Coin Relief 42 – Bronze coinage of the Roman Republic

The earliest Roman coin types seen through the PAS, of the Roman Republic, are relatively abundant, at least in terms of the precious metal coinage, with over 2,000 denarii recorded to date. One of the problems with Republican silver coinage is that it all pre-dates the Claudian invasion in AD 43 so it remains unclear precisely when any or all of it made its way to Britain. That some did cross the channel prior to AD 43 is suggested both on archaeological grounds and through the re-use of silver from denarii in British Iron Age coin manufacture. Much, however, likely remained in circulation for an extended period of time post-AD 43 to be lost within the new province of Britannia. This makes sense, since the silver was intrinsically valuable and would still have a function in the payment of, for example, soldiers’ wages.

Base metal, bronze, coinage is a slightly different matter. Lacking the same intrinsic value of the silver, the appearance of bronze Republican coinage in Britain might not be expected in any volume, if at all. After all, Britain was already producing its own coinage by the 2nd century BC in gold, silver, and also bronze. Furthermore, after the Claudian invasion there is quite extensive production from the main imperial and auxiliary mints within the empire that brought contemporary bronze coinage to the new province. It would seem unlikely that the earlier, Republican, bronze types would either be needed or used prior to AD 43 in Britain, or still be circulating alongside the new imperial issues afterwards.It is interesting, therefore, that there are now 11 examples of Republican bronze coins 
recorded through the PAS, with an additional 7 unverified examples contained within the IARCW dataset. The first securely identified example on the database was an as from Essex (ESS-05C304) recorded in 2006, since when various examples of asses and other fractional denominations have been recorded and include the earliest Roman coin recorded to date within the PAS dataset (KENT-618FA2).

Arguably, some of these coins may be more recent or antiquarian losses, for example the sextans from Berkshire (BERK-7EF5E1) whose findspot, circumstances of discovery, and appearance suggest it was probably not lost in antiquity. However, it is only through recording all of these examples that we can begin to look at their distributions and relationships to one another (and indeed distributions of other early bronze coinage within Britain; see below) to identify those that are likely to be genuine ancient losses and comment on their movement within the province. That said, we still currently have a very limited dataset to work with!

Early Roman coinage

Aes signatum (cast bar or ingot), Republic, c.2580-250BC. BM: 1867,0212.2 (The British Museum, License: CC-BY-NC-SA 4.0).

Coinage appears in the Mediterranean by c.600 BC, with perhaps the earliest examples being the group of electrum coins excavated beneath the Temple of Artemis Ephesos (the Artemision), Turkey, in 1904-5. The Greek city-states of Magna Grecia (southern Italy) in the4th century BC had quite extensive coinage and by c.300 BC the first elements of Roman coinage appear, perhaps due to proximity and interaction between the Italian populations and their ‘Greek’ neighbours. Initially, large cast lumps of bronze, aes rude (‘unformed bronze’), served as currency, but were replaced from c.300 BC with rare struck silver (derived from Greek prototypes) and bronze coinages that carry the legend ROMANO (‘of the Romans’) alongside large cast bars or ingots – aes signatum (‘stamped bronze’) – that weighed c.1.5kg or more.

Aes grave (“heavy bronze”, Republic, c.225-217BC. BM: R.12254 (The British Museum, License: CC BY-NC-SA 4.0).

In addition, a cast bronze coinage was produced – the aes grave (‘heavy bronze’) – circular in shape but with detail in high relief. This used as its base unit the Roman as, equivalent to 1 Roman pound of bronze or c.324g, with various fractions working on an uncial standard – 12 ounces to a Roman pound. There were several aes grave series with various weight revisions, but, importantly for the later bronze coinage, each denomination
carried a mark of value based on its weight in uncia relative to the as. The as was identified with an I, down to a single uncia marked with a circle/dot, as follows:

The denarius coinage

The early systems of Roman currency were quite disparate to begin with. However, by the mid-3rd century BC the aes signatum disappears and there is a subsequent reorganisation and integration of the struck silver, struck bronze, and aes grave into a more unified currency system in the period leading up to the Second Punic War (218-201 BC) against Hannibal. There is a standardisation of types depicted on the coins too, notably with the use of the janiform head on the obverse – characteristically, the silver didrachm depicts a quadriga on the reverse, which often leads to it being called a quadrigatus. 

Quadrigatus, Republic, c.225-212BC, with Janiform head of Dioscouri. BM: 1931,0504.17 (The British Museum, License CC BY-NC-SA 4.0). 

These early coinages prior to the Second Punic War are unlikely to be found in Britain (although if you do find one we definitely want to know about it!), but they do provide important background to the emergence of the denarius, which formed the backbone of Roman coinage for several centuries after its introduction. War with Hannibal placed huge economic strain on Rome that brought her to her knees. This was evident in the huge debasement of the quadrigatus from a very pure silver coin to one with less than 40% silver and a corresponding reduction in weight standard of the bronze coinages. Out of monetary collapse came reforms that replaced the quadrigatus with the denarius. There has been some discussion as to when this occurred, but the consensus, based on the discovery of the earliest issue of denarii in sealed contexts beneath the destruction layers of Morgantina (Sicily), suggests a date in c.211 BC.

The standard denarius type that emerged with this reform has a helmeted head of Roma on the obverse with a reverse type depicting the Dioscouri, the heavenly twins Castor and Pollux – there are a few examples on the PAS of this early type but they are less common than later issues. It was tariffed at 10 asses and the denominational marks seen on the bronze coinage prior to 211 BC were carried over onto the denarius and other new silver and bronze denominations that appear with the monetary reforms. Thus, the denarius is marked with an X (10 asses – literally ‘a tenner’), while the quinarius carries a V (5 asses or half a denarius), and the sestertius – a silver coin during the Republican period(!) – is marked IIS (2½ asses or a quarter of a denarius). We will look at the silver Republican coinage, in a
future blog post(s), since the types are extensive and do change, the denarius also undergoes various adjustments including a revaluation to 16 asses with the corresponding mark of XVI in c.141 BC.

Denarius, Republic, c.211-206BC?. Record ID NCL-012704 (All rights reserved, license: CC-BY-SA).

The as experienced various weight reductions prior to 211 BC and with the introduction of the denarius coinage the bronze denominations now employed a sextantal weight standard, the as based on a theoretical ⅙ of the Roman pound (libra) – the weight of the sextans of the previous system (initially c.54.5g). These were now struck rather than cast coins and typically have standard types that were issued until c.80 BC. The as itself
continued the types of the later aes grave issues, depicting Janus on the obverse and a ship’s prow on the reverse. For the fractions, the prow is usually standard as a reverse type with the obverses bearing busts of different deities or personifications and their corresponding marks
of value, as follows:

The combination of head and ship’s prow is distinctive of Republican bronze coinage and shouldn’t be mistaken for issues later in the Imperial series. It also lends itself to the phrase ‘heads or ships’ (‘capita aut navia’) used in the Roman period rather than ‘heads or tails’ when tossing a coin (see Macrobius Saturnalia 1.7.21). Bronze coinage was struck until c.80 BC, although during this time underwent several key changes. The production of large numbers of asses resulted in revaluation against the denarius in c.141 BC (the denarius now 16 rather than 10 asses) and a corresponding change in relation to the sestertius, which was now valued at 4 asses – a relative value maintained in the imperial period. In c.91 BC the Lex
Papiria de assis pondere (the ‘Papirian Law concerning the weight of the as’; see Pliny, NH 33.46) adjusted the bronze coinage to use a semuncial (half ounce) standard with some coins even marked L.P.D.A.P to reflect the introduction of the new law.

Republican bronze coinage on the PAS

Given the unlikely use of Republican bronze coinage in Britain, it is not surprising that there are very few PAS examples recorded to date. The 11 identified coins cover most of the standard denominations, with the as best represented in 6 identified examples and a further coins within the IARCW dataset:

I have not included here the small number of early bronze coins from provincial mints in the Mediterranean region that have also been recorded through the PAS (e.g. Punic coinages and those from mints in Spain). These need further work and warrant a separate blog post themselves. In this roundup, I have included only those PAS coins struck in the main Roman
mints and covered by the types listed in M.H. Crawford’s Roman Republican Coinage (RRC), which should be your standard reference when identifying any coins of the Republican period. These essentially run to the c.80s BC, although there is small group struck by Pompey the Great in the 40s BC, of which there appear to be no PAS examples.

A semuncia

Semuncia, Republic, c.217-215BC. Record ID KENT-618FA2 (Kent County Council, License: CC-BY).

Oddly, the earliest Roman bronze coin, and seemingly the earliest Roman coin on the database, is a semuncia (half an uncia or ⅟24 of an as) from the period immediately prior to the monetary reforms in 211 BC. This is an extremely unusual find as we would not normally expect to see any coinage of such early date in a British setting – indeed, its manufacture pre-dates the majority of the British Iron Age coinages too. Although we cannot discount the possibility of a more recent loss, its findspot and preservation suggest this is plausibly a genuine ancient find. If we consider that similarly early bronze Punic coins are found in Britain in increasing volume, and with a concentration in the south east, notably in Kent, it is plausible that this example could have reached Britain by the same mechanisms that are affecting the movement of the Punic coins (see below).


There are six examples on the PAS database, plus 3 IARCW coins.

As, Republic of the moneyer Q Marcius Libo, c.148 BC. You can just about make out the laureate head of Janus. Record ID LON-3D39D8 (Portable Antiquities Scheme, License: CC-BY).

Triens, Quadrans and Sextans

There is just one example of each of these on the PAS database.

Triens, Republic, c.2nd century BC. Helmeted head of Minerva (obverse) and ship’s prow (reverse) are just about visible. Record ID SUR-E28078 (Surrey County Council, License: CC-BY).
Quadrans, Republic, c.2nd century BC, with head of Hercules design. Record ID BH-91A0C3 (License: CC-BY-SA).
Sextans, Republic c.2nd century BC, with head of Mercury design. The findspot and preservation of this coin suggest it is perhaps not an ancient loss but possibly a modern or antiquarian loss. Record ID BERK-7EF5E1 (Oxfordshire County Council, License: CC-BY-SA).

Distribution of Republican Bronze coins on the PAS

The number of Republican bronze coins recorded to date through the PAS doesn’t allow for huge comment on their distribution or use within the province, or for that matter clarify the nature of their arrival here (whether in antiquity or more recently). However, the importance of recording these coins is evident when we plot their distribution, especially when considered in comparison with other early bronze coin types that are traditionally often seen as being modern rather than ancient losses. This serves two useful functions. On the one hand, we can identify the areas where these coins appear more precisely and therefore suggest whether their distribution (or individual findspot) suggests potential ancient loss –
clearly the more coins we have recorded the better in this regard. On the other hand, it also allows us to identify outliers whose findspots instead probably indicate different mechanisms for their appearance in Britain, potentially in more recent times. An excellent example of how the PAS data can be applied in this way was produced in 2020 by a former colleague at the British Museum, Robert Bracey, in a video looking at Dutch and Indian coins recorded on the database and how these can be used to identify whether examples of ancient coins from the area of India, Pakistan, and Afghanistan can be interpreted as ancient or modern losses. This references also the work of Caitlin Green who has looked at similar patterns for Indian, Ptolemaic, and Numidian coinage recorded through the PAS.

For the Republican coins, it is notable that their distribution falls within the main areas we might expect to see other early bronze types, albeit most of these other issues are slightly earlier in date. Given that much of this material is of Mediterranean origin, it is perhaps not outside the realms of possibility that early Roman bronze types are reaching Britain in much the same way as these other Mediterranean types are, and potentially prior to AD 43. While there are likely to be some examples within the data above that are more recent losses (such as the triens), what the PAS data is showing, I think, is possible evidence for the early movement of these coins to the region in antiquity. There seems to be a clear focus on the south east and Kent. This would make a lot of sense given that the Massaliot bronze coinages provided a prototype for the earliest Kentish potin types, so we might see movement of coinage here demonstrated by the large group(s) of material, particularly the Massaliot bronzes, at quite an early date. The cluster in Oxfordshire is curious – could this be the result of riverine activity from Kent and London west? Finally, the small group of material focussedon the Humber is interesting. Again, this has a coastal or riverine focus and on the east coast.  It wouldn’t be outside the realms of possibility that if material was reaching south eastern Britain at an early date it didn’t also find its way further north by similar mechanisms. This is a fascinating glimpse at the early bronze material and perhaps raises more questions than answers. Clearly, none of the above is definitive and much of it is (very!) conjectural and requires much more work in conjunction with data from other sources too. But, we can perhaps begin to see the patterns in the data that mean we shouldn’t immediately write off as ‘antiquarian’ the Republican bronze coins presented for recording (although this possibility both for the Republican and other early bronze issues cannot of course be ruled out entirely). Equally, this highlights the importance of recording those that do appear so that we can begin to understand their distribution within Britain. If some, or many, of the early bronze coins from various parts of the Mediterranean did reach Britain in antiquity, and perhaps relatively soon after they were struck, then why not Roman Republican bronze coins too…

References and further reading:

P. Walton and S. Moorhead ‘Coinage and the Economy’ in M. Millett, L. Revell, and A. Moore eds. The Oxford Handbook of Roman Britain (2016): 834-849

M.H. Crawford Roman Republican Coinage (Cambridge, 1974)

A.M. Burnett Coinage in the Roman World (London, 1987);

A.M. Burnett ‘Early Roman Coinage in its Italian Context’ in The Oxford Handbook of Greek and Roman Coinage (Oxford, 2012: Chapter 16)

E. Ghey and I. Leins Roman Republican Coins in the British Museum

Coinage of the Roman Republic Online (CRRO)  – which provides a digital version of Crawford’s RRC