1.1 Introduction
Social scientists are not accustomed to thinking in terms of systems. Although the concept formed part of a theoretical consensus in the mid-twentieth century, at least amongst a subset of mainly American social theorists, the notion has been largely purged from the Anglo-Saxon canon since the 1970s. There are many reasons for this. In reaction to the alleged scientism of social systems theory, coupled with the postmodern cultural turn, systems theory came to be viewed as a conservative relic. It justifiably came under criticism for its neglect of historical context and specificity, its inability to handle change, its normative content suggesting that given social relations represented equilibrium conditions – and consequently, its implicit endorsement of inequality. And yet the history of systems theory in the twentieth century is relevant for several reasons. Its ambition mirrors post-war optimism in other realms such as public welfare reform, and its interdisciplinary ambition reveals a remarkable degree of collaboration between disciplines which have become more separate today.
It also raises important questions for understanding natural-social interdisciplinarity today – why did the general systems theory programme not succeed, and why does interdisciplinarity appear further today than it did since the post-war era? The fate of systems theory over the twentieth century also parallels the fragmentation of sociology into sub-specialties over this period, with many based on a clear rejection of systems principles. Understanding this process is therefore key to appreciating why cross-discipline communication remains difficult today. This chapter outlines this history of ideas, focusing mainly on twentieth century social theory. It suggests that by separating the objectives of structural and conceptual modelling from the identification of functional requisites and relations for system functioning – as well as misguided applications of the concept of equilibrium – we can recast the interdisciplinary promise of systems theory from a new perspective of complexity.
1.2 The ‘First Wave’ of General Systems Theory
According to Graeme Snooks, ‘…the holy grail of systems theory is the development of a general dynamic theory that can explain and predict the emergence of order and complexity in a universe of increasing entropy’ (2008: 12). Skyttner describes general systems theory as operating on an abstract level ‘…with general properties of systems, regardless of physical form or domain of application…as an epistemology [which] structures not only our thinking about reality but also our thinking about thinking itself’ (Skyttner 2005: 40). Systems theory in its most general sense is therefore a multifaceted enterprise – a theory of structure (the essential concrete elements of a social or physical system), function (the inter-relation between these elements), and epistemology (a model of how knowledge of the system is best acquired through analysis). This includes more recognisable works such as Kenneth Boulding’s Hierarchy of Systems Complexity (pub. 1956), James Grier Miller’s General Living Systems Theory (pub. 1978), and later works such as James Lovelock’s Gaia Hypothesis (pub. 1979), Kenneth Bailey’s Social Entropy Theory (Bailey 2006, 2008), and Immanuel Wallerstein’s World Systems Theory (Wallerstein 1974, 2006). Our brief history begins with what may be referred to as the ‘first wave’ – in some respects the most ambitious – born of collaboration between scientists of several different backgrounds.
Thomas Kuhn (1962) suggested that the history of scientific practice is that of a succession of revolutionary paradigms, in which new modes of thought replace old, incompatible forms. A ‘normal science’ of consensus amongst practitioners is defined by Kuhn as one which ‘…like an accepted judicial decision in the common law, is an object for further articulation and specification under new or more stringent conditions’ (1962: 23). Kuhn’s view of paradigm succession is one of ‘consensuses followed by “crisis”’, through which practitioners divide allegiances between the new and the old. The success of the new paradigm is dependent on the explanatory capabilities of the new, giving rise to a period of ‘normal science’. These dominant ‘normal science’ paradigms are irreconcilable with the views of their predecessors (1962: 103).1 The history of systems-based thinking does not quite fit this neat model of succession, although some have suggested that a growing multidisciplinary emphasis on systems and complexity theory may slowly be creating a new ‘Kuhnian’ paradigm (Urry 2005; Castellani and Hafferty 2009: 119). The lure of a coherent research programme underpinned by an encompassing theory, is evident within the works of early general systems theorists, and its resilience in various forms across the twentieth century attests to its lingering appeal, if not it’s practical utility.
Writing in 1956, in response to Ludwig von Bertalanffy’s General System Theory: A New Approach to Unity of Science, (1951) Kenneth Boulding described general systems theory as a ‘…skeleton of science…on which to hang the flesh and blood of particular disciplines’ (1956: 208). The ‘skeleton’ of systems theory to which Boulding referred was generating some interest across numerous disciplines as a potential ‘gestalt’ communicative device. Boulding suggested that by adopting a systems-based way of thinking, practitioners from economics to physics to sociology, could appreciate the mutual affinity of their subject matter and better contribute to each other’s work. Seven years previous in 1949, a meeting of researchers under the broad rubric of the behavioural sciences convened at the University of Michigan to discuss the possibility of formalizing an empirically testable general theory of social and natural systems (Miller 1955).2
The idea of constructing such a collaborative model had already been mooted by von Bertalanffy in 1937 during a seminar delivered at the University of Chicago. Here, he articulated a view of general systems theory as a corrective to reductionist scientific methodology and mechanist reasoning, which had, in his opinion, remained in place since the industrial revolution (Hammond 2003: 104). Three years after Miller’s initial meetings in 1949, the group, now comprised of representatives from history, anthropology, economics, political science, sociology, psychology, medicine, physiology, and mathematical biology, had managed to sketch a remarkably comprehensive working programme of general systems principles (Miller 1955). Attesting to their organizational stability, if not their programmatic agreement, Kenneth Bailey delivered his presidential address to the 48th International Society for the Systems Sciences Annual Conference in 2005, remarking on the continuing remit of Miller’s precursor group to ‘…encourage the development of theoretical systems which are applicable to more than one of the traditional departments of knowledge’ (Bailey 2005: 365).
Although Ludwig von Bertalanffy’s General Systems Theory is typically identified as the first comprehensive work in the field (Castellani and Hafferty 2009), it is James Grier Miller’s Living Systems Theory that has served as a point of departure for much subsequent debate. Von Bertalanffy’s earlier work in biology is generally viewed as an ‘organicist’ attempt to move beyond the ‘mechanicist-vitalist’ debates of the early twentieth century. Mechanicism , with its origins in Newtonian mechanics and the Cartesian separation of mind and matter, assumed that the organisation of matter was inherent, implying the best way to understand complex organic structure was from the molecular (Hammond 2003: 36). Early organicists such as von Bertalanffy challenged this conceptual approach through Darwin’s suggestion that open systems tended toward higher-order complexities. In short, organicism mandated that an organism (including a community or society) was an emergent entity, irreducible to its constituent parts, and as such, must be studied as a system (Castellani and Hafferty 2009: 114).
There are two general ways in which classical theorists spoke of systems: closed systems, and open systems. The distinction is important in appreciating how early social and natural scientists compartmentalised their subject matter. Closed systems are those which do not display ‘import of energies in any of its forms such as information, heat, physical materials, etc.’, whereas open systems ‘exchange materials, energies, or information with their environments’ (Hall and Fagen 1956 cited in Bailey 1984: 8). von Bertalanffy’s key contribution was thus to advance further the concept of an open system as applied to the study of biological life, by suggesting that such systems be conceptualised as existing in a state of perpetual flow, continuously exchanging energy and components with their environments (Hammond 2003: 116).
According to this definition, human societies belong to the category of open systems, as they require constant exchange (appropriation of material and energy and expulsion of waste by-products), to successfully reproduce. Open systems differ from closed in that additional inputs are admitted as required, unlike closed systems where the constituents of the system are fixed (Skyttner 2005: 53). This difference is essential to bear in mind for sociologists, as several issues with functionalist’s use of concepts intended for closed systems will later be encountered. Conceptualizing organic life as a hierarchy of open-systemic organisation was therefore essential for early theorists to overcome the restrictions of thermodynamic law to closed systems; the ‘steady state’ of the open system was thus maintained by a combination of processes at various levels of organisation, which functioned to regulate matter-energy exchange.
Appreciating the dynamic nature of a system is one matter, describing and categorising its structure is another. Description of system structure is an important step in any macro or meso-analysis, and despite the lack of appeal of structural models in modern social science, sociologists routinely invoke notions of structure when identifying social properties such as institutions, measuring inequalities, or deploying concepts such as markets. Each implies a relative concreteness and degree of historical stability to the phenomenon in question. In terms of system structure, it is Miller’s Living systems theory which stands as exemplary of systems theorist’s attempts to establish an exhaustive conceptual framework with which ...