Introduction
Little to nothing has been written on the correspondence between Victoria Welby and Bertrand Russell despite recent attempts to blur the line between Russell's analytic philosophy and pragmatism.Footnote 1 I argue that Welby's “Significs”Footnote 2 — defined as “the science of meaning or the study of significance”Footnote 3 — embodies a pragmatist attitude insofar as it challenges the strict division between, on the one hand, a purely logical account of meaning devoid of contextual and psychological import, and, on the other hand, a pragmatic account of meaning that seeks to understand the way that signs are used to convey meaning in particular contexts. I aim to show that, in a segment of their 1905 correspondence, Welby brings this pragmatist attitude to bear on Russell's theory of descriptions.
I begin by outlining two suggestions that Welby offers for Russell's theory and Russell's replies. I then consider that Welby's suggestions overstep the scope of Russell's “On Denoting,” which is narrowly focused on matters of logic. Finally, by elaborating some key elements of Welby's own work, I expose the upshots of her suggestions and show how the later Russell begins to address Welby's concerns.
The First Suggestion
Welby's first suggestion draws attention to the distinction that Russell makes between “acquaintance” and “knowledge about.”Footnote 4 For Russell, “the distinction between acquaintance and knowledge about is the distinction between the things we have presentations of, and the things we can only reach by means of denoting phrases.”Footnote 5 For example, Russell claims that we can know that a phrase denotes unambiguously and still have no acquaintance with what it denotes:
[W]e know that the centre of mass of the solar system at a definite instant is some definite point, and we can affirm a number of propositions about it; but we have no immediate acquaintance with this point, which is only known to us by description.Footnote 6
While I have never been in direct contact with the centre mass of the solar system, I can nonetheless still know something about it through my acquaintance with a number of propositions pertaining to it. That is to say, I can be acquainted with the meaning of certain words like “mass” and “solar system,” without having been in immediate contact with the objects those words denote. Thus, Russell claims that “all thinking has to start from acquaintance; but it succeeds in thinking about many things with which we have no acquaintance.”Footnote 7 This final point, the idea that all thinking begins from acquaintance, is precisely the point with which Welby takes issue.
Welby's first suggestion asks whether something can be gained by extending Russell's distinction to make it threefold. She writes:
Then we should have A. awareness which belongs to sensation and is shared by the simplest forms of life, B. acquaintance as direct experience of, and C. knowledge as you define it.
We might be aware of other minds without being acquainted with them and without knowing them ‘through denoting.’ I venture to suggest this because the idea of awareness (earlier or lower than consciousness?) is associated in my mind with ‘Mother-Sense’ — as it were the starting-sense.Footnote 8
Here, Welby insists that Russell's distinction become threefold: that thinking begins with Mother-Sense, and not, as Russell suggests, with acquaintance. Welby makes this suggestion because Mother-Sense plays a crucial role in her own theory of meaning — it is that first that makes possible all else that follows from it. As Susan Petrilli describes, Mother-Sense (or what Welby also referred to as “primal sense”) is the “necessary condition for the human signifying capacity.”Footnote 9 Welby calls Mother-Sense a “primordial method of mind, one which is necessarily the precursor and condition of all forms mental activity, including even that of logic itself.”Footnote 10 We are, first and foremost, signifying animals; and we need to understand ourselves as such prior to developing our logic. Welby's first suggestion boils down to a worry that Russell, in claiming that “all thinking starts from acquaintance,”Footnote 11 has overlooked the very condition that makes both acquaintance and denoting possible — a Mother-Sense, or a capacity to signify in all of us.
A follower of Russell will be baffled by this suggestion, as Russell is after neither the conditions that make meaning possible, nor explaining the psychological makeup of the signifying animal. Rather, Russell confines his scope to a narrow analysis of denoting phrases in logic. Indeed, in his 25 November reply to Welby, he claims that the threefold distinction Welby is positing “is not relevant to my [Russell's] problem.”Footnote 12 Russell does not care to extend his analysis to the domain of psychology (via “Mother-Sense”) — his objective in 1905 was not to “give a complete account of (more or less) cognitive states of mind.”Footnote 13 In other words, Russell's “On Denoting” was never intended to provide a full-blown account of meaning that brings psychology and logic together, but rather, only to provide an account of how denoting phrases function in logic. However, Welby is interested in the science of meaning more generally. She sees logic as only one relevant branch of her more wide-sweeping Significs, which, in addition to logic, wants to make sense of both the speaker's and hearer's psychology as well as the pragmatic contexts in which communication and signification occur. By suggesting her threefold distinction, Welby has overstepped the scope of Russell's project and appears to have misunderstood the aim of his theory of descriptions. While Welby's and Russell's projects differ with respect to their aims, Welby should be understood as trying to convince Russell to broaden the scope of his project — to consider how his theory of denoting might contribute to a more general science of meaning.
The Second Suggestion
To grasp Welby's second suggestion, one must consider the central principle of Russell's “On Denoting”: “that denoting phrases never have any meaning in themselves, but that every proposition in whose verbal expression they occur has a meaning.”Footnote 14 Denoting phrases like ‘a man,’ “[e]verything, nothing, and something”Footnote 15 do not convey meaning independently, but only when put into a proposition of the form ‘C (x).’ For example, if taken in isolation, the denoting phrase ‘a man’ is “wholly destitute of meaning.”Footnote 16 However, Russell shows that the proposition ‘I met a man’ has the meaning: “‘I met x, and x is human’ is not always false.”Footnote 17 Hence, it is only after the denoting phrase ‘a man’ is put into the proposition ‘I met a man’ that the denoting phrase has meaning. Taken in isolation, denoting phrases convey nothing. Russell goes on to show how the same is the case for denoting phrases such as ‘all men’ and ‘the,’ however, it is beyond the scope of this paper to trace his arguments any further. The purpose of this brief exegesis is only to provide the context for Welby's second suggestion.
Welby's second suggestion shows how denoting phrases can express meaning when used by themselves. She writes:
‘What did you give to Smith?’ ‘Nothing.’ ‘And what to Jones?’ ‘Everything.’ The answerer here emphatically intends (means) to convey a fact. This would apply in the same way to ‘a man.’ ‘What did you see?’ ‘A man.’Footnote 18
According to Welby, denoting phrases such as ‘everything,’ ‘nothing,’ and ‘a man’ can have meaning, albeit, only when the one using the phrase intends for them to. Such is the case when the answerer replies to the questioner above.
However, Russell writes that in the case Welby describes the use of those words (‘everything,’ ‘nothing,’ and ‘a man’) no longer function as denoting phrases but as “mere abbreviations for propositions.”Footnote 19 When Russell claims that a denoting phrase has no meaning in isolation, he is considering the denoting phrase at a formal level, prior to being placed into a proposition. It is not clear that when the answerer replies “Everything” the phrase is still in isolation. To utter the phrase by itself seems already to place the phrase in a context. “Everything” only gains meaning when uttered as a response to the question “What did you give Smith?” In this context, “Everything” is no longer a denoting phrase, but itself, a proposition. It seems that Welby has misunderstood what Russell means when he says that denoting phrases are devoid of meaning when taken in isolation.
The Importance of Context
Welby, however, responds to Russell's concern in her 29 November letter:
… while words like ‘nothing’ are now as you say abbreviations from propositions, the case was originally and now is still in some minds, reversed. Once a word was the only sentence (as before a sound the only word!) now the sentence — or proposition — is virtually the word. That is why the context becomes, in judging the value of a word, so important.Footnote 20
Here, Welby is suggesting that context is crucial to determining the value of a word. The value of a word changes depending on the context in which it is put to use. When I say ‘Jack is a dog,’ and am referring to my furry quadruped, the phrase's significance is quite different from when I refer to my friend Jack, who is adventurous in the bedroom. In the first instance, ‘a dog’ functions as a denoting phrase — it picks out the properties ‘furry’ and ‘quadruped.’ However, in the second instance, ‘a dog’ functions as a connoting phrase — that is, the phrase extends its denoting capacity to suggest a metaphorical significance. One needs to appeal to context in order to tell the difference between denoting and connoting phrases. It may well be that denoting phrases do not have significance in isolation, but they certainly do when expressed through propositions, and the significance of propositions, themselves, depends on context to some extent.
Not only does the context effect the significance of propositions but so too does the meaning qua intention. There is a difference in intention that alters the meaning of my proposition, depending on whether I mean my dog or my friend Jack. A little bit of psychology seeps into the logic here, as when we consider meaning qua intention, we, by extension, require an explanation of the speaking subject. Part of this project involves bridging the gaps between both psychology and logic, and logic and rhetoric. Neither of these are easy tasks. Welby is, as she humbly claims, “keenly sensible of [her] own failure to present the case for ‘significs’ as an urgent need” but wishes to discuss its importance with Russell further.Footnote 21
Unfortunately, Russell responds to Welby only once more on this topic. In it, he seems to sever any fruitful connection between Welby's “Significs” and his own theory of denoting:
I am less concerned with what people do mean than with what things there are that might be meant. … Thus when a single word was the only sentence, I should doubt whether, so far as anything definite was meant, what was meant differed from what we should express by some sentence of many words.Footnote 22
Crucially, Russell's claim that he is not interested in “what people do mean” but in “what things there are that might be meant” seems to divide, what we now call, an ‘ideal language philosophy’ from a ‘natural language philosophy’ and, subsequently, a study of logical form from a study of language pragmatics. Whereas natural language philosophy is interested in the way that language is ordinarily used, ideal language philosophy is interested in developing a logically perfect language that reveals the shortcomings of natural language in its day-to-day use. As Graham Stevens notes, Russell can be understood as an ideal language philosopher because his “theory takes the surface grammar of English to conceal the true structure of the propositions these sentences express.”Footnote 23 Consider the following natural language sentence: ‘the current president of the United States is a dipshit.’ Rather than treat this sentence as it appears in natural language, Russell's theory will appeal to his ideal language and formalise it thus: ‘There is one and only one x such that x is currently the president of the United States and x is a dipshit.’ On Russell's account, the true meaning of the phrase is not picked out by the natural language, but by the formalisation of the utterance in terms of an ideal language. As Stevens argues, the purpose of Russell's project is not to replace natural language with a logically superior one, but rather, to cleanse natural language by distilling it into pristine logical form: “a logically perfect language will remove the disguise and unveil the true structure and semantic function of natural language expressions.”Footnote 24 Russell's theory of descriptions turns out to be about natural language expressions on Stevens’ account, not a strictly logical project.Footnote 25
However, on Stevens’ account, Russell is still not interested in language pragmatics, but rather, with semantics.Footnote 26 Welby's “Significs,” however, extends a semantic analysis of natural language to consider the pragmatics of language. Foreshadowing the later Wittgenstein, Welby writes in What Is Meaning? that “[t]here is, strictly speaking, no such thing as the Sense of a word, but only the sense in which it is used — the circumstances, state of mind, reference, ‘universe of discourse’ belonging to it.”Footnote 27 In other words, the meaning of a word depends on the way that it is put to work, or used, in a particular context. Welby reinforces this point when, in an earlier piece, she claims that meaning is always “changing on our very tongues.”Footnote 28 For Welby, language is not static but plastic — it is always on the move. Words adapt to their context as organisms adapt to their environment.Footnote 29 Since language is always changing, we cannot rely on a purely semantic analysis — we cannot expect an ideal language to account for each of these changes. Instead, we must adopt an approach that adapts to the movement of language — changes as language changes. We must preserve the “plasticity of language.”Footnote 30 Pragmatics is more effective than semantics in this respect.
Unlike Russell, Welby's project is anchored by a use-based approach akin to that of language pragmatics. With this in mind, we can begin to understand the true motive of Welby's suggestions and her subsequent insistence that context is crucial. Indeed, Welby can be understood, in this correspondence, to be bringing a kind of language pragmatics to bear on Russell's strictly semantic analysis. Welby does not misunderstand the scope of Russell's analysis, but rather, is calling on Russell to extend the scope of his analysis. For, as Welby teaches, if we want a full-blown account of meaning,Footnote 31 we must look to the different ways that language is used in specific contexts — “a systematic inquiry on the subject of meaning and its changes.”Footnote 32 Welby does not believe that there is a static ideal language floating beneath our expressions of natural language to which we can appeal as a foundation for meaning. The objective is not to uncover an ideal language chalked out in advance, underpinning our use of signs, but to work from within particular universes of discourse where meaning is made, understand how signs function in those contexts, and build up our account of meaning out of the use of signs in those contexts. Use comes first, and the ideal language is built up out of the various uses that function to create it — the ideal language is contingent on use and generated out of it.
Later Russell as Earlier Later Wittgenstein
Russell, while reluctant to take up such a position in 1905, seems to gesture toward one resembling Welby's in his later writings. In fact, Cheryl Misak notes in Cambridge Pragmatism that Russell later cites Welby as “one who sparked his interest in the topic [of signs and meaning], although at the time [she] had pressed him on the point, he resisted.”Footnote 33 Despite Russell's dismissal of Welby's suggestions in 1905, in The Analysis of Mind, he writes:
Understanding words does not consist in knowing their dictionary definitions, or in being able to specify the objects to which they are appropriate .… 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.Footnote 34
We learn more about the meaning of words by observing how they are used than we do by appealing to definitions of them. If the use comes first, then the definitions are slowly built up out of the use. We start with use, with pragmatics, and build our account of meaning out of it. Both Welby and the later Russell converge on the point that meaning is use. The later Russell, ironically, ends up beginning with “what people do mean” rather than, as the earlier Russell did, with “what things there are that might be meant.”
Conclusion
I have been trying to come to grips with a segment of the correspondence between Welby and Russell in 1905. I do not claim to have exhausted the investigation, only to have provided one possible lens through which we can begin to understand it. I have tried to show that, while an initial reading of the 1905 correspondence may yield the interpretation that Welby has misunderstood Russell, a more fruitful interpretation would be to think of Welby as bringing a pragmatist attitude to bear on Russell's theory of descriptions. Such a move is not, I will admit, ‘pragmatist’ in the classical sense of the term. However, there are pragmatist themes present in a use-based account of meaning, and I am convinced that a closer reading of her work would reveal a philosophical pragmatism worthy of the name. In light of this, Welby's correspondence with Russell can be understood as playing into Misak's recent narrative about the blur between “Russell and the Pragmatists.” On my view, their correspondence can be understood as contributing one more bit of evidence in favour of Misak's project in Cambridge Pragmatism.
It is a shame that Russell did not come around to a use-based account of meaning sooner, as, if he had, he may have been more open to Welby's suggestions during their 1905 correspondence. He may have been more willing to carry their conversation a little bit farther and begin to work out the connections between her “Significs” and his theory of denoting. Unfortunately, it did not turn out that way. Welby died in 1912 without ever knowing of Russell's eventual turn to use. Still, their correspondence was not for nothing. At the very least, Welby can be understood, not as misunderstanding Russell, but as attempting to re-orient his focus toward a use-based account of meaning as early as 1905.
