Perhaps the most significant of these caveats is that identities are not sin-
gular and exclusive, as the empirical work based on Eurobarometer data and
other analyses clearly demonstrates. Not only might individuals maintain af-
fective ties to several levels of political organization (e.g., municipalities, re-
gions, nations, states, supranational regions, the world) simultaneously, but
they may also do the same with respect to associations from other social do-
mains as well (e.g., family, religion, ethnicity, race), with some of the latter at
times even conflicting with or eclipsing political affinities.
14
Nor, for many in-
dividuals, are identity constellations necessarily static over the course of a life-
time. Charles Tilly makes the point eloquently:
Any actor deploys multiple identities, at least one per tie, role, network, and
group to which the actor is attached. That others often typify and respond to
an actor by singling out one of those multiple identities—race, gender, class,
job, religious affiliation, national origin, or something else—by no means es-
tablishes the unity, or even the tight connectedness, of those identities. That
sickness or zealotry occasionally elevates one identity to overwhelming
dominance of an actor’s consciousness and behavior, furthermore, does not
gainsay the prevalence of multiple identities among people who are neither
sick nor zealots. It actually takes sustained effort to endow actors with uni-
tary identities. (1998: 4; see also Smith 1992: 59)
Fortunately,the fact that people maintain complex identity structures does
notpreclude measurement of levels and changes with respect to any one or
more of these identities; but it does suggest caution, especially in trying to
affix onto individuals a single identity label.
As indicated above, there are also very good reasons to believe that iden-
tities are far less constant than, say,anationalist demagogue or a primordial-
ist-oriented scholar would care to admit. Indeed, a good deal of evidence—in-
cluding much collected in the field for this study—suggests that political
identities are fluid and contextual rather than static and fixed. One who is con-
sidered a Glaswegian in Edinburgh, for example, may well be a Lowlander in
Inverness, a Scot in London, British (or,perish the thought, English) in Spain,
and a European in Africa. Nor are these purely ascribed identities either; many
informants interviewed for this study reported alterations in their own feelings
of who they were, depending on their location at the time. Dormant identity-
related feelings may also be stimulated by perceptions or events, not least of
which include sporting matches pitting local or national teams against one
other.
15
Indeed, fieldwork for this project twice took place during the height of
the quadrennial World Cup football competition, the emotive power of which
should not be underestimated, deeply affecting as it does more than just the
“hooligans” who rioted across France during the 1998 tournament.
In part, recognition of this question invokes the debate between “primor-
dialists” (e.g., Geertz 1973; Smith 1992) and “constructivists” (e.g., Anderson
25Introduction
Green_Complete.qxd 1/25/07 10:52 AM Page 25