1. Top of page
  2. Abstract
  3. Carter's Complaint
  4. Criteria and Internal Relations
  5. The Methodology of Wittgensteinian Ethics
  6. The Inwardness of Concepts
  7. The Realm of Meaning
  8. Authority and Relativism
  9. Conclusion

Carter argues that Wittgensteinian moral philosophy – typified by the work of Raimond Gaita and Christopher Cordner – rests on shaky foundations because it vacillates between grounding moral judgements in grammar and in a form of life. In this article, I respond to Carter's criticism. I defend Wittgensteinian moral philosophy by showing that Gaita and Cordner specifically repudiate the purported dichotomy between grammar and a form of life. I then go on to explain why Wittgensteinian moral philosophers are right not to try to ground moral judgements in features of a shared form of life.

Carter's Complaint

  1. Top of page
  2. Abstract
  3. Carter's Complaint
  4. Criteria and Internal Relations
  5. The Methodology of Wittgensteinian Ethics
  6. The Inwardness of Concepts
  7. The Realm of Meaning
  8. Authority and Relativism
  9. Conclusion

Carter worries about the stability of “the footing of Wittgensteinian moral philosophy.” 1 His paper takes the form of a dilemma. Wittgensteinian moral philosophers think either that all moral judgements are founded on a particular conceptual scheme (a “grammar,” in Wittgenstein's terms), or else – the alternative which he favours – that moral judgements are grounded upon observable features of human nature (what Carter calls a “form of life”). 2 The purported instability in the writings of Gaita, Cordner et al. comes from their supposed vacillation between these alternatives. In what follows, I suggest that Carter's worry is based on a misunderstanding. Far from vacillating between treating moral claims as conceptually or empirically grounded, Gaita et al. specifically repudiate this dichotomy. I then go on to explain why Wittgensteinian moral philosophers are right not to try to ground moral judgements in features of a shared form of life.

Carter says that, for Gaita, “a moral claim is also, always and importantly a conceptual claim.” 3 In elaboration and defence of this contention, he distinguishes between judgements that are secured by grammar and those that are secured by a form of life. Grammar is to be understood as “an entire way of speaking, responding and valuing,” 4 whereas “our form of life is made up of those creaturely conditions we share that shape our grammar but are not, in turn, shaped by it.” 5 Gaita is supposed to endorse this distinction and to urge that moral claims (as conceptual) are secured by our grammar. However, this seems to set him on a slide towards relativism or subjectivism. After all, different individuals have different ways of “living, speaking and going on,” and will therefore diverge in those moral judgements that are, for them, well grounded. As Carter puts it:

With what authority do we foreclose on certain possibilities? Who determines what places a thing can have in human life? Even if we “look and see,” whose determination do we accept and why? 6

In order to avoid being stuck with intolerable and unanswerable questions such as these, Carter thinks that we should see morality as grounded in basic facts about our creaturely condition. Such facts, determined presumably by empirical investigation (when broadly understood so as to include anthropology, etc.), form a base that is common to all different grammatical practices. Such “basic facts” will, thus, provide us with a shared perspective from which to evaluate and criticise different ways of going on.

In order to substantiate this, Carter distinguishes the following claims made by Gaita:

  • (1) 
    One cannot love evil.
  • (2) 
    One cannot love cow dung.
  • (3) 
    One cannot do evil without incurring great loss.

He then says:

Gaita would claim that, like (1), (3) is guaranteed grammatically; that is, its truth is internal to a whole way of speaking, responding and valuing. But if there are basic facts, like (2), then precisely why is (3) not one of them? If (1) is in fact guaranteed by our form of life, then might (3) and, indeed, all the rest of morality be? In this case, morality would no longer lie in a particular grammar (or normativity) but in a grammar that is at least as extensive as our form of life. 7

In what follows, I argue that Carter's conclusion is misplaced; there are no “basic facts” in his sense, and therefore there are no moral “basic facts.” I will argue that Wittgensteinian ethics specifically repudiates a difference in kind between “grammar” and “form of life.” 8 I will then go on to suggest that we should not think of (1)–(3) as “secured by” either grammar or our form of life. 9 In the final section (VI), I briefly consider what to make of the charge that Wittgensteinian ethics is relativistic.

Before we can get there, however, some preliminary argument is required. In the next section, I discuss the role that appeals to criteria play in Wittgenstein's philosophy. I argue that the instability of criteria gives us good reason to think that Wittgensteinian moral philosophers do not intend them to play a foundational role in moral philosophy. In section III, I present four possible positions vis-à-vis the relation between moral concepts and their social context. In section IV, I show how Winch came, in the course of his thinking, to reject all four positions. Section V demonstrates how Gaita draws on Winch's work to provide an account of morality that repudiates the grammatical/empirical distinction.

Criteria and Internal Relations

  1. Top of page
  2. Abstract
  3. Carter's Complaint
  4. Criteria and Internal Relations
  5. The Methodology of Wittgensteinian Ethics
  6. The Inwardness of Concepts
  7. The Realm of Meaning
  8. Authority and Relativism
  9. Conclusion

Carter correctly observes the importance of the locution “X is part of the very concept of Y” in Wittgensteinian philosophy. In Wittgenstein's own terminology, this began as the concept of an “internal relation.” 10 Take Winch's example:

To say that red is darker than pink is to point to an “internal” relation between those two characters; to say that one piece of material is darker than another, however, is to indicate an “external” relation. 11

Thus, internal relationships “spring from, or even help constitute, what the related terms essentially are.” 12 For Wittgenstein, the terminology of the “internal” gave way to that of the difference between what is symptomatic and what criterial of a thing. 13 The presence of a symptom of an object (or condition) is consistent with the object's being absent, and similarly the absence of the symptom is consistent with the presence of the object. On the other hand, the presence of a criterion of y is definitive of the presence of y; it is “that without which not,” and vice versa. As with internal relations, “x” and “y” in “x is a criterion of y” can refer either to a relation between two concepts (“having an unpleasant feeling is a criterion of being in pain”), or else between two concrete particulars (“his unpleasant feeling is a criterion of his being in pain”).

The first use of the notion of a criterion in a moral context can be traced to Peter Winch:

When I'm horrified at the way someone is behaving, at his cruelty to another person, say, I may … ask “Don't you understand what you're doing?” That is, I may take his indifference to what he is doing as itself a criterion for his not understanding the nature of what he is doing. 14

But perhaps its most striking use came in Cora Diamond's article “Eating Meat and Eating People”:

We learn what a human being is in – among other ways – sitting at a table where we eat [animals]. 15

In the same article she says:

[T]here are some actions… that are part of the way we come to understand and indicate our recognition of what kind it is with which we are concerned. […] [I]t is not “morally wrong” to eat our pets; people who ate their pets would not have pets in the same sense of the term.” 16

Diamond and Winch are both emphasising the connection between certain concepts, judgements that employ those concepts and the social context in which they occur. We make sense of other people's behaviour by seeing how it may follow from, or be an expression of, their beliefs. At the same time, we come to understand those beliefs by placing them in their behavioural context. Radical divergences from the norm in an individual's attitudes or behaviour may betoken confusion rather than voluntarily chosen deviance, hence the puzzlement which is a natural reaction to callousness or evil. It is that puzzlement which Diamond gestures at when she says that it is not morally wrong to eat one's pet. 17 In these cases, a certain behavioural response is held to be criterial of our recognition of the object (viz., a human being, or an animal) by which we are confronted. To fail to respond in that way is, in some cases at least, to have failed properly to see – or fully to have understood – the nature of the situation. 18 And, more generally, to fail to see why this kind of response is called for in circumstances such as these is to lack the concepts (of the human, the animal, a pet and so forth) that we have. Adapting a phrase from Cavell, someone who could eat their pet with a blasé demeanour “do[es] not live in our world.” 19

I suggest that we might understand such conceptual appeals as having the same status as criteria in Wittgenstein's philosophy. Adopting Cavell's reading of the Investigations, we may say that references to criteria are “everywhere controlled by … a response to the threat of scepticism.” 20 However, they are not designed to show scepticism to be false. Rather, “the fate of criteria, or their limitation, reveals … the truth of skepticism.” 21

What truth is that? It is, in part 22:

the astonishing fact of the astonishing extent to which we do agree in judgment; eliciting criteria goes to show therefore that our judgments are public, that is, shared. What makes this astonishing … is that the extent of agreement is so intimate and pervasive … and that since we cannot assume that the words we are given have their meaning by nature, we are led to assume they take it from convention; and yet no current idea of “convention” could seem to do the work that words do – there would have to be, we could say, too many conventions in play, one for each shade of each word in each context. 23

Applying that to moral cases, we get the thought that appeals to what we say or do are attempts to draw attention to the ways in which, despite the manifold differences between us in inclination, outlook and upbringing, we do agree in our reactions to circumstances. 24 This agreement, as Cavell suggests, is grounded neither on the natural properties of those things to which we respond nor on conventions that settle, in advance, what responses are called for.

Transposing that to the moral case, we may say: there is no property shared by all (and only) those animals which we make our pets that warrants our treating them as we do. Equally, there is no property shared by all (and only) those members of the class homo sapiens in virtue of which they are owed respect or dignity. On the other hand, there is no convention that settles what animals are to count as pets, and to determine the limits of appropriate and inappropriate behaviour towards them. Similarly, no convention underwrites the compassionate response of a passer-by to the suffering of a stranger, which would settle the conditions in which responses of this sort are called for. 25

As Cavell suggests, we can see the inability of conventions to determine the limits of our practices by looking at the inability of criteria to set necessary and sufficient conditions for the use of an expression. 26 Take the case of the eating of one's pet. We cannot say, flatly, that eating the corpse of an animal is never consistent with its being one's pet. Not only might one do so in times of great need, but also it is conceivable that, in certain contexts and when done with a particular demeanour, eating the body of one's pet might be a means of expressing grief, or respect or love. 27 Similarly, in certain contexts, one might leave a wounded person (perhaps even leaving them to die) without therefore being indifferent to the reality of their situation. 28

Note, however, that in imagining circumstances in which we might (without contradiction in terms) eat our pets, appeal was made to an unspecified “special context” and “particular demeanour,” which would render it consistent with our understanding of what it means for a creature to be a pet. This secures consistency only through making the behaviour or the circumstances exceptional. We might say, then, that it is impossible to eat the body of one's pet in just the same way as one would eat the flesh of an animal bought from a market (or bought, pre-packaged, from a supermarket). But what counts as eating one's pet in the same way as one eats a supermarket chicken? What kind of behaviour counts as suitably different, such that it manifests respect or grief?

There are two reasons why we should not expect a resolution to this question to come in the form of an appeal to conventional agreement. First of all, as Cavell suggests, to have conventions play that role, we would have to cut them hopelessly fine – to have one convention for each potentially novel use of an expression, each novel context. 29 Not only do we not have conventions of this sort, but such conventions would be either hopelessly arbitrary or else so numerous as to defeat the purpose for which they were introduced. 30

Second, any difference that could be settled by reference to convention would not guarantee its meaningfulness in the sense in which we are interested. Thus, it might be stipulated that the eating of pets is appropriate so long as certain rituals are observed. (Say, that special utensils are used in the preparation of the flesh, or special phrases spoken before one eats.) But so long as those conventions are treated simply as barriers to be navigated, or niceties to be observed (in other words, so long as they command no real respect or no genuine participation), then it can fairly be objected that they make no real difference to one's relationship towards one's pet. Simply acting in conformity with a convention is not enough to guarantee (or undermine) the meaningfulness of one's behaviour. 31

Therefore, appeals to criteria cannot settle, ahead of the fact, what responses are or are not appropriate in situation types (such as when confronted by a wounded person on a roadside). Because of this, they cannot determine when the question “but what should I do?” (as it may appear in, say, “I know this is the body of my pet, but what should I do with it?”) can be legitimately raised, and if it is legitimately raised they provide no guarantee of a satisfactory answer. To the response “you ought to bury it, because this is how we treat our pets,” it is always open to the individual to demur: “but why ought I do what we ordinarily do?”

The upshot of this is that we should be suspicious of Carter's notion of moral claims [such as (1) and (3) above] as “secured by grammar.” Given this, one may be tempted to recoil to a position on which appeals to ordinary practice have no relevance to morality whatsoever – that we ought to dispense entirely with references to the parochial, to what we are inclined to do and say, when determining how we ought to act. Dealing with this position – and, more broadly, placing the relevance of appeals to practice in Wittgensteinian approaches to ethics – is the concern of the next two sections.

The Methodology of Wittgensteinian Ethics

  1. Top of page
  2. Abstract
  3. Carter's Complaint
  4. Criteria and Internal Relations
  5. The Methodology of Wittgensteinian Ethics
  6. The Inwardness of Concepts
  7. The Realm of Meaning
  8. Authority and Relativism
  9. Conclusion

At this point, we are confronted by two common complaints raised against the methodology of Wittgensteinian ethics. The first is that there are no conceptual truths in morality, and that therefore appeal to what “we” do can never be relevant to the question of what ought to be done in a given case. The second (related) complaint is that making moral judgements depend on practices compromises their objectivity, and therefore opens the door to an intolerable subjectivism or relativism.

We may distinguish four alternatives regarding the relation between practices and our concepts. On the one hand, there are those who think that (at least certain) concepts do not depend for their intelligibility on their appearing upon the background of certain way of life. This view is typified by the utilitarian insistence that we must sharply distinguish the moral from the psychological; in other words, we must distinguish how things ought to be from how things, in fact, are. As Peter Singer puts it:

Which parents could give away their last bowl of rice if their own children are starving? To do so would seem unnatural, contrary to our nature as biologically evolved beings – although whether it would be wrong is another question altogether. 32

Within this family of positions, we may distinguish between (i) those that take the “foundational” ethical concepts (thin concepts, in Williams' terminology) to reflect rationally necessary, and hence objective distinctions; and (ii) those that conceive of the application of these concepts as determined solely by an individual's will. 33

The former view is expressed in Singer's contention that moral judgements are to be justified by deriving them from certain, rationally self-evident intuitions. As he says:

the [claim that the] death of one person is a lesser tragedy than the death of five … is a rational intuition. 34

The common mark of this viewpoint is a desire to found morality on considerations that are not arbitrary and so avoid what is seen as the taint of the parochial. That means either discerning constraints that apply to all thinkers merely as such, or else showing how only a given system of moral concepts accurately reflects distinctions laid down in the nature of things.

View (ii), by contrast, denies that the world provides independent warrant for our moral practices. Rather, moral norms are to be located within the subject, as constituted by some desire or act of will of a given agent. In this way, morality becomes no more than an exercise in the activity of a faculty of choice, and can have no more objectivity than similar activities, such as expressions of preference or interest. Both Hume and Nietzsche can be read as espousing views of this sort. 35

Although these positions diverge in crucial respects, they share the common contention that at least some of our concepts can be explicated without reference to the particular practices in which they are embedded.

On the other hand, another family of views acknowledges that our concepts are given content by virtue of their place within a social context, and cannot be understood outside of it. This observation can be taken to be offering a justification of the conceptual distinctions that we make, on the grounds that, from within our practices, this is simply what we (must) do. Views of this kind fall under one of two categories. On the one hand, (iii) there are those views that accept that concepts (and judgements that are based on them) are relative to forms of life, and accept that the potential for variance in forms of life leads to potential variance in what judgements are true, what inferences valid and so on. This means accepting some form of conceptual relativism. On the other hand, (iv) there are those views that try to delimit certain concepts that are dependent on forms of life that are (necessarily, for beings such as us) universal. Views of this sort try to capture the practice dependence of our concepts, while avoiding the cultural relativism that it would seem to imply.

As I understand Carter, he defends a version of option (iv) above. In other words, the impossibilities of loving cow dung and loving evil are both to be grounded in facts that are basic to human life as such – that it is, in some sense yet to be explicated, our human nature that makes it impossible for us to love either evil or dung. He sees Wittgensteinian moral philosophers as oscillating between (iii) and (iv). In the next section, I will explain why this is mistaken. While Peter Winch did, in his early writings, subscribe to (iv), he came later to see all of options (i)–(iv) as unsatisfactory. Recognising this fact (and seeing why he changed his views) is crucial to understanding the work of contemporary moral philosophers, such as Gaita and Cordner.

The Inwardness of Concepts

  1. Top of page
  2. Abstract
  3. Carter's Complaint
  4. Criteria and Internal Relations
  5. The Methodology of Wittgensteinian Ethics
  6. The Inwardness of Concepts
  7. The Realm of Meaning
  8. Authority and Relativism
  9. Conclusion

In his early work, Peter Winch looked for features of human life that were inevitable for beings such as us and that would guarantee the inescapability (and, in that sense, objectivity) of certain moral concepts. 36 As he put it:

the very conception of human life involves certain fundamental notions – which I shall call “limiting notions” – which have an obvious ethical dimension, and which indeed in a sense determine the “ethical space” within which the possibilities of good and evil in human life can be exercised. […] Their significance here is that they are inescapably involved in the life of all known human societies … The specific forms which these concepts take, the particular institutions in which they are expressed, vary very considerably from one society to another; but their central position within a society is and must be a constant factor. 37

In his article “Nature and Convention,” he applied this sociological methodology to the concept of truth. In it he argued that “there could not be a society in which truthfulness were not in general regarded as a virtue.” 38 (Let us call this, following Holland, a “life-form argument.” 39) He gave quasi-genealogical reasons to think that there are certain moral concepts that all societies will share, and concluded that:

moral conceptions arise out of any common life between men and do not presuppose any particular forms of activity in which men may engage together. 40

Were his argument successful, that would provide just what Carter is looking for in his reading of Wittgensteinian moral philosophy. We could demonstrate the inescapability of certain moral concepts by seeing them as essential constituents of any human forms of life. 41 However, Winch's argument was – by his own admission – a failure. It failed for two reasons.

First of all, although there are certain empirical features of human beings that are common to all cultures, these features can be taken up into cultural mores in radically different ways, and it seems unlikely that there can be any demonstration of a “core” of concepts that are common to all possible human forms of life and that are yet substantial enough to form what we would recognise as “morality.” Although in that paper, Winch claimed that variance of this sort is consistent with his central thesis, he did not take seriously the prospect that cultural and personal variance might deprive the notion of a “core” or “fundamental” concept of truthfulness of any meaningful content. 42

Thus, while he cites remarks from Wittgenstein in defence of this procedure, at this point Winch did not recognise that, for Wittgenstein, the results of his “remarks on the natural history of human beings” were purely negative. 43 When, in the Investigations and elsewhere, Wittgenstein appeals to imagined or real human practices, he does so not in order to find some common element between them, but rather to disabuse us of the notion that there is some necessity (whether logical or metaphysical) in the distinctions that we make, in the concepts that we use. Notable by its absence are any positive specifications of necessary structures of human society:

I am not saying: if such-and-such facts of nature were different, people would have different concepts (in the sense of a hypothesis). Rather: if anyone believes that certain concepts are absolutely the correct ones, and that having different ones would mean not realizing something that we realize then let him imagine certain very general facts of nature to be different from what we are used to, and the formation of concepts different from the usual ones will become intelligible to him. 44

Second, such sociological observations fail to account for what, following Hertzberg, may be called the “inwardness” of concepts, such as truthfulness. 45 This point was put to Winch by R. F. Holland:

[T]he only truth … explained [by life-forms arguments] is the sort that is told in the degree to which it supports the surrounding organisation or gets you by without disrupting the social pattern. […] [A]longside [this] there could co-exist for at least some people in the society a concern with truth of an altogether different character, in which not to falsify became a spiritual demeanour. Where then could this spirit come from? 46

There are two ways in which a defender of a “life form” account of morality may try to account for a spiritual demeanour towards truthfulness. The first alternative is to treat the individual with a workaday concept of truth and the individual for whom it is a spiritual demeanour as sharing the same concept of truth. The second alternative is to take the divergence in individuals' attitudes to indicate that there are two separate concepts of truth in play in such a society.

Trouble lies at the end of either path. If one thinks that there are two separate concepts of truth here and that life-form arguments only account for one but not the other, then it seems as if such arguments have left morality itself entirely unaccounted for. This sense will be heightened by anyone who has experienced a feeling of truthfulness as a spiritual demeanour, as from that perspective, a workaday conception of truth seems inimical to genuine morality. (This, I take it, is what Kant is gesturing at when he insists on the distinction between hypothetical and categorical imperatives being one of kind rather than degree. 47)

On the other hand, there are problems with thinking that the difference between these individuals is not to be accounted for in terms of different understandings of “truth,” but rather in terms of different depths of commitment to the same concept. Such a view makes it seem like the person for whom truth is a spiritual demeanour is just someone who has a strong interest in its promotion. Since that interest outruns the functional purpose of truth, that person starts to seem unusual at best and, at worst, cranky. More to the point, it undermines the difference between these perspectives to make them out as one of degree. The difference between (say) Hume and Kierkegaard is not to be explained along the lines of the difference between a philatelist and someone who takes occasional pleasure in the design of their stamps.

Thus, in failing to account for the ways in which ethical concepts may be informed by the distinctive attitudes of especially committed individuals, life-form arguments fail to give a convincing account of ethics at all. 48 Indeed, Winch soon realised that without an understanding of the inwardness of the concept – i.e. of the importance that it held for the individual within their life, the ways in which it connected to other things that they found important, and so on – one would simply be unable to understand the behaviour of the person whose commitment to truth was exceptional:

If one looks at a certain style of life and asks what there is in it which makes it worthwhile, one will find nothing there. One may indeed describe it in terms which bring out “what one sees in it”, but the use of these terms already presupposes that one does see it from a perspective from which it matters. The words will fall flat on the ears of someone who does not occupy such a perspective … 49

However, once we think that there are ethical concepts that can only be understood by seeing them in the context of the lives of their adherents, we immediately lose the idea that impersonal sociological reflection alone could demonstrate the inescapability or desirability of a given ethical concept. That left Winch (and as I will try to show, Gaita following him) needing to emphasise both the dependence of our moral concepts on the background of social practices (practices that, in turn, are shaped by natural biological facts and our reflection upon them) and the underdetermination of our moral concepts by such practices.

Since the arguments that applied to (iv) apply equally to (iii), 50 if we are to avoid the recoil to alternatives (i) or (ii) above, we must give a different account of the relation between practices and concepts. In the following section, I will suggest that, in Gaita's writings, just such an account can be found.

The Realm of Meaning

  1. Top of page
  2. Abstract
  3. Carter's Complaint
  4. Criteria and Internal Relations
  5. The Methodology of Wittgensteinian Ethics
  6. The Inwardness of Concepts
  7. The Realm of Meaning
  8. Authority and Relativism
  9. Conclusion

Both Winch in his later writings and Gaita following him are careful to avoid the problems Winch encountered in his early conception of the relation between practices and concepts. Those problems were encouraged by terminology that invited a dichotomy between the “grammatical” (i.e. conceptual) on the one hand, and “forms of life” on the other. That, in turn, led to the thought that either we could ground conceptual distinctions in natural facts (including practices and instinctive human reactions) or else those distinctions were arbitrary. 51

The way out of this difficulty is to deny this dichotomy, and to undermine the thought that grammar's autonomy makes it in any meaningful sense arbitrary. (After all, as the etymology of arbitrary makes clear, for something to be arbitrary it must be subject to an individual's will, and grammar is not (and cannot be) under the control of any individual. This is the upshot of Lewis Carroll's fantasy of Humpty Dumpty. 52)

In this section, I will suggest that Gaita's appeal to “the realm of meaning” and related phrases like Diamond's talk of our “lives with language” are intended precisely to avoid the empirical/grammatical dichotomy upon which Carter relies. I will argue that claims (1)–(3) are, therefore, not different in kind from one another. 53

Gaita characterises the realm of meaning in various different ways. He says that it is a cognitive realm in which form (including tone) and content cannot be separated. Discussions within the realm of meaning are modulated by critical categories such as depth and shallowness, and by intellectual vices such as sentimentality, pathos, hyperbole and so on. Notably, as these vices occur within such forms of thought, they are forms rather than causes of failure. 54 The realm of meaning includes considerations that are moral (“can't you see what it means to suffer as he does?') and personal (“how can I face death without cowardice?”).

We can see what this comes to if we consider an example. I will take that of love. Love is informed by our physical makeup and the environment; love for others is influenced by all sorts of accidental features of the world and of our physical constitution. Romantic love, in particular, depends (at least in normal contexts) on regular physical proximity, and on reciprocal and mutually enjoyable intimate gestures. The sight of the beloved (again, normally) sets off complex chemical reactions in the body, making their company pleasurable and their absence unpleasant. If, for whatever reason, one begins to find the presence of the other burdensome rather than pleasant, or if – say through some accident or another – intimacy becomes difficult, painful or impossible, then one's love for them comes under threat and may with time die out entirely. 55

However, although love is influenced by empirical conditions in this way, it is not entirely determined by them. 56 This is the upshot of Gaita's discussion of the Nun's behaviour towards the patients in a psychiatric ward in Melbourne. Those individuals had been deprived of everything that we take to be necessary for living a flourishing life, and had been so reduced by affliction that they had lost all human dignity. 57 Nevertheless, the Nun was able to respond to them as fully her equals. Her love towards them showed that it was possible to treat other human beings without condescension no matter how much they had been reduced by suffering.

Thus, although the language of love makes possible attitudes towards the world (and its parts) which would not otherwise exist, it does not settle the limits of such attitudes. 58 To think that it does would be to believe that the range of possible thoughts that one can think are fixed a priori, and to disclaim the idea that we might, through confrontation with the world and reflection upon it, deepen our understanding of life and its possibilities. (As Wittgenstein urged, we have no more reason to think that the class of possible thoughts/attitudes is fixed a priori as to think that there is a similarly determinate class of all possible games that human beings might play. 59 Both suppositions depend on a misguided conception of possibility.)

With this in mind, we can return to Carter's discussion of claims (1)–(3). Carter doesn't know whether to call these ‘grammatical truths’ or ‘facts that are basic to our form of life’. The first thing to notice is that, as sentences abstracted from a particular context of use, we cannot assess the truth of (1)–(3). Consideration of the role that context plays in giving these expressions their sense will show that his dichotomy is misplaced.

What are we to make of the first of Carter's putative impossibilities – that one cannot love evil? Upon reflection, apparent counterexamples present themselves. From one direction, Milton's Satan declares “Evil be Thou my Good” (and what is that if not an expression of a love of evil?). 60 From another direction, Christian doctrine teaches us that even the worst sinners, even if unrepentant, are loved by God and are worthy objects of the love of saints. 61

Of course, we might qualify our original statement to take these putative counterexamples into account. We might say that evil is not the kind of thing that can sustain a genuine and deep love – that the nature of evil actions is inconsistent with the self-sacrifice that love requires. Equally, we might say that while God loves the sinner, She does not love the sin – that evildoers are loved for their innate capacity to be good, a capacity that no amount of wrongdoing can entirely remove.

But these qualifications should make it clear that (1) is not simply “guaranteed grammatically.” Carter claims:

If one means by “love” that one can love even evil, then one is either decidedly confused in relation to one's own grammar or in possession of a concept of love and attendant grammar that will seem to others monstrously strange. 62

This makes the issue out to be more straightforward than it actually is. While we may feel that Satan's attitude is flawed, we cannot simply dismiss his example by saying that either he is confused or else he means something different by “love.” Only a perspective that has a vivid sense of the possible depths of love, and the relative shallowness of evil, is one on which it will seem impossible to love evil. Achieving this perspective is not simply a matter of competent participation within our culture. It takes, rather, a distinctly personal form of understanding, supported by argument and sustained by rhetorical vision. It is, therefore, not simply a matter of reading off moral conclusions either from our physical proclivities (as if evil caused universal physical revulsion) or from our shared grammar (as if an unloved evil were on a par with an unmarried bachelor).

Importantly, this conception of love depends on one's physical behaviour as much as one's postures of mind. An inability to love evil will manifest itself in, among other things, an inability to take pleasure in otherwise enjoyable experiences that are ill-gotten. 63 Similarly, certain physical states or dispositions (such as a tendency towards jealousy) may undermine one's ability to hold onto this conception of love.

Regarding (2) – it should be noted that Gaita never says that dung cannot be an object of love, only that it would be absurd for an individual to treasure a piece of dung as a token of their beloved's affections. The nature of faeces is such that it cannot play that role in our lives. But that is perfectly consistent with Carter's suggestion that someone might come to love dung in another context – such as that which he describes in his essay. Indeed, Gaita's description of his friend Hora's attitude towards the natural world makes use of a similar experience of an individual whose vision of life is free of sentimentality, someone who accepts the existence of shit and does not take it as an obstacle to a joyous participation in life:

[W]hen I try to understand why [Hora's] stories have nourished me throughout my life, I know the answer must include the way Hora was so distinctively present in their telling, his openness to the world and the quality and many tones of his laughter And, strange though it may sound … it must include his sensuous love of the sun and the water, and how, in between stories, he plunged from the boat into the reservoir, swimming sideways, forwards and backwards, splashing and whooping. From none of that can I abstract the distinctive quality of his humanism – as a set of principles, for example. In that sun-drenched, summer-coloured humanism I found food to nourish hope. 64

Of course, the way in which dung plays a role in Hora's love for the natural world is markedly different from that which Gaita imagined when he pictured an individual treasuring dung as a token of their beloved's affections. The absurdity of the latter suggestion, together with the manifest possibility of a love which encompasses other roles for dung within a loving attitude towards the world, shows that there is no straightforward inference from physically derived attitudes to faeces (revulsion, etc.) to conclusions about how it can be loved. It is as much a matter of the grammar of loving, of what can be treasured and what not, as of our biological inheritance. The gestures of treasuring (carrying the object around one's neck, perhaps, caressing or kissing it, admiring it in the light, etc.) cannot be sustained with an object of this kind.

But how are we to understand the status of this “cannot”? It is not as if a biological study of the human body, together with a scientific investigation into the kinds of objects of our environment, will yield a specification of the class of caressable objects. Certain physical limitations (such as revulsion) may be overcome by someone whose experiences and demeanour allow them to sustain tenderness in confrontation with an object which would usually be taken to be beyond such reactions. 65 The limits to what limitations can be overcome in this way are that of what one can say and mean. One can compare this to an individual's ability to say, and mean, a declaration that they are happy. Anyone can say, at any point, that they are happy. Whether that declaration falls flat or not is another matter, and is not something that is under their control. 66

The strangeness of claim (3) is slightly different to that of (2). We can all surely relate to a sense of incredulousness at Socrates' contention that the good man cannot be harmed. 67 The source of this incredulousness is the fact that on a natural (i.e. ordinary) understanding of harms and benefits – the pleasantness of pleasures and the unpleasantness of pains – wrongdoing often leads to far more benefit than harm. Socrates claims that the harm of becoming a wrongdoer undermines the putative benefits of wrongdoing, such that no one would choose to do wrong knowingly. From the perspective of someone who lacks an imaginative conception of wrongdoing, and who is preoccupied with physical pleasures and pains, this view will appear to fly in the face of common sense.

In this case, it is not because of our natural proclivities but in spite of them that we can judge the good individual invulnerable to harm. And yet the Socratic position cannot be tenable if we imagine Socrates to have done nothing more than offered a redefinition of “harm”; to have stipulated that (for him) one is only harmed by the products of one's wrongdoing. If this definition is not supported by his example, then it is nothing more than mere words. In this way, although we may characterise the Socratic thesis as an insight into one possible understanding of the concepts of goodness and of harm, that thesis only has sense because there is a way of life in which it can be expressed. The conceivability of such a point of view depends on it being humanly possible to retain a sense of goodness on which harms incurred by goodness no longer appear harmful.

Thus, the impossibilities averred to in (1)–(3) all depend on features of our “lives with language.” We cannot say of any given impossibility that it is grounded either in our “form of life” or in our “grammar,” when these are conceived of as distinct alternatives. The limits of love are fixed by a combination of biological and grammatical features of our lives. Even if certain objects or individuals happen to be beyond the reach of human love, this is a product not merely of features of a shared creaturely condition, but rather of the features of that condition as they occur within the lives of the individuals and as they are understood by them. 68

Authority and Relativism

  1. Top of page
  2. Abstract
  3. Carter's Complaint
  4. Criteria and Internal Relations
  5. The Methodology of Wittgensteinian Ethics
  6. The Inwardness of Concepts
  7. The Realm of Meaning
  8. Authority and Relativism
  9. Conclusion

The previous section answered Carter's worry by arguing that there is no fixed distinction between the grammatical and the empirical, and that the sphere of possible objects of love is fixed by the responses of exceptional individuals. It remains to be seen whether this leads to relativism.

There are two reasons for thinking of the view I have attributed to Gaita as entailing relativism. First of all, what is compelling and what not becomes a personal matter, and not one on which two individuals need (if competent and fully informed) agree. As Gaita puts it:

We sometimes find something morally impossible for us, but not think that people who find it possible are thereby mistaken, let alone morally deficient. Even examples of great and pure authority can rightly speak differently to people. 69

This seems to undermine the universal and impersonal nature of morality, and as a result make its claims no more secure than matters of taste.

Secondly, there are no standards, external to the practice itself, that settle genuine from counterfeit instances of authority. It may, therefore, seem as if there is no way to distinguish between legitimate persuasion and cultism or brainwashing. After all, from the perspective of the cultist, the head of the cult is an exceptional individual. As a result, one may worry that our inclination to dismiss the individual who claims to treasure a piece of dung, and to exalt the saintly individual who loves her neighbour, is merely arbitrary, an expression simply of our preferences and/or upbringing.

A full response to this worry is outside of the scope of this paper. 70 It would involve undermining the thought that a practice whose standards are internal to itself must necessarily be less than fully objective. That is a lesson we might take from the rule-following passages of the Philosophical Investigations. 71

It would also involve showing how an individual's commitments and attitudes within such practices are not to be associated with the mere expression of preference, as if nothing spoke for or against one ethical attitude or another except desire, interest or force of will. A confrontation between two ethical outlooks is not always between two individuals with different desires or interests, but can instead be between two different ways of looking at things, two different ways of seeing and understanding the significance of one's life, and of the world and one's place in it.

In confrontation with another there is both risk and opportunity. One's sense of the rightness of one's current path might be upset, or one's conception of the possibilities for life might be deepened. Thus, we may say that the problem with the cultist is not that they defer to an authority figure, but rather that they assign to that authority figure an inappropriate role in their lives. For them, the cult leader acts as a surrogate for genuine confrontation with life's difficulties. Participation in the practices of the cult, therefore, becomes a form of avoidance of others. An explanation of the inauthenticity of the cultists' judgements will, therefore, appeal to their inability to see other points of view, their willingness to allow the cult leader to make decisions for them and so on. Such an explanation would allow us to draw a distinction between us and the cultist, without denying the importance of authority figures in determining the scope of moral possibilities. 72

We may then respond to Carter's question (“who determines what places a thing may have in human life?”) as follows: we do. And we do so on grounds often no more (nor less) secure than our sense of the manifest absurdity of the alternatives. We may acknowledge that we are not always infallible judges of what attitudes and responses are possible for human beings, without thinking that our confidence in moral judgements, our rejection of certain putative authorities, can only ever be provisional in the absence of some more secure (perhaps objective and impersonal) criteria for separating legitimate from illegitimate forms of persuasion. 73


  1. Top of page
  2. Abstract
  3. Carter's Complaint
  4. Criteria and Internal Relations
  5. The Methodology of Wittgensteinian Ethics
  6. The Inwardness of Concepts
  7. The Realm of Meaning
  8. Authority and Relativism
  9. Conclusion

For these reasons, Carter's worries about Wittgensteinian ethics are misplaced. Since there are no “mere facts” within the realm of meaning, there can be no moral “mere facts” of the sort for which he hopes. However, this shouldn't unnerve us. To recognise that moral distinctions are made within the realm of meaning involves accepting that they may be no more, nor less, secure than an individual's entire way of going on. 74 Nothing guarantees that others will share one's conception of value, and the activity of justifying one's moral convictions (either to oneself or to others) is not one of appealing to shared or impersonal grounds.

  1. 1

    D. Carter, “ ‘Part of the Very Concept’: Wittgensteinian Moral Philosophy,” Philosophical Investigations Vol. 36, Issue 1, pp. 3755 (2012): doi: 10.1111/j.1467-9205.2011.01467.x

  2. 2

    Ibid., p. 19. “Form of life” is a Wittgensteinian expression, but as should hopefully become clear, I do not think that Carter uses it in the same way that either Wittgenstein, or Wittgensteinian ethicists such as Gaita, do.

  3. 3

    p. 7.

  4. 4

    pp. 5–6.

  5. 5

    p. 16.

  6. 6

    p. 18.

  7. 7

    Ibid., p. 19; my emphasis. Another way to put this is as follows: for Carter, we should be able to show that moral incapacities are, at root, a species of physical incapacity.

  8. 8

    Since a “form of life” in Carter's sense is to be defined by empirical investigation, in what follows I will also speak of this as the “grammatical/empirical dichotomy.”

  9. 9

    Putting it in a nutshell: since the moral does not mark out a class of concepts, a moral observation is a reflection of an individual's entire sensibility; Cf. Winch, Ethics and Action, (London: Routledge and Kegan Paul, 1972) 188ff. This follows from the fact that we cannot define the class of the moral in terms of a series of concepts [C. Diamond, “Ethics and Mathematics: Resisting the Attractions of Realism,” in H. Sluga and D. Stern (eds.), The Cambridge Companion to Wittgenstein (Cambridge: Cambridge University Press, 1996) 242ff.], a particular depth of commitment [P. Foot, Virtues and Vices (Oxford: Oxford University Press, 2003) pp. 112113], or a particular subject matter [R. Rhees, Without Answers (London: Routledge and Kegan Paul, 1969) 103ff.].

  10. 10

    Cf. L. Wittgenstein, Wittgenstein's Lectures, Cambridge 1930-32 (Chicago: University of Chicago Press, 1989) p. 31.

  11. 11

    Winch (1972), p. 26.

  12. 12


  13. 13

    In the Philosophical Investigations, there is only a single reference to an internal relation – in “Philosophical Psychology: A Fragment” (formerly “part 2” of the Investigations. Hereafter, I will refer to it as PPF) §247. By contrast, the concept of a criterion is widely employed – cf. the index to L. Wittgenstein, Philosophical Investigations, P. M. S. Hacker (ed.) (London: Wiley Blackwell, 2009).

  14. 14

    Winch (1972), p. 32. It is interesting that, in this article, Winch slides from talking about internal relations to criteria.

  15. 15

    C. Diamond, The Realistic Spirit (Harvard : MIT, 1991) p. 324.

  16. 16

    Ibid., p. 323, my emphasis.

  17. 17

    In this context, she also talks about the importance of the concept of piety; cf. Diamond (1991), p. 323.

  18. 18

    This is an idea that Winch takes from Simone Weil. Cf. “The Iliad, Poem of Force,” in Simone Weil: An Anthology (London: Penguin 1995).

  19. 19

    S. Cavell, The Claim of Reason, 2nd edition (Oxford: Oxford University Press, 1999) p. 90. On the difference here as one of perception cf. Winch, “Professor Anscombe's Moral Philosophy,” 193 in L. Alanen , S. Heinämaa and T. Wallgren (eds.) Commonality and Particularity in Ethics (London: Palgrave MacMillan, 1997).

  20. 20

    Cavell (1999), p. 7.

  21. 21


  22. 22

    Only “in part” because of the complexity of the sceptical tradition. As Cavell suggests, a number of different and contradictory feelings and drives are bound up in scepticism. These include wonder (that there should be a world and that we should have access to it), aspiration (to know things with the completeness and certainty that characterises God's state of awareness) and fear (to avoid confronting the independent reality of others and one's vulnerability to them). Cf. his (1999), passim.

  23. 23

    Cavell (1999), p. 31.

  24. 24

    Wittgenstein (2009), §242.

  25. 25

    On the inadequacy of the conventional/natural dichotomy to account for the formation of our concepts (of what is better called our “lives with language”), cf. C. Diamond, “The Importance of Being Human,” Royal Institute of Philosophy Supplement 29 (1991): 35ff.

  26. 26

    Cf. Cavell (1999), 5ff.

  27. 27

    This is a point made in J. McMahan, “Our Fellow Creatures,” Journal of Ethics 9 (3–4) (2005): 353380. McMahan takes it to be fatal to Diamond and Gaita's methodology. That, I take it, is because he has misunderstood the role that appeals to what “we” do are intended to play on such views.

  28. 28

    In this respect, cf. Gaita's discussion of the behaviour of mountaineers; R. Gaita, The Philosopher's Dog (London: Routledge, 2002) 141ff.

  29. 29

    On the inability of convention to settle meaning, cf. C. Travis, Thought's Footing (Oxford: Oxford University Press, 2006) 113ff. More generally, his point is that any attempt to supply a theory that will determine the meaning of a sentence outside of its context of use is bound to fail. Such attempts fall prey to Wittgenstein's observation that “any interpretation would have to have another interpretation standing behind it”.

  30. 30

    After all, at least part of the reason for introducing conventions is to avoid making a decision in every new case.

  31. 31

    Cf. D. C. Lau (trans.), The Analects of Confucius (London: Penguin Classics, 1974) §§1.12 and 2.8 for a discussion of the difficulties of maintaining an enlivened sense of ritual.

  32. 32

    P. Singer, Practical Ethics (Cambridge: Cambridge University Press, 1993) p. 233; my emphasis.

  33. 33

    Cf. B. Williams, Ethics and the Limits of Philosophy, 2nd edition (London: Routledge, 2006) pp. 140142.

  34. 34

    Singer, “Ethics and Intuitions,” The Journal of Ethics 9 (2005): 331352.

  35. 35

    It underlies the Humean idea that two individuals with differing ethical views may agree on all the facts and differ only with respect to their attitude towards utility; D. Hume, An Enquiry Concerning the Principles of Morals, L.A. Selby-Bigge , ed., (Oxford: Oxford University Press, 1975) 286ff. And cf. F. Nietzsche, Beyond Good and Evil, R. J. Hollingdale, trans., (London: Penguin, 2003), passim.

  36. 36

    The idea of objectivity as indispensability or inescapability is now a common one – cf., e.g., H. Putnam, “Pragmatism and Moral Objectivity,” in M. Nussbaum and G. Glover (eds.), Women, Culture, and Development: A Study of Human Capabilities (Oxford: Oxford University Press, 1995).

  37. 37

    Winch (1972), pp. 42–43; emphasis mine.

  38. 38

    Ibid., p. 69.

  39. 39

    R. F. Holland, Against Empiricism (Oxford: Basil Blackwell, 1980) 101ff.

  40. 40

    Winch (1972), p. 59.

  41. 41

    For a modern version of this kind of argument, cf. B. Williams, Truth and Truthfulness: An Essay in Genealogy (Princeton: Princeton University Press, 2004), passim.

  42. 42

    This is a point that Winch gestured at, but did not fully develop, in his (1958), 98ff.

  43. 43

    Wittgenstein (1999), §415

  44. 44

    Ibid., PPF §366. One might ask whether there is a physical necessity in us having the concepts that we do – whether, in other words, they are best suited for reflecting (or coping with) the “very general facts of nature” that we have to deal with. In what follows, I will suggest not. It should also be noted that there is a slight awkwardness in this passage, stemming (I think) from an underlying incoherence in the notion of concepts being “absolutely the right ones.”

  45. 45

    Cf. “Philosophy as the Art of Disagreement: On the Social and Moral Philosophy of Peter Winch,” in J. Edelman (ed.) Sense and Reality: Essays out of Swansea (Frankfurt: Ontos Verlag, 2009). My exposition of Winch is indebted to Hertzberg's fine paper.

  46. 46

    Against Empiricism, 107.

  47. 47

    Cf. I. Kant, Groundwork for the Metaphysics of Morals (Cambridge: Cambridge University Press, 1997) 4:415.

  48. 48

    For the sake of convenience, I am using “ethics” and “morality” interchangeably.

  49. 49

    Winch, Trying to Make Sense (Oxford: Basil Blackwell, 1987) p. 190. We could characterise this as the guiding point of Winch's moral philosophy as a whole. Winch is constantly urging us to take seriously moral attitudes and practices which might, on the surface, appear to us to be bizarre or irrational. And since his seminal On the Idea of a Social Science and Its Relation to Philosophy, 1st edition (London: Routledge and Kegan Paul, 1958), he has insisted that the activity of coming to understand alien practices involves an imaginative and personal engagement with them. Meaningful behaviour cannot be understood through investigations which treat it as analogous to the physical sciences.

  50. 50

    Cf. Winch (1987), pp. 181–193.

  51. 51

    On the arbitrariness of grammar, cf. Wittgenstein (1999), §§497, 520, PPF §367; Winch (1987),165–166.

  52. 52

    L. Carroll, Through the Looking Glass (London: Penguin, 2007) p. 72.

  53. 53

    Cf., e.g., Gaita (2002), pp. 96–103.

  54. 54

    This is one way in which Gaita distinguishes considerations within the realm of meaning from “the factual, the logical or the discursively metaphysical” (2002: 102). However, this distinction ought to be treated with care. Discussions within the realm of meaning are still factual (insofar as we can make any sense of “factual”) and logical (again, insofar as we can make sense of that). Trying to get clear on, e.g. whether leaving someone counts as an abandonment, and if so what that means in the context of the relationship, is a different activity from determining the physical consequences of the individual's being left in this position. However, the two activities bear on one another. Just how these two activities are related, which in general may be characterised as the question of the relation between the scientific and literary images of mankind, remains an open question.

  55. 55

    On the precariousness of those features of our lives that give them meaning. Cf. S. Weil, “The Iliad: Poem of Force,” passim.

  56. 56

    This is a point at which McMahon baulks; cf. his (2005), 356ff.

  57. 57

    R. Gaita, A Common Humanity (London: Routledge, 2002) 17ff.

  58. 58

    The same lesson can be found in Primo Levi's account of the relationship between Charles and Lakmaker. Cf. Levi, If This is a Man (London: Abacus, 1987), 172ff., and Gaita, Good and Evil: An Absolute Conception, 2nd edition (London: Routledge, 2004) xvff. Here the wonder of Charles' behaviour is that he can treat Lakmaker with “the tenderness of a mother.” That involves an extension of the concept of motherly love beyond its ordinary contexts. That such an extension is possible is neither required nor prohibited by either the nature of the case or the grammar of “motherly love.”

  59. 59

    Wittgenstein (1999), §§65–69.

  60. 60

    J. Milton, Paradise Lost, revised edition (London: Penguin, 2003) ln.110.

  61. 61

    E.g. The King James Bible, Matthew 7:16.

  62. 62

    Carter, (2012), p. 15.

  63. 63

    Cf. Gaita (2004), 43ff.

  64. 64

    Justice and Hope,” in D. Modjeska , (ed.), The Best Australian Essays, 2006 (Melbourne: Black Inc., 2006) pp. 404405. Similar attitudes towards the natural world are brought out in Romulus, My Father (London: Text Publishing, 1998), passim. Hora's attitude is, therefore, the opposite of that sketched by Milan Kundera in The Unbearable Lightness of Being (London: Harper Collins, 2008), who takes the existence of shit to be the disproof of God.

  65. 65

    That is part of the point of Gaita's example of the Nun.

  66. 66

    Compare with Wittgenstein's deathbed declaration: “tell [my friends] it's been a wonderful life.” Cf. N. Malcolm, Wittgenstein: A Religious Point of View? (London: Routledge, 1997); Gaita (2004), pp. 198–199.

  67. 67

    Plato, Apology, 41d2ff.

  68. 68

    It might be objected that there are certain necessities that are secured by our creaturely condition alone, e.g. that a human being cannot jump 20 feet in the air. But (without denying the differences between this case and that of love) even a claim such as this depends on grammar – specifically, the grammar of “jumping.” Thus, human beings can jump 20 feet into the air if given a trampoline. Cf. Diamond (1991), 267ff.

  69. 69

    Gaita (2002), pp. 213–214.

  70. 70

    Cf. Winch (1987), pp. 181–193 for his own attempt to answer this charge.

  71. 71

    Travis has read a similar theme as the guiding ideas of McDowell's books – cf. his “Taking Thought” Mind, v. 109, n. 435, July, 2000, pp. 533–557.

  72. 72

    Indeed, an account on these terms must be possible if we are to reconcile Gaita's appeals to the role of authority figures in morality with his insistence on the necessarily personal nature of moral decisions. On the latter, cf. Gaita (2002), p. 278.

  73. 73

    On the standards that distinguish legitimate from illegitimate forms of persuasion, cf. Winch, “Persuasion,” Midwest Studies in Philosophy 17 (1992): 123137.

  74. 74

    Cf. L. Wittgenstein, On Certainty (New York: Harper and Row, 1969) §559.