five from Hadrian. This gives an
idea of the slow but regular development of public law in the Antonine age.
Ulpian cites constitutions in any number only from Trajan onwards. The
figures are: Caesar 1, Augustus 5, Claudius 2, Nero 2, Titus 1, Domitian 1,
Nerva 1, Trajan 14, Hadrian 48, Pius 113, Marcus and Verus 43, Marcus alone
65, Marcus and Commodus 2, Severus alone 46, Severus and Caracalla 61,
Caracalla 46, imperator 1, princeps/principes 9. The list suggests that Ulpian’s
reading went back to Trajan and Hadrian. The first constitutions he cites ver-
batim come from a chapter in Trajan’s official instructions (mandata)358 and
from two rescripts of Hadrian.359 Ulpian is selective about emperors as he is
about private authors. He certainly knew some rescripts of Commodus and
Pertinax, but he does not cite any, despite the fact that Callistratus,360
Papinian,361 Marcianus,362 and Modestinus363 cite Commodus, while
Callistratus364 and Modestinus365 cite Pertinax. A certain fastidiousness or
caution is at work here. Paul shares it.
It can be assumed that when Ulpian cites the text of a constitution he has
consulted the text. This could be found in the Book of Petitions and Rescripts
(Liber libellorum rescriptorum, evidenced for Gordian III),366 which probably
refers to the posting-up of a batch of petitions and replies in public, later
stored in an archive. Another source was provided by private collections of
laws. One collection that gave at least some texts of rescripts of Marcus and
Verus or of Marcus alone was that of Papirius Iustus, whose twenty books of
imperial constitutions (De constitutionibus) were available to Justinian’s com-
pilers.367 Pernice thought that Ulpian did not use this collection, but the texts
he cites are on different points from those recorded by Papirius.368 It is true
352
D 1.16.6.3 (1 off. proc.); 26.5.12.2 (3 off. proc.); 50.4.6 pr; 50.7.7 (4 off. proc.); 48.8.4.2 (7
off. proc.); 1.6.2; 48.6.6; 48.18.1.1, 16, 22, 27; 47.14.1 pr (8 off. proc.); 48.20.6 (10 off. proc.)
353
D 35.1.50; 40.12.27 pr (1 off. cons.); 25.3.5.7, 14; 34.1.3 ( 2 off. cons.); 1.7.39; 27.3.17;
50.12.8 (3 off. cons.)
354 355
D 50.12.1.5; 50.10.5 pr (1 off. cur. reip.) D 1.12.1.4; 1.15.4 (1 off. urb. pr.)
356 357
D 50.13.1.10, 12 (8 omn. trib.) D 49.1.1.1 (1 appell.); 49.9.1 (4 appell.)
358 359
D 29.1.1 pr (45 ed.) D 37.10.3.5 (41 ed.); 29.5.1.28 (50 ed.)
360 361
D 12.3.10 (Call. 1 quaest.); 35.3.6 (4 cogn.) D 22.3.26 (Pap. 20 quaest.)
362
D 40.10.3 (Marci. 1 inst.); 49.14.31 (4 inst.)
363 364
D 25.3.6.1 (Mod. 1 manual.); 27.1.6.8 (2 excus.) D 50.6.6.2, 13 (Call 1 cogn.)
365
D 40.5.12.2 (Mod. 1 manual.)
366
Honoré (1994) 47f.; Nörr (1981) 27–9; d’Ors–Martin (1979); W. Williams (1986) 201–4;
Liebs (1997) §411.3
367
Lenel (1889) 1.947–952; Kunkel (1967) 216–7; Liebs (1997) §415.5
368
Pernice (1885) 456 citing D 4.4.9.5 (Ulp. 11 ed.) and 39.4.7.1 (Papir. 2 const.). The last text
concerns a ward (pupillus) not a minor and simply grants an indulgence in the particular case. It
does not lay down a general rule and Ulpian was not bound to cite it: D 1.4.1.2 (Ulp. 1 inst.)
6. Sources and Scholarship 155
that Ulpian does not expressly mention Papirius. He cites one rescript from
the half-yearly collection (semenstria) published regularly by Marcus and
Commodus,369 a source that Tryphoninus also uses.370 That Ulpian con-
sulted some collection or collections of imperial texts seems likely from
phrases with ‘there exists’ (extat), for example ‘there exists a rescript’ of the
deified Pius,371 Marcus,372 Marcus and Verus,373 the deified brothers,374 or
Severus375 or ‘there exists the senatusconsultum Vitrasianum’.376 The bulk of
the constitutions cited will probably have come from an official archive of
some sort. A secretary for petitions (a libellis) needed to find out what preced-
ents there were on the points raised by the petitioner and on which the
secretary must advise the emperor. Whether, as Pernice asserts,377 Ulpian’s
reading was unsystematic and confined to rescripts bearing on the problem
before him, we cannot know. A secretary for petitions who held office for six
years (203–9) would over time have to deal with the whole range of topics on
which rescripts might be sought.
One of Pernice’s charges is that Ulpian passed off as his own statements of
the law which from other sources are known to have come from an imperial
constitution.378 In the case he refers to Ulpian reproduces379 what we know
from a text of Paul380 to be the words of a law of Marcus and Verus. But for
all we know Ulpian did make this clear and Justinian’s compilers have, in the
interest of brevity, struck out the relevant words. Since he is in general care-
ful to cite original texts when he can, we should not lightly assume an inten-
tion to pass off imperial law as his own invention.
Ulpian aims at a balance between private authors and imperial sources. He
is ready to cite either or both when appropriate. How does he perceive their
relation to one another? Which prevails in case of conflict? At first sight he
seems to put them on a level: ‘Julian wrote and it is provided by rescript
. . .’;381 ‘Papinian replied and it is provided by rescript’;382 ‘hence it has been
stated and provided by rescript’;383 ‘this is provided by rescript and juristic
opinion’;384 ‘as is provided by imperial constitution and juristic reply’;385 ‘for
this is recorded (by a legal author) and by rescript’.386
369 D 29.2.12 (11 ed: est et in semenstribus Vibiis Soteri et Victorino rescriptum)
370 D 2.14.46 (Tryph. 2 disp., citing Marcus); 18.7.10 (note on Scaevola, citing Marcus)
371 D 27.8.6 (1 ed.) cf. 36.3.1.11 (79 ed.); 27.10.1.1 (1 Sab.); 49.1.1.1 (1 appell.)
372 D 27.8.1.2 (36 ed.) 373 D 40.5.30.13 (5 fid.) 374 D 48.18.1.27 (8 off. proc.)
375 376 377
D 26.10.1.7 (35 ed.) D 40.5.30.6 (5 fid.) Pernice (1885) 457
378 379 380
Pernice (1885) 455 D 27.8.1.10 (36 ed.) D 26.5.24 (Paul 9 resp.)
381
D 13.7.13 pr (28 ed: scripsit Iulianus et est rescriptum)
382
D 48.18.4 (3 disp: ut Papinianus respondit et est rescriptum)
383
D 2.15.7.2 (7 disp: ut inde sit et dictum et rescriptum)
384
D 4.4.7.9 (11 ed: et hoc et rescriptum et responsum est)
385
D 3.2.13.7 (6 ed: ut et constitutum est et responsum)
386
D 50.1.2.5 (1 disp: hoc enim et relatum est et rescriptum)
156 6. Sources and Scholarship
Juristic and imperial law do not come in any particular order, as is shown
in the table compiled by Coriat.387 Juristic law is not confined to replies
(responsa) but extends to what has been written, said, or reported in a treatise.
Ulpian does not specially esteem replies. He cites none of his own. Nor does
he cite those of Marcellus, one of his main sources,388 or of Scaevola.389 Only
Papinian’s are treated with particular respect, but the respect is for the per-
son, not the form of expression. Though juristic opinion is a source of law,
Ulpian does not treat it as binding on himself. He is free to dissent from even
a unanimous opinion of his colleagues. Juristic opinion can, as Gaius says,
bind judges390 but it does not bind lawyers.
Imperial constitutions are different. Ulpian praises391 and occasionally
criticizes392 them, but they are binding and settle the law. If a testator
bequeaths to a slave the slave’s private fund (peculium) by way of legacy the
legacy should include, Ulpian thinks, any debt that the slave’s owner owes
him.393 But, he admits, this is contrary to a rescript of Severus and Caracalla,
of which he gives the text. He goes on to ask: ‘What if it was the testator’s
intention (that the slave should be able to claim what the owner owes)? . . .
Again, if the owner records in his accounts what he owes the slave, does the
debt form part of the legacy?’ There is then a discussion of the opinions of
Pegasus, Nerva, and Atilicinus. The latter held that the debt was not part of
the legacy, which is correct, says Ulpian, because it is consistent with the
rescript (quod verum est, quia consonat rescripto). So the rescript, even if open
to criticism, is decisive, and the principle that underlies it can be applied in
similar cases.
The view underlying this text is perhaps that rescripts on points if law are
declaratory of the true state of the law. Often they confirm Ulpian’s personal
opinion,394 which they can do in advance. It is this way of looking at legal
problems that makes it possible for Ulpian to say of a rescript of Marcus that
it conforms to his own opinion.395 Thus, on a point in the law of return from
captivity (postliminium) ‘there is no doubt after the rescript of the emperor
Antoninus and his deified father’ that a son born in captivity has the rights of
a son on return.396 But at other times Ulpian implies that a rescript has
changed the law. Of a rescript of Marcus that allowed a slave owner to accuse
his slave of adultery, he writes ‘After this rescript the owner has a duty to
387
Coriat (1997) 526–7, citing Collatio 12.7.4–6 (18 ed.); D 27.3.1.15 (36 ed.); 42.8.10.1 (73
ed.); 28.6.2.4 (6 Sab.); 38.17.2.2 (13 Sab.); 40.5.24.9 (5 fid.); J Inst. 2.20.12, where rescript and
constitution agree and D 9.2.29.1 (18 ed.), where they do not
388 Above n.9–10, 18–9 389 Above n.8–10, 109–15 390 Gaius, Inst. 1.7
391 D 17.1.6.7 (29 ed.); 28.6.2.4 6 Sab.); 50.2.3.1 (3 off. proc.); 48.19.8.1 (9 off. proc.); 50.15.3.1
(2 cens.)
392 D 33.8.6.4 (25 Sab.); 42.1.5.9 (3 off. cons.), 393 D 33.8.6.4 (25 Sab.)
394 36.1.19.1 (15 Sab: ut est et rescriptum); 26.8.5.3 (et ita est rescriptum); 49.1.1.2 (1 appell:
accuse his own slave.’397 A ward who has meddled with the affairs of another
can ‘after the rescript of the deified Pius’ be sued to the extent of any profit he
has derived.’398 Whether it declares existing law or makes new law a single
rescript is conclusive. It does not follow that rescripts are immune from crit-
icism. Caracalla ruled that debts could be taken in execution, if necessary.
This applies, on the better view, says Ulpian, only to debts that the debtor
admits are due. Yet one might argue that the judge should decide whether
they are due, as happens with execution against tangible property. ‘But the
contrary has been laid down by rescript’.399
Just as juristic opinion binds judges but not lawyers, so rescripts bind
judges and lawyers but not emperors. An emperor may depart from his own
prior view of the law or that of his predecessors.400 If rescripts are at variance
a lawyer may choose between the differing views.401 A lawyer advising an
emperor as secretary for petitions or as imperial councillor may properly urge
him to take a different view of the law from that presented by a previous
rescript or line of rescripts. The rules about the sources of law establish a hier-
archy. The rulings of emperors if unanimous bind lawyers and judges, the rul-
ings of lawyers of authority if unanimous bind judges. But emperors do not
bind themselves, and lawyers of authority do not bind themselves or one
another. The emperor’s decision has the force of law,402 says Ulpian in a
famous and characteristically forceful phrase. But it ceases to be law if the
emperor or his successor can be induced to take a different view. Moreover
within imperial laws there are the more or less weighty.403 Decisions in judic-
ial proceedings (decreta) are not on a level with general constitutions and the
same is true to a lesser extent of rescripts. So it is possible to say of a point of
law that it has been judicially decided and (separately) laid down as law,404 or
that it has been decided by rescript and (separately) laid down as law.405
Judicial decisions and rescripts make law most clearly when they express a
general principle, and of this an imperial edict or other general constitution is
the best evidence.
397 D 48.2.5 (3 adult.) cf. 31.61.1 (18 Iul. Pap: Iulianus quidem ait . . . sed post rescriptum Severi)
398 D 3.5.3.4 (10 ed.) 399 D 42.1.15.9 (3 off. cons: sed contra rescriptum est)
400 D 37.14.17 pr (11 Iul. Pap.) 401 D 49.14.6 pr (63 ed.)
402 D 1.4.1 pr (1 inst: quod principi placuit legis habet vigorem) 403 Crifò (1983) 420
404 D 4.4.26.1 (5 fid: et est decretum ab imperatore Severo et constitutum); 49.14.25 (19 Sab:
T H I S and the next two chapters try to fix the dates of Ulpian’s various works,
their order and the rate at which he wrote them. Though a sympathetic critic
has described this aim as ‘not worth the effort’1 the enterprise is instructive
from more than one point of view. It affords a paradigm of the Roman genius
for discipline and method. It shows how it was possible to achieve so much in
so short a period. Its conclusions have been largely accepted in a standard
work of reference.2 Moreover, an improved understanding of the Roman cal-
endar helps to modify the account in the first edition.
I begin with Ulpian’s eighty-one books On the Edict (Ad edictum prae-
toris),3 which for short will be called Edict. Once we assign these books to the
right period and fix the blocks of text in which Ulpian composed them, the
other works can be fitted in. To summarize the argument, Edict books 9–56
were composed in Caracalla’s sole reign, but that reign falls from Ulpian’s
point of view into two parts, one up to about book 19, the other starting
about book 22. Some of the later books (61–73) were written under Macrinus.
As to the rate at which Edict was written, the 76 books from 6 to 81 were
composed in three blocks, two of 251⁄2 and one of 25 books. These blocks rep-
resent annual stints and are independent of the political conditions that pre-
vailed at the time of writing.
To discover the dates of composition of Edict, and the blocks in which it
was written, we have, apart from some special clues, two resources. One
comes from references in Edict to reigning or recently dead emperors. These,
closely studied, are seen to reflect political conditions at the time of writing.
They can be used to date, within limits, the different parts of the commentary.
A second resource comes from the study of Ulpian’s style. Over the period
needed for such a large-scale work as Edict this slowly evolved. Had Ulpian
composed it in a continuous tract, it would still be useful to note how, as time
went by, he discarded certain expressions and adopted others. This would
help to date other works by their relation to the different parts of Edict. But
we can go further. Instead of being gradual, changes in the style of Edict are
in places abrupt. They occur between one book and the next, perhaps even in
the middle of a book. Such abrupt changes point to discontinuous composi-
tion. Ulpian has, for a time, put Edict aside in order to write or do something
else. Given the strain, and at times tedium, involved in writing on such a scale
this is hardly surprising. The breaks that can be detected help to fix the blocks
in which Edict was composed. These, in turn, give a clue to Ulpian’s method
of composition, and suggest a method, set out in the next chapter, for dating
his works and fixing the order in which he wrote them.
The Index Florentinus4 ascribes to Ulpian 83 books On the Edict. That is
a mistake. The headings of the Digest texts go up only to 81.5 The two books
on the edict of the market-masters or curule aediles formed a separate com-
mentary.6 Texts from these books are headed Ad edictum aedilium curulium
libro primo or libro secundo. They are not treated as the 82nd and 83rd books
of Edict. It was a mistake on Lenel’s part to treat them as such.7 As will
emerge later,8 these two books were probably composed at a time when
Ulpian had not got beyond Edict book 31. They were not left till the end.
I. REFERENCES TO EMPERORS
As regards references to living and recently dead emperors the 81 books, assum-
ing that they were composed in the numbered order, divide into five groups:
1. Books 1–6
2. Books 9–19
3. Books 22–57
4. Books 61–73
5. Books 74–81
The gaps cover books in which references to emperors are missing or ambigu-
ous. In any case these groups do not represent the blocks of text in which
Ulpian composed the commentary. They reflect rather the political condi-
tions obtaining at the time of composition. In regard to each group I set out
and seek to interpret the references to emperors.
1. Books 1–6
4 5
Ch. 24.1 D 21.2.52; 39.1.21; 39.2.24, 26, 28, 30
6 7
Lenel (1889) 2.884–98; Liebs (1997) § 424.2 Lenel (1889) 2.884–98
8
Ch.9 n.26–34
160 7. Dates and Plan I
These first six books do not settle who the reigning emperor or emperors are.
We cannot conclude from the phrase imperator Severus that Severus was alive
when book 6 was composed. Although Mommsen, on the basis of his epi-
graphic studies, believed that imperator is properly used only of a living
emperor,9 Fitting pointed out that, on Mommsen’s own showing,10 lawyers’
writings often violated this pattern.11 A treatise by a lawyer is not like an
inscription on a column, in which it is important to adhere to the exact title
and to observe the convention that a deified emperor should be called divus.
When a lawyer cites the exact words of the law of a dead emperor, he often
introduces the emperor as imperator, as in this text of Marcianus written after
Caracalla’s death:12
Imperatores Severus et Antoninus in haec verba rescripserunt:
On the other hand Papinian’s Replies (Responsa), apart from one text,14 call
dead emperors divus.15 What factors influence the choice of divus or impera-
tor in referring to a dead emperor? If the lawyer has consulted the text of the
constitution he cites, he is more likely to favour imperator. If, again, he is writ-
ing in a less formal mode (Quaestiones, debates about points of law, being less
formal than Responsa, written opinions that could be cited in court), or if he
wishes to show special respect for the emperor, he may adopt the imperator
form. But, as Papinian’s examples show, practice was not uniform. To avoid
monotony a writer might vary the form of reference. Perhaps for this reason
lawyers sometimes give the dead emperor’s name without any title: Traianus,
Hadrianus, etc.16 Apart from this, Mommsen listed twenty-one violations of
his convention in the Digest by writers as various as Callistratus, Menander,
9 Mommsen (1904–13) 2.156–7 cf. d’Ors (1942–3) 10 Mommsen (1904–13) 2.157–61
11 Fitting (1908) 5–8. 12 D 28.1.5 (Ulp. 6 Sab.)
13 D 1.5.8 (3 quaest.) cf. 50.1.11 pr (2 quaest.); 3.1.8 (2 quaest.); 31.64 (15 quaest.); 31.67.10
(19 quaest.); 12.6.3; 22.1.3 pr; 36.1.57.1 (20 quaest.); 36.3.5.1, 3 (28 quaest.); 35.2.11.2 (29
quaest.); 1.7.32.1 (31 quaest.); 48.5.39.4–6, 8; (36 quaest.) cf. Frag. Vat. 224 (11 quaest.)
14 D 20.2.1 (10 resp.) 15 e.g. D 29.2.86 pr (6 resp: divus Pius)
16 There are a number of examples from early emperors: Augustus: D 1.17.1 (Ulp. 15 ed.);
D. 23.2.14.4 (Paul 35 ed.); 40.1.14.1 (Paul 16 Plaut.); 49.16.12.1 (Macer 1 re mil.); 50.15.1.1 (Ulp.
1 cens.): Claudius: Gai. Inst. 1.32 c; D 4.4.3.4 (Ulp. 11 ed.); 16.1.2 pr. (Ulp. 29 ed.); Trajan: Gai.
Inst. 1.34; D.26.7.12.1 (Paul 38 ed.); 29.1.1 pr. (Ulp. 45 ed.); 49.14.13.6 (Paul 7 leg. Iul. Pap.);
49.14.16 (Ulp. 18 leg. Iul. Pap.); Hadrian: Gai. Inst. 1.47; Ulp. Tit. 24.28; Collatio 1.6.1 (Ulp. 7
off. proc.); D 11.8.3 (Ulp. 8 off. proc.); 37.1.6.8 (Mod. 2 excus.); 34.1.14.1 (Ulp. 2 fid. twice);
36.1.31.5 (Marci. 8 inst.); 40.12.27. 1 (Ulp. 2 off. cons.); 42.4.7.16 (Ulp. 59 ed.); 47.14.1.3 (Ulp. 8
off. proc.); 49. 14.13. 5 (Paul 7 leg. Iul. Pap.); 49.16.5.6 (Men. 2 re mil.); 49.17.19.3 (Tryph. 18
disp.); 50.8.12.3 (Papir. 2 const.)
On the Edict 161
Ulpian, Paul, Macer, and Modestinus.17 In these texts the author wrote
imperator for divus when speaking of a dead and deified emperor. Fitting
added two more instances from Paul18 and Marcianus,19 and overlooked two
from Macer.20 Given this weight of exceptions, it was bold of Mommsen to
argue that his rule remained intact. Nevertheless he relied on it to support a
mistaken hypothesis about the composition of Ulpian’s Edict.21
The safe course is to assume that, while the use of divus is usually a sign that
the emperor is dead (not infallible since a text can be altered retrospectively),
its absence is no sure sign that the emperor is alive. So the reference to impera-
tor Severus in book 6 of Edict does not tell us whether Severus was then alive
or dead.
2. Books 9–19
17 D 49.16.6.7 (Men. 2 re.mil.); 4.4.45.1 (Call. 1 ed. mon.); 48.15.3 pr. (Marci. 1 iud. pub.);
25.3.6.1; 40.5.12 pr (Mod. 1 man.); 27.1.6.8, 17 (Mod. 2 excus.); 27.1.13.6, 7 (Mod. 4 excus.);
27.1.15 pr; 19.2.49 pr (Mod. 6 excus.); 40.4.56 (Paul 1 fid.); 4.6.8 (Paul 3 brev.); 34.9.5.9 (Paul 1
iur. fisc.); 5.3.43 (Paul 2 Plaut.); 49.14.49 (Paul 1 resp.); 42.8.10.1 (U1p. 73 ed.); 49.14.25 (Ulp. 19
Sab.); 1.5.1.8 (Ulp. 21 Sab.); Frag. Vat. 119 (Ulp. 2 off. proc.); D 23.1.16 (Ulp. 3 leg. Iul. Pap.);
Ulp. Tit. 26–7
18 D 40.9.15 (Paul 1 leg. Iul. Pap.) 19 D 49.14.30 (Marci. 3 inst.)
20 D 48.21.2 pr; 49.14.34 (Macer 2 iud. pub.) 21 Mommsen (1904–13) 2.159
162 7. Dates and Plan I
3. Books 22–57
From book 22 onwards, however, the texts belong to a new period in which
Caracalla has priority over his father. This period will be called Caracalla B:
22 ed. D 1.5.17 ex constitutione imperatoris Antonini
22 ed. D 12.2.13.6 imperator noster cum patre rescripsit
24 ed. D 11.6.7.3 nam et divus Severus decrevit
31
= CJ 7.49.1, 19 Dec. 212
32 33
The heading is 1 ed. de rebus creditis, which corresponds to 26 ed. 13 June 195
164 7. Dates and Plan I
reign. The emperor’s death caused Ulpian to put his pen aside. By the time he
resumed book 52 Caracalla was referred to as divus Antoninus. But the two
texts belong to a continuous passage, as printed in the Palingenesia, of fifty
lines.42 The shock of Caracalla’s death would hardly have made Ulpian stop
in the middle of the passage he had started dictating.43 The correct explana-
tion is surely that of Mommsen.44 In the second text divus is a retrospective
insertion. A copyist will be tempted to insert divus if by the time he is copying
it the emperor in question has been deified. He is especially likely to do so if
the original referred to the emperor just by his name. So the likelihood here is
that the original read constitutio autem Antonini. The copyist has inserted divi
before Antonini. It is no surprise to find bare Antoninus in the original at this
point, since Caracalla has just previously been called imperator Antoninus
Augustus,45 which, in a legal text, is a trifle formal. A formal reference can
properly be followed by an informal one.
This presupposes that imperator Antoninus Augustus is indeed Caracalla,
not Elagabal. The constitution mentioned in book 52 is referred to in a
rescript of Alexander of 225,46 and there attributed to divus Antoninus pater
meus. Alexander was referring to Caracalla, from whom he claimed descent,
not to the discredited Elagabal.47 So book 52 of Edict was composed under
Caracalla. In any case Caracalla’s deification came not at once, but after an
interval, at the end of Macrinus’ reign or the beginning of Elagabal’s.48
If, therefore, divus was genuine in the second text, we should still have to sup-
pose the sort of break that Fitting proposed—a year’s interval between the
beginning and the end of a text that reads as continuous.
The third text49 presents more difficulty. Fitting, followed by d’Ors, held
that imperator noster in this text is Alexander.50 Gualandi, however, points
out that their arguments, drawn from the use of divus in book 52, are unreli-
able. One cannot rule out the possibility that the emperor referred to was
Caracalla. In my view imperator noster here is indeed Caracalla. But the argu-
ment for preferring a later emperor can be strengthened. This later emperor
could not be Alexander, because during the short time that he lived under
Alexander, between AD 222 and 223/4, Ulpian, as a prefect, did not have time
to compose twenty-five books of edictal commentary. One could, however,
adapt Fitting’s idea by substituting Elagabal for Alexander. As explained
later in this chapter,51 there appears to be a break in Ulpian’s Edict in the mid-
dle of book 56. This break, if it existed, occurred shortly before the text in
book 57 that we are now considering. There could have been a substantial
42 Lenel (1889) 2.741–2 43 Crifò (1985) 610 suggests this
44 Mommsen (1904–13) 2.169 45 D 36.4.5.16 (52 ed.)
46 CJ 6.54.6 (8 Jan. 225)
47 Herodian 5.3.10; 5.7.3; CJ 12.35.4 (Antoninus pater meus); 6.50.5 (17 Nov. 223: constitutio
interval between the composition of book 52, when there is reason to suppose
that Caracalla was still alive, and book 57. So book 57 might have been com-
posed under Elagabal.
There are, however, two reasons for rejecting this suggestion. There is no
sign that the text refers to Elagabal. Indeed, there is no positive evidence that
Ulpian composed anything under Elagabal. Secondly, there is positive evid-
ence, to be detailed in a moment,52 that some books of Ulpian’s Edict later
than book 57 were composed in the reign of Macrinus. If so, book 57 cannot
have been composed under Elagabal, unless it was written out of order. But
why should it have been? I therefore adopt the suggestion of Gualandi that in
this book 57 text imperator noster is still Caracalla.
But if Caracalla was sole emperor during the period of composition of
books 22 to 57, his relation to Severus had changed from what it was up to
book 19. Ulpian now treats him as senior to Severus, in the sense that he is
consistently mentioned first. Various formulae are used:
imperator noster cum patre53
imperator noster et divus pater eius54
imperator noster cum divo Severo55
imperator noster et divus Severus56
imperator Antoninus cum patre57
imperator Antoninus et divus Severus58
imperator Antoninus cum divo Severo59
In contrast with the Caracalla A period, where we found four texts in all of
which Severus preceded his son, we now find seventeen texts, in sixteen of
which one of the formulae just listed are used.60 The one exception comes in
book 35.61 Here the text runs Severus et Antoninus rescripserunt Epicurio.
Whatever the explanation of this exception, in which Severus comes first, it
cannot obscure the contrast between Caracalla B and Caracalla A. Respect
for or fear of the tyrant now made it standard practice to mention him first.
Other writers, such as Paul62 and Tryphoninus,63 do the same, so that the
works in which they do this should be assigned to the same part of Caracalla’s
reign. It is not easy to say whether this change (which, I shall argue, occur-
red in 213) reflects some striking political event,64 or whether it defers to a
52 53
Below n.88–92 D 12.2.13.6 (22 ed.); 26.7.3.4 (35 ed.); 47.4.1.7 (38 ed.)
54 55
D 27.9.3 pr (35 ed.); 27.3.1.13; 12.3.4 pr, 1 (36 ed.) D 29.1.13.4 (45 ed.)
56 D 48.18.3 (50 ed.) 57 D 11.7.12 pr (25 ed.); 19.2.9.4 (32 ed.); 26.7.9.6 (36 ed.)
58 D 18.3.4 pr (32 ed.) 59 D 19.2.9.1; 19.2.19.9 (32 ed.); 26.10.1.4 (35 ed.)
60 Above n.53–9 61 D 26.10.3.13 (35 ed.)
62 D 27.1.46.2 (cogn.); 40.8.7 (1 lib. dand.); 27.9.13 pr (1 orat. d. Sev.); 47.15.6 (1 pub. iud.)
63 D 27.1.44 pr (2 disp.); 49.15.12.17 (4 disp.)
64 Quite possible. In 213 come the first inscriptions that call Caracalla magnus imperator (CIL
V 28, X 5286). The Fratres Arvales call him Germanice max(ime) as early as 19 May 213: CIL VI
2086(2) line 15
On the Edict 167
65 D 18.3.4;, 19.2.9.1; 19.2.19.9 (32 ed.); 27.9.3 pr (35 ed.); 26.10.1.4 (35 ed.); 27.3.1.13; 12.3.4
pr; 12.4.3.1 (36 ed.); 29.1.13.4 (45 ed.); 48.18.3 (50 ed.)
66 D 26.10.3.13 (35 ed.)
67 D 12.2.13.6 (22 ed.); 11.7.2 pr (25 ed.); 12.2.9.4 (32 ed.); 26.7.3.4 (35 ed.); 26.7.9.6 (36 ed.);
27.2.1.3; 49.14.27 (34 ed.); 26.7.7.4; 26.10.1.7; 26.10.3 pr (35 ed.); 27.3.1.3; 27.5 1.2 (36 ed.)
70 D 16.1.2.3 (29 ed., twice)
71 Mommsen (1904–13) 2.158–9; Karlowa (1885–1901) 1.743; Pernice (1885) 444; Jörs (1903)
1505
72 Above n.9–21 73 Fitting (1908) 8 74 Ch.8 n.33–7
75 D 4.4.18.2, 3 (11 ed.) 76 Mommsen (1904–13) 2.15923
168 7. Dates and Plan I
But this argument is also weak. The phrase non ita pridem need not rep-
resent a very short period. It could continue to be appropriate for at least a
year or so. The idea that Ulpian could not compose sixteen books in a year is
ludicrous, his normal rate of composition being, we shall see, over forty
books a year. It is like the theories once held about Justinian’s Digest, which,
some scholars thought, could not have been compiled in three years without
earlier drafts or partial codifications to work on.86 That has been shown to be
wrong,87 and it is wrong for Ulpian too. Ulpian’s rate of work is best under-
stood, I suggest, if we suppose that on average he composed about a book a
week, with vacations of sixty days in all, during the period of about five years
when he concentrated on writing. If so, he wrote or dictated each week about
10,000 to 12,000 words or the equivalent of 25 to 30 pages of modern print. A
methodical writer can keep this up week after week if two conditions are pre-
sent. One is that the source material has been collected beforehand and is easy
to refer to. We should picture Ulpian during the reign of Severus not as mak-
ing a first draft of Edict but as collecting and analysing the material he intends
to use later. Secondly, we must suppose that Ulpian dictates to scribes who
have become familiar with legal terms. There is no reason why Ulpian’s work-
ing method should not have been the same as that of Anthony Trollope and,
if it comes to that, myself. We set ourselves to write so many words a day, but
do not pedantically adhere to the routine. Instead, we check our progress to
make sure that over a period we have kept up the pace.88
There is no solid basis, then, for the theory that the whole of Edict, or the
first thirty-five books of it, were first drafted under Severus and later revised
under Caracalla. So far as books 22–57 are concerned, the references to
emperors show rather that they were drafted in that part of Caracalla’s reign
that we have termed Caracalla B. This began ‘not long after’ 19 December
212. What is more, it began after the Constitutio Antoniniana, which is men-
tioned in book 22.
4. Books 61–73
In the next period Caracalla is called plain Antoninus, without a title, and
Severus, in most texts, is also plain Severus.
61 ed. D 26.5.18 et ita Severus rescripsit
64 ed. D 42.6.1.3 et ita Severus et Antoninus rescripserunt
68 ed. D 43.4.3.1 constitutum est ab Antonino
71 ed. D 43.30.1.3 et divus Pius decrevit et a Marco et a Severo rescriptum
est
73 ed. D 42.8.10.1 et ab imperatore Severo et Antonino rescriptum est
86 Peters (1913) and others 87 Diósdi (1971); Honoré–Rodger (1970); Honoré (1978)
88 Mullen (1990) 320–1
170 7. Dates and Plan I
It is this dilemma of Macrinus, surely, that Ulpian in his own way reflects.
Unsure how to refer to Caracalla, he cannot call him divus, since he has not
been deified and perhaps never will be. He is dead, so he cannot be noster,
Augustus, or princeps. Imperator might do, but in the writings of a lawyer that
title, used of a dead emperor, is perhaps too respectful. So, though Macrinus
chose imperator in his letter to the senate as a non-committal expression,
Ulpian’s solution is to adopt the plain form Antoninus. As to Severus, though
he is not in disgrace, he, too, suffers to some extent from his association with
Caracalla, and, in three texts out of four, becomes plain Severus.90 In the
fourth, on the other hand, he is imperator Severus.91 The text runs ab impera-
tore Severo et Antonino. This may again reflect Caracalla’s dubious status.
Severus can safely be called an imperator, but it is still possible that his son’s
memory will be condemned.
That Caracalla is now dead is clear from the fact that Severus is now men-
tioned before him,92 in contrast with the practice in the period we have called
Caracalla B. Severus resumes his natural seniority. During this period, as
Fitting pointed out,93 fewer laws and lawyers are cited. In contrast with the
first fifty-two books, Ulpian no longer cites writers by work and book as he
does earlier. An example of the earlier practice is:
apud Marcellum libro quarto digestorum relatum est . . .94
89
Dio 79.17.2–3 cf. Honoré (1962a) 209
90 91
D 26.5.18 (61 ed.); 42.56.1.3 (64 ed.); 43.30.1.3 (71 ed.) D 42.8.10.1 (73 ed.)
92 93 94
D 42.6.1.3 (64 ed.) Fitting (1908) 107 D 5.3.13.10 (Ulp. 15 ed.)
On the Edict 171
Such references, with one uncertain exception,95 are missing in the later
books. Jörs thought that the whole commentary was first drafted during
Severus’ lifetime in this skeletal form.96 Then it was revised under Caracalla,
and the work and book citations added. The revision, however, petered out
towards the end. If the revision theory is rejected, two other explanations lie
to hand. Ulpian may have been pressed for space towards the end of his great
enterprise. In Paul’s Edict, which runs to 78 books, citations by work and
book cease at book 41.97 The compilers of Justinian’s Digest may also have
been pressed for space. The middle section of edictal commentary was appar-
ently read by the Sabinian committee, the last by the edictal committee.98
According to Bluhme and Krüger the division falls in the middle of Ulpian
book 52 and Paul book 48. Perhaps the edictal committee decided that the
time had come to reduce the volume of excerpts by omitting these work and
book citations. But two such citations occur in Ulpian, Edict book 5299 in a
passage that Bluhme and Krüger assign to the edictal committee. This is not
a strong objection, since we do not know exactly when one committee took
over from another, and the apparent change of style may be largely due to the
compilers of the Digest. Even if Ulpian did reduce citations in order to finish
in the space available—and there is evidence that the commentary was
planned to finish at book 81100—this would not explain why Caracalla is
called plain Antoninus in books 64–73. The explanation is political, not a mat-
ter of style.
5. Books 74–81
101 102
Lenel (1889) 2.619–29 Ch.2 n.114
103
Ch.2 n.106–7 (three examples, rather than four as stated in the first edition)
104 105 106 107
Ch.2 n.48 Ch.2 n.49 Ch.2 n.511 Ch.2 n.244
108
Ch.2 n.117
On the Edict 173
109 Below n.123–4; ch.8 n.16–7, 43–59 110 Ch. 2 n.315–6 111 Ch.2 n.296
174 7. Dates and Plan I
112 113
D 39.2.4.5 (1 ed.); 2.4.10.5; 2.8.2.3 (5 ed.) D 2.2.1.2 (3 ed.); 2.4.10.2,9 (5 ed.)
114 115
D 39.2.4.8 (1 ed.); 50.1.1.1 (2 ed.) D 2.2.1.2 (3 ed.)
116 117 118
D 2.14.7.5; 2.14.10.2 (4 ed.) D 2.8.2.3 (5 ed.) D 2.8.2.5 (5 ed.)
119 120
Ch.2 n.320–3,420–431 D 2.13.4.5 (4 ed.) is closest
On the Edict 175
Three sentence endings out of four in this sequence are in est, used both as a
connective and as a participle.121 This fondness for est endings is not repeated
in the later books of Edict. The statistics for sentences ending in est etc. in
Edict are as follows:
Books Sentences in est Lines Frequency/1000 lines
1–5 50 1,024 49
6–31/1 308 10,753 29
31/2–56/1 240 8,287 29
56/2–81 269 6,934 39
The tendency to end a sentence in this unemphatic way falls off after book 5
but returns to a lesser extent in the last block of edictal commentary. Another
habit that is found from book 6 onwards is that of the author’s referring back
to what he has written previously with diximus.122
6 ed. D 3.1.1.7 ut initio huius tituli diximus
6 ed. D 3.1.1.11 sub titulo de in ius vocado plane diximus
9 ed. D 3.3.17 quae supra diximus
11 ed. D 4.2.7.1 ea quae diximus
11 ed. D 4.2.14.4 secundum quod supra diximus
11 ed. D 4.2.14.10 quatenus autem diximus
11 ed. D 4.2.16 pr quod diximus
A writer cannot refer back to what he has written until he has written a fair
amount. But it is striking that the use of diximus appears first in book 6 and
then, for a time, becomes frequent.
The existence of a break between books 5 and 6 of Edict has been
doubted123 but is in my view clear. The last seventy-six books (6 to 81) were
121
As Birks points out (1983) 177–8 in reply to Watson (1983) the feature of style to which I
draw attention concerns the lack of emphasis at the end of the sentence, not the syntax
122
One reference back with rettulimus occurs earlier: D 2.14.7.18 (4 ed.). Rettuli/-mus returns
in some later texts D 24.1.7.6 (32/31 Sab.); 24.1.32 pr (32 Sab.); 41.1.23.3 (43 Sab.)
123
Watson (1983) but Birks (1983) 177–8
176 7. Dates and Plan I
divided by Ulpian into blocks of 25 to 26 books, either two blocks of 251⁄2 and
one of 25, or a block of 26 and two of 25. Between blocks he broke off to do
or write other things. Then he came back to Edict. The reason why this plan
was adopted may be left for the next chapter. That chapter will examine
whether, armed with what has been learned about the political conditions
when Edict was written and the way in which it was divided into blocks, we
can date this and Ulpian’s other works more precisely. For the moment it is
enough to summarize the results so far reached.
The first block of Edict (books 1–5) cannot be dated exactly, but must be
earlier than 213. It is possible that it goes back to the early weeks of 211,
before the death of Severus at York on 4 February. The second block (books
6–31) belongs, at least from book 9, to Caracalla’s sole reign. Book 10 was
composed after, book 22 not long after 19 December 212. Book 22 was writ-
ten after the Antonine constitution, which was issued not earlier than 212.
This block therefore seems to belong to 213, or perhaps a few months before
and after. It straddles Caracalla A and B. During the course of it Ulpian
moves from treating Severus as the senior to giving Caracalla priority over his
father. The third segment (books 31–56) also belongs to Caracalla’s sole
reign, and must come between 214 and early 217. The fourth segment (books
56–81) begins in the reign of Caracalla (book 57) but books 61–73 were com-
posed under Macrinus. This block therefore belongs to 217 and perhaps also
218.
Useful in themselves, though rough, these results provide the springboard
for a more ambitious hypothesis about Ulpian’s survey of Roman law.
8
T O find how Ulpian set about his major survey of Roman law the next step
is to see how his 51-book treatise On Sabinus1 relates to the 81 books On the
Edict. Sabinus, as it may be called for short, deals with the older civil law
whereas Edict deals with praetorian law, but there is some overlap. Some top-
ics could as well be treated in either or both. If we begin as before with the
analysis of references to living or recently dead emperors, it is clear that
Sabinus was largely written in the period that I have called Caracalla B. In this
period, which ran from some time in 213 to the end of Caracalla’s reign in
April 217, Caracalla takes precedence over his father Severus when joint laws
are cited. Here are the relevant references:2
6 Sab. D 28.6.2.4 rescripto imperatoris nostri
6 Sab. D 29.2.6.3 ut divus Pius et imperator noster rescripserunt
6 Sab. D 24.1.23 Papinianus recte putabat orationem divi Severi . . .
12 Sab. D 38.16.1.1 ut est a divis Marco et Vero et imperatore nostro
Antonino Augusto rescriptum
12 Sab. D 38.17.1.2 secundum rescriptum imperatoris nostri et divi patris
eius ad Ovinium Tertullum3 (= CJ 8.50.1, Severus et
Antoninus, undated)
13 Sab. D 38.17.2.2 idque et Iulianus scripsit et constititum est ab impera-
tore nostro
13 Sab. D 38.17.2.47 et invenimus rescriptum ab imperatore nostro
Antonino Augusto et divo patre eius4
19 Sab. D 49.14.25 et est decretum ab imperatore Severo et constitutum
21 Sab. D 30.37 pr quae sententia rescripto imperatoris nostri et divi Severi
iuvatur
21 Sab. D 30.41.3 et ita imperator noster et divus Severus rescripserunt
Schulz (1906); Koschaker (1907); Wolff (1951) 145–71; Wieacker (1953) 241–61; (1960) 283–326,
446f.; Schulz (1961) 264–9; Astolfi (1983) 57–192; Luchetti (1987) 49–87; Liebs (1997) §424.3
2 Fitting (1908) 111–2, omitting three of these texts; P. Krüger (1912) 243
3 CJ 8.50.1 (Impp. Severus et Antoninus, undated)
4 Mammiae Maximinae 12 Apr. 203
178 8. Dates and Plan II
5 6
D 26.8.5.3 D 49.14.25 (19 Sab.)
7
D 24.1.23 (6 Sab.); 38.17.1.3 (12 Sab.); 38.17.2.47 (13 Sab.); 30.37 pr; 30.41.3; 30.41.5 (21
Sab.); 24.1.3.1 (32 Sab.); 24.1.32 pr (33 Sab.); 26.8.5.3 (40 Sab.); 46.3.5.2 (43 Sab., twice)
Quinquennium Ulpiani 179
If there has been an engagement when the woman was under age it continues
to be valid even if she has lived in the man’s house as if married while still
under age. So it is valid when she comes of age. It looks as if in Sabinus book
33 Ulpian could refer to Edict book 35, which comes in the middle block of
edictal text (books 31–56). This part of Sabinus could therefore have been
composed after the break in Edict that occurs at book 56. Sabinus book 33
would be later not merely than Edict book 35 but than Edict book 56.
What about the length of Sabinus? It stops, incomplete, at 51 books, break-
ing off at the point that, according to Lenel, corresponds to Paul Edict book
13 and Pomponius Edict book 29. Paul handled the missing part of the com-
mentary in his Sabinus books 13–16,13 and Pomponius in his books 29–36.14
8 D 33.8.6.4; 33.8.8.7 (25 Sab.); 23.3.9.3 (31 Sab.); 24.1.7 pr, 5, 6 (32 Sab.); 24.1.32.19 (33
(37 Sab.)
11 D 24.1.32.27 (33 Sab.); P. Krüger (1912) 245
12
D 23.1.9 (35 ed.). Frier (1984) 85811 objects that ‘always’ excludes a reference back to his
own published opinion. But with Ulpian ‘always’ only implies ‘a couple of times previously’:
Kalb (1890) 132
13 14
Lenel (1889) 2.1257–60; Krüger (1912) 244 Lenel (1889) 1.1286–93
180 8. Dates and Plan II
The hypothesis put forward in the first edition and to which I adhere is that
these blocks of text represent four annual stints of 251⁄2 books and one of 25
books of major commentary between the years 213 and 217. Seventy-six books
on the Edict and 51 on Sabinus make 127 in all over the five years 213–7, which
divides into four lots of 251⁄2 books and one of 25. We know that, if the idea of
blocks of text is accepted, most of the first block, the whole of blocks 2 to 4,
and the beginning of block 5 fall in Caracalla’s sole reign. Some of the later
books of block 5 fall under Macrinus.20 If these blocks represent annual stints,
the fifth must belong to 217, because in that year Caracalla was emperor at the
beginning of the year and Macrinus at the end. Working backwards, the first
block should belong to 213. This seems historically plausible. Caracalla’s sta-
tus is enhanced during the course of 213, which is also the year of the first
block of text. Edict book 10 is later than the rescript of 19 December 212.21
Book 2222 is later than the Antonine constitution, which cannot be earlier than
212. So the only set of dates that fits is 213 to 217.
20 Ch.7 nn.88–92 21 D 3.6.1.3 (10 ed.) cf. 12.5.2.2 (26 ed.) 22 D 1.5.17 (22 ed.)
23 Lenel (1889) 2.966–91; AE 1966.436 (Ephesus c.300); Lactantius, Inst. 5.11.18f.; Ind. Flor.
24.6; Jörs (1903) 1452; Schulz (1946) 139, 243–6, (1961) 310–3; Wieacker (1960) 391–408, 447f.;
Dell’Oro (1960) 117–211; Schönbauer (1965) 108–15; Marotta (1988) 251–351, 381–92;
Mantovani (1992); Liebs (1997) §424.14
182 8. Dates and Plan II
2 off. proc. Frag. Vat. 119 Imperatores Augusti Iulio Iuliano rescripserunt
...
3 off. proc. D 50.2.3.1 Imperator Antoninus edicto proposito statuit24
3 off. proc. D 50.2.3.3 Divi Severus et Antoninus permiserunt
4 off. proc. D 50.4.6.2 ita imperator noster cum divo patre suo rescrip-
sit
4 off. proc. D 27.1.6.6 imperator noster cum patre rescripsit
4 off. proc. D 50.7.7 imperator noster cum patre rescripsit
5 off. proc. D 50.12.6.1 imperator noster Antoninus rescripsit
5 off. proc. D 50.12.6.2 ita rescripto imperatoris nostri et divi patris eius
continetur
5 off. proc. D 50.12.6.3 rescriptis imperatoris nostri et divi patris eius
continetur
8 off. proc. D 48.18.1.10 imperator noster cum divo patre suo rescripsit
8 off. proc. D 48.18.1.15 imperator noster cum divo patre suo rescripsit
8 off. proc. D 48.18.1.16 Severus Spicio Antigono ita rescripsit
8 off. proc. D 48.18.1.17 Divus Severus rescripsit
8 off. proc. D 48.8.1.18 imperator noster cum divo patre suo id non
admiserunt
8 off. proc. D 47.9.12 pr imperator Antoninus cum divo patre suo
rescripsit
9 off. proc. Collatio 14.3.3 imperator Antoninus constituit/constitutione
imperatoris Antonini
9 off. proc. D 48.19.8.5 epistula divi Severi exprimitur
9 off. proc. D 48.19.8.12 imperator Antoninus rectissime rescripsit
9 off. proc. D 48.22.6.1 epistula divi Severi expressum est
10 off. proc. D 48.22.7.10 imperator noster cum divo patre suo huic rei
providerunt
Caracalla is mentioned with his father in the first and last books of this treat-
ise. There are thirteen such allusions in all. In the first three books two texts
out of three place Severus before Antoninus, but in the remaining books
Caracalla always precedes his father. Is the inference that the first three books
belong to the Caracalla A period, and the last seven to Caracalla B? If so, The
Proconsul must belong to the same year as Edict books 6–31, 213. For in that
block of Edict Caracalla A runs up to book 19, and Caracalla B starts from
book 22.
But do the citations in the first three books point to Caracalla A? Of the two
texts in which Severus precedes Antoninus, the first presents no problem. It
comes from book 1 and runs divus Severus et imperator Antoninus . . . sunt
moderati.25 Severus is dead, Caracalla is sole ruler, but Severus’ seniority has
not yet been displaced. The other text in which Severus is senior, from book
3 does present a problem.26 It runs divi Severus et Antoninus . . . permiserunt.
This implies that Caracalla and Severus are both dead. But in the later books
Caracalla is consistently imperator noster 27 and so living. Mommsen is surely
right to think that divi is here a gloss.28 Originally the text ran Severus et
Antoninus permiserunt or divus Severus et Antoninus permiserunt. Thinking
this too bald, the copyist, at a time when Caracalla was in fact dead and dei-
fied, inserted divi or turned divus into divi. If Severus was not originally
referred to as divus the original might seem a little bleak. But though Ulpian
does not often refer to emperors without some title (imperator, divus, noster)
he occasionally does so.29 At any rate, the order in which the emperors appear
is not in doubt. When book 3 was composed, Ulpian gave Severus preced-
ence. These two texts show Severus as senior, and are evidence that books 1
and 3 fall in Caracalla A. But what of the other text from book 1 (D 1.16.4
pr)? This runs:
Observare autem proconsulem oportet, ne in hospitiis oneret provinciam, ut impera-
tor noster cum patre Aufidio Severo rescripsit.
I am not sure that this can be satisfactorily explained. But when Ulpian ends
a sentence by citing a rescript in the form ut . . . rescripsit, ut rescripsit can per-
haps sometimes be an afterthought, inserted when the text is read over or
revised. A parallel comes in Sabinus book 6, where a sentence ends with the
phrase ut Marciano videtur.30 Sabinus was written under Caracalla 31 but
Marcianus is not known to have published any work until after Caracalla’s
death and deification.32 We know that there was a second edition of
Sabinus,33 and this phrase could have been put in at the time of revision. On
the whole Ulpian prefers the stronger ending et ita rescripsit,34 idque rescrip-
sit,35 or gives the substance of the rescript in the accusative and infinitive after
rescripsit.36 It is possible, then, that the phrase we are considering from The
Proconsul book 1 is, like the Sabinus text, a later insertion. This would be
to postulate a second edition of The Proconsul, for which there is no other
proc.); 8.3 (8 off. proc.); D 47.14.1.3 (8 off. proc.); 49.14.16 (18 leg. Iul. Pap.); 34.1.14.1 (2 fid.);
40.12.27.1 (2 off. cons.); 50.15.1.1 (1 cens.); Frag. Vat. 235 (1 off. praet. tut.). None of these refers
to Severus or Caracalla.
30 D 28.1.5 (6 Sab.) 31 Above nn.2–12
32 Lenel (1889) 1.6391; Kunkel (1967) 258 33 C Cordi (16 Nov. 534) 3
34 e.g. D 48.18.1.15 (8 off. proc.) 35 e.g. D 47.9.12 pr (8 off. proc.)
36 e.g. D 1.16.4.5 (1 off. proc.)
184 8. Dates and Plan II
The bulk of this work from books 3 to 8 belongs to the Caracalla B period,
though we have no evidence about the first two and last two books. If it was
written in a single year, it should go in 214, 215, or 216, but the choice between
these years must for the moment be left open.
Next comes Tribunals, also in ten books.39 Four texts cite Caracalla or his
father or both:
1 omn. trib. D 26.10.7.2 ex epistula imperatoris nostri et divi Severi
8 omn. trib. D 50.13.1.10 ita rescripto imperatoris nostri et patris eius con-
tinetur
8 omn. trib. D 50.13.1.12 ita est rescriptum ab imperatore nostro et divo
patre eius
8 omn. trib. D 50.13.1.13 divus Severus prohibuit
39 Lenel (1889) 2.992–1001; Jörs (1903) 5.1454; Pernice (1893); Fitting (1908) 119; P. Krüger
(1912) 246185; Wlassak (1919) 62–79; Schulz (1946) 256, (1961) 329; Wieacker (1960) 65; Dell’Oro
(1960) 271f.; Liebs (1966) 26344, showing that de omnibus tribunalibus complements the treatises
on the duties of various officials; Liebs (1997) §424.21
40 Lenel (1889) 2.939–50; FIRA 2.315f.; Ferrini (1901/1929–30); Schulz (1961) 231; Wieacker
(1960) 146; Levy (1963); Astolfi (1986) 366–71; Liebs (1997) §424.7
186 8. Dates and Plan II
How do the shorter works that form part of Ulpian’s corpus fit the chron-
ological scheme we have set out? Many of them should do so, because, as the
next chapter shows, they seem from references to ruling or recently dead
emperors to belong, like Taxation, to the Caracalla A period or, like Adultery,
to the Caracalla B period. But did Ulpian have time to compose more than
the 351⁄2 books a year we have so far attributed to him?41
The number of lesser works attributed to Ulpian that are clearly genuine,
or at least not shown to be spurious, comes to 39 or 40. These comprise,
with the number of books recorded for each work: Taxation 6; Trusts 6;
Adultery 5; Appeals 4; The Aelio-Sentian Law 4; The Consul 3; Edict of the
Market-Masters 2; Teaching Manual 2; The Quaestor 1; Consular Judges 1;
The Community Treasurer 1; The Urban Prefect 1; The Prefect of Police 1; The
Praetor for Guardianship 1; Engagements to Marry 1; and possibly The Speech
of Marcus and Commodus 1. We cannot count Excuses 1, which was written
under Severus, or at any rate before Caracalla’s sole rule, and so before the
end of 211.42 The first five books of edictal commentary must also be
excluded. They too were composed earlier, perhaps early in 211.
If 40 books are distributed over 5 years, they come to eight a year. On top
of the 351⁄2 books of major commentary and medium-scale treatise this would
take Ulpian’s annual stint to 431⁄2 books a year.43 The figure seems arbitrary,
but it is not. The Roman calendar provided for two judicial holiday periods
of 30 days each, intended for the grain and wine harvests. The Lex Irnitana
chapter K requires the officials (duumviri) of the municipality to announce in
good time the period, not to exceed twice thirty days, when judicial matters
are to be postponed on account of the grain and wine harvest (messis vin-
demiae causa).44 This municipal provision was based on the practice at
Rome,45 but the local magistrates could adapt the periods to suit local condi-
tions.46 Essentially the same arrangement existed in the fourth century, when
a law of 389, seeking to reduce the number of judicial holidays, makes an
exception for the twin months (gemini menses) that a more indulgent age
adopted to mitigate the heat of summer and the allow time for the harvests.47
41 What follows draws on Honoré (1988) 1433–40
42
Lenel (1889) 2.8991; Mommsen (1904–13) 2.16955; Schulz (1946) 249–50; (1961) 381–20
argues that it is a post-classical compilation. At any rate, it does not count for our purposes.
43
To be precise 431⁄2 books in four years and 43 in the fifth
44
González (1986) 161, 187: De rebus proferendis
45
Lex Irnitana ch. K: perque eos dies . . . ius ne dicunto nisi si (de) is rebus de quibus Romae
mes(sis) vindemiae causa rebus prolatis ius dici solet
46
RE Vindemia 17 (1961) 17–24 (Schuster)
47
CTh 2.8.19 (7 Aug. 389: Illos tantum manere feriarum dies fas erit, quos geminis mensibus ad
requiem laboris indulgentior annus accepit, aestivis fervoribus mitigandis et autumnis fetibus decer-
pendis). The drafter is probably Nicomachus Flavianus, who was an antiquarian
Quinquennium Ulpiani 187
This treats the two thirty-day periods as in some sense providing a genuine
rest from labour and not merely an opportunity to attend to matters con-
cerned with the harvest, such as collecting rents and fixing them for the next
year.48 In Tribunals Ulpian mentions the judicial holiday period for grain and
wine, but without stating the length of the period.49 The senatorial holiday
laid down by Augustus was different: a continuous period of 61 days in
September and October when only those drawn by lot were obliged to attend
meetings.50
Assume that Ulpian stopped writing during the two thirty-day holiday
periods and instead attended to his estates or rested. Subtracting 60 from 365
leaves 305 days, which is 43 weeks and 4 days. That fits a programme of 431⁄2
working weeks a year, with one book to be written each week. The numerical
coincidence is striking. Is it just a coincidence? If it is something more it
implies that Justinian’s commissioners possessed nearly all the works written
by Ulpian in the five-year period 213–7.51 In any event, in order to get his
magisterial survey of Roman law in 217 books completed, largely under
Caracalla, Ulpian must have worked to a schedule close to what is here sug-
gested. He must have disciplined himself more or less as Anthony Trollope
did, though, like the novelist, he might at times fall behind schedule and at
others catch up.52
Is it plausible to suppose that he took the week as his working unit? The tra-
ditional Roman week was the inconvenient eight-day week, nundinae.53 Not
until the fourth century do we find mention in legislation of Sunday (dies
solis) 54 and not until the fifth century the seven-day week (septimana).55 But
the seven-day week is virtually a universal phenomenon. It prevailed among
the Babylonians, Jews, Egyptians, Persians, Indians, and Greeks.56 From the
age of Augustus the Graeco-Roman world increasingly, though unofficially,
resorted to the seven-day week.57 Dio Cassius, contemporary of Ulpian,
attributes the planetary seven-day week to the Egyptians but stresses that ‘it
48 D 7.8.10.4, 12 pr (Ulp. 17 Sab., Gai. 2 rer. cott.). In the first edition I took the holiday to be
a continuous eight-week period. This was four days too short. Moreover the period was not con-
tinuous but should be divided into two separate stretches of thirty days
49 D 2.12.1 pr (Ulp. 4 omn. trib: ne quis messium vindemiarumque tempore adversarium cogat
ad iudicium venire, oratione divi Marci exprimitur . . . 2. sed excipiuntur certae causae, ex quibus
cogi poterimus et per id temporis, cum messes vindemiaeque sunt, ad praetorem venire.
50
Suetonius, Augustus 35.3
51 With the exception of one book: Ad orationem divi Antonini et Commodi: ch. 9 nn.137–140
52 What Trollope says in An Autobiography (1950 ed.) 118–22 is not entirely accurate but, if
he fell behind from time to time, his method served to warn him that he had done so and provide
an incentive to catch up
53 Boll (1912) 7.2549; Kroll (1937)
54 CJ 3.12.2 (3 Mar. 321); CTh 8.8.3 = 2.8.18, 11.7.13 (3 Nov. 386); 15.5.2 (30 May 394, MS
20 May 386)
55 CTh 15.5.5 (1 Feb. 425: Dominico, qui septimanae totius primus est dies)
56 Boll (1912) 2550, 2556–8, 2571
57 Boll (1912) 7.2547, 2558; Sontheimer (1967) 18.2.2463
188 8. Dates and Plan II
is now found among all mankind’, including the Romans, among whom it is
already treated as a sort of ancestral tradition.58 There is therefore no reason
why Ulpian should not have adopted the seven-day week as the period within
which he would, when composing his survey, plan to write one book of, say,
on average 10,000 words. In the next chapter I consider how far the 40 books
of lesser works can be fitted into the five-year plan59 that I attribute to him.
In the first edition I supposed that one could assign the major, medium-
scale, and lesser works to different parts of the calendar year. But it is clear
from what has been said about the parallel composition of Edict and The
Proconsul that this is not so. Earlier-numbered books were written before
later-numbered books, and this may give us a rough clue to the time of year
at which a particular book was composed. But in general Ulpian arranged his
work schedule in whatever way suited him best. Did his year begin on 1
January or did he stop writing on 17 December for the Saturnalia and begin
a new year’s stint straight afterwards?60 It is impossible to tell and makes lit-
tle difference.
I . T H E E A R L Y P E R I O D : A D 2 13 A N D 2 1 4
For the early period the background assumption is that On the Edict books 6
to the middle of book 31 together with Proconsul books 1 to 10 were com-
posed in 213. On Sabinus books 1 to the beginning or middle of 26 and ten
books of Disputations or Tribunals were written in 214.
It may be as well to begin by strengthening the assumption that the first big
block of commentary On the Edict and Proconsul belong to the same year,
213. A number of expressions occur only between Edict books 6 and 31 and
in Proconsul. Thus Ulpian has four texts with destricte (‘indiscriminately’).2
Discutere, in the sense ‘investigate or decide as a judge’, is found in three
texts.3 Commendatio, ‘recommendation’, comes twice.4 The same is true of
improbitas, ‘want of scruple’.5 Periculosus, ‘dangerous’ in a legal rather than
a physical sense, is found in a quartet of texts.6 All of these come, apart from
a single text from Edict book 1, from Edict books 6 to 31 and Proconsul.
These parallels, though not individually of much weight, together tend to
confirm that, if Edict books 6 to 31 were composed in 213, so was Proconsul.
What other works belong to this year? One sign will be that the work
was wholly or partly composed in the Caracalla A period, before about the
middle of 213. If, in a joint citation, Severus precedes Antoninus, this is some
1
I draw on Honoré (1988) but note that at p.1440 of that study 217 should be 216 and 218
should be 217.
2
D 3.3.13 (8 ed.); 4.4.7.8 (11 ed.); 4.8.15 (13 ed.); 48.18.1.26 (8 off. proc.)
3
D 4.8.13.2; 4.8.25.1 (13 ed.); 48.2.6 (2 off. proc.)
4
D 4.1.1 (1 ed.); 1.16.4.3 (1 off. proc.)
5
D 4.9.3.1 (14 ed.); 1.16.9.4 (2 off. proc.)
6
D 39.2.1 (1 ed.); 3.1.1.3 (6 ed.); 5.1.61 pr (26 ed.); 48.18.1.23 (8 off. proc.)
190 9. Dates and Plan III
evidence that the work from which it comes belongs to that period.7 If, later
in the same work, Caracalla precedes Severus, the evidence is strengthened.
On the other hand, a text in which Severus precedes Caracalla, and Caracalla
is given no title, may be from 217.8 Again, such a text might be one of the half
dozen exceptions to the general practice in the Caracalla B period.9
Additional clues are needed. These will often consist in parallels, such as those
set out above,10 between two or more works to which certain expressions are
confined or virtually confined.
1. Taxation (6 books)
The first lesser work to be examined is Taxation, from which Lenel prints
ninety-six lines.11 This contains a number of references to contemporary or
recently dead emperors:
1 cens. D 50.15.1 pr huic enim divus Severus et imperator noster ius Italicum
dedit
1 cens. D 50.15.1.2 quae a divo Severo Italicae coloniae rem publicam accepit
1 cens. D 50.15.1.3 cui divus Severus ius Italicum concessit
1 cens. D 50.15.1.4 imperator noster ius coloniae dedit
1 cens. D 50.15.1.7 divus quoque Severus coloniam deduxit
1 cens. D 50.15.1.9 a divo Severo ius coloniae impetravit
2 cens. D 1.9.12 pr ut scio Antoninum Augustum Iuliae Mammaeae con-
sobrinae suae indulsisse
2 cens. D 50.15.3 rescripto imperatoris nostri ad Peligianum recte expres-
sum est
The first text shows that Severus is dead and Caracalla sole ruler. The order
of mention of the emperors points to the Caracalla A period, and so to the
year 213. The text about the indulgence shown to Julia Mammaea is of inter-
est from the point of view of Ulpian’s political career, which Julia Mammaea
later on supported.
In a few texts Ulpian condescendingly remarks of a rescript, as he does in
the above text from Taxation book 2, that it is ‘correct’ or ‘quite right’: imper-
ator recte/rectissime rescripsit:
12 13
D 1.9.1 pr (2 cens.) Lenel (1889) 2.8991; Jörs (1903) 145 14
Below n.51–4
15
D 2.14.7.5 (4 ed.); 4.2.9.3; 4.3.2 (11 ed.); 14.1.1.12 (28 ed.)
16
D 43.13.1.7; 43.8.2.33 (68 ed.)
17
D 48.19.3 (14 Sab.); 34.2.19.8 (20 Sab.); 30.39.6 (21 Sab.)
18 19
D 48.13.7 (7 off. proc.); 48.22.7.9 (10 off. proc.) D 1.9.1 pr (1 cens.)
20 21
D 28.5.35.1 (4 disp.) D 27.1.15.16; Frag. Vat. 242 (1 off. pr. tut.)
22 23
D 49.2.1.4; 49.1.3 pr (1 appell.) Frag. Vat. 321 (8 ed.)
24 25
D 3.2.2.2 (6 ed.) D 50.15.1.2, 3,5, 10, 11 (1 cens.)
192 9. Dates and Plan III
26 Lenel (1889) 2.884–98; Ind. Flor. xxiv 1; Schulz (1946) 198; (1961) 244; Jörs (1903) 1439 (all
mistakenly treating this work as an appendix to Ad edictum praetoris); Liebs (1997) §424.2
27 D 21.1.19.4 (1 ed. cur.); 15.1.3.7 (29 ed.); 28.1.20.3 (1 Sab.); 29.2.30.1 (8 Sab.)
28 D 5.3.25.8 (15 ed.) 29 D 21.1.4.6 (1 ed. cur.); 7.8.4 pr (17 Sab.); 1.16.9.4 (1 off. proc.)
30
D 21.1.31.18 (1 ed. cur.); 3.2.6.6 (6 ed.); 5.3.31.4 (15 ed.); 15.3.7.3 (29 ed.)
31
D 21.1.4.5 (1 ed. cur.); 3.6.3.3 (10 ed.); 4.6.21.1 (12 ed.)
32
D 21.1.31.23 (1 ed. cur.); 3.3.27 pr (9 ed.); 4.4.13 pr, 16 pr (11 ed.)
33
Ch. 8 n.38–9; Schulz (1961) 305–7; Wieacker (1960) 385–8; Liebs (1997) §424.24
34
Lenel (1889) 2.903–26; Ind. Flor. xxiv 9; Schulz (1946) 255; (1961) 328; Jörs (1887) 5.1451;
Fitting (1908) 117; Liebs (1997) §424.19
35 36
D 28.5.9.5 (5 Sab.); 36.2.5.1 (20 Sab.) D 19.1.11.3 (32 ed.); 24.3.22.6 (33 ed.)
37
D 35.2.82 (8 disp.)
Lesser Works 193
the middle block of Edict begins at book 31/2 or 32. From the parallels in
these texts, the Disputations passage would fit the later months of 214.
A similar picture emerges from texts with dicet quis ‘someone will argue’.
These come in Sabinus books 13 and 25,38 Edict book 5,39 and Disputations
book 4.40 There are two texts with intellegendum erit ‘it will have to be under-
stood’, one from Sabinus book 21 and one from Disputations book 4.41 Then
we come to seven texts with pari ratione ‘for the same reason’. These run from
Edict book 29,42 presumably latish 213, to Sabinus books 17 and 26,43 in 214,
two texts from Disputations,44 and one from Trusts.45 Here the Edict text
should belong to the later part of 213 and the Sabinus text to the later part of
214, so that Disputations again fits 214 best. Is the same true of Trusts?
One parallel between Disputations and Trusts consists in the fact that
Ulpian uses dicebam only in these works. The use of the imperfect points to
oral argument.46 Of twelve texts with dicebam eleven come from Disputations
and one from Trusts.47 The expression is to be expected in Disputations, which
record oral argument, but not in a work on Roman trust law. The most plau-
sible explanation is that Ulpian was working more or less concurrently on
these two works and recorded an oral discussion, present to his mind at the
time of writing, in the treatise on trusts. The same correlation is found with
three uses of referebam ‘I argued that. . .’. Two come from Disputations 48 and
one from Trusts.49 Another expression that points to a link between
Disputations and Trusts is benignum est, ‘it is fair . . .’, which occurs once in
each work.50
So far, then, the evidence rather favours the idea that Trusts and
Disputations were both composed in 214. What of the evidence to be drawn
from references to emperors?
1 fid. D 32.1.4 epistula divi Severi et imperatoris nostri
1 fid. D 32.1.9 et ita imperator noster rescripsit
2 fid. D 32.11.19 imperator noster rescripsit
2 fid. D 35.1.14 imperator noster rescripsit
3 fid. D 36.1.1.13 et est decretum a divo Severo
3 fid. D 36.1.3.4 imperator noster rescripsit
4 fid. D36.1.15.4 et ita invenio ab imperatore nostro et divo patre eius
rescriptum
38 D 38.17.2.41 (13 Sab.); 32.55.7 = 50.16.167 (25 Sab.) 39 D 27.9.7 pr (35 ed.)
40 41
D 41.1.33.1 (4 disp.) D 30.39.1 (21 Sab.); 28.5.35.3 (4 disp.)
42 43
D 15.1.11.9 (29 ed.) D 7.1.12.3 (17 Sab.); 23.2.12.4 (26/1 Sab.)
44 45
D 17.1.29 pr (7 disp.); 28.7.10 pr (8 disp.) D 36.1.17.6 (4 fid.)
46
Ch. 1 n.144–5
47
D 27.8.2; 44.3.5.1 (3 disp.); 28.5.35 pr; 28.4.2; 29.1.19; 49.17.9 (4 disp); 36.1.23 pr; 33.4.2.1
(5 disp.); 46.7.13 (7 disp.); 35.2.82 (8 disp.); 34.1.14.3 (2 fid.)
48 49
D 29.1.19 pr; 49.17.9 (4 disp.) D 35.1.92 (5 fid.)
50
D 21.1.49 (8 disp.); 32.5.1 (1 fid.)
194 9. Dates and Plan III
spread over 213 and 214, with three of each in the first year and another three
of each in the second.
The two books of elementary teaching, Institutiones, run to 120 lines in the
Palingenesia.57 One text refers to a reigning emperor:
2 inst. Collatio 16.9 Sed imperator noster eas solas personas voluit admitti,
quibus decimae immunitatem ipse tribuit
The emperor is Caracalla and the text is concerned, it seems, with legislation
on his part restricting to a narrow circle of close relatives the right of intestate
succession to freedmen. Arguments from affinity of style suggest that
Teaching Manual is an early work. Ulpian has two texts in it with aequissimum
putavit, ‘he thought it the fairest solution’.58 There are also two with aequitate
motus, ‘moved by reasons of fairness’.59 Both these point, in view of the Edict
texts, to 213. There are three texts with ut, cum, or si libuerit, ‘as’ or ‘if he
pleases’.60 These rather suggest 214. So do the texts with utpote ‘namely’.61 It
is hard to decide between 213 and 214.
Taken as a whole, the writings of these first two years are self-confident, even
exuberant. Ulpian is relaxed, talks informally to the reader,62 draws on per-
sonal experience,63 and feels free to compliment Caracalla on his rescripts64
and Marcus on agreeing with the author’s own opinion in advance.65
Ulpian’s ambitious undertaking gets off to a powerful start. The detailed pro-
gramme of 431⁄2 books a year for 213 and 214 may have been something like
this:
213
Major commentary On the Edict book 6 to middle of book 31
Medium-scale treatise Proconsul 10 books
Lesser works Taxation 6 books
Market-Masters’ Edict 2 books
Total: 431⁄2 books
57 Lenel (1889) 2. 926–30; Ind. Flor. 24.4; Bremer (1863); Mommsen (1904–13) 2.56–63;
Collectio 2.157–9; Seckel–Kübler 1.492–4; FIRA 2.305f.; Girard–Senn (1967) 540–2; Wieacker
(1960) 206–16; Schulz (1946) 171–2; (1961) 207; Liebs (1997) §424.23
58 Collatio 16.7 (2 inst.); D 2.10.1 pr (7 ed.)
59 Collatio 16.9.2 (2 inst.); D 12.4.3.7 (26 ed.)
60 D 43.26.1.2 (1 inst.); 40.5.46.3 (6 disp.); 50.16.164.1 (15 Sab.)
61 D 1.1.4; 1.4.1 pr (1 inst.); 28.5.17.5 (7 Sab.); 34.4.3.2 (24 Sab.); 46.1.8.8 (47 Sab.)
62 Accipias, accipies: above n.27–8; dicet quis, above n.38
63 Scio, invenio, memini, dicebam, non ignoro: above n.12, 15–8, 45–6, 51–2, 54
64 D 48.19.8.12 (9 off. proc.) 65 D 29.1.3 (2 Sab.)
196 9. Dates and Plan III
214
Major commentary On Sabinus book 1 to middle of book 26
Medium-scale treatise Disputations 10 books
Lesser works Trusts 6 books
Teaching Manual 2 books
Total 431⁄2 books
There are uncertainties but, whatever the exact plan, the general picture pre-
sented here cannot be far from the mark.
The middle year of Ulpian’s five-year plan came in 215. We assumed that he was
then writing the middle segment of Edict, from the middle of book 31 to the mid-
dle of book 56. The medium-range work for the year was thought to be either
Disputations or Tribunals. If the arguments for holding Disputations to have
been written in 214 are accepted, it follows that Tribunals was composed in 215.
There are positive parallels between Tribunals and the thirties and forties
of On the Edict. Ulpian often uses non tantum . . . verum etiam, ‘not only but
also’. One variant of this, nec tantum verum etiam, ‘nor only but also’ occurs
in five texts, all from the middle segment of Edict.66 Another variant, this time
a pleonastic one, is non tantum, verum etiam quoque, ‘not only, but, in addi-
tion, also’, which is found in four texts.67 Ulpian soon eliminated this super-
fluous phrase. But the distribution of these four texts suggests that the early
part of the middle section of Edict and the early books of Tribunals were being
composed at much the same time. If the evidence of these texts can be trusted,
one minor work that occupied his attention at this time was the three-book
work on The Consul.
This picture is strengthened by his use of the diminutive of minutus, viz.
minutulus,68 ‘tiny’, and of minutatim, ‘piecemeal’.69 These rather precious
expressions seem to have enjoyed Ulpian’s favour briefly, presumably in late
214 and 215. This suggests that Tribunals and The Consul should go in 215.
Ulpian’s use of certain polysyllabic, impressive-sounding words falls in the
same period and suggests that other treatises on public officials, besides the
consul, were composed at much the same time. One type of word of this sort
are adverbs in four or more syllables beginning with the prefix in- and ending
in –ter: inconsideranter,70 indubitanter,71 incunctanter,72 indifferenter.73 Not
66 D 9.4.42.2 (37 ed.); 38.2.16.5 (44 ed.); 38.7.2.1; 38.7.2.2 (46 ed.); 38.15.2.4 (49 ed.)
67 D 26.5.7 (1 omn. trib.); 27.4.1.5 (36 ed.); 37.4.3 pr (39 ed.); 25.3.5.12 (2 off. cons.)
68 D 50.16.192 (37 ed.)
69 D 32.55.2 (25 Sab.); 2.15.8.9 (5 omn. trib.); 34.1.3 (2 off. cons.)
70 D 26.10.3.17 (35 ed.) 71 D 37.11.2.7 (41 ed.) 72 D 40.2.20 pr (2 off. cons.)
73 D 1.13.1.3 (1 off. quaest.) 74 D 26.10.3.16 (35 ed.)
Lesser Works 197
far removed is pervicaciter.74 The distribution of these words suggests that the
monographs on the office of consul and quaestor belong together. The next
group of words brings in two more books on offices, those of the urban pre-
fect and the community treasurer (curator reipublicae), an imperial commis-
sioner appointed to a municipality. This group of assonant words consists of
five-syllable words in -osus or -ose, apart from iniuriosus and iniuriose, which
are omitted because of their technical connection with the delict of iniuria.
They include perniciosus,75 perniciose,76 facinorosus,77 contumeliosus,78
incuriosus,79 and ambitiosus.80 These pretentious expressions suggest that
Ulpian, for whatever reason, indulged at this period in unusual rhetorical
excess. Here are other examples from works that appear to belong to 215:
D 26.7.7.1 (35 ed.) non quidem praecipiti festinatione, sed nec morato-
ria cunctatione
D 22.1.33 pr (1 off. cur. reip.) non acerbum se exactorem nec contumeliosam
praebeat sed moderatum et cum efficacia benignum
et cum instantia humanum
D 1.12.1.8 (1 off. pr. urb.) si verecunde expostulent, si saevitiam, si duritiam,
si famem, qua eos premant, si obscaenitatem, in
qua eos compulerint vel compellant, apud praefec-
tum urbi exponant
There are further links between the middle section of Edict, the treatise on
Tribunals, and the monographs on public offices. Raro or perraro faciendum
est ‘it should be done only rarely’ comes only in Tribunals 81 and The Consul.82
Liberalis, in the context of liberal arts or studies, comes in Tribunals 83 and
The Community Treasurer.84 Licentiam habet ‘he is allowed’ comes in this
block of Edict 85 and The Urban Prefect.86 Licentia erit ‘it will be permissible’
is similar and comes only in this block of Edict.87 Accipere nos oportet ‘we
must take it that . . .’ occurs mainly in this edictal block88 and in The Consul.89
There is perhaps a link between acerbum ‘harsh’ and acriter ‘harshly’, the first
in Tribunals 90 and the second in The Community Treasurer.91
Three works on the duties of public officials,92 the consul, the urban
prefect, and the community treasurer therefore seem to have links with the
middle block of Edict, and with Tribunals. Did the whole programme of
75 D 14.6.3.3 (29 ed.); 24.3.22.7 (33 ed.); 47.2.50.4 (37 ed.) 76 D 26.10.3.5 (35 ed.)
77 D 2.1.3 (1 off. quaest.) 78 D 1.12.1.10 (1 off. pr. urb)
79 D 22.1.33 pr (1 off. cur. reip.) 80 D 22.1.33 pr; 50.9.4 pr (1 off. cur. reip.)
81 D 26.10.7.3 (1 omn. trib.) 82 D 5.1.82 (1 off. cons.)
83 D 50.13.1 pr (8 omn. trib.) 84D 50.9.4.2 (1 off. cur. reip.)
85 D 24.3.22.8 (33 ed.) 86 D 1.12.1.3 (1 off. pr. urb.)
87 D 24.3.22.7 (33 ed.), 26.7.1.3 (35 ed.)
88 D 9.2.5.1 (18 ed.); 26.2.3.1 (35 ed.); 38.5.1.4 (44 ed.); 38.8.1.6 (46 ed.)
89 D 50.16.99.1 (1 off. cons.) 90 D 47.10.35 (3 omn. trib.)
91 D 22.1.33 pr (1 off. cur. reip.)
92 On which see Jörs (1903) 1452; Solazzi (1920–2) = (1957–63) 2.521–6; Dell’Oro (1960)
minor works for 215 consist of monographs on the duties of public officials?
If we assume a total of eight books of minor works, these three monographs
together come to five books. We can possibly add three out of The Quaestor
(probably one book), The Prefect of Police (one book), The Praetor for
Guardianship (one book), and Consular Judges (one book).
This gives eight books. On the duties of consular judges (circuit judges in
Italy) we have only three lines, on the duties the prefect of the watch only one.
It is impossible to link them linguistically with the other monographs. But it
seems reasonable to suppose that the works on public offices were mostly
composed in the same year and were regarded by Ulpian as forming a group.
Notes on individual works follow.
93 Lenel (1889) 2.951–8; Ind. Flor. 24.13; Jörs (1903) 1452; Fitting (1908) 119; Schulz (1946)
243; (1961) 314; Solazzi (1957–63) 2.521–6; Dell’Oro (1960) 31–85; Liebs (1997) §424.16
94 D 27.3.17 (3 off. cons.)
95 Lenel (1889) Pal. 2. 959–60; Ind. Flor. xxiv 19; Schulz (1946) 246; (1961) 313; Jörs (1903)
1454; Fitting (1908) 120; Dell’Oro (1960) 239–49; Mantovani (1988) 171–223; Liebs (1997)
§424.17
Lesser Works 199
1 off. pr. urb. D 1.15.4 Imperatores Severus et Antoninus Iunio Rufino ita
rescripserunt
Severus, divus in three references, is dead. We are not told who is currently
emperor, but Caracalla is not ruled out. In the last text, however, Severus is
mentioned before Caracalla, which points to the Caracalla A period. The
rescript mentioned is however a letter (epistula) of which the exact words are
cited.96 In this type of case there is sometimes an exception to the order of
emperors even in the Caracalla B period.97 In all there are five or six excep-
tions of this sort to set against seventy-one texts that follow the standard pat-
tern.98 The work may therefore have been composed in 215 along with other
monograph on public offices.
104 Lenel (1889) 2.960–6; Ind. Flor. 24.22; Jörs (1903) 1452; Fitting (1908) 118–9; Schulz
(1946) 249, (1961) 318–20; Wieacker (1960) 412–6; Dell’Oro (1960) 88–98; De Filippi (1984);
Liebs (1987b) 151; (1997) §424.10–11
105 D 27.1.7 (1 excus.) = Frag. Vat. 240 (1 off. pr. tut.); Frag. Vat 145 (1 excus.) = Frag. Vat
222 (1 off. pr. tut.); D 27.1.15.16 (Mod. 6 excus.–Ulp. 1 excus.) = Frag. Vat. 189 (off. pr. tut.)
106 Frag Vat. 125, 147, 159 (all 1 excus.); Fitting (1908) 115 but Ebrard (1917) 144; H. Krüger
(1930) 303.
Lesser Works 201
Of the eight texts in which Severus and Antoninus appear jointly, Caracalla
precedes his father in seven. In one Digest text,107 however, Severus has pri-
ority. The Praetor for Guardianship looks early and we have already come
across two texts in it with scio,108 and one each with invenio,109 inveni,110 and
memini.111 It is strong on first-person words referring to mental activity.
There is a case for allotting it to 213 or 214, but, along with other works on
public offices, it is perhaps better placed in 215.
The only sentence that survives from this work comes from a constitution of
Marcus. It contains nothing to indicate the period in which the work was
composed.
There are only five words from this work, which give no clue as to its date.
Certainty is not possible, but it seems likely that the 43 books composed in
215 consisted of:
Major commentary On the Edict from middle of 31 to
middle of 56
Medium-scale work Tribunals 10 books
Lesser works The Consul 3 books
The Urban Prefect 1 book
The Community Treasurer 1 book
The Quaestor 1 book
The Praetor for Guardianship 1 book
either The Prefect of Police 1 book, or Consular Judges
1 book
Total: 43 books
In fixing the detailed programme for 216 and 217 we can begin by eliminating
the minor works already assigned to the first three years of the quinquennium.
107 108
D 27.1.9 (1 off. pr. tut.) Frag. Vat 189, 242
109 110
Frag. Vat. 189 cf. D 27.1.15.16 (1 excus.) Frag. Vat. 177
111
Frag. Vat. 220
112
Lenel (1889) 2.950; Jörs (1903) 1452; Schulz (1946) 247, (1961) 315; Dell’Oro (1960) 260–3,
299; Liebs (1997) §424.12. Not in Ind. Flor.
113
Lenel (1889) 2.960; Ind. Flor. 24.20; Jörs (1903) 1454; Schulz (1946) 246; (1961) 314;
Dell’Oro (1960) 250, 254f.; Vriesendorp (1984); Liebs (1997) §424.18
202 9. Dates and Plan III
This leaves:
Adultery 5 books
The Aelio-Sentian Law 4 books
Appeals 4 books
Engagement to Marry 1 book
The Speech of Marcus and Commodus 1 book
Consular Judges/Prefect of Police 1 book
This makes sixteen books, eight each year, which is the number required to
bring Ulpian’s total up to 431⁄2 books for each of these years. In 216 the block
of major commentary is thought to have been 251⁄2 books On Sabinus, from
the middle of book 26 to the end of book 51 and the middle range treatise for
that year is thought to have been the first ten books on the Julio-Papian Law.
In 217 the block of major commentary is thought to have run from the mid-
dle of Edict book 56 to the end of book 81, which again comes to 251⁄2 books.
The middle-range treatise for that year is thought to have been the last ten
books of The Julio-Papian Law. If eight books of lesser works are added each
year we arrive at the required total.
If this outline is correct, a pattern emerges. Ulpian’s commentaries on
statutes (leges) belong to the last two years of the quinquennium. No work of
this sort seems to come in the first three years. Counting twenty books on the
Julio-Papian Law, five for the Julian Law on Adultery, and four for the Aelio-
Sentian Law, we have twenty-nine books on statutes in the last two years.
Somewhat as in the Digests (Digesta) of Celsus, Julian, and Marcellus,114 the
analysis of statutes formed an appendix to the discussion of other branches of
the law.
The various lesser works are now dealt with in turn.
Ulpian often uses et magis est, ‘and the better view is’.116 One variant of this,
magisque est, occurs only in thirteen texts, concentrated in later 216 and early
217.117 This distribution of texts tends to confirm the view that the last books
of Sabinus, ending at book 51, and the first books from the final block of Edict
texts, beginning from the middle of book 56, belong to the same period,
114
Lenel (1889) 163–9 (Cels. 28–9 dig.), 1.464–84 (Iul. 59–90 dig.), 1.627–32 (Marc. 21–31
dig.)
115
Lenel (1889) 2.379–84; Ind. Flor. 24.12; Jörs (1903) 1452; Schulz (1946) 256; (1961) 329;
Fitting (1908) 120; De Giovanni (1989) 95–101; Liebs (1997) §424.22
116
Eighty-eight texts: Honoré–Menner (1980) fiche 56 ET MAGIS EST
117
D 49.4.1.9 (1 appell.); D 37.14.16.1 (10 leg. Iul. Pap.); D 45.3.7 pr (48 Sab.); D 42.4.3.3 (59
ed.); D 40.5.4.15, 16, 22 (60 ed.); 28.8.8 (61 ed.); 29.2.71.1, 9 (61 ed.); 44.2.7.3 (75 ed.); 46.7.3.3, 8
(77 ed.)
Lesser Works 203
late 216 and early 217.118 The tenth book of The Julio-Papian Law fits this
pattern, since it should belong to late 216. Appeals could, consistently with
this bunching, belong to either year, but perhaps fits 217 rather better.
Some evidence can be drawn from a study of another construction, palam
est (‘it is clear’). In earlier years Ulpian uses this at the end of a clause. Palam
est follows the object or accusative and infinitive that it governs, as in ex con-
ducto actionem transire palam est.119 In a later group of texts, however, it is
the other way round: palam est precedes the object or accusative and infinitive
which it governs.120 Appeals is again associated with texts that belong to late
216 or 217. One text from Appeals refers to a reigning emperor:
4 appell. D 49.5.5.3 idque rescriptis imperatoris nostri Antonini declaratur
118
Further evidence of this linkage is provided by texts with the phrase praeterea sciendum est:
D 46.1.8.1 (47 Sab.); 42.6.1.12 (64 ed.); 42.8.6.9 (66 ed.); 43.19.1.10; 43.20.1.23 (70 ed.); 42.8.10.6
(73 ed.); 44.4.4.10 (76 ed.)
119 D 19.2.19.8 (32 ed.)
120 D 49.4.1.11 (1 appell.); 19.1.10 (46 Sab.); 18.4.2.3; 45.1.38.3 (49 Sab.); 42.8.3 pr (66 ed.);
43.16.1.24 (69 ed.); 43.21.1.7 (70 ed.); 44.4.2 pr (76 ed.); 46.7.5.3 (77 ed.)
121 Lenel (1889) 2.3791
122 Lenel (1889) 2.931–9; Ind. Flor. 24.11; Jörs (1903) 1446; Schulz (1946) 188, (1961) 232;
These links suggest that Adultery was composed in 217. What of the evid-
ence from references to emperors? There are two:
2 adult. D 48.5.14.3 Divi Severus et Antoninus rescripserunt
2 adult. D 48.5.14.8 poterit accusari ex rescripto divi Severi, quod supra rela-
tum est
Severus is clearly dead. What of Caracalla? One could with Lenel and Fitting
take the first text as showing that Adultery was composed after Caracalla’s
death and deification. But the second reference is to the same rescript as the
first. It implies that, though technically a joint rescript, it was in substance the
ruling of Severus. This suggests that Ulpian is writing after Caracalla’s death
when it was no longer necessary to treat him as the real author of constitu-
tions issued by Caracalla jointly with his father. Besides it is easier to insert
divus or divi into a text retrospectively than to eliminate it.131 The first refer-
ence could originally have read Severus et Antoninus rescripserunt. In that
case the mode of citation is appropriate to the reign of Macrinus132 and can
properly be placed in the year 217. This seems to me the correct interpreta-
tion. Adultery should be assigned to 217.
This work on the Aelio-Sentian law concerns a statute of AD 4 about the man-
umission of slaves. Thirty-seven lines have been preserved but they do not
refer to a reigning or recently dead emperor. A possible date is 216, and there
is a slender piece of evidence in its favour. Verum est followed by the
accusative and infinitive comes only in The Aelio-Sentian Law134 and in a late
book On Sabinus.135
The two excerpts from this contain only seven lines as printed in the
Palingenesia. They do not point to any particular year.
(1997) §424.9
Lesser Works 205
Though this work is not in the Florentine Index and is not confirmed by any
text heading, it must have existed.138 The quotation from it in Paul’s mono-
graph contains the phrase ‘the point at issue is’ (quaestio in eo est).139 This is
found otherwise only in four texts of Ulpian.140 These point to 216 or 217, the
years of the later books On the Edict and On Sabinus.
This brings to an end the discussion of the dates of Ulpian’s lesser works.
Which of them are to be attributed to 216 and which to 217 is doubtful. The
following is a possible scheme:
216
Major commentary On Sabinus middle of 26 to
end of 51
Medium scale treatise The Julio-Papian Law books 1–10
Lesser works The Aelio-Sentian Law 4 books
Engagement to Marry 1 book
Consular Judges/Prefect of Police 1 book
Appeals books 1–2
Total: 431⁄2 books
217
Major commentary On the Edict middle of 56 to
end of 81
Medium-scale treatise The Julio-Papian Law books 11–20
Lesser works Adultery 5 books
Appeals books 3–4
Speech of Marcus and Commodus 1 book
Total: 431⁄2 books
The overall picture is that of 217 books composed in the five years 213–7.
According to the scheme set out, 431⁄2 books were written in four of these
years and 43 in the fifth (assumed to be 215). Given the sustained quality of
the text this constitutes a monumental achievement.
137 Not in Digest or Ind. Flor. but cf. Paul, Ad orat. d. Ant et Comm: Lenel (1889) 1.1145–6;
Spurious Works
false was virtually impossible, since hardly anyone had time to read through
their voluminous works.
We have very few texts from these five works outside the Digest. Three
replies (responsa) of Ulpian are cited in other sources, but they are not attrib-
uted to any published work of his.5 In fact none of these works is mentioned
outside the Digest with the exception of Rules in one book, which will for con-
venience be given its Latin abbreviation, LSR for Liber singularis regularum.
Three texts from LSR6 appear in the Collatio (Romanorum et Mosaicarum
legum Collatio, or Lex dei quam deus praecepit ad Moysen). This compilation
of biblical and legal sources, which was intended to show that the law of
Moses said the same thing as Roman law but at an earlier date, was composed
between 390 and 410.7 Hence LSR was known in the late fourth or early fifth
century.
5 D 19.1.43 (Paul 5 quaest.); 50.5.5 (Macer 2 off. praes.); CJ 8.37.4 (Alexander 31 March 222)
6 Collatio 2.2.1; 6.2.1–4; 16.4.1–2 7 Liebs (1987b) 162–74
8 Lenel (1889) 2.1016; Krueger, Collectio 2.1–38; Seckel–Kübler 1.436–491; FIRA
a version that breaks off about two-thirds of the way through. The revised
version probably belongs to the fourth century, since Tituli omits references
to the penalties imposed on unmarried and childless people by Augustus’
laws. The omission probably reflects Constantine’s law of AD 320, which
removed the penalties.16 The fourth-century version of LSR should therefore
be later than 320 but perhaps earlier than 342, when Constantius II repealed
the law, designed to accommodate the emperor Claudius, that allowed a man
to marry his brother’s daughter.17 Tituli does not take account of this last
repeal. Indeed it contemplates gifts by will to certain pagan tutelary gods.18
Schulz went further. He argued in the introduction to his edition of Tituli,
which he called Epitome Ulpiani,19 that LSR was a post-classical work
derived mainly from Ulpian but also from other classical authors, such as
Modestinus. But Tituli appears, apart from the disabilities of the unmarried
and childless, to state classical law, though with some omissions. Indeed in its
original form it antedates the Antonine constitution of AD 212, since it refers
repeatedly to foreigners (peregrini) and those of Latin status (Latini) in a way
that implies that their status is still of daily concern.20 In short Tituli seems to
come from a fourth century edition of a classical but pre-212 LSR, marked by
some omissions but few if any additions.
What was the nature of the original LSR? On one view it was a reworking
of Gaius’ Institutes. There are some passages that look rather similar,21 and
one unusual common phrase.22 The strongest verbal parallel concerns a reso-
lution of the senate on the appointment of guardians.23 But this perhaps
merely shows that both authors followed the wording of the resolution.
Though the order in which topics are treated is on the whole similar to that of
Gaius, sometimes there are omissions, alternative ways of expressing things,
or a different order of exposition. The author of LSR is an independent
teacher and author.24 In my view his work can be taken as a ‘nutshell’ type
handbook that follows in the footsteps of Gaius but tries to improve on
him.25 In particular LSR outdoes Gaius by dividing more subjects up by
numbers (e.g. ‘three types’,26 ‘four ways’27) and providing more lists.28 This
technique, designed to make the law easy to remember, succeeds in providing
a clear guide to the main rules of Roman civil law, a guide that earned
16 CTh 8.16.1 (31 Jan 320) 17 CTh 3.12.1 (31 March 342) 18 Tit.22.6
19 Schulz (1926) 8–9
20 Tit. 1.5, 12, 16; 3.1–6; 5.4,8–9; 7.4; 10.3; 11.16, 19; 17.1; 19.4; 20.8, 14; 22.3,8; 25.7;
29
Mommsen (1904–13) 2.48
30
Tit. 1.13a; 3.3; 8.2f.; 11.18, 20; 11.14, 27; 12.1, 3; 25.26; Liebs (1982) 284
31 32 33
Most recently Mercogliano (1997, 1998) Tit. 26.7 Tit. 22.34
34 35
Tit.13.2; Liebs (1982) 284. Nelson (1981) 91
36
Tit. 1.8, 10, 21; 5.6; 8.5; 11.8, 20, 17.2; 18.1; 20.2; 22.34; 24.12, 13, 31; 26.7; 29.3,6
37
Schönbauer (1956) 308f.
38
Tit. 17.2: hodie ex constitutione imperatoris Antonini omnia caduca fisco vindicantur, sed ser-
vato iure antiquo liberis et parentibus
39 40
Dio 78.9 Tit. 26.7
210 10. Spurious Works
Mauricianus,41 who wrote under Pius.42 The case for an early date is made by
Avenarius in a forthcoming study based on an analysis of individual texts in
LSR.43
Various aspects of LSR set it apart from Ulpian’s work. The author calls
the Sabinians Cassiani.44 He refers to Neratius as Priscus,45 which is contrary
to Ulpian’s uniform practice in 76 texts where he calls the author Neratius.46
In his Teaching Manual (Institutiones) Ulpian refrains from vague appeals to
authority of the sort found in LSR such as ‘most lawyers think’ (plerisque
placet),47 ‘it is held’ (placuit),48 ‘it is agreed’ (constat),49 ‘some say’ (sunt tamen
qui dicunt),50 and ‘that is the general practice’ (et id observatur magis).51 Apart
from one mention of Celsus52 and one of imperator noster (Caracalla)53 he
does not cite authority at all in this elementary work.
Whether in general the style of LSR is consistent with Ulpian’s authorship
has led to a conflict of opinion. To a number of authors including myself LSR
is plainly not his work.54 Its unadorned definitions and dry numbered lists do
not fit any of Ulpian’s admittedly genuine writings. Even a juvenile Ulpian
would have explained himself more fully. There are virtually no linguistic par-
allels with genuine works, apart from the phrase ut puta (‘for example’), which
occurs four times in LSR.55 In the Digest it comes from texts of Ulpian 229
times out of 244, so that it is clearly a mark of his style, though it is also found
occasionally in Callistratus,56 Papinian,57 and Paul.58 But a single trait of this
sort is not enough to stamp LSR as Ulpian’s, though as all four authors are
Severan lawyers it perhaps fits that period. Statements of the law by Ulpian
and LSR sometimes diverge. According to LSR a trust in contrast with a
legacy can be in Latin or Greek,59 whereas according to Ulpian it can be in
any language.60
Other authors, however, take the view that LSR could be a work of Ulpian,
or at least a work put together from his writings by a pupil or associate.61 To
leave out connective tissue was correct, says Mommsen, in a work aimed
at practice.62 Nelson argues that rules (regulae), nominally the genre of
LSR, can be composed in a different style from more extensive works. But his
41 42 43
Tit. 13.2 Liebs (1997) §421.3 In preparation
44 45 46
Below n.73 Tit. 11.28 Honoré–Menner (1980) fiche 64
47 48 49 50 51
Tit. 1.18 Tit. 1.21 Tit. 1.21 Tit. 1.21 Tit. 22.22
52 53
D 1.1.1 pr (1 inst.) Collatio 16.9.3 (Ulp. inst.)
54
Schulz (1926, 1961); Arangio-Ruiz (1921, 1927); Liebs (1982, 1987b); (1997) §428.5; Honoré
(1982, 2000); Marotta (2000) 2556
55 Tit. 2.7; 24.17, 25, 27 cf. 19.18; 25.14; 29.1 (puta) 56 D 14.2.4.2 (Call. 2 quaest.)
57 D 11.7.43 (Pap. 8 quaest); 22.1.3.2 (20 quaest.); 45.2.9 pr (27 quaest.); 48.5.39 pr (36
argument is not convincing, least of all for Ulpian, the style of whose
Teaching Manual (Institutiones) resembles his large-scale works like the edic-
tal commentary and his monographs on special subjects like trusts.63
The author of LSR has his own turns of phrase, such as praetor urbis 64 for
the urban praetor where other lawyers use praetor urbanus. A person can give
a gift by will ‘imperatively’ (imperative)65 or ‘by request’ (precative).66 Neither
word is found in the Digest. He is fond of the preposition per: per eminen-
tiam,67 per consequentiam,68 per similitudinem,69 per formulam.70 There are
some parallels with Paul’s vocabulary. Paul and the author of LSR, along
with Pomponius71 and Pliny the Younger,72 call the well-known school deriv-
ing from Cassius Longinus and Masurius Sabinus Cassiani.73 Ulpian74 and
Marcianus75 call them Sabiniani. Abalienare for ‘to alienate’ is found, apart
from LSR, only in Paul and his teacher Scaevola.76 Infinite (‘without limit’) is
found in LSR and otherwise only in a Paul comment on Labeo.77 ‘But it is
safer’ (sed tutius est), a sign of interpolation according to an earlier genera-
tion,78 also has a Paul parallel.79
The LSR does not much resemble other collections of rules, but is more of
a teaching manual. In any case rules are not all composed in a special style.
Licinius Rufinus, a pupil of Paul and a Severan lawyer, composed twelve
books of Rules with pithy sentences such as ‘a ward who takes a loan is not
bound even by natural law’.80 A Thyatira inscription81 records Licinius as π
τν ποκριµτων, which in my view means secretary for petitions (a libellis),
in the 220s.82 The most likely period is October 222 to October 223, roughly
when Ulpian was sole praetorian prefect.83 The rescripts of this period display
the same well-turned, pithy phrases that are found in Licinius’ Rules. Thus,
‘no law forbids a woman from giving her whole estate to her
husband as a dowry’.84 To a limited extent authors certainly adapt their writ-
ing to the particular genre. But though rules and rescripts belong to quite dif-
ferent genres, Licinius’ Rules and the rescripts composed for Alexander
(12 ed.)
77
Tit. 5.6; D 22.3.28 (Lab. 7 pith. Paul epit.)
78
Gradenwitz (1887) 133; Beseler (1910–20) 2.164, 4.159; Schulz (1926) ad Tit. 22.22
79
Tit. 22.22; Paul 19.5.5.4 (Paul 5 quaest.)
80
D 44.7.58: pupillus mutuam pecuniam accipiendo ne quidem iure naturali obligatur
81 82
Herrmann (1997) 111f. Millar (1999) 90–108
83
Honoré (1994) 98–101 (secretary no.7)
84
CJ 5.12.4: nulla lege prohibitum est universa bona in dotem marito feminam dare cf. Honoré
(1994) 98
212 10. Spurious Works
Severus in 222–3 have this common feature, which supports the notion that
he composed the rescripts.
Can LSR, if not a work of Ulpian himself, be a selection from his works put
together by a pupil or associate?85 One difficulty with this fence-sitting com-
promise is that up to 212 there would not be much material of Ulpian from
which to choose.86 An alternative that keeps a link with Ulpian is that early
in his career the great lawyer encouraged a pupil or associate to produce this
manual as a guide to study and practice.87 The work was written but not
re-edited after 212, perhaps because Ulpian’s own elementary Teaching
Manual (Institutiones) in two books supervened in about 214.88 But in that
case how did LSR come to be attributed to Ulpian? Perhaps through
Modestinus, Ulpian’s pupil89 and the most likely recipient of his papers when
he was murdered. As the probable secretary for petitions from October 223 to
October 225,90 Modestinus must have been in Rome at the time of Ulpian’s
death or shortly afterwards. There are close parallels between the definitions
of acquisitive prescription (usucapio) in LSR and Modestinus.91 The same is
true of the passages on the enactment, amendment and repeal of laws.92
Schulz thought that the supposedly post-classical author of LSR borrowed
from Modestinus. But it could be the other way round. Perhaps Modestinus
borrowed from the author of LSR when compiling his Encyclopaedia
(Pandectae) and Rules (Regulae). LSR, a product of Ulpian’s circle of pupils
or associates, came to be treated as a genuine work of his.
This is speculation. Given the dates, and despite the traditional name, a pupil
or associate of Paul is not ruled out as the author of LSR. Or the work may
have been anonymous until it was attributed to Ulpian in the fourth century.93
Much depends on when the canonical list of works of the leading lawyers was
settled. Perhaps under Constantine. If one useful elementary work, Pseudo-
Paul’s Sentences (Pauli sententiae) could then be falsely assigned to Paul in
order that it might be used in court,94 why not another to Ulpian?
The Digest has two fragments95 ascribed to Ulpian liber singularis pan-
dectarum.96 The Florentine Index, however, mentions ten books, of which
The word condicticia is suspect, for it has in several texts been interpolated by
the Digest compilers.100 But not all mentions of the word are spurious. Ulpian
cites Julian in a text that draws a valid distinction between the penal action
for theft (actio furti) and the personal action (actio condicticia) to recover the
value of the stolen property.101 In the text set out, condicticia may be authen-
tic, but it is another matter to defend it as genuine Ulpian. The sentence states
the obvious. If the stipulation was for a certain thing, then the action must of
course be for a certain thing. The phrase actione per quam certum petitur is
also pleonastic. You do not sue by bringing an action. To bring an action,
agere, is to sue for something, petere.
These two texts do, however, have affinities of style with the rescripts of
March to October 222, when Ulpian was prefect of supply and later praetor-
ian prefect to Alexander. The composer of these rescripts, my secretary
no.6,102 favours assonances with sibilant sounds, like the esset, gessisset of D
40.1.34.1. Rescripts of this period have the sequences fuistis . . . adquisistis,103
97 Lenel (1889) 2.10135: Liebs (1997) §428.6 p.209 98 Lenel (1889) 2.1013
99 D 12.1.24
100 Ind. Interp. ad loc. (Seckel, Triantaphyllopoulos, Naber, Pernice, Lenel, Beseler)
101 D 12.2.13.2 (Ulp 22 ed.) 102 Honoré (1994) 95–8 103 CJ 4.50.2 (20 March 222)
214 10. Spurious Works
104 105
CJ 7.64.1 (25 March 222) CJ 7.56.1 (7 May 222)
106 107
CJ 7.8.4 (10 May 222) CJ 3.32.3 (30 Oct 222)
108 109
Honoré (1994) 97 CJ 2.18.10 (20 Nov 222)
110 111
CJ 9.1.3 (3 Feb 222) D 8.4.13.1 (6 opin.)
112
It is found in a Marcellus text: D 33.2.15.1 (Marc. 13 dig.)
113 114 115
CJ 4.14.3 (13 Sept 222) D 12.1.24 (1 pand.) CJ 4.44.1 (1 March 222)
116 117
CJ 3.35.1 (7 Nov 222) CJ 2.3.9 (28 Sept 222)
118
Cod. Greg. Visi. 13.14.1 (30 Dec 218)
119 120
Liebs (1971) 51f.; (1984) 446; Liebs (1997) §428.6 Eutropius 8.23
121 122
Festus 22.3 HA Severus Alexander 26.6
10. Spurious Works 215
In Lenel’s reconstruction there are 107 lines attributed to this work. The frag-
ments deal predominantly with private law. The order of treatment is obscure
and there are no references to legal authors or emperors. The work includes
some famous maxims, such as that justice is the constant and perpetual will
of giving each his own;124 that the precepts of law are to live honestly, not to
harm others, and to give each his own;125 that jurisprudence is the knowledge
of things divine and human, of the just and unjust;126 and that gross negli-
gence consists in not understanding what everyone understands.127 The
author clearly had philosophical leanings.
The work’s authenticity has long been assumed. But, as I pointed out in the
first edition,128 there are no grounds for attributing it to Ulpian, painful as it
may be to jettison him as author of these famous maxims.129 Rules contains
none of his favourite phrases. It has expressions and constructions foreign to
his style. For example the author uses voluptuosus,130 as does the author of
LSR, whereas Ulpian consistently uses voluptarius for ‘luxury’ expenses
etc.131 Another spurious expression is ‘on a favourable interpretation’
(benigna interpretatione), as in ‘on a favourable intepretation it is replied
rather by most lawyers that no one is disinherited’.132 Apart possibly from
one text with ‘on a humane interpretation’ (humana interpretatione),133 which
has plainly been altered by the compilers,134 Ulpian does not use this type of
expression at all, though it occurs in Scaevola.135 In any case the pleonastic
‘rather by most lawyers’ is impossible in such a forthright author as Ulpian.
And ‘it is replied’ (respondetur) is a vague expression not found in Ulpian
apart from this text.136 ‘According to the opinion of all’ (secundum omnium
sententiam) is, again, a stereotyped appeal to authority of the sort that points
to a provincial lawyer with limited access to original texts. Another text
123 Lenel (1889) 1.1013–1015; Honoré (1982) 111f.; Frezza (1983) 416; Liebs (1982) 282–92;
(1984) 446; Huchthausen (1985) 719f.; Mayer-Maly (1961) 569; Liebs (1989) §507.3
124 D 1.1.10 pr (1 reg.) 125 D 1.1.10.1 (1 reg.) 126 D 1.1.10.2 (1 reg.)
127 D 50.16.213.2 (1 reg.) 128 First in a lecture on 27 February 1981: Liebs (1982) 282
129 How could giving each his own be both the criterion of justice and one of three precepts of
law?
130 D 25.1.14.2 (5 reg.)
131 Tit. 6.14, 17 (voluptuosus) as against D 25.1.1,7,9, 11 pr (36 Sab.); 7.1.13.4 (17 Sab.), all
voluptarius
132 D 28.2.2 (6 reg.) cf. 39.5.16 (2 resp., also spurious)
133 D 38.17.1.6 (Ulp. 12 Sab.); Crifò (1985) 608–9 134 Index interp. 3.72
135 D 34.1.20.1 (Scae. 3 resp.)
136 It occurs in Tryphoninus: D 37.4.20.1 (Tryph. 19 disp.) and Paul: D 19.4.1.1 (33 ed.)
216 10. Spurious Works
speaks of a husband being able to choose which remedy to use against his
wife: ‘and it is in [the husband’s] power which action he chooses to use’ (et in
potestate est, qua velit actione uti).137 In all other passages of Ulpian ‘in potes-
tate est’ means that someone, such as a son, is in the legal power of another,
such as his father. The phrase is ungrammatical. It should run ‘and it is in his
(or the husband’s) power’ (et in eius/viri potestate est).
There are also sentences that run two points together in a confusing way
that Ulpian avoids. ‘If someone knowingly buys a free man a capital crime
arises’138 runs together two rules. The knowing purchase amounts to kidnap-
ping and kidnapping is a capital crime. It was not, however, a capital crime in
Ulpian’s day.139 ‘When the instititution of an heir is made subject to a impos-
sible condition consisting in an abstention’ is expressed as Si in non faciendo
impossibilis condicio institutione heredis sit expressa.140 It ought, in the inter-
ests of clarity to run Si institutione heredis impossibilis condicio, quae in non
faciendo consistit, sit expressa. Even so, ‘when an heir is instituted’ (cum heres
instituitur) would be more like genuine Ulpian than institutione heredis. The
author has again run together ideas that Ulpian would have expressed in sep-
arate clauses, and moreover has put them in the wrong order. ‘Apart from the
fact that’ (praeterquam quod),141 an ugly expression, is not otherwise found in
Ulpian except in a quotation from the praetor’s edict.142
The author is fond of asyndeton: honeste vivere, alterum non laedere, suum
cuique tribuere;143 aggeres facere, flumina avertere, aedificia vetera fulcire
itemque reficere, arbores in locum mortuorum reponere;144 et bonorum posses-
sionem dare potest et in possessionem mittere, pupillis non habentibus tutores
constituere, iudices litigantibus dare.145 He likes to invert the conditional:
Communis servus etiamsi . . .;146 procurator si quidem . . .;147 servus communis
ab extero heres institutus si . . .;148 servum meum heredem institutum cum liber-
tate si vivus vendidero . . . 149 Another inversion is: statulibera quidquid peperit
. . .150 None of these traits is a feature of Ulpian’s writing.
Can we fix the date and place of composition of Rules in seven books?
What type of author is likely to be responsible for the work? Frezza showed
that the author had links with Origen’s circle.151 Liebs suggests that the
author was a late-third-century provincial lawyer with philosophical inter-
ests, perhaps a Beirut teacher in touch with that circle.152 That the magistrate
having jurisdiction can appoint guardians to those under age points away
from Rome, where the praetor for guardianship (praetor tutelaris) had this
special jurisdiction. The author may have been an Ulpianus, that name being
common in Syria, or the name may have been falsely attached to the work in
order to boost circulation, something that, to judge from Pseudo-Paul’s
Sententiae, was a popular technique around AD 300 and a little later. The
book was the right length and the author or his agent could safely pretend
that it consisted of a collection of texts from the voluminous writings of
Ulpian, whose work may indeed have been one of the sources used.
4. Opinions (6 books)153
As printed in Lenel’s Palingenesia, we have 645 lines from six books of texts
attributed to the Opinions of Ulpian. The title (Opiniones) is unique in juristic
literature and seems to echo the statement of Gaius that the opinions and
views of lawyers of authority have the force of law, rather as does the title of
Pseudo-Paul’s Sententiae.154 The authenticity of the work has been in doubt
since Gothofredus first questioned it.155 Lenel and others held it to be a post-
classical compilation from genuine writings of Ulpian.156 Rotondi argued
that it was an anthology composed in the fourth century on the basis of frag-
ments of Ulpian.157 Wieacker pointed to the absence of Ulpian’s syntax and
vocabulary, to borrowing from the chancery style of the fourth century, to
contents that can be explained only on the basis of post-Diocletianic imperial
law, and to an arrangement that conforms more to collections of imperial
constitutions than to the order of the edict.158
Santalucia, however, in an extensive study, argued that the Opiniones are,
despite differences of style, a genuine work of Ulpian.159 In the first edition I
rejected his view but argued that the work, though not Ulpian’s, might belong
to the third century, even to the reign of Alexander Severus. The author could
perhaps be identified with secretary for petitions no.7, who held office from
October 222 to October 223. Liebs rightly rejected this identification on the
ground that, though the author may well have included material from the
rescripts of this period, none of the parallels I pointed to was compelling.160
It has since become clear that there is a better candidate for secretary no.7 in
153 Lenel (1889) 2.1001–13; Jörs (1903) 1450; Schulz (1946) 182; (1961) 22; Rotondi (1922)
1.453–85; Santalucia (1971); Volterra (1972); Wieacker (1973) 196–207; Liebs (1973); (1976) 321,
332–8; (1981); Honoré (1981) 120–8; Liebs (1983a) 499–501; (1984) 446; Honoré (1994) 101;
Liebs (1989) §507.4
154
Gaius, Inst. 1.7: Responsa prudentium sunt sententiae et opiniones eorum quibus permissum
est iura condere
155
Gothofredus (1653) 259 ad D 50.17.61; Rotondi (1922) 1.453f.; Arangio-Ruiz (1946) 169f;
Schulz (1961) 223
156 157 158
Lenel (1889) 2.1001–13 Rotondi (1922) 1.453 Wieacker (1960) 68f
159
Santalucia (1971) criticized by Liebs (1973) 279–310; (1976) 332–8; (1983) 499–501 but
contra Wacke (2000) 513–8.
160
Liebs (1983) 499–501
218 10. Spurious Works
There are eight words between ministeria and procedentia in this portmanteau
explanatory phrase. Others texts in Opinions have nominal phrases that come
close to it in length.165 Opinions also has alliterative assonances that are foreign
There are grammatical ineptitudes. ‘As regards civil disturbances, though the
local authority often suffers harm from them, yet . . . ’ (In civilibus dissension-
ibus quamvis saepe per eas res publica laedatur, non tamen . . .)168 should run
‘Though the local authority often suffers harm from civil disturbances.’
‘Ignorant of everything that was in truth’ instead of ‘everything that was true’
or just ‘ignorant of the truth’ (ignorans universa, quae in vero erant)169 is inept.
The author of Opinions has his own slightly offbeat way of expressing ideas.
Documents and proceedings can be inanes (legally worthless).170 Conduct can
be incivile (wrongful)171 and people can act civiliter (rightfully)172 or inciviliter
(wrongfully).173 He is guilty of clumsy and technically inept Latin phrases:
ipsa non habendi necessitate;174 munera quae in homines obiri possunt;175 equi
alieni a sciente possessi;176 curator iuvenis;177 numerus liberorum aut septuag-
inta annorum;178 quae solutioni debitarum ab eo quantitatium profecerunt.179
The author’s heart is in the right place, as is shown by frequent appeals to
equity180 and nature or natural law.181 But Ulpian could not have composed
this work.
The nature of Opiniones shows it to have been composed in a province.
Its content is largely provincial and especially municipal law. The provincial
pro sterilitate temporis boni viri arbitratu in solvenda pensione cuiusque anni pacto comprehensum
est); 50.9.1 (3 opin: medicorum intra numerum praefinitum constituendorum); 50.10.2 (similes
civium in patrias liberalitates); 3.5.33 pr (4 opin: sumptus honeste ad honores per gradus pertinentes
factus)
166
D 1.18.6.2 (1 opin.) cf. 2.14.52.3 (1 opin: produci ad perpetuam praestationem id pactum
postulabatur); 49.18.2.1 (3 opin: onera sollemnia omnes sustinere oportet); 27.9.9 (5 opin: tam cal-
lidum commentum); 8.4.13.1 (6 opin: solitum solacium)
167
D 4.4.40.1 (5 opin.) cf. 1.18.6.9 (1 opin: reprehensa exactorum illicita avaritia); 10.1.8.1
(oculisque suis subiectis locis); 27.9.10 (6 opin: illicite post senatus consultum pupilli vel adulescen-
tis praedio venumdato)
168
D 49.15.21.1 (5 opin.) cf. 1.18.7 (3 op: inspectis aedificiis . . . ea instead of inspecta aedificia)
169 170
D 2.15.9.2 (1 opin.) D 50.8.2.2 (4 opin.); 4.3.38 (5 opin.); 10.4.18 (6 opin.)
171 172
D 4.2.23.3; 50.13.3 (5 opin.) D 4.2.23.3 (5 opin.)
173 174
D 50.13.2 (1 opin.); 4.2.23.2, 3 (5 opin.) D 50.4.4.2 (3 opin.)
175 176 177
D 50.5.2.7a (3 opin.) D 50.13.2 (1 opin.) D 12.1.26 (5 opin.)
178 179
D 50.5.2.1 (3 opin.) D 4.2.23.3 (5 opin.)
180
D 2.14.52.2; 2.15.9.3; 50.13.2; (1 opin.); 50.5.1 pr (2 opin.); 50.8.2.9 (3 opin.); 3.5.44.2 (4
opin.); 10.1.8.1; 10.2.50; 27.9.10 (6 opin.)
181
D 37.15.1.1 (1 opin.); 9.2.50 (6 opin.)
220 10. Spurious Works
182 D 1.18.6 (1 opin., ten times); 47.9.10; 50.13.2 (1 opin.); 50.2.1; 50.4.3.15 (2 opin.); 1.18.7;
50.5.2.7,8; 50.8.2.3, 10; 50.9.1; 50.10.2.1, 2 (3 opin.); 4.2.23.1, 3; 47.13.1 (twice); 27.9.9; 50.13.3 (5
opin.); 10.1.8 pr (6 opin.)
183 D 3.3.74 (2 opin.) 184 D 50.8.2.4,6 (3 opin.) 185 D 49.1.12 (2 opin.)
186 D 37.15.1.2 (1 opin.) cf. above n.63 (praetor urbis) 187 D 2.1.17 (1 opin.)
188 D 50.4.3.1 (2 opin.) 189 D 8.4.13 pr (6 opin.) cf. Gaius, Inst. 2.31
190 Ch.8 n.23–37
191 CJ 10.44.1 (Alex: Veterani, qui, cum possent se tueri immunitate his concessa, decuriones se
fieri in patria sua maluerunt, redire ad excusationem quam reliquerunt non possunt, nisi certa lege
et pacto servandae immunitatis vel partem eius oneris agnoverunt)
192 D 49.18.2 pr (3 opin: Honeste sacramento solutis data immunitas etiam in eis civitatibus apud
quas incolae sunt valet: nec labefactatur si quis eorum voluntate sua honorem aut munus susceperit)
193 D 49.18.5 (Paul 1 cogn.); 50.6.2 (Ulp. 4 off. proc.); CJ 10.44.2 (Dio et Max.); D 50.4.17.1
(Herm. 1 iur. ep.); 50.6.6.13 (Call. 1 cogn); 50.4.4 pr (3 opin.); 50.2.2.8 (Ulp 1 disp.)
194 CJ 10.44.2 (Dio. et Max.) 195 Liebs (1973) 280–5 196 CTh 7.20.2 (Apr 325?)
10. Spurious Works 221
Another Opinions text says categorically that their sex relieves women of
physical public duties (onera corporalia).197 If this was taken to mean that
they are relieved of all personal duties (onera personalia) one would have to
suppose that the text belonged to the period before Philip198 and indeed to
that before Marcus and Verus, who rule that women are liable to personal
duties.199 But while in some contexts personal and physical duties, as opposed
to pecuniary or patrimonial ones, are treated as one and the same,200 in other
contexts such as ill health a physical duty is with good reason treated as some-
thing narrower than a personal duty.201 So understood, the statement that
women are free from physical public duties can be taken to state a permanent
principle, not something progressively eroded. The statement in Opinions202
that the office of ten leading councillors (decemprimatus) in a municipality is
a purely pecuniary and not a mixed personal and pecuniary one reflects a clas-
sification settled at the time of Diocletian’s tetrarchy,203 though earlier in
Diocletian’s reign the office is said to be mixed.204 A law of Constantine of
324 laid down that when a father is excused from a personal burden on the
grounds of having five legitimate children, but one son is of full age, that son
is subject to the personal burden.205 This law is reflected in an Opinions text,
which adds that the reason for the father’s exemption is that his sons will in
due course fulfil the personal burden.206 The fact that the reason is given,
which is unusual, suggests that the law was fresh in the mind of the author of
Opinions. In that case he was writing not long after 324.
The last plausible date for Opinions is 4 August 331, when a law of
Constantine reduced the minimum age for the office of decurion from 25 to
17 years.207 An author so concerned with municipal law as ours would not
overlook this important change. Opinions should therefore be dated to the
period between AD 324 and 331.
The sources used by the author of Opinions have to be gathered from indi-
vidual texts. Most of them form a loose chain of elementary or problemati-
cal rulings on particular facts such as we find in collections of imperial
rescripts. Santalucia rightly points to analogies with the style of rescripts.
197 D 50.4.3.3 (2 opin.) 198 CJ 10.64.1 (Phil.); 10.52.5 (Dio. et Max AA et CC)
199 D 50.1.38.3 (Pap. Iust. 2 const.); 50.1.37.2 (Call. 1 cogn.)
200 D 50.4.1.3 (Herm. 1 iur. epit.); 50.4.18.1 (Arcad. Char. 1 mun. civ.)
201 CJ 10.51.3 (Dio. et Max. AA et CC); CTh 6.35.3.2 (Constantine 27 Apr 319); D 50.5.2.7a
(3 opin.)
202 D 50.4.3.10 (2 opin.)
203 D 50.4.1.1 (Herm. 1 iur. epit: pecuniary); CJ 10.42.8 (Dio et .Max AA et CC: pecuniary)
204 D 50.4.18.26 (Arcad. Char. 1 mun. civ: mixed), Arcadius Charisius being slightly senior to
Opinions contains little in the way of general principle. Very occasionally the
author expressly mentions a rescript (rescriptum est)208 or an imperial con-
stitution (constitutum est) 209 but the second of these seems to be a rework-
ing of a text of Ulpian On the Edict.210 Another Opinions text seems to
derive from a rescript of Marcus and Verus as reported by Scaevola,211 a
third from a rescript of Diocletian.212 Then there is the one, already men-
tioned, that reflects a law of Constantine.213 Many of the others may be
taken from rescripts of the Constantinian period, which continued to be
issued though not on the previous scale.214 No lawyer is expressly cited, but
there is the general statement ‘learned lawyers have taken the view that . . .’
(viris prudentibus placuit).215 There are some parallels to Ulpian texts, but
many more to imperial constitutions, to Pseudo-Paul’s Sentences and to
Hermogenianus’ Summaries of the Law.216
Santalucia is entitled to credit for showing that, contrary to Lenel’s view,217
Opinions does not conform to the edictal or Digest pattern. His positive sug-
gestions about the order that the author followed have been cricitized by
Liebs as arbitrary, and Liebs has put forward an alternative scheme.218 On
any view the work starts with criminal law and the court system, and books
two and three are occupied by municipal law. The central topics of private
law, property and succession, come only in book 6. The law of succession,
which has for centuries occupied centre stage, is now dethroned.
Opinions is a useful but poorly written fourth-century work, one of the last
items of private legal writing to scrape into the canon by dint of being attrib-
uted to a famous author.
5. Replies (2 books)219
The two books of Replies (Responsa) attributed to Ulpian have been gener-
ally regarded as consisting of excerpts from genuine replies by the jurist. Jörs
says they are taken from Ulpian’s private archive.220 Schulz argues that the
collection is post-classical, but remarks that the editor does not seem to have
altered the substance of the law as stated by Ulpian.221 A text in the Vatican
Fragments corresponds closely to a surviving excerpt from Replies in the
208 209
D 2.14.52.3 (1 opin.) D 5.2.29 pr (6 opin.)
210
D 49.1.14 pr (Ulp 14 ed.); Liebs (1973) 298
211
D 5.2.29.2 (6 opin.) cf. 2.15.3 pr (Scae. 1 dig.)
212
D 50.4.3.5 (opin.1) cf. CJ 10.32.5 (Dio. et Max. 286)
213
D 50.4.3.6 (2 opin.) cf. CTh 12.17.1 (19 Jan 324)
214 215
Frag. Vat. 32–4, 36; P. Krüger (1912) 324f. D 50.1.6.2 (2 opin.)
216
Liebs (1973) 30090 217
Lenel (1889) 2.10012 218
Liebs (1973) 301–8
219
Lenel (1889) 2.1016–19; Ind. Flor. 24.17; Jörs (1903) 1438, 1446; Fitting (1908) 116;
Wieacker (1960) 384f.; Schulz (1946) 241; (1961) 307; Honoré (1982) 113–17, 128: Liebs (1997)
§424.26
220 221
Jörs (1903) 5.1446 Schulz (1946) 241; (1961) 307
10. Spurious Works 223
Digest.222 The Vatican Fragment collection was first put together about AD
320.223 Hence the two books of Replies attributed to Ulpian must have existed
by the first half of the fourth century.
There is no doubt that Ulpian gave formal replies, three of which are
attested.224 In discussing a case about trusts Paul says that a ‘reply of
Domitius Ulpianus was read out’.225 Aemilius Macer mentions another.226 A
rescript of Alexander refers to a third.227 There are, however, serious obsta-
cles in the way of accepting that the thirty-two short fragments attributed to
this work in the Digest are Ulpian’s. The fragments do not contain any phrase
typical of his style. They are arranged according to no known system. They
often mention the name of the person who consulted him in the form ‘He
replied to Aurelius Felix etc.’, giving the name of the consultant followed by
the substance of the reply in the accusative and infinitive.228 This feature is
not found in any other collection of Replies of the imperial age. In a substan-
tial number of texts, however, the editor or Justinian’s compilers have left out
the name of the consultant and kept just the accusative and infinitive.229
There are three texts in which Ulpian is reported as writing ‘I replied that he
could have an action’ (habere posse respondi) or something similar.230 Then
there a few statements of the law that are not in the form of a reply at all.231
Mention of the person consulting the lawyer makes these Replies unlike
any other surviving collection. The normal way of recording an author’s
replies is Modestinus respondit, Paulus respondit, or just respondit, without
any mention of the person to whom the reply is addressed. The editor seems
to be treating these replies as if they were like rescripts of the emperor, which
incorporate the name of the petitioner. Could they, all the same, have been
edited from genuine replies found in Ulpian’s papers? 232
Even this suggestion is not easy to accept. An editor can abbreviate the text
he is editing, but can hardly introduce expressions not in the original. A text
from the first book of Replies refers to societas universarum fortunarum, a uni-
versal partnership.233 This is what Ulpian in On the Edict calls a societas
omnium bonorum,234 the partner being a socius omnium bonorum.235 Societas
222 Frag. Vat. 44 (Ulp. 2 resp.)
223 Mommsen, Collectio 3.1–106; Seckel–Kübler (1908–27) 2.2.191–324; Wenger (1953)
543–5; Raber (1965); Gaudemet (1979) 75f.; Liebs (1987b) 150–62; (1989) §506
224 Above n.5 225 D 19.1.43 (Paul 5 quaest: lectum est responsum Domitii Ulpiani)
226 D 50.5.5 (Macer 2 off. praes.) 227 CJ 8.37.4 (31 March 222)
228 D 30.120 = Frag. Vat. 44 (2 resp: respondit Aurelio Felici . . .) cf. 17.2.73; 22.3.30; 32.68 pr;
45.2.8; 49.14.33 (1 resp.); 10.2.53; 11.8.4; 15.4.3; 23.3.51; 24.3.37; 30.120 pr, 1; 39.5.16; 40.11.1;
43.26.20; 49.1.13 pr, 1 (2 resp.)
230 D 27.4.5 (1 resp.); 27.1.23 pr, 1 (2 resp.)
231 D 2.15.10; 32.68.3 (1 resp.), 24.1.40 (2 resp.) 232 Liebs (1997) §424 p.185–6
233 D 17.2.73 (1 resp.)
234 D 17.2.5 pr (3 ed.); 17.2.52.16, 17, 18 (31 ed.); 47.2.52.18 (38 ed.); 42.1.16 (63 ed.)
235 D 17.2.52.16, 17, 18 (31 ed.); 47.2.52.18 (38 ed.); 42.1.16 (63 ed.)
224 10. Spurious Works
universarum fortunarum, on the other hand, is not found in any other legal
writer or imperial constitution. Indeed Ulpian nowhere uses the plural for-
tunae, though Scaevola does.236
In the second book of Replies the word persolvere is twice used for solvere,
‘to pay’.237 Apart from these texts, Ulpian uses persolvere only twice. In one
text persolvere means not ‘to pay’ but ‘to pay back’. A seller may have a choice
to forgo the price or to take back the thing sold and pay back the price.238 The
other text has plainly been altered by the compilers. Two people go surety for
ten. The debtor pays three, and then the sureties pay five each. The surety who
pays last can recover three, because if the debtor pays three only seven remain
due. If these are paid three have been paid that were not owing.239 As the text
stands ‘if these are paid’ (quibus persolutis) refers to the remaining seven being
paid, whereas the sense of the text requires that ten should be paid in addition
to the three paid by the debtor. So we cannot attribute to Ulpian the use of
persolvere for ‘to pay’. On the contrary, he prefers shorter words. Solvere,
which he uses in 504 texts,240 is shorter than persolvere. And why, if Ulpian
used solvere, should an editor change it to persolvere?
The Replies have a sentence beginning 241 and two ending 242 with ablative
absolutes, a construction that Ulpian avoids at the beginning and end of sen-
tences, though it would be going too far to say that he never resorts to it. But
he certainly never uses the ablative absolute termination with salvo or salva to
express a proviso, as in two texts from Replies.243 On the other hand Scaevola
ends a reply with a proviso of this sort.244 A text in Opinions book 2 has ‘there
is no reason why’ (nihil proponi cur), an expression otherwise absent from
Ulpian but found 38 times in Scaevola245 and 9 in Modestinus.246
Though Ulpian undoubtedly gave written replies to those who consulted
him, he does not seem to have regarded this as an important part of his work
as a lawyer. Apart from the three texts from Replies already mentioned,247 he
does not use ‘I replied’ of himself, whereas other writers do so, such as
Scaevola, who has 112 texts with respondi,248 Paul, who has 53 texts with
respondi,249 and Modestinus, who has five.250 They do this mainly in works
other than their Replies, which, in the published form, are normally couched
in the third person (respondit). Ulpian once uses ‘I wrote back’ (rescripsi) of
236 D 32.34.1 (16 dig: pater exiguarum fortunarum) 237 D 43.26.20 (2 resp.)
238 D 38.5.1.13 (44 ed: utrum malit de pretio remittere an potius rem quam vendidit recipere per-
soluto pretio)
239 D 12.6.25 (47 Sab: quia tribus a reo solutis septem sola debita supererant, quibus persolutis
Legal authors could then no longer write in their own name. They needed
a borrowed identity. Legal literature fades out with these summaries and epit-
omes, of variable quality but all misattributed to classical writers so as to
attract a wider sale or greater authority in court. In an age when the canon of
legal works was being settled Ulpian’s name carried weight.
11
Epilogue
A L E A D I N G writer and lawyer in government, Ulpian was famous in his own
age. His pupil Modestinus calls Cervidius Scaevola, Paul, and Ulpian the
leaders among lawyers,1 but gives Ulpian a special pre-eminence. A ruling of
Severus and Caracalla, he says, laid down that to be the curator of an insane
person counted towards being excused from further such offices and ‘the
excellent Ulpian said the same about three guardianships’.2 Modestinus uses
the phrase ‘the excellent Ulpian’ ( κρτιστος Ολπιανς) in two other texts.3
It is not used of any other lawyer. It is the tribute of a loyal pupil. A rescript
composed by Arcadius Charisius in the time of Diocletian is to much the same
effect.4 Town councillors and their children should be free from torture, as
‘the most learned Domitius Ulpianus in his book on public disputations
relates as a permanent item of learning’.5 Arcadius also cites ‘what Herennius
Modestinus with very good reason laid down in his notes and in public dis-
putation’, viz. that the public obligation of the ten or twenty leading members
of a municipal council to collect taxes is a mixed personal and patrimonial
obligation.6 These are the two predecessors whom Arcadius in the tetrarchy
cites in flattering terms.
In the late third century Ulpian was seen, then, as the source of a line of
authority, especially on public law, carried forward by Modestinus and
Arcadius. Both of them probably and Arcadius certainly held the office of
secretary for petitions7, as Ulpian had done. The three share a balanced, prag-
matic approach to the solution of legal and administrative problems. Further
evidence of Ulpian’s standing, probably from the tetrarchic period,8 comes
from an inscription in Ephesus. In it a proconsul urges the city to collect
together ‘things said on the basis of ancient laws in Ulpian’s De officio’,9 in
imperial constitutions, and in resolutions of the senate.10
1 D 27.1.13.2 (4 excus: ο κορυαοι τν νοµικν)
2 D 27.1.2.9 (2 excus: κρτιστος Ολπιανς)
3 D 26.6.2.5 (1 excus.); 27.1.4.1 (2 excus.)
4 Honoré (1994) 160–2 esp. 161300; Liebs (1989) §508.1
5 CJ 9.41.11.1 (27 Nov 290: vir prudentissimus Domitius Ulpianus in publicarum disputationum
ratione decrevit)
7 Honoré (1994) 104–7, 190–1 8 Robert (1967) 46
9 τ τε κ τν παλαιν νµων ν τος ∆ ι[κω παρ’ ] Ολπιαν ερηµ!να
10 JÖAI 45 (1960) Beibl. 82 no.8; AE 1966.436; IK Ephesos 2 (1979) no.217; Millar (1986) 279
228 11. Epilogue
11 D 1.16.4.5 (1 off. cons: ingress<ur>um etiam hoc observare oportet, ut per eam partem
provinciam ingrediatur, per quam ingredi moris est, et quas Graeci πιδηµας appellant sive
κατπλουν observare, in quam primum civitatem veniat vel applicet: magni enim facient provinciales
servari sibi consuetudinem istam et huiusmodi praerogativas)
12 Liebs (1997) 179 13 Church Teaching (Div. inst.) 5.11.19
14 CTh 1.4.1 (28 Sept 321/4: Perpetuas prudentium contentiones eruere cupientes Ulpiani ac
Pauli in Papinianum notas, qui, dum ingenii laudem sectantur, non tam corrigere eum, quam
depravare maluerunt, aboleri praecipimus)
15 CTh 1.4.3 (7 Nov 426)
11. Epilogue 229
decided by counting opinions. A lesser lawyer is sometimes the only one to get
the law right.16
When Tribonian applies these principles to the Digest material Ulpian
emerges as the author whose account is most often chosen as the best.
Excerpts from his work account for just over 40 per cent of the Digest, more
than those of any other author. But this figure gives too modest an idea of
Ulpian’s impact on the law. In each sub-chapter (title) the Digest compilers
normally put Ulpian’s exposition of the law first. Sixty per cent of the Digest
titles17 begin with an Ulpian text. From this point of view he dominates all the
other thirty-eight authors combined from the beginning to the end of the
great compilation. The result is that law students, whether in the late Roman
or Byzantine empire or the European middle ages and Renaissance began
their study of almost any legal topic with Ulpian. Nor was his role in legal
education merely the result of Justinian’s Digest. The Sinai Fragments
(Scholia Sinaitica),18 which seem to belong to the fifth or early sixth century,
consist of Greek comments, presumably deriving from an eastern law school,
on books 36–9 of Ulpian’s On Sabinus, part of the second-year law syllabus.
By a process of osmosis Europe’s view of the law has been formed more by
Ulpian than by any other lawyer. This is true as regards substance, style,
method of reasoning, and background philosophy. Papinian and Julian dis-
play greater subtlety, Labeo and Celsus are more inventive, but it was Ulpian
who expounded Roman law as a universal system capable, as it turned out, of
being adapted to the needs of the radically different societies that emerged
from the breakdown of empire.
16
C Deo auctore 6 (15 Dec 530: sed neque ex multitudine auctorum quod melius et aequius est
iudicatote, cum possit unius forsitan et deterioris sententia et multos et maiores in aliqua parte
superare)
17
Out of 432 titles 260 begin in this way
18
Girard–Senn (1967) 591–604; FIRA 2.637–52
TABLE I
List of Latin Words and Phrases Referred to in the Text
and Footnotes
References in the form 2107 are to chapters and footnotes. When the word or phrase is
not the subject of a footnote, the reference takes the form of citing the chapter and the
footnotes preceding and following the word or phrase, for example 745–6.
5.12.11 2627
DAUBE, D. (1973). ‘The Compilers’ Use of a Revised Paul and Ulpian’: SZ 90: 359–60.
DE FILIPPI, M. (1984). ‘Il titolo “de excusationibus” dei “Vaticana Fragmenta” ’, in
Sodalitas. Scritti A. Guarino 3: 1159–79.
DE GIOVANNI, L. (1989). Giuristi severiani. Elio Marciano.
DE MARINI AVONZO, F. (1982). Review of Honoré (1982): Iura 33: 194–201.
DE VISSCHER, F. (1961). ‘La constitutio Antoniniana et la dynastie africaine des
Sévères’: RIDA 8: 229–42.
DELL’ORO, A. (1960). I libri de officio nella giurisprudenza romana.
DIERAUER, U. (1977). Tier und Mensch im Denken der Antike.
DISDI, G. (1971). ‘Das Gespenst der Praedigesten’: Labeo 17: 187–192.
DITTENBERGER, W. (1903). ‘Athenaeus und sein Werk. Apophoreton’: Vers. deut. Phil.
und Schul. 47: 21.
DOWNEY, G. (1961). A History of Antioch in Syria.
DU TILLET, J. (Paris, 1549). Tituli XXVIIII ex corpore Ulpiani.
DURRY, M. (1938). Les Cohortes prétoriennes.
DUšANIć, S. (1964). ‘Severus Alexander as Elagabal’s Associate’: Hist. 13: 487–98.
EBRARD, F. (1917). Die Digestenfragmente ad formulam hypothecariam und die
Hypothecarrezeption.
ECK, W. (1974). ‘Beförderungskriterien innerhalb der senatorischen Laufbahn,
dargestellt an der Zeit von 69 bis 138 n. Chr.’: ANRW 2.1: 158–228.
ECKARDT, B. (1978). Iavoleni epistulae.
EDELSTEIN, L. (1966). The Meaning of Stoicism.
EISSFELDT, O. (1939). ‘Tyros (2)’: RE 2.7: 1876–1908.
ELLEGÅRD, A. (1962). A Statistical Method for Determining Authorship.
ENSSLIN, W. (1954). ‘Praefectus praetorio’: RE 22.2: 2391–502.
FALCHI, G. L. (1985). ‘Sui “fragmenta Berolinensia incerti auctoris de iudiciis” ’:
SDHI 51: 189–214.
FALCONE, G. (1997). ‘D 1.3.13: Pedio, Ulpiano e la lex contractus’: Labeo 43: 240–58.
FERRINI, C. (1886). ‘Postille esegetiche ai frammenti del commentario di Ulpiano alle
formule edittale ad legem Aquiliam’: Rend. Ist. Lomb.2 19: 245–60 = (1929–30) 2:
95–111.
—— (1901). ‘I commentari di Ulpiano e di Paolo ad legem Iuliam et Papiam’: Rend.
Ist. Lomb.2 34: 394–405 = (1929–30) 2: 237–49.
—— (1929–30). Opere, 5 vols.
FIRA (1940–3, 1972). = Fontes Iuris Romani Anteiustiniani (3 vols. ed. S. Riccobono,
G. Baviera, and V. Arangio Ruiz).
FITTING, H. (2 ed. 1908). Alter und Folge der Schriften römischer Juristen von Hadrian
bis Alexander.
FLEMING, W. B. (1915). The History of Tyre.
FLUSS, M. (1923). ‘Septimius (32)’: RE 2: 2434–53.
FREZZA, P. (1968). ‘La cultura di Ulpiano’: SDHI 34: 363–75.
—— (1983). ‘La persona di Ulpiano’: SDHI 49: 412–22.
FRIER, B. W. (1984). ‘Law on the Installment Plan’: Michigan Law Review 82: 856–68.
GARNSEY, P. (1967). ‘Adultery Trials and the Survival of the Quaestiones in the
Severan Age’: JRS 57: 56–60.
—— (1970). Social Status and Legal Privilege in the Roman Empire.
GAUDEMET, J. (1967). Institutions de 1’Antiquité.
286 Bibliography
GAUDEMET, J. (2 ed. 1979). La Formation du droit séculier et du droit de l’Église aux IVe
et Ve siècles.
—— (1987). ‘Les droits de l’homme ont-ils été reconnus dans l’empire romain?’:
Labeo 33: 7–23.
—— (1998). ‘Le monde antique et les droits de l’Homme. Quelques observations’, in
H. Jones (1998) 175–83.
GEFFCKEN, J. (1931). ‘Meleagros (7)’: RE 15.1: 478–88.
GIANELLI, C., and MAZZARINO, S. (1956). Trattato di storia romana.
GILLIAM, F. (1965). ‘Dura Rosters and the Constitutio Antoniniana’: Hist. 14: 74–92.
GILLIAM, J. P., and MANN. J. C. (1976). ‘The Northern British Frontier from
Antoninus Pius to Caracalla’: Arch. Aeliana4 48: 1–44.
GIRARD, P. F., and SENN, F. (7 ed. by F. Dumont 1967). Textes de droit romain.
GIUFFRE, V. (1974a). ‘Arrio Menandro e la letteratura de re militari’: Labeo 20: 27–63.
—— (1974b). La letteratura de re militari.
GONZALEZ, J. (1986). ‘The Lex Irnitana: A New Flavian Municipal Law’: JRS 76:
148–243.
GORDON, W. M. (1984). Review of Honoré (1982): Class. Rev. 98 = 342: 232–4.
GOTHOFREDUS, J. (1653). Novus in titulum Pandectarum De diversis regulis iuris antiqui
commentarius.
GRADENWITZ, O. (1887). Interpolationen in den Pandekten.
GROSSO, F. (1968a). ‘Ricerche su Plauziano e gli avvenimenti del suo tempo’: Att.
Acad. Lincei. Rend.8 23: 7.
—— (1968b). ‘Il papiro Oxy. 2565 e gli avvenimenti del 222–224’: Att. Acad. Lincei
Rend.8 23: 205–20.
GRUPE, E. (1899). ‘Gaius und Ulpian’: SZ 20: 90–8.
GUALANDI, G. (2 vols. 1963). Legislazione imperiale e giurisprudenza.
GULICK, C. B. (7 vols. 1927–41). Athenaeus. The Deipnosophists.
GUNDEL, H. G. (1966). ‘Papyrologisches zur constitutio Antoniniana’: Kurzberichte
aus den Papyrussammlung 22: 8–14.
HAGENDAHL, H. (1971). ‘Die Bedeutung der Stenographie für die spätlateinische
christliche Literatur’: Jb. Ant. Christ. 14: 24–38.
HAMMOND, M. (1940). ‘Septimius Severus, Roman Bureaucrat’: Harv. Stud. Class.
Phil. 51: 137–73.
HANNESTAD, K. (1944). ‘Septimius Severus in Egypt. A Contribution to the
Chronology of the Years 198–202’: Class. et mediaev. 6: 194.
HANSLIK, R. (1939). ‘Valerius (134)’: RE 7A 2411–2.
—— (1959). ‘Publicius 43’: RE 23.2: 1905.
HARTKE, W. (1951). Römische Kinderkaiser. Eine Strukturanalyse römischen Denkens
und Daseins.
HASEBROEK, J. (1921). Untersuchungen zur Geschichte des Kaisers Septimius Severus.
HAY, J. S. (1911). The Amazing Emperor Heliogabalus.
HEBERDEY, R. (1912). ‘Inschriften’: Forschungen in Ephesos 2: 95–203.
HEINECCIUS, J. G. (1765–71). Opera, ed. J. C. Heineccius, 9 vols.
HERRMANN, P. (1972). ‘Überlegungen zur Datierung der constitutio Antoniniana’:
Chiron 2: 519–530.
—— (1997). ‘Die Karriere eines prominenten Juristen aus Thyateira’: Tyche 12:
111–23.
Bibliography 287
WINKEL, L. C. (1993) ‘Einige Bemerkungen über ius naturale und ius gentium’, in Ars
boni et aequi. Festschrift W. Waldstein 443–50.
WLASSAK, M. (1919). Zum römischen Provinzialprozeß.
WOLF, J. G. (1959). ‘D 20.1.27 Marc. 5 dig.’: SZ 76: 520–34.
WOLFF, H. J. (1949). ‘Ulpian 18 ad edictum in Collatio and Digest and the Problem of
the Post-Classical Editions of Classical Works’: Scr. Ferrini 4: 64–90.
—— (1951), ‘Zur Überlieferungsgeschichte von Ulpian libri ad edictum’, in Festschrift
F. Schulz 2: 145–71.
—— (1959). ‘Zur Palingenesie und Textgeschichte von Ulpians libri ad edictum’: Iura
10. 1–12.
—— (1976). Die constitutio Antioniniana und Pap. Giessensis 40.
YARON, R. (1987). ‘Semitismus in Ulpian’: TRG 55: 3–17.
ZECCHINI, G. (1995). ‘La Constitutio Antoniniana e l’universalismo di Roma’, in
L’ecumenismo politico nella coscienza dell’Occidente.
ZIEGLER, R. (1978). ‘Antiocheia, Laodicea und Sidon in der Politik der Severer’:
Chiron 8: 493–514.
Index
Didius Julianus (emperor) 3, 6 Julia Domna 5–7, 17, 21–2, 24, 81–2
Didius Marinus 32 Julia Maesa 5, 27, 30–1
Dio 3, 8, 25, 30–3, 84, 170, 187, 209 Julia Mammaea, see Mammaea
Diocletian (emperor) 9, 221–2 Julia Paula 29
Domitian (emperor) 81, 154 Julia Soaemias, see Soaemias
Domitius 13, 35 Julian (emperor), see Didius
Domitius Honoratus 32 Julian (lawyer) 16, 91, 98, 100, 102, 128–34,
Domitius Ulpianus, see Ulpian 136–7, 139, 141–3, 146–50, 153, 155,
Du Tillet 207 177, 202, 229
Julius Caesar 154, 225
Egypt 10, 15, 18, 30–2 Julius Flavianus, see Flavianus
Elagabal (emperor) 6–8, 11, 26–30, 33, 35–6, Julius Paulus, see Paul
165–6, 203 Juventius Celsus, see Celsus
Emesa 7, 11, 27 Justinian’s compilers viii, 19–21, 23, 27,
Epagathus 32, 36 40, 43–4, 105–6, 124, 131, 134–5,
Ephesus 98, 209, 227 140–1, 144, 150–1, 155, 171, 213,
Euphrates 79 223, 229
Eutropius 8, 18, 215
Labeo 85, 91, 128–30, 132–5, 137, 142–3,
Festus 8, 18, 215 145–53, 179, 211, 229
Flavianus 30–2, 35–6 Lactantius 8, 225
Florentinus 88 Laetus 6
Fulcinius Priscus 134, 144, 147–9 Laodicea 10
Larensis 12
Licinius (emperor) 220
Gadara 9
Licinius Rufinus 33, 207, 211, 218
Gaius 40, 44, 72, 79, 84, 134–7, 153, 156,
Lorenius Celsus 32
207–9, 217, 228
Lydia 25
Galatia 10, 17
Galba (emperor) 31
Macer 161, 223
Galen 1, 12, 77, 89, 225
Macrinus (emperor) 4, 11, 14, 22, 26–7, 31,
Gallus Aquilius 145, 147
35, 158, 165–6, 170–2, 179, 181, 185
Gaul 14, 17
Maecenas 84
Gellius 83
Maecianus 130, 134, 139–40
Geminius Chrestus, see Chrestus
Mammaea 5, 7, 24, 27, 30–1, 34, 36, 83, 190
Gessius Marcianus 24
Marcellus 23, 41, 55, 59, 124, 128–32, 134,
Geta (emperor) 2–3, 6, 21–3, 25, 35
136–7, 147–8, 156, 170, 202–3
Gordian III (emperor) 154
Marcianus 17, 49, 72, 75, 80, 83, 88, 91, 124,
Greece 85
129, 134, 136–8, 152–4, 161, 183,
Gregorius viii
211
Gregory (wonder-worker) 83
Marcus Aurelius (emperor) 1–2, 6, 21, 23, 27,
Grotius vii, 35
59, 80, 89, 139, 154–6, 164, 169, 177–8,
191, 195, 209, 221–2
Hadrian (emperor) 10, 45, 86, 91, 149, 154, Marius Maximus 8
160, 184 Masurius Sabinus 130, 134, 141, 143, 145–50,
Herennius Modestinus, see Modestinus 211
Hermogenianus vii, 28, 206, 222 Mauricianus 148, 150, 210
Herodian 3, 8, 10, 23, 34 Maximian (emperor) 9
Hippolytus 7 Mela 147–9
Homer 90 Meleager 9
Honorius (emperor) 33 Menander 22–4, 152, 160
Messius 4
Iavolenus Priscus 10, 45, 55, 124, 129, 145–6, Minicius 131, 147
150–1, 153 Modestinus 4, 8, 17, 30, 33, 35, 136, 153–4,
Ilium 209 161, 207–8, 212, 215, 224–5, 227–8
Irnerius vii, 35 Moses 207
Italy 4, 6, 17 Mucius 130, 134, 142, 145, 147
Index 299
Valerius Comazon, see Comazon Vivianus 131, 134, 140, 144–5, 147, 149
Venuleius Saturninus 130, 142–3
Verus (emperor) 23, 154–5, 177, 191, 221–2 Xenophon 90
Vespasian (emperor) 141, 150
Victor 8, 28–9, 225 York 2, 22
Vindius Verus 150
Virgil 225 Zonaras 8
Vitellius 141–2, 145–7 Zosimus 8, 31