Thomas Jordan

Structures of geopolitical reasoning

Outline of a constructive-developmental approach

[Published as: Occasional Papers 1998:9, Kulturgeografiska Institutionen, Handelshögskolan, Göteborgs Universitet.]


1. Introduction


In this paper I propose an actor-oriented theoretical and methodological framework for studying the political geography of violent conflicts. More specifically, my objective is to develop a research strategy for studying the structures of geopolitical reasoning among political leaders, political activists and ordinary citizens.

Much can be said for the relevance, even urgency, of this topic in the contemporary world. Most violent conflicts involve geopolitical strategies and territoriality (i.e. behaviour aiming for establishing and defending territories against outgrips). [1] Recent examples include the civil war in former Yugoslavia, the violence in Chechnia, the issue of Jewish settlements in the Israel/Palestine conflict, and the Hutu/Tutsi atrocities. Many of the violent conflicts raging in the world have proved very resilient to conflict resolution efforts. While we often can find historical, economic, socio-political, and religious explanations to the conflicts, it seems that the costs in terms of lost lives, injuries and social and material destruction far exceed any reasonable expectations the parties might have of the benefits of using violence. It is therefore important to explore how the parties to such conflicts perceive the issues involved, the conflict process, themselves, and their counterparts.

The general hypothesis of my approach is that it is possible to identify several distinct modes of reasoning, differentiated by certain generic qualities of cognitive operations, or consciousness structures. If such differences exist, they are bound to have major implications for how geopolitical problems are perceived and interpreted, as well as for how individuals and groups perceive their identities, their interests, and their alternatives of action. In other words: a particular mode of consciousness may be associated with a typical mode of geopolitical reasoning. This would imply characteristic constructions of geopolitical interests, of the ingroup ("We"), of outgroups ("the Others") and of the relationship between them, as well as characteristic ways of perceiving the possible outcomes or strategies in a conflict of interests. The general question I would like to explore is consequently: What is the relationship between general modes of consciousness and specific patterns of geopolitical reasoning? The chosen approach for analysis is derived from constructive-developmental psychology. My hope for this study is that the theoretical framework I adopt will permit me to identify and analyse something we might call "deep structures" of geopolitical reasoning. The theoretical assumptions regarding consciousness structures and consciousness development will be discussed in depth in section 2. The approach I propose might yield results of immediate relevance for conflict resolution strategies in violent social conflicts by offering insight into how the conflict parties reason, and how a third party might create a favourable environment for the evolution of modes of reasoning favourable to conflict transformation and/or conflict resolution (see also Jordan, 1997b). In a more theoretical perspective, I would like to put forward the controversial suggestion that the perspective presented here has the potential of providing a platform for formulating a normative political geography [sic!]. A drastic, but seriously meant way of formulating a central question for a normative political geography is: What would an "enlightened" geopolitical strategy look like?


A note on the character of this paper

This paper is explicitly written as a part of my own explorative process of formulating a distinct research topic and a tenable research strategy. I use this opportunity to test out different ideas and to explore alternatives. I will give an outline of different theoretical and methodological alternatives, and discuss their advantages and disadvantages, limitations and strengths. At the same time I provide the reader some access to my way of reasoning about this topic, e.g. which kinds of reflections are most important to me in formulating a more precise framework for empirical studies. I would warmly welcome constructive criticism (in fact, any criticism) and suggestions. [2]

When I started this explorative endeavour, my thinking used the notion of consciousness structure as a cornerstone. I had read a lot about cognitive structures à la Piaget, and consciousness structures à la Gebser or Wilber. Consequently, I aimed at developing an adapted framework specifying consciousness structures as they might be encountered in the arena of geopolitical action. Surveying the literature, I found a number of theoretical frameworks that seemed relevant. I started to distill the relevant essence out of these frameworks, always probing for what could be of use in analysing geopolitical reasoning. However, during this work I gradually came to call my basic assumptions into question. Is it really very productive trying to specify well-defined consciousness structures which could be expected to show up neatly as coherent and consistent wholes in interviews? [3] Postulating the existence of discrete and consistent consciousness structures is a very vulnerable strategy, inviting all kinds of counter-evidence. There is a certain neatness in the assumption of a limited number of basic consciousness structures, making the notion attractive, but in the final analysis not much is won even if the effort to find evidence for such structures would receive some support in empirical studies. Consequently my approach to the whole field started to shift: away from coherent structures, toward specifying relevant dimensions of the development of consciousness systems. Such a framework does not require empirical complexity to be squeezed into a severely limited number of rigidly defined consciousness strucutres, but is open to various combinations of characteristics. I believe that it is productive to conceive a theoretical framework that is as open-ended as possible, while retaining as much stringency and precision as possible. By specifying a number of relevant dimensions for consciousness evolution (e.g. perspective-taking, scope of self-definition, ability to handle abstract notions, self-reflexiveness) rather than assuming the existence of a small number of well-delimited stages of over-all consciousness development, we are still invited to explore how different dimensions of consciousness development relate to each other. Are there close correlations between stages of development of cognition and of self-sense or not?

The structure of this paper is still very much determined by my initial approach, and I regard it as notes made during a journey, rather than as a finished report. I did not find it worthwhile to rework the entire study to de-emphasize consciousness structures, in order to focus on dimensions of consciousness development. The slightly uneasy fit between formulations of consciousness structures, and the subsequent disassembly of these structures into a number of dimensions of development is perhaps instructive in its own right. Or so I hope.


Theoretical backgrounds

This section describes my intellectual background, and thus provides a more explicit setting for later sections. However, if you want to jump directly into the main course, you can skip this part.

The perspective I present below draws on discourses developed within four different academic fields: political geography, international relations, cognitive-developmental psychology, and conflict resolution. I have become familiar with these traditions in slightly different ways.

I first met the field of political geography during my period as a doctoral student, through a post-graduate course on geopolitics. Political geography is a rather heterogeneous field, covering the analysis of global configurations of power, nation–state issues, boundary studies, the geography of political elections and much more. The approaches used range from descriptive studies, quantitative analyses in the positivistic tradition and model-building, to the world-system framework of Peter Taylor (1993). Recently, some scholars have started to explore the implications of new social theory for political geography, e.g. by discussing the role of social constructions of national identities (see e.g. "’Tuathail, 1996; Paasi, 1996). However, very few studies in political geography are based on an actor-centered approach that analyzes the reasons for the behaviour and strategies of the individuals and groups active in conflicts. Taylor’s influential approach uses systems analysis, with a focus on the political and economic processes inherent in the world-system (Taylor, 1993).

The core of the geographic perspective is putting the unique, real-life geographic manifestations at the centre of the analysis. Geographers normally insist on analyzing how various phenomena are actually situated on the earth’s surface, whether on a local, regional, national or global scale. Analyses solely dealing with abstract or disembodied entities (e.g. the mind, the rules of the world order, models of economic systems) can seldom pass off as geography. Geographers insist that situatedness in actual geographic constellations on the earth’s surface makes a crucial difference. This is a valuable point, I think, and my ultimate aim is to avoid separating my analysis of geopolitical reasoning from the actual and unique context in which persons and groups form their strategies. The word "ultimate" means that few traces of this aim will be found in the present study, which is focussed on a generic theoretical and methodological framework.

A review of the political geography literature is beyond the scope of this paper, since the main topic is exploring ways of adapting the cognitive-developmental approach to geopolitical problems. At a later stage I shall have to relate my approach to the mainstream approaches of political geography. However, in this study current theorizing in political geography will play a minor role.

My acquaintance with international relations stems from my involvement with the Peace and Development Research Institute, Gothenburg University (PADRIGU). During the period 1985-1994 I was a guest lecturer at PADRIGU’s course in International Relations, teaching a course on the geopolitics of the global raw materials supply system. I also participated regularly in the research seminars at PADRIGU, and developed close links to some of the researchers there. PADRIGU focussed on the North-South issues, developing a strong inclination for the world-system theory of Immanuel Wallerstein and his associates. Some of the researchers at PADRIGU found the world-system approach too abstract, and started to explore more action-oriented approaches to peace studies, in particular conflict resolution theory and methods. This led to the establishment of a praxis-oriented undergraduate course in Conflict Resolution, of which I was one of the co-founders, and the first director. In the course of planning his course, I made a thorough survey of literature on conflict resolution. Several different traditions exist in this field. One large group of scholars have a background in law studies. This group concentrates on developing alternatives to litigation, i.e. by developing new negotiation strategies (e.g. Fisher & Ury, 1981) or methods for alternative dispute resolution (ADR, see e.g. Goldberg et al., 1992). A second group, with a background in organization studies, focusses on conflict resolution in organizations (e.g. Glasl, 1997). A third group, with a background in psychology, focusses on communication and dealing with emotions in conflicts (see e.g. Volkan, 1988; Rosenberg, 1983; Mindell, 1992). A fourth group is made up of peace and conflict researchers, focussing mostly on the resolution of international conflicts (e.g.; Burton, 1990; Azar, 1990; Wallensteen, 1994).

My acquaintance with the field of international relations is important for this study as a general frame of reference. Even though the approach outlined here focusses on the operations of the individual’s consciousness, the ultimate aim is to use insights gained from this perspective to better understand international relations. The conflict management literature provides a wealth of insight into the dynamics of conflicts, not least from the subjective point of view. The work of Friedrich Glasl (e.g. 1997) has been particularly important for me, especially his detailed analysis of conflict escalation processes.

The field of cognitive-developmental psychology has been my major field of interest during the last few years, and I have been in the lucky position of being able to spend a lot of time reading the literature of this field (e.g. Kohlberg, 1969 and 1971; Selman, 1980; Loevinger, 1976; Neumann, 1949; Kegan, 1982 and 1994; Wilber, 1980, 1995, 1997; Wilber et al. 1986; Arieti, 1967; Gilligan, 1982; Torbert, 1987; Fisher & Torbert, 1995; and Schroder et al., 1967). Cognitive-developmental psychology, whose "founding father" was Jean Piaget, focuses on how the modes of reasoning of the individual change in the course of psychological development. A number of methods for investigating cognitive structures have been developed, e.g. analysis of semi-structured interviews (Kohlberg, Selman, Kegan, Rosenberg) and text analysis (Loevinger, Schroder et al., Tetlock, Suedfeld). Since this perspective is the focus of this study, I will leave further considerations to later sections.

I conceive of this study as a contribution to political geography, where the research issues are formulated in the traditions of political geography and international relations, whereas the theoretical and methodological approaches are borrowed from developmental psychology and conflict theory.


2. Theoretical frameworks

The cognitive-developmental approach

The cognitive-developmental approach relies heavily on the pioneering research of Jean Piaget, who started publishing in the 1920’s. Piaget sometimes called his endeavour "genetic epistemology", i.e. the study of how understanding, or meaning-making, develops in the human individual from birth to adult age. [4] Piaget asked questions about how cognition is structured, and how the structure of reasoning changes as the individual develops. In order to study this issue, Piaget talked to children of different ages, observed their behaviour, and invented and carried out a number of ingenious experiments in order to test how individuals reason about different topics. On the basis of this research, Piaget formulated a theoretical model comprising four main stages of the development of cognitive structures (sensorimotor, preoperational, concrete operational and formal operational reasoning). These structures were conceived by Piaget as consistent wholes, giving a characteristic and stable pattern to all kinds of reasoning of a particular individual, whether in physical, social or psychological spheres. An important aspect of Piagetian theory is the assertion that all individuals go through the same cognitive structures in the same sequential order. Extensive research, including cross-cultural studies, supports the notion of universal developmental stages, but does not support Piaget’s assumption about strict structural consistency. [5], [6] Piaget concentrated his efforts on studying children, and assumed for a number of reasons that all individuals reach the same end-stage (formal operations) during their teens. Later research has indicated that this assumption was premature. Several stages of cognitive development have been identified in adult age (see below), and extensive evidence indicates that a large number of individuals never develop the highest stages.

As is evident from the introduction above, cognitive-developmental psychology focuses on the operations of the individual mind. This implies that even if a particular set of ideas has become dominant in a specific society, we would expect individuals operating at different cognitive structures to understand these ideas in quite dissimilar ways. One good example of this is a study by Dana Ward (1988) in which he asked a number of subjects about their understanding of the concept "democracy." Ward could demonstrate that "democracy" had structurally different meanings to people reasoning at different cognitive structures. However, this focus on the individual mind raises a number of questions about the relationship between the individual and the collective in the analysis of social processes. Many social scientists insist that discourses (e.g. about national identities) are social constructions, implying that the psychological processes of the individual can have no decisive role as an explanatory factor. [7] I will postpone further exploration of this interesting issue here, because we first need a comprehensive understanding of cognitive-developmental frameworks.

An extensive survey of the literature on cognitive-developmental psychology has left me with four related theoretical frameworks which seem relevant for my purposes. [8] These are:

(i) the integrative complexity framework,

(ii) Rosenberg’s structures of political reasoning,

(iii) Kegan’s subject/object framework, and

(iv) a variant of the consciousness structure framework derived from Habermas and Wilber.

The former two follow Piaget in looking at the mind "from the outside," i.e. analysing the structure of reasoning. The latter two take a more decidedly phenomenological perspective, in exploring how it is to be a person with a particular consciousness structure (i.e. the experienced identity). [9] I will present the first three of these frameworks in order of increasing complexity. These three rely on well-defined theoretical specifications, offer research methodologies, and have been empirically validated. The fourth framework is more or less speculative, and is offered as a heuristic model only.

The first framework presented below, the integrative complexity framework, is only partly derived from Piaget’s theory. The framework was developed in the 1960’s, and got its most comprehensive statement in a book by Harold Schroder, Michael Driver and Siegfried Streufert (1967). During the 1980’s and 90’s the framework has been developed, modified and used in empirical studies on political psychology issues in a large number of publications, in particular by Philip Tetlock and Peter Suedfeld and their associates. [10] The integrative complexity framework focuses on one particular aspect of cognitive development, the cognitive integration of competing perspectives. While less sophisticated than the frameworks of Rosenberg and Kegan, the integrative complexity framework is stringently formulated, intelligible to people less familiar with psychological theory, and comparatively easy to use in empirical studies.

The second framework I will draw on is Shawn Rosenberg’s characterization of three structures of reasoning: sequential, linear and systematic (Rosenberg, 1988). Rosenberg is a political scientist with a thorough understanding of Piagetian theory. He has revised Piaget’s cognitive-developmental framework in order to accommodate for the interactions of social and individual processes. Rosenberg, working in Piaget’s tradition, assumes that there are a number of general structures of reasoning. The basic characteristics of a person’s structure of reasoning is therefore expected to generate similar types of thinking in different spheres, e.g. in the physical, the social, the political, and the psychological worlds. Rosenberg’s and Kegan’s frameworks are largely compatible, although there is a distinct difference in the perspective taken on cognitive development.

As the backbone of my theoretical framework I have chosen to draw on the work of Robert Kegan, a Harvard psychologist. Kegan has provided a consistent and penetrating elaboration and extension of Piaget’s theory, the Subject/Object framework. Kegan’s framework provides an in-depth analysis of the most basic characteristic of consciousness structures, with the potential to provide a powerful theoretical platform for exploring various research issues. However, the profundity of his theory is at the same time a limitation, in the sense that it is difficult and takes a long time to gain a thorough understanding of the theory. A further limitation is that Kegan’s framework mainly has been applied to understanding personal relationships (as opposed to inter-group relationships). [11]

The fourth framework is a sketch of three different consciousness structures based primarily on work by Jürgen Habermas and Ken Wilber. While Kegan offers a stringent and well validated theory, its implications on the macro-social level have yet to be explored. Habermas (1976), drawing on Piaget and Kohlberg, made an effort to use cognitive-developmental theory to delineate a theoretical model of social evolution. His model is little more than a sketch on the co-evolution of cognitive structures and societal forms, but formulates some potentially important ideas. Wilber (1981) used a similar approach to explore the role of consciousness structures in the history of cultures, but tried to integrate a broader range of theories into his model, including buddhist and psychoanalytical perspectives.

In the following four sections I will present the four frameworks in their own contexts. In each section I will briefly comment on how the frameworks might be applied to geopolitical reasoning. However, in order to develop a suitable research strategy, the relevant dimensions of each framework have to be extracted and integrated into an analytical framework adapted to the specific research issues I have in mind.


The conceptual/integrative complexity framework

The main focus of the integrative complexity framework is the extent to which individuals cognitively relate different perspectives to each other. This may apply to competing opinions, ideologies, theories, interests, or any other case when different viewpoints contradict each other. There are two quite dissimilar approaches to conceptual/integrative complexity:

According to conceptual complexity theory (Harvey, Hunt, and Schroder 1961; Schroder, Driver and Streufert 1967), differentiation and integration are stable personality traits of cognitive style. Integrative complexity theory emphasizes differentiation and integration as aspects of information processing that vary not only among individuals but also from situation to situation for each individual […]. (Guttieri et al., 1995)

Unfortunately the term "integrative complexity" has been used both to designate the structural and the situational interpretation, which means that a clear terminology is missing. Below, I will start presenting the original formulations, and thereafter review some of the themes in recent research.

Seen from a broader perspective, the integrative complexity framework only addresses one of many elements of cognitive development. There is no direct analysis of the actual structure of reasoning (as by Rosenberg), neither of the construction of the self (as by Kegan). However, the ability to cognitively handle contradictory perspectives is an important element in cognitive development, and might possibly be regarded as something of a proxy for consciousness development in general. Later research has shown that such a contention is very problematic, and I will return to this issue after presenting the original formulations.

Schroder, Driver and Streufert (SDS) defined four sequential levels of integrative complexity, representing increasing abilities to handle conflicting perspectives (Schroder et al., 1967). Below I have summarized the most salient characteristics of the four levels. [12] The description draws mainly on Schroder et al. (1967), with some minor complementary elements drawn from Harvey et al. (1961) and from comments by Tara Santmire [13] (personal communication). The actual method for assessing a person’s level of integrative complexity (the Paragraph Completion Measure) is presented in section 4.


1. Low integrative complexity

Information is interpreted and evaluated according to simple fixed rules. No alternative interpretations are considered. The low level of cognitive differentiation means a very restricted capacity for maintaining ambiguity or tolerating conflicting interpretations. Information that does not fit into existing patterns of evaluation is ignored or rejected. The values, opinions, assumptions, etc. of the individual are not systematically related to each other, but are compartmentalized in the mind. Mutually contradictory cognitive elements may be entertained simultaneously without awareness of contradiction. Information is evaluated according to yes-no/black-white categories, where little allowance is made for gradual differences. Persons reasoning at this level believe that there are right and wrong solutions to problems. There is little room for uncertainty, the individual consequently feels very certain about his/her evaluations. At this level, the propensity is high that individuals react to uncertainty with aggressiveness, thereby warding off a dilemma that can’t be resolved cognitively. Individuals at this level also tend to rely on and refer to external authorities supporting a certain stand.

The simplicity of information processing means that once a piece of information has been categorized in a certain way, it stays there. The room for reinterpretation is extremely restricted. New information is mostly assimilated into existing categories, or is completely rejected. The categories themselves are seldom accommodated to fit experience better. Reasoning at this level tries to get rid of cognitive conflicts as quickly as possible. Information contradicting established interpretations is reinterpreted to fit these patterns or is rejected. Held beliefs are therefore highly resilient to change.

The weakness of interpretative faculties leads to rather stereotyped reaction patterns. Certain types of information lead more or less automatically to certain fixed conclusions. The behaviour is therefore to a high degree controlled by external conditions, with a small degree of freedom to adapt to changing circumstances.

A limited capacity to differentiate categorizations creates sharp limits between categories. Within a category, individual elements are seen as similar. If a person is assigned to a particular group, e.g. the ingroup, a set of stereotyped attributes is attributed to this person. As long as this categorization persists, all information that indicates deviation from this set of attributes is rejected or reinterpreted. However, if a certain threshold is reached, the categorization shifts abruptly, and the person is attributed to another category, with an inverse set of attributes.

Faced with conflict, the low level perceives the positions of the conflicting parties as an either/or problem: the perspectives are (implicitly) irreconcilable.


2. Moderately low integrative complexity

At this level, the individual can recognize the existence of different perspectives, opening up the possibility of different interpretations of the same informations. This ability reduces the determinate character of cognitive processes and evaluation, and more flexibility in the interpretation of certain perceptions. However, the different perspectives remain compartmentalized, they cannot be related to and integrated with each other. The individual is embedded in either one or another set of interpretations in an either-or fashion. One can develop rules for when to use a particular perspective rather than another, but when the choice has been made alternative perspectives are no longer considered. This means that the categorizations or conclusions made are not modified by reflection according to the alternative perspective. Since the different perspectives are not integrated with each other within a holistic framework, a certain arbitrariness in the choice of perspective may result. A tendency to reason probabilistically is typical of this level: outcomes can be either this or that way according to some estimation of probability. However, in this level there is no consideration of the joint outcomes of different perspectives.

In relation to the low complexity level, this level implies the possibility of choice: shall I choose this or that way of interpreting my information? This possibility of choice undermines evaluations according to a simple good-bad dichotomy, since the evaluation is relative to the perspective one chooses. The decreased propensity to evaluate alternative viewpoints according to a strict good/bad categorization also reduces the tendency to react aggressively when confronted with different perspectives. The reliance on external authorities also decreases, since the individual develops cognitive procedures to choose among different modes of interpretation.

Faced with conflict, the moderately low level can perceive that there might be legitimate reasons for both positions, but they are still seen as irreconcilable.


3. Moderately high integrative complexity

The third level of cognitive complexity is characterized by the development of an ability to compare different perspectives with each other. This means that even though one particular interpretative perspective is chosen, the implications of other perspectives can be considered as well. The ability to consider different viewpoints simultaneously greatly increases the flexibility in interpretations and evaluations. A person at this level can therefore be open to relativations and revisions of cognitive decisions, according to different points of view.

The second level permitted the existence of alternative interpretations, but the third level allows for varying combinations of several perspectives simultaneously. Conflicting aspects of different interpretations are not avoided, but are used to gain further understanding.

A most important consequence of third level cognition is the ability to consider the influence of behaviour according to one perspective from another perspective, and adjust interpretations and evaluation according to this insight. One can take the role of another, and imagine how this person perceives a situation. For example, one can reflect on how one’s own behaviour is interpreted by another person, and adjust one’s behavior in order to reach a desired outcome.

The ability to use different principles of interpretation increases the freedom to make decisions, which in turn increases the sense of being an intentional being with a high degree of responsibility for what happens.

Faced with conflict, the moderately high level is capable of assessing the the effects of one position on the other, and can therefore adjust the resolution of the conflict by considering both sides. This can take the form of strategical games or compromise solutions.


4. High integrative complexity

In the fourth level, thinking is based in hypothetical reflection. In the third level, different interpretations of the same set of information could be employed. In the fourth level the principles of different cognitive perspectives are generated mentally, and these perspectives are used as guidelines in searching for new information. Different perspectives can not only be used in conjunction with each other, but they can also be integrated within a holistic framework where their functional relations to each other are specified.

Paradox is not avoided, but can be appreciated as an accurate representation of real experience. For example, both autonomy and dependence can be seen as desirable goals simultaneously and without contradiction. Furthermore, paradoxes and incompatibilities among perspectives are actively used to derive new insights and frameworks.

Faced with conflict, the high level is capable of considering different perspectives as an interacting system. This allows a creative search for solutions to conflicts that satisfy both parties through a rearrangement of their mutual relationships.


Integrative complexity applied to geopolitical reasoning

Can we derive predictions from the integrative complexity framework about what to expect from an analysis of geopolitical reasoning? Or, in more general terms, what contribution could this framework make towards an analytical framework for studying geopolitical reasoning and destructive territoriality? As a preliminary hypothesis I would formulate the following expectations:

A person reasoning with a low level of integrative complexity would tend to:

• Construct stereotyped images of ingroup and outgroups.

• Attribute collective properties to members of the groups.

• Argue in simplistic terms about geopolitical problems, only drawing conclusions from one relevant dimension.

• Reject the need to consider the perspective of the counterpart in geopolitical conflicts.

• Derive preferred solutions to geopolitical problems from an exclusive ingroup perspective.

• Be highly susceptible to the values, opinions, and strategies adopted by perceived ingroup authorities.

• Be unable to take a critical stand on conventional ingroup values.

• Take ingroup values for granted, and regard them as universal and inherently right in a moral sense.

• Attribute great importance to rigid boundaries separating inside from outside.

• See solutions to geopolitical problems in either/or terms (win-lose).

A person reasoning with a moderately low level of integrative complexity would tend to:

• Recognize that an outgroup has a right of existence and a right to have a different perspective.

• Reason in either/or terms, or in terms of compromises.

• Be unable to think creatively about developing integrative solutions to geopolitical problems.

• Be unwilling to consider non-conventional ideas about solutions to geopolitical problems.

• Reason and act strategically in order to gain advantages in strategical games (orientation to bargaining).

A person reasoning with a moderately high level of integrative complexity would tend to:

• Be able to consider the perspective of the counterpart in geopolitical conflicts, and to calculate with the mutual influence and reactions from both parties.

• Be prepared to reconsider conventional strategies and solutions in the light of new information.

• Be interested in the perspective of the other party in conflicts in order to arrive at reasonable solutions.

A person reasoning with a high level of integrative complexity would tend to:

• Explore many different points of view before making decisions about strategies and actions.

• Have differentiated and flexible images of ingroup and outgroups.

• Take many different dimensions of the relationship between ingroup and outgroups.

• Have a high level of tolerance for diversity.

• Search for creative solutions to geopolitical problems.

In general, the integrative complexity framework draws our attention to the complexity of reasoning about geopolitical issues, to the nature of ingroup and outgroup images, to the preparedness of considering the other party’s perspective, and to the propensity to consider non-conventional proposals. It does not, however, address a range of other geopolitically relevant issues, e.g. the scale of identity groups or the affective attitudes to boundaries.


Integrative complexity as a structural or as a situational variable

SDS conceived the level of integrative complexity as a durable characteristic of a person, implying that one and the same individual consistently operates from one particular level. Subsequent research has shown that an individual’s level of integrative complexity can vary considerably over time (see e.g. Guttieri et al., 1995; Tetlock, 1981; Suedfeld, 1994; Suedfeld & Tetlock, 1977; Suedfeld et al., 1993; Wallace et al., 1993). One of the most salient reasons for such variations is the level of stress in the environment (e.g. in decision-making during political crises). Suedfeld and Tetlock have therefore largely abandoned SDS’s conception of integrative complexity as a personality trait, and favour a contingency perspective. The question of structural consistency vs. situational determinance is both complex and important, and of course I will not be able to give a satisfactory final solution to the dilemma here. Perhaps the most reasonable attitude is to regard the structure vs. situation issue as an unresolved and interesting question for further empirical research. I will return to the consistency hypothesis after having presented the three other theoretical frameworks.

Suedfeld et al. also point out, with some emphasis, that there is nothing inherently desirable in high integrative complexity as such: [14]

[… ] the integrative complexity approach is not a normative or prescriptive model. In our view, neither high nor low complexity is more moral, more perceptive, or more likely to lead to successful solutions to international (or other) problems. [… ] Complex approaches require a greater investment of time, energy, emotion, and sometimes even material resources (e.g., to collect more information). The result may be a better understanding of the problems and various possible solutions; it may also be information overload, self-contradiction, and confusion. Simple strategies are more economical, and may lead to crisp, neat, rapid solutions. (Suedfeld et al., 1993:185)

That simplicity may be superior to complexity is an important point, but I think it must be considered in a wider perspective. Recent research on integrative complexity focuses on the analysis of speeches and texts generated by parties in conflicts, where the researchers look for signs that the subject considers and integrates the perspective of the counterpart in his/her reasoning. There is no analysis of the structure of reasoning, nor of the character of the "epistemological self" (Kegan, 1994; see below). If such concerns are included in the analysis, the case for normative statements regarding cognitive structures is strengthened considerably. However, this issue cannot be discussed further without a presentation of the other frameworks.


Rosenberg’s reconstruction of Piagetian theory

Shawn Rosenberg is a political scientist who has made a serious attempt to reconceptualize Piaget’s theoretical model of cognitive development in order to adapt it to the study of political phenomena (Rosenberg, 1988). He criticizes Piagetian theory for having an excessive focus on the individual, and a naive conception of the role of the social environment in the construction of cognitive structures. In his delineation of three different modes of reasoning, Rosenberg describes the general structures of reasoning as well as the implications of these basic structures for how the individual interprets political events and structures. His main concern is to create a theoretical framework for analysing how different persons perceive, evaluate and act in the realm of politics. The framework has been applied in studies of people’s understanding of domestic American politics, the US-Iranian hostage crisis in 1979/80, and the US bombing of Libya in 1986 (Rosenberg, 1988, ch. 5).

I have summarized below the salient aspects of the three structures of reasoning Rosenberg describes: sequential, linear and systematic reasoning (Rosenberg, 1988, ch. 4-5). In contrast to Kegan, Rosenberg exclusively focuses on adults, and he does not extensively discuss the process of cognitive development. Since all references are made to the same book (Rosenberg, 1988), I only give the page numbers in parentheses.


The sequential thinker

The reasoning of the sequential thinker is bound to the concrete, to the world as it immediately appears. The reason for this is that the sequential thinker has not developed an abstract mental world of generalized concepts and relations (103). That which can be observed, e.g. a political event, is not interpreted as an example of generalized categories. The sequential thinker cannot interpret concrete events by reflecting on how they fit into a general order of things. The absence of a mental field for exploring possible relationships precludes the use of principles to evaluate events. The world appears to the sequential thinker in the form of a collection of concrete events, persons, and circumstances with little coordination. To the extent that inferences are made, they are bound to concrete particularities not extended to generalized conclusions. The notion of causality, e.g. that events are caused by necessary and sufficient preconditions, does not play a salient role in the sequential mind. Events transpire, without much interpretation of how they come about. The attention is occupied by one item at a time, and there is little spontaneous effort to relate them to other items or to a general context (104).

The sequential thinker is not really aware that the world may appear differently to other people, and he or she has therefore a limited ability to take the perspective of others. The limited ability to hold abstract categories in the mind means that the sequential thinker does not operate with categories such as class or nationality (105). The time horizon is limited, since there is no understanding of long-term processes or systemic properties. Thinking about the past or the future is mainly restricted to the circular character of daily, weekly or seasonal routines (107f).

People are constructed as singular isolated entities, but they are not seen as active agents shaping their own fates. Rather, they are perceived as appendages to observed events (109). When interpreting political events or the behaviour of conflict parties, the sequential thinker does not extrapolate to consider what might have occurred beyond the immediately visible (111). Politics is understood in a person-centered fashion: leaders are held to be responsible for the state of affairs. The interpretations of sequential thinkers will lack consistency and understanding of long-term and systemic relationships.

When reasoning about politics, the sequential thinker focuses on particular actors and present or very recent events (163f). Politics is understood as concrete interactions, and there is no sense of durable relationships, or a general context in which concrete events are situated (163). The actors do different things, but the sequential thinker has no conception of the underlying causes, relationships, or structures that regulate and influence these activities. When talking about politics, the sequential thinker offers descriptions rather than explanations (175). Membership in political entities, such as countries, is thought of in very concrete terms: "Citizenship is defined in terms of doing something concrete and specific for the country. A good citizen does something which is helpful - he or she does the country a favor - much as one person would do for another." (176)

Rosenberg conducted a special study on how international conflicts are understood (185-187). The sequential thinker observes events as they are passing, but has little sense of the strategical aims of the parties to conflicts. Even if (immediate) aims are understood, the sequential thinker is unable to cognitively coordinate the interests of the parties, i.e. consider both sides simultaneously. He/she does not perceive the enduring properties of the relationship between the parties. Conflicts are therefore seen as depending on concrete interactions, and are regarded as transient (185). The inability to consider underlying relationships and hypothetical scenarios means that the sequential thinker does not consider alternative courses of action (186). He/she does not spontaneously think about conflict resolution, but tends to see resolution as that which eventually happens: the active party gets what it wants or does not get it (187).


The linear thinker

The reasoning of linear thinkers predominantly analyzes sequences of events (116). From observations of concrete events, the linear thinker constructs mental representations of cause-and-effect relationships. This generates an ability to reason hypothetically, to mentally analyze possible casual relations. However, these relations tend to remain linear and unidirectional. When two factors are seen as causally related to each other, the linear thinker assigns priority to one, and regard the other as dependent (117, 123). Even though the linear thinker is capable of hypothetical reasoning, the generalizations remain close to the concrete. Possible events may be conjectured by extrapolation and analogy. However, he or she does not construct a set of abstracted principles that can be used to critically evaluate the existing reality. The accumulated experience of how things use to be comes to be regarded as "normal" in the sense of how things ought to be (118). Linear reasoning does not involve highly abstract entities, such as generalized moral obligations or social systems. Therefore the linear thinker is unable to view concrete beliefs and values as expressions of a specific culture (121).

Causal relationships are understood as one factor acting upon another, not as a result of a system of interrelated factors, mutually determining each other (132). This means that persons, events, and other phenomena are not systematically related to each other. Connections between them are understood if they can be observed, explained by others, or imagined in concrete terms.

In considering ingroup-outgroup relationships, the linear thinker does not construct a general context (e.g. a political system; a world order) defining both groups (123). The restricted mental space for constructing hypothetical scenarios and general principles constrains the ability to take a critical stand on conventional beliefs, values and structures (124).

Political events are seen as caused by actors driven by their own internal motivations, rather than as consequences of a complex social system where individual drives, role expectations and systemic interrelationships interact. The cause-and-effect reasoning engenders an image of politics as a game where actors compete for implementing their own wishes, or as a hierarchical structure of command and obedience (125f).

When reasoning about politics, the linear thinker considers causal relations and organizational structure, although mostly in a cause-and-effect, unidirectional fashion (163). Political organizations are understood as hierarchical structures where control flows from the top downwards. There is little understanding of the constraints on the power of leaders. Groups may be regarded as differentiated political structures, but it is difficult for the linear thinker to conserve this image of differentiation when thinking about interactions between groups. In such interactions organizations are viewed as an homogeneous entity, speaking with one voice (163). Politics is interpreted as a cause-and-effect game, like a tennis match (164). The linear thinker looks for causes of political events (and foresees consequences of political actions). However, he/she tends to be satisfied when one likely explanation is found, and does not spontaneously look for complex causality (175).

Citizenship is thought of as a set of duties of the citizen in relation to his/her country. In contrast to the systematic thinker, the linear thinker does not develop a sense for how the political environment shapes the individual, or how the individual participates in shaping the political life (176).

International conflicts are interpreted in terms of parties that strive to realize incompatible aims. There is a tendency to see the parties as homogeneous entities involved in a bipolar opposition. In contrast to the sequential thinker, the linear thinker situates the conflict in an enduring relationship, where strategies, loyalties and oppositions may persist over time (185). Thinking about the causes of conflicts, the linear thinker tends to see one party as responsible and the other side as reacting (186). The ideal conflict resolution is when both parties are satisfied, but this is regarded as unlikely, due to the existence of incompatible aims. Therefore, the linear thinker regards domination of one party over the other as a more likely outcome of conflict processes (187).


The systematic thinker

The systematic thinker constructs a mental conception of the general rules or the system that regulate the particular. Concrete events are regarded as expressions of general principles (137). The versatility of the systematic mind allows the individual to compare the actual reality to many hypothetical possibilities, which means that the conventional wisdom might be critically scrutinized using abstract principles. The systematic thinker use both observation and deductive reasoning in order to make sense of the world (139).

The broad and versatile mental context constructed by the systematic thinker allows him/her to evaluate propositions or relationships from many different perspectives, taking both actual and hypothetical aspects into consideration. The systematic thinker can "induce the general from the particular, define those generalities in their own abstract terms, and then use them as a basis for redefining and constructing particular events" (139). In contrast to the linear thinker, who reasons in terms of one factor acting upon another, the systematic thinker can freely consider reciprocal relationships. The basic unit of reasoning is no longer separate entities, but a complex of mutual relationships. The system of relationships is here the primary unit of analysis, not the particular elements of the system.

Rosenberg also characterizes systematic reasoning as dualistic, since there is a difficulty in reconciling the general and ideal conceptions derived from deductive reasoning with the particular and real conceptions derived from observations of the concrete environment (142). However, he does not indicate how this dualism might be resolved, even though he hints at the possibility of a structure of reasoning beyond the systematic level (235).

In contrast to the linear view of politics as a game or a hierarchical structure, the systematic thinker sees politics as regulated by collective rules, norms and expectations, and the political system as a complex web of mutual relationships (144). Individuals are not seen as separate entities driven by one-dimensional motives, but as complex personalities with many internal and external loyalties. The thoughts and behaviour of an individual is seen as expressions of a distinct personality, rather than as free-standing properties (147). Political organizations, e.g. governments, are viewed as parts of a larger system in which they have a particular role. The systematic thinker has a differentiated image of such organizations, and can consider the roles, reciprocal relations, and mutual influences of various levels, groups, and individuals simultaneously (163). He/she can also recognize the constraints on the exercise of power, e.g. the need for leaders to attend to the perceived legitimacy of their position (164). Politics is seen as highly influenced by the properties of the political culture and institutions in which actors are embedded (175). For the systematic thinker, the workings of the political system as a whole is a variable which can be influenced and can influence events. He/she spontaneously considers multiple causes and multiple consequences of political events (175). Citizenship is regarded as a reciprocal relation, where the citizens must adhere to certain norms in order to reproduce the political order, but where also the political system must be responsive to the needs and concerns of the citizens (176).

Reasoning about international conflicts, the systematic thinker situates the events in the context of national and international systems (164). The conflict is interpreted as an expression of the properties of the international political system. The parties to the conflict are seen as complex entities, where internal conflicts may play a role for the course of events. The conflict itself can be interpreted using a multidimensional perspective, where the relations between the parties may have a different character depending on the dimension one considers. Hence, parties to a conflict on a specific issue may at the same time be allies when dealing with another issue (185). The most desirable form of conflict resolution is when the parties can reach a mutual understanding, perhaps by revising the very order regulating their relationship. Win-lose outcomes are likely to be regarded as unfortunate and unstable, since the losing party will tend to regard the solution as illegitimate (187).


Rosenberg’s model applied to geopolitical reasoning

Rosenberg’s framework is more obviously applicable to analysis of geopolitical reasoning than the other frameworks considered. However, he has focussed on how political events are understood by ordinary citizens, rather than on the structure of reasoning of the makers of international politics. However, if Rosenberg’s characterizations of the three types of reasoning are summarized, we might expect the following patterns of geopolitical reasoning:

A person operating with sequential thinking would tend to:

• Consider only recent and separate events

• Interpret geopolitical problems in terms of concrete interactions

• Be unable to consider causal relationships beyond the immediately observable sequences of concrete events

• Be unable to consider long-term consequences of events, e.g. of conflict resolution proposals

• Have undifferentiated and transient images of the parties involved

• Be unresponsive to identification with generalized national identities

• Tend to focus on single persons, e.g. leaders

• Be unable to consider structural or long-term constraints on the power of leaders, and would there fore tend to hold leaders responsible for the state of affairs

• Be unable to construct a consistent conception of the motives and strategies of a counterpart in a conflict.

A person operating with linear thinking would tend to:

• Consider chains of cause-and-effect events

• Be embedded in the ingroup perspective, with little ability to take a decentered perspective on geopolitical conflicts

• Develop durable ingroup and outgroup images

• Have an understanding of concrete role expectations, e.g. on leaders

• Focus on concrete interactions

• Be embedded in conventional norms and interpretations, with little ability to reflect on the over-all system

• Be satisfied with one reasonable explanation of events or situations, no consideration of multiple causality

• See one party as offender and the other as victim

• Think about resolution of conflicts in terms of dominance and subservience (win-lose)

A person operating with systematic thinking would tend to:

• Consider complex (systemic) causal relationships

• Construct differentiated ingroup and outgroup images

• Be able to consider constraints on the room of maneuver of leaders

• Think in terms of abstract factors, such as "culture," "legitimacy," "ethics," etc.

• Think about geopolitical conflicts as multi-dimensional relationships

• Consider the role of the systemic context in which the parties are embedded

• Be able to envision changes in the way the systemic context operates

• Be able to observe geopolitical problems from a decentered perspective (analysing the conflict using generalized principles)

Rosenberg’s framework focuses on the structure of the units of reasoning and on the structure of reasoning itself. It might therefore have relevance in studying how people make meaning of events, and how they reason about alternative courses of action. Since it does not explore the nature of people’s identities, however, it does not make any predictions about motives and feelings people might have in geopolitical crises. Rosenberg’s framework might therefore contribute to an understanding of how people interpret specific situations and how they might react as a result, but it has little to offer in explaining initiatives taken without external prompting.


Kegan’s subject/object framework

Robert Kegan’s framework draws heavily on Piagetian cognitive-developmental theory, but also goes beyond Piaget in one important way. Kegan does not restrict his analysis to the structure of reasoning peculiar to different stages of cognitive development, but focuses on the core identity from which a person constructs her/his meaning out of various experiences: "Kegan’s theory is about the general structure of the perceiving mind. The theory focuses on how our evolving "epistemological self" structures our experience" (Steiner, 1996:7). Whereas Rosenberg tries to capture the general structures of reasoning, Kegan draws our attention to the implications of the general structure of reasoning for how a person conceives her/himself (or more precisely what a person is) in various stages of development. In order to do this Kegan uses the notions of what is "subject," and what is "object" to a person, where: "‘subject’ refers to the [person’s] basic principle of organization, and ‘object’ refers to that which gets organized" (Lahey et al., 1988:13). The central question of the subject-object framework is: "[… ] from where in the evolution of subject-object relations does the person seem to be constructing his or her reality?" (Lahey et al., 1988:10). In one introduction to Kegan’s framework the subject/object distinction is put in this way:

‘Object’ stands for that which we consciously may know or have as our experience in all its ‘physical, social, and personal’ manifestations. ‘Subject’ stands for the part of us which does the experiencing, the knowing, the constructing. Epistemologically, for Kegan, the ‘self’ is an organizing principle. Whatever epistemology the self is at a given time, the self cannot reflect upon that way of constructing at that time. As we develop, that which was ‘self’ or ‘subject’ gradually becomes ‘object.’ (Steiner, 1996:22).

Kegan asserts that the subject-object relationship (or the consciousness structure) undergoes a series of reconstructions in the course of a person’s psychological development from infancy to maturity. Kegan defines and describes five subject-object balances, or "orders of consciousness," the first of which evolves during the second year of life. The infant has, according to Kegan, no subject-object relationship, since it is not able to consciously operate upon any object. [15] The infant cannot cognitively differentiate between subject and object, and is fully embedded in reflexes of sensing and moving. When the first order of consciousness evolves (normally during the second year of life) an object world is created in the infant’s mind. The infant becomes able to operate upon the reflexes (walk, talk, play), but remains embedded in the perceptions and impulses. Somewhere around the sixth year the second order of consciousness evolves, based upon the emergence of durable categories. The external world becomes populated by objects with durable characteristics, and the self identifies with needs, preferences and a self concept. This allows the child to subject perceptions and impulses to conscious control, e.g. controlling immediate impulses in order to attain longer-term desires. Most people develop the third order of consciousness during the teens, subjecting egocentric desires to a shared reality of values, roles and relationships. According to available evidence, only a minority of the adults in Western societies evolve into the fourth order of consciousness. The fourth order individual has developed an internal system for generating values, opinions and goals, and relies on this system rather than on the prevailing cultural canon. The fifth order, which only a small percentage ever attain, implies transcendence of the self-as-system personality. The fifth order individual is capable of taking the self-system as an object of reflection without being fully identified with it.

Kegan’s framework is difficult to explain or grasp in a few words, both because of its complexity and because it analyses the very core of our identity. I will therefore make a rather comprehensive presentation of each of the orders of consciousness, emphasizing the third and fourth order, since these are the most common ones among contemporary adults in Western societies.


The first order of consciousness

The first order of consciousness operates according to the principle of "independent elements" (Kegan, 1994:29). The self is made up of impulses and perceptions. If the self is the impulses, there is no possibility of regulating the impulses, or mediating contradictory impulses (Kegan, 1982:88). Perceptions are not linked in a systematic way, but experienced in the moment in an atomistic way. Therefore there is a characteristic lack of continuity and consequence in the thinking and feeling of the first order.

Adults operating at the first order are very rare, and are generally regarded as insane or mentally retarded.


The second order of consciousness

The crucial characteristic of the second order is the ability to perceive things, other persons, and oneself as phenomena that have durable properties, and to organize mental categories that have these durable properties. An object, say a dog, has its own properties, which are independent of my momentary perception of it. The earlier experience of the world, where anything could magically be transformed, is replaced by a concrete world of objects with durable properties. The world therefore becomes predictable and explorable. Other persons constitute a category having, for example, the property of intention (Kegan, 1994:21), i.e. they have their own purposes. This means that one can recognize that different people may have different points of view, and that one has to calculate with this fact. The self is made up of durable properties such as preferences, habits, and abilities. This enables the individual to subordinate immediate impulses to longer-term preferences and needs, generating a new sense of identity and freedom (Kegan, 1994:23). The properties observed and used to categorize things and people tend to be concrete rather than abstract. This applies also to the membership groups the second order individual constructs (Steiner, 1996:27, 145f). The emotional life becomes more related to durable wishes and dispositions than at the first level. However, the self is experienced as a set of needs, preferences, wishes, abilities, etc. which is separate from the social environment. The motivation only draws on one’s own isolated interests, not on negotiating a satisfactory relationship between one’s own and others’ perspectives.

The second order entails an increased concern for self-esteem, because the self is now perceived as something having enduring properties, while in the first order there was no notion of permanent properties (Kegan, 1994:20).

The person operating at the second stage can recognize that other people have other perspectives, but is not able to construct a durable relationship between the different perspectives (Kegan, 1994:23f). This inability means that one is unable to take care of the relationship, coordinating and mediating the perspectives in order to maintain a balanced relationship. The motivation is therefore linked to one’s own egocentric and disconnected perspective (Kegan, 1994:39). One will try to expand as far as possible, to see how far one can go without encountering resistance from other people. The perspectives of other people are not a constituent element of one’s own emotional and motivational life. One is unable to see one’s own interests, intentions, and preferences in the context of the relationships between oneself and others. There is no sense of guilt, because the perspective of the other has not yet become an internal voice (a conscience):

I am able to understand how they might feel about being betrayed, but how they will feel is not a part of the very source of my own feeling or meaning-making. [… ] Without the internalization of the other’s voice in one’s very construction of self, how one feels is much more a matter of how external others will react, and the universal effort to preserve one’s integrity will be felt by others as an effort to control or manipulate. [… ] Instead of seeing my needs I see you through my needs (Kegan, 1982:90-91).

Consequently, an important characteristic of the second order is the ability to calculate with others’ different perspectives, in combination with the inability to incorporate a care for the quality of the relationships to others into one’s own motivation. One will try to anticipate and avoid others’ negative reactions, but one will not perceive the maintenance of good relationships to others as a centre-piece of one’s own motivation. In fact, one is not able to think in terms of what happens to these relationships (Kegan, 1994:25). At the second order one is not in a position to be able to "subordinate one’s own interests on behalf of one’s greater loyalty to maintaining bonds of friendship, or team or group participation (Kegan, 1994:75). The absence of true guilt paradoxically means the absence of a shared reality. Capacity for guilt is a sign that the perspective of the other is present in the mind, and is a part of the self. This permits a person to consider the perspective of other people, not only in terms of how to pursue one’s own interest, but also in terms of caring for the relationships. The second order individual is consequently not socialized by internalizing a concern for social relations. However, at this stage the person is capable of constructing rules and roles:

[…] rules are thus constructed as exchanges with others, the same for everyone. […] But we know rules as we know measurements and categories – as objective realities, not as internalized principles or beliefs. [… ] With the new abilities we begin to join and embed in the culture as it is known outside the immediate family. As the first step, the self becomes the agent of that outside-ourselves’ order, the rules and roles of society [… ]" (Steiner, 1996:28-29).

A person at this stage might develop loyalty to a larger group, but this loyalty is subordinated to getting the self’s concrete needs met (Steiner, 1996:146).

If internal states, such as feelings, values, wishes, etc. cannot be coordinated into some kind of framework linking them to each other, it is impossible to reflect on one’s own motivation. Such a person cannot tell others about why they acted in a specific way in other terms than a narration of what concretely happened (Kegan, 1994:39).

Thinking in the second order operates with ideas, thoughts, facts, descriptions, but cannot construct abstract principles for relating them to each other in a systematic fashion. Reasoning is narrative, and uses examples without being able to formulate the general principles of which the examples are expressions. The absence of hypothetical reasoning precludes creative imagining ideals for future realization, i.e. visions of things and states beyond the concrete experiences one has made.

A very characteristic feature of the second order mind is the "embeddedness in the short-term, immediate present–a present lacking a live relation to the longer-term future" (Kegan, 1994:28). The future is not a reality one has to consider in the present, in the sense that present actions may influence one’s future situation in radical ways (e.g. that one can get aids, or that one can become a drug addict). The future is rather just a present that has yet to be experienced, it is conceived of in terms of present circumstances. Kegan puts it this way:

Concrete thinking, governed by the durable category, is simultaneously capable of holding several pieces of information together and incapable of subordinating the category of the actual to the cross-categorical realm of the possible, which is required for the construction of long-range plans, patterns, or generalizations" (Kegan, 1994:40).

Most people transcend the second order of consciousness in their teens, but a small proportion of contemporary adults still display the characteristic way of reasoning of the embeddedness in one’s own needs and interests.


The third order of consciousness

The core feature of the third order is what Kegan calls "cross-categorical coordinations," i.e. relationships between the durable categories dominating the preceding order. The self is made up of ideals, values, role conceptions, mutual loyalty, and norms, and can evaluate own needs, interests, and feeling states according to this self-conception. Reasoning at the third order can use definitions, i.e. extract the general principles that subsumes particular examples (Kegan, 1994:26). This level of abstraction introduces a space for mental constructs allowing hypothetical reasoning: a general principle might be held in mind, generating a grid of possibilities containing empty cells for facts or categories that one has not encountered in concrete experience (Kegan, 1994:359). Inference and deductive reasoning becomes possible and expands the range of reasoning enormously.

Cross-categorical experiencing in the realm of feelings implies being able to experience feelings about feelings, and the relationship between different feelings and wants. Such psychological states as "feeling conflicted," "guilty," "insecure," self-confident," and "depressed," all imply a self that is no longer embedded in the immediate experience of feelings, but a self that can relate to feelings in a larger context (Kegan, 1994:361). At earlier stages, one is subject to the feeling states, and cannot relate to them, interpret them, and do something about them on an inner level.

At the third order, one can recognize inner motivations and experience internal emotional conflicts without suppression, acting-out, or identity turmoil. Cross-categorical consciousness permits the perception of various elements of internal psychological life, in oneself, as well as in others.

Values and ideals can be constructed, and can be objects of attention and reflection, rather than just be elements of one’s self. However, although one operates with hypotheses, opinions, inferences and generalizations, one either has them or not. They cannot be put into the context of a system where they are related to each other and where they are compared and analysed. At the third order, one does not have a self-conscious system (e.g. an ideology, a coherent world-view) for generating beliefs, opinions, values, etc.

The third order self is inherently social: "The self constructs a social world relationally or mutually, rather than being embedded in one point of view at a time" (Steiner, 1996:30). The ability to construct and consider relationships opens the world of loyalty to groups and norms. The very self of the third order person is situated in the role, i.e. a self-definition determined by one’s relationship to significant persons or groups in the environment. How one appears to others, i.e. how one lives up to role expectations, is the ultimate measure of one’s identity. Disapproval related to the role expectations cannot be evaluated in terms of a personally evaluated set of standards (Kegan, 1982:201). There is no distinct self apart from the role, therefore the third order person has great difficulties in experiencing disagreements or conflicts that challenge the very nature of the role. "Being different" or "being separate" from others (or being perceived in this way by others) in the surrounding culture is profoundly threatening (Steiner, 1996:36). The third order person does not have a distinct sense of personal boundaries, and therefore often confuses who is (psychologically) responsible for what. A fourth order person can separate the self from the role, and therefore also experience being in relationship even if there is conflict. The sense of who one is at the third order is inextricably bound up to the smooth functioning of the role. The values, quality criteria, norms, etc. are derived from the group context, with little ability to take an independent position towards them (Kegan, 1994:224ff). Perceived threats to the fulfillment of role expectations of the role one is identified with are threats to the very core of the self at the third order (Kegan, 1982:102). Such threats therefore mobilize strong defenses (aggressiveness, grandiosity, etc.). The third order mind cannot construct generalized and abstract moral principles that permits evaluation and an independent stand towards the immediate group norms and loyalties (Kegan, 1982:57).


The fourth order of consciousness

The fourth order self is conscious of itself as a complex system with a certain distinctiveness and independence in relation to roles, norms and mutual relationships. The construction of the self as a (separate) system raises the maintenance of the system’s coherence and integrity as a central concern (Kegan, 1982:101). Threats to the sense of being in control of oneself are therefore perceived as really dangerous. Excessive control might be a problem for the fourth order mind.

The fourth order mind is capable of creating and operating with values and ideals, relating them to each other into a system (a personal world-view or system of belief) (Kegan, 1994:91). At this level one is no longer unconditionally dependent on the values defined by the social environment: "The fourth order self constructs its own authority" (Steiner, 1996:31). Guilt appears when one violates one’s own standards, rather than when one doesn’t live up to the expectations of others.

The third order mind is embedded in relationships, but the fourth order mind is capable of having a relationship to the relationships, enabling the fourth order person to manage incompatible claims associated with different relationships according to an overarching value system (Kegan, 1994:92):

This is a self that has gone from being identified with and made up by the trans-categorical structures of belief, value, and role to relativizing these structures, being aware of a stance toward or a relationship to these structures, that is, of having them as object rather than being them as subject." (Kegan, 1994:111).

The fourth order person can construct a conscious view of how a relationship ought to be, and evaluate ongoing relationships on this basis. She or he can also recognize and handle different levels of relating to the same person simultaneously, e.g. a private relationship and a formal relationship in an organizational hierarchy (Kegan, 1994:165). However, the limits to the fourth order is that the self is identified with the ego, and derives all meaning from the structure of the ego. There is no internal platform for transforming the ego structure, evaluating its way of functioning and setting up goals for changing the way the ego system is constituted.

The fourth order person can have negative feelings generated by the actions of another person, without necessarily perceiving the other person as responsible for creating the feelings (Kegan, 1994:121). This is only possible if one is capable of perceiving separately one’s own wishes and one’s feelings, and see the relationship between them. One is then no longer subject to one’s immediate feelings, but can relate to them, and, to some extent, manage them. Kegan’s summary of the fourth order:

This capacity [… ] represents a qualitatively more complex system for organizing experience than the mental operations that create values, beliefs, convictions, generalizations, ideals, abstractions, interpersonal loyalty, and intrapersonal states of mind. It is qualitatively more complex because it takes all of these as objects or elements of its system, rather than as the system itself; it does not identify with them but views them as parts of a new whole. This new whole is an ideology, an internal identity, a self-authorship that can coordinate, integrate, act upon, or invent values, beliefs, convictions, generalizations, ideals, abstractions, interpersonal loyalties, and intrapersonal states." (Kegan, 1994:185)

The fourth order person needs the ideology, or world-view, to have a platform for defending the personal integrity from claims arising out of roles and relationships. The identification with this system of meaning is so intimate that its corroboration and defense is a matter of maintaining one’s core identity. From this perspective, one might understand the animosity towards outgroup members, i.e. people who do not share one’s meaning system.


The fifth order of consciousness

The fifth order is characterized by the ability to relate systems to each other, see how they fit together. The self is identified with a decentered perspective, where the own personality as a system can be seen as a part of a larger whole, by which it is also created and transformed. Persons are regarded as systems that are partly determined and transformed by the relationships they have with each other. The internal world can be perceived as several systems which may be in conflict with each other, without perceiving these conflicts as a threat to the coherence of the self. Paradigms, philosophies, and theories can be seen as complementary perspectives. Incompatibilities between the perspectives can be used for gaining further insights and create new frameworks.

Among the central features of this new way of thinking seems to be a new orientation to contradiction and paradox. Rather than completely threatening the system, or mobilizing the need for resolution at all costs, the contradiction becomes more recognizable as contradiction; the orientation seems to shift to the relationship between poles in a paradox rather than a choice between the poles" (Kegan, 1982:229).

The experience of having a platform outside the ego as a coherent system decreases the need to have control. This might be experienced as a sense of flow and spontaneity (Kegan, 1982:231). Another consequence is that one can take responsibility for the undesirable aspects of one’s personality without feeling that critique of these elements threatens the very self. The motivation shifts from preserving the integrity of the ego, towards transforming the ego system (Kegan, 1982:247). The fifth order self also exhibits characteristic ways of relating to other people:

The capacity to coordinate the institutional permits one now to join others not as fellow-instrumentalists (ego stage 2) nor as partners in fusion (ego stage 3), nor as loyalists (ego stage 4), but as individuals–people who are known ultimately in relation to their actual or potential recognition of themselves and others as value-originating, system-generating, history-making individuals. The community is for the first time a "universal" one in that all persons, by virtue of their being persons, are eligible for membership. (Kegan, 1982:104)

Only a very small percentage of the population ever develop the full fifth order consciousness.


The Subject-Object framework applied to geopolitical reasoning

Extrapolating from Kegan’s framework we could expect to find a number of characteristic aspects of geopolitical reasoning at the four different orders of consciousness.

A person operating at the second order of consciousness could be expected to:

• Be embedded in his/her own needs and interests, and therefore instrumentalize other people to satisfy own needs.

• Derive motivation from concrete and palpable interests. Abstract notions like nation, loyalties, rights, justice, etc. may be exploited to justify and satisfy own interests, but mean little to the self-definition.

• Be susceptible to be motivated by experiences giving immediate gratification in the form of pleasurable feelings (e.g. experiences of personal power, wealth, sexual gratification).

• Be able to calculate with the consequences of other actors’ interests and perspectives, but not to derive own concern from the feelings of others.

• Be motivated by a drive to conserve and strengthen self-esteem.

• Be unable to reflect on, be concerned about and care for a durable relationship between ingroup and outgroup.

• Have an undeveloped sense of guilt in relation to own harmful actions, but only fear the concrete consequences, e.g. in the form of expected retaliation.

• Be unable to reflect and report on own inner motivation, due to the absence of a a coherent and durable conception of own motives.

• Have difficulties imagining long-term consequences of present acts, due to a time-sense restricted to the concrete passing present.

• Construct group membership as the sharing of concrete properties.

• Perceive rules and roles in an absolute either/or fashion.

• Have a very restricted ability to develop creative solutions to geopolitical problems using hypothetical reasoning.

A person operating at the third order of consciousness could be expected to:

• Be embedded in the shared norms, values and loyalties of the ingroup.

• Derive an important part of own identity from group membership.

• Derive motivation from generalized norms, e.g. in the form of acting unselfishly on behalf of the ingroup.

• Be able to experience guilt for violation of conventional standards.

• Have a strong conformist propensity, i.e. adhere to conventional values, opinions and interpretations, and be sensitive to approval by social authorities.

• Be unable to take an independent and critical stand on ingroup norms, and be susceptible to group pressure.

• Have a propensity to devaluate outgroups and ingroup non-conformists.

• Have a restricted ability to break out of conventional patterns to imagine creative solutions to geopolitical problems.

A person operating at the fourth order of consciousness could be expected to:

• Be embedded in an personal worldview constituting a platform for evaluating ingroup and outgroup values and opinions.

• Generate his/her own values and norms.

• Find allegiance to personal values more important than conformity to group pressure.

• Derive group membership on the basis of common worldview.

• Have a limited ability to critically reflect on the limitations and specificity of own worldview.

A person operating at the fifth order of consciousness could be expected to:

• Be identified with a self-reflective and transformational identity, enabling a critical perspective on prevailing structures.

• Consider different worldviews and perspectives, including the own perspective, from a decentered perspective.

• Conceive group membership as transient and multidimensional.


The Subject-Object framework in perspective

Rosenberg criticized Piaget for not situating his theory in a social context. The cognitive development of the individual is considered to be subject to universal processes irrespective of the historical and social context. Kegan pays a lot of attention to what he calls "cultures of embeddedness," i.e. how the culture supports and challenges the consciousness structures of the individual (Kegan, 1994). However, there is little consideration of how individual consciousness structures relate to societal structures and processes.

A salient weakness of Kegan’s framework derives from focussing on the individual. His analysis of the properties of the four consciousness structures is made in the context of a society whose institutions and culture are permeated with third and fourth order structures and practices. One implication of this is that he fails to specify how a second order person functions in a group setting dominated by second order structures. I would speculate that a collective dominated by the second order consciousness would have developed social mechanisms (e.g. external control, ritual, punishment for non-obedience to leaders) enabling a certain social order (division of labour, strategical collective action) without relying on third order capacities.


Consciousness structures - a societal perspective

Jürgen Habermas (1976) and Ken Wilber (1981, 1995) propose a view of social history as a dialectical and dynamic relationship between consciousness structuresand social structures. Social evolution must, according to this perspective, be regarded as a process of co-evolution of the internal (psychological) structures of the members of a society, and the external economic, political, and social structures. Few scholars have seriously tried to apply such an approach in empirical studies (see, however, Elias, 1971; Eder, 1991). It would take us too far from the present issues to discuss the many problems of this type of Grand Theory. However, the frameworks presented above all fall short on situating consciousness structures in a societal context. I will therefore offer a personal interpretation of the approach taken by Habermas (1976) and Wilber (1981; 1995) in delineating three consciousness structures seen from a societal perspective. Even though I use Wilber’s terminology, I do not claim to represent his conception faithfully. I have adapted and modified Habermas’ and Wilber’s formulations using elements from a number of other scholars as well, in particular from the Jungian psychologists Erich Neumann (1970; 1990a; 1990b) and Edward Whitmont (1982), as well as from the neopiagetian Georg Oesterdiekhoff (1992). [16]


The mythic-rational consciousness structure

The individual identified with the mythic-rational structure is embedded in a collective. The I-feeling is attached to the roles, values, norms, and lifestyles supplied by the social environment. The mythic-rational person has yet to develop a firm ability to reflect in terms of universal principles beyond the conventions of the society in which he or she is living. This means that there is no independent platform from which the individual can observe cultural values, reflect upon them, and either accept them as one’s own, or replace them with other, self-evaluated values. The mythic-rational mind can only with great difficulty regard its own worldview as only one of many possible worldviews. The "cultural canon" supplied by the society of which one is a member is accepted as a part of the natural order, as truth. One has the choice of complying with the cultural canon, or of breaking the conventions, but not the choice of adopting an alternative set of values, because this requires a platform for transcendence of the mythic-rational structure. A characteristic trait of the mythic-rational structure is the conviction that the beliefs and values of one’s own group ought to be adopted by all human beings. In the preceding mythical structure the world outside was largely irrelevant. But the mythic-rational structure can grasp such notions as historical time and vast spaces, and can envision how the outside world will be transformed according to a set of visions of how the world ought to be arranged.

The collective identity, the we-feeling, is constricted to the adherents to one’s own belief system and/or one’s own ethnic or national group. Outsiders are not accorded the status of fully equal human beings. Human rights and empathy are contingent upon group membership. Social interactions within the ingroup also suffer from deficits in terms of democracy in a deeper sense. In a conflict of interests, the mythic-rational mind can only with a great effort look at its own position and the position of the other from a decentered perspective (from the outside), and resolve the differences through a simultaneous consideration of all standpoints involved. This entails a propensity to fall into dominance/obedience relationships in social interactions: either my perspective prevails, or yours (win-lose).

The lack of a firm sense of having a self-evaluated value system makes the mythic-rational individual extremely susceptible to group values. A group made up of mythic-rational members can very easily derail completely if led by a destructive leader. This happens regularly in ethnic conflicts and in racist and other authoritarian movements. The members of such groups are not necessarily driven by strong destructive impulses, but they have no personal ethical principles which could guide them into refusing to go along with destructive acts. They may therefore very easily fall prey to the addictive thrill of experiencing power when victimizing others.

The conventional values about the right life occupies an important place in the motivational structure of the mythic-rational person. Individuals are strongly identified with social roles, i.e. with what the Jungians call the persona. Being successful in living up to the conventional expectations built into these roles is an important value. Many mythic-rational persons regard the achievement of personal fame as the ultimate goal. On a collective level the propagation of group belief and values, as well as the aggrandizement of the group/nation, permeate interaction with outgroups.

The cognitive world of a mythic-rational mind has difficulties in embracing complex and contradictory contents. Coherence of the worldview is achieved by massive dissociation and projection. The mythic-rational individual therefore tends to draw very firm and divisive boundaries between right and wrong, good and evil, inside and outside, we and them. That which is alien, deviant, discomforting, or threatening is promptly assigned to the outside, where it can be condemned and kept at a proper distance. A lot of energy is absorbed in maintaining the division between inside and outside, and the result is individuals and cultures living in a state of permanent conflict.

In the contemporary Western society the social fabric supporting mythic-rational individuals is disintegrating rapidly. This is partly due to the development of the mental structure, which revolts against conventional morals and rigid and constricted lifestyles. However, as the firm conformist norms lose their authority, a lot of individuals have not yet developed individual values that could guide them into a socially responsible way of life. With no external structures to hold them, they slide into delinquency, rampant egocentric pursuance of desires, or get caught up in authoritarian movements.


The mental-egoic structure

A significant attribute of the mental-egoic consciousness structure is the emergence of the individual out of the collective. The I-feeling of the mental-egoic person is attached to a personal, separate ego. There is a sense of being an individual beyond the socially defined roles, a sense of having a unique personality. This growing separate-self sense follows from an increased ability to reflect independently. The mind gains an agility in imagining all kinds of hypothetical possibilities. One can envision alternative futures, become enthralled by ideas and visions, but also fear for all the unpleasant things that might happen. Life is a field of possibilities, and is no longer confined to conventional patterns. "What do I want in this unlimited world," is a question which imposes itself on the mental-egoic mind. The values, roles, and worldviews of the social environment are no longer accepted uncritically, because the individual has acquired the ability to reflect on such patterns, using general principles, a sense of basic values beyond conventional moral rules, and modification of previous beliefs through personal experiences or different kinds of evidence. The subjective image of one’s own personality becomes central to one’s identity. There is a set of beliefs about who one is, what one can, which values are important in life, and so on. It is no longer spontaneous impulses or conventional role expectations that dominate the individual’s will, but a set of ideals about what ought to be. Identification with the self-conception as a mental mind may lead to a sense of threat from strong emotionality, which may be felt as overwhelming to the mind. Dissociation from the body is therefore a great risk.

The great limitation of the mental-egoic structure is that the individual believes him- or herself to be a separate person, and only that. This basic belief permeates the sense of identity and the individual’s basic motivation. If I am my separate personality, this is all I have, so I tend to find my own personal life project, my desire for satisfaction, my wishes, my needs, my relationships, my ability to live up to my ideal self-image, all-important. Everything refers to the ego. If you are your desires and your self-image, you are unable to see them in a larger perspective, as for example putting them aside in order to realize unconditional love. The motivation of the mythic-rational mind was strongly related to social conventions and norms. The versatility of the mental-egoic mind permits the formulation of personal visions and life projects. The mental-egoic individual experiences her- or himself as the author of of her/his life path. The concerns within which these visions and goals are formulated are, however, restricted to the ego perspective.

As long as the exclusive identification with the separate ego persists, there is always a risk for developing narcissistic tendencies, i.e. trying to accumulate power, success, prestige, etc. for your own person in order to appear grandiose and significant. Self-assertion comes naturally to this consciousness structure. Separateness entails an acute sense of vulnerability, hence a persistent drive to attain control over oneself and the environment. At its best, the highest satisfaction of the mental-egoic mind can be to give others satisfaction. However, it is still a question of one ego relating to another, referring this experience to the personal ego.

The mental-egoic mind can envision a multitude of hypothetical alternatives to the concrete reality, and is therefore able to compare the worldview of its own society with other worldviews. This versatility of imagination gives an ease of seeing events and situations from several different perspectives, and the ability to consider different perspectives at the same time. Where the mythic-rational mind was unable to reflect critically on social conventions, the mental-egoic mind can operate with universal principles, and can use these principles to evaluate conventions and norms. The propositions of social authorities are checked for internal consistency and are compared with accumulated evidence. The cosmology conveyed by the culture is no longer passively accepted as absolute, since it is now recognized that the cosmology is only a subjective and provisional image of the order of the world. The ability to shift perspectives and to take the role of others naturally leads to a living and meaningful conception of the "generalized other," from which follows a set of conceptions about how people in general ought to act towards each other ("the Golden Rule," for example). True democratic relationships become possible, where people might be willing to listen to each other with the goal of mutual adaptation. The mental-egoic mind is capable of perceiving the uniqueness of other individuals, where the mythic-rational mind tended to see each member of a particular group as basically similar (attribution of collective traits). However, even though reflection inevitably results in some kind of "live and let live" philosophy, the concerns of a person exclusively identified with a separate ego tends to speak for a "me first" mentality. Actual behaviour may therefore lag behind moral development in the realm of reasoning.


The existential structure

A key feature of the existential structure is the increasing ability to observe one’s own mind and its processes. The individual starts becoming able to behold the ego, its conceptual world and its life projects, so to speak from the outside. One starts to discern the inherent limitations of the specific forms of conceptual reason in which the mind is embedded. Because of a growing sense of being more than the ego, it is possible to let go of an anxious defence of the idealized self-image of the ego. The existential individual has an inner platform for the I-feeling that makes it possible to give up the need to control and secure the ego’s position, allowing change to occur without threatening the basic sense of existing.

The recognition of the ego’s inherent limitations opens many doors that were kept shut in earlier structures. Feelings and impulses that could not be integrated or understood within the confines of the reason-centered mind can be allowed, e.g. longing for self-transcendence, intuitive images and impulses, intensive peak experiences of merging with nature or other people, etc. However, an increased sensibility for the vast spaces outside the narrow ego world also opens up for an intensive experience of one’s own smallness, vulnerability, and insignificance. It is no longer possible to hold on to narcissistic delusions about personal grandiosity and significance. This emergence from the embeddedness in the ego’s elaborate system of meaning confronts the individual with the question of the meaning of one’s life. A satisfying answer to that question can no longer be derived from conventions (as in the mythic-rational structure), from ideologies or ego-related life projects (as in the mental-egoic structure), but must be anchored in the individual feelings and values. The existential individual has seen through the futility of pursuing ego gratification, and seeks for a larger context of meaning. This search usually leads to a commitment to authenticity and altruistic values. The desires and wishes of the ego becomes subjected to the need to make a contribution to the well-being of others.

But even though the existential individual starts identifying with universalistic values and perspectives (human rights, care for the ecological system, global justice, etc.), the I-feeling is still essentially limited to a separate self-sense. This separateness is intensely experienced on a feeling-level, and it is exacerbated by the loss of the relative safety of being embedded in the mythic-rational collective, or the mental-egoic ideology.

In the wider cognitive sphere, the main progress of the existential mind over the mental-egoic is the decentering of discourses. The existential mind can observe and operate with different paradigms, even if they are logically incompatible. Different perspectives can be compared and embraced without a need to give one of them a privileged status. Vision-logic recognizes that all perspectives have inherent limitations due to the reductive nature of concepts. A fully developed vision-logic deconstructs the fundaments of ethnocentrism: no perspective is a priori privileged. There is no avoiding the conclusion that all human beings have the same right to have their needs satisfied and their claims considered, nor that we ought to face our own position in the cosmos with a certain humility. The human being is a part of a vast web, and we ought to care for the state of the social and natural environment.

In social interactions, the mental-egoic person is driven to defend the idealized self-image and to pursue egoic desires and wishes. The existential individual can to some extent let go of the need to control the environment, and may therefore find it easier to accept lasting differences in values, opinions, life styles, behaviour, etc. in social relationships. Other people can be allowed to be different, without a need to place them into a neat category.


The Habermas/Wilber framework applied to geopolitical reasoning

Persons operating from a mythic-rational consciousness structure could be expected to:

• Be identified with conventional roles and values.

• Regard ingroup values as ultimate, and regard own worldview as universally valid.

• Derive ingroup membership from adherence to belief system.

• Devaluate outgroup belief systems.

• Have a propensity to resolve conflicts through dominance or bargaining (rather than through win-win), due to a restricted ability to integrate perspectives.

• Be susceptible to group pressure, e.g. by going along with destructive acts towards outgroup members, due to the absence of a firm personally anchored ethical system.

• Perceive enhancement and defence of collective self-esteem as an important concern. This may lead to collective self-aggrandizement (group narcissism).

• Have a propensity to reason in dichotomies: right-wrong, good-evil, inside-outside. This tendency can translate into having a stake in maintaining distance and divisions.

Persons operating from a mental-egoic consciousness structure could be expected to:

• Derive a substantial part of their motivation from desires related to ego-gratification. The resulting personal agendas may coincide with collective goals but are not derived from them.

• Have a restricted ability to take perspective on own discourse.

• Accumulate "valuables," e.g. status, achievements, esteem, wealth, pleasure.

• Be able to take perspectives of outsiders, but not be able to take perspective on own discourse.

• Deal with conflicts in a way which takes the others’ perspective into consideration.

• Be committed to universalistic values, e.g. human rights, but bend these to fit own interests.

Persons operating from an existential consciousness structure could be expected to:

• Derive motivation from a worldcentric perspective.

• Be able to disidentify from ego-related desires.

• Have a weak identification with collective identities

• Be able to reflect critically on own worldview.

• Abstain from claiming to have a privileged perspective.


Comparison and evaluation of the four frameworks

The four frameworks presented in the preceding sections seem to focus on different dimensions of consciousness development. The "integrative complexity" framework focusses on the development of the ability to handle different perspectives. This is an aspect of the development of cognitive skills. Rosenberg’s framework also deals with the development of cognitive skills, but he focusses on the structure of reasoning, particularly the nature of the basic units of reasoning, and stages in the construction of causality in the sphere of politics and other social interactions. Kegan’s framework is grounded in a similar conception of the basic units of cognition, but his focus is identity rather than cognitive abilities. Kegan asks how the meaning-making self is structured. The last framework is something of an amalgamation of various lines of analysis. Habermas’ theory is grounded in the development of moral reasoning, and the socially institutionalized forms of communication. In my reading, all of these approaches have distinctive merits, and it would not be very productive to treat them as competing theories. The integrative complexity framework can help us to identify how parties to a conflict deal with the existence of highly different subjective perspectives on the conflict. Rosenberg’s framework can help us identify in what kinds of units the geopolitical actors think, and how they conceive the causal relationships in the conflict process. Kegan’s and Wilber’s approaches can provide insight into the nature of the motivation of the conflict parties. Habermas and Wilber can furthermore help us identify differences in the nature of the interactional logic in social relationships.

I believe in open-ended frameworks, and will therefore make an effort to extract a number of relevant developmental dimensions out the frameworks presented above. However, before doing this, I want to address a number of general methodological and theoretical issues.


Delimitations and theoretical problems

The four frameworks presented above all assume the existence of a developmental hierarchy of cognitive structures or consciousness structures. The most controversial aspect of this family of theoretical models is the claim that cognitive development takes the form of a succession of structured wholes. This assertion means that at a certain level of development, a person is expected to function according to the typical patterns of the particular level across different areas of experience, and at different occasions in the same area. This is a strong claim, and a large number of scholars have raised objections to it on theoretical and empirical grounds. In this section I will briefly review some of the problematic aspects of consciousness development models in general.


The consistency hypothesis

The consistency hypothesis states (i) that consciousness structures are coherent wholes, and (ii) that a person is expected to operate consistently according to the same mode of consciousness in different areas of life (e.g. work, intimate relationships, political views) as well as on different occasions in the same area. The consistency hypothesis is attractive because it offers a firm theoretical platform. However, it has proved difficult to defend the consistency hypothesis when it has been challenged on theoretical and empirical grounds. As far as I can see, the debate about the consistency of consciousness structures is far from concluded. At this stage, it seems unwise to make the consistency hypothesis a centre-piece of one’s theoretical framework. In fact, I really hope that the consistency hypothesis will prove untenable, at least in its most restrictive formulation. It would be far more agreeable if it turns out that situational factors indeed have some influence on the mode of consciousness, because it would mean that we can go on studying how favourable environments for consciousness expansion can be created.

As hinted at above, the consistency hypothesis concerns two different types of consistency. The strong formulation assumes that a consciousness structure is a coherent whole, an interdependent system where a single principle or form generates consistency in various areas, such as reasoning about physical causality (Piaget), interpersonal relating (Selman), moral reasoning (Kohlberg), faith (Fowler), political reasoning (Rosenberg), and leadership in organizations (Torbert). This assumption makes a theoretical framework extremely vulnerable, and I find it personally questionable if it is at all desirable to spend a lot of effort in making a generalized model of coherent consciousness structures work. Applied to my field of inquiry, the generalized approach would entail describing a limited number of coherent, and structurally different, "geopolitical logics," and trying to empirically substantiate or refine this conceptualization. As a heuristic model, such an approach might have some value. The main alternative is to use the different theoretical frameworks to identify a number of analytic dimensions which could be useful for analysing the properties of concrete cases of geopolitical reasoning. This would permit an independent analysis of, for example, reasoning about political causality, ability to integrate perspectives, and collective subject-object balances. The extent to which these dimensions are correlated, as would be predicted by the strong claims of consciousness structure models, would then become a matter of empirical analysis. In this way, we can avoid reducing the complexity and unique features of individual cases to fit into a simplified typology, but still be able to make developmental analyses of geopolitical phenomena.

The second type of consistency (similarity of functioning across fields of life and over time) cannot be abandoned without fundamentally undermining the relevance of developmental models. Some form of consistency in behaviour, reasoning and affective reactions must be demonstrated if the developmental perspective is to be accorded any status of meaningfulness. This does not necessarily imply that developmental stage must be completely uninfluenced by situational factors. However, the developmental perspective expects the individual to be "at home" in particular stages of development for extended time spans. Regression is possible, but should be recognizable to the person as such when reflecting on the occasion.

There is no comprehensive framework which spells out different "lines of development" and the nature of the relationships across the lines (e.g. the relationships between stages of self-sense development and stages of moral reasoning). [17] Perhaps such a framework would be too schematic and simplified to be useful. Anyhow, I believe that empirical research into different dimensions of consciousness development has a considerable potential to yield interesting results.

In my own research, I will not stress the importance of identifying or testing for coherent consciousness structures. However, I accept as an important assumption that within a specific dimension of consciousness development, individuals progress through a limited number of basic modes of functioning. Therefore, within a limited time-span, I expect a person to display a recognizable pattern of reasoning or motivation, consistent across different issues, and at different occasions. However, a further problem must be taken into account, namely the possibility of cognitive regression due to changes in situational factors.


The regression hypothesis

The regression hypothesis claims that environmental stress may lead to cognitive and affective regression to earlier stages of development. One salient form of environmental stress is becoming involved in a social conflict. The Austrian conflict facilitator Friedrich Glasl (1997) has developed a comprehensive 9-stage model of conflict escalation, which includes a detailed description of how the conflict parties regresses in the realms of cognition, affect, volition and behaviour (see also Spillmann & Spillmann, 1991). A succinct formulation of the regression hypothesis would run something like this: In stressful situations a person’s normal mode of consciousness may become overloaded, due to a combination of anxiety (or other strong emotions), time pressure, and inability to cognitively handle complex and contradictory information in matters which the person experience as emotionally important. When the normal, most mature, mode of consciousness fails to deal with the situation effectively, the I-feeling regresses into some kind of less complex and more "automatic" response pattern. Such a pattern can rely on tradition, habit, conventions, fixed rules, unconscious scripts, impulses or reflexes. Various psychological defenses against stress become active, such as cognitive complexity reduction, fixation of interpretation patterns and attribution of stereotypical propeties to counterparts. Few empirical studies testing the regression hypothesis exist.

A comparison of consciousness development theory and the regression hypothesis show that the former focus on a stable progression from stage to stage, whereas the latter implicitly assume that there is a "normal" state of consciousness from which persons may regress due to situational factors. These basic assumptions may be theoretically compatible, but it is probably extremely difficult to operationalize a framework comprising both theories for empirical research of people’s meaning-making in concrete events. Such research would require longitudinal studies of people who gradually become involved in conflicts. In the field of geopolitical conflicts, which usually have a long history and a prominent place in the socio-political climate, such longitudinal studies become very difficult.

For my own research, I conclude that the developmental properties of a particular example of geopolitical reasoning may reflect both long-term developmental stages and situational factors. This indeterminacy does not affect the analysis of the developmental properties in particular cases, but must be kept in mind when discussing why these particular properties appear in a particular case.


The reasoning vs. actual behaviour issue

The neopiagetian approaches emphasize how persons make sense of their experiences by focussing on the structure of their reasoning. A major problem with this approach is that it leaves unspecified the relationship between reasoning and actual behaviour. The sometimes glaring incongruence between the morality a person declares verbally and the actual behaviour displayed by the same person in real-life situations has been one important thread in the debate on Kohlberg’s research on the development of moral reasoning. The problem is not resolved by observing behaviour and subsequently asking people how they justify their actions, since actions can be rationalized a posteriori. From a psychodynamic perspective it is doubtful if the explanations people give for their own behaviour are sufficient, or sometimes even relevant, for analysing why they act in a certain way. We may devise ingenuous methods for analysing the structure of reasoning, exposing things about people’s meaning-making they have no idea of themselves, without having any guarantee that their behaviour result from their (conscious) meaning-making. A major factor of uncertainty is, of course, the role of unconscious motivation. It is difficult to prove the relevance of unconscious factors, but even more difficult to disprove it. It might be argued that structures of reasoning are epiphenomena to unconscious motivational structures (in the psychoanalytic sense), and that they therefore play a marginal role in explaining people’s (political) behaviour. [18] Apart from psychodynamic theories, there are several other approaches that advance explanations for the gap between actual behaviour and conscious discourse (e.g. in ethnomethodology, social psychology and social anthropology).

I don’t expect to be able to resolve this dilemma using the approach presented here. I think a reasonable position is to assume that a person’s conscious meaning-making is a significant factor contributing to actual behaviour, but perhaps not the only (or even the crucial) factor. I will focus on conscious meaning-making, knowing that there are limits to the explanatory power of meaning-making structures when studying actual behaviour. However, I would like to stress that the question of the nature of the relationship between unconscious motivation and modes of reasoning is potentially very important, since it influences the possibilities for transforming (destructive) modes of social interactions.


The dangers of hierarchies and "Grand Narratives"

The kind of perspective presented here is highly inopportune in an academic and political climate influenced by postmodern discourses. Postmodernists emphasize the need to transcend parochialism and claims of having privileged perspectives. Furthermore, they are highly critical of the modernist fallacy of trying to formulate Grand Theories that claim true representation of large chunks of reality. From this perspective, the developmental paradigm is highly suspect because it claims that it can define less and more developed consciousness structures. Developmental theory, e.g. when applied to moral reasoning, implicitly or explicitly claims ability to judge which ethical and ideological systems are more advanced than others. Therefore, it might be used for marginalizing certain ideologies, worldviews and persons, by designing them as less evolved. A critical attitude to normative hierarchies is, in my opinion, a highly desirable feature of the debate climate. It is in itself a sign that post-conventional consciousness structures are gaining ground. Developmental perspectives might attract people who have use for a perspective that legitimates discrimination of less privileged groups. This danger is real, and a serious problem for applying consciousness structure analyses to adult people. I don’t think it is possible to avoid this danger completely even though any kind of devaluation or elitism can be criticized from the highest levels of the frameworks presented here. There is, however, strong arguments for making use of developmental perspectives. One of them is that a developmental perspective can expose the limitations of parochialist positions. Another strong argument is the desirability of specifying steps of development that have the potential to contribute to a more humane society. A third argument is that the developmental perspective, if put to work effectively, has a strong potential to contribute to empowerment, and therefore to liberation of the human potential.


Next step

Having presented the theoretical raw material and having discussed some general theoretical issues, the next task is to wield this mass into an effective research instrument. In the next major section, I will make an outline of the developmental dimensions I find relevant for my research purposes. Thereafter, I will review and discuss available methods for collecting and analysing empirical material.



3. A preliminary conceptual framework


In section 2, I presented four theoretical frameworks, and extrapolated some implications for analyses of geopolitical reasoning. In this section I present a preliminary conceptual framework derived from the four frameworks presented above, but adapted to my specific concerns. As I have already indicated, I prefer to use dimensions of consciousness as the basic units of analysis rather than formulating a model of consistent consciousness structures. Human consciousness is an extremely complex phenomenon, and there are innumerable legitimate ways conceptualize its dimensions and developmental gestalts. A certain measure of arbitrariness will be inevitable in any analytical framework, I believe. However, while there may be no single right way to conceptualize consciousness development, there may be wrong ways. There may also be significant variations in the usefulness of various models.

I will soon proceed to formulate a tentative framework for studies of geopolitical reasoning. However, before I focus on this quite narrow field of application, I want to dwell on general dimensions of consciousness development for a while. I find it useful to group the analytical dimensions I have encountered in the literature into two broad categories: self and instruments. Most cognitive-developmental models, including the integrative complexity framework and Rosenberg’s framework, focus on cognitive abilities. Far less research has been made on other types of skills, such as skills relating to feelings and emotions (e.g. the ability to handle contradictory emotions), to exercise of intentionality (e.g. ability to decide to transform some of one’s own feelings), and to interactive abilities (e.g. to handle other people’s aggressivity in a constructive way). However, these dimensions of consciousness development can also be regarded as abilities, albeit in other realms. The other major category focusses on the nature of the subject, or, in other words, the structure of the self.

In the table below, I give examples of major consciousness development dimensions in these two broad categories. Abilities represent something a person is able to do, whereas self-embeddedness is the platform from which a person exercises the abilities. The self-embeddedness regulates a person’s motivation and basic outlook. The nature of the self-embeddedness is crucial for determining in which way a person makes use of his/her abilities. On the other hand, it is obvious that the cognitive abilities a person has access to are strongly related to how the self is constructed.


Reasoning about causality

Abstractness of concepts


Coordination of perspectives

Self-reflection (cognition of own psyche)

Differentiation of person cognition


Differentiation of perceptions and reality


Differentiation of feelings and judgements

Endurance of displeasure and ambiguosity

Multidimensionality of affect/feeling



Exercising will in one’s own behaviour

Exercising will in relation to own feelings


Applying intentionality to transforming relationships

Differentiation of authority relationships


Self-identity (Kegan’s subject-object)

Scale of commitment/concern/motivation (egocentric-worldcentric)

Collective identification

Embeddedness in belief systems

Embeddedness in perceptions/interpretative patterns (decentering)

Dominating fears

In adapting a general consciousness development approach to the specific field of geopolitical reasoning and interactions, some dimensions appear more relevant than others. In the field of geopolitical interactions, it is obvious that a major point of interest is the nature of the relationship between one collective and another. How do persons or groups perceive an outgroup? How do they define themselves in relation to this outgroup? How do they reason about the causes of the present and past events in the relationship between ingroup and outgroup? How do they construct the nature of the relationship between the groups, and between the perspectives of each group?

Tentatively, I have identified five broad areas for investigation:

1. Reasoning about socio-political causality

2. Gestalt complexity

3. Ingroup identification

4. Perspective-taking

5. Ingroup-outgroup relationship

Drawing on the theoretical frameworks presented in section 2, I have specified conceptual dimensions and concrete questions for each of these areas. The result is a kind of checklist of relevant dimensions of analysis in specific cases of geopolitical relationships or events. The structure of geopolitical reasoning of one or several parties to an intergroup conflict can be described and interpreted using this checklist as a guideline. At the end of section 5, this framework is further operationalized into an interview guideline.


1. Reasoning (about causality)

• The structure of reasoning itself

Rosenberg’s specification of sequential, linear, and systemic thinking is here a useful starting-point for recognizing the nature of the respondents’ reasoning about (political) causality. Does the subject reflect on causal relationships at all; in terms of cause-and-effect relations (unilinear relationships); in terms of the system in which events are situated (multiple causality)?

• Hypothetical/creative reasoning

To what extent does the respondent go beyond immediate appearances and conventional patterns in order to reason hypothetically and creatively about causal relationships and future alternatives?

• Concrete vs. abstract reasoning

Is reasoning bound to the concrete, e.g. in the construction of gestalts of groups and intergroup relationships? Does the subject construct categories on the basis of abstract properties (e.g. citizenship rather than ethnicity; values rather than organizational affiliation; direct interactions rather than mutual trust or role distribution)? To what extent are concepts such as democracy, loyalty, justice, and rights constructed as concrete vs. abstract?

• Closure vs. openness in interpretation of causality

Is the respondent open to questioning his/her interpretation of the present state of affairs? Is the political belief system fixed or open to critical reflection?

• Propensity to make the environment responsible for negative

experiences vs. taking responsibility for own situation

Does the respondent regard the self and/or the ingroup as a victim of circumstances over which he/she/they have no control?


2. Gestalt complexity (Outgroups)

• Unidimensionality vs. multidimensionality of gestalts

To what extent are the images of outgroups differentiated, e.g. in the sense of considering several dimensions which might be evaluated independently from each other?

• Tolerance of cognitive dissonance, contradiction, ambiguity,

indeterminacy, doubt.

Are there signs that the respondent stably can embrace contradictory feelings, evaluations, opinions, etc. regarding the outgroup?

• Unidirectionality vs. mutuality in group structures

Is the outgroup perceived as a monolithic entity or as a complex system? Is decision-making constructed as a simple leadership-follower (hiearchical) relationship, or as a result of complex mutual relationships between different factions and individuals?

• Closure vs. openness in gestalt formation

To what extent is the outgroup image fixed? Are there signs of awareness of the image-reality problem?

• Differentiation of outgroup gestalt

To what extent does the respondent perceive individual differences among outgroup members? Are there signs of attribution of collective properties to outgroup members?


3. Ingroup identification

• Egocentric, sociocentric, worldcentric identity

Do the political values and beliefs refer to an egocentric, sociocentric or worldcentric identity (preconventional, conventional, postconventional values)?

• Concrete vs. abstract membership criteria

What are the most important criteria for assigning membership of ingroup? Concrete, such as ethnicity, creed, formal membership; or abstract, such as adherence to certain values, commitment to certain principles of interaction?

• Scale of ingroup

Every person have ingroups at different scales (family, relatives & friends, organization, movement, nationality, religious community). Which types of ingroups are most important in geopolitically relevant contexts?

• Discreteness vs. flexibility of ingroup boundaries

To what extent are ingroup boundaries discrete and fixed? What meaning is assigned to the group boundaries?

• Conformism vs. independence in individual–ingroup relationships

To what extent is the respondent disposed to take a critical stand in relation to ingroup values and interpretations?

4. Perspective-taking


• Ability to coordinate and integrate perspectives

Can the respondent take the perspective of the counterpart with some adequacy? To what extent is the respondent capable of coordinating and integrating ingroup and outgroup perspectives? Is the political reasoning of the respondent a result of a capacity to consider other perspectives?

• Ability to take a decentered perspective

Can the respondent reflect on the conflict from a third party perspective? Can the respondent take the ingroup perspective as an object of critical reflection?

• Preferred modes of conflict resolution

Tendency to think of conflict resolution in terms of concrete outcome (Rosenberg’s "sequential thinking"), win-lose, bargaining, negotiation, win-win.

5. Ingroup-outgroup relationship


• Separation of feeling and evaluation/judgment

Propensity to keep apart value judgements from negative emotional reactions to counterpart.

• Propensity to schismogenesis and devaluation

Propensity to create and reproduce psychological distance to outgroups.

These are dimensions to look for, aspects of the generic meaning-making of persons involved in geopolitical conflicts. The dimensions refer to how concrete contents are structured, not to the contents themselves. The researcher must keep the tension between content and structure in mind during interviewing and interpreting, not confusing these two aspects of the statements made by the respondents.

The tentative character of this framework means that the empirical material must be interpreted with a keen eye for aspects which might necessitate a revision of the conceptual framework. I regard this as a natural and desired feature of the research process.



4. Methods



In the first section, the general purpose of my research endeavour was formulated as: What is the relationship between general modes of consciousness and specific patterns of geopolitical reasoning? In section 2 I presented the theoretical frameworks I have found relevant for exploring this issue, and in section 3 I formulated a tentative conceptual framework for applying a consciousness development approach to geopolitical problems. The purpose of this section is to survey potential methods for carrying out empirical research. I will start by discussing two general research strategies, then I will present five methods which might be relevant and evaluate their usefulness in the context of my research problem.


Research strategies

Given the problem formulation in section 1 and the theoretical framework outlined in section 2 and 3, we can choose between two different research strategies. The choice of strategy depends on the importance assigned to theoretical issues vs. practical applications. If the development and testing of a theory on the role of consciousness structures in geopolitical processes is regarded as important, the research strategy should be designed to test a set of theoretical propositions. If on the other hand the practical applications, e.g. understanding geopolitical strategies or events, or evaluating the effectiveness of conflict resolution methods, are emphasized, then the research strategy should be designed to generate insights into the dynamics of concrete geopolitical processes. In other words, the strategical choice depends on which aspect is given priority: theory building or understanding of actual reality.


Theory testing

The basic hypothesis underlying my endeavour is that characteristic patterns of geopolitical reasoning corrspond to specific stages of consciousness development. Extensive empirical evidence supports the assertion that different stages of cognitive development exist (Kohlberg, 1981; Loevinger, 1976; Kegan, 1982; 1994; Selman, 1980). The theory-oriented approach would have as its main objective to test if there is a significant correlation between specific consciousness structures and characteristic structures and contents of geopolitical reasoning. Thus, we would have to use a three-step procedure: 1. determining the consciousness structure of a number of subjects using any of the methods described below (subject-object interviews, sentence completion tests, or text analysis); 2. elicit reasoning about geopolitical issues from the same subjects; and 3. testing the correlation between level of consciousness and identifiable patterns of geopolitical reasoning.

This research strategy implies a strong need for unequivocal formulation of concepts and hypotheses, and a stringent method. This need for stringency might conflict with a wish for openness towards the nuances of geopolitical reasoning. It seems desirable to me not to close down hypothesis formulation at an early stage, therefore I am reluctant to adopt the theory-testing strategy in the first phase of empirical studies.


The pragmatic approach

The other alternative is to generate a sufficient empirical material comprising geopolitical reasoning by different subjects, in order to identify levels of complexity in the material itself. This strategy would shortcut the need for scoring the subject’s consciousness structures. It would also evade a number of theoretical questions about the link between an individual’s basic consciousness structure, and a concrete instance of reasoning. One example of such issues is whether individuals always consistently use the same consciousness structure over time, and across different areas of life (e.g. discussing politics, interacting in the work-place, in intimate relationships).

The more pragmatic research strategy would involve formulating an openended and flexible theoretical framework, suggesting interesting themes to look for while analysing interviews or texts. However, no stringent hypotheses would be formulated before the analysis. The preliminary aim of the study would be to arrive at a catalogue of relevant statements, and a preliminary formulation of how bits and pieces of geopolitical reasoning can be ordered into a hierarchical framework on the basis of cognitive complexity.


Sentence completion tests

Sentence Completion Tests (SCT) or Paragraph Completion Measures (PCM) have been developed by several researchers as convenient instruments for eliciting material scorable for developmental stage. The basic idea is to ask the test subjects to complete (in written form) a number of sentence stems, such as "When I am in doubt . . ."; "When I am criticized . . ."; or "Rules . . .". For example, Loevinger uses a form with 36 sentence stems (Washington University SCT, WUSTC), which the test subject is invited to complete in 20-30 minutes. The responses are then scored according to rules laid down in a scoring manual. Various techniques can be used to convert individual scores to a single score for a person.

The most used SCT’s have been developed by Loevinger and associates (Loevinger & Wessler, 1970; Hy & Loevinger, 1996), and within the integrative complexity framework (Schroder, et al., 1967; Phares & Streufert, 1969; Baker-Brown, et al., 1992).

The SCT approach assumes that developmental level is a structural rather than a situational variable. SCT’s are used to determine the developmental level of individuals. This information can then be used as input in analyses incorporating other material as well, e.g. how the same individuals perform in different experiments, or what they say in interviews on certain topics. If the developmental level turns out to be highly situationally determined – as asserted by Tetlock and Suedfeld (see above) – the SCT must be used with great caution.

The advantages of this method are:

1. The procedure can be highly standardized, yielding high transparency, comparability and stringency.

2. The administration of the test requires no skill on the part of the test functionaires.

3. The scoring process can be codified into a standardized scoring manual. This permits high transparency, minimizes scorer biases, and facilitates the training of scorers.

4. Since administration and scoring of the test can be made in quite simple ways, the costs and time consumption can be kept low.

5. The method is suitable for quite large samples.

The disadvantages are:

1. The depth of information is very limited.

2. The standardization of the test forms and scoring procedure implies an extremely low flexibility in adapting the test to situational variables.

3. The format is very obviously a psychological test, which might reduce the range of available test subjects.

4. The limited volume of material for each individual entails a risk of not eliciting the most mature modes of reasoning.



A SCT (in particular the WUSTC) is an efficient measurement instrument for assigning developmental stages to test persons. Its use is, however, limited to populations who agree to fill out the form. It is therefore primarily relevant in in studies where it is possible to recruit people to participate in a scientific experiment. It is more difficult to use an SCT in an interview-based study of parties involved in a serious geopolitical conflict.

Having access to a precise measure of stage of development is not a major priority for my project, since I aim to identify distinct types of geopolitical reasoning rather than test the theory of consciousness development. The SCT could be used with a group of volunteers, however, in combination with a Kohlberg-type interview (see below) about a scenario.

One suggestion for further development might be to attach a set of sentence completion tasks verbally at the end of an interview. The stems would have to appear relevant to the main topic of the interview. Giving the same sentence stems to a large number of interviewees to complete might yield interesting material.


The Subject-Object interview

The Subject-Object interview is an instrument developed by Kegan and his associates for determining the order of consciousness of test subjects. The rather open and skill-dependent format of the interview is determined by the need to elicit scorable material about the subject’s basic meaning-making principles. The test subject is given ten cards, each with a printed title, e.g. "Angry," "Torn," or "Change." The test subject is asked to make notes on each card about experiences made in the recent past that come to mind upon reflecting on the card title. The actual interview is about one hour, and uses the notes made by the test subject as starting-points for interviewing the test subject about how she/he makes sense of her/his own experiences. The main task of the interviewer is to ask questions that stimulate the interviewee to explain why certain dilemmas are dilemmas to her/him. In particular, the interviewer is sensitive to what the subject is unable to reflect upon:

We know what is "subject" for the interviewee by listening for what she or he is unable to take a wider perspective on. That is, we believe we are in "subject" territory when we hear the person talking as if, in the experience itself that he is describing, he is unable to construct any wider form of reference for the experience, despite opportunities to do so created by the interviewer.

People can take a perspective on that which is "object" to them by definition. We know what is "object" in the interview then by seeing what the interviewee is able to reflect on, control, take responsibility for, be in charge of, manipulate, or regulate. (Lahey et al., 1988:14)

The interview technique requires a high degree of experience. I won’t go further into the details here, but refer to the comprehensive interview manual published by the Subject-Object research group (Lahey et al., 1988). Eliciting scorable material is the first part of the process, while scoring the interviews is the second. A certain flavour of how the scoring is made is conveyed by the following citation from the above-mentioned manual:

We do not look for certain themes or topics or issues which we think to be consistent with specific subject-object stages. When people first begin to learn about the different stages they [… ] assign certain personalities to the stages: stage 2 is selfish; stage 3 is nice but easily taken advantage of; stage 4 is unable to have close intimate relations, etc. Or people attach certain themes to the speech of a given stage--e.g., stage 3 talks about relationships; stage 4 talks about career. Or they attach certain motives-- stage 3 is motivated by needs for affiliation, stage 4 by needs for achievement, and so on. Such a line of reasoning might suggest that you analyze interviews by identifying stage-typical themes, preoccupying concerns, motives. Actually, subject-object balances have nothing to do with specific themes, motives, issues of preference. Subject-object balances are principles of organization. These themes, motives, and issues are stereotypical contents of organization. People can talk continuously about relationships and be almost any stage; people can be stage 3 and talk about everything but relationships.

It would be closer to the truth to say that the first step in analyzing a Subject-Object Interview has nothing to do with stages at all. Instead of looking first for stage particulars, we look for any material which seems to be expressive of structure, any structure. (Lahey et al., 1988:11-12)

Because the interview format is unstandardized, the method is highly sensitive to the subjective interpretations of the scorer. This reliability problem cannot be evaded completely, but it can be reduced by parallell blind scoring of two or three trained scorers. However, this procedure is time-consuming and costly, which means that the number of interviewees must remain quite limited.

The advantages of the subject-object interview are:

1. It permits the researcher to explore the core of the subject’s meaning-making.

2. The material yielded by the interviews is complex and rich, permitting a differentiated analysis.

3. The method is flexible, and can be adapted to different purposes and situations.

4. The method is normally perceived as stimulating to the test subjects.

The disadvantages of the method are:

1. The lack of standardization implies a low level of transparency.

2. The interview technique requires a thorough understanding of the theory, as well as long training and experience.

3. The method is, because of its unstandardized nature, vulnerable to the biases of the scorer. In order to ensure reasonable reliability, each interview must be blindly scored by at least two skilled scorers.

4. The administration and scoring of the test is highly time-consuming, therefore costly.

5. The method is not suitable for large samples.


"Situated" Subject-Object interviews

Pamela Steiner (1996) developed a variant of the Subject-Object interview in the course of her PhD work. She analyzed how orders of consciousness influenced reasoning in a complex decision making process in the case of an environmental conflict. She made two interviews with each participant in the study, one standard Subject-Object interview, and one "situated" interview. In the latter, specific events in the conflict process were taken as starting points for an exploration of consciousness structure.


Subject-Object interviews via e-mail

I am currently exploring the suitability of conducting Subject-Object interviews via e-mail. E-mail interviews offer some important advantages, notably:

1. Test subjects can be recruited easily via newsgroups or discussion groups on the Internet.

2. Subjects from different geographic areas can be reached, provided they have e-mail access.

3. The interviews must not be transcribed, which saves time.

4. The interviewer has plenty of time to consider answers and formulate new questions during the course of the interview. Hypotheses can be formed and tested in a considered way.

5. The e-mail format means that the interviewer can pick up several themes at once, whereas in an oral interview one can only pursue one of several possible themes in a response.

However, there are also problems, e.g.:

1. It is more difficult to build rapport, to convey empathy, and to create a trusting atmosphere.

2. The population is restricted to persons with e-mail. This constrains the population both in terms of socio-economic variables and in terms of geographic patterns.

3. Some people need a lot of follow-up questions in order to get somewhere, i.e. a high frequency of questions-answers. This can be achieved using e-mail only if the interviewer and the interviewee agree to sit at the computer simultaneously.


I regard the subject-object interview as the most sophisticated instrument for analysing consciousness structures. Its open-ended character also makes it flexible. An intriguing task would be to explore if the method can be adapted to probe for the structure of collective identifications in real-life social conflicts. The basic techniques used in the subject-object interview can be adapted to free format interviews about various topics, in order to elicit more material on the meaning-making structure of the interviewee.


Kohlberg-type interviews

Lawrence Kohlberg, in his research on the development of moral reasoning, used an interview method that could be adapted to many different purposes. The subjects were presented with a short story involving a moral dilemma, which was used as a starting-point for a probing conversation about how the subjects went about reasoning about resolutions to the dilemma. One advantage with this method is that the story is standardized. Comparison of the reasoning of many different subjects is therefore facilitated. The method is also suitable for longitudinal studies, where the same subjects are surreptitiously interviewed about their reasoning about the same moral dilemma.

Shawn Rosenberg used a similar method, but asked his subjects to discuss recent political events. He describes his interview method in the following way:

American politics interview. The focus of the interview was the subjects’ understanding of the structure of government decision-making. Each subject was asked to suggest an issue which he or she felt to be of particular interest. The issue chosen provided the substantive concern for the remainder of the interview. The subject was then asked questions such as who is responsible for the problem, what is government doing about it, what role does the Congress, the President, the Cabinet or the bureaucracy play, and how do these various aspects of government relate to one another in the decision-making process. When the meaning of a subject’s response was unclear, the interviewer probed further with questions such as what do you mean by that, why do things work that way, etc. (Rosenberg, 1988:162)

The responses were coded according to a description of typical sequential, linear, and systematic reasoning patterns.



This general model could possibly be adapted to investigations into the structures of geopolitical reasoning. This could be done by presenting interview subjects with a standardized scenario depicting a geopolitical dilemma. The case can be adapted to different purposes, depending on what aspect of reasoning is deemed most interesting. For example, the researcher might want to focus on how different subjects reason about conflict resolution possibilities in geopolitical conflicts. The cases can be hypothetical, or simplified versions of real conflicts. However, the researcher must be aware that the results might be highly influenced by the extent to which the subject is member of (and identified with) one of the parties described in the case.



An obvious response to the reasoning vs. actual behaviour dilemma raised above is to study the behaviour of test subjects, rather than their reasoning, or a combination of both. This can be made in real life, or in controlled experiments using simulated conflicts, role-plays or games.

A group of researchers at the University of Maryland have developed a simulation for studying the relationship of cognitive structure to patterns of decision making in crises (Santmire et al., 1998). Their experiment uses a computer-based simulation of a hostage crisis, where the participants play the roles of Sikh terrorists, Pakistani government, and Indian government. The complexity level of the participants according to the integrative complexity framework is measured using a Paragraph Completion Measure.



This approach permits the study of the relationship between cognitive structure and various aspects of decision making under stress under controlled circumstances. A major advantage is that actual behaviour can be observed, rather than mere reasoning. The major draw-back is that the participants play roles they are not genuinely identified with. According to several theoretical frameworks, persons with less complex consciousness structures have great difficulties in taking the roles of others, which might give confusing experiment results. Given the time- and resource-consuming character of developing and using simulations in relation to their draw-backs, I regard this approach as a low priority option for my own research.


Text analysis

Suedfeld, Tetlock and their associates have made a large number of investigations into political psychology using a text analysis method based on the integrative complexity framework (see Baker-Brown et al., 1992). This method can be applied to statements, declarations, newspaper articles, transcribed speeches, etc. One important advantage of this approach is the possibility of studying historical events. Another is the possibility of getting access to high-level decision makers not normally available for clinical interviewing. The obvious disadvantage is the lack of interactivity, which means that the researcher cannot probe for underlying structures, but must content her-/himself with what is available. This means that the analysis must remain superficial in relation to such dimensions as Kegan’s subject-object balances and personal motivation. Only by accident it is possible to gain access to a comprehensive view of the ingroup/outgroup images of the subjects.



Text analysis may be an interesting complement to more interactive methods, but seems unsuitable as the major method in studies of the deep structures of geopolitical reasoning.



The survey of methods in this section shows that for my purposes, a variant of the situated subject-object interview is the most promising method. This method can be adapted to different settings, it does not require that interviewees agree to fill out psychological tests, and it does not reduce the depth of information. Among the drawbacks the high sensitivity to the biases of the interpreting researcher is the most important. This drawback can be partly offset by independent and blind analyses by two researchers. However, the very nature of this research is hermeneutic, so it is an illusion to strive to eliminate the interpretative moment.

In order to gain more precision in the research project, it might be fruitful to complement an investigation of a real-life geopolitical conflict with more controlled experiments involving voluntary test subjects and fictitious cases. These experiments could allow use of a combination of SCT’s and Kohlberg-type interviews.



5. Interviewing geopolitical subjects


Case studies

In this concluding section I will briefly outline what a concrete research project using the approach outlined above might look like. This includes a preliminary interview guideline intended for interviews with persons involved in ongoing geopolitical conflicts.


A preliminary interview guideline

Below I have compiled a preliminary version of a generic interview guideline. [19] There are two important considerations regarding this outline. First, "generic" means that the questions listed here must be reformulated to fit into the chosen case. In order to elicit useful information, questions should refer to important concrete conflict issues, named parties to the conflict, and recent and significant political events. Secondly, the key feature of this interview technique is probing for how the respondent makes meaning in a deeper sense. Most replies will therefore need follow-up questions tailored to probe for how certain parties, events, opinions, etc. are constructed. In the guideline I have indicated the most important issues to probe for. The interviewer must consequently have an acute sense for the structural nature of the conversation – a difficult task. This need for reflecting on answers while formulating follow-up questions is a strong argument for trying to keep the option open for making two interviews, or at least for getting the respondent to agree to clear certain issues by a later telephone call. The e-mail format would make the task easier, but is normally not practicable due because too few participants in conflicts have access to e-mail.


Generic interview guideline

The respondent

- Personal biography: political affiliation, role in organizations

Follow-up questions on:

Opinions about critical political issues and recent events

The conflict

- What is the ingroup-outgroup conflict about? Issues?

Follow-up questions on:

Explore multidimensionality

- Which are the most important reasons for the conflict?

Follow-up questions on:

Who/what is responsible for the present state of affairs?

Explore understanding of causal relationships behind concrete features

of the conflict (e.g. recent events, salient aspects of behaviour)

Explore the extent of openness/closure in reasoning about causality

The outgroup(-s)

- Who is the counterpart?

Follow-up questions on:

Explore extent of multidimensionality in characterizing the counterpart.

Explore extent of attribution of collective properties to outgroup


Explore heterogeneity/homogeneity of counterpart image

Explore criteria for membership: concrete vs. abstract properties

Explore openness/closure of counterpart image

- What are the most important goals of the counterpart?

Follow-up questions on:

Explore tolerance for ambiguity, contradiction; differentiation of image

- How are strategical decisions made?

Follow-up questions on:

Explore reasoning about decision-making (complexity); leadership roles;

internal structure of counterpart.

The ingroup

- What is the meaning to you of being [ingroup name]?

Follow-up questions on:

Explore membership criteria (concrete-abstract properties)

Explore the relative importance of different scales of identity groups

(family, direct relationships, organizations, symbolic communities)

Explore conception of identity: ultimate values

- What are the most important geopolitical goals of [group name]?

Follow-up questions on:

Basis of justification of goals;

explore egocentric/sociocentric/worldcentric values

Explore ability to reflect on the ingroup perspective and belief system

- What is your opinion about your own leaders and institutions?

- How do you interpret the importance of internal conflicts?

Explore degree of differentiation of collective self-image

The relationship between ingroup and outgroup

- What is the nature of the relationship between [your group] and the counterpart?

Follow-up questions on:

How would you characterize your feelings toward the counterpart?

Explore differentiation of affects in relation to counterpart

Conflict resolution

- How can the differences be resolved?

Follow-up questions on:

Which kind of process/action is most likely to lead to a stable solution?

[Use this to explore perspective-taking abilities and capacity to

coordinate perspectives]

Do you have any idea on desirable territorial arrangements?

What role do boundaries play?

Explore propensity to prefer separation and rigid boundaries



1. See also my previous study on the psychological functions of territoriality: Jordan, 1996, 1997b. back

2. I am grateful for constructive criticism on an earlier version of this text from Dr. Tara Santmire (Department of Government and Politics, University of Maryland). Many of her suggestions remain to be acted upon. back

3. It should be noted that a large number of researchers has followed this path, including Piaget; Gebser (1949); Bois (1955); Loevinger (1976); Kohlberg (1981); Selman (1980); Kegan (1982, 1994); Rosenberg (1988); Wilber et al. (1986); Whitmont (1982); Lauer (1983); Torbert (1987); and many others. back

4. Comprehensive introductions to Piaget’s theories (and their possible implications for social theory) can be found in Rosenberg, 1988, and Oesterdiekhoff, 1992. back

5. See the discussion and references in Oesterdiekhoff, 1992, p. 285ff, and footnotes 1033 and 1034 on p. 285. For critical discussions and references, see Krebs & van Hesteren, 1994:108 and Lahey, 1986. back

6. Piaget’s model has also encountered a lot of criticism, in particular his strong formulations about the "jump" character of transformations from one structure to the next, and the assertion that individuals consistently operate from one single cognitive structure in all different contexts. See Lahey, 1986, ch. 2. Another line of criticism has concerned the atomistic nature of Piaget’s entire theory. Critics (often referring to Vygotsky) argue that understanding is a social process, not an individual structure. For a discussion on the relationship between Piaget’s and Vygotsky’s approaches, see Bruner, 1997. back

7. See e.g. Eriksen, 1992. back

8. In addition to these four I have considered several other approaches, e.g. that of Loevinger (Loevinger & Wessler, 1970; Loevinger, 1976; Hy & Loevinger, 1996; Kohlberg (1981), Selman (1980), and Lauer (1983). These have certainly contributed to my general understanding of the field, but they seemed to be less easily adapted to my purposes. back

9. Wilber (1997:339f [footnote 9]) pinpoints some of the differences: "This ‘consciousness axis’ is vaguely similar to cognitive development, but they are not simply the same thing, especially given the biases in cognitive research, which include: (1) an almost exclusive emphasis on it-knowledge, which is called ‘cognition,’ and which in fact leaves out the I and we aspects of consciousness; (2) a consequent overemphasis on the acquisition of scientific it-knowledge as the central axis of development, with a concomitant attempt to (3) define a central axis in terms of Piagetian logico-mathematical competence; (4) a consequent failure to count emotions (and prana) as a mode of consciousness; (5) an almost total ignoring of the transrational structures of consciousness." back

10. Tetlock and Suedfeld and their associates have been very productive, and I cite only some of their more substantial statements: Baker-Brown et al., 1992; Suedfeld, 1994; Suedfeld and Tetlock, 1977; Suedfeld et al., 1993; Tetlock, 1984; 1985; 1988; Tetlock et al., 1994; Wallace et al., 1993. back

11. See, however, Pamela Steiner’s recent doctoral dissertation on the relationship of Kegan’s orders of consciousness to decision-making processes in the political sphere (Steiner, 1996). back

12. In order to reduce possible sources of confusion I have replaced some of the concepts used by SDS with concepts consistent with the terminology used elsewhere in this paper. back

13. Tara Santmire is a researcher at the Department of Government and Politics, University of Maryland, conducting empirical research based on the integrative complexity framework. back

14. See also the comprehensive discussion in Suedfeld and Tetlock, 1992. back

15. This conception is not consistent with Daniel Stern’s (1985) seminal analysis of the psychological birth of the self during infancy. However, this inconsistency is of minor importance to the usefulness of Kegan’s framework, since Kegan’s main focus is on adolescents and adults. Stern’s pathbreaking research might imply a need for modifying Kegan’s framework somewhat, esp. by differentiating more clearly between cognitive processes and sensomotorical intelligence in the first year of life. back

16. Wilber’s framework is in itself a synthesis of a great number of scholarly contributions and spiritual traditions. back

17. Many scholars refer to correlations between the stages of different models, some even try to specify some theoretical links (e.g. between general cognitive operations and moral reasoning), but noone has taken on the daunting task of specifying a detailed theoretical framework for how different dimensions relate to each other. Wilber (1997:212) enumerates 16 candidates for such dimensions. back

18. I would guess that many psychoanalytically inclined observers would take this position, see, for example, deMause, 1982. back

19. I have borrowed some ideas from the interview guideline in Lindholm Schulz’s (1996) study of Palestinian nationalisms. back



ARIETI, S. (1967) The Intrapsychic Self. Feeling and Cognition in Health and Mental Illness, New York: Basic Books, 1967.

AZAR, E. E. (1990) The management of protracted social conflict: Theory and cases, Aldershot, England: Darthmouth.

BAKER-BROWN, G., BALLARD, E. J., BLUCK, S., DE VRIES, B., SUEDFELD, P., and TETLOCK, P. (1992) ‘The conceptual/integrative complexity scoring manual,’ in C. P. Smith (ed.) Motivation and personality handbook of thematic content analysis, pp. 400-418, Cambridge (UK): Cambridge University Press.

BOIS, J. S. (1955) ‘The epistemological profile and semantic psychoanalysis,’ General Semantics Bulletin, no 18-19.

BRUNER, J. (1997) ‘Celebrating divergence: Piaget and Vygotsky,’ Human Development, vol 40, pp. 63-73.

deMAUSE, L. (1982) Foundations of psychohistory, New York: Creative Roots.

EDER, K. (1991) Geschichte als Lernprozeß ? Zur Pathogenese politischer Modernität in Deutschland, Frankfurt a. M.: Suhrkamp.

ELIAS, N. (1971) Ü ber den Prozeß der Zivilisation. Soziogenetische und psychogenetische Untersuchungen, Frankfurt: Suhrkamp.

ERIKSEN, T. H. (1992) Us and Them in modern society, Oslo: Scandinavian University Press.

FISHER, D. and TORBERT, W. R. (1995) Personal and organizational transformations: the true challenge of continual quality improvement. London and New York: McGraw-Hill.

FISHER, R. and URY, W. (1981) Getting to yes. Negotiating agreement without giving in, Boston: Houghton Mifflin Co.

GEBSER, J. (1949/1992) Ursprung und Gegenwart. 1. Teil: Die Fundamente der aperspektivischen Welt, München: Deutscher Taschenbuch Verlag.

GILLIGAN, C. (1982) In a Different Voice. Psychological Theory and Women’s Development, Harvard University Press.

GLASL, F. (1997) Konfliktmanagement. Ein Handbuch fü r Fü hrungskräfte, Beraterinnen und Berater, 5. Auflage. Bern: Paul Haupt Verlag.

GOLDBERG, S. B., SANDER, F. E. A. and ROGERS, N. H. (1992) Dispute resolution. Negotiation, Mediation, and other processes. Second edition, Boston: Little, Brown and Company.

GUTTIERI, K., WALLACE, M. D. and SUEDFELD, P. (1995) ‘The integrative complexity of American decision makers in the Cuban missile crisis,’ Jorunal of Conflict Resolution, Vol 39:4, pp. 595-621.

HABERMAS, J. (1976) Zur Rekonstruktion des Historischen Materialismus, Frankfurt a. M.: Suhrkamp.

HARVEY, O.J., HUNT, D. E. and SCHRODER, H. M. (1961) Conceptual systems and personality organization, New York and London: John Wiley and Sons.

HY, L. X. and LOEVINGER, J. (1996) Measuring ego development, Second Edition, Mahwah, N.J.: Lawrence Erlbaum.

JORDAN, T. (1996) ‘La psicología de la territorialidad en los conflictos’, Psicología Politíca, No 13, pp. 29-62.

JORDAN, T. (1997a) ‘Achtsamkeit im Umgang mit Konflikten: ein Weg zur Bewußtseinsentwicklung’, Transpersonale Psychologie und Psychotherapie, 1997:1, pp. 78-101.

JORDAN, T. (1997b) ‘Territorialität und ihre Funktionen in Konflikten – Eine psychogeographische Betrachtung,’ in P. Jü ngst (ed.) Identität, Aggressivität, Territorialität. Zur Psychogeographie - und Psychohistorie - des Verhältnisses von ‘Sub’jekt, Kollektiv und räumlicher Umwelt, Urbs et Regio 67, Kassel, pp. 35-75.

KEGAN, R. (1982) The evolving self. Problem and process in Human Development, Cambridge (Mass.) and London: Harvard University Press.

KEGAN, R. (1994) In over our heads. The mental demands of modern life, Cambridge (Mass.): Harvard University Press.

KOHLBERG, L. (1969) ‘Stage and sequence: The cognitive-developmental approach to socia-lization’, in D. A. Goslin (ed.) Handbook of socialization theory and research, Chicago: Rand McNally and Company.

KOHLBERG, L. (1971) ‘From is to ought: How to commit the naturalistic fallacy and get away with it in the study of moral development’, in T. Mischel (ed.) Cognitive Development and Epistemology, New York and London: Academic Press.

KOHLBERG, L. (1981) Essays in moral development, vol. 1: The philosophy of moral development, New York: Harper and Row.

KREBS, D. L. and VAN HESTEREN, F. (1994) ‘The Development of Altruism: Toward an Integrative Model’, Developmental Review, 14, p. 103-158.

LAHEY, L., (1986) Males’ and females’ construction of conflict in work and love, Thesis presented to the faculty of the Graduate School of Education of Harvard University.

LAHEY, L., SOUVAINE, E., KEGAN, R., GOODMAN, R., FELIX, S. (1988) A Guide to the Subject-Object Interview: Its Administration and Interpretation, Cambridge: Subject-Object Research Group.

LAUER, R. M. (1983) ‘An introduction to a theory of adult development or, After Piaget, what?,’ in M. Levy (ed.) Research and theory in developmental psychology, New York: Lexington.

LINDHOLM SCHULZ, H. (1996) Between revolution and statehood: Reconstruction of Palestinian nationalisms, [doctoral dissertation] Göteborg: Padrigu Papers.

LOEVINGER, J. (1976) Ego Development, San Francisco: Jossey-Bass Publishers.

LOEVINGER, J. and WESSLER, R. (1970) Measuring ego development, vol I: Construction and use of a sentence completion test, San Francisco: Jossey-Bass.

MINDELL, A. (1995) Sitting in the Fire. Large group transformation using conflict and diversity, Portland: Lao Tse Press.

NEUMANN, E. (1949/1990) Depth psychology and a new ethic, London: Hodder & Stoughton.

NEUMANN, E. (1949/1970) The origins and history of consciousness, Princeton University Press, Bollingen Series XLII

NEUMANN, E. (1973/1990) The Child, Boston: Shambala Publications.

OESTERDIEKHOFF, G. W. (1992) Traditionales Denken und Modernisierung. Jean Piaget und die Theorie der sozialen Evolution, Opladen: Westdeutscher Verlag.

O'TUATHAIL, G. (1996) Critical geopolitics. The politics of writing global space, Minneapolis: University of Minnesota Press.

PAASI, A. (1996) Territories, boundaries and consciousness. The changing geographies of the Finnish-Russian border, Chichester: John Wiley & Sons.

PHARES, J. O. and SCHRODER, H. M. (1969) Structural scoring manual for paragraph completion test, Princeton (N.J.): Princeton University.

ROSENBERG, S. (1988) Reason, Ideology and Politics, Cambridge: Polity Press.

ROSENBERG, S., WARD, D. and CHILTON, S. (1988) Political Reasoning and Cognition. A Piagetian View, Durham and London: Duke University Press.

SANTMIRE, T. E., WILKENFELD, J., KRAUS, S., HOLLEY, K. M., SANTMIRE, T. E. AND GLEDITSCH, K. (1998) Differences in cognitive complexity levels among negotiators and crisis outcomes, Paper presented at the 39th Annual Meeting of the International Studies Association, March 18-21 1998, Minneapolis MN.

SCHRODER, H. M., DRIVER, M. J. and STREUFERT, S. (1967) Human Information Processing. Individuals and Groups Functioning in Complex Social Situations, New York: Holt, Rinehart and Winston, Inc.

SELMAN, R. L. (1980) The growth of interpersonal understanding. Developmental and clinical analyses, New York: Academic Press.

SPILLMANN, K. R. and SPILLMANN, K. (1991) ‘On enemy images and conflict escalation,’ International Social Science Journal, Vol 127, p. 57-76.

STEINER, P. P. (1996) Conforming and non-conforming concurrence: Aspects of "groupthink" and orders of consciousness in democratic decision making, Thesis presented to the faculty of the Graduate School of Education of Harvard University.

STERN, D. (1985) The Interpersonal World of the Infant: A View from Psychoanalysis and Developmental Psychology, New York: Basic Books.

SUEDFELD, P. (1994) ‘President Clinton’s policy dilemmas: A cognitive analysis,’ Political Psychology, Vol 15:2, pp. 337-349.

SUEDFELD, P. and TETLOCK, P. E. (1977) ‘Integrative complexity of communications in international crises,’ Journal of Conflict Resolution, Vol 21, pp. 169-184.

SUEDFELD, P. and TETLOCK, P. E. (1992) ‘Psychological advice about political decision making: Heuristics, biases, and cognitive defects,’ in P. Suedfeld and P. Tetlock (eds.) Psychology and Social Policy, pp. 51-70, New York: Hemisphere Publishing Corporation.

SUEDFELD, P., WALLACE, M. D. and THACHUK, K. L. (1993) ‘Changes in integrative complexity among Middle East leaders during the Persian Gulf crisis’ Journal of Social Issues, Vol. 49:4, pp. 183-199.

TAYLOR, P. J. (1993) Political geography: World-Economy, Nation-State and Locality, London: Longman.

TETLOCK, P. E., ARMOR, D. and PETERSON, R. S. (1994) ‘The slavery debate in antebellum America: Cognitive style, value conflict, and the limits of compromise,’ Journal of Personality and Social Psychology, Vol 66:1, pp. 115-126.

TORBERT, W. R. (1987) Managing the corporate dream, Homewood, IL: Dow Jones-Irwin.

VOLKAN, V. D. (1988) The need to have enemies and allies: From clinical practice to international relationships, Dunmore, Pa: Jason Aronson.

WALLACE, M. D., SUEDFELD, P. and THACHUK, K. L. (1993) Political rhetoric of leaders under stress in the Gulf crisis,’ Journal of Conflict Resolution, Vol. 37:1, pp. 94-107.

WALLENSTEN, P. (1994) Från krig till fred. Om konfliktlösning i det globala systemet, Eskilstuna: Almqvist & Wiksell.

WARD, D. (1988) ‘The structure of the idea of democracy in Eastport,’ i S. Rosenberg, D. Ward and S. Chilton: Political Reasoning and Cognition. A Piagetian View, Durham and London: Duke University Press.

WHITMONT, E. C. (1982) Return of the Goddess, New York: Crossroad.

WILBER, K. (1980) The Atman project, Wheaton, The Theosophical Publishing House.

WILBER, K. (1981) Up from Eden, Boston: Shambala Publications.

WILBER; K. (1995) Sex, Ecology, Spirituality. The Spirit of Evolution, Boston and London: Shambala Publications.

WILBER, K. (1997) The Eye of Spirit, Boston and London: Shambala Publications.

WILBER, K.; ENGLER, J. and BROWN, D. P. (1986) Transformations of consciousness, Boston and London: Shambala Publications.