Download Rosenberg, S. - Faculty Web Sites at the University of Virginia

We Are Not The Same:
Individual Differences in the Quality of Political Judgment
Abstract
An attempt is made to examine the logic underlying how people evaluate social and
political phenomena. Adopting a structural pragmatic approach, two hypotheses are
explored: (1) with certain caveats, an individual will evaluate very different issues in the
same structured way; and (2) different individuals' may evaluate the same issue in
structurally different ways. To test these hypotheses, a highly varied group of 48 subjects
were interviewed in-depth on three topics: (a) the public display of homoerotic art, (b)
U.S. intervention in Bosnia, and (c) helping a friend or relative. The results strongly
support the hypotheses. The paper concludes with a consideration of the relationship of
this research to work on political attitudes, political cognition and cognitive development.
We are not the same
Page
1
What people value and how they judge political events are questions of ongoing interest
to political science. Typically the analysis of political values focuses on people's attitudes. In
empirical studies, subjects are presented with issues or events of current public interest and asked
to express their opinion, usually in some close-ended bipolar format. While acknowledging the
value of this work, the research reported here is oriented differently. Rather than looking at
attitudes, the final product of a reasoning process, an attempt is made to examine how people
evaluate political phenomena. Attention thus turns to what people focus on when evaluating an
event or policy and how they judge it. This is investigated through in-depth interviews in which
subjects are asked in an open-ended fashion to identify what is important to consider when
judging an event and to explain why these factors are significant or valuable.
We adopt a structural pragmatic view of political reasoning, one that draws on the
developmental theories of Jean Piaget, L. S. Vygotsky and Jurgen Habermas. We focus on the
logic underlying people’s evaluations of political phenomena and how the quality of this logic
may vary across individuals. To capture this variation, we define three types of political reasoning
and evaluation. We then report empirical research that investigates the adequacy of this typology.
We conclude with a discussion of the relationship of our work to three other approaches to political
judgment: the survey research on political attitudes, the social psychological research on political
cognition and the developmental psychological research on social and moral reasoning.
A Structural Pragmatic Perspective
Building on the developmental psychologies of G.H. Mead (1932), Jean Piaget (1970,
Inhelder and Piaget 1958, 1970) and L.S. Vygotsky (1962; 1978; Luria, 1978), we view
reasoning as a purposive activity, an attempt to operate on the world. This involves placing
objects (e.g. persons or things) and actions in relation to one another in the attempt to achieve a
We are not the same
Page
2
goal. In this process, these objects and actions are rendered meaningful relative to one another
and come to be represented in a particular way. This pragmatic conception of thinking is
supplemented by a structuralist claim that there is a distinctive quality to how a person reasons
and therefore to the kinds of meanings (conceptual relations) and representations that person will.
Consequently, the various specific representations, understandings, evaluative judgments and
action strategies an individual constructs will all reflect a common underlying structure or logic.
Although a person may have differing levels of interest in or information about different topics
(e.g., a local policy matter and her relationship with her best friend), there will be structural
similarities to the kinds of questions she asks, the answers she finds satisfactory and the
evaluations she makes. For example, a person may reason in such a way that she tends to focus
on specific events and tends to render them meaningful by connecting them to a single antecedent
cause. According to the more structuralist aspect of our approach, this person’s reasoning strategy
should be in evidence whether she is explaining Muslim hostility to the US, the persistence of
poverty in America, or the unhappy mood of her co-worker.
Despite this emphasis on cognitive structure, we do not view reasoning as simply a
psychological process, a quality of mind, but as a radically social psychological one that
incorporates both the subjective intent and understanding of the individual actor and the
regulative force and cultural definitions of the social context. Thinking is something that a given
person does and therefore necessarily reflects her cognitive qualities or capacities. However, at
the same time, how a person thinks also reflects the social direction and meaning that inhere in
the social organization of action and the cultural narratives to which the individual is exposed.
The latter effectively how regulate how actors can talk to and behave towards one another in
everyday social interaction. As a result, the individual’s attempt to operate upon and with other
We are not the same
Page
3
people (and consequently how she thinks about them) is necessarily affected by these social
practices and definitions, even if, as is often the case, she does not fully understand their social
meaning. Consequently, the way she makes sense of the world is necessarily socially guided as
well as personally constructed. In the interplay between the individual’s attempt to subjectively
construct her own thoughts, purposes and actions on the one hand and her attempt to follow the
rhetoric and action strategies dictated by the environment on the other, a dynamic tension is
created. It is a tension between what she spontaneously constructs and thus fully understands and
what she assimilates from her environment, but may only partially comprehend.
Over time, the aforementioned tension may produce a developmental transformation in
the individual’s thinking. Following Jean Piaget (1970), we suggest that this cognitive
development occurs, in part, through “reflexive abstraction.” This is a psychological process in
which the individual begins to recognize the inadequacies of her way of understanding. This
leads to a shift away from the unreflective use of specific knowledge to a reflective consideration
of how one gains and uses knowledge. This leads to the development of a new way of reasoning,
one that is less egocentric and more integrative and abstract. An example is the transformation in
the quality of reasoning that typically occurs in the transition from childhood to adolescence or,
alternatively, the change in reasoning that may be produced by a university education.
There is also a more social dimension to cognitive development, one that involves the
internalization of cultural definitions and social regulations of action. This occurs in what L. S.
Vygotsky referred to as the “zone of proximal development,” (Vygotsky, 1978; Wertsch, 1991;
Valsiner, 1992; Hogan and Tudge, 1999). This zone consists of action patterns and/or ideas that
are too complex to be constructed by the individual herself and yet are simple enough that the she
can recognize and imitate the basic elements of what is being presented. As a result, individuals
We are not the same
Page
4
can adopt speech patterns and action strategies that enable them to participate effectively in
discourses and interactions that they do not fully understand. An example might be a student who
can draw on the rhetoric of a theory to participate appropriately in a classroom conversation even
though systematic probing might readily reveal (to the student as well as the teacher) that the
concepts are misunderstood by the student using them. As in this example, the use of socially
modeled rhetorical or action strategies has two effects: (a) more immediately, individuals can
effectively operate (and thus think and act) in ways that they do not fully understand; (b) more
long term, this “use without understanding” provides both a guide and an impetus for the
development of the individuals’ reasoning.
In this view of cognitive development, culture and social structure may expand or reduce
the possibilities for that person’s cognitive development. Environments are structured differently
and therefore place different kinds of demands on and introduce different kinds of opportunities
for the individuals who inhabit them (Habermas, 1979; Chilton, 1988; Kegan, 1994; Wilson,
1992; Nicolopoulou and Weintraub, 1998; Rosenberg, 2002). As a result, individuals may
achieve different levels of development and consequently may reason in fundamentally different
ways. We will explore these differences in our empirical research.
A theoretical caveat must be introduced here. Although emphasizing the structured
quality of thinking, we do not claim that a given individual will always think in the same way.
Changes in the structure of an individual’s social context may elevate or reduce the quality of that
person’s reasoning. In addition, a particular social context may cue responses, emotional
commitments and understandings that were constructed at earlier stages in a person’s
development. (For more on emotion and cognitive development, see Dupont, 1994; Kegan, 1994;
We are not the same
Page
5
Noam and Fischer, 1996.) However we do suggest that, given the requisite time and interest, an
adult will generally think at the highest or most advanced way she has developed.
Extending the cognitive developmental tradition of Piaget, Vygotsky and Mead to the
study of political thinking, Rosenberg described three types of political reasoning: sequential,
linear and systematic. They represent a developmental trajectory in which thinking becomes
increasingly integrative, de-centered (less dependent on the individual’s egocentric view of the
immediate situation) and self-consciously interpretative. Several studies were conducted to
examine these developmentally based differences in adult political thinking. Typically subjects
both participated in an experiment and were interviewed in-depth about two or three very
different events.. The results provided strong evidence that a given individual reasoned about
substantively very different issues (e.g. the dynamics of a pendulum’s oscillation and the
American air strike on Libya) in the same way and that different individuals might reason about
the same issue in structurally different ways (Rosenberg, 1988a, 1988b, Rosenberg, et al. 1988).
More recent research has demonstrated the effect of cognitive structure on the processing of
television news (Braunwarth, 1997); tolerating social and ethnic differences (Hanks, 1998), and
responding to aggressive political messages (Golec, 2002). Here we build on Rosenberg’s
typology and go beyond considerations of political understanding to an analysis of judgment and
evaluation. In so doing, we expand the definition of the three types of reasoning to include an
analysis of differences in the quality of the values individuals hold and in how they deploy those
values to evaluate political phenomena. Three types of evaluation, sequential, linear and
systematic, are described below.
[INSERT TABLE 1 HERE]
Types of Political Evaluation
We are not the same
Page
6
Sequential evaluation - feeling and needing
Sequential reasoning is embedded in the immediacy of present circumstances and the
unfolding of current events. The most elementary of the three forms of reasoning discussed here,
sequential reasoning is distinguished by its shifting focus, its dependence on appearances and its
relative lack of clear causal and categorical considerations. Sequential evaluation consists of
currently evoked feelings and needs. The orienting questions are: What does this feel like? Do I
want what’s happening now to continue? Embedded in present experience or cued by memory,
this evaluation depends on sensory experience (e.g., pain and pleasure). These sensed emotions or
reactions are understood as an aspect of the sequence of events in which they are evoked. In this
regard, subjective and objective dimensions of experience are not well distinguished. The
objective dimension is infused with subjective feeling and that feeling is identified with the
objective condition under which it is elicited. The connections here something of the kind posited
by more traditional forms of learning theory.
Sequential evaluations are inextricably linked to concrete present events. The needs and
feelings that constitute sequential evaluations are produced as events occur. At times, events in a
single sequence may elicit different feelings. These simply co-exist in the context of the sequence.
For example, the performance of a religious ritual may produce both fear and a sense of comfort.
Evaluative responses are not only by evoked directly by present events, but also by the
connection of present to remembered events. Feelings associated with a current event may be
readily shaped when the memory of past event whose memory is cued by present circumstances
may readily be associated with present events. Response to a current event may be readily be
influenced by the feelings associated with a past event whose memory is cued by the present
event. Alternatively, present circumstances may produce responses that then may be associated
We are not the same
Page
7
with remembered events, changing how they are evaluated. In this way, prior evaluative
associations (except where learning trials are traumatic) may readily be reconstituted through new
learning. Consequently, sequential evaluations and the internal dispositions they produce are quite
fluid or changeable.
The objects of sequential evaluation, the things that are understood to require judgment,
are concrete events. Specific objects or persons are considered with reference to the specific
events in which they participate. As a result, actors may be evaluated quite differently in different
contexts. From the perspective of other types of thinking, sequentially structured evaluation may
therefore appear unreliable and a person who evaluates sequentially may appear moody or
spontaneous in a manner often regarded as child-like. More abstract and categorical entities such
as the self, groups, behavioral rules or ideas are not natural objects of sequential reasoning or
evaluation. They typically will be ascribed little or no value, except perhaps in concrete contexts
in which the voicing of support for these ideas is associated with either pleasant or aversive
results. Social identity is simply not an issue for sequential evaluation.
Based on the individual's direct experience of feelings, sequential evaluations are quite
egocentric. Social influences do, however, enter in several important ways. First, the experienced
sequences that are at the heart of sequential reasoning and evaluation often are part of interaction
patterns that are organized by existing social institutions. Second, sequential evaluation may involve
the unself-conscious internalization of the expressed feelings and reactions of others. When
observed in the course of events, the feelings or wants expressed by another person may be
understood as attributes of the event itself. These observed feelings may then cue a match to
memories of one’s own feelings and consequently evoke a similar reaction. The result is an
empathic response, one that is produced by the confusion of the other's feelings with one’s own.
We are not the same
Page
8
Third, sequential evaluations may be affected by labeling. For example, the terms “good” and
“bad,” as used and reinforced by other people, may acquire a reward value of their own. When
attached to events by others, these terms may affect the sequential thinker's own evaluations.
The following excerpts exemplify aspects of sequential evaluation. The first, drawn from a
discussion of stealing to help another, is an example of an evaluation where the question cues a
remembered sequence that is associated with negative reinforcement. Notably the learned sequence
is applied in a way that is unaffected by an attempt to introduce a new hypothetical outcome.
Intrv.:
Subj.:
Intrv.:
Subj.:
Intrv.:
Subj.:
Intrv.:
Subj.:
Intrv.:
Subj.:
Oh, okay. Let’s say that... how about if you didn’t have the money… Would you be willing to steal
the drug? Break into the doctor’s office to get the medicine for your brother?
No.
How come? Why not?
Because my parents raised me, you know, not to steal. It’s wrong. I got busted when I was a
young kid because we were... Me and two other kids were horsing around one day and we busted
into this place and messed it up, you know.. the juvenile hall line. I learned my lesson real quick.
Okay. You mentioned you wouldn’t steal the drug, in part because you learned your lesson when
you were a kid. What if you knew that somehow you could steal the drug and never get caught?
Would you be willing to do it?
I’d be tempted but I don’t think I’d do it.
How come? You wouldn’t get caught, there’d be no price to pay.
Yeah, but in my mind it would still be wrong.
And why’s that?
It would still be the wrong thing to do. Because it would still be stealing and you may not get
caught now, but further down the line you’re libel to do something else and get caught.
The following is an example of evaluation in an empathic consideration of the another’s
pleasure. When asked for justification, the initial claim of pleasure are simply reiterated. Moreover
the response, cues a memory and the initial question is left behind as the subject retells the story
which leads off in a direction of its own.
Intrv.:
Subj.:
Subj.:
Now... it’s kind of an interesting thing that your brother plays Santa. Do you think it’s a good thing
that he does that?
Oh yeah. He enjoys it. He loves it.
That he has so much fun doing it. He enjoys talking to the kids and like, most Santas, they try
and... like South Coast Plaza last year, they had Santa. And they had kids and, you know, rush
them through. Get your picture taken, next kid. And Rob says you can’t do that with the kids. The
kids, you have to... When they put them on Santa’s lap, the kids want to tell Santa what they want
and you’ve got to listen, you’ve got to talk to them, you know, and try to quiet them. He says, “I
found out the last couple years that sometimes its better if the kid doesn’t sit on Santa’s lap, if he
just stands next to Santa.” “Some kids you know are between two and four,” he said, “two year
olds will have nothing to do with Santa. When you’re getting to the age of four, it’s fine, it’s cool.”
We are not the same
Page
9
The next excerpt (from a discussion of the display of homoerotic art) has several
hallmarks of sequential evaluation. The request for justification yields a loose train of
associations that follow on one another. An additional request leads to a simple
reiteration of the bad sequence. When asked about causal relations in the sequence, it is
apparent that these are not part of normal consideration. There is difficulty in maintaining
the boundaries of the topic, allowing children to see the exhibit, as subjective associations
with child pornography lead the respondent in his own direction.
Intrv:
Subj.:
Intrv.:
Subj.:
Intrv.:
Subj.:
Intrv.:
Subj.:
Intrv.:
Subj.:
Do you think these photographs should be displayed or not?
No, I don’t.
How come?
Well, because for one thing it’s a regular art museum, parents are going to bringing their kids in,
the kids don’t need to see this type of thing. They see enough violence and everything on TV. And
they’ve been raising a big stink here lately on this bit on the Internet with these people are
contacting kids over the Internet and a lot of the kids are running away from home, and so on. And
I think it’s wrong. I think really that the sexual attitude that the country now is more liberal than it
was. More easy going. And they accept the homosexuality, same sex type marriages and stuff like
that. And this is fine. But when you start bringing kids into it, no.
Okay, why not?
Because, to me, the kids don’t deserve to see this. The parents don’t need to see this.
Why not?
In a way this is obscene. You’re bringing maybe a ten or twelve year old kid into a photograph like
that and it’s going to ruin the kid for the rest of his life. The kid may not realize at the time that
they’re doing anything wrong but later on in life it’s going to affect him.
How do you think it will affect him?
I have no idea... I’ve never really gone into it with anybody, but I’ve talked to people who’ve.... I
talked to a cop that was on a raid to a child pornography ring up here in L.A. and he’s a veteran of
twenty years and he said it turned his stomach to see some of these pictures.
In the last example, the object evaluation, foreign policy, is too remote for
significant judgment. Request for evaluation merely cues a remembered story sequence.
Again that sequence cues others and again the original concern is readily left behind.
Intrv.:
So should we have gotten involved the way that we did to try stop the conflict, or are we better off
not to get involved and let events just take their course. What do you think we should be doing?
Subj.: Well, I figured we were going to get involved when it first broke out.
Intrv.: How come?
Subj.: Because of the United Nations and that a lot of the countries, as you said, all the separate, the
Bosnians, the Serbs, the Croatians, and all, were going to ask for help from the US to try and avoid
all the conflict that they’ve got over there now. And there’s still going to be fighting. They’re like
the Middle East. They’re like Palestine and Israel. They’re going to be fighting for the next twenty
years the rate they’re going. They keep talking about peace, but nobody wants peace. Everybody
wants this piece of land, this is considered part of Israel, this is considered part of Palestine.
Palestinians want their own home land, separate from Israel and the other Arab states, and they’re
We are not the same
Page 10
still fighting. They talked peace five, ten years ago and they’re still fighting over it. That one little
strip of land, that Gaza strip or where ever it is in there.
Intrv.: Well, and what about Bosnia…?
Linear Evaluation: personal preference, normal practice and social norms of actions
Linear thinking emerges through a reflection upon and objectification of the means of
sequential thinking. Thus it involves abstracting actors and action from the observable flow of
events. As a result, linear thought focuses on specific actions, actors and objects. Unlike in
sequential thinking, a given actor or action can be considered across different event contexts and
a number of actors or actions can readily be considered simultaneously. Evaluating circumstances
in a linear manner, a person decides what she specifically wants to do, the desirability of
categories of persons, objects or actions, and the normative appropriateness of sequences of
actions.
Linear evaluation may be viewed as a three-step process. Initially there is a primary
evaluation of a novel object (an action or actor) that involves linking the object to some already
valued outcome. This may occur in several ways. First, evaluation may be established on the
basis of personal experience of the immediate satisfactions or dissatisfactions associated with the
object. Second, evaluation may depend on social judgment. Here the object is evaluated in a
manner that reflects the judgments of other people. Third, evaluation may focus on an entire
sequence of actions. The sequence is evaluated either with regard to regularly recurring patterns
of action one has observed (what is normal) or with regard to what the culture defines to be
appropriate (what is good). As such, linear evaluation tends to be the more conventional than
either sequential or systematic evaluation. These patterns and norms are themselves not subject to
evaluation.
Once established, the initial evaluation leads to a secondary creation of value. The value
of the action in question tends to flow to those actions or actors to which it is subsequently
We are not the same
Page 11
related. This relationship may be established through a causal link. For example, if an action is
positively valued, actors who initiate or receive the action may be positively valued. The
relationship may also be categorical. For example, if an actor is positively valued, all the
attributes of the actor (e.g. its actions, beliefs, preferences) may be positively valued. Similarly if a
group is positively valued, then all members and activities of the group will tend to be positively
valued. In either case, there tends to be a singular, univalent judgment of social categories (a
person or group is thus judged overall to be either good or bad, or likeable or not). The result is
definition and evaluation often imply each other in linear thought - common concrete identity or
categorization produces common evaluation, and common evaluation produces common identity
categorization. This suggests that the social categorization and identification processes discussed by
social psychologists Hogg and Abrams (1988) may be understood as manifestations of specifically
linear evaluation.
The third step in the evaluative process is that of application. Linear evaluations, both
primary and secondary, exist apart from immediate experience and are applied to it. In this
regard, linear evaluations appear to be driven by specific behavioral rules and social conventions
and the resulting judgments tend to be relatively enduring. When new actions and actors are
linked to actions or actors already evaluated, the tendency will be to judge the new objects in a
manner consistent with the old. Where there is a value conflict, it consists of a practical conflict.
A typical example would be where a single act has both desirable and undesirable consequences.
Generally this will be resolved by attempting to render the countervailing consequences more
consistent by redefining, diminishing or eliminating them. (In this light, the dissonance reduction
behavior documented by social psychologists (e.g., Abelson, et al. 1968) may be understood as an
expression of linear and not systematic or sequential evaluation.)
We are not the same
Page 12
The following interview excerpts exemplify linear evaluation. The first involves the
assertion of a conventional rule that is not itself subject to evaluation when judging an act
of stealing. Its value is just “obvious.”
Intrv.:
Subj.:
Intrv.:
Subj.:
Intrv.:
Subj.:
Okay. So here’s the scientist...Who has created this cure. And it’s sort of hard to tell what he has
in mind. It seems like he’s doing it for profit. Um, do you think it’s a reasonable justification that
you need the drug and he’s not willing to meet that social obligation?
Um, no I don’t think that justifies stealing.
Why not?
I think stealing is stealing, whether it be $1.50 off your taxes or um taking a box of cookies from
the supermarket, I mean it’s stealing. You want it now. I mean I you know I, it would be pretty
hard for me to justify stealing, the constant, just stealing.
Why is that so important?
Well, it’s just not the right thing to do, I mean (laughs) I mean for things that are obviously wrong
to do, you know everybody’s parents have told them not to steal which obviously is unacceptable
behavior, so to try to justify it by creating a circumstance in which it seems more permissible, I
don’t think it makes it anymore right.
The second is example drawn from a discussion of displaying controversial art of
where a conventional rule is understood as natural or normal.
Intrv.:
Subj.:
Intrv.:
Subj.:
Intrv.:
Subj.:
What do you mean by the basics?
Things like freedom of expression. You know, freedom of speech. Those are very basic human
rights.
Why is that?
Um one is that humans have the ability to express themselves. You know, they’re born, most
individuals at least, um are born with the ability to express themselves. And to speak and to write
for that matter. Um, and to censor that, those very human characteristics. Just I don’t think it’s a
very good idea at all.
Why not? What are the consequences to that?
Well the consequences of that is that we go to a very homogenous society and that wouldn’t be
very fun at all.
The third excerpt is an example of the difficulty of understanding (when talking about
Bosnia) an evaluative perspective that involves anchoring values and conventions of practice that
are different than one’s own.
Subj.:
Intrv.:
Subj.:
Intrv.:
I think we have an obligation to diplomatically try to and I don’t think the President has done this
or not, so I don’t think Warren Christopher has been effective enough, I just can’t believe that in
1990s, we can’t reason with people to stop fighting. And it’s just so stupid, we have case study
after case study on different types of wars that can show them that there isn’t a benefit, you know.
The people die, countries go broke, um and governments fall apart. So I just can’t believe that we
can’t reason with these people to negotiate some sort of peace treaty?
Well if the combativeness of the situation, the cultural differences in their opinion…
They’re weird.
Maybe that um dynamics of the situation prevents their recognizing a solution, and experience
would seem to show that it’s been difficult to achieve a lasting peace.
We are not the same
Subj.:
Page 13
But if this, but if the solution is to break up the country, that’s a solution you know? If indeed
they don’t want to live in harmony together, well let’s solve that problem. You know, there’s just
so much out there to do and to be done, I just can’t believe that those people want to spend the
next several years playing this war. And through mediation, I just can’t believe that we can’t
resolve this.
The fourth excerpt is an example of the assuming a necessary connection between
good practices and outcomes when discussing business practices.
Subj.:
Intrv.:
Subj.:
Um, excuse me …there is a happy medium which we call democracy where people should think
about their social obligations when they are practiced in business. I mean I think businesses that
only focus on their economic priorities eventually do fail.
Why is that?
Because I think um I just can’t imagine that a business would be able to succeed without thinking
about um the social realities of, you know in this country. I think it would have a tremendous
effect in the employees of that business, and um I just it’s hard for me to picture an organization
that doesn’t realize its social obligation just because we have so many problems right now. I mean
even when you talk about um big corporations such as IBM, you know oil companies, they do
understand about giving back to the community. I mean they understand that it’s part of their
economic development strategies to give back to the community…. So I think it’s very important
for businesses and individuals running those businesses to think about their social obligations. I
mean I cannot imagine how a business could succeed with, um, its only intent being profit.
The last excerpt from a discussion of foreign intervention exemplifies valuing something in
terms of its concrete effects. It is also an example of bolstering. When a negative consideration is
introduced, the subject responds by bolstering (adding new positive effects) and mitigating
(reducing the negative effect).
Subj.:
Intrv.:
Subj.:
I mean there’s great benefits to us to bring peace in the world, and financially, socially, we will all
benefits of having peace everywhere.
Um, just sticking to that for the moment, um assuming that our intervention could result in
ultimate peace, what about the costs involved. Um, in terms of whether those troops remain there,
um it could end up hurting the US economy quite a bit. So that people like you would earn less
money. Does it make sense to you to avoid that situation?
Um, (pause) it seems to me that wars are actually good for the American economy. It certainly
causes a boom in some of the defense activity that support a lot of private businesses, and it makes
a heck of a lot of people feel good about their country when we’re in a war for whatever reason.
Um, and you know we seem to forget a lot of their problems when we’re at war. I don’t see it as
being that negative with an economic impact. I mean maybe in the long run it would cause us to
increase taxes but um maybe a short term thing.
Systematic evaluation - judgment of function and consideration of principle
Systematic thinking is the most sophisticated form of thinking of the three analyzed here.
It emerges with the objectification of the reasoning and evaluation characteristic of linear
thinking. The concrete anchored relationships of linear thought are no longer the result of
We are not the same
Page 14
thinking but its starting point. Consequently systematic evaluation begins with a reflection upon
the specific beliefs, preferences or conventional practices that are espoused by a person or a
group. These personal beliefs and social conventions are then interpreted, that is they are located
within a system (e.g. a personality system or a social system). The result is a relativistic
determination of the subjective or the cultural meaning of a specific interaction, preference or
norm. Following this interpretation, a judgment is made regarding its contribution to the well
being of the system in question. The judgment is a functional one. The value of the interaction,
value or belief is evaluated relative to the role it plays in the maintenance of the coherence,
elaboration and development of the particular personality or social system of which it is a part.
When the relationship between two or more systems is at issue, judgment focuses on the
appropriate form that such relationships should take. Typically this entails a consideration of the
quality of the systems involved and the need to maintain the integrity of them all. Principles of
exchange are defined in this manner and specific patterns of interaction, beliefs and values are
judged accordingly.
Proceeding in the above fashion, systematic thinking naturally distinguishes between
two levels of evaluation, one of social preference and the other of ethical judgment. The first
focuses on an individual's preferences or a group's conventions. These concrete specific
preferences and behavioral norms are understood as an expression of the particular personality of
the individual or the particular culture of the group in question. This leads to a relativistic
conception of specific preferences and behavioral norms. This systemic view also leads to a
consideration of the positive and negative aspects of a person or group as mutually supporting
elements of the whole of an individual’s personality or a community’s social life.
We are not the same
Page 15
At the second level of evaluation, there is a focus on the overarching requirements of
systemic functioning or inter-systemic relationships. The emphasis here shifts to the systemic
consequences of the specific interactions and evaluations being judged. This raises general ethical
questions regarding what systems are understood to require. Examples of concerns raised at this
level include the integrity of individuals or communities, the openness and good sense of
argument, and the obligation of individuals or groups to respect and/or to care for one another.
Unlike more concrete considerations, these general ones have a more universal quality. Once
defined, these general functional requirements of personality or community constitute the basis
for judging the value of a specific interaction, preference or norm to the person or group in
question. Thus while recognizing that a person creates the meaning and value of a specific
preference she expresses, more general claims can still be made regarding the effect of the
preference on the integrity of that person or on the quality of the relationships she may have with
other people. In this manner, the relativistic consideration of the particular values expressed by an
individual may be overridden by universal ethical considerations.
Some interview excerpts that exemplify systematic evaluation follow. In the first there is
the invocation of a norm of motherhood, but rather than begin treated as a given, it is
justified in terms of the condition of special dependence of a child on its parents.
Intrv.:
Subj.:
Intrv.:
Subj.:
Assume that your daughter is very sick and cannot pay for help….Would you give her the money?
Oh absolutely.
Why?
I think my relationship with Heather is very like one of unconditional love, and the second is if
you’re somebody’s mother, I think it’s as if a certain level of total loyalty that you have to them.
And it’s as much as it’s more for me an emotional connection than an ethical connection, the
ethical rule. It’s not that I think I should give her the $10,000, instead my relationship with her is
that I wouldn’t hesitate though to give it to her. I mean there have been a lot of things. I mean I
commuted for three years to law school living in Irvine, because I thought she should be more near
her father. Well that’s a very small advantage in her life and a very large disadvantage for me in
my life, and I think that was important for me to do. I think that when you have a child, they’re
very much in your care. To me, the child only really has his mother and father to be devoted to
him. And you don’t have anybody else that’s absolutely necessary yours. And, therefore, to me
there’s a very large level of loyalty that you want to offer them.
We are not the same
Page 16
The second two excerpts (the first dealing with stealing and the second with the display of
art) are further example of reflecting on a convention, putting it in the context of other conventions
and then attempting to use some higher order principle in order to prioritize conventions and
determine how and when to apply them.
Intrv.:
Subj.:
But you don’t think you should be allowed to steal either?
It depends. I mean it depends on the situation, I know that the ethical rule is, I don’t think there’s
an absolute ethical rule that… Let’s put it this way, I think there’s an ethical rule that it’s wrong to
steal, but I think the wrong of stealing can be balanced against other wrongs so that, and I don’t
know that, I mean absolute ethical rules don’t particularly interest me. I do think, I think that there
should only be rules to the extent that if everybody faces these rules, we don’t do things we would
otherwise like to do. But I think if by looking at it in a rule-free environment, then I can concoct a
rule myself rather than having an absolute rule you know laid down by law in some simple
fashion. And in that sense, my ethical rule would be that you have to weigh the harms to people.
So there’s no absolute wrong of stealing. It may be legitimate if the balance of harm is in line,
doing the action.
Subj.:
What we really have to be concerned about is first of all, artistic…Freedom of artistic expression
has to do with being able to be critical in expressing what’s happening in society. And I think we
always have to safeguard its freedom, not any particular work of art, but the freedom to have
legitimate expression and criticism. Basically, I think it’s the expression of ideas and criticisms of
society, I think that’s what free speech is all about. But we also have to realize that you know we
live in a society which has certain power imbalances, and that there are certain individuals and
groups of people who are going to be harmed because they have no power….We have to
recognize when we tromp on the less powerful and that includes blacks, gays, children, and all
kinds of other people that can’t really stand up and defend themselves. Now part of the function
of art is to take non-mainstream views, and to bring them up and say hey this is it. But if you have
an exhibit that shows the denigration of Blacks, essentially what you have is you have an
additional weapon. It’s not really an interesting expression that criticizes or illuminates society,
The last excerpt (discussing war) entails a reflection on the genesis of
conventional rules and an attempt to construct a principled basis for rule making.
Subj.:
Intrv.:
Subj.:
Subj:
In my example, I’m saying okay you’ve got an international community that says the rules of war
and that you should not resort to war of any kind, internationally or domestically, to solve your
conflicts. Therefore, war as you define war in some way is illegitimate.
So the rules of war are there are no rules?
Right. No, I’m not saying that there are, but in my example, if the international community finds
it that way, then I think it’s legitimate. Then I think it has certain legitimacy then to intervene and
stop it. If we were to evolve a level, that would be legitimate. But I think where we are right now
in our international community is that we have some sort of rules of war that say you know thou
shalt not have, thou shalt not try to wipe out an entire population, thou shalt not kill and
disproportion a number of civilian, thou shalt negotiate in good faith; you seem to have these
concepts of what’s allowable and what’s not allowable. And we feel that these have been
breached and so, therefore, we can intervene And I think if the international community feels
that’s legitimate, which I think from a pragmatic point of view, what I want to see go on in the
world is the negotiation of legitimacy.
(continuing later) I think the line you have to draw is the difference between, you know, a
significant military oppression of a group and a situation in which the military conflict is
We are not the same
Page 17
damaging a large number of people. And the situation in which a government is trying to regain
peace and stability. I don’t have any good rules about where to cross that line. And I don’t think
there are any really good rules. And you know what the difference is between revolution and
crime?...I have my own concepts about who’s the aggressor and who’s not, but again I think that
it’s something that has to be worked out in the international community.
RESEARCH METHOD
The research focuses on two hypotheses regarding the nature of political evaluation: (1)
that an individual's evaluation of very different phenomena will reflect a single underlying
structure (sequential, linear or systematic), and (2) that different individuals will evaluate the
same social and political phenomena in structurally different ways.
Subject population. Forty-eight adults participated in the study. With the aim of
exploring the full range of political evaluation, subjects were selected according to factors that
might cause or reflect differences in development. These included level of education, income and
responsibilities on the job. Three groups of sixteen subjects were interviewed. The first group
included subjects that had a high school education or less, earned less than $20,000 a year and
performed jobs that involved little planning or supervision of other workers. The second group
included subjects that had some college education, earned between $25,000 and $35,000 a year
and did have limited responsibility for short-term job planning or for supervising a few coworkers. Subjects in the third group had at least some postgraduate education, earned more than
$60,000 a year and had employment that gave them control over a number of other people or
responsibilities for long-term planning.
Data collection. Subjects were interviewed individually at length. The interviews were
taped and then transcribed. Each interview consists of several parts. To begin, subjects were
surveyed using close-ended questions in order to collect basic demographic data (on age,
ethnicity, occupation, education, etc.). Subjects were then interviewed in-depth on three topics: a
personal issue, a domestic political issue and a question of foreign policy. Discussion of the
We are not the same
Page 18
personal issue focused on a significant other chosen by the interviewee. The issue was whether
the interviewee would be willing to give up her life’s savings or steal to save that other person's
life. In the second part of the in-depth interview, the subject was asked to consider a domestic
social and political topic. The focus here was on the controversial homoerotic photography of
Robert Mapplethorpe and the attempt of a town council in a overwhelmingly conservative city to
ban a public display of his work in a local museum. The third part of the in-depth interview
focused on a foreign policy issue. The question presented was whether the United States should
continue its involvement in Bosnia. The issues were chosen not only because of their different
content, but because it was presumed that they would vary in the amount of interest they evoked
and the amount of relevant knowledge subjects had at their disposal. As such the comparison of
individual subjects’ reasoning response across these three issues provides a particularly stringent
test of the singularity of the structure of an individual’s reasoning.
The aim of the interview was to have subjects express themselves sufficiently such that a
subsequent interpretation of the quality of their reasoning and hence the meaning of their
statements would be possible. The guiding methodological presumption is that the meaning of a
simple statement or judgment is inherently ambiguous. So for example the statement, “I like the
Republican Party,” in response to the question, “Which of the two parties, Democratic and
Republican, do you like the best?” may have structurally quite different meanings depending on
the thinking of the respondent. For example, for someone reasoning sequentially the question
addresses an institution, a political party, that is not readily understood and has no clear personal
immediate reward value. Any preference expressed is likely to be momentary, one cued by a
remembered experience in which a party member was involved or a party preference was
expressed. It is likely to be unstable and have no issue content whatsoever. A product of linear
We are not the same
Page 19
reasoning, the response may reflect a categorical connection made between the party and a
positively valued group (e.g., intellectuals or evangelicals) or individual (e.g. a party leader) or
causal connection between the party and a positively valued action (e.g. taking a strong stand on
crime) or outcome. In the context of systematic thinking, parties are likely to be evaluated in
terms of sets of issues positions they express and the relative role in they play in the political life
of the country. In this regard, a party preference is likely to be less a matter of social identification
and more a reflection of the respondent’s general ideological perspective. Clearly the same
statement has different meaning in these three cases and as a result the stated preference: (1) will
have different kinds of implications for other preferences, (2) will be affected by different kinds
of conditions, and (3) will have different implications for action.
With these considerations in mind, the interview was conducted so as to lead the
interviewee to make a number of related assertions and judgments. Subjects’ evaluations of all
three issues, personal, domestic political and foreign policy, were probed extensively. (To deal
with the possibility of a total lack of information, the interview regarding the domestic and
foreign policy issues was preceded with a brief description of the relevant facts relevant. If
additional information was requested or appeared necessary, it was provided.) The probing
included the use of follow-up questions to explore the subjective meaning of the expressed values
and associated justifications. This typically involved questions such as “What do you mean by
____?” and “You mentioned that ___ is important to you. Why is it so important?”
Probing also involved the introduction of additional scenarios to challenge and thereby
clarify the subjects’ expressed values. These scenarios were framed so as to highlight sequential,
linear or systematic considerations. For example in the interview about the exhibition of
homoerotic art, a subject who favored the allowing the exhibition would be asked to consider: (1)
We are not the same
Page 20
if she saw the photographs and found them personally repulsive (sequential concern), (2) if a
fairly chosen panel of experts found the photographs to be of no artistic value (linear concern),
and (3) if an exhibition of racist art demanding Hispanics should be exhibited in a city with a
large Hispanic population (systematic concern). In the interview on foreign intervention in
Bosnia, a subject favoring intervention would be asked to consider: (1) that Bosnia is a little
country of no possible benefit to the US and intervention may place the subject herself or her
loved ones in immediate danger (sequential concern), (2) cultural prescriptions such as “We
should not be the world’s policeman” and “We should not interfere in the private business of
another country” (linear concern), and (3) that social and political values are culturally relative
and therefore any intervention by us may be no more than an unwarranted attempt to impose our
values on other people (systematic concern). Apart from a providing a means for further probing,
the aim of presenting differently structured counter-scenarios was to discover if, by the nature of
the questions posed and counterclaims made, the interviewer could lead the subject to reason at
different levels during the interview. In this manner, the introduction of these counter scenarios
provides a more demanding test of the structural consistency of subjects' evaluative reasoning.
The interview typically required two hours. The order in which the topics presented was
randomly varied from interview to interview. The interview procedure is demanding on the
interviewer, requiring that she pay attention both to the content and the structure or logic of what
is being said. In recognition of this considerable time was spent educating the interviewers in the
theory and working through several trial interviews to insure that sufficiently rich data (instances
wherein interviewees made a number of interrelated claims and the meaning of these were
sufficiently probed).
We are not the same
Page 21
Data analysis. For purposes of coding, each of the three parts (personal, domestic
political and foreign policy) of each subject’s interview was coded in a double blind fashion. In a
manner consistent with theoretical approach adopted, the coding was interpretative. It required
considering multiple examples of the specific claims a particular interviewee made and how, in
the course of making arguments or elaborating points, she related these claims to one another.
The interpretation of these examples involved a recursive going back and forth between specific
observation and general abstract characterisation. On the one hand, this involved focusing on
specific instances of claims that were made and how they were strung together in a particular
argument or description. These examples were then drawn upon to make inferences about the
underlying quality or logic of the interviewee’s reasoning and the understanding she constructed.
On the other hand, the interpretation involved making tentative assumptions regarding the quality
of the interviewee’s thinking in order to make inferences about the subjective meaning of the
particular claims she made and the quality of the specific conceptual linkages that she established
among them. The result was a bootstrapping activity in which specific observations informed
general characterizations and the general characterizations informed the specific observations.
Such an interpretative exercise requires a thorough understanding of the guiding structural
pragmatic theory and the typology of thinking styles. Coders were trained accordingly. To
supplement the training in the theory and its application, a set of coding categories was provided to
guide the analysis of each part of the interview. We emphasize that these were intended as a
supplement to, not a substitute for, the requisite theoretical and methodological training. The utility
of this interpretative coding procedure is reflected in the acceptable level of inter-coder reliability
achieved. Each part was analyzed by at least two coders who were blind to each other's coding, the
identity of the interviewee, and the interviewee’s score on other topics. Inter-coder reliability was
We are not the same
Page 22
88%. In the 20 cases of disagreement (out of 144 total), the coders met and an agreement was
reached. The coding categories are presented in Appendix 1.
RESULTS
The aim of the empirical research was to explore two hypotheses. The first is that a
general structure underlies the way in which an individual evaluates social and political issues. To
test this hypothesis, subjects were asked to evaluate three very different questions: one involving
a close personal friend, relative or lover, another involving the display of controversial art in a
conservative community, and a third involving American involvement in Bosnia. The analysis of
the transcripts indicates that forty-two or 88% of the forty-eight subjects evaluated the three
different situations in the same manner. When considering this result, it should be kept in mind
that this high level of cross-situational consistency is evident across topics about which subjects
were likely to have highly variable levels of information and interest.
The claim regarding the structural consistency of subject’s evaluations was also
supported in a second, albeit more indirect way. The reader will remember that in each section of
the interview, alternative scenarios were introduced that were specifically designed to stimulate
structurally different considerations. To the degree to which the quality of the individual
interviewee’s evaluations did not change in response to these manipulations, a single reliable
assessment of the structure of the subject’s evaluation of the situation becomes possible and
assessments of different coders should more readily coincide. This was clearly the case as
indicated by the inter-coder reliability of 88%. For 124 of the 144 separate texts (the 3 different
parts of the interview for each of the 48 subjects), coders came to a single and common judgment
of the structure of the individual subject’s evaluation despite the attempts that were made to
induce the interviewee to consider the same question in qualitatively different ways.
We are not the same
Page 23
The second hypothesis examined was that different individuals might evaluate the same
issue in structurally different ways. In this part of the analysis all 48 subjects were included. Six of
the subjects were scored at the same level on only two of the three parts of the interview. For the
present purposes, these six were assigned an overall score based on their performance on those
two parts. Our results indicate that different individuals did evaluate the same issues in
qualitatively different ways. Eleven of the subjects responded in a sequential manner, twentyseven responded in linear manner and ten responded in a systematic manner.
It is worth noting that alternative interpretations of these results are suggested by a social
learning theory. We comment on two briefly. The first suggests that differences in reasoning
could be attributed to differences in exposure to relevant information. The greater the exposure,
the more complex one’s reasoning about the issue at hand. However this cannot provide a
persuasive account of the results obtained. Although we might reasonably assume that all adults
have extensive and comparable exposure to a significant other and to the meaning of stealing, the
results indicate that when asked to reason about these very familiar matters, different adults did so
in qualitatively different ways. Moreover, in an overwhelming portion of the cases, a given adult
evaluated three very different topics, helping a significant other, displaying Mapplethorpe’s art
and American involvement in Bosnia, in the same way although we might reasonably assume
that their interest in and information on these topics varied significantly.
A second interpretation focuses on differences in formal training and suggests that those
with similar educational background would use the same language and concepts to make sense of
various issues. The evidence does suggest that training is partly related to performance on the
interviews. There is a rough, but only a rough, relationship between educational level and type of
evaluation. In other research, the matter was addressed directly. Subjects who had virtually
We are not the same
Page 24
identical training (e.g., students in the same year at the same university majoring in the same
field) were asked to perform a task related to their training. Despite these similarities in
background, there were significant differences in thinking about the task. Whereas none of the
subjects reasoned sequentially, roughly two-thirds reasoned linearly and one-third reasoned
systematically (Rosenberg 1988a; 1988b). This said, it would be important to explore the
relationship between socialization influences and cognition. A key concern would be to develop a
measure of the quality of formal training that go beyond a count of years of education.
In a different vein, some attitude research suggests that, even if our analysis of types of
judgment is correct, it is not particularly relevant to important instances of political thinking. The
key claim here is that the quality of people’s judgmental processes depends on how much
attention and effort they devote to the situation at hand. A good example of this is the distinction
made by Petty and Cacioppo (1986) between more casual, schema driven “peripheral”
processing and more coherent, logical “central processing.” Drawing on this literature, some
political scientists have suggested that given citizen disinterest in politics, most political
judgment, especially that involved in voting, would be peripheral (e.g. Marcus, Neumann &
MacKuen, 2000). While agreeing with this conclusion, we suggest that unlike the models offered
in social psychology, our analysis points to qualitative differences in “central processing” and
thus in the ability to reason when conditions foster greater attention and effort.
CONCLUSION
Adopting a structural pragmatic approach to the study of political reasoning, we defined
three types of political evaluation, sequential, linear and systematic. To test this typology, we
conducted in-depth interviews with a diverse group of 48 adults. The results of the research
provide strong support for the claims that (1) an individual's political evaluation has an underlying
We are not the same
Page 25
structure which determines her judgment of both political (domestic and foreign) and nonpolitical
phenomena (the personal decision to help a close friend or relative), and (2) different individuals
may evaluate the same phenomena in structurally different ways. Furthermore, the results suggest
that these structural differences in evaluation are well captured by the threefold typology of
sequential, linear and systematic political reasoning.
In conclusion, we consider the implications of our research for three other political
psychological approaches. We begin with the survey based research on political attitudes and
knowledge. Most of this research is predicated on the assumption that a statement of preference
has a generally common meaning, one that is anchored in the objective qualities of the attitude
object. Thus the statements such as “I strongly support George Bush for President” or “I am
against government subsidies for private education” are assumed to have meaning by virtue of
their referential connection to an objective entity, in this case the person of George Bush, or the
policy of government support for private education. With meaning thus anchored in a commonly
represented reality, the research effort focuses on what is regarded to be the more variable
affective component of the attitude, the intensity of the preference expressed. To this end, a
variety of measures have been developed including Likert scales, rank orderings to “feeling
thermometers.” When the cognitive component of an attitude is addressed, the focus turns to the
adequacy of the representation of the attitude object. Individual differences are conceived as
differences in knowledge rather than meaning and are assumed to reflect differences in exposure
to information or differences in attention or memory of the individuals exposed.
While acknowledging the value of this research, our approach suggests that greater
attention should be paid to the subjective meaning of an attitude. In our view, not only the
intensity or the valence of an attitude, but also its meaning, may vary significantly between
We are not the same
Page 26
subjects. For example, consider the assertion “I am against racism.” As constructed by someone
reasoning in a systematic fashion, racism is likely to be understood in quite general and abstract
terms, as a form of interpersonal relationship which categorizes people by their physical
characteristics and regulates social interaction to the benefit of some and the disadvantage of
others. The negative evaluation of racism will be based on functional considerations (e.g. it
creates disruptive and violent divisions within a society and therefore is dysfunctional for the
society as a whole) or on principled ones (e.g. individuals are personality systems and the
relationship between people therefore must be based on reciprocal respect for the integrity of
those systems; racism violates this standard). As a product of linear reasoning, the assertion “I am
against racism” takes on a qualitatively different meaning. Rather than a type of interpersonal
relationship, racism is likely to be understood as a concrete category of actions that are commonly
regarded as racist (e.g. making certain kinds of demeaning remarks about blacks or Jews, or
engaging in certain kinds of behaviors such as not allowing members of those groups entry into a
club or not giving them equal treatment on the job). Thus defined, racism is likely to be rejected
either because prevalent social norms dictate that specific behaviors toward particular groups are
unacceptable or because authoritative or valued other people stipulate that this is the case.
It is important to recognize that this difference in the systematic and linear construction of
the attitude “I am against racism” has clear implications for predicting the conditions under which
the attitude will be expressed or changed. In the case of the systematic reasoning, the construction
of racism and the grounds of its denial are conceived in quite abstract and general terms. The
understanding is therefore not tied to the specific circumstances under which the attitude was first
developed. Consequently the anti-racist attitude is likely to generalize the consideration of novel
interactions, behaviors or group divisions that are subjectively interpreted to be exemplars of a
We are not the same
Page 27
racist mode of interpersonal exchange. Similarly the attitude is likely to change only if the face of
argument that demonstrates that racism is consistent with maintaining a given social system or
with a principled understanding of interpersonal relations. Unlike its systematic counterpart,
linear reasoning defines racism and denies it in rather concrete and specific terms. Thus we would
expect that anti-racist attitudes would be subjectively linked to the particular terms in which they
were initially learned and therefore apply only to the original acts and groups in question. The
inclusion of novel acts and new groups would require additional learning. This type of reasoning
might account for evidence that suggests that although the intolerance of specific targeted groups
in the U.S. (e.g. blacks) has decreased, intolerance in general has not (e.g., Sullivan et al., 1982).
Our research also has implications for the research on political cognition. Like in our
approach, this research emphasizes the subjective construction of meaning and explores the
processes whereby individuals recognize and integrate the information made available to them.
Despite this similarity, our approach differs in three important respects. Following work in social
psychology, the research on political cognition is oriented by the assumption that cognition is best
understood as a complex of information processing mechanisms, including dissonance reduction
(e.g. Abelson, et al., 1968), attributional errors (e.g., Jervis, 1976), schema processing (e.g.,
Conover and Feldman, 1984) and heuristics (e.g., Quattrone and Tversky, 1988). In constrast, the
structuralist dimension of our approach suggests that there is a general mode of cognitive
functioning that underlies an individual’s various cognitive strategies and structures the various
specific definitional, evaluative or problem-solving strategies that person employs. This suggests
that rather than studying attributional, dissonance or other cognitive mechanisms in isolation,
research should also focus on how the different mechanisms may be related to one another.
We are not the same
Page 28
Second, and in complementary fashion, our structural pragmatic approach does not adopt
the orienting assumption of most political cognition research that all people cognize in basically
the same way. Thus we do not conclude that all individuals are equally likely to reason in a
heuristic driven, dissonance distorted and attribution-flawed manner. Instead we suggest that
there are qualitative differences between individual in their general mode of cognition and this
will be reflected in qualitative differences in the specific information processing mechanism they
employ. Thus we hypothesize that the evidence of cognitive dissonance and attributional errors
do not describe the reasoning of all individuals, but rather reflect the linear reasoning
characteristic of the lower division college students typically used in those experiments. In this
vein we would further hypothesize that neither sequential nor systematic reasoning would be
likely to generate these kinds of judgmental or inferential mechanisms.
Third our structural pragmatic approach suggests a more socio-historically relative
conception of political cognition. Reflecting the individualist and biological biases of the
cognitive psychology upon which it draws, much of the work on political cognition assumes that
the information processing mechanisms studied are characteristic of the species and thus typical
of all individuals in all societies. Whereas the substance of what people consider will certainly
vary with the socio-historical circumstance, the basic nature of how they process information will
remain the same. Viewed from a structural pragmatic perspective, this psychological
universalism fails to recognize the full significance of the interplay between structures of social
exchange and structures of cognitive functioning. Structural pragmatism emphasizes that the
social context determines the nature of the social tasks the individual must perform to effectively
interact with others and thereby stimulates the development of cognitive functioning. Insofar as
different social environments are more or less complex and therefore differ in the quality of the
We are not the same
Page 29
social tasks they set, we expect they will tend to foster more or less cognitive development and
therefore the populations inhabiting these different environments will evidence significant
differences in the basic quality of their cognitive functioning.
Finally we turn to the research most closely related to our own, the developmental
psychological research on the social and moral reasoning such as Kohlberg’s work (1981/4) on
moral reasoning, Turiel’s work on social reasoning (1983) and Kegan’s work on adolescent
psychopathology (1994). Our research clearly builds on the insights of this other work. In
addition our evidence of cognitive structure and individual differences is consistent with the
results of this other research. That said, our research differs in three respects. First, our work has a
different substantive focus. Whereas Kohlberg focuses on moral reasoning and Turiel and Kegan
focuses on social reasoning, our research addresses specifically political reasoning and issues. In
this regard, our research complements their work in developmental psychology.
Of greater theoretical interest, our analysis offers a more formal definition of the structure
of each type of reasoning. Following the example of Kohlberg’s analysis of moral reasoning,
many of developmental psychologists (Kegan is an exception) differentiate types of reasoning in
a way that conflates the underlying structure of a way of reasoning and the substantive
considerations to which that form of reasoning gives rise. In our view, this explains their tendency
to characterize reasoning in a domain specific ways (e.g. reasoning about physical objects, social
reasoning, moral reasoning, etc.). Lacking a concept of structure that is independent of content
and an understanding of how structure and content then operate on one another, Kohlberg and his
students lack a theoretical basis for the domain distinctions they make. In a prima facie way, the
problem of distinguishing domains of thought becomes evident when we consider the difficulty
social scientists and philosophers have defining criteria for differentiating social, moral and
We are not the same
Page 30
political phenomena. In more strictly cognitive developmental terms, the problem with stipulating
distinctions among cognitive domains on substantive grounds becomes evident when we
remember that the construction of these kinds of distinctions is itself a cognitive developmental
achievement. (Sequential reasoning does not produce nor sustain these distinctions. Linear
thinking may operate to sustain these distinctions if they are prevalent in the culture, but cannot
generate them on its own. Only systematic thinking can construct these content domains and use
them to guide judgment.) We have defined types of reasoning more formally and therefore can
more readily theorize about cognition in a way that is free of content and therefore applies across
substantive domains. This is evident in the present effort in which we hypothesize and provide
supporting evidence that the individuals’ judgment will reflect the same underlying structural
qualities whether they are evaluating a relationship with a significant other, the display of
controversial art in a city museum or the conduct of American foreign policy.
Finally we adopt a more explicitly social psychological perspective. Although Piaget,
Kohlberg and scholars following in their tradition recognize the importance of the environment as
a necessary stimulus to development, relatively little theoretical attention has been paid to the
analysis of those environments or to the interplay between them and the cognition of the
individuals who inhabit them. For example, Piaget does little more than acknowledge that society
is necessary for the development of formal operations and in so doing seems to assume that all
social environments are essentially equivalent. Taking the more social psychological and sociohistorically relative position of the neo-Marxist psychology of Vygotsky (1962, 1978; Wertsch,
1991), we suggest that theoretical attention be paid to the existence and cognitive consequences
of differences in social structures of interaction across and within societies. We argue that these
social structures do penetrate how we think in a way that offers an alternative guide to cognition.
We are not the same
Page 31
This suggests that insofar as societies (or groups within a society) are structured in different ways,
the ways of thinking most prevalent among the actors in these different settings will also be
structured differently.
To conclude, the structural pragmatic perspective suggests a rather distinctive approach
to political psychological research, one that helps capture the complexity of political reasoning as
an activity that is structured both by cultural definitions and social institutions on the one hand
and by subjectively constructed understandings and purposes on the other. The present research
has focused on the psychological dimension of political life by attempting to characterize the
different forms of subjective judgment. Future research must examine the implications of these
differences for political behavior. Suggestive examples of this are provided by Golec’s (2002)
study of politicians’ approach to political conflict, Wilson’s (1992) analysis of Chinese cultural
development and Reykowski’s (2007) political psychyological studies of democratization. Work
on the relationship between moral reasoning and political activism is also suggestive (e.g., Haan,
Smith and Block, 1968; Tygart, 1984; and Muhlberger 2000). Future research should also
directly explore the interplay between social context and subjectivity posited by the theory. For
example, the current interest deliberative democracy with its assumptions about the reasonable
nature of group discussions and the rational nature of subjective reasoning offers fertile ground
for the empirical study of the interplay between culture and institutional contexts on the one hand
and the cognitive capacities of individual participants on the other.
We are not the same
Page 32
REFERENCES
Abelson, R.P.; Aronson, E.; McGuire, W.J.; Newcomb, T.M.; Rosenberg, M.J.;& Tannebaum, P.H.
(eds.) (1968). Theories of Cognitive Consistency: A Sourcebook. Chicago: McNally.
Bearison, D.J. and Zimiles, H. (1986). Thought and Emotion: Developmental Perspectives. Hillsdale,
NJ: Lawrence Erlbaum.
Braunwarth, J. (1996). “Cognitive Reasoning and the Representation of Reality: The Role of the
Television News.” Paper delivered at the Annual Meeting of the International Society of
Political Psychology, Vancouver, Canada.
Chilton, S. (1988). Defining political development. Boulder, Colo. : L. Rienner Publishers
Converse, P.E. (1964). The Nature of Belief Systems in Mass Publics. In D. Apter (ed.), Ideology and
Discontent. New York: Free Press.
Conover, P. and Feldman, S. (1984). "How People Organize the Political World: A Schematic
Model", American Journal of Political Science, 1984, pp.95-126.
Damon, W. and Hart, D. (1988). Self-understanding in Childhood andAdolescence. New York:
Cambridge University Press.
Dupont, H. (1994). Emotional Development, Theory and Applications: A Neo Piagetian Perspective.
Westport, CT: Praeger.
Golec A. (2002). Cognitive skills as predictor of attitudes toward political
conflict: A study of Polish politicians. Political Psychology, 23, pp. 731-759.
Geertz, C. (1964). “Ideology as a Cultural Symbol System.” In D. Apter (ed.) Ideology and
Discontent. New York: Free Press.
We are not the same
Page 33
Haan, N., Smith, M.B. and Block, J. (1968). Moral reasoning of young adults: Political-social
behavior, family background and personality correlates. Journal of Personality and Social
Psychology, 10, pp. 183-201.
Habermas, J. (1979). “Moral Development and Ego Identity.” In Communication and the Evolution of
Society. Boston: Beacon.
Habermas, J. (1984/7). The Theory of Communicative Action: Volumes 1 & 2. Boston: Beacon.
Hanks, C. (1998). “The Psychology of Principled Tolerance.” Paper delivered at the Annual Meeting
of the International Society of Political Psychology, Montreal, Canada.
Heider, F. (1958). The Psychology of Interpersonal Relations. New York: Wiley.
Hogan, D. M. and Tudge, J. (1999). “Implications of Vygotsky’s theory for peer learning.” In A.M.
O’Donnell and A. King (eds.) Cognitive Perspectives on Peer Learning. New Jersey:
Erlbaum.
Hogg, M.A. and Abrams, D. (1988). Social identifications: A social psychology of group intergroup
relations and group processes. New York: Routledge.
Inhelder, B. and Piaget, J. (1958). The Growth of Logical Thinking from Childhood to Adolescence.
New York: Basic books.
Jervis, R. (1976). Perception and misperception in international politics. Princeton: Princeton
University Press.
Kegan R. (1994). In Over Our Heads: The Mental Demands of Modern Life. Cambridge: Harvard
University Press.
Kohlberg, L. (1981/4). Essays on Moral Development: Volumes 1 & 2. NewYork: Harper & Row.
Luria, A. (1976). Cognitive Development: Its Cultural and Social Foundations. Cambridge, MA:
Harvard University Press.
We are not the same
Page 34
Marcus, G.E., Neuman, W.R. & MacKuen, M.B. (2000). Affective Intelligence and Political
Judgment. Chicago: University of Chicago Press.
Mead, G.H. (1934). Mind, Self and Society. Chicago: University of Chicago Press.
Muhlberger, P. (2000). Moral reasoning effects on political participation. Political Psychology, 21,
pp.667-695.
Nicolopoulou, A. and Weintraub, J. (1998). Individual and collective representations in social context:
a modest contribution to resuming the interrupted project of a sociocultural developmental
psychology. Human Development 41-215:235
Noam, G. and Fischer, K. (eds) Development and vulnerability in close relationships. Mahwah, NJ:
Lawrence Erlbaum, pp. 91-113.
Petty, R.E. & Cacioppo, J. (1986). Communication and persuasion: Central and peripheral routes to
attitude change. New York: Springer-Verlag.
Piaget, J. (1970). Structuralism. New York: Harper & Row
Piaget, J. (1971). Biology and Knowledge: An Essay on the Relations between Organic Regulations
and Cognitive Processes. Chicago: University of Chicago Press.
Quattrone, G. A. and Tversky, A. (1988). “Contrasting rational and psychological analyses
of political choice.” The American Political Science Review 82 (3): 719-736.
Reykowski, J. (2007). “ .” Political Psychology **************************************
Riker, W. (1995). The Psychology of Rational Choice. Political Psychology, 16, pp. 3-21.
Rosenberg, S.W. (1988a). The Structure of Political Thinking. American Journal of Political Science,
32, pp. 539-66.
Rosenberg, S.W. (1988b). Reason, Ideology and Politics. Princeton: Princeton University Press.
We are not the same
Page 35
Rosenberg, S.W. (2002). The Not So Common Sense: Differences in the Way People Judge Social
and Political Life. New Haven: Yale University Press.
Rosenberg, S.W.; Ward, D. and Chilton, S. (1988). Political reasoning and cognition: A Piagetian
perspective. Durham, NC: Duke University Press.
Sullivan, J.L., Piereson, J.; and Marcus, G.E. (1982). Political tolerance and American democracy.
Chicago: University of Chicago Press.
Turiel, E. (1983). The development of social knowledge: Morality and convention. Cambridge, UK:
Cambridge University Press
Turner, J.C.: Oakes, P.; Haslam, S.; McGarty, C. (1994). “Self and Collective: Cognition and Social
Context.” Personality and Social Psychology Bulletin 20:5, pp. 454-463.
Tygart, C.E. (1984). Moral autonomy and socio-political activism among the faculty and staff of a
West Coast university. Sociological Inquiry, 54, 16-25.
Valsiner, J. (1992). "Social Organization of Cognitive " In Demetriou et al. (eds) Neo- Piagetian
Theories of Cognitive Development. New York: Routledge
Vygotsky, L.S. (1962). Thought and Language. Cambridge, MA: MIT Press.
Vygotsky, L.S. (1978). Mind in Society: The Development of Higher Psychological Processes.
Cambridge, MA: Harvard University Press.
Wertsch, J.V. (1991). Voices of the Mind. Cambridge, MA: Harvard University Press.
Wilson, R.W. (1992). Compliance Ideologies: Rethinking Political Culture. New York: Cambridge
University Press.
Yanow, D. (2003). “Accessing local knowledge: policy analysis and communities of meaning.”In M.
Hajer and H. Wagenaar (eds). Deliberative Policy Analysis: Understanding Governance in the
Network Society.” Cambridge: Cambridge University Press.
We are not the same
Page 36
APPENDIX 1 – CODING CATEGORIES
The following coding categories were provided as guided guidelines that were used in
conjunction with exemplary excerpts of the kind included in the text. Given the requirements of the
interpretative method, the categories and excerpts were not stand alone devices, but were used in
conjunction with training in the theoretical approach and its empirical research application.
(1) The interview section focusing on stealing to help a valued other.
SEQUENTIAL. The focus is on sequences of events and the feelings they evoke. Evaluation is
guided by these reactive, circumstantial and egocentric feelings. Thus stealing may be judged
negatively because it leads to a sequence of being shot at, being caught, going to prison and feeling
miserable. Alternatively, stealing may be evaluated positively because it involves in saving another
who “I love.” Probing reveals that this is a matter the positive rewards the other confers (“she does
nice things for me or “makes me feel good”). The sequential nature of the evaluation is also
reflected in the consideration of how a “bad” sequence of events may be repaired by returning
events to their original position. This might involve suggesting that the stolen drug be returned or
that the druggist just make some more. An empathic confusion of the other's feelings and one’s own
in a given scenario may lead to evaluations based on what is reinforcing for that other person. Thus
another person’s pain or fear may consequently lead to a decision to steal. Occasionally the
language of higher order (linear or systematic) concerns may be used, but the rhetoric is understood
in sequential terms. For example, reference may be made to the role imperative of a husband to a
wife. However if that the relationship is dissatisfying and the other person is no longer liked, such
role concerns evaporate in the face of negative reinforcements. Similarly the issue of “fairness” may
be invoked, but in sequential reasoning this simply means “getting what I want.” Justifications for
evaluations are typically not offered spontaneously. In response to direct questioning, the subject
may respond by simply reasserting the initial claim or reverting to saying it is just my preference.
LINEAR. The focus is on specific actors and concrete observable actions that are evaluated as
desirable, good or correct. This may involve a establishing a causal connection between the act and
valued outcome. Thus the act of stealing that leads to saving a significant other’s life may be judged
to be a good act. Similarly a person who chooses to do a bad act, stealing, may be evaluated as a
bad person. Actions, actors and outcomes may be also evaluated by the category to which they
We are not the same
Page 37
belong. Where judgment leads to contrary evaluations of the same objects, an attempt is made to
render them consistent by mitigating the significance of either the positive or the negative
considerations. Thus following the evaluation of the choice to steal as a good act, the introduction of
negative considerations (such as the harm to druggist) are mitigated (the harm is minimal) and
additional positive benefits may be introduced. Linear evaluation may also introduce moral
considerations by invoking concrete standards of how one should specifically behave in the
circumstances in question. Here appropriateness is determined by reference to a particular cultural
convention, role requirement or normal practice. Thus, the act of stealing may be negatively valued
because it entails breaking the law. Alternatively, it may be positively valued insofar as it conforms
to the norm that friends should help one another. Typically only one convention can be considered
at a time, thus the introduction of an alternative may effectively change the frame of reference and
thus lead to a change of judgment. These conventions or normal practices are the bases of
justification rather than its object. The rhetoric of higher order concerns may be introduced, but
these are understood in linear terms. Thus a concern for justice may be voiced, but this is typically
equated with doing what is socially correct, that is following an accepted conventional procedure
for allocating rewards (e.g., an eye for an eye, or taking turns).
SYSTEMATIC. The activity of stealing is seen as one trajectory of activity or one interaction
interwoven among several others including helping another person, placing oneself at risk and
depriving another of her property. At this first level of consideration, actions may have multiple
effects and be assessed complexly - that is as having positive and negative ramifications. These are
assessed relative to the personal preferences of the people involved and the social and legal
conventions that apply. At a second level, these various dimensions of the interaction are judged by
their systemic consequences. Thus, the act of stealing or the failure to steal may be evaluated in
terms of its effect on the integrity and coherence of the personality or self-definition of the thief or
her beneficiary. Alternatively, the needs of the community, as an organization and regulator of
social life, might be invoked. In this context, the theft may be judged by its implications for the
functioning of the community as a cultural as well as a social system. The law governing theft may
be judged similarly. These pragmatic considerations may be complemented by moral ones that
revolve around the relations between systems - either at the same level (individual to individual) or
at different levels (individual to community). Here general principles of human sociality are evoked
and then applied deductively to the judgment of particular situations and social conventions. For
We are not the same
Page 38
example, this may involve an assumption of individual integrity and mutual obligation leading to
the view that breaking the law by stealing is legitimate under certain conditions. The issue of
fairness may also be constructed in these principled terms.
(2) The domestic political interview section - displaying art that violates community values .
SEQUENTIAL. The focus is on the unfolding story of the exhibition of Mapplethorpe’s work and
judgment is made in terms of the feelings evoked or the desires satisfied or frustrated. Although
removed from immediate experience, the story may evoke feelings if it cues the subject's own
affectively charged past experience. The painting will accordingly be judged in these egocentric
terms. For example, the exhibition may be negatively valued because other photos of
homosexuality have evoked personal disgust in the past. Alternatively if the subject empathizes
with the observed or imagined feelings of characters in the story (through a confusion of one’s own
feelings with the other's feelings), she may evaluate the exhibition with regard to how it makes
others feel. Thus the exhibition may be negatively valued because it leads to people being upset or
hurt. The labeling of the displaying of the art as “good” or “bad” may also affect evaluation of the
overall sequence. Thus, the exhibition may simply be regarded as “bad” because it involves photos
of homosexuality and the latter is “obscene” or “wrong.” It is not necessary to understand why it is
“bad,” it is just labeled as such. Evaluations are not normally justified and the request to justify a
preference typically just leads to a reiteration of the initial claim. Because sequential reasoning
involves playing out sequences of events, questions and an initial response that cues the elements of
other sequences leading the subject away from the initial issue being addressed. Judgment here will
tend not to focus on concerns such as the community or social norms. These are little understood or
valued. When they are mentioned, it is merely a matter of repeating slogans.
LINEAR. The specific actions and actors presented in the interview are evaluated in terms of their
linear causal or categorical links to other actions and actors (which may be remote in space and
time) that have already been evaluated as preferable (or not) or social desirable (or not). For
example, the exhibition may be negatively valued because it causes reactions or results that are
unwanted (e.g. people will withdraw support from the museum) or because it is categorically
associated with objects (e.g. art) or people (e.g. homosexuals) that are negatively valued. Claims of
correct behavior may also be introduced. These might include the notion that artists have the right to
express themselves, people have the right to choose what they see, the decision regarding
We are not the same
Page 39
exhibitions is the proper preserve of the museum director, people ought to be exposed to new
things, etc. Cultural mores, social traditions, moral imperatives and law all enter in here on an equal
footing as conventions regarding correct performance of roles and rituals. These judgments tend to
be global and univalent. For example, once the photographs are judged to be bad because they
depict a bad act, they may also be judged artistically worthless or aesthetically unattractive. Because
they focus on specific action relations, these global judgments may be altered when the interviewer
introduces alternative foci (e.g., the effects on other actors or the relation to other cultural norms).
The ensuing value conflict is confusing and will often lead to some form of dissonance reduction by
bolstering or diminishing one or the other conflicting consideration.
SYSTEMATIC. The focus is on the various dimensions of the problem in light of the multi-sided
relationships between the director, the artist, the city council and the public. At the same time, there
is a clear recognition of the place and needs of the community as the system that both regulates
social exchange and defines social meaning and value. The preferences of the various players and
particular social conventions are regarded as first order concerns, ones that have significance, but
must be judged in light of second order considerations. For example, the artist's desire to show his
photographs or the prevalent cultural norm of freedom of expression will be evaluated in terms of
the broader consequences for the integrity and functioning of the artist or the community. When
applied to varying circumstances, these overarching values may lead to very different judgments of
the same activity enacted in different systemic contexts. Thus, the exhibition of photographs may be
deemed appropriate and functional in the context of a multicultural progressive city with a
significant homosexual population, but unnecessary and destructive in homogeneous, traditional
city. A number of functionally valued considerations may be considered simultaneously (e.g.
freedom of speech, the need to maintain equality) in judging appropriate rules of action and how
they should be applied. There is also a self-reflective recognition that the subject’s response
constitutes one possible evaluation of the situation and thus judgments tend to be more tentative and
perhaps conditional on interaction with others.
(3) The foreign policy interview section - American intervention in Bosnia.
SEQUENTIAL. Typically international affairs are too removed to evoke much understanding or
judgment. If they do, the focus is on specific isolated concrete sequences of unfolding events seen
on television or vividly narrated by others. Such a sequence of events may be judged consequential
to the degree to which it is associated with pleasurable or painful sensations or feelings. The ensuing
We are not the same
Page 40
confusion of other-event-self may yield an empathic response. Thus being able to observe an
American soldier's joy at hitting a target or a mother’s pain at the maiming of her child on television
may evoke comparable feeling in the observer. A sequence of events may also evoke feelings of
another sequence that was personally experienced as rewarding or painful. For example, a
description of Serbia “bullying” the Bosnian Muslims may evoke a memory of a sequence in which
the subject was herself bullied. The two sequences may be subjectively fused and the international
events may be judged accordingly. An evaluation may also be expressed if the expression of the
evaluation is itself embedded in a pleasurable sequence. For example, if other people have rewarded
the subject for condemning the Serbs, then she will do so even if she does not fully understand who
the Serbs are or what they did. Again preferences are not typically justified and requests for
justification will tend to lead to storytelling that often leads off topic.
LINEAR. The focus is on action and actors as discrete units of analysis to be evaluated as good,
desirable or correct. These concrete objects are evaluated in three ways. First, they may be judged in
light of the already evaluated causes and outcomes with which they are associated. Thus
intervention may be negatively valued because it will cost money and lives or because other nations
will regard us negatively as a result. Second, actions and actors may be evaluated by their
categorical associations. For example, American intervention may be regarded as a good thing
because the U.S. is a good country and therefore the things that it does are good. Similarly, the
Serbs are viewed as part of the group of Russian allies. If this group is negatively valued, the Serbs
will be also. Third, actions are evaluated by the degree to which they conform to concrete
behavioral norms. For example, American intervention in Bosnia may be negatively valued
because it is linked to the norm that one should not meddle in other people’s private business. Once
made, linear evaluations tend to yield global, unidimensional evaluations of the actors or actions
considered. This overall evaluation is often sustained even in the face of specific contrary evidence
by reinterpreting, diminishing or simply ignoring the information being considered. For example,
the Muslims, already regarded as a good people, may be observed to engage in a bad act (torturing
Serbs). To maintain a consistent evaluation, this act may be redefined as a good act (it helped stop
Serbian aggression) or Muslim responsibility for the act may be reduced (the circumstance forced
them to do it).
SYSTEMATIC. The focus is on interactions, conventions of behavior, and cultural values and
definitions. Objects of evaluation would thus include the mutual aggression of Serbs and Croats, the
We are not the same
Page 41
convention that Europe should look after affairs in their own “backyard,” or the value of national
sovereignty. On the one hand, these are evaluated by examining them in relation to the functioning
of political systems. For example, the systematic thinker might evaluate the conflict in Bosnia in
terms of the stability of political systems in the region. In this light, the assertion of Serbian power
could be judged to have a disintegrating effect on the stability of the area and be judged negatively.
Alternatively, American involvement might be evaluated in terms of its consistency with the
cultural values of the United States. Thus involvement might be seen as an affirmation of the
American way of life and judged positively. On the other hand, interactions and conventions may
be judged by the application of general principles of how systems ought to interrelate. Thus, the
idea of assassinating the Serbian President might be judged to be an effective military option, but
ultimately condemned because it does not conform to a principle of international exchange based
on free and open negotiation oriented to establishing cooperation.