558 Walter Pohl
the Romance-speaking majority ofFranci(French), without even being aware that their
identity had changed fundamentally. Methodologically, the clear implication is (although
this insight has spread relatively slowly) that we cannot automatically infer from ethnic
or political identifications in written sources to archaeological culture and to linguis-
tic classification, or vice versa. Not all Ostrogoths in Italy were necessarily buried with
grave goods that could be classed as Gothic, or spoke Gothic at home (although some
certainly did).
Goths and Gothicness
For further questions of late antique ethnicity, the Goths have become something of a
contested test case. What is still being debated is the extent to which there was ethnic
continuity among the Goths, how much their Gothicness mattered to them, or whether
“Goths” was just a convenient label for outside designation and, perhaps, self-promotion.
These debates are based on two breakthrough monographs that appeared in German in
the 1960s/1970s: Reinhard Wenskus’ “Stammesbildung und Verfassung” (1961) and
Herwig Wolfram’s “Die Goten” (1979/2008, English translation 1988). Wenskus, who
taught in Göttingen, used the late antique evidence about “Germanic” peoples to show
that none of the elements used to define ethnicity—common origin, language, culture,
etc.—is universally attested. His solution was that only a subjective definition seemed
viable: according to him, ethnicity (he preferred the term “Gentilismus”) is basically the
result of self-identification. That clearly implies that it can change, even in an individ-
ual lifespan. Thus, for the first time, Wenskus could account adequately for the many
changes in composition of migrant peoples such as the Goths. According to Wenskus,
what they had in common was, apart from the name, a common “tradition,” which
consisted of norms and narratives that accounted for the distinctiveness of the group.
These ideas, Wenskus believed, were handed down by small, high-status groups (which
he calledTraditionskerne, kernels of tradition). Their prestige (and that of the “tradi-
tion”) could attract new members to the group, so that, in case of success, it could grow
into a new tribe (Wenskus spoke of tribes and of their formation,Stammesbildung). This
model allowed for a dynamic view of the ethnic processes of the migration period, and
soon became very influential. Today, we have to acknowledge that, in spite of all its mer-
its, it also has its limitations, due to the time when it was published: it is quite gendered
and elitist (elite males transmit ethnicity), it overemphasizes the solidity of the “kernels
of tradition,” and it still operates within the overall framework of traditionalGermani-
sche Altertumskunde, studies of Germanic antiquity, largely ignoring the Roman cultural
matrix in which the migrating groups increasingly moved.
When Herwig Wolfram adapted the Wenskus model in his history of the Goths in 1979
(Wolfram 1979/1988; see also Wolfram 2005), he most of all added the Roman perspec-
tive. Wolfram’s narrative of Gothic history traced the successive stages of the emergence
of Gothic groups from mythical origins to Goths well integrated in the Roman world
in the sixth century, when the Romanized Goth Jordanes wrote hisGetica(History of
the Goths). For this formation process, Wolfram used the term “ethnogenesis” current
in the ethnology of the time. This was not conceived (a point misunderstood by his
critics) as a linear development, let alone a teleological one, but as a long and broken