The Inherent Normativity of Functions in Biology and Technology 117
run better than they do? Once one has started off on this course, one may quickly fi nd any
organism a sorry failure.
For artifacts, then, a point of reference can be set independently of all actual items, such
that all items can perform worse than the norm. The idea of an ideal knife, an ideal pump,
or an ideal thermometer makes sense, whereas the idea of an ideal liver, an ideal rabbit,
or an ideal mating dance seems pointless.^18
In the present context, the important question concerning a scale of relative perfor-
mances of function is: what makes such a scale normative? Carrying the analysis of section
7.4 one step further, I suggest that normativity enters only when the notion of justifi cation
is added. We can objectively classify the performances of items on a scale and compare
them, and defi ne a norm either by establishing what the average performance, in the right
circumstances, is among members of the population (see for a similar suggestion Krohs
2004: 100) or by introducing an ideal performance. None of this is yet normative, however,
except in the very meager sense that we can base our expectations concerning the future
behavior of items on the outcome of this exercise. Such a measured norm becomes norma-
tive only when we make it a norm governing our actions. Once we have ordered a collec-
tion of functional items into relatively poor ones and relatively good ones, ceteris paribus,
we are justifi ed in preferring, or ought even to prefer, a better item to a worse one in using
it for its “normal” purpose. And once we have ordered existing items with respect to an
ideal and we have seen that they are all far removed from the ideal, we are justifi ed in
looking for a new specimen, or even ought to look for a new specimen. (The ceteris paribus
clause is there because, for more detailed purposes, the criteria can be reshuffl ed; there
are a lot of things you may want to use a knife for that are better done with a knife far
removed from the ideal knife than with a knife that comes close to it.) Again this makes
no sense for biological items in general, only for those that are available for being used
by us for a purpose of ours.
7.7 The Normativity of Function as the Presence of Reasons to Use
The interpretation that I give in section 7.4 of normative statements seemingly supported
by functions refers to the justifi cation of beliefs. In section 7.6 I argue that a norm or
standard is normative only insofar as it governs human action. Therewith we have reached
the native country of normativity, where life is all about the justifi cation of beliefs and of
courses of action. Only by traveling there can we come to understand how the notions
of “functioning,” “functioning poorly,” and “malfunctioning” can be seen as normative
notions. There, “ought” refers exclusively to what people ought to believe, ought to desire
or aim for, and ought to do. The notion that a person ought to believe or to do something
is currently most often explicated in terms of the reasons this person has for believing or
desiring or doing something. By saying that someone ought to do Z, we mean that this