Some considerations on the circulation of Augustan and Tiberian bronze coins in Gaul, in R. WIEGELS (ed.), Die Fundmünzen von Kalkriese und die frühkaiserzeitliche Münzprägung (Osnabrücker Forschungen zu Altertum und Antike-Rezeption, 3), Möhnesee, 2000, p. 153-170.

Rainer Wiegels (Hrsg.) Die Fundmunzen von Kalkriese und die friihkaiserzeitliche Munzpragung Akten des wissenschaftlichen Symposions in Kalkriese, 15.-16. April 1999 Bibliopolis Johan van Heesch Some considerations on the circulation of Augustan and Tiberian bronze coins in Gaul If we study the loss of coins during the first half of the first century AD it becomes clear that after the massive Augustan injection of bronze coins into Gaul (including the later provinces of lower and upper Germany), the number of coins struck under the following reigns decrease in number (Fig. 1). These data of course have to be studied with caution. Several pitfalls make their interpretation hard. So we always have to keep in mind that coins circulate for a long period. That Celtic coins and imitations circulate side by side with the Roman money. That the length of the reigns differs, from four years for Caligula to more than 20 years for Tiberius, and that gold and silver coins as site finds are rare. Fur- thermore coins were not minted continuously. Although Augustus reigned from 27 BC till AD 14, bronze coins (the majority of the site finds) reaching the north of Gaul were only minted during the 30 years between 16 BC and AD 14, though none between 2 BC and AD 10. Long periods of inactivity can be noted for other reigns also. Claudius struck all his bronze coins in the first three years of his reign and Nero coined no bronze from AD 54 till 64. Keeping in mind all this, it is still apparent that bronze coins from Tiberius onwards are less frequent on most, if not all sites. As far as I know this is true for all of Gaul; this phe- nomenon is illustrated in Fig. I1. Augustus Tiberius Caligula Claudius Nero Neuss (D) 1617 225 189 214 57 Dalheim (Lx) 386 29 63 57 98 Liberchies (B) 302 31 37 45 92 Conde/Aisne (F) 785 62 243 288 1083 Nimes (F) 184 X 19 33 Augst (CH) 965 403 196 299 114 1 Fig. 1: Coins from Augustus to Nero (imitations and forgeries included). What we will try to do in this paper is to have a closer look just at this transitional period between the reigns of Augustus and of Tiberius. We will compare several sites from all over Gaul and try to quantify the importance of the successive coin issues. When precisely and why did the amount of bronze coins decreased? How and when did these coins reach the northern frontier zone? All this is of some importance as it has consequences for the dating of sites that were only occupied for a short time. Before looking at the material and trying to formulate some conclusions I will give a short survey of the late Augustan and some of Tiberius' coinages and their chronology. Chronology Archaeologists often think of coins as well dated objects. This however is not always the case. The dating of Greek, Celtic or Roman republican coins sometimes varies considerab- ly. Take for example the Greek bronze coins of Sicily. In the main reference works some of Data: Chantraine 1982 (Ncuss), FMRL (Dalheim), doc. coin cabinet Brussels (Liberchies), Giard 1968 and 1969 (Conde sur Aisne), M. Amandry in Fiches/Veyrac 1996, 102 (Nimes), Peter 1996 (Augst). 154 Some considerations on the circulation of Augustan and Tiberian bronze coins in Gaul these are only very approximately dated i.e. between 212 and 49 BC! The same difficulty exists with Celtic coins, for example some potin coins were recently redated from 50 BC to 120 BC. Strictly speaking then Roman imperial coins form a well dated category indeed. As most of them have the emperor's head and name on the obverse their attribution to a limited period of time is not difficult at all, but dating them to a precise year is often harder. On Julio-Claudian coinages precise chronological references such as the fribunicia potestas are often absent; and those elements that are present, for example the imperatorial saluta- tions, are difficult to date. One can of course argue that it is of little or no importance to have a closely dated sequence of coins, but for any historian - even if the actual social relevance of his research is fairly limited - a good chronology remains necessary to under- stand and relate archaeological, historical, epigraphic and numismatic data to each other. In two earlier papers I discussed the dates of the main western bronze series of Augustus". Here I will focus on the last bronze issues of Augustus minted at Lyons and one issue of Tiberius. It is certain that during the early years of the Augustan campaigns in Germany the main Roman bronze series used by the soldiers were those struck in Nimes and Lyons. In the earliest camps from the time of Drusus (died 9 BC), such as Oberaden, Rodgen and Dang- stetten, the coins of Nimes (c. 16/15-c. 8 BC)3 dominate. In somewhat later camps such as Haltern, the well known asses of the first bronze issue of Lyons (c. 7-c. 3 BC) abound4. At the very end of Augustus' reign a second Altar-series was struck at Lyons5. Although the reverse of the coins remains almost identical to the first series, the obverses are com- pletely different6. One type has the head of Augustus styled Pater Patriae, the other type shows the head of Tiberius and mentions two or three imperial salutations on the obverse (imp v - vii). Four denominations were struck: the sestertius, the dupondius, the as and the semis. The most common one is the as made of copper, the other denominations are less numerous and were made of brass. This series is dated c. AD 9-14 by Sutherland and AD 10-14 by Giard, who suggests that the imp vii issue of Tiberius continues early into the reign of Tiberius as sole ruler7. The following obverse legends occur: - Caesar Avgvstvs divi f pater patriae, head of Augustus, laureate - Ti Caesar Avgvst(i) fimperat(or) v, head of Tiberius, bare and laureate - Ti Caesar Avgvst fimperat vi8 - Ti Caesar Avgvst(i) f' imperat(or) vii, head of Tiberius, laureate 2 van Heesch 1993 and 1996. RIC 155-157 and RPC 523. 4 RIC 230. van Hccsch 1993 (chronology) and Chantraine 1982, 30 for the circulation in the Rhine camps. 5 RIC 231-248. 6 There is however a substantial difference in style between the first and the second altar series. This suggests an interruption in the minting of the two series. For these differences see van Heesch 1992. 7 RIC, p. 57; Giard 1983, 45-47 and 107-11. S The existence of coins with imp vi remains very doubtful (Giard 1983, 45-47). The reading of the specimens (in the Ashmolean and the British Museum) cited by Sutherland (RIC 242-243) is uncertain. Only one as and one semis of this year are known in public collections. Johan van Hccsch 155 Some chronological elements on these coins permit us to determine their dating to a certain extent more precisely: - Augustus is styled pater patriae, an honour conferred to him on 5th February 2 BC, - Augustus died on August 19th AD 14, and - the obverse legends do not yet mention Tiberius augustus (September 17lh AD 14). Giard supposed that the coins of Tiberius imperator vii might also have been struck after the death of Augustus as they were minted in large numbers. This however is highly improbable, for when Tiberius became augustus on September 17th AD 14 he was styled Ti Caesar divi Avg f avgvstvs from that moment onwards. This is exactly the obverse legend that was used on Tiberius' gold and silver issues and the rare Lyons Altar-bronzes (asses and semisses) struck after the death of Augustus9. So the chronological limits for these coins are between 2 BC and AD 14. There are of course other elements that permit us to be more precise. Giard pointed out that the famous Varus countermark (VAR) never appears on the second Altar-series. So it is plausible to place them after the clades Variana of AD 9. Dating these different issues more closely is much harder. The Augustan and the Tiberian- imp v issues are less numerous than the Tiberian-zm/? vii issue (see Fig. 4). The only chronological indication here is the fifth imperatorial salutation of Tiberius. This honorific title is hard to date, but the proposal of Syme based on literary evidence (Dio) looks acceptable. He assigns Tiberius' 5th salutation to AD 9 and links it to the defeat of the Dalmatians exactly one year after the Bellum Panonicum10. Another argument that might allow us to date these series more closely comes from the Augustan bronze coins minted in Rome. This mint stopped issuing bronze coins c. 3/2 BC, almost at the same time as the end of the first Altar-coins. In the last years of Augustus' reign, just as in Lyons, two fairly small issues came out. They have the portraits of Augustus and Tiberius respectively11. These coins are well dated and show the following inscriptions: - Portrait of Tiberius: Ti caesar avgvst f imperator v /pontif tribvn potestate xii, sc = struck between June 26,h AD 10 and June 25lh AD 11. - Portrait of Augustus: imp caesar divi f avgvstvs imp xx / pontif maxim tribvn potxxxiiii, sc = struck between June 26th AD 11 and June 25lh AD 12. I would suggest considering the two Altar-series, the one with Tiberius-/m/? v and the other of Augustus, as series parallel to those of Rome. In that case they should be dated between June 10th and June lllh and between June 11th and June 12,h respectively. If Tiberian coins of Lyons with imp vi really exist, which I strongly doubt12, they must fall in AD 11 or 1213. The last Augustan issue to be dated is the one showing Tiberius-/mp vii. The year of this imperatorial salutation is impossible to determine with exactitude. It must be placed before September I7lh AD 14 (Tiberius augustus) and some time after June ' RIC Tib., 31 and 32; Eden 1995. 10 Syme 1979, 316. 11 RIC 469-471. 12 See note 7. 13 It has to be remembered that if coins arc dated between June 26th AD 11 and June 25lh AD 12 by the tribunician year, the issue could have taken place e.g. only during the few weeks following the 26th of June. It is highly improbable that minting took place continuously and for the whole period. 156 Some considerations on the circulation of Augustan and Tiberian bronze coins in Gaul 26' AD 11, as Augustus is still styled imp xx (= imp v for Tiberius) at that date and his imperatorial salutations go up simultaneously with those of Tiberius (imp vii for Tiberius is imp xxii for Augustus)14. Although the year 12 cannot be excluded, Syme proposed AD 13 or early 14. He sup- poses that when Germanicus went to the Rhine he was under the escort of Tiberius caesar and that 'some exploit or other (perhaps trivial)' not mentioned in our written sources 'earns a salutation and thus heralds a triumph'. 'Velleius and Dio were hastening towards the decease of Augustus. So too was Cornelius Tacitus he says15. Syme proves nothing but his arguments are as good as any, so AD 13/14 for this very last Augustan bronze issue is quite acceptable16. So much for the late Augustan issues. Tiberius' silver and bronze coinages are only partially dated. Most of his silver and his major bronze issue, the famous asses in the name of Divus Augustus, with the altar of Providentia on the reverse, are not dated at all. We will only speak about the so called Providentia-asses, as they alone are numerous on early imperials sites (71% of all Tiberian bronze coins)17. These coins show the radiate head of Augustus and the legend divus Augustus pater on the obverse, and the front wall of an altar-precinct with panelled doors on the reverse accompanied by the legend provident. These so called Providentia-asses are studied in a good and detailed article by Sutherland of 194118. He suggests an extended period of issue from c. AD 22/23 till (?) 30iy. His dates are mainly based on the study of the die-axis and on arguments of style. Giard in his catalogue of the coin cabinet of the French national library on the other hand, accepts the dates of J.-P. Martin. He proposes AD 31-37 and links the coining of these asses with the Providentia of Tiberius20. The dates of Giard and Martin were based on fragments of the acts of the Arval Brethren (fratres Arvales) for the year AD 38, mentioning this altar (in campo Agrippae ad aram Providentiae Augustae) and other inscriptions from AD 32 referring to Tiberius' providence in preventing Seianus taking over the empire (AD 31)21. Since the discovery of the senatus consultum de Cn. Pisone patre (AD 20) we now know however that an altar Providentiae already existed before October 10lh AD 19, or even in AD 1722. It commemorated the providence of Augustus himself, who adopted Tiberius as his heir and arranged the adoption of Germani- cus by Tiberius23. The exact date of the foundation of this altar is difficult to determine. Eck Syme 1979, 316-320. The idea of Schwartz 1945, 36, that Tiberius was still imp vi in AD 13 is based on an erroneous argument. Schwartz bases Ms dating on an inscription of £monfl/Ljubljana (CIL IE 10768) that contains false dates due to a mistake by the stone cutter. There can be no doubt that this inscription belongs to the period after the death of Augustus as Tiberius is styled [Ti caesar avJGVST1 FAVG[. 15 Syme 1979, 317-320. 16 Also accepted by Giard 1983. He follows Schwartz' chronology (Schwartz 1945) and places e.g. Tiberius' triumph ex Pannonis Delmatisque in AD 13 instead of the traditional dating in AD 12. 17 See below but also Rodewald 1976, 149. 18 Sutherland 1941. 19 Same dates in RIC p. 99. 20 Giard 1988, 24. 21 For these inscriptions see Ehrenberg/Jones 1955, 65 n° 51 (Interamna) and 52 (Gortyn ). See also Eck/etalii. 1996, 199. 22 Eck/et alii 1996, 44, lines 83-84 and pages 199-201. 23 Eck/et alii 1996, 200. Johan van Heesch 157 supposes that it happened during the last years of Augustus' reign. Whatever the correct date was, it becomes fairly evident that a logical start for the Providentia-issti.es with the portrait of divus Augustus would be at the beginning of Tiberius' reign24. For how long these asses were struck is difficult to determine. It is not to be excluded that several issues were minted during the whole reign of Tiberius. This might explain why some of these coins show early-Tiberian, mid-Tiberian and late-Tiberian portraits as Sutherland mentioned more than 50 years ago25. The main problem with these coins is that quite a lot of them are local imitations, some of which are in fairly good style. These local coinages, imitating not only the Providentia-coins, but also the Augustan Altar-asses, were still pro- duced under Claudius26. It is not clear at all how many local mints were active; nor do we know anything about their status. All that can be said is that the Roman government tolerated these coinages. Circulation Let us now look at the circulation of these coins in Gaul. We will start with two special sites (Fig. 2): Velsen I, a Roman camp in the Netherlands, presumably constructed under Germanicus and dated between AD 15-30 by Bosman27, and Aulnay de Saintonge in the south-west of France, also a military camp dated between AD 10/14-3028. The chronology of both camps is not very certain and it would of course be worthwhile to confront the nu- mismatic data with the pottery. Velsen I Aulnay Celtic 1 Augustus -NimesI(16-8BC) 1 3 - Nimes III (AD 10-14) 2 - Lyons I (7-3BC) 18 1 - Lyons H(AD10-14): MB 7+1/2 24 HS 1 1 Sm 5 33 -Rome I (16-15BC) 2 1 - Rome II (3/2 BC) 16 2 Tiberius - Providentia 10 3 - Others 2 Caligula 9 Claudius 1 Fig. 2: Coins found at Velsen I and Aulnay de Saintonge (coins not attributed to one particular series were excluded). Also suggested by Eck/et alii 1996, 201. Sutherland 1941, 111. For a set of interesting imitations ovcrstruck on coins of Caligula and others see von Kaenel 1972, 113-115 and Giard 1988, 24. See also Kraay 1962, 34 with plate 4 and 5 and Giard 1975. Bechert/Willems 1995, 99 and Bosman 1997, 247-271. I find the dating of this camp fairly strange as 9 coins of Caligula (AD 37-41) were also found (coins for Agrippa included). Tassaux 1983/1984. 158 Some considerations on the circulation of Augustan and Tiberian bronze coins in Gaul Both camps are situated in completely different regions of Gaul. Velsen I apparently re- flects the circulation pattern in the Rhine region for camps from the period Augustus-Ca- ligula. The pattern is not exceptional for the region: of the Altar-series Lyons I dominates, there are a lot of moneyers' asses of Rome, most of them with the early Tiberian counter- mark Caesar in ligature, and P rovidentia-usses^'. The finds from Aulnay are completely different. There are almost no coins of the first Altar-series, the second issue of Lyons dominates, and the number of semisses with the altar is impressive. In what follows we will concentrate on a few aspects of the bronze circulation only. We will have a closer look at the relation between Lyons I and II, the quantitative aspects of the different Lyons II issues, the phenomena of halving coins, and the circulation of the Providentia-coins and the moneyers' asses. Sites Lyons I/n Lyons I/n actual number Percentage Neuss (D) 301/141 68/32 X an ten (D) 46/11 81/19 Velsen 1 (N) 18/7 72/28 Titelberg (Lx) 86/32 73/27 Tongeren (B) 74/29 72/28 Braives (B) 11/8 58/42 Liberchies (B) 98/49 67/33 Augst (CH) 159/164 49/51 Vindonis.sa (CH) 320/526 38/62 Zurzach (CH) 16/21 43/57 Monterean (F) 55/70 44/56 Conde/Aisne (F) 75/228 25/75 Viuz-Faverge (F) 4/6 40/60 Roanne (F) 9/9 50/50 Gtanum (F) 5/7 42/58 Albias (F) 1/7 13/87 Cahors (F) 2/5 29/71 Aulnay (F) 1/24 4/96 Fig. 3: The proportion of Lyons I \o Lyons II coins. Only asses and dupondii. Sestertii, semisses and halves were excluded. Fig. 3 gives an idea of the representation of the Altar-bronzes at different sites in Gaul . Lyons T is the series struck between 7 and 3 BC; Lyons II are the bronze coins issued bet- ween AD 10 and 14. The ratio between the two series is expressed in actual numbers and in percentages. All of these sites are supposed to be at the very latest Tiberian, but most are 29 For this pattern see Chantiaine 1982, 30-31 and van Hccsch 1996. See also the very interesting tables in Wells 1972, 268-269. 30 Data: Neuss, Chantraine 1982; Xanten, Hanel 1995; Velsen I, Bosnian 1997; Titelberg, FMRL (1- 4); Tongeren, Braives and Liberchies, van Heesch 1996 and unpublished documentation coin cabinet Brussels; Augst, Peter 1996; Vindonissa, Kraay 1962; Zurzach, Dopplcr 1994; Montcreau, Giard 1978: Conde sur Aisne, Giard 1968-1969; Viuz-Faverges, Amandry/Remy 1997; Roanne, Remy 1985; Glanum, Brenot/Callu 1978; Albias and Cahors, Depeyrot 1985; Aulnay de Saintonge, Tassaux 1983/84. Johan van Heesch 159 Augustan. A very clear pattern emerges from this table. Lyons I dominates the finds in the camps on the lower Rhine (Neuss etc.) and the immediate hinterland (Titelberg-Liberchies). Roughly taken these sites are all situated in the area that was of the utmost importance for the military actions during the campaigns from Drusus to Germanicus (13 BC-AD 16). At all the other sites from Vindonissa and Augst to the centre and the south of France it is the second Altar-scries that outnumbers the first one. The first question one asks is: is the pattern on the lower Rhine reflecting the real output of the mint of Lyons or not? If so, why then is Lyons I less frequent in the centre of France, e.g. in Augst, one of the oldest colo- nies of Gaul? Although Lyons I is not rare in central Gaul, as is shown by the famous votive deposit of Villeneuvc-au-Chatelot ending c. AD 9, the abundance of the second series cannot be doubted and could be illustrated by numerous other examples. Or should we consider the pattern of central Gaul (e.g. Roanne just north of Lyons) and Augst as reflecting the real output of the mint? Before trying to answer this, let us first look at our other data. LYONS 11 (AD 10-14) Sites Augustus Tiberius Total of coins ad 10-1 1 imp v imp vii ad 11-12 ad 13-14 % % % Vindonissa (CH) 16 13 71 347 Neuss (D) 17 27 59 62 Xanten (D) 15 0 85 13 Tilclberg (Lx) 3 3 94 38 Braives (B) 0 20 80 5 Liberchies (B) 55 5 40 20 Montereau (F) 24 19 57 16 Conde/Aisne (F) 32 11 57 53 Roanne (F) 0 0 100 5 Aulnay (F) 9 0 91 11 Fig. 4: Only asses and dupondii (halves included in totals). In Fig. 4 we take a closer look at the second Altar-series. It should be noted that we took into account only coins that were attributed to one of the series. So the total number of usable coins often became rather small. Anyhow one general pattern emerges and the most numerous group, be it on a site in the north, on the middle Rhine or in central France, is always the imp v/i-issuc. The Augustan issue represents only 17%, the imp v-coins 14% and the last issue with imp vii an overwhelming 70%31. Altar-coins of Tiberius augustus are extremely rare, and we will have to answer why imp vii was struck in huge quantities and why Lyons almost stopped minting bronze coins after 14 AD. 31 The 55% of coins of Augustus at Liberchies arc an anomaly possibly due to the large number of worn imp v or imp vii coins that were not attributed to a specific series, consequently they do not figure in our statistics. 160 Some considerations on the circulation of Augustan and Tiberian bronze coins in Gaul Percentage of halved coins Sites Lyons I Lyons II Rome 11 Provid. 7-3 BC AD 10-34 3-2 BC AD 34-? Neuss (D) 19 12 7 [ Xanten (D) 6 0 0 0 Velsen I (N) 0 13 0 0 Titetberg (Lx) 23 22 13 0 Tongeren (B) 1 3 0 0 Braives (B) 0 11 0 Liberchies (B) 6 4 0 0 Augst (CH) 26 15 11 2 Vindonissa (CH) 33 18 5 3 Montereau (F) 0 0 0 0 Conde/Aisne (F) 0 0 0 0 Viuz-Faverge (F) 0 0 0 0 Roanne (F) 0 0 0 0 Glanum (F) 0 0 0 0 Albias (F) 0 0 0 Cahors (F) 0 0 0 Aulnay (F) 0 0 0 j 0 Fig. 5; Halved coins in Gaul and Germany: percentages. In Fig. 5 we present the differences in the frequency of halving of several coin series: be- sides the Lyons issues, we show the data for the second issue of moneyers' asses from Rome (probably minted 3-2 BC) and the so called Providentia-bronzes of Tiberius. The NJmes coins, very frequently halved, are excluded from this table. Although halved coins are often more difficult to identify than others and the number of well identified coins is often reduced, some clear differences appear. Halved coins of these series are apparently veiy rare or absent once we leave the mili- tary zones. Although halved coins from Nimes are very common in central Gaul, the issues of Rome and Lyons were rarely halved there (this is confirmed by hoards like that of Ville- neuve were of the 558 coins from Lyons I only 5 were halved [0.8%]). When we turn our attention to the Rhine area it is clear that the Tiberian Providetuia-coins. arc rarely halved; the same can be said of the moneyers' asses32. Although the coins of the moneyers' were struck under Augustus, it is generally agreed that the majority of these coins arrived in Gaul under the reign of Tiberius'11. The highest proportions of halved coins come from military sites on or near the Rhine area. It is strange that the percentages for Lyons I are almost always higher than those for Lyons II. This, I suppose, can be explained by the fact that their circulation period was longer and that the chance that they were halved was higher. Not so in Pannonhr, e.g. Bjwna/Ljubljana, 58 moneyers' asses and 27 halves! Kos 1986 35-36. 13 Kraay 1962, 8. Johan van Heesch 161 Sites TlBERIAN BRONZE COINS Pt'a\)j/i /rpct TihpniK nnmhpTX ii li 111 uvi a / t UvlLlCfll ltl JJCICCllltlgCN Neuss (D) 170/24 i-■■-=-=- 88 Xanten (D) oil IJ Velsen I (N) 10/0 100 Titelberg (Lx) _£2—1_L- 9/3 75 Tongeren (B) 5/0 100 Braives (B) 3/0 100 Liberchies (B) 7/4 64 Augst (CH) 267/59 82 Vindonissa (CH) 596/226 73 1 Zurzach (CH) 16/5 76 Montereau (F) 16/10 62 Conde/Aisne (F) 17/10 52 Viuz-Faverge (F) 4/0 100 Roanne (F) 4/2 67 Glanum (F) 11/14 44 | Aulnay (F) 3/2 60 Fig. 6: Tiberian bronze coins in Gaul and Germany. Let us turn to Tiberius' reign now. Fig. 6 gives the percentage of the Providentia-asses to the rest of Tiberius' bronze coins. It is clear that this was Tiberius' major bronze series (76%) but, as pointed out before, we should keep in mind that numerous imitations circu- lated and that the statistics might be slightly influenced by this. Sites Lyons U/Providentia Lyons U/Providentia actual numbers percentages Neuss (D) 141/170 45/55 Xanten (D) 11/6 65/35 Velsen I (N) 7/10 41/59 Titelberg (Lx) 32/9 78/22 Braives (B) 8/3 73/27 Liberchies (B) 49/7 88/12 Augst (CH) 164/267 38/62 Vindonissa (CH) 526/596 47/53 Zurzach (CH) 21/16 57/43 Montereau (F) 70/16 81/19 Conde/Aisne (F) 228/17 93/7 Viuz-Faverge (F) 6/4 60/40 Roanne (F) 9/4 69/31 Glanum (F) 7/11 39/61 Aulnay (F) 24/3 89/11 Fig. 7: The ratio between Lyons II and Providentia asses. Fig. 7 compares these Pi-ovidentia-asses with Lyons II. No very clear pattern emerges. In most cases Lyons II is more numerous but in some, mostly military sites such as Neuss, 162 Some considerations on the circulation of Augustan and Tiberian bronze coins in Gaul Velsen and Vindonissa, Providentia surpasses the Lyons IT coins. For Velsen and Vindo- nissa at [east this could be explained by the date of their main occupation (Tiberian). Sites Lyons IT/Moneyers IT Actual numbers Lyons II/Moneyers II percentages Neuss (D) 141/174 45/55 Xanten (D) 11/14 44/56 Velsen I (N) 7/16 30/70 Titelberg (Lx) 32/14 70/30 Tongeren (B) 29/20 59/41 Braives (B) 8/0 Liberchies (B) 49/14 78/22 Augst (CH) 164/56 75/25 Vindonissa (CH) 526/494 52/48 Zurzach (CH) 21/15 58/42 Montereau (F) 70/8 90/10 Conde/Aisne (F) 228/20 92/8 Viuz-Faverge (F) 6/1 86/14 Roanne(F) 9/1 90/10 Cianum (F) 7/3 70/30 Aulnay (F) 24/2 92/7 Fig. 8: The ratio between Lyons II and moneyers' asses. In a last table (Fig. 8) we compare Lyons II with the second series of the moneyers' asses. That most of these coins arrived early in the reign of Tiberius seems acceptable, as they are rarely halved and since the countermark Caesar in ligature (often found on the lower Rhine) is early Tiberianj4. It is striking that these moneyers' coins are especially numerous on the military sites and less in civilian settlements. Compare Neuss, Xanten, Velsen I with Liberchies or Vindonissa and Zurzach with Augst. The moneyers' coins are very rare in central Gaul (Montereau, Roanne, Conde etc). Conclusions How then should we understand all these data, and what do they contribute to our know- ledge of the monetary circulation in Gaul? This of course is the most difficult part of my contribution and I should like to stress that I am not only treading on dangerous ground, but also that my explanations will be mainly hypothetical. If we looked for proof my text would end here. The ratio Lyons I/Lyons II The first interesting phenomenon is that the Altar-I series of Lyons was much more nume- rous on the lower Rhine camps and their immediate hinterland, than in the rest of Gaul and in camps situated on the upper Rhine area. If the lower Rhine-pattern reflects the actual output of the issues, this must mean that Lyons I coins were mainly sent to the lower Rhine and were only sparingly put into circulation elsewhere. If on the other hand the original output of Lyons I and II was equal, as is suggested by the finds in France, then larger quan- tities were send to the lower Rhine area than elsewhere during the first period, or Lyons II Germanicus? Mac Dowall 1992, 48. Johan van Heesch 163 coins arrived there only in smaller quantities. These last options seem less probable, especially as the number of legions on the Rhine increased. Why should the need for bronze coins diminishes in one area only? In my view, the output of Lyons I was much higher than the output of Lyons II. Sites such as Vindonissa and Zurzach have less Lyons I coins as their occupation starts later and the area was not really supplied with Lyons bronzes bet- ween 7-3 BC. Most of these coins were shipped directly to the lower Rhine region to be put into circulation in the military camps. Even near Lyons and in central France, Lyons I re- mains less common than on the Rhine. Coin-drift This brings us to the very important subject of coin-drift. Were coins transported directly from the mint to the military camps on the frontier, or were they put into circulation in the town of minting or in a very limited range of administrative centres? In the later case their dispersion was gradual and coins spread - so to speak - from hand to hand over a longer period of time. City-coinages, especially abundant in Spain and the eastern Mediterranean, served in most cases local circulation only. Larger coinages however, in this case those of the Roman state, were certainly not exclusively put into circulation in their place of minting. Coins issued in Rome or in Lyons, the main mint of the Julio-Claudians for gold and silver33, were sent to the areas where troops were stationed as state expenditure, especially army- pay, was the main mechanism through which coinage came into circulation36. That even bronze coins were transported to the coin consuming regions seems only logical. Direct shipments of bronze coins to remote comers of the empire can be proved for several coin series. Some well known examples are the brass asses of Trajan and Hadrian struck in Rome and shipped to Antioch in Syria and some Alexandrian tetradrachms of Scverus Alexander, struck in Rome and shipped to Egypt37. Quite interesting also is the hoard of die-linked Claudian sestertii from Pobla de Mafumct in Spain38. Several other cases in the work of Hobley show that the examples cited above were no exceptions39. Once put into circulation bronze coins did not travel much more and stayed in the general area were they were distributed. This is clearly shown by the limited circulation of the countermarked coins; most have their own "territory*' as can be seen from the maps published by Kraay and Berger40. Several second century examples of the immobility of aes coins are known, for example the famous Britannia-asses of Hadrian and Antoninus Pius rarely found outside of Britain41. Of course coins will have travelled in the purses of the soldiers and in the money chests of the army officers, but I doubt if this really changed the circulation pattern in any area. The influence of trade and travellers on the dispersion of coinage will have been minimal also. Merchants probably spent their profits on local products which they took back with them to sell them in their hometowns. 35 36 Mattingly 1985, 256-57. For the Roman state budget and the importance of military pay see Marquardt 1884, 77-148; Duncan-Jones 1994, 33-46 and Harl 1996, 207-249. Carradice/Cowcll 1987 and Burnett/Craddock 1983. Campo/et alii 1981. Hobley 1998, 130. Kraay 1956 and Berger 1996, 49-55. Duncan-Jones 1990, 39-44 and Walker 1988. 164 Some considerations on the circulation of Augustan and Tiberian bronze coins in Gaul Augustan-Tiberian examples of direct shipments of coins to the military areas can be recognised in the presence of the Mmes I series on the Rhine. They were almost the only bronze currency used in very early camps such as Oberaden, Rodgen, Dangstetten etc.42. The same phenomenon can be seen with the moneyers' asses sent to the Rhine frontier in the early years of the reign of Tiberius43. It is hard to see this only as the result of troop movements from Italy towards the Rhine. Most of these moneyers' asses were counter- marked in the Rhine camps, which suggests to me that they arrived 'en masse' and not in individual purses. It is striking that these coins hardly spread into the hinterland or to the centre of Gaul. I find it difficult to believe that the distribution of bronze coins was exclusively in the hands of the nummularii or money-changers who could make a profit in changing silver for bronze as suggested by Hobley44. Nummularii are not particularly numerous in the frontier zone45, and I find it hard to accept that the shipment of huge quantities of heavy bronze coins could have led to any significant profit. Why should gold, silver and bronze be trans- ported to the army through different channels? I think that, during the early empire, most coins, in all metals, were put into circulation through the military "pay masters". The direct link of bronze coins and the military can not only be observed in the Augustan military camps, Marcus Peter has clearly demonstrated that the departure of the soldiers from Vin- donissa m AD 101 caused a drop in the provision of bronze coins in the region (but this perhaps proves little about the way the coins arrived)46. As has been stated before, it cannot be doubted that gold and silver coins, the main currency of the Romans, were transported to the military camps and I see no reason to be- lieve that this did not happen to the bronzes also. But as we saw earlier, coins were not continuously minted by the Roman authorities. In some periods only gold and silver was coined, during others only bronze. Often the mint did not issue any coin at all. This means that the Roman pay-masters were able to pay their expenses out of the reserves of the trea- sury, which must have been filled with old and recent coins. The proportion of new coins to old in state payments is difficult to determine, but it seems less probable that huge quanti- ties were minted just to be stored in the treasury. This especially must be true for bronze coins as these are not really suited to the storage of wealth. Why should a mint - even when manned by slaves - have made the effort and incurred the costs of striking bronze if there was no direct need for these coins? As one aureus equals 400 asses, it would be unbe- lievable that they did spend 400% more time and energy on the production of bronze just to keep these coins in the treasury for future expenditure. If this is correct it is reasonable to suppose that most of the newly minted bronze coins were spent in a short period of time following their issue. There is no reason then to believe that coin-drift played any important role in the supply of coins to the military zone. It will not have taken much more than a few weeks for newly minted coins to arrive on the Rhine. If we look at the speed of overland military expeditions between Lombardy and the low Countries in the 16th century, we see that the average speed for a journey of c. 1100 km was 48 days or nearly 7 weeks. The highest speed recorded was 42 Chantraine 1982, 30. 43 Wiggl997, 284; Kraay 1962, 8. 44 Hobley 1998, 139-40 (but correct his reference to Martial 12:57 8 in Suetonius, Galba 9.2). 45 Andrcau 1987, 318-20. 46 Peter 1996/2, 316-18. Johan van Heesch 165 32 days . Any coin transport in Roman times could have done the same voyage (e.g. Lyons-Nijmegen) in the same time or perhaps even faster as the roads were in better condi- tion than in 16th century Europe. Variations in the output of bronze coins We saw that especially the imp vii coins of the Altar type were very common (70% of Lyons II, see Fig. 5). Why is one issue (e.g. imp v) less common than the other (imp vii)? It is tempting of course to link huge coin issues with extra expenditure such as large military operations. This however must not always be the case. Whether at war or not, the Roman soldiers had always to be paid. The imperial army at war did not necessarily need more money as the pay remained the same. One reason why the Roman mint from time to time coined bronze and on other occasions only gold and silver might have been the need to provide "the market" or the camps with "coppers". But another explanation can be pro- posed. As the Roman government during the first century BC and under the Julio-Claudians rarely bothered about the provision of small change (e.g. see the emergency issues under Claudius) it is possible that paying a part of the soldiers' salaries in bronze was simply a way of economising on gold and silver. In periods of a shortage of precious metals the government struck bronze coins even if the effort to produce and to transport them was much higher48. The opposite situation, a rise in the availability of gold and silver, might then explain why bronze decreased under Tiberius. We know that the exploitation of several Spanish gold mines goes back to the reign of Tiberius49, and the large numbers of Tiberian gold coins found in Gaul contrasts to the rarity of his bronzes50. The minting of bronze, iron or billon coins instead of precious metals in cases of emer- gency or shortages is described by Aristotle in his Economics, it is also considered normal practice by Oresme, the 14lh century writer of the Tractatus de origine, natura, jure et mu- tationibus monetarum51. The abundance of the imp v/i-coins of Lyons (AD 13-14) might then, hypothetically, be explained by a temporary shortness in precious metals or the desire to economise on those metals. When Tacitus speaks about the troubles that arouse on the Rhine after the death of Augustus (AD 14), he mentions the financial grievances of the revolting troops. The soldiers in Pannonia and in Germany wanted a pay raise from 10 asses per day to one denarius (16 asses)52. Could it then be that they also insisted on being paid in precious metals instead of (partly) in bronze coins? The switch in words by Tacitus from 'asses' to 'denarius' might also imply this. It is in any way striking that the minting of bronze at Lyons almost ended from this point onwards. Gold and silver, on the other hand, were coined in important quantities there till the reign of Nero53. Parker 1972, 280. For a very interesting comment on coin-drift see now Wigg 1999, 340-342. One golden aureus equals 400 asses (7 g versus 4 kg !). Domergue 1990, 198-200; Andreau 1990, 86, 91. Callu/Loriot 1990, 82. Jones 1993, 363-4 texts nos. 536-41 (with references to Aristotle, Economics, II, ii, 16b, 20, 23 and to Polyaenus Stratagems HI, x, 14 and IV, x, 2). Oresme, Tractatus de origine, natura, jure el mutationihus monetarum, II, 9-10 and III, 4. Tacitus, Annals, xvii, 6-8 and xxxi, 5. Giard 1988. See also Strabo's attestation of gold and silver minting at Lugdunum under Tiberius, Strabo iv, 3, 2; on this passage see Sutherland 1976, 46-47. 166 Some considerations on the circulation of Augustan and Tibcrian bronze coins in Gaul To sum up: we believe that coins, especially those in bronze, were minted for immediate use. Military salaries were also paid in old coins, but whenever new coins were issued they could reach the troops in a few weeks of time. The phenomenon of coin-drift hardly in- fluenced the coin circulation in this early period on the Rhine; it might have been of more importance in other regions as in the centre of Gaul, where permanent military camps were a rarity and direct state expenditure was limited54. h ] ©0° c 0 ) 1 © ) © © © © © © ® © Map: Sites mentioned in the tables. 1. Velsen ; 2. Xanten; 3. Neuss; 4. Tongeren; 5. Braives; 6. Liberchies; 7. Titelberg; 8. Conde-sur-Aisne; 9. Montereau; 10. Augst; 11. Zurzach; 12. Vindonissa (Windisch); 13. Viuz-Faverges; 14. Lyons; 15. Roanne; 16. Aulnau de Saintonge; 17. Cahors; 18. Albias; 19. Nimes; 20. Glanum We should like to thank Dr. David Wigg for his collaboration. Johan van Heesch 167 Abbreviations FMRL RIC RPC Bibliography Amandry/Remy 1997 Andreau 1987 Andreau 1990 Becher/Willems 1995 Berger 1996 Bosman 1997 Brenot/Callu 1978 Burnett/Craddock 1983 Callu/Loriot 1990 Campo/etalii 1981 Carradice/Cowell 1987 Chantraine 1982 R. Weiller, Die Fundmiinzen der romischen Zeit im Grossherzogtum Luxemburg 1 - 5 (Berlin 1972-1996). C.H.V. Sutherland, The Roman Imperial Coinage I: From 31 BC to AD 69 (London 1984). A. Burnett/M. Amandry/P.P. Ripolles, Roman Provincial Coinage I: From the death of Caesar tot the death of Vitellius (44BC - AD 69) (Paris/London 1992). M. Amandry/B. Remy, Les monnaies de fouilles du sanctuaire de Viuz-Faverges (Haute-Savoie) (Faverges 1997). J. Andreau, La vie financiere dans le monde romain. Les metiers de manieurs d'argent (lYe siecle av. J.-C. - Hie siecle ap. J.-C). Biblio- theque des ecoles franchises dAthenes et de Rome 265 (Rome 1987). J. Andreau, Recherches recentes sur les mines a l'epoque romaine, II, Nature de la main d'oeuvre; histoire des techniques et de la produc- tion. Rev. num. 1990, 85 ff. T. Bechert/J.H. Willems, De Romeinse Rijksgrens tussen Moezel en Noordzeekust (Utrecht 1995). F. Berger, Kalkriese 1. Die romischen Fundmiinzen. Romisch- Germanische Forschungen 55 (Mainz am Rhein 1996). A. Bosman, Het culturele vondstmateriaal van de vroeg-Romeinse versterking Velsen I. Academisch proefschrift Universiteit Amsterdam (Amsterdam 1997). C. Brenot/J.-P. Callu, Monnaies de fouilles du sud-est de la Gaule (VF s. av. J.-C.-Vr s. ap. J.-C). Univ. Paris X Nanterre. Centre de recherches sur l'antiq. tardive et le haut Moyen-Age. Cahier 3 (Paris 1978). A. Burnett/P. Craddock, Rome and Alexandria: the minting of Egyp- tian tetradrachms under Severus Alexander; Museum Notes 28, 1983, 109ff. J.-P. Callu/X. Loriot, L'or monnaye II. La dispersion des aurei en Gaule romaine sous l'empire. Cahiers Ernest-Babelon 3 (Juan-les-Pins 1990). M. Campo/J.-C. Richard/H.-M. von Kaenel, El tesoro de la Pobla de Mafumet (Tarragona). Sextercios y dupondios de Claudio I (Barcelona 1981). I. Carradice/M. Cowell, The Minting of Roman Imperial Bronze Coins for Circulation in the East: Vespasian to Trajan. Numismatic Chronicle 147, 1987, 26 ff. H. Chantraine, Novaesium VIII. Die Antiken Fundmiinzen von Neuss, Gesamtkatalog der Ausgrabungen 1955-1978. Limesforschungen 20 (Berlin 1982). 168 Some considerations on the circulation of Augustan and Tiberian bronze coins in Gaul Depeyrot 1985 Domergue 1990 Doppler 1994 Duncan-Jones 1990 Duncan-Jones 1994 Eck/et alii 1996 Eden 1995 Ehrenberg/Jones 1955 Fiches/Veyrac (Eds.) 1996 Giard 1968/69 Giard 1975 Giard 1978 Giard 1983 Giard 1988 Hanel 1995 Harl 1996 van Heesch 1992 van Heesch 1993 van Heesch 1996 G. Depeyrot, Les monnaies antiques des departements des Hautes- Pyrenees, de la Haute-Garonne, du Tarn-et-Garonne et du Lot (Soreze 1985). CI. Domergue, Les mines de la Peninsule iberique dans Fantiquite romaine. Collection de l'ecole francaise de Rome 127 (Paris/Rome 1990). W. Doppler, Die Miinzen aus Kastell und Kastell-Vicus. In: R. Hanggi / C. Doswald/K.Roth-Rubi (Eds.), Die friihen romischen Kastelle und der Kastell-Vicus von Tenedo-Zurzach. Veroffentlichungen der Ge- sellschaft pro Vindonissa 11 (Basel 1994) 361 ff. R. Duncan-Jones, Structure and Scale in the Roman Economy (Cam- bridge 1990). R. Duncan-Jones, Money and government in the Roman Empire (Cambridge 1994). W. Eck/A. Caballos/F. Fernandez, Das senatus consultum de Cn. Pisone patre. Vestigia 48 (Munchen 1996). T.G. Eden, The re-emergence of an ''altar of Lugdunum" as from the reign of Tiberius. Numismatic Circular 103-4, 1995, 142. V. Ehrenberg/A.H.M. Jones, Documents illustrating the reigns of Augustus & Tiberius (Oxford 19552). J.-L. Fiches/A. Veyrac (Eds.), Carte archeologique de la Gaule. Nimes 30/1 (Paris 1996). J.-B. Giard, Le pelerinage gallo-romain de Conde-sur-Aisne el ses monnaies. Rev. num. 1968 and 1969, 76 ff. and 62 ff. J.-B. Giard, La penurie de petite monnaie en Gaule au debut du Haut- Empire. Journal des Savants 1975, 81 ff. J.-B. Giard, La trouvaille de Montereau. Bulletin du Groupement archeologique de Seine-et-Marne 18-19, 1977-78, 62 ff. J.-B. Giard, Le monnayage de l'atelier de Lyon. Des origines au regne de Caligula (43 avant J.-C - 41 apres J.-C). Numismatique Romaine 14 (Wetteren 1983). J.-B. Giard. Bibliotheque nationale. Catalogue des monnaies de 1'empire remain, II, de Tibere a Neron (Paris 1988). N. Hanel, Vetera I. Die Funde aus den romischen Lagern auf dem Furstenberg bei Xanten. Rheinische Ausgrabungcn 35 (Koln/Boim 1995). K.W. Harl, Coinage in the Roman Economy, 300 B.C. to A.D. 700 (Baltimore/London 1996). J. van Heesch, Note sur la representation de l'autel de Lyon sur les monnaies d'Auguste et les imitations. Bull. Cercle Etudes Num. 29, 1992, 81-84. J. van Heesch, Proposition d'une nouvelle datation des monnaies en bronze a l'autel de Lyon frappees sous Auguste. Bull. Soc. Francaise Num. 48-4, 1993, 535 ff. J. van Heesch, Les monnaies augusteennes sur- quelques sites beiges. Contribution a 1'etude de la chronologie de I'occupation romaine du nord de la Gaule. In: Lodewijckx (Ed.) 1996, 95 ff. Johan van Heesch 169 Hobley 1998 Jones 1993 von Kaenel 1972 Kos1986 Kraay 1956 Kraay 1962 Lodewijck (Ed.) 1996 Mac Dowall 1992 Marquardt 1884 Mattingly 1985 Parker 1972 Peter 1996/1 Peter 1996/2 Remy 1985 Rodewald 1976 Schwartz 1945 Sutherland 1941 Sutherland 1976 Syme 1979 Tassaux 1983/1984 A. S. Hobley, An Examination of Roman Bronze Coin Distribution in the Western Empire A.D. 81-192. BAR International Series 688 (Ox- ford 1998). J.R.Melville Jones, Testimonia numaria. Greek and Latin Texts con- cerning Ancient Greek Coinage (London 1993). H.-M. von Kaenel, Die Fundmiinzen aus Avenches. Schweizerische Num. Rundschau 51, 1972,47 ff. P. Kos, The monetary circulation in the southeastern Alpine region ca. 300 BC-AD 1000. Situla 24 (Ljubljana 1986). C. Kraay, The behaviour of early imperial countermarks. In: R. Car- son/C. Sutherland (Eds.), Essays in Roman coinage presented to Harold Mattingly (Oxford 1956) 113 ff. C. Kraay, Die Miinzfunde von Vindonissa (bis Trajan). Veroffentli- chungen der Gesellschaft Pro Vindonissa 5 (Basel 1962). M. Lodewijckx (Ed.), Archaeological and historical aspects of West- European societies. Album amicorum Andre Van Doorselaer. Acta Arch. Lovaniensia Monogr. 8 (Leuven 1996). D. W. Mac Dowall/ A.V.M. Hubrecht/W.J.A. De Jong, The Roman Coins: Republic and Empire up to Nerva. Description of the Collec- tions in the Provinciaal Museum G.M. Kam at Nijmegen 12 (Nij- megen 1992). J. Marquardt, Romische Staatsverwaltung 2 (Leipzig 18842). H.B. Mattingly, Money for an Empire: the Julio-Claudian Experiment. Num. Chron. 145, 185, 255 ff. G. Parker, The Army of Flanders and the Spanish Road 1567-1659. The Logistics of Spanish Victory and defeat in the Low Countries' Wars (Cambridge 1972). M. Peter, Augusta Raurica. Inventaire des trouvailles mon&aires suisses 3 and 4 (Lausanne 1996). M. Peter, Bemcrkungen zur Kleingeldversorgung der westlichen Provinzen im 2. Jahrhundert. In: C. King/D. Wigg (Eds.), Coin finds and coin use in the Roman world. Studien zu Fundmiinzen der Antike 10 (Berlin 1996). B. Remy, Les monnaies romaines decouvertes a Rodumna (Roanne, Loire). Essai de circulation mon&aire (Lyon 1985). C. Rodewald, Money in the age of Tiberius (Manchester 1976). J. Schwartz, Recherches sur les dernieres annccs du regne d'Auguste (4-14). Rev. de Philologie 19-1, 1945, 21 ff. C.H.V. Sutherland, Divus Augustus Pater. A study in the aes coinage of Tiberius. Num. Chron. 6.ser., 1, 1941, 97 ff. C. H.V. Sutherland, The emperor and the coinage. Julio-Claudian studies (London 1976). R. Syme, Some imperatorial salutations. Phoenix 33, 1979, 308 ff. D. and F. Tassaux, Aulnay de Saintonge. Un camp militaire augusto- tiberien en Aquitaine. Aquitania 1, 1983, 49 ff. and Aquitania 2, 1984, 105 ff. 170 Some considerations on the circulation of Augustan and Tiberian bronze coins in Gaul Walker 1988 Wells 1972 Wigg 1997 Wigg 1999 D.R. Walker, Roman coins from the sacred spring at Bath. In: B. Cunliffe (Ed.), The temple of Sulis Minerva at Bath, II: Finds from the sacred spring (Oxford 1988) 281 ff. CM. Wells, The German policy of Augustus. An examination of the archaeological evidence (Oxford 1972). D. Wigg, Coin supply and the Roman Army. In: W. Groenman-van Waateringe/et alii (Eds.), Roman Frontier Studies 1995. Proceedings of the XVIth International Congress of Roman Frontier Studies. Ox- bow Monograph 91 (Oxford 1997) 281 ff. D. Wigg, Die Rolle des Militars bei der Miinzversorgung und Miinzwirtschaft am Rhein in der friihen Kaiserzeit. In: W. Schliiter/R. Wiegels (Eds.), Rom, Germanien und die Ausgrabungen von Kalk- riese (Osnabruck 1999) 327 ff.