In Philosophical Investigations, 37 (2), pp.152-166 (2014)


Wittgenstein and the expression of feelings in psychotherapy


Campbell Purton    University of East Anglia


Abstract.  Effective psychotherapy is often held to involve the expression of feelings. Within the person-centred approach this view has been especially emphasised by Carl Rogers and Eugene Gendlin.  I am concerned with the question of why the expression of feelings can be therapeutically effective.  Many psychotherapists picture feelings as ‘inner experiences’ for which the client tries to find appropriate words, but the difficulties with this picture, which have been highlighted by Wittgenstein, seem to call for a very different approach.  Here I develop a Wittgensteinian account of the expression of feelings that avoids the difficulties to which Wittgenstein draws attention.


There are forms of psychotherapy that involve procedures that work directly on a person’s behaviour, for example gradual exposure to frightening objects or situations, so that the person’s anxious responses gradually subside.  However much of what is called psychotherapy does not work directly with behaviour but with speech.  It was Freud’s patient ‘Anna O’ who first called psychoanalysis ‘the talking cure’.  There is a puzzle about this, in that it is not obvious how merely talking about one’s difficulties can resolve them, and indeed one hears reports of therapy clients who say that while they have benefitted from their experience of therapy in the sense that they now understand their difficulties much better, nevertheless they have not changed in themselves; they still feel much the same.  That is a real possibility, but on the other hand many clients do find that as a result of the ‘talking therapy’ they have changed, and do ‘feel different in themselves’. There is a question about how that is possible.


One thing that is often said is that therapeutic change depends in some way on the client’s expression of their feelings, and in the first part of this paper I will discuss what I think is involved in the expression of feelings.  Then  I will turn to the question of how it is that the expression of feelings can involve personal change.


The expression of feelings

It is a widely-held view that the expression of feelings involves giving attention to our subjective experiences, that it is a matter of introspection, of turning our attention inwards.  Wittgenstein, however, has argued that to think in this way involves a misleading picture of how our talk about feelings works.  In this paper I want to take Wittgenstein’s critique seriously, and then consider how we might reformulate our ways of talking about what is involved when people express their feelings in therapy.  Examples of giving attention to feelings could be drawn from psychoanalysis, client-centred therapy and probably many other forms of ‘talking therapy’.  I will start from an example taken from Eugene Gendlin’s focusing-oriented therapy[1], which is a development within Carl Rogers’ ‘client-centred’, or ‘person-centred’ approach[2].  Gendlin’s approach is especially convenient for my purposes since he regards the facilitation of client expression as the central feature of effective therapy.


At the University of Chicago in the 1950s Gendlin, together with other colleagues of Rogers, were studying in some detail the different kinds of things that clients do in therapy sessions. (At the time Gendlin was also completing his PhD on the philosophy of experiencing,  and throughout his career he has written extensively on philosophical as well as psychological themes).  Gendlin’s studies of therapy sessions indicated that some clients talk a lot about events in their lives; they may ‘tell the story of their week’, or talk about events in their childhood.  Other clients try to analyse what is wrong with themselves, possibly in terms of psychological theories with which they are familiar.  Then there are clients who repeatedly ‘act out’ their difficulties in the session, or allow themselves to become overwhelmed by their emotions. Clients who only do these things, Gendlin noted, tend not to make very good progress in therapy.  The clients who make better progress often do something different in their sessions. Instead of simply reporting on events or feelings, or analyzing themselves, or becoming submerged in their emotions, they talk a bit about their difficulty and then pause.  After the pause they say something new, that is in some way connected with what they said before, but which does not obviously follow from their earlier utterance. This pattern is best appreciated by examples, such as the following[3]:


C[lient]:  I’m mad at Tom [her new husband] for doing that.  I want to push him away.

T[herapist]: What does that pushing away feel like?

C: Hmmm...Like I want to shove him into that big black case my cello is in.

I think I must want to shape and cut him to some mold of mine.  I do that to people.

T: Well, sense it a little longer.

C: Hmm...Oh! (surprise) It’s like putting him away just now, but knowing that I will want him back later -  keeping him in a safe place like a precious object.


Let me first describe what happens here in terms of a change in the client’s feelings, without examining for the moment what is involved in such talk of feelings. The client begins by noticing something that she feels, something that she expresses by saying that she wants to push her husband away.  The therapist responds in a way that encourages her to give more attention to what she is feeling. Then comes a pause as she attends to the feeling  -  ‘Hmm …’  Following the pause, her feeling becomes more articulated, more specific.   She uses the metaphor of wanting to shove him into her cello case – it is that sort of pushing away.  Then she does something different  - she begins to analyse or speculate about what is going on  -  ‘I think I must want to shape him and cut him to some sort of mold. I do that to people’.  The therapist realizes that this is unlikely to be helpful, so rather than engaging with her intellectual analysis he again encourages her to sense what she actually feels. Then comes a second pause, followed by a further change in her feelings, a change that surprises her.  She isn’t  trying to cut him to some sort of mould; she is putting him safely away for a while.


What the client is doing here is not very different from what someone might do on their own, when trying, as we say, to ‘sort out their feelings’. In fact Gendlin has developed a self-help system, known as ‘Focusing’, in which one is instructed to attend to what he calls the ‘felt sense’ of one’s situation; to let any familiar or intellectual formulations go by, and to await the coming of something fresh.   I imagine that some people have always done this sort of thing; consider for example I. A. Richards’[4] technique for determining whether one’s response to a poem is sincere,  or the spiritual exercises of Saint Ignatius, or some Buddhist mindfulness practices.  Iris Murdoch[5]  draws attention to something similar in the context of querying whether Wittgenstein’s approach does justice to what we think of as ‘inner’ or ‘personal’ experiences, which seem beyond the reach of what she calls ‘the impersonal world of language’. She writes:


Consider for instance the case of a man trying privately to determine whether something which he ‘feels’ is repentance or not.  Of course this investigation is subject to some public rules…[b]ut these apart, the activity in question must remain a highly personal one upon which the prise  of ‘the impersonal world of language’ is to say the least problematic….Here an individual is making a specialised personal use of a concept… Repentance may mean something different to an individual at different times in his life and cannot be understood except in context….There are two senses of knowing what a word means, one is connected with ordinary language and the other very much less so…


Murdoch is concerned with the ethical dimension of  such self-examination, but also with the theme of inner mental activity, and with the doubts that those sympathetic to Wittgenstein  may feel about the notion of ‘the inner’. She is concerned to show that what can be publically said has only a limited relation to what one is doing when one reflects on what one feels. I think there is something right about that, but I would like to explore the theme in a bit more detail.  I will begin with a simple example, where it is clear how the so-called ‘public rules’ work, and then look at a case that is more like that of the client who expresses herself by saying that she feels that she’d like to put her husband in her cello case.


A counselling client – call him Charlie - says something, but then realises that what he said does not articulate his response adequately.  For example, he says in connection with having knocked over his sister’s cup of coffee at a recent social occasion, ‘I felt really bad about it’. When the therapist encourages him to say more he says ‘I felt guilty [pause]  ….no, it’s more like I was embarrassed.’  How does he find these words?  A familiar answer is that he directs his attention ‘inwards’ towards his ‘subjective experiencing’ and notices whether this experiencing has the qualities characteristic of guilt or of embarrassment.  It is this picture of the logically private inner experience that Wittgenstein questions.  I will not review Wittgenstein’s arguments here, but simply note the point that a quite different account is available.  According to that account what is involved is something like this:  Charlie, having learned English, has learned the use of the word ‘guilty’ in contexts of someone responding with discomfort  to having done something wrong  -  that is, they normally use the word as English speakers use it.  But then, in the present circumstances,  he realises that this is not a context of having done something wrong, exactly, but of having done something clumsy and socially unfortunate. He might notice this explicitly, but more often people’s linguistic training enables them to say, in a situation like this, that ‘embarrassed’ is a better word than ‘guilty’, even though they may not be able to say why.  Someone who wasn’t at all inclined to reject the word ‘guilty’ after giving attention to the details of the embarrassing situation would thereby show an inadequate grasp of the language involved.  It is part of knowing English that one knows that ‘embarrassed’ and not ‘guilty’ is the word to use in this situation.  (I will leave aside until later the kind of case where Charlie says ‘I can see that it would be more appropriate to feel embarrassed, but the fact is that I still feel guilty).


Here we are clearly in the realm of ‘public language’.  But consider now the fact that therapy clients often use words in new, idiosyncratic and metaphorical ways, ways which Murdoch would no doubt have delighted in.  Indeed, as therapy progresses we often see an increasing use of metaphor. For example, a client may say ‘That’s it, my life is just an empty box’.  Without knowing more about her situation, and the tone in which she says this, we may have only a vague sense of what she means, and we might be completely wrong in our opinion. The therapist, however, knows that she had a family but her husband left her, and her children have now grown up and moved away.  She provided a protective container for them which was her main role in life.  She did it well, but now she has that role no longer.  She is a discarded, empty box.  There is nothing puzzling about how she comes to use the image of the empty box.  Empty boxes were once full, but no longer are, they are things which are put aside, thrown away; they are far less important than what they contained, they have no value in themselves.  She feels she has no value in herself, like that.  But of course she can use this image only because of her familiarity with boxes and packaging; the words she uses are all bound up with the culture in which she lives.  Someone from a hunter-gatherer community wouldn’t say what she says, or understand what she means.  She can say what she says not simply through knowing English but through the living context of her knowledge of English.


The case of the client who wants to put her husband in her cello case takes the idiosyncratic use of language a little further; here we need to have some background appreciation of the role that music plays in her life, the vulnerability of cellos, her valuing of her cello, and so on.  Someone who did not have this background understanding would not understand what she meant, but that has nothing to do with what she says being ‘inner’ or ‘hidden’ in a philosophical sense.  As a client reveals more about themself, and the therapist becomes more familiar with the idiosyncrasies of their ways of seeing things, the language they use may become more and more opaque to an outside observer who is not familiar with the context. But again, opacity in this context has nothing to do with what is essentially or philosophically hidden.  Nor does the opacity involve any ineffability of what is felt: the feeling is not ineffable; the client is sharing it, in words, with the therapist.


However, there may still seem to be a puzzle here, which might be formulated as ‘What, exactly, is the client sharing with the therapist?  The obvious answer is that the client is sharing ‘their feelings’, but the puzzling question then is how we are to understand the notion of  ‘a feeling’ before it has been put into words.  Isn’t that something intrinsically private? Certain of Wittgenstein’s remarks may seem to suggest that he would deny that there is any such thing as having a sense, or feeling, of what one is going to say before one says it.  For example[6]:


“The word is on the tip of my tongue”.  What is going on in my mind at this moment?  That is not the point at all.  Whatever went on was not what was meant by that expression.  What is of more interest is what went on in my behaviour.  -  “The word is on the tip of my tongue” tells you: the word which belongs here has escaped me, but I hope to find it soon.”  -  for the rest, the verbal expression does no more than some kind of wordless behaviour.

            On this, [William] James is really trying to say: “What a remarkable experience!  The word is not there yet, and yet, in a certain sense it is  - or something is there, which cannot grow into anything but this word. – But this is not an experience at all.  Interpreted as an experience, it does indeed look odd.  As does an intention, interpreted as an accompaniment of action; or again, like -1, interpreted as a cardinal number.


Wittgenstein is not denying here that one may have the experience of ‘having in one’s mind a sense of what one wants to say’, while being unable to find the words.  That would clearly be absurd, since the phenomenon is a very common one.  What he is denying, I think, is that the having of the sense in one’s mind is itself an experience, to which one might attend.  In Culture and Value[7] he writes:


[t]here really are cases where someone has the sense of what he wants to say much more clearly in his mind than he can express in words.  (This happens to me very often)…A mediocre writer must be aware of too quickly replacing a crude, incorrect expression with a correct one.  By doing so he kills his original idea, which was at least still a living seedling. 


Let us pause for a minute with this.  Wittgenstein is surely right to say that in seeking the right words one should be careful not to move too quickly. For if one tries to do so, words may come, but they will come because they are familiar (they are what one habitually says here), or they follow logically in an obvious way.  This connects with the fact I mentioned earlier: that in effective therapy clients typically pause in their speech. However it is hardly the speed with which the words come that is crucial; it is rather how they come.  In seeking for the words one is looking not for the standard everyday response in one’s situation, nor for the logical thing to say, but for something that, as we might say, ‘comes from the feeling’.  What comes must, to use Wittgenstein’s metaphor, be the growth of the seedling.  The words must be the natural development of that which was previously not articulated in words. But what account are we to give of that?  If we are sympathetic to Wittgenstein’s general approach we will not want to say that that is a private inner feeling which only the individual in question can access.  But what else could it be?


One possibility is this: in giving attention to what one wants to say one is attending to one’s ‘inner feelings’; not in the way philosophers understand that phrase, but in the sense of attending to one’s bodily sensations and reactions.  It might be said that one can feel one’s responding in one’s body, perhaps as a muscular tension, or a gnawing in the stomach.  Should we then say that it is bodily sensations such as these which constitute ‘that which has not yet been put into words’.  This seems to be Gendlin’s view, at least as expressed in his more psychological writings.  Like Wittgenstein, he uses the ‘tip of the tongue’ example[8]:


Suppose you have been listening to a discussion and are about to say something relevant and important… Now suppose that, as you wait for your chance to speak, your attention is distracted for a moment and you lose hold of what you were about to say.  The others are now giving you your turn, waiting for you to say what you wanted to say, but you cannot….What do you do to regain the sense of what you were going to say?  Where do you look for it?   You look within the body… you grope within yourself.  There you do have a felt sense … you have that feeling of what you forgot.


‘Felt sense’ is Gendlin’s term for that which Wittgenstein calls the ‘seedling’ – that which is felt, and which given favourable conditions, will develop into a particular form of words.  Gendlin says[9] that a felt sense is ‘a bodily awareness of a situation or person or event.  An internal aura that encompasses everything you feel and know about the given subject at a given time…’  Or, most succinctly,[10]‘It is a special kind of bodily sensation.’


Wittgenstein himself considers this possibility that the ‘feeling’ to be articulated, the sensing of what is to be said, is a matter of  our feeling our bodily reactions in the situation.  He discusses this in connection with sensing the plaintiveness of a melody, or the timidity in a person’s facial expression.  We say we sense these things, as we say we sense what we want to say, but is this sort of sensing to be understood in terms of what one might call ‘physical reactions’?  Wittgenstein thinks not[11].  He writes:


Just think of the expression “I heard a plaintive melody”!  And now the question is: “Does he hear the plaint?”

And if I reply: “No, he doesn’t hear it, he merely senses it – where does that get us? One cannot even specify a sense-organ for this ‘sensing’.

We react to a facial expression differently from someone who does not recognise it as timid (in the full sense of the word). – But I do not want to say here that we feel this reaction in our muscles and joints, and that this is the ‘sensing’.  – No, what we have here is a modified concept of sensing.


I think Wittgenstein is right about this.  The sensing that is involved in sensing the plaint or the timidity, and the sensing that is involved in sensing what one wants to say, are not to be identified with bodily reactions, even though such bodily reactions may be there. (Some difficulties with identifying the sensing with a certain kind of bodily reaction are that in many cases such a bodily reaction simply may not be there, that people may have different bodily reactions while sensing the same thing, and that we don’t need to know the nature of a person’s bodily reaction in order to know that they have sensed something.  Stanley Coval once wrote of Gilbert Ryle ‘Ryle sensed trouble where trouble was’, but Coval couldn’t have known anything about  Ryle’s bodily reactions).


We are returned to the question of what account we are to give of that which Wittgenstein calls ‘the seedling’  -- or that which, as William James puts it, ‘cannot grow into anything else but this word.’  It is not helpful to picture it as a philosophically private inner feeling, nor to think of it as a bodily reaction, but we may perhaps get somewhere if we think of it as a person’s non-linguistic response to their situation, without turning that notion into the notion of a bodily reaction.


In speaking of a non-linguistic response I mean to include such things as cringing or hesitating, welcoming or rejecting, and responses such as a half-smile or a shrug of the shoulders.  (I mean ‘non-linguistic response’ in the sense that the response itself is not articulated linguistically; one hesitates but does not say ‘I hesitate here’.  However the hesitation might be a hesitating in what to say).  In the course of human development non-linguistic responses may develop into, or be replaced by, linguistic responses.  John Canfield[12] (2007) has given a plausible Wittgensteinian account of the sort of way in which linguistic responses may have developed out of the non-linguistic, in the course of human evolution, and also in the course of child development.  What begins as a cry of pain is replaced by an utterance such as ‘It hurts!’  Or a linguistic response may develop alongside, or interwoven with, the non-linguistic, as when someone says in a shivery way ‘It makes me shiver.’  (cf Wittgenstein[13] on this).  These are cases where someone might be said to be expressing their feelings, but they are also cases where we might say that they are articulating their response.  To articulate one’s response is to put it into words, but ‘articulation’ also suggests, as I will discuss later,  an increased complexity or ‘jointedness’ of response.  For early human beings, as Canfield says, a simple gesture or sound in the appropriate context (such as the context of pointing to a spear) might indicate ‘That is mine!’, but with the development of language one becomes less reliant on the context.  Language comes to provide sounds for different objects, so that the early human being may now be able to say, as a young child might, ‘Spear mine!’  This matter of the development of language out of non-linguistic responses is no doubt complex and controversial, especially in connection with how the syntax of language develops, but it seems to me clear enough that our linguistic utterances can often be said to be articulations or elaborations of our non-linguistic responses.


My suggestion, then, is that to speak of someone expressing their feelings is to speak of them articulating their response to their situation.  However, there is a difficulty here in that the word ‘situation’ often includes something of the person’s response.  For example someone might say ‘Charlie’s situation is this: His wife has left him and his children, just as he has begun this new job.’  That would usually be taken to imply that Charlie’s situation is a bit of a mess, that he is in a predicament, in a pickle. But we can also imagine a case in which, in spite of the circumstances being as described, Charlie is not at all in a predicament; rather, he is feeling that at last he is free, and is enthusiastically looking forward to meeting the new challenges that he faces.  One might, in a familiar sense of the term ‘situation’, say that Charlie’s situation is very different from that of a man who responded to the same circumstances with fear and bitterness.  (‘Situation’ here is rather like ‘world’ – ‘the world of the happy man is a different one from that of the unhappy man’[14]).  In what follows I will try to avoid this difficulty by using ‘situation’ to refer to the more impersonal description, to what ‘everyone’ within a particular community would agree to be the situation, and use ‘felt situation’ so as to include the person’s individual response to their situation.  There is of course no very sharp dividing line here.


I think that what one does when one tries to express one’s feelings is to give attention to both one’s situation and to one’s response to the situation.  One tries certain articulations (such as ‘I feel guilty’, or ‘I want to push him away’), and then notices where one feels uneasy, or senses that the response is not quite right.  Such sensing is grounded in our early training in the use of words, in the finer discriminations that come with increased linguistic competence, but also in the creative ability to find new ways of using the language we have.                                


The question may be raised, though, of how exactly one is to attend to one’s situation.  Gendlin says that one needs to attend to one’s felt sense of the situation, and for him this is a matter of attending to a vague bodily feeling.  But he also emphasises that one must get a felt sense of the whole situation.  As I have indicated, I think it is misleading to picture a felt sense as a bodily feeling, but I think Gendlin is right in emphasising that it is one’s whole situation to which one needs to give one’s attention.  If something new is to emerge there is no point in attending to what one has already formulated, such as ‘My sister is likely to be angry with me for knocking over her drink.  I feel really guilty.’ What is needed is to ‘get beyond’ language for a short while at least, so that new articulations can come.  One needs to attend to the situation as a whole, beyond its current differentiation by language, but how is one to do that, if not through attending to a bodily feeling?


To answer that question we need to consider what is involved, in general, in giving attention to something.  I think that Alan White[15] is right when he suggests that to attend to something is largely a matter of not letting oneself be distracted by anything else.  ‘One minds one’s own business by not minding that of other people.’  To attend to what one person is saying in a noisy, crowded room is to be occupied with that conversation, while trying not to respond to what everyone else is saying. So to attend to the situation as a whole is a matter of being occupied with the situation, while not allowing oneself to be distracted by specific responses to it, such as ‘I’m angry with him’, or ‘I want to push him away’.  One lets these specific responses go by while still being concerned with the situation as a whole.  If we like, we can still say, with Gendlin, that we need to attend to our ‘felt sense’ of the situation, but having a felt sense is not a matter of having a special kind of bodily feeling.   To have a felt sense of the situation is, rather, to attend to the situation as a whole, and to one’s response to it. One can – and one does -- respond to the situation as a whole, but one can only partially articulate that response.  In that sense the situation as a whole is beyond the reach of language, but through giving attention to the situation as a whole, we may find that we can articulate our response in ways that that we had not articulated it earlier.


The transformation of feelings


In what I have said so far I have been mainly concerned with the question of how we are to think about the expression of feelings in circumstances where we initially cannot find adequate words to do this. 


I turn now to the question of what the point is of expressing our feelings in a therapeutic situation. I want to consider how the expression of feelings in psychotherapy can lead to  personal change.  It is clear that finding the words to articulate a response to a situation can, in a straightforward way, make a difference to how one proceeds in the situation.  Suppose Bruno sees an animal in a field that looks like a big rabbit.  He says ‘Ah! That’s a …’  but he can’t initially find the word for it.  It is on the tip of his tongue …. Then it comes. ‘Oh yes, it’s a hare!’   Realising that it is a hare might lead him to consider coming back to this field in March in order to see the male hares boxing with each other, which he has never seen, and it would also lead to his not searching for its burrow, for he knows that hares don’t make burrows. In such ways the finding of the word is not just a linguistic matter, but in a small way makes a difference to his life.  On the other hand I don’t think we would say that it makes a difference to how he is, to what we might call his character or personality.  However,  the finding of the right words in therapy does sometimes make a difference to how the client is, to their character or personality.  We might say that in the case of the hare, what has changed is Bruno’s view of things, but in the therapy case it is not just that the client’s view of themself has changed, but that they have changed. 


There are several further kinds of case that may help us in understanding what is involved in the therapy case.  First there is a kind of case where we don’t quite know what to say about another person.   For example, Anneka is applying to join the Territorial Army, and Bruno has been asked for his opinion of her. He is not sure whether to say that Anneka is intuitive, or is just quick to react.  Or whether she is introverted by nature or just shy in large groups of people.  Bruno gives his attention to Anneka, or to what he knows about her, and tries to find the best words to express his opinion of her.  This, I think, is not in principle different from the hare case.  When Bruno has the right words he may not only be able to help the TA selection board with their decision, but may himself be able to deal with Anneka more intelligently than before.  As with the hare, he now knows better ‘how she will jump’ in certain situations.


We may also, in a similar way, try to find the right words to describe ourselves.  Having pondered over whether Anneka really counts as ‘introverted’, Bruno may consider to what extent this term applies in his own case.  He may reflect on his behaviour in various situations, study Jung’s work on psychological types, and eventually arrive at a firm sense of whether he is introverted or not.   Sometimes therapy clients do this sort of thing, and the knowledge they acquire of themselves may help them in managing themselves, as it were.  In my example earlier, of Charlie coming to feel embarrassed rather than guilty, I was concerned with the case where there is a change in Charlie, a change in his response to his situation. But we can imagine a different kind of case where Charlie comes to think, or believe, that embarrassment rather than guilt is appropriate in this situation, yet still feels guilty.  Then he will have to consider whether anything can be done about this. It is clearly a situation in which something is awry – normally people don’t feel guilty in embarrassing situations; that logically couldn’t  be the case. In this revised example Charlie has come to know more about himself, perhaps realising now that he has a strong tendency to expect punishment, and that this can distort his view of situations, but such increased knowledge of oneself is not the same as personal change. 


Another very relevant kind of case is that where we are in some way personally involved with the person whom we are attempting to characterise in words, so that finding the right words is also a matter of morally assessing our responses to the person.  Iris Murdoch[16] considers the example of a mother, M,  whose view of her dead or otherwise absent daughter-in-law, D, changes as she gives attention to her memories of D. M initially feels that D is vulgar, undignified, tiresomely juvenile, but then reflects that her view of D may well be influenced by her own conventionality, by her admitted snobbishness and jealousy.  At this stage, we may imagine, M no longer quite knows what to say about D.  What happens then, as M continues to reflect on D, is that M comes to see D as not vulgar, but refreshingly simple; not undignified, but spontaneous; not tiresomely juvenile, but delightfully youthful.  As Murdoch says of the mother, “Gradually her vision of D alters”.


That is an example of ‘finding the right words’, but it is clearly different from the Anneka example.  In Murdoch’s example the mother’s finding the right words involves a change in her.  Murdoch says that M’s vision of D has changed, but the change of vision here is not like Bruno’s coming to see that Anneka really is intuitive, or that he himself is introverted; what has changed is the whole quality of the mother’s engagement with her daughter-in-law.  The way she responds to her memories of D is now very different; she is different, at least in relation to D.


Finding the right words in therapy is more like the mother’s finding of them in connection with her daughter-in-law, than Bruno’s finding them in connection with Anneka or himself.  Indeed, in therapy there are sometimes transformations of  just the sort that Murdoch describes, where a person’s sense of another radically changes.  At other times it is rather that one’s sense of one’s own situation changes, although, as in the Murdoch case, this ‘sense’, or what Murdoch calls one’s ‘vision’,  needs to be understood not as a changed view, but as a changed way of responding to one’s situation.  The relevant difference between Murdoch’s case and the Anneka case is that in the latter case Bruno tries out different reports that he might make in connection with Anneka  -  he is trying to find an accurate description of her.  He tries out different descriptions while keeping in mind all that he knows of her.   In this process his initial description ‘Anneka is introverted’ is replaced by ‘She is shy in large groups’.  In the Murdoch case M, the mother, is not seeking an accurate or apt description of D, the daughter-in-law.   She is concerned with the issue of what her response to D should be.  Initially she finds herself feeling that D is undignified, but as she gives her attention to the whole situation, and takes into consideration her own snobbishness and jealousy, she finds that her response has changed: now she feels that D is spontaneous rather than undignified.  Of course someone might say that ‘she now describes D differently’, but what the new description amounts to here is the expression of a new response or attitude, not of a new opinion.


I think the therapy case is very similar to Murdoch’s  example; however in therapy the moral dimension that is Murdoch’s central concern is not so prominent.  Charlie has talked in counselling about knocking over his sister’s drink, and moves from saying he feels guilty to saying that he feels embarrassed, but this is hardly a moral change. On the other hand it is a case of finding a better response, and trying to find better responses seems to be an important part of morality.  Whatever we should say about that point, it does seem clear that the change in Charlie’s case is not that he has found a better description of his feelings, but that his feelings have changed. The change does not involve a more accurate opinion as to the nature of what he has done, but a more appropriate response to it.  There are cases where one reflects on what one has done – perhaps a long time ago, ‘in another country, and besides the wench is dead’  – and simply comes to a new view or opinion of what one then did or felt (that is like the case of Bruno concluding that he is himself an introverted person), but it is not how it is for Charlie.  Similarly, in the case of my original example, it is not that the young woman’s opinion has changed from ‘I think I want to shape him to some mould of mine’ to ‘My opinion now is that I want to put him away, but keep him safe’.  She is not working on her opinion; she is working on her attitude, on how she is inclined to respond, although she may need the therapist’s gentle guidance in helping her to focus on attitude rather than on opinion.


One complication is that the therapy client may also come to have a different opinion about their situation.  For example, Charlie, as he reflects on the coffee-cup incident, might come to realise from something his sister said afterwards, that actually she had been relieved at the diversion the incident had caused .  This is a realisation that the situation as a whole was not quite what he had thought.  Here he might say ‘Oh… that’s a relief! I don’t need to feel guilty after all’.   Both kinds of change may occur as the client gives their attention to their situation as a whole; but it is important to see that not all, and probably not most, of therapeutic change is a matter of  a change in the client’s beliefs.  It is not new insight into their situation, or even into their own personality, that is crucial, but a transformation of  response and attitude. This fits with something that Carl Rogers[17] once said concerning what he thought is the most crucial element in psychotherapy:  ‘In years gone by I believed that insight, properly defined, was such a crucial element.  I have long since given up this view’. 


In summary, I have suggested that the ‘talking therapies’ encourage clients to express their feelings, but that the expression of feeling should not be pictured as involving attention to a ‘felt sense’, if by that is meant some kind of vague bodily sensation.  Rather it is a matter of giving attention to one’s whole situation and to one’s response to it.  One then tries to articulate one’s response, and such articulation will usually take a linguistic form, although gestures or other non-linguistic responses may also be involved.  Then there is the question of what is involved in the transformation of feelings in therapy. My account of this has been that we need to bring our attention back from our current responses, and attend to our situation as a whole, opening ourselves to the changes in response that may then come. We attend to our situation precisely by not allowing ourselves to be distracted by our previous articulations of our situation.  We attend to ‘all that’ and try to be receptive to the coming of new articulations, new ways of responding to the situation.  Thus we find new ways of expressing ourselves, and in finding such new ways we find that our felt situation is now different, and that we ourselves have changed.  The articulation of a response is a change in the response, but as Aristotle held, things don’t always change into something different; sometimes they change into being more fully themselves.  In Wittgenstein’s image above, our initial inarticulate response to our situation is the seedling which, when watered by care and attention, can grow into a developed and more articulate response. The change in what we say, when that change involves the articulation of our response to our situation, is also a change in what we feel and are inclined to do; but it is a change that involves becoming more ourselves. Hence Rogers’[18] favourite quotation from Kierkegaard[19], that the aim of the endeavour is ‘to be that self which one truly is’. [20]


10 Kilbrack



NR34 9SH


[1] E. T. Gendlin, Focusing-Oriented Psychotherapy (New York: Guilford Press, 1996)


[2] C. Purton, Person-Centred Therapy: The Focusing-Oriented Approach (Basingstoke: Palgrave, 2004).


[3] Op. cit. note 1, p. 40.


[4] I. A. Richards, Practical Criticism: A Study of Literary Judgement (London: Routledge, 1929), p. 290


[5] I. Murdoch, The Sovereignty of Good (London: Routledge & Kegan Paul, 1970), pp. 25, 27-9


[6] L. Wittgenstein, 'Philosophy of Psychology - A Fragment', Philosophical Investigations,  (Chichester: Wiley-Blackwell, 2009), §§ 298-9.


[7] L. Wittgenstein, Culture and Value (Oxford: Blackwell, 1980), p. 79.


[8] E. T. Gendlin, Focusing (London: Rider, 1978/2003), p. 85.


[9] Op. cit. note 8, p. 32.


[10] Op. cit. note 1, p. 18.


[11] Op. cit. note 6, §§ 229-231


[12] J. Canfield, Becoming Human: The Development of Language, Self, and Self-Consciousness (Basingstoke: Palgrave, 2007).


[13] Op. cit. note 6, § 6.


[14] L. Wittgenstein, Tractatus Logico-Philosophicus, (London: Routledge, 1961), § 6.43.


[15] A. White, Attention (Oxford: Blackwell, 1964), pp. 41-60.


[16] Op. cit. note 5, pp. 17-23


[17] C. R. Rogers, 'The essence of psychotherapy: moments of movement', first meeting of the American Academy of Psychotherapists,  (New York: 1956).


[18]  C. R. Rogers, On Becoming a Person (Boston: Houghton Mifflin, 1961/1967), pp. 166.


[19] S. Kierkegaard, The Sickness Unto Death (Princeton University Press, 1941), pp. 29.

[20] An earlier version of this paper was presented as the annual Wittgenstein lecture at the Welsh Philosophical Society conference at Gregynog Hall, Newtown, in May 2012.  I am grateful to David Cockburn for very helpful comments.