Providing Markets with Small Change in the Early Roman Empire: Italy and Gaul, in Revue belge de Numismatique, 155, 2009, p. 125-142.

REVUE BELGE DE NUMISMATIQUE ET DE SIGILLOGRAPHIE BELGISCH TIJDSCHRIFT VOOR NUMISMATIEK EN ZEGELKUNDE ´ publiee SOUS LE HAUT PATRONAGE DE S. M. LE ROI par la ´ ´ SOCIETE ROYALE DE NUMISMATIQUE DE BELGIQUE uitgegeven ONDER DE HOGE BESCHERMING VAN Z. M. DE KONING door het KONINKLIJK BELGISCH GENOOTSCHAP VOOR NUMISMATIEK Directeurs ¨ ¸ Francois de CALLATAY, Johan van HEESCH, Ghislaine MOUCHARTE, Jean-Luc DENGIS CLV - 2009 BRUXELLES BRUSSEL “FROM MINTS TO MARKETS: THE MECHANISMS OF COIN TRANSFORMATION IN ANCIENT TIMES” JOURNÉE D’ÉTUDE INTERNATIONALE ORGANISÉE LE 13 À LA BIBLIOTHÈQUE ROYALE DE BELGIQUE JUIN 2008 La première journée d’étude internationale, organisée par le Cabinet des Médailles de la Bibliothèque royale de Belgique le 17 mai 2005, eut pour thème « The transport of coins through the ages ». Les actes en ont été publiés un an plus tard dans la Revue belge de Numismatique (152, 2006, p. 1-94), ce dont nous sommes très redevables envers la Société royale de Numismatique de Belgique. Suivant le projet annoncé de tenir désormais une réunion internationale de ce type tous les trois ans, une seconde journée d’étude s’est tenue au Cabinet des Médailles le vendredi 13 juin 2008 sur le thème « From mints to markets : the mechanisms of coin transformation in ancient times ». Ce thème est une nouvelle illustration d’un parti pris auquel nous tenons : choisir des sujets proprement historiques et les traiter sur la longue durée. Les études numismatiques ont considérablement amélioré notre connaissance de la production monétaire dans l’antiquité gréco-romaine, prise ici de c. 600 av. J.-C. jusqu’à c. 400 ap. J.-C. Des centaines d’études de coins, des dizaines de monographies permettent de mieux répondre aux questions de la date et de l’importance des monnayages émis, avec une prédilection particulière pour les émissions d’or et d’argent. D’un autre côté, les fouilles archéologiques ont généré un riche matériel monétaire tout autour de la mer Méditerranée, dont une bonne partie a été publiée. C’est le bronze ici qui prévaut. Les monnaies d’or et d’argent y sont en moyenne représentées par moins d’un pourcent. Cette conférence s’est donnée pour but d’articuler ces deux types de données en se posant la question : qu’est-ce qu’un légionnaire romain ou un mercenaire hellénistique faisait de sa paie en monnaies d’or et d’argent ? Il est très probable qu’ils furent rapides à dépenser une bonne partie de leurs salaires. Mais par quels canaux la grosse monnaie d’or et d’argent, impropre à régler les transactions quotidiennes, s’est-elle transformée dans ces monnaies de bronzes que l’archéologie découvre aujourd’hui ? Neuf conférenciers furent invités à en débattre, dont on retrouvera ciaprès les textes pour huit d’entre eux. C’est un plaisir pour nous de remercier les institutions et les personnes qui ont facilité la tenue d’un tel événement scientifique : la Bibliothèque royale de Belgique qui a financé cette RBN, 155, 2009, p. 1-2 2 FRANÇOIS DE CALLATAŸ ET JOHAN VAN HEESCH journée en mettant, entre autres, la belle salle Lucien de Hirsch à notre disposition. Que son Directeur général, M. Patrick Lefèvre, qui a tenu à accueillir personnellement les participants, en soit vivement remercié. Notre gratitude va également à la Politique Scientifique Fédérale, à son Président, M. Philippe Mettens, et à M. Pierre Demoitié, qui ont accordé un généreux subside pour la tenue de cette manifestation. Enfin, nous sommes redevables à nouveau envers la Société royale de Numismatique de Belgique d’avoir bien voulu accueillir le fruit de ces réflexions au sein de la Revue belge de Numismatique. François DE CALLATAŸ* et Johan VAN HEESCH** * François DE CALLATAŸ, Cabinet des Médailles, Bibliothèque royale de Belgique, Boulevard de l’Empereur 4, B-1000 Bruxelles. E-mail : callatay@kbr.be. ** Johan VAN HEESCH, Penningkabinet van de Koninklijke Bibliotheek van België, Keizerslaan 4, B-1000 Brussel. E-mail: Johan.vanheesch@kbr.be. JOHAN VAN HEESCH* PROVIDING MARKETS WITH SMALL CHANGE IN THE EARLY ROMAN EMPIRE: ITALY AND GAUL Abstract: Coinage in the western parts of the Empire was organized in a completely different way to Roman Greece or the East. Small change (i.e. asses, and especially semisses and quadrantes) was not of prime importance to the Roman government. Asses were, however, indispensable for small-scale monetary transactions, as were the subdivisions of the as from Augustus to Nero in Gaul and for an even longer period in Italy. Semisses and quadrantes were still minted under Antoninus Pius, who died in 161. While asses and the higher denominations were almost always issued by mints controlled by the state; the smaller subdivisions were often left to the local authorities, be they cities, the military or perhaps even private persons or entrepreneurs. This was so in Gaul from Augustus to Claudius, and in Italy from the late-first century BC till the reign of Augustus. Afterwards (from Nero onwards in Gaul, and from Augustus in Italy) the central mints developed a policy of minimal production of small change and did actually send batches of it to some parts of the empire. The mechanisms by which these coins entered circulation are not known, but several hypotheses are formulated here. In this contribution I will discuss Roman monetary policy towards small change, before going on to look at the mechanisms by which it was put into circulation. I will examine two completely different regions, Gaul and Italy, and will limit myself mostly to the first century AD. 1. Small change Though it is not my intention to theorize about definitions, I think it wise to explain what I mean by small change. It is obvious that small change is a relative notion, but this paper is mainly concerned with bronze coinage, and especially the smallest range of bronzes, i.e. asses, semisses and quadrantes, denominations that are particularly useful in small-scale transactions at markets and other commercial activities.1 ∗ Johan VAN HEESCH, Coin Cabinet of the Royal Library of Belgium, Boulevard de l’Empereur 4, B-1000 Brussels. E-mail: johan.vanheesch@kbr.be. I am most grateful to Dr. David Wigg-Wolf for revising my English. 1 The denominational system in first century AD Rome was as follows: 1 gold aureus = 2 gold quinarii = 25 silver denarii = 50 silver quinarii = 100 brass sestertii = 200 brass dupondii = 400 copper asses = 800 brass semisses = 1,600 copper quadrantes. RBN, 155, 2009, p. 125-142 126 JOHAN VAN HEESCH We should not forget, however, that even quadrantes were more valuable than the small change in use today. In the first century AD the salary for one day’s work for an unskilled labourer lay between 5 to 16 asses, and common soldiers received 10 asses a day. Taking a working day of 10 hours, this would mean 1 as or 4 quadrantes per hour.2 There can be no question about the high degree of monetisation of the Roman economy.3 Proof comes from the coin finds themselves. From no other historical period are so many coin finds known as from the Roman Empire. Medieval coins are extremely rare till the introduction of billon mites in the 14th century and pure copper denominations in the 15th/16th century. This abundance of Roman coin, and the rarity of Medieval, not only has to do with the thickness and indestructibility of the Roman coins, but also with the fact that small change was almost nonexistant between about AD 500 and 1400. I do realize, of course, that there were enormous differences within the huge Roman Empire, differences over time and in space. But this is completely normal. The more sophisticated a society is, the more differentiated the economy. Barter, social transactions (determined by social obligations and tradition) and highly specialized monetary transactions do exist side by side, even in modern society.4 It is quite obvious that it is much more advantageous for the authorities to strike 1 aureus than 1,600 quadrantes, or even 400 asses. Striking quadrantes is very labour-intensive, expensive in the fabrication of dies as well as in transport costs! Moreover these coins could be taken to be of little political interest as is shown by the systematic omission of the imperial portrait! During the Late Middle Ages minting billon coins was also considered to be very dishonourable for the monarch and of little profit but, as is shown by their nickname “denier de l’aumosnerie” or “the beggar’s penny”, they were of undisputable use in some transactions.5 That the Roman government’s prime interest did not lie with bronze coinage can be shown by the fact that some emperors did not strike bronze For wages and prices see now: W. SZAIVERT and R. WOLTERS, Löhne, Preise, Werte. Quellen zur römischen Geldwirtschaft, Darmstadt, 2005. 3 John Chrysostom (d. AD 407), In Principium Actorum 4, 3 = P.G. 51, 99, 36-40. Quoted by J.-M. CARRIÉ, Aspects concrets de la vie monétaire en Province, in RN, 159, 2003, p. 188. 4 P. TEMIN, A Market Economy in the Early Roman Empire, in JRS, 91, 2001, p. 169-181; J. VAN HEESCH, On the Edge of the Market Economy: Coins Used in Social Transactions, as Ornaments and as Bullion in the Roman Empire, in A. BURSCHE et al., Roman Coins outside the Empire. Ways and Phases, Contexts and Functions (Collection Moneta, 82), Wetteren-Warsaw, 2008, p. 49-57. 5 P. SPUFFORD, Money and its Use in Medieval Europe, Cambridge, 1988, p. 330-331. See also “Moneta pauperum” in Scotland, 15th century: P. GRIERSON, The Monetary Pattern of Sixteenth-Century Coinage, in Transactions of the Royal Historical Society, 5th series, 21, 1971, p. 45-60 (esp. p. 50). 2 PROVIDING MARKETS WITH SMALL CHANGE 127 for very long periods. To name just a few: Augustus for 12 years between 2 BC and AD 10; Claudius and Nero for more than 20 years between AD 43 and AD 64.6 But that this is not the whole story is demonstrated by the simple fact that the Roman government did strike masses of bronze coinage in other periods, and that it paid great attention to the different type of reverses which are sometimes less monotonous than those on the gold and silver issues. To take just two examples: Nero’s market building, the harbour of Ostia, his coin distributions, his triumphal arch and Titus’ coin with the coliseum are types only to be found on bronze! So bronze coinage did interest Roman officials, but unfortunately we do not know, as we do for some 16th century monarchs, whether the emperor himself was interested in the actual representations.7 That pure cost-effectiveness was not always the ultimate goal can equally be easily demonstrated, as coins were shipped from Rome to the remotest corners of the empire. Several coin issues of silver and bronze, often with Greek inscriptions and meant to circulate in Cappadocia, Syria and Egypt, are known to have been struck in Rome, as they are not only identical in style to the coinage of the Roman mint, but were also made of exactly the same metal as the Italian coins.8 But these are only general remarks, and it is worthwhile looking at two distinct areas in more detail to put all this into perspective. 2. Northern Gaul and “private” initiatives in the early-1st century Let us start with Gaul, and more specifically Northern Gaul. When the emperor Augustus installed some 8 legions on the Northern edges of the empire, his soldiers received pay on a regular basis. That golden aurei and silver denarii were used as soldiers’ pay can be shown beyond any doubt. Suffice it to say that Julius Caesar as conqueror of Gaul coined only gold and silver, but no bronze at all, and that Flavius Josephus actually describes the paying out of gold and silver to thousands of Roman soldiers before the walls of Jerusalem in AD 70 – mentioning also that it took four days before the several legions had all received their pay.9 6 H. MATTINGLY et al., The Roman Imperial Coinage, 10 vols., London, 1923-2008, passim. 7 Benvenuto Cellini, My Life (Oxford Worlds Classics), Oxford, 2002, p. 120-121 (I, 71). 8 For example: W. METCALF, The Silver Coinage of Cappadocia, Vespasian-Commodus (ANSNNM, 166), New York, p. 83-90; I. CARRADICE and M. COWELL, The Minting of Roman Imperial Bronze Coin for Circulation in the East: Vespasian to Trajan, in NC, 147, 1987, p. 26-50; H.R. BALDUS, Mon(eta) Urb(is) - Antioxia: Rom und Antiochia als Prägestätten Syrischer Tetradrachmen des Philippus Arabs, Frankfurt/Main, 1969; A.M. BURNETT and P. CRADDOCK, Rome and Alexandria: the Minting of Egyptian Tetradrachms under Severus Alexander, in ANSMN, 28, 1983, p. 109-118. 9 Flavius Josephus, Jewish War, V, ix 1-2 [348-356]. 128 JOHAN VAN HEESCH The opening of a mint for gold and silver in Syria at precisely this time is further proof of this.10 We do not know for certain if bronze was used to pay soldiers in imperial times, but Mark Anthony’s bronze fleet coinage probably served this purpose, and thousands of bronze coins are also found in the Roman camps along the limes.11 Each camp and its immediate surroundings constituted in itself an artificial island where monetary transactions and market exchange were much more frequent than in the hinterland. The circulation of small change in these “monetary islands” is well documented, and it will not be necessary to go into any detail here.12 From the reign of Augustus to the beginning of the reign of Claudius (15 BC-AD 41), copper asses reached the camps of the soldiers in huge numbers. Though the money of account in that period was the sestertius, the as was the denomination that was struck in far larger quantity than any other bronze coin. Although small change, the symbolic value of the as should not be underestimated as the soldiers’ pay was still expressed in it at the time.13 Due to a lower level of prices in Northern Gaul, asses were apparently too large a denomination, and as the government did not react to the needs for smaller denominations, the matter was left to the local authorities. To obtain half asses or semisses the copper asses were regularly cut into halves (Fig. 1). To take just one example from the military camp in Neuss (Germany): out of 1,543 asses from Augustus, some 320 were cut 10 A. BURNETT, M. AMANDRY and I. CARRADICE, Roman Provincial Coinage, 2, ParisLondon, 1999, p. 269 (contribution of W. Metcalf); I. CARRADICE and T. BUTTREY, RIC II, part 1, London, 2007, 174-178. 11 Mark Antony’s “fleet coinage”: see A. BURNETT, M. AMANDRY and P.P. RIPOLLES, Roman Provincial Coinage, I, London-Paris, 1998, p. 284-286. Finds in Roman camps e.g. Neuss and Kalkriese (battlefield): H. CHANTRAINE, Novaesium VIII. Die Antiken Fundmünzen von Neuss, Gesamtkatalog der Ausgrabungen 1955-1978 (Limesforschungen, 20), Berlin, 1982 and F. BERGER, Kalkriese 1. Die römischen Fundmünzen (Römisch-Germanische Forschungen, 55), Mainz am Rhein, 1996. On this question see R. WOLTERS, Bronze, Silver or Gold? Coin Finds and the Pay of the Roman Army, in Zephyrus, 53-54, 2000-2001, p. 579-588. 12 See e.g. J.-M. DOYEN, Économie, monnaie et société à Reims sous l’Empire romain. Recherches sur la circulation monétaire en Gaule septentrionale intérieure (Bulletin de la Société archéologique champenoise, 100, n° 2 et 4 and Collection archéologie urbaine à Reims, 7), Reims, 2007 (2008) ; F. KEMMERS, Coins for a Legion. An Analysis of the Coin Finds of the Augustan Legionary Fortress and Flavian canabae legionis at Nijmegen, PhD thesis, Nijmegen, 2005 (also published in the series “Studien zu Fundmünzen der Antike”); D.G. WIGG, Die Rolle des Militärs bei der Münzversorgung und Münzwirtschaft am Rhein in der frühen Kaiserzeit, in W. SCHLÜTER, R. WIEGELS, (eds.), Rom, Germanien und die Ausgrabungen von Kalkriese (Osnabrücker Forschungen zu Altertum und Antike-Rezeption, 1), Osnabrück, 1999, p. 327-343. 13 Tacitus, Annals, XVII, 6-8 and XXXI, 5. PROVIDING MARKETS WITH SMALL CHANGE 129 into halves, and 3 even into quarters to obtain quadrantes.14 That even this didn’t respond to the needs of the market is shown by the numerous finds of very small copper coins of the same weight as Roman quadrantes and struck by local authorities. The most prolific of these series is the well known Avaucia type, but numerous other examples exist in “civilian” Gaul (Fig. 2). These local copper coins (Avaucia bronzes) disappear from circulation gradually, but were very probably replaced from the reign of Tiberius onwards (?) by other locally struck series of small change that copied Roman prototypes more closely, i.e. Roman asses of Augustus, Tiberius and Claudius (Fig. 3).15 These were probably struck in great numbers in or near Roman military sites, but possibly also in civilian contexts. So here again we do not see any active intervention on the part of the Roman mint responsible for the main coin issues. A lot of these local imitations have one thing in common in the North of Gaul and that is that, although they systematically imitate Roman asses, they are much lighter, indeed half their weight! This makes it very probable that they were used as half asses or semisses!16 The fact that “Celtic style” small bronzes are apparently replaced by half-weight copies of Roman asses is probably not only due to the gradual Romanization of the local elites, but possibly also reflects a rise in the general price level. How else can one explain this shift from quadrantes (if that was the accepted value of the “Celtic” small bronzes of course…) to half asses in the finds? This situation continued until the middle of the reign of Nero. In AD 64 the mint of Lyons resumed minting bronze coins again, and sestertii, dupondii, asses and even semisses circulated in fair numbers. Small change of local fabric and made in local workshops gradually disappeared from circulation and official issues dominated the coin circulation pool.17 This was the result of a deliberate monetary policy by the Roman government. A few years later, during the reign of Domitian, the North was quite unexpectedly provisioned with large quantities of copper quadrantes (Fig. 4), as is shown by the spectacular finds from Nijmegen pub- 14 H. CHANTRAINE, Novaesium VIII. Die Antiken Fundmünzen von Neuss, Gesamtkatalog der Ausgrabungen 1955-1978 (Limesforschungen, 20), Berlin, 1982. 15 D. NASH, Plus ça change...: Currency in Central Gaul from Julius Caesar to Nero, in R. CARSON and C. KRAAY (eds.), Scripta nummaria Romana. Essays presented to Humphrey Sutherland, London, 1978, p. 12-31; D.G. WIGG, The Function of the Last Celtic Coinages in Northern Gaul, in C. KING and D.G. WIGG (eds.), Coin Finds and Coin Use in the Roman World (Studien zu Fundmünzen der Antike, 10), Berlin 1996, p. 415-436. 16 The fact that the emperor’s head never appears on quadrantes but is often present on semisses (Lyons, Nero) is perhaps of importance here. See D.G. WIGG, op. cit. [n. 15]. 17 Abundant literature: see e.g. all volumes of FMRD and the data in J.-M. DOYEN, op. cit. [n. 12]. 130 JOHAN VAN HEESCH lished by Fleur Kemmers.18 In the military camp of Nijmegen some 303 specimens were found. Compared to the total of 6 small bronzes from the previous reigns of Nero, Vespasian and Titus, this number is quite remarkable. Though not a general pattern in Gaul, more small bronzes of Domitian are found throughout the region, and I do not see any reason why this shouldn’t be explained as a deliberate policy to provide markets with small change. Domitian is known to have been particularly interested in monetary matters. Pliny cites Domitian Caesar as one of his sources for his chapter on metals and coinage, and Domitian undertook several important reforms to improve his coinage and the organisation of minting.19 This growing concern towards bronze coinage from the second half of the first century AD is also reflected in the concentration of minting in Rome from the Flavians onwards. The interest of the official minting authorities in providing the population with small change and the awareness of local needs is well illustrated by some other second-century phenomena of which I will mention only one. From the late-first century AD semisses and quadrantes were not a real need in Gaul anymore, as can be shown by the ever growing presence of larger denominations such as dupondii and sestertii in the finds, without any doubt a reflection of a slowly but steady rise in price levels.20 In other areas of the empire, however, which were probably not as rich as Gaul, smaller coins were still regarded as useful, as is shown by the continued minting of the so-called “mine coins” from Trajan to Antoninus Pius (Fig. 5). These semisses and quadrantes, most of them without the letters S(enatus) C(onsulto) on the reverse, were struck in Rome with particularly interesting reverses mentioning the Roman mines in the Danube and Balkan area. That these responded to local needs is proved by the existence of imitations of quadrantes from these parts of the empire, and by the fact that batches of these mine coins were actually send to the Balkan and Danube provinces, as is shown by the finds.21 That does not mean that they were meant F. KEMMERS, Quadrantes from Nijmegen: Small Change in a Frontier Province, in SNR, 82, 2003, p. 17-35. 19 Pliny, N.H., I, 33 and R. DUNCAN-JONES, Money and Government in the Roman Empire, Cambridge, 1994, p. 100 n. 24. Also T.V. BUTTREY and I. CARRADICE, RIC II, 2nd ed., London, 2007. 20 A general phenomenon see e.g. J. VAN HEESCH, De muntcirculatie tijdens de Romeinse tijd in het noordwesten van Gallia Belgica, Brussel, 1998, p. 77, 118-120 and J.-M. DOYEN, op. cit. [n. 12], e.g. p. 187-190. 21 B. WOYTEK, Die Metalla-Prägungen des Kaisers Traian und seiner Nachfolger, in NZ, 111-112, 2004, p. 35-68 (esp. 52-53 for the finds); for the imitations: e.g. S. DUŠANIĆ, Aspects of Roman Mining in Noricum, Pannonia, Dalmatia and Moesia Superior, in H. TEMPORINI (ed.), ANRW, II, 6, Berlin, 1977, col. 60 n. 29 and Notes on the Anonymous Quadrantes (in Serbian), in The Numismatist [Numismatičar], I, 1978, p. 27 n. 22. For other copies see a quadrans with Diana/Stag (Woytek type XIII and 18 PROVIDING MARKETS WITH SMALL CHANGE 131 exclusively for circulation in these areas, but find evidence, though rare, seems to indicate that they are fairly frequent in central Europe.22 1. Augustus (27 BC-AD 14), halved as struck in Nîmes 2. “Avaucia” bronze, reign of Augustus, Northern Gaul 3. Imitation of a Claudian as, Northern Gaul, 1st. cent. AD 4. Domitian (AD 81-96), quadrans, Rome 5. Trajan (AD 98-117), quadrans of the mines with Metalli vlpiani on the reverse (Coin cabinet of the Royal Library of Belgium, Brussels) RIC II, p. 474 no. 1015) in Vienna (Inventory 11184 and illustrated in J. VAN HEESCH, Studie over de semis en de quadrans van Domitianus tot en met Antoninus Pius (Masters thesis), Gent, 1979, illus. 366) and Woytek, type IX, p. 68 illus. H and p. 50 n. 88. 22 Based on the finds listed by B. WOYTEK, op. cit. [n. 21], p. 52-53 and also Fatmir PEJA, Contribution à l’étude des monnaies de “metalli Dardanici”, in B. KLUGE and B. WEISER (eds.), XII. Internationaler Numismatischer Kongress Berlin 1997. Akten, I, Berlin, 2000, p. 607-609 (4 coins with Dardanici on the reverse, one of Trajan, the other with Roma on the obverse, and all found in Sočanica, near Kosovska Mitrovica [Kosovo]). Though Woytek does not seem to believe that they were actually sent to that part of the Empire, the find evidence, however, is clear: only two circulation areas are well documented, i.e. central-Europe (provinces near the Danube) and Italy. If we exclude coins without a secure provenance (i.e. those of which only the location of the collection is know) most finds come from central Europe (at least 29 against 3 for Italy). Note also that only two mine coins were found in Germany (Augsburg and Regensburg), although these finds are published in detail in the FMRD-series. 132 JOHAN VAN HEESCH 3. Italy and the centralizing role of the mint in Rome Let us now go briefly to the finds in Italy. The situation is comparable to that in Gaul for in Italy also the smallest bronze denominations were produced by local authorities or private persons. There is, however, a major difference to Gaul, and that is that local coinages disappear half a century earlier, as almost none are known after Augustus (the Tiberian issues of Paestum excepted).23 During most of the first century BC bronze coinage was not a priority at all for the Roman state. This is well illustrated by the fact that from c. 80 BC almost no bronze was minted in Rome till the monetary reform of Augustus in 19BC, a reform that introduced a completely new set of brass and copper coins mentioning the Roman senate and the tresviri monetales (Fig. 6).24 To satisfy the need for small change old Republican bronzes continued to circulate together with imported coins from Ibiza and Marseille. These issues were also imitated on a large scale in Italy.25 The excavations by the Anglo-American Project in Pompeii in Insula VI produced 997 individual coins that vividly illustrate how we should imagine the circulation of small change in the area, indeed probably in all of “central Italy”. Of these 997 coins (all lost before AD 79), 387 are Ebusus (Fig. 8), pseudo-Ebusus, Massalia and pseudo-Massalia coins (Table 1)!26 Imitations of Augustan coins, however, are rare and the last town to strike local small change was the city of Paestum (Fig. 7) under Tiberius (AD 14-37).27 Local imitative coinages appear to be replaced from the A. BURNETT, M. AMANDRY and P.P. RIPOLLES, op. cit. [n. 11], p. 158-160; K. RUT(ed.), Historia numorum. Italy, I, London, 2001, p. 112-117. 24 The cessation of bronze coinage: see M. CRAWFORD, Roman Republican Coinage, Cambridge, 1974. Italian coinages at the end of the Republic: A. BURNETT, M. AMANDRY and P.P. RIPOLLES, op. cit. [n. 11]; K. RUTTER, op. cit. [n. 23]. 25 See now in great detail: C. STANNARD and S. FREY-KUPPER, “Pseudomints” and Small Change in Italy and Sicily in the Late Republic, in AJN, s. 2, 20, 2008, p. 405458. Also: C. STANNARD, The Monetary Stock at Pompeii at the Turn of the Second and First Centuries BC: Pseudo-Ebusus and Pseudo-Massalia, in P.G. GUZZO and M.P. GUIDOBALDI (eds.), Nuove ricerche archeologiche a Pompei ed Ercolano. Atti del convegno internazionale, Roma, 28-30 novembre 2002 (Studi della soprintendenza archeologica di Pompei, 10), Napoli, 2005, p. 120-143 and C. STANNARD, Numismatic Evidence for Relations between Spain and Central Italy at the Turn of the Second and First Centuries BC, in SNR, 84, 2005, p. 47-80. 26 R. HOBBS, Coins from the AAPP Excavations, Pompeii: Update 1, in NC, 165, 2005, p. 377-381 (esp. 378). 27 See [n. 23]. On imitations of Republican bronzes, see M. CRAWFORD, Unofficial Imitations and Small Change under the Roman Republic, in AIIN, 29, 1982, p. 139-164. On the (rare) imitations of Augustus’ Italian coinage: R. MARTINI, Caesar Avgvstvs. Collezione Veronelli di monete di bronzo: catalogo critico (Glaux. Serie Speciale, 2), Milano, 2001, p. 390-408 and plate L nos 918-921. TER 23 PROVIDING MARKETS WITH SMALL CHANGE 133 Table 1: The AAPP excavations in Pompeii. Coin finds from Insula VI, I, 1995-2004 (data: R. Hobbs, NC, 165, 2005, p. 378) ___________________________________________________________________ Silver denarii & quinarii (mostly plated) 27 coins Janus asses (uncut) 73 Janus asses (fragmented) 74 Other republican aes 8 Ebusus & pseudo-Ebusus 229 (= 32.8%) Massalia & pseudo-Massalia 155 (= 22.2%) Pseudo-Panormus 3 “Greek” 17 “Celtic” 1 Neapolis bronze 2 Early imperial bronze (pre-AD 79) 72 Early imperial quadrantes 36 Unidentified bronze 300 Total 997 ___________________________________________________________________ reign of Augustus onwards by official state products minted in Rome. In Italy, just as in Gaul, the Roman copper as is the most popular bronze denomination everywhere, but smaller coins such as quadrantes and semisses are not uncommon either. It is difficult to determine the exact proportion of these denominations in relation to each other. Data from well documented archaeological contexts show a lot of variation even in Pompeii. It is perhaps useful to illustrate this with three different finds from Pompeii. These show how dangerous it can be to generalize from a single complex, and also how careful we should be when interpreting the accumulated data from finds without studying the actual archaeological context. Early Imperial “aes” found in the Anglo-American excavations produced 72 early imperial bronze coins (Augustus-Vespasian) and 36 quadrantes from the same period. Thus 33% were quadrantes (see Table 1). Near the entrance of another building in Pompeii excavated in 1822, a box was found with 35 denarii, 354 sestertii, 188 asses or dupondii and 586 small bronzes or quadrantes28 At another site, a tavern, a pot contained 374 asses or dupondii and 1237 quadrantes29, but in other recently published find of bronze coins from the Termopolio there were only 3 quadrantes in a lot of 1,361 imperial bronze coins!30 Another accumulation of finds on corpses from Oplontis near Pompeii, also deposited durSee previous note. See previous note. 30 D. CASTIELLO and S. OLIVIERO, Il ripostiglio del termopolio I, 8, 8 di Pompei, in AIIN, 44, 1997, p. 93-205. 29 28 134 JOHAN VAN HEESCH ing the eruption of Vesusius in AD 79, confirms this picture: 2 aurei, 189 denarii, 59 sesterii, 39 dupondii, 117 asses and no quadrantes at all.31 The same differences and extremes can be illustrated by other examples, and a general comparison by Satriano (Table 2) showed that the percentage of quadrantes ranged between 1.5 and 30%:32 Table 2: quadrantes in Italian finds (data: Annali. Instituto Italiano di Numismatica, 50, 2003, p. 151-152). <10% 3 finds / excavations 10-20% 3 finds / excavations 21-30% 5 finds / excavations The further from Rome, the sole mint in the 1st and 2nd cent., the rarer quadrantes appear to be, though the actual figures differ from reign to reign.33 Though it is clear from finds from all over Italy that asses were the most common bronze coin during the Early Empire, at least till the reign of Hadrian34, and that semisses and quadrantes played a minor role in circulation, it remains a fact that other impressive concentrations of these small coins are known from Italy, e.g. the famous lot of some 1,200 quadrantes and semisses found during dredging operations in the river Tiber in the early 1970s.35 It should be emphasized also that although quadrantes and semisses played a minor role in circulation, those responsible at the Roman mint apparently attached great importance to the messages and the types of even these very small coins. If, for example, the traditional interpretation of the quadrantes of Caligula with pileus on the obverse and Remissa ducentesima (R.CC) is correct (Fig. 9), then this is the only numismatic document on the remission of this half percent tax.36 31 V. CASTIGLIONE MORELLI (appendix by E. DE CAROLIS and M. TORINO), Un gruzzolo dalla stanza degli “Ori di Oplontis”, in Rivista di Studi Pompeiani, 11, 2000, p. 187234. 32 R. CANTILENA, S. PANTULIANO, M. PELLEGRINO and M. SATRIANO, Monete da Paestum (I-IV secolo D.C.), in AIIN, 50, 2003, p. 25-156 (esp. 151-152). 33 Augustan quadrantes appear to be rare in Northern Italy, but that is not so for those of Caligula and Claudius. See the well documented study of J.-M. DOYEN, op. cit. [n. 12], tables p. 110, 128, 132 and 140. 34 Data : see the tables in J.-M. DOYEN, op. cit. [n. 12], tables p. 110, 128, 132 and 140; for Ordona (Apulia): S. SCHEERS, La circulation monétaire à Ordona, in Ordona IX (Études de Philologie, d’Archéologie et d’Histoire ancienne publiées par l’Institut historique belge de Rome, 34), Bruxelles-Rome, 1997, p. 302. 35 C.E. KING, Quadrantes from the River Tiber, in NC, s. 7, 15, 1975, p. 56-90. 36 Tacitus, Annals, II, 42; Dio, LVIII, 16, 2 and Dio, LIX, 9, 6; Suetonius, Caligula, 16. On these and other quadrantes of Claudius also with singular messages (e.g. PNR) see C. KING, op. cit. [n. 35] and H. MATTINGLY, Coins of the Roman Empire in the British Museum, I, London, 1923, p. CXLVII and CLVIII. Also A.U. STYLOW, Die Quadranten des PROVIDING MARKETS WITH SMALL CHANGE 135 In short: during the early Empire local issues of small denominations and local solutions to satisfy the need for small change were allowed in Gaul to be provided by “private” initiative until at least the reign of Nero (AD 54-68). In Italy, on the contrary, the monetary reform of Augustus gradually eliminated local initiatives of this kind. It is worthwhile noticing here that in Spain too local bronze coinages disappeared gradually from Claudius and Nero onwards. 6. Augustus, Rome, quadrans 7. Tiberius (AD 14-37), small bronze, Paestum; 8. Ebusus (Baleares), bronze, 3rd. cent. BC 9. Caligula (AD 37-41), Rome, quadrans; 10. Nero (AD 54-68), Rome, semis (Coin cabinet of the Royal Library of Belgium, Brussels). 4. How was small change put into circulation, and was there any policy to provide markets with small change? There can be no doubt at all that in markets in the Western Roman Empire, be it in the North near the Roman camps on the limes, in towns or smaller settlements in inner Gaul, or be it in Italy, gold, silver and bronze coinages circulated side by side.37 An intriguing fresco from the Caligula als Propagandamünzen, in Chiron, 1, 1971, p. 285-290; D.W. MACDOWALL, The PNR Type of Claudius, in SM, 18, 1968, p. 80-86; W. ECK, Die staatliche Organisation Italiens in der hohen Kaiserzeit (Vestigia, 28), München, 1979, p. 113. 37 Two excellent examples: Pompeii, see R. DUNCAN-JONES, Roman Coin Circulation and the Cities of Vesuvius, in E. LO CASCIO (ed.), Credito e moneta nel mondo Romano, 136 JOHAN VAN HEESCH villa of Julia Felix in Pompeii illustrates coins in all metals in amazing detail. To the left of a purse is a pile of silver and gold coins (certainly denarii and aurei), to the right bronze coins that are visibly larger. The painter even distinguished between copper asses (reddish) and brass dupondii and sestertii which look like gold.38 Several questions about small change need to be answered. How was it put into circulation, and was there any policy to provide markets with small change? Were even these smaller denominations used by the Roman government to pay its expenses, or should we also imagine another mechanism by which these coins were put into circulation, and was this done to facilitate market transactions? This last aspect is the easiest to answer as it is obvious that these coins were only issued to facilitate small payments, be it to buy small quantities, as beggar’s pennies or just to respond to the need to prevent merchants round up their prices. In Italy and Spain the quadrans and the semis were the usual entrance fee for public baths39. It is generally believed (and it makes sense) that most of the coins issued by the Roman mint were put into circulation through public spending. The more important posts of the Roman budget were: army pay, wages for civilian employees, the costs of the emperor’s household, handouts to civilians and soldiers (congiaria, sportulae, epula, donativa), building activities, gifts and public spending, subsidies and gifts to allies, etc.40 There is no doubt that most of it was paid in coins made of precious metal.41 But huge quantities of bronze coins have also been found in Roman camps all over the empire and it is reasonable to suppose that occasionally bronze coins also entered circulation through the wages of soldiers and civilians. The use of bronze had several advantages. First of all that of providing communities with small change for use in markets. Another motive for using bronze could have been the scarcity of precious metals.42 The fact that bronze coins from the reign of Augustus Bari, 2003, p. 161-180 and Kalkriese (Germany), see F. BERGER, Kalkriese 1. Die römischen Fundmünzen (Römisch-Germanische Forschungen, 55), Mainz am Rhein, 1996. 38 Museo archeologico nazionale, inv. 8598; illustration: http://pompeya.desdeinter. net/GiuliaFelice011.htm; J.-M. CROISILLE, Les natures mortes campaniennes (Collection Latomus, 76), Bruxelles-Berchem, 1965, p. 28-29 no. 8D and plate CX no. 208. 39 Prices: W. SZAIVERT and R. WOLTERS, op. cit. [n. 2], p. 282 (with references to Horatius & Martialis); Spain: Lex metalli Vipascensis, 1, 23 (CIL II, 5181): a semis for men and an as for women. 40 R. DUNCAN-JONES, Money and Government in the Roman Empire, Cambridge, 1994, p. 33-46. 41 R. WOLTERS, Bronze, Silver or Gold? Coin Finds and the Pay of the Roman Army, in Zephyrus, 53-54, 2000-2001, p. 579-588. 42 J. VAN HEESCH, Some Considerations on the Circulation of Augustan and Tiberian Bronze Coins in Gaul, dans R. WIEGELS (ed.), Die Fundmünzen von Kalkriese und die frühkaiserzeitliche Münzprägung (Osnabrücker Forschungen zu Altertum und AntikeRezeption, 3), Möhnesee, 2000, p. 165. PROVIDING MARKETS WITH SMALL CHANGE 137 have countermarks mentioning the names of generals, e.g. Varus, might also suggest that these high officers had something to do with paying out bronze coins to their men.43 Almost exclusively asses are involved, and it is hard to believe that smaller denominations were used to pay soldiers’ “wages”. Semisses and quadrantes were probably used on “a lower level”. This brings us to another important question. Is it possible small change entered circulation by other mechanisms; or in other words is it possible that private persons, middleman or local authorities played a role in all this? If we regard the Celtic-style bronze coins, such as the Nervian bronzes (and many other “Celtic-styled” coinages), and the imitative issues as parallel to the civic coinages we know from the Greek cities (or even regional/provincial coinages), these coins might have been put into circulation by local/regional authorities. Bronze coins could have been useful in all kinds of transactions.44 Though no contemporary sources inform us about this, it is possible that mechanisms existed to attract gold and silver to the treasury of the civitates as they had to pay previously determined amounts of taxes to the provincial authorities.45 Soldiers probably needed to exchange some of their gold and silver with local bankers or money changers. Civitates might have received gold or silver for the supply of food to the army and paid their local farmers only in bronze.46 Local labourers could also have received coppers as wages from civilian authorities responsible for building projects and the construction and maintenance of roads, etc. 43 M.A. SPEIDEL and H.W. DOPPLER, Kaiser, Kommandeure und Kleingeld, Vier neue Gegenstempel aus Zurzach und Baden und ihr Beitrag zur Geschichte, in Jahresbericht Gesellschaft Pro Vindonissa, 1992 [1993], p. 5-16; R. WOLTERS, C. Numonius Vala und Drusus. Zur Auflösung zweier Kontermarken augusteischer Zeit, in Germania, 73-1, 1995, p. 145-150. 44 On Avaucia and other local “Celtic” coinages and imitations see e.g.: J. VAN HEESCH, Coin Circulation and Coin Use in Northern Gaul between the Mosel and the North Sea from Caesar to Augustus, in M.P. GARCÍA-BELLIDO, A. MOSTALAC and A. JIMÉNEZ (eds.), Del imperivm de Pompeyo a la avctoritas de Augusto. Homenaje a Michael Grant (Anejos de AEspA, 47), Madrid, 2008, p. 29-39; J. VAN HEESCH, Les Romains et la monnaie gauloise: laisser-faire, laisser-aller ? in J. METZLER and D. WIGGWOLF, Die Kelten und Rom: Neue numismatische Forschungen (Studien zu Fundmünzen der Antike, 19), Mainz am Rhein, 2005 [2006], p. 229-245; D. NASH, op. cit. [n. 15], p. 12-31; D. WIGG, op. cit. [n. 15], p. 415-436. 45 On taxes: L. NEESEN, Untersuchungen zu den direkten Staatsabgaben der römischen Kaiserzeit 27 v. Chr. bis 284 n. Chr. (Antiquitas, 1, 32), Bonn, 1980 and W. SZAIVERT and R. WOLTERS, Löhne, Preise, Werte, op. cit. [n. 2], p. 131-166; M. MILLETT, The Romanization of Britain, Cambridge, 1990, p. 66-67. 46 Military purchasing food supplies on the “free” market : P. ERDKAMP, The Corn Supply of the Roman Armies during the Principate (27 BC-235 AD), in P. ERDKAMP (ed.), The Roman Army and the Economy, Amsterdam, 2002, p. 47-69 (esp. p. 65, 67, 74-75; with e.g. references to Pliny, Pan., 29.5 and FIRA 3.137). 138 JOHAN VAN HEESCH But production of small change could also have taken place in military camps. That might explain why a coinage such as the Avaucia series or the light weight imitation asses of Claudius are found all over the military zone along the River Rhine.47 Soldiers received their wages only three times a year and most of it, between 50 to 80%, remained “at the standards”, that is to say in the savings bank of the camp.48 Huge amounts of money were gathered here, and these sums were a real gold mine for usurpers like Saturninus, who plundered the camp savings in AD 89. Even if not much is known about these military deposits (some papyri and a passage of Vegetius excepted49), and the sum deposited in the camp was limited by Domitian to 1,000 nummi per soldier (10 gold pieces), it is quite possible that these “banks or treasuries” played a major role in the injection of smaller denominations into circulation.50 They are also likely candidates for the coining of Celtic styled coin issues (such as Avaucia bronzes and imitations), and they could very well be the lieu par excellence where official small change (semisses and quadrantes) was handled. Thanks to the Vindolanda tablets and other recently published documents we realize how sophisticated the Roman administration was, and how incredibly accurately finances were managed.51 Even the exceptional quantities of quadrantes of Domitian in Nijmegen could be explained by the activity of the military bankers, possibly to be identified with the signiferi or standard bearers.52 So much for Gaul, but how did this work in Italy where the military presence is supposed to have been much more limited than in the provinces. Did all coinage spread from its single mint in Rome outwards to the rest of the peninsula? The injection of purely imitative coin issues (e.g. pseudo-Ebusus) could be explained, as in Gaul, by commercial activities and expenditure at the 47 See J. VAN HEESCH, op. cit. [n. 44]; D.G. WIGG, op. cit. [n. 44] and J.-M. DOYEN, op. cit. [n. 12], p. 120-126. 48 J. VAN HEESCH, Some Aspects of Wage Payments and Coinage in Ancient Rome, First to Third Centuries CE, in J. LUCASSEN (ed.), Wages and Currency. Global Comparisons from Antiquity to the Twentieth Century, Berlin, 2007, p. 77-96; G.R. WATSON, The Roman Soldier, London, 1969, p. 104-107. 49 R.O. FINK, Roman Military Records on Papyrus, Princeton, 1971, p. 266-267 document 71; PGen Lat.1 recto; Vegetius, Epitome of Military Science. Translated by N.P. MILNER, Liverpool, 1996, p. 54. 50 On the limitation to 1,000 ‘sestertii’ see Suetonius, Domitianus, 7 (after the revolt of Saturninus). There could be a link with the countermarking of bronze coinage. 51 A.K. BOWMAN, Life and Letters on the Roman Frontier. Vindolanda and its People, London, 1994 for a general introduction and G. MINAUD, La comptabilité à Rome. Essai d’histoire économique sur la pensée comptable commerciale et privée dans le monde antique romain, Lausanne, 2005. 52 Suetonius, Domitianus, 7; Vegetius, Epitome of Military Science. Translated by N.P. MILNER, Liverpool, 1996, p. 54; G.R. WATSON, op. cit. [n. 48], p. 104-107. Quadrantes in Nijmegen: see F. KEMMERS, op. cit. [n. 18]. PROVIDING MARKETS WITH SMALL CHANGE 139 level of cities or regional authorities. For those of us who still remember the shortage of small change in Italy in the mid 1970s, it is easy to understand how easily people accept substitutes for small change. In those days everyone accepted telephone tokens, sweets and finally paper money issued by local banks. The spread of “official” (= minted in Rome) bronze coin however requires another explanation. Though it is certain that Rome’s army was mainly situated outside Italy, we should not forget that there was some military presence spread throughout the peninsular. In Rome there were the praetorian cohorts and the urban cohorts, as well as the vigiles. Outside Rome we know of the fleets stationed in Ravenna and Misenum, military cohorts in Ostia and Puteoli, although other “garrisons” must have existed. Though knowledge of the military in Italy is very rare, and information on police forces are virtually absent, the information transmitted by Suetonius is informative.53 Writing about banditry he clearly says that Tiberius posted many more stationes militum all over Italy than before.54 Though the number of military in Italy cannot be compared to that on the limes, the existence of army posts could explain how freshly minted coins in all metals were injected into the coin circulation pool. On the other hand, we should not forget also that in the early empire the number of administrative posts grew continuously. At least 70 administrative departments are known for Italy from the Flavians onwards, and most must have worked with a staff which we can assume received regular pay, as was the case for the apparitores or public servants that assisted senators and magistrates.55 They are likely candidates to have been paid in small denominations. It is equally unimaginable that the procurators responsible for collecting the 5% tax on inheritances, for example, or that of 5% on the emancipation of slaves and the 4% tax on the sale of slaves, did not work with an army of functionaries that received pay. Perhaps not always directly from the government but by the publicani as intermediaries, who might have played a role in the injection of new and small coin into circulation.56 Other ways new coinage could be injected are those cited above: public works, government purchases of food, coin distributions to the people 53 Suetonius, Tiberius, 37: “Stationes militum per Italiam solito frequentiores disposuit.” 54 See also: F. MILLAR, Italy and the Roman Empire: Augustus to Constantine, in Phoenix, 40, no. 3, 1986, p. 295-318 (esp. 308, 313); W. NIPPEL, Policing Rome, in JRS, 74, 1984, p. 20-29. 55 W. ECK, The Growth of Administrative Posts, in A. BOWMAN et al. (eds.), The Cambridge Ancient History, second edition, The High Empire, A.D. 70-192, Cambridge, 2000, p. 244-246; F. MILLAR, op. cit. [n. 54], p. 299; Ch. DAREMBERG and D. SAGLIO, Dictionnaire des Antiquités grecques et romaines, I, Paris, s.d., p. 327-328 (apparitores merces). 56 F. MILLAR, op. cit. [n. 54], p. 295-318 (esp. 299-305). 140 JOHAN VAN HEESCH of Rome, etc. But can we imagine small denominations being used for this? After all it is hardly imaginable that large food supplies were actually paid for in bronze coins, and certainly not in asses, semisses or quadrantes, even when these coins are particularly numerous in central Italy. Roman emperors who distributed congiaria to the Roman people at regular intervals probably did not use bronzes for this purpose as the amounts are too high, ranging from 60 to 150 denarii per person. We know from inscriptions, however, that smaller amounts were distributed in towns outside Rome. Mrozek lists gifts (or sportulae in Latin) of amounts as little as 1, 2 and 3 sestertii. These sums are too small to be paid in regular silver coins.57 It is always possible that smaller towns bought copper coins from the central government to have small change, but there is no evidence to confirm this. On the other hand it is very possible that the Roman government itself used these very small denominations in distributions of lesser importance in the way they also used tickets that could be exchanged for other goods. Suetonius tells us how during the games he instituted Nero threw gifts (called missilia) into the public.58 These festivities are even commemorated on his brass semisses (Fig. 10), and only on these! The same reverse type, referring to games and prizes, can also be found on semisses of Trajan, Hadrian and Antoninus Pius.59 It is tempting then to associate these issues with distributions in coin on the occasion of these festivities. Another way freshly minted small change might have been brought into circulation is by bankers or money changers. The further from Rome, the less profitable this would be as transport costs were high. But rational economic behaviour is not always a characteristic of the ancient economy, and batches of small change could be transported with other merchandise to reduce the expenses.60 That there was no steady supply of new coins to local communities outside Rome can be shown from the finds from the towns and settlements that perished in the eruption of Vesuvius in AD 79. Although destroyed on August 24th, during the reign of the emperor Titus, and situated only some 200 km south of Rome, no gold, silver or bronze coin of this ruler 57 S. MROZEK, Les distributions d’argent et de nourriture dans les villes italiennes du Haut-Empire romain (Collection Latomus, 198), Bruxelles, 1987, p. 35, 96, 281. 58 Suetonius, Nero, 11; Ch. DAREMBERG and D. SAGLIO, Dictionnaire des Antiquités grecques et romaines, I, Paris, 1900, p. 1934-1938 (missilia). 59 Games of Nero: Suetonius, Nero, 12.3. Semisses with references to games (table of the games): RIC I, p. 164-165 nos. 228-248 (Nero), RIC II, p. 292-293 nos. 685-688 (Trajan), p. 421 no. 621 (Hadrian), RIC III, p. 104 no. 572 (Antoninus Pius). 60 On bankers in general: J. ANDREAU, La vie financière dans le monde romain. Les métiers de manieurs d’argent (IVe siècle av. J.-C.-IIIe siècle ap. J.-C.) (Bibliothèque des écoles françaises d’Athènes et de Rome, 265), Rome, 1987 and ID., Banking and Business in the Roman World, Cambridge, 1999. PROVIDING MARKETS WITH SMALL CHANGE 141 was ever found at any of these sites.61 Titus became emperor on June 24th and it is unthinkable that he did not start coining in his own name right away. But during the first two months of his sole reign, not a single coin arrived in these Campanian towns. In short: although not much is known with certainty about the function of the smallest denominations (semisses and quadrantes), and we do not really know how they were put into circulation, we propose several possibilities. There can be no doubt that small change was indispensable in an advanced monetary economy. Local coinages, possibly minted not so much by individuals as by local authorities or the military (in Gaul), were used to cover small expenses such as gifts, wages for labour and in exchange transactions (silver for bronze). The phenomenon of locally minted coinages disappeared earlier in Italy than in Gaul, and by the reigns of Augustus and Tiberius only the state coinage was minted. This entered circulation through the same mechanisms as before: the exchange of larger denominations for bronze, wage payments and gifts. Roman semisses and quadrantes were used in Italy, and sometimes even sent to Gaul and the Danubian provinces to respond to local needs. Although economically speaking these measures were of little importance, they show that from time to time even the central government had a monetary policy towards small change. 61 R. DUNCAN-JONES, op. cit. [n. 37], p. 161-180 and L. BREGLIA, op. cit. [n. 28], p. 41-59.