84 Pieter E. Vermaas
accounted for in current theories of biological functions—etiological theories, in particular.
Its disadvantages are, fi rst, that it is an epistemic theory about agents ascribing functions,
thus violating the general intuition that biological functions are relations that biological
items have relative to context, and, second, that by this theory agents ascribe these func-
tions relative to goal-directed patterns intentionally identifi ed by other agents, thus intro-
ducing (teleological) mental states into the analysis of biological functional descriptions.
The fi rst disadvantage can be overcome by constructing the equally explorative ontological
counterpart of the generalized ICE theory.
The unifi ed ontological ICE-function theory
Item x has the capacity to φ as a function relative to a goal-directed pattern p for x,
iff:
- x has the capacity to φ in the execution of p;
- if this execution of p leads successfully to its goals, this success is due in part to
x’s capacity to φ; and - the agents d who designated p have intentionally identifi ed x for having the capacity
to φ in p and for contributing by this capacity to the success of p, and have inten-
tionally communicated p to other agents l.
In this ontological version, biological functions are again relations that biological items
have relative to context. Moreover, biological functions are in this theory epistemically
objective—judgments about functions depend on the three conditions that concern facts
whose truth is independent of the points of view, attitudes, or feelings about these facts
of the makers or hearers of the judgments. Yet the second disadvantage cannot be over-
come. Functions, including the biological ones, are in the unifi ed ontological ICE theory
ontologically subjective: the three conditions concern facts that refer to the mental states
of the designators since they have singled out the patterns relative to which items have
functions.
This reference to the mental states of the designators can to some extent be suppressed
by noting that the third condition in the unifi ed ontological ICE theory becomes redundant
when the communication between designators and laypersons is considered. If this theory
is applied to cases in which designators provide information to laypersons, it seems spuri-
ous to hold that the designators fi rst tell laypersons that the pattern can be singled out, that
x has a role in the effectiveness of this pattern, and second, add explicitly that they, as des-
ignators, have made all these discoveries. The third condition in the unifi ed ontological ICE
theory can in this case be taken as redundant, which simplifi es this theory as follows:
The unifi ed ontological ICE theory in designator-layperson communication
Item x has the capacity to φ as a function relative to a goal-directed pattern p for x,
iff:
- x has the capacity to φ in the execution of p; and