Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

date: 24 May 2019


Abstract and Keywords

This article elucidates the fundamentals of numismatic study. It defines a coin as “a piece of metal” certified by a mark or marks upon it to be of a definite exchange value, and issued by governmental authority to be used as money. These metals have to be scarce enough to have intrinsic value but plentiful enough to provide raw material. Virtually all other ancient coins are struck from gold, silver, or copper, sometimes alloyed with tin or zinc. The “mark or marks” have preoccupied numismatists since the beginning; this is probably owing to the interest in antiquity, roused in the Renaissance, that saw in coins miniature ancient monuments comparable to sculpture and other art forms. For the Greek world, discussion of economic history is surprisingly free of references to coinage. However, the numismatic evidence has not yet been exploited to the degree necessary for its proper appreciation.

Keywords: numismatic study, coin, money, metal, Greek world, coinage

Some aspects of the history of numismatics have been alluded to in the preface, and others are better read elsewhere (Babelon 1904). This introduction is intended to explain, mainly for the benefit of the uninitiated, some fundamentals of numismatic study, many of which are employed or cited again and again in the texts that follow.

We might begin with the definition of a coin, which the second edition of Webster's Dictionary describes as “A piece of metal (or, rarely, of some other material) certified by a mark or marks upon it to be of a definite exchange value, and issued by governmental authority to be used as money; also, such pieces collectively.”

All the objects discussed here are of metal, and all bear marks of some kind; these are anticipated historically by ingots, which are pieces of metal of irregular weight and shape intended to represent a store of wealth. These metals had to be scarce enough to have intrinsic value but plentiful enough to provide raw material, and in fact availability may have dictated such choices as electrum for the early coinage of Asia and copper for the first coinage of Rome. Virtually all other ancient coins are struck from gold, silver, or copper, sometimes alloyed with tin or zinc. The latter alloy, with its yellowish appearance, was referred to by the Romans as orichalcum.

The Webster's definition does not take account of weight, which was critical in ancient coinage—the value of a coin was intrinsic, so its tariff bore a direct relationship to its weight and fineness. The varying mints employed their own standards: just as the Athenian Currency Decree is about regulation of weights and measures as well as coinage, coin weights bore a direct relationship to the prevailing standard weight. They were also called by names deriving from weight systems: a drachm was normally 1/6000 talent, a tetradrachm 1/1500, and so on; and the Romans expressed their coin standards as so many units per pound (the Roman pound consisted of twelve ounces slightly lighter than the modern ounce).

(p. 4) The “mark or marks” might be almost anything. Early coins from the Artemision (Robinson 1956, and see chapter 3 here) bear nothing more than striations on one face and punches on the other. The striations are sometimes taken to represent imperfections in the flat surface on which a globule of molten metal was initially poured, and the punches—though later they come to contain figures—are nothing more than a device to ensure that the piece was of pure metal down to its core.

The “mark or marks” have preoccupied numismatists since the beginning; this probably owes to the interest in antiquity, roused in the Renaissance, that saw in coins miniature ancient monuments comparable to sculpture and other art forms. In any catalogue of coins produced prior to the advent of the Sylloge Nummorum Graecorum in the 1930s, and in many catalogues of Roman coins even today, most of the ink is used on descriptions of the two faces of the coin; some even omit the physical dimensions! The obverse (conventionally the “heads” side of the coin) is usually characterized by a bust or portrait; the reverse by an image—a god/goddess, hero, animal or plant, building, and so on. The principal device or image is known to numismatists as the type. Either side may have a legend (from Lat. legendum, something to be read) that describes or amplifies the image. The part of the legend that names the issuing authority is known as the ethnic.

These types and legends may take varying forms and bear different relations to one another. The earliest coins had no legends at all; doubtless the type was originally a sufficient badge to identify the issuing authority, and indeed it normally occupied almost all the available space in the field. With the advent of varied types, but above all of coins with types on both faces, it may have become both necessary and possible to place a word or words on the coins. Initially, these were simple designations of the issuing authority, either abbreviated (ΑΘE) or spelled our (ΣΥΡΑΚΟΣΙΩΝ). Most often, when the legend describes the authority behind the issue, it is in the genitive, and this custom survives the transition to Hellenistic coinage, on which the name of the king (ΑΛΕΞΑΝΔΡΟΥ, ΦΙΛΕΤΑΙΡΟΥ) is rendered in the genitive. Later additions to the content of legends include identifications of the deities or heroes portrayed and of the magistrates responsible for the striking.

The first Roman coins, the aes grave, were without legends, but it is no doubt owing to the origin of its silver coinage in Greek Italy that the first legends were simple mint identifications followed by magistrates’ names. These are, of course, in Latin. On Republican silver, other inscriptions occur as part of buildings or identifications of types. With the advent of empire, the imperial name, in simple (IMP CAESAR) or expanded (IMP NERO CLAVD CAES AVG GERM P M TR P P P) form, accompanies the portrait, often with iterations of imperial salutations, the tribunician power, or the consulship. These iterations are almost always absent from provincial coins of the period. The portrait/legend combination has been extremely useful to art historians attempting to identify or date portraits. Reverse legends in the imperial period often serve to identify the deity or personification portrayed, or to put it in context.

Apart from the visual and verbal content of a coin, what struck early students was a difference of style. This term, familiar mainly from the history of art, is at both (p. 5) its best and its worst when applied to numismatics. Granting that a consensus can often be reached on the style of an individual artist, how do we distinguish his work from the style of a period, or of a place, or indeed of a mint itself? How do we distinguish the archaic from the archaistic? The elusiveness of the criterion led Mattingly to the following formulation: “If two or three scholars readily agree on what they see, they obviously are seeing something that is really there, even if other scholars fail to see it” (1960: 42).

The vagueness—or rather, the difficulty of expressing a visual impression in words—was obvious. Even so, this was the principal mode of classifying coins, for chronological purposes, almost from the beginning. It strains credence today that Barclay Head (HN2 lxi) could assert that “in every period there are coins of which the dates can be positively determined; and around these fixed points a little experience enables the numismatist to group, within certain limits, all the rest.” He went on to name no less than seven periods (“Period of archaic art,” “Period of transitional art,” “Period of finest art,” “Period of later fine art,” etc.). None of these remains current, though well-defined criteria of style—most often, in Greek coinage, at particular mints, and in the provincial coinage of the Roman Empire occasionally applied to geographical areas—can be applied successfully.

Numismatics has come a long way, particularly with respect to portraiture. It is encouraging to see Hellenistic portraits on coins placed next to those in the round (e.g. Smith 1989) and even more to see coins, sculpture, and the historical record integrated (see especially Fittschen 1982), as well as the successive volumes of Das Römische Herrscherbild).

Still, it is fortunate that there are more objective techniques for determining the when and where of coin production. The terms obverse and reverse, defined above with respect to types, also apply to the means of manufacture, which in most cases involves striking by die (the alternative is casting in a mold). The obverse die is normally fixed in what numismatists call an anvil. This describes a heavy piece of wood or metal with a hollow that contained the obverse die and both fixed its position and protected it from expansion under pressure. The reverse die was held in the hand. A blank—probably heated, for images of coining implements usually include tongs, and heating would have rendered the metal more malleable and would have work-hardened the dies—was placed on top of the obverse die, and the reverse brought down on it under a heavy hammer. (See fig. 0.1)

This process has important consequences for the numismatist. Dies were hand-engraved; this means that each is unique. Each also had a finite lifetime, which has been variously estimated. When a die broke, a new one had to be substituted, but it would be the merest accident if both dies broke at once. Sometimes it is possible to be even more precise, since dies deteriorated before they broke, and visible cracks or other flaws developed. The more advanced the damage, the later in the sequence it must belong.


Fig. 0.1

In theory, with a large enough sample, it would be possible to document the entire history of a coinage based on the history of its dies. In fact some samples—for example the silver coinage of the Roman Empire—are so large that study by die is a (p. 6) practical impossibility. At the very least, the sharing of a die among two or more coins is an indication of proximity in time and space. We shall return to one other use of die linkage below.

One final aspect of die pairing is the die axis. This refers to the orientation of the dies vis-à-vis one another, and in modern times it is expressed in hours on the face of a clock. A 12:00 die axis indicates that when the coin is turned around its figurative equator, both images are “up”; if one is upside-down (as is the case in virtually all modern coinages, including that of the United States and Britain), that is a 6:00 die axis. Sometimes a particular die axis becomes entrenched at a mint; sometimes dies are paired as they come to hand, and the distribution of die axes is correspondingly random. A classic article by Sir George MacDonald (1906) explores the significance of the die axis in Greek coinages; modern applications can be found in Mildenberg (1989) and De Callataÿ (1996b). The axis is a standard part of catalogue descriptions and is ignored at one's peril.

We have so far examined the coin as the product of a manufacturing process; its behavior in circulation also provides information about its time and place of production and about the purpose of its issue and its use as money. Here we examine coins in context, whether carefully secreted or simply lost or discarded in antiquity.

A hoard is an aggregation of two or more coins put away by their owner with the intent of recovery. Through most of history there has been no accommodation for the thrifty saver. Carrying one's wealth around on one's person would have been imprudent, given the high levels of crime in antiquity; yet until recent times all but the wealthy have been excluded from the credit system and depository banks. In a world that lacked savings banks in the modern sense, burial was an easy and available option.

(p. 7) It is conventional to see hoarding as an instinct reinforced by a sense of impending danger; often enough the perception was justified by the event, with the consequence that there is a slight tendency for unrecovered deposits to cluster in times of political, social, or economic unrest. Historically, agriculture—the clearing of land—has accounted for the bulk of recoveries; in more modern times, urban excavation (for building foundations, utility lines, etc.) has had a role. And the twentieth century saw the advent of the metal detector.

The utility of hoards as a dating tool is straightforward: obviously, a hoard cannot have been deposited before the date of its latest coin. This normally provides only a terminus post quem, since most hoards are private stores whose manner of accumulation is strictly a matter for speculation. Conventionally a distinction is made between “savings hoards”—those accumulated over time—and “currency hoards,” which represent the conversion of assets to cash at a given moment. The savings hoard does not represent the currency of any given moment, since coins will have been added (and indeed subtracted—see Thompson 1962: 309) over time. Nor can the currency hoard give us an accurate picture of circulation over time. In practice, it is not always so easy to distinguish these classes, and some hoards (e.g., those caused by sudden destructions, as at Pompeii or Dura-Europus) may not be true hoards at all.

Sometimes a single hoard (Asyut in Egypt; Dorchester in England [Mattingly 1939]; Lohe in Sweden [Thordemann 1948]) is of surpassing importance for its own sake. But the real utility of hoards rests in the assembly and comparison of those that contain common bodies of material. When a large number of hoards from a fairly constricted period is available, they are invaluable for establishing chronologies. A sequence of hoards will reveal a relative order of appearance of issues. As Crawford has put it, “of two hoards with some issues in common, that which is later will contain issues which do not occur in the other hoard and which are less worn [than other coins in the hoard]. A relative order of issues follows automatically.” The principle can be extended as far as the hoards permit, and the resulting chronology is sturdy and reliable. This technique has permitted the arrangement of virtually the whole of the coinage of the Roman Republic (Crawford 1969, 1974).

Where it is abundant, the evidence of hoards is second only to that of die links in importance, with the caveat that is it susceptible to contamination in modern times. Hoards are infrequently found in controlled archaeological contexts, and there is almost always some doubt about the integrity of lots even when they are intrinsically plausible. A dealer will always have the inclination to skim the rarest coins, or those in freshest condition, because of the commercial premium these command. And it is no help that institutions are no longer competitive for acquisition of whole lots, on grounds of finance as well as specious ethical arguments.

Site finds are another matter. Unlike hoards, which can be placed in time with greater or lesser certainty, coins recovered in excavations (unless themselves in hoards) have no fixed chronology of loss. The very fact that these are lost coins, rather than hoarded ones, defines the nature of the material: mostly base metal and generally lower denominations. One category—discarded coins—may be represented by the unusually high number of false precious-metal coins in (p. 8) excavations, since it was illegal at most times to possess, much less to attempt to pass, such coins.

What is lost in terms of chronological control is regained in spatial control. From the numismatic point of view, the exact locus of find is seldom critical, and indeed is seldom reported in excavation catalogues. But the knowledge that a coin was excavated at, say, Antioch in Syria makes it a priori likely that the coin was produced there or in the neighborhood, since excavation finds themselves confirm that lower denominations did not stray far from their point of origin.

Both site finds and hoards have been put to new uses in recent times. In a famous and much-cited article, Keith Hopkins (1980) combined the evidence of hoards and site finds to create a model of monetary movement in the Roman Empire. This subject has caused considerable controversy, as scholars have confronted a universal coinage (the denarius) that may have functioned on a regionalized level (Howgego 1992; Duncan-Jones 1994). That issue is far from resolved, as is another one that has given a whole new tilt to numismatic method: the estimate of sizes of coinage.

Since the work of C. S. S. Lyon (1985) statisticians have used the evidence of dies surviving in a given sample to estimate the total number of dies used to produce a coinage. Other more sophisticated formulae have been developed since, but the goal is simply to estimate numbers of dies. Insofar as our sample is (charitably) treated as random, this is a simple application of established statistical method.

But it has been carried further, in two directions. The first is to use known die populations, as represented by their issues in hoards, to project the number of dies that might have been used in a sample in which the dies have not been compared. The second is to estimate the number of coins a die might produce, based on available resources: patchy records from antiquity, more extensive ones from the Middle Ages, but also results obtained empirically in an attempt to replicate ancient minting procedure. Famously, Crawford (1974: 641–707) used these combined methods to gauge the size of virtually the whole of the Roman Republican coinage, thereby spawning a wave of imitators. The initial skepticism culminated in the forceful arguments of Buttrey (1993, 1994), who stated flatly that the only thing we can be sure about in such calculations is that the result will be wrong. We cross the line from primary evidence into speculation in a way that is difficult for historians to discern; in short, we fabricate history. A conciliatory response from De Callataÿ (1995) seems to govern recent approaches.

Applied Numismatics

This book is intended to provide numismatic background; it would take a book of comparable length to scratch the surface of the ways the evidence of coins has been applied. Only a few of the categories can be mentioned here.

(p. 9) History. The importance of individual coins for history has long been recognized. The very introduction of coinage, still a matter of discussion as far as chronology is concerned, is a cultural watershed. Many smaller Greek poleis have left coins as the sole or major evidence for their existence. In an interesting interplay of archaeology and numismatics, the non-Roman coins from Morgantina date both its sack and the first denarii ever struck. The appearance of the temple of Janus on the coins of Nero helps to date its closing; recently discovered or rediscovered coins bear on the implementation of the Augustan “settlement” of 28/7 B.C., redating it to 28, and on the adoption of Hadrian. In the later third century, the appearance (or nonexistence) of coins of various short-term emperors or usurpers helps to document their historical reality; the recent emergence of two coins of the third-century Gallic ruler Domitianus gives life to a personality otherwise attested only by the Historia Augusta, which in the absence of the coins would have been grounds for doubting his very existence. The striking of gold and silver in the later empire has been used to date imperial movements. Far more sophisticated applications are illustrated in the essays that follow.

In a broader context, to take but a general example, the dominance of Athenian silver and later of the tetradrachm coinage of Alexander and his successors is clear from finds. This is a safe inference even in the absence of calculations, which, at least in the case of Athens, are a long way from implementation, considering the size of the coinage.

Art history. It is hardly surprising, given the original stimulus to study of ancient coins, that art history has been both a contributor to and a beneficiary of the discipline. The most often cited application of coins is for the identification of portraits: since portraits in the round are almost never accompanied by reliable identifications, coins—with their images accompanied by identifying legends—hold the key to the appearance of both Hellenistic rulers and Roman emperors and their families. Since the appearance of both these categories set fashion (bearded/beardless for men, a multiplicity of hairstyles for women) the information gleaned from coins, which are often very precisely dated, can be extended to private portraits as well.

Economic history. For the Greek world, discussion of economic history is surprisingly free of references to coinage. One recent exception is Von Reden (2007), a study of money in Ptolemaic Egypt in which coinage plays a significant part. One major work (Duncan-Jones 1996) has made considerable use of numismatic evidence in the context of the Roman economy, but somewhat surprisingly the role of coinage has been diminished as historians have come to appreciate the role of other forms of money in antiquity. The effort by Crawford (1974) to associate production of new coin with military expenditure has not proved convincing, even though the military must at most times have represented the largest single expenditure in the Roman budget. Clearly the numismatic evidence has not yet been exploited to the degree necessary for its proper appreciation.

(p. 10) Key to Illustration


Babelon, E. (1904). Traité de numismatique grecque et romain. Vol. 1. Paris.Find this resource:

Buttrey, T. V. (1993). “Calculating ancient coin production: Facts and fantasies.” NC 153: 335–351.Find this resource:

——— . (1994). “Calculating ancient coin production II: Why it cannot be done.” NC 154: 342–352.Find this resource:

De Callataÿ, F. (1995). “Calculating ancient coin production: Seeking a balance.” NC 155: 289–311.Find this resource:

——— . (1996). Les monnaies grecques et l’orientation des axes. Glaux 12. Milan.Find this resource:

Crawford, M. H. (1969). Roman Republican Coin Hoards. Royal Numismatic Society Special Publication 4. London.Find this resource:

——— . (1974). Roman Republican Coinage. Cambridge.Find this resource:

Duncan-Jones, R. P. (1994). Money and Government in the Roman Empire. Cambridge.Find this resource:

Esry, W. W. (1986). “Estimation of the size of a coinage: A survey and comparison of methods.” NC 146: 185–215.Find this resource:

Fittschen, K. (1982). Die Bildnistypen der Faustina II und die Fecunditas Augustae. Göttingen.Find this resource:

Fuchs, G. (1969). Architekturdarstellungen auf römischen Münzen der Republik und der frühen Kaiserzeit. Berlin.Find this resource:

Howgego, C. J. (1992). “The supply and use of money in the Roman world 200 B.C. to A.D. 300.” JRS 82: 1–31.Find this resource:

Jones, T. B. (1963). “A numismatic riddle: The so-called Greek imperials.” PAPS 107.4: 308–347.Find this resource:

Hopkins, K. (1980). “Taxes and trade in the Roman empire (200 B.C.–400 A.D.).” JRS 70: 101–125.Find this resource:

Kleiner, F. S. (1985). The Arch of Nero at Rome: A Study of the Roman Honorary Arch before and under Nero. Rome.Find this resource:

Lyon, C. S. S. (1985). “The estimation of the number of dies employed in a coinage.” NCirc 73: 180–181.Find this resource:

Macdonald, G. (1906). “Fixed and loose dies in ancient coinage.” In Corolla Numismatica: Numismatic Essays in Honour of Barclay V. Head. London: 178–188.Find this resource:

Mattingly, H. (1939). “The great Dorchester hoard of 1936.” NC5 19 (1939): 21–61.Find this resource:

——— . (1960). Roman Coins from the Earliest Times to the Fall of the Western Empire. London.Find this resource:

Mildenberg, L. (1989). “‘Those ridiculous arrows’: On the meaning of the die position.” Nomismatika Chronika 4: 23–27.Find this resource:

Price, M. J., and N. M. Waggoner (1975). Archaic Greek Silver Coinage: The Asyut Hoard. London.Find this resource:

(p. 11) Reden, S. von (2007). Money in Ptolemaic Egypt: From the Macedonian Conquest to the End of the Third Century B.C. Cambridge.Find this resource:

Rich, J. W., and J. H. C. Williams (1999). “Leges et ivra p. r. restituit: A new aureus of Octavian and the settlement of 28–27 B.C.” NC 159: 169–213.Find this resource:

Sellwood, D. G. (1963). “Some experiments in Greek minting technique.” NC7 3: 217–231.Find this resource:

Smith, R. R. R. (1989). Hellenistic Royal Portraits. Oxford.Find this resource:

Thordeman, B. (1948). “The Lohe hoard: A contribution to the methodology of numismatics.” NC6 9: 188–204.Find this resource:

Wegner, M., et al. (1939–). Das römische Herrscherbild. Berlin.Find this resource:

Weiss, R. (1968). “The study of ancient numismatics during the Renaissance.” NC7 8: 177–187.Find this resource: