Review of: The Golem: What You Should Know about Science, and The Golem at Large: What You Should Know about Technology, by Harry Collins and Trevor Pinch, Cambridge University Press, 1993 and 1998, respectively.
Most of the literature on the sociology and philosophy of science and technology reminds me of Mark Twain’s story about his cutting himself while shaving and then swearing a blue streak. His wife Livy overheard him, and, disapproving of his ‘cussing’, stopped at the bathroom door and repeated his stream of ‘swear words’, word for word, in hopes of shaming him. He looked at her and said: ‘My dear, you have the words, but not the tune’. Reading social science discussions of science is usually like reading the music criticism of the tone deaf: they get the notes, but not the tune.
The Golem books are better than most. When I first looked into them I was enthusiastic, particularly at the careful, and, for the most part, sensible way they look at real situations and cases as the principal content of the books. The books are simple and straightforward, addressed to a general public, and the writing is generally clear, concise and to the point. The second book (technology) is clearer than the first (science) and makes their message clearer. It is thus important that both books be read.
On later, closer reading I became less sure of my enthusiasm, since, while they get part of the process right, and keep saying they understand the rest, they don’t discuss all of the process, and may not clearly understand it. This incompleteness and somewhat biased emphasis will mislead readers. Some summing up sentences and paragraphs seem to me to me to be particularly misleading. I will discuss some of the general issues I see, and then discuss some of the individual cases and statements that either trouble me, or about which I have something particular to say.
The titles are a little misleading as the books really tell you only: ‘some of what you should know about science/technology’. You should know more than the early stage processes of science/technology. It is important to understand the way in which cumulative refinements and advances in theory and further experiments either consolidate or contradict early ideas, and how the whole process of ‘reworking’ bodies of scientific material can contribute to increased understanding of the body of material. At various points the authors refer to this later part of the process, in a ‘yes, yes we know’, tone, but then return to the principal, but incomplete, message.
The idea of the Golem is not really as neutral as the authors suggest: just a big dumb creature. In talmudic tradition and Jewish thought there is a taint of evil attached to the creature and the one who makes it, not much different than the resonances of Frankenstein’s monster. The titles are thus likely to be pejorative in flavor to those who know the word and the tradition, although the authors deny this intent.
the message of the books
Their principal message is that histories, textbooks and teaching present the processes of science and technology as deductive, neat and clear, while they are really craft processes, and, especially in their initial stages, are messy and uncertain. The message is correct, but while certainty of knowledge is certainly impossible, the messiness of the early part of the process of getting to reasonably certain knowledge is not a particularly good test of its later degree of certainty. The process is, after all, not necessarily the same as the product. Uncertain and very messy cooking processes frequently produce great meals. Few composers are Mozart, and can compose musical works in their head and then either play them or write them out rapidly and essentially completely.
The real question about the product is whether the results are reasonably testable and reliable. As noted above, while the authors note this, they make too little of the point.
They confuse how is the product is made (e.g., the dish cooked) with the nature and testability of the results.
Even the neat final results of mathematical and logical proof arise from complex and messy processes. Mathematics proceeds from imaginative conjecture through sketches of possible proof routes, with much backing and filling, to outlines of proofs, to building the proof, finding holes and difficulties, filling them in, and finally writing a complete proof. Later, the proof may be reworked to be as neat and simple as possible. It is often the case that later mathematics can be used to even further simplify the proof. The process can take years, involve the work of many brains, and in the middle seem wretchedly incomplete and illogical. (For examples see The Mathematician’s Mind: The Psychology of Invention in the Mathematical Field, by J. Hadamard, Princeton University Press, 1996 (reprinted).)
The authors’ lament that scientific history and textbooks often present uncertainty as certainty, and, in the process, gloss over the difficulties and messiness of the process of arriving at conclusions. The historical records, even many statements made at the time, often make things seem much clearer than they really are or were. (In the spirit of Claude Rains in the movie Casablanca, I was “Shocked! Shocked! Gambling here in Casablanca?”) This confusion sometimes occurs because what is being said and written at the time is an internal dialogue of individual scientists and conversation among scientists; part of the attempts to convince each other of an idea or a result that is, for one reason or another, believed.
The clean record is also used as a didactic tool, not for the methodology or history of the subject, but for the state the subject has gotten to at a given time. For this purpose the tale is made as clean as possible, so that the final train of reasoning and the place of the data in it are as simple and clear as possible. This works well for advanced students, who frequently then go back and find the historical warts and inconsistencies, but is really a poor tool for the teaching of beginners, who get an over simplified picture of the whole process. The criticism is really of the textbooks and teaching of secondary school science, and of much elementary university science.
This all has the unfortunate result, as the authors make clear, of leaving the impression that science and technology in process, and at the frontier, are far more certain than they are, or are seen to be by scientists and technologists themselves. When the science or technology become important matters in a public, political or policy debate this has unfortunate consequences. Certainty of knowledge is certainly impossible, but reasonably certain knowledge is often achievable, and too strong a statement of certainty can poison or confuse a public debate. On the other hand, the politics of real public debate can make admission of less than complete certainty an invitation to be ignored..
The same is true of their complaint about the ‘experimenter’s regress’; the difficulty of deciding whether an experiment provides data that fails to ‘decide the question’, or perhaps the experiment was just not ‘properly done’. On page 45 of the science golem they note: “Observation and prediction were linked in a circle of mutual confirmation, rather than being independent of each other as we would expect according to the conventional idea of an experimental test.” This is a correct picture of a difficulty: it can be difficult to be decide whether a ‘crucial’ experiment really proved the point, as the data may have a lot of variance, although scientists frequently talk as though every such experiment will have quite clear results.. (This may be another case, like marriage after the fifth divorce, of the triumph of hope over experience.)
However, the process does not really rest on an insistence on immediate certain proof. The sequence is closer to: idea, test, modification of the idea and the concept of the test, etc., continuing until there is a theory that fits a whole collection of tests, etc. The process is not that of two parallel systems of idea and test, but may be better described as a decreasing diameter helical process, in which the axis of the helix is time and events. In successful cases of science, as the ideas and the sequence of experiments evolve, one builds a set of ideas that explain more and more experiments and observations, and fit more and more coherently with a larger body of ideas and information from experiment and observation. The test is convergence to consistency and coherence over a larger body of theory and experiment or observation. If something like this doesn’t happen, the idea is likely to be discarded or put aside, although it may later be picked up again. The “we” in the quote in the paragraph above (the authors) are using an overly simplistic view of the process. Unfortunately, this simplistic view is what many textbooks present, presumably for initial didactic reasons.
Individual scientists often rework what they know continually, in a simplifying way, as a way of imprinting and codifying their knowledge. Further, the community reworks the material to make it as clearly and simply a coherent account of the theory and data as possible. This makes it easy to remember and easy to test against new ideas and data. However, the professional participants generally have clearly in mind whole volumes of uncertainties, exceptions, variations and caveats, which, in their minds, are like footnotes and placeholders of unfinished business. History is interesting, and valuable in learning, but doesn’t really matter much to the reconstituted, later account of the state of the science or technology. While the history of mathematics, controversies, warts and all is illuminating for understanding the development of the subject, the history of a proof neither validates nor invalidates it. There is now a sense in which the Michelson-Morley and oil drop experiments are irrelevant to the body of current relativity and quantum physics. In the author’s terms, the cable has too many strong strands for its strength to depend upon the weakness of some old and difficult experiments.
Unfortunately, the final paragraph of the science golem (page 149) simply sows confusion. After describing the problem of a children’s ‘science education’ class in which the students come up with wildly varying results which the teacher then explains away in favor of the ‘correct’ answer, the authors assert that this is a true picture of science:
“¼.with clean white coats and ‘Ph.D.’ after their names. They all come up with wildly varying results. There are theorists hovering around, like the schoolteacher, to explain and try to reconcile. In the end, however, it is the scientific community (the head teacher?) who brings order to this chaos, transmuting the clumsy antics of the collective Golem Science into a neat and tidy methodological myth. There is nothing wrong with this; the only sin is not knowing that it is always thus.”
This caricature is not too bad a picture of the various pieces of science in the early stages of their making. However, it ignores the roles of institutionalized skepticism and questioning, continual ‘reworking’ to add further observational and experimental results, search for broader consistency of theories with each other and with the broad field of observational and experimental results, and continual testing of the whole against the large body of results. This does not lead to certainty, but to a tested set of ideas and results that are reasonably certain, and that we call scientific knowledge of the world.
The caricature also ignores the requirement that new tested results and conclusions fit with previous sets over the domain where they were well tested: Einstein’s special relativity predictions reduce to Newtonian predictions when velocities are small compared to the velocity of light. In this sense Newton was not overturned, but extended. This test is not universally valid, but is a protection against rampant ad hoc-ery.
The authors correctly point out that science and technology are really only extreme expert extensions of ordinary craft and logic, but there is also the self- and community-discipline aspect of the craft to be understood. Individual and organized skepticism play key roles, even though sometimes, in the heat of battle, they are not given the scope they should have.
Comments on cases and statements in the science golem
The authors discuss a set of cases in each book to illustrate the point of uncertainty and messiness in the process. The cases all depend on secondary sources, but this may not matter as the points are clearly made through this literature.
Chapter 2: Two experiments that ‘proved’ the theory of relativity
p42: “The meaning of an experimental result does not, then, depend only on the care with which it is designed and carried out, it depends upon what people are ready to believe.”
I would add: “at the time and given what they then know.” This may change with later data and ideas, and with the ‘reworking’ of everything which is known about the subject.
p52: “Is this because science needs decisive moments to maintain its heroic image?”
No, because it later became clear that they were, as far as we now know, correct, and by being correct set science on a new and firmer path.
Chapter 5: A new window on the universe: the non-detection of gravitational radiation
p107: “The question is, will they gain the attention of the scientific community?”
No, the question is whether he or someone else will produce better explanations of the problems, or a better experiment with more data and less variance. Lots of scientists are still working on the problem, and there are new theoretical and experimental ideas. This may all add up to something, or may fizzle away, depending upon whether we are clever enough to find a way to untangle what it is the universe is doing.
Chapter 6: The sex life of the whiptail lizard
p119: “As always the facts of nature are settled within the field of human argument.”
No, the current human view of the facts of nature is settled within the field of human argument; nature remains what it was, we’re trying to find out what those facts are. The point is epistemological. I take it as a base hypothesis that there is a natural world outside of myself, and we’re trying to find out about it. I’m not sure whether the authors really agree with this hypothesis or not. We may build up a reasonably testable theory and set of experiments and observations, or we may miss the mark. The question remains open. Its like trying to go to a place whose existence and location are uncertain; we create the map as we go, and know we are there by seeing the place. This point may seem trivial, but it is an important difference in view. It is why all views of nature are not equivalent: some are backed by a broad consistency of ideas and data, some views cover only a narrow field of known observations and experiments.
Chapter 7: Set the controls for the heart of the sun: the strange story of the missing solar neutrinos
p134: “Feynman apparently advised the young Bahcall that he had done nothing wrong and that if there was a contradiction this made the result more rather than less important.”
Of course! Real progress is likely to be made precisely when experimental results appear to strongly challenge theoretical ideas. Weakly confirmatory results are likely to be much less helpful.
p138: “Science works the way it does, not because of any absolute constraint from Nature, but because we make our science the way we do.”
No, Science works the way it does because group effort and applied skepticism allows us to build up a picture of the collection of constraints that nature forces on us, bit by bit as we do our work. This is sometimes dramatic, but frequently appears as a cumulative constraint. No single result, no single constraint is absolute; there is always likely to be variance and error, but the cumulative effect can be quite convincing.
The cumulative result, or even a single apparent contradiction, may lead to a sudden shift in view, as a result of which much confusion may become sense as variation and complexity which were not previously understood collapse in a new picture of the world. The simplification of Ptolemaic astronomy into Copernican, Keplerian and then Newtonian astronomy is a classic example. It took a long time and a lot of data and argument, and even made a martyr or two along the way.
Conclusion: putting the Golem to work
p143: “To change the public understanding of the political role of science and technology is the most important purpose of our book and that is why most of our chapters have revealed the inner workings of science.”
If that is the purpose, it is well hidden, since it doesn’t seem to be explicitly discussed in either book. Besides, only part of the scientific process is discussed: the individual developing cases as they were at the time of initial development. Not nearly enough is said about the ‘reworking’ process, in which larger codification are attempted, theories are examined for consistency with each other, and new experiments are devised to test the new codification.
comments on cases and statements in the technology golem
Chapter 1: The role of Patriot in the Gulf War (p18)
The point that perceived performance of a weapon system will depend upon the criteria used is well taken. I know of a case in which military aircraft did not actually have to carry and fire defensive weapons on every mission. Since they sometimes had them and they were sometimes successful, it was usually only necessary to actuate the fire control system, thus sending a signal that they were ready to fire, in which case the attacking systems generally shut down to avoid being hit. Plausible bluffs sometimes work! This however is different than knowing what the technical performance of the weapon is against various targets.
It seems clear that the authors’ implied conclusion (p29) that systems built upon previous experience are always useless in the face of new experience is wrong: many systems (even weapons systems) can be used successfully in circumstances different from those for which they were designed. This is part of the problem for the designers: how can a system be flexible in the face of circumstances which are likely to be different from those I envision on the basis of past experience? Some systems meet this challenge fairly well, some don’t.
Chapter 2: The naked launch: assigning blame for the Challenger explosion
p30: “¼proclaimed the death of seven astronauts and the end of the space programme’s ‘can do’ infallibility.”
Given the death of three astronauts during the Apollo testing, and the near disaster of one attempted flight to the moon, and as a former Administrator of NASA, I know that NASA’s view of itself, whatever others may have thought, was of a ‘can do’ but hardly infallible organization. I would characterize the view as being: ‘we do radical things in a very conservative and careful way. That is, the attempt is always to do something very difficult, but the work is extremely careful and risk averse.
p33: “With twenty-twenty hindsight we can see they were wrong, but on the night in question the decision they reached was reasonable in the light of the available technical expertise.”
However, as Diane Vaughan points out in The Challenger Launch Decision, University of Chicago Press, 1996, page 383, they did not plot all their data, only the data from the failure cases, thus neglecting an important general practice. The interesting question is how this could have been done without someone noticing. This is the sort of error that outside peer review is good at catching, but immediate pre-launch conditions do not lend themselves well to such practices.
p47: “After this instance of blow-by, Roger Boisjoly in particular felt there might be a link between low temperature and the damage.”
Interestingly enough, in the year before the very first flight of any shuttle, towards the end of the development of the system, based on technical recommendations, a formal temperature restriction was put on flights by the Administrator of NASA at the time (me). Had that restriction been continued, the accident would probably not have happened. Why and how the restriction was either lifted or lost I do not know. (I also signed an instruction that there were to be no ‘tourists’, which would have included school teachers; only people with specific mission work to do were to fly.)
I have always felt that there might be some connection between the circumstances leading to the accident, and the fact that during the period leading up to it there was a period of change at the top. There was no Administrator at the head of the agency from December 4, 1985 until May 12, 1986. The Deputy Administrator in charge had been Deputy Administrator for only two months, and came to the position on November 25, 1985 with no previous NASA experience. The Challenger Launch accident occurred on January 28, 1986. It is at least possible that some of the continuity of practice and technical discipline of the agency was weakened during this period, although I have no evidence to that effect.
It is not clear to me why the ‘joint problem’ did not rise higher in the management chain, perhaps to the top of the agency. Such reporting upwards, and seeking for technical and managerial help, would be within the style of NASA as I knew it. The investigators of the accident do not appear to have paid any attention to this question. In any case, as the authors point out, the whole system is very complex, and its operation risky. It only works because of extreme care and attention.
Chapter 6: The Science of the Lambs
p116: “What went wrong?”
p124: “If scientists had treated the farmers as a group with relevant expertise in some areas, then both group might have learned to value the others’ contributions and seen the limitations of their own claims.”
The problem started in Whitehall. The government didn’t involve all of their scientists, but apparently only nuclear and radiation scientists.. I do not know the British system very well, but in the US it is at least possible that, once it was clear that there was a local issue of sheep farming, the Department of Agriculture and the ‘county agents’ for the area would have been involved. The ‘county agents’ are local, know local soils and agriculture conditions, and also know, and are known by, the local farmers. A more complete set of scientists might have done a better job. I would not necessarily have expected the nuclear scientists to know about this; the bureaucracy should have. However, the behavior of either scientists/technologists or bureaucracies in such situations does not necessarily reflect a broad view of science and technology, although it should.
Conclusion: the golem goes to work
p153: “Golem science and technology is a body of expertise, and expertise must be respected. But we should not give unconditional respect before we understand what the expertise comprises and whether it is relevant. To give unconditional respect is to make science and technology a fetish.”
Agreed, but then one should understand not only the sources of expertise, but also something about the principles, and the overall practices, and the reasons for belief. Otherwise all ‘experts’ look alike. The task of the scientists and technologists is to lay out carefully what it is that is being said, and why; what statements seem to them be scientific, which are well founded, which still debatable, and which statements seem to them to belong to other realms. They do not always do so, and do not always ‘stick to their knitting’, but then, it is common for everyone to express opinions about other’s areas of expertise. I would hope scientists and technologists would be disciplined about this.
Clearly non-scientists should be able to choose among possible alternatives, but they should do so knowing the consequences of the alternative choices as well as anyone can, given the current state of knowledge. That includes the scientist/technologist saying; ‘we don’t know, or we only know with such and such uncertainty, but based upon our expertise and experience we think thus and so.’ A purpose of science and technology is to say clearly when appropriate: ‘we don’t know¼’, or, ‘we don’t know how to¼’
Conclusion
I am in the peculiar position of agreeing with the message of the golem books, often disagreeing with how it is said, and thinking that it is dangerously incomplete. I found that reading the books, agreeing and disagreeing with them pushed me into a useful thinking and rethinking exercise, since both agreements and disagreements broadened my view of subjects I thought I knew well. I invite the authors to continue into the broader subject, rethink and try again. I would like to read the results.
Robert A. Frosch is a theoretical physicist by education. He did research in ocean acoustics and was among other things Assistant Secretary of the Navy for Research and Development, Assistant Executive Director of the United Nations Environment Programme (UNEP), Administrator of NASA (1976-1980), and Vice President of General Motors Corporation (GM) in charge of Research Laboratories. He is now at the Kennedy School of Government of Harvard University. Address: Robert_Frosh@Harvard.edu.
In recent years there has been a phenomenal interest in the growth of what some are calling ‘digital’, ‘online’, or ‘Virtual’ universities. Indeed, today, it is commonplace to read that information and communication technologies are radically reconfiguring the landscape of higher education, changing the very ‘nature’ of the university (cf. Cunningham et al 1998). You will already know the vision: a decrease in importance of the campus, as students ‘login’ from a distance to access ‘courseware’, new media technologies replacing traditional lectures, courses being delivered and assessed over the Internet, promising to make higher education available anywhere and at anytime.1
Depicted as a solution to the increasingly demanding problems of higher education, all of this has fired the imagination of academics, policy makers, university managers, and educational specialists alike, the assumption often appears to be that institutions can move straightforwardly toward this vision (cf. Newby 1999). Yet, our recent research – conducted under the UK’s Economic and Social Research Council’s Virtual Society? Programme – suggests that the universities which we have studied have found the introduction of new technologies, alongside their more traditional methods of providing teaching and learning, extremely difficult and that the actual model of Virtual University which we have seen emerging bears little relationship to the vision. What we have found, in short, is that the Virtual University works in theory but not in practice.
What are the difficulties of implementing the vision? Basically, the bottom-up, course-by-course approach to constructing a virtual university is slow, labour intensive and prone to failure. For example, we have followed the progress of a number of initiatives in one particular UK institution for over a year: there was a humanities degree course, where video-conferencing technologies were used to connect undergraduates based in the UK with students located in other parts of Europe, the idea being that they could present work and ideas to each other, and receive feedback, much like a traditional seminar; then there was the new ‘Cyber Culture’ course, available for credit as a self-study module over the Internet; and, finally, there was the ‘Information Skills’ module taught by library staff, which had also been translated into an online self-study module. Week after week we sat in on technical sessions and planning meetings as the academic material was gathered, the technology was developed, and the actual form of these initiatives began to take shape. Staff who were keen to be involved in the rolling-out of the projects were contacted, and possible groups of students willing to be part of the experiment were identified.
Yet, just a couple of months after everything had seemingly been put in place, each of the projects has – for want of a better word – ‘stalled’. The immediate reasons for this are varied: one of the partners pulled out of the video-conferencing project complaining of high telecommunication costs; only one distance student had enrolled for the Cyber Culture course; and library staff could not be convinced that the online version of the Information Skills course was sufficiently improved to warrant its introduction in place of their existing methods. What is common to each of these stories, however, is the failure to enrol (or to keep enrolled) all of those aspects of the university necessary to make the projects work (academic staff, students, computer services departments, libraries, validation committees, partner institutions, etc.,). Further, in the site that we studied, there were aspects of the University that were crucial for the success of the projects and that did not exist and therefore had to be built – for instance, the University lacked procedures for validating online courses – slowing the whole process. In short, initiatives were confounded by difficulties in co-ordinating a wide range of actors across a large organisation made up of diverse and disparate entities (i.e., departments and service units). It is, it seems, the very institution of the university which is at the heart of the problem.
This begs the question, if not this vision then what? If we want to understand the ways in which information and communication technologies are affecting higher education institutions in the UK, perhaps we need to look at this issue from a different point of view. And if the central problems of building the virtual university seem to relate to the university as an institution, then it is to this issue that we should perhaps redirect our attention.
The way in which universities are organised and operate is currently in a period of change. The introduction of new forms of Management Information Systems (MIS) and Enterprise Resource Planning (ERP) systems are crucial in all of this. Traditionally, administrative computer systems have been kept within the domain of the centralised administration and have had little influence on the primary functions of universities and their chalkface workforce. Most academic staff rarely come into direct contact with these systems, and students are hardly aware of their existence. The most obvious example here is the ill fated UK University Funding Council ‘MAC’ initiative which, through attempting to standardise the ways in which British Universities conducted their management and administration (cf. Goddard & Gayward 1994), seemed to reinforce divisions between departments and faculties and between those who managed and administered and those who engaged in academic work. However, the new generation of MIS and ERP systems, through a process of standardising certain roles and relationships, are being consciously used to ‘connect’ and ‘integrate’ processes that have traditionally been kept apart. In the words of one senior academic manager responsible for such a system, its function is ‘to bind the university together’ (cf. Pollock, forthcoming).
Let us focus on one example, that of students, and particularly how they are to be managed and administered. Until recently, the students relationship to computer systems has been mediated by academic and support staff. However, in one of the institutions that we conducted research, students are moving from being passive objects of administration to becoming the main groups of active users. From the beginning of the next academic year, online registration for new and returning students will be introduced. Students are to be provided with ‘smart cards’ that allow them to self-register for certain aspects of their degree course (i.e., to select options, and to enter personal information), and then, later, to validate the accuracy of other information (e.g., their academic results, their financial status, and so on). In a later stage of the implementation, students will also be able to pay bills on-line, book accommodation, schedule meeting with their tutors, and so on – all without coming onto the campus.
At one level, this is a rather mundane use for new information and communication technologies – hardly worthy of the notion of a revolution within the university. Yet, at another level, the practices, procedures, and processes, being laid down with the introduction of these new forms of management and administration systems have important consequences for the way in which the University is being reconfigured, and particularly the way in which certain key relationships are being developed. Arguably, we might suggest, it is these systems constitute a true Virtual University in the sense that the complex databases and sets of procedures that constitute the core of these systems are a model or simulation of the university: they have the form of a University without the thing itself. Instead of an organisation full of heterogeneous actors, with complex identities, everyone and everything is formalised, represented in a standardised form, with certain explicit roles and responsibilities towards the system. Students, for instance, are simply ‘users’ with particular duties towards the inputting and maintenance of information.
It is well known that information and communication technologies provide a powerful incentive to standardisation. It is for this reason that in our project –’Space, Place and the Virtual University’ – we have focused on the development and implementation of these systems, as well as more conventional technology-supported distance education. Indeed, as we see it, future developments in online learning and distance education technologies will be underpinned by the model of the university set down by these management and administration systems. It is these management or administrative computer systems which may be able, by clarifying and making explicit the various roles and responsibilities within the university, to facilitate the complex process of enrolling the various actors required to implement online learning and distance education technologies.
While we think that this standardisation is, in itself, not necessarily a bad thing, there are some possible dangers. To paraphrase the words of the sociologist, Phil Agre (1999), the problems begin when we standardise the ‘wrong thing’. In this respect, when replacing the administrative apparatus, we risk destroying or submerging those interactions that are tacit, informal, flexible – the very processes that might, for instance, offer important forms of support to students. Just how might a student ask for an extension on the payment of a late rent bill? Is this simply a matter of ‘functionality’ just waiting to be built into the system? Might cases that do not quite fit be supported – since we are now talking about computer systems and their users – by the increasingly ubiquitous ‘help desk’. Or will these ‘non-standard’ requests simply fall between the inevitable cracks in the system?
It is, of course, very easy to get it wrong. There are no adequate models to account for how an organisation as complex as a university is able to function. Here, we are reminded of the story told by the anthropologist Marilyn Strathern (1997) of how a group of consultants, after an academic audit of Cambridge University, complained of the ‘protean nature’ of the University, and of how its academic structure and system of governance could not be captured by their ‘managerial language’ or ‘organisational models’. The University of Cambridge, it appeared, worked in practice but not in theory. We should then be wary of the models (or visions) that we adopt and pay close attention not only to what is ‘standardised into’ the university but also that which is ‘standardised out’ of it. The key issue, then, is whether these centralised computer systems – the MIS and ERP – which are reshaping the institution of the university will work as well in practice as they do in theory.
Note
See the articles by Noble (1998), Winner (1998) and Newman & Johnson (1999) for critiques of this vision.
References
Agre, P (1999), ‘Infrastructure and institutional change in the Networked University’, paper prepared for the conference on New Media and Higher Education, University of Southern California, October 23-27 (available at www.rcf.urc.edu/~ics/ed/agre.html
Cunningham, S., Tapsall, S., Ryan, Y., Stedman, L., et al. (1998), New media and Borderless Education: A review of the Convergence between Global Media Networks and Higher Education Provision, Australian Government, Department of Employment, Education, training and Youth Affairs, Evaluations and Investigations Programme, Higher Education Division (http://www.deetya.gov.au/highered/eippubs/eip97-22/eip9722.pdf).
Goddard, A, D. and Gayward, P, H. (1994) ‘MAC and the Oracle Family: Achievements and Lessons Learnt’, Axix 1 (1): 45-50.
Newby, H. (1999), ‘Higher Education in the 21st Century: Some Possible Futures’, Discussion Paper, CVCP, London, March.
Newman, R. and Johnson, F. (1999), ‘Sites of power and knowledge? towards a critique of the virtual university,’ British Journal of Sociology of Education, 20(1): 79-88.
Noble, D. F. (1998), ‘Digital Diploma Mills: the automation of higher education,’ Science as Culture, 7(3): 355-368.
Pollock, N. (forthcoming), ‘The virtual university as ‘timely and accurate information’’, Information Communication & Society.
Strathern, M. (1997), ‘”Improving Ratings”: Audit in the British University System’, European Review, 5, 3: 305-321.
Winner, L (1998), ‘Automatic Professor Machine’ (available from: www.rpi.edu/~winner/apm1.html, downloaded June 6, 1999)
More information on the ‘Space, Place and the Virtual University’ project can be found at www.brunel.ac.uk/research/virtsoc.
The authors can be reached at tel +44 (0)191 2225876.
A workshop, “Local Knowledge and its Global Consequences: New Perspectives on Environment and Development,” took place at the Kennedy School of Government at Harvard University on April 7 and 8. Sheila Jasanoff, Marybeth Long-Martello, Clark Miller and Charis Thompson were the organizers. Support for the workshop came from a NSF grant Global Environmental Assessment Project based at the Kennedy School.
Prior to this workshop, a weekly seminar on local knowledge took place at Harvard during autumn 1999. Participants in the seminar explored the meanings, uses and implications of “local knowledge” in many settings. They discussed lay-expert clashes regarding agriculture and deforestation in Thailand (Tim Forsyth); the Convention to Combat Desertification (Elisabeth Corell, MIT); computer modeling of agriculture in the Great Plains (David Cash); the AIDS blood crisis in France and the United States (Jessie Saul), international forestry programs (Cathy Fogel), the Human Genome Diversity Project (Jenny Reardon); the Biosafety Protocol under the Convention on Biological Diversity (Aarti Gupta); and “quality of life” estimates under the World Health Organization (Dele Ogunseitan). Whereas local knowledge is often assumed to be place-based, these seminars elucidated ways in which local knowledge can be tied to particular communities, histories, institutional settings and expert cultures, as well as to geographic locations.
Building on the seminar series, the workshop on April 7 and 8 examined “local knowledge” as it relates to various environment and development contexts. The conference opened with case studies on the creation and commodification of local knowledge through parataxonomy training programs in Costa Rica (David Tackacs) and bioprospecting arrangements in Mexico and the United States (Cori Hayden). A second set of case studies examined local responses to the Orissa Cyclone disaster in India (Shiv Visvanathan) and the Thailand case (Forsyth). Constructions of locality and identity connected papers on elephant hunters and gun design in nineteenth century South Africa (William Storey); the role of missing expertise in the Bhopal gas disaster as a way to understand risk societies (Ravi Rajan); environmental discourses in the European Union (Brian Wynne); and interactions between local identity and pollution numbers in the United States (Alastair Iles).
On the second day, presentations on local knowledge and global politics explored various approaches to understanding and conserving biodiversity. These papers discussed the pastoral practices of the Masaai in conserving Kenyan savanna (David Western); how the new field of genomics localizes and universalizes DNA (Stephen Hilgartner); the Biosafety Protocol (Gupta); and constructions of local knowledge and indigenous identity under the Convention on Biological Diversity (Marybeth Long-Martello). Three papers on “disciplining the local” addressed monitoring practices for climate change and acid rain (Stephen Zehr); the roles of standards in North-South technology transfer (Judith Reppy); and the development by the World Wildlife Fund of sustainability indicators which neglect national boundaries (Clark Miller).
As the workshop presentations and discussions illustrated, local knowledge inhabits environment and development discourses, appears as an organizing principle in policy processes, features in the construction of identity for certain groups, provides a basis for asserting voices in political debates, and is part of material culture such as markets. In their diverse empirical material and theoretical approaches, the papers collectively suggest new conceptions of local (and global) knowledge, highlight previously overlooked sites of knowledge creation, draw attention to knowledge as a commodity, and investigate new interactions between local knowledge and local, national and transnational systems of governance. Notions of local knowledge are embedded in much theory and practice in environment and development arenas. Yet the varied meanings and implications of this concept remain largely under-theorized. This situation provides an opportunity for S&TS scholars to engage with new and diverse audiences, and to investigate issues outside the realms of most S&TS research.
The organizers plan to pursue an edited compilation of workshop papers and thus to continue exploring new forms of critical thought about local knowledge. The book’s organizing themes will likely address the role of technology in constructing, embedding and mobilizing local knowledge. It will focus on the means by which people in environment and development fields define and translocate the “local” and “global.” In addition, the book will examine international institutions as previously unrecognized sites for local knowledge creation.
author’s address: marybeth_long@harvard.edu