Goths and Huns 565others, which he took as keys to authentic myths. The Gothic origin story as told by
Jordanes includes a mythological migration from Scandinavia and a genealogy going
back to Gapt (attested as Geat in Anglo-Saxon genealogies), but also identifies the Goths
with Scythians, Getae, and Dacians, known from ancient history. Since Wenskus, our
understanding of these origin stories has changed (Wolfram et al. 2003), and we have
to acknowledge that they are Latin texts following ancient models and predominantly
based on classical myths and ethnography. In that sense, the Gothic origin myth from
Scandinavia is perhaps little more “authentic” than the Trojan origins of the Franks;
both stories had the same function for the retrospective legitimization of the ruling elite
of their kingdom. On the other hand, as argued earlier, it is remarkable that, in the
guise of classical literature, we also find traces of non-ethnographic material. We can only
speculate how old these traces may be. However, it is unlikely that the references to
thesemidei id est Ansesfrom whom the Goths were descended (Jordanes 1882, c.13,
76), or to the god Wodan from whom the Longobards got their name, were invented in
Christian kingdoms. We have to acknowledge that such “traditions” were not immutable
and as archaic as Wenskus thought. However, on the other hand, it is rather unlikely that
they were completely made up by their Latin authors, and that the Goths arrived on
Roman territory with a completely blank cultural memory. What we have is perhaps not
enough to fuel the “mythomoteur,” which is, according to the sociologist A. D. Smith,
the driving force for the construction of ethnic and national identities (Smith 1986).
However, it is there, and should not be brushed aside in an all-out attempt to deny any
barbarian element in the self-identification of the new barbarian elites (Pohl 2002c).
Still, considering the evidence, it is very hard to employ Wenskus’s subjective criterion
for ethnicity. It is more adequate to understand ethnic identities as the result of a series of
identifications: people identify with an ethnic group (and are accepted as members), the
group identifies itself through its representatives or in collective ritual, and others per-
ceive it as an ethnic group. Some texts and objects that have come down to us are traces
of such a series of acts of communication in which ethnicity was constructed (Pohl 2013).
This model has the advantage that it does not leave us with a binary decision—Goth/not
Goth, or ethnic/not ethnic—which creates the kind of conundrum around which some
of the less fruitful discussions of the past two decades have revolved. Instead, it offers
the option to describe ethnicity by degrees: Goths to whom Gothicness mattered more
or less; outside perceptions of individuals or groups as Huns who may have identified
primarily with a smaller ethnic group, but were also proud to be regarded as Huns; the
subtle modulation of royal ideology that relied on Roman, imperial, ethnic, or dynas-
tic representations according to audience and situation; or the mobilization of Gothic
identity and memories under military pressure.
Such contemporary formations of identity, and their textual representations, then cre-
ated the potential for further identifications. Even negative stereotypes, such as the iden-
tification of Goths and Huns with the apocalyptic peoples Gog and Magog, contributed
to the extraordinary prestige that they accumulated through the writings of their Roman
adversaries (and, later, advisors). The resources of Gothic and Hunnic identity were so
vibrant that they could be reused for much later, purely imaginary identifications—for
instance, of the early modern Swedes with the Goths or of the Hungarians with the
Huns. Goths and Huns (as many other barbarian peoples of their time) were not sta-
ble ethnic groups who emerged from the barbarian periphery and remained immutable