Truth, Politics and Psychological  Therapy – May The Lord Delivers Us from Professional Uncertainty

 Talk given to the Humanities and Mental HealthResearch Network in Nottingham, 1st March 2006

I feel quite diffident about airing my views in public these days.  Having retired from NHS clinical psychology nearly eight years ago, and from voluntary clinical practice about three years ago, I find it surprising how quickly things seem to change, and how almost dizzyingly out of touch one can get in what seems no time at all.  However, I’m not going to let that stop me holding forth this evening, in the hope that what a pensioner such as myself lacks in currency may to some extent be compensated for by length of perspective.

            I had thought until recently that I could fairly safely, even if tentatively, draw a few conclusions from nearly forty years’ experience of thinking and reading about, as well as practising and writing about psychological therapy. 

In many ways what seemed to me the most important of these conclusions followed from my observation from clinical psychological practice—my own and others—that the various theories put forward to underpin psychological therapies of pretty well all persuasions were not supported by what actually happened to patients/clients.

            I can best sum up my observations with the help of a reasonably simple table:-

Theory suggests

Clinical experience teaches

Insight leads to change

We are not in control of our conduct; therapeutic change is not demonstrable


People may ‘assume responsibility’


There is no such thing as ‘will power’


Thought (‘cognitions’) leads to action (‘behaviour’)


The causes of our conduct are frequently mysterious, and rationally unalterable


Characteristics/actions (real or imagined) of the therapist are central to change (e.g. ‘transference’; ‘warmth, empathy and genuineness’; behavioural and/or cognitive manipulations, etc.)

Patients’ conduct is controlled by more potent influences in their social environment


If in fact one wants to understand why people behave as they do and why also they so often find it very difficult or impossible to change—even when they desperately want to and bravely try to—it seemed to me very obvious that one would need to take into account the world they live in.  People don’t just exist in a consulting room with their therapist, they live in a  highly complex web of relations, personal and impersonal, within a society that is structured by powerful influences over many of which they have no control at all.  In this kind of situation ‘choice’ is not always, or even often, the option it seems to be.

Now this does really seem, does it not, pretty obvious.  So why on earth would the creators and theorists of so many therapeutic approaches overlook or ignore the importance of the social world in the formation of individual conduct?  Why not, for example, let experience shape theory, rather like this, perhaps?:-

Clinical experience teaches

Theoretical requirements

We are not in control of our conduct; significant therapeutic change is not demonstrable

Need to identify the causes of conduct beyond personal agency.  These likely to be biological as well as social

There is no such thing as ‘will power’


Need to establish the limits of ‘responsibility’;  factors that make free action possible

The causes of our conduct are frequently mysterious, and rationally unalterable


Rationalist explanations insufficient;  need to account for the disjunction between conduct and the accounts we give of it


Patients’ conduct is controlled by more potent influences in their social environment

Need to develop a multi-layered understanding of influence (including therapeutic influence), its modes of operation and the reasons for the permanence/impermanence of its effects


Well, I think the answer to this is fairly obvious as well, and that is that it is not in the interest of practitioners of psychological therapies to do so.  In order to justify making a living by holding out the promise or expectation of curing or alleviating clients’ distress, one has to individualize its causes as well as the mechanics of its cure: they just must be put down to what can be identified in the consulting room and what goes on between its occupants.  To admit the primary importance of social influences and the environmental conditions of people’s lives takes the whole question of ‘cure’ out of the hands of psychologists, counsellors and others, and identifies it as an essentially political issue.  In most important respects we do not choose to be who we are—the world (including biology) makes us such.

Now, up until a few weeks ago I was fairly happy with this broad conclusion.  It’s not that the vast literature that had grown since the mid-twentieth century was completely barren, by any means: there was plenty to show that most therapeutic approaches led to modest improvement for clients, as did drug treatments come to that.  It was just that such therapies tended to be nowhere near as potent as their practitioners would have wished, were on the whole pretty impermanent, and depended for their effectiveness not on the technical practices through which they were delivered, but on the quality of the human relationship between therapist and client. 

It seemed to me that the possible benefits of therapy could be summed up as clarification, comfort and encouragement.  You didn’t necessarily need trained professionals to provide these—and indeed there is a fair amount of evidence to suggest that they can be equally well if not better dispensed by lay people (ideally, of course, they should be woven as permanent features into the social fabric). It seemed, in fact, that roughly two-thirds of clients improved somewhat under pretty well any therapeutic regimen, while the other third stayed the same or got worse.  While not exactly disastrous for the practice of psychological therapies, this didn’t seem anything much to shout about either.

 Now in view of the extent to which their interests are challenged by what was the research evidence, it didn’t exactly surprise  me that professional psychological therapists and counsellors should simply carry on regardless, and this is certainly true of my own profession of clinical psychology, which has continued to develop and promulgate the virtues of ‘cognitive behavioural therapy’ (CBT) undeterred by any scientific scruples about the lack of really significant evidence for its effectiveness.  What I personally hadn’t reckoned on, however,  and which has only recently dawned on me, was the extent to which clinical psychology’s interests would flow together with other ideological and political interests to set up what threatens to become a kind of scientistic hegemony in the general field of talking therapies and mental health.

There seem to be a number of strands to this, the central engine of which is the drive towards the establishment of a market in health care that diverts public funds into private pockets (cf Pollock, 2005).  In order to achieve this, you need political, legislative power and a managerial bureaucracy to control both professional activity and the production of knowledge itself.

Rather amazingly, in view of its heretofore relative insignificance, clinical psychology now finds itself at the centre of a situation where all these elements are pretty well in place.  This has been revealed particularly clearly, I think, by the coming of the lord—in this case Lord Richard Layard, professor of economics at the LSE and (?former) adviser within the Prime Minister’s Strategy Unit, which, so it says, is a management unit of the Cabinet Office.

Lord Layard has produced two documents in particular that have caused great excitement among clinical psychologists.  The first is a Strategy Unit production entitled Mental Health: Britain’s Biggest Social Problem? produced about a year ago, and the second is a paper delivered at the Sainsbury Centre last September and called  Therapy for all on the NHS.  The more recent paper mainly rehearses the arguments of the earlier one but is somewhat less discreet in tone, and worth reading for that alone.

Now it’s pretty clear what Layard’s agenda is, indeed he sets it out in the first paragraph of Mental Health: Britain’s Biggest Social Problem?:

Mental illness is one of the biggest causes of misery in our society – as I shall

show, it is at least as important as poverty. It also imposes heavy costs on the

economy (some 2% of GDP) and on the Exchequer (again some 2% of GDP).
There are now more mentally ill people drawing incapacity benefits than
 there are unemployed people on Jobseeker’s Allowance
[My italics]


Scattered here and there in both documents are almost lachrymose references to, e.g., ‘the torment of mental illness’ (Layard 2005, p2), but its obvious that Layard’s focus is the money, not the misery.  This is implied in the introduction to his Sainsbury talk.  Meant perhaps to be disarming, it seems to me almost menacingly disingenuous:-

I am honoured to be asked to give this lecture, especially as I probably know less about mental health than any one here.


Most revealing, though, is in my view a passage from the Sainsbury lecture introducing the section ‘Pathways to Work’:

So we desperately need a better NHS, delivering more help and understanding to patients. But for many patients, work is also a major route to recovery.  And as taxpayers who pay for Invalidity Benefits, we can all say amen to this.


Luckily, especially for clinical psychologists, Layard has identified exactly what’s needed to deliver help and understanding, as well of course as get people off benefits and back to work, and that is CBT, delivered by teams headed by clinical psychologists located in treatment centres set up on a ‘hub and spoke’ pattern throughout the country.

To realize Layard’s vision, the number of psychologists would need to be doubled, and recruitment of psychiatrists would also need to be drastically stepped up.  So pretty well everybody is likely to be cheered by these proposals.

The key to all this, of course, is the pre-eminent effectiveness of CBT, of which Layard has become convinced largely through the good offices of our, in my view, absurdly named National Institute of Clinical Excellence (NICE).

Now it’s with NICE that we come down from the political stratum to the managerial bureaucracy that’s needed to realize political aims.

Despite his self-confessed ignorance of such matters, Layard doesn’t hesitate to prescribe what his proposed therapeutic professionals would do.  For example:-

The staff would operate under clear NICE guidelines relating to number of sessions, and patient progress would be monitored using a standard national system of recording completed at the beginning of each session. (p4)


            Since NICE guidelines are bandied about—not only by Layard—as almost infallible indicators of so-called ‘best practice’, I thought I’d better steel myself to reading some, so I had a look at the recent guideline on “OCD and ‘body dysmorphic disorder’” (that caused a brief flurry of excitement in the media a few weeks ago), containing 53 pages of recommendations, and the slightly earlier full, 165 page guideline on the ‘management of anxiety’.

            In many ways these make fascinating, if somewhat numbing, reading, since they appear to wipe out almost everything I thought I knew about psychological therapies.  In their way they’re masterpieces of pseudo-scientific bureaucracy, taking the Taylorization of intellectual endeavour and clinical practice to an extreme I wouldn’t really have thought possible. 

They do indeed assert that CBT is the treatment of choice for the ‘disorders’ in question, though it is noticeable that ample room is left for drug treatments.  There is a kind of algorithm of recommendations for the various DSM categories under consideration, and the recommendations are scrupulously backed up by recitation of the ‘evidence base’ that has been identified for them.

 I simply haven’t got the heart or the spirit to attempt a detailed critique of this ‘evidence base’, though relevant critiques do exist (I think in particular of the chapter on CBT in a forthcoming book by William Epstein (in press).  What’s striking about it—the evidence base, that is—is its near total reliance on rationalized/mechanized research methodology together with an equally near total indifference to the actual content of the research (this is demonstrated in reliance placed on review articles and meta-analyses).  A kind of unholy alliance is built up involving the DSM IV, the Cochrane Library and various limiting methodological requirements concerning control groups, double-blind trials, etc., which results in the virtually automatic churning out of ‘results’ that with a kind of deadening inevitability support the relative superiority of CBT.  Let me just give you a taste of this – we haven’t got time for a measured critical analysis (see appendix – there are 151 references in the guideline, most of them described in this format).

None of the factors that I had always taken research in psychological therapies to point up as being important receive any real consideration or influence the NICE evaluation of the so-called evidence. Very little attention is paid to who ‘delivers’ the treatments or what their theoretical allegiances are or what the characteristics, personal or demographic, of clients are beyond their age and diagnosis. Nothing about the quality of relationship between treaters and clients.  Huge reliance is placed on DSM IV and on self-report questionnaires to indicate improvement or otherwise.  Reflecting a total faith in mechanization, ‘measurement’ is everywhere and meaning nowhere; questions concerning reliability are taken for granted while validity is on the whole simply not considered. There is no indication of what practitioners actually do beyond the ascription of an orientation (mostly variants of behavioural and cognitive approaches).

This is in fact a strangely unhinged, make-believe world in which entirely hypothetical constructs—mere words—are taken as necessarily pointing to valid entities in the real world (the DSM productions are of course prototypical in this respect).  The combination of an almost metaphysical set of beliefs about the potency of ‘cognitions’ and the nature of ‘disorder’ with  obsessively detailed procedures of ‘measurement’ and statistical analysis is in fact uncannily reminiscent of the procedures of 17th century astrologers as recounted by Keith Thomas in his Religion and the Decline of Magic: interestingly, astrologers pointed to the complex methodology and mathematical intricacy of horoscope production as an argument for its scientific validity.

What emerges at the end of all this, anyway, is that CBT appears to be ‘effective’ in about two thirds of cases, which, as I said earlier, is what we already knew to be the case with pretty well all therapies, talking or otherwise.  The apparent triumph of CBT is thus a kind of sleight of hand performed, so to speak, on a darkened stage where the world beyond its rationalist/mechanist limits is simply not visible.

The NICE version of things may not be valid or remotely true in any significant sense, but it is certainly useful in maintaining an individualized model of ‘disorder’ as well as enabling the central control and direction of professional activity, whether in research or practice.  As you can imagine, it’s also not uninteresting to those practitioners who stand to gain from its undisputed rule.

 Prominent clinical psychologists Tony Roth and Pam Stirling, for example, writing in the clinical psychology house journal Clinical Psychology Forum (November 2005), though they feel that Layard may be a bit too enthusiastic about what CBT can achieve (‘…there is a real risk that the efficacy of CBT is being over-sold.’), are nevertheless quite clear about which side their bread’s buttered:

The government will only be persuaded to move on these proposals on the basis of hard-headed arguments, especially because this is—at the end of the day—an exercise in transfer of costs between departments, justified by a projection that this will be a cost-neutral exercise with major social benefit.  It is for this reason that the rationale of offering evidence-based treatments of known efficacy is pragmatic, even if not completely consonant with clinical opinion.  The case for the profession pulling together on this is clear.  (p 48, my italics).


In a piece in The Psychologist (January 2006) entitled ‘Responding to Lord Layard’, the Chair of the BPsS Division of Clinical Psychology, Graham Turpin, is also just a little nervous that Layard’s proposals could mean jobs for only some of the boys and girls, but feels nevertheless that:

Whereas the evidence for the efficacy of CBT interventions exists, the contribution of other evidence-based psychotherapeutic approaches could also be factored within the design of psychotherapy services to ensure that clients have a real choice of talking therapies on offer by the NHS. (p 12)


Turpin proposes therefore that the Society should formulate a response ‘as to how psychologists can help drive this agenda forward’.

            What all this amounts to is that we have lost any semblance—indeed any pretence—of pursuing scientific inquiry, what is the case, what is true.  Just listen again to those sentences of Roth and Stirling:  ‘…the rationale of offering evidence-based treatments of known efficacy is pragmatic, even if not completely consonant with clinical opinion.  The case for the profession pulling together on this is clear.’  This is almost classic in its near-phobic avoidance of considering anything that could possibly be construed as speaking the truth about the world.  What we deal with, rather, are rationales, pragmatics and opinions.  What is not disputed, of course, is that these latter constitute what one might call a clear interest-base for professional action.

            The reluctant conclusion that I find myself forced to contemplate is that truth plays only a very minor role in shaping the conduct of human beings, and one should not be puzzled by the observation that no one seems to take much notice of it.

That human behaviour is best understood as the upshot of practical and ethical choices based on reasoned consideration of reality, is, I suggest, a kind of myth that inhabits the very core not only of our everyday thinking—what has been called popular psychology—but also of almost all formal, psychological theories, including broadly clinical ones (including indeed, despite appearances to the contrary, psychoanalysis).  If, as it has professed to do, psychology has pursued the ‘prediction and control of behaviour’, this essentially rationalist/cognitivist model has got it precisely nowhere.  Control of behaviour seems to me a vain as well as a disreputable aim, but, in my view, anything like a valid understanding of human conduct—maybe even with a bit of accurate prediction thrown in—will not be possible without the recognition that what most of us do most of the time is react to the almost blind operation of power and interest within an inescapably social framework.  The ‘almost’ here is very important however, as in certain unusual and protected circumstances the pursuit of truth, as well as of justice, may indeed be possible.  Such circumstances are unusual because they require an extremely artificial control of the operation of power and interest, whereby certain kinds of interest are suppressed (e.g. financial gain or personal glory) and certain others maximized (e.g. technical, practical or emancipatory interests of the kind that Habermas wrote about in Knowledge and Human Interests).

As far as everyday life is concerned Niccolo Machiavelli and Karl Marx probably make much more convincing psychologists than do Sigmund Freud, B.F. Skinner or, God help us, Aaron Beck.

If I was nominating people for best 20th century psychologist, theoretical and applied, I certainly wouldn’t think of anyone who emphasized cognitivist rationality in their scheme of things.  For ‘best theoretical psychologist’, I think  I’d nominate Mandy Rice-Davies.  Not everyone here, I imagine will remember or even have heard of her, though most will know the pronouncement she was famous for.  She was a kind of up-market prostitute who gave evidence in the court case that arose out of the infamous Profumo scandal in 1963.  When prosecuting counsel pointed out to her that the then Lord Astor denied having had an affair with her, as she claimed, or even having met her, she replied ‘Well, he would, wouldn’t he’.  That demonstates to me a more profound understanding of human motivation than just about any ‘official’ theory I can think of.

                My nominee for best applied psychologist goes to someone everybody knows:  Margaret Thatcher, who demonstrated within weeks of coming to office that one can change the behaviour of virtually a whole population by the ruthless application of power.  Thatcherism, in fact, introduced a world, now firmly established, in which the operation of crude economic interest plays the fundamental, almost the only, role.  The social and cultural, not to say psychological/emotional devastation this causes is maintained by a web of nearly impenetrable institutional power/interest relations and obscured by a haze of Business gobbledegook, academic claptrap, and psychobabble.


God knows how we begin to get a grip on all this. I’d have thought that, to start with, and even if we have to maintain an element of secrecy about it, the best policy might be honesty.



Habermas, J.  1968.  Knowledge and Human Interests.  Heinemann.

Epstein, William Psychotherapy as Religion: The Civil Divine in America, to be published by

the University of Nevada Press

Layard, Richard. 2004. Mental Health: Britain’s Biggest Social Problem?


Layard, Richard. 2005. Therapy for all on the NHS.   

Pollock, Allyson. 2005. NHS plc.Verso.

Roth, T. & Stirling, P.  2005.  Expanding the availability of psychological therapy.  Clinical

Psychology Forum, no.155 (November), 47-50.

Thomas, Keith.1973. Religion and the Decline of Magic.  Penguin Books.

Turpin, G.  2006.  Responding to Lord Layard.  The Psychologist, 19, no. 1 (January), 12.



NICE Guideline on Anxiety, pp132-3

CBT versus supportive counseling

Barrowclough et al 2001
This randomised controlled trial sought to measure the effectiveness of cognitive behavioural therapy in older adults with a range of anxiety disorders. Patients were aged over 55 years and had a diagnosis of an anxiety disorder according to DSM IV including GAD, panic disorder with or without agoraphobia, social phobia and anxiety disorder not otherwise specified. Patients on medication had to maintain constant dosage through the study. Randomised patients entered a 6 week baseline phase in which no treatment was administered before being randomised to between 8 and 12 1 hour sessions of either CBT or Supportive Counselling. Each patient completed a credibility questionnaire to assess treatment credibility at sessions 2, at the end of treatment and then at 3, 6, and 12 month follow-up. Of 225 referrals, 55 fulfilled the inclusion criteria. After baseline, 9 dropped out before therapy and a further 3 became seriously ill and 3 dropped out by the 4th therapy session. Thirty nine patients were available for 3 and 6 month follow up and 40 for 12 month follow-up. 51% of patients had Panic Disorder and 19% had GAD. The primary outcome measures concerned Global anxiety and included three self-report questionnaires (the Beck Anxiety Inventory, The Spielberg, State Trait Anxiety Inventory, Trait version and a 20 item measure of anxiety). The Hamilton Anxiety Rating Scale was also used. Depressive Symptomatology was measured but is not reported here. The study design did not employ an intention to treat analysis. Results were tested for skewness and parametric and nonparametric tests were applied as appropriate. ANCOVAs were performed using pretreatment scores as the covariate. There was no significant difference between groups on treatment credibility. CBT had a significantly better outcome than the SC group on the BAI (F(1,42)=5.29, p<.05 and the Geriatric Depression Scale. There was a significant improvement within treatment on all measures apart from depression which did not improve. On all the measures, CBT, showed a significantly better outcome than the SC group with CBT also demonstrating a significant time by treatment interaction (F(3, 105)=2.39, p<.08).
To assess clinical significance, 2 measures of treatment response or magnitude of change were taken as being meaningful. This with endstate functioning being within the normal range. The 20% reduction was found to be a meaningful cut-off in earlier documented research (Stanley et al 1997). Seventy one per cent of CBT and 39% of SC patients met the criteria for responders for anxiety at 12 month follow-up. More patients in the CBT group showed clinically significant response (?2 (1, N=40) = 3.88 P<0.05). There was no significant difference between groups in proportion of responders on depression symptoms. Neither was there any significant differences in endstate functioning between groups on either anxiety or depression. 41% of the CBT and 26% of the SC group had high endstate functioning at 12 month follow-up. The authors state that results from this study show that CBT treatment may be effective when delivered in a format of a mean of 10 sessions with a primary emphasis on cognitive techniques.

David Smail
Nottingham March 2006

Back to:-
Unpublished Talks