Language and What Constitutes the 'Verification Principle'

The verification principle was the central thesis of the logical positivists (or logical empiricists), initially those comprising the Vienna Circle (including Moritz Schlick, Otto Neurath and Rudolf Carnap), originally named the Ernst Mach Society, after that positivist. Their form of empiricism was a development on the fundamental empiricist notion that all our knowledge comes from sensory experience.

Influenced by the work of Moore, Russell and Wittgenstein, their focus was on language and they held that for any non-tautological statement to have factual content, it must be capable of empirical verification. Disconfirmation relegated that statement to nonsense.

On this basis, the logical positivists concluded that, due to ‘misuse’ of language, most previous philosophy had no factual content and was mere ‘metaphysics’. Statements that didn’t meet their criteria, though for their intent meaningless, may still have meaning in the ordinary sense (e.g. statements of religious belief, ethical and aesthetic statements). Motivated by scientism, the logical positivists held that the role of the philosopher was to be confined to ‘a more modest’ analysis of ordinary language (some including Carnap and later Quine disagreed with this, arguing that philosophers should try to improve on ordinary language to try to frame a proper scientific theory) and to develop the logic of science. 

A.J.Ayer went to Vienna in 1932, spent four months with the logical positivists, returned to England and in 1936 published his Language, Truth and Logic. This book brought logical positivism to a wide audience and, aided by the force and simplicity of its style, was very influential - particularly in England and the United States. His argument was built around the verification principle.1

Ayer, in ‘expounding the method of philosophy’,2 argued that the only source of knowledge of the world is experience, that all propositions that have factual content are empirical hypotheses the function of which is to provide a rule for the anticipation of experience3 and which could be verified in ‘strong’ and ‘weak’ forms.4 If a proposition failed to satisfy this principle and was not a tautology (true by the meanings of its terms, a merely verbal truth with no factual content), it was ‘metaphysical’ and therefore neither true nor false but senseless (e.g. ‘This tomato is red’ from the position of the ‘logically perfect’ language comprises a particular - the tomato, and a universal - ‘redness’. From this perspective, the tomato has no redness - the universal ‘redness’ has been added to it.) Ayer argued that metaphysics shouldn’t be read into the features of language. Since the theories of metaphysics are incapable of confirmation/disconfirmation by experience and are not analytic, according to the verification principle, they are nonsense.

The verification principle could also settle disputes between competing claims that appeared to be different, by showing that the disputes were merely verbal (e.g. the dispute between belief in and rejection of ‘God’. Because no observation could be relevant to either argument, the dispute was merely verbal.) Two statements would not differ in factual content unless experience confirmed/disconfirmed one but not the other. Reason alone cannot give us anything with content.

The role of the philosopher was not to derive consequences from ‘first principles’, it was to analyse the criteria of empirical propositions, but not to affirm or deny their validity - that was for the scientist. It was to do a transliteration of the statements of ordinary language into an ‘ideal’ language of experience called reductive or translational analysis or synonymy. Not all verificationists thought that precise translation was possible (e.g. ‘England declared war’ is about the people of that country.)

Ayer’s analysis was both phenomenalist (statements about the speaker’s experiences are equivalent to statements about the external world) and behaviourist (statements about the ‘minds’ of others are equivalent to statements about their publicly observable behaviour - e.g. ‘He is in pain’ is synonymous with statements about that person’s observable behaviour.) Some verificationists argued that ‘minds’ cannot be analysed in terms of behaviour.

For Ayer, the philosophers’ propositions were linguistic, not factual, and their concern must be with the formal consequences of definitions and ordinary linguistic usages. Ayer, consistent with logical positivism, wanted to bind philosophy to science, which together would address different elements in a unified project.5

In the Appendix of the second edition, written ten years after the first publication of the book, Ayer very significantly retreated from a number of his positions.6 He wrote that his original formulation in chapter one for what is verifiable was not satisfactory7 and discussed this, concluding that if the principle of verification is to be considered as a criterion of meaning, it must be interpreted as to admit statements that are not strongly verifiable. He asked ‘But how then is the word ‘verifiable’ to be understood?’8

He put forward a re-worked verification principle: ‘a statement is verifiable, and consequently meaningful, if some observation-statement can be deduced from it in conjunction with certain other premises, without being deducible from those other premises alone.’9 He then immediately criticised this, arguing that this criterion was far too liberal since it allowed meaning to any statement whatsoever.10

He wrote ‘Another difficulty which I overlooked in my original attempt to formulate the principle of verification is that most empirical propositions are in some degree vague. ... there is never any set of observation-statements of which it can truly be said that precisely they are entailed by any given statement about a material thing.’11

His position continued to weaken: ‘every significant statement about a material thing can be represented as entailing a disjunction of observation-statements, although the terms of this disjunction, being infinite, can not be numerated in detail.’12

His final re-formulation of the verification principle in his book was that for ‘a literally meaningful statement, which is not analytic ... it should be either directly or indirectly verifiable’13

Ayer referred to scientific theories that are expressed in terms that do not themselves designate anything observable. In doing this, he undermined the empirical basis of his verification principle. He continued, of ‘the older empiricist principle that no statement is literally meaningful unless it describes what could be experienced’ that for him, this principle imposes ‘too harsh a condition upon the form of scientific theories; for it would seem to imply that it was illegitimate to introduce any term that did not itself designate something observable. The principle of verification, on the other hand, is, as I have tried to show, more liberal in this respect, and in view of the use that is actually made of scientific theories which the other would rule out, I think that the more liberal criterion is to be preferred.’14

He wrote that he wished the principle of verification to be regarded ‘not as an empirical hypothesis, but as a definition’,15 that he defended the criterion of verifiability as a methodological principle and he acknowledged that even metaphysics needed to be engaged with: ‘for the effective elimination of metaphysics it needs to be supported by detailed analyses of particular metaphysical arguments.’16

In his review in 1949 of Ayer’s book,17 Alonzo Church wrote that Ayer’s accounts of a priori propositions, what is meant by ‘existence is not a predicate’, his remarks in relation to Russell’s analysis of ‘The author of Waverley was Scotch’ and his amended definition of verifiability after Isaiah Berlin’s criticism that the definition of verifiability given in the first edition would make all statements verifiable, would all have benefited from the use of formal logic, on the basis of which Church argued (by taking three propositions such that no one of them alone entails any of the others and showing that of any statement that either it or its negation is verifiable) that Ayer’s amended definition of verifiability was open to almost the same criticism as his original definition.

Church pointed out Ayer’s ambivalence on the use of formal logic, his having stated both that the philosopher is always concerned with an artificial language and that he is in favour of ordinary speech in connection with the analysis of descriptions.

Carl Hempel’s ‘Problems and Changes in the Empiricist Criterion of Meaning’18 argued that previous formulations of logical empiricism needed clarification and amplification and for the development of a language that would give meaning to ordinary language sentences that were translated into it. 

He proposed observation sentences that he believed made precise the logical positivist idea of a sentence asserting something that is observable. Whereas the positivists of the Vienna Circle had accepted that a sentence had empirical meaning if it was capable of verification by observational evidence, Hempel argued that a sentence has meaning ‘if and only if it is possible to indicate a finite set of observation sentences ... such that if these are true, then (the sentence) is necessarily true, too.’19 He wrote that verifiability is meant to be a criterion of cognitive significance rather than of truth.’20

Hempel wrote that the criterion of verifiability has several serious defects:

- it ruled out statements of general laws because they cannot be conclusively verified 

- ‘the requirement of complete verifiability is too inclusive’

- ‘the denials of certain empirically - and thus cognitively significant sentences - are empirically meaningless and ... cognitively meaningless’21

He argued that hypotheses require subsidiary hypotheses for the prediction of observable phenomena.22

In 1951 the behaviourist W. Quine published ‘Two Dogmas of Empiricism’ in which he argued that since we can’t engage in the analysis of meaning because ordinary theories of meaning were without content, philosophy can’t be concerned with conceptual analysis, that the individual statements that the logical positivists had focused on lacked verification conditions and that these could only be found in theories, which are a conjunction of statements (confirmation ‘holism’). Pragmatically, theories are good if they work. But there was much in common between Quine and the logical positivists - he was still a verificationist when it came to theories, and he argued that philosophers (through metaphysics) and scientists should work together towards a theory of reality.

It is argued that Quine’s holism brought about the downfall of the verification principle but this is not so. Not only was that principle fatally flawed from the beginning - both non-tautological and non-verifiable, it was ‘meaningless’ - the philosophical current to which Quine and the others contributed never had a materialist base.

The very same philosophical idealism, failure to understand and refusal to accept the primacy of objective reality over thought and the importance of practice to knowledge held by Bogdanov and Mach that Lenin exposed in his Materialism and Empirio-Criticism ran into and through the history of twentieth-century linguistic analysis. One of the founders of linguistic analysis and proponents of a ‘logically perfect, ideal language’, the mystic Wittgenstein, later correctly denied there is such a thing - a language divorced from the imperfect, infinitely complex and changing world, and recognised that language is a lived, profoundly social practice, that it cannot be studied removed from that - a lesson Church might have learnt regarding his review of Ayer’s book.23

Ayer’s phenomenalism was a divorce of knowledge from the objective world and the criterion of practice. Time and again through his book, he asserted that sensations and what he based on them (linguistic and logical constructions) comprise the objects of knowledge.24 He wrote that ‘all empirical hypotheses refer ultimately to our sense-contents’.25 They in fact refer to the objective world, to which our sense-contents are secondary and on which they are utterly dependent.

Quine’s holism was the inevitable result of the recognition that statements about the world function in the world and not in abstraction - that they have their place with others such and which, together, are crucial to our living in the world and knowing it. Yet Quine, from his pragmatic perspective, also could not accept the primacy of matter, of that which is independent of thought, and he did not understand the relationship between ‘matter’ and ‘thought’. He wrote: ‘in point of epistemological footing the physical objects and the gods differ only in degree and not in kind. Both sorts of entities enter our conception only as cultural posits. The myth of physical objects is epistemologically superior to most in that it has proved more efficacious than other myths as a device for working a manageable structure into the flux of experience.’26 Physical objects and the gods differ profoundly in kind - the former are matter, the latter the product of conscious matter - as are ‘posits’ and myths.

The analysis of language is that of a changing, imperfect practice in a changing, imperfect world. For the endeavours of those engaged in that analysis to be profitable they require a materialist basis - one that accepts and understands the primacy of matter over thought, the dependence of the latter on and inter-relationship with the former and the centrality of practice to our knowledge of the world.

--

Notes

1. It should be noted that in the first sentence of the preface, Ayer stated that the views in his book derive from the doctrines of Russell and Wittgenstein, whose apophaticism book-ended his Tractatus and underlay his position regarding ‘the limits of language.’ Despite repeatedly dismissing mystics (‘It is no use [the mystic’s] saying that he has apprehended facts but is unable to express them. For we know that if he really had acquired any information, he would be able to express it.’), Ayer neither mentioned nor analysed Wittgenstein’s mysticism in relation to his subject. A.J.Ayer, Language, Truth and Logic, Penguin, London, 2001, 125

2. Ibid., 49 

3. Ibid., 23

4. ‘A proposition is said to be verifiable, in the strong sense of the term, if, and only if, its truth could be conclusively established in experience. But it is verifiable, in the weak sense, if it is possible for experience to render it probable.’  Ibid., 18

5. ‘For our part we are concerned to emphasise not so much the unity of science as the unity of philosophy with science. ... (the function of the philosopher) is simply to elucidate the (scientific) theory by defining the symbols which occur in it.’ Ibid., 168; ‘But if science may be said to be blind without philosophy, it is true also that philosophy is virtually empty without science.’ Ibid., 169

6. ‘I have come to see that the questions with which it deals are not in all respects so simple as it makes them appear ... I think now that much of its argument would have been more persuasive if it had not been presented in so harsh a form.’ Ibid. 171

7. Ibid., 176

8. Ibid., 179

9. Ibid., 179 This relation of an observation-statement to other premises foreshadows Quine’s holism.

10. Ibid., 179 Ayer refers to Isaiah Berlin’s ‘Verifiability in principle’, Proceedings of the Aristotelian Society, Vol. XXXIX on this point.

11. Ibid., 180

12. Ibid., 180-1

13. Ibid., 181

14. Ibid., 182-3

15. Ibid., 185

16. Ibid.

17. Alonzo Church, Review of Language, Truth and Logic, Alfred Jules Ayer, The Journal of Symbolic Logic vol.  14 1949, 52-3

18. Carl G. Hempel, Problems and Changes in the Empiricist Criterion of Meaning, Revue Internationale de Philosophie vol IV, No. 11, January 15, 1950, 41-63

19. Ibid., 4

20. Ibid., 4-5

21. Ibid., 5

22. ‘the cognitive meaning of a statement in an empiricist language is reflected in the totality of its logical relationships to all other statements in that language and not to the observation sentences alone.’ Ibid., 15

23. Ayer wrote ‘a synthetic proposition ... cannot be either proved or disproved by formal logic.’ Ayer, op. cit., 128

24. ‘material things are constituted by sense-contents ... the sense in which a material thing is reducible to sense-contents is simply that it is a logical construction’ ibid., 153, ‘The question, “What is the nature of a material thing?” is, like any other question of that form, a linguistic question’ ibid., 54, ‘the table at which I am writing is a logical construction out of sense-contents ... to say anything about a table is always to say something about sense-contents.’ ibid., 53-4

25. Ibid., 167

26. W. Quine, ‘Two Dogmas of Empiricism’, in From a Logical Point of View, 1961 (first published 1951), 44