Foundations Shaken but Not Stirred 103concepts as an open invitation for ‘tidying up’. Unlike the ‘reportive’ and ‘stipulative’
approaches of essential contestability, reconstruction favours explicative definitions,
which ‘can be appraised as good and bad in terms of their suitability for scientific
communication’ (Oppenheim 1981: 179). Constructing such explicative accounts
parallels constructing good scientific theories for Oppenheim, in terms of accuracy,
simplicity, and fruitfulness. A good explicative account reveals the inner structure of
concepts and facilitates generalizations. Although this sounds something like a quali-
fied return to a positivist perspective, Oppenheim is insistent that he rejects the older
forms of positivism and behaviouralism. Yet, to reject the old fashioned positivism
is not to abandon empirical theory. Oppenheim however does want to separate the
realms of values and facts (qua value non-cognitivism, as opposed to the value cog-
nitivism of essential contestability). Justificationisdistinct from description. Moral
beliefs are not descriptive. This does not mean that political study is value free, rather
that the theorist has to be conscious of values as distinct elements which are not
subject to truth or falsity claims.
Thus Oppenheim sees endless conceptual analysis of ordinary language in all its
general confusion and vagueness, and commitment to value cognitivism, as a false and
damaging pathway.^17 He also tries to return the whole argument back to a much more
positivistically inclined theory which actually sees language as something to control,
tidy up, and use with greater technical precision. Relativism should be avoided and
objectivity sought. There is, though, a lurking sense in Oppenheim of a neutral
metatheory—a form of subtle technocratic ideal that is to be imposed on political
language. All, or most of the old problems of logical positivism and behaviouralism
rise again here.
The final negative reading, like Oppenheim, sees essential contestability again as
far too much inclined to extreme relativism and incommensurability. However, in
this case, the alternative sought is not value non-cognitivism, but a variation of rig-
orous value cognitivism, which returns the discussion to the perceived older tasks
of normative classical political theory. In this case, the argument modifies essential
contestability. John Gray, for example, sought, for a moment in his diverse theoretical
career, to revise essential contestability in order to distance it from ‘sceptical, relativ-
ist, historicist and conventionalist traditions’. This would facilitate the possibility of
future ‘conclusive rational resolution’ to political theory conceptual debates. Essential
contestability allowed for the possibility that, although certain concepts were deeply
contested, this did not mean that there could not be ‘good reasons’ and some kind
of reasonable resolution to philosophical problems. This weakens the whole classical
Wittgensteinian essential contestability account. It does, however, leave open the pos-
sibility, once again, as Gray is keen to urge, for ‘perennial political problems’. He
comments that, ‘the revised essential contestability thesis endorses a classical concep-
tion of political philosophy as an intellectual activity capable of yielding determinate
results, and, so, of assisting reflective agents in their search for a good society’ (Gray
1977: 346; see also Gray 1978: 394–5). This allows the theorist, such as Gray, to endorse
much of the analytical philosophical dimension, and then to put it to work to defend
a singular account of concepts such as justice and liberty. It also still preserves the