Observing International Relations takes the recent emphasis on globalization as an opportunity to explore potential links between Niklas Luhmann’s Modern Systems Theory (MST) of “world society” and the international relations (IR) approach. The hope is to provide new perspectives on reoccurring issues of contemporary world society. In the introductory section editor Mathias Albert describes the problem – “world society” dynamics features prominently in international relations (IR) conceptualization, but are not satisfactorily handled by the traditional Westphalian state-oriented models. The thrust of the book is to investigate of the utility of Niklas Luhmann’s MST and its concepts, as a framework to understand IR issues and to bridge to alternative IR theories. This seems plausible, if only because MST, in contrast to a fragmented, bottom up approach of current IR theories, takes world society on top as ontologically primary. Albert also argues that a review of MST will be particularly useful to Anglophile audiences because relatively few of Luhmann’s books have been translated into English. The collection of edited articles is divided into 3 parts. Part 1 called “Luhmann and IR: A worthwhile encounter?” addresses this question providing diverging opinions of the utility of bringing MST and IR together. Part 2 “ Competing Notions of World Society and World Society as the ‘Largest Social System Possible’” develops the Luhmannian theory of world society and contrasts it with competing notions of international society, critically discussing the use of MST in international relations theory. Part 3 called “Bringing Modern Systems Theory to the Study of IR: Concepts and questions” more directly confronts MST’s core concepts summarized in Part 1 assessing its treatment of central concepts within IR, such as power, sovereignty, governance and war. Taken as a whole the book will particularly appeal to international relations specialists, historians and sociologists for its application of various theories to conceptualizations beyond the “post-Westpahian world”. However, it is a more difficult road for those looking for a succinct summary of MST, because MST theory is embedded in a breadth and depth of IR material. Chapters need often simultaneously introduce the more abstract structure of MST, its methodological instruments, its embodiment in social theory as well as covering and contrasting traditional and alternative IR theory with these while seeking for points of integration. As a result of this broad span, readers may find a chapter excellent in one aspect of MST but less so in others. For the reader Albert’s first 2 chapters are particularly challenged in this regard because they have to introduce weaknesses in IR theory as well as introducing relevant MST concepts. In chapter 2 (On the Modern Systems Theory of Society and IR) Albert pithily overviews Luhmann’s MST fundamental concepts and approach to social systems this way:

“For MST, all social systems are constituted by a difference between system and environment and are communicative systems ..Communication forms the basic operation of social systems … In contrast to sender-receiver models which ascribe communicative ‘acts’ to persons, perlocutors… etc., communication here is conceptualized as being produced and reproduced in recursive networks of communication. A recursive network defines the unity of a system. Taking up the insights on self-referentiality and autopoiesis originally developed in the natural sciences by Maturana, Varela and others, communication is thus seen as being produced within the system alone. No communication ‘enters’ a system save it being observed and selected and thus ‘produced’ within the system itself. ‘Action’ and ‘causality’ in such a view then do not form basic non-communicative processes in society, but rather are merely forms of observation and communicative ascriptions of action and causality within social systems. If social systems are constituted by communication and by communication only, then society is the highest-order social system which comprises all communication. There is no communication outside of society or between society and systems in its environment. In order to establish what society as the comprehensive system being formed by and comprising all communication ‘is’, it is thus not possible to draw on any form of membership or a population, a geographic feature, or for that matter, any externality not constituted by communication: the notion of ‘society’ in modern systems theory is ‘radically constructivist’”

Buried in this section are many of the key relations to IR detailed in subsequent articles — communication, differentiation, the role of observation etc. These primarily reflect Luhmann’s later systems theory where the observer-relative viewpoint is seen as inadequate to understand society as a unified entity “in itself.” Society is a self-organizing communication network with an autopoietic unity leading to differentiated autonomic sub-systems through communication. That is, society changes of its own structures, producing its own components based on autopoietic differentiation. To Luhmann such differences are generated by communication, which also means that world society produces and reproduces itself via that medium. Society is able to “produce itself” by reproducing its elements while maintaining an organization of these elements. Thus MST proposes that the way to obtain a “true” understanding of society system – by focusing on the process of self-production and self-organization rather than the ‘classical notion’ of sociological theories which see society as something ‘held together’ by shared norms and a collective identity. The utility of this idea is perhaps the single biggest reoccurring topic throughout the book. As is clear from the title of the book the concept of observation is also central to MST and refers to the handling of distinctions. Luhmann draws heavily on George Spencer Brown concept of ‘observation’ in terms of distinctions covered in The Law of Forms. Thus, making a difference/distinction in MST involves an observing system. Society’s distinctions lead from a pre-existing, united world society to ‘functional subsystems’ — functions like law, economics, military and politics. These functional subsystems are each closed, operating according to internal logics, that Luhmann calls ‘media of communication’. These media in turn generate ‘codes’ – for example, military codes versus financial codes. Functional differentiation and observation are of obvious importance to IR. Starting with Albert’s chapter we are exposed to MST’s constructive concept of communication involving a selection process. Selection of “information” (what the communication is about) comes from a repertoire of possibilities along with behavior that expresses this communication and “understanding,” which involves a sense of meaning that a communication generates. In this process misunderstanding even paradoxes may also arise in a “receiver.” Although functional sub-systems are autonomous, selection types must be “coupled” for communication to appear as an emergent occurrence seen less as a process of growth, then as a process of differentiation and structural coupling.

These ideas are developed and illustrated in different chapters and much has to be put together by the reader. Searching for core MST concepts in the text is made difficult by their absence in the index. Missing topics for a person interested in Luhmann’s main concepts are indexes to such things as “code, communication, differentiation, ontology, structural coupling, and world opinion.” Each is well discussed in places, but assembling these can be challenging and a reader seeking a succinct summary the collection won’t easily find it since key concepts tend to be distributed across the volume. For example, functional differentiation, one of the core concepts showing how a world system changes its structure, is well discussed, but chiefly in Dieter Kerwer’s “Governance in a World Society: The perspective of systems theory” (chapter 13).

I found Gorm Harste’s “Society’s War: The evolution of a self-referential military system”, for example, a surprisingly good application of Luhmann’s theory of media communication and structural coupling concepts to explain how a war and peace code like “struggle” has evolved using self description to characterize not only itself but the larger world. The history covered ranges from “selecting” Hoplite formations to “selecting” the atomic bomb with a key change occurring in the 17th century as the “military system” observed itself by means of specialized (differentiated) self descriptions that in turn evolved to new military codes. A key notion of autopoiesis theory and of Luhmann’s social systems theory covered in this chapter is that of self-description (also translated as self-reference) —designating the unity that an element, a process, or a system provides for itself. Self-reference is independent of observations by others, and implies that unity, or “self,” can only be produced through relational operations.

Chapter 3. Politics, Modern Systems Theory and the Critical Purpose of International Relations Theory by Thomas Diez discusses Luhmann’s radical constructionist view of observation — “the role of science is to observe”. Thus, the status of ontology and knowledge in MST is quite different from most IR approaches. As a closed system/system environmental disturbances do not exist, as such, so it is the sub-system that determines what is, or is not an influence. This leaves us with a view of society very different from the traditional one. Society is essentially without a core or fundamental division driving it, and there is no privileged position from which one single, rational global view dominates. Instead we have self-defined autonomous sub-systems in a constant process of renewal and redefinition. These are described in terms of system harmonization proposed as a process relatively insensitive to society’s external environment. This view, and the fact that the language of MST is a closed system, leads Diez to see MST’s initial value as a means of discoursing about IR rather than directly adding to it.

Stefan Rossbach’s chapter is called “Corpus Mysticum”: Niklas Luhmann’s evocation of world society”. It further explores the anti-essence ontological perspectives of MST in a critical fashion, calling them ontological metaphysics. Readers will find a critical but interesting history of the concept of autopoiesis which Rossbach relates to Gnosticism. In many ways the chapters which attack some of MST’s core concepts also provide the clearest view of these concepts and I found them as a whole major contributors to a better understanding of the MST superstructure apart

Part 2 on Competing Notions of World Society starts with Chris Brown’s skeptical chapter, “The ‘English School’ and World Society”. Brown is largely skeptical about the links that might be made between MST and the ‘English School’ ( ES) as well as its utility for understanding “international society”. Brown starts by highlighting ES’ revitalized idea of primary institutions as the central contribution of ES theory differentiating it from IR realism and neoliberal institutionalism. Considering Luhmann’s basic premises of functional differentiation and its resulting subsystems, there seem to be useful parallels in the newer aspects of ES theory. However, Brown challenges this and several central features of Luhmann’s theory as they apply to IR. First, he disputes the basic assumption of MST placing world society. Brown presents ES’s arguments that the states are ontologically prior to other societal forms such as the world society. He also challenges Luhmann’s idea that society is not normatively integrated through human interactions. To Brown, MST and the ES approach have unbridgeable issues because 10 MST erodes the central ES system-society distinction and 2) fails to address important processes such as human interactions from action-centered theory.

Sociological Institutionalism and the Empirical Study of World Society by George M. Thomas discusses cultural/institutional structures, such as in international governmental organizations. These Thomas views through sociological institutionalism (SI) theory. While SI is constructionist and does not view world society as a bottom up integration, it is not a radical constructionist approach as is MST. Instead SI theory sees change driven not by differentiation but by rationality as part of institutions in conformance to world culture. Thomas’ chapter, in contrast to the others, is a balanced mix of theory, prediction and empirical results.

Lothar Brock in “World Society from the Bottom Up” attempts to cojoin bottom up processes with MST’s top down view of society across 3 issues: world society formation as cumulative change, the conceptualization of change and the interrelationship between society and community formation beyond the state. As a starting observation he sees some similarity in the idea of differentiation in both approaches. However, in reality there are substantial differences. World society formation is seen as less a fact than an emerging reality taking place through bottom up rationalization of social relations. In this view world society is emerging as “debordorization” and “denationalization” proceeds. In MST’s top down stance, the complexity of social change is viewed as due to systemic differentiation while in a bottom up approach this change is seen as arising from deficiencies in lower parts. Finally, in considering the role of community in forming world society, Brock considers the normative aspects important to IR and finds MST’s abstractions lacking in the role of symbolic integration. In summary the bottom up approach gives more space to agent-based diversification and thus complements MST.

In “World Society, Systems Theory and the Classical Sociology of Modernity” Dietrich Jung explores the roots of world society in theories of modernity supplemented by Habermas’ paradigm of system and lifeworld. In opposition to MST Jung finds a central role for rational analysis and norms.

Starting Part 3 Anders Esmark’s “Systems and Sovereignty: A systems theoretical look at the transformation of sovereignty “ considers some issues in MST’s idea that sovereignty arises from functional differentiation. As background Esmark locates Luhmann’s semantic in terms of Spenser Brown’s Law of Forms, a logic of semantics. This is then applied to discuss the meaning of sovereignty understood through MST’s concept of “Paradoxification”. Because MST does not accept law as a hierarchy of rules with lower normative acts legitimated by higher level rules functional differentiation of the meaning and semantics of a concept like sovereignty may arise through communication paradoxes. For example, the sovereignty of law arises in response to certain political paradoxes inherent is certain forms of differentiation. The Law of Forms also allows the process to be understood, seeing semantics in the background and social structure as foreground. When observed this way the meaning of sovereignty arises from territorial segmentation.

“World Opinion’ and the Turn to Post-sovereign International Governance” by Hans-Martin Jaeger provides a history of how world opinion, and to a lesser extent sovereignty, emerged from the medium of 19th and 20th century communication. Jager’s discussion makes clear how communication on “world opinion” determines what can be observed by the political system.

As previously noted Gorm Harste “Society’s War: The evolution of a self-referential military system” is informative on a number of points including structural coupling, the nature of self-reference and how codes evolve within communications. Harste makes a potent argument that rather than states being the primary sovereign systems arising in the post monarchy world, it was more likely the abstract systems of finance, law, the military etc which were primary. This chapter serves informally as a bit of an integration point to prior sections.

“Organizations in/and World Society: A theoretical prolegomenon by Mathias Albert and Lena Hilkermeier serves as a further integration point reviewing MST and neo-institutional accounts of world society. Noting the advantages of each (for example MST’s coherence and neo-institutional approach’s ability to integrate data) they conclude that while we might not be ready to integrate these 2, their relative strengths can be cojoined.

In Dieter Kerwer “Governance in a World Society: The perspective of systems theory “ we have a very critical, but I think fair assessment, of some of MST abstract concepts. Kerwer proposes that MST main contribution to the issues of governance is negative — governance is unlikely to succeed with world society.

“Constructivism and International Relations: An analysis of Luhmann’s conceptualization of power by Stefano Guzzini is another excellent short chapter that applies a constructionist stance to MST’s concepts of communication and self reference. Not only is theory succinctly covered well, but the application to the evolution of power is enlightening. The book concludes with a short remarks section by Mathias Albert underlining areas of agreement and disagreement and pointing to future work.

Taken as a whole, readers interested in Luhmann’s work on world society will find this a challenging but thoughtful set of chapters, covering difficult topics in individually thoughtful ways.

Feature image (rear cover)

Robustness and networks

My recent research has been concerned with the relationship between the structure of a complex system and its dynamics. The data shown here comes from a computer experiment I performed on one million random Boolean networks – which are relatively ‘simple’ complex systems and yet can display all the trademark behaviors of complexity. This particular experiment was to look at the relationship between the robustness of a particular Boolean network and its reduced form (or its dynamic core). The reduced form of a Boolean network is simply the part of the network that is responsible for the gross characteristics of the network’s phase space. For example, if a network’s phase is characterized by three period-four attractors, then so will the network’s reduced form. Details of how to find the reduced form a complex network can be found in Richardson (2005). The robustness of a Boolean network is a quantitative measure of how sensitive the network’s qualitative behavior is to small external perturbations. So, a network for which any external perturbation whatsoever causes a bifurcation – i.e., a jump from one attractor basin into another – will have a robustness value of zero. At the other extreme, a network that is totally insensitive to external perturbations – i.e., a network with only one attractor basin – will have a robustness value of one. The image shows the robustness of the unreduced networks (x-axis) compared to the reduced networks (y-axis). The coloring refers to the frequency of a particular data point. For example, the blue points refer to an occurrence of 20 – or 1 – data point, whereas the red points refer to an occurrence equal or greater than 210 – or 1024 – data points. The black points refer to the average for a particular value of unreduced robustness.

The data show at least two important features. The first is that, although the average clearly shows that the reduced robustness is generally lower than the unreduced robustness, it also clearly shows the dangers of relying on averages when doing nonlinear data analysis. This is because the data is not mono-modal and so the average obscures other interesting artefacts within the data. The message: Beware of averaging! The second feature, then, is the multi-modal nature of the data. On further inspection I found that the fan-like structure results from networks with different dynamic core sizes. The smaller the dynamic core, the greater the reduction in robustness is when the network is reduced. So, although a network reduced to its dynamic core displays the same qualitative range of behaviors – i.e., it is functionally the same – it tends to be less robustness when exposed to external perturbations – i.e., it becomes increasingly qualitatively unstable. It would seem from this simple analysis that complex networks comprise at least two different types of element: those that contribute to a network’s function, and those that provide a buffer – or container – so that the network might function as ‘designed’ without being at the mercy of all environmental signals. I’m sure this simple picture will become more complex as the research progresses. The initial results are being written-up for publication and there is a draft paper available on request (Richardson, 2006).

Kurt A. Richardson

[email protected]