An Unknown Coin of Pupius Rufus Struck in Cyrenaica, Wiadomości Numizmatyczne 53, 2009

Wiadomosci Numizmatyezne, R. LIII, 2009, z. 2 (188) M.ATERIAtY PIOTR JAWORSKI AN UNKNOWN COIN OF PUPIUS RUFUS STRUCK IN CYRENAICA Two bronze coins representing a type that had not yet been noted in the literature have recently been recorded by the author in one of Poland's private collections of ancient coins. Both specimens were part of a large ensemble of coins brought from Cyrenaica in the 1970s and 1980s, a time when Libya was the workplace of thousands of Polish contract workers. Many of them brought home coins usually bought from local children, who sold them to tourists as souvenirs near large archaeological sites.1 Unfortunately, it was impossible to determine the location at which these coins had been purchased. The first of the specimens of interest — the better-preserved one — was presented at the exhibition "The Hoard from Ptolemais" at the Royal Castle in Warsaw at the turn of 2008/2009 and published in the exhibition catalogue: Obverse: Head of Libya (or Apollo)2 right, [in left field: L?] Reverse: Il-P, serpent coiled right. 1. 16.5 mm, 2.64 g, * (Fig. I)3 2. [n]-P, 16 mm, 2.73 g, ^ (Fig. 2) 1 These practices may be observed in Eastern Libya also today, despite decisive action of the local antiquities service. Although they lack archaeological context, coins of such provenance are of interest to academic researchers because, many local issues being poorly represented in museum collections, they offer a possibility of adding to the knowledge in this respect; cf. the ensemble of 173 coins purchased in Ptolemais, published by Kraeling: C. H. Kraeling, Ptolemais. City of the Libyan Pentapolis [The University of Chicago, Oriental Institute Publications, XC] (Chicago, 1962), pp. 268-269. 2 See discussion below. 3 P. Jaworski, E. Walczak, 'Catalogue',SkarbzPtolemais. Katalogwystawy, Zamek Krolewski w Warszawie, 15 grudnia 2008-15 stycznia 2009, ed. P. Jaworski (Warszawa, 2008), cat. no. 61 (pp. 100-101). 183 Figs. 1-2. Unknown quadrantes of Pupius Rufus (2:1 and actual size) As will be demonstrated further in this paper, because of the unquestionable Cyrenaican origin of these objects, as well as the abbreviated legend on the reverse and the typical imagery, these coins should be classified as quadrantes of Aulus Pupius Rufus, a Roman official responsible La. for the minting of coins in Cyrenaica in the period preceding the battle of Actium (31 B.C.). Only two types of his asses, two types of semisses and one type of quadrans have been known hitherto. A previously unrecorded second type of quadrans, analysed in this paper, adds a new element to the coinage system of Pupius Rufus in Cyrenaica. Aulus Pupius Rufus is a figure known solely thanks to coins struck in Cyrenaica.4 The office he held is designated as follows by Greek legends on bronze coins of the highest denomination: 'AvxLOXQdx(rjyog) km xa\Liaq and xap-iag avxioxQd(xrjyog),5 as well as by images: the sella castrensls between fasces known from asses of Pupius Rufus is an attribute of a propraetor, and the sella quaestoria, hasta and sacculus found on the semisses of this official constitute the attributes of a quaestor.6 Traditionally the activity of Pupius Rufus is dated to a period between two other officials known from coins of Roman Cyrenaica: Crassus and Scato. However, there is no certainty as to the absolute dates of his term in office, although it is commonly agreed upon that it must have been before 27 B.C. when, as a result of the reorganisation conducted by Augustus, the province of Creta et Cyrene was included in the competencies of the senate, and the governor in charge of it was not, as before, a quaestor pro praetore,1 but a praetor pro consule. 4 G. P e r 1, Die romischen Provinzbeamten in Cyrenne und Creta zur Zeit der Republik, Klio 52,1970, pp.349-351; cf. A. Laronde, 'La Cyrenai'que romaine, des origines a la fin des Severes (96 av. J.-C. — 235 ap. J.-C.)', ANRW II, 10.1 (Berlin-New York, 1988), pp. 1013-1014. 5 E. S. G. Robinson, Catalogue of the Greek Coins of Cyrenaica [in the British Museum] (London, 1927) (below: BMCCyr), p. ccxxiii. 6 A. Burnett, M. Amandry, P.P. Ripolles, The Roman Provincial Coinage, vol. 1.1 (London-Paris, 1992) (below: RPC I), p. 221. 7 The first governor in the rank of quaestor appointed to Cyrenaica was Lentulus Marcellinus. As it is believed, the reason for sending a quaestor to Cyrene and keeping sub- 184 E.S.G. Robinson (and others) dated the issues of Pupius Rufus to 30-27 B.C.,8 to a period immediately after the battle of Actium, while T.V. Buttrey proposed the years 34-31 B.C.,9 a period when Cyrenaica was handed over by M. Antonius to Cleopatra Selene, ending the parallel issues for Crete and Cyrenaica. The earlier issues meant for both parts of the province featured the same types of obverse and reverse, differing only in legend: Greek for Cyrenaica, Latin for Crete. Pupius Rufus was the first Roman official to strike coins with a Greek legend for Cyrenaica alone.10 The letters A and L, found on this official's coins, were to indicate — as is believed — Libya as their area of circulation or place of minting (Cyrene). Buttrey's dating precedes not only Scato's coinage, but also wartime coins released by Antony and Cleopatra VII11 and by Scarpus (the latter are unattested among finds from Cyrenaica and therefore classified as Roman coinage).12 Among the arguments in favour of such a dating of the coinage of Pupius Rufus, Buttrey presents metrological data (a notable, gradual decrease in weight of coins of particular denominations with every new issue) and the fact that as a rule coin issues after Antony have legends and images that refer to Augustus or the imperial family, whereas the coins of Pupius Rufus lack such references. Agreeing with Buttrey's reasoning, one may add one more argument, which has to do with a discovery made several years ago in Ptolemais. A recently published quadrans of Pupius Rufus excavated by the Polish mission13 (Fig. 3) represents the only type of this denomination accounted for in today's literature and has the letters ITT instead of the letter L in the left field of the obverse. If we were to interpret these letters as an abbreviation of the word Ptolemais, which indicating the place where the coins of this series were struck, this would mean that Pupius Rufus continued the practice of his predecessor Crassus who — in unknown circumstances — struck coins bearing the unabbreviated name of Ptolemais: IITOAEMAI14 (although it is believed that the coins for Cyrenaica were struck in Cyrene, and for Crete in Knossos). sequent officials of this rank there were financial issues: S.l. O o s t, Cyrene, 96-74 B.C., Classical Philology (1963), vol. 58, no. 1, p. 21. 8 BMCCyr, loc. cit. 9 T. V. Buttrey, 'The Roman coinage of the Cyrenaica, first century BC to first cen- tury AD', Studies in Numismatic Method presented to Philip Grierson, ed. C.N.L. Brooke et al. (Cambridge, 1983), pp. 31-32; Buttrey's dating was supported, but not without certain objections, also by: RPC I, loc. cit. 10 Although about 34 B.C. the common issues for Crete and Cyrenaica were terminated, it is possible that for some time the circulation of coins between the two parts of the province may have continued, as can be concluded from a half-as of Kydas, very worn from long-term use (struck in Knossos after 37-36 B.C.), found by the Mission of the Institute of Archaeology of University of Warsaw in Ptolemais: P. Jaworski, Antyczne mennictwo Cyrenajki. Polskie odkrycia numizmatyczne w Ptolemais', Skarb z Ptolemais, o.c, p. 34. 11 RPC I, cat. nos. 924, 925 (pp. 221 and 222). 12 M. H. Crawford, Roman Republican Coinage (Cambridge, 1974 [1995]), vol. I, no. 546 (pp. 542-543), pi. LXIV. 13P. Jaworski, 'Rzymskie mennictwo Cyrenajki (I w. p.n.e.-I w. n.e.). Odkrycia misji Instytutu Archeologii UW w Ptolemais', Biuletyn Numizmatyczny 2006, no. 4 (344), pp. 256-257; recently also: Jaworski, Walczak, o.c, cat. no. 60 (pp. 100-101). 14 Today we know four specimens of coins of this series; on the obverse the head of Tyche Ptolemais and on the reverse — a crocodile: RPC I, cat. no. 916 (pp. 220-221 — here refer- 185 Fig. 3. Quadrans of Pupius Rufus probably struck in Ptolemais, Polish excavations in Ptolemais The nomina of Aulus Pupius Rufus as the issuer were given in Greek next to his titulary (cf. above) on all types of bronze coins struck by this official. The legends always contain two nomina or less: AYAOC nOYIIIOC, nOYIIIOC/POY^OC, nOYIIIOC. However, no coin has the full tria nomina. It is also striking that in a pair of types representing each of the denominations, the legend on one was reduced to the nomen gentile (nOYIIIOC), and in the type of quadrans described in this article the abbreviation was limited to the first letters of the second and third nomen. The letters: II-P can therefore be read as the first letters of the nomen gentilicium and the cognomen of the issuer: n(ovmoq) P(ov<pog). The characteristic separation of the letters by the iconographic motive on the reverse (the serpent) is also attested on a known type of quadrans on which the letters forming the name of the issuer were divided as follows: n-OY/TI-I/O-C. Also notable when comparing the legends on both types of quadrantes of Pupius Rufus is the similar shape of the letters, especially the way n was written (Fig. 4). Fig. 4. Reverse of quadrans of Pupius Rufus found in Ptolemais (drawn by A. Dluska) The first of the abovementioned specimens, which lacks the original patina, as well as the less well-preserved second specimen, do not permit us to determine whether or not the letter L for Libya was struck in the left field. Such a supposition may, however, be formulated, if one considers the fact that all known coins of Pupius Rufus carried this symbol. A question hitherto unanswered is whether — as in the case of the earlier known type of quadrans — also the second type was struck not only in Cyrene, but also in Ptolemais. The motives depicted on the coins of Pupius Rufus were mostly of local origin. Such an image was certainly the head of Zeus-Ammon, which constituted a kind of a national emblem of Cyrenaica,15 the head of Libya (or Apollo, cf. below), and the ences to all specimens); J. Svoronos, Ta Nomismata tou Kratous ton Ptolemaion (Athens, 1904-1908), nos. 1901a (Athenian specimens) and 1901b (Vienna specimen), pi. LXIII.27-28; BMCCyr, o.c, p. ccvi, pi. XLII.10 (Vienna specimen). 15 Zeus-Ammon was placed on coins struck in Cyrenaica since the beginning, along with silphion: BMCCyr, p. ccxxxiii-ccxxxix; his bust facing right was placed on seals with which documents and goods, most probably from Cyrene, were sealed, cf.: J a w o r s k i, Antyczne mennictwo', o.c, p. 32 (fig. 18). 186 ram. An iconographic element that did not appear in Cyrenaica prior to the coinage of Pupius Rufus was the coiled serpent. The style of the depictions on the coins of this official clearly appears to be somewhat clumsy, even barbarised. The head of bearded, invariably horned Amnion depicted on the obverses of the asses of Pupius Rufus, is shown in profile, facing right. Above the forehead there is ;i pair of vertically fixed feathers. Based on these stylistic traits it is possible to distin- guish two types of depictions of this god's head: the first type — similar to the images known from the coins of Lollius16 — is characterised by classical, soft facial features (Fig. 5), the second type has a sharp countenance, the eye was depicted frontally and the horn is much thicker at the base (Fig. 6). The image represented on the reverses of both types is a sella castrensis between two fasces. Earlier, in Lollius' coinage, the reverses of the highest denominations showed a sella curulis. The imagery of the reverse is identical on both types of asses, but the obverses and reverses bear different legends. Fig. 5. Asses of Pupius Rufus (RPC 920): A — Polish excavations in Ptolemais, B — Museum of Ptolemais, Libya Fig. 6. Halved as of Pupius Rufus (RPC 919), Polish excavations in Ptolemais On the obverses of semisses of Pupius Rufus there is a figure of a ram (Fig. 7), standing right or left depending on the type. The reverses bear axonometric views of a quaestor's attributes — the sella quaestoria on the reverses of the first type of coins is viewed from the left, and on the second type from the right. As in the case of coins of higher denominations, the legends on the semisses too differ from one type to another. s BMCCyr, p. 116 (cat. no. 19-20, pi. XLI.6-7). 187 Fig. 7. Semis of Pupius Rufus (RPC 921), private collection in Poland Likewise, the two types of quadrantes struck by Pupius Rufus had — like the higher denominations — common imagery and different legends. Portrayed on the obverse was a head of Libya or, as the author of this paper assumes, of Apollo facing right, and on the reverse a coiled serpent (Fig. 8). The identification of the head depicted on the obverses of both types of quadrantes of Pupius Rufus as Apollo, as well as Libya, is disputed. Already Miiller identified it as a head of Apollo,17 although Robinson and Buttrey, and recently also the authors of RPC believed — perhaps due to a certain resemblance to the barbarised depictions on the reverses of small Late Ptolemaic bronzes struck in Cyrenaica — that the head is a depiction of Libya. A similar stance was taken by the abovementioned authors regarding the heads on the obverses of the smallest denominations struck in Cyrenaica by Lollius and Scato. According to the author, in both cases it is difficult to ascertain this identification. The typical hairstyle should not be considered an identifying trait, as in ancient art it was an attribute of various figures, including Libya and Apollo.18 Despite the simplified and somewhat clumsy style of the coins of Pupius Rufus, the head shown in profile might be interpreted as portraying masculine features. It is also worthwhile to compare the heads from the quadrantes of Pupius Rufus with the aforementioned heads of Libya, as well as with the head of Apollo among coins struck by other Greek mints. The definite identification of the head depicted on obverses of quadrantes of Pupius Rufus requires specific research on this subject, taking into account not only iconographic arguments. Fig. 8. Quadrans of Pupius Rufus (RPC 923), private collection in Poland The image of the coiled serpent, depicted in different ways on the two types of quadrantes, still awaits discussion. The religious character of these depictions is beyond doubt, although their connection to a specific cult will require in-depth studies taking local idiosyncrasies into consideration.19 17 L. M ii 11 e r, Numismatique de I'ancienne Afrique, vol. 1: Les monnaies de la Cyrenalque (Copenhague, 1860), p. 162, cat. no. 428. is Cf. W. Lambrinudakis et al., Apollon', in: LIMC II.l, Zurich-Munchen 1984, pp. 249-250 (Miinzen). 19 Connecting the serpent with the cult of Asklepios or Zeus-Ammon was suggested by Robinson: BMCCyr, o.c, p. cexxiii. 188 Imagery on the coins of Pupius Rufus was adopted by Scato several years later, perhaps in order to distinguish between denominations.20 He used the image of a camp chair on the reverse to mark the as, on the obverse of the semis he put a ram standing right, and the quadrantes struck by him were given a head of Libya (or Apollo) on the obverse and a coiled serpent on the reverse. The flans of both specimens under discussion were mould-cast and as a result of this they have a characteristic bowl shape and diagonal edges. Apparently there was no rule for placing the obverse and the reverse on a particular side; on the better- preserved first specifnen described above the obverse is on the flat side of the flan, while on the convex side the reverse image was struck in a sunk circle. On the second specimen the placement is reversed. Considerable freedom in placing the obverse and reverse dies on either side of the flan seems to have applied also to the known examples of the first type quadrans of Pupius Rufus, although the arrangement in which the obverse is on the convex side and the reverse on the flat side is definitely dominant.21 Considering the average weight of 2.69 g (set based on only two known examples), the denomination of the coin described in this article fits within the limits established for the quadrantes of Pupius Rufus, belonging to the previously attested type.22 Also the diameter of the two specimens (16-16.5 mm) corresponds to the standard known from literature.23 The smallest bronze coins of Pupius Rufus having the abovementiohed parameters have been identified as quadrantes for a relatively short time.24 Before, Robinson25 (like Muller)26 believed them to be semisses. As a result of a rather intricate discussion concerning weight standards and the reconstruction of the denomination structure of the coinage of Roman Cyrenaica it has been determined that the coins in subsequent series, bearing the names of Roman officials appointed for the province, introduced Roman weight standards into the circulation in both Crete and Cyrenaica (first the uncial, then the semuncial, finally a standard that respected the coinage reform of Augustus). What is more, Pupius Rufus, like his predecessor Crassus, struck bronze coinage in three denominations: the as, the semis, and the quadrans. Lastly, starting from Crassus and up to the time of Tiberius all of these denominations had the 20 Ibidem, pp. 119-120 (cat. nos. 40-43, pi. XLIII.8-10). 21A reverse layout, in which the obverse is on the flat side and the reverse is on the convex side can be observed on the specimen uncovered during the Polish excavations in Ptolemais, attributed to the local mint, cf. also: C. H. V. Sutherland, CM. Kraay, Catalogue of Coins of the Roman Empire in theAshmolean Museum, vol. I — Augustus (c. 31 BC-AD 14) (Oxford, 1975), cat. no. 835 (pi. 20). 22 The weight of 7 specimens known to Robinson varied from 2.25 to 3.25 g (BMCQv, p. ccxxvii), while the average weight of 9 specimens of this type known to the authors of RPC, equalled 3.04 g (RPC I, p. 227). 23This standard varies from 14-15 mm (RPC I, ibidem) to 16.5 mm (Buttrey, o.c, p. 35). The quadrans of Pupius Rufus found in Ptolemais is 16 mm in diameter (l a w o r s k i, Walczak, o.c., cat. no. 60, p. 100). 24 The decisive opinion in this matter was stated by B u 11 r e y, ibidem, pp. 34-36, his conclusions were supported in RPC I, p. 226. 25 BMCCyr, o.c., p. ccxxvii. 26Muller, o.c, tabl. V. 189 fixed weight and diameter of the subsequent standards, which, however, tended to decrease.27 Of the coins found in Cyrenaica that were cut in halves or in quarters due to the scarcity of change on the internal market, the much smaller denominations (semisses and quadrantes) were obtained also by dividing large bronze coins (asses) in halves or quarters. This practice is known and described for the western part of the Empire; for the East — including Cyrenaica — studies of this phenomenon are just beginning.28 Of the cut coins of Roman Cyrenaica known to the author, the majority, as much as half of the halves — semisses (18 specimens) and quarters — quadrantes (five spe- cimens), are issues of Pupius Rufus. In comparison with the number of known spe- cimens of the individual denominations of coins minted by this official (according to RPC I: 45 asses, 23 semisses and 9 quadrantes29 which is incomplete but gives a sense of the proportions), these data show what an interesting and important period it was in the history of the local economy. The discovery of an unknown type of quadrans of Pupius Rufus struck in Cyrenaica in 34-31 B.C. has allowed us to supplement the coinage system of this official with an important element it hitherto lacked. Currently it can be stated that this system — a developed one compared to that of other issuers in this region — comprised two basic types of each of the three denominations: the as, the semis and the quadrans. This system, supplemented with the coin analysed in this article, is shown on Fig. 9. Earlier Lollius and Crassus had also issued two types of each denomination. The difference was that the predecessors of Pupius Rufus struck them separately, one for each of the areas constituting the province, Crete and Cyrenaica. They carried the same motives on the obverse and reverse and differed solely in legends: Greek for Cyrenaica and Latin for Crete. The coinage of Pupius Rufus, whose issues bore only Greek legends, was meant exclusively for Cyrenaica. Despite the gradual increase of our knowledge of Cyrenaica in the time of Pupius Rufus, the extent of his coinage is still obscure. The evident differences in the known dies used to strike the individual coins of this official indicate the existence of at least a few issues of each type. Most probably, to a limited extent, minting activity was continued, not only in Cyrene, but also in Ptolemais after Crassus. Also the conclusions made to date about the coinage of subsequent Roman officials in Cyrenaica — Scato, Capito and Palikanus — await verification by archaeological discoveries. Translated by Dorota Dzierzbicka 27 RPC I, p. 226. 28 Currently, partly thanks to the discoveries of Polish archaeologists, it was possible to add a few dozen new specimens to the number of ten-odd cut coins from Cyrenaica known to date. The author is preparing an extensive study on this subject. 29 RPC I, p. 227. 190 Denomination Obverse Reverse As KAITAMIACL nOYIIIOC ANTICTPAT head of Zeus Ammon, r. sella castrensis between fasces; in the field: L-A AYAOC nOYIIIOC L TAMIA2 ANTICTPA head of Zeus Ammon, r. sella castrensis between fasces; in the field: L-A Semis noYnioc ram, r.; beneath belly: L POYTOC TAMIAC sella quaestoria, hasta, sacculus; in the field, r: L ANTICTPA ram, 1.; beneath belly: L nOYIIIOC TAMIAC sella quaestoria, hasta, sacculus; in the field, r: L Quadrans head of Libya nOYIIIOC (or Apollo), r.; behind: L coiled serpent, r. head of Libya (or Apollo), r.; behind: L (?) n-p coiled serpent, r. * Drawings: 1.-5.- L. Muller, Numismatique de l'ancienne Afrique, vol. 1: Les monnaies de la Cyrenai'que, Copenhague 1860; drawing 6. - P. Jaworski Fig. 9. Coins of Pupius Rufus 191 PIOTR JAWORSKI NIEZNANA MONETA PUPIUSA RUFUSA WYBITA W CYRENAJCE (Streszczenie) W ostatnim czasie udaio si$ zarejestrowac w jednym z prywatnych polskich zbiorow monet antycznych dwie brazowe monety nalezace do typu nienotowanego w dotychczasowej literaturze. Ze wzgl^du na niewatpliwe pochodzenie obiektow z obszaru Cyrenajki, jak rowniez na podstawie zredukowanej legendy na rewersie oraz charakterystycznych motywow ikonograficznych, przedstawiane monety naleiy uznac za kwadranse Aulusa Pupiusa Rufusa — urz^dnika rzymskiego, ktory w okresie poprzedzajacym bitwe^ pod Akcjum (31 r. p.n.e.) odpowiadal m.in. za emisjs monet na terenie Cyrenajki. Dotychczas znane byfy dwa typy bitych przez niego asow, dwa typy semisow oraz jeden typ kwadransa. Nieznany wczesniej dragi typ kwadransa, poddany analizie w niniejszym artykule, uzupemia schemat mennictwa Pupiusa Rufusa w Cyrenajce o brakujacy element. Author's address: Instytut Archeologii UW Krakowskie Przedmiescie 26/28 PL — 00-927 Warszawa p.jaworski@gmail.com Indeks 373494 ISSN 0043-5155 Naklad: 300 egz. WIADOMOSCI NUMIZMATYCZNE ROK LIII, 2009 - ZESZYT 2 (188) DOM WVDAWNICZY ieupsai War sz aw a 2009