Jason Oakes reviews Hunter Heyck, Age of System: Understanding the Development of Modern Social Sciences (Johns Hopkins University Press, 2015)
Jason Oakes is a postdoctoral scholar in the philosophy department at UC Davis, where he works on the history of growth models in 20th century biology and the human sciences. We asked Jason to review Hunter Heyck’s new book on the social sciences in the mid-twentieth century. Here’s what he had to say.
You can reach Jason at oakes(at)ucdavis(dot)edu
There exist tree or root structures in rhizomes; conversely, a tree branch or root division may begin to burgeon into a rhizome.
-Giles Deleuze and Felix Guattari, A Thousand Plateaus (1981)
The notion of “system” is a commonplace background concept. Many research specialties have a systems name for one of their problems areas, and it lives in popular and public cultures as well. There is the financial system, the ecosystem, the solar system, and the social system, sometimes simply called “the system,” standing in for social arrangements that feel stultifying yet seem outside of immediate personal control. The idea’s continued presence in the early 21st century makes it difficult to remember that not too long ago it was even more pervasive. Hunter Heyck’s Age of System: Understanding the Development of the Modern Social Sciences, re-acquaints us with the span of time from the interwar period to the Ford Administration when “system” was not just a piece of familiar intellectual furniture, but one of the central organizing concepts in economics, anthropology, sociology, and political science.
According to historian of economics E. Roy Weintraub, the use of the term “social system” depended on following economics in treating social processes as decomposable into hierarchically related sub-modules. Philip Mirowski has charted the path of the notion into economics from mathematical physics in his More Heat Than Light (1989). One of the subjects of my own research, Lawrence J. Henderson, was a popularizer of systems-talk in the interwar social sciences at Harvard University. Henderson picked it up fplrom his idiosyncratic reading of Vilfredo Pareto’s Traité de sociologie générale (Pareto, known waggishy as “the Karl Marx of the bourgeoisie,” had replaced Leon Walras, developer of general equilibrium theory, at the University of Lausanne). Henderson in turn pushed the notion of society as a quasi-physiological system to his allies in the so-called Pareto Circle at Harvard, and through his role as patron of the Harvard Society of Fellows. Barbara Heyl and Joel Isaac have interpreted these events rather differently, but they agree on the basics: system was a hot ticket at Harvard, both in the college and in the business school, and this would have ramifications for the development of the systems approach in the American science after World War II.
Heyck names this package of practices, models, and standards of evaluation “high modern social science and high modern style”, defined as “… a new perspective on science and nature, one that conceived of all things in terms of organization, structure, function, and process.” (1). Now, this is very broad definition, and one that could, with a little cajoling, be made to serve for many disparate schools and movements in the natural as well as the human sciences. But this is not the whole of Heyck’s definition. He expands: high modernist social science is distinguished by its practitioners
…framing all subjects of study as complex, hierarchic systems defined more by their structures than by their components. The goal of science, in this view, was to construct formal models of system behavior, and its chief method was to develop models that could enable one complex system, such as a digital computer, to simulate the behavior of another, such as the human mind (1).
This is much more specific, but it brings up difficulties. If the aim of high modernist social science is to model phenomena of interest as a system, and then use another system (such as a formal structure, an algorithm, or computer program) to simulate it, then the high modernists were confronted with two thorny methodological problems. First, what is the relationship between the phenomenon of interest and the model which purports to represent it? Second, how good a job does the simulation of that system do in simulating it? The question of realism and efficacy in modeling eventually became one of the shoals that the high modernist social sciences ran aground on, but this is getting a little ahead of ourselves.
In Age of System, Heyck has given science studies and the history of science a useful and needed book. It will be an important resource for those of us who study the history of the human sciences, and for anyone whose work is touched by the rise and fall of the “New Deal/Cold War consensus”. Heyck begins by describing how high modernism in the social sciences grew from its origin in the 1920s to its dominant position in the decades after the Second World War. He shows how two distinct regimes of patronage supported the growth of the high modernist worldview and the bureaucratic assumptions about the world that went with them. In Chapter Six, Heyck appraises high modernist social science in the light of some of the recent literature in the philosophy of modeling. In the book’s conclusion, “History and Legacy, the Tree and the Web,” he outlines the decline of the high modernist approach and its succession by another package (“late modernist”) which kept some of the models and methods of systems (such as game theory), but downplayed others (like input-output models).
System, and the high modernist approach, was not limited to the social sciences. Indeed, it was a pervasive theoretical perspective across the institutional board in the post-WWII era. Out of the successes of operations research and the solution of coordination problems in industrial production, the ‘one big optimized system’ way of doing things found adherents and converts in economics, war fighting, and strategy (see William Thomas’s recent book Rational Action). Sharon Ghamari-Tabrizi’s The Worlds of Herman Kahn (2005), and Jennifer Light’s From Warfare to Welfare (2003) give valuable context for the successes and failures of high modernist research, planning, and policy. Nancy Slack’s biography of G. Evelyn Hutchinson, G. Evelyn Hutchinson and the Invention of Modern Ecology (2011) shows how ecology took up the challenge of systems. Hutchinson and his fellow ecologists began to conceive of ecological processes (or if you prefer, the ecosystem) as a nested hierarchical structure that regulated itself with feedback loops that could be modeled with formal logic, mathematical expressions, and computer programs. Seen from this angle it can appear as though the whole complex world of nature and society was being encompassed in a highly rationalized worldview focussed on optimization, control, and simulation through machines. Audra Wolfe’s Competing with the Soviets: Technology and the State in the Cold War (2012) provides much-needed context on how the high modernist package fit into federal funding priorities for Big Science during the Cold War. Seen from this angle, it can appear as though the whole complex world of nature and society was being encompassed in a highly rationalized worldview focused on optimization, control, and simulation through machines.
As a somewhat satirical look back on the later stages of the organizational revolution from the perspective of 1977, John Kenneth Galbraith’s mordant series The Age of Uncertainty, produced by the British Broadcasting Corporation, is a guilty pleasure. Episode 9, “The Big Corporation” 14:00 – 17:00.
Some historians of science take another perspective here. While acknowledging systems’ tendency towards rationalization and decomposition, they saw another side to the systems sciences. On this view systems were not high modernist, but rather an alternative, holistic approach to dealing with complex social, biological, and physical phenomena. The promise of system was to reassemble a coherent picture of complex phenomena that mere analysis left in pieces. Deborah Hammond’s The Science of Synthesis (2003), and Andrew Pickering’s The Cybernetic Brain (2010) fall into this category. We can argue over whether cybernetics and systems science were symptoms of a rationalization process, or an attempt to break out of the same, but there were a lot of unlikely allies doing business under the name of “system” who agreed to not pick fights with each other at conferences. Some actors seemed to have a foot in both camps, practicing rationalization while they preached holism, as it were. Stafford Beer’s managerial mysticism is one example of this, as is the ambiguous figure of Buckminster Fuller. Arguably so was embryologist Conrad Waddington, who mixed Alfred Whitehead’s process philosophy with operations research and information theory. No less an enemy of planning than Friedrich Hayek was also an enthusiast for cybernetic metaphors of self-organization.
Heyck’s insights into the link between organizational change and social scientific research rightly guide him to look at patronage. He identifies two patronage “regimes”, arrayed around different constellations of supporting and enabling institutions. The first regime, centered on philanthropic foundations such as Carnegie, Ford, and Rockefeller, the Social Science Research Council, and military research agencies like the Office of Naval Research, the Air Force’s RAND corporation, and the Army’s Operations Research Office. This first regime dominated funding and support for the human sciences from 1945 through 1960, but was progressively displaced by a second regime in the 1960s. The second regime, which Heyck identifies as a response in the US national research and education agenda to the Soviet launch of Sputnik, had as its core the National Science Foundation and the National Institutes of Health.
Heyck’s method for measuring the growth of high modernism in the social sciences was to survey “flagship” journals in the relevant disciplines: anthropology, economics, sociology, and political science between 1925 and 1975. Using keyword searches and charting word frequencies, he shows how systems talk (such as “structure,” “system,” “function,” and “model”), present in around 7% of the articles of these journals in 1930, showed up in 60% by 1970. Subsequently the frequency of these use terms dropped off as high modernism lost ground to its critics and competitors. Heyck assigned papers to categories (“Core”, “Outer Core”, “Affiliates”, “Margins”) based on the number of keywords they matched. From my understanding of his methods, Heyck used the bibliometric groupings to confirm his argument about two systems of patronage, rather than to initiate it, but it is good to see the kind of robustness that comes from getting the same result with multiple methods.
Heyck’s methods (which he includes in an appendix) reinforced in my own mind the challenges of doing the history of the human sciences in the latter half of the 20th century. There is just so much paper that it seems nearly impossible to do anything above the level of case study without bibliometrics. Here the developments in metadata, network analysis software, and the cooperation of the big academic publication database owners like Web Of Science are valuable. So too is the services of a research assistant (for this project that person was Sylwester Ratowt, whose work Heyck acknowledges.) I am interested to see what kinds of scholarship our specialty will produce with these mixed methods going forward as more historians receive training in what is now being called digital humanities.
Trends and cycles are inevitable in intellectual movements, as in any social movements. After documenting how high modernism occupied the commanding heights of the social sciences, Heyck shows it was set up for a fall. Criticisms mounted in many specialties, from economics where the systems approach came under fire from neo-classicists demanding rigorous microfoundations, to rational actor approaches in political science and sociobiology. The New Left, feminism, and environmentalist social movements questioned the presumed rationalization and soullessness underlying systems. Heyck characterizes the critics and displacers of the high modernists as “late modernists,” and frames the decline of systems in the social science as part of the story laid out in Daniel Rogers’ Age of Fracture (2011). Things fell apart, the center could not hold, and many of the dominant social arrangements in the United States were greatly transformed.
Additionally, we have to return to the problem of the relationship between the model, the simulation of the model, and the problem area under study by the modeler. I don’t mean to get into any deep metaphysical issues about realism in the sense of representational fidelity to the relevant features of a phenomenon. Rather, I simply mean realism in the sense of whether the high modernist models of society, minds, and behavior performed as well as their practitioners needed them to in order to carry on their work. The systems critic Ida Russkoff Hoos’s acerbic essays from the 1960s to the ’80s capture the gist of the complaint: even according to its own criteria of efficiency and control over complexity the systems approach came up short in lines of work like waste management, education, health care, and information systems. Its successors built up many alternative ways of doing things, including an increased reliance on markets for resource allocation, networked organizational structures facilitated by advances in communication technology, and an increased willingness to embrace messiness and complexity in theorizing and calculation.
One category that I kept waiting to see in Age of System was capitalism, or some cognate term. Heyck does mention marxist geographer David Harvey, neoliberalism, and “the marketization of the everyday” in his discussion of the late modernists. However, I would have expected a much more central place for the role that the debates over planning, markets, rationality, and efficiency played in the funding and promulgation of the high modernist social sciences. Instead, Heyck situates the high modernist worldview within a related constellation of institutional arrangements he calls the Organizational Revolution, in the vein of organizational historians like Louis Galambos and Alfred Chandler. The organizational revolution depended, in turn, on “industrial modernism”(16). But this is not a big problem: whether you call it Keynesian managerial capitalism or “the continuing organizational revolution,” the periodization that obtains is similar. Many of these arrangements changed in the 1970s, and we are living with the results of those changes today.
Crucially, the viability of the systems approach in the social sciences, as well as in policy, depended on the ability of experts to plan the proper allocation of resources, and the capacity for human organisms to be molded by their environment (as Jamie Cohen-Cole argues in The Open Mind: Cold War Politics and the Sciences of Human Nature, 2014). This is true whether the planners were apolitical technocrats, socialists, or anticommunist liberals. It hard to see how the shifts in funding and support that Heyck lays out in detail could not be related to the larger concerns of national development, urban stability, race relations, economic growth, and the revolt of the youth. These were certainly on the minds of some high modernists with divergent political commitments, from Walter Rostow and Talcott Parsons to Noam Chomsky.
As I mentioned above, I wanted a more forthright position about the relationship between regimes of funding in the high modernist social sciences and the broader economic and institutional shifts that coincided with them. Heyck is alive to this question, but he reluctantly concludes that it is difficult to see whether patrons followed new developments in research or if researchers were led by the agendas of their patrons. I endorse very strongly, however, his closing thought: “The Web of Networks is woven among the branches of the Tree of System.” If this conclusion is in any doubt, I recommend reading Paul Baran’s On Distributed Communications: Introduction to Distributed Communications Networks (1964). If you like, read it while you watch Douglas Englebart’s demonstration of computing and communication systems at the 1968 Fall Joint Computer Conference in San Francisco. As you read and watch, think of the networks that grew in the branches of Heyck’s tree of system, and then think about what came next.
A network that grew in the branches of the tree of system: the computer workstation and communication suite as a sociotechnical system for human augmentation.