M ICH EL FOUCAULT
607
extent it is equivalent to a description. When one says "Aristotle" one em–
ploys a word which is equivalent to a precise description, like' 'the author of
the
A nalytics
,,
or "the founder of ontology. " But one cannot limit oneself to
this. A proper name has a complicated meaning; when one discovers that
Rimbaud did not write "La Chasse Spirituelle" one cannot claim that this
proper name or that of the author have changed meaning . The proper name
and the author's name are situated between the two poles of description and
designation; they are connected to what they name, but neither completely in
the mode of designation, nor completely in the mode of description. How–
ever-and it is here that the particular difficulties of the author's name arise–
the connection of the proper name with the individual named, and the con–
nection of the author's name with what it names, are not isomorphic . Here are
some of the difficulties .
If! observe, for example, that Pierre Dupont does not have blue eyes, or
was not born in Paris, or is not a doctor, nonetheless, this name, Pierre Du–
pont, will continue to refer to the same person. But the problems posed by the
name of an author are more complex: if I discover that Shakespeare was not
born in the house which one visits today, this obviously is not going to alter
the function of the author's name; but if someone proves that Shakespeare
did not write the sonnets which pass as his, that is a change of another kind,
one that affects the function of the author's name. And, if someone proved
that Shakespeare had written Bacon's
Novum Organum
simply because the
same author wrote the works of both Shakespeare and Bacon, that, too, is a
change which completely modifies the function of the author's name. The
author 's name, therefore, is not exactly a proper name like any other.
Many other facts draw our attention to the paradoxical singularity of the
author 's name . One cannot say that Pierre Dupont does not exist in the same
way that one can say that Homer or Hermes Trimestigus did not exist: in one
case, we mean that no one bears the name of Pierre Dupont; in the other that
several authors have been grouped under a single name, but that the real
author does not have any of the traits attributed to the figure of Homer or
Hermes. Nor is it the same thing to say that the real name of X is not Pierre
Dupont butJacques Durand, and that Stendahl's name was Henri Beyle. One
could also question the meaning and function of a statement like "Bourbaki
is so and so," and "Victor Eremita, Climacus, Anticlimacus, Frater Taci–
turnus , Constantin Constantius are Kierkegaard."
Perhaps we owe these differences to the following fact: the author's
name is not simply a subject or predicate which can be replaced by a pronoun.
The author's name plays a certain role in any discourse: it assumes a classifi–
catory function; it allows us to group together a number of texts, to delimit
them,
to
exclude others , and
to
contrast them with others. The fact of having
