Biological Approaches to Mental Representation 557it lacks the disposition to perform. CR-functions are not normative, because an
item cannot have a CR-function that it lacks the disposition to perform, and so
CR-functions do not underwrite talk of malfunctioning.
It has, as I say, become a near commonplace to think that there is a division of
explanatory labor between SE-functions and CR-functions, and the usual under-
standing of this division seems sensible on the surface. The usual understanding
is that SE- functions explain how items with functions evolved, since SE- function
ascriptions tell us what items were selected for (i.e., that which, after selection,
becomes their function), whereas CR- functions explain how complex organic sys-
tems operate, since they tell us what items are disposed to contribute to the
operation of the systems in which they are embedded. On this view, SE- functions
play an important role in evolutionary explanations, whereas CR-functions play
an important role in operational explanations in physiological biology.
This seems sensible on the surface but there are two main problems with it. One
is that, although it is true that SE- function ascriptions make evolutionary claims,
this does not amount to an important explanatory role for them. On this view,
all the SE-function ascriptions are doing is summarizing evolutionary claims that
are reached independently, and such claims could easily be made in other terms
(instead of saying thatthasxas its function, we could say thattwas selected for
x-ing). On this view, the notion of an SE-function is one that is easily eliminated,
so if SE-functions play an important explanatory role in evolutionary biology, this
is not it.
The second problem is that it is physiological biology, not evolutionary biology,
that most obviously makes use of a normative notion of function. Physiology is the
study of the functional organization of living systems and physiology is steeped in
talk of organic systems’ functioning properly or malfunctioning, of organic systems’
functioning normally or abnormally, of functional impairment, dysfunction, and so
on and so forth. Anyone who doubts this should just take a look, since it can be
seen at a glance at any physiology, or for that matter neurophysiology, text. This
normative notion is not the notion of a CR-function. It is, according to many, the
notion of an SE- function. Therefore, there must be something wrong with this
standard account of the division of labor. Unless biologists are terribly confused,
normative functions belong in operational explanations of how organic systems
work, or fail to work.
What are they doing there? The explanatory role of normative functions in
physiological biology is in part the role attributed to CR-functions by Robert
Cummins, but it adds an element of idealization. This is how Cummins puts it:
The biologically significant capacities of an entire organism are ex-
plained by analyzing the organism into a number of “systems” — the
circulatory system, the digestive system, the nervous system, etc., —
each of which has its characteristic capacities. These capacities are in
turn analyzed into capacities of component organs and structures. Ide-
ally, this strategy is pressed until the analyzing capacities are amenable