Skip to content

The Tripartite Theory

One bugaboo that haunts our and others’ thoughts about the gods is Georges Dumézil’s Tripartite Theory.


Briefly, based on an analysis of “mythologies” of the Indo-European language family, and using texts such as Rígsþula as keys, Dumézil grouped the gods (and as an afterthought the goddesses) into one of three functional niches: sovereignty, war, and fertility (often referred to as first, second, and third function respectively). He saw the first function as characteristically divided between two gods: a “sovereign god of law” and a “sovereign god of magic” (his type examples are Mitra and Varuṇa, respectively). The second function is pretty easy to understand: gods like the Hindu Indra, the Greek Ares, and the Roman Mars. For the third function, he was heavily influenced by Greek and so regarded fertility gods as also having to do with death: chthonic. Dumézil saw the fertility gods as characteristically paired and associated with twin-ness; this partly because of the Greek Dioscuroi and Vedic Aśvins, both divine twins, and partly because sexual relations require two bodies. The Vanir seemed to him to obviously fit this third function, and this is probably why so many heathens have absorbed the notion that Freyja and Freyr are not just sister and brother, but twins. It was in fact the Germanic pantheon that led him to divide the first function, that of sovereignty, between two gods: in order to accommodate both Týr and Óðinn. Some of the details of his theory show this quite clearly: the first kind of sovereign deity (Mitra, Týr) embodies divine Law and is implacable, while the second (Varuṇa, Óðinn) manipulates things and is from a human point of view cruel and capricious. Týr is god of the þing and does not laugh; Óðinn is called furor (fury) by Adam of Bremen and rather often kills his followers. Whereas Mitra and Varuṇa tend to be contrasted by calling one day and the other night, one the god of the visible and the other of the invisible, and so forth; a lot more abstract.


Dumézil’s theory was developed primarily based on gods of ancient India and the Zoroastrian figures that are successors to them and on Roman history, legend, and custom that tell us about the effaced pantheon of ancient Rome. In these materials, there is quite a clear element of social stratification, the Indian castes being quite blatant. One way of defining mythology that is widely accepted in academia is as stories that teach social relations in religious terms. In some sense of course, this is obvious; we learn a little bit from every story, in particular about human relations, behavioral options, and morals, and a story about the gods, being important, is thus particularly influential. But Dumézil glommed onto the threefold class heirarchy presented in Indic texts in particular, and welcomed Rígsþula as an almost unambiguously similar statement among Germanic texts. So there is a definite element of classism behind his analysis: he does indeed regard the gods of the first function as superior to those of the second, and those of the second as superior to the agricultural gods of the third function. Applied to heathenry, this leads to the idea that Thor is a god of churls and that followers of Óðinn are inherently better – as sometimes seen in the works of Edred – and to dismissal of the Vanir. It’s also been the main point of protest against Dumézil now that the shine has worn off his work since he died: he was personally enamored of class and authoritarianism, and anti-imperialist nativists are among the main attackers of the whole Indo-European enterprise, so they don’t like that at all.


But in any case – this doesn’t fit heathenry at all well. To begin with Rígsþula: that’s an anomalous poem about a god otherwise unmentioned in the lore, whose name is the Irish word for king, and who may or may not be Heimdallr. It also breaks off after defining four classes: Kon Ungr, the prototype of the king, is Rígr’s grandson, the son of the last of his three sons. And where is the support in the rest of the lore for social stratification related to different gods? All I can think of is the gibes in Harbarðsljóð.


“Gods of the peasantry” is a poor way to think of the Vanir. They were winning the war against the Æsir at one point – they devastated Asgard so that it had to be provided with a new wall by the giant builder. Freyr is betstr allra ballriða (best of all warriors – Lokasenna), folkvaldi goða (leader of the host of the gods – Skírnismál), and even ása iaðarr (lord of Æsir – Lokasenna again). Lest we think these terms are tongue-in-cheek, let’s recall that in addition to being the or one of the leaders of an army that breached the defences of Asgard, he not only killed a giant with a stag’s antler – he could have done it with his bare hands (Gylfaginning). And he’s called veraldar goð (god of the world, a term that lived on in Sami religion) in both Ynglingasaga and Flateyjarbók. That’s your “warrior function” as well as your “sovereign function.” And in any case, Germanic armies have always been fairly meritocratic. They started as levies of all the suitably aged men of the tribe and continued as levies of the menfolk of a shire or a country. That’s the same guys who farmed the fields. And in the Viking Age, the viking went out harrying to get treasure – with which to settle on a farm. (Unless of course he made himself king of some place like Sicily.) Classes of people who don’t farm, let alone a middle caste who only make war, are a late development in Germanic society, only really coming about under feudalism. As we see from both the sagas and the lore of the gods, such a division of labor is just not Germanic, especially the brushing off of farming. What mattered more was how free the farmer was.


And our gods all have something to do with sex. Thor blesses the marriage with his hammer! In fact, Thor single-handedly casts the tripartite division into doubt. He’s not a muscle-bound brute – he’s djúphugaðr, the deep-thinker (Haustlǫng), and he bests the dwarf Alvíss (all-wise, heh) handily in a contest of wits. Conversely, pretty much all our gods have to do with war – Týr, Óðinn, and Thor are often grouped together as “war gods,” and of course Týr is particularly identified that way. Not with rulership; some people don’t even known he is also god of the þing. Dumézil was troubled from the start by this “flattening” effect whereby “first function” gods of sovereignty were “pushed down” to the “second function” of war, and he became so troubled by the variety of manifestations of strength in Germanic “myths” that he bisected the second function, too, to account for Thor’s associations with fertility and the earth and for the frequent occurrence of chaotic warriors in Germanic stories. Of course, that meant he had each of the three “functions” divided between two types,  making it simpler and potentially more accurate to resolve the system as a dualistic opposition between constructive and destructive.


But in Germanic terms, it simply won’t wash anyway. Freyr won’t fit well into his system; Thor won’t fit well into his system; Óðinn frankly won’t fit well into any system; and what about Freyja taking half of those who die in battle? If that is simply the death part of being a “chthonic” deity, then what about the other half, that Óðinn takes? And looking at sovereignty, don’t Baldr and Forseti fit that model quite as well as Týr? One of the big problems is that war was the ultimate way of settling judicial questions in Germanic culture; this remains so even after the custom becomes accepted of saving general bloodshed by having a duel between designated representatives stand in for battle between armies. So it simply isn’t true that sovereignty and legal judgment resided in a sphere distinct from war. War was a legal judgment and the ruler had to be a warrior, not just a general. Dumézil’s class stereotypes don’t fit, and Germanic tribes did not set roles apart even in the manner of the three Vedic castes. Also, Freyja’s magic (and many names) are a counterpoint to Óðin’s. And where does Ullr fit in the neat system? Never mind Loki . . .


One reason the Germanic pantheon busts out of Dumézil’s system may simply be that we have an unusually large number of gods and goddesses recorded. Snorri tried to say there were 13, presumably on the model of the Greek pantheon, but there are far more in his own work. But those we know most about fit least well into the niches. And the whole thing presupposes that the Germanic pantheon changed only slightly after the tribes left the Indo-European homeland. That flies in the face of a lot of evidence. The Æsir-Vanir war and subsequent treaty. The substantial indications that Týr and Ullr are older gods. (One cannot have it both ways – if Týr was originally the same figure as the Greek Zeus, and a sky god, then it’s plain his role in the pantheon has changed.) In fact the rigidity of Dumézil’s theory and its problems accommodating our gods and goddesses cast into question the closeness of our pantheon to other Indo-European ones. It reminds one of how different the Ṛg-Veda and the Greek and Roman myths are from our lore. They have very different themes and obsessions (gods and goddesses manipulating people; gods giving people a whole raft of rules to live by . . . ) That’s why Rígsþula is so anomalous – it’s more like those other scriptures than the rest of our stuff is. And it’s the contact point with Dumézil’s theory.


So I think we’re better off not letting it color our view of the gods; and rolling back the influence it has had in our thinking, such as Edred’s assertions about Odinists being superior to Thorians, the comparative neglect of the Vanir, and views of kingship as inherently Germanic.

Posted in Forn Seðr.

0 Responses

Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.

Some HTML is OK

or, reply to this post via trackback.

Note: Click "@" to notify commenter of your reply.

Bad Behavior has blocked 88 access attempts in the last 7 days.