20 Mark Perlman
basis of assigning function. Although it is tempting to say that the function of an artifact
comes from what its maker meant for it to do, there are many instances where this is not
the case. Thus, 3) DON’T make designer’s intentions essential to artifact function. Finally,
I use an analogy with theories of knowledge and epistemological justifi cation to put the
teleofunction theories in perspective. The result of this is fi rst a warning, 4) DON’T let
theories of teleology spiral out of control into defi nitional oblivion, and then a more sub-
stantive positive recommendation of where we should go in thinking of teleofunctions—
Pragmatic Teleo-Pluralism.
2.1 DON’T Make a Hard Line between Natural Functions and Artifact Functions
It was often thought to be relatively easy to identify the functions of human-produced
artifacts—just look at the intentions of the designers of the objects. The designers built
the objects for some purpose, so the objects’ purpose rests in what they were intended to
do (even if they do not often or ever do it). It was thought that the natural objects, for
which naturalism deprives us of the Creator’s intentions, are the ones for which it is diffi -
cult to clearly identify teleofunctions. However, this dichotomy assumes a hard line
between natural objects and artifacts, and this line has been eroded. It might be thought
that modern technology is the culprit, but in fact humans have been responsible for creat-
ing “natural” objects for thousands of years.
Consider the shape of the landscape. Fly over the Midwest of the United States and
you’ll see geometrical shapes everywhere—farm fi elds cut the plains into squares, rect-
angles, and circles. The whole state of Iowa is almost one giant cornfi eld. This is obviously
not natural—humans have manipulated the environment for their own purposes. Consider
also the elaborately manicured gardens of most of the royal palaces of Europe. From
Windsor Castle to Versailles to the gardens of Hannover to the gardens of Schloss Schön-
brunn in Vienna, plants are meticulously molded to fi t human designs (see fi gure 2.2).
Not only gardens but individual plants can be shaped by human design—Japanese
bonsai trees are a prime example. These plants have parts with functions, but given the
degree of human intervention in their development, they seem as much artifacts as natural.
Consider also the vast varieties of roses—www.everyrose.com lists more than 7,250 variet-
ies of roses, with names from Aafje Heynis and Admired Miranda to Zephirine Drouhin
and Zwemania in colors from red, yellow, pink, and white to coral, peach, deep burgundy,
and black. Doubtless some of these are close to what could be considered “natural,” but
the vast majority has been bred by humans for specifi c color or other characteristics.
Humans have also had a hand in steering the evolution of many animal species. In fact
all domesticated animals are designed by humans to a large degree. Dairy cows are bred
to produce vastly more milk than they would need in any natural setting—so much so
their health can be endangered without regular milking by people, for no calf could drink