Masato Fukushima
One of artist René Magritte’s favourite surrealistic tricks is to remove objects out of their proper context. An apple of extraordinary size that takes up an entire vacant room (in The Listening Room 1958) strikes us, rather paradoxically, with the impact of the image, despite the simplicity of the trick. Together with, say, Ron Mueck’s photo-realistic sculptures of human figures of similar enormity, this painting offers a universal lesson about the meaning of scale in our everyday lives where things of ordinary size are peacefully immersed.
Scale does matter in many circumstances, including interdisciplinary collaborative work, if it is to be performed properly. Suppose, for example, that a country’s president has been shot by international terrorists and is rushed to an emergency operation. The ER staff might dream of collecting a team of talented surgeons – the allure of doing so is undeniable – but such a team of 100 members of the brightest specialists from all over the world jostling one another in an overcrowded operating room – like Japan’s world famous rush hour on public transportation – would create the kind of apocalyptic nightmare that Magritte might have been amused to paint.
In his classic work on software project management, Brooks (1975) argues that the idea of a ‘man month’ – that time and manpower are interchangeable – is a myth, especially in software development. Brooks convincingly maintains that even if time is short, it cannot always be complemented by the increase of manpower, most notably in software development. What might happen instead is that infusing additional manpower into an already tardy project further delays the process because a great deal of trouble is involved in providing instruction to the newcomers, setting up a more complex path of communication, and eventually, increasing the difficulty of managing such a project. Just remember the poor president who lost his life in an operation room packed with top-class doctors!
The lesson here is that scale is often essential in determining the success or failure of interdisciplinary collaborative work. These examples hint that to scale up the size of a group tends to result in decreased efficiency. This can be lethal if the needed operation is conducted in hazardous situations, such as the control tower of an airport or the command room for military operations, similar to the routines in an operation room. In such situations, the whole undertaking often requires concentration of mind and the tight collaboration that results from control by a single person (LaPorte and Consolini 1991).
One might wonder, however, if the cases thus far presented are adequate for considering the proper scale for interdisciplinary research practices, as the former require decisions within minutes or even seconds, not months or years, whereas research collaboration might not need to be as tightly organized as observed in these cases. Though the circumstances are not as dramatic, scale matters in research collaboration too. Collingridge (1996) enumerates a negative inventory of all the possible troubles that may be encountered by practitioners of any project on a large scale, ranging from the sheer difficulties of management to lack of flexibility, slow response, and inevitable bureaucratization.
Behind this view of the malaise inherent in scaling up projects lies a pivotal issue, generally referred to as the problem of ‘transactional cost’, which was originally conceived to prove that enterprises must exist to deal with market transactions so that all related costs are lower than those of individually handled transactions (Coase 1960; Williamson 1981). This concept can be extended to analyse required costs in organizations for interdisciplinary coordination, negotiation and collaboration when their scale grows. The man month myth also arose out of such analyses of ‘economies of scale’ for larger organizations.
Despite such insights, in interdisciplinary collaboration, the scale of number seems to continue to be important, as represented by so-called big science (Galison 1992) and, more specifically in recent years, big biology (Parker, Penders and Vermeulen 2010; Fukushima 2016). This bigness derives from the growing size of both the research organizations and the scientific instruments needed for more extensive research, with examples ranging from bigger accelerators and superscale radio telescopes to the more advanced second-generation sequencers of DNA. Global issues, such as climate change, also demand such enlargement of size in research collaboration (Edwards 2010). The once lively argument about ‘the end of science’ stemmed partially from the pessimistic prediction that the growth of such a scale of instruments in terms of both cost and manpower would necessarily have limits.
There are at least two interesting corollaries to this issue of the rising size of interdisciplinary research. One is that coordination is needed not only for possible conflicts among the different instruments and even different epistemic cultures (Knorr-Cetina 1999) – such as, say, between the wet (traditional) and the dry (informational) approaches to biology (Fukushima 2016) – but also for managing the data that is produced from such research. Edwards, Mayernik, Batcheller, Bowker and Borgman (2011) refer to such conflicts between the different formats of data and metadata, as well as the difficulty of coordinating them, as ‘science friction’. Their tentative but intriguing conclusion that such coordination is always ongoing and only temporary echoes Brooks’ (1975) similar conclusion regarding the software project, as mentioned above. However, the somewhat pessimistic wisdom gained about restricting the size of the developing team of software – as well as that from realizing the impossibility of a dream team of 100 surgeons for operating on the president – apparently does not transfer well, as the growth in the number of scales (size, velocity, heterogeneity, speed) of data and metadata seems unstoppable.
Another issue, though having a slightly different focus, are the contradictory vectors of values used for evaluating scientific research at large: namely, the pressure to upgrade the scale(s) of interdisciplinary research and the intrinsically individualistic view of scientific feats, which becomes most evident in the reward system that is the essence of its valuation. Historical tragicomedies related to the Nobel Prize – namely, who is in and who is out when the feat itself has been carried out collectively – exemplifies such contradictions. Intriguingly, this kind of contradiction demonstrates a striking kinship with the rising discussion about so-called ‘organizational accidents’ and issues relating to responsibility for them.
Researchers of organizational accidents have repeatedly emphasized that surface causes of large-scale accidents are, in fact, the outcome of an accumulation of chain reactions arising from smaller causes distributed throughout the organization (Reason 1997). The lethal mistake of a nurse who injected a patient with the wrong medication, for example, could have resulted from a series of blunders that went unchecked through various stages in the organization. Because of our individualistic assumptions concerning the cause of such accidents, those who are at the very end of this long chain of causal relations are often accused as the person most responsible.
The reward system in scientific endeavour seems to reflect this individualistic bias. The only difference is that scientists at the very end of such a large chain of causation are individually hailed, rather than put in jail. Alexander Fleming, who has been elevated to the status of a sort of scientific saint, is a case in point, as I realized in my observations of antibiotic labs: the impact of his antibiotic discovery is scaled out of proportion because of its ultimate global impact on various diseases, whereas the more substantial contributions by those who have made antibiotics available for medical purposes have never made the spotlight of the popular imagination (Macfarlane 1984).
In recent media coverage, Gregory Perelman also attracted my attention as the Russian mathematician who finally solved the Poincaré conjecture. Nonetheless, he rejected the Field’s Prize, allegedly because of his distaste for the individualistic bias of the present academic award system (O’Shea 2008). A TV programme in Japan blatantly promoted this cult of individualism in science by not even mentioning Richard Hamilton, whose Ricci flow technique Perelman improved before applying it to prove the conjecture, in its dramatic presentation of Perelman as an extraordinary genius coming out of nowhere (NHK Enterprise 2010). These cases above are but small examples of the adamantly predominant individualistic image of scientific practice, which endures even against the growing scale(s) of accumulated contribution of past and present efforts that have culminated in success.
In sum, scales matter in various aspects of our everyday lives, but even more so in interdisciplinary research with the growing collaboration and historically accumulated efforts of the past. Such scaling up of research is Janus-faced: it is necessary for solving the more complex issues that now apply on a global ‘scale’, but it also confronts us with internal problems of disorganization and friction, along with the persisting individualism in the reward system. The imaginary case of the over-congested operating room easily demonstrates the internal constraints of proper organization in some situations. In other cases, however, the network of collaboration may be further extended, even if a re-examination is needed of the foundations that afford such expansion, either in the form of material infrastructure, or of our belief system that is intrinsically individualistic, if both accidents and scientific success are very often organizational.
Perhaps what Magritte wanted to convey to perplexed viewers with The Listening Room was exactly this: the dire consequences of large-scale apple growth in a vacant room – to whose creaking sound we should listen as a symptom of its acute need for remodelling – though, in fact, nobody seems to care in the typical manner of his paintings.
Brooks, F. (1975). The Mythical Man-Month: Essays on Software Engineering. Reading, MA: Addison Wesley.
Coase, R. (1960). The problem of social cost. Journal of Law and Economics, 3: 1–44.
Collingridge, D. (1996). The Management of Scale: Big Organizations, Big Decisions, Big Mistakes. London: Routledge.
Edwards, P. (2010). A Vast Machine: Computer Models, Climate Data, and the Politics of Global Warming. Cambridge, MA: The MIT Press.
Edwards, P., Mayernik, M., Batcheller, A., Bowker G. and Borgman, C. (2011), Science friction: data, metadata, and collaboration. Social Studies of Science, 41: 667.
Fukushima, M. (2016). Constructing ‘failure’ in big biology: the socio-technical anatomy of the Protein 3000 Program in Japan. Social Studies of Science, 46(1): 7–33.
Galison, P. (1992). The many faces of big science. In P. Galison and B. Hevly (Eds.) Big Science: The Growth of Large-Scale Research (pp. 1–17). Stanford: Stanford University Press.
Knorr-Cetina, K. (1999). Epistemic Cultures: How the Sciences Make Knowledge. Cambridge, MA: Harvard University Press.
LaPorte, T. and Consolini, P. (1991). Working in practice but not in theory: theoretical challenges of ‘high-reliability organizations’. Journal of Public Administration Research and Theory, 1: 19–47.
Macfarlane, G. (1984). Alexander Fleming: The Man and the Myth. Cambridge, MA: Harvard University Press.
NHK Enterprise (2010). One Hundred Years’ Struggle for Poincare Conjecture: Mathematicians Dream of Mushroom Hunting. (DVD) Tokyo: NHK Enterprise (in Japanese).
O’Shea, D. (2008). The Poincaré Conjecture: In Search of the Shape of the Universe. London: Penguin Books.
Parker, J. N., Penders, B. and Vermeulen, N. (Eds.) (2010). Collaboration in the New Life Sciences. Farnham, Surrey: Ashgate.
Reason, J. (1997). Managing the Risks of Organizational Accidents. Aldershot: Ashgate Publishing Ltd.
Williamson, O. (1981). The economics of organization: the transaction cost approach. The American Journal of Sociology, 87(3): 548–577.