CURRE CY I ROMA A D BYZA TI E EGYPT BY LOUIS C. WEST AND ALLAN CHESTER JOHNSON AMSTERDAM ADOLF M. HAKKERT - PUBLISHER 1967 PREFACE IN THE preparation of these studies we have had the benefit of counsel and criticism from friends and colleagues. Professors E. W. Kemmerer and Frank D. Graham helped us in some of the economic problems. Mr. Sydney P. Noe, Secretary of the Amer ican Numismatic Society, Professors David Magie, William K. Prentice, and Dr. H. V. M. Dennis gave us the benefit of their scholarship on many points. To them we tender our gratitude for their constructive criticisms and hereby absolve them from any responsibility for heresies or errors which may appear in these pages. We also express our deep sense of obligation to Princeton University in providing a grant from the Andrew Fleming West Foundation and to the American Council of Learned Societies for providing the funds for publication. LOUIS C. WEST ALLAN C. JOHNSON Unchanged Reprint of the Edition Princeton 1944 Princeton University Studies in Papyrology 5 V TABLE OF CONTENTS PREFACE V ABBREVIATIONS IX I. THE SILVER COINAGE OF THE ROMAN PERIOD 1 II. THE CoPPER STANDARDA ND BRONZE CuRRENCY 13 III. THE AI KAI FORMULA 26 IV. PAYMENTS IN PTITAPAI DRACHMAE 30 V. ACCOUNTINGP RACTICESA ND MATHEMATICALC ALCULATIONS 43 VI. MoNETARY TERMS IN THE RoMAN PERIOD 65 VII. EGYPTIAN HoARDs 74 VIII. PRICE LEVELS IN THE ROMAN PERIOD 76 IX. EGYPTIAN AND IMPERIAL EXCHANGE 86 X. BYZANTINE ISSUES FROM THE'. ALEXANDRIAN MINT 97 XI. MoNETARY TERMS IN BYZANTINE PAPYRI 111 XII. LOCAL "GOLD STANDARDS" 140 XIII. THE RELATION OF GOLD TO BRONZE 157 TABLES: I. Tetradrachms of Tiberius 171 II. The Alexandrian Tetradrachm 172 III. Monetary Equivalents 174 IV. Frequency Table Showing Sizes of Bronze Coins 175 V. Bronze Coins--Sizes and Weights 176 VI. Hoards 178 VII. Silver Coinage of A.D. 305-450 180 DOCUMENTS: I. P. Baden 37 181 II. P. Giess, 47, 11. 28-9 181 III. Gnomon of the Idiologus 106 182 IV. CIG. 5008, 5010 182 V. P. 0. 1411 183 VI. P. 0. 9 V 184 VII. P. Ryl. Inv. 650 184 VIII. SB. 6086 V., 11. 1-8 185 IX. P. 0. 1918 V 186 X. Ju stinian Edict XI 187 SELECTED BIBLIOGRAPHY 192 INDICES 193 vii ABBREVIATIONS THE various collections of papyri are cited according to the system instituted by Wilcken in his Grundzüge, and continued by his reviews in the Archiv für Papyrusforschung. Other works fre quently cited are abbreviated as follows. AJA. = American Journal of Archaeology. AJP. = American Journal of Philology. JEA. = Journal of Egyptian Archaeology. NC. = Numismatic Chronicle. = NZ. Numismatische Zeitschrift. RIN. = Rivista Italiarta di Numismatica. RN. = Revue N umismatique. ZN. = Zeitschrift für Numismatik. Annals = Liverpool Annals of Archaeology and Anthropology. Archiv = Archiv für Papyrusforschung. Dattari = G. Dattari, Numi Augustorum Alexandrini. Cata logo della collezione G. Dattari, Cairo, 1901. Hunter = George MacDonald, Catalogue of Greek Coins in the Hunterian Collection, University of Glasgow, 1899-1905. Mickwitz, Geld = Gunnar Mickwitz, Geld und Wirtschaft im römischen Reich des 4. Jahrhundert n. Chr. Soc. Scient. Fenn. 1932. Milne, Alex. Coins = J. G. Milne, Catalogue of Alexandrian Coins in the Ashmolean Museum, Oxford, 1933. Newnham Davis = Sylloge Nummorum Graecorum, Vol. I, pt. 2, London, 1936. Rohde = Rohde, Die Münzen des Kaisers Aurelianus, seiner Frau Severina und der Fürsten von Palmyra, Miskoloz, 1881. Schotten = Huhl, Die Münzensammlung des Stiftes Schotten in Wien, 1910. Svoronos = "'1:-ßopwvbsT, a voµlO"µa,ra,T OV Kparovs TWII IlroAEµa'Lwv, Athens, 1904-8. Vogt = Vogt, Die Alexandrinischen Münzen, Stuttgart, 1924. Weber = Forrer, The Weber Collection, London, 1922-6. ix CHAPTER I THE SIL VER COINAGE OF THE ROMAN PERIOD W HEN Augustus 1 obtained possession of Egypt, he found a fiduciary currency in use throughout the country. Until the time of Diocletian no real effort was made to provide for Egypt a coinage equal in intrinsic value to the contemporary imperial issues. Imperial issues of gold and silver were apparently not allowed to circulate in Egypt, and Egyptian issues seem not to have been current elsewhere in the Empire. 2 This policy of mone tary isolation, followed consistently for approximately 300 years, probably represents the longest-lived experiment with a purely fidu ciary coinage that the world has seen. For three centuries the average Egyptian had no practical way of obtaining any coin whose intrinsic value corresponded to its nominal worth. The coins in circulation were billon tetradrachms and bronze of various denominations. In keeping with the common practice these coins, except for the first two issues of Augustus, had no marks of value. While there is no general agreement among scholars as to the names of the various denominations, all agree that billon and bronze were fiduciary. Perhaps the outstanding difference in the monetary system of the periods before and after Diocletian is in the use of gold. While hoards of gold have been found (see Chap. VII), there is Httle evidence that they were part of the currency in use in Egypt. Gold is frequently recorded in dower contracts, usually by weight; once two gold denarii are mentioned. 3 Gold and silver jewelry was common. The alimentary contracts published by Boak spe ; cify amounts reckoned in gold, but paid in silver.4 A temple account, written in uncials, and dated by the editor in the second century, has prices in gold and silver (ap-yvpovv,[ x]pvuovv), but this 1 To avoid confusion, Octavian is called Augustus even before he received the official title. For the purposes of this study, the Augustan period is considered as beginning with 30 B.C. and extending through the first six years of Tiberius during which he followed the Augustan system. 2 Broughton (Econ. Surv. IV. 888) quotes an inscription from Rev. Philol. LXIII (1937), 334 which he dates before Vespasian to show that "Alexandrian" silver drachmae were in use in Chios. The inscription, however, speaks of "drachmae of Alexand_er" not of Alexandria. 3 BGU. 1045 (A.D. 154). 4 P. Michigan 121 R (p. 29) (A.D. 45). 2 ROMAN SILVER COINAGE ROMAN SIL VER COINAGE 3 date is very questionable and probably the document belongs to knowledge the coins of the year 7 could be arranged in sequence, the fourth century. 5 In the third century expenses of priests at a and that this date may possibly be commemorative rather than temple near Philae are given in gold aurei and drachmae. 6 It is the year of actual issue. 9 Apparently Tiberius experimented with probable that gold was brought here for the Ethiopian trade and his :first issues in an attempt to find a coin of a size and fineness in this district it may have had a limited circulation. Gold c~in that would satisfy Egyptian needs. These tetradrachms range may have been in circulation at Alexandria where it was needed in weight from less than six to over thirteen grams, the percentage for foreign trade, especially with India. of silver decreasing as the weight of the coin increased. I t would The papyri show evidence of a silver drachma of 6 obols and seem that he was attempting to make the Egyptian tetradrachm one of 7 or 7¼ obols. This has led some scholars to suggest that the exact equivalent of the imperial denarius 10 for the lightest the latter is the Roman denarius which circulated in Egypt at a coin analyzed contains 3.60 grams of silver and the heaviest 3.62 premium. 7 They have however failed to notice that the denarius grams. This is extremely close to the average weight of 3 .64 grams is almost completely lacking from Egyptian hoards, and that the in his denarii. Ju dging from the very scanty evidence afforded mention of the denarius in papyri is extremely rare, limited pri by weights of the Egyptian tetradrachms issued in his 11t h year marily to military accounts where there is little doubt that the and later, Tiberius himself may then have set the weight of this denarius is the Egyptian tetradrachm of 28 or 29 obols. The tax coin at about thirteen grams. No analyses of these later coins imposed by Vespasian on Je ws is sometimes called nµ0 ovoO rJvaplwv seem to have been made. but this was paid in tetradrachms. The Gnomon of the Idiologus lt is clear that at least by the time of Claudius the weight of the mentions the sesterce in connection with estates of Roman citizens, tetradrachm had been fixed at about thirteen grams and the silver but this is a legal usage and does not prove that the coin was in content at 16 to 18 percent. This arrangement lasted without actual circulation. Moreover those who believe that the denarius appreciable change until about the middle of the reign of Marcus was required in the payment of taxes have failed to note that the Aurelius. To judge from the few coins available, there was then six-obol silver drachma is limited to official accounts and that 1 a slight decrease in weight of about 5%, and a very marked de taxes are for the most part calculated on the basis of the smaller crease in silver content, apparently a reduction of 75%. Unless unit. this decrease in size, wy-ight and fineness is peculiar to the small The history of the Alexandrian tetradrachm from a metrological number of coins available, it would seem that Marcus Aurelius point of view is shown schematically in two tables (see Appendix). was planning a change in the monetary system. Table I gives a chronological picture of the various silver issues of The sudden reduction in size and weight introduced by Marcus Tiberius; Table II gives the average weight and the silver content Aurelius was not continued. In his last year he evidently doubled of the tetradrachm struck from the time of Claudius to 296 A.D. the silver content, while Commodus went back to the standard of when the minting of a separate coinage for Egypt ceased. the early second century. Until the time of Decius in the middle The tetradrachm struck by Tiberius shows a surprising range of the third century the weight of the tetradrachm was generally of weights as well as distribution by years. For the seventh year maintained between 12 and 13 grams, but the silver content (A.D. 20) the :first in which tetradrachms were coined, 143 weights showed marked variations if we may trust the evidence of the are shown as against 20 for all the later years. 8 If one may judge small number of analyses available. Septimius Severus reduced from such a tabulation, it appears that soon after his seventh year the silver again to about 10%; Elagabalus to about 7½%. With Tiberius determined on a heavy tetradrachm, approximating in some variations this percentage was maintained for about thirty size and weight those issued in Asia Minor and Syria but with a years, or until the joint reign of Valerian and Gallienus, though low silver content, in place of the earlier better tetradrachm of shortly before under Trebonianus and Volusianus, there had oc lighter weight. I t is possible that with further study and fuller curred the first substantial decrease in weights, a cut of some 20%. 6 SPP. XXII. 157. 9 The case of the bricks made at Rome that are dated in A.D. 123 seems to be 6 CIG. IV, 5005, 5007-10 (A.D. 232-248). analogous. 7 Mommsen, Archiv I (1901), 273. Mitteis-Wilcken, Grundzüge I, LXVI. 10 Pointed out by Milne in Annals, VII (1914-6), 57. Giesecke (Ptolemäergeld, 80) 8 This table includes all available published weights and also weights obtained from says that the tetradrachm was 1/24th of lb. as the denarius was 1/96th of lb., but the the collections of the Amer. Numis. Society and of several members of that organization. denarius at this time was 1/84th of lb. ----- 4 ROMAN SILVER COINAGE ROMAN S!LVER COINAGE 5 The beginning of the final debade is to be seen in the period year until the 12th of Diocletian when the series was abandoned when Gallienus was sole ruler. Then weights were reduced by and a universal system for the empire inaugurated. another 10% and the silver by almost 50%. Even this drastic Unless the similarity in the amounts of silver in his denarii and reduction brought no immediate change in the local price structure, in his Egyptian tetradrachms of the year 7 is purely accidental, judging from the comparatively few prices found in contemporary Tiberius planned to equate the Egyptian tetradrachm with the papyri. People apparently still had general confidence in the denarius. Nero reduced the silver content of the denarius appa good faith and solvency of the government, although an edict of rently without changing the silver content of the tetradrachm. A.D. 260, as generally interpreted, indicates some incipient dis Theoretically this change would increase the value of the tetra trust.11 But it may well be that the bankers who were unwilling drachm as expressed in terms of Nero's new denarius, but for this to accept the current coins were in reality questioning the right of there is no evidence in papyri. By the time of Valerian the Macrianus and Quietus to issue any coins at all. Egyptian tetradrachm, poor as it was, ~tained more silver than The reduction in the weights of the tetradrachm beginning with there was in half of his "good" antoninianus. In other words, the Trebonianus may reflect the displacement of the denarius by the Alexandrian tetradrachm was a better coin than the denarius, if antoninianus and an effo rt made in Egypt to keep the local tetra one assumes that the latter was then rated as half of the antonini drachm in reasonable relation to the decreasing weight and silver anus. Both coins, however, had so little value as bullion that the content of the imperial coin. only reason for hoarding either rested on belief in the continuing The mint record of this coin, as it is known today, is of interest. good faith and credit of the government. Under Tiberius tetradrachms were issued in years 7, 11, 14, and N either the Ptolemaic silver of Augustus nor the billon tetra 18 to 23 inclusive. None can be definitely assigned to Caligula. drachms of his successors had any value outside Egypt except as Claudius struck coins in years 1 to 6 inclusive but not later. bullion. They did not circulate elsewhere and with the exception In his third year he issued a "silver" didrachm and a "silver" of two published hoards of billon found in Britain, both of the drachma, both now exceedingly rare and apparently unpopular in Iate third century, and a few isolated coins of the Ptolemies and Egypt, for no later ruler ever repeated the experiment. Nero later Alexandrian issues reported in British and Belgian collec issued tetradrachms in years 3 to 6 and 9 to 14. Galba, Otho, and tions, there are no accounts of Egyptian coins found outside Vitellius continued the issue which was again broken in the 4th Egypt. 12 Whether these finds came there in the way of trade year of Vespasian, who later struck tetradrachms only in his 8th may be doubted. They may have been brought by soldiers or year. Titus issued them in two out of three years, Domitian only by travellers as curiosities. in three, the second, sixth and eighth; Nerva in his first year, while The motives which led Augustus to maintain the Ptolemaic Trajan beginning with his 5th year struck tetradrachms regularly system in Egypt cannot be determined, but the obvious advantages except for the 13th and 17th. Hadrian omitted coinage only in may be pointed out. At first the uncertainty of his status in the -his first year; Antoninus Pius only in his first and twenty-fourth; political reorganization of the empire may have prompted him to while Marcus Aurelius issued tetradrachms every year until his continue the Ptolemaic system until his future position was clari 11t h and thereafter only in his 17th and 20th. Commodus issued fied. However, it is clear that he did not intend that Egypt them every year, Septimius in every year except his 1st and 15th, 1 should be just another province of the Empire, but he intern;l.ed to while Caracalla struck them in the first four years of his reign and • rule the country as the successor of the Ptolemies and he kept it then abruptly stopped all coinage both of "silver" and bronze at t as a personal possession. No member of the. senatorial dass was the Alexandrian mint. lt is tempting to see in this sudden cessa 1.. ever allowed to enter Egypt without his special permission. Evi tion of coinage in 214 A.D. some connection with the introduction l!J dently he meant to take precautions against the rise of another 1 of the new imperial coin, the so-called antoninianus, that made its i Antony to challenge his power. first appearance Iate that year or in 215 A.D., although there is The Ptolemaic silver was much debased under the later Ptole no evidence that this coin ever circulated in Egypt. Beginning 1 mies. According to Milne the silver which Augustus found in with the second year of Macrinus tetradrachms were issued every ~1 12 NC. Series 4, XI (1911), 357; NC. Series 5, X (1930), 335; NZ. XXIII (1930), 182; 11 P. 0. 1411 No. V; see also Ancient Egypt 1917, 160. i NC. Series 5, XVII (1937), 135; Rev. Beige, 5 ser. VI (1874), p. 186. i 6 ROMAN SIL VER COINAGE ROMAN SIL VER COINAGE 7 circulation was about 20% fine.13 This had been coined in units tetradrachm is equated with the denarius and we may reasonably of the tetradrachm, didrachm and drachma. In private docu infer that the Ptolemaic stater was equal to the Roman denarius. 21 \ ments of the Augustan period the silver issues are frequently At some time during his reign Augustus began to issue bronze designated as Ptolemaic (ap"fvpws IlTOAEµaiKos). Sometimes they coinage, apparently discontinuing the so-called copper issues are called coined silver (frlu1Jµos) and once as legal tender (ooKiµos).14 which had stood at varying ratios to silver.22 Henceforth the lt is more usual, however, to specify amounts in drachmae without bronze drachma (xaXKi11~)w as fixed in value at six obols and thus any qualifying adjective. In all such cases it is safe to assume that apparently had the same value as the Ptolemaic silver. There is, silver is implied. The tax-receipts simply specify that the tax is however, some evidence that the silver coins commanded a pre paid either in silver or in bronze (or copper). Three receipts mium in the payment of taxes. A receipt for the uv11ra~iµovr eads (A.D. 10-19) use the term ap"fvplov pv~apov, once in a payment of as follows: xa (XKov) Kn1ri M-y(ov)o pax(µos) 'TEO"Uap(e-sy l11onai) (opax o. poll-tax, and twice in an adaeratio of wheat requisitioned for µ,a/.,) The 27 units of bronze can hardly be anything eise than official visits. 15 obols, and it is reasonable to suppose that silver commanded a The Ptolemaic silver drachma had six obols. This is clear from premium over bronze as it had in the Ptolemaic period, and con a lease of a papyrus marsh which stipulated that six monthly pay tinued to do so later in relation to the billon tetradrachm. 23 This ments of 250 dr., and six of 583 dr. 2 ob. made up the total of is the only evidence of a premium in the Augustan period. 5000 dr. for the year. 16 The silver drachma used in the payment The new billon tetradrachm of Tiberius and his successors is of taxes also had six obols. This is shown by the rate of the sup called imperial silver (ap-yvplov "2eßauTOv or T,eßaurw11) in contracts plementary tax of one and a half obols per stater which is evidently and the same descriptive adjectives (ooKlµov, e1ru:rnµov)a re applied one-sixteenth.17 Similarly the taxes on garden-land were calcu as was the case with Ptolemaic silver. In private accounts these Iated on the basis of a six-obol drachma. 18 Arrears of the dyke-tax terms are never used. Sometimes ap-yvplov is used, usually to dis were paid by four priests for four years to the amount of 108 dr. tinguish from payments which may be made in bronze, but in Since the tax was 6 dr. 4 ob. per man, the total should have been general even this is omitted. 106 dr. 4 ob. The difference of eight obols is probably to be In official records taxes are usually paid in silver (ap-yvplov) and explained as a charge of two obols each for receipt. 19 wherever possible the tetradrachm was required. Often the silver The proof of the equation of the Ptolemaic stater to the Roman was qa~~ified by the adjective pv1rapov, and this term is found only denarius is indirect. The Ptolemaic silver was clearly equated in tax-receif}ts and ledgers. 24 So far as we can determine the with its successor, the billon tetradrachm issued by Tiberius and silver drachma used in tax-records was always reckoned on the his successors from the Alexandrian mint. A contract of deposit basis of six obols as in the Augustan period. lt is hardly necessary of 600 dr. of Ptolemaic silver and billon (ap"(vplov "2EßauTOvK ai to cite the evidence of numerous calculations of supplementary IITOXeµaiKov) made in the reign of Nero provided that it be repaid taxes, taxes on garden land, additions, and official banking trans with coin of legal tender (a p"(vpwv E1rlu7Jµo1010 Kiµo112)0. The billon actions all of which show that the silver drachma (or the billon tetradrachm) was officially valued as of six obols (or 24 in the 13 Alex. Coins, XVI. This statement seems to rest on Poole, Catalogue of Greek tetradrachm). The outstanding exception is the tetradrachm paid Coins in British Museum, The Ptolemies, LXXX, where the statement is made that the late Ptolemaic issues of silver resemble the potin of Tiberius. There seem to be no in instalments of the poll-tax in the Caranis tax-rolls, where it is analyses of the Ptolemaic silver. dear that this coin had a value of 29 obols.25 14 ap-yvpws IIroXeµa<Kos. See Preisigke, Wörterbuch, s.v. brlcrrJµosi s found in SB. 5244 There is little evidence to determine whether Tiberius required (8 B.C.), 5243 (A.D. 6); BGU. 189 (A.D. 8), 911 (A.D. 18); P. Cornell 6 (A.D. 17). 1 For ap-y( ) IIroXeµa,Kov brw7Jµou cf. BGU. 1050 (Ca. 5 B.C.). 1 a premium when taxes were paid in bronze as Augustus seems ' 16 See Chap. IV. to have done. A document dated in the reign of Tiberius by 15 BGU. 1121 (5 B.C.). 1 Grenfell and Hunt (P. Teb. 401) seems to afford evidence of a 17 0. Strass. 54, 61-5, 68-70; W. 0. 362-3, 367, 370-5. 18 W. 0. 356 (18 B.C.), 1364 (15 B.C.), 375 (A.D. 34). 1 21 P. Achmim 8 (A.D. 197). 19 BGU. 1198 (5 B.C.). 22 See Chap. II. 20 P. Hamb. 2 (A.D. 59). Does the use of b·lcrrJµos imply that transactions were f sometimes carried on in bullion (6.crrJµos)? The only evidence is the frequent use of i 23 BGU. 1590 B (A.D. 6). 1 24 See Chap. IV. gold in dowries where the amount is usually given by weight. 1 26 Michigan Papyri, Vol. IV. 1 ROMAN SIL VER COINAGE 9 8 ROMAN SIL VER COINAGE Private accounts, however, prove very clearly that they were drachma valued at six and a. half obols. Lines 25-6 read as reckoned in drachmas of 6¾ (once), 7 or 7¼ obols. Numerous follows: (-ylvovrai) xb(es) <TKE< Soe---K-,a L eµßo}..(f/s)- x1rß< Spe, (-ylvonai) ro [1r]av SP.,,., Here the conversion of 682½ ob. into examples beginning with the accounts of the record-office at Tebtynis in the reign of Claudius and continuing to the third 105 dr. indicates that the drachma contained six and a half obols. century show that a tetradrachm was valued at 27. 28 or 29 obo}§. How~ver, the p_riceo f the beer is undoubtedly two obols per pint, After A.D. 215 the tetradrachm always had 28 obols.29 In a few and 1ts value 1s clearly reckoned in a six-obol standard. The cases there is a calculation which might indicate that the elerk embole, whatever that may mean, and the price of beer are added was reckoning in a 24-obol tetradrachm but these may be due to together as if on the same standard. There is no other instance carelessness in the accounts. 30 where drachmas of different standards are so added without the A few official documents also use the tetradrachm of 29 obo!s. conversion of one to the other. Moreover, brewing was a govem Thus the reports of the fisheries in the Fayum, which probably ment monopoly, and as was indicated above, all official accounts should be regarded as semi-official since half the revenue seems were kept on the six-obol drachma. For this reason it is suggested to go to the state and half to the fishermen, were reckoned on this that a deduction of one-thirteenth was made from the cost of the basis about the middle of the second century. 31 The tax-rolls frorn embole (so also in line 28 where a similar conversion is found), and Caranis also indicate that, no matter whether the taxes were calcu then the obols were converted into six-obol drachmae by the usual lated in silver drachmae of six obols or not, the payments to the official procedure. I t may be objected that the scribe makes no local office or bank were in tetradrachms of 29 obols.32 mention of such a deduction; but if we had the complete document When the taxes were paid in units of the tetradrachm it made or knew what embole meant, the answer might be forthcoming. 1 little difference whether it had 24, 28, or 29 obols. We can only 11:t ax-receip!s i_ti s not. unusual for the scribe to add other charges affirm that the supplementary charges were always ca)culated on w1thout spec1fymg their nature, as is frequent in examples of the the basis of a six-obol standard for the drachma. But if the 1rpo<TÖLa-ypacpbµEavnad, there seems no reason why he should not peasat?-t :hose to pay the tetradrachm in smail change of bronze, make deductions as well, especially if this was a familiar practice there 1s httle doubt that he had to pay a premium. This at least in the brewery trade. was true under Augustus and it was true at the end of the second From the beginning of the reign of Claudius we have clear century at Caranis. In this connection it seems that additional evidence of two different values of the silver (billon) coinage. In evidence for a similar premium is found in a fragmentary register official documents the value of the. drachma is with rare exceptions of taxes (P. Ryl. 187) dated near the end of the first century. six obols in reckoning prices of govemment grain, in calculating Each entry in line 3 has a sixth of the tax and supplement entered the supplementary fee, etc. above it, and the totals are carried across. In line 4, the amount In pdvate accounts the drachma of silver is valued at 6¾, 7, or of the tax, being less than a drachma, requires no conversion but 7¼ obols. That is to say the billon tetradrachm in private reck 1 this sum is carried across in both the upper and lower entri~s in oning contained 27, 28, or 29 obols. One exception should be line 5. Howe~er, the entry in line 6, whatever <Tvµ1ravnm ay mean, made to the rule in private accounts. Interest rates continued, seems to reqmre .no conversion (perhaps because this amount had apparently, to be calculated on the basis of a six-obol drachma. 26 1 already been paid with the premium), and is carried over to both In two cases interest rates are quoted in silver obols; but no such coins existed.27 Wilcken suggests that the calculation is based on 1 29 For examples see Chap. VI. There is only one example of a 27-obol tetradrachm the silver drachma of 7 or 7¼ obols; but since there was also a six aft~r Cl~udi~s, and this is found i~ P. Teb. 589. By the courtesy of the University of Califorma L1brary, the photograph1c reproduction of this document has been examined obol drachma of silver in official accounts it is probable that the f and the reading of Grenfell and Hunt is as usual confirmed. ' latter is meant. 28 In private contracts, where amounts are usually 1 30 The clerks were notoriously careless (Chap, V). Examples of 24 obols may be • quoted in multiples of the tetradrachm, there is no evidence to found in P. Mich. 123 R IV. 22-23 (see note); P. Lond. 131, 1177. Boak's commentary 1 on P. Mich. 123 R. shows how carelessly these accounts were kept. determine the number of obols given to that coin. . 31P . Oslo 90-1; PSI. 160 (A.D. 149), but the account in PSI. 735 (A.D. 138) was kept i m the 6-obol drachma. 26 PSI. 688 R; P. Lond. 202. In the latter document the rc1,teso f interest seem open 3 32 P. Mich. 223-5. On the method of payment see Youtie and Pearl AJP. LVII to question. 1 1 (1937), 465 ff. 27 CPR. 12; BGU. 362. !!' 28 Gr. Ostr. I. 735. i 1 1 1 10 ROMAN SILVER COINAGE ROMAN SILVER COINAGE 11 entries in line 7. The remainder of the document is too fragmen every-day value of the tetradrachm. Similar sums for some un tary to interpret, but evidently there is a similar accounting prac known tax are recorded on a roll in the Columbia University collec tice carried on throughout. Apparently the tax was calculated in tion. a six-obol drachma and the sixth written above the line indicates In so far as taxes were paid in actual tetradra:chms, as was the that it was paid in a seven-obol drachma. If so, the charge for case of the poll-tax, there was no necessity for a premium. But conversion found at Caranis in A.D. 171-4 was similar in theory when the peasant had accumulated his bronze obols from his daily to that used in the first century, though a different system of wages, he must go to the local bank and buy his silver tetradrachm, bookkeeping was employed in the Rylands' accounts. or if he offered to pay his tax in bronze currency, he was compelled Thus it seems that we must reckon with a six-obol silver drachma to pay a premium. Under Augustus this premium was three obols used in official calculations of accounts, supplementary charges, on a tetradrachm of Ptolemaic silver. There is no evidence as and in tax-registers generally, while in private accounts the silver yet for any premium in the time of Tiberius. Possibly the tetra drachma had 7 or 7¼ obols. drachm was valued officially at 28 obols, but an extra obol may Were there two different silver drachmae in Egypt? Had the have been charged when exchanged at the banks. At any rate denarius been current, it would inevitably have found its way into the private accounts of the registry at Tebtynis under Claudius hoards, and it is not quite reasonable that it would have changed show that the clerks converted the fees for registration into tetra its name in the documents to appear there as four drachmae. drachms of 29 obols. The great farm account of Hermopolis in How could it ever have been distinguished from the billon tetra A.D. 79 shows that the steward of the estate sometimes converted drachm? The only "silver" coin current in the Roman period his money at the rate of 28, sometimes at 29 obols for the tetra was either the debased Ptolemaic silver or the equally debased drachm even in the same month. 33 The rates of exchange vary silver tetradrachm of Tiberius and his successors. throughout the second century, but the official rate in the Caranis The simplest way to explain the six-obol silver drachma in the rolls is 29 obols, even though this is in the period when the tetra accounts kept by the bureaus or banks in official calculations is to drachm of Marcus Aurelius had lost 75% of its silver content. assume that the bureaucratic officials had inherited this tradition The latest example of the tetradrachm of 29 obols is found in a from the Ptolemies and continued to follow it as an accounting temple-account of A.D. 215. Thereafter the value seems to be device long after the silver commanded a premium in daily life stabilized at 28 obols.34 Thus the accounts of a !arge estate pub and usage. The silver drachma of six obols never had a real lished by Viereck (0. Strass. 662-771) are all reckoned .on this existence in Roman Egypt. The billon tetradrachm commanded basis. For example (No. 672), the wages of 9 men at l dr. l ob. a premium, and the custom grew up of regarding the silver drachma amount to 10 dr. 2 ob. The tetradrachm of 28 obols naturally led as of 7 or 7¼ obols. Thus the metrological papyrus (P.O. 9V) to regarding a drachma as of seven obols. The workman, 'if he mentions a bronze drachma (xaAKtvn) of six obols, but the opaµxn was paid daily, was either given eight obols, or more probably a of seven obols was evidently an accounting device. This papyrus xaAKwno f six obols and two obols in small change. must be dated about the middle of the third century when the To sum up our conclusions. From 30 B.C.-A.D. 20 the Romans tetradrachm had 28 obols and the drachma would have a value of retained the Ptolemaic silver as legal tender, and issued no silver seven obols if such a coin existed. Hence the appearance in at Alexandria. The drachma was apparently equated with the accounts of sums in silver which are not multiples of the tetra Roman sesterce in value, but neither Roman nor provincial issues drachm. Five silver drachmae meant, in popular usage, a billon were permitted to circulate in Egypt. This Ptolemaic silver tetradrachm and seven obols. drachma was valued at 6 obols, but apparently commanded a There is some evidence that new taxes introduced by the Romans premium when exchanged for bronze. Then Tiberius introduced were not always calculated in the traditional way. Thus the the billon tetradrachm with a silver content approximately that Ieases of various concessions by the state such as reed-mats, oil of the Roman denarius. This billon was called silver, with a mills, fisheries, etc. were not bound by the old rules and sums like qualifying adjective frequently added in private contracts, and it 4 dr. 15 ob., (P. Milan 6) 80 dr. 80 ob. (P. Amh. 92), and the like, 33 P. Lond. 131. imply that these were calculated in accordance with the normal 34 BGU. 362'; sec Chup. VI.