> From: "Thompson, Fiona" <FIONA92@psy.soton.ac.uk>
> Date: Fri, 2 Jun 1995 16:25:05 GMT
>
> Can you clarify for me what the hermuneutic circle is? Is it the
> notion that psychotherapy is circular in its reasoning so that the
> psychotherapist may give a reason for someone's behaviour that has not
> been explained itself. Given this reasoning is it true therefore to
> assume that the reason psychotherapy is effective is because the
> client believes it is so and this is therefore why the parallels with
> the placebo effect have been drawn?
That's more or less right. Hermeneutics comes from scriptural
interpretation (and in the end, the interpretation of literature).
Hermes is Mercury, the messenger of the gods. When you interpret
a text, there's no real right or wrong of the matter. You either
accept the interpretation or not. If you do, it will confirm itself
in countless ways, like a self-fulfilling prophecy, but the confirmation
will only be of the fact that once you have accepted it, it does indeed
fit all the data. A horoscope always fits (no matter if it's your
horoscope or someone else's, as long as you believe it's yours).
Now the difference between hermeneutics of written texts and
hermeneutics of live discourse between a therapist and patient is that,
over and above the inclination of either of them to accept the
interpretative system in terms of which the discourse is interpreted
(say, Freudian psychodynamics, or some current "cognitive" scenario),
there is the element of suggestibility (on the part of both). That
makes the therepeutic/hermeneutic circle potentially a much more
dynamic and binding one than the one for inert texts.
That having been said, this course is on Debates in Psychology: The
critics of the critique that psychotherapy is just hermeneutics (with the
patient and therapist just interpreting things, and nothing really
getting better, or at least not because of the therapy) hold the contrary
view that:
(2) Psychotherapy is NOT just hermeneutics, but the testable (and
tested, and confirmed) application of a testable (and tested, and
confirmed) theory, not an interpretation. (Prioleau et al. would dispute
this.)
or
(3) Psychotherapy IS just hermeneutics, i.e., interpretation, but it is
the RIGHT interpretation, as attested to by the fact that it helps!
(Grunbaum would dispute this.)
In your essay you need to sort out the evidence and arguments for these
three kinds of positions (and perhaps others, as you read the panaorama
of views in Prioleau et al and Grunbaum and their commentaries and
repsonses). Classify the commentaries according to their points of view
and what supports them. Use that as a framework for your survey of the
different sides of the debates.
> Also whilst on this topic I am unsure how the peer review research
> fits in and relates to psychotherapy. Is it to do with the notion as
> with psychotherapy frequently scientists do not look at things
> objectively and are influenced by what is considered by culture,
> society to be the right thing to believe, ie therapy is trendy.
That's right. And there's also the general problem of human judgments
when they are not based on objective measures, but on human judgments
(peer review). The topics are even more closely linked, because it is
peer review that is meant to evaluate whether a theory or therapy has
been successfully tested and supported or refuted. (Is our faith in peer
review just a hermeneutic one?)
> When I was reading the Peters and Ceci article, not being a great
> mathematician I was unclear on what they called the regression
> effect. They gave this as one of their reasons why the second time
> round only one out of the nine journals was accepted.
They were referring to "regression on the mean." All that means is that,
the way averages work, they always have to average out. So about as many
times (and to as great an extent) as you overshoot the true mean, you
will also undershoot it, and keep getting closer to it, as you take
more and more measurements. An average of averages will always be
closer to the overall average (obviously) than any particular average.
So let's suppose that there is an average rating that any given paper
will get from referees. Sometimes it will be higher, sometimes lower,
but it will average out on the average. Then if a particular paper
happened to be accepted by the first set of referees, but in reality was
borderline, as many papers are, with big variation in referees'
judgments, the next time it might be rejected. But since P & C were
were only looking at previously ACCEPTED papers, you never got to see
the other side of the regression coin: previously rejected papers might
well get accepted the next time round (which is why papers usually just
get submitted to another journal once one rejects them, so they almost
all eventually appear somewhere in some form or other).
The standard example of regression is that the taller the parents, the
less likely that their children will be as tall as or taller than them:
This is not just a ceiling effect (so to speak), but regression on the
mean: The further the parents are from the average, the more likely that
their children will be closer to the average (and not just to the average
of the heights of the two parents, but to the overall population average).
Same for children of very short parents, who tend to be a bit taller
than their parents.
Nothing magic about regression; just everything averaging out in the long
run.
Chrs, Stevan
mailto:harnad@soton.ac.uk
This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:17 GMT