Why Did Byzantine Coins Become Cup-Shaped?
By Mike Markowitz for CoinWeek …….
Ancient Coin Series ……
On March 27, 2014 the US Mint released its first curved coins: five-dollar gold, one-dollar silver and 50-cent copper-nickel clad proof commemoratives celebrating the 100th anniversary of the Baseball Hall of Fame. The reverse shows the stitching of a baseball (convex side), while Cassie McFarland’s catcher’s mitt is on the obverse (concave side.)
These are not the first cup-shaped coins in history. Beginning in the 11th century, and continuing for about three centuries, the coinage of the Eastern Roman ( “Byzantine”) Empire shows an increasingly cup-shaped fabric.
Figure 1 Gold HOF Baseball Commemorative
Beginning about 312 AD under Constantine the Great, the gold coins struck by the Eastern Roman Empire were a standard of purity and value, accepted in trade throughout the known world. They turn up in hoards in Scandinavia, brought home by Vikings who served in the emperor’s guard. They’ve been found in Central Asia, India and even China. In the West, as long as Latin was the language of commerce, these coins were called solidi. As the use of Latin declined, they came to be called bezants, and many other names, such as nomisma, which simply means “coin” in Greek.
During the reign of emperor Michael IV (1034-1041), the gold coins became slightly concave. By the middle of the 12th century most of the coinage in all metals was deeply cupped. The reason for this bizarre transformation has baffled and intrigued generations of numismatists.
In this article we review some of the theories and discover an explanation that makes sense, based on what we know about metallurgy and structural engineering.
Coinage is normally a very conservative thing. People are skeptical about changes in their money, thinking the government is trying to put one over on them. In the case of Michael IV, this was probably true. Michael “the Paphlagonian” came from a humble peasant family. His brother John was a palace eunuch who rose to a powerful position in the Imperial court. Michael, who had worked as a moneychanger (and possibly a counterfeiter), became chamberlain to the Empress Zoe. He soon became her lover and eventually her successor, after arranging the murder of her husband, the Emperor Romanos III (11 April 1034). That was how the game of thrones was played by Byzantine rules.
Figure 2: Slightly dished - Gold Histamenon Nomisma of Michael IV Constantinople before 1041. Sear 1824
Over the course of his six-year reign, Michael IV apparently reduced the purity of the gold coinage from 23½ karats (.9792) to about 19½ karats (.8125). Some of Michael’s coins are perfectly flat, but some are struck with a slight curvature. It is likely that the later, more “debased” coins are the curved ones. Michael’s death, possibly from complications of epilepsy in 1041, was followed by a period of instability, when the coinage resumed its normal shape but continued a gradual debasement. Under Constantine IX (1059-1067) the cupped shape became dramatic.
Fig 3. Deeper cupping — Gold Histamenon Nomisma of Constantine IX 1059-1067
By 1092, when emperor Alexios I was forced to reform the currency, the “gold” nomisma was being struck from “billon” (a greyish alloy of copper and silver) without a trace of gold. To restore confidence, Alexios created a new concave coin, the hyperpyron, struck at the old weight of nearly 4.5 grams, in an alloy of 20 ½ carat gold.
In the medieval world, rulers were strongly tempted to adulterate or “debase” the currency, to stretch a limited supply of precious metal to pay more soldiers, build another palace, or distribute more diplomatic bribes (oops, I meant “gifts!”) But debasing the coins undermined public confidence, contributed to inflation and created all sorts of technical problems for the mint.
Gold (specific gravity 19.32) is much more dense than silver (specific gravity 10.49), for example. So if you want to hold the weight of a coin constant, mixing a higher proportion of silver into the gold will give you a bigger coin. Either it will become thicker (making it harder to strike), or it will be greater in diameter (requiring bigger dies to strike.)
In the year 1000, the average diameter of the gold nomisma was about 23mm (0.9 inches). By 1085, the coin had the same thickness but measured about 31mm (1.22 inches) in diameter. All things being equal, a bigger coin is more impressive. The typical design bore a portrait of Christ on the convex (obverse) side, and the figure of the emperor on the concave (reverse) side.
A large, thin coin is easily bent, and because a gold-silver alloy is more brittle than soft, pure gold, it’s also more likely to crack.
Alongside the gold hyperpyron, Alexios created an “electrum” (mostly silver) coin worth one-third as much, the aspron trachy (“white rough”), and a low-value silvered copper alloy coin, also confusingly called a trachy. Initially overvalued at 1/48 of a hyperpyron, the trachy gradually depreciated to as little as 1/184 of a hyperpyron or less.
Fig 4 Extreme cupping — Andronicus I Comnenus. 1183-1185. Billon (copper-silver alloy) Trachy. Constantinople mint. DOC 3; SB 1985.
How They Did It
Curved coins are more difficult to strike. Modern coin dies are precision-made machine tools of hardened steel, but medieval coin dies were individually handcrafted in soft iron. The artisans who made them could not maintain close tolerances. Even if they could, the process of hand hammering would have quickly destroyed tight-fitting die sets. It might seem logical that the lower die (fixed in an anvil) would be for the concave side, and the upper (“punch”) die would be for the convex side, but this would make it nearly impossible to extract a finished coin from the die. It is clear that the lower die must have stuck up from the anvil, and the upper die fitted — loosely — over it.
In a brilliant series of articles, the British numismatist Simon Bendall explained how cup-shaped coins must have been struck. If the lower die had greater curvature than the upper die, then the central area of the coin would be fully struck but the outer margin would be weakly struck or even blank. There are many such coins. If the upper die had greater curvature that the lower die, then only the edges of the coin would be fully struck and the center would be weak, or blank. No such coins are seen.
Fig 5 : how they did it
On many 12th century coins, there is an obvious mismatch or mis-alignment between the left and right-hand sides of the coin. This suggest that the upper die was rocked, or even re-positioned for at least two blows of the hammer at different angles, to ensure that as much of the design as possible would be transferred to the coin. On some coins the mis-match is so great that it is likely that two completely different obverse dies were used to strike the left and right sides.
Fig 6. Left-Right Die Mis-alignment on Hyperpyron of Alexios III (1195-1203) Sear 2008 (From the author’s collection)
Why did they go to all that trouble? A variety of explanations have been offered. For many years, the consensus of historians was that a curved shape indicated to Byzantines that a coin was alloyed, while flat coins were of pure metal. Some suggested that concave coins stacked better (they don’t) or that they jingled more pleasantly in a purse (doubtful.) Some claimed that the coins are harder to counterfeit (definitely true, because of the curvature of the dies, although there are examples of cast cup-shaped bronze counterfeits that were probably meant to be gilded or silvered).
A 19th century French numismatist suggested that the shape improved the playability of the coins in tsikaki, a Greek version of tiddly-winks, because he had seen contemporary small copper coins hammered into shallow cups by players.
It seems clear that the main reason for the emergence of the concave fabric, and its survival for almost three centuries, was to make thin coins strong enough to resist bending and breakage. When the fabric became too thin and the alloy was too brittle, edge cracks inevitably became a problem–something familiar to every collector of these difficult coins.
Fig 7: Reaching the limit: Severe edge cracking. Manuel I Comnenus. 1143-1180 “Electrum” Aspron Trachy (Silver, 3-5 kt gold). Struck 1164 – 1167 Constantinople Sear 1960.
Thin curved shells are inherently strong and light. This is how Mother Nature designed the eggshell. The Mint of Constantinople was inside the Imperial palace compound; and every day going to work, the mint workers were reminded of this as they passed the soaring dome of the church of Hagia Sophia, which still stands.
Fig. 8: Thin curved shell – the dome of Hagia Sophia Constantinople (rebuilt 562)
Medieval technology was rarely documented; everything technical was treated as a “trade secret.” But the basic principles of metallurgy and engineering don’t change. In the cup-shaped Imperial coinage, we see how the mint masters of Constantinople creatively applied these principles to solve a complex technical, economic and political challenge.
Mike Markowitz is “Second Consul” of the Ancient Numismatic Society of Washington. He has been a serious cllector of ancient coins since 1993. He is a wargame designer, historian and defense analyst, who writes for StrategyPage and Defense Media Network. He designed the game Alexandros, which won the 1991 Charles Roberts Award for best pre-WWII wargame. He has degrees in History from the University of Rochester, New York and Social Ecology from the University of California, Irvine. He has worked as a technical writer, editor and trainer for a variety of aerospace and defense firms. Born in New York City, he lives in Fairfax, Virginia.
In addition, Mike presents his expert insights on three numismatic video features shot by CoinWeek’s David Lisot. They are: Ancient Roman Coins for Beginners, Mike Markowitz, WHIT14-001, 70:00m; Introduction to Collecting Ancient Coins, Michael Markowitz, WHIT13-003, 69:57; and Coins in the Bible, Michael Markowitz, WHIT13-009, 44:26. Mike’s videos are available for purchase at $25 each plus $4 per order for S/H at: www.coinvideo.com or by emailing David directly.