changes in russian nationalist public opinion 2013–14tion of Crimea in March 2014. Putin appears to have tapped
into these sentiments, rather than engendering them through his
actions and the Kremlin’s public relations campaigns. Indeed,
while many Russians internalised Kremlin- supported narratives
about dangers of fascism, of Russophobia and of Western aggres-
sion emanating from Ukraine, they generally did not support
Russia interfering in the foreign policy choices of Ukraine or other
neighbouring countries.
In conclusion, we may venture to say that the patterns of public
opinion found in Russia in 2013 and 2014 are largely consistent
with ‘diversionary theories of war’, according to which leaders
recognise that a short victorious war can bolster their short- run
popular support and thus undertake such action in order to effect
it (see Levy 1989). That said, the Russian evidence also suggests
several interesting avenues for developing this line of theory in
future research.
One such question is how long this effect is likely to last – a
point central to its expected utility for state leaders. In the United
States, for example, analysts with the Gallup polling organisation
have estimated that, in the twentieth century, surges of public
support for incumbent presidents related to foreign affairs and
military action lasted an average of ten weeks (Hugick and Engle
2003). This is far shorter than the effects we have observed in
Russia. One plausible explanation is that the duration of rallying
effects may vary across political systems, with state media control,
repression of opponents and agenda- setting power enabling them
to last longer. Another, however, draws on comparative research
on ‘rallying’ effects in public opinion related to conflict, some of
which has found that conflict ties presidential popularity closely
to nationalist identity, although this typically fades as other issues
eventually come to the fore (Kam and Ramos 2008). This may
imply that countries with strong and enduring nationalist senti-
ment prior to the war are more conducive to longer- lasting ral-
lying effects, even though these effects are likely to fade as other
issues again become salient. We may also speculate that regime
type conditions the kind of rallying seen in the event of a short,
victorious war: in pluralistic polities based on the rule of law and
media freedom, the populace may be more likely to rally first