De re necessity and the idea of a necessary being
One objection to modal arguments for God begins before any premise about causation, contingency, perfection, or possible worlds. It says that the key expression "necessary being" is not merely unproved, but unclear. If there is no coherent sense in which a thing, as opposed to a sentence or description, can be necessary, then modal cosmological and ontological arguments do not fail at a later step. They fail to get started.
The most important source for this worry is W. V. Quine, especially in "Reference and Modality" and "Three Grades of Modal Involvement." Quine's objection was technical, but its target needs stating carefully. He accepted that modal claims can be intelligible at the sentence level: "Necessarily, 2 + 2 = 4" is true, while "Necessarily, the number of planets is eight" is false. His concern was that ordinary substitution of co-referring expressions can fail inside modal contexts. That creates a problem when we move from de dicto modality, concerning what is necessary under a description or within a sentence, to de re modality, concerning what is necessary of an object itself. Quine took this problem to put pressure on quantified de re modality and on essentialism unless a principled account of reference and essential properties is supplied.
Quine's famous example involves ordinary identity substitution. The number 9 is necessarily greater than 7, and the number of planets was, at one time, 9. But it does not follow that the number of planets is necessarily greater than 7. The expressions "9" and "the number of planets" may pick out the same object, yet modal contexts do not permit unrestricted substitution of co-referring expressions. The point is not that one object literally has incompatible modal properties, or that necessity straightforwardly belongs to a description. It is that definite descriptions are often non-rigid and that a theory of de re necessity must explain why some reference-fixing practices preserve modal truth while others do not. Quine regarded this as a reason to resist essentialist claims unless they could be given a non-question-begging basis.
Applied to theism, the objection is sharp. Modal cosmological arguments often distinguish contingent beings from a necessary being: contingent things might not have existed; a necessary being cannot fail to exist. Modal ontological arguments go further and reason from possible necessary existence to actual necessary existence. But if "x exists necessarily" lacks description-independent truth conditions, then such arguments seem to turn on an equivocation. They may have shown that some sentence is necessary, or that some description is necessarily satisfied if satisfied at all. They have not shown that there is an object whose non-existence is impossible.
The objection in its strongest form
The strongest version of the argument is not that all modal language is nonsense. It can allow that some necessity claims are harmless: logical truths, mathematical truths, or truths generated by linguistic convention. The problem is specifically with robust de re necessity. To say that a particular thing has necessary existence seems to require that there are modal facts about the thing independently of our descriptions of it. Quine doubted that we can make sense of such facts without smuggling in essentialism.
The objection can be set out like this:
- Modal arguments for God require de re modality. On the intended reading, they need an object that exists necessarily, not merely the claim that necessarily some object satisfies a divine description.
- But it is unclear what licenses the move from a modalized description or proposition to a claim about an object itself. The argument must distinguish a sentence such as "necessarily, something satisfies description D" from "there is something that necessarily exists."
- A Quinean critic can argue that, unless essential properties and reference across possibilities are independently explained, claims of necessary existence may reflect features of a concept or description rather than a genuine property of an entity.
- Therefore, the theist bears a prior burden: before using a necessary being in an argument, he must state the relevant notion of necessity and explain why it applies to an entity itself rather than merely to a proposition about it.
This is a serious challenge because it targets the grammar of the argument, not just its conclusion. A critic of the cosmological argument need not show that every contingent thing has no explanation, or that the universe is brute, or that infinite regress is possible. He can say: the contrast between "contingent being" and "necessary being" has not yet been made intelligible.
It is also important to note what Quine was not doing. He was not refuting theism by science, and he was not claiming that physics has shown there is no necessary being. The problem of evil is a distinct argument, and unlike the Quinean challenge it can provide genuine prima facie evidence against classical theism: the scale and apparent gratuitousness of suffering are difficult to reconcile with an omnipotent, omniscient, and perfectly good God. The logical problem of evil is not widely regarded as a strict contradiction after the development of free-will defenses, but the evidential problem remains a serious objection even though theists dispute its force and it does not deductively prove that God does not exist. The Quinean objection is instead conceptual: modal theistic arguments rely on a kind of necessity whose coherence and truth conditions require defense.
The best reply: Quine did not have the last word on modality
The strongest reply is not that Quine was simply refuted, but that later analytic philosophy showed that de re modality cannot be dismissed as unintelligible. Ruth Barcan Marcus helped develop quantified modal logic; Saul Kripke's Naming and Necessity challenged descriptivist theories of reference; and Alvin Plantinga used possible-worlds semantics in his modal ontological argument. David Lewis, while rejecting theistic conclusions, developed a detailed modal-realist account, and Kit Fine argued that essence should not simply be identified with necessity. These thinkers disagree sharply about the metaphysics behind modal claims, so their work does not establish that a necessary concrete being exists. It does show, however, that serious philosophers have offered technically developed accounts of de re modality rather than treating it as mere verbal confusion.
Kripke's account of rigid designation is especially relevant. In ordinary modal reasoning, proper names are not usually disguised descriptions: "Aristotle" need not mean "the teacher of Alexander" or "the author of the Metaphysics. Aristotle could have failed to teach Alexander and still have been Aristotle. If the name designates Aristotle, it continues to designate him in possible situations in which Aristotle exists; it does not follow that Aristotle exists in every possible situation. Rigid designation therefore helps distinguish a contingent description from the object it happens to identify, but it does not by itself establish that any concrete object has necessary existence.
For example, the person who won a particular election might have lost that election. Under the description "the winner," victory is built in. But the person himself is not essentially the winner. This does not show that de re modality is incoherent. It shows that we must be careful about scope, reference, and description. The very phenomenon Quine identified can be interpreted not as a reductio of de re modality, but as evidence that ordinary descriptions are often non-rigid.
Possible-worlds semantics also gives a formal framework for "necessary existence," but the framework does not settle its metaphysics by itself. In a varying-domain Kripke model, an individual has necessary existence if it exists in every relevant possible world; in a constant-domain model, every object is already included in every domain, so domain membership alone cannot express a substantive claim about necessary existence. A theory must therefore specify how existence, accessibility, and identity across worlds are being interpreted. One may dispute those choices, but after they are made the phrase is not obviously meaningless. Philosophers seriously debate whether numbers, sets, propositions, laws, substances, or persons have essential properties, even though no consensus follows from the formalism alone.
This reply has real force. It would be misleading to write as though Quine's critique simply destroyed de re modality. Contemporary metaphysics contains serious, technically sophisticated accounts of essence, necessity, rigid designation, and modal quantification. A theist who uses "necessary being" is not automatically relying on a notion that analytic philosophy has exposed as confused.
Where the reply is not enough
Still, the reply should not be overstated. Showing that de re modality is intelligible in some cases does not automatically vindicate necessary existence. There is a difference between relatively modest or theory-dependent claims such as "this person could not have been a prime number" and stronger claims such as "water is necessarily H2O," "Socrates is essentially human," or "there exists something whose non-existence is impossible." The former claims may be defended using Kripkean views about identity, essence, or natural kinds, but they remain philosophically disputable and do not entail that any concrete object exists in every possible world. The last claim is stronger and more metaphysically loaded.
Kripke helps explain how an object can have necessary properties, but he does not by himself show that any concrete object exists necessarily. Plantinga's modal ontological argument proceeds from the premise that it is possible that a maximally great being exists, where maximal greatness is understood to include maximal excellence in every possible world and therefore necessary existence. In an S5-style modal system, if it is possible that a necessarily existent being exists, then such a being exists necessarily, and hence exists actually. The controversial premise is precisely the possibility premise. Critics are therefore entitled to ask whether "possibly, a necessarily existent being exists" has been independently motivated or merely built into the definition.
Similarly, modal cosmological arguments often proceed from contingent reality to a necessary foundation. Even if the category of necessary existence is coherent, further questions remain. Why think contingent reality requires an explanation of the relevant kind? Why must the explanation terminate in something necessary rather than in a brute fact, an infinite explanatory structure, or a fundamental physical reality with modal status unlike ordinary objects? Why identify the necessary ground with anything recognisably divine? These are not settled by defeating Quine.
There is also an ongoing dispute about the interpretation of possible worlds. If possible worlds are taken as concrete realities, as in David Lewis's modal realism, ordinary individuals are world-bound and de re modal claims are handled through a counterpart relation rather than literal trans-world identity. On that view, "exists in every world" is not straightforwardly a claim about one object persisting across worlds. If possible worlds are abstract representations, necessity may depend on the structure and correctness of those representations. If possible worlds are heuristic devices, they may clarify modal talk without by themselves supplying deep metaphysical truth conditions. None of this makes necessary existence unintelligible, but it prevents the formalism from delivering an easy metaphysical victory.
What the objection does and does not establish
The Quinean objection establishes an important burden of clarity. A theist cannot responsibly rest an argument on the words "necessary being" while leaving unexplained whether necessity is logical, metaphysical, conceptual, explanatory, or something else. In most contemporary theistic arguments, the intended notion is metaphysical and de re: roughly, given the relevant nature of the entity, there is no metaphysically possible world in which it fails to exist. Some defenders ground this in essence, while others connect it with a broader theory of explanation. Neither route is secured by definition alone. Defining God as necessary introduces a concept; it does not by itself show that anything in reality satisfies that concept.
The objection also warns against sliding between descriptions and things. From "the ultimate explanation, if it exists, exists necessarily" it does not immediately follow that anything actually has necessary existence. From "God is conceived as non-contingent" it does not follow that there is a being whose non-existence is impossible. Modal arguments must earn their modal premises.
But the objection does not establish that theism is false. It does not even establish that all modal arguments are invalid. Quine's scepticism about de re modality is a contested philosophical position, not a settled result. Much of later analytic philosophy developed tools precisely to make sense of de re necessity. A defender of necessary being can appeal to rigid designation, quantified modal logic, essence, or possible-worlds semantics without thereby speaking gibberish.
The fair conclusion is modest. The phrase "necessary being" is not self-explanatory, and arguments that depend on it must give a serious account of de re modal truth. Quine gives atheists and agnostics a legitimate challenge to press: what are the truth conditions for "x exists necessarily," and why should we think anything satisfies them? But once that challenge is answered with the resources of post-Quinean modal metaphysics, the debate has to move on to the substantive premises. The Quinean objection is a demand for conceptual discipline. It is not, by itself, a disproof of God.