Constructivist epistemologies and therapies
Erwin E. (1999) Constructivist epistemologies and therapies. British Journal of Guidance & Counselling 27(3): 353–365. Available at http://cepa.info/3836
Table of Contents
Language and the world
The construction of meanings
Constructivist epistemology and therapy: is there a connection?
Constructivists link their therapeutic approach to arguments against an objectivist epistemology. These anti-objectivist arguments are examined: it is concluded that none of them is cogent. The question of whether acceptance of a constructivist, non-objectivist epistemology is likely to affect constructivists’ therapeutic practice is considered.
Many psychotherapists and counsellors are sceptical about the idea that effective therapy always requires the discovery of ‘what really caused’ the client’s problems. Some of the grounds for this scepticism are empirical; others are philosophical. For example, many constructivists argue on philosophical grounds that the quest for finding out whether the client’s ‘narrative’ corresponds to a mind- and language- independent reality presupposes a discredited objectivist epistemology. Some see the rejection of that sort of epistemology as the main philosophical basis for constructivist approaches to therapy: ‘Like the broader post-modern zeitgeist from which it derives, constructivist psychotherapy is founded on a conceptual critique of objectivist epistemology’ (Neimeyer, 1993, p. 230; for a discussion of this quotation and similar remarks by other post-modern theorists, see Held, 1998).
An ‘objectivist epistemology’ is one which holds that propositions are generally true or false independently of any particular paradigm, or school of thought, or language, or indeed of what any human believes; and, furthermore, that they can often be warranted independently of what anyone believes. Those who reject such an epistemology often accept other so-called ‘post-modern’ views such as: ‘Reality is constructed rather than discovered’, or ‘Words do not mirror reality’, or even ‘Apart from language, there is no objective reality’. In this paper, I intend to challenge some of these claims (although I do not discuss here any positive proposals for an objectivist epistemology, there are some modest proposals made in Erwin, 1996, chapter 2; and Erwin, 1997, pp. 74-80). In the closing section of the paper, I consider the question of what connections there might be between accepting a constructivist anti-objective epistemology and constructivist therapy.
To get a better fix on some of these ideas, I will begin with a quotation from one of the editors of the book Constructing Realities: Meaning Making Perspectives for Psychotherapists:
‘While there are a variety of constructivist models, they all hold in common the epistemological belief that a totally objective reality, one that stands apart from the knowing subject, can never be fully known. They reject the correspondence theory of truth, which postulates that our mental representations mirror an objective reality, ‘out there’, as it truly is. Rather, knowledge, and the meaning we imbue it with, is a construction of the human mind’ (Rosen, 1996, p. 5).
Rosen’s statement nicely captures the views of many constructivists, but one part of what he says would also be acceptable to non-constructivists. If ‘can’ is interpreted to mean what is practically possible, then who would disagree with the thesis that a totally objective reality can never be fully known? Even if there is a reality independent of mind and language – something that many constructivists do not concede – there are things about it that are practically impossible to know. Even if we were motivated to try, we cannot keep track of every twitch of every leg of every centipede. That reality cannot be fully known is a platitude accepted by all sides of recent epistemic controversies.
In contrast, the rejection of a correspondence theory of truth is contentious, and is of fundamental importance. The negation of the correspondence theory does not by itself entail anything about therapy, but when it is combined with other theses, it can make an important difference as to how one views both research and therapeutic practice. One constructivist, for example, links the rejection of truth as correspondence to doubts about cognitive therapies (Neimeyer, 1993). Some who are in the narrativist tradition conceptualise a client’s account of his or her problems as a ‘story’ that is true or false only in a poetic sense (Sass, 1992). Whether or not the client’s narrative reflects what ‘really happened’ is not, on this view, of major clinical importance. This conflicts with the traditional Freudian view that the client’s achieving genuine insight into the etiology of his or her problems is crucial, even if it is not sufficient, for resolving one’s psychoneurosis.
Constructivists are not alone in having doubts about the correspondence theory of truth. Some who describe their position as ‘social constructionism’ also reject an objectivist epistemology, including a correspondence theory of truth, even though in other respects their views are quite different from that of the constructivist (the latter position is called cconstructivism’; the social theory is called ‘social constructionism’). One leading social constructionist, Gergen (1994), writes: ‘Words are not mirrorlike reflections of reality, but expressions of group conventions’ (p. 119).
Some who reject the correspondence theory offer an alternative, such as a pragmatic theory, while others argue that we should not talk about the truth at all, but should use some substitute notion. The latter position is taken by Mahoney (1989) who says that modern constructivists are essentially asserting that viability should replace validity and ontological correspondence in epistemic inquiry. Polkinghorne (1992, p. 151) endorses a similar position when he claims that on a post-modern view one asks not whether a knowledge claim is true but whether acting on it produces successful results: ‘One does not ask if a knowledge claim is an accurate depiction of the real – is it true? One asks, rather, does acting on this knowledge claim produce successful results?’
Some parts of the above view seem reasonable enough, but other things are problematic. For example, there is nothing objectionable in inquiring in certain cases about a belief’s viability rather than its truth. What is not so clear, however, is that one can even apply a viability criterion without presupposing that certain things are true and others false.
Suppose I say: ‘I do not care whether the new narrative that the client and I jointly constructed is true, if this means that it describes what actually happened prior to the development of his obsessional behaviour. What I care about is whether the client’s adoption of the new story about his problems produces a diminishment of the problematic behaviour.’ This may be a defensible point-of-view, but in determining that the client’s new belief is ‘viable’, does not the therapist have to determine the truth or falsity of certain propositions (regardless of whether truth terminology is used explicitly)?
For example, suppose that the client smiles and pays his bill at the termination of therapy, but in fact his condition has deteriorated. He is more obsessive than when he entered therapy. In that case, the client’s new belief was not viable, if improvement with respect to his obsession is the criterion of viability. Even if there is a successful outcome, does not the therapist who favourably judges the client’s new narrative have to presuppose that it is false that something other than the adoption of the ‘new story’ brought about the improvement? It would be incoherent to say: ‘The client’s new belief was viable in that acting on it produced successful results, but it is not true that acting on it really did produce successful results. Something else deserves credit for the outcome.’ If it is not true that the adoption of the new narrative helped cause the outcome, it logically follows that the new belief was not viable.
In short, to apply a viability criterion, one has to establish certain truths: first, that there really is a successful outcome; and second, that there is a causal connection between adopting the new belief (or acting on it) and the outcome.
One could reply that there is simply no way to tell whether accepting a belief really did lead to successful results. This appears to be an implication of Polkinghorne’s view. He says (1992, p. 151) that on the post-modern view there is no way to tell whether a picture (he makes clear that by this he means a knowledge claim) corresponds to the real. How, on this view, does one ever know that a belief is viable? In Polkinghorne’s opinion, the answer must be: we cannot ever ascertain whether a belief has this property. Of what use, then, would Polkinghorne’s neopragmatic criterion be? As far as we could ever know, there would not be a single case in which it would be applicable.
A more modest view than that of either Polkinghorne or Mahoney, one that has gained adherents among narrative therapists, is not that we dispense with the concept of truth altogether, but just that in the one case of evaluating a client’s narrative we ask about its utility rather than its truth or falsity. If a client comes to believe that she was primarily responsible for the nasty family quarrels, it may not matter to the counsellor whether she is right or not. What matters, or at least what may matter, is whether the effects of believing the new view are beneficial. Even if the client was not solely responsible for the family fights, by taking responsibility she may lead the way to family peace and tranquillity.
In some cases, such as in the above example, the accuracy of the client’s beliefs may not matter to the outcome of the therapy, but in other cases, it may matter to something else of great importance. A woman client who seems to have remembered being raped by her father may now have a theory that appears to explain her failed marriages and troubled sexual relationships; accepting this theory could have therapeutic value even if it is false, but if it is false, the father may suffer grievous, unjust harm. ‘What the consequences are of a client’s constructing a false ‘narrative’ is obviously an empirical matter, and may vary from case to case. In any event, this more modest view, that at least sometimes it does not matter whether the client’s new narrative is wrong, is not an epistemological view, nor does it have any obvious, interesting epistemological consequences, even though it may have significant therapeutic implications.
Instead of doing without the concept of truth, some theorists adopt a pragmatic theory of truth. For example, a recent president of the Association for the Advancement of Behavior Therapy, Hayes (1993, p. 16), attempts to reduce truth to ‘successful working’. On this view, to say that a belief or verbal statement is true is to say that it works successfully. One problem with this idea is that it, too, requires in its very application covert appeal to truth in a correspondence sense. Suppose that my stated goal is to do something that will help the client avoid painful arguments with her spouse, and I believe that I have succeeded in achieving my goal. Am I right? I am right only if it is in fact true that the goal was reached and the therapy did, in fact, make a positive difference (for further details, see Erwin, 1997, pp. 66-67, 90-91).
Apart from the difficulty of applying a pragmatic theory, there is a more serious problem: false beliefs sometimes work in leading to successful outcomes. So, successful working is no guarantee of truth. It is somewhat ironic that the philosopher most often associated with a pragmatic view, Rorty, now admits defeat for all pragmatic theories of truth (Rorty, 1995) precisely because of this objection.
Even if, as Rorty apparently believes, a pragmatic theory of truth is demonstrably incorrect, many constructivists and social constructionists will still have doubts about a correspondence theory. First, as Gergen (1994, p. 23) puts it, our language about the world does not operate as a mirror of that world. Second, it is claimed that adoption of a correspondence theory leads to scepticism; for on such a theory, to know that one of my beliefs is true, I would have to do what is impossible – compare my representation of reality to reality itself. Why is that impossible? Because when I try to make the comparison, I can compare only one representation of the world to another representation. I cannot, it is said, get at the world itself.
Let us take each of these problems in turn. There are different versions of the correspondence theory of truth; most of the standard ones say nothing about truth being a mirror of reality. For example, if I say that what makes a proposition true is that it says what is the case, what I say does not entail, in any literal sense, that if P is true, it mirrors reality. Speaking of minoring reality may be wrong or obscure, but a reasonable correspondence theory does not commit one to such talk.
What about the charge that adoption of a correspondence view leads to scepticism? Here the dubious premise is that determining the truth of some proposition requires, on the correspondence theory, a comparison of a representation of reality with reality itself. A client says to me that she feels sad now that her terminally ill child has died. Is not the fact that she says this, or at least seems to have said this, in a particular set of circumstances, good evidence that she really does feel sad? The evidence may not rise to the level of certitude (it is possible that I did not hear what she said, or that she is lying), but as long as certitude is not required for knowledge, I may be in a position to say correctly: I know that she feels sad, whatever the true cause of her sadness. Even if I did not know this, she knows, does she not, that she feels sad; yet she has not managed the impossible, namely the comparison of a mental representation of reality with reality itself.
One could argue that there is no way to get evidence that a client feels sad, and even, more radically, that a client is never in a position to know whether she herself feels sad after her child has died – for all she knows, one might say, she feels elated. There are various ways that one could try to argue such things (none of them likely to be successful), but one has to go a great deal beyond saying that mental representations cannot be compared to reality. That proposition by itself does not entail that we can never find supporting evidence for anything.
A further complaint is that the key notions of a correspondence theory, that of ‘fact’ and ‘correspondence to reality’, have never been satisfactorily explained. This complaint can be handled by offering what is called a ‘deflationary’ account of the correspondence theory (Erwin, 1997, pp. 68-69), which is not metaphysical and does not purport to explain anything. It says merely that a proposition is true exactly if it says what is the case. No attempt is made to reduce the concept of what is the case to anything more basic. Another alternative is to switch to what is called a ‘disquotational’ view. On this view, neither the concept of ‘fact’ nor that of ‘correspondence’ is employed. So, the failure to explicate these concepts is not a mark against the theory. Truth conditions are formulated for each proposition by ‘dropping quotation marks’ as in the following example. The proposition expressed by The client’s problem was caused by a childhood trauma’ is true if and only if: the client’s problem was caused by a childhood trauma. More generally, ‘P’ is true if and only if: P. Such a formula does not, of course, give us any way of determining whether a proposition is true, but it is not intended to. The disquotational view tells us the conditions that are necessary and sufficient for a proposition being true. How we come to know whether it is true, or whether we can ever know it to be true, are different issues (for more details, see Horwich, 1990).
Language and the world
There is obviously a lot more that could be said about truth and the possibility of obtaining evidence that some proposition is true, but let us put these issues aside, and examine a different constructivist thesis: that there is no world independent of language. The philosopher most commonly associated with this view is Goodman (1978, 1984). On Goodman’s view, there is not one world but many. How many there are depends, in part, on how many world versions we construct.
As an illustration of his view, Goodman gives the example of the ‘Big Dipper’. It is we, he claims, who made the Big Dipper by constructing a version of how things are that contains the name ‘Big Dipper’. This may be true of the Big Dipper, someone might say, but what of the stars which constitute it? To this Goodman replies: there is no difference between the two cases that makes a relevant difference. If we brought the Big Dipper into existence by creating a world version in which there is a Big Dipper, then we did the same for stars. Because Goodman is committed to this view, a recent volume on his work is appropriately called Starrnaking (McCormick, 1996).
Is Goodman saying, then, that we make a world merely by constructing a version of it? He is not. We make a world, he claims, only by making a right version of it. A version in which there are stars, neutrinos and viruses is a right version; a version in which there are ghosts or unicorns or the ether is not a right version.
Other writers have expressed views similar to Goodman. In a section referring not to scientific discoveries about what is real, but to ‘the Scientific Creation of Reality’, Gergen (1994, p. 23) talks about concepts such as ‘schizophrenia’, ‘learned helplessness’ and ‘mid-life crisis’ as lenses supplied by the theorist to colleagues and society alike. He then concludes that the world is not so constituted – that is, does not contain schizophrenia, learned helplessness, or mid-life crises – until the lenses (i.e. the concepts) are employed.
Another writer, Maturana, who has had some influence among family therapists (Efran & Greene, 1996), holds that different people have different world views, but there is no single universe about which each person has a viewpoint. In fact, on his view, outside of language, nothing exists: ‘Outside of language nothing (no thing) exists because existence is bound to our distinctions in language’ (Maturana, 1988, p. 80). Maturana draws therapeutic lessons from his radical metaphysical views. Suppose that I am doing family counselling, and the family members I am working with have divergent views about the family dynamics. Should I try to figure out who is right? On Maturana’s view, each family member has an equally valid construal of the family system; there is no independent objective family system that exists in addition to the varied personal interpretations (Rosen, 1996, p. 7).
Working out an epistemology to fit the multiple worlds view has not yet been done in any detailed way (but see Neimeyer, 1993, and my reply, Erwin, 1997, pp. 62-74). If one believes, however, that there is no reality to be discovered but instead that we construct realities, it would not be surprising that one would see the purpose of empirical inquiry to be something other than the discovery of objective truth. One prominent social constructionist says that empirical research can continue, but only to provide rhetorical support for a theory (Gergen, 1994).
An obvious problem that confronts the view that we create such things as stars is that many things of this kind existed before human beings did. If they already existed, how can it be that we brought them into existence? In the same vein, how can Maturana be right in saying that existence is bound to our distinctions in language? Stars existed before humans used language to distinguish them from other things such as planets. How can Maturana possibly be right, in addition, in saying that outside of language nothing exists? If stars did exist before there was language of any kind, then their existence does not depend on language, and they do exist outside of language. Of course, we would not refer to them as ‘stars’ unless we used language, but that is a different point, and an entirely trivial one.
One might reply that stars are different in kind than the sorts of things that Gergen is talking about. Stars existed before humans did, but that is not true of such phenomena as schizophrenia, or learned helplessness, or mid-life crises. The point is correct but lends no support to the idea that by using language we construct reality.
Even though schizophrenia did not exist before humans came on the scene, the existence of the phenomenon predated the introduction of the relevant terminology. People were schizophrenic, and suffered mid-life crises, and experienced learned helplessness, before anyone described their problems in this way. How, then, can Gergen be right in saying that until the linguistic constructions were introduced, the world was ‘not so constituted’?
There may be other so-called ‘social realities’ that did not come into existence until after concepts that denote them were introduced. For all I know, there were no labour unions or parliaments until the respective concepts of ‘labour union’ or ‘parliament’ were first employed. It does not follow, however, that labour unions or parliaments are language-dependent except in a causal sense. Unlike stars, labour unions were created by humans, and it is unlikely that this would have occurred if no-one had possessed the concept of a labour union. Still, introducing the concept does not automatically bring into existence the thing denoted by the concept, and whether labour unions exist, and what their roles are in particular societies, are empirical questions. An earlier type of constructivist view also ran into problems in talking about things that existed before humans ‘constructed’ them. Phenomenalistic philosophers, such as John Stuart Mill and the early Bertrand Russell, said that physical objects were constructed out of sense data, which would be hard to do if the items in question were already in existence before humans came on the scene. Some phenomenalists at least attempted to deal with the problem by analysing statements about such things into hypothetical statements about what would have been observed had humans been there to make the relevant observations. To say, for example, that stars existed before humans did really means, on this view, that if someone had looked in their direction they would have had star-like sensations. These attempted phenomenalistic translations have long been known not to succeed, which is one reason that philosophers rejected phenomenalism a long time ago.
Neither Gergen nor Maturana, nor anyone else who has expressed similar views in the psychotherapy literature about our constructing reality, has offered a substitute for phenomenalistic translations; constructivists, at least those who write in the psychotherapy literature, tend to write as if there were no problem at all (in explaining how things that already exist could be constructed by us) that even needed to be addressed. Goodman does discuss the problem, but his replies do nothing to resolve it (Erwin, 1998a).
In the absence of adequate phenomenalistic translations of what it means to say that stars existed before we did, and of any other plausible solution to the problem of explaining how we could possibly create something that already exists, the most reasonable conclusion is that Goodman, Gergen, Maturana, and other constructivists and social constructionists are guilty of a simple logical error. When we invent certain terminology, such as ‘the Big Dipper’, ‘stars’, ‘mid-life crisis’, and ‘learned helplessness’, we create a way of speaking. Say, if you like, that the theorist who invents such terms gives others a ‘lens’ for viewing the world. What does not follow, however, is that the theorist also creates what the terms refer to. Whoever first named the Big Dipper did not also bring it into existence; it existed for aeons before we did. Whoever coined the expression ‘labour union’ might also have created the first labour union, but not merely by introducing the expression.
Sometimes the logical mistake of moving from a statement about our constructing certain terms to a very different one about our constructing reality is hardly noticed. Some writers move back and forth in the same paragraph between talking about terms and talking about their referents without apparent notice of the switch they have just made.
Words, on the other hand, are linguistic creations, and they are part of reality. So, a constructivist might say that at least some things are linguistic constructions, namely words. To say that, however, is to utter a triviality that no-one denies.
The construction of meanings
In the passage quoted earlier from Constructing Realities (Rosen, 1996), the author speaks of constructing not realities but knowledge and meaning: ‘Rather, knowledge, and the meaning we imbue it with, is a construction of the human mind’ (p. 5). This raises a very different issue from what I have been discussing so far.
We should begin by distinguishing talk of constructing knowledge and talk of constructing meaning. The second makes sense, but the first does not. To say that I know that P is true logically implies that P is true. For example, I may think that I know that the client’s problem originated in the pre-oedipal period, but if I am wrong, I do not know that this is so. I only think that I know. To know that P, it must be the case that P is true.
That being so, it makes no sense to speak of constructing knowledge unless we can construct truth. But if we do not generally construct reality, neither do we construct truth. There is a client waiting in my outer office. I form the belief that this is so. One might say, then, that I constructed the belief, or at least the theory, that the client is waiting. What I did not do, however, is construct the truth of the proposition that the client is waiting. I did not, then, construct, invent, or in any fashion make up my knowledge that the client is waiting in my outer office.
As to our constructing not reality or knowledge, but the meaning we imbue knowledge with, this, I grant, can make sense. If this is conceded, why cannot a constructivist reply as follows: ‘Let us put aside these deep philosophical disagree‑ments about truth and the construction of realities. Without presupposing any controversial doctrine about any of these things, I will make my case by talking about the construction of meanings’.
But what exactly is being claimed here? If a client interprets his wife’s refusal to engage in sexual intercourse as meaning that she is having an affair, then there is a sense in which he has constructed that meaning. Reality does not necessarily force this interpretation on him; rather, he constructs a theory to explain his wife’s behaviour. If this is all that is meant by ‘constructing a meaning’, then clients and therapists alike construct meanings. They construct theories about what certain events mean. What epistemological or therapeutic implications does this platitude have? None that I can see.
But perhaps I am moving too fast here. One thing that might be said is that in doing therapy, we should not ask if the client’s interpretation of events is true. The important point, from a purely therapeutic standpoint, is whether the client’s commitment to the new ‘narrative’ will have positive therapeutic effects. I have already discussed this claim; it is not unimportant but, as noted earlier, it is not an epistemological point. Freud and his followers may have been wrong in thinking that correct etiological insight was generally necessary for a lasting cure. Perhaps false interpretations given by either the client or therapist can sometimes be of more help than getting things right. Perhaps, and perhaps not. These issues can be resolved only through empirical inquiry, and, as far as has been demonstrated, resolving them requires no new, special epistemology.
In talking about the creation of meanings, some constructivists may wish to stress a different point. The important point, some will say, is that meanings are not causes; this, in turn, it may be argued, does have important epistemological implications. If the phenomena of interest in psychotherapy are primarily non-causal, then it is wrong to use certain standards of the natural sciences in judging psychotherapeutic hypotheses. In particular, it is wrong to demand experimental evidence for confirmation. Some have gone even further and claimed that intuition rather than observational evidence of any kind is what is needed (Taylor, 1971, p. 51).
Suppose, to take one example, that the client dreams of being a small baby. From a traditional Freudian perspective, what is important is the latent meaning of the dream. If we can figure this out, we may be able to discern something important about the client. From a non-Freudian viewpoint, what is important is not what the dream really means but what it means to the client. On either view, controlled experiments seem irrelevant in explaining what needs to be explained.
To take another example, suppose that a client in his mid-50s has suddenly become sexually impotent. Here we can ask a causal question: does the impotence have a physiological cause? Yet we can also ask a different sort of question – one that may be of more interest to the psychotherapist. What does the impotence mean to the client? Does it serve some symbolic function? Even if it does not, how does the client interpret its significance?
If such cases are typical, or at least common, in psychotherapy and counselling, does this not argue for a type of non-objectivist epistemology? The philosopher Taylor (1971, p. 57) claims that if we take into account the meaning that an event has for an agent, we must also recognise that the disciplines that deal with these meanings, the hermeneutical sciences, are not sciences of verification. We have to accept the radical idea that they are founded on intuitions.
Cannot the constructivist, then, make a good case for a new epistemology by falling back on a plausible assumption that clients do create meanings? I think not.
Begin with the first case. What is the basis for a traditional Freudian saying that the latent meaning of a dream is such and such? In making this judgment, the Freudian is tacitly relying on a causal claim: that the causal instigator of the dream is a repressed wish. This raises an empirical, causal question about Freud’s repression theory and theory of dreams. Furthermore, there are cogent reasons for thinking that in the absence of experimental evidence, or at least epidemiological evidence, confirmation of Freud’s theoretical claims is not likely to occur (Grünbaum, 1984, 1993; Erwin, 1996).
Suppose that we do not rely on Freudian theory, and simply ask how the client interprets the dream. If that is all we ask, then so far we are not attempting to explain anything. If we go on to explain why the client gives the particular interpretation she does, or what it means for her problem, then we are making a causal claim. We are implying that something made a difference to her giving that interpretation, or that her constructing that meaning will make a difference to her problem. Both of these are causal claims.
In the second case, why would the therapist be interested in what meaning the client gives to his impotence unless it is hypothesised that in giving it a certain meaning (for example, that he is now useless as a lover), the client is doing something that will make a difference in his life? If it makes no difference, then the creation of that meaning does not explain anything; if the therapist claims that it does explain something, then a causal hypothesis is being put forth.
So, saying that meanings are not causes can be seriously misleading. Clearly, the meaning of a word or sentence is not a cause, but a client’s interpreting events in a certain fashion is itself an event, one which may have causal consequences. Where it – that is, the interpreting – makes no difference to anything of clinical interest, it also explains nothing of clinical interest. Where it does explain something of clinical interest, it does make a difference, and, consequently, is a cause.
I conclude that the thesis that clients construct meanings, if this means that they interpret events in a certain way, provides no basis for a constructivist epistemology. To know that the client puts a certain interpretation on events in his or her life history, we need empirical evidence. To know, in addition, that the interpreting, the constructing of a particular meaning, explains anything about the origin or maintenance of the client’s problems, or about the course or success of the therapy, requires the confirmation of a causal claim. Not all confirmation of causal hypotheses require experimental evidence, at least not on my view, but many do (for epistemological criteria for distinguishing the two sorts of cases, see Erwin, 1998b).
Constructivists have additional reasons for rejecting an objectivist epistemology, but these reasons have been criticised elsewhere (Held, 1995; Erwin, 1997).
Constructivist epistemology and therapy: is there a connection?
I have talked mainly about constructivist epistemology rather than therapy, but are the two connected? Held (1998, p. 269) rightly asks if the adoption of an ‘anti- realist’ (or what I have been calling ‘an anti-objectivist’) epistemology affects the way therapy is actually practised. Two therapists may have divergent theoretical views and yet do pretty much the same kinds of things when they do therapy. Yet theory can sometimes make an important practical difference. A cognitive – behaviour therapist is not likely to insist that her client lie down on a couch, or to spend much time interpreting dreams or free associations, while a psychoanalytically oriented therapist might do all of these things precisely because of the clinical theory he holds.
Perhaps epistemological views, because of their highly abstract character, are different in their impact from other theoretical views. Even if Neimeyer (1993, p. 230) is right in saying that constructivist psychotherapy is founded on a conceptual critique of objectivist epistemology, perhaps the acceptance of that critique does not really influence the practice of the average constructivist therapist. Although this is possible, I think that it is likely to have some effect in some cases. Neimeyer himself links his rejection of cognitive therapy to his views about epistemology. It does not matter whether the two have any logical connection; as long as he believes that they do, his epistemology may well be influencing his therapeutic practice. Whether in fact that is happening depends on whether he would have rejected a cognitivist approach to therapy even if he had not rejected objectivist epistemology.
Another possible example concerns Maturana’s thesis that in family therapy, we should consider each family member’s construal of the family situation as ‘equally valid’. How can this be right in cases where the beliefs of two family members flatly contradict each other? If being ‘equally valid’ implies being equally correct, then the viewpoints cannot be inconsistent and equally valid. Whether we should try to figure out who is right is another matter; that depends on the facts of the individual case. If, however, we assume a priori that each point-of-view is equally valid, we are likely to go drastically wrong in cases where it is important to find out the truth. If a young woman has come to despise her father because the therapist has unwittingly suggested seeming memories of sexual abuse, and such abuse never happened, it may be vitally important to reject the idea that her viewpoint and that of her father are ‘equally valid’.
Without having the requisite empirical evidence, one cannot know exactly what effects the acceptance of Maturana’s views have on the practice of family counselling, but it is a reasonable conjecture that counsellors who become convinced that there is no mind-independent reality, and that, consequently, the viewpoints of all family members are equally valid, are likely to have a different approach than counsellors who believe that contradictory viewpoints cannot both be correct.
More generally, the idea that there is no language- or mind-independent reality, or at least none that can be known, is likely to have profound and pervasive effects in determining how one thinks about the entire field of psychotherapy. Exactly how this thinking about the field works its way into one’s therapeutic practice cannot be known a priori, but it would be surprising if there were no important effects at all. To generalise the example discussed above beyond family counselling, there may often be sound empirical reasons for not trying to determine the truth of a client’s beliefs, but someone who holds that such a determination is always impossible is more likely to forego the effort even when getting at the truth does matter to the success of the therapy, or does matter to something else of great importance.
Even if it were to turn out that there are no direct connections between the acceptance of non-objectivist epistemological views and therapeutic practice, there are likely to be important indirect connections. Epistemological views matter greatly in determining attitudes towards research, which in turn can affect practice, even if that does not always happen. A cognitive – behaviour therapist who believes that there are objective methods for determining which therapies work under which conditions may well pay more attention to experimental research than a constructivist who thinks that controlled experiments in psychotherapy are generally useless.
There is, finally, a deep division in the field between those who favour a kind of natural science approach to the study of psychotherapy and those who have serious doubts about that approach. There are many reasons for this disagreement, but some depend crucially on whether one accepts or rejects an objectivist epistemology. Acceptance does not automatically thrust one in the direction of a natural science approach, but rejection would appear to raise an insurmountable roadblock.
Despite the possible examples discussed above, I fully agree with Held (1998) that to determine the effects of epistemological views on the practice of therapy, we need empirical studies comparing the therapeutic practices of those who explicitly reject an objectivist epistemology and the practices of those who do not do this. It would be useful, however, to keep in mind two distinct layers of possible influence. When Neimeyer (1993) speaks of constructivist psychotherapy being founded on a conceptual critique of objectivist epistemology, he is presumably referring to the types of therapy that constructivists accept. Without taking a stand on this issue, I find his claim plausible. However, therapists claiming to use a certain type of therapy sometimes deviate considerably from the standard version. This is one reason why there has been a movement toward the development of therapy manuals and the standardisation of therapies. Consequently, there is room for slippage in two places: acceptance of a particular sort of epistemology might not affect the choice of a therapy; but even if it does, it might make little difference to actual clinical practice in cases where the therapist does not faithfully follow the prescribed practices for that therapy.
Efran J. & Greene M. (1996) Psychotherapeutic theory and practice: Contributions from Maturana’s structure determinism. In: H. Rosen & K. Kuehlwein (eds.) Constructing Realities: Meaning Making Perspectives for Psychotherapists. San Francisco CA: Jossey-Bass.
Erwin E. (1996) A Final Accounting: Philosophical and Empirical Issues in Freudian Psychology. Cambridge MA: MIT Press.
Erwin E. (1997) Philosophy and Psychotherapy: Razing the Troubles of the Brain. London: Sage.
Erwin E. (1998a) Anti-science theories in psychotherapy. Paper read at Annual Meeting of the American Philosophical Association, San Francisco CA.
Erwin E. (1998b) The challenge of philosophy to psychotherapists. Paper read to a Scientific Meeting of the Psychotherapy Section of the British Psychological Society, London.
Gergen K. (1994) Toward Transformation in Social Knowledge. London: Sage.
Goodman N. (1978) Ways of Worldmaking. Indianapolis MN: Hackett.
Goodman N. (1984) Of Mind and Other Matters. Cambridge MA: Harvard University Press.
Grünbaum A. (1984) The Foundations of Psychoanalysis: A Philosophical Critique. Berkeley CA: University of California Press.
Grünbaum A. (1993) Validation in the Clinical Theory of Psychoanalysis: A Study in the Philosophy of Psychoanalysis. Madison CT: International Universities Press.
Hayes S. (1993) Analytic goals and the varieties of scientific contextualism. In: S. Hayes L. Hayes H. Reese & T. Sarbin (eds.) Varieties of Scientific Contextualism. Reno NV: Context Press.
Held B. (1995) Back to Reality: A Critique of Postmodern Theory in Psychotherapy. New York: Norton.
Held B. (1998) The antisystematic impact of postmodern philosophy. Clinical Psychology: Science and Practice, V: 264–273.
Horwich P. (1990) Truth. Oxford: Blackwell.
Mahoney M. (1989) Participatory epistemology and psychology of science. In: B. Gholson W. Radish JR., R. Neimeyer & A. Houts (eds.) Psychology of Science: Contributions to Metascience. New York: Cambridge University Press.
Maturana H. (1988) Reality: The search for objectivity or the quest for a compelling argument. Irish Journal of Psychology 9: 25–82. http://cepa.info/598
McCormick P. (ed.) (1996) Starmaking: Realism, Anti-Realism, and Irrealism. Cambridge MA: MIT Press.
Neimeyer R. (1993) An appraisal of constructivist psychotherapies. Journal of Consulting and Clinical Psychology 61: 221–234.
Polkinghorne D. (1992) Postmodern epistemology of practice. In: S. Kvale (ed.) Psychology and Postmodernism. London: Sage.
Rorty R. (1995) Is truth a goal of inquiry? Davidson vs. Wright. The Philosophical Quarterly 45: 281–300.
Rosen H. (1996) Meaning-making narratives: Foundations for constructivist and social constructionist psychotherapies. In: H. Rosen & K. Kuehlwein (eds.) Constructing Realities: Meaning-Making Perspectives for Psychotherapists. San Francisco CA: Jossey-Bass.
Sass L. (1992) The epic of disbelief: The postmodernist turn in contemporary psychoanalysis. In: S. Kvale (ed.) Psychology and Postmodernism. London: Sage.
Taylor C. (1971) Interpretation and the sciences of man. Review of Metaphysics 25: 3–51.
Found a mistake? Contact corrections/at/cepa.infoDownloaded from http://cepa.info/3836 on 2016-11-12 · Publication curated by Alexander Riegler