The Overcoming of Metaphysics through the Logical Analysis of Language (often Englished as “The Elimination of Metaphysics through Logical Analysis of Language”) is a short, polemical philosophical essay rather than a book, and its argument is tightly concentrated on a single thesis: that the sentences of metaphysics are not false but meaningless. Carnap frames the essay as an application of the logical analysis of language developed within the scientific, anti-metaphysical wing of Viennese philosophy, and he states at the outset that the aim is not to refute metaphysical doctrines one by one but to show that they fail to be statements at all. What looks like a claim about reality is, on his account, a pseudo-statement (Scheinsatz) — a string of words that is grammatically well-formed and emotionally charged but asserts nothing that could be true or false.
The argument proceeds by asking what makes a sentence meaningful. Carnap divides meaningful discourse into two exhaustive classes. On one side stand the analytic sentences of logic and mathematics, which are tautological and tell us nothing about the world; on the other stand the sentences of empirical science, which are meaningful because they can be tested against experience and can in principle be confirmed or refuted by sentences describing what is directly given. Metaphysics, however, claims to deliver knowledge of a reality lying beyond all experience, while its sentences are neither tautological nor empirically checkable. Since every meaningful sentence belongs to one of the two classes, metaphysical sentences belong to neither, and are therefore pseudo-statements. Carnap stresses that this is a logical verdict about the status of the sentences, not a moral or psychological criticism of the people who utter them.
He then distinguishes two ways in which a pseudo-statement can arise. In the first, a word occurs that has no meaning in the context, because no criteria have been fixed for its application. Carnap illustrates this with invented example-words: one can attach a new word to a sentence and even lay down a rule that it applies to a thing only if some specified condition holds, but if no such test is ever given, the word is a mere sound and the sentence built from it says nothing. He draws the general conclusion that the meaning of a word lies in the rules governing its use — above all, for empirical words, in the possibility of deciding whether sentences containing it hold. In the second kind of pseudo-statement, every single word is meaningful, but the words are combined in a way that offends against logical syntax: predicates are attached to subjects of the wrong logical category, so that the sentence has the outward form of an assertion while the combination is excluded by the grammar of meaningful language. Carnap treats this “counter-syntactical” nonsense as the characteristic error of speculative philosophy.
The essay’s most famous pages apply this analysis to a single extended example: Heidegger’s inaugural lecture What is Metaphysics? Carnap quotes a series of its sentences about “the Nothing” — sentences in which the Nothing is said to “nothing,” to negate itself, to be more originary than negation. His diagnosis is that the ordinary word “nothing” functions in language as a logical particle, the expression of a negation or a quantifier, not as the name of an object; Heidegger converts it into a noun, supplies it with a pseudo-verb coined from that noun, and thereby produces sentences whose words are individually familiar but whose combination is logically inadmissible. Carnap is careful to say that Heidegger serves as an illustration rather than as a uniquely bad case, and he points out that comparable confusions run through the history of metaphysics, naming Hegelian dialectic among other examples. He also argues that many metaphysical theses arise from a confusion between two ways of speaking — between talking about the objects a language describes and talking about the language itself — so that a statement whose real subject matter is linguistic form is mistaken for a statement about reality.
Having denied metaphysics any theoretical content, Carnap does not conclude that it is simply noise to be despised. In the later part of the essay he argues that metaphysical sentences have an expressive function: they give voice to a Lebensgefühl, a total attitude toward life and the world, in something like the way lyric poetry and music do. The trouble is that the metaphysician expresses this attitude in the misleading form of theory, so that a cry of feeling is dressed up as a piece of knowledge — hence Carnap’s remark that “metaphysicians are musicians without musical ability.” Art, he suggests, is an adequate vehicle for such expression; metaphysics is a substitute for art that misunderstands itself. He adds that the elimination of metaphysics from the domain of knowledge leaves untouched the emotional needs that drive people toward it, and that the historical rise and influence of metaphysical systems is a legitimate subject for psychology, sociology, and history rather than for logic. He also explicitly refuses to treat his analysis as a judgment on the moral seriousness or sincerity of metaphysicians: the question is what their sentences mean, not what kind of people they are.
The essay closes on this division of labour. Logical analysis dissolves the apparent cognitive claims of metaphysics and assigns what remains of them to the sphere of feeling and expression, while empirical science and logical syntax divide between them the whole field of meaningful speech.
The essay operates with a criterion of cognitive meaning drawn from the practice of science: a sentence says something only if it is either analytic — a tautology of logic or mathematics — or empirically testable, in the sense that observations can be specified which would count for or against it. Metaphysical sentences satisfy neither condition. This yields a distinction the essay uses repeatedly: between words that have no meaning at all and combinations of meaningful words that violate logical syntax. The metaphysician’s vocabulary — “the absolute,” “the unconditioned,” “the infinite,” “being,” “the thing-in-itself,” “the principle” — consists largely of pseudo-concepts: expressions with the grammatical form of names but no application-conditions. Carnap asks what would count as identifying an instance of “the principle” or “the absolute,” and finds that the metaphysician can only answer by invoking further pseudo-concepts. Such words are not vague but empty of cognitive content.
The essay’s central term for the metaphysical sentence is the pseudo-statement (Scheinsatz): a sentence with the form of an assertion that asserts nothing. Because it is not false, it cannot be contradicted. This is what the word “overcoming” in the title signals: logical analysis does not refute metaphysics, it shows that there is no cognitive claim there to refute. A false statement can be met with evidence; a pseudo-statement can only be exposed as failing to meet the conditions under which any sentence can say something. Carnap therefore declines to argue with the metaphysician on the latter’s own ground and asks instead whether the sentence belongs to the class of sentences capable of truth or falsity.
The essay also allows that some metaphysical sentences are not pure nonsense but disguised empirical or logical statements. In such cases logical analysis can sometimes recover a meaningful content by translation — for instance, a metaphysical assertion about the unity of the world may be a confused way of stating the unity of science or the unity of the language of science. Carnap treats this recovery as the exception rather than the rule. Theological sentences fall on the other side: statements about a transcendent God have no empirical application-conditions and thus no cognitive meaning, though they may express an attitude.
Philosophy, on this account, is not a science with its own subject matter but an activity of logical clarification. The essay assigns to philosophy the task of analysing the language of science — distinguishing meaningful from meaningless combinations — rather than adding to the stock of knowledge. Carnap explicitly rejects the charge that this amounts to a new “dogmatic positivism”: the claim is not that metaphysics is false but that its sentences fail the conditions under which any sentence can be true or false. The essay’s significance lies in the consistency with which it applies this method to a contemporary text and in its insistence that the critique of metaphysics is a matter of logic rather than of taste or temperament.
