How Biological, Cultural, and Intended Functions Combine 53
attractive classifi cation of functions into artifact functions and biological functions is
indeed a good one.
4.2 Etiological Theories and the Current Classifi cation of Functions
First of all, let us be more specifi c about the standpoint assumed here. Our perspective
is the one adopted by etiological theories of function since the time of Larry Wright’s
seminal article (1973). According to this perspective, to have a function is to have a prop-
erty of a quite peculiar nature, a property that can serve to ground both etiological expla-
nations and normative claims.^3 Thus, an etiological theory of biological functions is meant
to elucidate what a function is in order to explicate two things: 1) why assertions such as
“the function of the heart is to pump blood” may offer an explanation for the present
existence of hearts (their etiology) and 2) how normative statements such as “this heart is
malfunctioning” make sense. Thereby advocates of etiological theories take a realist stance
towards biological functions. In contrast both with Hempel’s and with Cummins’ positions,
they defend that functions are genuine properties that cannot be dispensed with in scientifi c
theories since they differ substantially from non functional properties.^4 The intentionalist
theory of artifact functions can also be seen as an etiological theory of function since the
effect for which an artifact has been invented or a feature has been designed—its function
according to such a theory—explains the etiology of the functional item (why it exists)
and fi xes a norm. In fact, the artifact (or the feature) envisaged exists because it has been
made in order to produce that particular effect, and tokens of this artifact type that cannot
produce such an effect are judged defective. Nevertheless, it is doubtful whether the
intentionalist theory entails the same realist stance as the selectionist theory. However, it
is not necessary to clarify this point for the discussion that follows. So, let us leave it at
that.
As we argue elsewhere, an in-depth analysis shows that the nature of a function hinges
much more on whether an objective selection mechanism has had a role in establishing
or maintaining it, rather than on whether there has been an intentional element involved
at some stage or another.^5 As a matter of fact, functions that are supposed to be simply
determined by somebody’s intentions (intended functions), and functions that are supposed
to depend on some objective mechanism of selection (selected functions) turn out to be
quite different regardless of whether intentional elements have been involved in the selec-
tion process.^6 Roughly speaking, according to classical etiological defi nitions, “X has
selected function F” will mean X is there because previous Xs have been selected for
having done F, and “X has intended function F” will mean that X has been planned or
produced because someone thought it would do F.^7 Now not only do these two defi nitions
look dissimilar but the sorts of properties they capture are quite different. Let us consider
this in more detail.