can’t believe I only just learned about the attested Gaulish word “marcosior” (< PCeltic *marko- ‘horse’ but as a passive verb in the desiderative or future) which means “I want to be ridden like a horse” and occurs in exactly the context you’re imagining
that was fast. easy website
im seeing people fact-check this in the notes so i'll add that the full inscription is "marcosior maternia" and if you search that in google scholar you'll find lots of sources. it might not be true but it's definitely real, by which i mean that scholars don't know for sure what "marcosior maternia" means but this is a common interpretation. it's written on a spindle whorl (little weights that are used in spinning) and there's actually a lot of spindle whorls from roman gaul that have sort of amatory inscriptions on them, so they might have been gifts for lovers but it's not clear, but anyway a sexual meaning is perfectly expected for this type of object, and maternia looks like a woman's name in either the nominative or instrumental case, and marcosior looks like a passive verb related to the word for 'horse' in a desiderative or future form. so "i want to be ridden by maternia", "i will be ridden by maternia", and "maternia will be ridden" are some of the plausible interpretations.
also i didnt emphasize in the post just how fragmentary this language is. there's only a few hundred inscriptions in gaulish and theyre short. we dont know if there was an optative mood. but we do know how to say something about getting ridden by maternia
someone in the notes writes asterix fanfiction but isn’t putting sexy spindle whorls in it. coward
OK @esoanem I'll bite, how much evidence do we have for deponency in Gaulish? Given the behaviour of the corresponding r-passives in Latin it would seem sensible to suppose there was some but is there any actual inscriptional confirmation of that?
Not sure of the specific examples, but the Cambridge Encyclopedia of the World's Ancient Languages in its chapter on Continental Celtic by Joseph Eska says that "[t]here are not a few forms which terminate in -r, but the majority are deponents, and hence active in voice."
Given how well-attested they are in Old Irish, it would also be a little surprising if they'd dropped out entirely in Gaulish given how much earlier our sources generally are.
That said, in Old Irish, deponents are almost all of the W2 class which forms an f-future not an s-future. Likewise, whilst there is limited evidence of deponent f-futures, there doesn't seem to be an Old Irish evidence of a deponent s-future (3rd person passive forms are known, but in Old Irish the formal identity of deponents and passives was not well maintained).
Obviously Old Irish is a complicated informant here, both because of it hardly being a close relative (within Celtic) and from the complications inherent in doing anything to do with its verbal system.
As deponent s-futures are attested elsewhere in Indo-European (e.g. Greek), and given the distribution of deponents in Old Irish, it wouldn't be too surprising if they existed there too at some stage, but later got displaced by the innovative f-future (and, provided the f-future is young, we would then expect deponent s-futures to survive in other early Celtic languages, like Gaulish).
However, the W2 verbs go back to primary thematic verbs in *ey- and athematic statives in *eh1- (or thematic in *eh1ye-), neither of which are good candidates for this verb.
We might expect instead a W1 verb (which includes thematic denominal in *-eh2ye-, parallel to the Latin 1st conjugation), or maybe a denominal in *ye- which ended up analogically drawn to the W2 class (or rather the Gaulish equivalent of it). W1 verbs also generally form f-futures, and can be deponent (although are less often so than W2 verbs), so this doesn't overly complicate the picture.
I do think the passive interpretation is more likely correct, especially given the context of other sexual spindle whorls mentioned in some of the papers on this inscription, but a deponent one definitely doesn't seem implausible.
