Copyright holder: Tyndale University, 
3377 Bayview Ave., Toronto, Ontario, Canada M2M 3S4 
Att.: Library Director, J. William Horsey Library

Copyright: This Work has been made available by the authority of the 
copyright owner solely for the purpose of private study and research 
and may not be copied or reproduced except as permitted by the copyright 
laws of Canada without the written authority from the copyright owner. 

Copyright license: Attribution-NonCommercial-NoDerivatives 4.0 International 
License

Citation: Gould, Paul M. and Richard Brian Davis. “Modified Theistic Activism,” in Beyond the
Control of God? Six Views on the Problem of God and Abstract Objects, edited by Paul
M. Gould, pages 51-79. New York: Bloomsbury Press, 2014.

***** Begin Content ******


TYNDALE
UNIVERSITY
3377 Bayview Avenue
Toronto, ON
M2M 3S4

TEL: 416.226.6620

www.tyndale.ca

Note: This Work has been made available by the authority of the copyright owner solely
for the purpose of private study and research and may not be copied or reproduced
except as permitted by the copyright laws of Canada without the written authority from
the copyright owner.

Gould, Paul M. and Richard Brian Davis. “Modified Theistic Activism,” in Beyond the
Control of God? Six Views on the Problem of God and Abstract Objects, edited by Paul
M. Gould, pages 51-79. New York: Bloomsbury Press, 2014.

[ Citation Page ] 



Bloomsbury Studies in Philosophy of Religion

Series Editor:
Stewart Goetz

Editorial Board:
Thomas Flint, Robert Koons, Alexander Pruss, Charles
Taliaferro, Roger Trigg, David Widerker, Mark Wynn

Titles in the Series

Freedom, Teleology, and Evil
by Stewart Goetz

Image in Mind: Theism, Naturalism, and the Imagination
by Charles Taliaferro and Jil Evans

Actuality, Possibility, and Worlds
by Alexander Robert Pruss

God’s Final Victory: A Comparative Philosophical Case for Universalism
by John Kronen and Eric Reitan

The Rainbow of Experiences, Critical Trust, and God
by Kai-man Kwan

Thinking Through Feeling: God, Emotion and Passibility
by Anastasia Philippa Scrutton

Philosophy and the Christian Worldview: Analysis, Assessment and Development
edited by David Werther and Mark D. Linville

Goodness, God and Evil
by David E. Alexander

Well-Being and Theism: Linking Ethics to God
by William A. Lauinger

Free Will in Philosophical Theology
by Kevin Timpe

The Moral Argument (forthcoming)
by Paul Copan and Mark D. Linville

[ Series Listing Page ]



Beyond the Control of God?

Six Views on the Problem of
God and Abstract Objects

Edited by Paul M. Gould

B L O O M S B U RY
NEW YORK • LONDON • NEW DELHI • SYDNEY

[ Title Page ] 



[ Chapter ] 2

Modified Theistic Activism

Paul M. Gould and Richard Brian Davis

It is not easy to give an illuminating definition of an abstract object. According to
Morris and Menzel, an abstract object is (at the very least) an object “so firmly rooted
in reality that [it] could not possibly have failed to exist” (1986, 161).1 This characteri-
zation is not sufficient of course, since if classical theism is true, God is a necessary
being; it is not possible for Him not to exist. And no one thinks God is an abstract
object. Philosophers working in this area have therefore tended to proceed by giving
examples. Abstract objects are said to be impersonal entities: such things as properties,
relations, propositions, numbers, sets, and the like. Plantinga expresses the view of
many here: An abstract object, he says, is “an object that (like God) is immaterial, but
(unlike God) is essentially incapable of life, activity, or causal relationships” (1985,
88)?	

   Of course if you are wondering what nontrivial relationship holds between God
and abstracta, this is not the most helpful way to put things. For it implies right up
front that God cannot be causally related to abstracta, in which case it is difficult to
see how there could be anything but a trivial two-way relation of logical dependence
between them. This is just to say that in every world in which God exists (that is, in
every world), abstract objects also exist. But there is no deeper (asymmetrical) sense in
which abstract objects depend for their existence on God. There are powerful reasons
for thinking this sort of “platonistic theism” is incompatible with classical theism,
though we will not take the time to survey them here.3

   Conceptualism is the attempt to identify abstracta with divine ideas or concepts,
which by the very nature of the case would depend on God as thoughts on a thinker—
that is, causally and asymmetrically. In this respect, so-called theistic activism is
simply a strong form of conceptualism: One holding that “all properties and relations
are God’s concepts ... the contents of a divine intellective activity” (Morris and Menzel
1986, 166, emphasis added). However, a proper theistic activist also maintains that

	All necessarily existent propositions ... can be thought of as “built up” out
	of properties. Thus, in the way in which we characterize properties as God’s
	concepts, we can characterize propositions as God’s thoughts... . And taking
	numbers to be a variety of property, we thus have all necessarily existent abstract

[ Page 51 ] 



[ Page ] 52

	reality, from necessary mathematical objects to haecceities, to nonmathematical
	universals, to propositions, deriving existence from God. (ibid.)

So the suggestion is that divine concepts are ontologically basic; everything else in the
platonic horde is a derivative entity: a complex construction on God’s mental concepts.

   In what follows, we shall argue two things. First, it is plausible to think that
conceptualism holds with respect to propositions; in any event, it does a much better
job than its closest competitors (platonism and nominalism) in accounting for the
truthbearing nature of propositions. Secondly, it is wholly implausible (so we say) to
take the added step and equate properties and relations with divine concepts. Thus, a
modified theistic activism (MTA) emerges as the most natural and defensible way for
a theist to think about Gods relation to abstract objects.

Recipe for a proposition

Let us begin with propositions. What are they? Here we can do no better than to
begin with Lord Russell's insight: “The fundamental characteristic which distinguishes
propositions (whatever they may be) from objects of acquaintance is their truth or
falsehood” (1984, 108). Propositions are truthbearers. We know this (or can at least
stipulate it) pretheoretically. So consider the proposition

   (1) Quine is wise.

The question arises: How is it that (1) manages to be true? Naturally, the truth-condi-
tions for this proposition must obtain: Quine himself must exist and have the property
of being wise. But surely there are conditions to be met on the side of the proposition
as well—truthbearer conditions, as we might call them. Surely (1), to have a truth
value at all, whether true or false, has to be a claim or an assertion of some kind. It
must represent reality as being a certain way. Thus Alvin Plantinga:

	Propositions are claims, or assertions; they attribute or predicate properties to or
	of objects; they represent reality or some part of it as having a certain character.
	A proposition is the sort of thing according to which things are or stand a certain
	way. (1987, 190)

Similarly G. E. Moore:

	propositions, in the sense in which I have been using the term, are obviously a sort
	of thing which can properly be said to be true or false... . Every proposition is,
	as we constantly say, a proposition about something or other. Some propositions
	may be about several different things; but all of them are about at least one thing.
	(1953, 62, 68, emphasis added)

In short, propositions (again, whatever they are) are intentional objects; they are of or
about things. And this is an essential property of propositions; for if they lacked this
property, they could not possibly be claims or assertions of any kind, they could not



[ Page ] 53

represent anything, in which case they could not be true (/false). Regardless of how
things stood in the world, propositions just would not have anything to say about
those states of affairs. How then could we say of them that they were true (/false)? For
surely a proposition is true only if it represents the world as it is. And just as surely the
way things stand in reality is depicted as being thus and so only if something is being
claimed about the way things so stand. We do not say that Ayer's memoir Part of My Life
accurately represents his life—or rather, part of his life—if he is not mentioned, nothing
is claimed about his person, and no comment at all is made about his comings or goings.

A puzzle about parts

Well then, what does it take to be a truth-claiming proposition? According to one
venerable tradition—represented by Bertrand Russell (1903, §47-8), David Kaplan
(1989), and others—a proposition is a complex whole; it has internal constituents
(parts), and those constituents have a specific arrangement. This thesis can be
developed along platonist or nominalist lines.

   Consider, first, a broadly platonist approach to the situation. Here, we suppose,
things could go in a couple of directions. Following Russell, we might see (1) as an
admixture of concrete particulars and abstract objects.4 On the concrete side of things,
(1) would contain Quine himself as a constituent. But its other ingredients would
include the (abstract) property of wisdom, along with the (abstract) exemplification
relation Quine stands in to that property. This is a familiar story. The question, though,
is whether it even slyly suggests that (1) might be about Quine. And it is hard to see
that it does.

   Note first that whatever intentionality this proposition enjoys is inherited or
derived; it will be a function of (l)’s parts, each of which is essential to it. If (1) is a
Russellian proposition, for example, then if Quine had failed to exist, there would
have been no such proposition as Quine is wise. Moreover, if any of the constituents
of (1) were different than they are—say, if we substituted Obama for Quine—then an
entirely different proposition would have resulted: certainly not one about Quine, and
perhaps even with a differing truth value. So clearly, on the Russellian view, if (1) is
about Quine at all, it can only be because it contains him as a constituent.

   But here, we think, there is confusion. Although it is not uncommon to see
Russellian propositions presented as “complex abstract entities” (Pelham and Urquhart
1994, 307), the fact is they are neither abstract nor even propositions (at least as we
are thinking of them). They are not claims or assertions that represent certain states of
affairs; they are states of affairs. As the Russell scholar Anssi Korhonen notes:

	there is no condition in the world whose obtaining would be necessary for the
	truth of a [Russellian] proposition. Or, at any rate, this condition cannot be seen
	in any way distinct from the proposition itself. And this means that there is no gap
	between a propositions being true and somethings being the case: facts simply are
	(that is, are identical with) true propositions. (2009, 165)



[ Page ] 54

In other words, Russellian “propositions” are actually the truth-conditions for what we
are thinking of as truth bearers (propositions). On the Russellian view, (1) is not the
claim that Quine has the property of being wise; rather, it is a concrete state of affairs
that we can represent as follows:

   (1*) : Exemplification (Quine, wisdom).5

(1*) denotes a certain truthmaking fact, one consisting in the specific way Quine is
related to wisdom. But as Vallicella points out, we must not confuse this with either the
proposition Quine is wise or the state of affairs Quines being wise—both of which are
in need of something to make them true (/actual).6 (1 *) simply designates the concrete
fact that does that.

   This should make it clear that simply “plugging” Quine into a Russellian propo-
sition like (1*) is not going to create a vehicle for representing the way Quine is.
For (1*) just is the way Quine is. (1) represents the way things are; by contrast, (1*)
denotes the things that are that way. Or think of it like this: Since Quine is not of or
about anything, since he does not represent anything (even himself), the mere fact
that something contains him will not make that thing about Quine. You might as well
argue that Harvards philosophy department was about Quine because it contained
him as a member. The point is: Just in himself, Quine, while impressive in many ways,
is an intentional flop.

   Now here the platonist is not without reply. To avoid collapsing truthbearers into
their truth-conditions, while at the same time securing intentionality, she might try
removing Quine as a constituent of (1). Why not opt instead for a pure mix of platonic
properties and relations? In that case, perhaps (1) would be rendered best as

   (1**) Being Quine is coinstantiated with being wise,

a proposition consisting of two properties standing in the relation of coinstantiation.
Since it does not include Quine, (1**) is distinct from the concrete state of affairs it
represents—or at least so the platonist might claim. But does it represent? Why should
we think that? Here we can imagine someone arguing as follows. Since being Quine
uniquely characterizes Quine, how could it not be about him? His having this property
guarantees that it is about him. Now of course we can all agree that

   (2) An object has the property being Quine if and only if it is Quine.

But how does it follow that

   (3) Being Quine is about Quine?

It does not follow. For the slide from (2) to (3) obviously confuses being P (where
“P” names some property) with being about what has P. The two notions are not
equivalent. Something could have P without P's being about what has it. For example,
you could have being the author of Word and Object without that property being about
you. If you are Quine, that property qualifies you all right; it is predicable of you, but
it is not in itself of or about you; it is not directed upon you. Similarly, being wise is
neither about nor represents any of its instances. This is especially evident when one



[ Page ] 55

considers the fact that, for the platonist, there are worlds in which this property exists
but nothing is wise. So we do not secure intentionality for a purely abstract property
merely by pointing out that it has instances. Nor can we say being Quine is about
Quine because it contains something that is: Quine himself. For as we have already
noted, Quine is not about anything—even himself. To insist otherwise seems no more
than a basic category mistake.

   So what is a platonist to say at this point? Here is a final possibility. We treat Quine
is wise as a simple, brutely intentional Platonic Form—end of story. The problem,
though, is that this hardly squares with what we all learned at Socrates’s knee. The
Forms are not propositional.7 On the contrary, like the self-thinking Thought of
Aristotle, they are wholly indifferent to their concrete instances. A Form—say,
Circularity—is not directed toward any of the imperfect circles of the sort we routinely
encounter. Rather, the “direction of fit”8 is entirely the other way around. In order to
count as a circle, says Socrates, an object must “imitate” or “participate in” Circularity:
being an enclosed line each point on which is equidistant from a fixed point (the center).
It must possess a world-to-Form direction of fit, so to speak, to qualify as a thing of
this kind. As a Platonic Form, however, Circularity is a mere ontological target, a
certain standard of perfection to be approximated or achieved. And it can perform this
function very nicely without possessing any direction of fit toward the lower world
of concrete particulars. As a timeless, spaceless, metaphysical exemplar, Circularity
is no more about the crop circle that mysteriously appeared on your front lawn this
morning than the number five is about the fingers on your hand.

   So whereas we know by introspection that such things as thoughts, beliefs, and
desires are intentional, we have no basis at all in reason or experience for thinking that
Platonic Forms (including propositions, if we smuggle those in) are about anything.
Why then are we tempted to think otherwise? For this reason, we believe. If you start
with the idea that truthbearing requires intentionality, then (as Michael Jubien warns)
it is very difficult not to

	confer a strictly derivative dose of intentionality upon something that is otherwise
	intentionally inert [that is, the abstract platonic proposition]. It may be very
	convenient to suppress this and proceed as if the entity had the intentional feature
	on its own, but doing so would be a matter of efficiency of thought, not ontology.
	(Jubien 2001, 53-4)

We need propositions to do a certain amount of philosophical work. For the platonist,
their job is to serve as truthbearers and objects of our propositional attitudes. It is
therefore a matter of expediency to proceed as if they had the requisite intentional
properties, but without taking the time to explain how this could be. But this is
simply to smuggle intentional propositions into the nonintentional world of abstract
Forms—a move that has all the appearance of theft over honest toil.

   Suppose we turn, then, to a nominalist approach to these matters; perhaps it will
prove more promising. Now there are several varieties of nominalism with respect
to propositions, and there is not the space here to deal with all of them.9 Instead, we
will simply make some general remarks about what we take to be the core assumption



[ Page ] 56

underlying the approach; and hopefully, this will offer hints about how to think of the
versions we do not consider.

   Now we are operating, you recall, under the assumption that there is such a thing
as truth, that there are truthbearers, and that a thing must have ofness or aboutness to
possess truth. What sorts of things can the nominalist bring forward to fill the role
of propositional truthbearer? Here sentences stand out as the most likely candidate.
In the early Wittgenstein, for example, we are presented with the idea that proposi-
tions are sentences or linguistic items of a sort. “’The totality of propositions,” he says,
“is language” (2001, 22). Furthermore, a proposition (sentence) is said to represent
reality; for a “proposition shows how things stand if it is true. And it says that they do
so stand” (ibid., 25). It makes a claim about things, and it does so in part by virtue of
its syntactic and logical form: “there must be exactly as many distinguishable parts [in
the proposition] as in the situation that it represents” (ibid., 26).

   Let us assume all of this is in order. Then (1) is not an abstract object; it is the
concrete sentence inscription “Quine is wise”—a linguistic item with parts (that is,
“Quine,” “is,” and “wise”). And of course these parts have parts of their own (letters).
The sentence "Quine is wise” is therefore a concrete particular; it is a string of shapes
and characters. Call this string “S.” Why think that S is a primary bearer of truth? A
truthbearer must represent things. But neither S nor its parts is about anything just
in itself; shapes (even sequences of shapes) are not about things just by virtue of what
they are. Accordingly, it is obvious that if S possesses any intentionality, it is borrowed
or derived: not of course from its constituent parts, but rather (as Plantinga (2006,
17) notes) from our deciding to use it in a particular way, to express a proposition
(that is, (1)) about Quines being wise. Indeed, Wittgenstein concedes this very point
when he remarks, “a proposition is true if we use it to say that things stand in a certain
way, and they do” (2001, 28, emphasis added). This makes it abundantly clear that the
intentionality of a sentence lies outside of it.

   But where does this leave us? We have seen that “free-floating” abstract objects, that
is, Platonic Forms existing in Platos heaven, are impotent to account for the intention-
ality of (1); but concrete objects—ordinary particulars and linguistic entities—are really
no better. Does that mean that (1) is not about Quine, that it makes no claim about
him, and is therefore neither true nor false? Not at all. Fortunately, there is a better
way forward. The problems faced by platonism and nominalism lead us to a third,
more compelling view of the matter. The reason for the problems is by now obvious:
You cannot force intentionality upon objects whose natures simply defy the imposition.
Would it not be better, then, to begin with something known to be intentional,
something on which intentionality does not have to be imposed? Indeed, there is a
whole class of objects of this sort: ideas/concepts.10 By their very nature, ideas, regardless
of who has them, are of or about things. John Locke actually refers to them as signs; and
of course signs point to things. We speak of Descartes’s idea of a perfect being, Darwins
ideas about Nature, and Russell’s idea of a proposition. It is incoherent to suppose that
there are ideas that are not about anything. If it is not about anything, it is not an idea.

   This gives us a very handy solution to our puzzle about parts. That puzzle raised the
question of how (1), the complex proposition Quine is wise, could be about Quine. On



[ Page ] 57

the present suggestion, the answer is embarrassingly easy. (1) is about Quine because
it contains a part (the idea of Quine) that is essentially about him. Neither a Platonic
Form nor a concrete particular can pretend to credentials like this.

A puzzle about arrangement

But now a different puzzle emerges. Locating intentional parts—even the right
parts—for (1) is crucial. But it does not automatically issue in a truth claim. For while
ideas are certainly about things, they do not, strictly speaking, represent anything.
Thus we cannot say that propositions are representative truthbearers because they
are composed of true or false ideas. As John Locke observes, “truth and falsehood
belong, in propriety of speech, only to propositions” (2004, 345), not to ideas. One
can easily see why. For until “the mind passes some judgment on” its ideas, until it
“affirms or denies something of them” (ibid.), there simply is no claim or assertion, no
representation of reality as such. To judge or affirm that Quine is wise, to express that
proposition, the ideas of Quine (Q) and wisdom (W) have to be joined to one another.
To judge truly in this respect is to join Q and W in a way that reflects the actual relation
of the things these ideas are about.

   So a proposition is far more than an unrelated list of items, as is the case, for
example, with the ordered pair <Q, W>. The elements of this pair are indeed inten-
tional objects; they are ideas o/things. Still, <Q, W> does not represent Quine as being
wise. To do that, Q and W must be joined so that a specific claim is made. This requires
that these ideas be put together in a very precise way. Now <Q, W> is indeed ordered
since, by hypothesis, <Q, W>≠ < W, Q>. But the thing to see is that it is not ordered in
the right way. For while there is an asymmetrical “directionality” to the arrangement,
it is not affirmative; there is not the slightest hint that we have a claim or assertion on
our hands. For there is no predicational tie between Q and W. And given that Q and
W are ideas, is not that sort of tie going to require a mental knot?

	So what explains this arrangement of ideas? Well, according to Thomas Aquinas,

	Every composition ... needs some composer. For, if there is composition, it is
	made up of a plurality, and a plurality cannot be fitted into a unity except by some
	composer. (1995, 1:103)

Given this principle of composition, propositions will need a composer, since
they are composed entities with an orderly arrangement of parts. But why is this
necessary? Perhaps it will be helpful to remind ourselves of the obvious. Ideas, as we
are viewing of them, occupy neither Plato’s detached realm of abstracta, nor Ayer’s
world of empirically verifiable concreta. Ideas as such “cannot be seen, or touched,
or smelled, or tasted, or heard” (Frege 1997, 224). Locke speaks of “those invisible
ideas” of an individual “which his thoughts are made up of” (2004, 363). If we think
about it, we can see that an essential property of an idea is that it is had or possessed.
Ideas “need an owner” (Frege 1997, 334); they are ontological parasites on thinkers.



[ Page ] 58

And just as there cannot be thoughts without a thinker, ideas (which are nothing but
materials for thinking) cannot exist apart from the minds that have them. But then
what better explanation could there be for the orderly arrangement of ideas than the
mental activity of thinkers? The obvious conclusion to be drawn here is that the things
properly said to be true or false (propositions) actually result from mental activity—
from the joining or separating of ideas. Thus Locke once more:

	Every one’s experience will satisfy him, that the mind, either by perceiving or
	supposing the agreement or disagreement of any of its ideas, does tacitly within
	itself put them into a kind of proposition affirmative or negative, which I have
	endeavored to express by the terms putting together and separating. (2004, 510)11

Thus it follows straight away that propositions are mental effects. For propositions
have parts, those parts are best construed as ideas, and their being properly related
(that is, “fitted into” truth claims) requires a mental arranger.

An objection and reply

“If Q and W are just your ideas, then the proposition Quine is wise is only your
thought or proposition. Indeed, there is no such thing as the proposition Quine is
wise, in which case (as Frege (1997, 335-6) argues) no two people ever think the same
thoughts. They only ever think their thoughts; nothing is in common.”12

   Frege is on to something here. Propositions are not simply reducible to our
thoughts (yours or mine). If they were, they would suffer from a deplorable fragility.
And in any event, there just are not enough of us to think up all the propositions
associated with, say, the natural numbers. Propositions are in some sense objective;
even if we had not existed, they would have been none the worse off. So what is to be
done? Freges way out here was to recognize what he called a “third realm”: a realm
beyond the world of physical objects, beyond the realm of human ideas—in fact, a
platonic realm of abstract thoughts without owners. But as we have already seen, such
abstract platonic entities are no truthbearers; they lack all the necessary intentional
ingredients.

   The proper thing to do, therefore, is to go one step further and recognize a fourth
realm. Things belonging to this realm would have in common with human ideas that
they were intentional objects. But like ordinary particulars, their existence would also
be objective; that is, even if there were no human minds, these things would continue
to exist nonetheless. Thus, to borrow Frege’s example, the Pythagorean Theorem is
not just true from the time we discover it. It is more like a planet, which "even before
anyone saw it, was in interaction with other planets” (Frege 1997, 337). Here we may
happily agree. Propositions are indeed independent of human minds. Where Frege
erred, however, was in thinking that propositions do not require owners. This is
simply false if they are truthbearing claims composed of intentional parts. In this case,
what we must grant, it seems, is a supranatural realm of divine ideas and thoughts,



[ Page ] 59

not subject to the relativistic vagaries of human mental activity. The best way to look
at things, perhaps, is to see we human beings as arranging our ideas—which are
sometimes complete, but often only partial graspings of God’s ideas—into thoughts
which approximate (to varying degrees) those of God himself. We can then reserve
the term “proposition” for referring to Gods thoughts.

The perils of unbridled theistic activism

Naturally, much more remains to be said. But hopefully this fills in at least some of
the details (and motivation) behind Morris and Menzel’s brief assertion that “we
can characterize propositions as God’s thoughts.” However, we do not think that the
activist project should be extended to properties and relations, thus, in what remains,
we shall argue for a modified theistic activism (MTA).

   The linchpin of theistic activism is Morris and Menzel’s principle of Property-
Concept Conflation (PCP): “All properties and relations are God’s concepts.” It is
easy to see that this principle undermines basic attempts to explain the notion of a
substance. Consider, for example, the bundle theory: conjoined with PCP, it leads to a
form of Berkeleyan idealism. For suppose a material thing is nothing but a bundle of
compresent multiply exemplifiable properties. Then if properties are God’s concepts,
and if relations are God’s thinking concepts together, every material object is a mere
collection of divine concepts or ideas, in which case we shall have to say that a
substance changes just when God starts or stops “thinking together” His own concepts
or ideas. Not only does this eradicate the material nature of reality, it smacks of an
objectionable divine determinism (more severe than anything Calvinism has to offer).

   On a nominalist approach to bundles, things look only slightly better. Instead
of bundling platonic properties, we might try tropes or property instances—such
concrete entities as the wisdom of Socrates or this spot’s redness. A substance would
then consist in a bundle of concrete particulars, thereby sidestepping the reduction of
material things to abstract universals (and thus, by PCP, to divine ideas). Still, there is
the matter of how these tropes hold together to form one substance. The typical expla-
nation posits a basic, unanalyzable relation of compresence or collocation that serves as
the ontological “glue” for these otherwise disparate tropes. According to PCP, however,
this relation is not primitive; it is analyzable in terms of God’s conceptual activity. The
tropes that comprise Socrates hold together because God thinks them together. At first
glance, this strikes one as a tidy illustration of Paul’s words in Col. 1.17—that “in him
all things hold together.” Strictly speaking, though, Paul does not say that things hold
together because God thinks them together. In fact, would not the explanation go the
other way around? God thinks of the tropes comprising Socrates as being compresent
because He causes them to be so. In any event, it is (almost) trivial to say that God’s
thinking these tropes together is the cause of their being together. For causation is
a relation, and on PCP all relations are divine concepts. Therefore, to say that God
causes trope compresence by thinking tropes together amounts to the tautologous



[ Page ] 60

claim that God thinks of His thinking tropes together by thinking tropes together. The
concept of causation as an extramental relation goes completely by the wayside here.

   Next, consider bare particular theory. Following Bergmann (1967, 22-6)—and
more recently, Moreland (2001, 148-57), Pickavance (2009), and Sider (2006)—we
might hold that a proper ontological assay of a substance (Socrates, let us say) must
reference a bare particular, the relation of exemplification, and certain external
property “ties.” To say that Socrates is human just means that the property of being
human is “rooted in” Socrates as a constituent, but also “tied to” Socrates’s bare
particular (call it “b”) by the exemplification relation. Unlike Socrates, however, b has
no internal property constituents, and thus counts as a “thin” or “bare” particular.
Proponents of this view go on to say that bare particulars have no categorical or
kind-defining properties of their own; they only ground the properties of the concrete
particulars of which they are constituents.13 But then given PCP, it looks as though
Socrates is the result of Gods conceiving of & in terms of the divine concept being
human, which is strange enough. However, matters are even worse; for b is a thing
with no properties of its own, which (on PCP) simply means that God does not have a
concept of it; in which case Socrates is the consequence of God’s conceiving of a thing
of which He has no conception. It is not a coherent picture.

   Someone will object that we need only turn to Aristotelian substances and the
problem dissolves. On the Aristotelian view, a concrete particular like Socrates is a
basic entity; he is not “built up” from constituent properties “tied to” a constituent
(bare) substratum. Rather, as Loux notes, the subject of ontic predication is Socrates
himself. This is made possible by the simple fact that he belongs to a natural kind (that
is, being human) that marks him out as a particular sort of thing, “countably distinct
from other members of that kind and from members of other kinds” (2006, 113).
Moreover,

	in virtue of being an instance of its proper kind, a concrete particular can be the
	subject for attributes—properties—that are external to its core being. So concrete
	particulars do have a structure that the ontologist can characterize: there is a
	core being or essence furnished by a kind and a host of properties that lie at the
	periphery of that core and, hence, are accidental to concrete particulars, (ibid.)

Here there is no need to call up an incoherent (explanatorily prior) bare particular for
purposes of property support. That has to be an advantage.

   The deeper issue, for whatever notion of substance used is that, given PCP, divine
concepts are called upon to play a role that seems inimical to their nature. As essen-
tially intentional mental objects, divine concepts (and concepts in general) mediate
between mind and world.14 The primary role for properties, however, is that of making
or structuring reality. As George Bealer observes, “[properties] play a fundamental
constitutive role in the structure of the world” (1998a, 268). As we argued earlier,
properties (that is, Platonic Forms) are intentional flops—they are not about anything,
and essentially so. But, if divine concepts are called upon to play both the mediating
and making role, it seems that they can only do so at a high cost in terms of theoretical
economy. To see why consider: Divine concepts/properties are essentially intentional



[ Page ] 61

(in their mediating role) and not intentional (in their making role). But, this picture
seems incoherent. To make sense of this story, the activist employing PCP will need to
argue that divine concepts/properties have (exemplify?) the following nature (essential
property?): Being intentional when a constituent of thought, being nonintentional when
not. Such a nature (property?) of divine concepts is surely unlovely (at best) and (at
worst) smells of ad hocness.15 We say, best to leave properties and relations alone—they
are sui generis, and admit their own ontological category.

Can God create abstract objects?

The version of MTA we wish to defend is a kind of modified platonic theism. Thus far
we’ve argued that some abstracta are located within the divine mind (propositions and
concepts) and some are external to God (properties and relations not exemplified by
God). Thus, abstract objects exist in two realms: the divine mind16 and Plato’s heaven.
This picture of the world opens us up to a number of worries, worries that we will
attempt to discharge in the remainder of this chapter.

   If the existence of abstract objects is admitted into ones ontology, a prima facie
problem arises for the traditional theist. The problem is this. Abstract objects, it seems,
are best understood as uncreated entities. But, if abstract objects are uncreated, then
God is not the creator of everything, a view that appears unacceptable to the tradi-
tional theist constrained by Scripture and tradition.

   Some platonists (e.g. Peter van Inwagen (2009)) argue that theists need not
worry—there is no actual tension between traditional theism and the existence of
abstract objects since abstract objects are not the kinds of things that can enter into
causal relationships. It is our judgment that if sense can be given to the causality by
which one necessary being (God) would cause another necessary being (abstract
objects) to exist, then there is no good reason to think that God could not create
abstract objects and hence, no good reason to think that abstract objects cannot enter
into causal relations.17

   Assume anti-reductionism regarding causation.18 Anti-reductionism is not the
view that causation is primitive. Primitivism regarding causation denies that there
are any concepts more basic than causation. All we wish to endorse here is that there
are no noncausal terms that can adequately explicate the notion of causation. Given
anti-reductionism regarding causation, God’s creating abstracta can be understood as
follows:

   (C^) God caused abstract object P if He brought it about that P exists.

(C^) specifies a plausible account of causation in which such creation is possible (in
which a sufficient, but not necessary condition for such creation is realized). Still, one
might object: (CA) is hardly illuminating in terms of how God creates abstract objects.
Specifically, the right-hand side of the conditional doesn’t explain the left-hand side—
as it stands, they are virtually synonyms at face value. Fortunately, we’ve already seen



[ Page ] 62

a more illuminating anti-reductive account of how God causes abstract objects, the
theistic activism of Morris and Menzel:

	(CTA) God caused abstract object P if P is (i) a constituent of God’s mind (ii)
   	brought about by the activity of thinking.

We also think that the following is an acceptable understanding of God’s creating
abstracta:

	(CMTA) God caused abstract object P if P is brought about by the activity of divine
	willing.

(CMTA) allows that God creates abstract objects wholly distinct from His being—
existing in a platonic heaven even. And if (CTA) and/or (CMTA) are possibly true, then
it is reasonable to think that God can indeed create abstract objects. We argue that He
has in fact done so.19

Divine bootstrapping?

But, can our MTA avoid the charge of incoherence because of divine bootstrapping?
Recall that the bootstrapping worry is usually advanced as follows: "God has
properties. If God is the creator of all things, then God is the creator of His properties.
But God can’t create properties unless He already has the property of being able to
create a property. Thus, we are ensnared in a vicious explanatory circle. God causes
His nature to exist—a nature He must already possess to do the causing.” We think
that the incoherency worry due to divine bootstrapping can be successfully avoided
by endorsing the following two claims:

	[A] God’s essential platonic properties exist a se (i.e. they are neither created nor
	sustained by God, yet they inhere in the divine substance); and

	[B] Substances are Aristotelian.

Endorsement of claim [A] allows us to avoid the unwanted view that God creates His
own nature; claim [B] ensures that the divine substance is a fundamental unity that
is the final cause of its constituent metaphysical parts (including divine concepts and
essential properties)—which ensures God’s ultimacy. Thus, the divine substance, along
with all of its essential properties exists a se and everything distinct from God (that is,
everything external to God’s borders) is created and sustained by God.

   What about the rest of the platonic horde: numbers, sets, quantifiers, states of
affairs, and possible worlds? No doubt each of these alleged abstracta requires a paper
in its own right. However, if our discussion of MTA has taught us anything, we should
not expect any one approach here to single-handedly resolve the issues raised by the
platonic horde and its relationship to God. Still, we have good reason to think that
each of these abstracta can be safely brought either into the mind of God or located in
Platos heaven, without violating God’s aseity or sovereignty. Thus, one kind of platonic



[ Page ] 63

theist can have it all: an attractive (realist) theory of the mind-world-language nexus
and fidelity to Scripture and tradition.

Notes

1  This is not quite true of course. Sets are presumably abstract; however, arguably, some
   sets—e.g. Obama’s singleton—exist in only those worlds in which Obama does. For
   more on the contingency of sets, see Alvin Plantinga (1976, 146-7).
2  Plantinga now thinks that abstract objects can enter into causal relationships, see
   (2011,32).
3  But see the Introduction where they are nicely laid out.
4  Compare Russell: “Whatever may be an object of thought, or may occur in any true or
   false proposition ... I call a term.... A man, a moment, a number, a class, a relation, a
   chimaera, or anything else that can be mentioned, is sure to be a term” (1903, 43).
5  Following D. W. Mertz, we use the colon locution for the operator “The fact that.”
6  On the importance of distinguishing between concrete and abstract states of affairs,
   see William F. Vallicella (2000, 237).
7  As Matthew McGrath notes, if Plato believed in propositional Forms, he could have
   had Socrates solve the problem of how a false belief (e.g. in Pegasus) could actually
   be about something by saying that there is an object of belief here: the (false) abstract
   proposition Pegasus exists. However, no such proposal is even considered. See
   McGrath (2012, §1).
8  A term coined by J. L. Austin. See Searle (1998, 101).
9  For an excellent overview of nominalist options, see Michael Loux (2006, 130-9).
10 We shall use the terms “ideas” and “concepts” synonymously.
11 Locke actually distinguishes two sorts of propositions: mental (thoughts) and
   verbal (sentences). Both are composed of signs, he says: the former consist of ideas,
   the latter of words. Here we focus exclusively on mental propositions. For as we
   noted earlier, even if we can speak of sentences as being true or false, it is only in a
   derivative sense.
12 Thus Michael Jubien: “Somehow or other, [conceptualism] has to make room for the
   fact that you and I can 'believe the same thing’” (1997, 52).
13 Compare Moreland and Pickavance: “the properties said to be necessary for bare
   particulars are not genuine properties; these include simplicity, particularity,
   unrepeatability, and those of the three categories of transcendental, disjunction, and
   negative properties” (2003, 10).
14 Dallas Willard states, “[concepts] form the ‘bridge’ that connects a thought and its
   object” (1999, 13).
15 Part of the ad hoc character of the necessary fix is that (as we mentioned earlier) it
   is typically thought that properties/Forms are wholly indifferent to their concrete
   instances, but now the activist is forced to admit otherwise. One of us has argued
   that it seems possible for the theistic activist to employ PCP, even if it is unlovely and
   uneconomical—at any rate, it is not logically impossible. See Gould (2011a).
16 Below we will allow that God exemplifies properties and stands in relations to
   constituent parts, thus this first realm of abstract objects is best understood as the
   divine substance (of which the mind is a part).


[ Page ] 64

17 We are not alone in this, e.g. Plantinga thinks that theistic activism gives us good
   reason to think that abstract objects can enter into causal relations: “[if] sets,
   numbers and the like ... are best conceived as divine thoughts ... then they stand to
   God in the relation in which a thought stands to a thinker .... If so, then [abstract
   objects] stand in a causal relation” (2011a, 32).

18 Anti-reductionism regarding causation is plausible, enjoys independent motivation,
   and has been ably defended recently by inter alia John Carroll (2010) and James
   Woodward (1990). Typical arguments for anti-reductionism involve (i) detailing
   the repeated failures of reductive analysis; (2) the fact that there is a sparse base of
   noncausal concepts that can be employed in providing a reductive analysis; and (iii)
   the case of preemption.

19 Granted, there are other issues looming in the background that would need
   to be addressed in articulating a robust doctrine of God’s creation of abstract
   objects, including providing an explication of eternal creation—for in the case of
   God creating everlasting abstract objects, the cause is not temporally prior to its
   effect. This topic would require another paper, but for now let us state that many
   contemporary philosophers working on the metaphysics of causation agree that
   causation need not involve reference to the relation of temporal priority. Indeed
   contemporary discussions of causal asymmetry deal routinely with cases in which
   cause and effect are simultaneous and, we are told, physics takes seriously the
   possibility of backwards directed time-travel and the accompanying backwards
   directed causation. See John Carroll (2010) and Huemer and Kovitz (2003). For a
   robust defense of the claim that God can create abstract objects, see Gould 2013.



[ Page ] 65

Response to Paul M. Gould and
Richard Brian Davis

Keith Yandell

In my first reply, I wish to say something that I will not repeat. It has been a pleasure to
read the papers of my colleagues. I have learned from them and respect their positions.
However much I sometimes disagree with views expressed, I do not think our discussion
concerns a point of Christian orthodoxy. I turn to putting my points briefly, concerning
both Professor Gould’s fine introduction and the fuller essay by Gould/Davis.

   Regarding the Augustine passage quoted by Professor Gould (17, note 4), what he
spoke strongly against was holding the view that God has to appeal to Platonic Forms
in order to have a recipe for making things. It is no part of propositionalism to say that.

    Modified Theistic Activism (MTA) holds that some abstracta (e.g. propositions,
concepts, and maybe more) are the product of Gods intellectual activity, but others
(such as properties and relations distinct from God) are the product of Gods creative
activity, but not His intellectual activity. MTA is “modified” in another way as well:
Those properties (and relations) essential to God exist a se within the divine substance
as uncreated entities, thus waiving a uniform account of properties (and relations).

   Whether God strictly creates propositions is another matter. If propositions are
the propositional contents of God’s thoughts, then if God is necessarily omniscient,
having those thoughts are part of, or at least entailed by, God’s nature. If necessary
omniscience requires being unrelentingly aware of every truth, then God does not
create them; thinking them is part of, or at least entailed by, God’s nature. If necessary
omniscience permits merely being able to bring every true proposition to mind
without always having it “in mind,” then without independent abstracta, God must
recreate whatever propositions are not “in mind.” This of course is not possible if
God is eternal rather than everlasting, and the eternalist will embrace an eternal
unrestricted attendance of all truths in the hall of divine thought.

   Suppose there are abstract objects. Then consider the sovereignty issue. Either it is
possible that abstracta are caused, or it is not. If it is, then God created them. If it is not,
then it no more restricts God’s sovereignty that God did not create them than that God
cannot make other contradictions true. Either way, on any logically consistent notion
of sovereign creation, God is sovereign regarding creating if and only if, for everything
X that exists, is not God, and can be created, X is created by God.

   The claim that abstract objects are ideas necessarily had by a necessarily existing
God cannot be literally true. Granting that there is an abstract-versus-concrete
distinction, abstract objects are not concrete. God is a concrete being. The ideas of
a concrete being are concrete, as it is that in virtue of which they have propositional
content. Hence abstract objects are not God’s ideas. The person who says that abstract



[ Page ] 66

objects are ideas in the divine mind is saying that what exists instead o/abstract objects
are ideas necessarily had by a necessarily existing God.

   There is a problem with saying that propositions are assertions as opposed to saying
that they are what can be asserted. “Blueberries are not moonbeams” is a perfectly fine
assertion, but I doubt that anyone (including God) has asserted it.

   The idea that God creates abstract objects that are distinct from God and neces-
sarily exist has attracted some. I doubt that a necessarily existing thing can be created
by, or be dependent on, anything, but I waive that. Strictly, what must be meant here
is that God emanates necessarily existing abstracta (see my lead essay for further
discussion). To say that God necessarily has some property Q, or necessarily engages
in some activity A, or necessarily undergoes some process P resulting in the existence
of X, is to say that God does so eternally or everlastingly. Given that abstracta are
distinct from God, and exist necessarily, God too must exist necessarily. If God
emanates abstracta, God is not free to refrain from so doing. So God cannot fail to be
a creator. On this view, divine sovereignty does not allow God to be otherwise than
being the maker of one kind of thing. Further, God depends for God’s existence on
there being the abstracta that God must create. This is not merely a case in which,
since any necessary truth entails every other, the denial of any necessary truth entails
the denial of any other. If God cannot but produce abstracta, then their absence would
entail God’s absence without appeal to entailment among necessary truths, even
assuming that God exists is a necessary truth.

   Taking “world” in a very broad sense, propositions represent the way that the
world is, and are true if the world is that way. They are thus about the world whose
content determines their truth value. It is not obvious that propositions are made
up of concepts as sentences are made up of words. Consider the claim that S has the
concept of being a cow insofar as S knows what is true if and only if there are cows.
Then concepts are propositional abilities that persons have, or the results of using
those abilities, and propositions are primary.

   Appeal to tradition is not so straightforward as assumed. MTA is inconsistent with
the traditional divine simplicity doctrine. Many contemporary Christian philosophers
reject divine eternity. Does this raise questions about MTA and divine everlastingism?
Why, and to what extent?

   I do not think Scripture is against propositionalism. The question of the existence
of propositions is a technical philosophical issue and on one view such things are
necessarily not causable. The dispute as to whether that view, combined with the view
that there are propositions, can be held by Christians who think consistently is contro-
versial—too philosophically, theologically, and exegetically controversial for any “side”
to justifiably claim exclusive exegetical basis for its view.

Greg Welty

Anyone reading the Gould/Davis chapter and my own can see that our respective
approaches to these matters have much in common. We have both argued that the



[ Page ] 67

intrinsic/derived intentionality distinction is relevant to critiquing nominalism, and
that the intrinsic intentionality of propositions is best explained as the intrinsic
intentionality of thoughts. In addition (going beyond the scope of my own chapter), I
agree with Gould/Davis that the Russellian platonist analysis of propositions fails, that
unbridled theistic activism (TA) is perilous, and that the “God cannot create necessary
beings” objection to TA is bogus. Despite all this, here are a few areas of potential
disagreement.

   Gould/Davis claim that conceptualism “does a much better job than its closest
competitors (platonism and nominalism) in accounting for the truthbearing nature
of propositions” (52). While their case against nominalism seems secure, their case
against platonism needs some work. First, while I agree that the Russellian analysis
fails for the reasons noted, Gould/Davis seem to think that the only platonist alter-
native to Russell is to “treat Quine is wise as a simple, brutely intentional Platonic
Form—end of story” (55). They continue: "The problem, though, is that this hardly
squares with what we all learned at Socrates’s knee. The Forms are not propositional”
(55). But this criticism unwarrantably restricts platonist options to whatever Plato
actually wrote. Why can't the modern platonist stipulate that there are non-spatiotem-
poral, intentional truthbearing entities—the kind expressed by sentences and serving
as the referents of that-clauses, among other things—and be done with it? What,
exactly, is the problem here? Platonists simply infer that there must be such entities
because this ontological specification best satisfies the functional concept of “propo-
sition” that emerges from realist arguments for propositions. What Plato said about
“Forms” doesn’t matter.

    Second, Gould/Davis claim that “whereas we know by introspection that such
things as thoughts, beliefs, and desires are intentional, we have no basis at all in
reason or experience for thinking that Platonic Forms (including propositions, if we
smuggle those in) are about anything” (55). But surely platonists will say that we do
have reason for thinking that propositions are about things: The realist arguments give
us reason to think there are such intentional entities. Following Jubien, Gould/Davis
apparently think that platonists start with an “intentionally inert” entity, “the abstract
Platonic proposition,” and then they “confer a strictly derivative dose of intentionality
upon” (55) it. (They “smuggle intentional propositions into the nonintentional world
of abstract Forms” (55). They “force intentionality upon objects whose natures simply
defy the imposition” (56). “Platonic entities ... lack all the necessary intentional ingre-
dients” (58).) But as far as I can tell, platonists just infer that there are these intentional
entities, and that is that. Who says these entities must start out “intentionally inert,”
like bits of graphite or ink, only to be baptized by aboutness later?

   Third, Gould/Davis claim that “Given [Aquinas’s] principle of composition, propo-
sitions will need a composer, since they are composed entities with an orderly
arrangement of parts” (57). But while this would pave the way for conceptualism, it’s
not clear why we should regard propositions as “composed entities with an orderly
arrangement of parts.” This is part and parcel of Russell’s (failed) analysis of propo-
sitions. Must it be an element of any platonist account? Why? (Propositions make
reference to entities. But are they therefore composed of parts? One might as well say



[ Page ] 68

that since God refers to entities He is composed of parts. More argument is needed
here, I think.)

   Fourth, Gould/Davis claim that platonists don’t take the time to “explain how”
propositions can have “the requisite intentional properties” (55). This is a worry for
platonists, but I wonder how strong it is. I agree that “it is incoherent to suppose that
there are ideas that are not about anything. If it is not about anything, it is not an idea”
(56). But does stating this “explain how” thought has intentionality? Are we not saying
that thoughts are intentional by definition? In this respect, do platonists claim any less
for their entities?

   In contrast to all this, my alternative case against platonism and for conceptualism
involves conceding that platonists can easily infer their denizens of the abstract realm
from the various realist arguments on offer (I canvassed four). The problem is not that
their entities are not intentional or are not composed of parts, but that their entities
multiply ontological kinds beyond explanatory necessity. This appeal to simplicity is
not open to Gould/Davis, as they are happy to posit “Platos heaven” in addition to the
divine thoughts.

   About modified theistic activism (MTA) I will explain just one worry. MTA is
“modified” in order to avoid bootstrapping worries. It retains the characteristic claim
of TA that abstract objects like propositions and properties causally depend upon God.
He “brings about” these things in some sense of nonreductive causation. But MTA
exempts God’s essential properties from this causal process, so that He is not stuck
creating His nature. This is certainly an improvement. But as Matthew Davidson has
pointed out (1999, 288-90), even if God only creates propositions the bootstrapping
worries remain. If God causes the proposition God exists to exist, He causes its
essence (which includes the property being true) to be exemplified. Thus, in causing
this proposition to exist, He makes it true, and thus God causes His own existence. In
response, Gould/Davis could further restrict MTA, such that propositions (or at least
the troublesome ones) also inhabit platonic heaven rather than being caused by God.
But then this frustrates their sustained, carefully crafted case that platonic proposi-
tions are inherently unworkable. I would make the change in the other direction
(surprise, surprise!): The relation between propositions and God is one of constitutive
dependence rather than causal dependence.1

William Lane Craig

Gould/Davis’s modified theistic activism is a hybrid view: Conceptualism with respect
to propositions and absolute creationism with respect to properties and relations.
Their essay, like Weltys, illustrates the way in which the two different debates over
so-called nominalism (see pp. 115-16 of my lead essay) can run together.

   Consider first their absolute creationism. They affirm that God causes abstract
objects (like properties) that are wholly distinct from His being. The key question in
assessing their view is whether they successfully answer the bootstrapping objection.
In reply to the objection, they endorse two claims (62):



[ Page ] 69

	[A] God’s essential platonic properties exist a se (i.e. they are neither created nor
	sustained by God, yet they inhere in the divine substance) 

and

	[B] Substances are Aristotelian.

Their view seems to be that God does not stand in a relation of exemplification to
certain platonic properties existing extra se, in virtue of which He is powerful, wise,
good, and so forth. Rather God just is a powerful, wise, good substance, and any
properties He has are immanent universals. But this Aristotelian view seems to me
to yield the palm of victory to the anti-platonist. For this just is anti-platonism with
respect to God. But then no justification remains for giving things the “ontological
assay” that the platonist wants to give. Things can be powerful or wise or brown
without being so in virtue of standing in an exemplification relation to an abstract
property. Once the absolute creationist has gone this route, no rationale remains for
platonism about properties. If this is not their view, then I frankly do not understand
it.2

   Now consider their conceptualism with respect to propositions. I applaud their
emphasis on the centrality of intentionality. Any aboutness associated with either
sentence tokens or abstract propositions is derivative, the result of their being used by
intentional agents.

   Now considerations of intentionality immediately raise the question of the
ontological status of intentional objects. It is a datum of experience that we often think
about things that do not exist—Santa Claus, the accident that was prevented, the hole
in my sock, last summer's vacation, and so on. Gould/Davis tell us, “It is incoherent to
suppose that there are ideas that are not about anything” (56). This was a characteristic
emphasis of Alexius Meinong’s Theory of Objects (Gegenstandstheorie). According
to Meinong the intentional objects of some of our ideas are nonexistent objects.
Do Gould/Davis mean to endorse Meinong’s view? If so, then why not say with the
neo-Meinongian that abstract objects like propositions, numbers, and properties are
nonexistent objects? Then divine aseity and creatio ex nihilo remain inviolate.

   On the other hand, if they mean to endorse a view of intentionality more akin
to Meinong’s teacher Brentano, according to which intentionality is not a relation
between a mental state and an object, but a monadic property of a mental state (e.g. I
have the property thinking-of-Santa Claus), then why not say with the anti-realist that
our ability to think of propositions, numbers, and properties does not imply that there
actually exist such objects?

    Gould/Davis would say that we need something more than mere sentence tokens
to serve as truthbearers. But why? The truth-predicate “is true” may be seen simply
as a device of semantic ascent which enables us to talk about a statement rather than
to assert the statement itself. For example, rather than say that God is triune, we can
ascend semantically and say that it is true that God is triune. Similarly, whenever
the truth-predicate is employed, we can descend semantically and simply assert the
statement said to be true. For example, rather than say that it is necessarily true that



[ Page ] 70

God is self-existent, we can descend semantically and simply assert that, necessarily,
God is self-existent. Nothing is gained or lost through such semantic ascent and
descent	

   Why is a device of semantic ascent useful or needed in natural language? The
answer is that the truth-predicate serves the purpose of blind truth ascriptions. In
many cases we find ourselves unable to assert the statement or statements said to
be true because we are incapable of rehearsing them due to their sheer numerosity,
as in “Every theorem of Peano arithmetic is true,” or because we are ignorant of the
relevant statements, as in “Everything stated in the documents is true.” In theory even
blind truth ascriptions are dispensable if we substitute for them infinite disjunctions
or conjunctions like “Either p or q or r or ... .” While such infinite disjunctions and
conjunctions are unknowable by us, they are known to an omniscient deity, so that
God has no need of blind truth ascriptions. Hence, He has no need of semantic ascent
and, hence, no need of the truth-predicate.

   Finally, I am not convinced that Gould/Davis’s conceptualism can avoid collapse
into anti-realism. I concur with Oppy (pp. 178-9 that thoughts are not abstract objects
nor are they composed of abstract objects (ideas). Otherwise, it seems unintelligible to
say that thoughts (abstract objects) exist in God’s mind. They are concrete mental events
or states. Like psychologism, against which Frege inveighed, divine conceptualism, as
Welty rightly sees (p. 90), is really an anti-platonic realism. As different versions of
concrete realism (see Fig. 1 on p. 40), psychologism and conceptualism posit mental
tokens rather than physical tokens in the place of abstracta. But the anti-realist theist
also believes that God has thoughts to which truth can be ascribed. So what gain is
there in conceptualism? One may as well just rest content with anti-realism.

Scott A. Shalkowski

Gould/Davis begin with the wholly sensible task of asking and answering the
crucial, but often overlooked, questions regarding propositions. What are proposi-
tions supposed to be? What are they supposed to do? In their critique of platonism,
they bring together what philosophers say in its favor, showing that even though it is a
highly recommended philosophical view, it cannot perform the central tasks for which
it was invented. If one does not accept an identity theory of truth (McDowell 1994;
Hornsby 1997), then there must be some distinctions between the ways things are and
the ways things are represented to be, even for truths. Their discussion of Russellian
propositions containing Quine (himself) and non-Russellian propositions containing
being Quine highlights the idea that things in themselves do not represent. Purely
ontological accounts of representation and truth fail, whether platonist or nominalist.

   Their rejection of nominalism, however, is too swift. No concrete item intrinsically
represents, but nominalists need not think that they do. All that follows from the
intentionality requirement on representation is that there are no unintended claims
and, hence, no unintended truths. Nominalists and the divine conceptualists are in
league here.



[ Page ] 71

   Nominalists might part company when Gould/Davis insist that “propositions are
mental effects” (58). The problem is not that propositions are parasites (57). All hangs
on the nature of the parasitism. If the extension of “proposition” goes no further
than beliefs held, assertions or claims made, then the differences are merely stylistic.
Gould/Davis want to make propositions God’s thoughts. Nominalists need not object.
Insofar as we can make sense of the common philosophical vocabulary, God is
neither abstract nor physical but is concrete. So, Gods thoughts are no less within the
nominalist's framework than are ours.

   What separates nominalists from MTAists is whose intentions are relevant to our
comprehensive philosophical accounting of things. Nominalists may agree that if
truths antedate the creatures sufficient to intend the relevant signage, then human
signage is insufficient. If one can sometimes think the same thought as another, then
those thoughts cannot be wholly private and the acts of thinking must have something
in common.

   Two pauses, though. Why think that we do, as we commonly say, think the same
thought? Why is it not sufficient that you think that Quine is wise and so do I? Both
platonists and MTAists share the idea that besides (a) Quine being wise, (b) you and
your thoughts, (c), me and my thought, there must still be an object—a represen-
tation—to which each of us and our thoughts must be related. Nominalists deny that
there are sufficient grounds for thinking this. We can, if we like, say that you and I
think the same thing when you think Quine is wise and so do I, but each of us repre-
sents things to ourselves—and to others when we decide to go public that Quine is
wise. To the extent that we have ideas and we intend to direct them as signs one way
and not others, we simply intend the relevant mental state and its public exhibition
to be directed toward what is (Quine) and how he is (wise). No grounds are apparent
to me for taking “the same” to mean anything more, ontologically speaking, than that
you and I each think/speak/write about Quine and, furthermore, that he is wise. If
this is what “the same” comes to here, then nominalists agree that there are proposi-
tions (though the claim will invite inevitable confusion), but deny that there is another
object involved whether external or internal to the mind of God. If one insists that “the
same” requires an object besides Quine and any of our signage relevant to isolating
and characterizing him, then nominalists parts company and ask why “the same”
signals anything more than what the nominalist already recognizes—you and I agree
about Quines wisdom.

   If the nominalist's complaint is answered sufficiently, the identity secured by each
of us thinking God’s thought that Quine is wise brings with it a new problem. My
thought, on this account, is no longer primarily related to Quine (though it may be
so directed). Lapsing into metaphor perhaps, my thought must be God’s, where God’s
thought is “in” God’s mind. Now, I am no more in God’s mind—even when God is
thinking of me—than Quine is in mine when I think of him, lest the view lapse into
the Berkleyan idealism Gould/Davis seek to avoid (59). How exactly is it that the very
same thought that is God’s is mine? If my thought is a physical going-on in my brain,
I do not see how it can be, but even if my thought is in my mind beyond space and
time, I still do not see how my thought can literally be God’s. Platonists under-tell



[ Page ] 72

the story regarding how we are related to sui generis platonic objects and MTAism
requires something beyond the assertion that we all think the same (divine) thoughts
when we think anything. Since the argument for MTAism begins with the failure of
platonists to put the whole of their platonistic picture together into a coherent whole
by incorporating it with what we think we know about truth, truthmakers, etc.,
MTAism must not commit the same error. Since we now have God, Quine, you, and
me involved, we require a bit more to put the pieces together. How do I get access to
God’s thought? How is my getting access to that thought different from when God
reveals to me that Quine is wise? How is it different when I falsely (let us say) think
that Quine is foolish when God thinks (truly) that Quine is wise? When I think that
Quine is wise, I do not think that Quine is foolish. If propositions are Gods thoughts,
then assuming that mistaken beliefs whose contents are propositions are possible,
then God must think Quine foolish, since there is no independent proposition for
God to entertain, while also thinking him wise. How, exactly, does all of that go,
especially the identity claim?

Graham Oppy

Gould and Davis defend a conceptualist account of propositions: They claim (*) that
propositions are thoughts in the mind of God. In arguing for (*), they make a number
of controversial assumptions, including: (1) that there are propositions; (2) that
propositions are claims or assertions; (3) that propositions are truthbearers; (4) that
propositions are essentially intentional objects; and (5) that propositions are mind-
dependent entities.

   Ad (1): In the context of a general inquiry into abstract objects, it cannot just be assumed that there are propositions. After all, we might decide that we should be
fictionalists about them. And it is not hard to see the utility that talk of propositions
brings even if there are no propositions. In particular, there are various kinds of
generalizations that are much more easily stated if we allow ourselves to talk about
propositions as objects. Consider for example: If the Pope says it, then it must be true.
Without the fiction of propositions, we would be left with an infinite conjunction, or
an infinite number of instances of a schema, or the like. Or consider: Mary believes
what Tom doubts. Even if we could have said Mary believes that so-and-so but Tom
doubts that so-and-so, the wheels of conversation are greased if we pretend that there
are propositions over which we can quantify (and, in any case, we can say Mary
believes what Tom doubts even if there is no sentence of the form Mary believes that
so-and-so but Tom doubts that so-and-so that is available to us). At the very least, in
the light of these observations, a case needs to be made that we should prefer realism
to fictionalism about propositions.	.

   Ad (2): If there are such things as propositions, then propositions are the contents
of claims and assertions. If I assert that it is warm, then the content of my assertion is
the proposition that it is warm. If I claim that it is warm, then the content of my claim
is the proposition that it is warm. But, equally, if I wonder whether it is warm, then



[ Page ] 73

the content of my wondering is the proposition that it is warm. And yet, of course, my
wondering involves no claim or assertion. This seems to me to be sufficient to establish
that, if there are such things as propositions, then propositions are not assertions or
claims. Moreover, it also seems to me to be sufficient to establish that propositions do
not represent reality as being some particular way: rather, someone’s taking an appro-
priate attitude towards a proposition may represent reality as being some particular
way. If, for example, I assert that it is warm, then my assertion represents (local) reality
as being warm; but if I wonder whether it is warm, then my wondering does not
represent (local) reality as being any particular way.

   Ad (3): If there are such things as propositions, it may be so that there is some
sense in which propositions are truthbearers. But there are other good candidates for
truthbearers, and there are important questions to ask about which candidates are
base-level truthbearers. Some truthbearers are verbal tokenings. If I say, at a particular
time and place, that it is warm, then, if it is warm at that particular time and place, I
speak truly. More carefully: If I produce a verbal token—“It is warm”—at a particular
time and place, then, if it is warm at that particular time and place, I speak truly in
producing that particular token. Of course, my verbal token must be produced with
appropriate intent: If I produce my token while acting on the stage, then, even if it is
warm on the stage, I shall likely not have spoken truly in producing my token. And
doubtless there are other qualifications that should also be introduced. But, at the very
least, it is often so that I speak truly, in producing a particular verbal token, just in case
things are as my verbal tokenings represents them to be.

   Some truthbearers are other kinds of tokenings. If I produce a written token—“It is
warm”—at a particular time and place, then, if it is warm at that time and place, I write
truly in producing that particular written token (given the satisfaction of other relevant
conditions). If I produce a particular mental token—“It is warm”—at a particular time
and place, then, if it is warm at that time and place, I think truly in producing that
particular mental token (perhaps given the satisfaction of other relevant conditions).
And there may be other kinds of tokenings that are also truthbearers.

   On the view that I favor, tokenings are the base-level truthbearers, and—perhaps—
tokens and propositions are derivative truthbearers. (Whether verbal tokenings are
more basic truthbearers than mental tokenings is a question that need not be decided
here.) If there are propositions, then tokenings involve both tokens and propositions—
“vehicles” and “contents”—but the base-level truthbearers are the tokenings, and not
either the tokens or the propositions. (A view to which I might be forced to retreat is
that tokenings, tokens, and propositions are all truthbearers, none of which should be
supposed to be more fundamental truthbearers than the others.)

   Ad (4): Tokenings are intentional—tokenings are about things. So, for example,
thought tokenings are intentional—thought tokenings are about things. If I have a
thought that Quine is wise, then I have a thought about Quine. If we think of my
thought tokening as being a tokening of “Quine is wise,” then my thought tokening
is about Quine because it contains a tokening of “Quine” and—in appropriate condi-
tions—tokenings of “Quine” are about Quine (because they refer to Quine). Given
that tokenings are intentional, and that propositions are the objects (or contents)



[ Page ] 74

of tokenings, we can allow that there is a sense in which propositions are inten-
tional objects—but it seems to me that this intentionality of propositions is clearly
“secondary” or “derivative.” (If you are worried about talk of thought tokenings,
it should be clear that I can run exactly the same line in connection with verbal
tokenings. Verbal tokenings are intentional, and their intentionality depends upon the
referential properties of sub-tokenings of names, pronouns, and so forth.)

   Ad (5): Given the distinction between tokenings, tokens, and propositions, it is,
I think, quite clear that, while tokenings are mind-dependent objects, tokens and
propositions need not be. Certainly, once produced, written and verbal tokens are
not mind-dependent; but, if there are mental tokens then, of course, they are mind-
dependent. However, crucially, if there are propositions, then there is simply no reason
at all to suppose that propositions are mind-dependent. If there are propositions, then
propositions are the contents of tokens involved in particular tokenings—but, as just
noted, it is the tokenings that are the clearly mind-dependent entities.

   Ad (*): As argued in my original paper, the claim, that propositions are thoughts
in the mind of God, is ambiguous. Thoughts can be, variously, tokenings, tokens, or
contents. For reasons already argued, propositions cannot be tokenings or tokens in
the mind of God: Each of these suppositions simply involves a category mistake. But,
if God exists, and there are propositions, while propositions will be the contents of
divine thought tokenings, propositions simply will not be ontologically dependent
upon those divine thought tokenings. After all, propositions could not play the role of
being the contents or objects of divine thought tokenings unless they were ontologi-
cally independent of those divine thought tokenings.

Notes

1 Davidson’s argument (ibid.) seems to preclude a TA account of possible worlds as
  well. If God creates them, He causes God exists to be true in all possible worlds. Since
  this includes the actual world, God causes His own existence in the actual world.

2 Neither do I understand Gould/Davis’s claim that “[B] ensures that the divine
  substance is a fundamental unity that is the final cause of its constituent metaphysical
  parts (including divine concepts and essential properties)—which ensures God’s
  ultimacy” (62). If by “final cause” they mean a final cause in Aristotle’s sense, then
  I do not understand how God is the final cause of His parts. If they mean by “final”
  something like ultimate, then they seem to countenance that God is the efficient
  cause of Himself—the ultimate bootstrapping trick!



[ Page ] 75

Response to Critics
Paul M. Gould and Richard Brian Davis

The Bootstrapping Worry

Craig, Welty, and to a lesser extent, Yandell, press worries regarding MTA’s suggested
way out of the bootstrapping objection. In our lead essay, we suggest that the
bootstrapping objection can be avoided by endorsing two claims:

	[A] God’s essential platonic properties exist a se (i.e. they are neither created nor
	sustained by God, yet they inhere in the divine substance); and

	[B] Substances are Aristotelian.

In his reply, Craig suggests “this Aristotelian view ... yields the palm of victory to the
anti-platonist” (69) for “things can be powerful or wise or brown without being so in
virtue of standing in an exemplification relation to an abstract property” (69). Craig
admits that if his characterization is not accurate, then he does not understand our
view. We concur—his characterization is not accurate and we welcome this oppor-
tunity to further clarify our position.

   Minimally, to exemplify a property is to possess or have a property. Broadly
speaking, two distinct styles of metaphysical explanation—best explained in terms
of the concept of “ontological structure” (van Inwagen 2011)—can be discerned for
understanding property possession by ordinary concrete objects. The Relational
Ontologist thinks that ordinary concrete objects have no ontological structure; that
the only parts an ordinary concrete object has are its everyday mereological parts;
properties are possessed/tied-to/exemplified by the substances that have them even
as they stand "apart” from them. The Constituent Ontologist thinks that ordinary
concrete objects have an ontological structure; that the familiar objects of our
everyday experience have constituent metaphysical and physical parts; properties
enter into the being of the substances that have them by being nonmereological parts
of it (see also Wolterstorff 1991).

Both approaches tell us that substances exhibit whatever character they have in
virtue of properties had by it. Properties characterize substances. How, on either
approach, is property possession assayed? Consider the following sentence:

	(1) God is divine.

On the relational approach, (1) can be further analyzed as:

	(2) God exemplifies being divine.

God stands in a relation or tie to the property being divine. On the constituent
approach, (1) can be further analyzed as:



[ Page ] 76

	(3) Being divine inheres in God as a constituent.

and

	(4) God’s individuator exemplifies being divine.

Sentences (3) and (4) are understood as follows: the divine substance (that is, God)
has as a constituent the property being divine. Further, the property being divine
is exemplified by an individuator that is also a constituent of the divine substance
(perhaps a bare or thin particular).1 The property inheres in the substance and is
exemplified by some individuator (which also inheres in the substance). Thus, Craig
is mistaken in his claim that [A] is “just anti-platonism with respect to God” for
God (and other substances) exhibit the character they do in virtue of standing in the
exemplification relation to the abstract properties they possess.

   Still, our parenthetical remark in [A] tips our hat: we are inclined to think the
constituent approach is the way to go here. But, we need not limit ourselves to a
constituent approach in avoiding the bootstrapping objection. Rather, we could (and
perhaps should) have said the following:

	[A*] God’s essential platonic properties exist a se (i.e. they are neither created nor
	sustained by God, yet they are exemplified by the divine substance)

[A*] highlights the fact that both the relational and constituent approach to property
possession is open to the defender of MTA: Either God exemplifies platonic properties
that exist extra se (on the relational approach) or that inhere within the divine
substance (on the constituent approach).2

   Further, on [B], to say that God is an Aristotelian substance is to endorse the claim
that God is a final cause of all His parts. As a final cause, the divine substance explains
why it possesses the essential properties that it does and why it possesses the essential
properties it does as a deep unity. God is not the “efficient cause of Himself” (74n. 2)
as Craig wonders, rather the whole—as a final cause—is logically prior to its parts and
the fundamental ground of its parts—hence, God is ultimate in explanation.

   Welty suggests that God’s creating propositions such as (E) God exists simply
relocates the bootstrapping worry. If propositions are divine thoughts brought about
by the activity of thinking, and causing something to exist is to cause its essence to
be exemplified, then in bringing about (E), God causes the property being true to be
exemplified by (E), thereby causing His own existence. Formally stated, the argument
seems to go as follows:

	(5) If God causes (E), then God causes the property being true to be exemplified
	by(E).

	(6) If God causes the property being true to be exemplified by (E), then God causes
	His own existence.

	(7) God causes (E).

	(8) Therefore, God causes His own existence.



[ Page ] 77

The principle grounding (6) seems to be that “x makes true y” entails “x causes y” or
(more accurately and awkwardly, since y is a proposition) “x causes the object y is
about.” Far from it. Socrates’ drinking the hemlock makes it true that Xantippe is a
widow, but it doesn’t follow that Socrates’ drinking the hemlock causes her to become
a widow (Davidson 1999, 283). Hence, we reject (6), and thus avoid the incoherency
charge in this iteration of the bootstrapping objection. On MTA, God is the creator of
all propositions, and in the Biggest Bang (see Leftow 2012) God freely (and eternally)
thinks up all possible creatures and possible states of affairs thereby fixing all necessary
alethic truths and falsehoods. On this picture, it makes no sense to say that God
thereby brings about His own existence!

   Yandell claims that the nontrivial truth of

   (9) If abstracta did not exist, then God would not exist

entails that “God depends for God’s existence on there being the abstracta that God
must create” (66). In our reply to his lead essay we respond to this worry in greater
detail. Here let us say the following. On MTA, the nontrivial truth of (9) is grounded in
the fact that God is the creator of all abstracta (setting aside His essential properties).
While it is true that the absence of abstracta entails God’s absence, it is false that God
thereby depends for His existence on abstracta. The existential dependence is the other
way around given the causal asymmetry between God and abstracta. There is no otiose
dependency in view, and no divine bootstrapping on MTA.

Worries regarding divine conceptualism

There are two substantial objections to MTA that remain. Both concern the concep-
tualist understanding of propositions. A proposition, we say, is a truthbearer. Further,
possessing truth entails being an intentional object: An object that is of or about
something, that represents the world (or some part of it) as being thus and so. This
is precisely what concrete sentence inscriptions and platonic3 abstracta cannot do—
not in themselves anyway.4 Thoughts, by contrast, are indisputably and essentially
intentional. They are, says the proponent of MTA, composite wholes: specific mental
arrangements of (brutely) intentional parts—Locke’s “invisible ideas.”

   Of course if all we have at our explanatory disposal is human mental activity, the
lifespan of a proposition will be fleeting indeed, lasting only as long as we happen to
be thinking it, and existing in just those worlds where we do so. In short, there won’t
be any necessary truths (/falsehoods). The best course is to recognize a supra-natural
realm of divine cognitive activity. A Proposition (capital ‘P’) is a divine thought: an
ordered arrangement of divine ideas. Propositions are abstract in the sense of being
multiply instantiable in human minds; but they’re not mind-independent abstracta of
the platonic variety. (Thus contra Craig and Yandell, there is no incoherence in saying
that God is a concrete being whose thoughts are abstract.) It doesn’t follow though,
as Shalkowski fears, that our thoughts are literally identical with God’s. Instantiation



[ Page ] 78

doesn’t entail identity. Our thoughts, as we say, are mere approximations and partial
graspings of those in God’s mind. Now on to the objections.

   The Composition Objection. It is “not obvious” (says Yandell) and “not clear” (says
Welty) that propositions have parts—namely, concepts or ideas. This may be so. But then
again we don’t so much as slyly suggest that our “parts proposal” has that irresistible and
evident lustre Locke extolled. It’s not a deliverance of introspection; it’s an inference to
the best explanation. That it lacks Lockean luminosity in no way counts against it.

   Unfortunately, Yandell and Welty claim that the “parts proposal” isn’t even a
good explanation. Thus Welty comments: “This is part and parcel of Russells (failed)
analysis of propositions” (67). But that’s not why Russell’s analysis failed: because it
held that propositions have parts. It failed because it invoked the entirely wrong sort
of parts—parts that were painfully nonintentional (e.g. Quine himself), and thus
impotent to account for a proposition’s being about anything.

  Yandell’s concern lies elsewhere. We should reject the “parts proposal,” he tells us,
because “propositions are primary” and concepts are merely .“propositional abilities
that persons have.” This follows if we accept this claim: “that S has the concept of being
a cow insofar as S knows what is true if and only if there are cows” (66). Here Yandell
seems to say that you have the concept being a cow only if you know there are cows is
a true proposition. But surely that is incorrect. Graham Oppy, we may assume, grasps
or apprehends the concept of God. But it hardly follows that he knows there is a God.
He’ll be the first one to tell you that! In general, to grasp the concept of being an F, you
don’t have to know that there are any Fs. There is no threat to the “parts proposal” here.

   The Content Objection. Oppy isn't inclined to think that propositions are truth-
bearers; nor are they to be identified with thoughts. Rather, on his favored view, there
are tokens (e.g. the inscription “Quine is wise”), propositions (e.g. the content of “Quine
is wise”), and tokenings (e.g. my producing the token “Quine is wise”). Oppy’s claim is
that only tokenings are truthbearers, since only tokenings are intentional.

   There are two principal difficulties with this suggestion. First, tokenings aren’t
truthbearers. A tokening—whether verbal, written, or thought—is an act or event.
However, as C. S. Lewis so aptly put it, “events in general are not ‘about’ anything
and cannot be true or false” (1947, 21). The slamming of a door, for instance, isn’t
“about” anything. It’s the sort of thing that “happens” or “occurs.” But it’s nothing
even remotely like an intentional object that could represent a state of affairs. Thus it
couldn’t possibly be true or false. The same thing goes for tokenings.

   Secondly, why think that thoughts and propositions are distinct? On Oppy’s view,
a thought tokening is the producing of a thought—a mental token. Suppose we agree.
Now consider this token:

	(T) The thought that Quine is wise.

The propositional content of (T) is presumably

	(P) Quine is wise.

Oppy declares that (T) and (P) are distinct. This isn’t obvious. If propositions can
exist in worlds where there are no human beings, and indeed no concreta at all (as the



[ Page ] 79

proponent of MTA holds), then (P) won’t be a physical object. It will be something
nonphysical: an abstract platonic object, perhaps, or an immaterial thought. Now (P)
is about Quine and represents him as being wise; hence, it can’t be a platonic object.
Like it or not, then, we must say that (P) is a thought. Which thought? Well, obviously,
the thought that Quine is wise. But notice that’s just (T) itself. Contra Oppy, therefore,
there really is no difference between (T) and (P). At any rate, thus sayeth the modified
theistic activist.

Notes

1 There are sophisticated defenses of bare particulars in Moreland and Pickavance
  (2003) and Sider (2006). For those who feel a little queasy, think of an individuator in
  Wolterstorff’s (1991, 543) way: a particular that just is a nature.

2 For more see Gould (2011a and 2013) and Moreland (2001 and 2013).

3 The astute reader will notice that we use “platonic” in a more technical sense when
  talking about intentionality. In one sense, as Gould makes clear in the Introduction
  (n. v), “platonism” is just the view that abstract objects necessarily exist. In this sense,
  we agree that propositions are in fact platonic. But, we deny that propositions are
  “platonic” in the sense that they exist in Plato’s heaven as brutely intentional entities.

4 Welty asks why platonists can’t “just infer that there are these intentional entities,
  and that is that” (67). The answer is that the inferences involved in indispensability
  arguments simply don’t deliver “intentional entities.” They give you—at best and if
  they work—only indefinite entities that play the role of truthbearers. Intentionality
  is “bestowed” after the fact. But simply to announce here that platonic abstracta are
  intentional is wholly gratuitous and unsupported. Not so, of course, for thoughts,
  whose intentionality is known by introspection, not indispensability.

***** This is the end of the e-text. This e-text was
brought to you by Tyndale University, 
J. William Horsey Library - Tyndale Digital Collections *****