Naming and Necessity

Context

Naming and Necessity began as a series of three lectures delivered by Saul Kripke at Princeton University in January 1970. The lectures were delivered without a written text — Kripke, then 29 years old, spoke from notes — and their impact was immediate and electric. Among those in attendance were some of the most distinguished philosophers of language in the English-speaking world, and the lectures' arguments against the established consensus of the preceding half-century were so persuasive that the field of philosophy of language was, in the assessment of many observers, never the same again. The lectures were published in 1972 as a paper in the volume Semantics of Natural Language and then, with minor revisions and a substantial postscript responding to critics, as a book by Harvard University Press in 1980. The book version is the standard text.

The intellectual context against which Kripke's lectures were directed was the dominance of what is known as the descriptive theory of names. This theory, which had been the consensus view in analytic philosophy of language since the early twentieth century, held that proper names are not directly referential — they are not labels attached to individuals. Rather, a proper name is equivalent to a definite description: a phrase that picks out an individual by specifying properties that uniquely identify that individual. "Aristotle" means "the student of Plato who wrote the Nicomachean Ethics and taught Alexander the Great." To know the meaning of the name "Aristotle," on this view, is to know the description associated with it.

The descriptive theory had deep roots. It can be traced to Gottlob Frege's distinction between sense and reference, developed in the 1890s. Frege argued that a name has both a reference (the object it points to) and a sense (the "mode of presentation" of that object). Two names can refer to the same object but have different senses: "the morning star" and "the evening star" both refer to Venus, but they present it differently, and someone can know that the morning star is the morning star without knowing that it is the evening star. The sense, not the reference, is what determines the cognitive significance of the name — it is what we grasp when we understand a name.

Bertrand Russell took Frege's insight further. In his 1905 paper "On Denoting," Russell argued that ordinary proper names are, in effect, disguised definite descriptions. When we say "Aristotle was a philosopher," what we are really saying is "the student of Plato who wrote the Nicomachean Ethics was a philosopher." The name "Aristotle" has no independent meaning; it is a shorthand for whatever description we associate with it. This position, known as descriptivism, became the received view in analytic philosophy. It was accepted, with variations, by Wittgenstein in the Philosophical Investigations, by John Searle, and by most of the leading figures in philosophy of language. It seemed to solve a number of puzzles: how names can be meaningful even if their bearers don't exist (we can talk about "Santa Claus"), how names can refer to the same object through changes in our knowledge about that object, and how different speakers can use the same name to refer to the same person even if they associate different descriptions with it.

Summary

Rigid Designation

Kripke's central concept is that of the rigid designator. A rigid designator is a term that refers to the same object in every possible world in which that object exists. Proper names, Kripke argues, are rigid designators. The name "Aristotle" refers to Aristotle — the very same individual — in every possible world in which Aristotle exists. It does not refer to "whoever wrote the Nicomachean Ethics," because there are possible worlds in which Aristotle didn't write the Nicomachean Ethics (perhaps he died young, perhaps someone else wrote it). In those worlds, "Aristotle" still refers to Aristotle, even though he lacks the properties that the descriptive theory says define the name.

This contrasts with definite descriptions, which are typically non-rigid. "The author of the Nicomachean Ethics" refers to different people in different possible worlds: in the actual world, Aristotle; in another possible world, someone else might have written it. The description picks out whoever satisfies it, and that can vary from world to world. The name, by contrast, always picks out the same individual. This is the heart of Kripke's challenge to descriptivism: if names were equivalent to descriptions, they would be non-rigid, but names are rigid, so they cannot be equivalent to descriptions.

Proper Names and the Causal-Historical Theory

If names are not equivalent to descriptions, how do they refer? Kripke's answer, sketched in the lectures and developed more fully in the postscript, is the causal-historical theory of reference. A name is originally bestowed on an individual in a "baptism" — an initial naming, which may involve pointing at the individual, giving a description that identifies them, or some other act of dubbing. The name is then transmitted from speaker to speaker through a causal-historical chain: each speaker who uses the name does so because they learned it from someone who learned it from someone who... ultimately traces back to the original baptism. As long as the chain is unbroken, the name refers to the original individual, regardless of what descriptions the later speakers associate with the name.

This theory has striking consequences. A speaker can use the name "Aristotle" to refer to Aristotle even if every description they associate with the name is false — even if they believe Aristotle was a Roman general or a Chinese sage. As long as they acquired the name through the right causal chain, it refers to Aristotle. This seems counterintuitive to the descriptivist, who holds that the name means whatever the speaker takes it to mean, but it accords with our actual practice. We routinely use names to refer to people about whom we know almost nothing, and our reference does not fail just because our descriptions are wrong.

Natural Kinds

Kripke extended his theory of rigid designation from proper names to natural kind terms — terms like "water," "gold," "tiger," and "heat." These terms, he argued, function like proper names: they are rigid designators that refer to the same kind in every possible world. "Water" refers to H2O in every possible world, not because we define "water" as "the stuff with the chemical structure H2O," but because water is H2O, and the term "water," introduced by a baptismal act pointing at samples of the liquid, rigidly designates that kind.

This has profound implications. It means that "Water is H2O" is not just an empirical discovery about the actual world; it is a necessary truth. In every possible world, water is H2O — there is no possible world in which water is not H2O. But this necessity was discovered empirically, through chemical investigation. It is not knowable a priori — before the eighteenth century, no one knew that water was H2O, and there was no way to know it without doing chemistry. Kripke's conclusion: "Water is H2O" is necessary a posteriori — necessary, but knowable only through experience.

Necessary A Posteriori Truths

The category of the necessary a posteriori is Kripke's most original and far-reaching contribution. Before Kripke, philosophers had generally assumed that necessity and apriority go together: what is necessary is knowable in advance of experience, and what is empirical is contingent. Hume had argued that all empirical truths are contingent ("relations of matters of fact"), and Kant had argued that necessity is a mark of the a priori. Kripke shattered this assumption. "Water is H2O," "Hesperus is Phosphorus" (the morning star is the evening star), and "Aristotle was a human being" are all necessary truths — true in every possible world — but they are also empirical discoveries, knowable only through investigation. The distinction between necessity and apriority, which philosophers had conflated for centuries, turns out to be a genuine distinction, and it has consequences for virtually every area of philosophy.

Key Arguments

Against Descriptivism: The Modal Argument

Kripke's modal argument is the most direct assault on the descriptive theory. Consider the name "Aristotle." According to descriptivism, "Aristotle" means "the teacher of Alexander the Great" (or whatever description we associate with the name). Now consider a possible world in which Aristotle never taught Alexander — perhaps he died young, perhaps someone else got the job. In that world, the description "the teacher of Alexander the Great" refers to someone else, or to no one. But the name "Aristotle" still refers to Aristotle. If it did not — if "Aristotle" referred to whoever taught Alexander — then the sentence "Aristotle might not have taught Alexander" would be false or trivial, when in fact it is true and informative. The name, unlike the description, picks out the same individual in every possible world. Descriptions are non-rigid; names are rigid. Therefore, names are not descriptions.

The Epistemic Argument

Kripke's epistemic argument challenges descriptivism from a different angle. Consider the sentence "Aristotle was the teacher of Alexander the Great." If "Aristotle" means "the teacher of Alexander the Great," then this sentence is analytic — true by definition — and it would be inconceivable that Aristotle did not teach Alexander. But we can perfectly well imagine Aristotle not teaching Alexander; it is a genuine epistemic possibility. This shows that the name "Aristotle" does not mean the same as the description "the teacher of Alexander the Great." If it did, denying the description would be a contradiction, not a genuine possibility.

The epistemic argument reveals that descriptions, even when true of an individual, are not part of the meaning of a name. They are things we believe about the bearer of the name, not things that constitute the name's meaning. This distinction between what a name means and what we know about its bearer is the key to Kripke's positive account: the name means the individual, and our beliefs about that individual are separate from the name's semantic function.

The Semantic Argument

Kripke's third argument concerns the way names are actually used in communication. Suppose two people, A and B, both use the name "Cicero." A believes that Cicero was a Roman orator; B believes that Cicero was a Roman general. On the descriptivist view, A and B are using "Cicero" with different meanings — A means "the Roman orator" and B means "the Roman general." But they are clearly talking about the same person, and they can disagree about whether Cicero was an orator or a general — a disagreement that would be impossible if they were using "Cicero" to mean different things. This shows that the reference of a name is not determined by the descriptions associated with it but by something else — the causal-historical chain connecting the speakers to the original bearer.

The Circularity Problem

Kripke also raises an internal difficulty for descriptivism. If "Aristotle" means "the student of Plato," how do we know who "Plato" is? By the descriptivist's own account, "Plato" must mean "the teacher of Aristotle." But then "Aristotle" means "the student of the teacher of Aristotle," which is circular. The descriptivist cannot define every name in terms of other names without eventually going in a circle, and the only way out is to admit that at least some names refer directly — not via descriptions. But if some names refer directly, the descriptivist's general thesis is false. This argument, though compact, is devastating: it shows that descriptivism cannot account for the entire network of names without either circularity or an appeal to direct reference that undermines the theory's own foundations.

Legacy

Naming and Necessity has been called the most influential work in analytic philosophy since the Second World War, and the assessment is not hyperbolic. Its impact has been felt across virtually every area of philosophy.

In the philosophy of language, the book overturned the descriptive theory of names that had held sway for decades and replaced it with the causal-historical theory of reference. The details of the causal theory have been debated, refined, and challenged — by Gareth Evans, by Michael Devitt, by Tyler Burge — but the central insight, that names are rigid designators whose reference is not determined by descriptions, has become the new orthodoxy. The book also raised new questions about the relationship between reference and meaning, about the role of the community in fixing reference, and about the nature of the "baptism" that initiates a naming practice.

In metaphysics, Kripke's distinction between necessity and apriority reopened questions that had been considered settled since Kant. If some empirical truths are necessary, then the category of the necessary is not confined to logic and mathematics. This opened the door to a revival of metaphysical modality — the study of what is possible and necessary — as a respectable philosophical enterprise, after decades in which it had been dismissed by logical positivists as empty speculation. Kripke's framework, extended in his later work on the essentialist implications of rigid designation (if "Aristotle" rigidly designates Aristotle, then it is essential to Aristotle that he is Aristotle — he could not have been someone else), has reinvigorated metaphysics and given it new tools and problems.

In the philosophy of mind, Kripke's arguments have been applied to the mind-body problem. If pain is identical to a certain neural state (C-fiber firing, say), then this identity is necessary — just as the identity of water and H2O is necessary. But we can imagine pain without C-fiber firing and C-fiber firing without pain, which means the identity is not a posteriori necessary but, on Kripke's framework, suggests that the identity claim is false. This argument, which Kripke sketched briefly in the lectures and developed in the postscript, has been one of the most influential challenges to physicalist theories of mind. David Chalmers has used Kripke's modal framework to develop the Hard Problem of consciousness, and the debate over whether mental-physical identities can be necessary a posteriori truths continues to animate the philosophy of mind.

In epistemology, the discovery of necessary a posteriori truths has reshaped the landscape. The traditional Humean picture, in which all empirical truths are contingent and all necessary truths are a priori, no longer holds. This has opened new questions about the nature of knowledge, the relationship between empirical inquiry and modal truth, and the status of philosophical thought experiments. It has also influenced the philosophy of science, where the necessary a posteriori framework has been used to analyze theoretical identities in science ("water is H2O," "heat is molecular motion," "light is electromagnetic radiation").

The enduring significance of Naming and Necessity lies in its combination of philosophical depth with extraordinary clarity of argument. Kripke writes in plain English, with minimal technical apparatus, and his arguments proceed step by step from premises that any reasonable philosopher would accept to conclusions that overturn the assumptions of a century. The book is short — the lectures themselves run to fewer than a hundred pages — but its arguments have filled libraries of commentary. It remains, more than fifty years after its delivery, a work that no serious student of philosophy of language or epistemology can afford to ignore — a text that changed the questions philosophers ask about knowledge, truth, and the understanding of reality.