A Complicated Form of Life
A Complicated Form of Life


This paper explores Wittgenstein’s remark at the beginning of the second part of the Investigations that an animal such as a dog cannot hope, nor feel grief. Here Wittgenstein presents a cogent argument against the idea that these feelings are essentially ineffable, since they can only be felt by those who can talk, by those who engage in a rather complicated form of life that can only be set up through language.

Table of contents

    One of Wittgenstein’s strongest arguments – and nonetheless one of his less explored arguments in the secondary literature – against the idea that I can never let others really know what is going on within me because the most I can do, for instance, is to say I’m happy and smile (that is, show an outward sign of my happiness), never show my happiness itself, is that there are feelings we can only have after we have mastered a language, after we have already learned how express them. This is what Wittgenstein remarks about hope. “Can only hope those who can talk?,” he asks at the beginning of the second part of the Investigations (Wittgenstein 2001, 148). And his answer is quite simple: “Only those who have mastered the use of a language. That is to say, the phenomena of hope are modifications of this complicated form of life” (Wittgenstein 2001, 148).

    But what does he mean by “complicated form of life”? To answer this question, we have to remember that Wittgenstein acknowledges two kinds of criteria for ascribing thoughts, desires, feelings and experiences in general to others. The first one is behavioural. The second one is verbal. It is also important to note that both of them are always measured against wider surroundings. Thus the cry is a criterion for pain. But if a baby starts crying without any evidence that it is hurt (or that it has some illness that causes pain), then we will probably say it is not a cry of pain, that it is more likely to be a cry of hunger or thirst. For both are behaviorally displayed by the baby in the same way. Or better, they all are behaviorally displayed in the same way when we consider a rather short section of time. For if the baby is crying with hunger, it will not quiet down when we give it water. Nor will it be pleased with food if it is crying because of thirst. And if nothing satisfies it at all, we shall probably think it was not hungry, nor thirsty, but actually in pain. Besides the general context, we also take into consideration what it is that cries (if it is a what or a who). If it is a living being, then we shall say it is in pain. But if it happens to be a machine programmed to make the sound of a cry when we press a button, then we won’t say it is in pain. We will not accept its behaviour – no matter how lively it is – as a criterion for pain. For it does not satisfy the basic condition of being something that can feel pain. So we have two criteria, a general context and a basic condition that have to be met in order for us to ascribe pain to others.

    Why do we take someone’s verbal expression “I’m in pain” as a criterion for saying “He is in pain”? Can’t he who says “I’m in pain” be lying? But the fact that the verbal expression of pain can be insincere doesn’t show that we can’t rely on them to ascribe pain to others (in fact, we should note that the person who wishes to deceive us in making us think that he is in pain exhibits precisely what we call a ‘pain behaviour’), for it is anchored in non-verbal behaviour that we have no choice other than to take it as being authentic. Although an adult can always lie when he says “I’m in pain” (that is, when he exhibits a verbal pain behaviour), we can’t easily say a baby may always be faking when it exhibits non-verbal pain behaviour. In fact, that would be as misguided a move as pointing to a sample of blue and saying “That colour is not blue”. Precisely because the baby’s non- verbal pain behaviour is a paradigm for the application of the word ‘pain’.

    If we can trust non-verbal pain behaviour (if it doesn’t make sense to mistrust non-verbal pain behaviour), then we can trust verbal pain behaviour (then it doesn’t make sense to mistrust verbal pain behaviour). They are tied together by an internal relation. Linguistic expressions of pain – and that holds both for those in first person as for those in third person – are taught by reference to non verbal pain behaviour. Therefore we don’t require that a person who says “I’m in pain” also behaves as if he was in pain so that we believe him – especially if he is already an adult. It’s not because he doesn’t behave as if he was in pain that we cannot say he is in pain. The verbal expression of pain is enough. It replaces non-verbal pain behaviour. It is so perfectly fair to pass from a calm verbal pain behaviour to “He is in pain”, as it is to pass from a wild non-verbal pain behaviour to “He is in pain”.

    So let’s come back to the expression “complicated form of life”. When we teach a child to say “I’m in pain” instead of crying, what we do is modify its behaviour. Furthermore, after teaching it to say “I’m in pain”, it can eventually say “I have a throbbing pain”. That is, when we teach it to talk, we give it the necessary basis for it not only to express, not only to manifest its pain, but also to describe it. We therefore expand its possibilities of expression. We expand the complexity of its behavioural repertoire. But that of course doesn’t mean that we necessarily expand the universe of its sensations. Just as we don’t expand it when we teach the child to say “I’m in pain”, we also don’t expand it when we teach it to say “I have a throbbing pain”. To deny that it could feel throbbing pains before learning to say “I have a throbbing pain” would be like denying that it could feel pain before saying “I’m in pain”. In fact, it would be like denying that it could have five fingers in its hand before learning to say “I have five fingers on my hand”.

    But with the feelings of grief and hope, for instance, matters are quite different. They are not in the least like the sensation of pain. Although we can say we are feeling grief or hope at a given moment (just as we can say we are feeling pain at a given moment), that is, although we can feel grief or hope, there isn’t much sense in saying we experience these feelings before we can express them. After opening the second part of the Investigations discussing hope, Wittgenstein makes exactly this remark in relation to grief:

    “But don’t you feel grief now? (...) The answer may be affirmative, but that doesn’t make the concept of grief any more like the concept of a sensation. – The question was really, of course, a temporal and personal one, not the logical question which we wanted to raise.” (Wittgenstein 2001, 148)

    Both hope and grief are feelings that a newborn child cannot feel. The same holds for all feelings that presuppose relatively complicated forms of life. If a father gives one of his teenager sons a new bicycle and the other one an old bicycle, it is possible that the second one will feel envy (or jealousy) towards the first one. He may even feel envy towards his brother without giving any sign of it. If in some other occasion he had expressed envy that could be remembered in order to ascribe envy to him despite his calm and indifferent behaviour. But if this father now gives one of his newborn nephews a big teddy bear and the other one a matchbox toy, we will not in this case say that the second one would probably feel envy towards the first one. Even if he strikes him in the face or bites him.

    Therefore, we can say that grief, hope and envy are not feelings that babies have within them from the moment they are born. They presuppose a form of life too complex to be experienced by a newborn. Here we meet a fundamental argument against the idea that we can never let others really know what we are feeling since we can only give them external signs of our private experiences, never put them on the table for free public examination. For if we can only feel grief or hope after “mastering the use of a language” and after making our entry into a relatively “complicated form of life”, then it doesn’t make much sense to say that they are ineffable.

    So grief and hope are private in the sense that we can hide them, but not in the non-sense that we can’t show them. Furthermore, even if a newborn child can feel a throbbing pain without being able to express that it is a throbbing pain it is feeling (for we can’t tell by its cry whether it is a throbbing pain or not), we can’t say that an adult can feel grief, hope or envy with phenomenal properties he can’t express. The truth of the matter is that he can say how deep his grief is. He can tell us the intensity, the strength and the duration of his grief. In fact, he can indeed manifest the intensity of his grief (and that holds also for his envy and his hope) in his actions.

    To say that a pain can have phenomenal properties that a person may not know in a given moment how to express is not to say that a pain can have phenomenal properties that deflect any possible expression. The way we learn to express pain is different from the way we learn to express hope. But that doesn’t entail either that we can’t express pain, or that we can’t express hope. When we learned the concept of pain, what happened, for instance, was that we accidentally hurt ourselves, started crying and then an adult asked us to stop crying and simply say “I’m in pain”. We don’t learn to feel pain, only how to express it, only how to name it (better, only to replace our behavioural expression of it by an appropriate verbal expression). It isn’t a sensation that exists only in our specific form of life and that we could simply cancel if we adopted a different form of life (unless it happened to be a form of life in which we never got hurt and never got painful diseases). The fact that pains and throbbing pains could already be felt before we could express them doesn’t mean that we can never get to talk about them, it doesn’t mean that they are essentially ineffable. Stones already existed before we started to talk about them. And they aren’t for that reason indescribable.

    But when we learn words like ‘grief’ or ‘hope’, what we go through is rather different. Neither the feeling of grief, nor the feeling of hope are ineffable. For totally different reasons from those given for the pain sensation, of course. The fundamental point is that we don’t simply learn to express our grief or our hope. We also learn to feel them. Here language doesn’t only function as a means of communication. It is one of the ingredients of that which we communicate. It is true that, as Hacker puts it:

    “The possession of a language extends the will and affections no less than the intellect. (...) A dog can want to go for a walk now, but it cannot now want to go for a walk tomorrow or next Sunday; it can want a bone now, but not a bone for Christmas.” (Hacker 2004, p.62-3)

    But what happens in the case of the feelings of hope and grief goes a lot deeper. They are not just extensions of feelings which the baby brings within itself from birth. They are not built up on other more basic feelings or sensations through some sort of psychic composition. They are added to our inner world. Just as there may be a concept that “refers to a character of human handwriting” that cannot have any “application to beings that do not write”, so too there may be concepts that refer to a very “complicated form of life” and that therefore can’t be applied to beings that don’t engage in it (PI II i). This is the case of the concept grief. This is the case of the concept of hope. They are internally related to our form of life.

    We can therefore say that contrary to what we are inclined to think, we are not born equipped with all the feelings and states of mind we experience in our adult life. The idea that there are innate feelings is so awry as the idea that there are innate truths. We are not born with an already made inner world, and what happens when we learn to talk and to act in accord with the form of life of our environment is not just to name what before remained locked up in us and to associate our innate feelings (like hope) to their appropriate contexts (like being in deep trouble). What we learn is not just to call grief ‘grief’ (instead of ‘happiness’) and to feel grief precisely when we are deceived, forgotten or betrayed by others (and not when someone gives us a present, threatens us or gives us a rather difficult task). What we learn when we learn our language is to name what didn’t exist before. What we learn when we engage in our complicated forms of life is to have their likewise complicated feelings.


    1. Hacker, Peter 2004 Wittgenstein: connections and controversies, Oxford, Clarendon Press
    2. Wittgenstein, Ludwig 2001 Philosophical Investigations, Oxford, Blackwell Publishers.
    Murilo Rocha Seabra and Marcos Paiva Pinheiro. Date: XML TEI markup by WAB (Rune J. Falch, Heinz W. Krüger, Alois Pichler, Deirdre C.P. Smith) 2011-13. Last change 18.12.2013.
    This page is made available under the Creative Commons General Public License "Attribution, Non-Commercial, Share-Alike", version 3.0 (CCPL BY-NC-SA)


    • There are currently no refbacks.