7 The Inherent Normativity of Functions in Biology and Technology

7.1 Introduction

Functions are attributed routinely in biology to organs, to traits, and to forms of behavior,
and in technology to artifacts. We say, for example, that the function of a particular pump
is to make the water of a central heating system circulate through the pipes and radiators,
and that the function of a particular heart is to circulate the organism’s blood through
its arteries and veins. Apart from these two core domains of functional talk, functions
are attributed in the social sciences to social traits such as marriage systems, religion,
and the like, although much less generally and less enthusiastically. Such social traits are
partly similar to biological traits, in that some have emerged as historical accidents, and
partly similar to artifacts, in that some may have been designed, either for the function
that is attributed to them or for some other function that they may or may not actually
perform.
Despite the apparent centrality of the notion of “function,” there is little consensus,
in the philosophy of the sciences and of technology, on how to understand the concept,
the sort of work it does, and the conditions governing the legitimate attribution of function.
My aim in this chapter is not to settle the matter as to what is the correct theory of
function, or which theory is to be preferred to which on what grounds, or whether a unifi ed
theory of function is possible at all. My aim in this chapter is to investigate the role
that is played, in the debate on what is and what is not an or the adequate theory of
function, by the claim that “function” is an inherently normative concept. It is widely held
that functions are normative in the sense that an item can have a function but at the
same time be physically incapable of performing that function. In such a case, we speak
of malfunction. A pump that has broken down is not able to circulate the water through
the central heating system, but most people would agree its function is still to do so. This
is expressed by saying that, although the device does not pump, it is supposed to pump
or ought to pump. Similarly many people would say that a heart that (momentarily)
fails to circulate the blood still has the function to do so and accordingly ought to do so.
Another aspect of the normative character of function is that we distinguish between

Maarten Franssen