The Roots of the Concept of Naturalised Epistemology: Russell and Quine

Introduction

A few words are needed to clarify the title and purposes of this article. The assumption that Russell’s philosophy was at the origin, directly or indirectly, of the Quinean concept of naturalised epistemology is at first paradoxical, or at least highly questionable according to historiography on the matter. For was it not, to a large extent, precisely against a foundationalist type of epistemology, such as Russell’s, that this concept was conceived? To say that such an epistemology was somehow on the path, philosophically speaking, to Quine’s concept thus seems to be a contradictio in adjecto, if not a provocation. The problem becomes extraordinarily more complex, as far as Russell is concerned, if we take into account that, according to some historiography, the philosophy of this author, after the Principia Mathematica, completely lost its relevance and originality with Wittgenstein’s critiques [1]. It lost those qualities, in particular, despite the publication of some very important works after those criticisms had been received, such as The Analysis of Mind and, a few years later, An Outline of Philosophy [2, 3]. I will suggest, however, that while Russell is not an author of the idea of naturalised epistemology, his own conception of epistemology in a series of works from the 1920s paved the way, positively speaking, for Quine’s, even if, ultimately, their respective and more or less distant results were incorporated into the latter conception. My theory is, from the outset, that we should not accept a substantial part of what Quine himself tells us, generically, about the origins of his conception [4], and, more generally, because I refuse to subscribe to what the known historiography, or what we might call the “official version of the origins of the concept of naturalised epistemology,” says about the matter [5, 6].

In this article, following my own research on the subject:

1. I will provide an important set of suggestions that point to the enormous advantages, from the point of view of studying the history of analytic philosophy, in all its complexity, of both of these attitudes.
2. Secondly, I seek to contextualise the works of Russell’s philosophy, to which I alluded, on the problem of naturalised epistemology, especially in relation to the theory of meaning and the philosophy of language in general. (It is a complex task, from the hermeneutic point of view, that I cannot develop here: the theory is that, in one way or another, we always go from the present, or from our current contexts of interpretation, to the past, not the other way around). And, also on this subject, I will argue, resuming my own investigations in the past, that we should not share the interpretations of the known historiography, which are closely related to the theory of the history of analytic philosophy that informs the official view of the origins of the concept of naturalised epistemology [7, 8, 9].
3. Finally, I will try to establish a parallel between Russell’s epistemology and Quine’s, which, far from opposing or contrasting them, as usually happens, highlights their affinities and complicities.

In part, this parallel was already partially made by Russell scholars as early as the last quarter of the 20th century [10, 11], but nevertheless remains generally ignored, as shown by the negative, in the 21st century, other investigations into this great English philosopher [12].

The Official Version of the Story

That there is a “standard” or official view of the origins of the concept of naturalized epistemology is something evident in the current analytical literature. I understand by “official view” a theory of the history of philosophy that constitutes the dominant grid of reading and interpretation, by the English-speaking university apparatus, of the place of Quine’s philosophy in the analytical movement, especially with regard to logical positivism, and particularly from the late 1960s [13]. Quine began to travel the route of “naturalised epistemology,” as we know, in the 1950s, with “Two Dogmas of Empiricism” [14] and other works [15] entering a decisive stage in the late 1960s with the publication of a series of collected essays in Ontological Relativity and Other Essays (the main essay, which gave the title to the work, dates from 1968) [16]. The interval between Quine’s relative obscurity in the philosophical world and his heyday after the 1970s can be explained by the definitive domination of the philosophical agenda, at that time, by British philosophy or by the so-called “ordinary language philosophy” (Ayer, Strawson, Ryle, Austin, and others). In such a context, Quine was a minor and irrelevant figure; but after the same philosophical current collapsed, his philosophy moved to the foreground of the international philosophical agenda. This is reflected, indirectly at least, in the place occupied by Quine in the collection of articles published by Rorty in The Linguistic Turn, and, in particular, in the thesis of the latter author, defended in the “Introduction” to that book, which, in line with Quine’s conceptions, involves the rejection of epistemology [17].

Now, when speaking – above – of a reading “grid” about Quine, I am talking about a set of fundamental historical and philosophical assumptions that more or less comfortably guide interpretations and are far from being evident by themselves, not because they are false in isolation, or considered per se, but because they are reductive and simplistic, pointing, in fact, to a much more complex and profound storyline woven into the history of the concept of naturalised epistemology. We could substantiate them in the following theses:

Theses:

1. This epistemology arises, in a revolutionary way, against the traditional epistemology of Carnap and of logical positivism in general, distinguishing itself from Carnap’s reductionism and verificationism, to which it opposes, for the first time in the history of analytical philosophy and in Western philosophy, a consequent holist and naturalistic view, owing to foundationalism, in the matter of epistemology and philosophy of language.
2. Such an epistemology argues against the assumptions of Western philosophy as a whole, from Descartes onwards, inaugurating an entirely new perspective for the development of philosophical reflection.

As one of the defenders of this thesis will say, in short: It has become customary for epistemologists who profess allegiance to a “naturalistic” conception of knowledge to pay homage to Quine as the chief contemporary provenance of their inspiration – especially to his influential paper “Epistemology Naturalized”. Quine’s main argument in this paper against traditional epistemology is based on the claim that the Cartesian foundationalist program has failed (...). While this claim about the hopelessness of the Cartesian “quest for certainty” is nothing new, using it to discredit the very conception of normative epistemology is new, something that any serious student of epistemology must contend with” (p. 36-

I will have the opportunity, I hope, to try to demystify these theses in some of their various doctrinal components, in line with some recent historiography both in terms of logical positivism, and Viennese in particular, and of Russell. The decisive point is not so much that Russell or Carnap

adopted a reductionist and verificationist programme in epistemology but, to reiterate, that Quine introduced for the first time in the history of philosophy, without any precedent or background in the matter, the problematic of naturalism and holism, and the matter of foundationalist epistemology “versus” naturalised epistemology. Such a vision establishes, therefore, a clear rupture between Quine’s naturalised epistemology problematic and traditional philosophical problematics, as if there were no relationship, continuity, or compromises between them. I would add that it is precisely this sign of rupture or discontinuity that characterises what I have called the “official vision,” and which, as far as Russell in particular is concerned, we should not subscribe to today. It is not surprising that, by precisely but misleadingly demarcating apparently clear historical-philosophical limits or boundaries, under the new problematic of naturalised epistemology, this vision had become a real programme of interpretation, easy and convenient, of the analytical historiography practiced in the English-speaking universities of the last quarter of the 20th century onwards. Not even the fact that Quine was a frequenter of the so-called “second Circle” of Vienna [4], that he never hid his enormous debt to Russell and Carnap, for example, in spite of all disagreements [19, 20], seems to suggest the suspicion that, after all, it may not be quite so, and that ultimately there is no innovation without tradition and crisis, no rupture without some continuity or compromises. But for now, and with a view to developing the theme of this article, let me comment on Quine’s own acceptance or not of the theses I mentioned above. Such an objective involves making some remarks on the Quinean view of the history of analytic philosophy.

Developments

It is known that Quine wrote very little about the history of philosophy and of analytic philosophy in particular when considered in themselves. The theories of Russell, Carnap, Tarski and others appear in his work in the context of his own philosophy, providing an excellent example of the model of historical and rational reconstruction with which Rorty, years later, would come to identify one of the fundamental genres of analytical historiography [21]. Be that as it may, the fact is that there seems to be little doubt that he, with some hesitations that I will explain later, was the first subscriber to the above-mentioned theses, suggesting here and there that Russell was not only ignorant of holism and naturalism regarding the theory of meaning but he was at the origin of the reductionism and verificationism of logical positivism and of Carnap in particular [14]. Not because Carnap, of course, according to Quine, had simply developed Russell’s theories; but because, historically and philosophically speaking and setting aside the invaluable contribution to the philosophy of logic and mathematics from men like Gödel or Tarski, Carnap’s sought to follow the same essential epistemological assumptions of a model of “first philosophy” as Russell’s, and in particular Russell’s after the first edition of The Principles of Mathematics [22]. Putnam, who has always been more prolix than Quine about the history of philosophy, carefully developed this perspective [23]. Having said that, it is no less true, as has already been pointed out, that Quine never failed to pay due homage to Russell [4, 20], both from a philosophical and a personal point of view, strategically distancing himself as much as was possible at the time from Russell’s critics and certain followers of the so-called “ordinary language philosophy,” as is the case of Strawson in particular [24].

Incidentally, the connection with the latter philosophy is quite relevant for characterising Quine’s view of the history of the analytic movement. All in all, Quine never had his own, autonomous view of the history of analytic philosophy and, most especially, of Russell’s place in it. In fact, as I suggested in my doctoral dissertation and in some more recent works, already alluded to here, he was, as indeed were the other members of what Dummett termed the “American school” [13], limited, broadly speaking, to adopting and developing the characteristic view of British philosophy during the 1950s and 1960s [8]. This will most likely explain certain ambiguities on Quine’s part about Russell, which I have just alluded to, and, more decisively, his silence and his omissions about Wittgenstein, who, a decade before the American philosopher launched the agenda of a whole new problematic of the theory of meaning centred on naturalised epistemology, had reached very similar conclusions regarding this theory [25, 26].1 British philosophy saw Russell and logical positivism in both its Viennese and American versions as its main historical enemies, because both would essentially be an expression of the “classical tradition” in philosophy, which, in contrast to a new one it would itself represent, made the theory of knowledge and epistemology in general its main philosophical concern [27, 28]. Wittgenstein, on the other hand, was accused of, at least initially, having collaborated (at the time of the publication of the Tractatus) with such a subversion of the philosophy of analysis, which had the familiar Russellian name of “philosophy of logical atomism” [29]. The anti-positivist critique of the late fifties by British philosophy and its followers, to which Quine seems to have adhered, interpreted the history of analysis generally from the point of view to which I have just alluded: it maintained that historical assumption of this philosophy according to which it was Russell’s philosophy, and particularly his doctrine of atomism logical, which is at the origin of positivists evils, and therefore generally subscribed historical responsibility to Wittgenstein’s first philosophy in relation to logical positivism, although Wittgenstein’s

1 Regarding the parallel between Wittgenstein and Quine on the matter, see, for example, Heal (1989).

known work after Tractatus, from the perspective of some authors of this same anti-positivist criticism, seems to have completely redeemed him from such a responsibility. Such an interpretation has remote echoes in authors such as Popper [30]. We have to wait until the last quarter of the 20th century for the anti-positivist critique to rehabilitate Wittgenstein’s first philosophy, divorcing it from the supposed Russellian influence, according to the theory by which such a philosophy will present a pure view of logic, uncontaminated by the spurious epistemological presuppositions of the logical atomism of Russell and his positivist followers [31].

The modern anti-positivist critique that follows the impact of the problematic of holism and naturalism on contemporary philosophy, and in which we must include, in addition to Quine, Feyerabend, Hanson, Toulmin, Popper and Putnam, generally adopted as to logical positivism and its origins the characteristic representation of British philosophy in the 1950s and 1960s, with the exception, as far as the first Wittgenstein is concerned, of the reservation to which I have just made reference. But as R. Haller, that is, the pioneer of the renewal of studies on Viennese logical positivism, showed in a series of memorable works, the image of a positivism reduced to classical, phenomenalist and verificationist foundationalism has no real historical and philosophical correspondence among the generality of the positivists [32, 33, 34]. On the contrary, Quine’s holism and naturalism had among their main antecedents or precedents Neurath’s philosophy since the first two decades of the 20th century [33]. This result of the Austrian historian’s investigations was, in part as a consequence and as is known, later extended to certain theories of science characteristic of French philosophers at the dawn of the 20th century, such as P. Duhem, giving rise to a whole new single historiography on the subject [35, 36]. However, on the other hand, studies on both the connection between Russell and Viennese logical positivism and on the question of what might be the most remote origins of that school in the German philosophy of the second half of the 19th century, have also undergone considerable development in the last thirty years [37]. The image of a positivism essentially inheriting the historical legacy of Russell’s philosophy, and of logical atomism in particular, has been frankly shaken by the investigations of A. Richardson, M. Friedman and others, which showed it to be the result of a vast historical-philosophical entanglement of German thought from the last half of the 19th century (still basically unstudied), centred on the justification of the possibility not only of mathematics and physics (as in Kant) but also of the new “life sciences” (energetics and biology) and sciences of the “spirit” (the so-called “human sciences”), and where neo-Kantianism is a dominant philosophical feature [9, 38, 39, 40]. More than that, and worse than that, for those investigations that we have mentioned: the contemporary anti-positivist critique, of which Quine’s philosophy is a prime example in a series of works since the 1950s [41, 42], when accusing Viennese logical positivism of reductionism and verificationism it not only apparently ignores the fact that it was, as we said, essentially holistic in terms of the philosophy of science (a holism that is not only naturalistic, as in Neurath, but also properly logical or logico-syntactic, as in Carnap, Hempel and others), and that such a holism clearly anticipates its own holism, as it continues to ignore that positivist epistemology, particularly Carnap’s, is not only not foundationalist in the classical sense in which it opposes relativism, but also, long before Kuhn, seems to anticipate the anti-foundationalist philosophical relativism of the new philosophy of science currently in vogue [43, 44].2 As a result, the historical portrayal of the genesis of logical positivism by contemporary anti-positivist criticism is not only generally erroneous and misleading, but is actually false.

Some researchers of the philosophy of anti-positivist critics, and in particular that of Quine, have been confirming this revision of the history of the concept of analysis of these critics. Quine, whose philosophical training was largely carried out with the Viennese logical positivists and with Carnap in particular, generally took it for granted that his own naturalistic holism in matters of philosophy of science would be opposed to positivist epistemology in general, ignoring the holism that characterises it and, above all, belittling the fact that his “naturalised epistemology” is basically a reformulation in new terms, as I suggested earlier, of Neurath’s naturalism, that is, nothing less than the great promoter of positivist holism from the beginning of the twenties until practically the institutional dissolution of the Circle. Quine, in fact, does not fail to stress the importance of Neurath’s philosophy, in contrast to logical positivism as a whole. But his observations on the matter were cryptic, episodic, and therefore in need of further development [4, 42]. However, for Quine to have done so, that is, to have omitted to say that he was yet, in a sense, in the 1950s and 1960s to take sides in the famous positivist dispute over the theory of meaning implied in the famous problem of the status of “protocol sentences” in favour of the one who, to some extent, had already won it as early as the mid- thirties, is not philosophically indifferent or secondary. D. Koppelberg notes that Quine, by attacking logical positivism as he did, was doing a favour for “British philosophy,” which had long since (in M. Dummett’s words) elected Carnap (not Heidegger!) as its “chief enemy” [45].3 This is, from the points of view of psychology and history, a good interpretation, but

2 The parallel between Carnap and Kuhn was initially made by Reish (1991). It was taken up, among others, by Earman (1992).

3 On the disagreements between English “ordinary language philosophy” and logical positivism, see Urmson (1962).

philosophically speaking it seems insufficient. An alternative explanation, which is suggested here, would be that Quine’s dominant interest does not consist in the critique of the supposed positivist reductionism in itself, nor merely in intervening in the disputes between “British philosophy” and logical positivism, although both this critique and this intervention seem to be indisputable facts, but, instead, in relaunching the more general debate between foundationalism and naturalism in part (and I underline “in part”) in the wake, as I will suggest below, of theories such as the one Russell presents to us in such pivotal works as An Inquiry into Meaning and Truth and Human Knowledge [46, 47]. However, it seems to be perfectly evident that the historical-philosophical assumptions Quine presents for this debate are, in general, historically misleading if not false. Any philosopher needs to recreate the history of philosophy in this or that way in accordance not only with the demands of his own philosophy but also with those of the historical context in which he finds himself. It is a fact that, as Rorty would say, well illustrates the meta-historical and meta-philosophical character of a large part of the problems current in analytic philosophy from its very beginning. What a philosopher, as an author, says or does not say about his own conceptions, after they are known, is a matter that, par excellence, falls within the scope of historiography or philosophical commentary. It may well happen, as has abundantly happened in the past and continues to happen today, that some commentator is absolutely right as opposed to the author.

Counter-Arguments in Relation to the Official Version

Russell, rightly, was victim throughout the history of analytic philosophy of such a type of meta-historical representation of what true philosophical analysis should be, starting, from the outset, with his relationship with Wittgenstein at the time of the Tractatus and continuing later with his influence on Viennese logical positivism [7, 8]. The erroneous and misleading assumption of analytical historiography on the subject is that his philosophy was essentially foundationalist, in the classical or traditional sense, in addition to being atomist and reductionist, and, therefore, alien to the problematic of holism and naturalism in philosophy. I have persistently tried to defend the theory, against what I have called the “Wittgensteinian reading,” that such an assumption has no historical sustainability in Russell’s philosophy from the 1920s onwards, as his books and other papers show [48, 49], and that what we actually find there, as in the case of some logical positivists at the time, such as Reichenbach and Schlick, is an anticipation of the problem, and of its philosophical implications, which Quine came to call, from the 1960s onwards, “indeterminacy of translation” (chap. 3) [16, 50, 51]4 Such a problem appears in Russell in the form of the semantic relativity between theory and the observation, and leads him to a version of the problematic of holism which we can call a “partial semantic holism.” In fact, at the beginning of the 1920s, in works such as The Analysis of Mind and a few years earlier, more decisively, by Wittgenstein himself, we find the defence of the fundamental conception according to which meaning results from the use of the language itself,5 in the more general framework of an approach, in Russell’s case, that, recognising the limits and difficulties of traditional foundationalism, seeks a middle ground between this and naturalism (chap. 1) [8]. All this, of course, is far from being evident for traditional readings of Russell’s philosophy, and particularly for either the Wittgensteinian reading or the reading undertaken in the light of the official view of the origins of the problematic of naturalised epistemology. Compare the following passage from The Analysis of Mind (1921), which clearly anticipates, in its own way, the Wittgenstein of the Philosophical Investigations and Quine himself from the 1960s onwards: Understanding language is more like understanding cricket: it is a matter of habits, acquired in oneself and rightly presumed in others. To say that a word has a meaning is not to say that those who use the word correctly have ever thought out what the meaning is: the use of the word comes first, and the meaning is to be distilled out of it by observation and analysis (p: 197-198) [2].

Or even, more decisively, as far as Russell’s naturalism and behaviourism is concerned, and regarding his theory of meaning as early as the early 1920s:

1. On suitable occasions you use the word properly.
2. When you hear it, you can act appropriately.
3. You associate the word with another word (say in a different language) which has the appropriate effect on behaviour.
4. When the word is being first learnt, you may associate

4 This is very clear in Reichenbach’s philosophy. In 1926 he states, on the subject of semantic theory/data-observation relativity: “There are no facts, proclaims the idealist, who views the whole conception of the world as a construction of reason. In a certain sense, this is true (...) even the simplest facts of daily life are to some extend theory-laden. (...) But how can facts decide between theories if they themselves presuppose theories? Instead of overthrowing the theory ‘refuted’ by experiment, may we not alter the theories that first made the ‘thing’ observed into ‘this’ particular fact? Could we not arbitrarily establish ‘any’ theory in just this way, by interpreting every fact accordingly? Are there any facts at all that claim to characterise something objective?” (Reichenbach & Cohen (Eds.) (1978: 289). On the holisms of Schlick and Reichenbach, and their relationship with Russell, see Demopoulos & Friedman (1989).

5 This thesis led Russell to present in 1921, long before Popper, an anti- verificationist argument in terms of the theory of meaning. See Russell (1978/1921: 268-269).

it with an object, which is what it “means”, or a representative of various objects that it “means” (p. 199- 200) [2].

This naturalism and behaviourism was especially developed by Russell in An Outline of Philosophy [3]. Therefore, if there is any problem regarding these matters in Russell, it cannot be that the philosopher was or might have been unaware of their relevance, which, moreover, he knew very well given his privileged knowledge, at the time, of the philosophy of authors such as the pragmatist J. Dewey, as An Inquiry into Meaning and Truth retrospectively illustrates. A fundamental aspect of my remarks above, about Russell’s attitude towards Viennese logical positivism from the 1930s onwards, is that – after his reception of Wittgenstein’s Tractatus – he feared the damaging implications of a radical holism in philosophy both in the logical or logico-syntactic version and in the sociological and naturalistic version [46, 52], although in 1940 he sometimes confuses, apparently, the two versions, because it was not clear to him (as it was never clear until very recent philosophers and historians, as is the case of Friedman [40, 53]) what, strictly speaking, would distinguish the semantic and naturalistic version of Neurath from the properly logico-syntactic version of Carnap and Hempel. In both cases it is holism, but the meaning of it is very different if we vary from case to case or from one case to another.

For Russell, in works such as An Inquiry into Meaning and Truth (1940), it was precisely a radical holism, in the two versions I have just mentioned, that the positivists’ proposed solutions to the crisis of justification of the so- called “protocol sentences,” such as those of Neurath, Carnap and Schlick, were concerned with [54, 55, 56, 57, 58, 59].6 Russell thought that these proposals led to a logical-linguistic reduction of philosophical problems, as was the case with Carnap, or a sociological and naturalistic reduction, as was the case with Neurath, and that both could lead to the end of epistemology and philosophy itself in general. This problem – the impact of holism on philosophy – had always been a fundamental matter for the English philosopher since his reception of Wittgenstein’s Tractatus. As argued in An Inquiry, holism would be the most pernicious consequence of the logical positivists’ denial of the existence of a pure and virgin datum of interpretation, or of something independent of language itself, such as so-called “facts” or “perceptions” (p. 117-118) [46]. Russell was convinced that such a kind of denial had led the positivists to Hegelianism and, in particular, to a kind of Neo-Hegelian panlogism, and identified similar radical holist attitudes either on the part of Dewey’s pragmatism or on the

6 On this crisis, see Carnap (1987), Neurath (1959/1932-1933); Hempel (1935); Barone (1986); Jacob (1986); and Cicera (1994).

part of the ordinary language English philosophers (p. 272) [46]. As he stated, raising to the foreground of reflection the problem of holism in philosophy: There are some schools of philosophy – notably the Hegelians and the instrumentalists – which deny the distinction between data and inferences altogether. They maintain that in all our knowledge there is an inferential element, that knowledge is an organic whole, and that the test of truth is coherence rather than conformity with “fact.” I do not deny an element of truth in this view, but I think that, if taken as the whole truth, it renders the part played by perception in knowledge inexplicable (p. 117) [46].

It is evident in several passages of An Inquiry that, according to its author, it was essential to maintain the distinction between language and facts, or, as Quine will say (in “Two Dogmas of Empiricism”), from an opposite perspective, between “the empirical elements” and the “logical elements” of the theory of meaning, in order to avoid a logical-linguistic reduction of philosophy. Not that Russell had, at the time, the perception that was already required of a problematic such as analyticity, which would only become part of the philosophical agenda, with Quine, a decade later. The core of his argument has as a background, instead, that same set of problems raised by the justification of the protocol sentences that had occupied the Viennese positivists. And it is no less clear that Russell’s critical arguments in these texts are not based on foundationalist premises, as his critics and opponents claim, but precisely on an attempt to find an alternative path between traditional foundationalism and naturalism. Russell’s idea, in terms of language, for example, is that language learning begins from birth, that is, in the early stages of childhood, and through what he calls “noticing,” that is, a process that basically consists of the ostensive designation of the objects or contexts to which the names refer in general, and which appears now, in his philosophy, instead of what before (until 1919 and the impact of the Tractatus) he had called “acquaintance.” But, applied to human experience as a whole, this process is interpreted by him in naturalistic and behaviourist terms, namely through the notion that meaning comes from the use of language (p: 59) [46], as mentioned above. Faced with the question of how one can reconcile the thesis that reference and meaning are ultimately based on “noticing,” since childhood, with the thesis that, later and independently of contexts, they result from the use of language itself in general, Russell would certainly argue that his linguistic “noticing” is a primitive and more or less ideal stage, ontogenetically and anthropologically speaking, of the possibility of reference and meaning in general (p. 70) [46]. From this point of view, he himself could subscribe to a large part of the theory developed by Quine in his early works on the philosophy of language [60], as An Outline of Philosophy

[3] shows.

The essential point, against Quine’s interpretation of logical positivism but in line with the contemporary rehabilitation of the latter that I alluded to above, is that for Russell the positivist theory of science was, without knowing it, essentially holistic, not reductionist. And, in addition, that reductionism in philosophy subscribed to by some positivists, such as Carnap and above all Schlick, is a positivist escape from the implications or pernicious consequences of this holism in philosophy, particularly those that seemed to follow from a theory like Neurath’s [53]. In other words, for Russell, the Viennese positivists would not have had the philosophical awareness that was required of their holistic assumptions in the philosophy of science, and, therefore, they would have sometimes, inconsequently, adopted reductionism. From this point of view, the idea that positivist verificationism, far from eliminating metaphysics, reintroduces it in another way, ultimately being based on it, is recurrent throughout An Inquiry: “A great deal of metaphysics”, states Russell, “is involved (...). I cannot imagine any way of discovering whether the metaphysics in question is true or false, but I think it is worthwhile to state the assumptions involved.” (p. 220) [46]. More decisively for Russell, still in 1940 and anticipating Quine (1953c), there were not only “dogmas” in empiricism, as this philosopher would eventually claim, but empiricism itself, as a whole, constitutes a dogma: (...) I will observe, however, that empiricism, as a theory of knowledge, is self-refuting. For, however it may be formulated, it must involve some general proposition about the dependence of knowledge upon experience; and any such proposition, if true, must have as a consequence that itself cannot be known. While, therefore, empiricism may be true, it cannot, if true, be known to be so (p. 221) [46].

It follows clearly from what Russell tells us in An Inquiry about Neurath that what Quine will come to call “naturalised epistemology” [61] is basically equivalent to the negation of philosophy in the first and essential sense in which it (philosophy), once the limits and unavoidable difficulties of traditional foundationalism are recognised, should somehow continue to be able to found scientific knowledge under penalty of disappearing completely. It is a theory that he insisted on later in Human Knowledge and, finally, in My Philosophical Development (p. 229-230) [47, 62]. In this sense Russell pertinently notes in the first work that the ultimate justification for Neurath’s radical semantic empiricism is not philosophical but sociological. Another thing, presumably, he would not say about the role Quine entrusted to philosophers in matters of science within the framework of naturalised epistemology (p: 28-30) [61]. It was essential for Russell to conserve and maintain the autonomy and independence of the theoretical domain of philosophy in relation to science, accepting, if you will, the Cartesian conception of the tree of knowledge but at the same time reinterpreting in such a context, in the sense of naturalism, everything that was possible to reinterpret. And that was precisely what he tried to do, in his own way, particularly from the 1930s onwards (when the impact of logical positivism on the international scene took place), with a series of important articles, generally ignored even today, that constitute, as early as the first half of the 20th century, the first systematic critique of logical positivism [63, 64, 65]. At the basis of such an undertaking, some years before Quine and the contemporary anti-positivist critique, is already that same problematic of holism in philosophy that guided both this critique and that American philosopher.

On the other hand, it also follows from Russell’s anti- positivist critique that, while it is true that we must accept Quine’s indeterminacy of translation and a large part of its implications (chap. 3) [16], we should in no way feel obliged to accept his inscrutability of reference, if it is precisely that conception that is at the base, philosophically speaking, of the concept of naturalised epistemology. (The indeterminacy of translation, strictly speaking, should not necessarily lead to the inscrutability of reference but, in the Quinean argument, it is indissociable or inseparable from it.) In fact, since his first works on the concept of vagueness in the early 1920s, he maintained that the inevitable indeterminacy of meaning that results from the use of language has to do not only or exclusively with the indeterminate occurrence, but also with our cognitive relationship to it. Russell stated: When knowledge is vague, this does not apply to the knowing as an occurrence; as an occurrence it is incapable of being either vague or precise, just as all other occurrences are. Vagueness in a cognitive occurrence is a characteristic of its relation to that which is known, not a characteristic of the occurrence in itself (p.147-148) [66].

Which means, in Quinean terms, that if it were the occurrence or reference itself that were indeterminate, we would not even be able to understand the possibility of the indeterminacy of translation, and the latter would therefore lose all philosophical pertinence. In this indeterminacy that Russell has in mind there is no trace of the idea of generality, in the sense that some critics misleadingly accuse his conception of language. The indeterminacy of translation (the general vagueness of cognitive occurrences) in itself is a fact that, in turn, cannot be indeterminate, otherwise the concept of indeterminacy itself becomes completely empty and self-contradictory. The referent (or the “known” as Russell calls it), on the other hand, is also not indeterminate in itself and is only properly indeterminate in the context of translation (or of the relation between the “cognitive

occurrence” and the “known”). Aware of the Kantian origins of semantic holism in general and in contrast to what was generally happening in his time, Russell persistently insisted on this point, which I have just alluded to. He would therefore accept the Quinean indeterminacy of translation but reject that it could necessarily lead, as Quine thought, to any inscrutability of reference. Thus, it was once again the need for an alternative path between classical or traditional foundationalism, which does not accept the legitimacy of the first indeterminacy, and a naturalism more or less radical, which, in one way or another, essentially militates for the inscrutability of reference, on which Russell focused on his last philosophical works.

Conclusions

It has been suggested throughout this article that Russell’s philosophy since the 1920s is at the origins of the Quinean concept of “naturalised epistemology,” and it is certain that he would not subscribe to such a concept without many reservations. With the idea that there were inappropriate associations either between Russell and logical positivism or between that philosopher and philosophical modernity from Descartes onwards which having been demythologised, and the naturalistic and behaviourist orientation of his philosophy after the aforementioned period exemplified, the conclusion reached is that Russell was perfectly aware not only of the importance but, above all, of the dangers of the consequences of such an orientation, because he himself, around the 1940s, became aware of them within the framework of his own philosophy. In fact, the first major typology on doctrines regarding the concept of truth (namely: correspondence theory vs. coherence theory) was originally made by Russell himself, in a pioneering way, in An Inquiry into Meaning and Truth (19731940). The decisive problem, in the relationship between Russell and Quine, is/ was the following: whether or not to accept that holism and/ or a semantic and radical version of it leads to the end of philosophy as systematic investigation, as seems to happen with that epistemology. The English philosopher would not subscribe to such a consequence, which led him to not rehabilitate the well-known Cartesian and foundationalist perspectives in philosophy, contrary to what his critics usually argue, even today, but to adopt, in new ways, some of the respective premises, as is the case of those that accept, on new terms, the role of psychology and epistemology. He would argue against Quine for the same reasons he had already argued against Wittgenstein in the “Introduction” to the Tractatus (chap. 1) [52, 67, 68]. In any case, to claim that Russell was not part of the path to the Quinean concept of naturalised epistemology and/or that he ignored such a path, as is often invoked, is completely unacceptable. In contrast, what this article has suggested is that, in some way, the philosopher’s views in books such as An Inquiry into Meaning and Truth have more or less decisively influenced those of the author of “Two Dogmas of Empiricism,” “Ontological Relativity” and “Naturalized Epistemology” [69].