6
INTRODUCTION TO JEAN BURIDAN’S LOGIC
‘resemblance’ by interpreting it pragmatically. Each inscription-token is
unique. Logical laws are stated for inscriptions that are similar (similes),
that is, sufficiently resemble one another in the relevant respects. But there
is no saying what respects are the relevant ones, or which degrees of re-
semblance are sufficient; these factors depend on the context, on our in-
terests and aim. Inscriptions treated as the same in a given context are
called “equiform,” but there is no such thing as equiformity tout court,
and so Buridan’s nominalism is not compromised. Logical principles are
therefore restricted to equiformity-classes of inscriptions.
In some cases
we may be forced to be very restrictive in our choice of equiformity-class:
The sentence printed below this one is false
The sentence printed below this one is false
2 + 2 = 17
The first two inscriptions cannot be equiform with respect to truth-value,
for the first is false and the second true (QM 5.1 fol. 26vb). Equally, when
discussing the Liar, for example, we have to carefully distinguish individual
inscription-tokens (Soph. 8: Hughes [1982] 8.4.3, 13.5–6, 15.8.2).
Buridan seems to have kept an open mind on how arbitrary an
equiformity-class may be. There may be an eventual limit, imposed by
the causal theory of concept-formation and the requirement that concepts
be natural likenesses of the things of which they are concepts, but Buri-
dan does not say so explicitly . For most cases a handy practical criterion
will serve to demarcate equiformity-classes; Buridan frequently uses the rule
that terms supposit and appellate in equiform sentences just as in the orig-
inal sentence (or part of a sentence: this will be important for the theory of
consequences): see TC 3.7.41.
Buridan’s pragmatic interpretation allows him to deal with classes
of possible inscriptions. In general the laws of logic are not stated for actual
sequences of inscriptions but rather specify what sequences of inscriptions
would be acceptable, if formed.
11
This is, of course, an aspect of the nor-
mative element of logic, but for nominalists the point is deeper: logical laws
cover not only actual inscriptions but possible inscriptions as well. There is
no other way to describe basic principles of logic. For instance, any sentence
can have a contradictory, but in actual fact such a contradictory may never
have been inscribed; how then can we say that a non-actual inscription must
be true if its actual contradictory is false? Buridan sensibly states his prin-
ciples for actual and possible inscriptions, and, in some cases (such as the
11
See TC 1.3.8 and TS 1.8.72, in which Buridan states how his logical laws apply to
the Mental correlate of inscriptions, and the discussion of consequences in Section 7.2
below.
c Peter King, from Jean Buridan’s Logic (Dordrecht: D. Reidel 1985) 3–82.
INTRODUCTION TO JEAN BURIDAN’S LOGIC
7
distinction of the possible and they possibly-true), they must be carefully
distinguished, shedding new light on philosophically unexplored terrain.
A final point. The term ‘equiform’ might be read as ‘having the same
(or similar) logical form,’ which is narrower than the usage I have given it.
To be sure, Buridan does sometimes use it in this way, but then only as a
special case of the more general sense: the relevant respects of similarity
in question are the series of syncategorematic terms. Unless specifically
noted, I shall follow Buridan in his practice of taking ‘equiform’ to mean any
similarity-class of inscriptions, where the respects in which the inscriptions
are similar are defined contextually.
3.3. Mental as an Ideal Language
Mental Language is of the utmost importance for Buridan’s logic.
For while Mental is a natural language, in a way in which Spoken or Writ-
ten are merely conventional, we should not understand this the way today
“natural languages” such as English or French contrast with symbolic lan-
guage in rigor; rather, Mental is a natural language perspicuous in rigor.
The basic claim, common to Buridan and many other fourteenth-century
logicians, is that Mental is a canonical language, an ideal or logically perfect
language.
12
This claim involved five theses:
[1] Mental is a universal language.
[2] Mental is adequate in expressive power.
[3] Mental is disambiguated.
[4] Mental is nonredundant.
[5] Mental sentences display their logical form.
Each thesis calls for further comment.
Ad [1]. The universality of Mental is a matter of its structure, not its
content. That is, we do not necessarily all have the same stock of concepts;
I may completely lack the concept of lion, which you possess, due to our
different past interaction with the world. But the structure is the same for
all, meaning (roughly) that we all have similar mental abilities: we can all
combine simple concepts into complex concepts, for example. Moreover if
two people each have the same term of Mental, then their concepts differ
only numerically: your concept of a lion and mine may have been acquired
through the experience of different lions, but the concepts are equiform,
that is, for most practical purposes they may be regarded as the same.
12
This point was first made with respect to Ockham, not Buridan, in Trentman [1970],
but it holds generally for fourteenth-century logicians. Trentman does not consider
[1]–[5], but they seem the necessary requirements for a language to be counted as
‘ideal.’
c Peter King, from Jean Buridan’s Logic (Dordrecht: D. Reidel 1985) 3–82.