existence? There is an argument, inspired by remarks by Peter Geach (1969) and
Anthony Kenny (2005a), for the conclusion that we cannot understand Aquinas as
talking about specific existence. Consider the following passage by Kenny (2005a: 43–4):
Now let us ask whether, when it is said that God’s essence is existence, it is specific or individual
existence that is in question. The fact that the doctrine is supported by the phoenix argument
suggests that Aquinas had specific existence in mind; but as Peter Geach has shown, if
interpreted as referring to specific existence, the thesis is an absurdity. Statements of specific
existence can be paraphrased in terms of the quantifier: so if God is pure existence, then“God”
must be equivalent to“For some x, x... .”—a quantifier with a bound variable attached to no
predicate. So understood, the thesis reduces the divine name to an ill-formed formula.^69
This is a weak argument. First, if it were sound, a similar argument would suffice to
show that God cannot be identical with His goodness: statements about the goodness
of something can be understood as subject-predicate sentences in which the predicate
“is good”appears, as in“pleasure is good.”So if God is His goodness, then“God”
must be equivalent to“is good,”an unsaturated predicate. But pseudo-sentences like
“Is good loves us”and“I pray to is good”are ill-formed, nonsensical constructions.
So understood, the thesis that God is His goodness reduces the divine name to an ill-
formed formula.
Perhaps Kenny would also welcome this conclusion, but he shouldn’t. It’s a bad
argument too, but it is easier to see why it is a bad argument. Suppose there are
properties, such as the property of being good. Perhaps there is even more than one
property of being good: perhaps there is the property of being good in the way that
God can be good, and there is the property of being good in the way that a creaturely
thing can be good, and perhaps these properties are only analogously related. Set this
aside for now—let’s assume that“is good”can be used to ascribe a single property of
goodness to both God and creatures. That’s the function of the predicate“is good,”
and its semantic content is the property of being good. We can if we like introduce a
name for the property as well—we’ve already been referring to it with the term“the
property of being good”but let’s give it a proper name too. We’ll call it“Bob.”The
semantic content of this name is the same as the semantic content of the predicate
“is good.”But it in no way follows from this that the sentence“Bob is identical with
is good”is a well-formed sentence. Although the semantic content of“Bob”is identical
with the semantic content of“is good”—put less blandly, the thing denoted by“Bob”
is the property expressed by“is good”—the way in which that content is attached to its
different linguistic representations is correlated directly with the different ways in
which it can occur in propositions about it.^70
(^69) See also Klima (2013: 160) for a similar complaint.
(^70) Aquinas thought that one and the same thing can appear in the subject position of a judgment and
the predicate position, but when it does, it does so under different guises. See, for example,Summa
TheologiaI, question 13, 12th article (Aquinas 1948: 71). This is in fact noted by Kenny (2005a: 156–8), so
it is surprising that he does not allow for a similar move for quantification phrases.