Using a consistently constructivist approach, the formation of prejudices, their
role in identity, and their resultant resilience are examined. Next, the strategies which may be used to persuade
people to change such constructs are considered. For the person who owns the prejudice these amount to experiences
which invite reconsideration of existing ways of construing: for one who desires to invite another to change, these
strategies include distancing, circular questioning, circulation, parenthesising and orthogonality of facilitator
A talk based on an earlier version of this paper was given at the American
Society for Cybernetics conference at Amherst Massachusetts on July 19 1991.
In this paper prejudice is examined from the point of view of constructivist theory. In
the following we should like to provide a perspective on the following issues: the genesis of prejudice, its role in
the individual, the conditions under which prejudice changes, and tactics which may be useful in changing prejudiced
ideas. Included in this discussion are a number of interconnected constructivist ideas on the nature of thinking:
first, the recursive nature of thought (the consequences of thinking become part of future thinking); second, the
biological and ontological limits to knowledge (which limits have their own ethical implications); and third, the need
to re-examine assumptions embedded in language concerning the nature of thought and its interconnectedness with nature.
It is our intention to consider changing prejudice generally, whether it be focused on
race, religion, age, or gender. In earlier work one of us examined these issues in relation to gender (Gash &
Morgan, in press) and more recently in relation towards children with mental handicap (Gash 1992, Gash in press).
Through history the way in which people have interpreted differences between social groups has played a role in the
stability and structure of their societies. At present many countries have serious political and social problems which
are intimately connected with the way people think about human differences. We will give two examples; one local and
one international. First, the local example. In Ireland, in the six counties known as Northern Ireland, there are
serious problems commonly seen to arise out of religious differences. There are serious differences too between groups
of people in their views on how to cope with the situation in Northern Ireland. These differences have become so
important to the people holding them that some members of the groups regularly resort to violence, firmly believing in
the correctness of their actions. The international example is taken from the current situation in the U.S.S.R. where
many different Republics have different views on how their future development can best be planned. Our hope is that
insights, from constructivism and cybernetics, into the nature of thought and how prejudice is formed and changed will
help people understand each other better and to be more tolerant of their differences.
While Hans Kung (1990) has argued that world survival requires a world ethic, we join
with other cyberneticists in thinking that insights like these into human thinking and human systems will form part of
such a world ethic.
Levels of thinking; how and what.
It is Piaget's account of the process of thinking which remains perhaps the most
enduring feature of his theory. The biological affinities of his "equilibration process" are clear. The
purpose of equilibration is to maintain an equilibrium between the organism's ongoing experience and its previous
experience. Ongoing experience is structured (assimilation) by previously acquired structures, which in turn change (accommodation)
in response to new features noticed. There is continual dynamic interaction between these two aspects of the process of
What is often unnoticed in this account of experience is that it does not allow for
direct experience of the environment. Any consistent or radical constructivism does not allow for direct experience of
the environment Von Glasersfeld (1987, 1991). This is a point which we would regard as requiring emphasis because it is
central to appreciating the practical implications of this way of understanding people in their social lives. Piaget
(1970) himself was notoriously ambiguous on this point. Consider his quote (1970, p. 706) "..assimilation is the
integration of external elements into evolving or completed structures of an organism." There is nothing in this
to alert the reader that interpretations of experience are constrained by existing structures and may vary profoundly
according to an individual's past experience. It cannot be assumed others have similar interpretations of any event.
A case can be made for the giving Vico (1710) the credit for being the first
constructivist (e.g. Gash & Von Glasersfeld (1976); Gash (1980)). A critical feature of Vico's epistemology was the
implicit refusal to allow knowledge of the world to people on the grounds that one only knows fully (scientia) that
which one has made oneself. This provides a solution to the recurring problem in the history of philosophy - how can
one be certain that one's knowledge is a true representation of reality? The constructivist's solution is to see the
problem as insoluble: one cannot get outside one's experience to make the comparison, one can only compare experience
with existing structures. Differences in experience are noticed, learned, and then participate recursively in later
noticings and learnings. What is important is that the cognitive items thus created are viable (Von Glasersfeld, 1987).
Constructivist theories imply levels of performance from simple to complex. The
earliest activities described in Piaget's theory are the primary circular reactions. During the first months after
birth, these circular processes provide the infant with knowledge of its own hands, and more generally of its body.
While developmental psychologists are familiar with 'levels' and the controversy over whether such levels imply stages,
Bateson (1972) argued that theories of learning
have been particularly slow to draw attention to some implications that such 'levels' have for learning. Bateson (1972)
drew attention to the parallels between levels and logical typing and argued that what was known about learning of one
logical type did not necessarily apply to learning of a higher logical type.
We have found Bateson's (1972) ideas about levels of learning to be very useful in
thinking about prejudice. We will begin with a quote concerning his levels of learning:
"Zero learning is characterised by specificity of response which
-right or wrong is not subject to correction. Learning 1 is change in specificity of response by
correction of errors of choice within a set of alternatives. Learning 11 is change in the process of learning
1, e.g., change in the set of alternatives from which choice is made, or it is a change in how the sequence
of experience is punctuated. Learning 111 is change in the process of learning 11, e.g., a corrective
change in the system of sets of alternatives from which choice is made." (Bateson 1972, p. 293,
italics in original).
Level zero which is only included for the sake of completeness, occurs when the learner
is pre-programmed to respond in fixed ways. Hunger, thirst, and probably some levels of sexual interest are examples of
such "wired in" behaviours. Their satisfaction is mediated by the maintenance of physiological equilibria
including hormone levels, blood sugar levels and so on. There is no choice in experience at this level, one is hungry,
or thirsty. Bateson (1972, p. 287-288) gives numerous examples of learning 1 with simple laboratory learning
such as is described in psychological experiments: habituation, Pavlovian conditioning, instrumental learning, rote
learning, and the disruptions which can occur when reinforcement is withdrawn. Each of these phenomena illustrate a
change in behaviour which is not characteristic of level zero learning. More importantly, in each of these cases of
level one learning context is held constant. Let's consider an example. A man learns to shoot with a rifle. One can
describe the skills involved. In fact, Bateson (1979, p.195) has devoted some pages to distinguishing the difference
between the type of learning involved in learning rifle shooting (level 1 where error correction works on single
actions) and shotgun shooting (level 11 where correction can only work on classes of actions) (1979, page 195). Learning
1 is about learning an item of behaviour.
Taking William Powers (1973) cybernetic idea that behaviour corrects errors in
perception according to goals, then we can say that in learning 1 the learner adjusts behaviour to control
specific items of behaviour, aiming with a rifle, tennis shots, or ways of hitting the ball in golf, or football. One
could also include items from the cognitive domain ranging from spellings, to scientific and literary facts. At level
1 what is learned is items of behaviour or facts.
Learning 11 can be seen to have happened when there is a change in learning one.
When one has learned a certain number of facts about word processors at level one one then learns that they work in
certain ways. For example, in earlier times we learned that the operation of saving one's document requires care in the
choice of "filename". Carelessness in this operation could result in wiping out a previous document with the
same name. After their initial struggles with word processors one can reasonably expect learners to have learned how to
learn about them. This second level learning will be manifest in the questions they ask, and generally in their
expectations about the functioning of novel computer systems.
Consider the man who has learned to shoot game (learning 1, rifle shooting) and who
announces he enjoys shooting game (elephant and hippopotamus). He risks verbal attack from people who disapprove for
ecological reasons. On the first occasion that this happens the hunter will realize that s/he is amongst an anti-blood
sports group. The time spent hunting which is appreciated with hunting groups has another meaning for this group. A new
context is learned. Learning at this logical level is not about behaviour but about the categories of contextual
organization of behaviour (e.g., Bateson 1979, p.134).
Identifying context, [the context that gives identity] therefore, plays an important
part in self identity because the way in which one anticipates events comes to be part of the way in which others
expect one to behave. The hunter has some choices now that the views of this group of people have been made public.
Let us provide another type of example: two men meet at a social gathering and find
common interests. Later at work one of them mentions the other and is informed that his new acquaintance is
Catholic/Protestant. In many contexts this item of information might have minor social consequences. However in certain
areas in Belfast there are very serious social and personal consequences to being a Protestant or a Catholic. Let's
give an example. In a discussion at work Andrew (a possibly Protestant name) mentions his friend Ciarán (a possibly
Catholic name) and encounters disbelief from his friends (who are Protestant/Catholic); he realises that his friends
are prepared to treat him differently because he associates with a member of the "other" religion. In this
case the friendship is contextualized by the attitudes of the colleagues at work. Additional meanings have been added.
There are now perceived values added to the fact of the relationship. So in my examples, an interest in hunting and
having a type of friend, have been given a context by the peer group, a context which colours the meaning of the
individual's previous learnings.
What is at stake here is the forced choice between the individual being a constitutive
component of one organization as opposed to constituting another incompatible organization. Here we use the term
'organization' in Maturana's sense to underline the religious or ethnic social system being specifiable as an entity
with closure manifested as a third-order network of conversations.
A description of an individual can be given in terms of level two learnings. If one
learns that being friendly works in certain contexts one learns to be friendly. If one was to describe an individual as
friendly, playful, and opportunistic, or another as mean, thrifty, and unscrupulous one can see that these descriptions
arose out of situations in which the behaviour could be explained in terms of learning one. In other words, being
thrifty in one context is generalized to other contexts by an individual who found that being thrifty worked in the
initial case. It is easy to see that the characteristics of "personality", such as those mentioned above,
form part of the stereotypes which can be used as ways of organizing one's thoughts about members of different groups
of people, whether one is thinking about racial groups, or groups based on gender, or some other potentially
One must understand constructions partly in terms of their intent, that is, their intent
to the construing subject. From the point of view of Kelly  the 'intent' of any constructive anticipation is to
optimally relate the person to his medium in such a way that he 'fits' and simultaneously conserves his personal system
coherence [Kenny, 1989]. Furthermore a construct is structured as a distinction so that it simultaneously affirms and
negates the polarised alternatives of which it is comprised. With a prejudice therefore, it is the case that part of
the effect of the intent in expressing one's identity [ie., of relating oneself to the medium] is to denigrate an
object or person and thereby affirm or empower the self. If conditions are such that it is an easy option to feel good
by putting another down then this may become a recurrent pattern. Indeed, applying the analysis above, people can
become prejudiced when they are rewarded for making prejudiced statements.
Where historical events have drifted in such a manner that two ethnic or religious
groupings have distinguished themselves from one another [by this process of simultaneously affirming and mutually
negating] using exactly the religious distinction to achieve this mutual characterisation, then to remove this
distinction of mutual prejudice is to dissolve the very basis upon which each tribal group founds its existence.
Obviously, this is a threatening prospect which is always refused.
Change and identity.
Second level learning is central in Bateson's description of the process of acquiring
stereotypes and prejudices. In this constructivist approach what are the reasons for the enduring intransigent quality
of prejudice which causes so much conflict in the world? There are three sorts of answers, not entirely independent
from each other. First, such traits are modes of assimilation or "ways of looking" at events; second, they
have a self-validating quality; and third, efforts to challenge these types of learnings may simply serve to identify a
new context in which the prejudice will be merely suppressed.
First, ways of looking at things are not easily challenged. This reflects what we tried
to bring out in the account of Piaget's process of thinking; assimilation by individuals will be based on their prior
learnings and assumptions cannot be made that different people will assimilate an event in the same way. Whether one is
for or against a group of people will have a bearing on the interpretation one gives to their behaviours. Take the case
of the Birmingham Six, it took sixteen years to get the British Legal Establishment to admit that the evidence used to
convict these men was faulty and unreliable.
Second, people tend to see the things which they want to see. From this point of view
constructs are 'self-fulfilling prophecies' [Kenny, 1989]. This illustrates the self-validating nature of this type of
learning. Writing in the latter part of 1992 it is tempting to cite an example involving the different construals of
the recent French referendum on Maastricht. President Mitterand was quoted as hailing the result as a triumph for
French democracy: the British press were not slow to emphasise the small size of the majority, and infer that the
Maastricht treaty was in need of reexamination. Bateson (1972) considered second order learning as difficult to change,
and if learned in infancy almost impossible to eradicate. Allowing President Mitterand some pride in his victory, the
following quote is illustrative (c.f., Bateson 1979, p. 134):
"Moreover, like all themes of contextual learning, these themes of
relationship are self-validating. Pride feeds on admiration. But because the admiration is conditional - and the
proud man fears the contempt of the other - it follows that there is nothing which the other can do to diminish
the pride. If he shows contempt, he equally reinforces the pride."
Third, we suspect most people have tried to change prejudices or stereotypes by giving
information, so it need hardly be pointed out that this is generally an ineffective way of trying to change peoples'
ideas. What seems to be likely is the creation of a new context for the learner who discovers that here, e.g., in the
Protestant/Catholic (note the order) work context, "I must be careful about what I say about Catholic/Protestant
This account goes some way to providing an explanation of why prejudice is so difficult
to change. Part of the difficulty is that second level learnings are about the ways in which behaviours or information
are classified or patterned, and this is independent of any one specific context. Efforts to change a prejudice which
fail to take account of this by providing specific counter-examples err in their appreciation of what is needed to
change prejudice. Second, identity is involved and people change the way their friends expect them to behave at the
risk of altering their status in their social group. We should now like to consider how prejudices and stereotypes may
A first consideration is that an individual's prejudices are personal constructions with
social implications for the individual's identity and group membership. So as we have just seen - telling any
individual that their second level learnings are wrong is unlikely to have the desired effect, and may even strengthen
the learning. In fact some of the efforts to change gender stereotyping in children have actually increased it (see the
review by Liben and Bigler, 1987).
More generally, the conditions under which one is likely to successfully reduce
prejudice are well known and stringent. They include the notions that cooperation should be maximized and competition
minimized; that roughly equal numbers of minority and majority group members should be involved; and that differences
in competence should be avoided (Stephan, 1984). These observations indicate that changing prejudices will need careful
planning and appropriate transactions.
Thinking about relationships in a circular way allows insight into why threat may not
lead to change in a person's construct system. There are two reasons for this, one is that it is important for anyone,
indeed any organization, to conserve their identity as a way of fitting with their environment. Secondly, as Kelly
(1955), has pointed out, conditions of threat do not facilitate change in a person's construct system. So, in trying to
change an individual's constructs it is important to consider carefully how the individual one is interacting with is
going to interpret the interaction. If what one is proposing as a change helps this individual to organize his or her
experience in a better way then the reorganization may be forthcoming, otherwise the proposed change is unlikely to be
Secondly, it is possible that an individual will learn new facts using level one
processes without engaging in level two learnings - for example at different points in time. Similarly it is possible
that some change may occur at level two without engaging level three processes. Presumably this is more likely during
critical developmental periods in the formation of the individual's identity. So during childhood and adolescence the
situation is likely to be more flexible than it would be for aspects of identity which are expressed from infancy in
some form or other. It may also be the case, however, that the same processes are needed even if the learning may be
Level three learning is about "a corrective change in the system of sets of
alternatives from which choice is made." (Bateson, 1972, p. 293). Amongst the examples given of learning 111
are changing the habits learned via learning 11, and limiting and directing learning 11. So applied to an
example mentioned earlier it is about learning not to be ashamed of one's friendship with members of a
"different" group. These learnings involve a shift in one's identity. Because of this Bateson cautioned about
the dangers of level three learning, because in it one's individuality may become blurred. Here one is hopefully on the
edge of growth, but because of the alteration to one's expectations one is going to change. So people desiring to
promote such changes need to be aware of the dangers. Clarity about the identity of context, of the shared meanings in
social encounters, is necessary to be oneself and to act authentically.
It seemed to us that people may change their thinking when they are required to
reconsider the viability of their thoughts by (1) being asked questions, or (2) by discussing counterexamples.
So if we consider the general case of the prejudice "members of the outgroup are unreliable" - questions may
be of the form "is this always true in your experience?" or "can anybody think of a case where this is
not true?" Counterexamples can be introduced in various ways - with people, on video, or via anecdote (e.g. Gash
1991). We have referred earlier to the conditions which are conducive to altering prejudice.
Distancing strategies (Sigel, 1982) are one type of questioning which may be
found useful in reducing prejudice. We find it useful in explaining distancing to use Maturana's (1988) distinction
between experience and explanation. Experience just happens, whereas explanations require a frame of reference and they
require language. Moving from experience to explanation is initially a creative process in which language is put on
experience in accordance with an individual's existing language structures. Distancing strategies are questions which
are put to a person about their experience which are designed to invite them to put new words on their experience.
Talking about the process in this way draws attention to the biological-cognitive constraints leading to statements in
Sigel (1982) sees distancing as critical in the development of representational thinking
in children. The higher level distancing strategies referred to by Sigel and his colleagues include open questions
which place demands on children to infer causality, to draw inferences, and generate alternatives. Each of these types
of demands may well be useful in trying to get older people to reconsider their ideas about people, their qualities and
potentials - especially when such ideas are prejudiced.
In illustrating distancing one can observe that it is often difficult to describe
intense emotional experiences, for example a bereavement, just after the event. The process of talking about such
events in a caring atmosphere is therapeutic. In other words, in time, a person learns to speak about the bereavement
without being overcome with emotion. Feelings can be expressed in ways which come to feel appropriate, to do justice to
one's sense of loss, and to the memory of the other. At this point one has, following Sigel above, distanced oneself
from the immediate emotional impact of the experience: one is no longer overwhelmed by the experience: and strictly one
no longer acts in the same way. In the process one has changed, though this may appear to be overstating what has
happened. It is a point to which we will return in the next section.
Caution is in order however, because some types of experiences are very difficult to
reconsider or to distance oneself from in the present. Experiences which put in jeopardy a well established identity
will be far harder to reconsider. Those of us with insight or therapeutic experience will be more aware of those areas
which for us (or others) are taboo, sacred, or threatening. It may well be that for some prejudiced individuals the
threat to identity involved in changing a well ingrained prejudice would be so profound and involve so much work as to
be impossible or 'unwanted' by the prejudiced individual. Consider the remark in the film Gandhi as to how to solve the
conflict between Hindu and Moslem in India; Give a Hindu family a Moslem child and have them bring the child up as a
Moslem and give a Moslem family a Hindu child and have them bring the child up as a Hindu. Gandhi's idea here was that
if a family were to do this they would learn the mutual respect necessary to allow mutual coexistence.
The process of producing a viable psychological / emotional distance between the person
and his core prejudices is obviously more complex and difficult than creating such a viable distance between the person
and some particular 'external objects' like other religious groups. Thus we have the need to have a four-part strategy
to generate the requisite distance within which strategic frame the desired changes may occur. This strategic framework
involves the four principles of Parenthesizing, Circularity, Circulation and Orthogonality. These will be
outlined and discussed below.
In this section we have put the learning process at centre stage as a necessary prelude
to discussing ways of reducing prejudice. It is our view that the process of putting together constructs with which to
reframe experience, is inevitably constrained by the necessary requirement that individual's change in ways which
conserve their identity. If identity is threatened then respect [orientation for the positive coordinations of actions]
is lost and it is likely that the threatened individual will feel a need to strengthen the stereotype or prejudice.
Bateson's (1972, 1979) emphasis on context in distinguishing levels of learning provides insight into why this should
be so. Sigel's (1982) notion of distancing provides insight into the role of questioning in trying to get people to
reconsider their prejudices. In a study which invited children to reconsider their gender stereotypes one of us found
that the teachers' own experience had taught them to be respectful of the children's feelings in this and other areas
where the children's identity is involved (Gash 1991). It was good to find that the teachers' own classroom experience
coincided with the more theoretically derived experiences of the importance of considering identity as pivotal in
thinking about ways of helping others reconsider their expectations or prejudices.
Implementation in Teaching-Learning
Having considered how the learning process works, let us consider how it can be
implemented. Teachers have a critical role to play because in spite of what we have said about distancing - prejudiced
statements appear in networks of conversations from which it is impossible to extract oneself. One is embedded in one's
conversational networks because they are part of one's identity. We think help is needed to rework one's being. The
trust and openness needed is necessarily excluded by the thrust of arguments marshalled with a view to insisting on
right and wrong. So, what are the insights which help in implementing the ideas with which this paper has been
concerned? They are insights which come from constructivist therapeutic approaches. However this work can be seen as an
extension of the educational implications of Piaget's account of the process of thinking.
There are four elements in this: circularity, parenthesising, circulation, and
orthogonality. Here these four principles [elsewhere called the 'principles for the generation of space for novelty',
see Kenny 1989] have been framed as ways of destabilising the conversational network's coherence. There now follows a
summary of the use of the four generative principles in a classroom situation where the focus is on perturbing changes
in preemptively held attitudes concerning the topic of 'Men / Women' as the prejudiced focus.
CIRCULARITY - The Introduction of Alternative Observers
A main assault on 'TRUTH' is to demonstrate that there are viable alternatives to the
received wisdom. There are other positions from which one could see things differently if one was to move from one's
own position and take up a perspective on the issues from some alternative vantage point. Having to listen to someone
else's version of the truth forces one to call into question the 'obvious nature' of one's preferred 'truth'. It forces
one to ask questions not about the 'facts' of a case but about how it is that other viable versions came about. This is
to raise questions about the different ways in which people invent reality, and the differing criteria for the
acceptability of explanations that exist for different groups of observers.
The principle of circularity can be seen as providing a framework for opening up of the
recognition of the existence of valid alternative ways of thinking. What this means is that we contextualize the fixed
prejudices within a wider domain [Dilated domain] which brings forth alternative voices, opinions, and observers.
Instead of allowing the prejudiced opinions to stand as the source of referential reality, these prejudicial anchoring
points are placed within the context of other observers who see things differently. In other words, what happens here
is that the prejudicial text is not allowed to stand as an independently existing object which is self-evident.
Instead, an observer is introduced as a necessary part of the domain of discourse. It is the relationship among the
particular observers that must be brought into focus. This is achieved by introducing into the field a variety of
different Observers, each with their own observing preferences. What this allows us to do is to bring into question the
particularities of any one observer in the ways in which he or she relates themselves to the other observers in the
context of the prejudicial text. It is the peculiarities of these inter-relationships that will explain the differences
in observer position in the text.
This is a first assault on the idea that -
(a) everybody sees this issue in the same way (Uniformity); and
(b) that everybody is sharing a common identity as revealed by the prejudicial orientations (Homogeneity).
One of the things which the facilitator of the conversation, whether therapist, teacher,
or negotiator, must work on is the ways in which the assumption of homogeneity of views must be conserved. This will
include attempts to silence the 'dissident voices'. The task will move inevitably from the narrow focus, whether it be
religious prejudice or gender prejudice, to the wider focus on the ways in which systems of conversations and beliefs
operate to conserve themselves intact as an ideology. This will lead to an analysis of the emotional convictions that
underlie these attempts at conservation of invariance in the group and of the forms of inter-relationships which are
oriented to the conservation of a given group identity.
The following is an example taken from the area of gender prejudice. This area was
chosen because it seems to us to be somewhat easier to illustrate the principles in this area rather than in the more
volatile and often lethal conversations of religious and racial prejudice.
The actual way in which circularity is manifested is in terms of "circular
questioning" which is done simply by asking one person what they think another person thinks or feels about
TEACHER: Who wants to give an opinions on this issue?
ROBBIE: Well I think that girls are just plain stupid and that's all there is to it!
TEACHER: That's a clear statement from Robbie describing his own view of this
matter. Who else has a view on this?
ROBBIE: Everybody agrees with me.
TEACHER: There are many different views and opinions among this group on all kinds
of matters, and what I first want to know from all of you is why you think that Robbie thinks that you all agree with
him on this?
PAUL: Robbie thinks that we all hate our sisters the way that he does!
MIKE: Yeah, some of us don't even have sisters to hate!
TEACHER: Paul, do you think that if Mike had sisters that he would hate them the way
Robbie hates his sisters?
PAUL: No, because he gets along with mine - especially my older sisters!
TEACHER: Mike, what does Paul feel about his sisters?
MIKE: Well he gets on with them OK - he plays tennis with his twin sister often.
TEACHER: Does he like his twin sister better than the others?
MIKE: Its hard to say, but I guess twins always stick together more - even if they
also fight with one another more too.
TEACHER: Who has any theories as to why Robbie hates his sisters so much?
PAUL: Its because they are all older than him and pushed him around when he was
small. Now they have him boxed in and he can't do anything about it.
TEACHER: Any other ideas from the group?
BILL: I think its because his parents allowed them to leave school early to get jobs
and he is still here.
TEACHER: Why was that?
BILL: Because his parents didn't see the point of schooling for girls after the
minimum amount. Its better if they are earning money for the family.
TEACHER: OK. Any other ideas about Robbie's relation with his sisters? So far we
have a theory about being controlled and dominated by sisters when he was younger; and another theory that says he is
envious of his sisters having left school and got jobs. Who else has another view on this?
JOE: I was wondering if it had more to do with the fact that his parents are
treating his sisters differently to how they treat him. I mean, they can leave school and he is stuck with it, it
TEACHER: Why do you think that this is important to how Robbie gets on with his
JOE: Well, he wants what they have and he doesn't like them getting away with
something that he can't get away with.
PAUL: Yeah, I go along with that. Robbie thinks his sisters get away with murder,
while he is always getting a hard time from his Dad.
TEACHER: Why is that?
PAUL: He's the only boy in a family of girls, and his dad treats him tough, while
the girls don't really matter that much to him.
TEACHER: Is there another reason for thinking girls are stupid and so on?
JOE: Yeah his Dad wants him to turn out tough like him, and that means that he has
to be able to take rough treatment. The girls don't have to take any rough treatment since they are expected only to
get married one day and that's all. But Robbie will have to make it as a man in a tough world. Women have it easy.
TEACHER: So you think that women have it easy in this world. How would anyone in the
group describe Joe's feelings about women?
ROBBIE: I think he agrees with me. Girls have it soft and it is not fair to expect
that they should be treated equally with us.
TEACHER: So here we have an idea that there is an injustice being done to men, who
have it tough, and yet are expected to treat women (who have it soft) as equal to men? In what way does this injustice
show itself in society?
ROBBIE: Well I gave the example of my sisters being allowed to do things that I
JOE: Women demand the same opportunities as men in work, but it is men who have
built up the industries and frameworks of work over the years. Why should women just come along and expect to have an
equal slice of the cake?
TEACHER: Can anyone in the group think up some possible reasons as to why it is
reasonable that women should expect an equal slice of the cake?
PAUL: Well personally I don't feel any sense of injustice about this. I don't see
that women have it soft, because my sisters work as hard as me for exams and all that . Also my mother works at a part
time job as well as working at time. We all have to help otherwise it is a mess at home.
TEACHER: Paul, how do you explain the sense of injustice voiced by Robbie and Joe?
PAUL: Well we heard Robbie's resentments about his sisters. With Joe it seems to be
an argument about how men came to have more rights at work than women. But it leaves out the other kind of work in the
household that someone has to do. the women had to do that type of work with kids and stuff throughout history. Now
things have changed.
In these interchanges we can see the teacher eliciting inferences and alternative views
through the process of Circular Questioning. The objective is to "re-contextualize" the domain of discourse
from that assumed as being homogenous to one that is populated by observers of divergent opinions. This does not
necessarily include a full drawing out of the lines of argument from each point of view. the main task is to clearly
bring forth the awareness that there are a number of different Observers with different ways of describing and
explaining the (apparently) same phenomena.
What will be seen in moving through these levels is that the phenomena being described
are in fact not the same at all. What is being described is different according to the values and intentions of the
observer who is doing the describing.
PARENTHESIZING - The Introduction of the Multiverse
The term parenthesising is taken from Maturana (1988) although we use it here in a more
extended sense. It is introduced here to create ways of calling into question apparently independently existing facts.
So when two people are in disagreement one does not allow the argument to descend to the level of "objective
reality" which would allow an arbitrator to decide which person had the correct version of the truth. This means
creating and maintaining a series of positions from which different observations are made by focusing on the ways in
which we communicate together and the intentions which we each have in such communication. Different observers, the
different selves in communication, often have the identical purpose of finding validating instances of their own
existences in the so-called domain of objective reality. When people operate like this it means that someOne will find
his or her self validated and someOne else will find his or her self invalidated or negated.
To avoid this conflict of conserving self-interest parenthesising is crucial. In
parenthesising the desire is to show that any so-called "object of discussion" has an entirely
subjective context - that it is brought forth by the operations of distinction of some observer. It is not simply 'just
there in reality'. The intent is to cut off all attempts to appeal to some 'independent' referee or fact which can
prove one person right and another wrong.
The principle of parenthesising, then, is an attack on the assumption that there is only
one way of viewing or describing the reality of any prejudice. Having introduced alternative Observers with
Circularity, we now introduce and maintain the alternative Observations using Parenthesizing. The tactic
proceeds by the explicit exploration of the implications and constraints of the current prejudicial stance, followed by
the searching for alternative constructions, ideas, theories, possibilities, conjectures upon the ways things could be
configured between people with different prejudices. It is based on an effort to be creative and constructive in order
to construe both in a positive manner, and to construe in a way different to the conventional wisdom. In doing this the
facilitator must be mindful of the group constraints against creativity of this sort which threatens to remove a
fundamental orienting construction from the group's currency.
The assumptions being tackled through the use of parenthesising are:
(a) that there is only one true version of reality,
(b) that this is a simple truth which is easily asserted, and
(c) that some one person is actually privileged to know the truth about this manner, and therefore privileged to
"know better" than anybody else (Forms of Authority being claimed).
Often as a result of applying circularity certain observers will begin to fight for the
right to be the 'one who knows best'. When this occurs the facilitator is precipitated into this second process of
parenthesising where the facilitator focuses upon the status of the actual prejudicial ideas, using various sources and
materials to invalidate the "fact-uality" or "accuracy" of these prejudices. The group must also
examine its presumed sources of "authority" and to speak the prejudices it espouses. In this case the focus
must be on the relation of the particular observer to the prejudicial prescriptions.
TEACHER: Why do you think Robbie that girls are more stupid than boys?
ROBBIE: Because they like dolls and dolls are stupid?
TEACHER: In what sense is a doll "stupid" do you think?
ROBBIE: I don't like them.
TEACHER: So because you don't like them dolls are stupid?
TEACHER: Is it also the case that because you think girls are stupid that you don't
ROBBIE: That's right, they make me feel bored. They're a waster of time and that
makes me mad to have to be with them.
TEACHER: So you have a lot of feelings about girls then?
ROBBIE: No I don't, I just told you they're stupid.
TEACHER: You feel dislike for girls; they make you feel something when you call them
ROBBIE: OK, but I don't feel anything nice about them.
TEACHER: Are there other things that you don't like and that you feel are stupid?
ROBBIE: Yeah, school!
TEACHER: Why do you think school is stupid?
ROBBIE: I'd prefer to be out in the street with my pals, and anyhow my Dad says its
a waste of time!
TEACHER: Why does your Dad think school is a waste of time?
ROBBIE: He thinks all of you teachers are a bit soft headed, and anyhow none of you
know anything about surviving in the real world with real jobs and doing something important.
TEACHER: What would you like to be when you grow up?
ROBBIE: Rich and tough!
TEACHER: Like your Dad?
ROBBIE: Yeah! Well he's tough at least.
TEACHER: We seem to have come along a track from discussing "stupid girls"
with their dolls to successful touch men. do you think there is any connection between your feelings about girls and
your ideas about you growing up to be tough?
ROBBIE: What do you mean?
TEACHER: Well, you have two kinds of strong feelings which you experience: first,
strong feelings about girls being stupid and a waste of time; second, strong convictions about how you should be
spending your time - e.g., being in school is a waste of time, in order to become the kind of tough adult that your
father is. What I wonder is if you need to believe that girls are indeed a waste of time, and school is a waste of
time, in order for you to feel that you will be able to become a tough adult in the future? Or to put it the other way
around - if you started to like girls or school would you still be able to grow up tough?
Here is an example of a teacher exploring the internal logic and implications of the
prejudicial ideas of Robbie. the teacher begins to draw out the polarized possibilities for acting within Robbie's
frame of reference. Only when we come to see the constraints upon Robbie's choices of action can we begin to generate
alternative possibilities. The rest of the group will be very useful in posing alternatives that could liberate Robbie
and help him make more positive choices. Prejudices hang together in constellations and have varying amounts of
meaningful implications for the identity of the person. The idea is to show how the person's use of prejudice is
intrinsic to the person's own way of making sense, and is not an intrinsic quality possessed by the object of the
prejudice. Further, it is easy to expose the person's assumed source of authority to speak to a process of invalidation
by showing the "subjective" origins of the particular prejudice in question.
CIRCULATION - The Introduction of Alternative Text to the same
The third step in the process is the use of circulation. Having firstly established the
forms of group inter-relationships at stake, and secondly, the creative possibilities for alternative perspectives to
the prejudicial text, the facilitator is now in a position to analyse more closely the relationship between the
observer and the texts being produced by these observers. Seeking to loosen up and experiment with alternative ways in
which these observers might constructively relate themselves to the alternative materials being produced. More so, we
are after the possibilities for using these alternative construals to relate the self to the social network in novel
ways and particularly to use these alternatives to inter-relate the observer community together in ways which generate
a common identity not premised upon the prejudicial negation of some members or of certain non-members of the
collective. What we are ultimately aiming at here is to generate new ways for the participants to become a constituent
member of a 'parallel' or alternative system with organizational closure.
This works largely through a role-playing procedure where all of the observers are asked
to enact roles which are premised upon the novel alternatives. This is done in a playful way where the role-taking
observers and the observer audience combine their efforts to materialize and elaborate the implications of these new
ideas in their networks of conversations. The important point here is that the observer community should be able to get
a feeling for the possibilities of reframing their collective identity under a different pattern of values. The success
of this depends in great measure also on the abilities of the facilitator to orient the enactments towards this sense
A useful method here is to ask those who have started from the position of asserting
prejudices to write, as a group, a script for one of the actors advocating one of the alternative opinions which
emerged in the group discussions. This helps them to deepen and make more complex their understandings of the
constellation of constructs involved in the alternative viewpoint. In this case textual interactions are not supplied
since they would amount to a small play describing the alternative descriptions and explanations of human conduct which
would replace the interactions based on the prejudiced text.
ORTHOGONALITY - The Introduction of the Non-Constitutional Position
The fourth and final part of the process is to have implemented from the outset, the
process and 'position' of orthogonality. This is the most complex aspect of the whole process since all of the other
tactics of re-contextualization are themselves contextualised within the frame of orthogonality. To create an
orthogonal frame (or Bias in the Batesonian sense) means to create a domain of interactions (with any individual
or group of individuals) within which the facilitator, teacher, consultant or therapist can relate effectively with the
other participants but in such a way as to conserve his own professional 'neutrality' and to not interact in a way that
would be constitutive of the conversational group’s [whether it be tribal, cultural, or family], value system in
which the participants are constitutive members.
This raises very important questions for the personal status of the facilitators’
self-critical awareness and capacities for self-critical monitoring. If the facilitator has not engaged in some
processes of self-critical analysis, at least in the areas under discussion, then there is little or no possibility for
him or her being able to successfully generate the necessary orthogonal contextualization for this entire process. It
is obviously important for the facilitator to embody the parenthesizing of prejudice in a reflexive manner.
One of the most important aspects of the orthogonal positioning of the facilitator is to
continually extricate him or herself from the tacit and explicit attempts of the various speakers or factions within
the group to get the teacher to agree with, side with, endorse or otherwise validate the prejudicial attitudes in
question. If this enmeshment should occur, the teacher has lost the powers for the open facilitation of this process,
for unfolding the prejudicial issues in a way that can lead to changes of attitude.
The teacher can fall into this trap in many ways. The most common being the non-verbal
communication of agreement and disagreement with the different observer positions. Unless the facilitators have
undergone a specific awareness-making process using, for example, audio-visual feedback, they are unlikely to sustain
an orthogonal positioning for very long - since human groups are premised upon the politics of who agrees and
dis-agrees to be coordinated by the group's characteristic assumptions. Facilitators frequently signal their agreements
in a variety of non-verbal ways. It is to avoid any ambiguities that in meetings we are asked to make the unambiguous
non-verbal sign of agreement/disagreement by raising hands in the air to vote.
TEACHER: We were listening to the different opinions in the group about women
"having it easy" or not.
ROBBIE: Well what do you think sir? do you think it is fair that Ms. Brown gets time
off work because she is having a baby? If she's going to have a family and miss out on work its better that she does
not take up a job that a man could have instead.
TEACHER: Do you think that there should never be any women teacher's then?
ROBBIE: Not if they are going to try to do two things at once that do nothing
together. If they have kids then they should stay at home and look after them.
TEACHER: So you think that all women should be made to choose either to work or to
have a family?
ROBBIE: Yeah! they can't have it both ways. Why should they?
TEACHER: Can anyone help with this issue? Can anyone think of some reason why women
not only can have it both ways but must have it both ways?
JOE: Yeah, what if it was the case that men had to have the babies, that it wasn't
women. What would we be saying then? What's it like to be the one in the household who has to have the babies, and to
have to work as well? Lots of families these days can't afford to live unless both parents are working. Does this mean
that these people shouldn't be allowed to have kids? That seems too extreme to me. do you agree with that sir?
TEACHER: Let's imagine for a minute the type of world we would be living in if
there was a law forbidding poor people from having any
kids. How many of you would have been born if, where both your parents have some kind of work, they would have been
forbidden to have kids?
PAUL: Well, I for one wouldn't be here, nor any of my sisters. My parents like
family life so much I can't imagine them ever being happy without kids. It would be a very lonely world for them and
people like them. Is this what you (to the teacher) are getting at? You are against this type of inhuman society aren't
TEACHER: What we are trying to do here is to imagine what it would be like living
under the type of conditions which follow from forcing women to choose to work or to make a family. then we have to
decide what type of society we prefer to live in and why.
This scenario is to try to illustrate some of the obvious ways in which the teacher's
own preferred positions are ‘invited’ into the discussions by the students. The teacher must avoid these
invitations if he or she wishes to avoid:
(a) polarizing the discussions into rigid groupings of "for" and
(b) adopting an authority position of he or she who knows best;
(c) becoming an advocate for one of the sub-groupings of the conversational system against the others;
(d) constituting the conventional wisdom or politically correct position on the topic in question.
None of these options will serve to elaborate, evolve or change the pre-existing
strongly held positions of the conversational participants.
The Copernican revolution decentered "people" from being at the centre of the
universe and said that everything did not revolve around "people". Similarly parenthesizing can be read as
another revolution in which we must come to realize, paradoxically, that "persons" are at the centre
of the universe in the sense that everything we see and know is entirely dependent on our creative acts as observers
So when we hear a person saying an extremely prejudiced statement one can respond with a
phrase such as - "I know that that is the way it looks to you; but are you sure now; let's look at the way in
which this idea is very important for you, and how you came to hold this idea in the first place?" One emphasises
that it is the person's own view and its validity is only his or hers (and it may perhaps have no further relevance).
In doing so one is accepting the person's right to make the statement, one is accepting the statement for the moment,
but one might well be hoping that it will be possible for the person to see limits in its applicability - soon. So
parenthesizing is the process of putting parenthesis on statements - to take away from the person [who made the
statement] the illusion that such a statement is the only existing perspective or value.
Summarizing these four aspects of therapeutic encounters: orthogonality is the
facilitator's or therapist's deep awareness of the validity of all perspectives, and the avoidance of being
simultaneously aligned with one as 'obviously correct' and of negating another as 'obviously mistaken'. Circularity
is to generate, for the clients, the awareness of a differentiated field of observers, no one of which can claim to see
the 'whole story' or the 'truth' but can only offer one version of the story. Parenthesizing is the closing off
of the possibility of refuge in any 'privileged source' or 'external authority' and focusing the speaker back upon the
operations of distinction by which some 'subject' creates some 'object'. Finally, Circulation is to introduce a
variety of 'voices' or 'text' to the different observer / speaker positions taken up within the network of
conversations constituted by the participant observers. Here each person learns that he can take up the text of
different voices at different times without fear of contradiction or fear of losing his legitimate position as a bona
fide participant within the conversationally closed system.
The key for cognitive change seems to require a type of mutual respect so that
differences in perspective between individuals can be explained with the consequence that either a common understanding
can be reached or there will be a respectful decision to differ.
We will conclude with a quote with which Maturana (1991) ended his recent ontology of
"Wisdom breeds in the respect for the others, in the recognition that power
arises through submission and loss of dignity, in the recognition that love is the emotion that constitutes
social coexistence, honesty and trustfulness, and in the recognition that the world that we live is always, and
unavoidably so, our doing."
Bateson, G. Steps to an ecology of mind. New York; Ballantine, 1972.
Bateson, G. Mind and nature. New York: Dutton, 1979.
Deissler, Klaus, G. Recursive creation of information: Circular questioning as
information generation. (translated by Steven Awodey) Marburg; Infam 1987.
Gash, H. Reducing prejudice: constructivist considerations for special education. European
Journal of Special Needs Education, 1992, 7, 146-155.
Gash, H. A constructivist attempt to change attitudes towards children with special
needs. European Journal of Special Needs Education, (in press).
Gash, H., & Morgan, M. School-based modifications of children's gender-related
beliefs. Journal of Applied Developmental Psychology, (in press).
Glasser, W. Control theory in the classroom. New York: Harper Row, 1986.
Kelly , G. A. . The Psychology
of Personal Constructs. 2 Vols. [New York: Norton]
Kenny, V. . Life, The Multiverse and Everything: an introduction to the ideas of
Humberto Maturana. In Goudsmit, A.L. [Ed] 'Self-Organization in Psychotherapy', [Heidelberg: Springer-Verlag].
Kenny, V. . Anticipating Autopoiesis: Personal Construct Psychology and
Self-Organizing Systems. In Goudsmit, A.L. [Ed] 'Self-Organization in Psychotherapy', [Heidelberg: Springer-Verlag].
Kung, H. trans. John Bowden. Global responsibility: In search for a New World Ethic.
Liben, L.S., and Bigler, R.S. Reformulating children's gender schemata. In Children's
Gender Schemata, eds Lynn S. Liben and Margaret L. Signorella. New Directions for Child Development, Vol. 38.
Winter 1987. San Francisco: Jossey Bass, 1987.
Maturana, H. Reality: The search for
objectivity or the quest for a compelling argument. Irish Journal of Psychology, 1988, 9, 25-82. (Special issue:
Radical constructivism, autopoiesis and psychotherapy. Ed. Vincent Kenny.)
Maturana, H. Science and daily life: The ontology of scientific explanations. In F.
Steier, (Ed.), Research and Reflexivity. London: Sage, 1991.
Sigel, I.E., and Cocking, R.R. Cognitive development from childhood to adolescence: a
New York: Holt, Rinehart and Winston, 1977.
Sigel, I.E. The relationship between parental distancing strategies and the child's
cognitive behaviour. In L.M. Laosa and I.E. Sigel (Eds.), Families as learning environments for children. (pp.
47-86). New York: Plenum, 1982.
Sigel, I.E., Stinson, E.T., and Flaugher, J. Family process and school achievement.
To be published in a book by D. Reiss and R. Cole. (text available from I. Sigel. ETS
Princeton, NJ. 08541.)
Stephan, W.G. Intergroup relations. In Lindzey, G., and Aronson, E. Eds. Handbook of
Social Psychology, Volume 2. New York: Random House, 1984.