18 Mark Perlman
paper “Functional Analysis” broke new ground in giving a naturalistic and ahistorical
analysis of functions based on causal role in a system, an approach that had little trace of
the metaphysical, and proposed an account acceptable to science. The historical, or etio-
logical, side to naturalizing teleology was brought to prominence with Ruth Garrett Mil-
likan’s landmark 1984 book, Language, Thought, and Other Biological Categories
(hereafter abbreviated as “LTOBC”), and subsequent papers (1986, 1989a, 1989b, 1989c,
1990, 1993, 2002). Following her book, there was a fl ood of different accounts of func-
tions, how they would or would not serve as the basis of this and that, and the functions
literature took off.
We now have various streams or families of views on teleology, with reductionist theo-
ries dividing into systematic (causal-role, Cummins-style) functions on one end, and
various historical/etiological views on the other. The historical versions further divide into
those that focus on evolution by natural selection in the distant past (Millikan [ibid.];
Neander 1991a, 1991b; Papineau 1987; Griffi ths 1993; Allen and Bekoff 1995) and those
looking only to goal-contribution in the more recent past (Nagel 1977; Woodfi eld 1976;
Nissen 1997; Boorse 1976, 1977, 2002; Godfrey-Smith 1994). We can also look to the
future for a basis of functions—functions as propensities (Bigelow and Pargetter 1987).
Then there are the nonreductionist views of functions, from Emergentism (Bedau 1990,
1991, 1992a, 1992b; Cameron 2003) to Conventionalism (Searle 1998). In my 2004 article
“The Modern Philosophical Resurrection of Teleology,” I present the following taxonomy
diagram of theories of function (see fi gure 2.1).
The terrain is getting thick with competing views of teleological function. In one sense
this is a good thing—we’re reexamining an area that seemed virtually closed off from
philosophical consideration only half a century ago. From the viewpoint of theory of
science, it is unclear whether we are in what Kuhn called the chaotic “Pre-Paradigm”
phase, or whether we have competing paradigms. Or perhaps the paradigm is naturalistic
reductionism, and the proliferation of views is just the philosophical side of “normal
science”.
Things become even more complicated when we look to the issue of functions of arti-
facts. It may have appeared that the functions of artifacts were easily explainable by refer-
ence to the intentions of the makers of the artifacts. But we now see diffi culties with such
a simplistic account, and philosophers are making use of the developments in the teleologi-
cal view of organisms to answer questions about artifact functions.
However, rather than pledge my allegiance to one or the other camp in the debate, I
want to look at the debate itself, and see if perhaps some of the confl ict between theories
is misplaced or misguided. Among these various views, the counterexample game is now
going strong—clever exceptions lead to revisions, extensions, exceptions, or rejections of
the functional account. Perhaps the problem lies in the mission of seeking “the function”
in the fi rst place. I argue that many of these theories have something right about them—the
mistake is thinking that we can use only one. By pursuing a more multifaceted approach