Leavis and Wittgenstein
The literary critic, according to Leavis, “ought to be an anti-philosopher.” Leavis ranked Wittgenstein, in this regard, with other philosophers. This, I argue, was a mistake. Wittgenstein’s philosophy is anti-Cartesian and anti-Lockean in ways that not only mirror Leavis’s distrust of Locke and Descartes but also advance his efforts to argue against them. This essay specifically explores the commitment of both Leavis and Wittgenstein to the idea of “the life of language”: to the idea of language as deeply involved in the constitution of “human worlds,” rather than a mere system of notation for the recording of natural regularities.
I think of myself as an anti-philosopher, which is what a literary critic ought to be.—F. R. Leavis, Thought, Words and Creativity
For a number of years my work has been partly occupied with the examination of various points of contact between philosophy and literature. It involved, however, no more than marginal reference to the work of F. R. Leavis, certainly because of a culpable lack on my part of extended acquaintance with his work, but also to some extent, no doubt, because of Leavis’s own resolute denial of the possibility of any useful interchange between philosophy and literary criticism as he conceived it.
Recently, however, one or two readers, in particular Danièle Moyal-Sharrock, to whom I am much indebted for the tip, took me to task for this, pointing out that some of the notions developed in my [End Page 206] work—notably that of literature as one of the main activities involved in the constitution of what I, like Leavis, had taken to calling “human worlds”—bear a strong family resemblance to related ideas advanced, with similar goals in view, by Leavis in two late works: The Living Principle: “English” as a Discipline of Thought (1975), and Thought, Words and Creativity: Art and Thought in Lawrence (1976). There is no doubt a certain oddity about this parallel, or convergence, given that my ideas have been deeply influenced by the thought of Ludwig Wittgenstein, a philosopher toward whom Leavis’s attitude was, in the last analysis, dismissive. But its very oddity, perhaps, makes the convergence worth looking into a little more deeply. I shall begin that process here.1
In the two books I mentioned, Leavis, as is usual with him, combines passionate, searching, and often profoundly inventive thought with equally passionate denunciation of ways of thinking hostile, as he sees things, to the requirements of a living, creative culture.
Here, as elsewhere in his work, a main representative of the forces of darkness is “philosophy,” by which he tends to mean (i) the tradition of “analytic” philosophy inaugurated, at the start of the last century, by Russell and Moore; (ii) the earlier tradition of British Empiricism (against which Blake and Coleridge also reacted, and for not altogether dissimilar reasons), whose main ideas analytic philosophy took up and carried forward into the twentieth century; and (iii) the tradition of empirical social science to which British Empiricism, historically speaking, also gave rise.
However, Leavis had too good a natural talent for philosophy not to see that, behind much of what he finds damaging in all these intellectual currents, stand the hoary, seventeenth-century outlines of the Cartesianism upon whose foundations the entire edifice of modern philosophy has been erected.
Hence, much of what he has to say in the more philosophical chapters of these two books revolves around an attempted critique of Descartes, and more particularly of the Cartesian dualism of Mind and Body, with all (and it is a great deal) that that entails. Such a critique, of course, is a major enterprise of “technical” philosophy. The peculiar talents and strengths of a literary critic, even a major one, are not adequate to such a task, and Leavis recognizes that they are not. Accordingly, he turns for support to philosophical allies. The chief ones he finds are the historian of philosophy Marjorie Glicksman Grene, and the physical-chemist–turned–philosopher and social theorist, Michael Polanyi. In the seventies of the last century these were, and remain, respected names [End Page 207] in philosophy and social theory, but no one even then would have put either in the first rank of their chosen disciplines (except, in Polanyi’s case, of physical chemistry), and neither, I suspect, is widely read today.
What is odd, in fact, to a philosopher about Leavis’s attack on Descartes is that he does not seem to be aware of the series of major critiques of Cartesianism mounted in the twentieth century by philosophers who definitely are in the first rank. The list, even confining it to writers whose work was freely available at the time, includes, at the very least, Martin Heidegger, Maurice Merleau-Ponty—and Wittgenstein.
Given that, and given the fact that Leavis knew Wittgenstein personally, as he could hardly have failed to do in the Cambridge of the nineteen thirties, it is perhaps surprising that he failed to realize the latter’s significance as an opponent of Cartesian dualism, and of the epistemological scepticism, equally unsympathetic to Leavis’s outlook, that has proved one of Descartes’s most enduring legacies to subsequent philosophy.
But, as Leavis’s brief account of their relationship, published in The Human World,2 makes clear, that relationship remained in crucial respects disappointingly shallow. What is fascinating about that memoir is the capacity Leavis there displays for acute and sympathetic personal observation of a man with whom he also confesses to “an intellectual incompatibility, and perhaps something like an antipathy of temperament” (CAP, p. 144). Leavis’s recognition of Wittgenstein’s “genius” even extends to a severe attack on Bertrand Russell as “having no glimmering of Wittgenstein’s immense superiority to him as a person—as a centre of life, sentience and human responsibility” (CAP, p. 132). But there is in Leavis no real sense of what this genius amounted to; of what, in philosophical terms, it was in the process of producing. Any hope in that direction is dashed by the opening sentence of the memoir: “I had better say at once that I didn’t discuss philosophy with Wittgenstein.” Nor was there much discussion of literature. “Cultivated as he was, his interest in literature had remained rudimentary” (CAP, p. 129).
In the two books that concern us, published twenty years after Wittgenstein’s early death, Wittgenstein’s ideas, in tune with much philosophical opinion at the time, are presumed to be either impenetrably obscure (“literary students … will hardly be expected to tackle Wittgenstein’s very difficult text”3) or else to be adequately represented by the “ordinary language philosophy” still at that point flourishing at Oxford. “There is today a cult of Wittgenstein,” observes Leavis, commenting acidly on the suggestion that “seminars on his later philosophy should be arranged for students of English.” [End Page 208]
He is an enemy of linguistic “science” whose philosophy is called linguistic: here we have what prompted the suggestion and still prompts it. But the conception of language implicit in that “linguistic” is of no interest at all to the intelligent student of English—no interest and no use.(LP, p.101)
It emerges a page or two later that Leavis’s main guide to the despised account of language peddled by the “cult of Wittgenstein” was David Pears’s little paperback Wittgenstein (Fontana Modern Masters), which had appeared in 1971. Pears’s book remains a good beginner’s guide for those interested primarily in philosophy. But for those with Leavis’s interests, a very different guide is needed. That guide I will now try to make at least some initial moves toward supplying.
I said earlier that Leavis’s ideas concerning “the human world” or “the third world” prompt comparison with Wittgenstein. That, I think, is true enough, and I pursue some of those links in my new book. But there is a more fundamental set of connections between the two, which proceed at the level of their respective accounts of language as “living”—a living and growing human enterprise, linked organically to a multitude of other human enterprises, rather than a mere system of descriptive notation—and their common discontent with kinds of philosophy that either ignore, or work actively to constrain and falsify, the life of language.
We can best commence the exploration of Leavis’s stance in this connection by examining his dismissal of a critic whose approach he regards as viciously “philosophical.” In Thought, Words and Creativity, at the start of his essay on Lawrence’s “The Captain’s Doll,” he considers, and dismisses,
a philosopher’s criticism of myself as one who, in my characterisation of Lawrence’s genius, makes great use of the word “life,” but seems quite unaware of the need to attempt some distinguishing between the diverse values it covers. Since the philosopher has a post in a university English department and is articulate and confident, I had remarked that, his assumption being that the distinctions must be defined philosophically, he must have done, educationally, a good deal of damage. But even when they have not also earned a standing as philosophers, can we trust lecturers in English to point out—or to know—that the most important kind of thought is decidedly not philosophical?4 [End Page 209]
The manner in which, in this rather characteristic stretch of Leavisian prose, irritable invective seems to be usurping the place of argument shows Leavis, here as elsewhere, needlessly playing into the hands of his detractors. He disposes, however, of a perfectly sound reply to his unnamed critic, which he duly offers a page or two further on: “The limitation, the very much less intricate complexity [of The Captain’s Doll relative to Women in Love], lends itself to the expository treatment of Lawrence’s basic thought; for the treatment of the love-theme in ‘The Captain’s Doll’ is equally a treatment of the life-theme—bears, that is, directly on the philosopher’s and the critic’s objections to the indeterminateness of the word ‘life’” (TWC, p. 94).
What a writer “means by” a word, Leavis is saying, is not something that we have to be taught at the outset in order that we may read him (or her) with understanding, but something that we learn by reading him (her).
The same point is put by Merleau-Ponty, in La prose du monde, in terms of the latter’s distinction between constitutive and constituted (or sedimented) language. “One can have no idea of the power of language until one has taken stock of that working or constitutive language which emerges when the constituted language, suddenly off center and out of equilibrium, reorganises itself to teach the reader—and even the author—what he never knew how to think or say.”5
We begin to read, Merleau-Ponty argues, by the light of the everyday meanings of words constituted by, or sedimented out of, long usage in many contexts. But a great writer can set before the reader imagined circumstances in which the worn faces of common words begin to melt into an indeterminacy from which begin to unfold new possibilities of meaning. Thus, Merleau-Ponty tells us, knowing the common meaning of the word rogue (coquin), we know what Stendhal means when he tells us that the fiscal general, or chief justice, Rossi (in La Chartreuse de Parme) is a rogue. But by the time we have finished the book, Stendhal’s treatment of Rossi has enlarged the meaning of the word “rogue”: “a rogue is Rossi” (PW, p. 12).
In the same way, as I have argued elsewhere, Shakespeare’s treatment of the relationship between Lear and Cordelia enlarges our conception of what is involved in filial piety. He shifts the notion away from the standard Renaissance conception emphasizing mere passive obedience to parental decree, thereby associating it with something both more independent and more actively compassionate.6 [End Page 210]
If these arguments go through, then Leavis is plainly right to suggest that we cannot read such works as Women in Love or The Captain’s Doll with any hope of grasping what they have to offer us, unless we are prepared to allow words to retain the indeterminacy of meaning, inherent in what Merleau-Ponty calls the constituted language of the everyday, since that very indeterminacy provides the indispensible basis for such constitutive extensions and augmentations of constituted or “everyday” meaning.
But if Leavis is right about that, he is also right to object that the demand for exactness of meaning—in effect, the demand that every word that a writer introduces, or that a critic introduces in analyzing that writer’s work, be given the type of exact verbal or operational definition appropriate in mathematical or scientific studies—misunderstands not only the nature of literary language, but the nature of language itself.
Continuing the argument along these lines, one might reflect that the reason why mathematics, the sciences, and certain types of philosophical inquiry have recourse to exact verbal or operational definition is precisely, after all, that the “ordinary language”—the language in which everyday life is conducted—is, beyond a certain point, indeterminate; and indeterminate, precisely, in ways that make it unfit for the purposes of those types of inquiry. Exact definition establishes a language whose powers of unforeseen extension and exfoliation of meaning have been clipped and confined, by definition, to fit it for a specific range of purposes, outside which it can no longer function; much as evolution gradually tailors certain species inhabiting special environmental situations to fit the requirements of such an environment in specialized ways. For such species, loss of environment commonly results in extinction, while less specialized species, able to adapt to new environments, survive. The special languages of mathematics and the sciences, artificially preserved by exact definition from change or accretion of meaning, resemble the flightless birds of inaccessible islands; the indeterminate argot of everyday life, species such as Homo sapiens or the crocodile, which avoid extinction precisely by retaining the power of adaptation to new circumstances.
Because what interests the creative writer is not the pursuit of an exact science but, on the contrary, the potentialities of meaning in everyday life, he cannot concern himself with specialized languages of that sort. It is precisely the commonplace language of everyday life, the unchastened and unrefined argot whose indeterminacy and consequent potentiality for accretion and exfoliation of meaning are two faces of the same thing, that he must use. [End Page 211]
Leavis is, moreover, by no means the only literary figure to have conducted an attack on exact definition with this point at its heart. Henry Fielding, for example, though not a favorite author of Leavis’s, makes his Sophia passionately resist her aunt’s attempt to demonstrate that respect for the “proper” meaning of words—as exemplified in this case by the definition of “hatred” in Bailey’s Dictionary—can allow her no rational ground for her hatred of Mr. Blifil. I have in mind the following passage, in book 7, chapter 3, of Tom Jones:
“If I was not as great a philosopher as Socrates himself,” returned Mrs. Western, “you would overcome my patience. What objection can you have to the young gentleman?”
“A very solid objection, in my opinion,” says Sophia, “I hate him.”
“Will you never learn the proper use of words?” answered the aunt. “Indeed, child, you should consult Bailey’s Dictionary. It is impossible you should hate a man from whom you have received no injury. By hatred, therefore, you mean no more than dislike, which is no sufficient objection against your marrying of him. I have known many couples, who have entirely disliked each other, lead very comfortable genteel lives.”
Since the events of the novel, as they have developed to this point, have supplied both Sophia and the reader with ample evidence of meanness, hypocrisy, self-love, and sadism as leading features of young Blifil’s character, one of the things Fielding may be suggesting here, pace Bailey, is that hatred may have a moral as well as a personal dimension.
And certainly Fielding shows elsewhere, for example in the construction of his cod dictionary A Modern Glossary7 (which includes such entries as “HONOUR: Duelling”; “MODESTY: Awkwardness, rusticity”; and “FOOL: A complex idea, compounded of poverty, honesty, piety and simplicity”), that his and Sophia’s contempt for the niceties of exact definition rest on exactly the same grounds as Leavis’s. Fielding holds, in effect, with Leavis, and for that matter with Orwell, that an insistence on exact definition in everyday context, though it may at times serve the purpose of Orwell’s “Newspeak” in crippling the ability of certain minds to think beyond the limits so established, can in the end do no more than disguise, for a time, the intrinsic capacity of meaning in a natural language to transgress any limits we attempt in this way to impose upon it.
But if the meaning of a word can never be completely and finally specified, someone might object, how is it possible for us to communicate by means of language? Without the aid of exact, agreed-upon [End Page 212] definition, how can we know that we are talking about the same thing, or more simply, understand what we are supposed to be talking about in the first place? Call this the Exactness Objection (EO).
Put in this way, EO combines, without clearly distinguishing them, two subordinate questions that deserve to be distinguished. On the one hand there is a question of fact: is the nonspecialized language of every day actually, as Leavis puts it, “indeterminate”? On the other hand, there is a question of the type that philosophers since Kant have called “transcendental”: How is it possible that meaning, in the language of the everyday, could be indeterminate? (How could we hope to know what we were talking about if it were?)
Leavis’s arguments in The Living Principle, along with the supporting arguments I have adduced from voices as various as Merleau-Ponty, Fielding, and my own earlier work, all address the first of these questions. But, given the interplay of the two questions within the logical structure of EO, the lack of a prior, and persuasive, answer to the second, transcendental, question must cast doubt on the plausibility of any affirmative answer to the first. And once general doubts about the intrinsic possibility of the alleged indeterminacy have been raised, specific doubts about each of the affirmative answers we have so far canvassed are easy to formulate.
Thus, for instance, if the meaning of “life” is, as Leavis seems to want to say, simply “indeterminate” in everyday usage, why does it not follow that Lawrence, or anybody else, can give the word any arbitrary, private meaning he pleases? And how, if that is indeed the conclusion to be drawn, could Lawrence’s private use of the term in itself attract, without circularity, the respect due to a great writer? For that matter, why shouldn’t Fielding’s satirical fun at the expense of Bailey’s Dictionary be taken to impugn, not the lexicographical enterprise per se, but simply bad lexicography? Again, could one not reply to Merleau-Ponty that, while Stendhal may present his reader, in the person of Rossi, with an unfamiliar example of a rogue, he cannot by such means change the meaning of the word “rogue”? And why, if it comes to that, should Shakespeare’s treatment of Cordelia be thought of as changing, or transcending, rather than as dependent for its dramatic force upon, our everyday understanding of such terms as “daughterliness”?
I was once foolish enough to suggest tentatively, in correspondence with a fellow philosopher, a classicist by training and a Platonist by philosophical predilection, that although the conception of daughterliness entertained by many of Shakespeare’s contemporaries or by, say, [End Page 213] a conservative Japanese of the same period, might have differed from the kind of daughterliness exemplified by Cordelia, all three might, nonetheless, have shared a common understanding of the basic meaning of the term “daughterliness.” My Platonist friend replied, wither-ingly, that the meaning of a general term just is a concept; that concepts not being constructs of human convention, but inherent in reality, are discovered, not invented, and that if Shakespeare were right about the content of the concept of daughterliness, then his contemporaries, and for that matter the Japanese, were just wrong about it, in the same sort of way in which those who believed the planetary obits to be circular were just wrong about the geometry of the solar system.
The issue of what we are to make of Leavis’s dismissal of “the philosopher’s and the critic’s objections to the indeterminateness of the word ‘life’” thus seems to turn entirely on the question of whether an answer can be found to EO, understood as what Kant would have called a transcendental question [transzendentale Frage]: How is it possible that meaning in everyday language should be indeterminate, and yet that we should be able to understand one another?
The only plausible answer to the latter question that I am aware of is to be found in the later work of Wittgenstein. In part 1 of the Philosophical Investigations, from around paragraph 75 to around 89, Wittgenstein embarks upon what is, in effect, an extended attack, which he continues to develop elsewhere, upon the very thesis to which Leavis’s “philosopher and critic” has nailed his flag: namely, the thesis that exactness in the prior specification of meaning, besides serving whatever specific practical purposes in our lives it may from time to time serve, is a necessary, and therefore completely general, condition of understanding.
I.84 sets the tone for this attack: “I said that the application of a word is not everywhere bounded by rules. But what does a game look like that is everywhere bounded by rules? Whose rules never let a doubt creep in, but stop up all the cracks where it might?—Can’t we imagine a rule determining the application of a rule, and a doubt which it removes—and so on.”8 And, at I.85: “A rule stands there like a sign-post.—Does the sign-post leave no doubt open about the way I have to go? Does it shew which direction I am to take when I have passed it; whether along the road or the footpath or cross-country? …—So I may say, the signpost does after all leave no room for doubt. Or rather, it sometimes [End Page 214] leaves room for doubt and sometimes not. And now this is no longer a philosophical proposition but an empirical one” (PI, p. 34e).
Finally, at I.88:
If I tell someone “Stand roughly here”—may not this explanation work perfectly? And cannot another one fail too?
But isn’t it an inexact explanation?—Yes; why shouldn’t we call it “inexact”? Only let us understand what “inexact” means. For it does not mean “unusable.” And let us consider what we call an “exact” explanation in contrast with this one. Perhaps something like drawing a chalk line round an area? Here it strikes us at once that the line has breadth. So a colour-edge would be more exact. But has this exactness still got a function here: isn’t the engine idling?(PI, p. 35e)
Wittgenstein’s point is that, for a rule-governed activity to serve in most cases the purposes of communication, it is neither necessary nor possible to equip it with rules so comprehensively specified as to rule out in advance every possibility of misunderstanding. The point is generalized, with specific application to language at, among other places, I.142. “It is only in normal cases that the use of a word is clearly prescribed; we know, are in no doubt, what to say in this or that case. The more abnormal the case, the more doubtful it becomes what we are to say” (PI, p. 48e).
Leavis’s “critic and philosopher” might reply that Wittgenstein’s talk of signposts and directions about where to stand is irrelevant to the issue: we were talking about meaning, about understanding or failing to understand what is meant by an expression in language, not about understanding or failing to understand how to make use of a mere conventional device, such as a signpost.
It is precisely the validity of that distinction, however, that is at issue in the Investigations. At I.8, Wittgenstein offers the example of a simple “language game” involving the letters of the alphabet reinterpreted as numerals, color samples, and four words for objects, “block,” “pillar,” “slab,” and “beam”: “A gives an order like ‘d—slab—there.’ At the same time he shews the assistant a colour sample, and when he says ‘there’ he points to a place on the building site. From the stock of slabs B takes one for each letter of the alphabet up to ‘d,’ of the same colour as the sample, and brings them to the place indicated by A” (PI, p. 5).
At I.10, Wittgenstein raises what amounts to the question that bothers Leavis’s “critic and philosopher”: putting aside the use A and B make [End Page 215] of them, what do these words mean; what do they signify? “Now what do the words of this language signify?—what is to shew what they signify, if not the kind of use they have?—and we have already described that” (PI, p. 5).
The importance of this move is, among other things, that, as Wittgenstein goes on to show, it radically weakens the criteria that a speaker must satisfy in order to be said to understand a word or sentence.
We are apt to suppose that to understand a general name, or a sentence, is to possess the ability to recognize, or pick out, items or characteristics of the sort to which the name applies, or states of affairs of the kind that the sentence asserts to obtain. And the ability to recognize, to pick out, these things, we presume, reasonably enough, must presuppose knowledge of their nature. Thus, if one understands the meaning of “daughterliness,” one must be able to recognize, or pick out, daughterly acts or natures, and therefore one must know what daughterliness is; must be, in other words, privy to the nature, or essence, of daughterliness.
Professional philosophers have, on the whole, confined themselves to formalizing and intellectualizing this folk-philosophical account of what it is to understand. Thus, for instance, the two main alternative accounts disputed among analytic philosophers over the past thirty years or so, of what it is to understand the meaning of a general name, are, on the one hand, that it involves possession of a description sufficiently precise to single out all and only those things to which the name applies, and on the other, that it involves baptismal acquaintance with some sample or samples of a “natural kind,” such as water, or tigerhood, together with knowledge of what it is that makes a given sample a sample of that kind; in the case of putative water, perhaps, its turning out to consist of H2O; in the case of a putative tiger, perhaps, its turning out to possess a certain genetic constitution.
On any such view, plainly, Leavis’s “critic and philosopher” is in the right. If it is a necessary condition of merely understanding the meaning of such general names as “daughterliness,” or “rogue,” or “life,” that one grasp the nature, or the essence of the things those terms pick out; then it follows, contrary to Leavis, Merleau-Ponty, et al., that nothing a writer might do with the words “rogue,” or “life,” or for that matter “daughterliness” could possibly change the reader’s opinion concerning the nature of the things to which those words refer, since without already fully grasping the natures of the things in question, the reader could have no idea of what the writer is talking about. [End Page 216]
On Wittgenstein’s view, however, the picture changes. If Wittgenstein’s arguments go through, then understanding the meaning of a word or sentence does not necessarily require one to fully grasp the nature of the realities picked out by word or sentence. All it requires is that one know how to deploy—to use—the word in the context of engaging in whatever human practice or practices serve to confer a use, and therefore a sense, upon it.
What about “daughterliness,” then, or Leavis’s “life”? We need, perhaps, to take the advice that Wittgenstein offers at I.77: “In such a difficulty always ask yourself: How did we learn the meaning of this word (‘good’ for instance)? From what sort of examples? In what language-games? Then it will be easier for you to see that the word must have a family of meanings” (PI, p. 31e).
We learn the meaning of expressions like “daughterliness,” I take it, as part of learning what it is to take part in human relationships of one sort and another. In that process we learn quite early on that some ways of behaving tend to disrupt, and in the end erode, relationships, while other kinds of behavior work to sustain and develop them. “Daughterly,” “fatherly,” “sisterly,” “brotherly,” and so on are terms used to characterize behavior that falls into the latter category. That way of using them, Wittgenstein suggests, is all that their “meaning,” their “sense,” does, or could, consist in. That is all we need to grasp in order to be capable of understanding other people’s use of the word, and to use it ourselves in ways that others will understand. As to what daughterliness may or may not actually consist in, as to what the nature, or the essence, of daughterliness might be, we need possess only the vaguest and most sketchy ideas; need, indeed, if it comes right down to it, know nothing at all.9 The stage is set, in other words, for Shakespeare to show us how it is that the apparent undaughterliness of Cordelia’s conduct in act 1, scene 1 of King Lear in fact manifests a deeper, more profound daughterliness.
But, someone might object, if the nature of daughterliness is not fully determined by the meaning of the word, do we not have to say that the sense of “daughterliness” exemplified by Cordelia represents merely the arbitrary appropriation of the word to stand for something having no connection whatsoever with anything it meant previously?
This objection gestures toward a vast background of recent debate in philosophy. It is possible to interpret various remarks of Wittgenstein’s, taken in isolation, as entailing something that can be represented as an implausibly extreme nominalism or conventionalism. Critics of this type attribute to Wittgenstein such views as (i) meaning is a function of [End Page 217] collective agreement and nothing else, or (ii) the validity of each step in a mathematical proof is solely a function of collective decision. Both interpretations have spawned an extensive philosophical literature over the past half century.
One thing that has emerged with some certainty from these discussions, however, is that, whatever the intrinsic merits or flaws of the extreme skepticism attributed to Wittgenstein concerning rules, meaning, and necessity, it is entirely unclear either that Wittgenstein intended to advance such views or that the passages generally cited are sufficient, in context, to support such an interpretation. Hence, beyond noting the existence of this larger debate, and footnoting, as I have just done, a few of the more important contributions to it, I shall ignore it here in favor of directly answering the specific objection in hand.
The answer to that objection (though beyond citing I.77 I shall shirk the exegetical labor needed to demonstrate it to be not merely mine, but also Wittgenstein’s) is that the nature of the employment, or “use,” that originally introduced a given word into the language continues to control subsequent extensions and mutations of its meaning. Thus, in the case of “daughterliness,” the originating use of the word is to provide a general label for ways of occupying a certain role in a certain relationship that sustain rather than disrupt the relationship. It is that which allows Shakespeare to equip Cordelia with words and actions that demonstrate how apparently disruptive conduct can be the reverse of what it seems; the parallel with Kent’s combination of love, fidelity, and opposition to Lear being a part of the demonstration.
Similar things can be said, along Wittgensteinian lines, about (Leavis’s favorite expression) “life.” “Life” and “death” function, basically, as opposed terms marking a transition. “Life” is not, in other words, the name of an aspect of reality—a natural kind, as philosophers like Quine and Hilary Putnam like to say. One could hardly explain the meaning of “life” ostensibly, for what would one point at? Nor would it even make sense to suggest that one might need to know the nature of life, or what life consists in, in order to understand the meaning of the word. At the same time, the logical nature of the basic use we make of “life” and “death,” to mark the poles of a transition, is enough to direct the extension of each into further, related uses. Thus we can speak by analogy of what has never lived as “dead,” as in the phrase “dead matter,” because a dead body is no more than that (thus Antony: “O pardon me, thou bleeding piece of earth”). [End Page 218]
Then there are cases in which, as Wittgenstein put it at I.84, “a doubt creeps in.” One such is the case of viruses, which behave like living things in the sense that they divide and multiply, invade and destroy living cells, mutate and so on, but are in essence simply complex molecules. Are we to call them living things or dead matter? Are viruses living organisms? As Wittgenstein says, we have no option but to decide. But the decision is not an arbitrary one. The behavior of viruses mimics in certain respects the behavior of living cells; in addition, the molecules that constitute viruses are molecules of nucleic acids, molecules of the sort, that is, which control development and inheritance in living organisms. Hence it is less misleading to extend the term “living organism” to cover viruses; but the extension, although it may be motivated by the basic use accorded to the terms “life” and “living” in English is—this is the Wittgensteinian point—an extension, which rests on a decision: it does not constitute a mere application of the meaning of “life” or “living organism” to the specific case of viruses.
In related ways, the basic use of “life” and “death,” as terms marking a transition, both attracts and controls a large and open-ended range of such extensions. We come to use “life” and “living,” for example, as general terms to cover all that lies on the hither side of the transition in question. Both terms then begin to build up a range of connotations based upon metaphor and metonymy.10
Thus the idea of life comes to be associated with change and the new; death, accordingly, with stagnation and repetition. We call “lively” a debate rich in new ideas and argument; we call a village or town in which the daily life of the inhabitants barely sustains itself in leaden, diurnal repetition a “dead-alive hole.” And at a certain point in this process we reach Lawrence’s deployment of the words “life” and “death” in The Rainbow, of which Leavis says, among much else, “[Birkin] is the man who, though he would like to save Gerald from his mechanistic fate, and have him for his friend, diagnoses Gerald’s disease and sees him as modern civilization speeding towards the death-break—the cessation of life (which is creativity) in a ‘frozen world’” (TWC, p. 130).
That thought brings us back to the point from which we began to forge these connections between Leavis and Wittgenstein: the moment at which we came up against Leavis’s bitter scorn at the idea that “distinctions must be defined philosophically,” his ex cathedra pronouncement that “the most important kind of thought is decidedly not philosophical,” his offhand insistence that meaning in ordinary usage is “indeterminate” in ways that “philosophical” definition cannot capture, along [End Page 219] with the equally offhand suggestion that a study of the bearing upon the meaning of “life” of “[Lawrence’s] treatment of the love-theme in ‘The Captain’s Doll’” should be enough to show the “critic and philosopher” why meaning is indeterminate, and why that doesn’t matter!
It is worth noticing in passing that Leavis’s objection to “philosophical” definition is one main source of his anti-Cartesianism. Descartes’s Discourse on Method, with its admiring espousal of the “long chains of perfectly simple and easy reasonings by which geometers are accustomed to carry out the most difficult demonstrations,”11 as a general model for rational thought, and the accompanying resolution to “divide each problem into as many parts as … feasible” (PhW, p. 20), lies at the root of every form of reductionism in post-Renaissance thought.
However, reductionism—the idea that advances in the understanding of complex phenomena always involve showing the latter to be constructed out of simple elements—is now very deeply rooted in the practice of both science and philosophy. Hence it is not even necessary to be a foot soldier in the literary culture wars of the past half century and earlier to see that Leavis’s contempt for “philosophical” definition must, in most people’s minds, expose him to a charge of antiscientism, and perhaps even irrationalism.
“Reductionism” can mean two different things, however. Explanatory reductionism, the form most often seen in natural science, involves no more than the application of Occam’s razor to the choice of explanatory principles, the rule being that, where possible, one dispenses with a complex explanation for phenomena when a simpler one presents itself.
“Logical” or “definitional” reductionism, on the other hand, the kind most often met with in philosophy, is the claim that some class of statements can be replaced systematically, and without loss of meaning, by another class of statements. A typical example, encountered in most beginning classes in the subject, is the so-called sense-datum theory: the perennial thesis that the meaning of statements concerning physical objects can be fully expressed by means of statements mentioning only sensations.
This second type of reductionism is essentially concerned with meaning. Its claim is that entire categories of discourse can be shown to be definitionally dispensable, in favor of simpler ways of speaking more in touch with empirical reality. Its attraction—or one of its attractions—is that it seems to offer a way of both detecting and deleting inflated, unmeaning discourse from public debate. It came into English thought originally by way of the anglicized Cartesianism of Locke’s Essay Concerning Human [End Page 220] Understanding (1690), with its insistence that “complex ideas,” if they are not to be dismissed as phantoms of the mind, must be capable of analysis in terms of “simple ideas” originating in experience.
Locke calls ideas once traced, in this way, to their origins in experience, “determined.” His ideal is that “when any man uses a term he may have in his mind a determined idea which he makes it the sign of.” Where a man cannot do this, he says, “He in vain pretends to clear and distinct ideas; it is plain his are not so; and therefore there can be expected nothing but obscurity and confusion where such terms are made use of which have not such a precise determination.”12
The stance of Leavis’s “philosopher [who] has a post in an English Department” is plainly identical with Locke’s. The former thinks, in effect, that Leavis has no right to the term “life” unless it can be shown to be what Locke called a “determined” idea by being supplied with a precise definition in terms of ideas simple enough to be considered the direct deliverance of experience.
Since Leavis’s quarrel is merely with reductionism of this philosophical type, rather than with the explanatory reductionism characteristic of the natural sciences, his antireductionism offers no grounds at all for a charge of antiscientism, let alone irrationalism. Leavis’s quarrel, like—it could equally well be argued—Blake’s, is far more with Locke than with Newton. And in the engagement with Locke’s version of Cartesianism, Wittgenstein’s views on meaning and use offer important lines of support both to Leavis and to his present-day disciples. If grasping meaning is primarily a matter of grasping the role played by an expression in the conduct of a practice, such as measurement, or the institution of marriage, or family relationship, then knowing what a word means is something intrinsically distinct from—though it may often include—knowing how to apply it.
I may, for example, know what “micron” means—that it is the name of a module of measurement—without possessing the least idea of how to set about assessing the length of anything in microns. Hence, on the one hand, knowledge of the meaning of N may leave entirely open, and to be determined, the nature of Ns; while, on the other hand, investigation of the nature of Ns may lead to drift in our sense of the meaning, in the sense of the connotations, the implications of one context of other, of N, without that drift being arbitrary, or uncontrolled; since it continues to be controlled by the continuing pattern of use, of practice, from which N originally acquired and continues to derive its basic sense. [End Page 221]
The determination of meaning, pace Locke, has nothing to do with establishing equivalence relationships between the “complex” and the simple, and everything to do, as Leavis thought, with the power of language to reveal, in creative discourse of all kinds, including major creative literature, new aspects of the choices we originally made in stipulating roles, and hence senses, for its component expressions: to reveal, in short, the gradual broadening of those originating choices into what Wittgenstein, at I.77, calls a “family of meanings” (PI, p. 31e).
To explore the structure of such a “family of meanings” is not only to explore the place of the various practices and extensions of practice on which successive uses depend but also, in exploring that, to explore the roles of the practices in question, and of the terms and connotations they license, in the constitution of our common life. That is precisely the type of exploration that interests both the creative writer and the literary critic. To explore the roots of the common language we speak every day, as the best creative writing both invites and compels us to do, in practices we take for granted as the common foundations of our lives—to examine its powers of expression, the limits of those powers, and their potential for further development—is to engage, in effect, in an act of personal, but also collective, self-examination. It is to think about our own “nature,” collective and personal alike, in a way that requires us to bring before the bar of critical consciousness—rather than simply going through the steps of the various dances they prescribe for us—the common practices of the historic society to which we belong, practices constitutive both of the meanings of the words and sentences of language we speak, and of the life that that language informs, and to inquire what those practices have made and might make of us. It is this sense that the critic Harold Bloom was in no way fanciful to choose to title one of his books Shakespeare: The Invention of the Human.
Wittgenstein’s thoughts about meaning and use, I suggest, by allowing us to get clearer in the above ways about how and why something worth calling “thought” might enter into the reading of creative literature, also allow us to give a perfectly respectable sense (a sense neither irrationalist nor antiscientific, that is to say), to Leavis’s defiant “the most important kind of thought is decidedly not philosophical.”
The question of the relative “importance” of one kind of thought over another we may safely leave on one side. My own view is that Leavis would have done well to replace “important” in that sentence with “indispensible,” as elsewhere he speaks of it as “necessary”: “What we have to get essential recognition for is that major creative writers are concerned with a necessary kind of thought” (LP, p. 20). [End Page 222]
“Philosophical” thought, in the strict sense, is almost without exception critical and reductive in character. Its aim is to root out intellectual confusion. It pushes intellectual dilemmas to their logical limit in the hope either of deciding or transcending them. In the process it forces one endlessly to pursue Locke’s ideal of analyzing the complex in terms of the simple, of sharpening small differences of meaning, in order, precisely, to deny confused thought the refuge of obscure language.
“Thought” of the kind Leavis has in mind, on the other hand, the kind of thought deployed equally in creative writing and in criticism, tends to concern, rather than to use, language. And, because of that, it requires both reader and critic to keep before their minds the full “family of meanings”; meaning, that is, in its full and native power endlessly to mutate and transform itself, and us. Leavis puts the point more wryly, and in a more balanced way, in the essay on “imagery and movement” in The Living Principle: “The point to be stressed is that, whatever was gained by the triumph of ‘clarity,’ logic and Descartes, the gain was paid for by an immeasurable loss: you can’t, without basic reservations, subscribe to the assumptions implicit in ‘clear’ and ‘logical’ as criteria without cutting yourself off from most important capacities and potentialities of thought, which by its nature is essentially heuristic and creative” (LP, p. 97).
It is a point beautifully exemplified in his analysis of the lines (in Antony and Cleopatra, act 3, scene 2):
Her tongue will not obey her heart, nor canHer heart inform her tongue—the swan’s down-feather,That stands upon the swell at full of tide,And neither way inclines.
The author of a note in the Arden edition suggests that it is not clear (with the implication, Leavis suggests, that it should be clear) whether what is being compared to the “swan’s down-feather” is Octavia’s heart or the inaction of both heart and tongue. Logic chopping of this kind may be faithful to the Cartesian conviction that complexes are to be understood, if at all, in terms of their mode of constitution out of simples. But, as Leavis shows in his analysis of the lines, it is the last thing we need, if we are to respond intelligently to writing whose entire modus operandi rests upon the holding open and chiming together, in the mind of the reader, of multiple possibilities of meaning and reference. [End Page 223]
The implied criterion, “clarity,” entails an “either/or”: does the image mean this or that? The reductive absurdity of the conception of language behind the criterion thus brought up is surely plain. It wouldn’t be enough to say that the image has both meanings: no one really reading Shakespeare would ask to which it is, or to what, that “the swan’s down-feather” is meant to apply metaphorically, because it would be so plain that the relevant “meaning”—the communication in which the “image” plays its part—is created by the utterance as a totality, and is not a matter of separate local “meanings” put together more or less felicitously. The force and precision with which Shakespeare’s English imparts its meaning here depend on the impossibility of choosing one of the scholar’s alternatives as right and the clear inapplicability of the question he puts.(LP, p. 102)
What I have been arguing, in effect, is that Leavis was perfectly correct to regard the type of “thought” evident here as both legitimate and “necessary.” It is the type of thought characteristic of the better creative writers and critics, one that pursues essential kinds of collective and personal self-examination through the examination, in the full richness of its powers and potentialities, of the language in which we explain ourselves to ourselves, and beyond that, of the multifarious underlying practices that give sense to the terms of that language, and simultaneously form us as users and speakers of it. It is indeed, as Leavis believed, in real danger of being deleted from university studies, and from our lives, by the conviction that only the social sciences, philosophy, or “theory” can contribute any insight worth having concerning the human condition. And the situation is rendered all the worse for the fact that it is a type of thought capable of delivering sharp and essential criticism of some of the more absurd and damaging illusions engendered by the supposedly more “scientific” or “intellectually demanding” styles of discourse.
It is this sense of impending loss that fuels Leavis’s entirely reasonable bitterness when he says, speaking of the climate of opinion associated with the Robbins Report on Higher Education (1963), in whose aftermath, in British universities, we are all still living:
Nevertheless, it is politic to insist on [literary studies as a] “discipline,” and necessary to be able to justify the insistence—so important is the [literary] kind of intelligence in the present phase of human history, when Lord Robbins, surveying the needs of education at the university level, recognizes that the natural sciences must be complemented by the study of human nature and brings out the force of this recognition by pointing to psychology and the social sciences. These are disciplines; his [End Page 224] concessive gesture towards Literature and the Arts makes it plain that they are to be regarded as pleasing adjuncts to what really matters—graces and adornments in the margin of life that shouldn’t be discouraged; they contribute dignity and amenity.(LP, p. 20)
1. Further development of the ideas presented here can be found in Bernard Harrison, What Is Fiction For? Literary Humanism Restored (Bloomington: Indiana University Press, 2015).
2. F. R. Leavis, “Memories of Wittgenstein,” The Human World 10 (1973); and in The Critic as Anti-Philosopher: Essays and Papers by F. R. Leavis, ed. G. Singh (London: Chatto and Windus, 1982); hereafter abbreviated CAP.
3. F. R. Leavis, The Living Principle: “English” as a Discipline of Thought (London: Chatto and Windus, 1975), p. 103; hereafter abbreviated LP.
4. F. R. Leavis, Thought, Words and Creativity (London: Chatto and Windus, 1976), p. 92; hereafter abbreviated TWC.
5. Maurice Merleau-Ponty, The Prose of the World, trans. John O’Neill (London: Heinemann, 1969), p. 14; hereafter abbreviated PW.
6. Bernard Harrison, Inconvenient Fictions: Literature and the Limits of Theory (New Haven: Yale University Press, 1991), pp. 54–57; hereafter abbreviated IF.
7. Henry Fielding, A Modern Glossary, in The Covent Garden Journal 4 (January 14, 1752).
8. Ludwig Wittgenstein, Philosophical Investigations, 3rd ed. (Oxford: Blackwell, 2001), pp. 33e–34e; hereafter abbreviated PI and cited by both paragraph number and English-language page numbers in this bilingual facing-page edition.
9. For further development of this point, see Patricia Hanna and Bernard Harrison, “The Limits of Relativism in the Late Wittgenstein,” in A Companion to Relativism, ed. Steven D. Hales (Oxford: Blackwell, 2011), pp. 179–97.
10. For an account of the nature of metaphor in terms of a Wittgensteinian “use” or “practice”-based view of meaning, see Bernard Harrison, “The Truth about Metaphor,” Philosophy and Literature 10, no. 1 (1986); and IF, pp. 261–77.
11. René Descartes, Philosophical Writings, ed. and trans. Elizabeth Anscombe and Peter Geach (London: Nelson, 1963), p. 21; hereafter abbreviated PhW.
12. John Locke, An Essay Concerning Human Understanding, ed. Peter Nidditch (Oxford: Oxford University Press, 1975); p. 13. [End Page 225]