11 Epilogue: the evidence of things not seen
Throughout these chapters I have repeatedly invoked the formula that works
of art present a subject matter as a focus for thought and emotional attitude,
distinctively fused to the imaginative exploration of material. This formula
proposes that works of art typically have representational, expressive, and
formal dimensions, all of which, both independently and in interaction, are
normal foci of attention in making and responding to a work. I have
attempted to outline debates about how original works might be made and
what their interest is, how works of art distinctively call for interpretation,
how they engage our emotions, how they explore the exercise of agency, and
how they enter into and comment on wider social developments.
What, then, is the status of this formula that undertakes to sum up the
dimensions of art and to lend some order to the debates? Is it a definition of
art? Does it specify conditions that are individually necessary and jointly
sufficient for anything being a work of art?
I do propose this formula as a definition, but not as a specification of
necessary and sufficient conditions. Instead this formula is proposed as a
specification ofcriteria, in Wittgenstein’s sense of that term, for calling
something art. Pain-behavior, for example, is a criterion for pain, according
to Wittgenstein. It is, first of all, inconceivable that in general pain should
have no relation to pain-behavior. Our grip on what pain is arises out of the
fact that pain-behaviors such as crying, wincing, withdrawing, and so on are
natural expressions of pain. Infants and small children produce a range of
pain-behaviors immediately in expression of pain, without intervening con-
ceptualization, and so, sometimes, do adults. That pain-behavior is a natural
expression of pain is part of the grammar of the conceptpain.
But pain-behavior is neither a sufficient nor a necessary condition for
pain. Pain-behavior can be feigned or simulated, present when the pain itself
is absent. Pain-behavior can be suppressed, absent when the pain itself is
284