Foundations Shaken but Not Stirred 89rigid categories of the analytic and synthetic—combined with verification—ordinary
language philosophy viewed meaning in terms of common linguistic usage. The
diverse varieties of usage of words could not be simply swept aside with the cat-
egory of verification (although it is clear thatnotall logical positivists wanted this).
The emphasis therefore shifted from the precise meaningful definition of words (and
their contextualization as empirical or analytic) towards elucidating concepts in their
diverse rich uses.
Underpinning the above conception is a particular view of language. Language is
seen as an activity. Words are deeds and tools. Typical of this approach is J. L. Austin’s
speech act theory (Austin 1962, 1971). Speech acts are what Austin termed ‘per-
formatives’. Thus, in many everyday utterances, we perform an actinspeaking. For
example, the performative, ‘I promise’, is not reporting or describing the practice
of promising. Rather, it is itself invoking the conventional practice of promising. It
is doing by saying.^7 People perform acts with words all the time in ordinary lan-
guage and yet the same sequence of words can perform different acts. The full sense
of one sequence of words can mean different things in different conventional con-
texts. Even non-verbal acts depend on conventions. The speech act thus depends
crucially on conventions in ordinary language usage.^8 A successful speech act takes
place when conventions subsist which standardize the use of words and listeners are
fully conscious of them. The gist of what Austin is claiming is that we are, first,
always in the midst of language, and second, that meaning relates to conventional
use within ordinary language. Third, there is an enormous variety of meanings con-
tained within ordinary usage. Therefore, the philosophical emphasis moves away from
defining and tidying up concepts, to the complex task of elucidating conventions
and uses of words in particular linguistic contexts. This did not mean, for Austin,
that ordinary language provided all the possible answers. As Austin noted, ‘ordin-
ary language isnotthe last word; in principle, it can everywhere be supplemented
and improved upon and superseded. Only remember, it is thefirstword’ (Austin
1971: 185).
However, one should not overdo the distinction between ordinary language philo-
sophy and logical positivism. Although ordinary language theory did concede, unlike
logical positivism, that normative questions about justice were legitimate to ask, it
was still at one with logical positivism in resisting the idea that philosophy could
provide any solutions to such normative issues. Ordinary language was still in agree-
ment with logical positivism that philosophy did not include anymodus operandi
for justification or prescription. All that Austin might have suggested is that if we
wanted to understand justice, we should look at the conventions. Political concepts
could not,per se, be recommended. Political theory was still a second-order activity,
distinct from direct normative claims. Thus, both schools of thought acknowledged
the ‘second’ and ‘first-order’ distinction. Philosophy did not generate any knowledge.
In addition, both philosophies were suspicious of metaphysics.^9 Ordinary language
theory had an almost conservatively proprietorial sense of the subtle conventions
embedded in ordinary language. Logical positivism was simply hostile. Some analytic
philosophers, like Strawson, in the 1970s, did try to modify the understanding of