Science Policy and Science Publishing

John Ziman

  • Scientists in Whitehall by Philip Gummett
    Manchester, 245 pp, £14.50, July 1980, ISBN 0 7190 0791 7
  • Development of Science Publishing in Europe edited by A.J. Meadows
    Elsevier, 269 pp, $48.75, October 1980, ISBN 0 444 41915 2

‘Reallocation of Responsibilities of Research Councils: Royal Society opposes Reform’ was probably the runner-up to ‘Small Earthquake in Peru: Not Many Dead’ in the famous competition for the least sensational newspaper headline. Nevertheless, here we are, a generation into the nuclear era. The scientific bureaucrat can no longer be laughed off as a cross between Professor Branestawm and Dr Strangelove. Exactly how scientific expertise should be employed in the affairs of the nation is a central question of modern politics.

The government research laboratory is an intermediate social habitat, not quite self-sufficient in its traditions or its mission, lying across the boundary between the very different domains of academia and the Civil Service. The sociologists of science have given their attention to its academic characteristics: in what way is it an institution where research scientists can seek personal recognition for their contributions to knowledge? There is a substantial literature on the internal management of research organisations: in what way does one get such individualists to work together on appropriate problems? There is, of course, a babble of incoherent argument about science policy: on what problems, at what expense, should government research and development be concentrated?

But very little has been written, from a scholarly point of view, about the interface between the scientific and administrative realms, and about the people whose job it is to make the connections across this interface. Old hands like Solly Zuckerman tell us how it was for them, and perceptive novelists such as Nigel Balchin, William Cooper and C.P. Snow expose much that is true in the guise of fiction. Drawing upon a variety of sources – reports of Royal Commissions, obiter dicta of scientific notables, newspaper editorials etc – Dr Gummett gives us the first coherent, objective, yet sympathetic analysis of the current situation and its historical background.

Most important, our scientists are still, as the famous phrase goes, ‘on tap’: they have never succeeded in getting ‘on top’. Whether academically employed in very basic research, or organised directly for technological development, they now depend almost entirely on funding by the state or by large industrial corporations. There is not really a government science system, or a science budget, because these funds flow along a variety of channels, but their eventual use is more closely controlled by non-scientific authorities than ever.

Long ago, when all research was done on a shoestring, anyway, scientists who were directly employed by the government, along with industrial research workers, had a very inferior status in the scientific community. Authority was firmly in the hands of the ‘academics’, who worked in universities or in the quasi-academic laboratories of Quangos such as the Medical and Agricultural Research Councils. Indeed, scientific notables such as the officers of the Royal Society were treated with great deference – on a par with the archbishops of the Established Church – in Whitehall.

After the Second World War, the money came pouring in, to buy far more science for society. For about twenty years the system expanded, largely under the leadership of its traditional élite. Although of course a great deal of this money was spent on the harshly technological development of such intractable devices as nuclear reactors and aero-engines, the ideology of basic research was still dominant: knowledge was to be valued as much for its own sake as for any benefits it might bring. This ideology was even fostered in strictly mission-oriented institutions such as the Atomic Energy Research Establishment at Harwell, and could thus claim lip-service throughout all government science. The peasantry of applied science could in this way hope to enrich itself, so to speak, and to unite with the aristocracy of pure science in an autonomous estate of the nation.

This hope has not been fulfilled. The traditional snobbery of academia towards government and industry has been much muted, but mainly because academia has itself been cut down to size. Dr Gummett gives a full account of the two major battles where the scientific community learnt the limits of its aspirations. The first was over British participation in the construction of a new 300 GeV particle accelerator at the European Centre for Nuclear Research at Geneva – at a cost of £20 million, or so. A conflict between ever-expanding research costs and the finite resources of the Exchequer was inevitable: but the government decision against this project in 1968 took a surprisingly painful toll in disillusioned nuclear physicists. This long-drawn-out conflict not only pushed the science policy machine into a permanently defensive posture against Treasury economisers: it also demolished the notion of a united front for science, as chemists split with physicists on the fundamental issue of the benefits to be derived from the project.

The second significant event was the debate over ‘Rothschild’ – that is, the report in 1971 by Lord Rothschild, head of the government ‘Think Tank’, on the organisation and management of government R & D. His crude demand that research results should be supplied, on ‘contract’, for the ‘customers’ that were buying them, was a direct challenge to the academic ideology, and was bitterly opposed in a rearguard action by the scientific élite. As Dr Gummett correctly reports, the resulting modifications in the machinery of the research councils, in the administration of research within various government departments, and in notional cash flows between these diverse organs of the body politic, have made very little difference to the detail of who does what research on which problems. Nor does it seem to have had much effect on the morale of research workers at the laboratory bench, whose main cause for concern in the past ten years has been serious deterioration in their terms of employment and pay. But the scientific notables now understand that they can no longer justify large expenditure on basic research with the vague promise of beneficial ‘spin-off’. However eloquently they may continue to preach the simple gospel of the untrammelled search for knowledge, they know in their own hearts that this is not a faith on which they can rely for their daily bread. This disillusionment was inevitable: yet it may have a more significant effect on the social role of science than direct administrative and financial constraints.

The human qualities of the leaders of science is much more important than one might think. The scientific community is not very hierarchical in its internal affairs, but has very narrow gateways into other communities. Such gateways, especially those leading into the government machine, are kept by members of the scientific élite, in their official roles as advisers, chief scientists, council chairmen etc. One of the most instructive chapters of this book discusses the curricula vitae of such notables and the qualities they must bring to their tasks.

The trouble is, of course, that these ‘political’ qualities are not closely correlated with research capability. It is true that, just as a good general must have proved his personal courage in action, so must administrative authority in science be made credible by evident research competence. On the other hand, it is quite hopeless to allot political and social responsibility to scientists in strict proportion to their scientific eminence. I accept Dr Gummett’s implication that fellowship of the Royal Society comes too late, is too specialised, and is a little too chancy as a standard qualification for membership of the ‘advisory élite’, although at least it provides a fair number of suitable names from which to choose. What is worrying is evidence of the conventional prejudice of the British governing establishment in favour of its own type: Cambridge still gets the pick of science students at A-level, but they do not all live up to their promise in later life and it is surprising that they form such a high proportion of the scientific advisers. We should be seriously concerned at the career gap between the average scientific worker in the average research laboratory and those whose climb to power has been along narrower, more aristocratic paths. For all that, we are remarkably well-served by our scientific top brass – certainly by comparison with many other countries. They seem quick to pick up the techniques of court politics (for they never have to offer themselves in public for votes), and learn to perform adeptly in the committee rooms and corridors.

Dr Gummett emphasises the intrinsic compartmentalisation of British government, which continually defeats every move towards a unified apparatus for science policy at the centre. But research itself is a very uncertain business, where an improvised response may prove far more effective than a far-sighted plan. The incorrigibly unwieldy machinery of research councils, departmental chief scientists, university grants committee and so on, works better than one might expect because the people who run it know, respect and trust each other fairly well, and can accommodate flexibly to unexpected stresses.

The real trouble, it seems to me, is that most working scientists are very isolated from their contemporaries, even within the same government department. The most obvious sign of this is the barrier against scientific specialists moving into general administrative posts in the Civil Service. Dr Gummett recounts the sad tale of unsuccessful efforts to break down this barrier: but acknowledges that it is built upon a massive bedrock of educational practice and cultural tradition. My own modest proposal for reform would be to make every civil servant show that he could do a proper job as a narrow specialist and then pick out those who showed some capability for taking general administrative responsibility – but that goes completely against the grain of our amateurish gentlemanly style, which was excellent for ruling India but is quite the wrong way to run a microelectronics industry.

As a consequence, scientific ways of thought do not diffuse rapidly into the operations of government. Rothschild argued for research to be undertaken on some relevant problems but made no provision for communicating the outcome of this research to those who could best apply it. This is not just a matter of making the executives and administrators sit at the feet of the scientists: scientific civil servants are almost as isolated from operational experience as the ivory tower academics, so they often have little conception of what is really worth investigating or inventing. Formal liaison procedures cannot make up for the lack of personal mobility across the boundaries between specialised and generalised domains of responsibility.

Another characteristic structural defect, which Dr Gummett sharply identifies on a number of occasions, is the inadequate staffing of policy organs such as the research councils, Parliamentary Select Committees and Royal Commissions. The scientific and political notables come to these committees as authorities in their own right, powerful in personal influence and opinion, but with over-full office diaries and almost no time to read widely, and think the issues through in depth. There is a striking disparity between the meticulous attention which they would give to a scientific argument in their own research, and their sloppy treatment of economic, educational, managerial and political issues, where almost any outworn doctrine or prejudice is allowed its head. They do not seem to feel the need for carefully-researched background papers on the deeper facts, the cognate cases in other countries, or the well-founded theoretical implications of the issues they are to decide. I share his scepticism about making a science of science policy, but there is no excuse for the inadequate support for science policy study both inside and outside the government.

I come back to a key question: how can the scientists function best as a community within society? The managerial structures of the research councils and government laboratories give little scope for influences to flow up from the grass roots. Trade unions are too preoccupied with labour relations to contribute significantly to the formation of research policies at each level. Scientists define themselves primarily in terms of their disciplines or technical specialities, and discover their most natural affiliations in the corresponding learned societies and professional institutions. In many fields, such as engineering and medicine, these organisations are strong, and fairly represent the views of their members. But the Royal Society is no longer broad enough in its membership and disciplinary coverage to perform this role for the scientific community as a whole. As Dr Gummett puts it, ‘in the absence of any one organisation concerned to protect the interest of all scientists within the public sector, questions about the overall working of science-government relations tend not to be raised in any coherent way. The initiative rests with the governmental agencies and committees, and there are no strong countervailing forces.’ Like so much in this admirable book, he says it succinctly – and I couldn’t agree more.

The British research system is not really quite so ramshackle as we are led to believe by its critics. But as Dr Gummett himself emphasises, it is not a rational organisation of the sort which most people think that science – that ideally rational activity – ought to have. It is as well to bear in mind the vanity, jealousy and exploitation, as well as overloaded conscientiousness, it involves: much more than a tidy-minded bureaucrat would think he had built into his model.

What we are now beginning to understand is that science policy is just policy: every government apparatus in every country makes it – or fails to make it – in its own way, just as it raises taxes or makes war in its own peculiar fashion. I learnt this lesson myself in Madrid a few years ago, when, more or less by accident, I heard the principles of science policy being expounded to a large gathering of Spanish scientists. A distinguished French civil servant drew charts of an interlocking structure of bureaux and conseils. A Harvard professor explained that the only way to do it was to exploit the creative rivalry between autonomous Federal agencies. Sir Brian Flowers (now Lord Flowers – one of the biggest wheels in our system) insisted that whatever official apparatus you set up, the decisions would really be made by chaps who happened to have been up at Oxbridge together and who happened to bump into each other in the cloak-room of the Athenaeum. In each country the research system functions according to the local administrative tradition, and presumably produces the science that the country deserves.

To the conventional administrative mentality, science is both peculiar and ordinary. It tends to be treated as ordinary, which is probably wrong: but nobody, anywhere, has yet discovered how to match up its peculiarities to the standard government machine. Scientific expertise is not well-employed within Whitehall, because the traditional role of the specialist in the British Civil Service is that of a non-executive subordinate. There is no general policy for science in the United Kingdom, because the British government has inadequate facilities for making policies that transcend departmental boundaries. But the relations between science and government are no worse than in most advanced countries because they are competently managed by sober, sensible, industrious chaps who went to school together, deal honestly enough with each other and have a modest capacity for accepting a compromise which can be represented as being for the public good. What more should we hope for, or expect?

Ever since the invention of printing it has been worth the labours of many dedicated lives to communicate what has just been discovered, or wildly conjectured, to others not yet in the know. Science publishing is a very tangible enterprise, manifesting itself along miles of library shelves, in ‘computerised databases’ ‘accessed in real time’ through ‘intelligent’ terminals, and in the balance sheets of several great commercial corporations. It is also a major social institution, whose theme is already apparent in the very first issues of the Philosophical Transactions of the Royal Society of London, put together by Henry Oldenburg in 1665, out of miscellaneous items of news that had come to him, as Secretary of the Society, from his wide circle of correspondents at home and abroad.

Simply as the latest in magic, science has always had its public. There have been scientific books since, let’s say, Aristotle. But the scientific journal was something new. Who were Oldenburg’s readers, and what were they after? It may be that one should think of them as akin to present-day readers of New Scientist: practical people, with a bent for observation, or full of insatiable curiosity, trying to keep up to date with the latest surprising discoveries. In every century, there has been this same audience, eager for enlightenment or astonishment at what research might reveal. It is right to see modern science journalism, for all its fads and follies, as a major sector of science publishing, stemming from the same historical roots as its more proper academic cousins.

The Phil. Trans. was soon given over entirely to more formal scientific communications, of less doubtful validity. This seems to have been a continuing process over the centuries. As the opaque core of hard science has expanded, it has engulfed these outer layers of popularisation and incorporated them into its own substance. This happened with Nature, for example, which began life more than a hundred years ago as a popular-science magazine and is now far beyond the best-educated layman. The same trend is observable in Scientific American, whose articles have grown distinctly longer and more technical in the past few years – and which is now being superseded in the popular market by several new publications. Mechanics Magazine was selling 16,000 copies a week in 1824 and English Mechanic claimed sales of 30,000 in the 1870s: what is the secret formula that can bring commercial success to such ventures and then be forgotten until it is rediscovered in the next generation?

There is nothing new, either, in the semi-popular book about science, whether composed with mid-Victorian reverence for the works of the Creator, as revealed through the study of nature, or with contemporary distrust of the ‘works’ of Capitalism, as revealed in every ecological disaster. The full-time ‘science writer’ is in a new profession which ought to be more highly esteemed, for it is a considerable art to make high science and advanced technology both intelligible and interesting. John Tyndall and T.H. Huxley were distinguished scientists who had this gift – which was not shared by all their contemporaries, whose books often drifted off into dullness or idiosyncratic technicalities. There is no convincing evidence that our own scientific notables would do so much better even if they were less burdened with administration and other irreducible responsibilities.

It is just such recognition of historical uniformity that makes this book so extraordinarily interesting. One might be chary of a work by a dozen different authors, commissioned by the Elsevier Publishing Company to celebrate two anniversaries: the quadri-centenary of the foundation by Lodewijk Elsevier of the great 17th-century academic publishing house and the centenary of its ‘rebirth’ in its present form. But Professor Meadows is both a scholarly historian of science and a moving spirit in contemporary studies of the communication system of science. The various chapters are not only authoritative and well-informed: they are drawn together by a common understanding that science publishing is not just a technical or commercial enterprise like making and selling beer bottles, but is cognitively, psychologically, sociologically and culturally inseparable from the research activity that it serves. Actualities focus our attention on the challenges of the 1980s as if they had no antecedents. In the historical dimension we can see the communication system of science in much deeper perspective.

For example, a scientist who today cries out against the ‘excessive proliferation of the literature’ is only echoing the 1789 reviewer who exclaimed that ‘one should seek to limit the number [of journals] rather than to increase them, since there can also be too many periodicals.’ They are both swimming against the tide of expanding knowledge, increasing technical specialisation, and the consequent subdivision of traditional disciplines. Then, as now, individual scientists, in close partnership with commercial publishers, have been more enterprising than the established learned societies in setting up appropriate media of communication for new ‘invisible colleges’. Then, as now, editors under pressure to publish material of doubtful scientific quality have had to erect a screen of anonymous editorial advisers and ‘referees’ to sieve the wheat from the chaff. Then, as now, the financing of scientific journals has been slightly uncertain, delicately balanced between the very different needs and resources of individual scientists, institutional libraries and learned societies. Then, as now, the system of scientific communication has been divided between commercial publishers and so-called ‘not-for-profit’ organisations, such as university presses and learned societies. It is as incoherent and fragmented, as diverse in scope and scale, as the whole book trade, of which it must seem to many to be merely a specialised segment.

But there is still a distinctive uniformity of style and format. Apart from typographic and bibliographical conventions, a modern scientific paper says much the same sort of thing, in much the same language, whether it appears, belatedly and obscurely, in the Quarterly Proceedings of the National Academy of Sciences of Ruritania or, rapidly and visibly, on pages 3943-7 of Volume 243 (Jan.-March 1979) of the International Journal of Semiotechnography, published (it could be) by the great Elsevier Publishing Company, the world’s leading publisher of scientific journals. The essential character of every modern scientific communication can already be detected in those first Philosophical Transactions: it is a message concerning an observable event, or a connected chain of reasoning claiming no more than it can substantiate within a limited context.

In three centuries, particular features of scientific papers have been emphasised and refined, but without fundamental alteration in their indicative, informative, quasi-objective style. The language has become more formal and technical, adopting peculiarly impersonal conventions at the expense of comprehensibility: we can no more express ourselves scientifically with 17th-century grace or 19th-century limpidity than we can now pray convincingly in the language of King James’s Prayer Book. The text must be more strictly ordered, not just to make it that bit more convincing, but because a modern scientific paper is only a tiny fragment of knowledge that must somehow be given meaning within the immense edifice to which it contributes its millionth part. Nowadays it must satisfy higher standards of superficial intellectual civility, as befits a modest newcomer to the solemn circle of its elders in one of the innumerable areas of specialised interest to which it seeks entry: that is not to say that it has vastly improved in rationality, accuracy or fundamental validity. The control of ‘quality’ in scientific publishing applies less to inner essentials than to maintaining an agreed general level of outward form.

A significant development has been the continuous improvement and increasing use of diagrams, maps, figures, pictures and other ‘illustrations’. Technical innovation, from woodcuts, through copperplate, lithography and wood engraving, to photo-reproductive processes, has made such material cheaper to produce, and opened up entirely new channels of non-verbal communication. In a recent issue of Nature, picked up at random, I found an average of about one illustration per page – graphs, circuit diagrams, electron micrographs, satellite photographs, chemical structures and so forth. In parallel with the transformation of the public media from verbal to pictorial tidings of good and bad cheer, the traditional media of science have evolved the means to convey information that could scarcely be made intelligible in words or algebraic symbols. This development, seemingly born of innocent technical facility, has revolutionary implications for the philosophy of science, which has never come to logical terms with the irresistible credibility of a recognisable visual pattern.

But the ‘literature’ of science is more than a collection of messages from scientist to scientist, or from scientists to the general public: it is a social institution, central to the internal functioning of the scientific community. In the beginning, little trouble was taken to identify the sources of the information that appeared in the scientific journals. But discovery does not come without individual effort: the labour of research has to be recompensed with ‘recognition’, whose just allocation depends upon secure claims of priority. A major trend in the evolving protocols of scientific publishing has been towards precise attributions of authorship and date to every distinct ‘contribution’. In the 18th century, for example, the formal date of publication of a journal issue was deemed sufficient evidence of priority – until disputes arose about the long delays between the editor’s receipt of a manuscript and its appearance in print. If a piece of intellectual property is to be registered at its full social value it must be published in a ‘reputable’ journal subject to editorial quality control. So now an article carries both the date on which the manuscript was received and the date on which it was ‘accepted in final form’, perhaps after months of argy-bargy with those uncomprehending, anonymous, cowardly, yet potent referees. And the author’s modest signature – as it might be, ‘G. White, Selborne’ – becomes a list of 57 names, whose alphabetical order coyly conceals vast distinctions of authority within the team, drawn from seven institutions (addresses complete with zipcode, please), not to mention the all-important footnote reporting that this work was supported by grant number ABC 12345 XYZ of the PQR agency – and so on. The evolution of these rubrics mirrors the historical transformation of science itself, from amateurism to professionalism, from the academic mode to the industrialised and collectivised research of today.

Most scientists, bless them, take the system of science publishing for granted. But they know very well that scientific publications are precious and must be preserved, for that is where, if anywhere, scientific knowledge resides. A library of scientific books and journals is an ever-growing archive, where original, primary scientific information is stored, and from which it must always be retrievable. That, again, is no new problem. Indexes, lists, collections, abstract journals, catalogues, data compilations, and other instruments of information-retrieval, go back a couple of centuries. The exponential growth of science, doubling every fifteen or twenty years in numbers of scientists, in scientific papers, in journals, in institutions, in expenditure, and in every other indicator, has always seemed explosive to the point of impending crisis. Decade by decade, the archive becomes more precisely categorised, with standardised citations to journal volumes and pages, standardised journal titles, prescribed formats for abstracts etc.

Where can I find that one little fact that I desperately need? Secondary services expand and combine to publish very elaborate and expensive abstract journals, deeply indexed by subject, author, biological species or chemical compound. And now, of course, these essential facilities are computerised, for immediate access by teleprinter, from continent to continent. Is all our thinking about messages and meaning to be superseded by the ‘electronic journal’, always available on the telescreen but never printed, crammed with continually updated data which can never be read by the human eye, a warehouse of ‘information’ without a scrap of understanding or knowledge? If I do find that little fact I need, will I recognise it or understand it, or be able to use it as I had hoped? Is the ‘journal’ a neutral channel for the rapid transmission of disconnected fragments of fact and conjecture going to its absurd extreme, superseding the more ancient and human device of the ‘book’, which attempts to speak coherently and asks to be swallowed whole?

There is much more, both of broad issue and of instructive detail, in this volume than I can possibly indicate in a short review. The separate chapters, on topics as diverse as ‘The Development of Commercial Science Journals in Victorian Britain’ or ‘The Changing Appearance of Research Journals in Science and Technology’, will take their place as distinctive contributions to the history and sociology of science through an underlying unity of purpose and theme, and they cohere into a ‘book’ that succeeds remarkably.