(Editor's Note) George Stricker had the
good idea of forwarding to the SEPI list a contribution on another
list by Larry Beutler, some key points of which are the ones that
follow: "Scholarly reviews of research by those within the broadly
based psychotherapy research perspective have long held that the psychotherapy
relationship, patient variables, and other non-technical factors,
are more highly related to outcome than the specific procedures and
theoretical model employed by the therapist…The low percentage
of variance contributed by differences in the model and procedure
used has been the basis for the Do-do Bird verdict---all psychotherapies
obtain essentially equivalent results, a finding that keeps being
repeated in the preponderance of meta-analytic reviews… It is
fair to conclude that the correlations between qualities of the therapeutic
relationship and outcome are almost always higher than the relationship
between type of therapy applied and outcome--i.e., the relationship
accounts for more of the variance than what the therapist does from
a theory of change perspective." This started a debate in which
mainly two positions were confronted. In one, as represented by George
Stricker, "The Do-do bird challenge is to find differential effectiveness
and, by doing so, get beyond common factors as the basis of all therapeutic
change." An alternative way would be not to get beyond common
factors in search of differential effectiveness, but on the contrary
to look right there for the basis of all therapeutic change. The debate
highlighted that these two positions are not to be thought as mutually
exclusive, but rather as complementary.
George Stricker, 5 June 2000
Larry Beutler published the following on another list to which I
belong. For those who haven't seen it and are interested, it presents
a very comprehensive and useful summary of a very important area that
has been part of the SEPI debate many times. I am forwarding it with
Larry's permission.
________________________________________________
Jeff Lohr has raised a very important question about the causal relationship
between the therapeutic alliance and outcome of psychotherapy. Distilled,
Jeff's comment connotes some concern with two issues: (1) that the
"therapeutic relationship/alliance" is touted as being more strongly
related to outcome than the specific things that a therapist does
or the particular model of therapy used, and (2) that is is assumed
that the therapeutic relationship/alliance is causally related to
outcome. Jeff wonders about the empirical evidence for these two assumptions.
The assumption that the relationship is a more powerful contributor
to therapeutic outcome than the procedures used by the therapist is
a common assumption among those who identify themselves with "Psychotherapy
Research" and this position is represented among the membership of
SPR (the Society for Psychotherapy Research, International). Jeff's
comment, and the responses to it, remind me, however, that there is
a different perspective represented among those who largely identify
themselves with a specific, usually behavioral tradition. These individuals
may be inclined to be more closely affiliated with AABT than with
SPR. The latter viewpoint was expressed reasonably well by Bill Follette
who expressed a lack of confidence in the research on therapeutic
alliance or relationship. He expressed the view that (1) the measurement
of therapeutic alliance or relationship was confounded with outcomes,
and (2) the role of enhancements to the therapeutic relationship could
be understood as a process in which the therapist enhanced their reinforcing
power.
In the hope of increasing dialogue on these points, across the perspectives
that differentially characterize these different scholarly traditions,
I'd like to summarize some of the reasons for the two assumptions
that Jeff initially brought into relief and in the process to address,
through available research, why Bill's analysis of the relationship
as a reinforcement enhancer does not do justice to the complexity
of the process.
First, let me point out that these two different traditions, what
I will simplistically refer to as "the psychotherapy research" and
"the behavioral analysis" traditions, often differ very widely in
what they select and focus on as the predictor and independent variables
in treatment research. Admitedly, this is a simplistic distinction
because there are a lot of overlaps between them, and frequently there
are people who keep their hats in both camps. But, there are also
some very clear differences between them. I remember David Barlow,
for example, at an APA meeting a year or so ago, pointing out that
in the circles in which he travels, the idea of "non-specific" effects
or the Do-Do Bird verdict (assumedly related to the supremacy of relationship
factors over technique and model factors) were seldom mentioned. David's
view, I suspect, would be shared by most of those who identify themselves
with "behavioral research" more than "psychotherapy research". And,
this view leads people, I suspect, to place great emphasis and faith
in the techniques of intervention and the theoretical model from which
it comes. It is from this tradition that RCT research makes the most
sense and the common assumption is that there are specific treatments
for specific disorders.
While the RCT tradition, and the model of specific diseases and specific
treatments that underlie it, certainly is conducive to the effort
to define Empirically Supported Treatments initiated by Dave Barlow
through SSCP, psychotherapy researchers of the SPR tradition are more
persuaded by the influence of variables beyond specific, theory-derived
procedures (e.g., relationship, therapist characteristics, etc.).
Hence, John Norcross's development of a task force that is complementary
to the Division 12 effort and that is designed to identify "components
of the empirically supported relationship" (Incidentally, I speak
as a member of both the Division 12 and the Division 29 Task Forces).
David Barlow's experience is very different from mine and I suspect,
from the experiences of many of those who work within a broadly based
"psychotherapy research" tradition. The assumption of the pre-eminence
of techniques and specifically of cognitive and behavioral techniques
over these external variables, is one that is often disparaged by
broadly based psychotherapy researchers. These latter researchers
are concerned, in their research, with a wider range of contributing
and correlating variables than those that derive from a given theory
like either IPT or CT. Their concerns are reflected in the chapters
on Therapist, Client, and Process variables in the HANDBOOK OF PSYCHOTHERAPY
AND BEHAVIOR CHANGE (Bergin & Garfield, 4th Ed., 1994, Wiley).
While they conduct and have widely contributed to RCT research studies,
psychotherapy researchers of this ilk are also drawn to methodologies
that allow the study of variables that cannot be randomly assigned,
such as therapist relationship factors, patient and therapist predisposing
variables, personalities, etc. Some of Ken Howard's naturalistic research
as well as his articulated concerns with the problems of RCT methodlogies
are widely and highly valued among this group of scholars, probably
much more so than they are among those committed to a theory-specific,
and more particularly, a behavior analysis tradition.
Scholarly reivews of research in this area by those within the broadly
based psychotherapy research perspective have long held that the psychotherapy
relationship, patient variables, and other non-technical factors,
are more highly related to outcome than the specific procedures and
theoretical model employed by the therapist. Lambert & Bergin
(1978) expressed this view years ago, and it has not substantially
changed (see Lambert, 1989; 1994; Lambert & Bergin, 1994; Lambert,
et al, 1986). Lambert & Assay (1992) apportioned the percentage
of variance attributed to various therapist, client, and treatment
factors and reached the conclusion that client variables are most
strongly associated with outcome, followed by the quality of the therapeutic
relationship (therapeutic alliance, facilitative relationship, etc.).
The use of specific procedures and models (e.g., CT, IPT, etc.) accounte
for less than 10% of the total outcome in their analysis. This proportion
was similar to that reported in meta-analyses by (Smith, Glass, &
Miller, 1981; Shapiro & Shapiro, 1984). The low percentage of
variance contributed by differences in the model and procedure used
has, as you know, been the basis for the Do-Do Bird verdict---all
psychotherapies obtain essentially equivalent results, a finding that
keeps being repeated in the preponderance of meta-analytic reviews
(see Wampold, et al, 1997). Thus, it is fair to conclude that the
correlations between qualities of the therapeutic relationship and
outcome are almost always higher than the relationship between type
of therapy applied and outcome---i.e., the relationship accounts for
more of the variance than what the therapist does from a theory of
change perspective.
Now the evidence for a causal relationship between relationship and
outcome is a little less clear as Jeff has pointed out. Before it
can be addressed adequately, Bill's concerns with the measurement
of therapeutic alliance must be addressed. He observed that some items
in relationship questionnaires often ask the rater (patient or therapist
or clinician) if the client's goals are being achieved or if they
are pleased with the way things are going. There has been much written
about this apparent confound, but four points can be made: First,
there are many measures of the therapeutic alliance, by various names,
many of which do not include this confound (see Greenberg & Pinsof,
1986) but results seem quite independent of the specific measure used.
Second, these measures are highly intercorrelated, frequently to the
point of approximating their own reliability indices (Tichenor &
Hill, 1987; Salvio, et al, 1992). Third (and there are several studies
of this, but I can't bring the specific references to mind, so I'll
use our own data), it makes little difference whether the relationship
items that seem to reflect outcomes are excluded from the scales or
not. Recently we submitted a study of the sequence among three classes
of predictors--patient variables, therapy procedures, and matching
variables--on two outcomes (therapeutic alliance and symptomatic change),
and subsequently, the causal chain when relationship was inserted
as one of the predictors . We were critiqued because our measure of
alliance (an independent observer rating on the HAQ) included five
(5) items that were confounded with outcome. We eliminated these items
and correlated the scales using part-whole correlations, obtaining
correlations above .80, not bad for a scale that has interrater reliabilities
that are much lower than this, often. We are now looking to see if
the use of the non-contaminated scale changes the pattern of relationships
obtained, but I doubt it.
Finally, when measured as part of the therapeutic relationship, questions
about outcome are in the nature of asking the respondent to reflect
back and assess the degree of their progress. This single point in
time (retrospective) measure of "outcome" would never be accepted
in today's climate as a measure of improvement. In fact, one of the
major critiques of the CR study was that the authors measured outcomes
only at the end of treatment. The problem is that when measured in
this way, there is only a very low correlation of these ratings and
pre-post measures (Beutler, Wakefield, & Williams, 1994; Beutler
& Hamblin, 1986). To compensate, the CR authors developed a retrospective
pre-post measure (a procedure that our group had recommended in 1981).
I've recently seen a study, the reference to which I can't pull up,
that suggested that even these scores did not correlate well with
actual pre-post measurement. In fact, these retrospective measures
only correlate well with current state, leading to the conclusion
that they are measures of "satisfaction" not measures of improvement.
Thus, it is highly unlikely that the use of such items seriously contaminates
pre-post measures of outcome.
Assuming that I have dispatched the issue of contamination raised
by Bill Follette for the moment, let us look at the evidence for a
causal chain between patient variables, through treatment procedures
and relationships, to outcome. That evidence is relatively sparse,
but two lines do suggest a causal chain that links relationship quality
to outcome. First is a plethora of studies that have used lagged correlations
between early developed relationships and outcomes. In Vanderbilt
I, Strupp observed that the therapeutic relationship as measured before
the 10th session predicted distal outcomes (This was reported in an
SPR presentation---I don't know the published reference) and this
was consistent with the early findings of Luborsky in the Penn Psychotherapy
Project, in which ratings at session #3 were quite higly predictive
of distal outcomes, sometimes 2 years later (Luborsky et al, 1991).
(Both Les Greenberg and Adam Horvath have confirmed these findings,
I believe, though I don't have a specific reference). In the study
that I mentioned in an earlier paragraph, we also looked at early
alliance (first five sessions) and distal outcome, obtaining a value
of .36 after the influence of initial patient variables and therapy
procedures were factored out through MR procedures.
In a Structural Equation Modeling procedure that compared a simple
relationship --> outcome versus relationship + patient + matching
variables --> outcome model, we found that both models were consistent
with the data, but that process variables added variation that was
independent of the therapy procedures used (Beutler, Clarkin, &
Bongar, 2000). As you know, SEM doesn't care about the temporal sequence
of variables. That has to be built into the model. So, SEM could easily
confirm that a given data set is consistent with the view that distal
outcome produced an earlier measured relationship event, if the investigator
was foolish enough to propose this model to test.
As a final point, consider the repeated observation that different
therapies often are distinguished by a particular pattern of developing
the therapeutic relationship (Rounsaville, et al, 1987). I will send,
to those who are interested, a series of graphs from a current comparison
of three types of psychotherapy in the course of the development of
the therapeutic alliance. These graphs are from an RCT study of CT
and two new therapies, one focused explicitly on trying to develop
a therapeutic alliance that is devoid of patient resistance and the
other that matches specific treatments to specific non-diagnostic
qualities of the patient. To summarize, the growth of the relationship
differs quite distinctively. This morning, I got back a components
analysis of these differences. I don't have it all digested yet, but
it is clear that a given HAQ score in the three different therapies
is reflecting a very different experience for each, and that the pattern
of components over time is very different from therapy to therapy.
While the overall correlation between early alliance and distal outcome,
and even follow-up, remains in the .30-.40 range, therapies are distinguished
by what items load in the ratings of alliance and the pattern of development
is very different from therapy to therapy.
If all a good relationship did was improve the therapist's reinforcing
power, I would think that the NT (relationship-based) therapy would
have been the best on symptom measures, since the alliance was best
in this treatment, but this was not the case. CT and PT produced better
effects on several indices in this sample of co-morbid stimulant dependent
and depressed patients.
References
Bergin, A.E., & Lambert, M.J. (l978). The evaluation of psychotherapeutic
outcomes. In S.L. Garfield & A.E. Bergin (Eds.), Handbook of psychotherapy
and behavior change: An empirical analysis (pp. 139-190). New York:
John Wiley.
Beutler, L. E., Clarkin, J. F., & Bongar, B. (2000). Guidelines
for the systematic treatment of the depressed patient. New York: Oxford
University Press.
Beutler, L.E., & Hamblin, D.L. (1986). Individual outcome measures
of internal change: Methodological considerations. Journal of Consulting
and Clinical Psychology, 54, 48-53. (special edition)
Greenberg, L.S., & Pinsof, W.M. (l986). The psychotherapeutic
process: A research handbook. New York: Guilford.
Lambert, M. J. (1989). The individual therapist's contribution to
psychotherapy process and outcome. Clinical Psychology Review, 9,
469-485.
Lambert, M. J. (1994). Use of psychological tests for outcome assessment.
In. M. E. Maruish (Ed.) The use of psychological testing for treatment
planning and outcome assessment. Hillsdale, NJ: Lawrence Erlbaum Associates,
Publishers.
Lambert, M. J., & Bergin, A. E. (1994). The effectiveness of
psychotherapy. In A. E. Bergin & S. L. Garfield (Eds.), Handbook
of psychotherapy and behavior change (4th ed., pp. 143ˆ189).
New York: Wiley.
Lambert, M. J., Shapiro, D. A., & Bergin, A. E. (1986). The
effectiveness of psychotherapy. In S. L. Garfield & A. E. Bergin
(Eds.) Handbook of psychotherapy and behavior change, 3rd ed. (pp.
157-211). New York: John Wiley and Sons.
Rounsaville, B. J., Chevron, E. S., Prusoff, B. A., Elkin, I., Imber,
S., Sotsky, S., & Watkins, J. (1987). The relation between specific
and general dimensions of the psychotherapy process in interpersonal
psychotherapy of depression. Journal of Consulting and Clinical Psychology,
55, 379-384.
Salvio, M., Beutler, L. E., Engle, D., & Wood, J. M. (1992).
The strength of therapeutic alliance in three treatments for depression.
Psychotherapy Research, 2, 31-36.
Shapiro, D.A., & Shapiro, D. (l982). Meta-analysis of comparative
therapy outcome studies: A replication and refinement. Psychological
Bulletin, 92, 58l-604.
Smith, M. L. Glass, G. V., & Miller, T. I. (1980). The benefits
of psychotherapy. Baltimore: The Johns Hopkins University Press.
Tichenor, V., & Hill, C. E. (1989). A comparison of six measures
of working alliance. Psychotherapy, 26, 195-199.
Wampold, B. E., Mondin, G. W., Moody, M., Stich, F., Benson, K.,
& Ahn, H. (1997). A meta-analysis of outcome studies comparing
bona fide psychotherapies: Empirically, "All must have prizes." Psychological
Bulletin, 122, 203-215.
Tullio Carere, 12 June 2000
I realize that a taxonomic approach (a map of the common psychotherapy
factors organized in a comprehensive whole) has not much sex appeal,
but could at least the Do-Do Bird verdict (thank you George for reminding
us of it) be a firm point of departure? If we assume, according to
this virdict, a large supremacy of relationship factors over technique
and model factors as predictive of change, we are faced with a most
basic question: What makes a therapeutic relation out of an ordinary,
every-day relation?
One first possible answer would be the following: Therapeutic factors
are embedded in ordinary, every-day relations, and a therapeutic relation
is one led by a therapist who (a) knows which are these basic factors,
and (b) knows how to activate and employ them moment by moment in
the relation with her client. One second possible answer would be
instead: As what happens in ordinary relations is that everybody tries
to impose his theory or world view onto the other (and so called therapists
are no different in this respect), real therapy only begins when one
suspends his theoretical and personal preconceptions and persuades
the other to do the same--thanks to this suspension a space opens
where negotiations, new experience and genuine search can happen.
The two answers are not, to me, mutually ewclusive (but the second
comes first, and breaks the ground to the first). Any other responses
to the Do-Do Bird challenge?
Alan Javel, 13 June 2000
Differences are indeed easier to see, probably because they are the
most glaring. To understand the similarities, one would need to look
very closely at technique. In my opinion, what we do, how we do it,
and when we do it are the basic building blocks of any relationship,
therapeutic ones included. I am always disturbed by the idea that
my theoretical orientation, whatever it may be, may merely be window
dressing, a small part of the actual business of the therapy which
has been defined by my office environment, my verbalizations, and
my silences.
Hilde Rapp, 13 June 2000
Alas, the dodo bird was rendered down to illuminate the dark nights
of explorers of uncharted territories...
If it weren't for the differences in viewpoint between our two eyes,
our picture of the world would be rather one dimensional.
Similarly, to my way of working, much of the power of psychotherapy
derives from those aspects of our practice which are different from
the ways in which odinary relating takes place. In my personal experience,
meta-conversations are singularly unpopular among my loved ones, yet
they are the deadlock breakers in my therapeutic work.
I have no trouble with enthuisiastic clebrations of 'vive la differe/ance'
– bowing to Derrida slightly –but I suppose as a Sepi-ite
I see my task more as endeavouring to understand differences, and
to understand whether they are merely interesting, trivial or therapeutically
significant – in the way of Clarkin and Frances' differential
therapeutics. We are way off being able to map systematic differential
applications of therapeutic techniques to distinct clinical populations
with measurable different problems in ways which yield demonstrable
solutions. Even if none of these ways of describing our art,
science and craft chimes well with particular value base which puts
the clients meaning making in the centre, there still is a burden
of proof on us within Sepi that we do more than provide human
understnding and the equivalent of pastoral care.
This is were I like the postmodern stance: a modest approach to mapping
some differences which seem to us clinically important, intellectually
non trivial, and at least describable if not measurable in some coherent
and communicable way: to start with the particular, the human, the
subtle small gestures of human relating – and where we feel bold
enough to look for the larger social paterns that connect...
Sepi is to my mind about generating interesting hypotheses about
change factors and other things, to find really interesting ways of
talking about the complexities of the human condition, and to bandy
together with our colleages in SPR to devise intelligent ways of finding
out what we do well that helps clients to put their life back together
and to get on better with others...
How about the phoenix verdict?
Michael Basseches, 13 June 2000
I must have missed or forgotten the original do-do bird challenge--Could
you, George, or someone, remind me.
But to answer to the question Tullio poses would be to agree that
"therapeutic factors are embedded in ordinary, every-day relations".
I would add though that psychologically destructive factors are also
embedded in ordinary every-day relations. Clients often come to us
with a hope to remedy some of the damage that their relationships
have done to them, or ways in which their development has been constrained
by those relationships, but with a repertoire (of action/meaning-making
schemes) that may provoke others to repeat the damage that has been
done. Therefore, I view no goal as more important in therapist training
than learning how to create relationships which minimize the destructive
factors, as well as to optimize the therapeutic factors in "ordinary
every-day relations". You refer to the latter when you say--"a
therapeutic relation is one led by a therapist who (a) knows which
are these basic factors, and (b) knows how to activate and employ
them moment by moment in the relation with her client," but as
you know from what I have written, I think the former dimension--effort
and skill at avoiding harm or oppression (Leston Havens calls it "predation")
is equally important.
Your second answer -- suspending theoretical and personal preconceptions
-- may be one way of attempting to avoid harming clients within therapy
relationships, but I have trouble endorsing that one because I think
(as you do, I believe) that it is impossible to do so completely,
and also, because there are many other skills required in this effort
as well. Thank you for pushing/inviting us to keep addressing such
crucial questions in the service of psychotherapy integration.
George Stricker, 14 June 2000
Dear Mike,
I'll just respond to your question to me, and reserve any other comments
for another time. In the 1970s Luborsky wrote a summary of psychotherapy
research, found overwhelming evidence for effectiveness, but no evidence
for differential effectiveness among orientations. His article was
subtitled "All have won and all must have prizes," from Alice in Wonderland,
hence the term Do-do bird effect. The Do-do bird challenge is to find
differential effectiveness and, by doing so, get beyond common factors
as the basis of all therapeutic change. That is exactly what is being
attempted by people such as Beutler, who are looking for therapist-patient
matching keys to how to work most effectively, but I am not impressed
by the depth of their findings (even though I keep hoping that there
will be something there).
Michael Basseches, 14 June 2000
Thanks, George. So would the following qualify as a response to the
challenge or is it a side-stepping of the challenge?
While common factors may be one set of significant contributors to
therapeutic change, and aspects of therapist-patient match may be
another set of contributors, differential effectiveness may be found
in variations in therapists' intellectual and interpersonal abilities
(to which effective therapist training can hopefully contribute) to
understand and work through a less than optimal match and to transform
it into a successful therapeutic relationship.
I believe that this set of of perspective/ability/training/experience
factors represents a third very important set of contributors to therapeutic
effectiveness (to which SEPI has, and should continue to attend).
I think both Tullio and my posts were efforts to address this issue.
Does it indeed address the Do-do challenge.?
David A Shapiro, 14 June 2000
Actually it was Rosensweig in the 1930's who first applied the Dodo
metaphor to psychotherapy research. Subsequently, Bill Stiles, Robert
Elliott and I took up the theme in our 1986 American Psychologist
article, "Are all psychotherapies equivalent?".
George Stricker, 16 June 2000
To Mike Basseches:
I don't think that the set of factors you mention (and I agree with
your sense of their importance) speaks to the Dodo bird issue, because
those factors do not differentiate the orientations and the final
verdict, all must have prizes, still holds. However, whether or not
you read the Dodo as I do, you still are presenting a hypothesis,
and it would take empirical data to address the Dodo and provide contrary
evidence.
To David Shapiro:
I wasn't aware of Rosenzweig's use of the term, but the Luborsky
review was in the early 70s and preceded your article. When it comes
to the provenance of a term like Dodo bird, I don't think we can be
sure of anyone other than Lewis Carroll.
Tullio Carere, 17 June 2000
George Stricker wrote: "The Do-do bird challenge is to find
differential effectiveness and, by doing so, get beyond common factors
as the basis of all therapeutic change."
This is surely a way of reading the Do-do issue. Another way, in
my view no less legitimate, is not to get beyond common factors in
search of differential effectiveness, but on the contrary to stay
within the common factors, and look there for the basis of all therapeutic
change. I dare say that the latter reading is even more legitimate,
because if "All have won and all must have prizes", the outcome cannot
depend on the method-specific factors in the first place, but it must
relate primarily to those factors that are common across all orientations.
The standpoint, then, radically shifts. The focus is no more on a
theory of the mind, as in most psychotherapeutic traditions, but on
the therapeutic relation, and specifically on the question: what is
therapeutic in a therapeutic relation? (and its twin question: what
is pathogenic in a pathogenic relation?). Do we need empirical research
to know that an abusive relation is pathogenic? I don't think so.
By the same token, we should not need empirical research to know that
theoretical abuse is what turns a would-be therapeutic relation into
a pathogenic one.
But could a therapist who clings to his theory or manual not be abusive?
I don't see how he could. This means that manual- or theory-driven
treatments are fine for research aims, but they are no real therapy.
On the contrary, they are surely abusive in the long run. It follows
that the first and foremost common factor to any therapy is the freedom
from theory, which does not mean to have no theories--it means not
to be conditioned by them, that is to be able to suspend them, to
remain in a (relatively) theory-free space. Which corresponds, besides,
to the time-honored first duty of all therapists: primum non nocere.
Michael Basseches wrote: "Your second answer -- suspending theoretical
and personal preconceptions -- may be one way of attempting to avoid
harming clients within therapy relationships, but I have trouble endorsing
that one because I think (as you do, I believe) that it is impossible
to do so completely, and also, because there are many other skills
required in this effort as well."
The suspension of all theoretical and personal preconceptions cannot
be anywhere near perfection, but it must be "good enough" to allow
at least for the beginning of a negotiation--how could a negotiation
ever begin if one refuses to yield his/her truths? (And you are the
first among us to alert to theoretical abuse). We may then be even
more ambitious, if we don't content ourselves with negotiations but
we want to know what one person really needs (beyond all theories,
techniques, and constructions).
If the suspension of all theoretical and personal preconceptions
- the phenomenological epoché - is the basic common factor
of all therapies worth the name, it is also the base from which to
start, and to which always return, to investigate on all other common
factors. Failing this, we would only have theories and would be prisoners
of them (which has in fact happened to the psychotherapy schools--couldn't
SEPI be a remedy?).
Hilde Rapp wrote: "How about the phoenix verdict?"
That's another bird. As you know, the phoenix lives very long, and
goes into fire only at the end of its long life. Psychotherapy is
still very young. There will come a time for the phoenix, but it seems
the Do-do is our bird now.
George Stricker, 17 June 2000
I think Tullio and I were saying something very similar, but from
different perspectives. I spoke about the Dodo challenge (to find
differential effects) and he about the Dodo effect (the importance
of common factors). Two sides of the same coin, as I see it. I also
agree on the problems of manualizing that which should be more responsive
to idiographic occurrences. However, there are two points on which
we disagree, I think. The first is the value of empirical data. There
are many things which are obvious (e.g., the sun moves around the
Earth), but do not stand the test of investigation. The second is
Tullio's statement "It follows that the first and foremost common
factor to any therapy is the freedom from theory, which does not mean
to have no theories--it means not to be conditioned by them, that
is to be able to suspend them, to remain in a (relatively) theory-free
space." I think he is speaking about a wish rather than an empirically
verified finding, especially as it is the case that many therapists
are theory-driven, to the detriment of their work, but that means
that being theory-free clearly is not a common factor. I think we
also have to be aware of the distinction between the role of theory
in understanding what is happening (critical, in my view) and its
role in determining our interventions (secondary, again in my view,
and I think in Tullio's as well).
Tullio Carere, 17 June 2000
George Stricker wrote: "I think Tullio and I were saying something
very similar, but from different perspectives. I spoke about the Dodo
challenge (to find differential effects) and he about the Dodo effect
(the importance of common factors). Two sides of the same coin, as
I see it."
I totally endorse this view. On the one side of the coin I put theoretical
integration (the assimilative avenue), on the other side the common
factors approach (the accommodative avenue).
George: "I also agree on the problems of manualizing that which
should be more responsive to idiographic occurrences. However, there
are two points on which we disagree, I think. The first is the value
of empirical data. There are many things which are obvious (e.g.,
the sun moves around the Earth), but do not stand the test of investigation."
To me this is another coin, or better said the same coin at another
level. On the one side there is theoretical-empirical science, based
on hypothesis testing; on the other side there is descriptive-phenomenological
science, aimed at the uncovering of evidence. If this dialectic is
lost, science is lost too.
George: "The second is Tullio's statement "It follows that the
first and foremost common factor to any therapy is the freedom from
theory, which does not mean to have no theories--it means not to be
conditioned by them, that is to be able to suspend them, to remain
in a (relatively) theory-free space.": I think he is speaking about
a wish rather than an empirically verified finding, especially as
it is the case that many therapists are theory-driven, to the detriment
of their work, but that means that being theory-free clearly is not
a common factor."
This is right what I meant: when therapists are theory-driven, it
is to the detriment of their work. Saying that "the first and foremost
common factor to any therapy is the freedom from theory", I meant
any genuine, non theoretically abusive therapy.
George: "I think we also have to be aware of the distinction
between the role of theory in understanding what is happening (critical,
in my view) and its role in determining our interventions (secondary,
again in my view, and I think in Tullio's as well)."
My view here is slightly different. Instead of seeing the role of
theory as critical in understanding and secondary in determining our
interventions, I prefer to balance the theoretical-interpretive moment
and the phenomenological-intuitive one in both.
George Stricker, 17 June 2000
It seems as though Tullio and I have reached agreement, despite the
apparent differences in the original presentations. There is one point
he made well that I would like to reiterate: "To me this is another
coin, or better said the same coin at another level. On the one side
there is theoretical-empirical science, based on hypothesis testing;
on the other side there is descriptive-phenomenological science, aimed
at the uncovering of evidence. If this dialectic is lost, science
is lost too." I agree totally, although I, with Reichenbach,
prefer to refer to this as hypothesis generation and hypothesis testing,
and the dialectic nature of these processes, which Tullio refers to,
is crucial, and there is no need to choose between them.
Zoltan Gross, 17 June 2000
May was an extraordinary month for me. In the beginning of May, I
attended and presented at the SEPI Meetings in Washington, D.C. and
at the end of the month I went to the Conference on the Evolution
of Psychotherapy (CEP)in Anaheim, CA. At both meetings, the presenters
talked about psychotherapy in different languages. They had different
conceptions about what the nature of help was or should be. Underlying
these differences were different assumptions about the task of psychotherapy.
The presenters described what they thought psychotherapy "looked"
like and it looked different to all of them.
This is not an unusual observation. Minuchin at the CEP meetings
commented on the fact that the diversity of explanations about the
nature of psychotherapy was both enormous and intelligent. He went
on to say that all of the presenters believed that they were "effective"
therapists, including himself. He also noted that it is remarkable
that so many dedicated and talented men and women have examined the
two person therapeutic relationship have come up with so many different
explanations about its nature. Furthermore, he said that regardless
of their explanatory differences, they were all helpful (the Do Do
Bird). I agreed. I saw it with my own eyes. The interviewees invariably
left their fifteen minute sessions expressing gratitude for the help
they received.
The bedlam of voices caused me to think they were talking about different
interpersonal enterprises. This observation puzzled me. If they were
different kind of relationships, why, then, did all of the interviewees
express the same appreciation for the different kinds of help they
received? The clatter of explanations had such a dizzying effect on
me that I turned the sound off. The quieting peace that followed allowed
me to look at what these different therapists were doing with their
client/patients, without being distracted by my confusion about their
different explanatory systems. I was struck by how much alike they
looked.
Of course, the therapists were clearly different in their personal
presentations. Albert Ellis' aggressive assertiveness was different
from Eugene Gendlin's gentle empathy or Leslie Greenberg's "Santa
Claus" geniality. Yet they were all authoritative and had the presence
and wisdom that comes with years of practice and teaching. They were
all respectful and comfortable. They were clearly interested in being
helpful. At the beginning of each demonstration, the presenters gently
took care to put their clients at ease. All of this was done without
asking anything in the way of validational feedback from the interviewee.
At the same time all of the therapists were powerfully present with
their interviewees. They made eye contact and continually sending
nonverbal signals of approval and understanding to the client/patient.
They were all quiet, warm, nonjudgemental and intuitively empathic,
wanting to have a clear understanding of how the interviewees experienced
their distress. They asked the interviewee to describe the problem
that brought them to the interview. The therapist asked penetrating
questions until he or she understood the problem the interviewee about
which he/she wanted help. These questions demonstrated the intelligent
interest of the therapist and conveyed to the client/patient that
she/he was in competent hands. The therapist carefully proceeded to
reframe their interviewee's problem in his/her own terms. The therapist's
restatement of the problem caused the interviewee to think about his/her
problem in a new way and provided him/her with a way of thinking about
the problem in the therapist's language. The client/patients were
given a new cognitive perspective on the nature of her/his distress.
Where else in life can a person get this kind of undemanding, knowledgeable,
good willed attention? I disagree with Tullio's contention that therapists
are engaged in an "ordinary everyday relationship." They were engaging
in a relationship which I believe is a Twentieth Century interpersonal
invention. It takes years of clinical practice to become comfortable
and skilled in this kind of relationship. To me the therapeutic relationship
is related to but not identical with Buber's description of the "I-Thou"
relationship. The similarity of approach by the therapists I saw is
partially responsible for the Do Do Bird effect. Other variables also
operated to produce the differential effects to which George Stricker
calls our attention.
Three kinds of information will help us escape from the banality
of the Do Do Bird's prizes. We need a better understanding of the
dynamics of dyadic interaction. None of our current explanations of
dyadic interaction enables us to describe the moment to moment music
of emotional experience that arises when two people are personally
involved with one another. Related to this, we also need to know much
more about the contribution that the therapist's person plays in psychotherapy.
I look forward to results John Norcross' commission looking into this
matter. While there is evidence that therapists differ in their effectiveness,
there is no research studying the role of the therapist's person in
the therapeutic process. That information would provide us with valuable
information about the nature of therapeutic effectiveness. Finally,
more clearly specifying the goals of therapy will help us to see the
differences in the therapies as well as their similarities. There
is a real difference between therapies that are devoted to character
change and those that seek to provide relief from the distress of
affective disorders.
Hilde Rapp, 19 June 2000
The 'phoenix verdict' is just an invitation to consider a paradigm
shift – of the sort that would allow us to consider the Do-do
challenge (to do, to attend to what we do reflectively, and not to
become extinct) and the Dodo verdict (as explained by George) in the
same frame. To suspend and conserve both perspectives in a meta-frame:
what is common and how is this non trivial – ie distinctive to
therapy rather than any broadly emancipatory activity, and what is
distinctive. Core competencies and specialist competencies: core competncies
which transcend orientations, and specialist competencies which are
theory, technique, modality or context driven.
As you know Tullio: the bird which comes out of the fire is the same
bird which went in – just with some of the dross burnt clean
in the purifying flame – and that which was good distilled and
fortified...
PS: Let me know if you want me to stop being poet and mystic and
start being craftswoman and scientist – the philosopher cannot
be dispensed with.
Tullio Carere, 21 June 2000
