1
Koenraad Verboven
The Sulpicii from Puteoli and usury in the early Roman Empire
in:
Tijdschrift voor Rechtsgeschiedenis / Revue d’Histoire du Droit / The Legal History Review
71 (2003), p. 7-28
This is a prepublication version
For reference purposes please refer to the published version (available on line at:
http://booksandjournals.brillonline.com/content/10.1163/157181903100362490
for off-prints feel free to contact me at Koen.Verboven@ugent.be
The writing-tablets discovered in 1959 in the Agro Murecine not far from Pompeii are famous enough
not to need any further introduction1. Yet, in many aspects they remain as mysterious as ever. Who
were the Sulpicii who wrote and kept them – wealthy independent businessmen or simple front men
for the imperial court and the Italian aristocracy? What were they - argentarii or faeneratores ? How
were they organised 2? Why are there no tablets recording interest stipulations ? The scope of this
paper is inevitably more limited. I here propose to analyse the question of how the seemingly
gratuitous nature of the loans in the tablets can be explained from a legal point of view.
Interest regulations and realities.
The history of Roman interest regulations is complicated and goes back to at least the fourth century
BCE, when a long series of successive debt crises resulted in the complete interest prohibition by the
lex Genucia in 342 and the abolition of debt-bondage in 328 by the lex Poetelia. Somewhere in the
third or perhaps even as late as the early second century BCE, a lex Marcia confirmed or reintroduced
the interest prohibition3.
Traditionally, Roman laws were not formally repealed, but they could become obsolete. By the end of
the second century, when the formulary procedure was well in place, the enforcement of the old
interest laws was in the hands of the praetors. In 89 BCE the praetor Sempronius Asellio was
murdered by the faeneratores because he allowed debtors to proceed against their creditors. More than
anything else the event shows that it had become highly unusual for praetors to grant actions against
usurers4. The consuls of the following year, Sulla and Pompeius Rufus, carried a lex unciaria,
1
For a new full edition see now G. Camodeca, Tabulae Pompeianae Sulpiciorum. Edizione critica dell'archivio
puteolano dei Sulpicii, Roma, 1999 (Vetera 12). All references in this article to the tablets refer to this edition
(TPSulp.). See also here for further bibliography concerning the history of the tablets.
2
On these questions see K. Verboven, L’organisation des affaires financières des C. Sulpicii de Pouzzoles
(Tabulae Pompeianae Sulpiciorum), in : Cahiers Glotz, 11 (2000), p. 161-171. K. Verboven, The Sulpicii from
Puteoli, argentarii or faeneratores ? in : Hommages à Carl Deroux, Bruxelles, in print.
3
Gaius, 4,23. For a history of Republican interest regulations see Ch. T. Barlow, Bankers, Moneylenders, and
Interest Rates in the Roman Republic, Michigan University Microfilms International, 1978, passim; G. Tilli,
Postremo vetita versura, in: Bolletino dell'istituto di diretto romano, 86-87 (1984), p. 147-163; Y. Rivière, Les
quadruplatores: la répression du jeu de l'usure, et de quelques autres délits sous la république romaine, in:
Mélanges d'archéologie et d'histoire de l'Eco
le Française de Rome. Antiquité, 109 (1997), p. 577-631; A. Storchi Marino, Quinqueviri mensarii: censo e
debiti nel IV secolo, in: Athenaeum 81 (1993), p. 213-250; A. Pikulska, Problème de l’usure en rome
républicaine. Favor debitoris?, in: Orbis Iuris Romani., 6 (2001), p. 124-150.
4
Appianus, Bella Civilia 1,54 ; Livii Periochae 74; Valerius Maximus 9,7,4.
2
probably attempting to revive the archaic faenus unciarium5. Late in 51 BCE the senate decreed that
interest would be reduced to one percent a month (centesima) and that compound interest
(anatocismus) would not be allowed6. The limit seems to have remained in force under Caesar7.
In the Early Empire, interest regulations were governed by three basic principles. Firstly, interest was
limited to a maximum of 1% a month (usura centesima). Secondly, compound interest (usurae
usurarum) was not permitted. Thirdly, the total sum of interest paid by the debtor to his creditor could
not exceed the original amount of the loan (usurae supra duplum). The origins and codification of
these three principles, however, and their practical effectiveness is far from clear8. The first two
principles probably go back to the senatus consultum of 51 BCE9. Allthough the first unambiguous
evidence of the usura legitima / centesima under the Empire dates only to the mid second century10,
Plutarch’s De vitando aere alieno implies a legal limit on interest rates by at least the late first or early
second century11. On the other hand, the rule that a debtor could not be compelled to pay more than
double the sum he had borrowed is attested only from the early third century BCE onwards and may
be a Severan innovation12.
A multitude of texts show that money-lending at interest and usury were quite common in the Early
Empire. Thus, Trimalchio states in a matter of fact way that after accumulating a fortune, he withdrew
from trade and started to < per> libertos faenerare13. Persius suggests that money could easily and
5
Festus (ed. Lindsay), p. 516, s.v. unciaria lex. Tacitus (Annales 6,16) ascribes the archaic faenus unciarium to
the XII Tables, but it may have been introduced only by a lex Menenia Duillia in 357 BCE (Livy 7,16,1). See M.
H. Crawford (ed.), Roman Statutes. Vol. 2, London, 1996, p. 686-687 (Bulletin of the Institute of Classical
Studies, Supp. 64). On the faenus unciarium in general see H. Zehnacker, Unciarium fenus (Tacite, Annales,
VI,16), in: Mélanges de littérature et d’épigraphie latines, d’histoire ancienne et d’archéologie. Hommages à la
mémoire de Pierre Wuilleumier, Paris, 1980, p. 353-362; C. Appleton, Contribution à l'histoire du prêt à intérêt
à Rome. Le taux du fenus unciarium, in: Nouvelle Revue historique de droit français et étranger, 43 (1919), p.
467-543.
6
Cicero, Ad Atticum, 5,21,13 ; A. Pikulska, Anatocisme. C. 4,32,28,1: Usuras semper usuras manere, in: Revue
Internationale des Droits de l’Antiquité, 3e s. 44 (1998), p. 440-441 ; G. Billeter, Geschichte des Zinsfusses im
griechisch-römischen Altertum bis auf Justinian, Leipzig, 1898, p. 177. In the course of 49 BCE the tribunes
again intervened (Dio 61,37,2). Note, however, that Cicero (Ad Atticum 1,12,1) seems to imply that the
centesima usura was at least the ethical if not the legal maximum already in 61 BCE.
7
This is implied by Suetonius (Divus Iulius 42,3) who claims that when Caesar remitted all interest accrued
since the beginning of the war, late in 48 or early in 47, the creditors lost about a quarter of their money. See also
M. W. Frederiksen, Caesar, Cicero and the Problem of Debt, in: Journal of Roman Studies, 56 (1966), p. 134.
8
The earliest known example of a statute containing these three principles may be found in Lucullus’s provincial
edict for Asia promulgated ca. 70 BCE (Plutarch, Lucullus 20,3). But this was merely a temporary measure to
solve the devastating debt crisis cause by the enormous fine of 20,000 talents imposed by Sulla on the province
for its support of Mithridates. Cf. L. Solidoro, Ultra sortis summam usurae non exiguntur, in: Labeo, 28 (1982),
p. 165-168 ; Pikulska, Anatocisme (n. 6), p. 437-438 ; L. Migeotte, L'emprunt public dans les cités grecques.
Recueil des documents et analyse critique, Québec & Paris, 1984, p. 340-341.
9
Although strangely Tacitus (Annales 6,16) makes no mention of the usura centesima or the prohibition on
compound interest and seems to be implying that in 33 CE charging interest was still (in theory) illegal.
Pikulska’s theory (Anatocisme (n. 6), p. 441) that ‘postremo vetita versura’ would mean ‘a prohibition on
compound interest’ and refer to the senatus consultum of 51 is unconvincing in the light of Tacitus’s narration
(first faenus unciarium, then faenus semunciarium, finally vetita versura). On the presumed technical meaning of
versura as ‘a loan carrying compound interest’, on which Pikulska’s intepretation is based, see K. Wille, Die
Versur, Berlin, 1984. But contra (in my view convincingly) H.-P. Benöhr, Versura, in : Zeitschrift der SavignyStiftung für Rechtsgeschichte, romanistische Abt., 107 (1990), p. 216-248.
10
Dig. 45,1,90 (Pomponius) ; Gnomon of the Idios Logos §105 ; P. Fouad I, no. 45 anno 153 (= FIRA 3, no.
121).
11
Plutarch, Moralia (De vitando aere alieno) 829 C. See also infra.
12
Cf. Solidoro, Ultra sortis (n. 9), p. 169 ; Dig. 22,1,9,pr. (Papinianus) Dig. 12,6,26,pr. (Ulpianus); Dig. 22,1,29
(Marcianus); Dig. 22,2,4 (Papinianus); C. 4,32,10 (Caracalla); CTh. 4,19,2 (Gratianus & Theodosius, 380 CE) ;
Interpretatio CTh. 4,19,1.
13
Petronius 76,9. Note how the decision echoes the Sulpicii’s status as freedmen / middlemen and seems to
suggest that ‘professional’ faeneratio in Campania in the days of Nero was largely in the hands of freedmen.
3
safely be invested in loans at 5%, no doubt through professional of semi-professional middlemen14.
Pliny the Younger admitted that although most of his wealth was in land some was invested in loans at
interest15.
Usury doesn’t seem to have posed a serious social or political problem in the first two centuries of our
era. The only serious disturbance we know of is the Tiberian crisis of 33 CE, described in some detail
by Tacitus, which vanished when the emperor put 100 million sesterces worth of free loans at the
disposal of creditworthy debtors in need of cash16. It was only at the time of the Severan emperors that
governement again set out a policy to combat usury. This is clearly recognisable in the Corpus Iuris
Civilis, both in the Codex and in the Digest, where we find a landslide of imperial constitutions dating
to the Severan age concerning illegal interest rates17.
Nevertheless, during the entire period from the first Julio-Claudian emperors to the last of the
Antonines, faeneratores continued to be despised as greedy usurers and to be accused of illegal
practices18 Tacitus complains that although faeneratio had time and again been prohibited in Rome,
the practice kept re-emerging and he asserts that at the time of the Tiberian debt crisis not a single
senator was not deeply involved in usury19.
The Sulpician tablets, dating mostly to the years just after the Tiberian crisis, give us a glimpse of how
this was possible. Several tablets show that the Sulpicii acted as intermediaries, sometimes – as
independent faeneratores – borrowing in their own name and subsequently lending the money at their
own risk20, sometimes – as proxenetae – merely assisting in drawing up and safe-keeping the
necessary documents21.
14
Persius 5,149-150.
Pliny, Epistulae 3,19,8.
16
Tacitus, Annales 6,16-17 ; Suetonius, Tiberius 48; Dio, 58,21,4-6 ; H. Bellen, Die Krise der italienischen
Landwirtschaft unter Kaiser Tiberius (33 n. Chr.), in : Historia, 25 (1976), p. 217-234 ; M. K. Thornton & R. L.
Thornton, The financial crisis of AD 33: A keynesian Depression ?, in : Journal of Economic History, 50 (1990),
p. 655-662 ; R. Wolters, Die Kreditkrise des Jahres 33 n. Chr., in : Litterae Numismaticae Vindobonenses, 3
(1987), p. 25-58 ; C. Rodewald, Money in the age of Tiberius, Manchester, 1976, p. 1-17.
17
Cf. Solidoro, Ultra sortis (n. 9), p. 169 ; P. Gröschler, Banchieri e limite delle usurae, in : G. Crifo & S. Giglio
(cur.), Atti dell’Academia Romanistica Constantiniana XII Convegno Internazionale in onore di Manlio
Sargenti, Napoli & Roma, 1998, p. 345-352.
18
cf. e.g. Valerius Maximus 4,8,3; Martial 2,44 ; Seneca, Controversiae 9,1,12; Seneca, De Beneficiis 7,10;
Lucanus 1,181-182: hinc usura uorax auidumque in tempora fenus; Columella praef. 8; Tacitus, Annales 6,16 ;
Plutarch, Moralia (De vitando aere alieno, passim) 827D-832A. See the epitheta in the Thesaurus Linguae
Latinae for some more examples (Banier, TLL 6, 1912-1926, art. fenerator, col. 475 r. 45-50, art. fenus, col. 484 r.
70-84 - col. 485 r. 1-2).
19
Tacitus, Annales 6,16 : multisque plebi scitis obviam itum fraudibus quae toties repressae miras per artes
rursum oriebantur … sed tum Gracchus praetor, cui ea quaestio evenerat, multitudine periclitantium subactus
rettulit ad senatum, trepidique patres neque enim quisquam tali culpa vacuus veniam a principe petivere.
20
TPSulp. 69 (debitum in stipulatum deductum) ; 75 ; 76 ; 82 ; (apochae) 94 ; 95 (fragmenta rationum). For this
practise see K. Verboven, Le système financier à la fin de la république romaine, in: Ancient Society, 24 (1993),
p. 90-91 ; Suetonius, Augustus 39; Donatus, ad Terentii Phormionem 778-783 : Dicitur etiam versuram facere,
qui cum minore faenore acceptam pecuniam maiore occupat ; Martial 2,44 ; Ps.-Acro, in Horatii Sermones 2, 3,
18: Iani statuae tres erant; ad unam illarum solebant convenire creditores et feneratores, alii ad reddendum alii
ad locandum fenus. For faeneratores known to have practised this kind of mediation cf. Atticus (Cornelius
Nepos, Atticus 2,4-5) ; Rabirius Postumus (Cicero, Pro Rabirio Postumo 5; 25 ; 39) ; P. Sittius (Cicero, Pro
Sulla 58) ; Cluvius and C. Vestorius (Cicero, Ad Atticum 6,2, 3); Egnatius Rufus (Cicero, Ad Atticum 12,18, 3;
19, 2).
21
See mainly the dossier on Novius Eunus (TPSulp. 45 ; 51 ; 52 ; 67 ; 68) where Sulpicius Faustus first appears
as witness (TPSulp. 51), later as adiectus solutionis causa (TPSulp. 67 ; 68). Note also the numerous documents
in which no Sulpicius is mentioned, but that were obviously in safe-keeping with them. On proxenetae see
Seneca, Epistulae ad Lucilium 119,1; Dig. 50,14 (Ulpianus); H. Siber, Operae liberales, in : Iherings Jahrbücher
für Dogmatik der römischen und deutschen Privatrechts, 88 (1939-1940), p. 177-179; J. Michel, Gratuité en
droit romain, Bruxelles, 1962, p. 193-195, 533 ; J. Andreau, La vie financière dans le monde romain. Les
15
4
How astonishing, therefore, to find no trace of interest clauses in the Sulpician documents. The tablets
seem to pretend that the Sulpicii generously extended only gratuitous loans. Yet the numerous
documents of trials and arbitrations22 and the many tablets concerning the public sale of goods and real
securities of insolvent debtors23 tell a different story. There is hardly any doubt possible that the loans
extended by and through the Sulpicii were interest bearing. But how?
Nature of the stipulationes in the Sulpician tablets: stipulationes purae.
The stipulationes attested in the tablets of the Sulpicii, like their counterparts among the Tabulae
Herculanenses, are mostly stipulationes purae; i.e. the tablets merely record that a debtor had agreed
by stipulatio to repay his loan. Interest obligations are never mentioned and expiry dates are added in
only three or four cases24. This seems quite extraordinary compared to what we know from other
sources. The examples in the Digest suggest that condiciones were standard practice in the case of loan
stipulations. The loan stipulations attested in the Transsylvanian tablets invariably mention capital
sum, interest rate and expiry date25. A recently discovered wooden tablet from Vindonissa dating from
90 CE recording a stipulatio mutui dationis is very similar to the Transsylvanian examples26.
Yet, the Campanian cautiones are not wholly without parallels in other sources. A number of cases
adduced in the Digest confirm that interest was not alway stipulated. Thus Paulus mentions a
chirographum recording a mutuum cum stipulatione on which no interests had been stipulated. If,
however, the expiry date (the kalends of the following month) was not respected an interest poenae
nomine of one percent a month would be due27. A comparable case is mentioned in another fragment
of Paulus where an interest of 0.8% a month was due if the expiry date was not met 28. A very similar
arrangement is found in one of the Sulpician tablets, where a fine of 20 sesterces a day (poenae
nomine) is stipulated in case the expiry date is not met29.
It would seem, therefore, that we are dealing with two different types of cautiones, neither of which
was necessarily uncommon: one essentially recording stipulationes condicionales with interest clauses
and expiry dates, the other recording stipulationes purae or at least stipulations without interest
clauses.
So far, three plausible hypotheses have been put forward to explain the absence of interest stipulations
in the Sulpician tablets. The first, adhered to i.a. by Bove and Camodeca, states that interest was
deducted in advance from the sum paid out to the debtor30. The problem with this is that the absence of
expiry dates makes it difficult for either creditor or debtor to calculate their relative profit or cost
margin. The second, proposed by Guarino and recently elaborated by Gröschler, asserts that interest
métiers des manieurs d’argent, Rome, 1987, p. 434-435; Verboven, Système financier (n. 20), p. 90-91;
Verboven, Organisation (n. 2), p. 170.
22
At least 41 out of 127 documents relate to trials or arbitrations (TPSulp. 1-41).
23
9 testationes adfixi libelli for the anouncement of public sales (TPSulp. 83-88, TPSulp. 90-93) and 1 dilatio
auctionis (TPSulp. 89).
24
TPSulp. 50 (mutuum cum stipulatione) ; TPSulp. 56 (mutuum cum stipulatione) ; TPSulp. 68 (debitum in
stipulatum deductum) ; TPSulp. 69 (debitum in stipulatum deductum). Of these TPSulp. 50 is very fragmentary
and its reading remains uncertain.
25
CIL 3, p. 930, no. 3 ; p. 933, no. 4 ; p. 934, no. 5
26
AE (1996) 1124 ; M. A. Speidel, Die römischen Schreibtafeln von Vindonissa, Brugg, 1996, p. 98-99, no. 3
(Veröffentlichungen der Gesellschaft Pro Vindonissa 12).
27
Dig. 12,1,40 (Paulus).
28
Dig. 45,1,126,2 (Paulus).
29
TPSulp. 68: … in dies sing(ulos) HS XX nummos obligatum iri …. The main debt is the balance due of
previous debts amounting to 1250 sesterces. Note that the poena is about 48 times higher than the usura
legitima, a strong indication that at least in the first century fines for non-payment were not subject to the limit of
the usura legitima.
30
L. Bove, Documenti di operazioni finanziarie dall'archivio dei Sulpici, Napoli, 1984, 41-42, 47-48; G.
Camodeca, L'archivio puteolano dei Sulpici, Napoli, 1992, p. 174-177 (Pubbl. Dipartimento diretto romano e
storia delle scienza romanistica dell'università degli studi di Napoli Frederico II, no. 4) ; Camodeca, Tabulae
Pompeianae Sulpiciorum (n. 1), p. 134.
5
was paid monthly on the basis of an informal agreement (a pactum conventum). Although the creditor
could not legally force his debtor to pay interest, he could – in the absence of an expiry date– demand
immediate repayment in case the debtor stopped paying interest. If the debtor failed to hand over the
money without delay, he was legally obliged to pay usurae morae until the loan was fully repaid31.
The third hypothesis, advanced by Purpura and Andreau, asserts that interest stipulations were
contained in separate cautiones, none of which have survived32.
As such, all three hypotheses are possible and we should allow for the possibility that the Sulpicii
resorted to all three, depending on the exact circumstances of each case or the wish of their clients.
Nevertheless, I will try to show that not all three are equally convincing.
Separate cautiones usurarum
In favor of the Purpura – Andreau hypothesis we may note that the Sulpicii were obviously in the habit
of drawing up dossiers consisting of a variety of documents to back up their operations. In the case of
Cinnamus’s loan to Marius Iucundus, for instance, we see that the datio pignoris is recorded in a
separate cautio (TPSulp. 46), while the cautio recording the actual mutuum cum stipulatione (TPSulp.
53) doesn’t even mention the existence of the pignus. The chirographum recording the loan to
Niceros, servus arcarius of Puteoli (TPSulp. 56), is supplemented by a testatio (TPSulp. 114)
recording both the drafting of the cautio and the actual loan (per chirographum hac die dedit credidit).
It seems reasonable to suppose that the so-called nomina arcaria, as well, were only parts of larger
dossiers containing a variety of documents, since they only record that a certain sum had been payed
out, not that this had been done as a loan33.
The Digest confirms that agreements concerning conditions or details of transactions could be
recorded separately. Thus, according to Julianus, a formless agreement (pactum conventum) between
creditor and debtor that repayment of a loan would not be demanded as long as the interest was duly
paid, had to be treated as being part of the actual stipulatio, even though the latter had been formulated
as a stipulatio pura. Presumably, in the first century CE such arrangements merely provided an
exceptio, but the effect was largely the same34. The case confirms that a stipulatio pura could in fact be
merely part of a larger dossier, in which other documents could signicantly change the nature of the
‘naked’ agreement contained in the stipulatio.
But why would the Sulpicii bother drafting separate cautiones for capital and interest? There was
surely no legal reason, as is shown by the Transylvanian tablets and the Vindonissa tablet. Although
strictly speaking interest had to be stipulated separately, a stipulatio of different obligations was
legally considered to consist of as many stipulationes as there were obligations stipulated and Ulpian
explicitely confirms that this rule was valid for stipulation of capital and interest 35. Moreover, if their
aim was to cover up interests exceeding the legal maximum, the method can hardly have been
effective.
31
A. Guarino, Giurisromanistica elementare, Napoli, 1989, p. 205; P. Gröschler, Die Tabellae-Urkunden aus
den pompejanischen und herkulaneischen Urkundenfunden, Berlin, 1997, p. 165-177; Camodeca, Tabulae
Pompeianae Sulpiciorum (n. 1), p. 134.
32
G. Purpura, Tabulae Pompeianae 13 e 34 : due documenti relativi al prestito marritimo, in : Atti XVII
Congresso Papirologico 3, Napoli, 1984, p. 1245-1266; J. Andreau, Banking and business in the Roman world,
Cambridge, 1999, p. 98-99 (Key themes in ancient history). See also Gröschler Die Tabellae-Urkunden (n. 31),
p. 162-163.
33
Contra see Gröschler Die Tabellae-Urkunden (n. 31), p. 193-194
34
Dig. 2,14,4,3 (Paulus libro tertio ad edictum.) Ex facto etiam consultus, cum conuenisset, ut donec usurae
soluerentur sors non pet< ere> tur, et stipulatio pure concepta fuisset, condicionem inesse stipulationi, atque si
hoc expressum fuisset. Cf. Dig. 12,1,40 (Paulus libro tertio quaestionum): pacta in continenti facta stipulationi
inesse.
35
Dig. 45,1,75,9 (Ulpianus, libro vicensimo secundo ad edictum): Qui sortem stipulatur et usuras quascumque,
certum et incertum stipulatus uidetur et tot stipulationes sunt, quot res sunt. Cf. also Dig. 45,1,86,pr. (Ulpianus);
1,140,pr. (Paulus); 1,29,1 (Ulpianus).
6
Usurae ex pacto nudo
According to the Guarino – Gröschler theory, monthly interest payments would have been agreed
upon in a pactum conventum. Such a formless agreement was in itself unenforceable, but in the
absence of an expiry date, the creditor could effectively put pressure on his debtor by threatening to
demand immediate repayment36.
Although a pactum conventum was legally unenforceable, formless agreements on interest payments
were not illegal. It was only in the early third century that Roman jurists developed a coherent system
of how to deal with usurae ex pacto nudo. Severus decreed that a debtor could not reclaim interest he
had paid ex pacti conventione unless it transgressed the modus legitimus, in which case it would be
considered as an undue part of the capital that could be reclaimed with the condictio indebiti37. In
calculating the balance of a loan in case of default, interest paid on the grounds of a pactum nudum
could be reassigned to meet the interest obligations arising from a stipulatio usurarum or be deducted
from the capital of the loan if a stipulatio usurarum was lacking38. However, a creditor remained fully
entitled to sell pignora given to guarantee usurae ex pacti conventione39.
The situation in the first two centuries seems to have been much more favorable to the creditor. The
rule that interest had to be stipulated in order to be enforceable40 was interpreted and applied in a very
narrow and restricted sense. Since the condictio indebiti applied only to sums paid per errorem,
‘illegal’ interests paid on the basis of informal arrangements could not be reclaimed 41. Severus’s
innovation was twofold. On the one hand, he allowed debtors to use the condictio indebiti to reclaim
interest paid beyond the legal maximum. On the other hand, in cases of default, interest paid solely on
the grounds of a pactum nudum would be considered as downpayment on the capital sum, whereas
before such undue interest payments were presumably ignored.
Clearly, pacta nuda were a suitable and safe way to avoid interest regulations in the Early Principate,
provided a creditor found an extra-legal way to assure that his debtor would honour the agreement.
36
Gröschler, Die Tabellae-Urkunden (n. 31), p. 156-177 (p. 165 : Die Zinsen brauchten in den Urkunden also
nicht dokumentiert zu werden, weil zu ihrer Durchsetzung bereits andere Mechanismen zur Verfügung standen).
Followed by D. Johnston, Roman Law in Context, Cambridge, 2001, p. 86-87 (Key Themes in Ancient History).
Note in this respect TPSulp. 99, a pactum de solutione in diem cum stipulatione poenae dupli adiecta and
TPSulp. 63, a nomen arcarium with a stipulatio poenae and a iusiurandum added in case a non stipulated
(therefore informally agreed) expiry date was not met.
37
Dig. 12,6,26,pr. (Ulpianus libro uicensimo sexto ad edictum) : Si non sortem quis, sed usuras indebitas soluit,
repetere non poterit, si sortis debitae soluit: sed si supra legitimum modum soluit, diuus Seuerus rescripsit (quo
iure utimur) repeti quidem non posse, sed sorti imputandum et, si postea sortem soluit, sortem quasi indebitam
repeti posse. proinde et si ante sors fuerit soluta, usurae supra legitimum modum solutae quasi sors indebita
repetuntur. quid si simul soluerit? poterit dici et tunc repetitionem locum habere. ; C. 4,32,3 (Septimius Severus,
a. 200) : Quamvis usurae fenebris pecuniae citra vinculum stipulationis peti non possunt, tamen ex pacti
conventione solutae neque ut indebitae repetuntur neque in sortem accepto ferenda e sunt. ; Dig. 46,3,5,2,3
(Ulpianus libro quadragensimo tertio ad Sabinum): ex pacti conuentione datae repeti non possunt.
38
Dig. 46,3,5,2 (Ulpianus): for the text and a discussion of this highly complicated text see infra appendix.
39
Dig. 13,7,11,3 (Ulpianus libro uicensimo octauo ad edictum) Si in sortem dumtaxat uel in usuras obstrictum
est pignus, eo soluto propter quod obligatum est locum habet pigneraticia. siue autem usurae in stipulatum sint
deductae siue non, si tamen pignus et in eas obligatum fuit, quamdiu quid ex his debetur, pigneraticia cessabit. ;
Dig. 46,3,101,1 (Paulus libro quinto decimo responsorum) Paulus respondit aliam causam esse debitoris
soluentis, aliam creditoris pignus distrahentis: nam cum debitor soluit pecuniam, in potestate eius esse
commemorare, in quam causam solueret: cum autem creditor pignus distraheret, licere ei pretium in acceptum
referre etiam in eam quantitatem, quae natura tantum debebatur, et ideo deducto eo debitum peti posse. ; cf. also
Dig. 46,3,5,2,1 (for the text cf. appendix).
40
Cf. Pauli Sententiae 2,14,1: ex nudo pacto inter ciues Romanos actio non nascitur and Dig 19,5,24 (Africanus
libro octauo quaestionum) usuras nisi in stipulationem deductas non deberi.
41
J. C. Van Oven, Leerboek van Romeins Privaatrecht, Leiden, 1948, p. 367-368 ; Gröschler, Die TabellaeUrkunden (n. 31), p. 174-176 ; cf. Dig. 12,6,1 (Ulpianus libro uicensimo sexto ad edictum) Nunc uidendum de
indebito soluto. Et quidem si quis indebitum ignorans soluit, per hanc actionem condicere potest: sed si sciens se
non debere soluit, cessat repetitio. Dig. 50,17,53 (Paulus libro quadragensimo secundo ad edictum) Cuius per
errorem dati repetitio est, eius consulto dati donatio est.
7
However, pacta nuda were not very suitable to arrange monthly interest payments for short term
loans, as most of the Sulpician loans obviously were42. Although absence of an expiry date implied
that the debtor had to pay praesenti die when the creditor requested it, it can’t have been very hard to
stall for a few weeks, a considerable amount of time for loans only meant to cover a few months. Of
course the debtor would have to pay usurae morae, but these never transcended the usura legitima. A
trader could safely stop paying interest, wait for his ship to return and then smilingly pay the usurae
morae demanded from him.
Expiry dates
The main argument produced in favor of monthly interest payments ex pacti conventione rather than
advance payment of the entire sum of interest due, is the absence of an expiry date in most of the
Sulpician documents43. Two cautiones among the tablets – one containing a mutuum cum stipulatione,
the other a debitum in stipulatum deductum – explicitly record that the debtor promised to pay cum
petierit (creditor) (TPSulp. 51, 67)44. According to Guarino, Gröschler and recently Camodeca, these
cases preclude that interest was deducted or paid in advance.
However, the Digest offers at least one counter-example, where a debtor promises (by stipulation) to
repay his loan at his creditor’s request (qua die petieris) and to pay the entire sum of interest agreed
upon within thirty days after the initial stipulatio45. The implication is that absence of an expiry date
did not per se preclude advance bulk payment of interest.
In principle, absence of an expiry date in the case of advance bulk interest payments was dangerous
only to the debtor, not to the creditor, who was entitled to claim payment of the entire stipulated sum
praesenti die46. The core-question is therefore: how could a debtor protect himself against his creditor?
As we already saw, expiry dates could be agreed upon in a formless pactum conventum, which – if
proven – provided the debtor with an exceptio pacti against premature claims by his creditor47.
Moreover, at least in theory, there was no need even for the expiry date to be fixed for a minimum
term to apply. In principle, any claim resulting from a stipulatio pura could be subject to a minimum
term, providing it was obvious that the contracting parties had such a term in mind when they
concluded their deal48. Interestingly, a fragment of Florentinus’s Institutiones in the Digest asserts that
42
This is not only obvious in the cautiones containing expiry dates, but may also be inferred from the pignora
consisting of perishables and the fact that most debtors were traders.
43
There are five or six exceptions : TPSulp. 56 (chirographum mutui dationis, a loan to Niceros, servus arcarius
of the city of Puteoli); TPSulp. 68 (chirographum debiti in stipulatum deducti, clearly connected, however, to a
case of defraud); TPSulp. 69 (chirographum debiti in stipulatum deducti); TPSulp. 63 (nomen arcarium),
TPSulp. 99 (pactum de solutione in diem cum stipulatione poenae dupli adiecta). Possibly also TPSulp. 50
(chirographum mutui dationis, a loan to a certain M. Antonius M. f. Maximus), but the reading of the tablet is
very uncertain.
44
Cf. also two Transylvanian tablets : CIL 3, p. 934-935 (FIRA 3, no. 122) ; CIL 3, p. 930-932 (FIRA 3, no. 123)
(both with montly interest payments) ; Dig. 45,1,48 (Ulpianus); 1.135,pr. (Scaevola).
45
Dig. 45,1,135 (Scaeuola libro quinto responsorum) : Si ita quis promiserit: 'decem tibi dabo, qua die petieris,
et eorum usuras in dies triginta', quaero, usurae utrum ex die stipulationis an ex die, qua petita sors fuerit,
debeantur. respondit secundum ea quae proponerentur ex die stipulationis deberi, nisi aliud actum manifeste
probaretur.
46
Dig. 45,1,41,1 (Ulpianus libro quinquagensimo ad Sabinum) Quotiens autem in obligationibus dies non
ponitur, praesenti die pecunia debetur, nisi si locus adiectus spatium temporis inducat, quo illo possit perueniri,
uerum dies adiectus efficit, ne praesenti die pecunia debeatur: ex quo apparet diei adiectionem pro reo esse, non
pro stipulatore ; Dig. 50,17,14 (Pomponius).
47
By the third century pacta made at the time of the stipulatio (in continenti) were considered condiciones. Dig.
2,14,4,3 (Paulus libro tertio ad edictum): cum conuenisset, ut donec usurae soluerentur sors non pet< ere> tur, et
stipulatio pure concepta fuisset, condicionem inesse stipulationi, atque si hoc expressum fuisset ; Dig. 12,1,40
(Paulus libro tertio quaestionum) : pacta in continenti facta stipulationi inesse creduntur.
48
Dig. 45,1,41,1 (Ulpianus libro quinquagensimo ad Sabinum.) Quotiens autem in obligationibus dies non
ponitur, praesenti die pecunia debetur, nisi si locus adiectus spatium temporis inducat, quo illo possit perueniri ;
Dig. 45,1,73 (Paulus); 1,115 (Papinianus); Dig. 22,2,6 (Paulus); Dig. 45,1,137,2 (Venuleius).
8
a creditor who received advance interest payments was held to have tacitely agreed (tacite pactus
videtur) not to demand repayment before the intended expiry date. This not only means that the debtor
could claim an exceptio pacti in case his creditor tried to exact payment before the intended date, but it
also indicates that the expiry date had not been stipulated or otherwise explicitly agreed upon by
pactum conventum49.
In the case of the Sulpician tablets, the dossier of Marius Iucundus (TPSulp. 46, 53, 79) shows that
absence of an expiry date in the actual loan contract does not necessarily imply that the parties had no
expiry date in mind. The cautio recording the datio pignoris in this case (13,000 modii of wheat)
(TPSulp. 79) also records a pactum conventum granting the creditor the right to sell the pignus if the
loan had not been repaid within two months50. Strictly speaking, the date mentioned in the pactum de
pignoris vendundo only prevented the creditor from selling the pignus before this date and did not
deprive him of the right to demand repayment sooner, but presumably the arrangement also provided
the debtor with an exceptio pacti against premature claims of his creditor. Nevertheless, the case of
Marius Iucundus is exceptional. Nothing indicates the existence of missing pacta containing expiry
dates for the other loans. A number of cautiones are garanteed by fideiussores, not by pignora.
The absence of expiry dates may appear less strange when we consider the probable purpose of the
transactions financed by the Sulpicii. Although we are not dealing with pecuniae traiecticiae sensu
stricto, the loans were obviously contracted by traders to finance their operations. The rithm of these
was determined by the unpredictable sea and winds. Thus, even the most able trader could never know
in advance when exactly he would be able to repay his creditor. Under these circumstances it may
have been preferable not to specify an expiry date. We learn from several passages in the Digest that if
a loan was contracted to finance a specific project or enterprise limited in time, repayment was not due
before completion of the project. We may note in this connection that the loan to Niceros, the slave
accountant of the city of Puteoli, who was obviously no trader, was subjected to an explicit expiry date
recorded in the cautio (TPSulp. 56).
However, the cautiones in the Sulpician archives (like their counterparts among the Herculanean
tablets), never mention the purpose for which the loan was contracted and although the case of Marius
Iucundus shows that separate documents may have provided adequate protection, there is absolutely
nothing to substantiate the idea that such was the rule. Was it possible then for the debtor-trader to
claim an exceptio on the grounds that his creditor had lent him the money knowing that it was intended
for a time bound project?
Perhaps, however, the problem is mostly theoretical. As financial mediators, the Sulpicii depended
very much on their reputation of trust. Breach of faith, even though legally possible, would have been
detrimental to their overall profit rate: pacta sunt servanda, even when they are legally unenforceable.
In the long run, the Sulpicii had no interest in breaking formless agreements, because this would
inevitably damage their reputation on which their income largely depended51.
Interest deduction and advance interest payments.
Let us return now more in detail to the deduction of interest, which according to Plutarch was a
common practice among money-lenders to avoid interest regulations52. The so-called Sententiae
49
Dig. 2,14,57 (Florentinus libro octauo institutionum) Qui in futurum usuras a debitore acceperat, tacite pactus
uidetur, ne intra id tempus sortem petat. The expiry date could probably be deduced from the way the interest
had been expressed, e.g. as usurae centesimae.
50
TPSulp. 79 (Tab. III, pag. 5, ll. 9-11) : Si idibus Mais primis ea HS ((I)) ((I)) q(uae) s(upra) s(cripta) s(unt)
non dedero solvero / satisve fecero, tum liceat tibi id triticum, quo de agitur, / sub praecone de condicione
pignoris quo d(e) ag(itur) vendere.
51
On the importance of reputation in economic activities see A. Greif, The Organization of Long-Distance
Trade: Reputation and Coalition in the Geniza Documents and Genoa during the Eleventh and Twelfth
Centuries, Ann Arbor (Mi.), UMI, 1989 ; A. Greif, Reputation and Coalition in Medieval Trade: Evidence on
the Maghribi Traders, in: Journal of Economic History, 49 (1989), p. 857-883 ; Verboven, Système financier (n.
20), p. 84-87 ; K. Verboven, The Economy of friends. Economic Aspects of Amicitia and Patronage in the Late
Republic, Brussels, (in print) (Collection Latomus).
52
Plutarch, Moralia (De vitando aere alieno) 829 C-D.
9
Hadriani mention a complaint that numerous creditors were in the habit of deducting 10% from the
capital they paid out, charging the usura centesima on the entire sum and allowing the debtors to
postpone interest payments until the expiry date of the loan, but charging compound interest (usurae
usurarum) for this ‘service’ at the rate of again 1% a month53.
This passage present two possible alternatives for monthly interest payments: the interest could be
deducted from the capital when the loan was paid out to the debtor or its payment could be forestalled
until the expiry date of the loan. The recently discovered tablet from Vindonissa, recording a
stipulation to pay both capital and interest at the stipulated expiry date provides another example of the
latter technique 54.
The first procedure – if taken to the letter – was manifestly illegal and the debtor had adequate means
at his disposal to defend himself. If he was sued by his creditor, he could use the exceptio doli (later
exceptio non numeratae pecuniae) because he had never received the money which had been
stipulated. Alternatively, by the second century at least, he could take the initiative by bringing a
condictio liberationis against his creditor to anull his obligation to ‘repay’ the amount exceeding the
sum he had actually received55.
The second possibility, designated by Paulus as usurae sorti mixtae was not in itself illegal56 and the
Digest offers numerous examples of the practise without implying any illegalities about it. But, of
course, the normal interest procedures and regulations applied. Thus interest had to be stipulated in
order to be enforceable and interest exceeding the legal maximum could not be claimed in court,
although it did not anull the loan as such and the capital and interest up to the legal maximum
remained due57.
Florentinus, however, provides a third alternative. In stead of deducting interest from the capital of a
loan or of postponing payment of interest, the debtor could agree to pay the total amount of interest in
advance. Florentinus adequately describes the technique from the viewpoint of the creditor as
53
Hadriani Sententiae 5 (ed. Böcking p. 204-205):
Codicillos aliquis Adriano tradidit, per quos dicebat, plures esse feneratores, qui iniquas usuras
exigebant, in quibus quosdam in denarios mille ex ipsa die denarios centum subducere et extrinsecus centesimas
accipere et easdem iterum faenerari. Adrianus dixit: vir clarissimus, praefectus meus, de ea re excutiet et
renuntiabit mihi.
54
AE (1996) 1124 ; Speidel, Die römischen Schreibtafeln (n. 26), p. 98-99, no. 3:… Ibi sortem et usuras probas
recte dari stipulatus est Sex. Carisius Maximus, quo spopondit L. Haterius Marius …
55
Dig. 12,7,3 (Iulianus libro octauo digestorum.) Qui sine causa obligantur, incerti condictione consequi
possunt ut liberentur: nec refert, omnem quis obligationem sine causa suscipiat an maiorem quam suscipere eum
oportuerit, nisi quod alias condictione id agitur, ut omni obligatione liberetur, alias ut exoneretur: ueluti qui
decem promisit, nam si quidem nullam causam promittendi habuit, incerti condictione consequitur, ut tota
stipulatio accepto fiat, at si, cum quinque promittere deberet, decem promisit, incerti consequetur, ut quinque
liberetur. See also Dig. 12,7,1 (Ulpianus libro quadragensimo tertio ad Sabinum).
56
Dig. 22,1,20 (Paulus). The illegality in the Sententiae Hadriani resided in the usurae usurarum which the
faeneratores charged.
57
Dig. 2,14,17.pr. (Paulus libro tertio ad edictum.) Si tibi decem dem et paciscar, ut uiginti mihi debeantur, non
nascitur obligatio ultra decem: re enim non potest obligatio contrahi, nisi quatenus datum sit ; Dig. 12,1,11,1
(Ulpianus, libro vicensimo sexto ad edictum) Si tibi dedero decem sic, ut nouem debeas, Proculus ait, et recte,
non amplius te ipso iure debere quam nouem. sed si dedero, ut undecim debeas, putat Proculus amplius quam
decem condici non posse ; Dig. 2,14,27,5 (Paulus, libro tertio ad edictum) Si cum decem mihi deberes, pepigero,
ne a te uiginti petam: in decem prodesse tibi pacti conuenti uel doli exceptionem placet. ; C. 4,30,2 : (Imp.
Antoninus A. Maturio) Minorem pecuniam te accepisse et maioris cautionem interposuisse si apud eum qui
super ea re cogniturus est constiterit, nihil ultra quam accepisti cum usuris in stipulatum deductis restituere te
iubebit D. id. April. Antonino A III et Balbino conss. (a. 213). Dig. 22,1,20 (Paulus): Usuras illicitas sorti mixtas
ipsas tantum non deberi constat, ceterum sortem non uitiare.
10
‘receiving future interests’ (in futurum usuras (accipere))58. This procedure was perfectly legal and
required – at least in principle – no formalities, while the creditor retained the right to demand full
payment of the capital. As such, it offered an attractive solution to the limitations inherent in the first
two techniques. It assumed that the entire capital of the loan was first paid out to the debtor, who then
‘voluntarily’ paid his interests in the form of a lump sum. Once the debtor had voluntarily paid his
interest ex pacto nudo, he could not reclaim it as indebitum on the grounds that it had merely been
informally agreed upon, while the mutuum cum stipulatione recorded in the cautio remained valid and
enforceable. The absence of an expiry date in the cautiones of the Sulpician archive makes sense in the
light of these formalities. It allowed the creditor to hand over the money safely to the debtor. If the
latter refused to pay in advance the entire sum of interest agreed upon, the creditor could demand
immediate repayment of the loan’s capital.
Scaevola offers an illuminating example of how the practice could be organised. A creditor had lent
his debtor a sum of money in Rome, which was to be repaid in three months time in some distant
province. For some reason the debtor cancelled his voyage and wanted to repay his creditor
immediately in Rome ‘after deducting the sum which he had paid to his creditor as interest’ (detracta
ea summa, quam creditori suo usurarum nomine dederat). The creditor did not agree and he consulted
a lawyer, who answered that he could only claim the entire sum at the place and date agreed in the
stipulation, not in Rome itself59.
What is interesting about this case is that it suggests that the debtor had first formally received the
entire sum, commiting himself by stipulation to repay this sum at a certain designated place in three
months time, then – on the grounds no doubt of a pactum nudum – he had given part of the money
back to his creditor as interest (creditori usurarum nomine dederat). A variant of this procedure is
provided by a fragment of Scaevola’s Responsa, where a debtor promises (this time formally by
stipulatio) to pay the entire sum of interest agreed upon within thirty days 60. Interestingly the passage
in the Sententiae Hadriani we already mentioned, carefully specifies (both in Greek and Latin) that
), as if the illegality resided rather in the timing
10% had been deducted ex ipsa die (
than in the actual arrangement.
No one was fooled of course and we may surmise that in many cases (although not necessarily in all
cases) debtors acknowledged having received the entire sum of their loan without this actually being
the case. Legally, however, the procedure was watertight and provided a convenient loophole to evade
interest regulations. Probably until the time of Septimius Severus the debtor had no means to recover
his money.
Exceptio non numeratae pecuniae
The formalities of voluntary advance interest payment draw attention to the highly problematic
exceptio / querela non numeratae pecuniae. The exceptio could be used by the debtor against his
creditor when the latter demanded payment of a loan which had not been paid out in full. The querela
gave him the right to sue his creditor in case he had already paid the debt in question. While the
querela almost certainly does not predate the third century, there has been a lot of disagreement about
the dating of the exceptio. Since the early twentieth century, the exceptio has mostly been considered
58
Dig. 2,14,57 (Florentinus libro octauo institutionum) Qui in futurum usuras a debitore acceperat, tacite pactus
uidetur, ne intra id tempus sortem petat. See also Andreau, Banking and Business (n. 32), p. 98.
59
Dig. 45,1,122 (Scaeuola libro uicensimo octauo digestorum.) Qui Romae mutuam pecuniam acceperat
sol< u> endam in longinqua prouincia per menses tres eamque ibi dari stipulanti spopondisset, post paucos dies
Romae testato creditori dixit paratum se esse Romae eam numerare detracta ea summa, quam creditori suo
usurarum nomine dederat. quaesitum est, cum in integrum summam, qua stipulatione obligatus est, optulerit, an
eo loco, in quo soluenda promissa est, sua die integra peti posset. respondit posse stipulatorem sua die ibi, ubi
soluenda< m> stipulatus est, petere.
60
Dig. 45,1,135 (Scaeuola libro quinto responsorum.) : Si ita quis promiserit: 'decem tibi dabo, qua die petieris,
et eorum usuras in dies triginta', quaero, usurae utrum ex die stipulationis an ex die, qua petita sors fuerit,
debeantur. respondit secundum ea quae proponerentur ex die stipulationis deberi, nisi aliud actum manifeste
probaretur. Note that the exact amount of interest is not mentioned.
11
as an innovation of the Severan jurists, but recently Trofimoff and Busca argued that it originated in
formulary procedury somewhere in the late first or early second century61.
Although the tablets from the Agro Murecine, understandably, offer no new evidence to solve this
question, they do illustrate the advantages of the exceptio non numeratae over the exceptio doli. While
the latter was not allowed against parents or patrons because it was dishonouring, the former, as an
exceptio in factum was allowed62. The tablets from the Sulpicii show us a world dominated by
independent freedmen. C. Sulpicius Cinnamus was a freedman of C. Sulpicius Faustus, with whom he
frequently co-operated on an independent basis63. There can be no doubt that Cinnamus at least
occasionally (if not systematically) served as an intermediary for Faustus. If the latter wanted to cheat
Cinnamus, the exceptio doli would have been of no use. The utility of a special ad hoc exceptio in
factum, granted causa cognita, in such a case hardly needs further arguing. The tablets suggest,
therefore, that the exeptio non numeratae pecuniae originated in a world of freedmen financiers.
Wether the proposed new dating is correct or not, the exceptio n.n.p. was essentially no more than an
off-shoot of the exceptio doli64 elaborated at least (if not conceived) by the Severan jurists into a new
more comprehensive system to combat usury. The most remarkable and controversial feature of the
Severan system, saw the light in 215, when Caracalla reversed the burden of proof, both for the
exceptio n.n.p. and the exceptio doli in cases where a debtor denied having received the money of the
loan he had promised to ‘repay’. The innovation postdates the destruction of Pompei by over a
century, but it draws our attention to the problem of proof.
Proof in the nomina arcaria.
If we look at the Sulpician tablets we find a type of document clearly intended mainly (if not solely) to
offer proof of the creditor’s actually paying out the money of a loan to his debtor: the so-called nomen
arcarium. According to Gaius nomina arcaria were used as legal documents to provide proof of
61
A. M. Busca, Ancora in tema di excepto non numeratae pecuniae, in: Studia et Documenta Historiae et Iuris,
51 (1985), p. 489; H. Trofimoff, La cause dans l’exception et la querelle non numeratae pecuniae,, in:
Zeitschrift der Savigny-Stiftung für Rechtsgeschichte, romanistische Abt., 103 (1986), p. 343-344. Note however
that Gaius applies the exceptio doli in cases of non numeratae pecuniae, not the exceptio non numeratae
pecuniae (Gaius 4,116 and 4,119, cf. infra for the text). Justinian’s Institutiones (Inst 4,13,2) copy Gaius 4,116
almost word by word, but replace the words exceptio doli mali by exceptio non numeratae pecuniae.
62
Dig. 44,4,4,16 (Ulpianus libro septuagensimo sexto ad edictum) Aduersus parentes patronosque neque doli
exceptio neque alia quidem, quae patroni parentisue opinionem apud bonos mores suggillet, competere potest:
in factum tamen erit excipiendum, ut, si forte pecunia non numerata dicatur, obiciatur exceptio pecuniae non
numeratae. This may explain why the exceptio doli at least for a while seems to have continued to be used in
cases of non numeratae pecuniae. See Dig. 17,1,29,pr. (Ulpianus) ; C. 4,30,3. See also J. Ph. Lévy, “Exceptio
non numeratae pecuniae”. La loi de 215: une révolution dans le droit de la preuve?, in: IURA 36 (1985), p.
107-114.
63
Cf. K. Verboven, Organisation (n. 2), p. 161-163.
64
See Van Oven, Leerboek (n. 41), p. 210. Cf. Gaius 4,116 : Conparatae sunt autem exceptiones defendendorum
eorum gratia, cum quibus agitur. saepe enim accidit, ut quis iure ciuili teneatur, sed iniquum sit eum iudicio
condemnari. uelut < si> stipulatus sim a te pecuniam tamquam credendi causa numeraturus nec numerauerim.
nam eam pecuniam a te peti posse certum est. dare enim te oportet, cum ex stipulatu tenearis; sed quia iniquum
est te eo nomine condemnari, placet per exceptionem doli mali te defendi debere. Gaius 4,119: Omnes autem
exceptiones in contrarium concipiuntur, quam adfirmat is, cum quo agitur. nam si uerbi gratia reus dolo malo
aliquid actorem facere dicat, qui forte pecuniam petit, quam non numerauit, sic exceptio concipitur: si in ea re
nihil dolo malo avli agerii factvm sit neqve fiat. Dig. 44,4,2,3 (Ulpianus libro septuagensimo sexto ad edictum):
Circa primam speciem, quibus ex causis exceptio haec locum habeat, haec sunt, quae tractari possunt. si quis
sine causa ab aliquo fuerit stipulatus, deinde ex ea stipulatione experiatur, exceptio utique doli mali ei nocebit:
licet enim eo tempore, quo stipulabatur, nihil dolo malo admiserit, tamen dicendum est eum, cum litem
contestatur, dolo facere, qui perseueret ex ea stipulatione petere: et < si> cum interponeretur, iustam causam
habuit, tamen nunc nullam idoneam causam habere uidetur. proinde et si crediturus pecuniam stipulatus est nec
credidit et si certa fuit causa stipulationis, quae tamen aut non est secuta aut finita est, dicendum erit nocere
exceptionem.
12
obligations65. This is confirmed by the examples found among the Sulpician and the Herculanean
tablets. The text is always very formulaic and is obviously derived from a model designed by an
unknown Roman jurist.
TABELLAE A. AGERI
Exp(ensa)
N. Negidio
HS tot
petiit et numerata accepit
domo ex arca
Ac(ce)p(ta)
Arcae
HS tot
Unlike the chirographa recording mutua cum stipulatione or debita in stipulatum deducta, the nomina
arcaria merely record that a payment had taken place, not that the receiver was obliged to repay the
money. Usually in the documents the actual nomen arcarium is followed by a second clause recording
a personal66 or real suretyship67 or a stipulatio poenae68, from which the loan aspect can and could be
deduced (which would have sufficed to provide proof of the loan aspect of the transaction if
necessary). The actual nomen arcarium clause, however, provides proof of something else, viz. that
the money had been taken out of the arca of the creditor and had been booked as expensum in his
account books, had then been given to the debtor on the latter’s request in order to be booked as
acceptum in his account books and to be deposited in his arca 69.
Contrary to the nomina arcaria, the chirographa merely testify that money had been given and
received as a loan. In many cases the cautio doesn’t even specify that the loan was paid out
(numeratio), while it never precludes that part of the ‘loan’ was immediately (ex ipsa die to paraphrase
the Sententiae Hadriani) returned to the creditor usurarum nomine. The accounts or the money boxes
of creditor and debtor are never referred to in the chirographa.
Seen from the formalities pertaining to voluntary advance interest payments, the nomina arcaria make
sense: they show a preoccupation to prove that the money was not just fictitiously paid out or
momentarily ‘received’ (accepisse) by the debtor, but had truly been taken out of the money-box of
the creditor and been given in full to the debtor to put in his money-box. If the debtor’s account books
showed otherwise, the creditor could refer to the nomen arcarium, signed and sealed by the debtor and
up to nine witnesses70. Nothing had been deducted in advance – if the debtor voluntarily agreed to pay
a lump sum of interest later, that was an entirely different matter. It would have been very hard for any
debtor to claim that he had not been fully paid out. We might say that while the chirographa mutui
dationis aim at proving an obligation, the nomina arcaria aim at proving that an obligation was iusta
causa.
Were the Sulpician loans usurious?
The final question still remains: were the loans extended by or through the Sulpicii usurious and if so
is this the reason why the cautiones record mostly stipulationes purae or at least stipulations without
interest clauses ? Camodeca believes that this was the case. At least one tablet would seem to confirm
this theory. TPSulp. 75 is a double receipt (apocha) recording an acceptum of 6000 denarii (24,000
65
Gaius 3,131-132: Alia causa est eorum nominum, quae arcaria uocantur. in his enim rei, non litterarum
obligatio consistit, quippe non aliter ualent, quam si numerata sit pecunia; numeratio autem pecuniae rei facit
obligationem. qua de causa recte dicemus arcaria nomina nullam facere obligationem, sed obligationis factae
testimonium praebere. Unde < non> proprie dicitur arcariis nominibus etiam peregrinos obligari, quia non ipso
nomine, sed numeratione pecuniae obligantur; quod genus obligationis iuris gentium est.
66
TPSulp. 60, 61, 62, 64 ; TH 70+71 ; TH ined. III 6 (see Gröschler, Die Tabellae-Urkunden (n. 31), p. 145) ;
TH 67 III 6.
67
TH 74 (fiduciae), TH 73.
68
TPSulp. 63.
69
On a more detailed analysis the nomina arcaria see Verboven, Sulpicii from Puteoli (n. 2); cf. also Gröschler,
Die Tabellae-Urkunden (n. 31), p. 138-146 ; Camodeca, Tabulae Pompeianae Sulpiciorum (n. 1), p. 151-152.
70
Cf. TPSulp. 64.
13
HS) by Sulpicius Cinnamus from an unknown debtor and an acceptum of at least 20,000 HS by one C.
Aponius Stephanus from Cinnamus. Clearly, Cinnamus had acted as an intermediary between Aponius
Stephanus and the unknown debtor. Unfortunately, the tablet in question breaks of just after the XX
mark of the 20,000+ HS paid out by Cinnamus. There is, however, very little room left to the right of
the tablet to add 4 strokes to make up the missing 4,000 sesterces. The curious fact also that Cinnamus
receives a sum expressed in denarii (VI (milia) X) but pays out a sum expressed in sesterces, suggests
that the sum payed out by Cinnamus could no longer be expressed in a round sum of denarii. If so,
Cinnamus would have paid out between 21,000 and 23,000 sesterces, realising a profit margin in this
case between 4.2% and 15.5 %71. The actual interest rate for the debtor would have been even higher
since Aponius Stephanus, as well, had surely demanded interest. Unfortunately we have no idea of the
duration of the loan72.
In his de vitando aere alieno Plutarch sharply criticises the illegal lying practices of the faeneratores
who ‘write that they give so-and-so much to so-and-so, though they really give less’. And so the
debtor is cheated, because he ‘receives less than what is written’. Plutarch asserts that the system was
devised to defraud the law (paranomoos)73. But his preoccupation is a moral one, not a legal one and
illegalities are only noted in passing. The passage in question smoothly passes from the alleged
illegalities to the moral of cheating.
The concept legal versus illegal interest rates as such is absent from Plutarch’s essay because it is
irrelevant for his purposes. Even when he refers to anatocismus he fails to note the practice as illegal74.
Legally, Plutarch’s ‘usurer’ seems well armed: when the debtor tries to take him to court, he meets
him unabashed75. This could mean that usurers hid their illegal practices well, but it also calls to mind
Tacitus’s complaint that no matter how many laws were enacted against lending at interest,
faeneratores found ways to avoid them76. Plutarch’s text is well in line with what we found in the
Digest: until Severus interest paid ex pacto nudo could not be reclaimed by the condictio indebiti, even
if the rate substantially transgressed the legal maximum.
However, Plutarch’s essay on borrowing is focused strictly on non-productive loans. The mere
existence of productive loans is virtually denied. The tablets from the Agro Murecine prove otherwise
and Seneca confirms and generalises the picture of the tablets by claiming that ‘if you want to do
business, you will need to borrow’77 Surely, other moral standards applied in cases where a debtor
used his loan to enrich himself.
71
Expressed in terms of the 24,000 HS repaid by the unknown debtor, not in terms of the (unknown) capital.
TPSulp. 75 : C(aius) Sul[picius Cinnamus scripsi] / Me a[ccepisse ab ---] (denarios) VI (milia) n[ummum] /
Fa[u]sto Corn[nelio] Sulla [Felice] / L(ucio) Saluio O[thone Titiano co(n)s(ulibus)] / III idus M[a]ias / C(aius)
Aponius Step[han]us scr[ipsi me accepisse] / ab C(aio) Sulpi[cio] Cinnamo HS XX [millia n(ummum) debita] /
mihi. Act(um) Puteol[is]. Camodeca suggests HS XX [ m(illia) n(ummum) debita], but ‘millia’ is never
abbreviated in the tablets except when the entire sum has already been recorded in full before. The second X of
XX is written smaller and almost between lines 8 and 9.
73
Plutarch, Moralia (De vitando aere alieno) 829D:
72
Note that this view is well in line with the dolus sanctioned through the
exceptio doli.
74
Plutarch, Moralia (De vitando aere alieno) 829C :
. ; Plutarch, Moralia (De vitando aere alieno) 831E :
On
anatocismus see Pikulska, Anatocisme (n. 6), passim.
75
76
Plutarch, Moralia (De vitando aere alieno) 828f :
Tacitus, Annales 6,16: multisque plebi scitis obviam itum fraudibus quae toties repressae miras per artes
rursum oriebantur.
77
Seneca, Epistulae ad Lucilium 119,1: Ut negotiari possis, aes alienum facias oportet. See also K. Verboven,
Damasippus, the story of a businessman?, in: C. Deroux (ed.), Studies in Latin Literature and Roman History.
VIII, Brussels, 1997, p. 195-196 (Collection Latomus 239); Verboven, Système financier (n. 20), p. 75-78.
14
For traders advance interest payments or postponed interest payments had an advantage over monthly
interest payments. It allowed them to wait for the return of their ships and the sale of their cargoes.
Although the loans attested in the Sulpician tablets don’t appear to have been pecuniae traiecticiae
stricto sensu, the debtors were mostly traders and it should not come as a surprise that they preferred
the regime of advance interest payments to that of monthly payments.
Lastly, we should allow also for the nature of the debtors. These were no oppressed penniless poor,
nor were they extravagant luxurious aristocrats. They were well-to-do traders. The trade ventures for
which they borrowed from (or via) the Sulpicii were risky but highly lucrative. Significantly, although
faeneratio is characteristically portrayed as a sign of avaritia, when Roman authors write about
oportunities for inordinate enrichment they refer to sea trade not to usury78. The latter was mainly an
investment option when a succesfull negotiator withdrew from trade ‘satieted or rather I should say
satisfied with their profits made’79, like Trimalcio who, after accumulating more riches than all of his
fellow citizens put together, withdrew from trade and started to lend at interest to or through
freedmen80. Clearly, they were not powerless victims: they chose to borrow from (or via) the Sulpicii
and they chose to borrow ordinary loans in stead of maritime loans, presumably because the Sulpician
loans, although perhaps ‘usurious’ in a legal sense, were still cheaper than pecuniae traiecticiae. In
doing so, they preferred to take the risks of their trade ventures upon themselves rather than pay the
insurance premium (pretium periculi) attached to maritime loans81.
*
* *
I conclude that the absence of interest stipulations in the tablets from the Agro Murecine is best
understood if we assume that the Sulpicii charged interest beyond the legal maximum on the grounds
of legally unenforceable agreements (pacta conventa). Formally the debtor received the capital of his
loan in full, after which he ‘voluntarily’ payed back part of this sum as interest. Although we can’t
exclude the possibility that in some cases interest was in reality deducted from the capital of the loan,
there is no reason to suppose that this was necessarily always the case. The absence of expiry dates
provided the Sulpicii with sufficient means to put pressure on the debtors. From a strictly legal point
of view, it seems likely that the loans extended by or through the Sulpicii were usurious, but, given the
nature of the debtor and the probably purposes of the loans, it is far from self-evident that this was also
the case from a social and economic point of view.
Koenraad Verboven
Ghent University (Belgium)
78
On faeneratio as sign of avaritia see Verboven, Système financier (n. 20), p. 80-98. On trade by sea as a
source of enrichment see Verboven, Damassippus (n. 77), p. 214-215 ; K. Verboven, Mentalité et commerce. Le
cas des negotiatores et negotia habentes, in: J. Andreau e.a., Mentalité et choix économiques des Romains,
Bordeaux, Ausonius (in print).
79
To paraphrase Cicero, De officiis 1,150 (satiata quaestu vel contenta potius, ut saepe ex alto in portum, ex ipso
se portu in agros possessionesque contulit.), who of course refers to investments in land, not in interest bearing
loans.
80
Petronius 76,9 : postquam coepi plus habere quam tota patria mea habet, manum de tabula: sustuli me de
negotiatione et coepi < per> libertos faenerare.
81
Note also e.g. that Trimalcio financed his trade venture from his own means, rather than by taking pecunia
traiecticia.
15
Appendix 1
Structure of proof in the Tabulae Sulpiciorum.
Mutua cum stipulatione
(10)
Chirographa
Scripsi
Debita in stipulatum deducta
(4)
Chirographa
Scripsi
Nomina arcaria
(6)
Testationes
Expensa /-os
Petisse
numerata /-os (4 cases out of 10)
Accepisse
Numerata /-os
Accepisse
Domo ex risco vel ex arca
Mutua /-os (= causa debiti)
Causa debiti (3 out of 4 cases:)
Quaestum venditionis
Quos mutuos sumpsisse
reliquos mutui accepti)
Debere
Debere
Condicio diei (2 out of 10 cases82)
Condicio diei (2 out of 4 cases))
Stipulatio / fidepromissio proba /-os Stipulatio proba / -os recte dari
recte dari
Accepta / -os
Risco vel arcae
Datio pignoris vel fideiussio (5 out
10 cases83)
Iusiurandum fideiussionis causa (1
out of 10 cases)
Fideiussio / fiducia / pignoris datio84
Iusiurandum redditionis et poenae (1 Stipulatio et iusiurandum poenae (1
out of 4 cases)
out of 6 cases)
Condicio diei (1 out of 6 cases)85
Sponsio ad stipulationem poenae (1
out of 6 cases)
82
In one other case, the so-called dossier of Marius Iucundus (TPSulp. 46, 53, 79), a condicio diei is recorded in
a separate cautio de pignoris datione.
83
In one other case, again the so-called dossier of Marius Iucundus (TPSulp. 46, 53, 79), the pignoris datio is
recorded in a separate cautio (TPSulp. 79) not mentioned in the cautio stipulatoria (TPSulp. 53).
84
Missing only in TPSulp. 63, but here replaced by a sponsio garanteeing a stipulatio poenae.
85
In connection to a stipulatio poenae.
16
Appendix 2
Rescriptum de modo redigendae pecuniae debitae distractis pignoribus
Dig. XLVI.3.5.2 (Ulpianus libro quadragensimo tertio ad Sabinum)
1. Imperator Antoninus cum diuo patre suo rescripsit, cum distractis pignoribus creditor pecuniam
redigit: si sint usurae debitae et aliae indebitae, quod soluitur in usuras, ad utramque causam
usurarum tam debitarum quam indebitarum pertinere: puta quaedam earum ex stipulatione,
quaedam ex pacto naturaliter debebantur. si uero summa usurarum debitarum et non debitarum
non eadem sit, aequaliter ad utramque causam proficit quod solutum est, non pro rata, ut uerba
rescripti ostendunt.
2. sed si forte usurae non sint debitae et quis simpliciter soluerit, quas omnino non erat stipulatus,
imperator Antoninus cum diuo patre suo rescripsit, ut in sortem cedant.
3. eidem autem rescripto ita subicitur: 'Quod generaliter constitutum est prius in usuras nummum
solutum accepto ferendum, ad eas usuras uidetur pertinere, quas debitor exsoluere cogitur: et
sicut ex pacti conuentione datae repeti non possunt, ita proprio titulo non numeratae pro solutis
ex arbitrio percipientis non habebuntur'.
This highly complicated text has been thoroughly analysed by Gröschler, whose interpretation I
largely, although not completely, follow86. The passage contains a comment by Ulpian on a rescriptum
from Severus and Caracalla. Gröschler has rightly seen that the key to understanding the text lies in
the expression ad utramque causam usurarum ... pertinere in stead of ad utramque summam
usurarum.
The opening passage should be read as : ‘When there are interest (obligations) due and others not due,
then what has been been paid to meet (these) interest obligations should be ascribed to both grounds,
i.e. both for the interest which is due and that which is not due. For instance when some (interest) is
due on the grounds of a stipulatio and some is due (merely) on account of the natural obligation
arising from an informal agreement.’
Now there were two possibilities: either the summa usurarum arising from the enforceable obligatio
verbis was equal to the summa usurarum arising from the unenforceable obligatio naturalis, or both
summae were not equal. In the first case, there was no problem; the obligatio naturalis was fully
backed up by the obligatio verbis and the pactum conventum could be largely ignored except in so far
as it created exceptiones pacti. Suppose however that both summae were unequal, e.g. when the debtor
had agreed by stipulatio to pay usurae centesimae, but by pactum conventum to pay usurae binae
centesimae. In that case the summa indebita (which was only debita naturaliter) was twice as high as
the summa debita ex stipulatione. Of course the amount of the summa indebita exceeding the summa
debita was in itself unenforceable, but as we have seen, interest paid solely on the grounds of a pactum
conventum, even exceeding the legal limit, could not be recovered. So the question now was what had
to be done if a debtor failed to meet his interest obligations arising from the stipulatio but had before
this time already paid interest indebita on the grounds of an informal agreement. Until the time of
Severus, this money was probably lost. In calculating the balance due, the creditor could simply ignore
what his debtor had ‘voluntarily’ paid on the grounds of his obligatio naturalis. Severus changed this:
henceforth in calculating the balance of a loan, all interest payments had to be ascribed to both the
obligatio verbis and the obligatio naturalis. The consequence of this, according to Ulpian (from puta
... onwards), was that the money which had been paid fulfilled both obligations equally in stead of pro
rata as had been customary before.
An example to clarify this: Suppose a debtor borrowed 1000 sesterces for one year, agreeing by
stipulatio to pay 10 sesterces a month but by pactum conventum 20 sesterces a month. The total
summa usurarum debitarum would then be 120 sesterces, the summa usurarum indebitarum 240
sesterces. The first 6 months the debtor duly paid 20 sesterces a month, then he stopped paying
anything. Until Severus in calculating the balance due the creditor could claim that the sums had been
86
Gröschler Die Tabellae-Urkunden (n. 31), p. 167-169. See also here for arguments against long assumed
interpolations.
17
paid pro rata and therefore the total of 120 sesterces paid had to be divided into 60 sesterces to meet
the usura debita ex stipulatione and 60 sesterces to meet the usura debita naturaliter ex pacto nudo.
60 sesterces remained due ex stipulatione, 60 sesterces ex pacto nudo. The latter were lost, but the
former remained due. Severus, however, decreed that the money paid had to be fully ascribed to both
grounds. Therefore the 120 sesterces which had been paid were counted equally (aequaliter) to meet
both obligations. The obligatio verbis was thereby fully met, 120 sesterces remained due ex pacto
nudo, but this created merely an unenforceable obligatio naturalis and were therefore lost to the
creditor.
Thus far I agree with Gröschler. However, I don’t follow his interpretation of paragraph 2. This
second paragraph states that if interest had not been stipulated at all, money paid to meet interest
obligations would be considered as downpayment on the capital of the loan. According to Gröschler
this would be in contradiction to paragraph 1, which recognizes the validity of interest payments ex
pacto nudo once interest obligations ex stipulatione had been fulfilled. Therefore, he proposes the
emendation < pactus> [stipulatus].
Gröschler’s proposition, however, is hard to follow. Surely interest payments which had neither been
stipulated nor otherwise agreed upon by pactum conventum, could only be construed as paid in error,
thereby giving rise to the condictio indebiti. The emendation is furthermore incompatible with the
subiectio mentioned in § 3 where the rule that payments should first be ascribed to the fullfillment of
interest obligations is said to be valid only for usuras ... quas debitor exsolvere cogitur; i.e. only for
interest due ex stipulatione not for interest due merely on the ground of an unenforceable pactum
conventum. Ulpian adds that altough money paid on the grounds of an informal agreement could not
be recovered, it should not be left to the discretion of a creditor receiving money (ex arbitrio
percipientis (sc. creditoris)) to determine that the money he had received was given in fullfillment of
the unenforceable interest obligation ex pacto nudo by booking the (usurae) ex pacti conventione as
paid (pro solutis) while leaving the capital of the loan booked as due.
In fact § 2 makes perfect sense. The first paragraph lays down what had had to be done when there
were interests due ex stipulatione and ex pacti conventione, viz. the money paid would be considered
as given in fullfillment of both grounds equally. The natural question now arising, is what to do if a
debtor had fullfilled his interest obligations ex stipulatione – either because no interest had been
stipulated at all (as in the case of the Sulpician tablets), or because the summa usurarum debitae ex
stipulatione had been fully met (§1) – had further paid usurae indebitae on the grounds of a pactum
nudum, but failed to repay the capital of his loan ? Until the time of Severus, the money paid to cover
the usurae indebitae would undoubtedly have been lost for the debtor. Severus, however, decreed that
in this case the sums paid would be deducted from the outstanding capital of the loan. This is what we
find in § 2: ‘But if it should happen that interest was not due and someone simply paid in fullfillment
of an obligation87 which had not been stipulated at all, then the emperor Antoninus and his divine
father decreed that the money would fall to the capital of the loan.’
To return to our previous example. Suppose our debtor had payed 20 sesterces a month for 8 months,
totaling 160 sesterces. The usurae debitae (120 sesterces) were fully met, 40 sesterces were paid as
usurae indebitae. These 40 sesterces were then deducted from the 1000 sesterces capital of the loan,
leaving only 960 sesterces still due.
This rule is perfectly in line with another reform of Severus, we already mentioned: interest paid in
excess of the legal maximum on the grounds of a pactum conventum had to be considered as
downpayment on the capital of the loan, thus providing the debtor with a condictio indebiti against his
creditor. Neither is there is a contradiction with the imperial decree mentioned in C. 4,32,3, which
merely denies the condictio indebiti to a debtor to recover interest paid on the grounds of a formless
agreement not in excess of the legal maximum88.
Dr. Koenraad Verboven
‘Paid in fullfillment of an obligation’ since solvere differs from numerare in the sense that it always refers to
the fulfillment of some kind of obligation (c.q. an obligatio naturalis).
88
Cf. supra n. 37 for the text
87
18
Ghent University (Belgium)
Koenraad Verboven