Philosophy of the Performing Arts

54 performance and the classical paradigm
But types, as we have seen, are entities of a very different stripe from the
performances that are claimed to be their tokens. They are, we are told, abstract
entities that exist eternally, lack any determinate spatial location, and are argu-
ably changeless in their intrinsic properties. If this is the case, we may ask what
sorts of properties types can possess. More crucially, if performable works are
types, we may wonder whether they can possess the same sorts of appreci-
able properties as the performances that are their tokens. If we ask what we
appreciate in performances of performable works, what immediately come
to mind are audible properties such as the ones cited by Day in his account of
the Second Symphony – the “varied sonorities,” and the expressive properties
that we hear in those sonorities. These are properties of particular experi-
enced events that take place at particular times. But how could such prop-
erties be possessed by something abstract and eternal? Indeed, if we accept
some accounts of the nature of types,^4 they lack internal structure, and thus,
it would seem, cannot possess even the structural properties to which Day
refers. And lacking such structural properties, they cannot possess expressive
properties that depend upon internal relations such as “recapitulation.”^5
It might be thought that there is really no problem here. Consider our initial
linguistic example of types and their tokens. If I write the word “sheep,” the
word-token that I produce contains five letters, two of which are e s. But surely
the word-type “sheep” possesses exactly the same properties. And, indeed,
these two facts seem to be intimately related. It is because the word-type “sheep”
has these properties that its tokens have these properties, we might say: the
type prescribes that its tokens possess certain of the properties that it possesses
itself. Such a view is proposed by Richard Wollheim in his elucidation of the
idea – discussed in Chapter 2 – that multiple works, including performable
works, are types (1980, 74–84). Types, for Wollheim, are generic entities that
can have other entities as elements “falling under” them. And generic entities, he
maintains, may share properties with their elements. Sometimes this is merely
a contingent matter. For example, the class of average-sized things may itself be
average-sized. But sometimes a generic entity possesses a property because its
elements possess that property, or vice versa. Where this is the case, Wollheim
suggests, we can say that the property is transmitted between the generic entity
and its elements. In the case of types, he claims, all, and only, properties of a
specific kind are transmitted between a type and its tokens: namely, those shar-
eable properties that its tokens must have in order to be tokens of that type. For
example, the properties of being rectangular and bearing a particular design in
red, white, and blue are transmitted between the Union Jack and its tokens.^6
If types and their tokens transmit properties to one another in this way,
then we can easily explain how both the word-type “sheep” and particular
tokens of that word-type can possess the property of containing five letters.
And, similarly, there is no problem explaining how the performable work