The Bertrand Russell Society Quarterly

more on russell and quine: a reply to stevens^*

Andrew Lugg

In ‘Russell as a Precursor of Quine’, I argued that from 1912, if not earlier, Russell was “a naturalistically-minded epistemologist in the Quinean mould”.¹ I drew attention to Quine’s view of Russell as a kindred spirit and expanded on a remark from Our Knowledge of the External World, which Quine quotes at the end of ‘Russell’s Ontological Development’, his most important discussion of Russell’s philosophy: “There is not any superfine brand of knowledge, obtainable by the philosopher, which can give us a standpoint from which to criticize the whole of daily life”.² My central point was that though differing from Quine in many ways, Russell hewed to a similar philosophical line and was no less concerned to develop a system of the world from within the framework of scientific theory. I was of the opinion then, as now, that an appreciation of the naturalistic cast of Russell’s thought is essential for understanding his philosophical views, and I wrote the paper in the belief that this is all too often overlooked.

Graham Stevens agrees that “the naturalistic elements of Russell’s philosophy are important” and “Russell’s naturalism is an important precursor of Quine’s”. But he believes I go astray since “Russell cannot be accurately characterised as an empiricist”, even a Quinean empiricist, and “Russell’s greatest influence on Quine’s naturalistic project did not stem from his epistemology but from his semantics”. In Stevens’s view, Russell’s change of heart regarding propositions in 1919 prompted him to adopt a naturalistic standpoint and he was not a naturalist, Quinean or otherwise, earlier in the decade.³ The trouble with this, as I see it, is that it assumes I place Russell in the empiricist camp as well as the naturalist camp and neglects that I reckon the empiricist elements of Russell’s philosophy to be secondary to his naturalism.

Like Stevens I think it wrong to read The Problems of Philosophy and Our Knowledge of the External World, as many commentators do, as empiricist works. No doubt in these works Russell takes there to be much in the mind that is not first in the senses – for one thing he thinks we apprehend relationships among universals. What I dispute is only the further suggestion that this excludes him from the ranks of the naturalist. There is a world of difference between holding that Russell’s thinking was empiricist in thrust or intent and holding, as I do, that it was naturalistic in inspiration and execution.

When considered without the surrounding text, my remark about Russell being “an empiricist in the Quinean mode” is doubtless misleading. But I think it fairly clear that I was emphasising that Russell construes the problem of the external world in much the same way as Quine, i.e., as a scientific problem about the relationship of scientific knowledge to its sensory basis. (Russell deemed the sensory basis of knowledge to be part of the physical world and took this to be revealed by scientific inquiry.) In the offending passage I was summarising how I read Russell. I was out to stress, as I put it in the preceding sentence, that “however much Russell differs from Quine about the nature of natural knowledge, he agrees with him in taking epistemology to be a branch of natural science and in regarding the problem of our knowledge of the external world as a scientific problem”.

Similarly in the only other remark I in which mention empiricism – “[Russell’s] empiricism is integral to his naturalism” – I was not implying that Russell was an empiricist pure and simple, still less equating his naturalism with empiricism. I was noting that, like Quine, he took the picture of knowers as surfaces across which energy travels to be a finding of empirical science. I did not, and would not, dispute that “Russell’s attitude to empiricism was variable and rarely resulted in unconditional subscription”. I would only add that Russell revised his views about what people know and how they know it in accordance with his understanding of the findings of natural science. My main contention was that in the 1910s, as well as later, he regarded his speculations as contributions to our theory of the world as a going concern.

Nor would I want to be thought of as believing the principle of acquaintance is “nothing more than an elaborate statement of empiricism”. I believe – and would attempt to show if pressed – that Russell took his view that we are directly acquainted with sense data, universals, and maybe ourselves to be, if not a clear-cut scientific result, a reasonable inference given what is known about human knowledge.⁴ Indeed, as I remarked in a footnote, I take Russell to have revised “his views about acquaintance along with his understanding of the deliverances of natural science” (p. 17). On my reading of the relevant texts Russell regarded acquaintance as a scientific notion comparable to absolute simultaneity and later discarded the idea because he came to think of it – as Einstein thought of absolute simultaneity – as scientifically problematic and superfluous.

In this connection I would take exception too to Stevens’s objection that I am wrong to discern an echo of Russell on meaning and acquaintance in Quine’s view of meaning as resting on sensory evidence. What I was after was the idea that there is a similarity between Russell’s conception of immediate knowledge and Quine’s conception of an observation sentence (as expressed in Word and Object). It was not my intention to deny the obvious – that Quine and Russell differ regarding meaning. Rather I was pouring cold water on the common assumption that acquaintance is antithetical to naturalism and pointing out that science is reasonably thought of as revealing the existence of two sorts of knowledge, direct and indirect.

Stevens also intimates (see his note 6 and discussion of Russell on descriptions) that I fail to notice that Russell made significant use of set theory in the constructions in Our Knowledge of the External World and related works. In particular Stevens seems to think that I believe that during the years in question Russell’s universe was, “if slightly more populated than that envisioned by the nominalist, then nonetheless comparable to the ‘desert landscapes’ relished by Quine”. This doubly misses the mark. I take both Russell and Quine to be robustly Platonist in their thinking, and I do not presume that nominalism, or something close to it, is a prerequisite for naturalism – after all Quine, a naturalist if ever there was one, posits abstract objects and appeals to resources of set theory in his own constructions.⁵

In any event, I wonder how Stevens is able to square his picture of Russell’s taking “the naturalist turn” in 1919 with the remarks I quote from The Problems of Philosophy, Theory of Knowledge, Our Knowledge of the External World, and Russell’s 1919 review of Dewey’s Essays in Experimental Logic, a paper in which Russell restates his earlier thinking about empirical knowledge in an especially uncompromising fashion. In these works Russell commits himself unequivocally to naturalism, and there is, I would argue, every reason to regard him as developing his new view of propositions in 1919 within the context of his naturalism. Even if “overly hasty”, the shift in his thinking is one that would have come naturally to a naturalistic philosopher.

As for Stevens’s insistence that “Russell took the naturalistic turn when he looked to psychology to provide a new home for propositional content”, I think I see what he is suggesting and why. Taking Russell’s post-1919 “psychologised theory of content [to be] much more in tune with Quinean intuitions [than his earlier thinking]”, he concludes, none too surprisingly, that Russell “was only really engaged in a project that can be usefully labelled as “naturalistic” after he abandoned the anti-psychologism that was central to his early philosophy”. For Stevens, Russell was antipathetic to psychologism from early on and he embraced a naturalistic (Quinean) line only after he had made a place for psychology in his philosophy.

This is an attractive story but I remain unpersuaded. As I argued, before 1919 Russell treated epistemology as “contained in natural science” (to borrow a phrase from Quine) and viewed the problem of the external world as “a question of physics” (to put it as he does in a passage in his review of Dewey’s Essays in Experimental Logic, quoted in my article). Moreover, and more importantly, Russell’s anti-psychologism prior to 1919 did not extend as far as Stevens suggests.

In Our Knowledge of the External World, a work composed in 1913/1914, for instance, Russell not only observes that “psychologists . . . have made us aware that what is actually given in sense is much less than most people would naturally suppose” (p. 75), he also stresses that the distinction between hard and soft data, crucial to his discussion, is “psychological and subjective” (p. 79) and speaks of his “hypothetical construction” as effecting a “reconciliation of psychology and physics” (p. 104). Furthermore he avers in Theory of Knowledge, as I noted in my paper, that “it is impossible to assign to the theory of knowledge a province distinct from that of logic and psychology” and he devotes the first part of his Dewey review to “Logical and Psychological Data”.⁶

Neither Russell nor Quine pay much attention to the distinctions and categories of contemporary philosophy, and it is important that they not be read as if they do. Stevens is right that “Russell’s epistemology . . . infiltrates his logic, his semantics, and even his metaphysics” – in Our Knowledge of the External World, for instance, he candidly acknowledges a “somewhat puzzling entanglement of logic and psychology” (p. 76). And Quine is equally cavalier regarding the divisions among subjects that good philosophers are supposed never to transgress – in ‘Epistemology Naturalized’, for instance, he declares that “epistemology ... becomes semantics” and “merges with psychology, as well as with linguistics” (pp. 89-90). This may or may not be as deplorable as Stevens implies. But it is, I think, pretty uncontroversial that Russell would applaud Quine’s addendum that the “rubbing out of boundaries could contribute to progress ... in philosophically interesting inquiries of a scientific nature” (p. 90).

Finally regarding Stevens’s claim that Russell’s “greatest influence on Quine’s naturalistic project ... stem[s] ... from his semantics”, I shall only say this does not seem to be how Quine himself saw things. As far as I am aware, Quine never spoke of being influenced by Russell’s “semantics”, never mind extolled Russell’s “psychologising of propositional content”. Rather he dwelt on the problem of our knowledge of the external world, the theme I focused on in my paper. Thus in ‘Russell’s Ontological Development’ he refers to Russell’s attempt to construct the world from sense data (using the resources of logic and set theory) as “a great idea” (p. 83) and in ‘Homage to Rudolf Carnap’ refers to the task of “deriving the world from experience by logical construction” that Russell “talked of” and Carnap “undertook ... in earnest” as “a grand project” (p. 40).⁷

NOTES

^* I am grateful to Graham Stevens for writing up his thoughts about my paper. He has helped me to get clearer – at least in my own mind – about the complex relationship between Russell and Quine. In addition I should like to thank to Paul Forster and Peter Hylton for helpful comments.

¹ Bertrand Russell Society Quarterly 128-129, 9-21.

² Cited in context in the first paragraph of my paper.

³ For more on Russell on propositions I’d recommend Stevens’s The Russellian Origins of Analytical Philosophy and ‘Russell’s Repsychologising of the Proposition’.

⁴ This is perhaps clearest in Russell’s discussion of the (epistemological) problem of “mixed psychology and logic” (‘Professor Dewey’s Essays in Experimental Logic’, p. 234). See also Theory of Knowledge, especially p. 46, and Our Knowledge of the External World, pp. 72-80. In his review of Dewey, Russell notes that he “agree[s] entirely” with Dewey when he says: “To make sure that a given fact is just and such a shade of red is, one may say, a final triumph of scientific method” but disagrees with him when he adds: “To turn around and treat it as something naturally or psychologically given is a monstrous superstition” (p. 235).

⁵ As an aside, I might mention that in The Russellian Origins of Analytic Philosophy, Stevens speaks of Russell’s pre-1919 philosophy as having “an empiricist flavour in the sense that [his] justification for admitting universals is that we have acquaintance with them” (p. 109).

⁶ Nor, incidentally, is it entirely obvious what Quine means when he says epistemology is “a chapter of psychology” (‘Epistemology Naturalized’, p. 83). While he sees the problem of the external world as a problem for the psychology of human animals, he also treats it as one of rationally reconstructing how we manage to get from the stimulations of our neuroreceptors to scientific discourse, something that can be “schematized by means of little more than logical analysis” (Pursuit of Truth, p. 2).

⁷ Significantly, ‘The Scope and Language of Science’, the paper in which Quine sets out his philosophical programme for the first time, and Word and Object, Quine’s greatest work, both begin with a discussion of Russell’s epistemological problem of the transition from sense to science. And in his last book, From Stimulus to Science, Quine again praises Russell’s attempt to realise “the dream of empiricist epistemologists: the explicit construction of the external world, or a reasonable facsimile, from sense impressions” (p. 10).

REFERENCES