Realism and World Politics

the model. Application of theory then takes the form of assessing the fit between
the model and things in the world. The semantic conception can cover quite a
spectrum of philosophies of science.^44 Even within a traditional understanding of
explanatory, causal natural science, a strong case was made for changing the
understanding of theory towards the semantic conception.
With this move came a strengthened role for visual elements like pictures. The
received/syntactic view saw science as carried by linguisticrepresentation:

In the framework of Logical Empiricism, then, there can be no fundamental

role in science for non-linguistic entities like picture or diagrams. Such things

might, of course, play some part in how scientists actually learn or think about

particular theories, but unless their content is reduced to linguistic form, they

cannot appear in a philosophical analysis of the content or legitimacy of any

scientific claims to knowledge.^45

In the semantic view, in contrast,

the primary representational relationship is not the truth of a statement rela-

tive to the facts, or even the applicability of a predicate to an object, but

the similarity of a prototype to putative instances. This is not a relationship

between a linguistic and a nonlinguistic entity, but between two nonlinguistic

entities.^46

Models have no empirical content. Theories as such make no empirical claims; they
define predicates such as ‘is a pendulum’, ‘is a game of bluff’ or ‘is an instance of
securitisation’. They are used to make empirical claims about reality. Truth value
can only be attached to theoretical hypotheses applying the theory to an empirical
situation, basically taking the form ‘The Cuban missile crisis is similar to a game of
bluff’.^47
Where discussions in the philosophy of social science often gets polarised over
naturalism – whether to model social science on natural science or emphasise the
contrast – this updating of our picture of how natural science actually works
promises to get us beyond the dominant dichotomy in IR: Americans/mainstream
scholars take theory to be ultimately causal laws that explain, while Europeans
(notably including rather mainstream Europeans^48 ) and some dissident Americans
take theory to be roughly ‘interconnected sets of concepts that make sense out of
something’ – one excessively narrow, the other extremely wide. The model-
theoretic approach could get us beyond this (without everybody having to sign up
to demanding philosophical dogma with, for instance, critical/scientific realists, the
currently strongest ‘third way’). Processing theories of theories will clarify that
diverging IR approaches all contain explanations, but non-mainstream ones should
(robbed of the excuse of not being positivists) tighten conceptual sets, and main-
stream work would have to spell out their theories (no longer hide behind laws and
hypothesis testing).

Waltz’s theory of theory 73