The last coin of Taras?: A study of a late Tarentine coin in the collections of the Uppsala University Coin Cabinet.

(1)

Department of Archaeology and Ancient History

The last coin of Taras?

A study of a late Tarentine coin in the collections of the Uppsala University Coin Cabinet

Karl Appelgren

BA thesis 15 credits in Classical archaeology and ancient history Spring term 2021 Supervisors: Filmo Verhagen & Ragnar Hedlund

(2)

Abstract

Appelgren K. 2021. The last coin of Taras? A study of a late Tarentine coin in the collections of the Uppsala University Coin Cabinet.

Appelgren K. 2021. Taras’ sista mynt? En undersökning av ett sent tarentinskt mynt i Uppsala universitets myntkabinetts samlingar.

In this thesis, a coin from the Hannibalic occupation of Taras is analysed and discussed. The method applied in the analysis is Panofsky’s iconological method, and the theoretical framework has been derived from the research questions themselves in dialogue with modern numismatic research. The focus of the discussion is on the relationship between the coin and its historical context. In the thesis, it is argued that the coin is a didrachm with heavily reduced weight, and that the weight reduction is a result of the financial difficulties caused by the Second Punic War.

Denna uppsats är en analys av ett mynt from Hannibals ockupation av Taras. Den metod som tillämpas i analysdelen är Panofskys ikonologiska metod. Det teoretiska ramverket har sin utgångspunkt i uppsatsens frågeställning, och har utarbetats i dialog med modern numismatisk forskning. Diskussionsdelen fokuserar på förhållandet mellan myntet och dess historiska kontext. I uppsatsen framförs argument för att myntet är en didrachm med kraftigt reducerad vikt, och att viktreduktionen är en följd av de finansiella svårigheter som orsakades av Andra puniska kriget.

Keywords: Tarentine coinage, Second Punic War, Hannibalic occupation of Taras, Carthaginian influence, Roman conquest of Italy.

BA thesis in Classical archaeology and ancient history 15 hp. Supervisors: Filmo Verhagen &

Ragnar Hedlund. Defended and passed 2021-06-15.

Department of Archaeology and Ancient History, Uppsala University, Box 626, 75126 Uppsala, Sweden

Front cover. Silver coin from the Hannibalic occupation of Taras, ca. 212–209 BC (the coin discussed in this thesis). Uppsala University Coin Cabinet. Inventory number: 400871. https://www.alvin- portal.org/alvin/attachment/download/alvin-record:190910/ATTACHMENT-0001.tiff

(3)

1. Introduction

Between 218 and 201 BC, the Second Punic War raged in the Mediterranean. In the opening year of the war, Carthaginian general Hannibal Barca took the Romans by surprise by crossing the Alps into Italy. After the crushing defeat of the Romans at Cannae in 216 BC, Hannibal’s campaign settled in the southern part of the Italian peninsula, where it was to remain until 203 BC, when Hannibal was recalled to Africa by the Carthaginian senate.¹

It has been argued that Hannibal invaded Italy with the hope of detaching Rome from its subordinate Italian allies.² This strategy was only partly successful; whereas some Roman allies rebelled, others remained loyal to Rome. One city that did revolt was Taras, known to the Romans as Tarentum, the old Spartan colony that had formerly been one of the most influential powers in Italy. Interestingly, the revolt did not take place immediately after Cannae, but three and a half years later. In the winter of 213/12 BC, the city defected from Rome and joined forces with Hannibal. The Carthaginian takeover of the city was only partial, however, since the citadel remained in Roman hands throughout the war.³ Under these peculiar circumstances, coins were struck in the Tarentine mint.

1.1. Aim, goal, and research questions

Even though the scholarship on the coinage of the Hannibalic occupation of Taras can be traced as far as to the latter half of the 19^th century, the discussion has been limited to numismatic details. No study has yet given due notice to the relationship between the coinage and its historical context, which is in this case unusually complex. In this Greek city, coins were struck while the city was being occupied by Punic forces, although the Roman enemy was still in control of the citadel. Given these circumstances, it is astounding that the only comprehensive account of the coinage was given by Sir Arthur Evans in 1889.⁴ Later publications rely on him, often without adding much of their own. Whatever the merits of Evans’ work, it has a certain tinge of connoisseurship and is mainly focused on chronology and categorisation.

The aim of this thesis is to contribute to the understanding of how ancient coinage and economy were affected by crises. The goal of the thesis is to fill the gap outlined above and add numismatic evidence to the knowledge of the Hannibalic occupation of Taras. In the thesis, this is achieved through a qualitative analysis of a single Tarentine coin from the Uppsala University Coin Cabinet. To facilitate the fulfilment of the aim and the goal, the following research questions have been devised:

• Why was the coin produced?

• What impact did the Second Punic War and the Hannibalic occupation have on the shaping and production of the coin?

There are, of course, no simple answers to such questions, which by nature call for discussion rather than a straightforward reply. For this reason, the following subset of questions have been

1 Livy 30.19.12.

2 Fronda 2010, 37.

3 Lomas 1993, 59–76.

4 Evans 1889.

(6)

added, whose specific character will facilitate the achievement of a tangible result and guide the analysis:

• What is the coin’s denomination?

• What weight standard underlies it?

• Who was responsible for its production?

• Who was the issuing authority?

• Is there a Carthaginian influence on the coin?

• Does the coin signal financial desperation?

• Is it reasonable to interpret the coin as a sign of political independence?

By this arrangement, the subordinate questions will function as a foundation for the discussion of the two main research questions.

1.2. Material

The primary material of the thesis is restricted to one coin only. This coin belongs to the tenth and last period of Tarentine coinage as outlined by Evans, a period which coincided with the Hannibalic occupation of the city (212–209 BC).⁵ It is difficult to estimate the total quantity of this coinage, since its constituents are shattered throughout the world and no up-to-date publication exists. It is important to stress that neither the coin nor its coin type are tantamount to the coinage that was minted during the Hannibalic occupation of Taras, and that this thesis does not purport to give a comprehensive account of all Tarentine coinage from the period.

Rather, it is a detailed qualitative study of a single specimen. On the other hand, the thesis assumes that this specimen is representative for its coin type and, to a very large degree, for all Tarentine coinage from the period. The reasons for this assumption are given below.

The coin is an example of the first of the five coin types that were distinguished by Evans in his account of the tenth period of Tarentine coinage.⁶ To this day, there have been no additions to these types.⁷ The five types are very similar to each other, and share the same image types: a horseman on the obverse and a dolphin rider on the reverse.⁸ The main criteria for their division into five coin types are the names and monograms inscribed on them and the minor differences in the depiction of the image types. In all essentials, therefore, the coin can be regarded as representative, not only for its coin type, but for all Tarentine coinage dated to the Hannibalic occupation. Where its coin type diverges from the other four coin types of the period, this will be noted.

The coin is kept in the collections of the Uppsala University Coin Cabinet.⁹ Its acquisition can be dated to 1955, when Arne Furumark, professor in Classical history and archaeology at Uppsala University, bought it from Mr Robert E. Hecht, an American antiquities art dealer. The price of the coin was 12 dollars.¹⁰ Unfortunately, it has not been possible to trace the whereabouts of the specimen any further.

5 Evans 1889, 196–211.

6 Evans 1889, 210–211.

7 Rutter 2001, 106.

8 Evans 1889, 210–211.

9 Inventory number: 400874.

10 This information has been derived from the archives of the Uppsala University Coin Cabinet.

(7)

1.3. Structure

The thesis is structured as an abductive argument. By abduction, I mean a form of logical inference where the best explanation for a certain phenomenon serves as its conclusion, with the phenomenon as the premise.¹¹ In practice, the results of the iconographical analysis will function as the premise, whereas the iconological interpretation and the following discussion provide the conclusion. There, the results of the analysis are furnished with the best explanation for their occurrence by means of contextualisation. Contextualisation is not an end, but a means of understanding the coin on a deeper level than a mere analysis would allow. In its turn, the contextualised coin will be conducive to the thesis’ aim: to contribute to the understanding of how ancient coinage and economy were affected by crises.

1.4. Historical background

Before the Pyrrhic war (280–275 BC), Taras was the leading power of southern Italy and the leader of the Italiote league. The war had begun as a conflict between Taras and Rome, and the Tarentine declaration of war can be understood as a reaction to the increasingly marked Roman presence in the region. When Taras called on the Epirote king Pyrrhus for help, this was in line with a previous policy of dealing with external threats.¹² No less than three times in the previous century, the Tarentines had employed foreign generals as mercenaries in their dealings with the indigenous peoples of southern Italy.¹³

When Pyrrhus left Italy after his defeat at Maleventum in 275 BC, the Tarentines were able to hold their city for another three years. In 272 BC, it was finally taken by the Romans.¹⁴ The peace terms seem to have been relatively favourable for Taras. For instance, it retained the right to strike its own coins.¹⁵ On the other hand, Taras had to accept a new status as an independent Roman ally, and a Roman garrison was installed in the citadel. Kathryn Lomas suggests that the political consequences may have involved the exile of anti-Roman politicians and the installation of pro-Roman governments.¹⁶ It is however important to stress that the city’s new status as a Roman ally did not imply that it had become Roman, nor that it had lost its independency. The key to the success of the Roman conquest of Italy was complex, and is best understood as a combination of many factors, including both the founding of coloniae and the creation of independent allies. In the words of Andrew Wallace-Hadrill:

To put it provocatively (and with conscious exaggeration): the founding of colonies is part of the hellenisation of Italy (Roman power expressed in a hellenic idiom), the creation of independent allies is the true romanisation (Roman power expressed in a Roman way). Too paradoxical, of course: but it is necessary, to counter the sheer deadweight of historical inevitability that makes it appear to us that Rome always wanted to make Italy fully Roman, that we should underline the consistency with which for at least three centuries, the Romans ignored, and eventually actively resisted, the idea of making the allied states Roman.¹⁷

According to Lomas, the impact of Rome on the Greek South before 200 BC was relatively slight.¹⁸ Rome’s control of Italy was primarily a military alliance, and the main duty of the allies was to provide troops to give military support.¹⁹ In the case of Taras, however, this obligation may have been demanding enough to cause deep dissatisfaction. More importantly, perhaps,

11 Bowell, Cowan & Kemp 2020, 176‒179.

12 Lomas 2017, 254–256.

13 Brauer 1986, 61–86.

14 Lomas 2017, 259.

15 Brauer 1986, 169.

16 Lomas 2017, 259.

17 Wallace-Hadrill 2008, 80–81.

18 Lomas 1993, 76.

19 Lomas 2017, 272.

(8)

the Roman alliance had greatly diminished Taras’ influence in the region. As suggested by Lomas, Taras might have been entering a more expansionist phase at the outbreak of the Second Punic War, which could explain the impetus of their agreement with Hannibal.²⁰ It is noteworthy that Taras was the only city to make an entirely voluntary approach to Hannibal.²¹

The Tarentine revolt and the ensuing Carthaginian conquest of Taras are well attested in Livy. From him, we learn that the city was conquered with help from the inside.²² The conquest was only partial, however, since the Tarentine citadel remained in Roman hands throughout the war. With respect to the sympathies towards the belligerents in the Second Punic War, the city seems to have been divided in a pro-Roman and an anti-Roman faction, to which the conspirators belonged.²³ According to Livy, parts of the pro-Roman population managed to escape to the citadel when the city was conquered by Hannibal.²⁴

Whether or not there had been a break in Tarentine coinage before the Second Punic War, as many scholars think, it is interesting that coins were struck under these unusual circumstances – in this divided city.²⁵ Although the issuing authority was obviously in control of the mint, it was not in control of the whole city, which of course included the citadel as well.

In this thesis, the relationship between the coinage and its highly complex historical context will be examined.

1.5. Theory

Since the aim of the thesis is to contribute to the understanding of how ancient coinage and economy were affected by crises, a general theory of how numismatic evidence can be used as a historical source is necessary. But rather than presenting ready-made theory with loose or no connections to the research problem at hand, this theory section will proceed from the research questions themselves. This is done by reformulating the questions as more general questions, upon which general answers are sought. The purpose of these answers is to provide the thesis with a general theory, against which the (particular) results of the analysis can be compared.

The first research question, “Why was the coin produced?”, can easily be translated to the following general question: “Why were (ancient) coins produced?” In his book Ancient history from coins, Christopher Howgego addresses this question within the chapter ‘Minting.’ There, he challenges the traditional view that coins were struck for no other purpose than to allow states to make payments, and sheds light on the considerable amount of ancient evidence for alternative motives. According to Howgego, there was a connection between coinage and political autonomy, and the pride involved in the assertion of civic identity should not be underestimated. There are strong indications that the production of coinage was a common reaction to restored freedom throughout antiquity.²⁶

It should be emphasised that pride and assertion of political autonomy do not replace state payments as motives for coin production in this model. Rather, they are part of a complex picture of coinage, which involves many coexisting components that affect each other. In some cases, it is certainly possible to imagine that state payments were the actual motive for a city’s coin production, whereas the assertion of political autonomy was a lucky consequence, of which the citizens were proud. In other cases, the assertion itself might have been the motive, with the economic benefit as the side effect. It is, of course, often hard to decide which of these scenarios was the case. The point is that simplistic models should be avoided.

We now turn to the second research question, from which we get: “What was the impact of war (and other crises) on ancient coinage?” Just like today, ancient times of trouble caused

20 Lomas 1993, 71.

21 Lomas 1993, 75.

22 Livy 25.8.1–6.

23 Brauer 1986, 183–197.

24 Livy 25.10.

25 Rutter 2012, 138. Burnett 2012, 308.

26 Howgego 1995, 33–41.

(9)

pressure on the economy. In such conditions, financial solutions were called for. A well- documented example of such solutions can be found on the other side of the Second Punic War, in Rome, whose financial expedients included a loan from Hiero of Syracuse, a doubling of tribute, and borrowing from her own citizens. Monetary manipulations, such as debasement and lowering of weight standards, were additional solutions.²⁷

For the sake of clarity, these theoretical considerations can be labelled “Howgego’s theory of the connection between coinage and crises.” It is however important to emphasise that such a theory is nothing but a synthesis of my own readings of Howgego’s numismatic works; it is not presented as a comprehensive theory by Howgego. What is lost in consistency is gained in aptness, however, since the theory has been derived from the research questions themselves.

Arguably, it is the most appropriate theory for the research problem at hand.

1.6. Method

The method applied in the analysis is Erwin Panofsky’s iconological method as outlined in Meaning in the visual arts.²⁸ Although this method, strictly speaking, belongs to art history, it can easily be adjusted to include special aspects of numismatics, such as weight standard, denomination and inscriptions. In contrasting iconography with iconology, the analysis follows a distinction made by Panofsky himself, who in his later works describes iconology as “an iconography turned interpretive.”²⁹

The method can be analysed as a tripartite process, which consists of pre-iconographical description, iconographical analysis, and iconological interpretation. In the words of Panofsky, these are not unrelated operations, but intertwined methods of approach that “merge with each other into one organic and indivisible process.”³⁰ A keyword for Panofsky is ‘stratum,’ and the three methods of approach correspond to the three strata of subject matter or meaning in a work of art. For our purposes, it will suffice to think of them as stages, since the structure of the analysis benefits from such an arrangement. It is however important to remember that this is a matter of presentation rather than strict methodology. In actuality, the three stages merge and presuppose one another.

One of the greatest benefits of Panofsky’s method is that it enables a natural transition from the empiric material to its context. Moreover, its tripartite division of the pre-iconographical, iconographical and iconological is surprisingly applicable to different kinds of materials, including coins.

1.7. Source criticism

Numismatic sources have their problems. The lack of archaeological context is not unique for the coin discussed in this thesis, but a common feature of much numismatic evidence.

Moreover, the coinage portrays a limited part of the society, and is intrinsically elitist. Most often, the issuing authority is found in the upper strata of society and can barely be considered as representative for the whole community. This picture is even more complicated in the case of occupied Taras, where the remaining elements of the pro-Roman faction in the citadel likely were completely excluded from the minting process.

Although material evidence is at the centre of the analysis, a word about the usage of literary sources is needed as well, since the iconological method involves the study of texts in its second and third stages. Moreover, the historical nature of the research questions also presupposes an acquaintance with the textual remains. The main figure in this picture is Livy (64/59 BC – AD 12/17), whose third decade of books Ab urbe condita is the main source for the historical

27 Howgego 1995, 111–112.

28 Panofsky 1955.

29 Panofsky 1955, 32.

30 Panofsky 1955, 39.

(10)

knowledge of the Second Punic War. To name only a few of the problems associated with using Livy as a source, he lived at a much later time and had a Roman bias towards Roman history – a bias that was probably shared with most of his sources. For many different reasons, his historicity cannot be taken for granted.³¹

Even though numismatic evidence cannot compare with literary sources in revealing the complexities of political life, they have advantages for the historian.³² In contrast to Livy, the coin discussed in this thesis is contemporaneous to the Second Punic War, and it obviously lacks a Roman bias. If anything, it represents the line of the issuing authority, and can be interpreted as an expression of its will – whatever this authority might have been. To summarise, the problems of numismatic sources are of a very different nature than the problems of its literary counterparts.

31 Fronda 2010, 5–13.

32 Howgego 1995, 62.

(11)

2. Previous scholarship

A beginning for the scholarship on the coinage of the Hannibalic occupation of Taras can be placed in 1887, when Barclay V. Head, the British numismatist and keeper of the Department of Coins and Medals at the British Museum, successfully dated them to this period in his Historia Numorum.³³ Prior to Head, the coinage had been dated to much earlier periods. In 1863, for instance, L. Sambon had dated them to the period preceding the Pyrrhic wars.³⁴ This confusion ended with Head, who firmly attested that “the short period when Tarentum shook off the Roman yoke during the Second Punic War is the only time to which they can be attributed.” His argument for this is that none of the names on the coins appear on the didrachms and drachms of previous periods. Interestingly, Head claims that the coinage has a high weight in comparison with earlier coins, while the opposite is the actual case.³⁵ Head writes that the weight of the didrachms from this period was 120 grains, when the actual weight varies between approximately 50 and 60 grains (3.2–3.9 grams). Arthur Evans suggests that Head was misled by an error in an earlier Italian publication.³⁶ Be that as it may; as a result of this mistake, Head was excluded from the interesting discussion of the coin’s denomination. On the other hand, given the nature and scope of Historia numorum, which is presented as “a manual of Greek numismatics,” Head’s account would probably not have been a suitable place for such detailed discussion anyway.³⁷

Two years later, Arthur J. Evans, who would later strike fame as the excavator of the Palace of Minos at Knossos, included the coinage from the Hannibalic occupation of Taras as the tenth and last period in his 220-page article ‘The “horsemen” of Tarentum.’³⁸ The purpose of Evans’

article was “to distribute the long array of Tarentine horsemen for the first time into successive chronological divisions.”³⁹ Given the difficulty of this pioneering undertaking, it is astonishing how accurate his predictions turned out to be.

Evans corroborates Head’s dating of the coinage to the Hannibalic period. In contrast to Head, he is fully aware of the low weight of the coins, which he explains by arguing that a massive weight reduction took place during the war. By acknowledging the bond between the Tarentine didrachms and their iconography, he manages to resist the temptation of adopting a different denomination for them. He argues that the stress of the Hannibalic war had revolutionised the conditions of the Italian money-market, and in this he clearly foreshadows modern views of the impact of the Second Punic War on the coinage of Italy.⁴⁰

As stated before, Evans distinguishes five different coin types from the Hannibalic period (see section 1.2.). In addition to these, he also mentions a sixth type, but since this belongs to a lesser denomination, it will not be discussed any further in this thesis.⁴¹ The five types share their traditional Tarentine silver didrachm imagery. On the obverse, there is a horseman, and on the reverse – a dolphin rider. Together with the Greek inscriptions that figure on the coins, the variations in the imagery constitute the criteria for Evans’ division. In general, these

33 Head 1887.

34 Sambon 1863.

35 Head 1887, 54.

36 Evans 1889, 197.

37 Evans 1889, 5–6.

38 Evans 1889, 191–211.

39 Evans 1889, 8.

40 Evans 1889, 191–211.

41 Evans 1889, 211.

(12)

variations relate to the poses, attributes, and symbols of the horseman and the dolphin rider, but in the case of the horseman, there are in-depth variations as well: the first three types depict naked boys, whereas the fourth and fifth types depict adult men.⁴²

In 1908, Evans’ classification of the tenth period of Tarentine coinage was confirmed by the Taranto 1908 hoard. According to Michael P. Vlasto, who relates the testimony of unnamed sources, the 114 coins of the hoard were found in a fractured vase at a depth of three yards (ca.

2.7 m). Of these coins, 25 are of the exact same type as the coin studied in this thesis, that is to say, the first of the five types distinguished by Evans.⁴³

For the eighty years or so that follow Vlasto, scholarship seems to be astoundingly quiet about Tarentine coinage from the tenth period. Mentions of it seem to be restricted to museum catalogues, such as the 1942 Sylloge nummorum Graecorum of the National Museum in Copenhagen, Denmark, where the name on the first type is erroneously rendered as ΣΥΡΑΜΒΟΣ. This situation is somewhat altered by the appearance of Brauer’s monograph on Tarentine coinage and history in 1986, where two pages are devoted to the coinage in question.

In all essentials, Brauer follows the account of Evans, without adding much of his own, except the negative remark that the coinage is “degenerate” and that it is “sad” that it bears names that are “not even Greek.”⁴⁴

The next appearance of the coinage is in 2001, when N.K. Rutter devotes a short section to it in his Historia numorum. Italy.⁴⁵ Interestingly, the whole section is labelled “Punic standard,”

based on the weight standard that Rutter assumes for the coinage. Rutter expands this notion to the actual denominations as well, calling the coins “Punic half-shekels.” Rutter also mentions Tarentine coinage from the Second Punic War in his chapter ‘The coinage of Italy’ in The Oxford handbook of Greek and Roman coinage, where he repeats the claim that the coinage reflects a Punic standard.⁴⁶

To conclude, the literature on Tarentine coinage from the Hannibalic occupation is meagre, with the sole exception of Evans. Other scholars tend to reproduce his scholarship without adding much of their own. In adopting a Punic standard and describing the coins as Punic half- shekels, Rutter deviates from previous scholarship, according to which the coins are Tarentine didrachms with a heavily reduced weight standard. Although Rutter seem to give no argument for this deviation, it is in line with a widespread tendency in modern numismatics to give priority to weight and weight standards. By doing so, it often disregards other aspects of coinage, such as imagery and inscriptions.

The title of this chapter, “Previous scholarship,” is significant for the character of previous dealings with Tarentine coinage from the tenth period. As stated before, Evans’ work has a certain tinge of connoisseurship, and one could add that it is concerned with the historical context insofar as it affects the arrangement of the horsemen into successive chronological divisions. It is the goal of this thesis to fill this gap by means of a detailed qualitative study of a single coin, which expands into a discussion of the historical context.

42 Evans 1889, 210–211.

43 Vlasto 1909, 253–254.

44 Brauer 1986, 195–197.

45 Rutter 2001, 106.

46 Rutter 2012, 138.

(13)

3. Analysis

The analysis is divided into three sections in accordance with the three strata of Panofsky’s iconological method. In the first section, a pre-iconographical description of the coin is given.

This is followed by an iconographical analysis, where the cultural conventions that made the coin meaningful are outlined and employed to deepen the understanding. The third and final section amounts to an iconological interpretation. This also marks the beginning of the historical contextualisation that continues into the following discussion.

3.1. Pre-iconographical description

The diameter of the coin is 18.96 millimetres. Its weight is 3.2 grams. It is made of silver, but the fineness is unknown. The axis clock is 5.

3.1.1. The obverse

Figure 1. The obverse. Naked boy crowning horse.

The motif of the obverse is a horseman (Fig. 1). The corporal language of the horse suggests that it is standing rather than moving. This can be explained by the observation that it is crowned by its rider. The body shape and relatively small size of the rider indicate that it is a boy. Since there is no sign of clothes, one might infer that he is naked. The head is heavily leaned backwards, on the boundaries of the anatomically possible. Below the head to the left, the Greek inscription ΚΛΗ (klē) is visible. And between the legs of the horse, the inscription ΣΗΡΑΜ (sēram) is discernible, which continues with ΒΟΣ (bos) to the right of the front legs.

(14)

3.1.2. The reverse

Figure 2. The reverse. Dolphin rider.

The motif of the reverse is a dolphin rider (Fig. 2). He is naked but seems to wear some kind of headgear. In contrast to the horseman, the rider and the dolphin seem to be moving. They move in direction to the left – the opposite direction of the horseman on the obverse. The nose of the dolphin is missing. From the physical appearance only, it is difficult to tell whether the dolphin rider is a man or a woman, but since the iconographical analysis will make clear that we are dealing with a man, I will consider him a man already at this stage. In his left hand, the dolphin rider holds a trident. In his right hand, he extends an acrostolium, the ornamental part of the prow of an Antique warship.⁴⁷ Between the feet of the dolphin rider, an Α is discernible, which continues with ΡΑ (ra) on the right side of the trident. In addition to this, there is a monogram to the right of the trident.

3.2. Iconographical analysis

The pre-iconographical description above raises several questions. What is the significance of the horseman and the dolphin rider? What do the Greek inscriptions add to the picture? It is the task of iconographical analysis to answer questions such as these.

3.2.1. The obverse

In the preceding section, we concluded that the obverse depicts a naked horseman that is crowning his horse, which is standing. The horseman is probably a boy. As already mentioned, the horseman is a traditional image type of Tarentine didrachms. To understand the horseman as an image, we will have to investigate the cultural conventions that underlies this tradition.

Horsemen started appearing on Tarentine silver didrachms by the middle of the fifth century BC.⁴⁸ The plural is important, since it is from early on possible to distinguish different kinds of horsemen. In addition to the naked boy, there are adult horsemen as well (Fig. 4). The naked boy seems to be associated with the act of crowning the horse (Fig. 3), whereas the adult horseman often (but not always) are depicted in a much more militaristic manner. An example of this can be seen on a coin type dated to the Pyrrhic period, where the horseman charges right and lances downwards (see Fig. 4). Variations are found in the depiction of the horses as well, which stand, prance and gallop on different occasions (compare Fig. 3 and Fig. 4). These are only a few examples of the vast iconographic repertoire offered by the horsemen of Taras. It is however beyond the scope of this thesis to give a comprehensive account of it here.

47 A dictionary of architecture and landscape architecture, s.v. Acrostolium.

48 Brauer 1986, 36–37.

(15)

Figure 3. Tarentine silver didrachm, ca. 302–281 BC. Naked boy crowning horse.

Figure 4. Tarentine silver didrachm, ca. 281–272 BC. Horseman charging right, lancing downwards.

Initially, the horsemen were concentrated to the reverse, but soon they were transferred to the obverse.⁴⁹ The horsemen quickly became a symbol for Taras, much like the owls were symbols for Athens.⁵⁰ It is symptomatic that Evans named his monograph on Tarentine coinage after them.

To make sense of the horseman, some prima facie contextualisation is needed. In the Hellenistic period, Taras was famous for its cavalry. Tarentine horsemen where probably employed as mercenaries, and “Tarentine” even became synonymous with a certain type of light cavalry. In the case of the naked boy crowning his horse, however, it is of course other connotations that come to mind. Rather than of soldiers, one thinks of the competitors in equestrian religious games. Unfortunately, we lack evidence of any such Tarentine practices.⁵¹ On the other hand, it is debatable if additional evidence for athletic and agonistic practice is really required. It could be argued that the crowning of the horse and the boy’s nudity speak for themselves.

In addition to the horseman, the obverse includes two inscriptions written in the Greek alphabet. The meaning of ΚΛΗ is yet to be deciphered, but earlier traditions in Magna Graecia seem to suggest that it might be the name of the engraver.⁵² When it comes to ΣΗΡΑΜΒΟΣ (Sērambos), more is to be said. As suggested by Evans, this is probably the name of the magistrate who was responsible for the production of the coin.⁵³ However, the name diverges from the names of previous Tarentine magistrates in its non-Greek flavour. As noted by Evans, the name has a certain likeness with the name of the vintner Sarambos mentioned in Plato’s Gorgias.⁵⁴

The other four coin types from the Hannibalic period contain different names on their obverse. On type C and type D, we find the Greek names Kritos and Philiarchos, respectively.

In contrast, the inscriptions on type B and E read ΣΩΓΕΝΗΣ (Sōgenēs) and ΣΩΚΑΝΝΑΣ (Sōkannas) respectively.⁵⁵ The latter share their non-Hellenic character with Serambos, and as proposed by Evans, it is possible that they are the names of Hannibal’s officers.⁵⁶ On the other hand, it should be emphasised that the names are written in the Greek alphabet and that they do

49 Rutter 2012, 133.

50 Brauer 1986, 37.

51 Brauer 1986, 175–176.

52 Jenkins 1972, 185–208.

53 Evans 1889, 205.

54 Evans 1889, 203; Pl. Grg. 518b.

55 Evans 1889, 210–211.

56 Evans 1889, 204.

(16)

have the Greek noun endings -os, -ēs, and (perhaps Doric) -ās, all of which lack parallel in Punic and other (Northwest) Semitic languages, such as Hebrew.⁵⁷ Despite the uncertainties of their origin, the names clearly show an impact of Hellenisation. It is however difficult to decide whether this can be regarded as an in-depth cultural influence; it might just reflect the standard procedure of Greek renderings of non-Greek names.

The fact that none of these names were present on previous Tarentine issues, whose inscriptions were of an unmistakably Greek character, led Evans to the assumption that there had been a break in the activities of the Tarentine mint.⁵⁸ This assumption was strengthened by a discontinuity in the style of the coinage, whose execution differs from previous Tarentine issues.⁵⁹ Another divergence, the difference in weight, will be discussed in detail below.

That there had been a break in Tarentine coinage is confirmed by modern numismatic research.⁶⁰ Evans had originally placed the break in 228 BC, but Rutter and Burnett go further, and date the break to ca. 240 and ca. 250 BC respectively. Whenever and if such a break took place, the coinage from Evans’ tenth period clearly marks a new beginning with respect to the names, style and weight standard.

3.2.2. The reverse

Like the horseman, the dolphin rider is a traditional type of Tarentine silver coinage. It appears already on the incuse coinage of the sixth century BC and is present on Tarentine issues ever after. In opposition to the horseman, it seems to have made the opposite journey, that is to say, from the obverse to the reverse. Together, they remain the two core images on Tarentine coins until its cessation after the Roman reconquest in 209 BC.⁶¹

There are at least three possible candidates for the identity of the dolphin rider. First, there is the legendary poet Arion, who was saved by a dolphin when the crew on his ship plotted against him and threw him overboard. According to Herodotus, Arion was active at the court of Periander at Corinth, but voyaged to Sicily and Italy, where he made a fortune by singing.

He has also a connection to Taras, since he hired and boarded his ship there.⁶²

Figure 5. Dolphin rider. Reverse of Fig. 3.

Figure 6. Dolphin rider. Reverse of Fig. 4.

57 Hackett 2008, 90.

58 Evans 1889, 191.

59 Brauer 1986, 196.

60 Rutter 2012, 138; Burnett 2012, 308.

61 Brauer 1986, 20.

62 Hdt. 1.24.

(17)

Second, there is Phalanthus, the oecist (founder) of Taras. To keep a long story short, Phalanthus was the leader of the Partheniae, or “children of unmarried mothers,” who left Sparta in the aftermath of the First Messenian War.⁶³ Legend has it that before reaching Italy, Phalanthus suffered shipwreck in the Crisaean sea and was brought ashore by a dolphin.⁶⁴ After reaching Italy, Phalanthus and the Partheniae conquered Taras from the indigenous Iapygians and refounded it as a Spartan colony.⁶⁵

Third, there is Taras himself, the eponymous hero of the city, who also shares his name with a neighbouring river. According to legend, Taras was the son of Poseidon and the local nymph Satyra.⁶⁶ Being the son of the sea god Poseidon, Taras obviously had a good reason for riding a dolphin. On the other hand, it seems difficult to tie him to any particular dolphin legend, such as Arion’s or Phalanthus’.

Arguably, the Greek letters on the reverse speak for the latter, since they clearly form part of the word (Τ)ΑΡΑ(Σ) (Taras). That the ΑΡΑ should be supplied with a Τ and a Σ can be made clear by comparison with other specimens of Tarentine coinage from the tenth period (Fig. 6).

Moreover, this ethnic inscription had been coupled with the dolphin rider since the fifth century BC.⁶⁷ However, one should think twice before equating the ethnic with the name of the dolphin rider. If anything, the copresence of the ethnic and the dolphin rider could have contributed to the eventual identification of the dolphin rider with the eponymous hero of the city.

Scholarship seems to prefer Phalanthus to Taras for the identity of the dolphin rider.⁶⁸ But rather than settling for one of these figures and identifying the image with him, I would like to keep the complex picture. There was clearly a conflation between Phalanthus and Taras, oecist and eponymous hero, in Tarentine culture, especially in later times.⁶⁹ A similar situation can be seen in the Tarentine mixture of Apollo and Hyacinthus, which resulted in the syncretic Apollo Hyacinthus. ⁷⁰ Where the evidence speaks for such a conflation, syncretism must be acknowledged as an explanation. Personally, I find it plausible that the dolphin rider symbolises both Phalanthus and Taras, and maybe aspects of Arion as well.

In comparison with the horsemen, the depictions of the dolphin rider are more consistent.

The variations on the five Hannibalic coin types of Evans are restricted to the attributes held by the dolphin rider. The acrostolium in the right hand of the dolphin rider occurs only on type A, whereas the trident in his left hand is present on all five types except type B, where the dolphin rider holds cornucopiae instead (see Fig. 6 for an earlier, Pyrrhic example of this).⁷¹

3.2.3. Weight standard

We now turn to the more specifically numismatic aspects of the coin, which, albeit technical, will turn out to be compatible with the iconological method. Although the coin type represents traditional Tarentine silver didrachm iconography, the weight of the coin (3.2 g) is much too low for a traditional Tarentine silver didrachm (see Table 1). Arguably, this anomaly could be explained by a drastic weight reduction. Such a reduction would have a predecessor in the Pyrrhic period, when a weight reduction by 1.3 g had taken place. This post-Pyrrhic weight would become a new Tarentine standard that persisted throughout the period of Roman domination preceding the Second Punic War.⁷²

63 Strabo 6.3.2–3.

64 Paus. 10.13.10.

65 Brauer 1986, 4–5.

66 Paus. 10.10.8.

67 Brauer 1986, 26.

68 Brauer 1986, 33.

69 Brauer 1986, 32–35.

70 Brauer 1986, 88.

71 Evans 1889, 210–211.

72 Rutter 2012, 138.

(18)

Table 1. Some relevant weight standards.

Attic stater (didrachm)

Punic half-shekel (“drachm”)

Traditional Tarentine didrachm

Post-Pyrrhic Tarentine didrachm

Hannibalic

“standard”

8.6 g 3.8 g 7.9 g 6.6 g 3.2–3.9 g

The lowering of the weight standard was much more drastic this time, however. From the post- Pyrrhic weight of 6.6 g, the coins from the Hannibalic occupation of Taras dropped to between 3.2 and 3.9 g, or approximately half their former weight. The extremity of this weight loss has led scholars such as Rutter to assume that the Tarentine weight standard was abandoned completely, and replaced by a new, “Punic” standard.⁷³ Such a standard would be compatible with the heaviest specimens of Hannibalic coinage, but not with the lighter coins, to which the coin in Uppsala (3.2 g) belongs. At closer scrutiny, however, the Uppsala specimen seems to have been struck on a small flan, since the nose of the dolphin and the Τ and the Σ of the ethnic ΤΑΡΑΣ are missing on the reverse. Arguably, the existence of lighter specimens such as this one could be explained by lacking craftmanship in the production. To put it bluntly, the craftsman might have taken less metal than he was supposed to.

It is, however, possible to reject Rutter’s view and argue for a massive lowering of the weight standard. The conditions of Taras during the Second Punic War were extreme. That the war had a negative impact on the economy, can be inferred by analogy with Rome, whose financial difficulties during the war are well known.⁷⁴ In the Greek world, to which Taras still belonged, the lowering of weight standards was a more common expedient than debasement.⁷⁵ Carthaginian coinage, however, seems to have been systematically debased.⁷⁶ Needless to say, a scientific analysis of the silver content of this coinage would be especially interesting in the case of this coinage, given the circumstances and the copresence of two different traditions with regards to debasement.

Whether or not the coins were debased, the question of whether a lowering of weight standard took place as well remains. By adopting a completely new standard, the city would miss out on the opportunity to issue coins at a lesser value than its denomination suggested.

And if the Hannibal coinage reflects a well-established Punic standard, to which it confirms, no negative impact of the war is detectable. The advocates of a Punic standard seem to ignore these considerations for the sake of numismatic precision.

3.2.4. Denomination

Whereas the iconography of both obverse and reverse clearly speaks the traditional language of Tarentine didrachms, the low weight of the coin points in a completely different direction. As a result, the denomination of the coin is controversial. Following his adoption of a Punic weight standard, Rutter describes the coin as a Punic half-shekel.⁷⁷ There is also the alternative view that the coin is a (Tarentine) drachm, which would be approximately consistent with the pre- Pyrrhic Tarentine weight standard, but inconsistent with the post-Pyrrhic weight (with the possible exception for the lighter specimens).⁷⁸ I find this view problematic, since a sudden recourse to the pre-Pyrrhic weight standard, which was considerably heavier than the standard that dominated the post-Pyrrhic part of the third century BC, seems highly implausible given the financial difficulties caused by the ongoing war. Another problem for the Tarentine drachm view is that Tarentine drachms were associated with a different imagery, which did not include

73 Rutter 2001, 106.

74 Crawford 1985, 52–74.

75 By ’debasement’, I mean a certain kind of devaluation by means of reducing the content of precious metals in the coinage.

76 Howgego 1995, 113–114.

77 Rutter 2001, 106.

78 Brauer 1986, 196.

(19)

horsemen.⁷⁹ To summarise, there are three views of the denomination of these coin types: The Tarentine didrachm view, the Tarentine drachm view, and the Punic half-shekel view. Whereas the first view draws on the iconography, the other views draw on the weight.

In my opinion, the latter views (and the third view in particular) underestimate the importance of iconography with respect to denomination. Since the imagery clearly speaks the language of Tarentine didrachms, and given the occurrence of the Tarentine ethnic (ΤΑΡΑΣ) on the reverse, the coin should be understood as a Tarentine didrachm, I think. There are however two problems facing all advocates of the Tarentine didrachm view: the low weight and the non-Greek names (on three out of five of the coin types), in this case Serambos. Since the Punic half-shekel view is able to account for both of these, the proponents of the Tarentine didrachm view face a twofold challenge: to solve the two problems and provide a convincing argument for why the Tarentine silver didrachm view is the more plausible view in light of all the evidence.

3.3. Iconological interpretation

In the iconographical analysis, it was concluded that the iconography of the coin reflects Tarentine traditions of silver didrachms. Questions were raised, however, by the name Serambos on the obverse, as well as by the low weight of the coin, neither of which is consistent with previous issues. In the concluding section of the iconographical analysis, three views on the coin’s denomination were presented. In the third and final stage of the analysis, two of these views, the Punic half-shekel view and the Tarentine didrachm view, will provide the basis for the contextualisation, which will reach its full extent in the ensuing discussion.

The Punic half-shekel view is able to explain the low weight of the coin by the imposition of a Punic weight standard in Taras. As stated before, this is the position taken by Rutter.⁸⁰ Since Rutter offers no argument for his position, I have attempted to make my own reconstruction of the Punic half-shekel view, which I interpret as follows:

During the Hannibalic occupation of Taras, a Punic weight standard was imposed on the Tarentine mint by the issuing (Punic) authority. For some reason, the magistrates responsible for the coinage, whose names were non-Greek and probably Punic, decided to manufacture new dies with traditional Tarentine imagery. They could have chosen Punic imagery, but since they were in Taras, they chose Tarentine imagery – perhaps as a sign of benevolence for the occupied city. Since the coins in the Tarentine mint were struck on a Punic weight standard, the coins were now Punic half-shekels, albeit with the traditional imagery of Tarentine didrachms. On this view, the inscription ΤΑΡΑΣ on the reverse is not to be understood as an ethnic, but rather as a mere indication of the location of the mint.

A possible objection against this view is that its explanation for the Tarentine imagery is weak and unconvincing. Is the mere location of the mint sufficient to account for the Tarentine imagery? The dies were new and could have been engraved in other ways to accommodate with Punic (half-shekel) traditions. If the coin would be a Punic half-shekel, this would be reflected in the imagery.

This objection does not seem to pose a significant threat to the Punic half-shekel view, since its proponents could simply give the answer “yes” to the question. The dies could of course have been manufactured in a number of ways, but since the mint was in Taras, Tarentine imagery was chosen. Instances of such borrowed imagery are known from other places in southern Italy, such as the monetary landscape in Campania during the First Punic War.⁸¹ One might conjecture that Hannibal’s officers needed half-shekels urgently and were indifferent to the imagery used. It is possible to think of a Punic magistrate, say Serambos, who entrusted a Tarentine engraver with the task of producing dies for Punic half-shekel for him. Since the engraver – let us call him Kleon – was accustomed to Tarentine dies, he manufactured a half-

79 Brauer 1986, 196.

80 Rutter 2001, 106.

81 Termeer 2016, 59.

(20)

shekel with the outward appearance of a Tarentine didrachm. Kleon preferred the imagery of the didrachm to the imagery associated with drachms, whose weight tallied with the half- shekel’s, even though didrachms were twice as heavy this these. After all, the didrachm was more important and valuable than the drachm.

It is clearly possible for the proponents of the Punic half-shekel view to account for the traditional Tarentine imagery in such a way. Possible arguments are not lacking, and it seems difficult for the adversaries of the view to turn their opinion that the importance of imagery is underestimated by the half-shekelists into an argument. A stronger argument against the view can be found in the inscription ΤΑΡΑΣ. According to the Punic half-shekel view, this is a mere indication of the location for the mint. The problem with this claim is that it lacks support from the evidence. From other studies of the coinage of southern Italy, it is clear that such inscriptions are ethnics, not mere indications of the location of the mint. In Termeer’s Campanian example, coins from different cities that shared their imagery were distinguished by the ethnics alone.⁸²

What about Serambos? According to the Punic half-shekel view, he was the magistrate responsible for the coin’s production. But is his presence on the obverse a sufficient condition for the coin’s being Punic? Certainly not. A magistrate of Carthaginian origin could of course be responsible for producing a Tarentine didrachm without making it a Punic half-shekel.

The inscriptions on the coin discussed in this thesis are interesting and must not be forgotten, but since the inscription on the obverse and inscription on the reverse point in different directions, the evidence must be considered as neutral. It can hardly settle the question of the coin’s denomination.

My initial assumption that the proponents of the Punic half-shekel view underestimate the importance of iconography has proved insufficient for a rebuttal of the view. However, the biggest problem for the Punic half-shekel view is not to be found in the numismatic data, but in the context of the Second Punic War. Let us recall that the Carthaginian conquest of the city was facilitated by anti-Roman elements of the population. According to Livy, the anti-Roman party was composed of thirteen young Tarentine aristocrats (nobiles iuvenes Tarentini), whose leaders (principes) were called Nico and Philemenus. In 213 or 212 BC – both Livy and modern historians are unsure of the exact date – Nico and Philemenus met with Hannibal to make arrangements for his entering the city, which should take place under the condition that the Tarentines would keep their own laws and possessions, without paying any tax to the Carthaginians nor receive a Carthaginian garrison against their wishes (invitos).⁸³

As noted by Kathryn Lomas, Taras was the only city to make an entirely voluntary approach to Hannibal.⁸⁴ In the light of all this, it is questionable that the Carthaginian presence in the city was as an occupation at all, at least in the ordinary sense of the word. Rather, the Carthaginian presence could be interpreted as a defensive measure against the Romans in the citadel, who could attack at any time. As an ally to the anti-Roman party of Taras, the Carthaginians provided troops to prevent hostilities from the common enemy of the Tarentines and the Carthaginians.

Given the unbalance in the military strength of Taras and Rome, such assistance was certainly welcome. I cannot find any reason to believe that Hannibal should have violated his agreement with the conspirators. By keeping a Carthaginian garrison in Taras against their wishes, he would certainly have done so.

Departing from Livy’s account, I propose that the issuing authority were not Cartaginians at all, but Tarentines, more particularly Tarentines of the anti-Roman party. The reason for their striking of coinage was to defray the expenses in the war against the Romans, a war in which the Tarentines and the Carthaginians fought as allies. There is literary evidence that suggests that the Tarentine war effort was significant, not to say remarkable. In the sea-battle of Sapriportis in 210 BC, a Tarentine fleet of approximately twenty ships led by a certain Democrates crushed a joint Romano-Greek fleet of equal size.⁸⁵ At this point, the war had already turned in Rome’s favour, and it is questionable whether the Carthaginians played any

82 Termeer 2016, 68–69.

83 Livy 25.8.2–3.

84 Lomas 1993, 75.

85 Livy 26.39.1–19.

(21)

role at all in the Tarentine victory, since Livy speaks only of Tarentines. This victory must have been an immense source of pride for the Tarentines and the anti-Roman party. But the happiness was short-lived. When the Romans reconquered the city in 209 BC, Democrates was killed alongside Nico.⁸⁶ It seems likely that Democrates was one of the thirteen young aristocrats mentioned above.⁸⁷

Remarkable war efforts such as this one certainly required money, which would explain why a new series of coins was struck. On this view, the combination of war expenses and the general stress on the economy caused by the war could account for the low weight of the coin, which would then be the result of a massive weight reduction as a means of devaluation. By this reduction, the Tarentine didrachm lost approximately half its weight. It is true that such an extreme weight reduction had never before taken place in the Taras’ history, but on the other hand, Taras had never before experienced such an extreme situation, which, arguably, called for extreme (and novel) measures. Once again, Taras was at war with a superior enemy that controlled its citadel. In such a small area and for such a limited amount of time, the occurrence of a successful devaluation is not unthinkable.

Given this interpretation, where the low weight of the coin is accounted for by a massive weight reduction, the coin does indeed signal financial desperation. But there is also another aspect, untouched as of yet. This aspect is not excluded by the reasoning above, but rather coexists with it. I speak of course of the rhetorical power of coinage. The anti-Roman party needed to assert its authority over Taras, and what better way could there possibly be to do this than to revive the once so glorious coinage of the city, with horsemen, dolphin riders, and – not the least – the ethnic ΤΑΡΑΣ? To make this possible, they maybe occasionally had to resort to the expertise of Carthaginian magistrates and engravers, especially if Tarentine coin production had ceased around the middle of the third century BC, as Rutter and Burnett think.⁸⁸ This would account for Serambos’ appearance on the obverse – but remember that the Greek names Kritos and Philiarchos appear as well – rather than the names of the leaders of the anti-Roman party, like Nikon, Philemenos and Demokrates (the Livian renderings of the names have of course been Latinised). To interpret the coinage as a sign of political independence is likely to be superfluous, since Taras seems to have been independent as a Roman ally as well (if independence is compatible with the presence of a Roman garrison, that is). Rather than celebrating their independence, the anti-Roman party made use of coins to reassert their former power, perhaps aspiring to be the reborn leader of the Italiote league. To speak of a “Roman yoke” that was shaken off, as Head and Evans does, is probably misleading, but on the other hand, the outcome of the Pyrrhic wars and Taras’ post-Pyrrhic status as a Roman ally had certainly diminished the city’s influence in the region. Money was needed, yes, but so was political influence. In this context, the rhetorical power of coinage should not be underestimated.

86 Livy 27.16.3.

87 Fronda 2010, 190.

88 Rutter 2012, 138; Burnett 2012, 308.

(22)

4. Discussion

In the preceding analysis, it was argued that the coin discussed in this thesis is a Tarentine silver didrachm with a heavily reduced weight. In the iconological interpretation that concluded the analysis, an initial attempt at contextualisation was made. It is the purpose of this discussion to situate the results of the analysis in the context of the Second Punic War and make way for a broader discussion of how coinage and economy are affected by crises.

As stated before, the financial difficulties on both sides in the Second Punic War are well known. The economic crisis of Rome has been described by Claude Nicolet, who emphasises that monetary manipulations were only one of many financial expedients employed by the Roman state to defray the expenses of the war. Other expedients included a doubling of the tribute and a loan from Hieron of Syracuse. More importantly, the Second Punic War forced the Roman state to resort to credit. For the first and last time in its history, the Roman state had to borrow from its own citizens.⁸⁹

Following the chronology of the Danish numismatist Rudi Thomsen, which was confirmed and refined by the American Morgantina excavations, Nicolet dates the introduction of the Roman denarius to the Second Punic War, more exactly somewhere between 214 and 211 BC.

Nicolet insists that the introduction of the denarius be explained against the background of a

“sextantary devaluation” (dévaluation sextantaire), in which the weight of the bronze as, whose weight originally had been equal to a libra, was equated with a sixth of the libra.⁹⁰ At the same time, the value of the new (silver) denarius was fixed at ten asses (thence the name denarius).

These intertwined monetary manipulations, the devaluation of the as and the fixation of its value at a tenth of a denarius, should be understood as a reaction to the financial difficulties caused by the Second Punic War.⁹¹ And their effect was lasting. The Romans had experimented with silver coinage before, but for the first time, silver and bronze coinage were now tied together. The system of denominations based on the denarius was to remain for the five hundred years to come, albeit with changes.⁹²

Thomsen had originally placed the introduction of the denarius between 214 and 209 BC, but the time frame was shortened by two years when a votive coin hoard that included denarii appeared during the excavation of the sanctuary of Demeter and Persephone at Morgantina.

According to Nicolet, it is known from other sources that the Romans destroyed this sanctuary between 214 and 211 BC.⁹³ Therefore, the introduction of the denarius must have taken place before this.⁹⁴ Besides providing a terminus ante quem for the introduction of the denarius, the excavation of the sanctuary is noteworthy in another respect, which is even more relevant to the matter at hand. On the floor of the temple, coins that had been cut in two pieces, seemingly deliberately, were found. To a large extent, the coins had their origins in Syracuse, where Hiero II had issued them. Interestingly, some of these coins had been overstruck by the Romans according to the denominational system of the sextantary as. Nicolet suggests that this is the result of a dévaluation à moitié, a devaluation by half, through which the Syracusans responded

89 Nicolet 1963, 419–432.

90 The weight of the libra was approximately 324 g. See Burnett 2012, 302.

91 Nicolet 1963, 432.

92 Howgego 1995, 11.

93 Nicolet’s account is actually a bit misleading. Morgantina was the Roman base in Sicily, and in 213 BC, it was conquered by the Carthaginians. In 211 BC, the Romans reconquered the city. The destruction must be dated to this year. See Marchetti 1993, 30.

94 Nicolet 1963, 434.

The last coin of Taras?: A study of a late Tarentine coin in the collections of the Uppsala University Coin Cabinet.