(P1)God has many genuinely distinct essential attributes.
+Scripture presents God through a rich vocabulary of perfections: he is holy, loving, omnipotent, omniscient, just, and self-existent. An essential attribute is one a thing cannot lack and still be itself, the way a triangle cannot lack three sides, and each of these belongs to God that way.
(1) Scripture describes God through many distinct perfections.
- He is holy, set apart from all creation and from every moral imperfection (Isaiah 6:3, Revelation 4:8). He is loving, directing his care and covenant faithfulness toward creatures and supremely toward those made in his image (1 John 4:8, Deuteronomy 7:9). He is omnipotent, sovereign over all things with unlimited power (Revelation 19:6, Job 42:2).
- He is omniscient, knowing all things down to the thoughts and intentions of the heart (Psalm 139:1-4, Hebrews 4:13). He is just, never acting against moral perfection (Deuteronomy 32:4, Romans 3:26). He is self-existent, drawing his being from no source outside himself (Exodus 3:14, Acts 17:25).
- These are not one feature approached from six angles. Learning that God is holy does not already tell you what learning that he is loving adds, and a reader who knew only that he was omnipotent would still have to be told that he is just. Scripture treats them as distinct because they are distinct.
(2) Any doctrine of divine simplicity has to answer to that vocabulary.
- Ask a child what a bicycle is made of and she can take it apart for you: wheels, frame, chain, saddle. Ask what God is made of and the church's old answer is that he is not made of anything. He was never put together. That teaching is divine simplicity, and everyone who holds it, in any version, agrees on at least that much.
- The disagreement is over how far to push it. The strong doctrine, the classical form taught by Aquinas, adds the identity thesis: each of God's attributes is identical with his essence (what he is, the nature that makes him God rather than anything else) and with every other attribute, so that his wisdom simply is his power and his power simply is his holiness. That claim protects God's independence from everything outside him, and it pays for the protection in philosophical difficulty. The weak doctrine stops at "never put together." It keeps the attributes distinct and asks how much simplicity can be affirmed without giving any of them up.
- Whichever version one holds, Scripture sets the floor. Any doctrine on which God has no genuinely distinct attributes, or on which positive language about him says nothing, fails on biblical grounds before its philosophical problems are even reached.
(3) The attribute-words keep their ordinary core meaning when said of God.
- When Scripture says God is wise and a wise person is wise, "wise" is not being used in two unrelated senses. If it were, "God is wise" would tell a reader nothing at all, like using one word for a colour and for a temperature: grammatical, and empty.
- Nor is the word used in exactly the same way, since God's wisdom is unlimited and ours is not. The core meaning is shared and the degree is maximal. Duns Scotus pressed this point seven centuries ago: unless "wise" keeps some one meaning across its human and divine uses, no reasoning from creatures could reach God at all, and every argument of natural theology would fail at its last step (Ordinatio I, d. 3, pars 1, qq. 1-2).
- The strong doctrine does not deny this much. Aquinas held that names like "wise" and "good" are said of God properly rather than as metaphors, and that "wise" and "powerful" are not synonyms when said of him (Summa Theologiae I, q. 13, aa. 3-4). His theory of analogy, the view that a word said of God keeps its human meaning while stretching it, is his account of how far the stretch goes. Whether the stretch goes too far is one of the things the two doctrines argue about (D13). P1 needs only what both grant: the attributes differ in content, and each is truly said of God.
Therefore, any acceptable doctrine of simplicity must leave God with many genuinely distinct attributes, each truly and informatively said of him.
(P2)A thing's properties are not parts or constituents out of which it is composed.
+Everything in the weak doctrine turns on what a property is. If wisdom, power, and holiness are ingredients God is put together from, the way a wall is put together from bricks, then a God with many attributes is a God made of parts, and this premise says that whatever a property turns out to be, it is not an ingredient of the thing that has it.
(1) Philosophers disagree about what, if anything, a property is.
- Socrates is wise. Gold is yellow. God is omnipotent. Each sentence has the same shape: a subject, then a predicate, the part that says something about it ("is wise," "is yellow," "is omnipotent"). Saying it is a predication. The question is what has to exist in reality for the predication to be true, and that is an ontological question, ontology being the study of what there is, as opposed to what we merely say or think.
- Property realists say a real abstract entity answers to each predicate. An abstract entity exists outside space and time; it has no location, no weight, and no physical features, and it is what a mathematician has in mind who thinks the number 7 would exist even if there were nothing left to count. Wisdom-itself, on this view, is such an entity. Socrates exemplifies it, meaning he is an instance of it, the way a red ball is an instance of redness, and wisdom-itself would exist whether or not anyone had ever been wise. A property many things can share at once is also called a universal.
• Some realists go further and make the property a constituent of the thing that has it: a genuine building-block, as bricks are constituents of a wall and hydrogen and oxygen are constituents of water, except that this block sits at the level of being rather than of matter. On that picture Socrates is partly made of his wisdom.
- Property anti-realists deny that any such extra entity is needed. A true sentence of the form "a is F" requires the concrete individual a, and nothing called F-ness over and above it. The truthmaker for "gold is yellow," meaning whatever it is in reality that makes the sentence true, is simply gold, being the kind of stuff it is.
(2) A predicate describes its subject without adding an entity to the world.
- Start with something W. V. Quine noticed in 1948. When you say "Socrates is wise," the words "is wise" do their whole job without your having to believe in a separate thing called wisdom-itself. Reading every predicate as the name of an abstract object, he argued, is a grammatical illusion rather than a metaphysical necessity. That half of his view is exactly what anti-realism needs: a predicate describes, and it does not add an entity to the world.
- Quine did not stop there, and here the weak doctrine parts company with him. A criterion of ontological commitment is a rule for reading off, from what a person says, what they have thereby claimed exists. Quine's rule says you are committed to whatever your best theory says "there is." In his famous slogan, to be is to be the value of a bound variable, which is a logician's way of saying that whatever your "there is" sentences range over, you have signed for.
• Physics cannot be stated without saying there are numbers, so Quine concluded that numbers exist. By the same rule, a theory that says "there is a property the two apples share" is committed to that property. Taken whole, his criterion is the engine of Platonism (the view, named for Plato's Forms, that abstract objects like numbers and properties really exist), not a way out of it.
(3) An ordinary "there is" is not a census of reality.
- The way out is neutralism, the view that neither the words "there is" nor a name commits you to the existence of anything. Listen to ordinary speech. There is a way you smile. There is a faster way to get to work than the highway. There is a lack of compassion in the world, and something the two of us have in common, and infinitely many prime numbers.
- Nobody who says these things believes they have just added five new residents to reality, and the neutralist says they are right not to. Ordinary "there is" is not a census. So "God is wise" and "wisdom is one of God's attributes" can both be true without wisdom-itself being anything at all.
- Neutralism did not begin as a theological convenience. Its roots run back through a century of work in logic and the philosophy of mind that had no interest in the divine attributes, and the charge that it was invented to rescue God is answered on its own ground (D6).
- Anti-realism is contested, of course, and property realism is held by serious philosophers on the other side. It is the more direct of two routes to this premise, since if there are no properties there are certainly no property-parts, and it is the route the weak doctrine's own defenders take. The objection that properties plainly exist is answered on its own (D1, D11, D12).
(4) Even a property realist can deny that properties are parts of things.
- One of the most careful property realists writing today believes there are abstract properties and still holds that they are not constituents of the things that have them. A red apple is related to redness, not built out of it. He offers no analysis of that relation (call it exemplification, instantiation, or having), takes it as basic, and finds the rival talk of properties as constituents of things unintelligible.
- This is the relational picture, the opposite of the constituent picture: a thing stands in a relation to its properties, which lie outside it, rather than containing them as parts. Think of it this way: having many properties no more composes a thing than having many admirers does. The admirers are real, you are genuinely related to every one of them, and not one of them is a piece of you.
- The Christian has a reason of his own to prefer the anti-realist route. If wisdom-itself existed on its own, uncreated and necessary, God would not be the only uncreated reality, and his aseity, his existing from himself and depending on nothing outside him, would be in question. The realist who takes the relational route owes an answer to that worry from somewhere else (D10); the anti-realist does not.
- The simplicity conclusion does not wait on that answer. Whether a reader ends up an anti-realist or a relational realist, nothing in the properties makes God a compound, and that is all this premise claims. Its denial is constituent ontology, the view that a thing's properties are ingredients it is built from, and that view is one camp among three rather than the default.
(5) Anti-realism does not make talk about God empty.
- You might think: "If there is no wisdom-itself in God, then 'God is wise' is just something we say about him." That is a real position, and it has a name. Theological nominalism holds that the attribute-words are only labels we attach to God, with nothing in him answering to them, so that "God is wise" reports our way of talking rather than anything true of God. It is emphatically not the weak doctrine's position.
- The distinction that matters is between "this predication is true" and "this predication commits us to a distinct abstract entity." The anti-realist holds that "God is wise" is genuinely, factually, informatively true. What he denies is the further claim that its truth requires a thing called wisdom-itself, either as a constituent of God or as a separate necessary being alongside him.
- Telling a friend that your grandmother was patient, funny, and stubborn gives three real and different pieces of information about one woman, and nobody supposes you have posited three abstract objects she was assembled from. In the same way God can be genuinely and distinctly omniscient, omnipotent, and holy, each predication fully meaningful and adding what the others do not, without any of them naming a separate entity that composes his being.
Therefore, having a property is not having a part, and God's many attributes are many true things about one God rather than many pieces of him.
(P3)If properties are not constituents of their bearers, having many distinct attributes does not entail composition.
+This premise is the hinge of the argument: once properties are not ingredients, there is no way to get from "many attributes" to "many parts."
(1) Without property-ingredients, the classical argument for the identity thesis loses a premise.
- The classical argument for the strong doctrine runs in five steps:
• God has many attributes.
• Those attributes are distinct abstract entities or metaphysical constituents.
• If they are distinct from each other and from God, then God is composed of them and depends on them.
• Composition and dependence are incompatible with aseity.
• Therefore the attributes must all be identical with God and with each other.
- The argument is valid. But its second step is constituent ontology applied to God: it treats the attributes as ingredients he is built from. If P2 is right, that step is false and the argument does not go through.
- The third step needs the ingredients too. God "depends on" his attributes only if they are something he is made of, or something outside him that does the work of making him what he is. On the anti-realist route there is nothing of either kind. On the relational route the properties do no such work: God is powerful, and power-itself merely answers to the word, so he leans on it for nothing. What the relational realist still owes is not an answer about composition but an answer about whether anything uncreated exists alongside God, which is the aseity question (D10).
(2) A truthmaker account shows how the weak view works in practice.
- A truthmaker is whatever in reality makes a true statement true (P2). The truthmaker for "the cat is on the mat" is the actual cat in its actual place, and truthmaker theory asks, of each true statement, what must exist for it to be true.
- The truthmaker approach to divine simplicity was worked out by a defender of the classical doctrine, who used it to make sense of the traditional identity claims: "God is his wisdom" becomes "God himself, and nothing else, is what makes 'God is wise' true." The weak doctrine borrows the account and declines the identity language, since on anti-realism there is no wisdom-itself for God to be identical with (D13).
- So "God is omniscient" is true because of God, "God is omnipotent" is true because of God, and "God is holy" is true because of God. There is no abstract entity for each predicate and no constituent for each attribute. One concrete reality makes all these distinct predications true by being the kind of being he necessarily is.
- The distinctions among the attributes are real distinctions in content and meaning. They are distinctions in what is said of God, not divisions in him.
(3) What is true of God essentially is made true by God alone.
- Many philosophers hold that a creature's contingent features, the ones it could have lacked, need truthmakers beyond the creature itself. One popular candidate is the trope, a particular instance of a property belonging to one individual: the wisdom of Socrates is his own particular wisdom, not wisdom-itself, and it is something he acquired and could lose. Trope theorists count it as a constituent of Socrates while he is wise. A thoroughgoing anti-realist says instead that Socrates, as he now is, makes "Socrates is wise" true and no further item is needed. The weak doctrine can leave that quarrel about creatures open.
- God differs from creatures in two respects that settle the matter for him whatever is said about them. For one thing, every one of his essential attributes is necessary: he cannot fail to be omniscient, omnipotent, or holy, so there is no wisdom he acquired and none he could lose. For another, God exists necessarily, so his being the way he necessarily is suffices to make every essential predication about him true, with no intermediate property-instances and no building-blocks.
- Whether God also has contingent states, his knowing that Peter exists or his willing this world rather than another, is a further question (D4). A contingent state is neither a separable part nor an ingredient, so the answer either way leaves P4 standing.
Therefore, a being with many distinct attributes is composed of nothing on that account, and the argument that drove the tradition to the identity thesis loses the premise it cannot do without.
(P4)God is a spiritual substance with no separable parts and no metaphysical constituents.
+This premise is the positive half of the doctrine: what simplicity, in its weak form, asserts about God. A substance is a thing that exists in its own right rather than as a feature of something else, a horse rather than the horse's colour, and God is a substance with no matter in him.
(1) Simplicity means non-composition, not attribute-identity.
- The weak doctrine does not identify the divine attributes with each other or with the divine essence. It requires only that God not be assembled from separable parts, whether physical or metaphysical.
- God is not made of matter. He is not a compound of form and matter in Aristotle's sense, where matter is the stuff and form is what organizes it into this kind of thing, as bronze and a shape make a statue. He has no parts that could be divided from each other and leave him diminished or destroyed by their removal. And given P2, he has no property-constituents either.
- Stated as a formula, the weak doctrine reads: "(DS₄) As a spiritual substance, God is not composed of separable parts; neither does he have metaphysical constituents." The label is the one its defenders use, the numeral marking it as one formulation among the several the tradition has produced, and the formula keeps the legitimate core of the historical doctrine without the contested identity thesis.
(2) Scripture and aseity together warrant the premise.
- Scripture's side is plain. God is spirit (John 4:24), so there is no matter in him to divide. He names himself as the one who simply is (Exodus 3:14), and he is served by no one as though he needed anything (Acts 17:25). Nobody assembled him, and he leans on nothing.
- Aseity carries the rest. Think of a house. It depends on its bricks, on its mortar, and on its builder, and take the bricks away and there is no house. Anything with separable parts is like that: it depends on those parts, on their arrangement, and on whatever arranged them. A being who exists from himself, with nothing before him, cannot be a thing of that kind, because there was nothing to supply the parts and no one to assemble them.
- The same reasoning bars ingredients at the level of being, if there were any. A God built from wisdom-itself and power-itself would depend on them as prior components, which aseity forbids. P2 says there are no such ingredients in anything; aseity adds that there could be none in God in any case.
(3) DS₄ says more than that God is immaterial.
- You might think: "Of course God is not made of physical parts. He is a spirit. So DS₄ says nothing." But every clause of DS₄ denies a picture of God that serious theologians have actually held, and a claim that rules out real positions is not a trivial one.
- Consider what classical theology said about angels. Aquinas held that even angels, purely spiritual as they are, are composed of essence and existence as distinct metaphysical principles. What an angel is (its essence) is one thing; the bare fact that it exists is another; and so something outside the angel must account for why that nature exists at all rather than remaining merely possible. God alone is not like this. In him, essence and existence are not two items needing something to unite them. God simply is.
- The weak doctrine need not adopt that analysis of angels, and it does not claim that simplicity by itself sets God apart from every created spirit. What sets him apart is aseity (P4.2): for every creature, angels included, the question why it exists has an answer outside it, and for God it has none, because there is nothing outside him to give one. DS₄ is what aseity requires at the level of composition. It is the same denial the tradition made when it said God is not compounded of form and matter, of essence and existence, or of substance and accident (an accident being a feature a thing can gain or lose while staying itself, like a tan).
- So the doctrine does real work without the identity thesis. It rules out God's being assembled, or depending on anything more fundamental, or having a unity that the separation of parts could threaten. It also explains why his attributes are not separable from him as a creature's are: Socrates can lose his wisdom and keep his strength, and God cannot lose his omniscience while keeping his omnipotence. Joined to the attributes established by the cosmological, ontological, and teleological arguments, it gives a God who is genuinely one and was never put together, at none of the strong doctrine's metaphysical cost.
Therefore, God is a spiritual substance with no separable parts and no metaphysical constituents, which is exactly what the weak doctrine means by calling him simple.
(C1)Therefore, God is simple and yet has many genuinely distinct attributes.
+God is simple, God has many genuinely distinct attributes, and there is no contradiction in saying both. The argument is deductive: if the premises hold, the conclusion cannot slip out, and each premise needs only to be more plausible than its denial. P1 gives God many distinct attributes, P2 and P3 keep them from being parts, and P4 says what simplicity then means: no separable parts and no metaphysical constituents. Put together, God is both genuinely many-attributed and genuinely simple. That conclusion fits four things a Christian doctrine of God has to fit, and it has one cost, which is worth stating plainly.
(1) Scripture's positive revelation of God stays meaningful.
- Scripture says God is omniscient, omnipotent, loving, holy, and just, and expects the words to mean something. The weak doctrine keeps that meaning whole: God's being omniscient is a truth about him genuinely distinct from his being omnipotent, conveyed in words that carry their ordinary meaning at unlimited degree (P1).
(2) The Trinity fits without strain.
- The strong doctrine has struggled with the Trinity from the start, since three persons who are each identical with one essence ought to be identical with each other. The thinkers most rigorous about the strong doctrine outside the church all concluded that it ruled the Trinity out: Eunomius among the Arians (the fourth-century party that denied the Son's full deity), the Muslim philosopher ibn Sina, and the Jewish philosopher Maimonides, who said that those who confess one God with many attributes "declare the unity with their lips, and assume plurality in their thoughts" (Guide for the Perplexed I.50).
- The weak doctrine has no such difficulty. The three persons share one simple, non-composite nature, wholly present in each. Their distinctions are real, not reducible to each other, and require no person to be a metaphysical part of the nature.
(3) God's freedom in creation is preserved.
- The strong doctrine faces the modal collapse objection, which its defenders dispute. "Modal" concerns what is possible and what is necessary, and the worry is that if God's act of creating is identical with his necessary essence, creation turns out to be necessary rather than free.
- The weak doctrine faces no such problem. God is necessarily omnipotent, omniscient, and holy; whether he creates, and what he creates, are contingent matters of his free will, and his creating this universe adds no constituent to his nature. It is a free exercise of the power he necessarily has.
(4) The weak doctrine is philosophically leaner.
- It needs no constituent ontology and no identity thesis, the claim that has generated centuries of difficulty. On its anti-realist route it carries no Platonic abstract objects either. On the relational route it carries the realist's abstract objects and still no property-parts.
- Either way it asks only that God be a spiritual substance with no separable parts, plus the independently motivated denial that properties are ingredients, and it keeps everything theologically essential in divine simplicity (D5).
(5) The case rests on the denial of constituent ontology.
- Its central moves, that distinct attributes do not entail composition and that God can be the sole truthmaker for every essential predication about him, do not go through if properties are ingredients. A reader who finds that picture more natural, so that a wise thing has wisdom in it the way a wall has bricks in it, will find that the dissolving move of P2 and P3 does not grip. For that reader the strong doctrine's route, which accepts the problem and resolves it through the identity thesis, is the one to weigh.
- The two doctrines are less rivals than alternative routes from different starting points, and neither is free of difficulty. The weak doctrine faces the grounding objection, the question of what makes "God is omniscient" and "God is omnipotent" two truths rather than one (D9), and on its anti-realist route it must defend anti-realism on its own merits (D1, D11, D12). The strong doctrine faces modal collapse (C1.3) and the difficulty of reconciling attribute-identity with the personal distinctions of the Trinity (C1.2).
Primary sources for the weak doctrine and its truthmaker account:
Brower, Jeffrey E. "Making Sense of Divine Simplicity." Faith and Philosophy 25, no. 1 (2008): 3-30. (The truthmaker reading of divine simplicity, offered in defense of the classical doctrine; the weak doctrine borrows the account and drops the identity language, P3, D13.)
Brower, Jeffrey E. "Simplicity and Aseity." In The Oxford Handbook of Philosophical Theology, edited by Thomas P. Flint and Michael C. Rea, 105-128. Oxford: Oxford University Press, 2009.
Craig, William Lane. Systematic Philosophical Theology: On God: Attributes of God, Volume IIa. Oxford: Wiley-Blackwell, 2025, chap. 5 ("Simplicity"). (Primary source: Craig's DS₄ formulation, the property anti-realist grounding, and the historical survey of the doctrine's forms, D5.)
Saenz, Noël. "Against Divine Truthmaker Simplicity." Faith and Philosophy 31, no. 4 (2014): 460-474. (The direct published objection to the truthmaker reading of simplicity, D9.)
On properties, abstract objects, and neutralism:
Azzouni, Jody. Deflating Existential Consequence: A Case for Nominalism. New York: Oxford University Press, 2004. (The neutralist account of quantification that Craig adopts.)
Bergmann, Michael, and Jeffrey E. Brower. "A Theistic Argument against Platonism (and in Support of Truthmakers and Divine Simplicity)." In Oxford Studies in Metaphysics 2, edited by Dean W. Zimmerman, 357-386. Oxford: Oxford University Press, 2006. (The theist's independent reason to resist Platonism, used in P2.)
Brentano, Franz. Psychology from an Empirical Standpoint. Translated by A. C. Rancurello, D. B. Terrell, and L. L. McAlister. London: Routledge, 1995; first published 1874.
Campbell, Keith. Abstract Particulars. Oxford: Blackwell, 1990. (Trope theory, the second anti-realist account of resemblance in D1.)
Craig, William Lane. God Over All: Divine Aseity and the Challenge of Platonism. Oxford: Oxford University Press, 2016. (Craig's own statement of neutralism, its lineage through Brentano, Husserl and Lambert, and where he parts from Quine's criterion of ontological commitment.)
Craig, William Lane. God and Abstract Objects: The Coherence of Theism: Aseity. Cham: Springer, 2017. (Craig's extended defense of anti-Platonism about abstract objects, the metaphysical foundation of the weak doctrine.)
Gould, Paul M., ed. Beyond the Control of God? Six Views on the Problem of God and Abstract Objects. New York: Bloomsbury Academic, 2014. (The six-view exchange recommended in D10; Graham Oppy replies from outside the faith.)
Hofweber, Thomas. "Ontology and Objectivity." PhD diss., Stanford University, 1999. (Source of the list of things we ordinarily say there are.)
Lambert, Karel. Free Logic: Selected Essays. Cambridge: Cambridge University Press, 2003.
Putnam, Hilary. Philosophy of Logic. New York: Harper & Row, 1971. Reprinted in Mathematics, Matter and Method: Philosophical Papers, vol. 1, 2nd ed., 323-357. Cambridge: Cambridge University Press, 1979. (With Quine's "On What There Is," the classic statement of the indispensability argument answered in D11.)
Quine, W. V. O. "On What There Is." Review of Metaphysics 2 (1948): 21-38. Reprinted in From a Logical Point of View. Cambridge, MA: Harvard University Press, 1953.
van Inwagen, Peter. "A Theory of Properties." In Oxford Studies in Metaphysics 1, edited by Dean W. Zimmerman, 107-138. Oxford: Oxford University Press, 2004. (His argument from unparaphrasable ordinary sentences to the existence of properties, D12.)
van Inwagen, Peter. "Being, Existence, and Ontological Commitment." In Metametaphysics: New Essays on the Foundations of Ontology, edited by David J. Chalmers, David Manley, and Ryan Wasserman, 472-506. Oxford: Oxford University Press, 2009. (The univocity theses and the Quinean criterion, the neo-Quinean objection to neutralism in D12.)
van Inwagen, Peter. "Relational vs. Constituent Ontologies." Philosophical Perspectives 25 (2011): 389-405.
Williams, Donald C. "On the Elements of Being: I." Review of Metaphysics 7, no. 1 (1953): 3-18. (The founding statement of trope theory, D1.)
Critics of the strong doctrine:
Hasker, William. "Is Divine Simplicity a Mistake?" American Catholic Philosophical Quarterly 90, no. 4 (2016): 699-725.
Mullins, R. T. "Simply Impossible: A Case Against Divine Simplicity." Journal of Reformed Theology 7 (2013): 181-203.
Mullins, R. T. The End of the Timeless God. Oxford: Oxford University Press, 2016.
Plantinga, Alvin. Does God Have a Nature? Milwaukee: Marquette University Press, 1980.
Wolterstorff, Nicholas. "Divine Simplicity." Philosophical Perspectives 5 (1991): 531-552.
Defenders of the strong doctrine engaged in the defeaters:
Dolezal, James E. God without Parts: Divine Simplicity and the Metaphysics of God's Absoluteness. Eugene, OR: Pickwick, 2011.
Dolezal, James E. All That Is in God: Evangelical Theology and the Challenge of Classical Christian Theism. Grand Rapids: Reformation Heritage Books, 2017. (The reading of "without parts" as including metaphysical parts, D5.)
Duby, Steven J. Divine Simplicity: A Dogmatic Account. London: Bloomsbury T&T Clark, 2016.
Feser, Edward. Scholastic Metaphysics: A Contemporary Introduction. Heusenstamm: Editiones Scholasticae, 2014.
Feser, Edward. Five Proofs of the Existence of God. San Francisco: Ignatius Press, 2017.
Leftow, Brian. "Is God an Abstract Object?" Noûs 24, no. 4 (1990): 581-598. (Reply to Plantinga's charge that the identity thesis makes God an abstract object, D3.)
On the grounding objection, the Incarnation, and the confessional history:
Hughes, Christopher. On a Complex Theory of a Simple God: An Investigation in Aquinas' Philosophical Theology. Ithaca: Cornell University Press, 1989. (Source for the grounding objection, D9.)
Morris, Thomas V. The Logic of God Incarnate. Ithaca: Cornell University Press, 1986. (The philosophical Christology behind D8.)
Muller, Richard A. Post-Reformation Reformed Dogmatics, vol. 3: The Divine Essence and Attributes. Grand Rapids: Baker Academic, 2003. (The standard study of Reformed Orthodoxy on the divine attributes; its finding of internal diversity on the attribute-essence relation is used in D5.)
Weinandy, Thomas G. Does God Change? The Word's Becoming in the Incarnation. Still River, MA: St. Bede's Publications, 1985. (The Incarnation as the coming-to-be of the assumed human nature, with no change in the divine nature, D8.)
Ancient and medieval sources:
Duns Scotus, John. Ordinatio I, d. 3, pars 1, qq. 1-2. In Duns Scotus: Philosophical Writings, translated by Allan B. Wolter. Indianapolis: Hackett, 1987. (The argument that without some one meaning shared across human and divine uses of a word, no knowledge of God from creatures is possible, P1.)
Fourth Lateran Council (1215), constitution 1. In Decrees of the Ecumenical Councils, edited by Norman P. Tanner, vol. 1. London: Sheed & Ward; Washington, DC: Georgetown University Press, 1990. (The conciliar statement that the divine essence is entirely simple, D5.)
Maimonides, Moses. The Guide for the Perplexed. Translated by M. Friedländer. 2nd ed. London: Routledge, 1904. I.50. (The strong doctrine turned against the Trinity, C1.)
Suárez, Francisco. On the Various Kinds of Distinctions (Disputationes Metaphysicae, Disputation VII). Translated by Cyril Vollert. Milwaukee: Marquette University Press, 1947. (The separability test for real distinction, D2.)
Thomas Aquinas, Summa Theologiae, I, qq. 3-13; q. 19, a. 3. (Primary text of the classical doctrine the weak doctrine is measured against; q. 13, aa. 3-4 on the divine names, q. 19, a. 3 on the contingency of God's willing.)
Biblical citations:
Exodus 3:14; Deuteronomy 7:9; Deuteronomy 32:4; Job 42:2; Psalm 139:1-4; Isaiah 6:3; John 4:24; Acts 17:25; Romans 3:26; Hebrews 4:13; 1 John 4:8; Revelation 4:8; Revelation 19:6.
+The whole argument rests on property anti-realism, but properties plainly exist. Two red things share something, redness, and without real properties nobody can explain resemblance, natural kinds, or the laws of nature.
1. Property realism is a contested position, not the default.
A property realist holds that two red things genuinely share a third thing, redness itself, an abstract entity that would exist with or without the apple and the fire truck. An anti-realist holds that both are red, that this is true, and that no third resident of reality is needed to make it so (P2). The objection treats the realist view as obvious. But it is one serious position among several, and its own central notion is famously obscure.
Think about it. What is the relation between this red apple and redness-itself? The apple is in the kitchen. Redness-itself has no location and no physical features; it is supposed to exist outside space and time, and the apple is supposed to be related to it somehow. Peter van Inwagen, a property realist and one of the most careful metaphysicians alive, says he cannot explain how a thing "has" a property, and he finds talk of properties as constituents of things unintelligible. The question is genuinely open.
2. Resemblance does not compel property realism.
That two red things share something is where the discussion starts, not where it ends, and anti-realists have done the work of explaining resemblance without universals. Nominalism (from the Latin for "name") is the family of views on which only particular things exist and general words are labels we apply to them, and it comes in more than one version:
- Resemblance nominalism: "red" applies to a thing because it resembles the paradigm red things, or the other red things, and no shared entity is required.
- Trope theory: what the apple and the fire truck share is not one abstract redness but two particular colour-instances, one belonging to each, that happen to be extremely alike. Their similarity is a basic fact rather than something explained by both pointing at the same universal.
Both are seriously defended in the published literature. Trope theorists usually go on to build things out of their tropes, which is the constituent picture P3 sets aside; what the resemblance question needs from them is only that resemblance calls for no shared universal. The objection opens a discussion. It does not close it.
3. Anti-realism is a claim about what exists, not about what sentences mean.
The anti-realism the weak doctrine uses says that a true sentence commits you to the concrete things it is about, and to nothing else. "God is omniscient" is made true by God himself, not by omniscience-itself. That is not a claim that the sentence means less, or is any less true. It is a claim about what reality has to contain for it to be true.
The position has two halves, and both are held by philosophers with no theological stake in the outcome (P2, D6). The first is Quine's: a predicate such as "is wise" describes its subject and adds no entity to the world. The second is the neutralist's: an ordinary "there is" adds none either, so that "there is a wisdom in God that he never had to learn" is no more a census of reality than "there is a way you smile." Whether the position was adopted just to rescue God is answered on its own (D6).
4. To deny P2 is to hold constituent ontology.
P2 does not claim that anti-realism is certainly correct. It claims that properties, whatever they are, are not ingredients things are built from, and that claim is denied only by the constituent ontologist. Anti-realists deny it because there are no properties to be ingredients. Relational realists deny it because properties, though real, lie outside the things that have them. That leaves the constituent picture as one camp among three, held by serious philosophers and by no means the default.
The point is not a headcount. To deny P2 one has to hold that the constituent picture is the only reasonable account of how a thing has its features, and nobody in the debate holds that: its most careful realist opponent finds it unintelligible (D1.1), and applied to God it makes him depend on ingredients, which aseity forbids (P4). A premise whose denial carries that burden is comfortably more plausible than its denial, which is all a deductive argument asks of it.
+If God has distinct essential attributes, there is a real distinction in him between his omniscience and his omnipotence, whatever we call them. Any real distinction in God is complexity, so the weak doctrine relabels the problem rather than solving it.
The strongest form of this objection comes from defenders of the strong doctrine such as James Dolezal and Edward Feser: even if the attributes are not abstract property-parts, distinct attributes are a real distinction within God, and a real distinction within God is composition under another name.
1. The strong doctrine resolves the same multiplicity in the same way.
The identity thesis does not remove the multiplicity of the attributes; it relocates it. Even on the strong doctrine, "omniscient" and "omnipotent" remain distinct in meaning, and the Thomist (a follower of Thomas Aquinas) must explain how one undivided reality grounds two distinct true predications. His explanation is analogical predication, the theory of analogy from P1: our finite minds approach the infinite God from different angles, and the multiplicity is in our knowing rather than in his being.
That is structurally the weak doctrine's own answer. Both hold that one undivided concrete reality, God himself, grounds many distinct true predications. The strong doctrine adds the identity thesis on top, and the weak doctrine says the shared answer already suffices. The charge that the weak doctrine has left unsolved a problem the strong doctrine solved fails, because both solve it in the same basic way.
2. The objection assumes that distinct predications need distinct items in the thing described.
The assumption driving the objection is that if two different things can truly be said of God, there must be two different things inside God answering to them. That is exactly what P2 denies. Consider a flawless diamond. It is beautiful, it is extremely hard, and it is chemically pure carbon: three distinct true predications telling you three different things, and it does not follow that the diamond has three internal items (beauty-itself, hardness-itself, carbon-itself) as constituents composing it. The diamond is simply what it is, and these are three true things said of one object from three angles of interest. God's attributes work the same way.
3. On no definition of "real distinction" does the weak doctrine put two things in God.
The medieval Scholastics used the phrase in more than one way, and on either usage the weak doctrine puts one thing in God, not two.
- On the separability test, which Suárez made standard (Disputationes Metaphysicae, disp. 7), A and B are really distinct only if at least one could exist without the other. Water and the glass it is in are really distinct: pour the water out and the glass remains. By that test God's attributes are not really distinct, since there is no possible situation in which God is omniscient but not omnipotent, or holy but not loving. A distinction that no possible separation could ever show is not a division into components.
- Thomists use a different test: a distinction is real if it holds in the thing prior to any mind's considering it. On that test they count essence and existence in a creature as really distinct although neither is ever found without the other. Applied to God, the question becomes whether there are two items in him, and the weak doctrine's answer is that there is one item, God, and two true things said of him (P3).
Aquinas himself held that the divine names differ in concept while naming one reality, and that the difference in concept has its foundation in God's perfection rather than in any division of him (Summa Theologiae I, q. 13, a. 4). Later Thomists called that a distinction of reason with a foundation in the thing: a distinction our minds draw, not a division in the thing, and yet not arbitrary, because something about the thing warrants it. Think of a single road running between a valley and a ridge. From the valley it is "the way up," and from the ridge it is "the way down." There are not two roads, and the two descriptions are still not made up, because the road really does run both ways. That is the distinction the weak doctrine affirms too. Where the doctrines differ is not over whether the attributes are separate things in God, since both say no, but over whether each attribute is to be identified with the divine essence (D13).
+The classical argument derives simplicity from aseity: a God with distinct attributes depends on them for what he is, which compromises his self-existence. On the weak view God still depends on his omniscience and omnipotence, whatever we call them.
1. The aseity argument against distinct attributes presupposes constituent ontology.
Aseity means that God depends on nothing outside himself for what he is. The argument that distinct attributes compromise it runs: if omniscience and omnipotence are distinct constituents of God, then God as a whole depends on each as something more fundamental than himself. But that dependence holds only if the attributes are building-blocks composing God the way bricks compose a wall. If they are not (P2), God does not depend on them as parts more fundamental than himself. He simply is what he is.
Think of an elementary particle, a genuinely indivisible unit with no sub-parts. It has a certain electric charge, which governs how it interacts with other charged particles, and a certain mass, which governs how it responds to gravity. Charge and mass are genuinely distinct quantities: measured differently, governing different interactions, and knowing one tells you nothing about the other. Yet the particle, having no internal parts, does not depend on charge and mass as components more fundamental than itself. It simply is the kind of thing it is, and charge and mass are two true things we say about it because our instruments meet it in two ways. God's attributes are like that: genuine truths about what he necessarily is, not building-blocks he depends on.
2. Aseity forbids dependence on anything outside God, not a describable inner character.
The doctrine of the Trinity already shows that aseity is compatible with real internal distinction. Father, Son, and Spirit are genuinely distinct within the Godhead, and that does not compromise aseity, because the relations among the persons are internal to God and involve nothing outside him on which he depends. In the same way, the fact that "omniscient" and "omnipotent" are distinct true descriptions of God implies no dependence on anything outside him. Both are necessary, both are internal to what he essentially is, and neither is an external item he needs in order to be what he is.
3. The identity thesis generates an aseity problem of its own.
Plantinga pressed the identity thesis into a dilemma. If God is identical with his attributes, and attributes are properties, then God is a property, and a property is an abstract object that cannot know, love, or act, which no person can be. If instead the attributes are distinct from God, then either God created his own nature, which is impossible, since he would have had to be wise before making wisdom, or his nature exists uncreated alongside him, which compromises his sovereignty over what exists.
Plantinga's own concern was sovereignty rather than aseity, and he set anti-realism aside as beside his point. But the dilemma bites on aseity just as hard, and it bites only on property realism. The weak doctrine's pairing of distinct attributes with anti-realism escapes both horns, since there is no property for God to be identical with, to create, or to depend on.
+God is contingently the Creator of this world and contingently knows what Peter is thinking. If God himself makes those truths true, and God exists necessarily, then they hold necessarily too, which is false, since Peter need not have existed.
The strongest form of this objection uses a rule most truthmaker theorists accept, called truthmaker necessitation: if a truthmaker exists, the truth it makes true cannot fail to hold. Apply it to a necessarily existing God who is the truthmaker for "God is the Creator of this world," and creation comes out necessary, which collapses the contingency creation is supposed to have. This is the hardest case for the truthmaker account, and the weak doctrine's defenders say so.
1. Only intrinsic predications are made true by God alone.
An intrinsic predication says what something is in itself, apart from its relations to anything else. "God is omniscient" is intrinsic: a truth about God's own nature and not about any creature. An extrinsic predication says something about a thing in terms of its relation to something else. "John is the father of James" is true of John, but its truth depends on James existing and standing in that relation; had James never been born, John would not be "the father of James," and nothing in John himself would be different.
"God is the Creator of this world" and "God knows what Peter is thinking" are extrinsic in that sense: they describe God in relation to contingently existing creatures. Their truthmaker is not God alone but God together with the creature or the state of affairs, and the contingency lies in the creature's existence, not in any change within God. So God's necessary existence does not by itself necessitate "God is the Creator of this world," because that truth also requires the world's contingent existence.
2. The strong doctrine faces the same problem and answers it the same way.
Contingent divine predications are a challenge for any doctrine of simplicity. If God's act of creation is identical with his necessary essence, creation looks necessary, and the strong doctrine's answer, distinguishing the necessary divine act from its contingent relational outcome, is structurally the answer just given (D4.1). Both locate the contingency in the creature. The objection gives the strong doctrine no advantage.
3. The weak doctrine has a second answer that the strong doctrine cannot use.
DS₄ forbids separable parts and metaphysical constituents (P4). It says nothing against God's having contingent states, his knowing that Peter exists or his willing this world rather than another, because a state of knowing is not a part that could be detached from him and not an ingredient he is built from. So the weak doctrine can take either of two routes, and Christians divide over which:
- A Christian who holds, with the classical tradition, that there is nothing contingent in God himself, so that he is intrinsically the same in every possible world, takes the extrinsic route (D4.1) and locates all the contingency in creatures.
- A Christian who holds that God freely and contingently knows and wills what he creates may say instead that the truthmaker for "God knows what Peter is thinking" is God as he contingently is, and that P3's claim about God alone as truthmaker concerns his essential predications only.
Both are orthodox positions, the disagreement between them is older than this dispute, and DS₄ is satisfied on either. The strong doctrine, which identifies God's knowing and willing with his necessary essence, has only the first route.
+The historical doctrine, affirmed by councils and confessions, means the identity thesis: God and his attributes are identical. A doctrine that says only that God has no separable parts is not divine simplicity at all, only a trivial claim wearing its name.
The strongest form of this objection is pressed by James Dolezal and Steven Duby: the church has always meant the identity thesis by "simplicity," and what the weak doctrine keeps is not a weaker version of the doctrine but a different claim under the same label.
1. The tradition never held one unified doctrine of simplicity.
There has never been a single doctrine of divine simplicity affirmed by the whole tradition. Thinkers have affirmed versions from the thin through to the full identity thesis. The Westminster Confession of 1646, the Presbyterian standard whose phrase "without parts" is often cited as the confessional anchor for the strong doctrine, is itself open to more than one reading. Dolezal argues that the Reformed Orthodox, the Calvinist theologians of the generations after the Reformation, consistently read it to include metaphysical parts and not only physical ones, and that historical claim has genuine force. But Richard Muller's exhaustive study of Reformed Orthodoxy finds significant internal diversity on exactly how the relation between the attributes and the essence was to be understood. The Confession's wording is a deliberately broad theological statement able to house more than one metaphysics, not a settled philosophical definition.
2. Historical precedent does not settle the question, as the Reformers themselves held.
Suppose the tradition did mean the identity thesis all along. What follows? The question is not primarily what the tradition affirmed but what is true and consistent with Scripture. The strong doctrine is not clearly required by Scripture, and it generates serious philosophical problems. When a traditional formulation turns out on careful analysis to be philosophically troubled and without clear scriptural warrant, the right response is reformulation rather than deference. The Reformers did not reject the medieval tradition wholesale, but they insisted that tradition answer to Scripture and to rigorous analysis, not the other way around. Applying the same principle to one item of scholastic metaphysics is not a betrayal of the Reformation. It is the Reformation's own method.
3. The weak doctrine is a substantive claim, not a trivial one.
DS₄ says considerably more than that God is not made of physical parts (P4.3). Each of its clauses denies a picture of God that has had serious defenders: that he is compounded of form and matter in some spiritual analogue, that his essence and his existence are two principles in him needing something to unite them, that his attributes are ingredients he is built from, or that the persons are pieces of the nature. It agrees with the tradition on the line that matters, that God alone exists from himself and depends on no principle more fundamental than himself. A claim that rules out real positions and carries that consequence is not a trivial observation.
4. The weak doctrine preserves everything the strong doctrine was protecting.
What was the tradition guarding with simplicity? Aseity (God depends on nothing outside himself), immutability (God does not change by gaining or losing attributes), uniqueness (God is not one member of a kind that could have other members), and the absolute unity of the divine being. The weak doctrine preserves every one. What it drops is the identity thesis, a philosophical articulation developed inside a particular Aristotelian-Neoplatonic framework, which Scripture does not state and the creeds all Christians share do not state either. The Fourth Lateran Council (1215) did call the divine essence entirely simple, and what that phrase requires is exactly what the two doctrines argue about; the weak doctrine reads it as uncomposed, which the tradition's own defenders agree is at least part of what was meant. Reformulating the doctrine for a new metaphysical setting while keeping its theological content is legitimate development, not abandonment.
+Property anti-realism is a general thesis about everything, yet ordinary talk about creatures (two things sharing a colour, laws governing real physical quantities) seems to need real properties. Adopting anti-realism only to rescue God from composition is ad hoc.
1. The anti-realism was adopted for all of reality, not for God.
The charge is that the position is ad hoc, a fix invented for this one job, and that mischaracterizes it. The anti-realism in question is anti-Platonism about abstract objects generally: the denial that numbers, properties, propositions, or possible worlds exist on their own, eternal, non-physical, and independent of any mind, as Platonism holds that the number 7 existed before anyone counted and will exist after the universe ends (P2). Craig, whose project the objection targets, does not hold realism for creatures and anti-realism for God. He holds a consistent anti-realism about abstract objects, developed at book length, and the order of motivation runs the other way from what the objection supposes. The general case for anti-Platonism is driven first by the worry that abstract objects existing independently of God would compromise his aseity across the board, for numbers and propositions as much as for properties (D10). Divine simplicity is one of several theological problems the position turns out to illuminate, not the reason for adopting it.
2. Neutralism has a pedigree that owes nothing to theology.
The particular anti-realism at work, the neutralism of P2 on which an ordinary "there is" commits nobody to anything, was not built for this job either. Its roots run back a century and more, through philosophers who had no interest in the divine attributes:
- Franz Brentano and Edmund Husserl treated reference, the way a word or a thought points at something, as an act of the mind aimed at an object, whether or not anything in the world answers to it, rather than as a link between a word and a thing that must exist.
- Karel Lambert's free logic is a logic built to work without assuming that every name picks out something that exists, so that from "Pegasus has wings" it does not follow that there is a winged horse. Free logic frees the names and still lets the quantifier (the logician's word for "there is" and "some," the words that say how many things a sentence is about) carry commitment. Neutralism takes the second step and frees the quantifier too.
- Jody Azzouni defends neutralism today with no theological stake in it, and he has a name for the move Quine's heirs make, coaxing an ordinary "there is" out of someone and then handing them a whole theory of what exists: philosophical chicanery.
Craig follows Azzouni on the quantifier and Brentano and Husserl on reference. A position with that lineage was not invented to rescue God from composition. It was found already standing and put to work.
3. Resemblance, natural kinds, and laws can all be handled without abstract properties.
The objection assumes that explaining why creatures resemble one another, why nature sorts into kinds, and why physical laws hold requires property realism. That is exactly what anti-realists dispute, and they have done the work:
- Resemblance. Two red things resemble each other. Quine treated that as a basic fact needing no deeper explanation, and resemblance nominalists and trope theorists explain it through the things themselves and their own particular colour-instances, never through a shared universal (D1).
- Natural kinds. A natural kind is a grouping nature makes rather than one we invent: gold or tiger, as opposed to "things on my desk." The word "gold" can apply to every concrete lump that resembles the others in the right ways without a shared abstract nature that every lump contains.
- Laws of nature. A law can be understood as a regularity among concrete events (every unsupported apple falls) without abstract entities standing at either end of the relation.
Whether these accounts succeed in full is a live debate. But they are serious proposals, defended by philosophers with no theological stake in the outcome, and that is all it takes for the charge of special pleading to fail.
+God cannot be omnipotent without being omniscient, or holy without being loving. The identity thesis explains why, since they are one and the same thing; the weak doctrine, with genuinely distinct attributes, leaves their necessary co-presence a brute fact.
1. The attributes are co-present because all are essential, not because they are identical.
Why can God not be omnipotent without being omniscient? The objection answers: because they are one and the same thing. There is a simpler answer. Both are essential attributes of God, belonging to his nature necessarily, and an essentially omnipotent, essentially omniscient being cannot fail to be both. Take away either one and you have not removed a feature of God; you have removed God. That is a principled and complete explanation of inseparability, and it needs only what both doctrines affirm, that every divine attribute is essential.
Consider the number 7. It is necessarily both prime (divisible only by itself and 1) and odd (not divisible by 2). Being prime and being odd are not the same thing to say about it: we define them independently, each applies to many numbers the other does not (9 is odd and not prime, 2 is prime and not odd), and we understand each without reference to the other. Yet in 7 they are necessarily co-present, because both are essential to 7 as the number it necessarily is. Their co-presence is explained by their both being essential to 7's nature, not by their being identical. God's attributes work in exactly the same way.
2. The strong doctrine uses the same explanation in the end.
Even on the strong doctrine, the identity thesis does not fully explain why these particular attributes are all identical to each other and to God's essence. Why are omniscience and omnipotence both identical to God, rather than one being identical and the other a genuinely distinct attribute? The strong doctrine's answer traces back to God's essential nature: because God's nature is what it necessarily is, and both are essential features of it. That is the weak doctrine's answer. If the strong doctrine can explain co-presence by appeal to God's essential nature, so can the weak, and the identity thesis is a further metaphysical claim that is not doing the explanatory work.
+After the Incarnation a new truth holds: the Son is human. If that predication is intrinsic, God's existence necessitates it and the Son was always human; if it is extrinsic, the Incarnation is only a change of relation, which is not what Chalcedon confessed.
The dilemma is real. Intrinsic predications say what a thing is in itself and extrinsic ones describe it by its relations (D4), and "the Son is human" seems to fit neither: read as intrinsic, the truthmaker account makes it necessary; read as extrinsic, the personal union looks like a mere relational description.
1. Chalcedon locates the change in the assumed human nature, not in the divine nature.
The Council of Chalcedon (451 AD) defined that in Christ two natures, divine and human, are united in one person "without confusion, without change, without division, without separation." The phrase "without change" guards the divine nature from any internal alteration in the Incarnation. What comes into being in the Incarnation is the human nature of Christ: a concrete human nature that begins to exist and is assumed by the eternal Son. The Son does not change internally; what changes is that a human nature comes to exist and is united to him in what theology calls the hypostatic union, the union of the two natures in one person (from hypostasis, the Greek word the councils used for a concrete individual, a person).
So "the Son is human" holds in virtue of that union: the Son is the person in whom a divine nature and a human nature are united. The predication is not purely intrinsic (it depends on the contingent coming-to-be of the human nature) and not purely extrinsic in the way "God is to the left of the Eiffel Tower" would be (since the union is genuinely personal, not merely relational). It is a unique predication grounded in a unique situation.
2. The truthmaker for "the Son is human" is the hypostatic union itself.
On the truthmaker account the truthmaker is not God considered alone in his eternal nature but the Son together with the assumed human nature that contingently exists and is personally united to him. The contingency lies in the coming-to-be of the human nature, which is a creaturely event. This is the answer given to God's contingent relations to creation (D4), applied to the one case where the relation is personal: as "God is the Creator of this world" is made true by God-plus-the-world, "the Son is human" is made true by the Son-together-with-the-assumed-nature.
3. Both doctrines carry the same Christological tension.
How divine simplicity in either form fits the full reality of the Incarnation is one of the deepest problems in philosophical Christology, and it remains under discussion. What can be said is that both doctrines must invoke the Chalcedonian framework, both must locate the newness of the Incarnation in what is assumed rather than in the nature that assumes, and both must hold that the personal union is real without any internal alteration of the divine nature. The weak doctrine asks nothing here that the strong doctrine does not.
+Even if God alone makes both "God is omniscient" and "God is omnipotent" true, what makes them two truths rather than one truth said twice? With nothing in God answering to the difference, the weak doctrine has a grounding gap.
The strongest form of this objection is Christopher Hughes's. A grounding question asks not whether something is true but what in reality makes it true and makes it distinct from other truths. Grant that God is one concrete reality who serves as truthmaker for everything true of him; there still seem to be two different facts, that God is omniscient and that God is omnipotent, and the property realist has an easy account of what grounds the difference (two properties, and God exemplifying each) that the anti-realist owes and has not obviously paid.
1. The difference is grounded in God himself and in what each predication says.
Ask the same question about something ordinary. "Gold is yellow" and "gold is heavy" are two truths, and the truthmaker for both is gold. What grounds their difference? Not two items inside the gold, one for the colour and one for the weight, but the fact that being yellow and being heavy are different ways for a thing to be, and gold is both ways. The difference lies in what is predicated, and the one lump suffices for both because it is the kind of stuff it is.
God's attributes work the same way, with one addition: everything true of him essentially is necessary. "God knows all truths" and "God can do all things" differ because knowing and doing differ, and both are true because God is the being he necessarily is. The demand for a separate item in God to answer to each difference in truth is constituent ontology asked a second time (D2), and once properties are not ingredients the demand has no force. The tradition did add that we come to name the attributes separately because we meet the corresponding perfections separately in creatures (Summa Theologiae I, q. 13, a. 4), but that is a point about how we learn the words, not about what makes them true: God would be distinctly omniscient and omnipotent in a world where he created nothing.
Think of white light through a prism. The prism does not manufacture the colours; every wavelength was already in the beam together, and the glass separates for our eyes what was one light all along. The attributes are like that: many truths about one undivided reality, distinguished by what they say and not by pieces of him.
2. The strong doctrine does not escape the grounding challenge either.
The strong doctrine seems to have a simpler answer: omniscience and omnipotence are identical, so there is one fact, not two. But then what grounds our having two concepts rather than one? Its answer, that the multiplicity is in our finite minds and not in God, is structurally the weak doctrine's answer (D2): many things are true of one undivided reality, finite minds grasp them one predication at a time, and no internal divisions in God answer to the count.
3. The grounding question is live for every doctrine of simplicity.
This objection is philosophically acute and should not be dismissed. The truthmaker account has drawn a direct published objection from Noël Saenz, and the philosophers who built it continue to refine it in reply; the question of what makes two truths about one thing two is a live question about truthmakers generally, not a peculiarity of theology. What can be said with confidence is that any account of divine simplicity must explain how one undivided reality grounds many distinct true predications, and the truthmaker account is a principled answer that invites refinement rather than refutation.
+Even granting property anti-realism, numbers, mathematical truths, logical laws, and possible worlds seem to exist necessarily and independently of God. If any abstract object exists independently of him, aseity is threatened whatever we say about properties.
1. The objection is right that property anti-realism is not the whole story.
Property anti-realism dissolves the specific problem of divine simplicity: if properties are not abstract entities, God's many attributes do not compose him out of abstract parts. It does not by itself address the rest of the apparent abstract objects. If the numbers 1, 2, and 3 exist necessarily and independently of God, or if the law of non-contradiction holds independently of him, then something exists outside him and independently of him, and aseity is threatened from a different direction. That much is granted.
2. The anti-Platonist project covers abstract objects generally, and every theist needs it.
This objection is one of the primary motives of the anti-Platonist project the weak doctrine draws on (D6), not an afterthought to it. The position is that God is the source and ground of all reality, abstract objects included, and it surveys the full range of anti-realist strategies, from fictionalism (talk of numbers is a useful fiction that commits you to nothing) through pretence theory to divine conceptualism (the abstract objects are God's thoughts), arguing that at least one is more defensible than accepting abstract objects that exist necessarily and independently of God. The strategies are set out and weighed in D11.
None of this is a problem the weak doctrine created. It is a challenge for any theism that affirms aseity in a world where mathematical and logical truths look necessary, and the strong doctrine faces it identically: if the number 7 exists independently and necessarily, the identity thesis about God's attributes does nothing about it. Both doctrines reach for the same anti-Platonist resources, and the objection gives the strong doctrine no advantage.
3. The broader debate is live.
Whether any anti-realist strategy fully succeeds is an open question in philosophy of religion and metaphysics. What is clear is that theists of many kinds have recognized the challenge and developed serious responses to it. The best short way in is a published six-view exchange in which Platonists, absolute creationists (who hold that God creates the abstract objects), divine conceptualists, and anti-realists answer one another directly, with a leading atheist philosopher replying from outside the faith.
+Our best scientific theories cannot be stated without saying there are numbers, functions, and shared properties, and we should believe what our best theories say exists. Property anti-realism leans on those theories daily while denying what they say is there.
The strongest form of this objection is the Quine-Putnam indispensability argument, so called because the abstract objects are said to be indispensable, impossible to do without, in stating our best science. To quantify over something is to say "there is" one of it, in the logician's way of putting it, and the argument runs: (I) if a sentence is literally true, then the things its names name and its "there is" ranges over exist; (II) there are literally true sentences of science and mathematics whose names and "there is" can only be about abstract objects; (III) therefore abstract objects exist. Putnam added the charge of dishonesty: to lean on the theories every working day while denying what they say is there.
1. The first premise is the contested criterion of commitment.
Premise (I) does not report what exists. It is a rule about what must exist if what we say is true, and it is exactly the rule neutralism denies (P2). So the argument cannot be turned against neutralism without begging the question: it assumes the criterion the neutralist rejects and then finds the neutralist committed. The Platonist needs an independent argument for (I), and the usual one, that ordinary "there is" carries commitment and translates straight into the logician's quantifier, is what the ordinary sentences in P2 call into question (D12).
Consider how the two sides weigh certainty. "2 + 2 = 4" is about as certain as anything a person believes. That there is a correct Quinean criterion of ontological commitment is a philosopher's theory, and a contested one. If the criterion says that anyone who asserts "2 + 2 = 4" has thereby asserted that the number 2 exists as a mind-independent object, and one is more sure that 2 + 2 = 4 than of any ontological consequence of it, the rational thing to give up is the criterion. That is the anti-Platonist's quick argument, and it is why the burden sits where it does.
2. Indispensability to the truth of a theory is not indispensability to what exists.
Grant that physics cannot be written without numbers, and grant that "spiders share some of the anatomical features of insects" is true and cannot be neatly paraphrased. The neutralist grants the truth and the quantification and denies only the last step, that the "there is" in such sentences is a census. Everyday speech quantifies constantly over things nobody takes for residents of reality:
- There is a chance of rain this afternoon.
- There is a shortage of parking downtown.
- There is a reason he was late, and a way you smile when you are about to say no.
If our not being ingenious enough to paraphrase these away committed us to chances, shortages, reasons, and ways as objects on a par with electrons, then ordinary English would be the most extravagant ontology ever proposed, and nobody believes that it is.
3. The Platonist must defeat every rival at once.
For the indispensability argument to establish Platonism, Platonism must be the only reasonable way to handle premises (I) and (II). Against (I) stand free logic (D6), neutralism (P2), and neo-Meinongianism, the view named for Alexius Meinong on which we can truly talk about things that do not exist, Sherlock Holmes and the golden mountain included, so that "there is" need not mean "exists." Against (II) stand several live positions:
- Fictionalism: the sentences are useful and not literally true.
- Figuralism and modal structuralism: the sentences are true under a paraphrase (as figures of speech, or as claims about what any structure of the right shape would have to satisfy) that quantifies over nothing abstract.
- Pretence theory: we are prescribed to imagine the sentences true, the way the Axiom of Infinity in set theory (the rule that says there is an infinite set) is adopted for its usefulness rather than its self-evidence.
- Divine conceptualism, a realist option: the abstract objects are God's thoughts.
The theist does not need to pick a winner. The argument is like a prosecution that must eliminate every other suspect before the jury can convict: it fails if even one of these positions is left standing as reasonable, and nobody in the debate holds that all of them are unreasonable.
4. Grant the objection everything it asks, and the weak doctrine still stands.
This is the objection the anti-realist route has to survive. Suppose it fails and abstract objects exist. The relational route is still open (P2, P3): a property realist who denies that properties are ingredients of things, the apple related to redness rather than built from it, has all the materials the weak doctrine needs, and what remains is the aseity question about uncreated abstract objects, which every theist faces on any doctrine of simplicity (D10).
See also:
Natural Theology: Applicability of Mathematics
+Neutralism is a confusion. "There is" and "exists" mean one thing, captured by the logician's quantifier, so "there is a way you smile" is either loose talk that can be paraphrased away or a real, if lightweight, commitment.
The objection is Peter van Inwagen's, and it is the sharpest published reply to the position P2 adopts. His theses run: being is the same as existence; the word is univocal, meaning it has one sense wherever it turns up, in "there are prime numbers" as in "there are cows"; that one sense is what the existential quantifier expresses, the logician's symbol for "there is at least one"; and so to say "there is an F" while denying that any F exists is to say "something is F and nothing is F," which is a contradiction. Philosophers distinguish heavyweight quantification, a "there is" that adds its object to the census of reality, from lightweight quantification, a "there is" that commits to nothing beyond the truth of the sentence. Van Inwagen's position is that the only "there is" is the heavyweight one, so the person who says "there is a way you smile" either means it, and is committed to ways, harmlessly, or is speaking loosely and could, with effort, say what she means without the "there is."
1. The univocity thesis does not touch neutralism.
Grant univocity. The neutralist agrees that "there is" and "exists" mean the same thing wherever they occur, and that no ordinary word changes sense between "there are cows" and "there is a way you smile." What he denies is that the shared sense is a census. The cow and the smile are alike in that "there is" applies to both; they differ in that the cow is a mind-independent object with causal powers and the way you smile is not. That difference is settled by asking about mind-independence and causal powers, not by the word. Van Inwagen's contradiction, "something is F and nothing is F," arises only if "exists" in the second clause has already been given the heavyweight reading, which is the thing in dispute.
- The position his argument does refute is quantifier variance, the view that "there is" has two senses, one committing and one not. The weak doctrine's defenders reject that view for the same reason he does.
2. The paraphrase demand cuts against the Quinean.
Van Inwagen's own best case is a sentence he holds true and unparaphrasable: spiders share some of the anatomical features of insects. Grant that nobody has produced a paraphrase. The question is what follows. On the Quinean rule, the failure of paraphrase commits the speaker to anatomical features as objects. On the neutralist reading, the sentence is true, it quantifies over anatomical features, and whether anatomical features are objects in the world is a separate question that the sentence does not answer. The neutralist's point is that the second reading is how the sentence is actually used: nobody who says it has taken a stand on universals. If the failure of paraphrase really committed speakers to whatever they could not paraphrase away, the same rule would hand them the chance of rain and the reason he was late (D11).
3. Van Inwagen grants the lightweight reading himself.
He allows that "The crack in the Liberty Bell is 62.23 cm in length" is true on an ordinary reading and may be false in what he calls the ontology room, a stipulated strict use of the quantifiers (some philosophers call the dialect Ontologese) in which cracks may not be among the things that are. That concession is most of neutralism. It admits that ordinary English "there is" does not settle ontology and that a special, stipulated usage is needed to ask the heavyweight question. The neutralist's reply is that the premises of the indispensability argument are written in ordinary English, and on the reading van Inwagen concedes to ordinary English they are true and commit nobody to anything. To reach the heavyweight conclusion, the Platonist has to move the premises into the stipulated usage, and there they are no longer obviously true. The neutralist adds that nobody actually speaks Ontologese, not even in the seminar room; there is only ordinary language, and in ordinary language one asks whether something is a real, mind-independent object by asking exactly that.
4. A standoff over the criterion is all P2 needs.
Honesty requires saying that a Quinean criterion of commitment is the usual starting point in philosophy of mathematics and that neutralism is a minority position. Its defenders have an argument that the criterion is false (D11), and a committed Platonist, glad to have the numbers, can resist it by accepting the consequence. What they have shown to everyone is that the criterion is a theory rather than a datum, a rule someone proposed rather than a fact anyone observed; that the intuitions usually cited for it (ordinary "there is" sentences) do not support it; and that its most famous consequence, that "2 + 2 = 4" commits its speaker to a mind-independent object, is less certain than the arithmetic. That is enough for P2, because the deductive standard asks whether the premise is more plausible than its denial, and P2's denial is constituent ontology, which the relational realist rejects along with the neutralist (D1, D11).
+Once God alone makes "God is wise" and "God is powerful" true, with no wisdom-itself and no power-itself, nothing in God is his wisdom rather than his power. That is the identity thesis; the weak doctrine differs only by refusing a form of words.
1. The two doctrines agree about what is in God.
The truthmaker account was built by a defender of the classical doctrine to say what "God is his wisdom" could mean once properties are not ingredients: God himself, and nothing else, makes "God is wise" true (P3). The weak doctrine says the same about reality: one concrete God, many true predications, no items in him answering to them one by one. Aquinas himself said the divine names are distinct in what they signify and are not synonyms, while the thing they name is one (Summa Theologiae I, q. 13, a. 4). So far there is no quarrel, and nobody should pretend there is.
2. The identity thesis says more than the truthmaker account.
The strong doctrine holds that in God there are no distinctions of any kind: not only that his wisdom is his power, but that his essence is his existence, that his knowing is his willing, that his act of creating is identical with his necessary essence, and that each of the three persons is identical with the one essence. Those are identity claims, not restatements of "one truthmaker," and each generates a difficulty the weak doctrine does not share:
- If God's act of creating is his essence and his essence is necessary, creation looks necessary, and the strong doctrine's defenders have to explain why it is not (Aquinas, Summa Theologiae I, q. 19, a. 3). The weak doctrine lets God's free act be contingent and distinct from his nature (C1, D4).
- If each person is identical with the essence, the persons are identical with each other unless the rule that two things identical with a third are identical with each other is restricted, which is what the classical defenders do. The weak doctrine lets the persons be really distinct without being parts (C1).
- If "wise" names nothing in God that "powerful" does not, the strong doctrine needs its theory of analogy to explain why the two words are not synonyms in his case, and it has to keep that theory from sliding into the view that we know nothing of what God is (P1). The weak doctrine says the words keep their core meaning.
3. On anti-realism the identity thesis is not false so much as empty.
"God is identical with his wisdom" is an identity claim, and an identity claim needs two names that might pick out one thing, the way "Mark Twain is Samuel Clemens" does. On anti-realism "his wisdom" is not the name of anything; it is a way of saying that God is wise. So the weak doctrine does not refuse a form of words for the sake of refusing it. It holds that there is nothing for the words to assert. What is left of the identity thesis after the properties are gone is the truthmaker claim, which both doctrines share, and the further identities the strong doctrine adds (D13.2), which they do not.
4. A difference with consequences is not verbal.
The two doctrines are closer than their partisans sometimes say, and both protect the same three things: God was not built, God depends on nothing, and God is not one instance of a kind (P4, D5). Where they part is over whether "simple" means uncomposed or means that every distinction in God collapses. That is the difference between saying a diamond was never glued together and saying that its hardness, its clarity, and its weight are all one and the same fact about it, and it reaches creation, the Trinity, and the meaning of our words about God.
See also:
PT / Theology Proper: Divine Simplicity (Classical / Strong)PT / The Trinity: Divine Simplicity and the Trinity