INVITED PAPER

1. Signature, substance and spirit

When Cora Diamond attempts to interpret Wittgenstein’s remark “Not empiricism and yet realism in philosophy, that is the hardest thing” – in her influential paper “Realism and the Realistic Spirit” – she begins by making certain connections between Wittgenstein’s aspiration towards philosophical realism and realism understood as a literary aspiration (one particularly closely associated with the genre of the novel).¹ According to Diamond, in a realistic novel we expect character to be built up out of observed detail rather than deduced from a label or a typology (“like all Russian officials, he had a weakness for cards”), we expect certain things not to happen (the absence of magic, myth and fantasy), and we expect an insistence on coherently grasped and displayed relations of cause and effect (an operative conception of how things actually work in our lives). Such expectations are consistent with a standard account of the origins of the novel as a genre – namely, in a rejection of the prior genre of romance on the basis that its conventions appeared to the novelist to misrepresent the nature of reality; they no longer facilitated a way of representing the world that had any chance of seeming accurately to capture its nature as opposed to presenting a mere appearance or representation of it, hence a kind of falsification of it (cf. Watt 1957). In short, the novel’s questioning of the generic conventions it inherits is in the name of a more faithful representation of the real.

In truth, however, the novel was no less a genre than was romance: the individuals it portrayed did not exist, the specific resources used to create the impression of their reality were no less conventional than those of any other literary genre, and the text as a whole could not but be anything other than a linguistic representation of the real rather than reality itself. And it was not long before the conventionality of the conventions of formal realism became apparent, both to readers and to other novelists, to inevitably subversive effect. By the 1760s, Laurence Sterne could write a text that simultaneously deploys the resources of formal realism to great effect in creating the impression of reality, but also makes their conventionality a thematic as well as a formal issue, to the point of parody – in Tristram Shandy. But precisely by pursuing the latently absurd implications of such conventions, by for example attempting to live up to the requirement that there be a one-to-one temporal correspondence between the novel and the reader’s experience of it, Sterne achieves an effect of realism. The reader thinks: if someone really did try to write in accordance with this convention, absurdity of just this kind would really result; and so Sterne manages to recreate the impression of Tristram as a real individual, precisely by allowing him to enact the deconstruction of a preceding convention for the representation of reality – thus showing it to be no more than a convention, and effacing the conventionality of the specific literary means by which this revelation is achieved.

The history of the novel since Defoe, Richardson and Sterne might therefore be written entirely in terms of the ways in which novelists repeatedly subject their inheritance of realistic conventions to critical questioning in order to recreate the impression of reality in their readers (in large part by encouraging those readers to see prior uses of convention to represent the real as merely conventional in contrast with their own, far more convincing ones). The realistic novel endlessly renews its claim to be an unprecedentedly faithful representation of individual human experience of the world in comparison with other literary genres precisely by claiming to be more faithful to that task even than its novelistic predecessors. Only by ceaselessly testing, criticising and otherwise innovating with respect to the conventions through which it represents reality can the novelist create the impression that, unlike her predecessors’ merely conventional efforts, she is conveying reality to her readers as it really is. And since her best efforts could only result in the recreation of new conventions, they – and so the impression of reality they make possible – will inevitably be vulnerable to the critical questioning of her own successors. And this parasitic process is not only open-ended; its reiterated criticisms of past conventions will sooner or later test the faith of both novelist and reader in the very possibility of a representation of individual human experience that is at once both convention-dependent and authentic. One might think of the difference between modernism and postmodernism in this tradition as a matter of whether, in the face of this realization, the novelist can keep his faith in realism, or finds himself forced to abjure it.

In this paper I want to suggest that Cora Diamond’s way of inheriting Wittgenstein confronts us with a philosophical version of this modernist predicament, and instantiates one radical way in which one might keep faith with philosophical realism in the face of its inherent propensity to undercut its own aspiration. To see why, we might begin by viewing the signature concepts with which Wittgensteinian work is so often identified (“language-game”, “grammar”, “ forms of life”) as representational devices forged by Wittgenstein himself in the service of simply putting things before us as they really are (as ways of ensuring that we look and see what is in front of our eyes despite our conviction that it must take a particular form (PI §66), of describing the very various ways language in fact works despite our urge to misunderstand them (PI §109), of delineating the actual use of words exactly as it is without either interfering with it or attempting to justify it (PI §124)). He took it that in many cases of philosophical confusion, our conflicted tendency to think that things must be a certain way whilst being unable to deny that they appeared for all the world to be otherwise might be alleviated if we were to ask ourselves how we actually used words in this vicinity; and such self-interrogation is very often facilitated by conceiving of our life with words in terms of language-games possessed of a grammatical structure and embodied in a specific form of life.

But if – like any other representational conventions – this set of signature concepts is sufficiently substantial or robust to acquire a life of its own, then they might on occasions stand between us and an ability simply to acknowledge how things really are; rather than helping to subvert our tendency towards the imposition of a philosophical “must”, they may actually subserve its further expression. And when a Wittgensteinian philosopher becomes so committed to the use of these signature concepts that he cannot conceive of another way of perspicuously representing the phenomena of our life with language when responding to a philosophical problem, then he has in effect imposed a set of philosophical preconditions on the reality he putatively aspires simply to describe. He has donned a set of Wittgensteinian conceptual spectacles; and by employing those concepts as lenses through which he views everything, he actively subverts the realistic spirit in which their creator forged and (at least attempted to) deploy them.

One might think that, if Wittgenstein’s signature concepts really are in the service of the realistic spirit of his philosophical enterprise, then there can be no tension between a commitment to that spirit and an equally resolute commitment to the deployment of those concepts.² After all, if they were forged precisely in order to facilitate the (re-)direction of our attention to the ways in which we actually use words in our lives, then they ought to possess the inherent flexibility and open-endedness needed to accommodate any pattern of word use that human beings have found and might conceivably find worth employing. So, if one really takes on board the idea that the whole point or purpose of Wittgenstein’s talk of “language-games”, “grammatical investigations” and “forms of life” is solely and simply to return us to our actual life with language, then one will recognize that conducting a grammatical investigation of a word by locating the language-game that constitutes its original home in our form of life just means attending to how that word is actually used. If the former way of describing the matter even threatens to suggest that the way in which that word is actually used must conform to certain prior conceptions (however minimal) of what legitimate word use is like, then its import – and so the true significance of the concepts employed to convey it – has simply been misunderstood.

In my view, this is not an objection to my way of stating the problem, but rather another way of articulating it. For where I invoked a willingness to sacrifice any Wittgensteinian signature concept whose substance (however minimal) might, in certain extreme or unusual contexts, limit our capacity to apprehend the full range of our uses of words in all its rich variety, this putatively alternative interpretation prefers to talk of a willingness to reshape or recast our uses of those signature concepts in each new context of meaningful word-use in such a way that they always subserve a perspicuous representation of its reality. But what, in the end, is the substantial difference between being willing to sacrifice any element of what we hitherto took to be essential to a signature concept in order that it might represent our life with language as it really is, and being willing to sacrifice the concept in order to represent our life with language as it really is? According to both interpretations, on those rare occasions in which the putatively independent substance of Wittgenstein’s signature representational conventions appears to threaten the realistic spirit of his enterprise, faithfulness to that spirit requires a willingness to put those conventions in question – whether by radically reshaping them or by discarding them altogether. And to this extent, both interpretations converge on the adoption of a distinctively modernist relation to the challenge of inheriting Wittgenstein’s realistic spirit in philosophy.

2. Spirit vs. signature

Over many years, Cora Diamond has demonstrated an unerring eye for, and a deep interest in, a range of linguistic phenomena that not only clarify the proposal that inheriting Wittgenstein’s work is primarily a matter of inheriting its realistic spirit, but also bring out the very real difficulty of doing so whilst continuing to employ his signature concepts. I want to focus on three such phenomena, which in effect subject the projectibility of those concepts to extreme stress-testing, and thereby raise the question of whether a real commitment to the realistic spirit might actually require a willingness to sacrifice not only certain prejudicial interpretations of its familiar conceptual or discursive embodiment, but that embodiment itself.

1: In her essay “Riddles and Anselm’s Riddle” (in RS), Diamond contests Norman Malcolm’s famous reading of Anselm’s ontological argument by exploiting Wittgenstein’s comparison of a mathematical conjecture that lacks a proof to a riddle for which we have not found a solution. The task of seeking a proof of the conjecture is given such orientation as it has, and so the conjecture has whatever meaning we may wish to say that it has for us prior to the proof’s construction, on the basis of our familiarity with other mathematical concepts and procedures on analogy with which the conjecture has been constructed. We play at using the “phrase” as an assumption, establish further conditions on that to which it may be held to apply, and then judge whether we are willing to accept that anything could meet all those conditions – whether the promise of a necessary connection articulated by the “conditions” we “established” might be fulfilled.

Diamond suggests that Anselm’s ontological argument is a working out of just such promissory connections. The riddle-phrase “that than which nothing greater can be conceived” (hereafter TTWNGCBC) is itself constructed on the basis of a familiar model (great, greater, greatest, greatest conceivable); and Anselm draws upon existing linguistic connections between lacking something, being limited, being dependent, coming into existence and having a beginning to establish that if we were to call anything TTWNGCBC, then it would be something that had no beginning. It is not that, on the basis of our understanding of TTWNGCBC, we know that it has no beginning, as if we were simply reminding ourselves of a language-game we know how to play; we are forging the outer shell of a necessary connection in a language we do not yet know how to speak.

Anselm’s emphasis on the difference between existence in the understanding and existence in reality can then be seen as a misleading way of distinguishing between ideas that we can, and those that we cannot, conceive of being the result of human inventive capacities. He wants to stress that our conception of what is possible might itself be shown up by reality, that reality might show us not only that something is the case that we imagined was not the case, but that something beyond what we had ever taken to be possible, something beyond anything we could imagine as possible, was actual. If so, then TTWNGCBC could not possibly be identified with anything we can imagine to be a product of the human imagination; for we can think of something greater than that – namely something that could not conceivably have been conceived by us, something in the light of which the products of our religious imagination appear as a queer collection of bloodless abstractions, sentimental projections and so on, something that reveals a logical space where none had seemed to exist.

But if anything we were willing to count as TTWNGCBC must be something that we cannot imagine having merely imagined, it must also be such that, if we were ever to encounter it, we could not imagine it never having existed. For if we could, then we could separate the idea of it from its actuality, could make sense of the possibility of making sense of it as a mere possibility to which nothing actual happened to correspond; and if so, then we could conceive of something greater than it – namely, something whose actuality is a condition for the possibility of conceiving it, something without which it is inconceivable that we could possess a language of any kind for it. Hence, anything we were willing to count as TTWNGCBC would have to be something whose non-existence could not be conceived, whose conceivability is itself conceivable only on condition of its actuality.

With ordinary riddles, and mathematical proofs, it is only when we discover that there is a solution to the riddle, and how it counts as a solution to that riddle, that we fully understand the question the riddle poses; before this, the relevant phrases or propositions have only promissory meaning. But with TTWNGCBC, Anselm has “established” that every statement we can make about it has, and can only have, a promissory meaning; the full transparency of that language to us is ruled out, because if it were to have a meaning we could fully grasp now, then we could conceive of something greater than whatever those words describe (namely, something whose nature exceeds the grasp of any concepts of which we can even conceive). And of course that form of words (“something whose nature exceeds the grasp of any concept of which we can even conceive”) is no more fully transparent to us than any other form of words to which it is “grammatically” linked, via the outer shell of a “necessary connection”. All are “allusions” to a “language” we cannot even conceive of speaking before actually finding ourselves in a position to speak it – a language given to us by the being to whom it applies, and whose revelation of himself will effect the radical conversion of all our existing concepts of him.

Accordingly, in the sense in which Wittgenstein normally claims that words have a grammar, these words do not; they are grammatically distinctive in that that they have no grammar, but only a “grammar”. On Diamond’s view, that is what a close attention to the way we employ such words will reveal; but if we instead assume that any meaningful use of words must have a substantial Wittgensteinian grammar, we prevent ourselves from acknowledging the reality of this way of employing words. She is not denying that we do talk of God in the context of honest, transparent language-games, whose grammar tells us what kind of thing is being spoken of; but she is claiming that whatever we are talking about in such games is not a possible solution to the riddle posed by the phrase TTWNGCBC – for that is a form of words that stands in need of a determination of meaning, and one which must come not from us but from whatever it turns out to apply to (which is why Diamond talks of Anselm’s riddle-phrase as embodying a great riddle, since ordinary riddle phrases are given meaning by us, insofar as we can find a way of meaning them). Moreover, if whatever TTWNGCBC turns out to apply to is capable of revealing everything we have hitherto imagined of God to be utterly misplaced, then part of what TTWNGCBC might reveal is the utter inadequacy of our present religious language-games, and the forms of religious life in which they are embodied. Anyone willing to recognize that possibility could hardly think (as Malcolm does) that people who regard the God of the Old Testament as a genocidal maniac are conceptually or philosophically confused simply because language-games are played in which saying such things of God is ungrammatical.

Does Diamond think that Malcolm’s error here simply one of operating with an unduly fixed or fixated notion of “grammar”? If so, one might have expected her to say that the phrase TTWNGCBC does indeed have a grammar, just not of the kind that Malcolm is inflexibly prepared to allow for; but instead, she persistently uses that signature concept in quotation marks whenever she applies it to TTWNGCBC, as if she is equally unwilling simply to apply it in this context or flatly to deny it any application. Her quotation marks rather imply that the projection of this concept into this context ought to be experienced as far from straightforward; if one doesn’t hesitate about whether or not to use it – if this aspect of our life with words doesn’t force one to question the usefulness of this concept in this context – then one simply has not got the linguistic phenomenon at issue fully in focus.

2: In her essay “Wittgenstein and Religious Belief: the Gulfs Between Us” (Diamond 2005), Diamond offers an innovative interpretation of Wittgenstein’s obscure remarks about the indispensability of a picture or pictures in the life of a religious believer. She links the centrality of the belief that God has a name in a certain tradition of Judaeo-Christian thought to the particular significance we attach to the individuality of those human beings we love or care for, our sense that their significance is unique and irreplaceable, and so not capturable in any general terms (say, by reference to their distinguishing characteristics). This conception, she points out, might include acknowledgement of the ways in which an encounter with a particular person might transform our concepts – as George Eliot, that magnificently ugly woman, gave a totally transformed meaning to the term “beauty” for Henry James. Eliot shows the concept up, moving James to use it almost as a new word, certainly as a renewed one. And Diamond suggests that part of the importance of the idea of God as having a name lies in the sense that his self-revelation similarly reorients our concept of “divinity”.

The Judaeo-Christian God is viewed as having revealed himself, in deed and word (those of his prophets and, in the Christian case, those of Christ himself), and as having thereby made it possible for his hearers to speak and act in response to this unprecedented self-revelation; it is only in the terms made newly available through God’s actions in history that the hearer can understand the kind of conduct truly expected of her. God himself has thereby given a transformed content to the word “God”: Diamond, following Rosenzweig, calls this the conversion of our concepts through God’s self-revelation. Think of this as a picture of God as speaking to his people. In what circumstances would it be natural to say that this picture is indispensable to the religious believer concerned?

Imagine a Wittgensteinian philosopher who wants accurately to characterize this aspect of our believer’s life with words. He will note that she describes her language-game as one in which God speaks and is responded to; but he will naturally want to ask what, in her game, counts as God’s having spoken – what, one might say, the grammatical criteria are that she and her fellow-believers employ in playing this language-game. From the believer’s point of view, however, the very form of that question implies that we – in our ways of speaking – are the ones who ultimately determine what counts as God’s speaking; whereas it is essential to her understanding of the God of whom she speaks that her ways with religious words have a kind of openness to God’s actions, an openness which means that it is not for her (or anyone other than God himself) to lay down rules for what counts as God’s speaking. To do otherwise would mean arrogating to ourselves the authority to determine the limits of God’s capacity to reveal himself, rather than remaining open to the ineliminable possibility that His self-revelation might show up our current ways of talking about Him as utterly shallow or misconceived.

The picture of God’s speaking thus lies at the basis of all this person’s religious thought, because it is not only central to the religious language-games that she plays but also to the way she regards or relates to those language-games – namely, not as practices in which what counts as God’s speaking is ultimately subject to determination by our rules. She will therefore resist any philosophical description of her religious language that makes its deployment ultimately a matter of our modes of speaking rather than God’s. As a result, any Wittgensteinian commitment to eliciting the grammatical criteria for her ways of speaking, where those criteria are taken to be hers or her tradition’s to determine and employ, will precisely misrepresent that way of speaking: this signature Wittgensteinian concept will here prevent us from simply acknowledging the reality of what lies before us, unless one radically recasts an aspect of its normal grammar that one might hitherto have taken to be both essential to it and uncontroversial in its implications.

3: In her paper “The Difficulty of Reality and the Difficulty of Philosophy” (Diamond 2006 – hereafter DR), Diamond provides a range of examples of what she wants to call “the difficulty of reality”:

[T]he phenomena with which I’m concerned [are] experiences in which we take something in reality to be resistant to our thinking it, or possibly to be painful in its inexplicability, difficult in that way, or perhaps awesome and astonishing in its inexplicability. We take things so. And the things we take so may simply not, to others, present that kind of difficulty, of being hard or impossible or agonizing to get one’s mind around. (DR, 99)

Diamond’s primary examples of this phenomenon are a Ted Hughes poem (about the agonizing incomprehensibility of the destruction of young men’s lives in the First World War) and J.M. Coetzee’s fictional protagonist Elizabeth Costello (with her maddened, isolating perception of the moral barbarism of our treatment of non-human animals). She also cites Czeslaw Milosz’ talk of beauty (the architecture of a tree, the dawn chorus) as something that should not exist, for which there are no reasons for and indeed reasons against, but which nevertheless undoubtedly exists; and Ruth Kluger’s Holocaust memoir, in which a young woman’s act of encouraging a terrified child seems to her to be incomparable and inexplicable, although many people to whom she tells her tale wonder at her wonder, seeing nothing mysterious in the fact that some people are altruistic. Roy Holland’s conception of the miraculous as the occurrence of something which is at one and the same time empirically certain and conceptually impossible (Holland 1980), as with Christ’s inaugurating miracle at the marriage-feast at Cana, also fits the bill. For whereas many people (and not only philosophers) would rule out any such “conception” in advance, because it violates the very idea of a conceptual order in the absence of which the possibility of genuine thought will vanish, others may be willing to take seriously the possibility that one’s experience might force one to violate one’s idea of what a well-ordered concept must be.

A difficulty of reality, then, is an apparent resistance by reality to one’s ordinary modes of life, which include one’s ordinary modes of thinking and talking; to appreciate that kind of difficulty “is to appreciate oneself being shouldered out of how one thinks, how one is apparently supposed to think” (DR, 105). But Diamond can properly acknowledge such difficulties only by once again sacrificing one of the supposedly defining features of a distinctively Wittgensteinian approach to philosophy. For its business of returning words from their metaphysical to their everyday use (PI, 116) is usually glossed as a matter of rehousing words in the Heimat of ordinary language games. But properly to register the essential nature of a difficulty of reality asks us to acknowledge the capacity of reality to shoulder us out from our familiar language-games, to resist the distinctively human capacity to word the world, and thereby to leave us as bewildered and disorientated as a bird that suddenly finds itself incapable of constructing a nest, or a beaver of building a dam. Would it be at all helpful in clarifying this highly distinctive aspect of our relation with our words to say that being shouldered out of our language-games is just one more language-game, or to declare that words have a grammar when they fail us just as they do when we effortlessly employ them to word the world, or to describe these uncanny encounters as just another element in the homely forms of human life? Surely difficulties of reality ought rather to resist the grammar of “language-game”, “grammar” and “form of life” (however flexibly they are projected) just as radically as they resist that of any other aspects of our thinking and talking?

Taken one way, these are just three particular examples of Diamond’s general commitment to Wittgenstein’s realistic spirit rather than to his signature ways of embodying that spirit. The notions of “grammar”, of “rule-governed language games”, and of “criteria” certainly do help us to put the ordinary or the everyday undogmatically before ourselves in many contexts; but in others, they risk impoverishing or narrowing down our sense of what the ordinary or the everyday might be, and thereby risk betraying our most fundamental inheritance from Wittgenstein. And whenever they do, we are obliged to dispense with them – to find other ways of embodying or maintaining the realistic spirit of our enterprise. Taken another way, however, the subject-matters of these papers exhibit a certain kind of family resemblance – quite as if the nature of the philosophical error being diagnosed in each case and the nature of the issues whose treatment exemplifies that error are non-accidentally related.

To begin with, these papers have a distinctively ethico-religious field of reference. The first two are explicitly concerned with aspects of religious uses of words, and the third includes a religious inflection of the concept of “the miraculous” as a prominent illustration of the phenomenon with which it is concerned. Moreover, by referring to Anselm’s TTWNGCBC as a “great riddle”, Diamond connects this aspect of religious language with Wittgenstein’s invocation in the Tractatus of “the riddle of life in space and time” (TLP 6.4312), and so to his early conception of ethics; and a similar focus on ethical matters is equally evident in the “Difficulty of Reality” paper.

This distinctively ethico-religious field of reference also exhibits a certain thematic unity or inter-relatedness. One might say that in all three contexts, Diamond is attempting to demystify or detoxify the concept of nonsense – to suggest to those who are likely to be deeply attached to “nonsense” as the key Wittgensteinian term of criticism that forms of speaking that either lack or transcend or refuse ordinary assignments of sense are not thereby emptied of human meaning. On the contrary: great riddle phrases and words failing us in the face of recalcitrant experience are both techniques of language, part of the inconceivably rich human tapestry of our life with words, and ones whose usefulness for us precisely depends on the specific ways in which each resists translation into transparently grammatical speech-forms. One might also say that they are variously inflected realizations of an idea of reality as capable of outstripping our means of making sense, of exceeding our best conception of how to conceive of it. Great riddle phrases are the outer shells of a possible language-game, whose proper inhabitation we cannot even imagine imagining, except on the basis of the authoritative actualisation of its putative referent; the idea of a conversion of our concepts involves reality’s capacity to reveal the radical impoverishment or even the utter emptiness of words like “beauty” or “divinity”, whose meaning we previously took ourselves to have fully mastered; and encounters with difficulties of reality disclose reality as essentially resistant to our ways of thinking and talking about it, as at once inexplicable or incomprehensible and isolating (making us incomprehensible to others, and even to ourselves).

Putting field of reference and theme together, their link with the methodological error under examination is hard to miss. For if we think of inheriting Wittgenstein as a matter of inheriting his signature concepts, we would be sorely tempted to view those concepts as constituting a field of philosophical discourse that is guaranteed to accommodate anything that reality might throw at us – or at least to licence in advance the categorization of anything that does refuse accommodation as surd or void. To do so would thereby amount to refusing in advance to acknowledge reality’s capacity to exceed our conceptual grasp. A truly rigorous and thorough-going commitment to dispense with philosophical “musts” surely requires a willingness to dispense with any imposition of conditions on reality; and that would seem to include the condition that it make grammatical sense – that it be housed in transparent, grammatically-governed language-games (and so within the grasp of any given array of philosophically-forged conceptual tools, such as that of “a transparent, grammatically-governed language-game”). If, then (following Diamond’s understanding of the matter), we rather think of inheriting Wittgenstein’s philosophy as a matter of inheriting this realistic spirit, we would be committed to perspicuously representing the way reality presents itself in all its actual variety, including the various ways in which it refuses to submit itself to any given or preferred means of perspicuous representation.

So understood, a connection emerges between Wittgenstein’s later philosophical project and Diamond’s way of interpreting his early conception of ethics, of aesthetics and of philosophy, in her paper “Ethics, Imagination and the Method of Wittgenstein’s Tractatus” (Diamond 1991 – hereafter EIM). She there glosses a distinction the Tractatus draws between the happy and the unhappy man:

I [characterized] an attitude to the world in terms of acceptance of the independence of the world from one’s will... [T]he ethical spirit is tied to living in acceptance of the fact that what happens, happens, that one’s willing this rather than that is merely another thing that happens... a sort of piety in action, in life, is possible, that looks with clear eyes at the happenings of the world, at the happenings of the world being whatever they are. (EIM 62-3)

Call this acceptance of the world’s independence of one’s will the orientation of the happy man: that of the unhappy man Diamond articulates as a matter of his being dissatisfied at the world’s refusal to meet the conditions he lays down, to submit to his control. Here, she invokes Hawthorne’s story “The Birthmark” and the Grimm’s tale “The Fisherman and his Wife”: in the former, Aylmer’s discomfort at his new wife’s birthmark leads him to a course of action which results in the destruction of her goodness, her beauty and her life altogether; in the latter, the wife of the fisherman who finds the wish-granting flounder begins by asking for a better home and ends by expressing dissatisfaction at the sun’s and the moon’s rising independently of her will.

The realistic spirit in philosophy looks remarkably like a further expression of the happy man’s orientation to the world – of a willingness to acknowledge the world’s refusal to conform to any conditions one might lay down for it, hence to acknowledge reality’s independence of one’s will. And by the same token, our desire to impose such conditions, to inhabit the conviction that reality must be a certain way or be wholly graspable within a certain conceptual scheme or discursive field or mode of representation, appears to amount to an inflection of the unhappy man’s dissatisfaction with the world’s independence of his will. So understood, Diamond’s way of inheriting Wittgenstein’s later philosophical work as an inflection of modernist realism discloses an ethical dimension to it that is continuous with the ethico-religious spirit to which his early work similarly aspires (and which it likewise regards as inseparable from a certain aesthetic imperative – “ethics and aesthetics are one” [TLP, 6.421]).

But are we willing to accept the full burden of that interpretation of the later Wittgenstein? What is the aesthetico-ethico-religious spirit of the Tractatus?

3. Refusals of sense

According to Diamond, the Tractarian conception of nonsense is austere: there is no dividing nonsense-sentences into two categories, those that are mere gibberish and those whose gibberish points to an ineffable truth. Those in the grip of philosophical nonsense have simply given no meaning to the sentences they proffer, but are under the illusion that those sentences are not only meaningful but of deep significance; and the author of the Tractatus aspires to liberate them from that illusion by self-consciously inhabiting it with a view to leading them to recognize its emptiness from within. But if these philosophical sentences have no meaning, we can attempt to understand those who proffer them under the illusion that they do mean something only by entering imaginatively into their taking nonsense for sense; and doing so seems to require that the diagnostician maintain herself within the same illusory logical space as her interlocutor, aspiring to specify what the interlocutor takes herself to be thinking or wanting or saying, whilst being fully aware that there is no such specification to be given.

On Diamond’s account, understanding the sentences of the would-be engager in ethics runs along parallel lines in the Tractatus. The ethical sentences themselves are nonsense, but the utterer of them is to be understood by imaginatively entering into the point of view from which such nonsense can be taken for sense; and such ethical sentences further resemble philosophical ones in reflecting the attractiveness of the idea of a point of view on the world as a whole (as if from without or sideways on), whatever may happen in it. However, the author of the Tractatus has very different designs upon the utterers of ethical nonsense sentences:

The attractiveness of philosophical sentences will disappear through the kind of self-understanding that the book aims to lead to in philosophers; the attractiveness of ethical sentences will not. But if we understand ourselves, ourselves the utterers of ethical nonsense, we shall not come out with ethical sentences under the illusion that we are talking sense. (EIM 74)

Suppose someone points out to us that talk of an “attitude of satisfaction or dissatisfaction with the world as a whole” is modelled on talk of attitudes to something in particular, some way that things are within the world that might satisfy or dissatisfy us, but that no sense is assigned by such familiar ways of talking to an attitude of dissatisfaction to the world as such, however things may be within it – that that last phrase only appears to make sense. Then (according to Diamond’s Wittgenstein) we may, for example, resolve to frame any future ethical sentences we are inclined to utter, with words like “I am inclined to say... ”, but we would not lose our inclination to say them. Such framing would mark a certain gain in self-awareness, a liberation from unself-consciously taking nonsense for sense; but continuing to use the sentences so framed would also mark our continuing to feel that just these sentences (modelled on meaningful ones but unmoored from their patterns of use) express the sense we want to make – or rather, that our intentions in uttering them were essentially incompatible with making sense: “every account of what he means which would make it out to be sense he would reject ‘on the ground of its significance’; the nonsensicality of what is said belongs to the essence of the linguistic intention” (EIM 75).

What makes a sentence an ethical sentence is thus the intention with which it is uttered – an intention that (whether its user recognizes this or not) would be frustrated by the sentence’s making sense. To achieve clarity about this kind of attraction to a sentence means becoming self-conscious about that intention; but achieving such clarity is entirely consistent with resolutely continuing to use the sentence with that intention. Likewise, achieving clarity about what one is doing, philosophically speaking, when one attempts to understand someone who is inclined to take nonsense for sense in this particular way involves becoming aware that, since one is imaginatively inhabiting the point of view of the possessor of such intentions, one is drawing upon one’s own capacity to inhabit linguistic intentions that are essentially incompatible with making sense, and hence that successful attempts to embody the understanding one seeks will necessarily involve the use of nonsensical sentences. Accordingly, achieving that methodological self-consciousness is entirely consistent with resolutely continuing to use those sentences when attempting to impart philosophical clarity about ethical sentence-users.

Any understanding we may achieve of the Tractatus’s way of engaging with the would-be engager in ethics will therefore necessarily draw upon the same sources in us – the same array of impulses and intentions – that we draw upon when we either proffer or respond to ethical sentences directly (if we do). Understanding the Tractatus treatment of ethics means recognizing that its author requires exactly the kind of understanding that is required whenever one understands any would-be engager with ethics – that of imaginatively entering into the intentions of a resolute (and in this case, self-aware) user of nonsense. To think that such an author’s intentions might be accurately articulated in intelligible or senseful sentences of commentary would thus betray a failure to understand him; or as Diamond puts it, “the book’s ethical intention includes the intention of the book not to be interpreted” (EIM 86). I am inclined to say: If Diamond’s interpretation of the Tractatus treatment of ethics made sense, we would have to reject it (on the grounds of its significance alone).

But recall: on a resolute reading of the Tractatus, for a sentence to be nonsense is simply for it to lack sense – for (at least one) constituent sign in that sentence to have no meaning assigned to it. Hence, to describe the intention of the would-be engager in ethics as necessarily frustrated by any intelligible sentence is simply to say that she refuses to accept any intelligible candidate articulation of her intention as an articulation of that intention. No such assignment of meaning to her utterance will satisfy her – not because she wants to assign it an ineffable meaning, and not because she wants to assign different meanings to it simultaneously, but because she finds satisfaction precisely in refusing to find any available assignments of meaning satisfying. Hence, to understand her requires understanding why she might find intelligibility (to others and to herself) essentially unsatisfying.

Diamond’s earlier paper on Anselm presents a helpfully analogous context in which satisfaction resides in the refusal of sense. “TTWNGCBC” is seen as embodying a great riddle precisely because it resists assimilation into any existing, or imaginable, transparent language-games. That one so regards it comes out in one’s refusal to accept that anything we can talk about in a language with a grammar (in the familiar Wittgensteinian sense of that term) could possibly be an answer to the question we’re asking; if we could even conceive of explaining its grammar, it wouldn’t be the phrase that those who use it want. And the point of that refusal is to mark a contrast between two conceptions of what it is to believe in God – one, according to which one can display what it is to have faith entirely in terms of the way a variety of religious language-games are played, and another according to which confining our attention to any such array would ensure that we entirely missed the point of talk about God.

In the Tractatus context, of course, the specific kind of meaning-assignment to ethical utterances that is being resisted is that characteristic of fact-stating, empirical discourse in general, and of empirical psychological discourse in particular. Hence, the absolute or unconditional nature of the refusal indicates a sense of absolute discontinuity between the ethical and the empirical world: “That which I take myself to see in myself or another if I think of that person as having a [good or] evil will – that thought of mine about a person – has no room in the sphere of thoughts about the world of empirical facts. Put there it is not about what I wanted it to be about” (EIM 85-6).

Such a refusal is not an objection to the very idea of intelligible evaluative uses of language. Evaluating strawberries or sewage effluent can be understood perfectly well without requiring imaginative participation in taking nonsense for sense; so can the use of language to alter people’s feelings and attitudes, or to express adherence to prescriptive principles, or to guide action. But for just that reason, such uses fail to capture what Wittgenstein means by “ethics”; that meaning is precisely given expression by his refusal of any such ways of assigning evaluative sense to his ethical sentences (and so would be obliterated by any philosophical account – whether Wittgensteinian or not – that took it for granted that any meaningful ethical sentences must fit within such familiar, transparent modes of language use). The point of his refusal is thus to draw a sharp contrast between two kinds of evil (and hence two kinds of good): “evil [that] is... inconsequential..., something close to home... something [not] very bad to which one might become accustomed, and [evil as] something terrible, black, and wholly alien that you cannot even get near”, as the brothers Grimm characterize the ethical vision of their fairy tales.

Take Rumpelstiltskin: the boastful miller and the greedy king are not at all nice decent folk, “but their badness is not connected by the tale with our capacity to respond to evil as unapproachable and terrible, as Rumpelstilzschen’s evil is” (EIM 87) – the evil of wanting not only to abduct another’s child, but (as his gleeful, dancing fireside song connecting the child’s arrival with cooking implies) to devour it. Little wonder the tale assigns a very different fate to the miller and the king than it does to Rumpelstiltskin, who tears himself in two with self-consuming rage. Sensible, well-meaning commentators have argued that Rumpelstiltskin is as much a helper as a villain in comparison with the girl’s father and husband, even that he is the traumatized victim of loneliness – hence someone whose immorality is of an everyday mediocre kind, and explicable in terms of familiar psychological syndromes. This is bringing Rumpelstiltskin’s evil into the domain of the empirical world with a vengeance; and it thereby evades the very contrast that the tale is designed to register. Refusing to draw such distinctions may appear as realism, expressive of down-to-earth rational disdain for mystery and mysticism; but in truth it simply obliterates their potential ethical significance, and thereby violates the realistic spirit. And the early Wittgenstein’s removal of thought and talk about the good and evil will from empirical talk as such is essentially another way of marking that contrast, one of a number of possible techniques of language through which it might be indicated and maintained. This, one might say, is an aspect of the aesthetic achievement of the Tractatus – call it the beauty of its willingness to cleave to words whose rightness consists in their emptying themselves of sense before our very eyes; and it is an achievement that is essentially continuous with the vision the text thereby advances of the internal relation between ethics and aesthetics (transcendentally considered).

But it is not just that the literary techniques of the Tractatus, on the one hand, and those of the fairy tales and short stories beloved of its author on the other, are different ways of marking a crucial articulation of one and the same ethical vision. For Diamond, the “purely” literary versions of these marking-off techniques work in highly specific ways to which analogies can be found in Wittgenstein’s philosophical writing both early and late (and which take centre stage in the other paper of hers we cited earlier). The representational techniques of “The Fisherman and His Wife”, for example, have three critical or criterial features:

i. the kind of evil Grimm means; ii. its presence, as something of which one may have a sense already in what is on the surface perfectly matter-of-fact; and iii. its explicit connection ultimately with the wife’s resentment that things in the world go as they do independently of her will (EIM 82)

The second feature on this list is particularly interesting – the sense Diamond has that one may intuit the wife’s cosmic dissatisfaction even in her initial desire to have a cosy little house rather than a privy (something the tale marks by noting that the sea is already faintly discoloured and mildly turbulent when the fisherman brings his wife’s first wish to the flounder’s attention, as if the world-annihilating storm she eventually unleashes is already gathering its energies). It is internal to these thoughts about terrible evil that they “seem to be justified by nothing that is as it were available on the surface of events. We have a sense of something dark and terrible ‘within’, as we might say” (EIM 83). This notion of the inner as discontinuous thus embodies the discontinuity between ordinary badness and terrible evil, the latter’s essential unrelatedness to everyday moral and psychological understanding; but it also connects it with a certain kind of resistance to our understanding – its blackness not only a marker of its depth, but also of its utter unapproachability.

Diamond’s conception of Tractarian, transcendental good and evil thus looks remarkably like an anticipatory example of what she will later come to call “a difficulty of reality” – a phenomenon which shoulders us out of our everyday ways of comprehending the world and what happens within it, presenting a traumatizing or agonizing resistance to our understanding. For an integral part of what is bewildering about such difficulties is precisely that what we see as incomprehensible is seen by others as utterly everyday – just as we can easily imagine readers of the Grimm tales who sense no cosmic evil in the initial responses of the fisherman’s wife (and are correspondingly more likely to want to explain away or dismiss the Grimms’ ways of connecting those initial responses to transcendental good and evil). Difficulties of reality thereby serve to isolate individuals, disclosing others as opaque to them and themselves as opaque to those others; reality’s resistance to our understanding reveals us as essentially resistant to one another’s understanding – just as those to whom ethics is a great riddle will seem incomprehensible to those who find nothing within them (no impulses, intentions or convictions) with which such deliberately nonsensical formulations might resonate.

And it is not as if those impulses, intentions and convictions are entirely transparent to their possessors. On the contrary: Coetzee’s Costello is periodically possessed by the conviction that she is going mad, becoming incomprehensible to herself as well as to others, incapable of understanding why these issues drive her to the brink of insanity. Likewise, those who recognize unapproachably terrible evil in Aylmer or the Fisherman’s Wife can do so only because they are capable of imaginatively entering into the perspective of someone they conceive of as dissatisfied by the world’s failure to submit to his will, which amounts to attempting to inhabit one kind of attachment to nonsense (even if the kind which survives its revelation as nonsense). And in so doing, as Diamond puts it, “I enter imaginatively into the seeing of it as sense, I as it were become the person who thinks he thinks it. I treat that person’s nonsense in imagination as if I took it to be an intelligible sentence of a language I understand, something I find in myself the possibility of meaning” (EIM 81). So to intuit terrible evil in the Fisherman’s Wife or Rumpelstiltskin depends upon finding in myself the kinds of impulses and intentions that find expression in the nonsense that articulates that kind of evil, and that I imagine another saying in her heart insofar as I do understand her. To identify incomprehensible evil in another thus depends upon a willingness to acknowledge a similarly incomprehensible possibility in myself (as Diamond notes, this is central to Wittgenstein’s later treatment of Frazer and the baleful rituals with which he was so preoccupied).

What, then, of the transcendental good will? In order to intuit its presence in another, which presumably means being compelled to characterize that other in terms of a kind of piety in action, an ability to look with a clear eye at the world’s vicissitudes and to acknowledge unconditionally its independence from his will, one necessarily resorts to nonsense phrases, and so registers a kind of resistance to the understanding in such goodness. But that resistance to sense surely goes further – it also involves a perception of the miraculousness of such goodness, the sheer incomprehensibility of its realization in the world, the utter inexplicability of such radical self-abnegation in terms of our best naturalistic patterns of moral and psychological explanation. And if one can imaginatively enter into the perspective of such a good will in another, that will be because one is able to acknowledge similar impulses and intentions in oneself – because one is willing to relate to oneself as inexplicably but undeniably capable of goodness beyond virtue as well as evil beyond vice (as Rai Gaita might put it).

4. “The sea was dark green with shades of yellow, and not nearly as calm as before...”

But is it really credible to regard the realistic spirit with which Wittgenstein works in the Investigations as essentially a further expression of such a transcendental conception of ethical value, as one way in which the orientation of the happy man towards the world as such might find expression (just as that of the unhappy man might find expression in resisting it)? The suggestion might seem less outrageous if we imagine reaching it in two steps.

The first is to recall that part of what the realistic spirit – with its absolute or unconditional refusal to impose conditions on reality – requires is a willingness to acknowledge such phenomena as great riddles and difficulties of reality, and so a willingness to suffer a kind of philosophical wound – to be shouldered out of our most familiar ways of making philosophical sense of our rich and varied ways of thinking and speaking. For doing so involves drawing on a capacity imaginatively to enter into these ways in which we might intelligibly refuse assignments of meaning or sense to our words, which means finding within ourselves the possibility of (as it were) becoming the person who finds these kinds of satisfaction in nonsense. But to find it possible to mean such refusals just is to discover that we are the kind of person who can find such satisfaction in nonsense; so realizing the realistic spirit in philosophy in these domains is only possible for those who can find in themselves the capacity to make sense of such phenomena as the transcendental conception of ethics that pervades the Tractatus.

Of course, even this initial, local connection between method and vision is contestable. For anyone incapable of making sense of such linguistic techniques will for that reason be incapable of seeing Diamond’s willingness to look with a clear eye even on the kinds of phenomena that manifest it as anything other than an incomprehensible attraction to mysticism and mystery-mongering. The very phrase “aspiring to an absolutely unconditional refusal to impose conditions on reality” is after all, itself nonsense, being constructed on the model of refusals that are context-specific, hence conditioned and relative; so if we can’t make any sense of these difficult realities of our life with words, then the philosophical practice of acknowledging them with a clear eye will inherit exactly the same riddling resistance to intelligibility that the phenomena themselves exemplify. Choosing to acknowledge rather than to deny such phenomena in philosophy is thus as capable of isolating us from one another, rendering us mutually incomprehensible, as are the phenomena themselves.

Suppose, however, that we do take that first, already isolating step; the idea of a more general connection between Wittgenstein’s later method whatever its subject-matter – the quiet, apparently trivial, weighing of linguistic facts – and such cosmic or transcendental ethical visions might still seem absurdly over-reaching. But consider: it may be that it is precisely when one’s commitment to a methodological orientation is under most pressure, when its realization requires of us the most extreme and demanding forms of self-denial, that its underlying general significance is most likely to declare itself – just as the extremity of the fisherman’s wife’s concluding demand makes manifest the spirit of her initial, apparently everyday, wish. So understood, the point of asserting a general connection between method and vision need not be to imply that the philosophical imposition of conceptual preconditions on our life with words is always and everywhere a manifestation of evil as black as that of Rumpelstiltskin’s heart; it may rather amount to sensing in any such stance the faintest turbulent discolorations of a gathering storm, and to intuiting a trace of something miraculously graceful in even the most modest and local resistance to such impositions. But of course, it is internal to something’s harbouring a difficulty of reality that where one person sees a matter of fathomless significance, another sees only banality.

Biographical note

Stephen Mulhall is Professor of Philosophy and a Fellow of New College, Oxford. He previously held positions at All Souls College, Oxford and the University of Essex. His main research interests include Wittgenstein, Heidegger, Sartre and Nietzsche; ethics and the philosophy of religion; and the relation between philosophy and the arts (especially film and literature). Recent publications include Wittgenstein’s Private Language: Grammar, Nonsense, and Imagination in Philosophical Investigations§§ 243-315 (2006),The Conversation of Humanity (2007) and The Wounded Animal: J.M. Coetzee and the Difficulty of Reality (2009).