The monetiform nature of lead tokens in Italy has repeatedly led scholars to conclude that these objects operated as a form of alternative currency. Dressel believed that an assemblage from the Tiber he published represented a privately issued emergency coinage, the ‘till money’ of an innkeeper or grocer.Footnote ¹ Thornton identified these objects as a form of ‘peasants’ money’.Footnote ² Rostovtzeff suggested some tokens acted as surrogates for money within small household economies and groups of clients. In this discussion Rostovtzeff cited Figure 5.1, a token that names two individuals, Olympianus and Eucarpus, as well as the sum of 1,000 sestertii. The portraiture style of the bust, as well as the dates of the tokens associated with the find (admittedly from the Tiber), suggests a date in the Julio-Claudian period or the first century AD. Rostovtzeff, acknowledging that the relatively large sum of 1,000 sestertii was unlikely to be represented by a single lead token, suggested the amount named on the series was the total value of the tokens issued by these two individuals.Footnote ³ The hypothesis is an intriguing one, although other interpretations are possible. The sum might represent the amount bequeathed by Olympianus and Eucarpus as capital to fund particular banquets or acts of euergetism (although it would be a modest amount in such a context), or the token issue might be an act of satire, given during the Saturnalia or another festival.Footnote ⁴ Others have seen this issue as evidence that tokens were used as calculi or counting pieces.Footnote ⁵

Figure 5.1 Pb token, 17 mm, 12 h, 3.95 g. Male bust right, OLYMPIANVS around / EVCARPVS around HS ∞. From the Tiber.

TURS 1460.

Lead monetiform objects elsewhere have been interpreted as emergency small change, but for Roman Italy the evidence does not appear sufficient to come to this conclusion.Footnote ⁶ As explored in the previous chapters of this volume, tokens might be utilised in a variety of ways. Some of these roles do overlap with the functions of money. Tokens could represent a particular value (i.e. a product or experience that was worth a certain amount), and, if issued in advance, form a type of credit. The findspots of tokens across Rome, Ostia and elsewhere (discussed in more detail below) do suggest that at least some were issued before a particular occasion or benefaction. An exploration of the use of tokens in the medieval period has demonstrated that they might act as IOUs to defer expenditure, or as a form of payment in kind rather than cash.Footnote ⁷ Some Roman tokens may have functioned in a similar way (with patrons issuing clients or customers tokens to be exchanged for goods rather than money), but there is no surviving evidence to confirm this was the case. This chapter explores the potential economic roles of tokens in Roman Italy, and the extent to which these objects facilitated the functioning of micro economies.

Was there a shortage of small change in Rome and Ostia during the imperial period that led to the creation of tokens similar to that shown in Figure 5.1? The Roman mint produced quadrantes and semisses, the smallest denominations in the imperial monetary system, throughout the first century AD and into the second, with semis production starting again in the later third century.Footnote ⁸ It has traditionally been thought that the quadrans denomination was utilised mainly in Italy, particularly in the region between Rome and Pompeii. However recent work has demonstrated that later quadrantes, particularly from the reign of Domitian, travelled beyond Italy to be used elsewhere.Footnote ⁹ Although the find evidence from Italy is patchy, the evidence as it survives suggests Julio-Claudian quadrantes were relatively common in the region. Later issues of small change are less frequently found and appear to have been shipped to the provinces in greater quantities.Footnote ¹⁰ From our current evidence, the production of small change in the first and second centuries largely overlaps with the zenith of lead token production in Rome and Ostia (which begins in the first century and continues into the second).

More research is required to better understand the role of the small bronzes within the Roman monetary system.Footnote ¹¹ A semis was half an as, a quadrans a quarter of an as. Surviving price lists suggest the quadrans was too small to have actually purchased anything on its own, apart from entrance into the Roman baths.Footnote ¹² Indeed, Reece suggested the denomination served as ‘change necessary for equity’, that is, solely to be provided as change, much like small one or two cent pieces today.Footnote ¹³ Significant numbers of quadrantes in boxes or dolia in shops in Pompeii suggest Reece may be correct – the quadrans may have existed to ‘provide change’.Footnote ¹⁴ King also pondered the oddity of this denomination, and suggested that perhaps quadrantes did not primarily serve as small change, but were instead issued to fulfil some other administrative requirement – imperial congiaria, for example, although she admitted there is little evidence to support this idea.Footnote ¹⁵ Weigel wondered whether some small bronzes (the so-called ‘anonymous’ issues) were struck for public religious celebrations.Footnote ¹⁶ This hypothesis provoked a rebuttal by Buttrey, who noted that the small number of issues mean that they cannot have been a regular feature of festivals held each year; he concluded we are unable, given the current state of knowledge, to know what the purpose of the quadrans denomination was.Footnote ¹⁷

Ancient literature, particularly satirists, use the word quadrans to highlight a particularly small amount; these same authors also use the phrase 100 quadrantes to refer to the dole given by a patron to his client.Footnote ¹⁸ Martial repeatedly connects this dole with bathing; the entrance to the baths was one quadrans.Footnote ¹⁹ Indeed, Seneca uses the phrase res quadrantaria when describing Roman baths.Footnote ²⁰ The denomination thus must have been useful in this regard – had there been no quadrantes, the purchasing of bath access would have been hampered by lack of appropriate specie. One additional item that might have been purchased for a quadrans seems to have been overlooked to date in modern scholarship: Varro tells us that this was the cost charged for ferry transport by raft.Footnote ²¹ That the Roman government issued small change in the form of quadrantes and semisses reflects that fact that some consideration must have been given to the role of small change in facilitating the economy and everyday transactions.Footnote ²²

During the second century AD the Roman mint ceased production of quadrantes, perhaps because price rises meant they became even less useful, or perhaps because the cost of their production now outweighed their value. We have little information in this regard. Since token production began in earnest in Rome in the first century AD, we cannot argue that lead tokens were produced to fill a void of small change after production of quadrantes ceased. But to what extent did tokens take on some of the roles played by small change, thereby easing any pressure on small change supply that might have occurred?

Quadrantes and semisses bear some similarity to lead tokens in that they did not always carry a portrait of the emperor, and at times quadrantes also lacked a legend referring to the imperial authority. Many quadrantes types, particularly of the Julio-Claudian period (altar, cornucopia, clasped hands and caduceus, modius), are also found on tokens. Indeed, one of the most common types amongst the tokens published by Dressel as a merchant’s ‘till’ was Figure 5.2, which carries an altar on one side and a lituus on the other. The ‘bowl-shaped’ altar appears on several quadrantes issues under Augustus (Figure 5.3) and the lituus also appeared alongside a simpulum on quadrantes of 9–8 BC.Footnote ²³ Of the collection of tokens published by Dressel, this type was by far the most numerous: of the 487 tokens, 205 were of this type.Footnote ²⁴ Dressel suggested that the tokens found in higher numbers were those of the till owners themselves and that the other tokens, present in smaller numbers, represented the issues of others. A significant quantity of this type (139 examples) was also among the Tiber finds published by Rostovtzeff and Vaglieri.Footnote ²⁵ This token issue, it seems, has survived in much higher quantity than any other. Beyond the two assemblages from the Tiber, however, no further specimens have come to light (in excavation publications or museum collections). This might be a reflection of the very small circulation area of tokens.

Figure 5.2 Pb token, 14 mm, 2 h, 2.12 g. Garlanded altar / Lituus.

From the Tiber, TURS 1072.

Figure 5.3 Copper quadrans, 18.5 mm, 6 h, 3.19 g, 8 BC. Cornucopia flanked by S C, PVLCHER TAVRVS REGVLVS around / Garlanded altar, III·VIR A A A F F · around. RIC I² Augustus 425.

But despite these similarities, we should not see Italian lead tokens as emergency or privately issued small change. To begin with, lead tokens simply do not survive in sufficient quantity to have functioned as a currency. The altar and lituus token issue discussed above is an anomaly in terms of the quantity that has survived – we otherwise find only a handful of each type preserved, or a single known specimen. The overall picture may have been affected by museum acquisition practice: for Roman coinage, museums historically have focused on obtaining rare pieces, or selected only one or two examples of a type to include in their collections. A similar phenomenon might have occurred with lead tokens. The token assemblage now housed at the Museo Archeologico Nazionale di Palestrina, coming from illicit excavations, contains far more tokens of the same type than other European national collections. It also contains far more examples of tokens that might have been overlooked by antiquarian collectors as not being of great interest – for example, tokens that carry only a single letter on each side. Like Roman coinage, the overall image gained from tokens in museum collections may not necessarily reflect the realities of what was in circulation. But even those tokens which appear in number (e.g. the 344 ‘altar/lituus’ tokens from the Tiber) still do not seem to have the characteristics of an alternative currency: their concentrated findspot, for example (the Tiber and nowhere else), suggests they did not ‘circulate’, even within a relatively small area. Indeed, the ‘altar/lituus’ series may have been used for an event or distribution in a particular moment in time, after which the tokens lost their value and were discarded.

The quantities and findspots of lead tokens in the archaeological record give the impression of accidental survival of artefacts that were otherwise utilised for a moment in time and then melted down for reuse – the tokens we find today are the result of accidental loss, rather than intentionally deposited for their value. Tokens of bronze and orichalcum also do not survive in large quantities, certainly not the numbers known for other issues of emergency small change in antiquity. The pseudo-coinages that acted as unofficial small change in the region around Pompeii and Minturnae, for example, with designs adopted from the coinages of Massalia and Ebusus, survive in significantly higher quantities.Footnote ²⁶ Moreover, lead tokens do not attempt to imitate official currency; in fact they are consciously different in design, even when coin imagery forms a point of inspiration. Tokens of Roman Italy, in bronze, orichalcum or lead, are also not generally found intentionally hoarded with official Roman coinage, discussed further below. This is an important point of difference from pseudo-coinage or emergency small change, which is found stored together with official coinage – in Pompeii for example.Footnote ²⁷ These characteristics suggest tokens served a different purpose to Roman currency, and were recognised as ‘different’ by users. Some functions of tokens, however, may have overlapped with those of currency.

Tokens and Roman Baths

Some Roman lead tokens specifically name Roman baths, or bath managers (balneatores) and scholarship has suggested a connection between these artefacts and the administration of bathhouses. Rostovtzeff suggested these types of tokens were used to gain access to particular bathing establishments.Footnote ²⁸ Lead tokens have also been interpreted as a method to carry out euergetism in this context: tokens may have facilitated the sponsoring of free access to the baths by particular individuals.Footnote ²⁹ Tokens have also been connected to the internal administration of a bathhouse – used to mediate access to particular services, for example, like massage, oil, food or drink.Footnote ³⁰ Scholarship, however, remains rightly tentative – it is extremely difficult to deduce the precise use of tokens in bathing contexts, and studies of bathing in the Roman world more broadly have revealed that we cannot generalise: customs were different from region to region and bathhouse to bathhouse.Footnote ³¹ Nonetheless, several new archaeological finds have come to light since the last published scholarly discussion of this issue, and it is worth revisiting the question in the light of the recent evidence.

Some of the earliest evidence for tokens in Italy is connected with bathing. During excavations in Fregellae, a settlement captured and destroyed by the Romans in 125 BC, tokens of both lead and bone were uncovered. In a stratum associated with the spoliation of a wall (probably of a private house), a bone token was found with the legend L. ATIN | MEM carved on one side and BALN (or BALIN) with ligate lettering on the other.Footnote ³² Palaeography and the archaeological context date the piece to c. 150–125 BC. The legend might refer to the position of balneator, or a bal(i)neum, bal(i)n(ae) or bal(i)ne; the abbreviated name has been resolved as Lucius Atinius Memmianus. Balneum, balnea, and balnae were words used to refer to baths in the Republic. The term thermae appeared only in the first century AD, although balineum continued to be used, likely to refer to establishments that were not as lavish as thermae.Footnote ³³

A structure that has been identified as the baths of Fregellae has been uncovered, thought to have been erected around 190/180 BC. Excavations of this structure uncovered five lead tokens, found close together in a ‘gruzzoletto’ in the central space of the south entrance of the baths.Footnote ³⁴ The context appears to be one of spoliation; the five tokens were all of the same design, and likely made from the same mould. One side of the tokens displayed the bust of Mercury wearing a petasus with three globules before him, while the other side was decorated with a dolphin swimming right. Pedroni noted the water associations of the dolphin, and wondered if the three globules were meant to represent a particular value: a triens or quadrans, which might have been the cost of entry to the baths.Footnote ³⁵ Globules are used to indicate particular values on Republican bronze coinage; RRC 97/5c and 98A/5 (c. 211–208 BC) have the head of Mercury on the obverse and a ship’s prow accompanied by three globules on the reverse, indicating the coin is a quadrans. As in the later imperial period, the tokens of Fregellae appear to have adopted contemporary numismatic imagery while avoiding straight imitation (choosing a dolphin design instead of a prow).Footnote ³⁶

The ‘pseudo-coins’ and lead tokens in Baetica and Italy in the second and first centuries BC also carry imagery of bathing. As Stannard has established, these lead and bronze pieces were produced in both Baetica and Italy, with issues in both regions carrying similar designs.Footnote ³⁷ Amongst the shared imagery is the type of a ring, two strigils and an aryballos, at times carried by a dog on Italian pieces.Footnote ³⁸ Another series carries an image that was previously identified as a ‘miner’ but which has now been correctly identified as a furnacator, the individual responsible for stoking the fire of a bath’s hypocaust.Footnote ³⁹ A type in struck lead has also come to light. This shows Apollo seated on an omphalos, offering two strigils and an aryballos to a small nude male figure before him; the other side of the token displays a prow and two globules (a design adapted from Roman Republican coinage).Footnote ⁴⁰ In spite of the bathing imagery, the sheer production volume of the Italo-Baetican series, and the fact that bathing imagery is often combined with other types that have no apparent connection to Roman bath culture, meant Stannard concluded that these are more than bathing tokens, although their precise function remains elusive. More recently Stannard has suggested the strigil type might be connected with euergetism – gifts of vinegar and oil to bathers, free entry, or something larger in scale, for example connected to bath building or management.Footnote ⁴¹

Although undoubtedly more work needs to be done, there does seem to be a connection between tokens and bathing in the Republic, which continued into the imperial period. The archaeological evidence, particularly from Ostia, suggests the use of lead tokens within bathing houses, although we should remain cautious in our conclusions. Bathhouses, after all, are places where small items like lead tokens are likely to be dropped and lost, and so it is difficult to distinguish between accidental loss and intentional use. What follows is a discussion of the relevant evidence that connects some tokens to bathing; although each individual instance may not be convincing on its own, the combined weight of the data does suggest a connection between bathhouses and lead tokens. The finds from the baths of the Cisiarii in Ostia have already been discussed in Chapter 3; all the tokens in that complex came from what is interpreted as the frigidarium (Room C), largely from fill contexts.Footnote ⁴²

Excavations in the Baths of the Swimmer (‘Le Terme del Nuotatore’) in Ostia uncovered lead tokens, a token mould half, as well as token manufacturing waste. In stratum 1 of area XVI (dated from the middle of the third century to the middle of the fourth century AD), a fragment of a token mould was uncovered, designed to produce circular tokens showing a beetle.Footnote ⁴³ In an earlier layer (stratum II, dated to the end of the second century and beginning of the third century AD), lead waste was found, which bears the characteristic ‘tree branch’ design of the channels carved into token moulds.Footnote ⁴⁴ The logical conclusion is that this is waste from the casting of lead tokens, which in all likelihood took place on site. For the period in which the casting waste was uncovered, however (Phase 4a, post AD 190–210), the area was likely no longer part of the bathhouse. Area XVI is in the western part of the complex, thought to have been converted into a place of habitation by this time, with XVI identified as a popina or wine bar.Footnote ⁴⁵

In area XXV of the baths, which was an open-air corridor in the northeast of the complex linking several service areas (ovens, a cistern), three lead tokens were uncovered. The first of these carried the imagery of Fortuna and a female figure (probably Juno with patera and sceptre) and was found in stratum VA (dated AD 80–90 and associated with hydraulic installations). The other two tokens were found in stratum III (AD 190/200–225). One was decorated with a furca or a stylised caduceus flanked by stars on one side and the legend F C on the other (TURS 3065; the meaning of F C is unknown). The second was decorated with a palm branch on one side and what is probably the letters PM on the other (TURS 691–2).Footnote ⁴⁶ An additional lead token was later excavated immediately outside the northeast area of the complex, carrying the imagery of Isis or a worshipper of Isis on one side and a semi-nude male figure on the other; the stratum of this piece is dated to AD 160–180/90 and is connected the levelling of the area in the late Hadrianic and Antonine period.Footnote ⁴⁷ The sum total of the token evidence from the Baths of the Swimmer reveals token use throughout the lifetime of the complex (which was built in AD 89–90), although we cannot definitively connect the finds with bathing.

The excavations of the Baths of Neptune (‘Terme di Nettuno’) in the early twentieth century also uncovered a lead token and a token mould, although as is typical of early excavations, we do not possess detailed stratigraphic data. A hexagonal mould half that created nine quadrangular tokens of 8×8 mm, each decorated with three globules, was found during excavations of the southern side of the peristyle of the baths near the via della Fontana.Footnote ⁴⁸ Unfortunately we cannot be certain of the precise context the mould half was found in – whether it was a stratum of use, abandonment or fill.

In the same excavations, conducted by Vaglieri from 1909 to 1910, a lead token was found in a place described as the ‘second taberna’ of the baths, if one counted from the via dei Vigili.Footnote ⁴⁹ The design of the token, as mentioned in Chapter 3, was the lighthouse of Ostia and the legend TI S on one side, and Fortuna seated left on the other side (TURS 61, Figure 3.10; the meaning of the legend remains a mystery). A token of the same type was found in the Terme Bizantine in Room D; Spagnoli suggested this piece likely dates from the earlier period of the baths, in the latter half of the second century AD.Footnote ⁵⁰ Pensabene records a third bath context for this token type, from the Terme di Serapide.Footnote ⁵¹ Although the discovery of three tokens of the same type in three different bath complexes may suggest the use of this token in a particular citywide benefaction (e.g. gifting of free bath access, granting of particular goods for use in the baths), the finds are solitary and the type is also found elsewhere (stray finds from illegal metal detecting, a token that seems to be of this type was also found in a late antique context from Portus).Footnote ⁵² The finds of the lighthouse token series may thus simply reflect that the piece was carried across the city. Apart from the example of the baths of the Cisiarii, tokens are found in very small numbers in bath complexes, and not found collected together as one would expect if they served a particular exchange function within the complex. That said, it may simply be that, since Ostia continued to be inhabited into late antiquity (when tokens appear to have been used differently), no abandoned or preserved ‘hoard’ exists to be discovered. Spagnoli is currently working on publishing a corpus of eight tokens from the area of the Terme sotto la via dei Vigili, which may serve as a complementary example to the relatively large amount of tokens from the Terme dei Cisiarii.Footnote ⁵³

A far more significant number of lead tokens (150 in total) were found during the 1981–3 excavations of a Roman bath complex in Alameda in Spain.Footnote ⁵⁴ The baths were constructed in the latter half of the first century AD. Twenty-seven coins were also found, dating from Augustus to the second half of the second century AD, with one outlier – an antoninianus of Galerius. The tokens, also thought to date to the first few centuries AD, were in a poor state of preservation, but carried a variety of designs, including male and female heads (including radiate male heads), animals, vessels, globules, stars and a cross with four globules design. Mora Serrano notes that they seem to fall into two diameter groups: one of 15–16 mm and one of 9–12 mm.Footnote ⁵⁵ The excavation records do not reveal much about the findspots of these tokens within the complex, and Mora Serrano refrains from interpreting their use – the bathhouse may also have had a cultural-religious aspect, indicated by a rectangular edifice at the foot of the adjacent La Camorra mountain and an altar found nearby dedicated to Isis Bulsae.Footnote ⁵⁶ Although the quality of the find information is not ideal, the existence of a large number of tokens from this site, in comparison with far fewer coins, does hint at use within the bathing complex, whether or not bathing was undertaken as part of a cultic experience.

No other published bath complexes from the Roman Empire carry such a high number of token finds, but lead tokens are found in these contexts. In Ephesus, four lead tokens were found in a debris layer of the frigidarium of the bath on the western slope of Mount Pion. The tokens are uniface and all carry the same design: the forepart of a boar being speared by Androklos, the mythical founder of the city.Footnote ⁵⁷ A theta is placed above Androklos’ spear. The same image (without the theta) is found on Ephesus’ third century civic coinage.Footnote ⁵⁸ Baier suggested these tokens might have been used to access the private bath complex, with the theta (Θ) perhaps referring to therma (θερμά), warm baths.Footnote ⁵⁹ A second example from outside Italy is the find of a single token with the legend L. II A within a wreath, which was uncovered in the sediments of the frigidarium drain at Caerleon in Britain.Footnote ⁶⁰ This is just as likely to be an accidental loss as an object connected to the functioning of the bathhouse.

Remarkably, the small amount of token finds from baths in Roman Italy contain no examples that make specific reference to particular bathing establishments, bathing activities, or individuals connected to bathing complexes (e.g. those with the position of balneator). And yet tokens of this nature exist, offering us further evidence for the connection between at least some tokens and this popular Roman activity. TURS 886, for example, carries the legend BALI|NEVM GER|MANI, which has been interpreted as the name of a particular bathing establishment that is otherwise unknown (Figure 5.4).Footnote ⁶¹

Figure 5.4 Pb token, 18 mm, 12 h, 4.09 g. BALI|NEVM / GER|MANI.

TURS 886. Reference Rostovtzeff and ProuRostovtzeff and Prou, 1900: no. 415b.

Other tokens types carry the legend BAL, thought to be an abbreviation of balineum or balneator.Footnote ⁶² TURS 888, for example, carries the legend BAL on one side and the legend TIC|ILL on the other. In his catalogue, Rostovtzeff suggested the legend might be understood as Bal(ineum) Ti. C(laudi) Ill(ustris) or similar. More recently, Bruun has suggested the type may have referenced the balneum Tigellini, an establishment also mentioned by Martial.Footnote ⁶³ Quite often the rendering of G on Roman lead tokens looks very close to (or indeed the same as) a C, so the suggested reading is very plausible. Indeed, on the specimen now in the Museo del Castelvecchio of Verona, the letter does appear to be a G.Footnote ⁶⁴ The individual named may be the Tigellinus who acted as praetorian prefect under Nero.Footnote ⁶⁵ A bal(ineum) nov(u)m is also mentioned on lead tokens (TURS 887). Rostovtzeff believed that a token carrying the legend L·DOMITI·PRIMIG, accompanied by the image of an amphora of oil and a jug, referenced the bathing establishment of one Lucius Domitius Primigenius (the other side of the token issue was decorated with a ring from which strigils and an ampulla were suspended).Footnote ⁶⁶ The existence of a name consisting of a tria nomina and the absence of the letters BAL on this series leaves open the possibility that these tokens may have been issued as part of a bathing benefaction by Primigenius – free access to a particular bath house, for example, or free oil.

Where a name occurs directly before BAL, it is likely that we have reference to a balneator, the manager or contractor of a complex. TURS 892 carries the legend IVL | BAL, which likely references a balneator named Jul(ius); Mercury seated on a ram is shown on the other side.Footnote ⁶⁷ Another issue carries BAL on one side and IV on the other – one cannot know whether this is a number (IV), a reference to a balneator named Julius, or a complex called the bal(ineum) Iu(lium).Footnote ⁶⁸ TURS 890 has the same ambiguity, carrying the legend BAL on one side and FEL on the other – this may reference a bath manager, a bathhouse or perhaps communicate a wish for the possessor of the token to have a good bathing experience (fel(iciter)). A similar sentiment is expressed on a mosaic from the bathing complex at Thamugadi in North Africa: the image of sandals is accompanied by the text BENE LAVA.Footnote ⁶⁹ TURS 893 carries the legend SVB | BAL, which likely refers to someone with the cognomen Sub(erinus) or similar; Victory is portrayed standing on the other side.

Although the full range of responsibilities associated with the position of balneator may never be known (and indeed, these may have varied from establishment to establishment), it seems that the word could reference an owner, leaseholder or manager of a private bathhouse.Footnote ⁷⁰ The position was normally found in bathhouses that could not afford numerous employees.Footnote ⁷¹ Nielson suggests that in the vast majority of cases the leaseholder employed a manager for the bathhouse, normally a freedman, a member of the lower classes, or a slave.Footnote ⁷² Ulpian’s Digest records balneatores and fornacatores (fire-stokers) as instrumenta balnei (‘bath equipment’) included as part of the property left in a legacy.Footnote ⁷³ Epigraphic evidence does little to clarify the role of these individuals.Footnote ⁷⁴ But textual evidence does provide some insight into at least one of their functions: the balneator is mentioned in texts as the person who, among other activities, accepts the fee from a customer and then allows them to enter the bathhouse.Footnote ⁷⁵ This does suggest that those tokens naming particular balneatores may have been used to gain access to particular bathing establishments. The fact that no thermae are named on Italian tokens may indicate that these artefacts were mainly used within smaller, private establishments, the type that attracted the term balineum and which employed balneatores. However, the presence of tokens in larger thermae in Ostia suggests otherwise; it may be then that smaller establishments had to take pains to be specific on their tokens, while larger establishments needed no such identifying legend. If the tokens discussed above were issued by particular balneatores, they provide a rare piece of material culture associated with these elusive individuals.

If tokens were used within some establishments as access tickets, or to facilitate the distribution of foodstuffs or oil as acts of euergetism, then these objects would have, intentionally or not, lessened the pressure on the Roman supply of small change. Lead tokens may have been supplied as ‘change’ in the absence of available official specie – customers, for example, may have been given a token for additional future access to the baths in lieu of change. Alternatively, patrons may have given their clients tokens for bathing, although in this case one would imagine the Roman preoccupation with prestige would have meant that the tokens carried direct reference to the benefactor. At any rate, the gifting of a token to use to access a bathhouse would explain why many of the tokens that carry reference to bathing have been found outside of bathing establishments. Tokens may then have, even unintentionally, lessened the pressure on the quadrantes in circulation in Rome and Ostia.

Did tokens used in bathing contexts encourage the user to identify themselves in the image? As explored elsewhere in this volume, this practice can be identified for tokens used in other aspects of Roman social life, for example in cults. Rostovtzeff recorded that an unpublished bronze token was in the Bibliothèque nationale de France showing a male standing right dressed in an exomis, with his right hand on his breast and carrying a strigil and ampulla in his left hand. The other side of the token, Rostovtzeff recorded, carried the image of a ring with strigils and an ampulla within a wreath.Footnote ⁷⁶ The token is shown here as Figure 5.5. There appears to be no hard or fast rule about whether bathers went nude or merely scantily clad; both are referenced in surviving evidence.Footnote ⁷⁷ Depictions of bathers in surviving material culture from the Roman world are rare, so there is little evidence to compare against the depiction in Figure 5.5.Footnote ⁷⁸ Unfortunately the items the male figure is holding, and those shown on the other side of the token, are not particularly clear. The figure may indeed be holding the instruments associated with bathing, though one wishes the details of the proposed strigils were clearer. The reverse of the token shows an ovoid vessel which looks very similar to a club, and if the item next to the vessel is to be interpreted as strigils, then the die engraver has run them together so that they form one continuous line. The representation of two strigils facing each other (and ultimately joined) in this way is unusual, if not unique, as is the representation of the individual on the obverse. The rarity of depictions of bathers, however, means the singularity of the representation should not surprise. If this is a bather, we should note that what is represented is not the elite experience of bathing (in which an individual was accompanied by attendants who carried various paraphernalia), but a rather more humble representation.

Figure 5.5 AE token, 13 mm, 6 h, 1.44 g. Male figure standing right wearing a very short tunic gazing upwards, right hand on his breast, left hand carrying an ampulla and strigils (?) / Ampulla (?) on left, next to two strigils (?) on right; dotted border.

Unpublished, BnF inv. FRBNF45877423.

Cohen connected a further bronze token in the Paris coin cabinet to bathing, reproduced here as Figure 5.6. Cohen reported that the legend, which is located on the lower part of the obverse, reads BAL on the left and LORVS on the right. Cohen suggested that the BAL referred to balnearius, and that the two individuals shown may be two bathers. The legend is no longer legible, unfortunately (although ‘LORVS’ on the right seems plausible with what remains), and the figure on the right hand side of the obverse appears to be leaning on a thyrsus and so is more likely Dionysus. It is not possible, therefore, to confidently place this piece among the tokens connected to bathing. The Paris coin cabinet also possesses a series of orichalcum tokens that show nude figures in a variety of poses, accompanied by numbers on the other side.Footnote ⁷⁹ Campana, noting that these pieces appear to be concentrated in one European collection, and that they have a regular die axis of 12, suggested that they were more modern creations, made during the Renaissance or even later.Footnote ⁸⁰ These tokens are certainly of a different style to the orichalcum pieces carrying imperial portraits and other designs that are securely dated to the Roman imperial period.

Figure 5.6 AE token, 22 mm, 6 h, 3.62 g. Male figure on left standing with right hand on hip, facing Dionysus (?) on right, who leans on thyrsus with left hand and has right leg crossed in front of the left, worn legend around / XV within dotted border within wreath. Cohen: vol. VIII, 266 no. 9.

Lead tokens present similar difficulties in interpretation. A series of tokens showing a nude male crouching with arms outstretched as if he were about to jump were assigned by Rostovtzeff to the section of his catalogue connected to baths, suggesting the figure stood as if he were about to jump into a body of water.Footnote ⁸¹ On Figure 5.7 the presence of Neptune on the other side of the token does evoke water, but it is uncertain if the nude figure is meant to be a bather. No other surviving material culture from Roman antiquity seems to show bathers acting in such a fashion, although the vibrancy and sense of fun is characteristic of tokens. Indeed, the image recalls Seneca’s description of the individual who jumps into the bathing pool with ‘unconscionable noise and splashing’.Footnote ⁸² Literary descriptions of such ‘characters’ with annoying habits at the baths are also found in Martial, and if Rostovtzeff is correct in seeing a connection to bathing here, the design might have been intended to bring a smile to the face of the user as they ‘recognised’ one such annoying individual.Footnote ⁸³

Figure 5.7 Pb token, 17 mm, no other data recorded. Neptune standing right holding trident in right hand and dolphin in outstretched left / Nude male bending at the knees with both arms outstretched before him.

TURS 901.

Tokens showing paraphernalia associated with bathing – strigils and oil flasks – are easier to identify. Several lead tokens amongst the assemblage from Rome and Ostia show strigils and a vessel for oil, at times hanging from a ring or anulus.Footnote ⁸⁴ Figure 5.8 shows an oil jug above a strigil; the combination probably references the practice in which oil would be applied to the body, and then dirt and the oil would be scraped off with the strigil after exercise. The same scene (a jug above a strigil) appears on another token series, TURS 900. This time the image is combined with a type on the other side that shows an altar inscribed with the letter S, a palm branch and wreath. What the S may signify here is unknown. The use of the same imagery, as well as the common use of S, palm branch and wreath, suggests that these token types may have belonged to a single series or issuer, although no specimen of TURS 900 is currently available for first hand examination, meaning we cannot know if they are of the same style as Figure 5.8.

Figure 5.8 Pb token, 19 mm, 12 h, 5.46 g. Jug above strigil / Wreath on left, next to palm branch and S.

TURS 899, Reference Rostovtzeff and ProuRostovtzeff and Prou, 1900: no. 105bis.

Since Rostovtzeff’s catalogue several additional token types referencing bathing have come to light. Amongst the tokens now in the Museo Archeologico Nazionale di Palestrina is a token that carries the image of a dolphin with the legend BAL above; the other side of the token carries an uncertain design (Figure 5.9). A more worn specimen of the same type exists in Milan (also quite heavy at 9.08 g) and was published by Overbeck, who read the legend as ‘CAE … ’ and wondered if the uncertain image was a portrayal of gladiatorial weapons.Footnote ⁸⁵ The reading on the Palestrina piece is quite certain; the representation of a dolphin accompanied by an abbreviation of balineum (BALN) is also known on a token found at Puente Melchor, located in Puerto Real in Spain.Footnote ⁸⁶ The Museo di Castelvecchio in Verona also possesses a lead token issue not known to Rostovtzeff: on this piece two strigils and an aryballos hang from a ring on one side, and the legend IVC is found on the other.Footnote ⁸⁷ TURS 563, which Rostovtzeff reported as bearing a comic mask on either side, may also carry a ring from which two strigils and an aryballos are suspended (Figure 5.10). The motif on the other side of the token has attracted varying theories without a definitive identification; Scholz entertained the idea it represented an oval-shaped loaf of bread, while Turcan wondered whether it represented the teats of a sow.Footnote ⁸⁸ A decorated oval shield is also a possibility.

Figure 5.9 Pb token, 22 mm, 12 h, 9.43 g. Dolphin swimming right, BAL above / Uncertain image (Column (?) above rectangular object, person holding spear or sceptre on right, horse with head turned back and foreleg raised on left?) cf. Reference OverbeckOverbeck, 2001: no. 50.

Figure 5.10 Pb token, 14.7 mm, 12 h, 3.51 g. Long oval object decorated with five pellets / Ring with two strigils hanging down on either side of a container of oil (aryballos).

TURS 563.

Of course, as has already been revealed by the discussion of the tokens in the baths of the Cisiarii in Chapter 3, other imagery may equally have been used on tokens within a bathing context. It is hoped that future finds may further elucidate the situation. However, the combination of bathing imagery with legends that may be abbreviations of names (e.g. S, MOF, PVR), the use of festive imagery on Figure 5.8 and, above all, the tiny quantities in which these tokens survive, suggests that euergetism is a likely context for their use. If tokens acted as entry tickets for bathhouses on any regular basis, one would imagine we would find them in much larger quantities. If, however, these items were used to facilitate occasional acts of euergetism (free entry, gifts of oil, or other goods), then their limited occurrence in the archaeological record can be better understood. Indeed, if lead tokens were used in a variety of euergetic acts (e.g. food distributions, banquets, bathing), then they form an important (and recurring) part of the material tradition of this practice within central Italy.

Tokens and the Tabernae in Ostia

The presence of tokens in the popina adjacent to the Baths of the Swimmer raises the question: were tokens manufactured and used by small commercial establishments in Ostia? The biggest concentration of tokens in connection to this question comes from the tabernae in the theatre. Unfortunately, however, the theatre at Ostia was used in one way or another from late antiquity into the Middle Ages and beyond, which means that the finds may have originally come from a multitude of places – we cannot assume the theatre was the original place of deposition. The wide variety of finds associated with the tokens, detailed below, demonstrates this. Nonetheless, since this evidence has not been gathered together before, it is worth presenting here. Wherever these pieces were originally deposited, they are nonetheless to be connected with Ostia since lead tokens did not frequently travel from their place of issue.

The theatre at Ostia was originally built by Agrippa during the reign of Augustus; it was later renovated under Septimius Severus, before being modified again in the fourth century to allow aquatic displays.Footnote ⁸⁹ Early archaeological investigations were begun in the latter part of the nineteenth century by Lanciani, but of relevance to our discussion here is the later work by Vaglieri, who focused on uncovering the outer structures in the early twentieth century. Vaglieri had an interest in lead tokens, publishing a series from the Tiber with Rostovtzeff in 1900, and making careful mention of any finds of this nature in his reports in the Notizie degli Scavi and Giornali degli Scavi.Footnote ⁹⁰

Vaglieri’s excavations of the shops running around the outside of the theatre uncovered tokens, moulds and casting waste. Excavations of the back room of one of the tabernae beneath the cavea uncovered a complete cast from a lead token mould, with the eleven tokens still attached to their casting. On one side the tokens carried the letter S and the other side the letter N; Vaglieri posited that perhaps this particular casting had not been to the maker’s satisfaction, and so it was discarded.Footnote ⁹¹ Small marble figures were also found (including portrayals of a male youth and Venus), as well as part of a marble ‘trapezoforo’ or table leg. Marble bearing graffiti and part of an inscription (M | ORDIE | IVS) were also found, as well as three bone handles and six coins (one of larger diameter and five smaller in size).

In 1913, Vaglieri turned his attention to the entrance to the theatre and what he labelled ‘taberna 2’. He observed that beneath a dump layer was a stratum of fine earth mixed with a dark substance and water, which he suggested might be the remains of marble working.Footnote ⁹² The layer did indeed contain a significant quantity of marble, including a female portrait of the second century AD, the torso of a satyr, the torso of a female maritime (?) divinity, a Hellenistic style head of Venus, two fragmentary inscriptions and a token mould for seven quadrangular tokens of c. 10 mm×10 mm size. The design of the tokens was only roughly sketched and not fully legible.Footnote ⁹³ A mould for circular blank tokens was also found – this mould may not have yet been incised, or else was intended to create a ‘blank’ design on one side of a token, a phenomenon known on several issues within Rome and Ostia.Footnote ⁹⁴ The moulds were likely present here because of their material, and were assembled alongside other marble objects.

The third taberna of the theatre also housed a significant quantity of marble, coming from a fire stratum and the stratum beneath this. The context here, according to Vaglieri, appears to be one of fill and dumped marble, containing numerous sculptures (including the torso of a Nereid and a seated philosopher) and some inscription fragments. Included among these was a further token mould that would have once created tokens carrying the number IV.Footnote ⁹⁵ The fill from around the theatre also uncovered a token mould amongst other marble finds – this mould have would have created circular tokens carrying the letter P.Footnote ⁹⁶

Although the tabernae furnished numerous token moulds, the archaeological contexts of these pieces means it is impossible to extrapolate much. But it is worth noting that tokens have been found associated with theatres elsewhere. As mentioned in Chapter 3, excavations from around the theatre at Nemi uncovered three lead tokens.Footnote ⁹⁷ Recent excavations of the theatre in Sardis also uncovered a handful of tokens.Footnote ⁹⁸ A bronze token found in the fill (no. 114.1) was decorated with the bust of Artemis and a stag accompanied by a legend referring to Ephesus (ΕΦΕ-CIΩΝ); DeRose Evans suggested that this piece, and the coins found in the fill, were likely used to buy food and drink in the theatre. As discussed in Chapter 2, Martial also seems to indicate that some sort of token was used to mediate the distribution of food and drink during spectacles.

The state of the evidence in Ostia, however, does not allow us to reach a definitive conclusion about whether particular tabernae at the theatre utilised tokens. During cleaning associated with excavations of a taberna along a street to the west of the Capitolium and south of the so-called Piccolo Mercato, a token (TURS 1478) was uncovered, but the context is not particularly secure.Footnote ⁹⁹ In 1914 two tabernae along the decumanus were excavated; the Giornale degli Scavi records that during this work a mould half was uncovered, which would have created tokens bearing the number IV (the design of the token cavities on the right side of the central casting channel) as well as tokens decorated with a palm branch (token cavities on the left side of the central casting channel).Footnote ¹⁰⁰ No assemblage of lead tokens in Rome or Ostia has been found inside a taberna, which would demonstrate they were used in a manner similar to ‘merchant’s tokens’ of later periods.

Temporal Value? Tokens and Find Contexts in Ostia

An examination of the findspots of tokens in Ostia suggests these objects circulated throughout the city and, as with other small finds of low value, were subject to casual loss.Footnote ¹⁰¹ Indeed, this impression confirms the picture presented elsewhere in this volume: tokens were artefacts that were manufactured sporadically and used for brief moments in time. The tokens that survive to the present day appear to be those that were lost or had lost their value, ending up in contexts of fill or abandonment. The findspots of tokens and token moulds from Ostia not already discussed in this volume are presented here. The overview brings this material together for the first time, and in doing so reveals how widespread tokens were in Ostia in antiquity.

Most of the published lead tokens from Ostia come from the early excavations of Vaglieri: his interest in the artefacts meant that he focused on recording them and that he made a particular note of their existence in the Notizie degli Scavi. The concentration is thus the result of Vaglieri’s interests, which were not shared by later excavators. The recent work of Spagnoli on tokens across the settlement has demonstrated that tokens have also been found elsewhere in Ostia, even if they are not yet published.Footnote ¹⁰² A preliminary list of moulds from the settlement has been compiled by the present author; more unpublished examples may sit in the stores.Footnote ¹⁰³ Spagnoli’s ongoing work will undoubtedly reveal further find contexts from other excavations, but the tokens and moulds published to date already illustrate the widespread nature of token use in Rome’s harbour.

Vaglieri’s excavations of Ostia’s decumanus in 1913–14 uncovered several lead tokens. Work in 1913 was carried out along the decumanus up to the via della Pistrina: a mould half was uncovered for circular tokens of 13 mm bearing an image of standing Fortuna, as was a token of 23 mm decorated with a rudder on one side and cornucopia on the other (TURS 2421).Footnote ¹⁰⁴ The written records of the excavation (the Giornale degli Scavi) record a further four lead tokens associated with the campaign along the decumanus; the information given does not provide a precise context.Footnote ¹⁰⁵

The tokens of the Baths of Neptune have already been discussed. Across the road from these baths tokens were uncovered around the so-called Caserma dei Vigili, the seat of Ostia’s fire brigade (vigiles). Excavations uncovered two token mould halves: one in 1911 that created tokens carrying the letter F, and one in 1912 along the via Fullonica which cast tokens showing a cornucopia within a wreath.Footnote ¹⁰⁶ As Vaglieri observed, the Caserma had already been subject to large amounts of spoliation in antiquity.Footnote ¹⁰⁷ Excavations of the porticus in front of rooms 1 and 2 on the north side (from which one enters the complex), uncovered two tokens of the same type. Vaglieri recorded that the tokens were worn and so the identification is not definitive, but from his description they appear to be of the type TURS 869.Footnote ¹⁰⁸ Figure 5.11 shows a token of the same type. Rostovtzeff had connected the token with the iuvenes of Ulubrae, having interpreted the legend as Cur(ator) Iu(v)e(nium) Ul(u)b(ris). The find of two tokens in the same location suggests that this series is better associated with Ostia. The three birds reference the Capitoline triad (Jupiter, Juno, Minerva). While CVR most likely references a curator, IVE and VLB remain difficult to resolve.

Figure 5.11 Pb token, 19 mm, 10 h, 3.72 g. Eagle standing right with head turned back left. CVR above left, IVE on the right / Peacock and owl standing right with closed wings, VLB above left.

TURS 869, Reference Rostovtzeff and ProuRostovtzeff and Prou, 1900: no. 247c.

Excavations along the via dei Vigili uncovered a lead token showing Fortuna on one side and P within a wreath on the other (TURS 2226).Footnote ¹⁰⁹ The fullonica flanked by the via di Fullonica and via delle Corporazioni furnished a bronze token carrying a palm branch on one side and a figure within a double wreath on the other.Footnote ¹¹⁰ The excavations of the Piazzale delle corporazioni to the west also uncovered tokens and token moulds. The 1912 work along the portico behind the theatre and of the east side of the Piazzale uncovered a token mould half for nine circular tokens of c. 5 mm (the token cavities were blank).Footnote ¹¹¹ A lead token was also uncovered with Salus on one side and Fortuna on the other (TURS 1993).Footnote ¹¹² During work on the large square in front of the ‘Quattro Tempietti’, Vaglieri reported that in the ruins from the street a lead token was found, decorated with a wreath on one side and an unidentifiable figure on the other.Footnote ¹¹³

Along the via della Fontana, which ran along the western side of the Baths of Neptune and the Caserma dei Vigili, several token mould halves were uncovered. One, intended to create circular tokens of 15 mm bearing the letter E, was found during the clearing of a house.Footnote ¹¹⁴ An antoninianus of Philip the Arab and a bronze coin of Valerian were also found, along with seals, pieces of lead, lamps, a palombino marble weight, nails, glass vessel fragments and other finds. In the same year a further mould for seven circular tokens of 9 mm was found, as well as a coin of Claudius. The design of the mould cavities on this piece proved too rough to make out.Footnote ¹¹⁵ Excavation work in the Casa dei Dipinti, a large apartment block dating to the second century AD, also revealed token activity. A palombino marble mould half was found above the covered corridor at the entrance to the house along the via delle Fontana; the mould was intended to create quadrangular tokens of 13×12 mm decorated with five globules.Footnote ¹¹⁶ In a room inside the block, which Vaglieri records as being located to the north of that already excavated, three lead tokens were found: one bearing a palm branch on one side and the legend SP on the other (TURS 3349), one with the number V on one side and a wreath on the other and the third illegible.Footnote ¹¹⁷ The room was decorated with frescoes of fish and goats, and contained a white mosaic pavement. Other finds included a gold ring, three denarii (one dated to the reign of Hadrian), bronze objects (e.g. fibulae) and an iron key.

In 1914 excavations began to the south west of the via dei Molini and three lead tokens were among the finds. The types were ‘AISV(?) | PAN(?) / Centaur’, ‘SVNV | IVLI / elephant’ and ‘cuirassed and helmeted bust / figure standing left’ (TURS 163).Footnote ¹¹⁸ Two further tokens were uncovered in October 1916 from excavations on the west side of the large tufa wall that runs along the west side of the street. The designs on the tokens were unrecorded; the area possessed a large amount of construction materials, which were perhaps accumulated in antiquity.Footnote ¹¹⁹ Tokens were also found inside the structures lining the via dei Molini. In Room 14 of the Caseggiato dei Molini, which housed a bakery, a lead token was found – the room originally held an oven but was converted into a shop (not associated with the bakery) in the later third century.Footnote ¹²⁰ The token was decorated with a male figure on one side and a quadruped on the other. On the other side of the via dei Molini stood the Grand Horrea, a large warehouse, where in 1956 a token was found, but neither the precise location of the find within the warehouse, or the design of the token is recorded in the Giornale degli Scavi.Footnote ¹²¹

An alley separates the Caseggiato dei Molini from the Caseggiato di Diana (House of Diana), which sits on the via di Diana and the via dei Balconi. Excavations in 1915 along the via di Diana uncovered three lead tokens, with a fourth uncovered nearby to the west in 1914.Footnote ¹²² To the east of the House of Diana is the so-called Piccolo Mercato, a storage structure. Work around this area in 1911–12 uncovered four lead tokens. In 1911 work began to remove the earth from Petrini’s earlier excavations; along the street parallel to the south side of the Piccolo Mercato a lead token decorated with a wreath on one side and an S on the other was uncovered, along with fragments of lamps, a weight, coins and a bone stud.Footnote ¹²³ A token of type TURS 623 (elephant / bull) was also uncovered in excavation work in the area that year.Footnote ¹²⁴ The continuation of the campaign in 1912 revealed a further token of type TURS 2372 (standing Fortuna / PS).Footnote ¹²⁵ In the same year excavations took place on the street parallel to the west side of the Capitolium and the south side of the Piccolo Mercato. Vaglieri records that during cleaning operations another token was found of type TURS 1478. Another specimen of the same design is shown as Figure 5.12; the type is found with relative frequency in museum collections. Earlier work on a street parallel to the Capitolium also uncovered a palombino token mould half, designed to create circular tokens carrying the legend CT.Footnote ¹²⁶

Figure 5.12 Pb token, 22 mm, 7.6 g, die axis not recorded. ROM|VLA / Shield (?) with A above G on left and A above S on right.

TURS 1478.

Moving further west, a token mould half was uncovered during the excavations of the Insula delle Ierodule in Regio III during 2003–4. The mould half was made of lunense marble, and would have cast quadrangular tokens bearing an anthropomorphic figure holding a sceptre (?) with a container (modius?) between its legs. The mould was found in a corridor (Area 10) in a stratum of abandonment. Clay moulds and glass were also found, leading the excavators to wonder if there was not some type of artisanal workshop located within the building. Two coins were also found, both of the third century AD; the building itself was destroyed in the latter part of the third century.Footnote ¹²⁷

Excavation activity in the sewers of the town, the Tiber and the harbour have also uncovered lead tokens. The 1908 excavations of a sewer that ran towards the Tiber uncovered a token decorated with a lion walking right.Footnote ¹²⁸ In 1910 the excavations of a sewer that ran beneath the pavement of a room that contained two ‘vasche’ (basins) of the baths along the via del Teatro uncovered two lead tokens; one with the design ‘H / CC’ and the other ‘A / C’ (TURS 3363). The excavation also uncovered bronze coins, a bronze token (type unrecorded), as well as terracotta and glass fragments.Footnote ¹²⁹ In 1910 excavations of the ‘piazzale’ between the two streets immediately behind the harbour of the town uncovered a 14 mm circular lead token of the type ‘palm branch / vessel’.Footnote ¹³⁰

The dredging of the Tiber riverbed between the Palazzo imperiale and Ponte delle Scafa uncovered thirteen lead tokens. Of those that were legible on both sides three were decorated with a patera on each side, one was of the type ‘cornucopiae / CPA’ (TURS 2432), one was of type TURS 2517 (Hercules standing with club and cup, A EV / TFS between two palm branches), one of type TURS 2792 (Ram standing right / M), with another token being described as uniface and decorated with a young male (Julio-Claudian?) head next to a lituus and praefericulum.Footnote ¹³¹ Twenty-one coins were also recorded, dating from Augustus to late antiquity, as well as other lead, bronze and iron objects.

A token mould half was found during excavation work in the area formerly owned by the Alobrandini family, which would have cast tokens showing an eagle.Footnote ¹³² The ruins to the northwest of the eastern gate of Ostia also furnished two tokens; the first a bronze quadrangular token said to show a horse running and the second a lead piece that was poorly preserved.Footnote ¹³³ A token decorated with a wreath on one side and palm branch on the other (TURS 3286) was found in the via delle Foce in 1955.Footnote ¹³⁴

What this brief survey illustrates is that tokens, as well as token moulds, appear to be scattered throughout Ostia. Further finds in the town have been mentioned elsewhere in this volume (e.g. the Sabazeum). Although many of the finds are from fill, disturbed contexts or abandonment strata, the presence of token moulds and tokens across the settlement suggests that these objects were in use amongst a wide segment of the population. If these objects were used to mediate time sensitive acts of euergetism or distribution, then the presence of moulds and tokens in these types of archaeological contexts is understandable – once they had been used for a particular occasion or period of time, the objects became worthless and were thrown away. That only a handful, or even a single, token from a particular series survives suggests an artefact that was used and then destroyed – the majority of tokens, one imagines, were handed in and then melted down for reuse; those that survive escaped this process in some way.

Can we estimate the frequency with which tokens were used in Ostia, and hence whether they served to lessen the pressure on small denominations in the town? The total tokens found in excavations in Ostia, indeed, even the total number of token types recorded by Rostovtzeff (which largely come from Rome and Ostia), seem small when one notes that these are the remnants of an estimated 200 or 250 years. It seems more likely, as this book has suggested, that the tokens were used in occasional acts of euergetism. Sponsoring entry to baths, or distributing particular goods on particular occasions, may have meant that for a small moment in time individuals did not need the specie to purchase these goods themselves, but this would not, one imagines, have had an enormous impact on the demand for small change within daily life.

Travel and the Secondary Lives of Tokens

According to current find evidence, the vast majority of tokens remained close to their place of manufacture. But some did travel, and it is worth exploring the circumstances that resulted in these objects being carried from one region to another. In some instances the materiality of tokens, particularly those made of bronze, which look very much like money, may have meant they acquired a secondary context as a coin to be used in the economy. But if this occurred, it did so as a series of isolated instances: the very small numbers involved suggest tokens were not shipped elsewhere en masse. The study of the shipment of coin blocks by Frey-Kupper and Stannard points out that the low value of bronze coinage (and by extension, tokens) meant that very large numbers of coins would be needed for long distance transfers of this type of specie to be worthwhile. Smaller numbers of finds suggest the movement of people and groups, who carried such pieces on their person.Footnote ¹³⁵ Indeed, the findspots of tokens that travelled, discussed in further detail below, suggest that in many cases they were recognised as something different from official currency or even pseudo-currency. The imagery of tokens, particularly spintriae, held evident appeal, which may have resulted in these objects being curated. Finds of lead and bronze tokens far from their place of manufacture, like the lead Egyptian tokens found in Italy discussed in Chapter 2, may reveal regional networks and commercial relationships.

Bronze and orichalcum tokens have been found across the Empire (for a discussion of these types and images, see Chapter 1). For the issues carrying portraits of Julio-Claudian emperors on the obverse and numbers on the reverse, recorded findspots are rare. Two specimens are recorded as part of the Sottosuolo urbano 2 (SSU2) from Rome (‘laureate Augustus / VI’, the second illegible); a further findspot is recorded in Pergamum (‘radiate Augustus / X’).Footnote ¹³⁶ A tessera reported to be of the Augustan period was found during a nineteenth century excavation in Nendorp-Wischenborg in Germany, in what has been interpreted as a grave context; the piece is now lost and we cannot know the original design.Footnote ¹³⁷ A token carrying the portrait of Tiberius and the number II appeared on the market with a possible find context of Germany; the piece was reported by Martini alongside a token carrying the radiate head of Augustus on one side and the number XIII on the other, said to have been found in the Garigliano in Italy.Footnote ¹³⁸ The majority of the tokens with Julio-Claudian portraits belong to museum and private collections and possess no find information. Martínez Chico has recently published bronze and orichalcum tokens carrying the portraits of emperors and sexual imagery in Spain; he states that the pieces in the collection of Gonzalo Cores Uría and the archive of Alberto Campana undoubtedly have an Andalucian origin; if the information is correct we have further find information (however vague) for many more specimens.Footnote ¹³⁹ More recently a token showing Drusus the Younger and the number XIIII (a previously unknown combination) came to light near the Giribaile reservoir in Vilches (Jaén, Andalusia), a region that possesses the remains of an Iberian oppidum. Martínez Chico suggested these pieces moved to Spain in connection with the movement of troops and elite to Baetica and the increasing Roman municipalisation of the area.Footnote ¹⁴⁰

What of other bronze and orichalcum tokens produced from this workshop? Two specimens of tokens released by Gaius Mitreius, both of the basilica type, have known findspots: one was found on the Saalburg in Germany and another on the island of Capri; both specimens are now lost.Footnote ¹⁴¹ The findspots of the so-called spintriae carrying sexual imagery have seen considerably more scholarly attention; this may explain why there are more recorded findspots for this series than any other. De Callataÿ, who has recently observed that most spintriae appear to have been known before 1800, noticed that many of these early finds have a provenance of Capri, but whether this was an invented findspot to enhance the value of these pieces cannot be known.Footnote ¹⁴² More securely, tokens with sexual imagery have been found in a potter’s workshop in Salles, and at Argenton-sur-Cruese, both in France.Footnote ¹⁴³ Two spintriae have been recorded as coming from the Thames in London, although it is uncertain if this was an ancient context or a more modern loss.Footnote ¹⁴⁴ Two further specimens, one pierced, were found in Croatia – one in Narona and the other at Majsan.Footnote ¹⁴⁵ Another pierced specimen was uncovered as a stray find at Caesarea Maritima.Footnote ¹⁴⁶ Crisà has recently published a piece found in a Roman villa in Patti Marina near Messina in Sicily.Footnote ¹⁴⁷ There is no pattern to the distribution of the finds (no one number or scene concentrates in a particular area); the overall picture is one of individual pieces that travelled to particular regions from Italy before being lost or deposited. The pierced specimens at Caesarea Maritima and Majsan suggest that some of the spintriae may have been cultivated for their imagery.

It is clear that the fascination with spintriae is not confined to the modern world. The same interest can be seen in the various imitations of these artefacts throughout the Roman Empire. Perhaps the most famous example is the imitative spintria found in a tomb at Mutina in Italy; the piece was covered in gold leaf and was found alongside four coins.Footnote ¹⁴⁸ A terracotta imitation is also known from Salone: one side shows a couple engaging in intercourse in a lavishly appointed room complete with drapes and kline, while on the other side the number V is depicted within a dotted border and wreath.Footnote ¹⁴⁹ Buljević posited that the imitation was directly copied from an original spintria, and that the terracotta piece was produced for the same reason as the original spintriae, whatever this might have been. This need not have been the case: the imagery might have simply appealed to the creator of the terracotta token, who may have used the resulting product in a different way. Martínez Chico’s catalogue of tesserae from Spain includes what appear to be two imitations in lead, as well as a possible token converted into a pendant; several of the tokens in museum collections are also pierced, suggesting later use in jewellery.Footnote ¹⁵⁰

A piece once in the Martinetti collection and now in the British Museum also draws inspiration from higher quality bronze and orichalcum tokens: this piece bears the head of Mercury on one side and the number V within a wreath on the other (Figure 5.13). Other pieces that may have functioned as tokens carry sexual imagery that is not so imitative in design. A lead piece measuring 1.8 cm × 1.5 cm found on the Lavanter Kirchbichl in Austria shows two lovers under a roof, with a column on either side.Footnote ¹⁵¹ A similar design, in bronze, is found on a quadrangular token now in Paris (Figure 5.14). A further piece in the BnF, shaped like a tabula ansata, carries an incuse erotic scene on one side and II on the other.Footnote ¹⁵² Sexual imagery was a popular motif within Roman visual and material culture, and the spintriae appear to have inspired the creation of further paranumismatic objects in different locations.

Figure 5.13 AE token, 12 mm, 1.24 g, 3 h. Head of Mercury right wearing petasus, caduceus over shoulder / V within wreath.

BM 1940, 0401.60.

Figure 5.14 AE token, 17×15 mm, 4.22 g. Nude female figure reclining left within domed canopy.

BnF inv. F 7917.

A specimen of the bronze token series carrying the bust of Dionysus on one side was found north of the baths during the excavation of a large Roman villa on the south-west shore of Lake Nemi.Footnote ¹⁵³ The villa was abandoned in c. AD 150 because of a natural catastrophe, providing a terminus ante quem for the type. No die connections between the Dionysus tokens (which appear to be relatively small issue, even for tokens) and the Julio-Claudian types have been found by the author. The numbers on the reverse dies are only placed within a dotted border, not a wreath, and the series may be a later issue, or from a different workshop.Footnote ¹⁵⁴

A concentration of token finds is present in Lepcis Magna in North Africa: four specimens of the orichalcum issue carrying a cantharus on one side and a modius on the other were found in four different tombs in the necropolis.Footnote ¹⁵⁵ The archaeological context of the finds gives the series a date of c. AD 50–150.Footnote ¹⁵⁶ In this context it is also worth noting that, like the Dionysus type discussed above, this series has a dotted border and no wreath. The concentration of four tokens of the same type in the same city is remarkable; Munzi suggested they might have come to circulate in Lepcis Magna on the basis of their metal value.Footnote ¹⁵⁷ Munzi also reported that seven anonymous issues of small bronzes have been found in the necropoleis of Lepcis: four specimens of RIC II (anonymous) 26 (head of Apollo / tripod with S C) were found in hypogea as well as three examples of type RIC II (anonymous) 19 (head of Mars / cuirass and S C). A fourth specimen of the Mars type was uncovered as a sporadic find in the funerary region of Uadi er-Rsaf.Footnote ¹⁵⁸ The concentration of particular designs (two coin types, one token type) amongst the small change found in tombs at Lepcis does suggest that the finds reflect currency that may have been shipped to Lepcis Magna (officially or unofficially) to act as small change – it is hard to explain the repeated recurrence of types if the finds reflected coinage that arrived in Lepcis in the pockets of merchants or other travellers.

The find contexts of these examples, however, may be significant. Munzi noted the presence of other Trajanic and Hadrianic small bronzes in the hypogea at Gasr Gelda, and observed that in general the earlier tombs seem to have contained asses and the later burials smaller denominations (e.g. semisses or quadrantes).Footnote ¹⁵⁹ As Kemmers demonstrated for the quadrantes of Nijmegen, the arrival of small denominations in North Africa may have been later than the use of the denomination in Rome; here it seems to be a phenomenon of the Trajanic and Hadrianic period. Each coin was a single find, not associated with other coins. It seems that these pieces acted as Charon’s obol; users selected the lowest value coins for this purpose, which, we might note, included the orichalcum tokens. The value of these orichalcum pieces must have been at the lower or lowest end of the monetary economy. Munzi observes that small bronzes are found elsewhere in Lepcis Magna, but in smaller numbers: of the thirty-eight coins coming from the excavations of the theatre, for example, only four were identified as quadrantes.Footnote ¹⁶⁰ An archaeological survey uncovered a further anonymous quadrans (RIC II (anonymous) 7) in the region from a mausoleum/villa site and a lead token (illegible) on a villa site.Footnote ¹⁶¹ Lepcis Magna should be added to studies exploring the role and circulation of small bronzes in the Roman Empire.

A specimen of the ‘cantharus / modius’ token series has also been found in the river Walbrook in London; although the token came from the silted riverbed, the associated finds suggest deposition sometime after the coins of Antoninus Pius arrived in the region, and before the arrival of coins of Marcus Aurelius.Footnote ¹⁶² A specimen was also found in Segobriga in Spain, during the excavations of the cryptoporticus south of the theatre.Footnote ¹⁶³ Munzi also recorded possible find locations in Germany and in Siscia.Footnote ¹⁶⁴ There are at least three obverse dies for this issue: one with the modius clearly displaying three legs (with the middle leg quite thick) (Figure 5.15), another displaying the modius with two smaller legs at either side (Figure 5.16) and a third on which the modius is flanked by the letters Θ Ε (Figure 5.17). Given that for the token found in Ephesus Θ is thought to refer to therma, one wonders whether this is how we are to understand Θ Ε here, but without a bath find context (as on the Ephesian token) any suggestion can only be speculative. Either these tokens were required in such quantity that multiple dies were needed during a single production, or else production was occasional, with different dies being engraved over time. A full die study may assist in coming to a definitive conclusion. Unfortunately, the specimens with find locations are either very worn or not illustrated, so it is not possible to assess whether the products of different dies travelled to different locations.

Figure 5.15 Orichalcum token, 19 mm, 6 h, 2.83 g. Modius with three corn-ears, dotted border / Cantharus, dotted border.

BnF inv. 17070.

Figure 5.16 Orichalcum token, 16 mm, 6 h, 2.57 g. Modius with three corn-ears, dotted border / Cantharus, dotted border.

The Hunterian Museum Glasgow, Tessera no. 25.

Figure 5.17 AE token, 19 mm, die axis not recorded, 3.41 g. Modius with three corn-ears, Θ on left, E on right / Cantharus, dotted border. Ex BCD collection.

It is evident that bronze and brass tokens travelled throughout the Empire. Their small numbers mean that they cannot have seriously impacted the supply of small change, but the monetiform appearance of these pieces meant that they might have occasionally been used in this context. It is worth noting, however, that these tokens are not found hoarded alongside other coins – their unusual appearance, and low value, might have prevented them from being used as stores of wealth. Instead, these pieces seem to have been selected for curation, being employed as pendants, Charon’s obol, and as inspirations for new creations. These pieces were intentionally designed to sit apart from official currency, through the choice of design, and the absence of the S C that appears on the anonymous quadrantes and other bronze from the Roman mint. It seems that most users recognised their differences from official currency and treated them accordingly.

The finds of the bronze and orichalcum tokens described above are rather dispersed, and largely appear to be the result of individual activity rather than evidence of a particular trade network. Other tokens, however, do seem to reflect on-going connections between particular places. Stannard’s exploration of the similarity in imagery between lead pieces in Baetica and Italy in the second and first centuries BC, as well as finds of Spanish lead pieces in Italy and lead pieces from Minturnae in Spain, is material evidence of contact between the two regions; Stannard has suggested this is a trade network.Footnote ¹⁶⁵ Stannard has also argued that the presence of uniface quadrangular bronze tokens in both Ostia and Minturnae during the imperial period is evidence of movement, perhaps of goods, between the two ports.Footnote ¹⁶⁶ For lead tokens of Roman imperial Italy, however, the evidence seems consistent with the movement of individuals rather than a sustained connection between two places.

Several examples of lead tokens travelling within Roman Italy, particularly within the region of Latium (Nemi, Minturnae and the Garigliano, Alba Fucens), have already been discussed in this volume. A further example can be found with the type TURS 919 (Figure 5.18), decorated with an erect phallus on one side and a pair of scales on the other. The combination of imagery here recalls the famous fresco at the House of the Vetii in Pompeii, in which Priapus is depicted weighting his phallus on a pair of scales; one imagines the token imagery evoked the same sense of abundance and wishes for wealth. The discovery of two specimens of this type in the Baths of the Cisiarii in Ostia dates the series to roughly AD 150–250.Footnote ¹⁶⁷ Pensabene records a further specimen of this type from Ostia; the archaeological museum at Palestrina has twenty-four specimens of this series from the seized collection acquired from illegal excavation activity.Footnote ¹⁶⁸ Another example was found in the river Liri at Minturnae during the excavation work undertaken by Ruegg.Footnote ¹⁶⁹ The presence of only a single specimen at Minturnae, in comparison to the higher number of examples associated with Ostia, suggests that it was the token, and not the token mould, that moved southwards. Of the cast lead tokens coming from the river at Minturnae, a handful seem to have come from Rome or Ostia; many more are not paralleled elsewhere and may have been local designs.Footnote ¹⁷⁰ Ruegg suggested the lead pieces might have been given as votive offerings in the river; like Charon’s obol, the very low value of these pieces may have been what made them attractive as votive offerings.

Figure 5.18 Pb token, 14 mm, 3 h, 3.34 g. Phallus / Pair of scales.

TURS 919.

The scattered finds of tokens from Rome and Ostia reported in Hadrumetum, and occasional Egyptian tokens in Italy, have already been discussed elsewhere in this volume. These also seem indicative of the movement of particular individuals rather than an intentional shipment. Several additional finds of Roman tokens are recorded in southern France. CIL XII, 5699.12 reports twenty-one lead tokens found in southern Gaul. Several of the types reported also have find locations in Rome. For example, one of the specimens from southern Gaul is of the type TURS 2817 (rooster / C C on either side of a palm branch). Two further examples of this type are known to have Italian contexts – they were once owned by Rostovtzeff, who said he acquired them in Rome.Footnote ¹⁷¹ The specimen found in southern Gaul thus likely travelled from Rome or Ostia. Similarly, two specimens were recorded to be of type TURS 692 (bull walking right / PM); the numerous specimens of this type in Italian collections, including that of Ficoroni, suggests this was an issue manufactured in Italy.Footnote ¹⁷² A similar case might be made for CIL XII, 5699.12m, of the type TURS 310 (eagle and B / LP|O; the meaning of the legend is unknown). Other tokens in the CIL entry may have slight misreadings (e.g. in the aforementioned token the O is reported as Q), which obscure the fact they are known types from Italy. Unfortunately, only two of the tokens found in southern Gaul are now accessible, but both of these have fabric consistent with the tokens of Rome and Ostia.Footnote ¹⁷³ Indeed, the impression given by CIL XII, 5699 is that of an assemblage of lead tokens from Rome and Ostia, but without being able to physically examine all the tokens we cannot be certain in this conclusion.

The CIL also reports two further lead tokens found at Perpignan, a Roman settlement site on the southern coast of Gaul.Footnote ¹⁷⁴ The types reported are also the same as known tokens found in Rome and Ostia, and the presence of these types in major museum and antiquarian collections suggest they were used in the imperial capital and/or its harbour. The tokens thus likely travelled from Italy to southern Gaul. The first token displayed Fortuna on one side and the legend TI|CE on the other (a token of the same type is shown as Figure 5.19), while the second carried the representation of an ithyphallic Priapus or Silvanus and the legend C PE|DANI (C. Pedani) (a token of the same type is shown as Figure 5.20). Rostovtzeff wondered whether TI|CE was a reference to Tyche, although he also entertained the possibility the letters might be an abbreviated name: Ti. C(laudius) E(utychus). Another token issue showing Fortuna on one side with the legend TIC|EV on the other was resolved by Rostovtzeff as a reference to someone of this name (TURS 1171); this token is much smaller than Figure 5.19 (11 mm) and so not directly comparable, but it might be they are tokens of different sizes issued by the same person.Footnote ¹⁷⁵

Figure 5.19 Pb token, 18 mm, 12 h, 3.09 g. TI|CE / Fortuna standing left holding cornucopia in left hand and rudder in right.

TURS 1502, Reference Rostovtzeff and ProuRostovtzeff and Prou, 1900: no. 430a.

Figure 5.20 Pb token, 15 mm, 12 h, 2.87 g. C PE|DANI / Ithyphallic Priapus or Silvanus standing right holding sickle.

TURS 1299, Reference Rostovtzeff and ProuRostovtzeff and Prou, 1900: no. 426a.

The location of these tokens in Gaul, close to the coast, as well as the finds of Roman or Ostian lead tokens in other ports (Minturnae, Hadrumetum) suggests that the pieces accompanied merchants and sailors on their voyages. Their loss abroad may have been accidental or votive in nature. The small number of finds, and the location of some specimens in rivers, suggest these pieces may have been selected to fulfil particular vows, perhaps because of their low value and monetiform nature. The presence of struck lead on the Isla Pedrosa shipwreck, as well as the Egyptian lead tokens found on an imperial period shipwreck off the Carmel Coast, are clear evidence that material of this nature was carried on naval vessels.Footnote ¹⁷⁶ The presence of imperial period lead tokens from Rome and Ostia in other regions (which largely seem to be ports) provides further evidence to suggest that these pieces were given out in advance of a specific event. That is, rather than being both distributed and used on a particular occasion or within a particular building, tokens appear to have been given to individuals to be redeemed at a later moment in time. This would explain why lead tokens managed to travel, and why some tokens appear never to have been redeemed.

The survey of the evidence, meagre as it is, does not support the idea that tokens acted as a form of supplementary or emergency currency. By mediating exchanges that might otherwise have required small change, tokens may have reduced some of the burden on Rome’s official currency, but this was surely not their intended function. As a recent exploration of euergetic acts within collegia has suggested, activities of this nature may have served to lower transactions costs for members, but whether euergetic acts were performed with economic aims in mind is less clear.Footnote ¹⁷⁷

But can a study of tokens bring anything to our understanding of Roman currency, particularly small change? The stark difference in imagery between Roman quadrantes and other denominations of the Roman mint has led to the suggestion that these pieces were distributed during festivals or on special occasions. Buttrey believed that the Domitianic quadrantes showing a two-horned rhinoceros, for example, might have been showered on crowds during spectacles.Footnote ¹⁷⁸ The anonymous quadrantes in particular (those without an imperial portrait or legend), have been interpreted as artefacts connected to distributions, religious festivals, or public games.Footnote ¹⁷⁹ Indeed, small bronzes carrying the imagery of prize tables are thought to have entered circulation as distributions during the events referred to on the coin types.Footnote ¹⁸⁰ If used for distributions during particular events, one might see how a quadrans or a semis gifted by the emperor could be seen as an imperial benefaction, buying the recipient a free bath, for example, while also contributing to the supply of small change (the latter could be seen as an act of euergetism, particularly in Greek cities in the Roman Empire).Footnote ¹⁸¹ Buttrey’s objection to this idea, that there do not appear to be enough issues of this type to have been regularly used during Roman festivals every year, does not preclude the idea that these coins may have featured occasionally at events, in the same way tokens of differing materials might be used occasionally.Footnote ¹⁸² Roman religion, and Roman festivals, remained a vibrant ever-changing affair, and we should not assume that an event took place in an identical fashion every time.

Greek cities in all likelihood struck coinage to facilitate transactions during particular festivals; the bronze coins carrying the legend ΕΛΕΥΣΙ produced from the fourth century BC have been interpreted as a festival coinage connected to the Eleusinia, and in the imperial period cities must have produced coinage during festivals even if the imagery of these coins only began to communicate these occasions from the reign of Commodus.Footnote ¹⁸³ The Roman mint may also have struck small change for use during festivals – the semis series struck from c. AD 62 under Nero that shows a prize table, accompanied by the legend CERTAMEN QVINQ(ennale) ROM(ae) CO(nstitutum), is plausibly connected to Nero’s introduction of the quinquennial games in Rome. These Neronia were a Greek style celebration involving musical, gymnastic and equestrian competitions.Footnote ¹⁸⁴ It is not often commented upon, but the appearance of the prize table on these Roman issues is extraordinary; Greek cities at this stage only rarely included such imagery on their coinage.Footnote ¹⁸⁵

There are parallels between quadrantes, semisses and tokens in terms of their imagery, low value and possible connection with festivals. But there are also clear differences, which throw light on our understanding of both types of artefact. Despite earlier scholarship seeing quadrantes as an Italian phenomenon, it is clear this denomination did travel, and in greater volume than tokens. Moreover, quadrantes and semisses are found stored alongside other coins, a phenomenon that is extremely rare for ancient tokens. Thus quadrantes and semisses were clearly intended to function as small change, however they entered circulation. Tokens, by contrast, were unequivocally not intended to function in this way. This is underlined by a key difference between tokens and official small change – the legend S C. While this legend appeared on Roman official coinage, tokens, by contrast, overwhelmingly did not carry these letters.Footnote ¹⁸⁶ Tokens, in the main, appear to be connected with acts of euergetism, and anonymous quadrantes and semisses functioned as small change. We cannot rule out the idea that some tokens, particularly those made of bronze, came to be used as small change in a secondary capacity. Likewise, we cannot rule out the idea that some coin issues of small change were struck and distributed at particular festivals. But we cannot classify tokens as a substitute small change.