No CrossRef data available.
Article contents
Donald T. Campbell on the institutions of scientific knowledge and the limits to interdisciplinarity
Published online by Cambridge University Press: 12 July 2022
Abstract
Extracts from an important article by the American psychologist, philosopher and social scientist Donald T. Campbell are reproduced here, with an introduction underlining the importance of his argument for today. Campbell identified disciplinary boundaries as enablers of specialization but often barriers to scientific innovation and shared knowledge. But instead of unbounded interdisciplinarity, Campbell argued for focused specialisms that cross disciplinary boundaries. This argument is particularly relevant for the development of institutional research in the future.
Information
- Type
- Fragment
- Information
- Copyright
- Copyright © The Author(s), 2022. Published by Cambridge University Press on behalf of Millennium Economics Ltd.
Footnotes
1 Editorial insertions are in square brackets. These include numbers indicating the start of a new page in the original.
2 The preparation of this paper has been facilitated by the Council for Intersocietal Studies, which operates under a grant from the Ford Foundation to Northwestern University. [Excerpts here are extracted from Campbell (Reference Campbell, Muzafer and Carolyn1969).]
References
Abbott, A. (2001), Chaos of Disciplines, Chicago and London: University of Chicago Press.Google Scholar
Bentkowska, K. (2021), ‘Response to Governmental COVID-19 Restrictions: The Role of Informal Institutions’, Journal of Institutional Economics, 17(5): 729–745.CrossRefGoogle Scholar
Camic, C and Hodgson, G. M. (eds.) (2011) Essential Writings of Thorstein Veblen. London and New York: Routledge.Google Scholar
Campbell, D. T. (1965), ‘Variation, Selection and Retention in Sociocultural Evolution’, in Barringer, H. R., Blanksten, G. I. and Mack, R. W. (eds.), Social Change in Developing Areas: A Reinterpretation of Evolutionary Theory, Cambridge, MA: Schenkman, pp. 19–49. Reprinted in General Systems, 14, 1969, pp. 69–85.Google Scholar
Campbell, D. T. (1969), ‘Ethnocentrism of Disciplines and the Fish-Scale Model of Omniscience’, in Muzafer, S. and Carolyn, W. S. (eds.), Interdisciplinary Relationships in the Social Sciences, Chicago: Aldine, pp. 328–348.Google Scholar
Campbell, D. T. (1974a), ‘Evolutionary Epistemology’, in Schilpp, P. A. (ed.), The Philosophy of Karl Popper (Vol. 14, I & II). The Library of Living Philosophers, La Salle, IL: Open Court, pp. 413–463.Google Scholar
Campbell, D. T. (1974b), ‘Unjustified Variation and Selective Retention in Scientific Discovery’, in Ayala, F. J. and Dobzhansky, T. (eds.), Studies in the Philosophy of Biology, London, Berkeley and Los Angeles: Macmillan and University of California Press, pp. 139–161.Google Scholar
Campbell, D. T. (1974c), ‘“Downward Causation” in Hierarchically Organized Biological Systems’, in Ayala, F. J. and Dobzhansky, T. (eds.), Studies in the Philosophy of Biology. London, Berkeley and Los Angeles: Macmillan and University of California Press, pp. 179–186.CrossRefGoogle Scholar
Campbell, D. T. (1976), Assessing the Impact of Planned Social Change, Hanover, NH: The Public Affairs Center.Google Scholar
Campbell, D. T. (1979), ‘A Tribal Model of the Social System Vehicle Carrying Scientific Knowledge’, Knowledge, 1(2): 181–201.CrossRefGoogle Scholar
Campbell, D. T. (1987), ‘Blind Variation and Selective Retention as in Other Knowledge Processes’, in Radnitzky, G. and Bartley, W. W. III (eds.), Evolutionary Epistemology, Theory of Rationality, and the Sociology of Knowledge, La Salle, IL: Open Court, pp. 91–114.Google Scholar
Campbell, D. T. (1988), Methodology and Epistemology for Social Science, Chicago: University of Chicago Press.Google Scholar
Campbell, D. T. (1989), ‘Rationality and Utility from the Standpoint of Evolutionary Biology’, in Hogarth, R. and Reder, M. (eds.), Rational Choice: The Contrast between Economics and Psychology, Chicago: University of Chicago Press, pp. 171–180.Google Scholar
Campbell, D. T. (1994), ‘How Individual and Face-to-Face-Group Selection Undermine Firm Selection in Organizational Evolution’, in Baum, J. A. C. and Singh, J. V. (eds.), Evolutionary Dynamics of Organizations, New York and Oxford: Oxford University Press, pp. 23–38.Google Scholar
Clift, B., Kristensen, P. M. and Rosamond, B. (2022), ‘Remembering and Forgetting IPE: Disciplinary History and Boundary Work’, Review of International Political Economy, 29(2): 339–370.CrossRefGoogle Scholar
Collins, H. (2010), Tacit and Explicit Knowledge, Chicago and London: University of Chicago Press.CrossRefGoogle Scholar
Crew, B. (2019), ‘The top 10 Countries for Scientific Research in 2018’, Nature Index, 1. https://www.natureindex.com/news-blog/top-ten-countries-research-science-twenty-nineteen. Retrieved 14 May 2022.Google Scholar
Fleck, L. (1979), Genesis and Development of a Scientific Fact, Translated from the German Edition of 1935, Chicago: University of Chicago Press.Google Scholar
Goldberg, S. C. (2010),Relying on Others: An Essay in Epistemology, Oxford: Oxford University Press.CrossRefGoogle Scholar
Goldman, A. I. (1999), Knowledge in a Social World, Oxford and New York: Oxford University Press.CrossRefGoogle Scholar
Goldman, A. I. (2009), ‘Social Epistemology: Theory and Applications’, Royal Institute of Philosophy Supplement, 64: 1–18, Published online.CrossRefGoogle Scholar
Gough, I. (2017), Heat, Greed and Human Need, Cheltenham, UK and Northampton, USA: Edward Elgar.CrossRefGoogle Scholar
Hayek, F. A. (ed.) (1935), Collectivist Economic Planning, London: George Routledge. Reprinted 1975 by Augustus Kelley.Google Scholar
Hayek, F. A. (1945), ‘The Use of Knowledge in Society’, American Economic Review, 35(4): 519–530.Google Scholar
Hayek, F. A. (1988), ‘The Fatal Conceit: The Errors of Socialism’, in Bartley, W. W. III (ed.), The Collected Works of Friedrich August Hayek (Vol. I), London: Routledge.Google Scholar
Hayek, F. A. (1989), ‘The Pretence of Knowledge’, American Economic Review, 79(6): 1–7.Google Scholar
Hodgson, G. M. (2019a), Is Socialism Feasible? Towards an Alternative Future, Cheltenham, UK and Northampton, MA: Edward Elgar.Google Scholar
Hodgson, G. M. (2019b), Is There a Future for Heterodox Economics? Institutions, Ideology and a Scientific Community, Cheltenham, UK and Northampton MA: Edward Elgar.Google Scholar
Jacobs, J. A. and Frickel, S. (2009), ‘Interdisciplinarity: A Critical Assessment’, Annual Review of Sociology, 35: 43–65.CrossRefGoogle Scholar
Kitcher, P. (1993), The Advancement of Science: Science without Legend, Objectivity without Illusions, Oxford and New York: Oxford University Press.Google Scholar
Kuhn, T. S. (1962), The Structure of Scientific Revolutions, Chicago: University of Chicago Press.Google Scholar
Longino, H. E. (1990), Science as Social Knowledge: Values and Objectivity in Scientific Inquiry, Princeton: Princeton University Press.CrossRefGoogle Scholar
Madsen, D. (2018), ‘Epistemological or Political? Unpacking Ambiguities in the Field of Interdisciplinarity Studies’, Minerva, 56(4): 453–477.CrossRefGoogle Scholar
Martin, B. R. (2012), ‘The Evolution of Science Policy and Innovation Studies’, Research Policy, 41(7): 1219–1239.CrossRefGoogle Scholar
Oğuz, F. (2010), ‘Hayek on Tacit Knowledge’, Journal of Institutional Economics, 6(2): 145–165.CrossRefGoogle Scholar
Ostrom, E. (2007), ‘Challenges and Growth: The Development of the Interdisciplinary Field of Institutional Analysis’, Journal of Institutional Economics, 3(3): 239–264.CrossRefGoogle Scholar
Paniagua, P. and Rayamajhee, V. (2022), ‘A Polycentric Approach for Pandemic Governance: Nested Externalities and Co-Production Challenges’, Journal of Institutional Economics, 18(4): 537–552.CrossRefGoogle Scholar
Polanyi, M. (1940), ‘Collectivist Planning’, in Polanyi, M. (ed.), The Contempt of Freedom: The Russian Experiment and After, London: Watts, pp. 27–60, chapter 2. Reprinted in the Journal of Institutional Economics, 15(6), December 2019, pp. 1055–1074.Google Scholar
Polanyi, M. (1948), ‘Planning and Spontaneous Order’, The Manchester School, 16: 237–268.CrossRefGoogle Scholar
Polanyi, M. (1958), Personal Knowledge: Towards a Post-Critical Philosophy, London: Routledge and Kegan Paul.Google Scholar
Polanyi, M. (1962), ‘The Republic of Science: Its Political and Economic Theory’, Minerva, 1(1): 54–73.Google Scholar
Stern, N. (2007), The Economics of Climate Change: The Stern Review, Cambridge and New York: Cambridge University Press.CrossRefGoogle ScholarPubMed
Sunstein, C. R. (2006), Infotopia: How Many Minds Produce Knowledge, Oxford and New York: Oxford University Press.Google Scholar
Surowiecki, J. (2004), The Wisdom of Crowds: Why the Many are Smarter than the Few and How Collective Wisdom Shapes Business, Economics, Societies, and Nations, New York: Doubleday.Google Scholar
Van Noorden, R. (2015), ‘Interdisciplinary Research by the Numbers’, Nature, 305: 306–307.CrossRefGoogle Scholar
Vatn, A. (2005), Institutions and the Environment, Cheltenham, UK and Northampton, MA: Edward Elgar.Google Scholar
Whitley, R. (1984), The Intellectual and Social Organisation of the Sciences, Oxford: Oxford University Press.Google Scholar
Zuckerman, H. (1988), ‘The Sociology of Science’, in Smelser, N. J. (ed.), Handbook of Sociology, London and New York: Sage, pp. 511–574.Google Scholar
You have
Access
- 4
- Cited by
Donald T. Campbell (1916–1996) was an American psychologist, philosopher and social scientist. He coined the term evolutionary epistemology and several of his works have achieved classic status. Much of his research centered on the nature, role, distribution, organization and evolution of knowledge. He was also a pioneer in the application of Darwinian evolutionary ideas in social science (Campbell, Reference Campbell, Barringer, Blanksten and Mack1965, Reference Campbell and Schilpp1974a, Reference Campbell, Ayala and Dobzhansky1974b, Reference Campbell, Ayala and Dobzhansky1974c, Reference Campbell1976, Reference Campbell1979, Reference Campbell, Radnitzky and Bartley1987, Reference Campbell1988, Reference Campbell, Hogarth and Reder1989, Reference Campbell, Baum and Singh1994). An earlier advocate of the application of Darwinian ideas to social science was the institutional economist Thorstein Veblen, although Campbell did not cite him (Camic and Hodgson, Reference Camic and Hodgson2011).
Campbell's (Reference Campbell, Muzafer and Carolyn1969) early paper on the ‘Ethnocentrism of disciplines and the fish-scale model of omniscience’ has received less attention than several of his other works listed above. But it should be of interest to readers of this journal, at least for the following reasons.
First, Campbell's essay is about the generally important role of institutions in the development of scientific research. Modern science is very much an organized endeavor (Whitley, Reference Whitley1984). This was understood by several earlier writers, but these social and structural dimensions were given more explicit attention by Kuhn (Reference Kuhn1962), Polanyi (Reference Polanyi1962) and several other authors who followed their inspiration. This was followed by a burgeoning literature on the ‘sociology of science’, with several contrasting positions being adopted (Zuckerman, Reference Zuckerman and Smelser1988). The study of (academic and related) institutions became part of the study of science. But much of the work in the sociology of science promoted versions of ‘social constructivism’ that denied the importance, or evaded the problem, of revealing truth (Goldman, Reference Goldman1999, Reference Goldman2009). This influential movement eclipsed the work of Polanyi and Campbell in this area. In contrast to social constructivism, they saw adaptive and responsive (imperfect but necessary) institutions as potential inquirers into truth.
Second, Campbell's paper addresses the problems involved in crossing disciplinary boundaries. As well as the general issues and difficulties with interdisciplinary research, his argument is particularly important for researchers into institutions, many of whom argue that their research should draw from multiple disciplines, including from psychology, philosophy, politics and law, as well as from economics (Ostrom, Reference Ostrom2007). Institutional research is focused one set of social phenomena, but the complex and multi-faceted nature of the subject matter drives it into several disciplines. Without a clearly bounded home of its own, it encounters problems of academic influence, security and funding, which are explored by Campbell in this essay.
At the core of Campbell's argument is the presumption that not all relevant knowledge can be brought together, shared and processed by a single agent or decision-making body. Much important knowledge is particularistic and context dependent, making it impossible for collective planners to embrace all relevant knowledge. As Friedrich Hayek (Reference Hayek1989: 4) put it in his Nobel lecture, the facts ‘in their totality cannot be known to the scientific observer, or to any other single brain’. Like Hayek (Reference Hayek1935, Reference Hayek1945), Polanyi (Reference Polanyi and Polanyi1940, Reference Polanyi1948) wrote similarly of the impossibility of bringing all relevant knowledge together for the purposes of central planning. Both authors referred to tacit knowledge – things we know but cannot adequately articulate – as one of the major and permanent impediments to collective omniscience (Collins, Reference Collins2010; Hayek, Reference Hayek and Bartley1988; Hodgson, Reference Hodgson2019a; Oğuz, Reference Oğuz2010; Polanyi, Reference Polanyi1966). As well as to the critique of socialism, Polanyi (Reference Polanyi1958, Reference Polanyi1962, Reference Polanyi1966) applied this understanding of knowledge to his analyses of science and to his critique of the idea that science could be comprehensively planned.
Campbell's line of thought was similar. He declared that a single scientist cannot grasp all that is going on his or her own discipline, let alone in science as a whole. He quoted Polanyi (Reference Polanyi1966: 72) on ‘chains of overlapping neighborhoods’ where ‘scientists keep watch over each other’ on matters of veracity and rigor. This became Campbell's optimal ‘fish-scale model of omniscience’, where, given the impossibility of centralizing all knowledge, scientific specialisms overlap with one another, providing opportunities for adjacent scrutiny and synergy.
Much of Campbell's article is devoted to an explanation of how actual science departs from this ‘fish-scale’ optimum. This ideal outcome is thwarted by ‘the ethnocentrism of disciplines’, with deleterious consequences for scientific progress. Disciplinary boundaries are often arbitrary, resulting from historical outcomes or peculiarities of lesser current relevance. Despite huge changes within the disciplines in the 20th century, their boundaries are remarkably resilient (Abbott, Reference Abbott2001). Increasing international competition between different university departments in the same discipline creates further incentives to keep the boundaries as they are. Yet the enduring powers and boundaries of departments and disciplines inhibit innovation and development in science.
Campbell (Reference Campbell, Muzafer and Carolyn1969: 331) emphasized that ‘the locus of scientific knowledge is social’ where ‘the discipline, the verification, the integration are all in the end social products, imperfectly and incompletely represented in the work of any one scientist’. This stress on the role of institutions in guiding the production and retention of knowledge, makes Campbell, along with Polanyi, as a pioneer of what is now called ‘social epistemology’ (Fuller, Reference Fuller1988; Goldberg, Reference Goldberg2010; Goldman, Reference Goldman1999, Reference Goldman2009; Kitcher, Reference Kitcher1993; Longino, Reference Longino1990; Sunstein, Reference Sunstein2006; Surowiecki, Reference Surowiecki2004). Fleck (Reference Fleck1979) had previously explored the social dimensions of knowledge, in a work originally published in German in 1935. Hayek deserves some credit too, but he focused exclusively on the role of markets in processing knowledge. In contrast, Polanyi (Reference Polanyi1962: 57) wrote more broadly that the ‘the coordinating functions of the market are but a special case of coordination by mutual adjustment’. Crucially, this ‘coordination by mutual adjustment’ existed in non-market institutional contexts as well, including – for Polanyi and Campbell – within the institutions of scientific enquiry. (Another relevant example is the legal system.)
Given this view of knowledge as simultaneously social and parcelized, Campbell argued that the popular goal of ubiquitous interdisciplinarity is an impossible dream. It is a ‘Leonardesque aspiration’ that cannot work with modern science, with its high degrees of professionalization and specialization (Whitley, Reference Whitley1984). Yet even today, incautious calls for ‘interdisciplinary research’ are commonplace. There have been ‘widespread efforts to promote interdisciplinarity in U.S. universities’ (Jacobs and Frickel, Reference Jacobs and Frickel2009: 43) and in academies in many other countries. Part of the problem is that what is meant by ‘interdisciplinary’ is capable of multiple interpretations (Madsen, Reference Madsen2018).
A study by Richard Van Noorden (Reference Van Noorden2015) found that the percentage of Web of Science research papers citing work outside the authors' own discipline has been increasing since the 1980s, in both the natural and the social sciences. He also found that interdisciplinary research took much longer to have an impact. Also, the countries that are world leaders in scientific research (Crew, Reference Crew2019) are not necessarily the most prolific in interdisciplinary terms. The USA, UK and Germany were behind India, China, Taiwan, South Korea, Brazil and Italy by one measure of their proportions of interdisciplinary output. Interdisciplinary rhetoric is fashionable, but the reality is uneven.
Instead of unrestricted interdisciplinarity, Campbell favored an approach involving interdisciplinary specialism. He did not develop this idea in much detail, but he favored it as a strategy for crossing and transcending disciplinary boundaries while avoiding the shallowness of wider interdisciplinarity. For him it was vital to attain a high degree of integration and communication between scholars from different disciplinary backgrounds, by focusing on well-defined problems, topics or areas of research.
Examples of interdisciplinary specialism could include the study of pandemics (such as COVID-19) and the design of policies to deal with them. Here it is important to combine the knowledge and insights of researchers from both the natural and the social sciences. Natural scientists need to understand the nature of the virus, the mechanisms of transmission and the possible impacts on human public health. Social scientists are also important, to analyze how people react to information about the disease and how they can be persuaded and incentivized to follow measures to limit its impact and spread. Researchers into institutions have a major role here (Bentkowska, Reference Bentkowska2021; Paniagua and Rayamajhee, Reference Paniagua and Rayamajhee2022).
Climate change presents other challenges that should involve both environmental and behavioral scientists. Environmental research is necessary to understand the extent and impact of greenhouse gas emissions, likely scenarios of climate change, and their impact on the natural environment. Social scientists come in to evaluate the risks, costs and benefits, to appraise popular perceptions of the problem, to design institutions to constrain harmful emissions, and to consider ways in which people can be persuaded and incentivized to change their behavior to reduce the deleterious consequences. This too involves matters of institutional function and design (Gough, Reference Gough2017; Stern, Reference Stern2007; Vatn, Reference Vatn2005).
Studies of pandemics and of climate change offer two cases where multi-disciplinary research can be focused on a major problem, with urgent needs for policy responses. A broader example of multi-disciplinary research is the field of innovation studies, which cohered and took off in the 1980s, and it now sports its own institutes, academic positions, journals and degrees (Martin, Reference Martin2012). Another example, of lower but still significant success, is the field of international political economy, which spans politics, economics, geography and other traditional disciplines (Clift et al., Reference Clift, Kristensen and Rosamond2022).
Campbell wrote his article before the international take-off in the number of business schools after 1980. In some respects these offer havens for interdisciplinary study. But international competition between business schools pushes them to fight for positions in each of multiple disciplinary rankings, several of which are global. If you are labeled as an ‘economist’ of any kind, then you are obliged to take the rankings for economics seriously, including those of journals in mainstream economics. Anyone in a business school is competing with colleagues over promotion, funding and so on. The replication of disciplinary compartmentalization within business schools is a barrier to interdisciplinary research (Hodgson, Reference Hodgson2019b: 155–156).
An alternative strategy would be to attempt to replicate the success of innovation studies and to set up multi-disciplinary institutes of institutional research (Hodgson, Reference Hodgson2019b: 172–174). This would build on the widespread acknowledgement that ‘institutions are important’ in the process of development, in dealing with corruption and other major problems. The problem here, as Campbell (Reference Campbell, Muzafer and Carolyn1969) and Abbott (Reference Abbott2001) pointed out, is that existing disciplines are notoriously protective and conservative, and they will guard internal university funds from possible competition. As with innovation studies, much of the funding for multi-disciplinary institutional research may have to come from outside.
Although it was written more than half a century ago, Campbell's essay is still of immense importance and value. About half of its text is reprinted here. Much of the remainder includes discussions of interdisciplinary initiatives in US universities in the 1950s and 1960s. Although these are interesting and important, they have been largely overshadowed and superseded by other developments since 1969. Campbell wrote before the rise of the Internet, which is having a growing impact on research networking and collaboration. The remaining text, reproduced here, is largely devoted to analytical issues that have retained their vibrancy and significance.Footnote 1
This paper is a preliminary exercise in the sociology of science – an exploratory application of principles of groups and intergroup organization to group processes in the institutionalization of science.Footnote 2 Our goal … is a comprehensive, integrated multiscience. The obstacle described in this paper is the ‘ethnocentrism of disciplines’, i.e. the symptoms of tribalism or nationalism or ingroup partisanship in the internal and external relations of university departments, national scientific organizations, and academic disciplines. The ‘fish-scale model of omniscience’ represents the solution advocated, a solution kept from spontaneous emergence by the ethnocentrism of disciplines. The slogan is collective comprehensiveness through overlapping patterns of unique narrownesses. Each narrow specialty is in this analogy a ‘fish-scale’. Figure 1 illustrates the title. Our only hope of a comprehensive social science, or other multiscience, lies in a continuous texture of narrow specialties which overlap with other narrow specialties. Due to the ethnocentrism of disciplines, what we get instead is a redundant piling up of highly similar specialties, leaving interdisciplinary gaps. Rather than trying to fill these gaps by training scholars who have mastered two or more disciplines, we should be making those social-organizational inventions which will encourage narrow specialization in these interdisciplinary areas.
(a) Present situation: disciplines as clusters of specialties, leaving interdisciplinary gaps. (b) Ideal situation: fish-scale model of omniscience.
The diagram of Figure 1 is of course an oversimplification, an analogy in two dimensions of what should be n-dimensional. Particularly likely to be dyscommunicative are the exaggerated gaps between the disciplinary clusters. The real situation is perhaps more often one of unrecognized overlap. The disciplinary clusters may at their edges overlap other clusters, but ships that pass in the night, they fail to make contact. The clusters, as it [329] were, may overlap but lie on independent planes. Such an alternate diagramming might be possible, and the reader is invited to substitute it in his mind's eye in what follows. The issue, too, is not one of total absence but of relative density of interdisciplinary specialties.
Comprehensive training and the Leonardesque aspiration
Too often in discussions of interdisciplinary training one hears calls for breadth, for comprehensiveness. Too often we attempt the production of multidisciplinary scholars, professionals who have mastered two or more disciplines, rather than interdisciplinary specialists. This orientation I will parody as the ‘Leonardesque aspiration’: the goal of creating current-day [330] Leonardos who are competent in all of science. As a training program it is bound to fail in one of two directions. At its worst it produces a shallowness, a lowest-common-denominator breadth, an absence of that profound specialization which is essential for scientific productivity. At best it is evaded in the direction of the interdisciplinary narrowness here advocated.
….
The myth of unidisciplinary competence
Lying behind many models of interdisciplinary competence is an unrealistic notion of unidisciplinary competence – the image of scholars competent in one discipline. It will clarify the discussion of interdisciplinary competence to recognize at the outset that there are no such persons. What we have instead is a congeries of narrow specialties each one of which covers no more than one-tenth of the discipline with even a shallow competence. Yet individual disciplines do have some integrity, some comprehensiveness – at least in comparison with social science as a whole or with specific interdisciplinary areas. What must be recognized is that this integration and comprehensiveness is a collective product, not embodied within any one scholar. It is achieved through the fact that the multiple narrow specialties overlap, and that through this overlap a collective communication, a collective competence and breadth, is achieved. This approach is our only hope for a unified and complete behavioral science. The present social organization of science impedes it. [331]
The locus of scientific knowledge is social
Philosophy of science and epistemology have not yet assimilated the fact that the problem of knowledge must, in the end, be stated at the social level – though Charles Sanders Peirce and James Mark Baldwin, for example, were making this point at the turn of the century. Moving the problem of knowledge from a solitary viewer's vision to language is a step, but the implicit model is still usually a single native speaker with perfect knowledge of a stable language. Sufficient attention is not yet given to the social and incomplete conditions of language learning, to the fundamental idiosyncracy and errorfulness of functional individual lexicons, to the very partial distribution of words that are still somehow ‘in’ the language, to the effective redundancy which makes imperfect language as competent as it is. When these have been assimilated, the locus of ‘truth’ and ‘knowledge’ will have clearly shifted from individual ‘minds’ to a collective social product only imperfectly represented in any one mind. Similarly in the philosophy of science, the competence, the discipline, the verification, the integration are all in the end social products, imperfectly and incompletely represented in the work of any one scientist. Michael Polanyi (Reference Polanyi1966: 72) writes to this point in The Tacit Dimension identifying the locus of scientific authority as the ‘Society of Explorers’ itself:
… the principle of mutual control. It consists, in the present case, of the simple fact that scientists keep watch over each other. Each scientist is both subject to criticism by all others and encouraged by their appreciation of him. This is how scientific opinion is formed, which enforces scientific standards and regulates the distribution of professional opportunities. It is clear that only fellow scientists working in closely related fields are competent to exercise direct authority over each other; but their personal fields will form chains of overlapping neighborhoods extending over the entire range of science. It is enough that the standards of plausibility and worthwhileness be equal around every single point to keep them equal over all the sciences. Even those in the most widely separated branches of science will then rely on each other's results and support each other against any laymen seriously challenging their authority.
Present disciplines as arbitrary composites
While it is probably not essential to the perspective, it is certainly relevant that the present organization of content into departments is highly arbitrary, a product in large part of historical accident. …
There are no doubt many natural divisions within the domain of the social or behavioral sciences – but they are not employed in the allocation of content to disciplines. A hierarchy of levels of analysis exists in which the focus of differential description at one level becomes the assumed undifferentiated atoms of the next; this is the atom-molecule-cell-organ-organism-social group-etc. model. On this hierarchy, sociology, political science, geography. and anthropology are all mixed across the individual and group levels, and so is experimental social psychology. …
Another natural division is between the descriptive-humanistic, on the one hand, and the scientistic, on the other. On this dimension, too, our departments and disciplines are mixed, with all save history having both strong scientific and strong descriptive-humanistic factions.
There are ways of describing the internal logic and coherence of disciplines, but this cannot be done with singular principles and still capture both the reason for the content within one discipline and the reason for the content's appearance or exclusion in others. And certainly the dimensions so used would be only a partial sample of potential classificatory criteria.
The specialties within disciplines are more coherent, and eventually specialization takes over, each scientist allowing the congeries of irrelevancies within his own disciplinary knowledge to atrophy, journals to go unread, subscriptions to lapse, etc. The temporary disciplinary breadth transiently achieved in graduate school is of course not undesirable – the objection here is rather to the repetitious duplication of the same pattern [333] of breadth to the exclusion of other breadths equally relevant but organizationally unsupported.
Ethnocentrism of disciplines: effects of organizing specialties into decision-making units
Consider what would happen if we took a large domain of specialties and aggregated adjacent dozens into ‘departments’, as collective-decision-making units. In this hypothetical example, let us suppose that the aggregation has been arbitrary except for the requirement of adjacency, that all specialties are equally well staffed to begin with, and that any specialty can belong to only one department. We are now interested in the effect of this second-level organizational structure, this superimposition of departmental boundaries upon the specialty boundaries. We are particularly interested in differential effects upon the future growth of the ‘central’ specialties versus the peripheral or marginal specialties, this centrality or marginality being in this hypothetical case a purely arbitrary byproduct of where the administrative boundaries happened to fall. Figure 2a is an effort to portray this starting point. (In Figure 2, Ap, Bp, Cp, are examples of peripheral specialties, Dc and Ec of central ones.)
(a) Hypothetical pattern of specialty overlap at the time of superimposition of arbitrary ‘departmental’ boundaries (heavier lines). (b) Resulting modification of specialty overlap as a result of organizing decision-making and communication along the arbitrary ‘departmental’ lines. (‘Departmental’ boundaries omitted to facilitate inspection of specialty overlap pattern.)
Consider first purely internal decision-making. For most issues there are differences in priority and preferences which are associated with specialty points of view, and overlapping specialties are apt to have overlapping preferences. [334] The situation will also be that collective decisions will be achieved only by a consensus of a plurality of specialties within the department. The accidentally central specialties have more natural allies within the department, and find it easier to achieve support for their concerns. The natural allies of the peripheral specialties lie in other of the arbitrary departments, and are organizationally prevented from effectively presenting their consensuses. The incidentally central specialties are also more frequently the compromise candidates, and when an ideology is needed to rationalize the historically arbitrary departmentalization it is the concerns of the central specialties that are chosen as epitomizing the true common denominator. The essence of the initially arbitrary aggregate. Centrality becomes reinterpreted as common root, trunk, and fountain head when initially it meant only remoteness from the boundaries with other departments.
Here are some considerations which would illustrate these purely internal dynamics. The selection of a chairman in a peaceful department will follow these lines – only if a peripheral is a compromise between two strong factions each with some central specialties will he be a chairman. Deciding a core curriculum to be required of all students, the minimum essential their being sound, well-grounded X-ologists, will go in this direction. Pressure there will be to require one course from each specialty. Demand, time to meet the needs of specializing will preclude this in any but the smallest departments. The process of choosing a smaller set will involve the elimination [335] of peripheral specialties because inherently in the arbitrary organizational structure there are fewer other specialists in the department who deem them important. This tautology may of course be expressed ideologically as ‘that's not really X-ology – it even comes close to being Y-ology’. The setting of qualifying exams and dissertation committees will exercise a constant centralizing bias on the training of graduate students in peripheral specialties. Going across departmental lines in study programs is wasted effort as far as these important hurdles are concerned. The peripheral specialist himself will be anxious that his students show up well on the central core content, and will be willing to see equally or more relevant crossdepartment content be neglected, since no punishment is involved in its neglect, no institutional reward in its achievement. This is a minor problem if these departmental hurdles are low. But there is great pressure on departments to achieve excellence, and there is a perverse tendency to see this as implemented by high standards in the achievement of passive regurgitative mastery of past achievements in the literature. Under the institutional decision-making arrangements here described, the inevitable effect of higher standards in training is greater neglect of peripheral and crossdepartmental content. And this is without the byproduct of increasing the profundity of specialized training for any but the central specialties.
Deciding who in the department merits a raise or is ready for promotion, whose competing offer it is essential to meet and whose the department has not the funds to match, whose specialty needs additional staff and space, all show the effect of the arbitrary location of departmental boundaries on what becomes defined as central or peripheral. In my professional career at four universities I have repeatedly seen men who were of exceptional creativity and competence, and who were absolutely central as far as social science or behavioral science was concerned (occupying positions like those designated as H or N in Figure 2a), be budgetarily neglected and eventually squeezed out because the departmental organization of specialties made them departmentally peripheral. That in some cases these were arrogant men does not explain the result, because their detractors were also arrogant men, but more centrally located. However, a selective feedback may well be at work, it may be that of all those who are initially attracted to peripheral, boundarycrossing specialties, only the arrogant persist in bucking the institutional pressures which would otherwise move them to more centrally defined specialties.
The dynamics just described are internal to our hypothetical departments. These arbitrary budgetary units compete with each other as budgetary entities, for budget increases, space increases, personnel increases. The specialties within the arbitrary departments thus come to share common fate, and become joint actors in competition with the other arbitrary aggregates of specialties. This common fate, though arbitrary in its initiation, is real enough in practice to provide the basis of an ingroup identification against [336] competitive outgroups. There also develops implicit or explicit competition for the most talented students, and indoctrination procedures designed to maintain the loyalty of those who have tentatively joined. Characteristic of ingroup–outgroup relations in other settings, these indoctrination procedures not only emphasize ingroup virtues and ideology but also contrasting outgroup faults. Philosophy and sociology departments have frequently maintained internal solidarity by teaching about the wrongness of behavioristic psychology. Sidney Aronson (Reference Aronson, Muzafer and Carolyn1969) in his paper for this volume has documented the manner in which historians build ingroup morale by deprecating sociologists. (It is even symptomatic of ethnocentrism that the first illustrations that come to his and my mind are ones in which our own departments are being wrongly disparaged by outgroups, rather than vice versa.)
If we add to our hypothetical example the feature that on most university campuses the arbitrary aggregations into departments be parallel as to which specialties are combined, further institutional pressures emerge. Departments' students must be prepared to appear adequate to centrally dominated hiring committees at other universities. The new faculty appointments to the department must be ones that inspire admiration on the part of the parallel departments of other universities.
Effect of departmental organization on scientific communication and specialist competence
In this topic we come to the most direct effect of departmental organization on scientific knowledge. Each scientist's competence, his participation in the collective activity of science, is based upon communication. The hypothetical departmental organization under consideration affects communication patterns in many ways. In our hypothetical example, suppose that those specialties aggregated into an arbitrary department be housed adjacently. But that departments be scattered at random. Incidental oral and paperpassing communication links thus become predominately intradepartmental. And the extradepartmental aspects of peripheral specialties suffer great relative neglect in comparison with their intradepartmental overlaps. Shop talk reading of dissertations, reading of each other's preprints and reprints, looking at laboratory setups and research instruments, all illustrate primary modes of communication seriously warped. The bias extends also to books and journals read. One is rewarded socially for shared detailed reading of exciting new developments. No such reward occurs for unshared reading and thus the literature in the crossdepartmental aspects of a specialty loses ground to the reinforced intradepartmental reading.
With a parallel arbitrary organization occurring at other universities and in national and international disciplinary organizations, still other boundary effects occur. Professional organizational membership and their journal discounts to members lead to stereotyped patterns of journal subscription with most members limiting themselves to journals within one field – these are [337] invariably so voluminous that one excuses himself for not reading in other fields by noting that this would be foolish when he is not even able to keep up in ‘his own field’. ‘Own field’, needless to say, tends to become defined in terms of these arbitrary departments whenever it goes beyond the narrow specialty. Abstracting sources, annual reviews, handbooks, etc., further the redundant repetition of intradepartmental patterns of breadth at the expense of equally relevant crossdepartmental ones.
(The annual reviews and handbooks do another disservice from the fishscale model. Even within departments, each scholar's coverage must be partial, collective competence being assured by different partialities for each. Annual reviews and handbooks become crutches leading large numbers of scholars to the same partiality, and the inevitable neglect of other partialities. The degree of this partialness is generally underestimated, impressed as we must be at the long bibliographies appended.)
The departmental grouping of communicators allows unstable language to drift into unintelligibility across departments. A basic law is that speakers of the same language, once isolated into separate communities, drift into local idiosyncrasies and eventually unintelligibility, once the discipline of common conversation is removed. This tendency produces departmental linguistic idiosyncrasy even for shared contents and referents. Furthermore, as Edmund Leach and others have noted, such idiosyncrasy may be exaggerated as an ingroup solidarity device. What is despised as jargon by the outgroup may be the shibboleth of adequate professional training by the ingroup.
Resulting ethnocentrism of disciplines
Figure 2b shows the results of these dynamics. A few isolated specialties have been lost and the peripheral specialties have allowed their crossdepartmental edges to atrophy, producing departmental discreteness and interdepartmental gaps.
What started out as a hypothetical case has occasionally in the previous paragraphs drifted into the description of current actualities, perhaps blurring the point being made: even if the true nature and the historical starting point had been one of a homogeneous texture of overlapping specialties, the organizing of specialties into departments and disciplines for decision-making and communication purposes would have produced disciplinary discreteness, cohesiveness, and interdisciplinary gaps such as exist at present. These features therefore should not be judged to justify as natural the existing arrangements of specialties.
The historical origins of departments and disciplines are of course quite other than those of our hypothetical example. The actual histories add historical depth to the tribal myths of origin and no doubt provide a greater true commonality to departments than in our hypothetical instance. Yet no matter what the degree of valid core and discreteness, the dynamics described here [338] would inevitably have the effect of artificially enhancing it. … [348] …
Summary
Interdisciplinary programs have been misled by goals of breadth and multidisciplinary training. Even within disciplines, disciplinary competence is not achieved in individual minds, but as a collective achievement made possible by the overlap of narrow specialties. This fish-scale model of collective omniscience is impeded in interdisciplinary specialty areas by the ethnocentrism of disciplines, by the organization of specialties into departments for decision-making and communication. For an integrated and competent social science, we need to invent alternative social organizations which will permit the flourishing of narrow interdisciplinary specialties.