Nicholas C. Burbules

University of Illinois


Paul Smeyers

Katholieke Universiteit Leuven



            What did Wittgenstein mean when he said ethics was a matter of which we could not (sensibly) speak? Such a claim seems implausible on its face, since people do in fact talk about ethical issues all the time, often to good purpose. Moreover, it is not clear what such a claim would mean for moral education: if we cannot speak about ethics, how can we teach young people about it?

            We want to suggest a certain continuity of Wittgenstein’s views on ethics, from his early statements in the Tractatus (1921) to his Lecture on Ethics (1929), to his later work in Philosophical Investigations (1953) and elsewhere. We believe that some of the central later Wittgensteinian ideas — language games, forms of life, how we learn to follow a rule, and family resemblance relations — can help to explain in what sense ethical understanding might be inexpressible. We will argue that conceptualizing ethics as a shared practice builds upon these Wittgensteinian insights, while providing a fruitful perspective on moral education.

            Early in his career Wittgenstein says that ethics is like aesthetics and religion, and cannot be spoken about (Tractatus, 6.421). It belongs to that realm where, as he says, things cannot be said but only shown (Tractatus, 4.1212). His Lecture on Ethics reasserted this basic view of inexpressibility, again linking ethics with religion: “Ethics, if it is anything, is supernatural and our words will only express facts” (Lecture on Ethics, pp. 40).

Yet his personal writings and correspondence with friends make it clear that he was highly concerned with being ethical himself. Indeed, as his friend Paul Engelmann put it, Wittgenstein believed that it was precisely those things about which one could not speak (within the strictly limited propositional discourse of the Tractatus) that were the most important things in life, including ethics.

            Yet, as is well known, he changed his mind in his later career about how language works and what it can do. The strict dichotomy between what can be said (expressions about the logical structure of the world and about logic and mathematics) and what he called “nonsense” (everything else) was replaced by a variety of diverse “language games.” Considering the many uses he says that language can have, it seems odd for him not to have included ethical discourse as one type.

            One possible answer for this omission is that he stayed faithful to his earlier philosophical intuition, that nothing could be said about ethics. Certainly, for Wittgenstein throughout his life his views on ethics were closely aligned with his religious beliefs, about which he was extremely private, and which do seem to pertain more to the transcendent and the ineffable.

            But here we wish to explore an alternative answer. In our view, the crucial concept in Wittgenstein’s later work is “practice.” He says that it “is not certain propositions striking us immediately as true, i.e. it is not a kind of seeing on our part; it is our acting, which lies at the bottom of the language-game” (On Certainty, # 204). The concept of practice is given shape in this notion of language-games, with their interwovenness of utterances and actions and how they find their home within a “form of life.” Wittgenstein presses us to adopt the view that in the end there is simply what we do; this does not mean that justifications cannot be given, but that justifications come to an end, and then he says, I have reached bedrock, then my spade is turned (Philosophical Investigations, # 217). In this respect, our ability to explain and justify ourselves ethically is limited.

            There is moreover in the idea of a language game the importance of a particular linguistic community, on the one hand, and the possibility of an individual who may come to give a new meaning to particular phenomena, on the other. As he says, sometimes we follow rules and sometimes we make up the rules as we go along go (Philosophical Investigations, # 86). By making clear that both dimensions have to be taken into account he avoids the danger of simple conservatism or conformism. Wittgenstein’s “theory” of meaning advocates neither a position of pure subjectivity nor of pure objectivity. In order to be understood (i.e., to make sense in what one says or does), the present use of language cannot be radically different from former ones. It is within this context of use that the meaning of a concept is determined. As there is no absolute point of reference, neither internal nor external, the community of language speakers (some community; in a pluralistic society there will naturally be more than one) forms the warrant for the consistency of meaning.

            In this epistemology, Wittgenstein made it clear that following a rule is not just a matter of mimicking a particular behavior from one situation to another. Though we follow rules, they cannot be fully made explicit; it is always necessary to take into account all the elements of the new situation one finds oneself in, which implies, among other things, communication, dialogue, and above all commitment. One can only be “certain” of the frame of reference itself; this is part of the life we have inherited, not the result of systematic (rational) teaching. When one has learned to follow a rule, Wittgenstein says, this is manifested only in the claim “Now I know how to go on,” which is a performative ability, not a rationally articulatable understanding (Philosophical Investigations, #151 and 179).

            Wittgenstein’s ideas concerning particular epistemological positions and language have far-reaching implications for the domain of ethics. To say that the meanings of “good” or “right” are not once and for all determined, does not imply that it does not matter what we do. But convincing someone on the ethical level is for Wittgenstein not (simply) a matter of giving them reasons. It is more like a practice in which other people are interactively involved. Here again he draws our attention to the importance of context and to how one has learned to use concepts like “good” and “right.” Because of the fact that every context is necessarily particular I am answerable for what I do. There are no ultimate foundations.

            The question is whether an ethical problem can be characterized in the Wittgensteinian sense as a philosophical problem, that is, a problem of the sort “I don’t know my way about.” He also says that philosophy is a kind of therapy, a kind of work on the self, that helps us get outside certain problems and see them in a new light. Can we see ethics as similarly involving a kind of work on the self? But can we only change ourselves? Why can no more be said?



            Wittgenstein seems to hold the position that though our actions are guided by rules and though people generally act in consistent ways, the reasons for this cannot be spelled out fully. It cannot exhaustively be made explicit, it remains inexpressible, yet art and evocative language may be able to touch upon it (this is what he tried to do in the Lecture on Ethics). It is about what one does, not about the reasons that may be given, which seem to be superfluous. Therefore, on his view, a systematic moral philosophy will always and necessarily fail.

            Wittgenstein often compared philosophical problems with being lost, with being trapped in a fly-bottle, with not knowing one’s way about. Rules, he says, are like signposts suggesting a way to go, but the notion of rule-following for him was complex and subtle, because there is no one correct way to follow a rule; and, as noted, he also says that sometimes we have to make up the rules as we go. At the same time, it must be possible to make a mistake; making the rules up does not mean just acting in any way one likes. So when can one say one has understood a rule, or knows how to follow it? When one can say, “now I know how to go on.” His simple illustration here is a mathematical rule, like “add 2.” If you give someone the sequence, 2, 4, 6, 8... they understand the rule when they can “go on,” when they can continue with the sequence (...10, 12, and so on). But not all rule-following is this simple. First of all, there may be more than one way to continue the sequence, or more than one rule that would generate it — it is the doing that matters for Wittgenstein, not the articulation of a rule.

            Second, and following closely on this point, the person may not be able to articulate a rule even if asked. Here rule-following is akin to Michael Polanyi’s idea of tacit knowledge: understandings that enable complex activity and decision making, but which cannot be put exhaustively into words. Such performative abilities are typically learned through observation and emulation, trial and error, making and learning from mistakes, not through explicit instruction or explanation: novices must watch and participate in activities with experts as gradually over time they begin to “get it,” until they reach a point where, again, they can “go on” on their own.

            Hence, this Wittgensteinian argument suggests a different way, not based on spirituality or the transcendent, in which some ethical understandings may be inexpressible in words — namely, that they are matters of conduct, learned in context through observation and emulation, and performed more or less consistently without being the result of conscious deliberation or rule-following in the strict sense of that term. Again, in Wittgensteinian terms, there is simply what we do.

            At a home in which he was a guest, Wittgenstein was asked by one of his hosts whether he would like some tea. Her husband, overhearing, called to her, “Do not ask — give!” This comment “most favorably impressed” Wittgenstein (Malcolm, p. 61), and this anecdote suggests a flavor of what we are talking about. For Wittgenstein, one should be gracious and generous without thinking about it, without asking, just by knowing what the proper thing is to do. There seems to be a touch here of Zen; somehow, without speaking or thinking, one simply intuits in the instant what the proper course of action should be. But Polanyi’s theory gives an alternative account of this process: that it has more to do not with mystical intuitions, but with learned habits and responses that arise from familiarity with a situation and a group of other people.



            What does it mean to conceive of ethics as a practice? First of all, it means that it is a constellation of learned activities, dispositions, and skills. We learn to engage in complex practices through observing and emulating others who are more skilled than we; through our own practice, trial, and error; through making mistakes, and learning from them; through deliberation and reflection on what we are doing and why; through creatively responding to new and unexpected situations; and so on. From the framework we are sketching here, ethics is no different: we learn to be good and to do good; we are initiated into a form of life that values these activities and that supports us in enacting them. This background of conditions is true even when we seem to be deliberating and acting entirely on our own; for however autonomous and self-directed our efforts might appear at that moment, we could not have been capable of such deliberation and action without a substantial set of interactions with others from the earliest stages of our lives. In this sense ethics always exists against the background of a form of life.

            The Wittgensteinian analysis of understanding and following rules also pertains here. Consider the following range of ethical situations, with simple illustrations:

what we do without thinking (e.g., spontaneously giving money to a beggar);

what we do when we know how to go on (e.g., filling one’s pocket with $5 in change in the morning, and giving 50Ę to each beggar one sees until the money runs out);

what we do when we do not know how to go on (e.g., encountering a beggar sitting outside a liquor store);

what we do when we are trying to teach someone else how to go on (e.g., encountering a beggar when we are with one of our children).

There is no reason to assume that our processes of thought and action will always work in the same way, in ethics as in any other complex practice. Sometimes the situation is highly familiar and our responses are well-rehearsed; sometimes it is a novel situation, but one in which we have an established repertoire of ways of coping with it; sometimes it is a highly problematic, confusing, or difficult situation, in which our ordinary repertoire either does seem to apply, or does not work in the way we expect; and sometimes we are consciously in a situation in which we are thinking not only of our own processes of deliberation and action, but also of enacting these in such a way that others might learn from us.

            These general characteristics of contexts of practice pertain equally to ethical situations, and they indicate something very important, which is that all ethical situations may not elicit the same responses from us in every instance, or from others in similar circumstances. Theories of morality that suggest general processes of deliberation, even of calculation (as in utilitarian theories), that lead to determinative conclusions about what to do — or those that assume a single model of action, such as certain kinds of virtue theories — make the error of abstracting ethical deliberation and action from the quite varied contexts in which we actually think and act morally. The situations listed above have elements in common that make them all ethical situations; but this does not mean that they all necessarily work in the same way. (Recall here also the notion that there might be different ways of following the same rule; or different rules that yield similar courses of action.)

            In what sense, then, are these all ethical situations? Here another concept from Wittgenstein helps to clarify the matter, his idea of family resemblances. Unlike the Platonic model of finding a common form that underlies all instances of a concept, the Wittgensteinian view is that, as with members of a family, there may be a number of overlapping shared characteristics that create a fairly distinct cluster of associated instances, without any subset of these characteristics being necessary and present in all cases: “a complicated network of similarities, overlapping and criss-crossing: sometimes overall similarities, sometimes similarities of detail” (Philosophical Investigations, #66). In the same way, the range of situations we characterize as “ethical” do not necessarily share any single common feature: broadly they concern issues of human well-being, but in different senses of that term; they concern activities that express and develop ethical identities, but they do so in different ways; they involve responsiveness to the demands of situations that call forth from us certain moral responses; they involve choices and actions in which we are aware that we are being observed, judged, or imitated by others, which expresses a responsibility for the practice itself. This last point is worth emphasizing, because all practices, insofar as they are practices, are never entirely personal and idiosyncratic; they are learned, they are taught, they are part of a shared legacy within a form of life. Hence enactments of those practices always exist against the background of implicit norms necessary for those practices to be exercised and maintained within a particular context and time frame, and for them to be carried forward and passed on to others over time — even when practices change, this larger context of norms is necessary for the very continuation of circumstances that allow change. In this sense it is not an exaggeration to call ethics the practice of practices.



            Within this anthropologized account, the matter of justification must be addressed. Wittgenstein in his later work tended to reject general models and theories, emphasizing the particulars of context and purpose; one can give justifications, he said, but they eventually reach an end, and he says at this stage one can only describe: “this is what we do.” The form of life is the bedrock beyond which explanations cannot go — nor can a form of life be asked to justify itself, because it sets the conditions for any possible justification.

            The conditions that make a practice, any practice, possible, are not arbitrary: they provide a set of constraints and norms that are generalizable in the sense that any alternative practice would need to rely on at least some of these same constraints and norms (for example, they must be replicable from generation to generation of practitioners, and this entails nonrelative processes by which such communication and teaching will be possible). Thus ethical justifications, while they must rely on the particulars of a moral situation, are not entirely free-floating. All the same, in this accounting there is a fundamentally different character between the questions, “Why did you tell the truth to Harry,” and “Why do you tell the truth?” The second question may only be answerable in certain situations with the reply, “because this is what I (or we) do.”

            Yet even within the framework of a particular situation and context, the task of moral justification can be extremely complex — more so than is given credit in some moral theories. When ethics is viewed as a practice, more dimensions of the question are revealed than simply how one means for one’s action to affect another. There is the crucial dimension of moral deliberation and the justification of one’s actions, to be sure. But alongside this, and inseparable from it, are other frameworks of justification. Each act, even the most simple, also has the dimension of moral learning — how the act will affect the development of one’s own moral character and identity; the image we have of ourselves, and how we reconstitute it over time. While it might seem that this consideration always falls in line with acting to serve the interests of others, that may not be so in every instance. Similarly, there is the consideration of moral exemplification, or moral teaching, in which we consider the influence of our act upon others who might observe or become aware of it. There is also the consideration of moral self-formation, different from the factor of moral learning, since it pertains to the aspects of caring for one’s self that allow the maintenance of one’s moral agency and capacity to act morally. These are not just distinct moral considerations that, as in any moral case, might create conflicts of principle or priority and so be difficult to reconcile; they are moral considerations of decidedly different character, and so are not amenable to simple comparison. Yet they always coexist, for as in any complex practice, the influence of our immediate actions upon their end always also has effects on the continuation and maintenance of the practice itself. To act in such a way that one’s immediate purposes are served, while the integrity and possibility of the practice generally are undermined, is self-defeating.

            As Wittgenstein says, ethical teaching cannot simply be reduced to training (Culture and Value, p. 93). Here we wish to suggest an illustrative example: the broken cup. One of the present authors worked for a while in a kitchen run by an older woman who told this anecdote. When she had been a young girl, she was helping her mother wash dishes, when she accidentally dropped a cup, which broke on the floor. “Without hesitation,” as she told the story, and before the young girl could burst into tears, the mother had taken another cup, thrown it on the floor, and said, “See? It doesn’t matter.”

            This is a nice, quaint story, and it seems simple in its outlines. But how many people would have had the wherewithal to respond similarly, and without hesitation? In what ways is the mother’s response a moral act? First, there is the immediate effect on assuaging an anxious child’s sense of remorse. There is the reassurance that the mother is not angry. There is the message that the loss of a cup (or two) is no great tragedy. This is all at one level. But beyond this, there is the way the mother chooses to give this message; after all, any of these things could simply have been said. By breaking a cup herself, she is also showing, “I love you more than material things. I understand how mistakes can happen. I break things too.” By performing this act, the mother also communicates an indelible message (it was recounted to this author when the woman was in her sixties, who had obviously remembered it, and he heard it twenty years before now). It was an enduring act of moral education, whether consciously or not. Moreover, and more subtly, there are the effects of this message, and choosing this way of expressing it, on the mother herself; she is also reminding herself (or even perhaps just realizing) that material things don’t matter, that anger is rarely the proper response to a child’s innocent mistake, and so on. Perhaps she reflected upon her own act, taken so spontaneously, and discovered something new and unexpected about the essence of the moral situation at hand, or about herself as a person. Perhaps she became a better person for it, and better able to confront similar situations in the future.

            If indeed she acted “without hesitation,” then it is inconceivable that she considered all these dimensions with forethought before acting. Rather, this example shows how a range of prior moral experiences, a general set of moral dispositions, a situation that calls forth a certain emotional empathy, and a sudden inspiration, can all combine to foster an act of moral genius. Certainly an appreciation of the implications of the act not only for the immediate problem, but for its enduring effects on the parties concerned, including the agent herself, might often come only through hindsight. But our point here is also that the considerations that led to this act cannot be summarized in any simple moral theory, or even less a calculus, that can either explain the act or justify it. While it is only one kind of moral act, it does happen to be a real example, and it illustrates nicely the point we are making about conceiving ethics as a complex practice in which several moral dimensions can be simultaneously present.



            Despite the importance we are placing here on the dimensions of moral deliberation, moral learning (one’s self), moral teaching or exemplification (with others in mind), and moral self-formation, from the perspective of the moral agent, there is a crucial way in which each of these considerations involves other persons as well. Social interactions are also dimensions of a practice as it exists and evolves over time. In our view, it is very important to consider these in the most natural ways: we are born into a world we do not make; others treat us ethically before we know how to act this way ourselves; our primary introduction to ethics is normally grounded in concern for the well-being of others; we often have ethical responses to others before we even have a language in which to describe them, let alone justify them. In all of these cases, and others, a relational approach seems the only one that can account for how we actually acquire the capacity for ethical conduct.

            First, moral deliberation often involves others in the process of how we reflect upon and decide what to do. Much of the time, this is social in an obvious way: we communicate with others as a way of reaching greater clarity or determination about what we should do. Both aspects are crucial here: clarity in the sense of intersubjectively working through the moral considerations until we reach a decision about what course of action is best; and determination in the sense that reaching this understanding in an intersubjective way can give us greater confidence that the course of action we have identified is a legitimate one. The support and encouragement of others may play a crucial role in animating our capacities; sometimes we are inspired, and taught, by their ethical example. But here again these actual social interactions may also take the Vygotskyan form of internalized deliberations that do not apparently involve others — our deliberations seem to be entirely personal and self-determined — yet which obviously derive from previous conversations with others, in which their voices and perspectives are represented in one’s own internal deliberations. Often this dynamic is what we call “conscience.”

            The role of others in moral teaching or exemplification is fairly obvious and does not need to be belabored here. Examples such as the broken cup make this kind of involvement clear.

            Moral learning and moral self-formation also involve others in the development and maintenance of one’s moral identity and agency. As Arendt and others have pointed out, this begins from the time one is born into the world (a condition she calls “natality” (1958). Even if later one becomes primarily autonomous and self-directed in one’s moral choices, the capacity to be such, and its particular narrative character, are grounded in the relations one has had, and continues to have, with others. The moral experiences and narratives that one has encountered personally, that one has heard about, and that one has shared with others, all go into the complex narrative told to one’s self in which one figures as a moral agent.

            On another level, the very capacity to act morally is tied up in many ways, only some of which the agent may realize, with a set of relations to others. Crucial moral qualities such as courage, determination, reflection, patience, integrity, and so on, even when they are at some stage internalized as driving forces of personal character, are in fundamental respects other-regarding as well. Moral theorists like to say that ethics is what you do when no one else is watching you; and at the moment of action this may be true. But the emergence of these capabilities, the examples from which they derive their importance to us, their imaginary force as if we were being seen and judged by others, all enter in subtle ways to moral deliberation and action even when they do not seem to be primary considerations. If ethics is a social practice this must be so, because it is only through experiences of participation and exemplification involving others that we master the capacities and dispositions of moral agency in the first place.

            Finally, other people engage us in many moral situations as the people whom we see concretely affected by our actions. Their presence to us often constitutes a calling-forth that draws from us sentiments and motivations that we had previously not experienced. This relational quality runs throughout the considerations just discussed here, but they are especially salient, we believe, in the simple moment when a look, an appeal, a moment of vulnerability in others suddenly opens a moral horizon, or a feeling, that in some sense comes to us; it is not a response we impose on the situation as much as an effect it has upon us that makes us, in a real sense, more capable of a moral response than we had been before.


            The proposed framework here, illuminating some of the dimensions of conceiving ethics as a practice, is indebted to key elements of Wittgenstein’s philosophy, especially his later philosophy, but is not exclusively Wittgensteinian. It is indebted as much to theorists who have also tried to decenter the ethical subject, particularly Foucault, and who have tried to problematize the idea of a stable, ethical self. In the view outlined here, there is no “method” to ethical deliberation and no single way in which the ethical agent arrives at a proper course of action. Elements of self-formation and learning interact with elements of emulating others and responding to their conceptions of us, which interact in turn with elements emphasizing our role as ethical exemplars or teachers for others. In the context of practices, when one is teaching, when one is learning, and when one is working on improving one’s own practice, cannot be easily separated. This is true of the practice of ethics also. Because the dimensions of this complex practice work themselves out differently in every concrete instance, any efforts at justification beyond an accounting of the considerations in this case, are highly artificial. Some people are generally honest, or generous, etc. But if you were to ask them why they were so, you would probably get something more like a personal narrative or an autobiography than what philosophers think of as a justification. Or you might simply hear, “This is just the way I am” (or this is just what we do, where the “we” refers to a particular family or community).

            What emerges from this map is a picture of a network of relations, present, past, and future, of which the ethical agent is, has been, or anticipates being a part. Our ethical identity is formed in the dynamic of how we treat others, how they treat us, and how we see them treating each other. This dynamic informs, influences, and sustains us as ethical subjects; in our responses we are often capable of more than we know or can articulate. This network of relations, “overlapping and criss-crossing,” represents the inseparability of questions of ethical conduct and questions of ethical teaching and learning. As Wittgenstein argues is the case for language, so we argue here for ethics: the essence of understanding a human practice is in understanding how it is learned.



Hannah Arendt, The Human Condition (Chicago: University of Chicago Press, 1958).


Norman Malcolm, Ludwig Wittgenstein: A Memoir (New York: Oxford University Press, 1958).


Ludwig Wittgenstein, Culture and Value (Chicago: University of Chicago Press, 1980).


Ludwig Wittgenstein, ”A Lecture on Ethics” in J. Klaage and A, Nordman (eds.) Ludwig Wittgenstein: Philosophical Occasions (1912-1951) (Indianapolis: Hackett, 1993): 115-155.


Ludwig Wittgenstein, On Certainty (New York: Harper, 1969).


Ludwig Wittgenstein, Philosophical Investigations (Oxford, Basil Blackwell, 1953).


Ludwig Wittgenstein, Tractatus Logico-Philosophicus (New York: Routledge and Kegan Paul, 1921).