Delahunty and Garvey
ones: dog represents all dogs, not just Lassie or Snoopy.
Categories need discussion. Let’s imagine that we are in a context in
which we are talking about technical matters and that we are expected to be
technically correct. A simple example of such a context might be a discus-
sion about plane figures in a geometry course. In such a context, when we
use the word square, we mean “a plane figure having four equal sides and
four right angles” (WNWD p. 1381), no more and no less. The elements
of the definition, “plane figure,” “four equal sides,” and “four right angles,”
are all necessary to define square and together they are sufficient for its
definition. When scholars try to define technical concepts they generally
try to define them in terms of necessary and sufficient conditions. If they
succeed, then, in principle it is possible to decide for any item whether it is
a representative of that category or not. Given our definition of square, we
can decide for anything whether it is a square or not. The world, however, is
not always as rigid as a geometry class.
Imagine now that we have been rescued from the math discussion, and
we go to a birthday party where there is a flat layer cake cut into squares.
If you noticed that the pieces did not meet the mathematical definition,
could you reasonably object that the pieces are not really squares? Anyone
objecting on those grounds wouldn’t deserve any cake. As far as we know,
there is no English word for the almost square pieces that a flat cake is cut
into, so until someone invents such a word and it is widely accepted, we
can use square and our audience will accommodate us. These kinds of ac-
commodations lead to rampant polysemy in much of the vocabulary. So it
is important to remember that polysemy, accommodation, and context are
inextricably intertwined.
If words and categories were all defined in necessary and sufficient terms,
then categories could be kept clearly distinct, as squares and triangles are in
geometry. But if we can bend these definitions, or if we cannot provide nec-
essary and sufficient definitions for categories, then the boundaries between
categories may get quite fuzzy. In fact, many natural categories are like this.
Where exactly does red end and orange begin? Where do animals end and
plants begin? We are unlikely to get unanimous agreement on the answers
to such questions. Indeed an article in the June 2008 issue of Scientific
American grapples with the problem of defining “species” (Zimmer 2008).
Nonetheless, in ordinary, non-technical company, we cut each other some
slack by not expecting words always to be used with technical rigidity. We
can also indicate when we are using words imprecisely by using hedges,
such as like or sort of; or we can indicate that we are being technically cor-
rect by including expressions such as technically—Technically, a phoneme is a