42
INTRODUCTION TO JEAN BURIDAN’S LOGIC
6.6 Determinate and Confused Supposition
A term is said to have determinate supposition in a sentence when a
sufficient condition for the truth of the sentence is that it be true for some de-
terminate singular falling under the common term (TS 3.5.1), where ‘some’
here means for at least one such singular (TS 3.5.3). Buridan gives two
conditions which must be met for a term to have determinate supposition
in a given sentence: rather, they are conditions which apply in virtue of in-
ferential connections which obtain between the original sentence and related
sentences:
(1) From any given singular falling under the common term the sen-
tence with the common term follows, all else remaining unchanged
(TS 3.5.5).
(2) All of the singulars can be inferred disjunctively in a disjunctive
sentence (TS 3.5.6).
Roughly, a common term will have determinate supposition if it is in the
scope of a particular quantifier which is not in the scope of another logical
sign. Thus the term ‘man’ in “Some man is not a sexist” has determinate
supposition, because (i ) “Socrates is not a sexist; therefore, some man is
not a sexist” is an acceptable consequence; (ii ) “Some man is not a sexist;
therefore, Socrates is not a sexist or Plato is not a sexist or Aristotle is
not a sexist et sic de singulis” is an acceptable consequence. The condi-
tions for determinate supposition, then, roughly correspond to existential
generalization and instantiation.
A term in determinate supposition is used to talk about at least
one of the things it signifies. To what, then, does it refer? The answer is
clear; ‘some man’ in such sentences refers to some man. Geach has noted
a problem here:
67
if ‘supposition’ is a relation of reference, then what does
‘some man’ in the sentence “Some man is ϕ” refer to when the sentence is
false? If ‘some man’ refers to some man, then we may ask which man. But
the sentence is false, and so it cannot refer to those men who are in fact
ϕ, for there are none. Perhaps it refers to every man, being false in each
particular case? But then there would be no difference between ‘some’ and
‘every,’ since ‘every man’ refers to every man.
68
Thus the very idea that a
67
In Geach [1962] 6–7 the problem is discussed at some length; I can treat it only briefly
here.
68
It is only fair to point out that existential quantification may be in no better shape:
the variables bound by a quantifier range over the entire domain, so that universal
and existential quantifiers equally ‘refer’ to the entire domain. The mediæval and
modern cases only appear different; the modern case seems simpler because the work
of fixing the reference of a term is embodied in the interpretation-function, after which
c Peter King, from Jean Buridan’s Logic (Dordrecht: D. Reidel 1985) 3–82.
INTRODUCTION TO JEAN BURIDAN’S LOGIC
43
term can refer to only part of its extension is incoherent.
69
Geach himself (unwittingly) offers a solution ([1962] 7):
Suppose Smith says, as it happens truly: “Some man has been on
top of Mount Everest.” If we now ask Smith “Which man?” we
may mean “Which man has been on top of Mount Everest?” or
“Which man were you, Smith, referring to?” Either question is in
order. . .
Without realizing it, Geach is pointing out that ‘some’ can have a referential
or an attributive use.
70
In its referential use, ‘some’ picks out a particular
man or men, so that we may immediately assess the sentence’s truth-value.
In its attributive use, though, ‘some’ commits us to the sentence’s truth-
value, so that we may determine the reference (extension) of the term. The
referential use roughly corresponds to ‘some (i. e. this or these)’ and the
attributive use to ‘some (i. e. some-or-other).’
To return to Buridan: a term used referentially has determinate sup-
position, and a term used attributively has non-distributive confused sup-
position. The particular sign of quantity ‘some’ normally has determinate
supposition, that is , is used referentially; the inference-rules given above
guarantee this. In TS 3.5.1 Buridan says that a term has determinate sup-
position when it is true for some determinate supposit, that is, referentially.
The very point of the first rule is to insist on the referential reading. Further
support is found in the uncommon idiom for negatives, which is equivalent
to branching particular quantification, each taken referentially though not
necessarily for the same item(s). But ‘some’ in certain sentential contexts
can, in combination with other logical terms, have non-distributive confused
supposition, that is, be used attributively. Such uses will be discussed in the
next section. The rules of supposition are primary; it is a mistake to think
that a term always has the same reference, no matter what the sentential
context. That is one of the central points of supposition theory, noted in
Section 6.1.
Hence ‘some man’ in the false sentence “Some man is ϕ” refers to
some man, i. e. to some particular man. The sentence is false because that
determining truth-value is simple. As we shall see, this is but to embrace one possible
solution. Mediæval logic will take another path.
69
This is Geach’s central claim, and he argues against the Doctrine of Distribution by
means of such puzzles. But Geach is more abusive than conclusive; his objections, as
I shall argue, are ill-founded.
70
This distinction was first noted in Donnellan [1966], who applied it only to definite
descriptions; he was able to do so in virtue of the fact that definite descriptions are
disguised existential quantification - which have an attributive and a referential use.
c Peter King, from Jean Buridan’s Logic (Dordrecht: D. Reidel 1985) 3–82.