It’s often said that Greco-Roman polytheism was additive. This is generally correct, but we cannot take this as if polytheism was a cosmology into which more and more gods could be put. Instead, we also have to think of other aspects as open to addition. In particular, there was an openness to more and more explanations of the same facts, not necessarily in the sense that two explanations were collectively more correct than a single one could be, but certainly in that a person who could refer to many opinions was thought to be more educated. Judgement between alternatives was seen as a sign of erudition as well, but in many circumstances, it was more important to show that one could judge well rather than that one could judge correctly.
This additive approach, which could flesh out the picture especially of such deities as had little definite character, is wonderfully demonstrated in a passage of Macrobius’ magnum opus, the Saturnalia. In a discussion of the twelve months and their names, he derives the name of May (Maius in Latin) from the goddess Maia, and then proceeds to go through a legion of propositions about who she is. In one sense, this is the inverse of the process by which already prominent deities collect the names of minor ones as additional titles: in that case, the question is “under how many names is this great god worshipped?”, in the former, “who is really meant by this obscure name?”. But the effect is not so different, as both minor and major gods become unique nexuses of identifications.
Something that distinguishes Roman religion, at least in the period for which we have records, from that of the Greeks is that questions of mutual identity among the local gods (and not just between native and foreign gods) are central from the very beginning. The contrast should not be overstated—Ares and Enyalios or Apollon and Paion, for example, had an unclear relation of maybe-identity-maybe-not from very early on. But if one can trust the testimony of Macrobius, the rituals contained in the pontifical books (admittedly a vague reference, from which no certain date can be extrapolated) already called on Maia, Bona Dea, Fauna, Ops, and Fatua as one. The large number of Roman gods, many with only a very vaguely defined nature or with an extremely specific domain, both encouraged priests and other thinkers to look for underlying structures, but also meant that many different equally convincing structures could be arrived at. But we should probably imagine that many gods were ambiguously maybe-identical-maybe-not with others from the very beginning, and not just secondarily as a result of speculation; or conversely, we might say that speculation is inherent in ritual formation, but that this kind of implicit theory is in need of and generates further theorizing. If speculation is inherent in the ritual texts and practices, it is also a fundamental part of reading and learning them. Roman pontifical learning, in other words, is not just reproduced or transmitted, but in a constant state of being revised and expanded upon. Yet this revision does not (exclusively) take the form of reformation, where one thing is replaced by another, but a kind of stability through seemingly chaotic pluralism.
But to return to Macrobius: he first (1.12.18) gives the opinion of Cingius, that the Maia after whom the month was named is the wife of Vulcan; and against this that of Piso, that Vulcan’s wife is named Maiesta. An alternative explanation (1.12.19) is that the Maia in question is the mother of Mercury; the association is no doubt derived from the Greeks, as Maia is the mother of Hermes. But the common worship of Maia and Mercury was widely popular among Romans. (In fact, there once was a small temple of the two in my hometown, Regensburg in Bavaria, set up by Roman merchants.)
A more speculative interpretation (1.12.20) is that of Cornelius Labeo, who bypasses the relative identifications (Maia wife of Vulcan or Maia mother of Mercury) and instead takes Maia to be a title of the Earth, meaning “the Greater”, and compares this to Mater Magna “Great Mother”, which is in his view also a name of the Earth. This is in line with his general tendency to take divine names as epithets of planets (the earth counting as a kind of honorary celestial body, but not as a planet in antiquity).
Labeo also uses the circumstantial evidence of the rite from the pontifical rites already mentioned (1.12.21), although the Earth is not explicitly addressed in that ritual, because he also takes all the other goddesses to be epithets of her (1.12.22).
Other alternatives, advanced on single argument each, are the Roman Juno and Proserpine; and, with no supporting evidence at all, the Greek Chthonian Hecate and Semele (1.12.23). But it seems that these are identified not with Maia directly, but with the Bona Dea and indirectly with (the Earth and) Maia (1.12.29). More elaborate is the reasoning for taking her to be the daughter of Faunus (i.e. Fauna, who we have already seen was equated with both Bona Dea and Maia in one ritual), which rests on a mythological explanation of the “divine law” in Bona Dea’s temple (1.12.24–25). Circular proof of this is derived from the fact that Bona Dea is is called theos gunaikeia, ‘women’s goddess,’ in Greek, and that Varro says that she is the daughter of Faunus (1.12.27). (Surely Varro says this about the Greek goddess because he identifies her with Bona Dea, and not the other way around?) Another explanation from features of this temple is that she is Medea (1.12.26).
In sum, we have two Maias, we might say one Greek and one Roman; and we have two Roman goddesses with whom she is identified directly, namely the Bona Dea and Fauna (as well as Ops, and Fatua who is barely distinct from Fauna in the first place).
Bona Dea is identified with three Roman goddesses, Juno, Proserpine and the daughter of Faunus, with the Greek Chthonian Hecate (perhaps by extension from Proserpine?) and theos gynaikeia (which in fact served as the usual Greek translation for Bona Dea, and I think was actually coined for this purpose), two deified heroines from Greek myth, Semele and Medea.
And finally, both Maia and Bona Dea are anchored, so to speak, in a “visible deity”, the Earth.
These multiple identifications of Maia, rather than explaining her away, make her irreducible to any of the deities she is identified with. Admittedly, the discussion of Bona Dea almost takes over the account about her, but inevitably some “knots” in the net of identifications must cluster closer together than others. Although we can attempt something of a chronology—for example, the Earth only comes in with Cornelius Labeo, in the 3rd (2nd?) century CE—, I think that we must not write that by assuming that there are original unique identities that then become confused. Even if Maia the wife of Vulcan was worshipped before the borrowing of the Greek Maia and the identification with Bona Dea, this does not mean that she was not identified with other deities at that time. What, for example, are we to make of her doublet, Maiestas? And if we wanted to say that Fauna and Bona Dea are one, then why are there two names in the first place, and why does Macrobius feel he has to justify the identification? To say that the Romans were more confused about their religion than we are is hardly reasonable. Instead, we should think of the conception of each god as always having internal tensions and external resonances. Only this, I think, explains the constant concern with differentiating and identifying them, and only that continual negotiation maintains the intelligibility of the gods.