Communication Institute for Online Scholarship
Communication Institute for Online
Scholarship Continous online service and innovation
since 1986
Site index
 
ComAbstracts Visual Communication Concept Explorer Tables of Contents Electronic Journal of Communication ComVista

What the Brits Got Out of the Q: And Why Their Work May Not Line Up With the American Way of Getting Into It!
EJC logo
The Electronic Journal of Communication / La Revue Electronique de Communication
**************************** EJC/REC Vol. 1, No. 1, 1990 ***

WHAT THE BRITS GOT OUT OF THE Q:  AND WHY THEIR WORK
MAY NOT LINE UP WITH THE AMERICAN WAY OF GETTING INTO IT!


Rex Stainton Rogers
University of Reading

Wendy Stainton Rogers
The Open University


        Abstract.  A view of Q methodology is proposed
     which is based on social constructionism and which
     regards Q factors as social representations,
     understandings, and accounts rather than
     self-referential attitudes or subjective
     viewpoints.  Questions are also raised concerning
     the way in which factors are typically extracted,
     the number extracted, and the way in which they
     are rotated by Q methodologists.  The principles
     at issue are illustrated in terms of a
     deconstruction of the concept of addiction based
     on a Q sort containing statements of
     understanding, and another containing policy
     statements.  The interlinkages among the resulting
     factors point to conjunctions of voice.

        CE QUE LES BRITANNIQUES ONT FAIT DE LA METHODE
     Q ET POURQUOI LEURS TRAVAUX NE CORRESPONDENT PAS
     AVEC LA FACON AMERICAINE.  En partant de la
     theorie de la construction sociale de la realite,
     une vision de la Methode Q considere les facteurs
     Q comme des representations sociales, des
     comprehensions et des motifs plutot que des
     attitudes auto-referentielles ou des points de vue
     subjectifs.  La facon par laquelle les facteurs Q
     sont generalement produits est questionnee ainsi
     que le nombre de ces facteurs et la facon dont les
     rotations sont realises.  Ces principes sont
     illustres par la deconstruction du concept
     d'intoxication sur la base de groupes de type Q
     comportant d'une part, des enonces de
     comprehension et d'autre part, des enonces de
     politique.  Les liens entre les facteurs qui en
     resultent indiquent des convergences de discours.


     Anybody who uses Q should not be surprised to hear that
there are several stories to be told about the recent
history of the use of Q in Britain.  This paper takes
advantage of this expectation to explore a muted voice, that
is, to argue against the view that the "doing of Q" has (or
can bring into being) a community of Q methodologists who
share a bedrock of agreement about what they are trying to
achieve, and what assumptions underpin the endeavor.  Under
our social constructionist heresy, to use Q is to employ an
effective pattern analytic for explicating diversities of
socio-cultural representations, understandings and policies.
Such accounts and voices are held to owe nothing to the Q
methodological axiom of "self-reference" which we regard as
problematized and compromised by post-structuralist and
post-modern theory.
     We use as an illustration of our radical social
constructionist use of Q a study which we carried out to
deconstruct the notion of "addiction."  Participants in the
study completed linked Q sets, one covering understandings
and one social policies.  In addition, we asked them to
provide open ended comments on the propositions which were
used to help explicate the sorting patterns.  In this way we
were able to explore not only the diversity of accounts of
addiction current in Britain at this time, but also to
consider the conjunctions and disjunctions between
understanding-stories and policy-stories.  We argue that our
approach can offer observations which are of theoretical and
empirical interest, whatever the ideological investment of
the reader.

            The "British Dialect" of Q in Theory

Reflections on Communicational Angst

     One of the differences between European and North
American culture is said to be that here, as we say, "titles
matter."  But only, presumably, when people understand what
they mean, and what they signify!  The reasons for the title
we have used to start this paper should become clearer over
the next few pages.  For now, we hope it will serve to
sign-post our concern about a language of "owning" or "being
owned," the kind of idea explored in the semantic praxis of
Ursula Le Guin's (1974) thought-experimental anarchy of
Anares.
     In consequence, in this paper we write as, and of,
British users of Q methodology, not as, or of, British Q
methodologists.  The drawing of this distinction is
important both to the story we wish to tell of research here
in the UK, and to the establishing of a sensitivity to text,
both of which are very much part of the "deconstructive" way
we work.  To worry the "taken-for-granted" (cf. Schutz,
1972) is perturbating: deconstruction is perturbation as
method.

Deconstructing Revivalism

     Shortly before he died, Will Stephenson travelled from
the USA, his adopted country of more than 40 years, to be
guest speaker at a "Q Workshop" in Reading, England (Brown,
1989).  As co-organizers of the conference we saw its
purpose as one of relating of the practice of the Q
methodology that William Stephenson had originated in
Britain in 1935 to a range of recent developments in the
socio-cultural disciplines.  These included
post-structuralism (cf. Parker, 1989); postmodernism (cf.
Featherstone, 1988) and social constructionism (cf. Shotter
& Gergen, 1989).  This rather complex agenda is a signal to
the possibility that what could be read as a renaissance of
interest in Q methodology in the UK (cf. Brown 1989, 1990)
is both less and more than such an analysis might imply.
     As part of a "William Stephenson Story" there is, of
course, a thematic good gestalt to such "...a savory
homecoming...," particularly one amounting to "...this first
sign of vindication in the country of his birth..." (Brown
1990).  In such a narrative, we would indeed expect him to
be "...especially surprised and pleased to discover the many
British theses and dissertations which had been produced in
the six or eight short years since his ideas had been
rediscovered" (Brown, 1990).  However, as a cultural
descriptor, a renascence can be read as a journey into the
future as well as a recapitulation of things past.
     So even within this story itself, there is another,
contrasting theme, hinted at in the description of Will's
"pointing out to each participant how he or she had
misunderstood entirely" (Brown, 1990).  We could make that
sound very reminiscent of Einstein's railing that "God does
not play dice with the Universe" -- once Schroedinger's Cat
was out of the bag he himself had enabled!
     The title above, of course, also signals an alert about
meaning and contrasted cultural locations.  The story we are
telling is intended to challenge the taken-for-granted,
"default option" assumption that the Q methodological work
happening around Reading, itself far from a single entity
(cf. Stainton Rogers & Stainton Rogers, 1989a), should be
seen as a new branch of the "family-back-home," carrying on
the traditions of their forebears, their work merely an
extension of what went before, and still very much a part of
its overall endeavor.
     Again, there is no denying that such a reading can
produce a gratifyingly ecumenical text.  It is one that was
clearly in the "minds" of those at the annual Q conference
1989 (the first after William Stephenson's death) who saw
all those who use Q as a "...community of scholars...bound
together by common values and common understandings"
(Crumley, 1990).  Subsequently this voice was to be
reflected in the formation of the "International Society for
the Scientific Study of Subjectivity."
     However, it is possible (and however uncomfortable, we
believe, necessary) to narrate alternative realities.  We
would say that sharing in common the use of Q methodology
does not imply that we are all, essentially, doing the same
thing, albeit, in slightly different ways.  We would argue
that there is no more a core essence to the doing of Q
methodology (once we exclude the trivial delineation of
operational rules) than there is to the doing of, say,
sonnets.  Neither do we regard what we are doing here in
Britain as a "development" of what happened in the USA.
Rather, to use our own kind of rhetoric, we see ourselves as
having appropriated Q for our own devices.  Indeed we would
portray ourselves more as upstarts who have stolen the
family silver to fund their own devious plans, than fine
scions in the family tradition, ready and eager to start up
the UK chapter of R Methodologists Anonymous.

Another Version of the Travellers' Tale

     We suggest that through multiple story telling and
story listening, it is possible to experience the "Rashomon
effect" (cf. Austin Locke, 1990) -- to recognize that each
one of the stories that can be told brings into being a
reality -- a plausibility structure (cf. Berger & Luckmann,
1967) which can acquire pressing credibility via its
telling.
     The notions that understandings are person-made
creations, and that one understanding can only be said to be
truer (or whatever) than another by the application of a
further human operation (such as judgement or critical
analysis), is part of the social constructionist critique of
the positivistic essentialism of so much social science (cf.
Giddens, 1979; Shotter, 1984).  It is also one reading of
the idea of political discursivity as operant subjectivity,
that  Brown wrote about in his 1980 book POLITICAL
SUBJECTIVITY:

        Behavior of this kind is both subjective and
     operant.  It is subjective since each person's
     viewpoint on political or other matters is simply
     that -- his (sic) viewpoint.  It is operant
     because it exists naturally within a particular
     setting. (p. 4)

     However, in an actuarial sense, there are always fewer
viewpoints than persons.  Far from being unique to the
individual, these "viewpoints" bunch together into clusters
of high similarity; they covary together into a limited
number of factors in simple structure.  Further, according
to Brown,  "These factors can be considered as attitudes
operantly defined..." (p. 23).  Hence, Q methodology enables
a best estimate or model of these shared "attitudes."
     Small wonder, then, that the methodology attracted us
as social constructionists, a stance that has had its
theoretical proponents aplenty, but has lacked any uniquely
appropriate method.  We had been, for some time, looking for
a way of explicating -- locating, identifying and describing
-- the multiplex of accounts, understandings,
representations and policy positions we assumed were there
to be found in effectively any aspect of collective life one
cared to focus upon.  In Q we discovered such a means for
exploring a whole plethora of images, ideas, debates and
explanations -- from lesbian identities to images of child
sexual abusers; from understandings of why we become ill to
ways of making sense of post-bereavement experiences.  (For
a sample of British Q studies, see Gleeson, 1987; Kitzinger
& Stainton Rogers, 1985; Stainton Rogers & Kitzinger, 1986;
Stainton Rogers & Stainton Rogers, 1989b; Stainton Rogers,
in press.)

Disolving the Subject

     In the research we have carried out, we have put an
emphasis upon the cultural bases to the products of Q
analysis, following in this way Brown when he says:

        ... we are not generally interested in what
     particular persons rendered a viewpoint -- the
     same factors could probably have been gotten from
     other subjects -- but in the ways in which the
     different viewpoints themselves differ. (p. 238)

     Where, however, we would take issue is with the use of
"viewpoints."  We would prefer to stick to terms such as
"stories" or "accounts," because we are troubled by (and
wish to trouble) the reality that the term "viewpoint"
brings into being.  A viewpoint requires a viewer, and it is
clear that Brown, at this stage in his argument, holds that
a completed Q sort reflects:

        ... the impact of a mind in operation, of a
     person thinking, evaluating, and interpreting.  As
     such the self is always and intimately involved,
     for they are always a person's own thoughts his
     (sic) evaluations and his interpretations....
     (Brown, 1980, p. 44)

     Subjectivity in this case has been both personalized
and reified.  Indeed, on the very same page we are also told
that we should not always believe a person's Q sort, for
they may be lying or self-deceived!  To a social
constructionist, such language is the very stuff of
confusion.  Indeed, to us it seems that it confuses Brown
himself, because earlier he told us that an operant approach
... has little use for such platonic concepts as validity.
There is no outside criterion for a person's own point of
view" (p. 4).
     The different stories about Q that you can hear from Q
methodologists on the one hand, and ourselves as users of Q
on the other, concern not the procedure but the theory.  The
phenomena, "data" (e.g., best estimates of factors) can be
moved between one camp and the other unaltered.  Thus what
to Brown are "attitudes" or "viewpoints" (which thereby can
be, in some instances, delusions or falsifications) to us
are "opinionations," or "accounts" (which thereby are
neither true, nor false, but just "stories").
     What has changed?  The notion of "self-reference," "...
central to Q methodology ...," (Brown, 1980, p. 43) is
theoretically discounted in the social constructional use of
Q.  This, we believe, is the crunch point  -- where our use
of Q, and approach to it, diverge catastrophically (in the
sense of catastrophe theory) from its origins.  We have no
interest at all in its use as a "royal road" to subjectivity
as viewpoint.  Instead we see it as a powerful form of
pattern analysis (cf. Williams, 1976) which can be used to
identify alternative propositional configurations.  In
itself, we see Q analysis as doing no more than that --
allowing us to express in Q sort form best estimate
components from covarying evinced patterns (the Q sorts
given to us by our participants).  Why we found Q so
attractive is what we can do with those best estimates.  We
can use them, once uncovered, as indications of the
tectonics that render our social texts and cultural
discourse -- the accessible heterogeny of understandings of
social phenonema, voices in policy debates, or images of
characters or personae in the social drama.
     Whereas with self-reference, the person providing the Q
sort "may be lying or self-deceived" (Brown, 1980, p. 44),
once self-reference itself is seen as a trap in language,
the issue of lying dissolves.  If the Q sort is not seen as
"theirs" but as "the one that they gave," and if all that
can be given are culturally available understandings, voices
or images, then deception is not an issue!  As Q (from
whatever camp) is not concerned with the "factual,"
disolving the subject (cf. Henriques et al., 1984) absolves
us from the specious worries about the truth value of Q
sorts that the doctrine of self-reference creates in the
first place.

How Many Patterns?

     The majority of Q methodological studies from the USA
report 2, 3 or 4 factors.  When we have described studies in
which we found many more factors (sometimes 10 or more), a
few eye-brows (hackles even) have been raised.  We see this
as another consequence of the  traditional Q methodological
focus on subjectivity.  To be in two minds (or even three or
four) has some kind of commonsense plausibility -- to be in
12 has no credibility at all!  No wonder there is surprise
at the number of factors.  If instead we see Q as as means
of gaining access to cultural forms, collectively expressed,
no such limits operate.  It makes perfectly good sense to
discover that within a plural community, there are a large
number of alternative storylines being peddled -- by the
mass media, by religious interests, by political parties, by
professional and by pressure groups.  Indeed, we might well
expect the number of such "stories" or "accounts" to itself
vary, according to the area of culture explored.  For
example, when considering images or representations (e.g.,
of characters who are "mad") we may plausibly expect to find
plenty of different instances.  On the other hand, a tight
and well-rehearsed area of social debate might be expected
to yield only three or four alternative accounts, at the
most.  And certainly, one of the ways that Q can be used is
to explore the empirical support for these kinds of a priori
expectations.
     This is facilitated by a tendency to have standardized
our procedures.  First, we have opted exclusively for
principal components analysis (cf. Williams, 1976).
Secondly, we have employed a benchmark for factor
(component) identification, by seeking to interpret all
factors with eigenvalues greater than unity.  Thirdly, at
least in the initial phase of analysis, we explicate all
interpretable (i.e., pure loaded, or, as we tend to call it,
uniquely exemplified) factors.  Finally, we have also
"deviated" from received Q methodological wisdom in opting
for a standardized "atheoretic" rotation (specifically,
varimax), rather than "theoretic" rotations (normally where
one Q sort and/or participant is given a theoretic
significance not accorded to others and so pre-set upon a
rotated axis).  We are not objecting to such an approach by
Q methodologists.  Rather, where Q is used as a pattern
analytic in the study of cultural form, the grounds for such
manipulations are unlikely to arise.

Connections

     The different theoretical substrate to social
constructionist Q studies has enabled us to tackle different
research questions to those Q methodology has addressed.
One example is the frequent use of designs in which
participants complete more than one Q pack.  This enables
the exploration of questions about how "accounts" or
"stories" may interlink across areas of cultural discourse.
We have adopted the term "voice" to express how a discursive
position may be articulated consistently across a number of
topics or domains.  For example, a feminist voice may be
expressed via understandings of the causes of illness, or of
child abuse.  Equally, a biomedical voice may be articulated
within both an understanding of madness, and an account of
social policy about what society should do about the mad.
     To put this in concrete terms, when we give people two
Q sorts to do, one on understanding of a particular topic or
issue, the other upon policy or action, we can look for
systematic links between the two.  For example, where we
find that participants whose Q sorts tend to exemplify
factor 1 with our pack about understandings of child abuse
also tend to be the ones whose Q sorts exemplify factor A in
terms of social policy about child abuse, whereas
exemplificatory Q sorts for understandings factor 2 tend to
link with exemplifications of social policy factor C, then
we can begin to explore not only "accounts" and "stories,"
but the ecology of "voice" too.  There is quite a subtle
issue of interpretation involved here.  We are not saying
anything about what individual people believe or about how
"their" cognitions and conations interrelate.  The procedure
is merely a means of exploring what cultural elements tend
in practice to be voiced in conjunction with each other.

Digest

     The social constructionist Q studies conducted recently
in the UK can be told as sufficiently distinctive in theory
and practice to merit their own accounting.  Having outlined
the theoretical picture, the remainder of this paper is
devoted to a brief report of a typical study.

           The "British Dialect" of Q in Practice

The Worrying of Knowledge-Making

     In common with a lot of British Q work, the study
reported here began with a concern we had about the
mongering of meanings.  We were disquieted by the all too
ready ease with which apparently consensually understood
concepts were being employed as explanatory-descriptors in
both popular and professional discourse.
     Two strands of argument have made us question such
seeming consensuality.  First of all,  Q work empirically
indicates that multiplexity not singularity is the norm --
whatever the topic studied, there is never a bedrock of the
consensual.  Secondly, singularity is challenged by the
critical social constructionist (indeed, neo-marxist and
post-structuralist) linkage of knowledge and power.  For
example, Young (1980) has  described the concept of "stress"
as an individualizing discourse which, in according
"thing-hood" to an ephemeral m'lange of events and
experiences, mutes argument about and understanding of
other, more threatening and much more complex reasons for
experiencing discomfort, such as oppression or exploitation.
"Stress," Young argues, is neither a unitary cause of
certain forms of illness (as it is portrayed as being), nor
is it, in itself, a monolithic condition (as it is diagnosed
and treated).  Rather, Young argues, it is part of an
ideologically motivated epistemology.
     From such an ideological stance, we have become
radicalized over the ties between our discipline
(psychology) as an arena of academic knowledge-making, and
its associated practitioner hegemonies.  Not surprisingly,
therefore, we were drawn to study concepts which have been
constructed, in part, within psychology.  We wanted to
explore the epistemological arena within which psychology
has promoted its orthodoxy as "empirically valid," and hence
of superior epistemological status.

Overcoming "Addiction"

     Within such a research reality, the concept of
"addiction" clearly meets the criteria for deconstruction.
It needs to be said at the outset that we are not claiming
to have discovered (nor to be further elucidating) its
merely semantic problematics or the difficulties these have
caused to its medical or paramedical users (cf. Warburton,
1985) and equally worry the substitute term "dependence."
Rather our position would be that the appropriation within a
medical model of any ordinary language term (habit,
dependence, abuse, indulgence) -- even the apparently
value-free "use" favored by Warburton (1985) -- is to open
their meanings to penetration by alien analytics predicated
upon the world view of that model -- for example, to
constrain understandings of the "difference" between "use"
and "abuse" to an empirical issue resolvable by empirical
methods (and, hence, to render mute, say, a moral analysis).
     A Q study is a specific against any such problem of a
priori foreclosure, for it is capable of eliciting not only
such understandings but also their alternatives, for it
samples into the locally available ecology of
understandings.  Further, by imposing a methodological
discipline of epistomological relativism it presents each
understanding as a labor of equal merit, coming from the
equal acts of work (Q sorting) of the participants.
Further, in sharp contrast to the self-fulfilling
nosographic concerns of its medical users, "addiction" in a
Q study is what the participants say it is, formed out of
their work alone without constraint by the researcher's
pre-existent theory of "addiction."

Sampling the Concourse

     To these ends, the sampling of propositions voiced in
professional and popular discourse, through which the Q sets
would be evolved, was pursued eclectically -- drawing upon
both oral and written texts varying in location from
interview protocols, through mass media output, to
scientific texts.  As has often been the case in British
studies, this sampling, when piloted, proved to bifurcate
into a set concerned with Understandings (e.g., "Addiction
is a disease which requires medical treatment") and a Social
Policy set (e.g., "Alcohol companies should not be allowed
to sponsor sport and other social events").
     It is also common for British studies to request of
participants the completion of inter-linked Q sorts.  This
feature also permits the consideration of the inter-linkage
of voicings.  While as anti-essentialists we do not equate
the story-teller with the story, as cultural analysts we are
prepared to allow as an empirical possibility that stories
told about Understandings contemporary with stories told
about Social Policy may share a common reality, a view of
the world with plausibility across the two domains.  In
several studies this has, in fact, proved to be the case.

Research Details

     Both Q sets contained 80 items and were sorted by 68
participants who also gave biographical information and
open-ended glosses upon the items.  The participants
included people with a professional involvement in
"addiction," "substance users" and those of other habitual
"indulgences," and people who had no particular biographical
experience of "addiction."
     Each of the two sorts was independently subjected to
principal components analysis and varimax rotation.  After
the elimination of factors with eigenvalues only marginally
above unity and with only one Q sort giving a distinguishing
loading, twelve accounts of Understanding (factors 1-12) and
three of Social Policy (factors A-C) were identified.  For
each a best estimate of the component expressed as a Q sort
was made by making a weighed average (by factor loading) of
the Q sorts exemplifying that factor.  Interpretation was
based upon these rankings and upon the comments made by the
exemplifying participants about the items.  For ease of
access to the constructions so revealed, a precis of each
viewpoint was also generated.  These precis are closely
based upon the actual language of both highly polarized
propositions and open-ended comments.
     The multiplicity of factors to many British studies can
lead to a very ponderous "Results" section if the treatment
of results is seen to require each factor to be reported in
full (i.e., with the positionings (weights) of many items
covered along with associated open-ended glosses).
Consequently in presentation heavy recourse is often made to
the precis, a step only adopted after it had been
established that such summaries were recognizable and
acceptable to participants.

Findings

     A full treatment of the results to this study, while
necessary to enable a report such as this, is inevitably
lengthy and unnecessary to establish that the concept of
"addiction" is indeed a suitable case for deconstruction.  A
number of independent themes were found to weave through the
12 Understandings with varying saliences and with different
aspects being given prominence.  Factors 2, 5, 7 and 8, for
example, each focus in quite distinctive ways on addiction
as a biological response to substances which alter brain
chemistry.  However, of these only factor 7 opted for a
simple disease model of addiction.  Factor 2, in fact,
expressed the competing behavioral sciences account.  While
factor 5 stressed both individual differences in
vulnerability and differences between substances in their
addiction risk, factor 8 saw all substances as potentially
hazardous.  Against all of these accounts, factor 1 took the
view that addictions are better regarded as social
constructions than as biological givens, while factor 10
refused to confine the concept of addiction to substance use
allowing that it could apply to other behaviors.  Factors 3,
4 and 12, while differing in other aetiological aspects,
agreed in placing weight upon the blame that should accrue
to "society" over the problem of addiction.  By contrast,
factor 9 focused upon the role and culpability of the
"pushers." Factor 11 was distinguished in stressing the
benefits of some substance use.
     This brief overview of the 12 Understandings factors is
enough to establish that each account placed a different
gloss upon the definition of addiction, upon how aetiology
is accounted for, and on the way addicts and users should be
perceived.  Some accounts were overtly permeated by moral
values, some by ideology and politics.  Some were informed
by biographical experiences and some by more distal sources.
     With Social Policy, far less diversity emerged, a
finding which has sometimes but not always also established
itself in our studies.  Here the factorial divide proved
largely to parallel a tripartate conceptual model proposed
by Ray (1985).  Factor A argued primarily for personal
freedom in the context of considerable suspicion of the
motives of those who seek to control and intervene.  Factor
B evinced a demand for State control and intervention with a
stress on the need to protect the weak and vulnerable.
Factor C was most concerned with protecting the rights of
non-indulgers against the risks to which indulgers expose
them and directed this particularly against tobacco and
alcohol use.
     Overall, those participants who exemplified factors 2,
4 and 5 tended to generate Q sorts which exemplified or
loaded strongly upon factor A.  Those exemplifying factors
5, 6, 7 and 8 linked into a similar way to factor B.  Each
of these patterns illustrates the way in which several
independent understandings can inform what might be seen,
without the additional analysis, as a singular political
agenda on policy.  It is also important to note here that Q
analysis also allows us to resist simplifying the tension
between factors A and B into contrasting poles of a bipolar
dimension such as freedom vs. control.  Factors A, B and C
are mutually independent, mutually complementary discourses
on addiction policy.
     Just as contrasted understandings can converge upon a
singular policy program, so, more rarely, a single
understanding can inform more than one social action
discourse.  This happened within the present study with
factor 1 whose exemplars proved to divide between factors A
and C.  Biographical information on the participants
concerned suggests that a shared distrust of the power that
constructs received definitions of addiction was linked by
some to the voicing of a State power distrusting
anarcho-libertarian social policy but by others to a power
distrusting assertion of rights to freedom from false needs
and from the impact of others' indulgence.

Advertiser's Announcement

     Radical social constructionist studies such as the
above provide empirical grounds for questioning both the
taken-for-granted claims of scientism to offer a uniquely
superior understanding of concepts with aetiologies in
cultural tectonics, and liberal-humanistic notions of the
achievability of consensual social action and social policy.
     First then, we see the "addiction" study as part of a
program of deconstruction aimed at troubling the
epistomological pretentions of the social hygiene industry
(those practitioner crafts such as "Health Education" that
claim to be applied bio-social sciences).  It faces such
researchers or practitioners with the counter claim that
their accounts of "addiction" are just that -- tales from a
library of stories, constructions among many complementary
alternatives with no special claim to objectivity save that
valorization which comes from the power of their users.  As
constructions of reality they can in no sense be decoupled
from the politico-economic structures which maintain, fund,
inform and utilize them.
     Second, along with a number of other investigations we
have conducted in the UK into a range of contemporary
debates (e.g., child abuse and human rights), we narrate
this study as demonstrating that debates about social issues
like "addiction" are predicated neither upon a consensual
bedrock of agreement nor upon simple dimensionalities upon
which view can be strung.  Rather, there are sedimented
within a culture as a wide variety of alternative
constructions, stories plausible in the telling and of
internal coherence.  While in this specific instance the
diversity over policy was less manifest (as indexed by sheer
numbers of accounts located), those which were identified
were equally in mutual tension.  What these findings further
suggest is that in this (and probably in any applied
setting), it is both naive and misleading to make the
pluralist assumption that policy can reflect a social
consensus -- not least because policy agendas are themselves
complexly discursively interlinked to understandings that
are in turn informed by divergent forces of knowledge
construction and maintenance.  At best, any statutory
response may be integrable with some of the understandings
accessible to a collectivity and congruent with some
constructions of action.


                         References

Austin Locke, J. (1990).  Rashomon: A social anthropological
   model.  BASAPP Newsletter, No. 5, 9-11.
Berger, P.L. & Luckmann, T. (1967).  The social construction
   of reality.  Harmonsworth: Penguin.
Brown, S.R. (1980).  Political subjectivity: Applications of
   Q methodology in political science.  New Haven: Yale
   University Press.
Brown, S.R. (1989).  The first British Q conference.
   Operant Subjectivity, 12, 110-114.
Brown, S.R. (1990).  Remembrances of things past.  Operant
   Subjectivity, 13, 53-56.
Crumley, W. (1990).  Looking to the future.  Operant
   Subjectivity, 13, 103-104.
Featherstone, M. (Ed.). (1988).  Theory, culture and
   society.  London: Sage.
Giddens, A. (1979).  Central problems in social theory.
   London: Macmillan.
Gleeson, K. (1987).  Representations de la Maladie Mentale.
   In G. Bellelli (Ed.), La Representation Sociale de la
   Maladie Mentale.  Naples: Naples University.
Henriques, J. et al. (1984).  Changing the subject:
   Psychology, social regulation and subjectivity.  London:
   Methuen.
Kitzinger, C. & Stainton Rogers, R. (1985).  A Q
   methodological study of lesbian identities.  European
   Journal of Social Psychology, 15, 167-187.
Le Guin, U. (1974).  The dispossessed.  New York: Harper &
   Row.
Parker, I. (1989).  The crisis in modern social psychology
   -- and how to end it.  London: Routledge.
Ray, L. (1985).  Problems of substance abuse: Exploitation
   and control.  Social Science and Medicine, 20, 1225-1233.
Schutz, A. (1972).  The phenomenology of the social world.
   London: Heinemann.
Shotter, J. (1984).  Social accountability and selfhood.
   Oxford: Blackwell.
Shotter, J. & Gergen, K.J. (Eds.). (1989).  Texts of
   identity.  London: Sage.
Stainton Rogers, R. & Kitzinger, C. (1986).  Human rights:
   Bedrock or mosaic?  Operant Subjectivity, 9, 123-130.
Stainton Rogers, R. & Stainton Rogers, W. (1989a).
   Subjectivity, representation and communication: Report of
   the first British Q conference.  Operant Subjectivity,
   13, 15-19.
Stainton Rogers, W. & Stainton Rogers, R. (1989b).  Taking
   the child abuse debate apart.  In W. Stainton Rogers, D.
   Hevey & E. Ash (Eds.), Child abuse and neglect: Facing
   the challenge.  London: Batsford.
Stainton Rogers, W. (in press).  Explaining health and
   illness: Explorations of diversity.  London: Harvester.
Warburton, D. (1985).  Addiction, dependence and habitual
   substance use.  Bulletin of the British Psychological
   Society, 38, 285-288.
Williams, W.T. (Ed.). (1976).  Pattern analysis in
   agricultural science.  Melbourne: CSIRO.
Young, A. (1980).  The discourse of stress and the
   reproduction of conventional knowledge.  Social Science
   and Medicine, 14B, 133-146.

------------------------------------------------------------
     Rex Stainton Rogers is in the Department of Psychology,
University of Reading, Bldg 3, Earley Gate, Whiteknights
Park, Reading RG6 2AL, United Kingdom.  Wendy Stainton
Rogers is in the Department of Health and Social Welfare,
The Open University, Walton Hall, Milton Keynes MK7 6AA,
United Kingdom.

                  Copyright 1990
Communication Institute for Online Scholarship, Inc.