James March in Irvine: A history of the ahistorical in organisation theory

Introduction

After receiving an award for his contribution to liberal arts education, the historian Niall Ferguson (2016) observed that ‘History at U.S. colleges is suffering decline and fall, and faster than Gibbon’s Roman Empire’ (p. 11). That decline is also evident in organisation studies if we trace its origins back to the late 19th century when the German historical school was highly influential (Jones and Monieson, 1990) or indeed to the writings of Max Weber. And there is little sign that things are changing, notwithstanding appeals for a ‘historic turn’ in organisation studies (Clark and Rowlinson, 2004) or initiatives such as this Special Issue of Management Learning. In this article, I argue that the marginalisation of history in organisation studies is connected with events that happened in the University of California–Irvine some 50 years ago and that recounting these events is important in reflexively constructing what we now understand as knowledge in the field.

I begin the article with a brief discussion about historiography and how the events in Irvine fit within current debates about what constitutes (or might constitute) historical organisational studies. I then narrate and discuss these events, which are centred on the years 1964–1969 when James G. (Jim) March was Dean of the School of Social Sciences in the University of California’s new campus in Irvine.

A historic turn?

History has typically had a marginal place in the field of organisation studies – especially since the mid-20th century – notwithstanding periodic attempts to assert its importance. Most recently, there have been suggestions that organisation studies might experience a ‘historic turn’ or at least that there might be more interaction between history and organisation. Part of this discussion has taken the form of taxonomies that purport to outline different ways in which history and organisational research might become more proximate to one another. The overarching sense of these exercises is to integrate both domains, giving more or less equal status to their quite different traditions. For instance, Maclean et al. (2016) present a taxonomy of four conceptions of history in organisation studies – evaluating, explicating, conceptualising and narrating – and posit a strategy of ‘dual integrity’ where the objective is to create a ‘unified, principled historical organization studies integrating organization theory and historical analysis’ (p. 611). In practice, this means that two acid tests for authenticity must be met: for organisation studies, it is theory development, which they understand as ‘making an explicit contribution to advancing generalizable knowledge within the field’ (p. 615), while for historical research the test is historical veracity. Similarly, Decker (2016) confines her taxonomy to different integrationist positions.

Curiously, this desire to integrate was not the dominant goal in other cases where different paradigms clashed, as, for example, in the ‘paradigm wars’ of the 1980s, where the common argument was that paradigms are incommensurable and hence not amenable to integration. Indeed, the two paradigms of history and organisation theory could be seen to be in conflict, with the rise of one not unrelated to the decline of the other. Thus, having integration as a goal could be seen as an attempt by organisation theory to further marginalise, if not co-opt, history. Another argument is that Maclean et al.’s ‘dual integrity’ requirement places, as they recognise, ‘exceptional conceptual and empirical demands on researchers’ (p. 616). This might partly explain the limited contribution of history to organisation studies up to now, and probably in the future as well. Moreover, if integration is the primary strategy, then other objectives must be downplayed, assumptions made and the complexities of different positions glossed over.

This brings us to the story at hand and how this might fit within any of these taxonomies. If we stick with Maclean et al.’s (2016) frame, it is not a case of using history to test a theory (Evaluating), nor is it using history to generate new theoretical constructs (Conceptualising), nor is it using history to apply and develop theory to reveal the operation of transformative social processes (Explicating). By default, therefore, it seems to fit within their understanding of history as ‘Narrating’ where the value to organisation studies lies in explaining the form and origins of significant contemporary phenomena. For now, let us ignore the fact that this mode cannot pass one of their dual integrity acid tests – the requirement to develop theory – because in this understanding of history ‘theory is largely offstage’ (p. 614). Instead, let us focus on the contemporary phenomenon that the ‘Irvine case’ might speak to, and whether or not this warrants being considered as significant.

The case for the ‘Irvine case’

This article relates one version of the ‘Irvine story’. The story is important to organisation studies for a number of reasons. First, it describes a significant period in the life of the recently deceased James March, who has been one of the most influential contributors to the field. The story focuses on the period from 1964 to 1969 when March was Dean of the School of Social Sciences in the University of California’s new Irvine campus. I set these temporal boundaries partly for pragmatic reasons, but also in explicit recognition of March’s influence locally on the School and more broadly on the field of organisation theory, where his contribution has been immense. In 1958, he and Herbert Simon – who was subsequently awarded the Nobel Prize in Economics – published the classic text Organizations (March and Simon, 1958), which Karl Weick (2017) recently described, in his review of the book 60 years after its publication, as ‘both a giant set of shoulders and an enduring source of collective omniscience’ (p. 9). In many ways, the book aspired to mark out a field of inquiry that subsequently came to be. In 1963, he wrote the equally seminal A Behavioural Theory of the Firm with Richard Cyert, and his subsequent career marked him out as a prolific, innovative and influential social scientist of the highest calibre. He has written countless articles, 21 books, eight poetry collections, and produced two films. His most recent publication was an Organization Science article published in 2015 on, appropriately, ‘intellectual outliers’ (Augier et al., 2015). As of November 2018, his top 5 publications had over 100,000 citations in Google Scholar. He has 17 honorary doctorates, while a Harvard Business Review study identified him as the second most influential management guru after Peter Drucker (Prusak and Davenport, 2003). He is best known for his work on organisational decision-making, though he has published in the fields of politics, economics, psychology, sociology, leadership and organisation studies. In particular, organisational adaption and learning have been major themes in his writing, while his highly cited Exploration and Exploitation in Organizational Learning was foundational to the field of organisational learning. Few would argue with Starbuck’s (2012) claim that ‘he was a father of the field called “organization theory”’ (p. 88). Surprisingly, while there are a few pieces on his contribution to the discipline (Augier, 2010; Starbuck, 2012), the story of his time in Irvine has yet to be told.

What has been written about March tends to focus on his time in the, now legendary, Graduate School of Administration (GSIA) in the Carnegie Institute of Technology in Pittsburg, which was an extraordinary hotbed of ideas and research from when he joined in 1953 to when he left in 1964. The group was led by Herbert Simon, a true polymath who is now recognised as a founding father of many scientific domains including artificial intelligence, information processing, decision-making, complexity theory and computer simulation. As well as Simon and March, the GSIA group included Richard Cyert, Abraham Charnes, William Cooper, Charles Holt, John Muth, Robert Lucas, Oliver Williamson, William Starbuck, Victor Vroom, Edward Prescott, Edwin Mansfield, Franco Modigliani, Richard Nelson and Edward Feigenbaum. Remarkably, six of the GSIA group – which fluctuated in number from 30 in 1955 to about 50 in 1964 – subsequently received Nobel Prizes in Economics (Lucas, Miller, Modigliani, Prescott, Simon and Williamson), 10 were elected to the US Academy of Sciences (including March) and 15 were elected to the American Academy of Arts and Sciences (again including March). GSIA was also the poster child for a new form of Business School, founded on a strong commitment to research, that, in line with influential reports of the time (Gordon and Howell, 1959; Pierson, 1959), emphasised empirical research, deductive reasoning and mathematical modelling (Augier and March, 2011). While the GSIA story has been well documented and retold in at least a dozen book chapters and papers, there is hardly any account of the events in Irvine even though it was an important sequel to GSIA with its own distinctive features.

Irvine was distinctive, but it was also riding intellectual fashions of the time. Figure 1 outlines some of the more important relationships and individuals around which these fashions were framed.

OPEN IN VIEWER

Figure 1.

Significant nodes in the mathematical modelling network.

As early as the 1920s, the social sciences, and economics in particular, were being heavily influenced by developments in mathematics. By the 1930s, the leading lights of the younger generation of economists, like Samuelson and Arrow, were mathematizers, with Samuelson depicting a mathematical representation of economics that privileged analogies with physics, biology and thermodynamic systems in equilibrium or near to equilibrium. These ideas underpinned the formation, in 1932, of the Cowles Commission for Research in Economics, which aimed to strengthen the links between economic theory and mathematics and statistics. Similar thinking was evident in Harvard during the 1930s and early 1940s, with the famous ‘Pareto Circle’ promulgating a mechanical-system model of society centred on the concept of stable equilibrium. The Pareto Circle included the leading sociologists Talcott Parsons and Robert K. Merton, with the latter subsequently collaborating with Paul Lazarsfeld when he moved to Columbia University in 1941. In turn, Lazarsfeld, became another influential 20th century American sociologist and was one of the founders of ‘mathematical sociology’. In 1954, he published a collection of papers entitled Mathematical Thinking in the Social Sciences (Lazarsfeld, 1954) which presented a clear picture of the emerging mathematical modelling paradigm. One of Lazarsfeld’s doctoral students, the engineer James Coleman (1964), published Introduction to Mathematical Sociology which again highlighted the commonality between mathematics and social theory: ‘The very character of much mathematics, as logically consistent systems of elements and transformations, parallels closely our general conception of what a “theory” consists of’ (p. 8).

Coleman’s book was part of the Prentice Hall Series on Mathematical Analysis of Social Behaviour, which he and March edited. Another important book in that series was Harrison White’s (1963) An Anatomy of Kinship; Mathematical Models for Structures of Cumulated Roles, which used a branch of abstract algebra, group theory, to analyse kinship structures.

Thus, by the 1950s, there was growing interest, across the social sciences, in the commonalities between biological and cultural evolution, in the concept of general systems, and in the application of sophisticated mathematics to understanding such phenomena. For instance, Simon, who was a colleague of Samuelson’s in Chicago, drew on Nicholas Rashevsky’s work in mathematical biology to present a mathematical analysis of social phenomena. In 1952, he reworked Homans’ model of group behaviour into a set of differential equations, from which various deductions could be made. Even though Homans’ original model was non-mathematical, it was still well suited to ‘mathematization’ not least because Homans was a central figure in the Pareto Circle.

The Second World War also highlighted the vital role of university research, leading to what March referred to as, ‘post-World War 2 enthusiasms in social and behavioural science’. Those enthusiasms, ‘relative to other times [both earlier and later], were strongly interdisciplinary, were strongly quantitative, strongly “scientific”’ [JM]. In that sense, the Irvine story merely ‘reflected . . . the dominant beliefs of a dominant group of social behavioural scientists at the time’ [JM]. As Figure 1 outlines, this dominant group was relatively compact, with significant nodes in the University of Chicago, Harvard (initially in the Pareto Circle and later in Harvard’s Department of Social Relations), Columbia, Stanford, and GSIA.

Methodology

The research is based on extensive interviews and email correspondence I had with 11 faculty and two students who were involved in the ‘Irvine experiment’ during or soon after March’s time as Dean. ¹ I used archival records and snowballing to identify potential interviewees. Some faculty were deceased, while others were uncontactable or did not reply to my invitation, while I was unable to contact any non-faculty staff. Even though I focused on faculty, I also interviewed two students and had email correspondence with several others. In total, I conducted 15 hours of interviews, mostly in 2004 and 2005, and these were subsequently taped and transcribed. In addition, some of my interviewees sent me their own recollections by email. The narrative and analysis also used interviews conducted by the UCI historian, Sam McCulloch, with William Schonfeld (1990), Ivan Hinderaker (1974), Fred Tonge (1974), Clark Kerr (1968), Arnie Binder (1989), Kim Romney (1989) Julian Feldman (1974), Daniel Aldrich (1974, 1989) and Anne Lage’s interview with Jack Peltason (1998). As well as these interviews – which are published on UCI’s Oral History website http://ucispace.lib.uci.edu/handle/10575/5251 – I also read McCulloch’s unpublished 1973 interview with James March, which is only available in the archives. The research also drew on the descriptions and analysis of Johan Olsen (1970, 1979 [1976]), who was a visiting scholar in Irvine at the time, and an unpublished paper by Jean Lave (2009) as well as March’s own writings. In addition, the research drew on correspondence with other individuals, and a detailed analysis of university catalogues, memos, reports, minutes, and so on, in the UCI archives. I collated and coded the interviews and archival material using Scriviner and I then organised the whole dataset thematically, based around emerging themes, chronologically and, finally, around individual life histories.

Story-telling warrants some reflection on issues of representation, which are routinely present in historical studies, especially in historical sociology. The story told here is just one interpretation and other stories about Irvine can surely be told. In particular, I only interviewed those who remained in and were successful in academia and I interviewed no non-faculty staff member, very few students and nobody who was seriously disaffected by their time in Irvine. While I was not seeking the ‘truth’ – when I said this to March he quickly responded, ‘If not the truth, which falsity do you want?’ – veracity was still important, and so I checked the story’s factual accuracy by reverting to my interviewees with my version of the narrative, making edits as appropriate. However, the range, depth and number of interviews, supplemented by the detailed archival records, meant that my interviewees rarely disputed my depiction of events.

Irvine’s School of Social Science: strategy

March’s commitment to a quantitative approach to social science was clearly articulated before he arrived in Irvine and right through his time there, and it is plain, from a letter he wrote to Hinderaker in November 1963, that he wanted to continue the GSIA ethos in Irvine. In particular, his vision was that the Division should become a ‘leader in the application of modern techniques for empirical investigation and theory building’, which meant that the ‘social sciences should be heavily laced with mathematics, statistics and computer methodology’.

And this is very much what he sought to do: for Mike Cole, ‘the big idea was that it was going to be a quantitative School of Social Sciences’, while for Jean Lave, ‘the mathematical social sciences group was certainly, a central identified, centre identity of the school, no doubt about it’. According to her, a fundamental belief of the group was that social science ‘should have a mathematical basis; so that mathematics was the universal language and that means history is irrelevant’.

Sam McCulloch (1996), who was appointed Dean of Humanities just before March and who subsequently wrote a history of the university, wrote that March had defined his social science programme as

the systematic observation, interpretation, and quantitative analysis of human behaviour. He planned that it would involve students in an ultra-mathematical, ultra-statistical, ultra-model-building type of curriculum in which the required course work would include two years of university level mathematics. This heavy maths orientation made his program unique and took advantage of the availability of high-speed computers. (p. 29)

March continued with this strategy throughout his time as Dean. In November 1968, just months before he left Irvine, he wrote a memo to Aldrich emphasised that there would be an increasing concern, across the social sciences, with quantification, formal theorising, and the application of more sophisticated mathematical techniques.

Hiring

March’s first appointment, in 1964, was Julian Feldman (34), who had been one of his graduate students in GSIA. Feldman had graduated from GSIA in 1959 and then joined the business school faculty in Berkeley. He was an expert on the computer simulation of human thought and decision-making, and, with Edward Feigenbaum, had just published Computers and Thought, the first collection of articles about artificial intelligence. Feldman believed that GSIA achievements could be replicated in Irvine, which attracted him to the venture, though he also thought it a bit strange:

One of the things which I thought was rather strange was that, well, people thought yes, you plunk a place down in the middle of nowhere and you divorce yourself from history. Nobody could divorce himself from his own history, was one thing. [JF, 1974]

Feldman and March took responsibility for the initial round of appointments. The first group of students were due to arrive in the Fall of 1965, so March and Feldman hired about 15 faculty during that year, including Acting Assistant Professors Kathleen Archibald, John Boyd, Barbara Foley, Joseph Hart, Alan Miller, Deane Neubauer, Karl Radov and John Weicher; Assistant Professors Duran Bell, Inga Bell, Isabel Birnbaum, Myron Braunstein, Gordon Fielding, Sheen Kassouf and Howard Rosenthal; and Associate Professors Lewis (Creel) Froman and Martin Shapiro. Most of these were in their early 20s and still had not completed their dissertations, though several were older, such as Kassouf and Shapiro. A total of 287 students registered with the School of Social Sciences in the Fall of 1965, including 13 graduate students. In 1966, a further 375 students registered, of which 20 were graduate students. More new staff were appointed including Arnie Binder, Charles Cnudde, Lyman Drake, Alan Gross, Mary Key, Jerry Kirk, Charlie Lave, Jean Lave, Duane Metzger, Volney Steffler and John Wallace.

Not all of the appointments were committed to mathematical modelling but many were. Myron Braunstein, Arnie Binder, Mike Cole and Albert Ahumada were quantitative psychologists, while John Boyd, Michael Burton and Kim Romney were mathematical anthropologists interested in developing mathematical models of kinship and social behaviour. Volney Steffler was a junior Harvard Fellow, having studied anthropology and psychology, and was also interested in mathematical and computational applications in anthropology and psychology. Kathleen Archibald and Barbara Foley were both ‘quantitative’ sociologists and were rare female faculty appointments at that time. Jerry Kirk was another mathematical sociologist, while Duran Bell, who had just received his PhD from Berkeley, was hired in 1965 and was one of the first Black faculty members at UCI. Charlie Lave joined in 1966 and his research interest was in quantitative economic history and transportation economics: ‘And the values that I particularly valued and many of the young people with whom I got associated, was the mathematical modelling, and that was one of the fairly strong centres of activity’ [Charlie Lave]. A couple of years later, Bill Sharpe joined – he was subsequently awarded a Nobel prize in Economics – but only stayed for 2 years.

Kim Romney was one of the few senior hires (he was a full Professor in Harvard having previously been a full professor in Stanford) but the other hires were almost all very young. According to Charlie Lave, many of these young faculty came from Stanford which had been a centre for mathematical modelling – ‘there was a mathematical behavioural science programme there in the ‘50’s, into the 60’s; Kenneth Arrow was involved, Patrick Suppes, Richard Atkinson’ – and the first Stanford Symposium on Mathematical Methods in the Social Sciences had been held in 1959. The average age of the faculty in 1966 was just 26, with five women in the faculty of 30, which was unheard of at that time:

I think that he [March] also believed that you should pick young people with a lot of promise . . . many of the people were very young. [Kim Romney]

. . . we had March as the Dean and a couple of senior people in Political Science, a couple of senior persons in Psychology and then mathematical types. Everybody else were very junior people. [Mike Cole]

There were a whole bunch of people who ended up here who were very young, they were just getting their PhD’s [Charlie Lave]

. . . most of these young people came in with no experience other than as graduate students. They really didn’t exactly know how Universities were run. [Duncan Luce]

Looking back, March believed that

the strategy of getting young people before they were identified by the market is a very good strategy, provided you have good judgement and provided you are prepared for the obvious fact that under the circumstances that most of your young people are likely to leave.

Curriculum

The strategy of developing a quantitative approach to the study of social phenomena was carried through in the undergraduate programme of study (March had little interest in developing a coherent graduate programme). This is clear from the first 1965–1966 catalogue which states that the

educational programs in the Division of Social Sciences . . . are built upon the new social science of systematic observation, interpretation, and quantitative analysis of human behavior. The availability of high-speed electronic computers; the development of mathematics oriented toward the problems of social science; and the refinement of techniques for sampling, observing, and modifying human behavior are creating new methods in anthropology, economics, geography, political science, psychology, and sociology.

The catalogue shows the central position given to mathematics in the curriculum (see Table 1).

OPEN IN VIEWER

Table 1.

Curriculum extracted from 1965 to 1966 Catalogue.

	Fall	Winter	Spring
Freshman	Intro to Analysis Maths Humanities English	Intro to Analysis Maths Humanities English	Intro to Analysis Maths Humanities Computer Science
Sophomore	Anthro or Economics Maths Natural Science Elective	Political Science or Psychology Maths Natural Science Elective	Geography or Sociology Maths Natural Science Elective
Junior	Major Field Statistics Outside Field Elective	Major Field Statistics Outside Field Elective	Major Field Statistics Outside Field Elective
Senior	Major Project Fine Arts Outside Field Elective	Major Project	Fine Arts

The mathematics requirement imposed on all social science undergraduates was ‘most unusual’, according to Arnie Binder, while Julian Feldman perceived that the requirements in computer programming, mathematics and statistics were ‘controversial’. Consistent with the broader fashion of the time, maths was perceived to offer a universal language on which inquiry could and indeed had to be based, but it also

acts as a screening device for improving the calibre of students who are attracted to the social sciences. So far, we seem to be very successful at discouraging anti-scientific humanists but only partially successful at attracting potential scientists, who now go into physics, chemistry and math. (Memo from March to Aldrich, 1968)

Another most unusual feature was the notion of just offering one degree, and not offering degrees in subjects such as psychology or economics.

An important, memorable and influential pedagogical innovation was the ‘Introduction to Analysis’ course led by March with contributions by other faculty. This was a mandatory course, taken by all undergraduate and graduate students and Deane Neubauer’s recollection is that most of the faculty ‘had to both teach a module in the course and had to audit it as well’. In essence, it provided a way of thinking about social phenomena that required one to create a ‘model’ of a phenomenon in a form that could, ultimately, be converted into a mathematical and/or computer model (March, 1970: 140). The main models explored were of choice, exchange, adaptation, diffusion and structure. Bill Sharpe describes it thus:

Jim had a required course, which everybody had to take, which was in effect models and social science. I audited that. It was brilliant. He did things like marriage, he had a contagion model; he took not just economic issues, but a series of applications, and in each case a model thereof. There was a lot of angst. He’d get these kids, who took it in their sophomore year; they were do-gooders, the social science majors, they were going to go out and save the world. He made them look at each of these problems with models, some equations, some graphs, maybe a computer programme. I can’t remember. There was a lot of no, no, no; this is not how you do social science. I don’t know if the students liked it.

March and Charlie Lave subsequently produced a text book based on the course (Lave and March, 1975) which, according to Mike Cohen – who took the course as a student,

was also used in high schools. A beautiful course. When I got to Michigan, I built a course around the text book. It was taught up until a few years ago with a reproduction of the book. Certainly the most beautiful course on thinking about social life – it had a huge effect on me – a way of thinking. ²

At its core, the ‘Introduction to Analysis’ course was built on a belief in mathematical modelling – ‘It was more mathematical than I’d have had it’ [Deane Neubauer] – though it was more than that; ‘It managed to convey a creative attitude and aesthetic criterion and a commitment to simultaneously holding multiple perspectives on phenomena . . . that’s great training for social science and life’ [Michael Cohen]. In essence, it provided students with tools of analysis, with March encouraging students to play detective by getting them to formulate models of miniature social problems – see March (1970) for examples – and through doing so to develop their skills of analysis, abstraction, speculation and critique. March then linked these miniature problems to wider social issues which presumably resonated with the students who were surrounded by and participated in intense debate about a range of contemporary social issues – especially around civil rights, feminism, the Vietnam war – which were being hotly debated at the time. Radical (often Marxist) critiques of society and social science were emerging, as were utopian anarchist alternatives to existing social orders. That said, March opined that the faculty was ‘socially and politically relatively conservative’ and ‘was a little more radical in life style than it was in politics’, while a number of interviewees commented that the wider Orange County community was ‘extremely conservative’.

March (1970) saw his pedagogical philosophy as very different from most other undergraduate social science programmes: ‘most first courses in [the social sciences] portray the fields as a large collection of definitions loosely connected by important empirical generalizations’ and emphasise the secondary skills of reading, organising category schemes and labelling phenomena (pp. 60–61). His argument was that ‘the skills required for a modern social or behavioral scientist are the skills of analysis, model-building, hypothesis-forming, data interpreting, and problem-solving . . . skills [that are] highly correlated (or identical to) the skills involved in mathematics and natural sciences’ (p. 59). But the social norm with respect to the social and behavioural sciences is ‘antianalytical’ and these sciences are ‘associated (quite inappropriately) with a rejection of things, quantities, abstractions. They tend (except for economics and political science) to be relatively “feminine”’ (March, 1970). His disregard for history is evident from his assertion that ‘virtually nothing of the behavioural and social sciences is taught in the first twelve years of school. The exceptions are insignificant and unfortunate: geography . . . civics, . . . and modern history comprise the normal fare’ (March, 1970: 58–59). Perhaps it is no surprise, therefore, that March quickly agreed that the Department of History – which, in the original academic plan, was to be located in the School of Social Sciences – should be moved to the School of Humanities (Peltason, 2001: 167).

Prior theory, for March (1970), only inhibits students from thinking analytically about social phenomena, a principle he playfully enunciated as ‘Gresham’s Law of Study: Reading drives out thinking’ (pp. 69–70). Interestingly, a similar sentiment underpins Glaser and Strauss’ (1968) contemporaneous formulation of grounded theory in that it too sidelines prior theory. (This connection between March and Glaser is also shown in Figure 1.) Yet, while these approaches might disparage history and prior theory, they are still centrally concerned with the practice – March would call it artistry – of theorising and the formulation of theoretical models.

Interdisciplinarity

If March was committed to a quantitative approach to social science inquiry, he was equally committed to the idea of interdisciplinarity. Interdisciplinary research was fashionable at the time and indeed Sewell (1989) has described the period from the Second World War to the mid-1960s as the ‘Golden Age’ of interdisciplinary social psychology. Specifically, the Irvine ‘experiment’ followed in the tradition of similar interdisciplinary academic experiments, such as GSIA (1950–1964), Yale’s Institute of Human Relations (1929–1950), Harvard’s Department of Social Relations (1946–1966) and the RAND Corporation. The link with Yale was clear: March had obtained his PhD there as had one of his early hires, Deane Neubauer, while another early hire, Mike Cole, was an Assistant Professor there before moving to Irvine. For Cole, Irvine ‘was a more or less faithful attempt in continuing [the] sort of interdisciplinary social science research’ pioneered in Yale, while interdisciplinarity was Neubauer’s ‘biggest “take away” from the Irvine experience’. Another early recruit, Kim Romney, had trained in Harvard’s Department of Social Relations and saw clear parallels between the three experiments in interdisciplinarity in Harvard, Yale and Irvine. March, in contrast, saw Harvard as a ‘different kind of thing’, probably because it lacked a commitment to mathematical analysis. Jean Lave, who also moved from Harvard to Irvine, felt that by the time she left Harvard it had lost its commitment to interdisciplinarity: ‘They were busy locking the doors between floors so that departments of sociology, and social psychology and anthropology should not touch each other anymore’, which supports Sewell’s assertion that the Golden Age of interdisciplinarity ‘had largely vanished’ (p. 1) by the mid-1960s. If the tide was turning, March’s own commitment to interdisciplinarity is clear from the letter he wrote to Hinderaker before joining Irvine, in which he stated that ‘faculty should [have] substantial disrespect for traditional disciplinary identifications’, that they should avoid ‘mimicry of standard views on what is subsumed under the terms “psychology,” “sociology”’, and that specialisation should be by problem areas rather than by traditional academic disciplines. Mike Cole concurred, ‘the idea of interdisciplinary scientific rigorous research was probably the single biggest value, coupled with these pretty wild social values’. And that value meant that most were hostile to the idea of having a conventional departmental structure. William Schonfeld, who led the departmentalisation of the School in the 1980s, captured the mood when he recalled that ‘the School of Social Sciences was founded not just without departments, but with a virulent belief that these were enemies of intellectual life’ [WS, 1990].

In a strategy report he sent to Aldrich in November 1968, March reiterated his commitment to interdisciplinarity, stating that the School ‘has been organized to facilitate the integration of research and teaching activities around the concept that the most crucial and exciting work in the social sciences is that which is interdisciplinary in nature’. In line with this, ‘Our guiding principle in acquiring future faculty will be to appoint people whose area of specialization are particularly conducive to interdisciplinary effort’. This principle meant that the focus on mathematical modelling was sometimes sacrificed in favour of a more straightforward question: ‘“Was this person interesting?” And I think clearly the dominant story was, “We don’t care whether you have the same kind of ideas as we have, as long as we can find your ideas interesting”’ [JM].

Almost inevitably, this eclecticism created a high level of variety:

as it developed it turned out to be a number of people who I would now describe as social constructivists, a number of people who turned out to be relatively pure mathematical modellers. There were some people who became committed ethnographers, and you are talking about a range – someone like Bill Sharpe at one end was creative and a little bit different, but a financial economist – and then you have people like Duane Metzger and Jean Lave at the other end, and who were fairly creative, constructivists, post-modern anthropologists. [JM]

Kim Romney took the view that March ‘also wanted variety. High variance. You wouldn’t have people coming out of the same mould’. Hence, ‘March hired a lot of people who were not mathematically inclined, so [the maths requirement] was never . . . I mean it was held more as a value than as a practice’ [KR]. Schonfeld again: ‘They did believe a lot in high variants, that’s why you had a number of faculty who were quite gifted and a number of faculty who were not. They were looking for anybody who was different’.

But an eclectic recruiting strategy would not necessarily convert into productive interdisciplinary work nor replicate what happened in GSIA. Irvine’s School of Social Sciences had faculty from many more disciplines than GSIA – which had no anthropologists, linguists, educationalists or sociologists – and many of its faculty had neither a grounding in, nor commitment to, quantitative analysis, which again was different from GSIA. This problematised interdisciplinary communication. To partly address this, all faculty were involved in the Introduction to Analysis course so as

to develop a sense of the overall rational/mathematical frame, but also for us to get to know each other at a substantive level and to break out of the disciplinary straight jacket that most of us had left graduate school with having traditional social science degrees. [DN]

In addition, people like Volney Steffler and Duane Metzger argued ‘that we [the faculty] should have a mathematical basis, so that mathematics was the universal language and that means no history; history is irrelevant’ [JL]. Moreover,

the deal was: if you were talking to somebody in a different discipline, it was not fair to bring in old work in your discipline or refer to it as an excuse for not explaining in one-word terms what it was you were trying to talk about. [JL]

For Jean Lave, this requirement that

we should speak in syllables of one word [meant there was] no appreciation of the notion that social theory makes any difference in the world. It was the most ahistorical, atheoretical attempt at a social science you could possibly imagine!

Lave noted that Inga Bell was the only person to quit at the end of the first year, out of the 15 who were hired in 1965:

She was a Marxist sociologist. And I bought into the kind of way they talked about her: she didn’t fit and didn’t belong there; it was a good thing she left. And it wasn’t until many years later that I thought I understood why she left and why the reaction was that way, but I think she was the only one among the first fifteen who had enough sophistication as a social theorist of whatever stripe – it didn’t matter – to make a critique of what was going on and nobody was prepared to listen to it.

March’s ‘experiment’ was to put about 30 young academics together and then see how they might organise without replicating existing structures, or as Mike Cole recalls it: ‘we created this rule that you cannot create an academic unit which was identifiable with an existing discipline [like sociology, anthropology or economics]’. In practice, this meant that once a year each faculty member had to declare whether he or she would be in a Programme or not: ‘You had to make a positive move to declare your intentions. Programs really did cut across academic disciplines. They absolutely violated standard divisions between disciplines’ [JL]. By 1967, three groupings had emerged, which came to be named as Programme A, Programme B and Programme C. Programme A, sometimes referred to as ‘Formal Models’ or, more officially, the ‘Program of Mathematical and Computer Models in the Behavioral Sciences’, followed through on the GSIA work, and the maths and computer programming requirements placed on the students was in harmony with the philosophy of this group. Programme B, which was sometimes known as ‘Language and Development’, included what would be recognised as anthropologists, sociologists and social psychologists. It sought ‘to provide sufficient understanding of complex cultural phenomena to produce significant cultural change [and to develop an] understanding of individual and small group behavior, as well as national, macro-level behavior’ [March’s report to Aldrich, November 1968]. In practice, this group was divided into two or three sub-groups, though the membership and identity was fluid. It was partly defined by its opposition to Programme A, though Mike Cole stressed its ‘degree of common purpose’ and that it was an ‘intellectual meeting place [for] understanding cultural variations in human thought processes’. Finally, Programme C was ‘a residual category for those faculty members in the Division who are not members of either program but hold appointments in the Division’ (memo from March to Aldrich, June 1967). It was ‘the set of those not belonging to any set’ (Lave, 2009). There was also a further group – the ethnomethodologists – who seemed to be outside this A, B, C structure. Harvey Sacks, who essentially founded the field of conversation analysis, was the leading figure here, but Harold Garfinkel, who was primarily based in nearby UCLA, was hired twice by the School, quitting both times after 6 weeks. Garfinkel, like Jean Lave, had been a student in Harvard’s Department of Social Relations where Talcott Parsons was his supervisor (see Figure 1). His programme of inquiry was distinctly different from Parsons, in that he was most interested in the empirical study of social order being created. And in radical opposition to the mathematical modellers, the ethnomethodologists foregrounded ‘indexicals’ – words like ‘here’, ‘there’, ‘me’, ‘now’, ‘after’ – whose meaning is always contingent on context.

History mattered little to any of the groups. As we have seen, Programme A was avowedly ahistorical, but there was little or no concern for history in any of the other groups either.

Disruption

The ethnomethodologists routinely used ‘breaching experiments’ that deliberately violated commonly accepted social norms as a way of highlighting the taken-for-granted rules that underpin social order. In some ways, Irvine’s School of Social Sciences could be considered a breaching experiment as March, in his initial letter to Hinderaker, stated that the Division (the unit was initially called a Division, before it became a School some years later)

should be conspicuously experimental and innovative [and] there should be major innovations with respect to curricula in the social sciences, instructional methods, academic organization, and staffing policies. The social science division should be viewed as an experimental laboratory rather than as primarily a production facility.

For Mike Cole, it was ‘an experiment . . . March was just letting it happen, to see what would happen’. William Schonfeld, who joined in 1970, concurred, ‘It was as though you were doing a real-life experiment with human beings’, and indeed the Course Catalogue (1969–1970) makes this clear: ‘Undergraduate and graduate education in the School of Social Sciences at UCI involve participation in an experiment. The program, faculty and students differ substantially from conventional counterparts elsewhere’ (p. 115). Deane Neubauer also saw it as an experiment:

Irvine as a pedagogy was an experimental project; there’s no doubt about that, and it was intoxicating . . . We used to call it, some of us, ‘Jim’s experimental universe’. We would tease him sometimes at dinner parties, ‘How did you manage to get the University of California to give you this real-life laboratory?’

Michael Cohen also observed how March was especially interested in

scrambling conventional organizational arrangements, to get people thinking in new directions. There were even some experiments done, where they shuffled the labels on everyone’s mailboxes so that everyone was getting everyone else’s mail. He was very interested in that period in what happens when the conventional patterns of behaviour get disrupted.

These practices could also be understood as local manifestations of broader movements that drew on, for instance, ideas in cybernetics about self-organisation (Wiener, 1961 [1948]), or McCulloch’s (1965) ‘principle of redundancy of potential command’ (which posits that command should pass to the region with the most important information).

Elsewhere, Bill Sharpe characterised it as ‘a really interesting experiment; but unfortunately, all the rats died!’

The strategy gave at least the appearance of chaos. According to William Schonfeld (who only arrived in 1970 so his perception should not be extended back into the 1960s), ‘the only rule was, there are no rules’, though Mike Cole, who arrived in 1967, says something similar: ‘There were no rules, and it was as close to a blank slate in an institution as you’re ever going to find’. In his interview, Arnie Binder emphasised the notion of disorganisation:

‘disorganization’ was a word that was used permanently . . . it meant that we were certainly never going to have departments here, above all; that we’re never going to assign offices according to discipline. So you have to have the psychologists here and anthropologists in the next office, and so forth, and the interactions had to be so that there were no organizations by disciplinary focus above all . . . if they moved in a direction of what some would call ‘responsible organization’, he [March] would oppose it.

Mike Cole recalls that

There was a lot of probably illegal relations between students and teachers, there was a lot of drug use, and a lot of alcohol abuse was going on . . . There were affairs between faculty, between couples on faculty, and between faculty and students. So you had these different things that would make it seem chaotic . . .

. . . part of the disorganization was that it seemed like, you know what you’re against but you don’t know what you’re for. Literally we would go in on Saturday to see what the hell we were going to do on Monday. And we would do that quite regularly.

Of course there were rules, not least the rules of the university regarding tenure, which became very real for some of the staff in 1968–1969. And March himself did not agree with the proposition that the School was rule-less:

Rule-less would not be quite right. We had some fairly strict rules. Obviously we viewed the University of California as having a whole bunch of procedures that got in our way, and we were arrogant enough to think that, by and large, we were being ruled by colleagues who were not of the quality equal to ours.

Author: Was this arrogance justified?

JM: Probably not. The University of California is a wonderful institution; but there is an awful lot of mediocrity in it.

Looking back, he paints it thus:

I think some people would describe it as a benevolent despotcy [sic]! I don’t think that it was, it didn’t have much structure, so in that sense it was democratic, you had to make your case, and anyone could make their case, and little subgroups formed, of people who liked to be with each other.

But the strategy could also foster dislike, disaffection, disharmony and disintegration. The strategy was also confounded by the intellectual and political currents of the late sixties, and, in retrospect, March felt that the experiment ‘would have been better timed a decade or two earlier’. Some of those currents were pushing academics into social activism and demanding that they apply social science knowledge to significant social problems (Crowther-Heyck, 2006; National Science Board, 1969: xi). This wider sentiment was articulated most forcefully in the School by Arnie Binder, a mathematical psychologist who had joined the group in 1966. He and some others felt

that the School of Social Sciences was largely irrelevant in the way that it was pursuing its own intellectual elegance, that there were chips on the table that were enormously important, and that there were a lot more ways to do interdisciplinary research other than through mathematical modeling. [Deane Neubauer]

By 1968, Binder had proposed a ‘Program in Social Analysis’ and, while March initially backed this idea, he then withdrew his support. Binder was clearly disaffected and worked to separate off his own unit from the School.

But the biggest disruption happened in 1969 when March announced that he was moving to Stanford:

. . . the major incident is Jim leaving, and that was significant because the guy who brought us here and was our intellectual leader was all of a sudden saying that he didn’t love us anymore. [Charlie Lave]

His leaving left a big vacuum. [Kim Romney]

One of the things that I was very unhappy about was March leaving when he did. I thought he’d stirred the pot and then just walked away from it. I mean I’m sure Jim has a different story about that. But then I followed. [Mike Cole]

Other key faculty left Irvine around this time, including Martin Shapiro, Bill Sharpe and John McCall, while, in 1970, Arnie Binder set up a new unit called the School of Social Ecology that brought staff and students away from the School of Social Sciences. Curiously, it mirrored the School of Social Science’s focus on the interdisciplinary study of social problems, though, importantly, it dropped the mathematics requirement. By 1973, that requirement had also been diluted in the School of Social Sciences and by the 1980s it had disappeared.

March remained in Stanford for the rest of his career, during which time Stanford became a key pillar in the development of organisation theory (Dobbin and Schoonhoven, 2010).

Discussion

Jean Lave, who joined the School in 1966, is best known for her work on situated learning, especially for her seminal text with Etienne Wenger (Lave and Wenger, 1991) who was a student in Irvine from 1984 to 1990. Lave’s argument is that knowledge is situated, that context is always implicated in theory and theory development, and that ‘there is a historically specific, circumscribed and local character to all thought’ (Lave and Wenger, 1991: 193). The counter-argument is that this invites a deluge of local descriptions and trivia-mongering instead of theory, though a reasonable response is that while not all local descriptions matter, some do. This, I argue, is the case with Irvine’s School of Social Sciences, especially during Jim March’s tenure as Dean. The story is important, not least because the events that happened there contextualise influential theoretical contributions, including ideas such as ‘organized anarchies’, the ‘garbage can model’ (Cohen et al., 1972) and ‘exploration and exploitation in organizational learning’ (March, 1991). The events are also an important part of the historical context of Lave’s own influential work on learning (which emphasises historical context). More broadly, if we accept Lave’s position, then the notion of ‘organizational theory’ is itself historically specific, circumscribed and highly contextual, which means that the events described in this article have a place in the discussion about the relationship between ‘organizational theory’ and ‘history’, which is this Special Issue’s particular concern.

First, let us briefly consider Cohen, March and Olsen’s paper on the ‘garbage can model of organizational choice’ which has been highly influential in decision theory and organisation studies. The article introduced the idea of an ‘organized anarchy’ which was a radically different way of thinking about organisations. From this point of view, and contrary to conventional understandings, an organisation is ‘a collection of choices looking for problems, issues and feelings looking for decision situations in which they might be aired, solutions looking for issues to which they might be the answer, and decision makers looking for work’ (Cohen et al., 1972: 2). Cohen, March and Olsen then develop a model of an ‘organized anarchy’ based on prior studies of universities, though the School of Social Sciences in Irvine is omitted, which is surprising as Cohen was a doctoral student there from 1966 to 1972, while Olsen was a research fellow there in 1968 and 1969. The notion of organised anarchy was further elaborated in Cohen and March’s (1974) study of American college and university presidents which also makes no reference to Irvine. The garbage can paper does reference Olsen’s (1970) thinly disguised report – which actually introduced the concept of the ‘garbage can’ – on the tortured and capricious process that the School went through in finding a replacement for March, though that process is not really reflected in the model they developed. Following in the tradition of GSIA and Irvine’s Programme A, their model takes the form of a computer program that sought to simulate the interaction of four streams (choices, problems, solutions and participant energy) plus a set of processing assumptions. What is interesting here is how the authors privilege a contrived computer simulation while ignoring what was surely an exemplary case of ‘organized anarchy’ that they all knew well. This likely reflects the position that March expressed when I asked him what he had learned from his time in Irvine:

I spent a lot of time writing about how one cannot learn from history and how people learn much more than is there. It’s possible to interpret history in any way you like so we tend to interpret it in some way that confirms our prior beliefs. There is lots of evidence for that and I don’t think I am immune to that, so in some sense it never seems to be particularly useful to learn anything from it, since I was not likely to be in that situation again or no one else is likely to be in that situation again.

An example of this sentiment can be found in his paper, Learning from Samples of One or Fewer, where he asserts that since ‘history offers only meagre samples of experience . . . the paucity of historical events conspires against effective learning’ (March et al., 1991: 1).

One could use this story about Irvine to test and refine the garbage can theory, which would be what Maclean et al. (2016) refer to as ‘evaluating’ in their classification of conceptions of history in organisation studies. Elsewhere I have done such an exercise (Kavanagh, 2013) but in relation to March’s (1991) famous ‘exploration and exploitation’ paper, which is also based on a computer simulation model. March’s paper now has over 20,000 citations in Google Scholar and indeed his model of organisational learning has been an important part of the process through which the phenomenon of organisational learning is performatively brought into existence. What the Irvine story highlights is the fact that while the computer model’s ‘initial conditions’ was remarkably similar to the Irvine set-up – for example, both had ‘high variability’ – the computer model described an evolution or form of organisational learning that was radically different from what happened in Irvine. In the model, the diverse group eventually acquires a shared, homogeneous and stable understanding of reality, while, in contrast, the School of Social Sciences (which, like March’s model, consisted of a group of around 50) fractured into an unstable, heterogeneous mix of individuals and factions, as represented by Programmes A, B and C, the ethnomethodologists and Arnie Binder’s group that eventually splintered off into Social Ecology. Moreover, the Irvine story runs a coach-and-four through the model’s basic assumption that a stable reality, exogenous to the group, exists. For instance, the historical story shows that ethnomethodology cannot usefully be understood as part of an exogenous, stable reality; rather, that reality was constructed, and people in, and associated with, Irvine were significant contributors to that process.

The Irvine experiment was founded on a belief in using quantitative approaches to understand social phenomena, which found expression in an ahistorical, atheoretical ideology that was widely shared among the group. For instance, Duane Metzger and Volney Steffler, both affiliated with Programme B, championed the

principles that all talk across disciplines must be in words of one syllable, or at least directly intelligible. No fair retreating into citations of scholarship in your own discipline. No appeals to historical contexts of contemporary work – that would be evasion. (Lave, 2009: 6)

Ironically, while the School was committed to mathematics as a universal language – with its ahistorical, abstract and decontextualised epistemology – some of Lave’s most important subsequent work highlighted the situated, context-embedded nature of mathematical practice (Lave, 1986, 1988; Lave et al., 1984). She now notes how she was ‘a delighted, enthusiastic participant in its collective search for interdisciplinary unity via a mathematical language, empirical modelling and anti-historical, anti-social-theoretical stance, and [yet] end up today working within a historical, materialist theoretical problematic’ (Lave, 2009: 2). Mathematical models do not do irony.

Yet, Lave still appreciated the paradigm of model building as articulated in the book that her husband Charlie wrote with Jim March (Lave and March, 1975), which she saw as a ‘way of thinking my way into problems’. She was also quite willing to use mathematical techniques when appropriate – ‘I did enormous amounts of regression analysis on the [Liberian] tailor’s data’. In contrast, she felt that

the Programme A guys, the maths guys . . . [were not using] . . . Jim and Charlie’s book [but] were trying to say, ‘Can we take some maths and make it look like some social phenomenon?’, in which case we have a mathematical model of something social. To do that they had to reduce, reduce to get to their math models. I think we [Programme B] were doing the opposite. To me that was a huge difference. [Jean Lave]

But, as Lave (2009) admitted, Programme B was almost as ahistorical as Programme A – ‘most of us absorbed this empirical ahistorical mode of inquiry’ (p. 7) – though later in her career she embraced a historical materialist theoretical stance.

History’s exclusion from the School fits with Turner’s (1921) ‘frontier thesis’, which posits that encounters with the frontier gave Americans a forward-looking focus, releasing them from European historical traditions and customs, and in so doing shaped a distinctively American set of values and form of democracy. Southern California was such a frontier in the early 1960s, as I was reminded by a photograph I came across in the UCI archives. Taken in 1963, the photograph is of three descendants of the Gabrielino tribe, in Native American clothes, hiking past a marker at the hub of the original 1000 acres donated to the U.C. Regents by the Irvine company. Having occupied Southern California for at least 2000 years, the tribe was effectively terminated in the 1950s under the Eisenhower policy of ‘assimilating’ Native American Tribes. This less than edifying part of American history might also go some way to explain the ahistorical nature of the intellectual enthusiasms of the time.

Conclusion

There is much to be learned through looking back at what March and his colleagues sought to achieve at Irvine. The tale could be interpreted as a case study of quite different phenomena – it was an organised anarchy, a breaching experiment, an attempt at resisting institutional authority, a learning organisation, a site of particular power relations, a case of exploration and exploitation, a story about gender, a case study of creativity and play, an experiment in interdisciplinarity, a study in group dynamics, a case of leadership and followership – while any number of theoretical frames might be employed to make sense of what happened there. However, for our purposes, it is perhaps best understood as a case that sheds light on the curious and evolving relationship between history and organisation theory. In particular, it provides a historical grounding for understanding how history and theory have been differentiated from one another, which then frames our understanding of the various attempts to integrate history and theory, or at least to have more interaction between the two domains. For instance, the fissure between Programme A and Programme B in the mid-1960s pre-figured organisation theory’s paradigm wars of the 1980s, while the ahistorical tenor of the different factions in Irvine helps explain why Burrell and Morgan’s (1979) highly influential taxonomy of four different paradigms of organisational analysis has routinely collapsed into two camps, positivism and interpretivism, neither of which gives a central place to history. It also foregrounds the shared ahistorical roots of what are routinely understood as opposing research traditions, while it also describes how the advocates of different epistemological positions confronted (and avoided) one another. Finally, it enriches our understanding of the various relational activities that have produced organisation theory’s particular forms of positivism and interpretivism, as well as contemporary understandings of what constitutes theory and history, and how these relate to one another.

In large part, the story is about the larger than life, yet humble, Jim March, and the influence he had on a whole field of study. Everyone I interviewed held him in the highest regard and always spoke of him with affection and respect, as the following quotations illustrate:

Jim was really charismatic . . .. He’s a remarkable guy. [Bill Sharpe]

If you met Jim March, you’d know that he can be an exciting guy, and he communicated that to the students. [John Payne]

He is one of the nicest, most honorable men. [Arnie Binder]

If you had to invent a mentor at that time in my life, he would have looked a lot like Jim . . . an exceptional man. [Dean Neubauer]

He was very greatly respected and almost revered by most of the faculty. [Kim Romney]

Volney [Stefflre] and Jim knew more about everything than any of us did. [Charles Lave]

I still consider him one of the most creative brilliant academics I have ever known. [John Payne]

In June 2018, I sent Jim an early draft of this article (which did not include these quotations) for comment. He responded saying that he ‘did not find any egregious errors and a lot of good data and analysis in the Irvine Story’ although in a subsequent email he opined,

As I thought some more about your excellent Irvine piece, it occurred to me that perhaps you have underestimated the counter-culture role. It divided both the community and the intellectual discourse, though the community was more tolerant than the discourse.

We corresponded a bit more on the topic until he took ill in July. He was inquisitive and generous to the end.