In 2011, Fleur Kemmers and Nanouschka Myrberg stated in their article “Rethinking Numismatics. The archaeology of coins” that coins, through their integration of text, image, and existence as material objects, offer profound insights not only in the ‘big history’ of issuers and state organizations, but also in ‘small histories’, cultural value, and the agency of humans and objects. Although coins had a value assigned by the mint, individuals or communities used them differently based on their own standards of value, producing “irrational” usage patterns. Coins as active agents can be used to make certain statements, and when they end up in contexts outside their primary context of production or use, they can also become motors of change. The authors point out that users of coins are able to make (political, social, religious) statements about the authorities that minted the coins through a particular use or non-use of these objects. Kemmers demonstrated this theoretical idea in another article where she looked at the use of Severan coins in three distinct functional contexts: military contexts, civilian contexts, and ritual contexts. Here, she pointed out that a disproportional large number of Severan bronze coins could be found in Western-European ritual contexts of the 2nd-3rd century, compared to military or civilian contexts. Thus, a conscious pre-selection of coins seems to have been made when dealing with religious functions. Even more surprising, a deliberate selection also seems to have been made in coin types. Dividing the Severan coins up into 6 types (Jupiter, Mars, Providentia, Sol, Spes, and Other), Kemmers was able to demonstrate that Mars coins were considerably more common in military contexts than civilian or ritual contexts, whereas Sol coins were more dominant in civilian and ritual contexts. In fact, in ritual contexts, Sol is by far the most dominant type. One theory that could explain these discrepancies could be that people adhered different “feelings” to different coin imagery, making certain types more or less suitable for certain functions.
Can something similar be observed for our magico-religious coins? In order to examine this question, I divided all synagogue coins up into 27 different types (Angel, Big E, Big I, Big K, Big M, Camp-Gate, Constantinopolis, Cross, Emperor, Emperor dragging Captive, Emperor on Galley, Fallen Horseman, Jupiter, Monogram, Roma, Securitas, She-Wolf, Sol, Three Emperors, Two Emperors, Two Soldiers One Standard, Two Soldiers Two Standards, VOT(A), Victory, Victory dragging Captive, Victory on Prow, and Other). Then, I compared the coin types that could be found in each specific context (Votive, Charity, Treasury, Emergency, Post-Destruction, and Magico-Religious). For the ease of overview, I removed the Other category (by far the largest category, containing coins from the Hellenistic period to the Medieval Period), so that differences between the other categories would be more visible. The graphs can be found at https://www.ancientsynagoguecoins.com/coin-agency/.
When looking at all synagogue coins (7,396 in total), Fallen Horseman coins form the biggest category (1,346 coins or 18% of total). However, when combining Victory coins (second largest with 1,214 coins) with Victory dragging Captive (third largest with 938 coins), this category forms the largest group with 34,5% of the total coins. Cross coins and VOT coins close off the top five. However, a caveat in this overview needs to be mentioned. The Caesarea deposit, interpreted by me as Treasury, is solely responsible for 1,142 of the 1,346 total Fallen Horseman coins, thus heavily skewing our graphs. When removing Caesarea from the total deposits, the Fallen Horseman type is actually only the 11th largest category of the assemblage and ‘Big M’ coins round off our top five.
More important, however, are the differences that can be seen when comparing the coin types found in our different functional contexts. When dividing the coins up in these categories, it becomes clear that Votive deposits follow our total assemblage the closest. Here, the top five coins types are Victory dragging Captive, Victory, VOT, Emperor, and Fallen Horseman. The two Victory types form 39% of the total assemblage. A totally different picture emerges from the Charity coins. Here the top five consists of Two Soldiers Two Standards, Two Soldiers One Standard, Camp-Gate, Victory on Prow, and She-Wolf. However, just as Caesarea is an outlier, so is Beth She’arim, which is responsible for 204 out of the 205 Two Soldiers Two Standards coins and all 169 Two Soldiers One Standard coins. This rare instance of large numbers of Two Soldier coins is intriguing. A possible explanation might be that this deposit was perhaps not a charity hoard that can be associated with the synagogue building, but a shipment of coins that still needed to be distributed and was temporarily stored in the basement of a building when it collapsed. A future die study of these coins could perhaps indicate if they were minted in the same batch or not, although none of the coins look freshly minted. If they do not come from the same die, then someone was clearly collecting these rare coins for an unknown reason.
When removing Caesarea from the Treasury deposits, Big M coins form the biggest category in this group, followed by Victory, Cross, Victory with Captive, Monogram, Angel, and VOT, for a total of 21 different types. We’ve already pointed out that Treasury coins have the largest range in date, but it now also seems that they have one of the largest ranges in types. A pretty similar picture can be observed in the Emergency deposits, which contains 18 different types. The top five, however, consists of Big M coins, followed by Two Soldiers Two Standards, Big K, Two Soldiers One Standard, and Camp-Gate. If we assume that this category is the most reflective of what people had on hand at any given time, then it seems that all the other categories show clear indications of pre-selection, as none of them have many Two Soldiers, Big K, or Camp-Gate coins in them. The very low number of Victory coins in this group clearly contradicts what we can observe in all the other categories.
The two deposits that could be interpreted as Post-Destruction coins only contain two types: Emperor (3) and Victory (44) coins. Since, however, this is our only group containing solely gold coins, this would explain the lack of many other types that were only minted in bronze.
Our last category is then the magico-religious coins. Here, the top five consists of Victory, Victory dragging Captive, Cross, VOT, and Emperor types. However, the category contains 25 different types, giving us a cross section of almost all bronze types that were around in Late Antique Palestine/Israel, perhaps indicating that people threw in what was around or that arbitrary coins were chosen to collect as tithing coins. Victory/Victory dragging Captive, however, form 42% of the total assemblage, possibly indicating that these particular coins were the preferred types to include in the tithing/ magico-religious deposits, especially since these types are almost lacking in the emergency deposits.
It would be very useful to compare our synagogue coin types to non-synagogue coin types, in order to see if our categories follow or deviate from the standard excavated coins from sites in Israel/Palestine. Unfortunately, no overview could be found on the distribution of types found at all, or a select group, of Late Antique sites in Israel/Palestine. I did, however, browse the bronze coins as preserved in the American Numismatic Society (ANS) collection, dating from 250 to 700 CE. Here, 28,511 bronze coins from all over the Roman/Byzantine Empire and beyond could be found. Of these, Big M coins make up 16.5% (4714 coins), Victory with/without Captive/on Prow 10.5% (3.012 coins), Fallen Horseman 6% (1769 coins), VOT 6% (1,684 coins), Jupiter 4% (1,238 coins), Sol 3.5% (985 coins), camp-gate 3% (925 coins), Two Soldiers Two Standards 3% (868 coins), Roma 2% (640 coins), Monogram 2% (618 coins), She-Wolf 1% (342 coins), Constantinopolis 1% (341 coins), Two Soldiers One Standard 1% (296 coins), Securitas 0.3 % (96 coins), Angel 0.1% (25 coins). Unfortunately, I was not able to search for Cross coins, Big E/I/K coins, or any Emperor coins, as the parameters could not be refined in this way. The database can also not be narrowed down to coins only found in the Eastern Mediterranean or Israel/Palestine (mostly because the origin of most of the coins in public and private collections are unknown). In how far we can thus compare synagogue coins to the Late Antique “average assemblage” is thus debatable. Perhaps a future targeted search in the IAA database would make for a better comparison. In any case, according to the ANS database, Big M coins make up the largest group of bronze coins minted between 250 and 700 CE (16.5%). In the magico-religious synagogue deposits, however, they only make up 1.2% of all coins. Does this mean that people wanted to keep the follis in economic circulation and preferred not to discard them? Was their value too high to be used as a symbolic gift? This hypothesis may be correct, as M-folles also only do not appear in the Charity deposits, and only form 0.68% of the Votive coins. In contrast, Big M coins are the majority in the Emergency hoards and the third largest category in the Treasury deposits, indicating that their value was appreciated in economic contexts.
Table showing all the synagogue deposit coins reverse types. Toggle the parameter to switch between absolute numbers or percentage of total. The “Other” category has been left out for clarity.
Table showing all the synagogue votive coin reverse types. Toggle the parameter to switch between absolute numbers or percentage of total. The “Other” category has been left out for clarity.
Table showing all the synagogue charity coin reverse types. Toggle the parameter to switch between absolute numbers or percentage of total. The “Other” category has been left out for clarity.
Table showing all the synagogue treasury coin reverse types. Toggle the parameter to switch between absolute numbers or percentage of total. The “Other” category has been left out for clarity.
Table showing all the synagogue emergency coin reverse types. Toggle the parameter to switch between absolute numbers or percentage of total. The “Other” category has been left out for clarity.
Table showing all the synagogue post-destruction coin reverse types. Toggle the parameter to switch between absolute numbers or percentage of total. The “Other” category has been left out for clarity.
Table showing all the synagogue magico-religious coin reverse types. Toggle the parameter to switch between absolute numbers or percentage of total. The “Other” category has been left out for clarity.
 Kemmers and Myrberg 2011, p. 87.
 We already saw this relationship through the fabrication of medals or amulets out of specific “potent” coins.
 Kemmers 2009.
 Kemmers and Myrberg 2011, pp. 94–96. Similar studies were performed by Kaczynski and Nüsse, who looked at the numbers of antoniniani coins found in specific sanctuary contexts in Germany, and concluded that coin types with military symbols were less often chosen in these contexts (Kaczynski and Nüsse 2009), and by Nathan Elkins, who looked at the circulation of Nerva coins to determine of some of his imagery was targeted to specific geographic audiences (Elkins 2017).
 Of course, these are artificial classifications. There is no textual nor material evidence that the ancients would have viewed coin images in such categorical fashion (See also Elkins 2009). However, these categories form a starting point in trying to see internal similarities and differences between the different functional coin groups.