4
INTRODUCTION TO JEAN BURIDAN’S LOGIC
Therefore, the things that are in speech are the marks (notae) of
the passions that are in the soul, and the things that are written
are [the marks] of those that are in speech. And just as letters are
not the same for all people, so the utterances are not the same. But
the first things of which these utterances are marks are passions of
the soul, the same for all people, and the things of which the latter
are likenesses (similitudines)—real things—are also the same [for all
people].
There are three distinct levels of language: Written, Spoken, and Mental.
Each is a fully developed language in its own right, with vocabulary, syntax,
formation-rules, and the like. These languages are hierarchically ordered;
elements of Written Language (inscriptions) are the “marks” of Spoken
Language, elements of Spoken Language (utterances) are the “marks” of
Mental Language. The elements of likenesses concepts, which are mental
particulars, “acts of the soul” (QM 5.9 fol. 33rb). The relationship of one
language to another is not holistic, but piecemeal. A particular inscription
will be related to a particular utterance, a particular utterance to a partic-
ular concept. Buridan calls the relations obtaining among these languages
signification. A written word immediately signifies a spoken expression; a
spoken word immediately signifies a concept of concepts of the mind; the
concepts of the mind signify those things of which they are they natural
likenesses. The written and the spoken term have an “ultimate” significa-
tion, namely what is conceived by the concept (TS 3.2.8, Soph. 1 Theorem
2). Clearly, a change in the signification of a term of one level will change
the (ultimate) signification of the lower-level terms signifying it. The re-
lation among the various levels, then, is rather like encoding.
7
Immediate
signification is conventional, but the signification of concepts is natural, and
indeed universal; concepts are “the same for all.”
3.2 Nominalist Semantics and Equiformity
Logic, for Buridan, will include Aristotelian psychology as a compo-
nent (QM 6.12 fol. 41vb): the constituent elements of Mental are concepts,
and Buridan holds substantive doctrines about (i ) the causal theory of con-
cept acquisition, and (ii ) the activity of the mind in combining concepts.
Logic, of course, is not simply psychology, for logic includes a normative
dimension.
The central theme dominating Buridan’s logic, and indeed his ap-
Boethius in Minio-Paluello [1965]; Boethius conflates certain distinctions present in
the original text: see Kretzmann [1974].
7
This suggestive analogy is made in Normore [1976] 13.
c Peter King, from Jean Buridan’s Logic (Dordrecht: D. Reidel 1985) 3–82.
INTRODUCTION TO JEAN BURIDAN’S LOGIC
5
proach to philosophy in general, is nominalism. Mediæval nominalism took
many different forms, but at root is the denial of universal or metaphysically
shared entities.
8
This has several forms in Buridan: he will deny the exis-
tence of universals; he will minimize the number of categories; he will reject
various ‘abstract entities,’ such as propositions. This last is important, and
will be discussed in more detail in Section 5.5; what is relevant to our pur-
poses here is Buridan’s attempt to create a nominalist semantics, a theory
of logic and language that is based on particular inscription, utterances, and
thoughts. The crucial notion in this enterprise is equiformity.
9
Buridan rejects any abstract notion of ‘proposition’ to serve as truth-
bearer and fundamental constituent of semantic analysis, and replaces it
with a theory of logic and language designed to apply to the individual
inscription.
10
Of course, all inscriptions are particular individuals, dis-
crete and different from one another. Buridan is thus faced with a familiar
dilemma: the laws of logic are general, and govern classes or types or sets of
inscriptions, not individual inscription-tokens; ‘class,’ ‘type,’ and ‘set’ are
abstract entities. Now whether mediæval nominalism ought to countenance
such entities is not clear: they are not obviously the sort of metaphysically
shared item it proscribes. But they do most of the same work, and they are
abstract, and suspiciously so; mediæval nominalists should look on them
with a jaundiced eye. But how is logic possible without recourse to such
abstract notions?
There are many familiar attempts to escape the horns of this dilem-
ma. Quine, for example, endorses sets, but argues that as abstract entities
go they are pretty well-behaved (so well-behaved that he insists on calling
himself a nominalist. Mediæval nominalism might take the same route,
although then it must account for the ontological status of such abstract
entities. Others have tried to give reductivist accounts of ‘set’ or ‘class’ or
‘type,’ suggesting that they are no more than reducible equivalence classes
based on resemblance or similarity. This is Buridan’s approach.
But according to the latter, is not ‘resemblance’ equally an ab-
stract entity, a universal of the sort proscribed? Buridan avoids reifying
8
Note that this formulation does not necessarily involve denying the existence of objects
not in space and time, or immaterial objects—God is an example; nor does it require
the denial of possible or future objects. What it does entail is that any such objects
must be particulars.
9
This useful term was coined by Hughes [1982] 5; we’ll use it to refer to expressions
that are sufficiently similar in relevant respects, although distinct by other criteria.
10
In the rest of Section 3.2 I shall, for simplicity’s sake, discuss only inscriptions; the
same points can and should be made for utterances and individual acts of thought.
c Peter King, from Jean Buridan’s Logic (Dordrecht: D. Reidel 1985) 3–82.