Ethics: Wittgenstein’s Radical Ethics

§1. Wittgenstein’s Radical Ethics

My aim in this paper is to show how a certain kind of understanding, what I will call “I-you” understanding, lies at the heart of Wittgenstein’s later philosophy. (Sometimes I speak about the I-you perspective.) That understanding is, I will suggest, essentially ethical. Hence, although it is rarely explicitly concerned with ethics, ethical concerns are nevertheless fundamental to Wittgenstein’s later philosophy. I will try to show this through considering some central aspects of Wittgenstein’s later work on our understanding of another person. Wittgenstein’s aim there, I will claim, is to enable us to see the importance of such understanding, in spite of our attempts to conceal it from ourselves. In doing so, I will suggest that the central aim of Wittgenstein’s later work on the mind has in this respect been misconstrued, and its ethical aspect eclipsed. His aims there are far more radical than we might otherwise think.

By “I-you” understanding, I do not mean some specific “kind” of understanding that is hard to detect. Rather, I am thinking of the understanding aspect of an (in a non-philosophical sense) fundamental aspect of our being that we fail to recognize because of its obviousness. We overlook it because it is always before our eyes (PI §129). Still, the difficulty of recognizing I-you understanding is not just a habitual failure; acknowledging it is the most central moral task. I-you understanding is the understanding that “is there” when two persons address each other in the infinitely many ways that this can take place. I will bring into focus some aspects of this understanding.

Though I-you understanding cannot be said to be hard to detect it has some “characteristics” that makes it prone to be overlooked. First of all, it cannot be accounted for by formal means. A person can look at another person in “that” particular way and in countless other ways, but the intelligibility of their understanding each other cannot be explicated in terms of any sort of “grammar” of glances. All the special ways of looking at each other already take place within understanding. This means that any kind of grammar that is established in communication between persons is established within understanding. Philosophers overlook that what can be said about forms of communications; discourses, cannot be said about the understanding that creates them. Any discourse, including language when seen as a discourse, has a grammar. One could call it the regularity of the definiteness of meanings; rules. But understanding itself cannot be understood in terms of grammar and rules. Thus, Wittgernstein’s remark, when he warns philosophers from being obsessed by rules and criteria, is in my view too weak: “For you would still have to apply the rule in particular cases without guidance” (PI §292). My point is that the understanding that makes it possible to apply the rule without guidance, is the one that makes all our systems of communication possible. But it cannot be accounted for in terms of any of those systems. Language and other discourses do not make possible or explain I-you understanding. Rather, they are expressions of, “consequences” of, this understanding. This understanding is something that takes place between persons; between “you” and “I”. I-you understanding is not social. Social understanding is possible only in terms of an established discourse, in which our thinking of ourselves as “us” – an anything but innocent move – becomes possible.

Secondly, I-you understanding is prone to be overlooked because it is through and through moral. Our difficulties with morality; with each other, reveals themselves also as difficulties to acknowledge the language that goes with morality, that is: the language of I-you understanding. I-you understanding and the language that goes with it is prone to be repressed.

That I-you understanding is ethical in character means that it is unmediated. This is important. If I-you understanding would be intellectual in character, it would of course have to have a morally neutral “vocabulary” of some sort; a vocabulary whose meaning-relation to the “mind” of the intending person (who on this view in fact becomes that isolated being that in philosophy is called the subject) must be assessed according to some account or theory of meaning. But I-you understanding is ethical and unmediated or, to use words that Wittgenstein used in connection to the unmediated character of understanding; “primitive” or “primary”. I will show some aspects of this fact and the way I think it is central to the thought of the later Wittgenstein.

In order to understand what I mean by I-you understanding it is important to see that it is not “lacking” in something; it is not something like “non-conceptual” understanding if this is thought of in terms of the ordinary philosophical concept of understanding minus conceptuality. The idea behind this kind of construal is precisely what I question. I-you understanding is that understanding which is able to create among other things language.

One more thing: In contrast to what is routinely and mostly thoughtlessly assumed, I-you understanding is not social in character. Rather, social understanding is a repression of I-you understanding. I cannot demonstrate this sufficiently in the present paper but everything that I say presupposes this relationship between I-you and collective understanding. This is particularly clear in §7 on apology and forgiveness.[1]

§2. Pity

In §287 of the Philosophical Investigations (PI) Wittgenstein asks: “How am I filled with pity for this human being? How does it come out what the object of my pity is? (Pity, one might say, is one form of being convinced that someone else is in pain.)” In the preceding remark he points out that “one does not comfort the hand, but the sufferer: one looks into his eyes.” In caressing someone’s hand one does not comfort the hand but the human being in pain. What is the philosophical purpose of these seemingly trivial remarks? If they have a philosophical point, it must touch upon really basic issues, where we have a tendency to overlook something of crucial importance to our lives. What could it be?

I will assume that Wittgenstein does not mean that in order to get hold of this importance one must be able to master conceptual intricacies that philosophers have so far failed to master. The problem is rather that most people want to hold on to a view from which the things that are “most obvious can become the most difficult to see” (PO p. 161). If what Wittgenstein is talking about really is obvious, how could we fail to see it? How should the above quoted remarks be understood? What is their obvious and yet hard-won point?

Failing to see what Wittgenstein is talking about gives rise among other things to the typical philosophical figure of the solipsist in its different guises. The confusion of the solipsist cannot be cured by common sense for the solipsist does not “disagree with us about any practical question of fact, he does not say that we are simulating when we complain of pains, he pities us as much as anyone else […]”. This is why, for Wittgenstein, “restating the views of common sense” is useless (BB pp. 58-59). Thus, when Wittgenstein points to how we feel pity for each other he does not invite us to think about what we ordinarily mean by pitying someone. He points at the philosophical role of a certain aspect of pitying, an aspect that we (for reasons that are as deep as they can be) do not want to acknowledge.

How are we to get sight of that aspect? How could we come to acknowledge that aspect? In this paper I will present an aspect of our engagement with each other that is present everywhere in our lives but whose importance we fail to appreciate. This aspect is a matter of what I call the I-you understanding: the understanding that is present when two persons address each other.

I am not claiming that Wittgenstein explicitly approached philosophical problems from the point of view of the I-you understanding. But he did make extensive use of this understanding, and my aim is to show how fruitful his remarks become when seen from this perspective. In my view, Wittgenstein’s philosophy is ethical in that it so systematically brings into play the I-you understanding. But the reason why he could never quite explain the role of ethics in his thinking is because the role and ethical character of the I-you understanding was nevertheless not fully clear to him.

§3. The Fly-bottle

In PI §288, Wittgenstein gives what I take to be a clear indication of what he sees as lying at the root of the philosophical problem of skepticism concerning another person’s pain: “if I assume the abrogation of the normal language-game with the expression of a sensation, I need a criterion of identity for the sensation; and then the possibility of error also arises.” Wittgenstein says here that if we do not take into consideration the kind of talk that the expression of sensation and feeling in general are part of, we will be inclined to ask for criteria for identifying sensations and feelings, and that it is this that gives rise to the possibility of error, and therefore also to skepticism.

What is involved in taking the “normal language-game with the expression of a sensation” into consideration? Many would say that Wittgenstein wants to remind us about how we actually use words: we do not normally ask for criteria for identifying, say, pain. But this does not take us anywhere. Everybody, even the solipsist, agrees that we do not normally ask for criteria for identifying pain. This fact, by itself, tells us nothing. We need rather to see how overlooking a certain, essential, aspect of the language-game of expression creates a need for criteria of identity for sensations. This is where the philosophical problem arises.

My aim is to show that we come closer to understanding the matter if we see how the “normal language-game with the expression of a sensation” is an instance of I-you understanding. I-you understanding is, in a certain, non-philosophical way, fundamental but despite this we have serious difficulties with acknowledging it. The difficulties we have with this understanding go as deep as the difficulties we can have with each other: they are moral difficulties.

Another way of saying this would be to say that the extent to which we can repress the I-you understanding shows itself in the kind of language we use. When we deny that we understand another person our language itself changes: we create an alternative language-game based on criteria that both denies the I-you understanding and attempts to legitimize that denial.[2] This also means that we deny the denial, which is a way of expressing what I mean by repression.[3] To the extent that we do not repress understanding, we will not create any system or any particular language-games but, instead, just use words freely. The “fly-bottle” again, is constituted by our adoption of a language of denial. Consider for instance the case of two siblings who do not talk to each other at all because of a quarrel over an inheritance. However, if they happen to meet when a third person is present, they will talk to each other politely enough to conceal their quarrel. Quite obviously the language they talk is heavily restricted by their enmity. Yet, in a certain sense there is nothing odd with the things they say in this polite conversation. How can such a terrible closedness pass without notice? Well, because the kind of public, or collective, discourse they engage in represses I-you understanding.

But is it not biased to say this? Is it not quite typical too, that enmities become visible in what I call public discourse? Yes, but my point is not about what “can” be dealt with in this discourse – anything can – but about how it is dealt with. If enmities turn up in public discourse, it is dealt with impersonally. The third person might try to calm down the adversaries in order to put an end to the quarrel. Or the third person might express her opinion about what she hears, which might lead one of the adversaries into thinking that she is partial. Further, the third person might recommend that the siblings contact a lawyer. And of course the siblings can act in these impersonal ways even if no third person is involved. My point is that as long as there is enmity, the siblings repress I-you understanding. This means that they cannot deal with their inheritance the way two loving siblings would. This of course also means that they cannot talk that way. However, and this is important, they can speak in terms of collective language. This is because I-you understanding is shut out from it.[4]

For the purposes of the present paper it suffices to point out just that whatever kind of collective discourses there are, they are characterized by the exclusion of I-you understanding and the language use that goes with it. Public language can consist of a good-humored jargon, collegial politeness, reconciliation, superficial chatter, deep intellectual discussion or, something I will come back to, apology. What “makes” a discourse collective is when, and to the extent that, it represses an underlying I-you understanding. It should be noted that this repression starts with the individual; in repression one’s self-understanding becomes collective and public language is the language that is available to this self-understanding. Because of our lack of love we do need public language; we do not, for good reasons, want to talk about, say, our marital problems with just any person so we talk friendly and smile happily with our neighbor even if our thoughts are elsewhere. And when there are difficult problems, you may not be able to talk on I-you terms with your spouse either. Instead you, for instance, call each other names, refuse to say anything at all, pretend to be good humored, etc. Shortly put, one could say that public or collective language involves those meanings – whether scornful, favorable or objective – where I-you understanding is repressed. I hope the meaning of this will become clear later on. The important point to see is that if we want to come into touch with I-you understanding, we get nowhere if we investigate collective language.

§4. The Magic of Reality

Is there a clear reason for saying that Wittgenstein’s notion “normal language-game with the expression of a sensation” is an instance of the I-you understanding? I will show how the I-you perspective can clarify the meaning of the above remark from PI §288 as well as other, related remarks from Wittgenstein’s later philosophy.[5]

What is it that Wittgenstein tries to make us see by referring to the importance of the “normal language-game with the expression of a sensation”? In the remarks that precede PI §287 and §288, Wittgenstein undermines the idea that sensations are private, showing the futility of pointing inwardly to one’s pain, how confused it is to think that inward pains could be transferred to outside objects, how it is not our body that feels pain but a person, and so on. Then come the points about pity and about the temptation to exclude the normal language-game with the expression of a sensation.

In PI §296, Wittgenstein’s interlocutor insists that “there is a Something there all the same, which accompanies my cry of pain! And it is on account of this that I utter it”. Wittgenstein asks: “Only to whom are we telling this? And on what occasion?” I want to pause and reflect on these remarks. In general, Wittgenstein tries to ward off our tendency to think in a way where we end up with a twofold picture in which it appears as if only I can know my feelings, with the corresponding compulsion to speak about sensations as if they were inner objects that make sensation language intelligible. With this picture we have no other option but to account for another person’s sensations and feelings by trying to transfer the inner sensation that only I can know onto an external object.

In response to the claim about the “Something there” of pain, Wittgenstein does not come up with a counter-argument. He does not even say that the idea is nonsensical. (Compare this with PI §278: “’I know how the colour green looks to me‘ – surely that makes sense! – Certainly; what use of the sentence are you thinking of?”) He just asks to whom one would say it and on what occasion. What Wittgenstein insists on here is that when a sentence is an instance of the “normal language-game with expression of sensation” it has a point in that it is said to someone. That is what “having a point” is all about. I think that it is along these lines that his idea about language-use should be understood. I do not think that he had any general idea about language-use as some kind of criterion for deciding meaning. In fact he seems to explicitly reject that idea.[6] His concern was, rather, to point to cases where what is said really has a point, and in almost all his examples he refers to cases where one person addresses another – or then he invites the reader to do so: “Just try – in a real case – to doubt someone else’s fear or pain.” (PI §303)

Wittgenstein rejects the idea that a natural expression of pain would be an outburst of a subjective sensation of pain. Rather, natural expressions of pain are an integral part of the language-game of pain. This language-game is characterized by the kind of sympathy for another person that is exemplified by pity. Pain is manifested by the person who feels pain, not by body-parts or sensations (§302), and pity is “one form of being convinced that someone else is in pain” (§287). This means that “pain” becomes intelligible in the kind of language-game where expressions of pain and pity are at home.[7] This is a language-game where “I” communicate with “you”: I ask you questions, you answer; you tell me something, I respond; I see something and you see it too; you suggest something and I hesitate; I say I’m hungry and you suggest a restaurant, etc. The understanding that manifests itself in this communication is what I call I-you understanding and it contains much more than words (though the tendency to say that what is “more” here is bodily in nature is just as misleading as saying that it is mental). My suggestion is that Wittgenstein’s notion of the “normal language-game with the expression of a sensation” refers to an aspect of I-you understanding. This means that I will take “the expression of a sensation” to be something that can only be understood in terms of an “I” communicating with a “you”, and that it does not make sense to try to account for it in purely first-personal terms for that inevitably means that one thinks of it as a behavioral counterpart to some inner mental episode. Neither can it be accounted for in the terms of public language as my discussion of apology and forgiveness below, will indicate.

The kind of communication that goes on between “you” and “I” cannot be captured by saying that we are expressing thoughts to each other, or saying that we are expressing both thought and feeling. Let me instead provisionally say that the communication between an “I” and a “you” is from the very beginning “engaged”. Though we may be very differently engaged in each other, there is no such thing as a completely non-engaged, neutral, communication between an “I” and a “you”. Sometimes we explicitly address our engagement with each other, but mostly it is just present in the things we say and do.

The important thing about the I-you understanding is that it is not based on anything more ultimate, such as criteria, facts, inner sensations, brain events, signs, concepts, etc. If I am irritated with a certain person I encounter, this feeling of mine is not grounded or made intelligible by some kind of inner sensation that would accompany my words, or by concepts, facial expressions, gestures, etc. Instead, my feeling of irritation manifests itself in my words, facial expressions, etc. The other person might perceive my irritation but that does not mean that she therefore reads it off from certain facts about my words, voice or face (c.f. PI §§285 and 292)[8]. It is because the other understands me that she can make sense of my words, voice and facial expressions at all. It is only under very special circumstances that we try to judge from the look of someone’s face what she thinks. And if there were nothing about the other’s face that I understand without interpretation, then neither could I perform any interpretation.

“But how could there be understanding if there are no common criteria for what does and does not constitute understanding?” We have serious difficulties with accepting the thought that we understand each other “just like that”: it seems like magic. And so we create an idea of reality in order to avoid an illusory magic. Here, however, Wittgenstein reminds us that even in the cases where we do use words according to rules we “would still have to apply the rule in the particular case without guidance” (§292). When the importance of I-you understanding is not seen, this remark becomes completely misunderstood. It is indeed true, as Saul A. Kripke says that Wittgenstein’s remarks on rules and rule-following constitute “the most radical and original skeptical problem that philosophy has seen to date” (Kripke 1995, p. 60). But one could also say that this skeptical problem reveals with striking clarity how hopeless problems we will run into if we ignore the fact that “I” understand “you”. I take Wittgenstein to mean that “understanding” involves that there is “something” beyond the rules (grammar, logic, language, etc.), that is: someone who cannot be captured by rules because she is a living being who, among other things, uses and establishes rules.

In the paragraphs preceding 288 Wittgenstein points out that expressions of pain are connected to living beings; in connection to living beings ‘[a]ll our reactions are different’ (§284). A living being is an understanding being which means that it expresses, and hence understands expressions of, pain. In the case of human beings this involves language, but the language-game of pain depends on our being living beings. If we were odd creatures with a solipsistic mind, it would indeed, as Kripke believes, be the case that it could never be established why one interpretation of a rule would be more correct than another. But, given that we understand and misunderstand each other, what establishes the meaning of a rule is what someone, say a teacher, had in mind with it. If the pupil does not understand, there is no way of getting her to understand it simply by virtue of further specifications of the rule. This is not a problem about rules of grammar but of understanding each other. In I-you understanding we have infinitely many ways of trying to get another person understand what one means, but there are no rules that, by virtue of their being rules, will lead to understanding. It is the one person who will have to understand the other person. Such is I-you understanding.[9]

§5. Sensations and Objects

With the above remarks in mind, let us now return to what Wittgenstein said in §288: “if I assume the abrogation of the normal language-game with the expression of a sensation, I need a criterion of identity for the sensation”. As noted, it does not make sense to assume that Wittgenstein is here thinking of first-person expressive statements in the sense this is usually understood. Rather, what he means is that if we ignore the way we are engaged with each other, and the understanding that goes with that, we will end up demanding criteria for identifying sensations and as a result doubt will creep in. Why is this?

If we are blind to the way our understanding of sensation words (among other things) is inseparable from our understanding each other, we are left with an abstract word; a word that somehow “has” a given meaning, and an isolated sensation that it is supposed to refer to. This sensation is also what is supposed to give meaning to the “corresponding” word. When this is the case sensations will appear as somehow less straightforward referents than physical objects, for one cannot compare the word with the qualities of the object it refers to (and for instance McDowell 2002, pp. 286-87, does see things this way). Thus, one faces the task of finding rational criteria for applying given sensation-words to the, as it is assumed, corresponding sensation, and it is this task that unavoidably brings in doubt.

Here one could give many examples, such as the total lack of philosophical consensus about what those criteria are, how they should be assessed, and about whether there are any such criteria to assess. More importantly, doubt enters the picture because sensations are in the above picture construed as objects. The problem is caused by the grammar of this move: it does not depend only on the thought that sensations are mind-independent givens. For problems to arise it suffices that one thinks that the intelligibility of sensations presupposes concepts.

Concepts presuppose criteria, but it is uncertain what criteria (behavior, facial expressions, exclamations, etc.) count as “necessary” criteria for a sensation such as pain. This uncertainty can persist only because behavior, facial expressions, exclamations, etc., that is, the “normal language-game with the expression of a sensation”, are seen as signs for the assumed sensation in the nervous system. What on this view would decide the issue is whether a sensation is really there, not whether a human being is really in pain. But all philosophical suggestions about how the sensation would decide the issue have produced only nonsense or empty speculations about causality.

The ethically indifferent inclination of this view should also be obvious. For the issue here is about concepts, inner sensations and ways of truthfully assessing necessary conditions for having a given feeling. If, by contrast, we consider the issue from the point of view of I-you understanding, the whole issue of someone feeling pain is moral. In those exceptional cases where the question whether someone really feels pain arises, it is a moral question: is this person in pain or is she pretending it for some reason? From the point of view of I-you understanding, it is absurd to claim that deciding whether the expression of a sensation or emotion is genuine is an intellectual problem where appropriate criteria are evaluated. Relating to another person’s expressions of pain or fear is not about applying certain concepts in order to assure oneself of the existence of a relevant “episode of consciousness”, to use an expression of McDowell (McDowell 2002, p.283). Rather, relating to another person’s pain involves I-you understanding. There can be specific uncertainties and certainties within this understanding, but the general fact that I understand you is neither uncertain nor certain: I just understand and so do you.

The most important aspect of “uncertainty” in the I-you perspective is when I close myself for the other, when I repress the fact that I understand you. Thus, the words that Wittgenstein uses in order to dispel skepticism are not coincidental: their illuminating power has its source in their moral character: comforting the sufferer, feeling pity for another person, reactions of sympathy, lack of sympathy, etc. (PI §§286-87, PO pp. 381-83.) It is important, too, to realize that “moral” here does not mean “morally justifiable action” or anything else of the sort discussed in moral philosophy, churches or ordinary life, though one encounters what I mean by moral understanding all the time in everyday life, for our addressing each other in the sense of an I to a you is a moral issue. The problems that we have with each other are also moral; not philosophical: “Those cases where the inner seems hidden from me are very peculiar. And the uncertainty which is expressed in this is not a philosophical one; no it is practical and primitive.” (RPP II, §558)

§6. The Nothing of a Something

I have suggested that Wittgenstein’s aim is to show how meaning resides between us, in the I-you understanding. That understanding is continuously present in Wittgenstein’s thinking. When he asks the reader or his interlocutor questions like “what use are you thinking of?” the point is not about what specific answer you might give, but about the fact that in order for there to be a use it would have to make sense when it is said to someone. He never treats a candidate for meaning as a claim, that is, as something in this sense general, in order to provide arguments for or against its meaningfulness in general.[10]

Though it would be odd to call I-you understanding a “kind” of understanding, it could be said to be a way of understanding in the sense that it is “present” only in certain kinds of encounters between an I and a you. (I mean: when the judge reads the sentence to the convicted person, or when a philosopher lays out her argument to another philosopher we do not have I-you understanding as far as the reading or the arguing goes.) Wittgenstein most often deals with his themes in a way that involves two persons in conversation. He tries to reach out to the understanding between him and another person. He does not argue for statements or positions. He does not build up a network of reasons that proposes to be compelling to anyone who is reasonable. He tries to make “you” see things in a different, more honest, way. This is not a difference only in address but also, through this, in “logic”. When seen in the new light, philosophical theses will be revealed as what they are; only “structures of air” (Luftgebäude, PI §118). Any account of Wittgenstein’s philosophy that takes him seriously must emphasize this aspect of his thinking.

Wittgenstein makes it clear that ignoring what I call the I-you understanding is connected to an inclination to fall into the kind of thinking where giving criteria for correct use of concepts becomes central. This brings with it the possibility of systematic error concerning another person’s sensations and feelings. Moreover, the idea of criteria for correct use is just the other side of the idea of an inner sensation that is supposed to legitimize pain-language.

The idea that an inner experience is indispensable is an expression of a craving for justification; for a Something on account of which any mental concept can have meaning. This Something can take on any shape that enables us to create the illusion of an independent justification. It can be a sensation, a fact, a concept, a criterion, a phenomenon, a principle, etc. Or, the independent justification could be “common usage”. In the present paper I cannot discuss either collectivity or the question why the need for a mediating Something is so pressing. This would involve discussing repression, understood as a denial of conscience, not, as Freud would have it, as a psychic mechanism. Here I will focus on the way in which this Something shows itself in philosophical discourse and how Wittgenstein’s thinking helps us get sight of it.

In the I-you understanding someone’s feeling pain is “not a Something, but not a Nothing either” (PI §304). That a person feels pain is not a Nothing because it matters to us (“us” in the sense of you, you, you…). And what matters is that someone is in pain. We suffer from the other person’s pain, we try to alleviate it, etc. The pain does not become intelligible as pain because we conclude that the other has an inner sensation or because we conclude from her grimaces, groans and words that she is in pain – or because we have become familiar with the concept of pain. The pain becomes intelligible because another human being behaves and talks in such and such a way. It is important to see that we do not read off the pain from anything (PI §§245, 246, 292); we respond to a human being. It is not words, facial expressions, etc. that “provide” us with the information necessary to conclude that another person is in pain. It is the fact that another person has pain, with all that goes with that, that makes pain intelligible. I understand a person, not geometric constellations in the face, sonorities of phonemes or meanings of signs. A word or a facial expression is not a sign in the sense in which philosophers usually think for there can be signs only where there already is I-you understanding. And there is no relevant episode of consciousness in virtue of which pain becomes intelligible. Pain is not a Something because there is nothing besides the human being in pain, that “makes” the pain intelligible. No object is involved. Thus, if we nevertheless try, as philosophers do, to “construe the grammar of the expression of sensation on the model of ‘object and name’, the object drops out of consideration as irrelevant.” (PI §293)

The whole discussion on pain, and sensations and feelings in general, in the PI is intelligible only when it is seen that what it is describing is an ethical discourse. That pain matters to us is an ethical issue and the ways in which it matters to us are the “ways” that make up the meaning of “pain”. To put it in philosophical jargon, the intelligibility of the concept of pain is ethically determined. That this understanding is ethical means that it is unmediated.

§7. Forming avoidance

There is a way of relating to things that is mediated and characterized as one in which we broadly speaking, “use our brains”. By this, I mean that the observations and conclusions we make have such a form that any rational creature that understands the notation in which they are put will also understand their meaning. They are in this sense third-personal, public or objective. If we want something from the things in the world and if we want to communicate this craving to someone else, there is no other way of doing this than by representing these things and what we want from them. The way of thinking that goes with this can be applied in many different ways and so we can also put our expectations of each other in the form of representations. In order for this to work, we need to have some kind of shared notion about what psychological, behavioral and moral concepts mean. So we construct words that allow us to express our expectations of each other (though of course the words are not only about this) on the same model as we construct words for representing things in the world.

Obviously, we need a lot of common concepts in order for society to function at all, and the complexity of modern society shows how effective such constructions, let me call them “social constructions”, are. Nevertheless, these constructions are secondary to the I-you understanding. These social constructions become repressive when they are taken to be primary, thus blinding us to the importance and character of I-you understanding.

“I” and “you” understand each other not because we want something of each other but because we matter to each other, and this is so no matter how reluctant we are to acknowledging this. Let me try to illuminate the difference between concepts working along the lines of social construction and concepts working within I-you understanding by comparing two concepts: apology and forgiveness. My comparison will reveal the sense in which it is misleading to call forgiveness a concept.

I think that the concept of apology can be specified in the way that J.L Austin does in his How to Do Things with Words, that is, by uttering the words “I apologize for…” to the relevant person in relevant circumstances. One could say that the word “apologize” functions only because we, as a community, have given it this meaning. If I, say, offend someone, an apology has been made if I go to the person I offended and say something like: “I’m sorry for what I said. I didn’t mean to…”. The complexities surrounding this matter do not change the basic fact that apologizing is about saying so and so to the relevant person. Here, the words used are essential. This means that the apology is performed by using relevant words. When words are used in this way they can be paraphrased without much loss of meaning and hence articulated in the third person: “He apologized and said…” Everyone will understand what happens on the basis of this third-personal utterance, and they understand it from the words. I say “without much loss” because even in this case the I-you perspective is present, though repressed.

It does not matter much if the person apologizing looks haughty and uncomfortable with what she does. If she utters the relevant words to the person she has offended, she has apologised. The relationship between the persons is in this case not of first importance, for the whole business is to restore the public image of the person offended and it is characteristic of public discourse that it hinges on formal procedures.

With forgiveness things are different. You cannot ask for forgiveness by saying certain words, or by performing certain actions, by behaving in certain socially agreed upon ways, etc. There is in fact nothing in particular that you can do in order to ask for forgiveness. Nor can the other person forgive you simply by saying “I forgive you for …” The need to ask for forgiveness arises in one’s conscience when one feels that one has in some sense betrayed or hurt the other. Asking for forgiveness is an expression of one’s understanding of this fact and the depth of one’s asking for forgiveness depends on the depth of one’s understanding of one’s betrayal or hurt. This again connects to the life that one lives with the person betrayed. Living a life is not only about practical issues, but also importantly about how one understands one’s relationship to the other. This “understanding the other” in its turn is not about the private thoughts one has about the other, but about the way one is present for the other with one’s thoughts. And if one betrays the other, one’s thoughts will give rise to pangs of conscience. Forgiveness expresses a recognition of one’s betrayal and the way it hampers the openness with the other.[11] Living with the other and living with one’s conscience are inseparable: it is all about a way of understanding oneself with the other. Though we long for this openness it is also very hard, causing moral problems: “We could […] say: Hate between men comes from our cutting ourselves off from each other. Because we don’t want anyone else to look inside us, since it’s not a pretty sight in there” (CV p. 46).

It should go without saying that the above observations go equally much for forgiving as they do for asking for forgiveness. Asking for forgiveness and forgiving are thus “played out” in the life of the persons, in the way they understand each other. One might even say: their shared life will show what forgiving means. “Only in the stream of thought and life do words have meaning.” (RPP II, §504) I do not think that Wittgenstein means that words have meaning only in life in contrast to philosophy. This would be odd an in fact quite wrong. What he means is that when we are philosophizing, for this is what he is doing and he is addressing philosophers, we should not involve ourselves in the hopeless task of construing theories about “necessary preconditions” for words to have meaning; theories about necessary conditions for intelligible discourse. We should rather see what role words have when people are addressing each other.

The “meanings” that the two people in our above example communicate to each other cannot be paraphrased, because the whole issue is about the way they understand each other. This “understanding each other” is not some special, isolated phenomenon that crops up in connection to “forgiveness” and other similar concepts. Rather, it is what their shared life is all about. The words and sentences they use when they talk can be paraphrased, but what they mean when they utter them cannot because it is not about words. The way in which the use of the word apology differs from the use of the word forgiveness indicates this difference. It is by reflecting on the way language works in connection to forgiveness that we see the ethical foundation of language and the reason why philosophers have had so much trouble in speaking about ethics.

One difficulty with getting sight of the role of the I-you understanding is that language does in part seem to function in a way that is akin to the picture that (for instance) McDowell suggests. The concept of apology, for instance, appears to be precisely a “limiting case” of the object-designation model that he proposes. It seems to have a given meaning that has a point only to the extent that it corresponds to a certain, specifiable state of mind, comparable to an episode of consciousness, which is here presupposed, namely that of regretting.

Austin explicitly wants to put aside this whole issue because it upsets his thesis. For if in one’s inwardness one does not feel any regret, the words one utters when apologizing seem to be empty. In order to fend off the “backstage artiste” that appears here, and in order to avoid the problem of matching inner states with outer expressions, Austin emphasizes what I have called the collective aspect of language: the meaning is in the words, so if you utter the relevant words an apology has been performed, no matter what you say to yourself inwardly. (See Austin 1990, p. 10.) We see how the very functioning of this collective language game depends on not just ignoring but repressing the understanding between persons: No matter what either of the two persons might think of each other, an apology has been performed when such and such words have been said in such and such circumstances. Unfortunately, it is this way of using words that has been the object of inquiry in philosophy. It is this un-ethical conception of language that I see the later Wittgenstein moving away from.

Whether he wants it or not, Austin’s view recreates the classical problem concerning the relationship between inner states and outer behaviour. This is because his account constitutes an “abrogation of the normal language-game with the expression of a sensation” and hence evokes a need for a criterion of identity for the sensation, to paraphrase Wittgenstein. Thus, in the traditional philosophical view, the meaning of the concept of apology presupposes a Something in the subject’s consciousness. It is this feature of the concept of apology and countless other concepts that lends McDowell’s account its apparent plausibility. Austin’s objection to traditional philosophy consists merely in thinking that the words themselves, in their public meaning, can perform the task without paying attention to inner states. From my point of view, this is an unimportant different. And on top of that, Austin’s suggestion is obviously implausible.[12]

I tried to show above that in forgiveness language works in a way that is different from apology. In forgiveness and in the whole “life of talking” that the persons in question are involved in, words are not used on the basis of their having essential (definitional, standardized, formal) meanings. Of course, some kind of regularity of meaning of words must be there but in the I-you understanding speaking is not about using standardized meanings combined in grammatically appropriate ways. Instead, words are aspects of reaching out and opening oneself to the other. This is the sense in which addressing another person is as such an ethical issue.

To the extent that one dares to be open with the other, one’s language use moves in a direction that I indicated with my outline of forgiveness. And to the extent that one closes oneself for the other one’s language use restricts itself to the kind of use I exemplified with my outline of apology. In the latter case the formality of the words used is the other side of taking a social attitude, such as politeness, and of being impersonal. This does not signify an absence of the ethical, but a repression of it.

My example with forgiveness aims at pointing at the sense in which everything we say becomes meaningful only in terms of I-you understanding. We often formalize our understanding for instance in such a way that we pretend that we do not understand the I-you address, and sometimes – as in a court of law – our practices presuppose such formality. We can also think of the way a rigorously formal language like, say, the one connected to courtly manners, precludes openness between persons.[13] But the formality is really an aspect of collectivity, which should become clear when one sees that to keep rigorously to a dialect is equally pernicious to openness.

Formalizing understanding occurs in many different ways in human life, but the common denominator for formalization is that what matters is the agreed upon formal meaning of a given signification. In this broad sense, formalization means that some relation is agreed to obtain or some procedure agreed to be valid simply in virtue of an agreed upon meaning of signs. In such cases, pre-established criteria decide the meaning of words, practices, habits, etc. Sometimes giving these meanings is explicit and (quasi-)legal or normative: “You can vote in an election only if…”, “You are supposed to address her as…”, “When the judge enters you are supposed to…”, “in this part of the world we always … when …”, and sometimes implicit expressions of value, for instance when the value of gardening, education, exercise, etc., is referred to.

Much of this formalizing aims at securing certain behaviour, facilitating intercourse and at veiling and moderating moral problems by determining that certain things are thus and so as a matter of course. We want to exclude certain aspects of meaning from entering the stage. My point with comparing apology and forgiving was to show how the relative, linguistic specificity of apology; the formal determination of its meaning, is the other side of excluding I-you understanding. You “just proceed” in a collectively established and thus generally accepted way and you will have performed an apology, without having, apart from the formal procedure, to engage with the addressee that much – or this is the repressive illusion that we take as “reality” insofar as we collectively try to uphold it. In connection to forgiveness we saw that it has no such specificity. This does not mean that it would be more uncertain what we mean by that word – quite the contrary. It means that what we mean by that word points to our life; to the way the persons concerned live with each other. It is this that gives sense to forgiveness. If the persons fail, it is not that they fail to live up to the meaning of a concept: they fail each other.

The I-you understanding is not simply absent in apology. After all, the whole point with apology is about an “I” saying something to a “you”. What I try to attach attention to is the way in which we try to escape a difficult aspect of the I-you understanding, in this case forgiveness, by formalizing it. Formalizing means here to obscure the I-you understanding by making the exchange more formal, that is: dependent on words. Saying sorry to another person in a formal kind of way – and the everyday practice of saying it in passing can be one way of formalizing it – “feels good” because asking for forgiveness is so hard. The addressee is equally often prey to this repression and also prefers the formal address. Mutuality as such does not equate to acknowledging I-you understanding.

One could say that in public language the I-you understanding is not eliminated as such; it is rather that I-you understanding is repressed, which means; substituted with public language, in the respects that we find it hard to take. It is always possible to come up with a substituted form, and often the difficulties with I-you understanding begin already with the way we address each other, and so in many languages the plural second person is used in polite discourse.

In §§654-659 of the PI Wittgenstein talks about an important sense in which we ought to think of language as something “primary”; as a “proto-phenomenon”; as something that is not in need of explanation. Seeing language in this way involves seeing how feelings (“and so forth”) are to be regarded as ways of “looking at” (Betrachtungsweise). He says this when he raises the question why it is that I want to tell someone about my previous intentions and wishes. I suppose that he means that we do so because we matter to each other. This involves that it matters what I have previously thought: “I reveal to him something of myself when I tell him what I was going to do.” (PI §659) This revealing is itself of great moral importance; it is part of what it is to love another person and the more we dare to love someone, the more we will open ourselves to each other. In hate, by contrast, we, as Wittgenstein notes, cut “ourselves off from each other” because “we don’t want anyone else to look inside us, since it’s not a pretty sight in there”.

Not only hate but all kinds of moral difficulties that trouble us incline us to cut ourselves off from each other and this is internally related not only to the language we speak but also to the way that we look at language. If we are embittered or envious we do not want to reveal ourselves to anyone. But not only this: if we have allowed bitterness or envy to form our lives, we will also question, or even ridicule, ideas about human beings understanding each other.

Wittgenstein speaks about the primary character of language in many places and in different ways, and sometimes he speaks about “primitive language-games” (LWPP II, p. 39), relating these ways of speaking about language to what he calls “primitive reactions” (RPP I, §§915-916, LWPP I, §§133-134, PPF §§ 161, 289). It has been said that Wittgenstein is not all that clear about what he means by these ways of speaking. It is for sure not possible to deduce what he means here by way of a detailed textual analysis of the passages with their surroundings. However, I do not find these remarks particularly obscure; quite the contrary. I think that he points at the immediate, ethical, character of understanding between persons and he tries to make us see the language that goes with it.

One of the decisive short-comings of philosophy – if one should not rather say that philosophy as such is a huge complex of short-comings – that the “philosopher delivers the word to us with which one //I// can express the thing and render it harmless.” (PO p. 165) But this occurs in ordinary life too. When I-you relationships are discussed in terms of apology this is a way of rendering forgiveness harmless and involves that the focus will be on what the criteria of making an apology are. We have here the whole business with which philosophers are occupied. In the unmediated, “primitive language” of I-you understanding, this is not possible. Here we will face what is anything but harmless: the openness between you and I the way it shows itself, in its hardship and love, for instance in forgiveness.

§8. Conclusion: Ethics and Interpretation

The meaning a you has for me is not as it were packed into concepts (or facial expressions, gestures, etc.) and then unpacked by me when I interpret the package of information that is, supposedly, delivered to me by you. Neither do I “have ‘meanings’ in my mind in addition to the verbal expressions; rather language itself is the vehicle of thought.” (PI §329) For if I had meanings in my mind, then of course I would have to translate them into words in order to transmit them to you. In this case you would of course have to interpret my message. This, still further, would mean that interpersonal understanding would depend on intellectual abilities. In this picture anything communicated must be a thing in the sense that it must be an “object that we can designate and classify”, to put it in the words of McDowell. Otherwise it would be completely impossible to decode the messages that the other is sending me. On this view, “external objects” would seem to have a certain robustness in that I could here compare the meaning of the message with my own experience of the object described in the message. (I say “seem” because in the end this view makes nonsense of everything.) Sensations, by contrast, become “limiting cases” because I cannot compare the linguistic expression of sensations with any objects of experience. The object “must”, on this view, nevertheless be there (see McDowell 2002, pp. 286-87).

It is the obsession with this object that pulls philosophers in the direction of both subjectivity (“what is in the mind”) and objectivity (“what is there independent of the mind”).[14] Thus, those who want to avoid the perils of subjectivism cannot but cling to some  objective “foundation” and so, as McDowell puts it, the “very idea of thought – the exercise of intellect – presupposes a notion of objectivity” (2002, p. 185) while speaking to another person is to be understood as “correct moves in a language-game” (McDowell 2009, p. 93). When this is said, then of course ethics must also, on pain of being meaningless, be in some sense objective and so, McDowell again, ethics is like science in that “in both of them it can be rational to say of a conclusion that logos itself compels it […]” (2002, p. 186). Logic and words “compel” us to do something! This is the way things look when I-you understanding is not acknowledged.  But for me to understand your pain I need no objectivity and no moves in any a language-games.

That language itself is the vehicle of thought means that insofar as we “are”, dwell, in the I-you understanding, words can become like cries and sighs, they can become hard to say, they can caress your soul, etc. Words become as it were extensions of the way we look at each other, touch each other and so on. (C.f. PI §§543-46, PPF §§83-85.) This is why saying “I understand you” means something entirely else than saying “I understand the words you speak”. By saying this I point at something I think constitutes the fundamentally ethical character of Wittgenstein’s philosophy.[15]