Today I’ve been working on a piece on very short inscriptions – so short that we’re not even sure what language they are in.
There are some obvious reasons why very short, abbreviated inscriptions are sometimes used. On a coin or a tile stamp, space can be so limited that an abbreviation of the name of the craftsman or workshop is all that can fit. In these cases, the abbreviation acts as a kind of “logo”, which identifies the maker. Because many different ancient languages used the same few alphabets, we can’t necessarily decide what language the inscription is supposed to be in. For example, Greek, Gaulish and Oscan all use the Greek alphabet, and were all in contact with each other through trade. So if we find an item that has been traded between the areas which spoke these languages, we can’t use the alphabet to decide what language the maker was using.
A good example of this problem in action is this silver cantharus, or two-handled drinking cup, excavated in the mid nineteenth century from Alesia in central Gaul. Its date is debated, but it may be from around the first century BC. Because of the cup’s similarity to silver goods from Italy such as the Boscoreale Treasure (although it probably pre-dates the Treasure by at least a generation), most scholars have agreed that the cantharus was produced in Italy, not Gaul.
Alesia is best known for being the site of a battle between the Gauls and Julius Caesar, who besieged and took over the town in 52 BC – right around the time when the cup was probably made. Who would have brought such a beautiful and unusual piece to Alesia? Could it have been Julius Caesar himself?
It sounds a little like wishful thinking, perhaps, but it has been suggested quite seriously that this was Julius Caesar’s drinking cup. That is not, of course, the only possibility. Gaul had strong trade links with Italy, especially with southern Italy, long before Julius Caesar came along. But the historical link to the siege is hard to resist.
The inscription on the cup certainly doesn’t solve the problem of who drank from the cup. The cup has three inscriptions underneath the foot. One is a series of circles and lines that denote the weight of the piece – and these suggest that it was weighed based on a system that was in use after approximately 91 BC. Another is almost illegible, and might have been added at a later date. The third is the most interesting, and shows how difficult abbreviated inscriptions can be. It reads:
It is not possible to establish whether the inscription was written at the time of manufacture, in Oscan, or was added later, in Gaulish. The Greek alphabet was used for both, and there are no morphological endings. Three solutions have been proposed:
Lejeune (‘Le canthare d’Alise’, 1983) – Gaulish
μεδα(μος ?) αραγε(νι? -νιος ?) ‘Medamos, son of Aragenos?’
Lejeune (‘Le canthare d’Alise’, 1983) – Oscan
μεδα(τιες) αραγε(τασις) ‘Medaties, silversmith’
Crawford (Imagines Italicae, 2011) – Oscan
με(τις) δα(τιες) αραγε(τασις) ‘Metis Daties, silversmith’
This is a great example of how inter-connected and multilingual the ancient Mediterranean was: this cup was produced in the south of Italy, was brought to Gaul through trade or warfare, and might have been used in either a Gaulish or a Roman town. The whole system is so interwoven that it is impossible to know who wrote the inscription and what language they were writing in.
I think, though, that we might be going about this problem in the wrong way – there’s no reason why an abbreviated inscription has to be in any particular language, in the same way that a company logo is not necessarily in one language. Craftsmen in the ancient world knew that they were creating items like this for export, and presumably had some idea of how widely the Greek alphabet was used. By removing the endings of the words, they were able to write inscriptions which could be read in whatever language the reader chose.
All the same, it’s pleasing to think of Julius Caesar relaxing after a long day with a drink, and taking a sip from a cup with an Oscan (or Gaulish!) inscription on it.
Edited to add:
I’ve had some questions about the letter shapes, so here’s a copy of the drawing from Lejeune, which is reproduced in Crawford’s Imagines Italicae. If you have any further questions on bibliography, Crawford has a short list of the recent work on this (there’s actually not that much).
Nick – ha, good catch! Edited now. (But I was tempted to leave it as it was!)
Katherine: “in the same way that a company logo is not nec.” I like the way you abbreviated this sentence as an instance of what you were saying.
David: Untermann does not include proper (human) names in his Woerterbuch: that’s probably why μεδα is not listed there.
I don’t feel too strongly about the letter shapes either. Since they are cursive, a lot of individual variation is possible. I only checked the register of letters in Recueil des Inscriptions Gauloises I (1985) for parallels. Many more texts have come to light since then, but I don’t have them ready at hand.
The idea about the maker’s logo makes sense, although I would still expect such a logo to make sense in a language, either in the producer’s own tongue (Oscan?), or in an international trade language (Latin, Greek?).
Thanks for your thoughts, David – much appreciated.
Oscan araget- (cognate with Latin argentum) does seem to be the most convincing explanation for the second word, though it’s only attested in Oscan as ‘money’, and not ‘silver’. That’s not too much of a problem, as it’s easy to see how the word could be used with both meanings, but it’s worth noting.
As I said in the post, though, we may not need to decide what language the inscription is in. If the cup was intended for the export market, then it might be intended as a kind of maker’s logo that was readable by speakers of many different languages. This is a different way of thinking about inscriptions than we normally do, as we tend to want to categorise everything – but perhaps we should sometimes embrace the uncertainty?
The letter shapes are interesting – they are not particularly typical of Oscan either, but they are quite cursive in style, and they may be based more on handwritten than inscriptional styles. (Alex Mullen talks in her book about the possibility that the Gallo-Greek alphabet seems to based more on Greek handwriting than on Greek monumental inscriptions, as well, incidentally.)
A few random thoughts:
1. I have never seen this inscription mentioned in the Gaulish scholarship after 1983, so I guess that Lejeune himself must have rejected the possibility of it being Gaulish. The (suggested) names on this object have not been taken into account in the most recent collection of Celtic names (Delamarre, Noms de personnes celtiques).
2. Lejeune’s suggested names are doubtful in their “Gaulicity”: the name Medamus is attested several times on Roman stones, but as far as I can see never in a context that could be proved to be Celtic; it could be any circummediterranean language. The other name is also uncertain: there is one *Arragenus in Cologne, but again the context need not be Celtic; there is also no Celtic etymology that springs to the mind for it.
3. As for the Oscan hypothesis, this seems much more plausible: Oscan has araget- “silver/money”, which fits perfectly. Although there is nothing that fits meda directly, the nominal stem med- is well attested in the language. Interestingly, Untermann (Wörterbuch Oskisch-Umbrisch) mentions arage, but does not include meda.
4. I am not aware of anaptyxis in the Gaulish word for “silver”, i.e. it is to my knowledge always attested as arg/canto-, but never arag-. Also, the second syllable of the Gaulish word has a, not e.
5. As for the letter shapes, although parallels can be found in the Gallo-Greek corpus, there is only a single parallel for the specific type of Δ (G-242), and one for this type of Α (G-241), both in Bibracte. The normal way to write Α in Gallo-Greek texts is to use a variant with horizontal stroke.
In summary, I have strong doubts that this inscription is Gaulish/Celtic.