
A History of Philosophy | 80 Philosophy of Language
What this covers
This is a single university lecture, one installment in a history-of-philosophy series from Wheaton College, that surveys twentieth-century philosophy of language. It begins with the logical positivist attempt to reduce language and clear out metaphysics. It then moves through Frege's split between sense and reference, Quine's view that meaning belongs to language rather than to mental states, and Quine's thoroughgoing nominalism about universals like "roundness," which he treats as collective terms for resembling things rather than names of real essences. The lecturers are not named in the source.
The middle of the lecture turns to structuralism. Following Ferdinand de Saussure, words are arbitrary signs whose relations differ across languages, so different languages construct different worlds of experience. Derrida's deconstruction presses this further by showing where verbal schemes fail. Noam Chomsky stands out as the dissenter, positing a universal deep structure in a more Kantian spirit. The lecture also follows anti-realism into philosophy of science through Mach, Kuhn's paradigm shifts, and Paul Feyerabend's radical relativism. Hilary Putnam's provisional realism is offered as a middle path, and Rorty's critique of the mirror of nature appears as the postmodern endpoint.
The final stretch is where the lecture becomes most contested. It treats possible worlds and the puzzle of transworld identity, setting Nelson Goodman's view of possible worlds as a semantic device against Nicholas Wolterstorff's account of universals as objective logical possibilities, which he extends to aesthetics through the "ping" and "pong" example. The lecturer then carries these ideas into theology: Plantinga's possible-worlds argument for God's existence, whether God has an essence and so could sin, and the claim that logic's laws are laws of being and therefore of God. The distinction between logical and metaphysical possibility also comes up, illustrated with a fairy giraffe.
Recent developments in philosophy of language have reopened metaphysical and ontological questions that logical positivism attempted to eliminate, particularly around logical entities, universals, realism versus anti-realism, and possible worlds.
- Frege and Popper distinguished logical objects from physical objects and mental states, establishing that language refers to non-physical logical entities
- Structuralist philosophy of language shows how different linguistic frameworks structure different worlds of experience, leading to anti-realist positions
- Debates about possible worlds and counterfactual statements require either accepting logical entities as real or reducing everything to empirical hypotheses
This asset isn't compiled yet
You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.
Logical positivism with its reductionist view of language eliminated metaphysics by denying the existence of logical entities or universals, but newer developments in philosophy of language have reopened these metaphysical questions by recognizing that linguistic categories relate to ontological categories.
“the logical categories uh evident in language uh would be related to ontology. Uh and particularly when we're talking of the uh demise of logical positivism and the way in which its reductionist view of language eliminated metaphysics it's understandable that um new views of language would um open up to metaphysics.”
According to Quine, meanings are functions of language and what speakers refer to, not of mental states or intentions; thought itself is verbal behavior or the formulation of language symbols in the mind, not uniquely psychological activity.
“meanings are functions of language, not of mental states. The question when something is uttered is not what are you thinking of, but what are you referring to? What are you talking about? You see. He takes language, that is to say, he takes language to be verbal behavior, a kind of overt physical behavior. He's only interested in thought in so far as it is expressed in the symbolism of language. Thought is not psychological activity uniquely. It is verbal behavior.”
Debate about counterfactual statements in philosophy of language raises the technical problem of transworld identity: determining whether an object in one possible world is the same object as in another possible world, which requires determining what properties are essential to an object's identity—whether Socrates without a hooked nose would still be Socrates.
“there's a debate goes um about counterfactuals, about um what's called transworld identity. That is to say whether the Socrates of another possible world in which um shall we say um uh he uh did not have a hooked nose whether that Socrates would still be the same Socrates. Now the question is what constitutes the same Socrates. Is it of the essence of Socrates that he have a hooked nose?”
Alvin Plantinga uses possible worlds logic in philosophy of religion, arguing that if a logically possible world exists in which God necessarily exists, then this removes objections to God's existence and shows that God's existence is logically possible, shifting the burden on objectors to identify their specific objection rather than proceeding from the assumption that God's existence is impossible.
“Alvin Plantinga in his book The Nature of Necessity worked with uh this sort of issue and argued that um there is a logically possible world in which God necessarily exists. Now you see Plantinga is not given to trying to demonstrate the existence of God. He's given to trying to remove objections to the existence of God. And if you can show that there is a logically possible world in which God necessarily exists then um you've removed any objection. You see? And it's logically possible then that God exists. And so uh his next question to the objector is okay, now what's your objection?”
Whether God could sin raises the question of whether God has an essence: if God has an essence (specifically, essential perfect goodness), then God necessarily cannot sin because sinning contradicts being God; if God has no essence, then the word 'God' is an arbitrary name with no conceptual meaning, making it meaningless to say God cannot sin.
“would it possible logically that God could sin? Right. Yeah and you see he would say now that resolves into the question as to whether God would still be God if he did. Because God by definition in theistic religions is that is one who is altogether good. You see? In other words it resolves into the question does God have an essence? Which is the title of some lectures that he gave at Marquette University... If God has an essence then there are some things God cannot do. Um God cannot be God. No, God cannot be not God. You see? God cannot be not God. If being perfectly good is of the essence of God if sinning is imperfection then God cannot sin.”
Subordinating God to the laws of logic is not a problem because the laws of logic are laws of being itself, and God is ultimate being and creator of all other beings; therefore the laws of logic are ultimately laws of God, and God as ultimate being cannot violate them.
“Now, you know, if you want to say, "Oh, that's subordinating God to the laws of logic." No, it's it's saying that God is a being. The laws of logic are laws of being, ultimately the laws of God. Because they're laws of being. God is ultimate being, creator of all other beings. So, the laws of being are laws of God.”
Every being has at least some essence—the essence of being itself, which is to be and not not-to-be, instantiating the principle of non-contradiction as fundamental to being rather than as an external constraint.
“every being has at least some essence, the essence of being, which is to be, and not not to be.”
Frege distinguished between sense and reference in language: the sense (or meaning) of a sentence is the logical proposition or objective state of affairs it expresses, while the reference is what the sentence is about—physical objects, sense data, or other entities. These are fundamentally different aspects of meaning.
“a sentence um stands for a proposition. Uh where a proposition is not just a sentence, a statement uh but is an objective state of affairs... An objective state of affairs which can be talked about in a variety of sentences.”
Deconstruction, as practiced by Derrida, attempts to undo what structuralism claims we have structured by showing that verbal schemes don't work completely or work inconsistently, demonstrating that language dominates our experience and prevents us from seeing and talking about things in alternative ways.
“What is deconstruction? Well, it's undoing what the structuralist says we have structured. Uh what Derrida, the deconstructionist, is doing, you see, is trying to deconstruct uh the verbal schemes which Derrida builds uh to show uh that that really doesn't work completely. Or that there are a variety of languages there that work inconsistently. Uh it is our language that dominates our world of experience and keeps us from seeing it and talking about it in other ways that we might.”
Wolterstorff applies his theory of logical possibilities to aesthetic theory, arguing that there are objective universal aesthetic associations between certain sounds and qualities across cultures—for instance, 'ping' sounds are associated with delicate qualities while 'pong' sounds are associated with boisterous qualities—demonstrating an objective basis for aesthetic values rather than purely subjective or relativistic responses.
“he argues for that by pointing out uh that there are similar associations to the same sounds in all sorts of different languages... he'll go running through scores of examples like that to illustrate the point. Similar associations of similar sounds... His point is that there is an objective basis for certain aesthetic associations and qualities.”
Richard Rorty's postmodern philosophy of language, in his book 'Philosophy and the Mirror of Nature,' critiques the representational theory of knowledge and the foundationalism associated with Locke's mirror metaphor, arguing instead that philosophy is simply a pragmatic conversation without access to truth about things in themselves.
“Richard Rorty in his um now famous book on philosophy and the mirror of nature... he is doing is um insisting on the failure of that representational theory of knowledge and the foundationalism with which it was associated... he sees philosophy not as giving us access to the truth at all about things in themselves, but as simply a conversation uh that has pragmatic value.”
Structuralism in linguistics, following Ferdinand de Saussure, holds that language consists of arbitrarily assigned words (signs) that refer to empirical objects, but these words are interrelated to each other in different ways across languages; since different languages structure the relationships between words differently, they construct different worlds of experience with different structured meanings.
“According to uh de Saussure, language consists of arbitrarily assigned words that are signs that refer to empirical objects. Uh these arbitrarily assigned terms are interrelated to each other. The point is that you get therefore a variety of different languages, not only because of the differences of words, but because of the differences about the relationships between words.”
Quine is a thoroughgoing nominalist who treats universal terms like 'roundness' and 'humanness' as syncategorematic terms—words used to refer collectively to all members of a category—rather than names for real essences; the category itself is merely a linguistic classification that organizes similar things according to empirical resemblances, not according to shared logical objects.
“universal terms like roundness, humanness, are not names for essences, logical objects. Uh they um are simply terms uh that refer to any member of a whole set of similar particulars. He's a thoroughgoing nominalist.”
Noam Chomsky differs from structuralists like de Saussure and deconstructionists like Derrida by arguing that beneath the surface structure of language (which varies across languages) there is a universal deep structure shared by all languages, representing a more Kantian position that preserves universalism even within structural linguistics.
“I mentioned the name of Chomsky, the structural linguistic. Um the difference with Chomsky is that he's much more Kantian in that he thinks there is a universal in-depth structure shared by all languages. You see, a universal depth structure in addition to this kind of surface structure, as he calls it, which Saussure seems to be talking about. But the deconstructionist sees no depth structure.”
Thomas Kuhn's paradigm shifts in his 'Structure of Scientific Revolutions' support Goodman's view that with each paradigm shift comes a new language and new way of structuring phenomena, reinforcing the structuralist idea that scientific knowledge is relativistic and theory-laden rather than about discovering objective reality.
“There's some influence here from Thomas Kuhn with his structure of scientific revolutions. You see. Uh Kuhn who has uh recognized that with paradigm shifts, you get, as Goodman would say, a new language, a new way of structuring things.”
Paul Feyerabend takes structuralist anti-realism in science to its extreme, asserting that all scientific knowledge is fundamentally relativistic and rejecting the possibility of any privileged or correct scientific picture, representing the most radical form of anti-realism in philosophy of science.
“The person who in philosophy of science has taken it um uh to the relativistic extreme is a man named Feyerabend. Uh Feyerabend, who um uh is much less given to talking about a correct picture or pictures. And is very blunt about the relativity of all scientific knowledge.”
Nelson Goodman treats possible worlds language as merely a verbal device or semantical trick, maintaining that all possible worlds are simply linguistic constructs where the reference points are things we experience or might possibly have experienced; thus other possible worlds are just hypothetical statements about what this world might have been, making Goodman an anti-realist about possible worlds.
“For Nelson Goodman, the language of possible worlds is uh merely a verbal device. It's a semantical trick. Um all possible worlds are simply linguistic constructs. Where the particular points of reference are things we all experience or might possibly have experienced. Like my first child being a girl. You see? So in that sense another possible world is simply a a hypothetical world. It's a hypothesis about what this world might have been or might yet be. You see? So the language of possible worlds is just the language of empirical hypotheses and nothing more.”
Nicholas Wolterstorff's theory of universals reintroduces logical objects by treating universals as objective logical possibilities or 'possibles'—unchanging states of affairs that cannot be other than they are, bound by the law of identity—distinct from actual physical entities but objectively real in constraining what can exist in any possible world.
“what uh Wolterstorff does... is to speak of universals as possibles. Uh that is to say um unchanging logical possibilities that are objectively real in the sense that they cannot be something other than they are. They're bound by the law of identity.”
Hilary Putnam at Harvard rejects both foundationalism and radical relativism, arguing for 'provisional realism' about science: while granting that alternative constructs are possible and theories are subject to revision, he insists that scientific theories are genuinely about reality and not merely conventional language games, because there are firm observations and material entities (like electrons and force fields) that all scientists observe and measure regardless of theoretical framework.
“Hilary Putnam... is quick to grant that alternative constructs are certainly possible. And that scientific theories of course are subject to revision... But he still wants scientific theories to be taken realistically. And he insists that our constructs are not just conventional ways of talking... he justifies it by saying that uh we have firm knowledge of certain observations and material entities... the sorts of things that all scientists observe and measure regardless of their theoretical constructs.”
Metaphysical possibility should be distinguished from logical possibility: something can be logically possible (no logical contradiction) without being metaphysically possible (causally or physically possible); for instance, a 'fairy giraffe with butterfly wings' is logically possible but may not be metaphysically possible.
“what is metaphysically possible is what is logically possible... To say it's logically possible is to say there's no objection. You see? There's no logical objection. It's logically possible. Now is it actually possible? If we were talking of um a fairy giraffe with butterfly wings would say it's logically possible. Uh but is it actually causally possible?”
Ernst Mach argued that scientific theories are merely economical ways of describing relationships between sense data, implying that there are multiple equally economical and equally valid alternative scientific descriptions.
“We heard Ernst Mach say that. When when Mach said that a scientific theory is just an economical way for describing the relationships between sense data. An economical way. Well, there are various economical ways of doing it.”