The Elimination of Metaphysics Ken Matheson, St. Francis Xavier University

The Elimination of Metaphysics Ken Matheson, St. Francis Xavier University

In Language, Truth and Logic, A. J. Ayer is harshly critical of metaphysicians engaged in theorizing about a reality that transcends the limits of sense experience and particularly of metaphysical propositions, declaring that metaphysicians produce nonsense. He argues that protracted metaphysical debates have not reached consensus nor resolution and have not generated knowledge comprising ‘matters of fact’. He wants to show that both the metaphysical propositions and the questions that the metaphysician investigates are literally senseless. His goal is to eliminate metaphysical debates from philosophical investigations.

In this essay, I summarize and comment on Ayer’s argument with emphasis particularly on demarcation–the boundary between that which is sensible and nonsense. I briefly review Karl Popper’s earlier proposal for demarcation and then review and comment on Ayer’s alternate proposal, and, finally, I show that, although the demarcation models distinguish metaphysical propositions as separate from empirical propositions, the models do not necessarily support the claim that there is no value in metaphysical debate. Notwithstanding Ayer’s goal, metaphysical debate continues.

Ayer declares that "no statement which refers to a ‘reality’ transcending the limits of all sense-experience can possibly have any literal significance.” In other 1 words, statements, such as metaphysical propositions, that are not, or cannot be, supported by empirical evidence are nonsense. He supports his declaration with a claim that longstanding metaphysical disputes are unwarranted and that they have produced no meaningful knowledge. He postulates that knowledge that transcends 2 the world of science and common sense is not possible and investigation of, and discussion about, such metaphysical knowledge is a waste of effort.

Ayer acknowledges that he is not the first to claim that metaphysics is meaningless. He cites Kant’s claim that the human mind is incapable of

understanding anything beyond the phenomenal world. Unlike Kant, Ayer does not 3 want to confront the practice of metaphysics, noting that attacking the soundness of a metaphysical argument cannot eliminate dispute and that showing the weakness of an argument, by itself, cannot refute a metaphysical conclusion. Weakness in an argument lessens the certainty of the truthfulness of a conclusion but does not prove that the conclusion is wrong. Ayer wants to attack the nature of metaphysical propositions to show that, as a matter of logic, they are meaningless and that a transcendent metaphysic is impossible.

He asserts that a metaphysical question that is the subject of dispute is not a genuine question and that a metaphysical proposition is not literally significant. Ayer argues that a proposition or claim (metaphysical or otherwise) is literally significant (or factually significant) to a person if, and only if, the person knows what observations and empirical evidence would confirm or refute the claim.4 For Ayer, observed evidence is anything that one could experience through sense perception. Questions, as well as propositions, are genuine only if the questioner knows what observations would lead to its answer; it is genuine only if it could be answered through observation of empirical evidence. He contends that a question such as “do substantial Forms exist” is not a genuine question because there are no experiential observations that could inform an answer. Plato postulated his theory of Forms as a possible explanation for consistencies that he observed in the world, but he never claimed to have seen, tasted, smelled, heard or touched a Form. Similarly, for the proposition “souls are eternal” it is unlikely that one could specify experiments or investigations that would yield direct evidence of the veracity of the claim. Hence, in Ayer’s opinion, debate over metaphysical questions and propositions such as these produces nonsense.5

In order to eliminate metaphysics and thereby avoid the waste of labour on the production of nonsense, Ayer wants to make a clear demarcation between nonsense and that which is literally significant. Demarcating the Empirical Science Popper’s Principle of Demarcation

In Karl Popper’s The Logic of Scientific Discovery , published a year earlier 6 than Ayer’s Language, Truth and Logic, Popper describes his principle of demarcation that establishes the boundary between the empirical sciences and non-

Ibid, 5.3

Ibid, 6.4

empirical fields of inquiry such as mathematics, logic and metaphysics.7 Popper’s intent is to “define the concepts ‘empirical science’ and ‘metaphysics’” in a way that one can clearly know whether a field of study is part of empirical science. He considers the view that a system (set of propositions) is part of empirical science if and only if it can be tested to produce experiential evidence, and he finds that it is an unsatisfactory demarcation. He argues that defining empirical science as “only those statements which are reducible to elementary (or ‘atomic’) statements of experience” is a demarcation based on verifiability, and the laws of nature, the foundation of empirical science, fall on the wrong side of the demarcation because the laws of nature are not practically verifiable. For example, the law of gravitation says that every particle attracts every other particle in the universe. One could make an inference based on a few observations, but certainty requires verification of attraction between all pairs of particles in the universe. As there is an infinite number of particles in the universe, verification requires an infinite number of observations–an impossibility. The laws of nature are not propositions, they are simply observations. An observation can verify a proposition, but an observation by itself is not verifiable, it is simply an instance of a sense perception. For the laws of nature, one can only say that, in the absence of contradictory evidence, they are probably true. Consequently, demarcation on the basis of verifiability, excludes the laws of nature from empirical science. Popper also points out that much of the practice of empirical science begins with a theorem for which there may be no empirical evidence and, as such, would be excluded by the demarcation based on lack of verifiability. Although theorems are essential to empirical science, 8 demarcation on the basis of verifiability excludes them from the science.

Popper proposes a demarcation based on falsifiability rather than verifiability. He asserts that a proposition is within the scope of empirical science if it is falsifiable. A claim of falsifiability asserts that testing for falseness of the proposition could yield empirical evidence. If one can specify what evidence would demonstrate that a claim is false, then the claim is falsifiable, and one could, theoretically, design an experiment to gather such evidence. Demarcation by falsifiability admits all empirical propositions and also the laws of nature. Consider, for example, Newton’s first law of motion–an object does not change its motion unless a force acts on it. The law is an elementary statement of experience, consistently observed but unprovable because, once again, one would have to observe an infinite number of occurrences of a body in motion to prove that the proposition is true in every case. However, if we observed an object changing its

motion without a force acting on it, that evidence would refute the law, and consequently, the law is a falsifiable proposition. It is theoretically possible that someone could design an experiment to show that an object changes its motion with no external force acting on it, and demarcation by falsifiability admits the law to empirical science.

By contrast, one cannot falsify a proposition if one cannot conceive of evidence that would refute the claim and such a proposition must, therefore, be excluded from empirical science. For example, the proposition “souls are eternal” is not falsifiable. It is impossible for one to imagine a testing scenario that would yield evidence, by sense perception, that demonstrates death or annihilation of a soul.

Popper’s claim is that falsifiability of propositions demarcates the full scope of empirical science.

Ayer’s Principle of Demarcation

Ayer seems to reject Popper’s demarcation by falsifiability because it requires conclusive falsification, and instead, he proposes demarcation by verifiability. Ayer’s argument against falsifiability does not seem to be in opposition to the purpose for which Popper uses falsification, rather, Ayer argues that, just as it is impossible to conclusively verify a proposition, so too, it is impossible to conclusively refute a proposition. He argues that empirical evidence is always gathered under specific conditions and that any change in conditions could affect the observations and thereby amend our belief in the truthfulness or falsehood of a claim. He asserts that “we cannot hold that the genuineness of a proposition depends on the possibility of its definite confutation.”9 Repeated, consistent supporting evidence improves confidence in the veracity of a claim, however, Ayer argues that claims of truthfulness or falsehood are only claims of probability and not conclusiveness, because, under different conditions, one could make contrary observations. In challenging demarcation on the basis of falsifiability, Ayer seems to be disputing an interpretation of Popper’s demarcation principle that Popper did not assert. While it may be true that conclusive confutation is not possible, for example, when one can imagine refuting evidence but cannot conceive of an experiment that might produce the evidence, Popper introduces falsifiability because it admits the possibility of empirical evidence. In my reading of The Logic of Scientific Discovery, I do not find that Popper suggests or implies that a proposition is falsifiable if, and only if, it could be conclusively refuted. Whether or not a proposition can be conclusively refuted is not part of Popper’s principle of demarcation. Ayer’s dismissal of demarcation by falsifiability, solely on the basis of the impossibility of conclusive confutation, may be unwarranted.

Ayer proposes that significant (genuine) propositions are demarcated by 10 verifiability. To set his parameters for what can be verified, Ayer specifies two types of genuine propositions: ‘a priori’ and ’matters of fact’. A priori propositions are either tautologies (statements that cannot be false) or conventions or definitions about the meaning of certain symbols or words. For example, the statement “velocity is the rate at which an object moves and is expressed as distance [travelled] in a unit of time” is a convention or definition of “velocity”, and it is literally significant because there is no doubt of its veracity. While a priori propositions cannot pass Ayer’s test of supporting empirical evidence, he nevertheless considers them to be literally significant because it is not possible for empirical evidence to refute the assertions.

For Ayer, ‘matters of fact’ are propositions that make a claim about the state of affairs in the material world and, he postulates, that empirical evidence can verify the probability of, but not the certainty of, the truth of such claims. He holds that propositions that are generalizations, induced from observations, are only probable and never certain. For example, the proposition that “all birds fly” is based on observations of birds in flight and it is held to be true until one observation of an emu that has no capability of flight. The proposition, prior to the observation of the emu, is literally significant because it is based on observation, but it is only probable because it is subject to refutation by subsequent observations, and more importantly because one can never conduct an infinite number of observations.

While studying in Vienna, Ayer had some contact with the Vienna Circle and 11 met some philosophers who demanded conclusive verifiability as the standard for literal significance. Ayer, however, adopts a less stringent standard that he terms “verifiable in the weak sense.”12 He requires only that “some possible senseexperience should be relevant to the determination of its truth or falsehood.” In 13 other words, he concedes literal significance of a proposition if one can imagine a scenario that could provide verification, even if no evaluator has completed, nor could complete, all possible verification tests. For verification in the weak sense, it is enough to imagine a verification scenario and to precisely and clearly specify the empirical evidence that the scenario would produce. It is not necessary that the testing scenario has occurred and has produced the empirical evidence. For example, the proposition “the earth’s core is molten lava” has literal significance because a scientific model supports the hypothesis, and, if one built a probe that

Ayer seems to treat “significant proposition” and “genuine proposition” as synonymous.

The Vienna Circle was a group of early twentieth-century philosophers who sought to reconceptualize empiricism.

could withstand the extreme temperatures of the core, it could provide confirming or refuting evidence. That one can imagine a testing scenario is sufficient for Ayer to consider the proposition “validated in principle” and to deem literal significance.

Ayer declares his principle of demarcation as follows:

If a putative proposition fails to satisfy this principle [that some possible sense-experience should be relevant to the determination of its truth or falsehood], and is not a tautology, then I hold that it is metaphysical, and that, being metaphysical, it is neither true nor false but literally senseless.14

By this principle, tautologies, because they are never false, and propositions that are verifiable comprise the totality of statements that have literal significance, and all other propositions do not have literal significance. Based on this principle, Ayer’s test for literal significance of a proposition is as follows:

Could some experiential proposition(s) be deduced from it in conjunction with certain other premises without being deducible from those other premises alone?15

In other words, if, based on a proposition and the rest of our scientific knowledge, one can imagine some actual or possible observation that could not have come from the rest of the scientific knowledge without this proposition, then the proposition has literal significance. If the proposition is not essential to actual or possible observation, then the proposition has no literal significance.

In a manner of speaking, Ayer eliminated metaphysics by the stroke of his pen. His principle of demarcation declares that all propositions that lack literal significance are metaphysical, thereby eliminating them from empirical science. He claimed at the outset he was going to show, as a matter of logic, that a transcendent metaphysic is impossible, however, what he has shown is that a metaphysical proposition is not verifiable and is not really a proposition.

Ayer claims that metaphysical propositions are not genuine propositions, but that statement seems to be self-contradictory and uncharitable. He asserts that “a metaphysical sentence [is] a sentence which purports to express a genuine proposition, but does, in fact, express neither a tautology nor an empirical

hypothesis.” By genuine proposition, Ayer means that one has provided, or could 16 provide empirical evidence to support the proposition. If one were to modify his assertion by replacing “genuine” with “empirically supportable” then it reads: “metaphysical hypotheses are sentences that purport to express an empirically supportable proposition”, and that seems like an internal contradiction. If a statement is a metaphysical proposition, then by Ayer’s definition, it could not legitimately purport to be an empirical hypothesis. Further, the assertion seems to be an uncharitable characterization of metaphysical hypotheses. In his Meditations, when Descartes said “God necessarily exists” (see Meditation III), he expressed a genuine belief, but it is unlikely that he thought he was expressing an empirically supportable proposition.

Although, on the surface, Ayer’s demarcation principle appears to eliminate metaphysics from empirical science, if we examine it carefully, we find that it says nothing at all. To simplify the statement, we could restate the phrase “fails to satisfy this principle [that some possible sense-experience should be relevant to the determination of its truth or falsehood]” as “is not empirically supportable”. He does not define metaphysics, other than to say that metaphysics comprises all propositions that are not empirically supportable, and therefore we could replace “metaphysical” with “not empirically supportable”. Making that substitution, his principle reads, “If a putative proposition is not empirically supportable, and is not a tautology, then I hold that it is not empirically supportable, and that, being not empirically supportable, it is neither true nor false but literally senseless”. Ayer attributes senselessness to metaphysical propositions, and, in common language, “senseless” has a pejorative connotation. However, in the context of the principle of demarcation, “senseless” can only mean that a proposition cannot be verified by sense-perception, or, to restate it, “senseless” can only mean that a proposition is not empirically supportable. If we make that additional substitution, then the demarcation principle may be restated as: “If a putative proposition is not empirically supportable, and is not a tautology, then I hold that it is not empirically supportable, and that, being not empirically supportable, it is neither true nor false but literally is not empirically supportable”. The demarcation principle does not eliminate metaphysics, in fact, it says nothing at all. It is literally senseless or a tautology.

Both Popper’s and Ayer’s principles of demarcation allow that empirical propositions are within the scope of empirical science and that non-empirical propositions are not. Ayer labels non-empirical propositions as metaphysical, but that demarcation, by itself, proves nothing about the value or lack of value in

metaphysical statements. The demarcation merely shows that metaphysical propositions are not empirically corroborated.

Ayer, A. J. Language, Truth and Logic. London: London, Gollancz, 1936.

Popper, Karl Raimund. The Logic of Scientific Discovery (2nd ed.). Routledge, 2002.