present. There are a number of different kinds of pain-behavior, from the
most natural and immediate (screaming and withdrawing) to the somewhat
more controlled and conventionalized (saying“I have a headache”). Typically
some but not all kinds of pain-behavior are present in any single case.
Yet the connection between pain-behavior and pain is neither merely
conventional, learned, and artificial, nor is it altogether causal and unrelated
to consciousness and human expressiveness. It is a mistake to think of the
pain as perfectly inner (in either the private mind or the brain) and the
behavior as “merely outer” (either simply caused or merely expressed
according to artificial convention). When pain-behavior is present and pain
is absent, we require a special story about what is going on (“x is feigning
pain in order to solicit sympathy”;or“it’s a play”). Likewise when pain is
present and pain-behavior is absent (“x is ignoring his pain and focusing on
the task at hand”;or“x does not like making a show of himself”). A grasp of
the connections between multiple kinds of pain and multiple kinds of pain-
behavior, including the behaviors of feigning and suppressing it, is central to
knowing what pain is. It is part of the grammar of pain to know that in
certain circumstances this kind of wince counts as an expression of this kind
of pain, while in other circumstances this tightness of the face expresses
another kind of pain and this cry yet another.
The representational, formal, and expressive dimensions of art function
similarly as criteria for calling anything art. This is a conceptual claim that is
put forward in order to elucidate and organize our linguistic and conceptual
practice, in a situation in which we are confused by the varieties of artistic
practices, by the varieties of things people say about them, and by the
powerful but obscure character of our own responses. In the grip of these
confusions, we are likely to emphasize one criterion too much at the expense
of the others.^1 In some works one of the criteria may be much less clearly and
obviously fulfilled. Purely instrumental music and abstract painting lack
conspicuous and obvious representationality, even if it can be argued
(as I have argued) that works of these kinds symbolize and explore abstract
patterns of human action and of emotional responses to developing contours.
Some political art lacks a conspicuous formal or aesthetic dimensions (even if
it can be argued that varieties of absorbing form may be much more plastic
(^1) See my discussion, explicating the work of Stanley Cavell, of criterial claims as claims of
reason in Eldridge,Leading a Human Life, pp. 107–08.
Epilogue: the evidence of things not seen 285