In Therapy Today 27 (1) February 2016  pp. 24-27


Why the Dodo got it right.         Campbell Purton


What the Dodo says


Almost eighty years ago Saul Rosenzweig1 suggested that the effectiveness of psychotherapy depends not on the specific theories and practices of therapists, but rather on features of the therapy process that are common to most therapies.  The title of his paper included a quotation from Alice in Wonderland, ‘At last the Dodo said “Everybody has won and all must have prizes”.’  This suggestion has been strongly confirmed by more recent research into the effectiveness of therapy. From the late 1970s the development of statistical methods of meta-analysis of outcome studies has led researchers to the firm conclusion that psychotherapy can be remarkably effective2; however, at the same time such studies have equally firmly established that the relative effectiveness of the different forms of therapy is about equal.


In spite of this latter finding, the traditional schools of therapy continue on their way, broadly relying still on the theoretical frameworks within which they grew up.  In practice, the truth of what ‘the Dodo’ says does not require them to change, since for each school of therapy the evidence is that its  practice is effective.  On the other hand, the present situation is theoretically very uncomfortable.  The various theoretical frameworks are incompatible with one another and cannot be integrated into an overall super-theory.


The consequences of this situation for the assessment and regulation of psychotherapy practice are very serious.  In assessing trainees’ competence to practice, BACP has always considered whether the applicant has demonstrated that their practice is consistent with their theoretical perspective. There has never been any requirement that the applicant’s theoretical position should itself be a sound one, since such a requirement would require consensus over which theories are sound  The growing demand for statutory regulation of therapy has highlighted the problem, since while many governments have understandably wanted there to be such regulation, neither governments nor bodies such as the Health Professions Council are in a position to judge which approaches to therapy should be recognised. Within Europe this had led to wide discrepancies: Austria’s legislation recognises sixteen different approaches to individual psychotherapy, Finland’s just two3.


There seem to be three possible responses to ‘the Dodo’.  One is to say that in spite of all the evidence, just one of the traditional theories is broadly correct, while the others are incorrect.  But given the strong empirical evidence for the equal effectiveness of therapies grounded in the different theories, this option seems very unattractive.


The ‘postmodern’ response to the Dodo


The second response is to invoke postmodernist ideas, and to say that while the theories cannot be merged into a super-theory, they can all be accepted, as they stand, in spite of their incompatibilities.   This idea is the relativistic one that the world can be seen in many different ways.  It can be combined with a pragmatic view of the use of theories: simply use whichever theory seems to work in any particular client situation.  Cooper and McLeod’s4 development of ‘pluralist therapy’ is a prominent example of this response.  However, postmodernism is a highly controversial philosophical position, and my suspicion is that the problems we face in connection with the assessment of psychotherapy approaches are unlikely to be resolved at some high level of philosophical argument.  The issues involved are much more down-to-earth. What matters is not the high-level question of whether incompatible theories can all in some sense be ‘true’, but the question of whether in practice a therapist can, for example, take seriously both psychodynamic and cognitive-behavioural theories.  Taking a theory seriously involves a commitment to a particular way of seeing things, and to a particular view of what sort of engagement with clients is appropriate. In common-sense terms, the psychodynamic therapist trusts in the general picture that Freud paints of the nature of human beings, and the CBT therapist trusts in a very different picture. They say these pictures are (more or less) true, but truth here is not an abstract philosophical notion; it is bound up in practice with trust (indeed, these words are etymologically related). 

Trust in a particular theory gives solidity and coherence to a therapist’s work.  A therapist needs a rationale for their work, and clients need to know (or at least believe) that the therapist has such a rationale.  Empirical studies5 of ‘therapeutic allegiance’ (therapists’  belief in their theories) and ‘client expectancy’ (clients’ belief that they are being treated in a way that has some effective rationale to it) strongly confirm these common-sense ideas.  However, it seems clear that a therapist can’t believe in (trust, give allegiance to, think mainly in terms of, habitually relate their clinical experience to) more than one theory at a time.  Nor does it make much sense to suppose that a therapist could be a psychoanalyst on Mondays, a CBT therapist on Tuesdays and a person-centred therapist on Wednesdays.  That would be like saying one could be an Arsenal supporter on Saturdays and a Chelsea supporter on Wednesdays. I think that someone who claimed to hold such a position wouldn’t really count as a football supporter at all! So, there is an important sense in which we can’t hold that all the theories are true.

The ‘placebo effect’ response to the Dodo

A third response to ‘the Dodo’ is to be found in the qualitative work of Jerome Frank6, and in the later quantitative research of Bruce Wampold2.  Their view is that the effectiveness of psychotherapy is to be explained by factors that have nothing to do with the truth of the theories in which the therapies are supposed to be grounded.  To put it bluntly, the theories in themselves are worthless, but the practices associated with them, and which they largely share, can be very effective.  In Frank’s view, the important factors in therapy are the human relationship between therapist and client, a therapeutic rationale that provides an explanation for the client’s difficulties, and a procedure which can be seen to be grounded in the rationale.  Frank argues that these conditions for psychological healing can be met by various traditional forms of religio-magical healing, as well as by most forms of psychotherapy.  What distinguishes psychotherapy from traditional methods is that in psychotherapy the rationale is a psychological theory.  But which theory is immaterial.

Wampold’s statistical evidence does much to support this suggestion. He concludes that the effectiveness of therapy depends on the client’s working alliance with the therapist (including agreement on the goals and tasks of therapy), the client’s motivation, and the ‘expectancy’ and  ‘therapist allegiance’ factors mentioned above.  There still needs to be a rationale, or theoretical framework, for therapy but, again, which rationale is employed makes no difference to therapeutic effectiveness.

What this kind of view amounts to is that psychotherapy works in the same sort of way that placebos presumably work in medicine.  The effectiveness of a placebo seems to depend on the human relationship that the doctor establishes with the patient, the client’s belief that the treatment will work, and the doctor’s belief that it will work. But here the doctor’s belief is not the same as that of the patient.  The patient believes that the pill will cure them, whereas the doctor believes that the pill will cure them, given the patient’s belief in it.  This difference can give rise to ethical difficulties about ‘informed consent’ in connection with the  medical use of placebos7 and such difficulties would be even more acute if it were generally accepted by therapists that the effectiveness of their procedures was analogous to that of medical placebos. There is evidence that effective therapy requires a trusting relationship between therapist and client, and agreement on the aims and methods of therapy. However, in placebo therapy, there cannot really be such agreement.

In fact I think it is very implausible that the whole of psychotherapy is analogous to a placebo effect.  I say this because it seems to me that in many cases we can see what it is that is effective in therapy, in addition to possible placebo effects.

A very brief sketch of some kinds of therapy

One common sort of therapeutic procedure involves ‘behavioural modification’. This might involve, for example, gradual desensitisation to situations which the client knows not to be really dangerous, or help in managing a client’s attempts to give up smoking through setting up contingencies in which smoking becomes difficult or unpleasant.  Gradually the client’s responses change as they ‘get used’ to the situation that seemed so alarming, or to life without smoking.  A significant range of client problems seem to fall into this category. Behaviourist learning theory had an ‘explanation’ of the effectiveness of such techniques, but that theory has long been abandoned in psychology8.  However it is not obvious that we need a theory here. The fact that we gradually get used to frightening things, or can adapt to altered circumstances, is basic to our understanding of human life  -- what would our lives be like if we didn’t adapt in such ways?  It makes sense to try to explain certain kinds of behaviour change in terms of ‘getting used to something’, but it is unclear what we should say to someone who wanted an explanation for the fact that people get used to things. They might be asking about what goes on in the nervous system when people get used to things, but that is a different kind of question.

A second common sort of therapeutic procedure involves clients reflecting on what they are feeling, considering their actual situation and how they are responding to it, and then, after trying out various ways of putting it, finally acknowledging that their actual feelings are not what they initially thought they were.  “So … it seems that I am jealous!”  A psychodynamic therapist might ‘explain’ this process in terms of ‘making the unconscious conscious’ and a person-centred therapist in terms of the client’s ‘increasing congruence’.  But, as in the previous kind of example, it is not obvious that we need the theories that are being invoked here. It is part of our common-sense understanding of human life that we sometimes deceive ourselves.  We don’t always give our full attention to things that we don’t want to acknowledge, and it is only when someone or something forces us to look more closely at our situation that we come to acknowledge how things really are.  We can easily understand, without the help of theory, how this exploratory, clarificatory kind of approach works.


I think that a significant proportion of therapy issues fits into one or other of the above two patterns.  In the first kind of case the client needs help in bringing their reactions into line with how they really see their situation. (‘Spiders aren’t scary after all!’).  In the second kind of case the client needs help in bringing how they see things into line with how they are in fact reacting (‘I’m jealous, not irritated!’).  The two kinds of case are the converse of one another, but in both cases the therapist is concerned with helping the client heal a split between their actual reactions and their considered view of their situation.  In both cases the therapeutic process works through the therapist helping to bring into greater prominence the side of their life that the client is neglecting.  In the first kind of case the client is caught up in their response to the situation (e.g. spiders are scary), and the therapist works with how the client really sees the situation, in order to help them in modifying their response.  In the second kind of case the client is caught up in a distorted view of things (‘I’m certainly not jealous’), and the therapist works at drawing their attention to what their response really is. We might say that in the first kind of case the therapist supports the client in ‘walking their talk’, and in the second in ‘talking their walk’.  Of course both kinds of case need much more discussion (to be found elsewhere9), but I hope to have conveyed both how different they are, and how they are, nevertheless, closely related.

Writers on CBT tend to choose examples from the first group of problems; writers on psychodynamic and person-centred therapy tend to choose examples from the second group.  Neither kind of procedure requires anything by way of theory in order to understand it, but nor can the effectiveness of such procedures be understood in terms of a placebo effect.  No doubt placebo factors such as ‘client expectancy’ and ‘therapist allegiance’ are important, but they are not the whole story.  More important, according to the research findings, is the human relationship between client and therapist.  But it is misleading to call that element ‘placebo’.  The personal interaction between therapist and client seems healing in itself, and I suspect that is partly due to the support it provides for clients to try out and/or articulate responses that they are uneasy about.

Why the Dodo is right

Taken together, I think the three closely-related elements of personal interaction, behavioural adaptation to situations, and release from self-deception, form the central core of much effective therapy.  My guess is that behavioural procedures work best for the first group of problems mentioned above, and psychodynamic/person-centred procedures for the second group, but in practice client issues are often a mix of problems, and if we lump all client problems together we should expect none of the major approaches to therapy to be, overall, more effective than any of the others.  They are all effective, but not because of the truth of their associated theories; they are effective because they can, in a very common-sense kind of way, heal the splits that bring people to therapy.


Campbell Purton is retired from his post as Senior Lecturer in the Centre for Counselling Studies at the University of East Anglia. His latest book is The Trouble With Psychotherapy: Counselling and Common Sense. Email:


1.      Rosenzweig S. Some implicit common factors in diverse methods of psychotherapy: "At last the Dodo said 'Everybody has won and all must have prizes.'". American Journal of Orthopsychiatry 1936; 20: 412-415.

2.      Wampold, BE. The Great Psychotherapy Debate. London: Routledge; 2001. Cooper, M.   Experiential Research Findings in Counselling and Psychotherapy. London: Sage; 2008

3.      Austrian Health Institute (ÖBIG) (2003).  Regulation of the professions of psychotherapist, clinical psychologist, health psychologist in the member states of the EEA and the Swiss Confederation 2003;   Downloaded 5 April 2014.

4.      Cooper M & McLeod J. Pluralistic Counselling and Psychotherapy. London: Sage; 2011

5.      Wampold BE. The Great Psychotherapy Debate. London: Routledge; 2001: 165-168.

6.      Frank JD. Persuasion and Healing: A Comparative Study of Psychotherapy. Baltimore: Johns Hopkins University Press; 1961

7.      Shapiro AK & Shapiro, E. The Powerful Placebo. Johns Hopkins University Press;   1997

8.      Gardner H. The Mind's New Science: A History of the Cognitive Revolution. New York: Basic Books; 1987

9.      Purton C. The Trouble with Psychotherapy: Counselling and Common Sense. Basingstoke: Palgrave Macmillan; 2014