I ONCE WROTE (LINE, 1989 ) THAT there was nothing that national libraries did that could not be done in some other way or by some other body or bodies, and was not so done in one or another country. This is true of even the most basic functions. The deposit and preservation of national imprints could be spread among several libraries, where they can be consulted; and the national bibliography (which some countries do without) could be produced by the private sector or co-operatively by other libraries. The big question was whether the most cost-effective way of performing necessary national functions was by the national library.
Since then many national libraries have grown weaker simply because governments have cut their funds: they have improved their efficiency, but that has not enabled them to maintain their former acquisition programmes or services. The Library of Congress is now the only national library that aims to have a comprehensive world wide collection. This does not mean that many countries do not still try to ensure that they have a fairly comprehensive coverage of the world's publications. Several have attempted co-operative acquisition programmes, which have never really succeeded for logistic and other reasons. Others have worked on the more sensible principle that their libraries collectively must achieve a wide coverage; as in Australia, their national libraries have made a virtue of necessity and proclaimed the Distributed National Collection (Henty, 1995 ). Their policy of aiming to collect all national printed publications remains intact, but they have had great difficulty in extending deposit to other forms of publication, including material available only online.
This last is an obvious example of an established function of the national library being threatened by IT. IT is changing the whole concept of publication; the Web contains, alongside research articles, vast quantities of trivia and also serious discussions of the kind that might previously have taken place in the press. The Web is in fact enabling new forms of communication to prosper. A national library that collected only printed matter would in a few years have a very incomplete record of the nation's published output. Similar factors apply to non-book media, which not all national libraries have collected in the past: if collection is restricted to tangible forms an increasing body of material will be missed. Even if the legal deposit law is all-inclusive, as in Norway, the problems of enforcing it are almost insuperable, and the burden of collecting is insupportable. Another problem is that globalisation affects publication like almost everything else, so that it is often hard to know what is published in or by a particular country.
This same factor of globalisation is one of several that is beginning to make national bibliographies an irrelevance. Most of them are in any case very incomplete, since they cover only books; and who wants a published list of works issued in a particular country, rather than information on works of interest to them in languages they can understand? However, a distinction should be made between national bibliographies and national bibliographic control, which is the only way to ensure comprehensive worldwide coverage.
Globalisation is also making it less necessary for countries to aim at worldwide collections even of printed matter, whether selective or comprehensive, concentrated or distributed (Line, 1996 ). Journal articles can be supplied by photocopy or by digitised text almost as quickly (often more quickly) from abroad as from the home country of the requester. This will never apply to the same extent to books, but international access is improving yearly as the catalogues of more and more significant libraries become accessible online and libraries come to take more seriously their obligation to supply books in their ownership to other libraries. This is weakening another function of national libraries: the construction of union lists (which can easily be put together from automated records), and the actual supply of publications, though only the British Library does this latter on a large scale.
All libraries are affected by IT. At the same time it both poses threats, particularly that of being bypassed in favour of direct access, and offers opportunities. The ultimate threat is non-existence, which some think is a real prospect: public libraries because there are other priorities for funding and other opportunities for enlightenment and entertainment; academic libraries because students and researchers will soon be getting everything online. I doubt if anyone who knows and has thought much about these matters really believes in the coming death of libraries, but some do believe in their transfiguration into new types of organisation. Others see the likelihood of gradual change, with some new activities added and some old ones fading away. The ultimate opportunity is transformation (rather than transfiguration) into information storers and providers, reaching a larger number and wider range of users in a wider variety of ways and playing an interactive part as information exchanges. Whether threats or opportunities dominate will depend on government policies and (not always the same) practices, on people - both librarians and their consumers - and on chance events.
National libraries too are faced with threats and opportunities. Threats come not only from governments but from other libraries and from the private information sector, both of which can use IT to do what national libraries do now. National libraries, like all big national organisations, usually attract a great deal of criticism (not to say abuse). The causes of this critical attitude differ from country to country, but they include resentment of their wealth, fear of their domination, impatience at their bureaucracy, scorn of their inefficiency, and irritation at their arrogance or aloofness. The fact that these features are usually exaggerated and often explicable does not change the criticism, which is often accompanied by an expressed desire to see them improved. IT at last gives ordinary libraries the chance to join together in competing with the national library - in collecting, in recording, in making bibliographic records accessible to users and available for use by other libraries, and in supplying material. Some national libraries expanded their activities in the postwar period; these extra activities, like business information services, are the most vulnerable, because if they make a profit they are targets for the private sector, while if they make a loss (as most of them do if all costs are taken into account) they may not be affordable. So we may see expansionist national libraries having to retrench and get back to basics (Line, 1995 ).
Opportunities for national libraries certainly exist, and they are being taken by many (Cornish, 1992 ); but they tend to constitute even greater opportunities for others. The only more or less unique opportunities they have are those based on exploitation of their collections. And there are opportunities that academic and public libraries have, in serving a prescribed body of users with personal services, that are not open to national libraries.
Yet there are things that good national libraries do that ordinary libraries in co- operation or the private information sector cannot do. One is to offer for consultation in one place a collection of material, past and present, from all over the world. There is room for more rapid acquisition and faster service, but no amount of access to remote resources can entirely substitute for this function, which is especially valuable for the humanities scholar. Likewise, distributed legal deposit collections are not a substitute for central collections, since boundaries between subjects and forms of material are often irrelevant to users, even when they are not out of date or artificial. It might be argued that near-comprehensive collections of national publications are becoming as much of an irrelevance as national bibliographies in a time of globalisation, but they represent a major part of the country's heritage, and unless in some inconceivably remote future nations become obsolete their symbolic importance and real value will continue.
Britain has a unique element in its national library - the British Library Document Supply Centre. There is no doubt that it would never be established now, but the fact is that demand on its services from the UK and abroad is huge and continues to increase. Is this a temporary phenomenon, due to the growth of literature and the simultaneous restrictions on library budgets? If so, will this soon be counteracted by a massive switch to online access to electronic publications? No-one dares to predict precisely what will happen to currently published journals and books. However, it seems likely that most books will stay much as they are, and that scholarly journals will fall into three categories: print only, print plus online, and online only, possibly in roughly equal proportions. If this is so it will still be immensely useful, not only for the UK but for the world, to have a vast collection of serials, conference proceedings, reports and books dedicated to supply to other libraries. To oversimplify the case: if the UK demand placed on the British Library was shared by the top 30 academic libraries, and if the items they did not possess were excluded, they would each have to deal on average with some 60,000 additional requests a year. Very large numbers of requests would not be accessible in the UK, and many would not be accessible at all. This scenario is not readily imaginable, and it will be a long time before IT makes much difference to the situation.
It is always useful to ask "If we did not have such-and-such, would we invent it?" From a strictly utilitarian point of view, it is doubtful if we would now invent monumental national libraries; we would find other and cheaper, if less effective, ways of performing national functions. But we do not start from scratch: big national libraries exist, and it is almost unthinkable, on the grounds of cost and logistics alone, to dismember them and distribute their resources among other libraries. Secondly, national pride is a major factor. National pride is not always a good thing; some terrible things have been done in its name. But national libraries are at worst harmless, and at best major contributors to civilisation: a good national library is a legitimate source of national pride. Otherwise, in an electronic age it is hard to explain the construction in recent years of huge new buildings for several national libraries, including those of Denmark, Estonia, France and the UK (Melot, 1996 ).
For such reasons as these they will continue to exist; and they need not exist merely as shadows of their former selves, so long as those that have worldwide collections are able to maintain extensive acquisition programmes, and those that confine their collecting to national materials are able to do so effectively. In either case, the selective archiving of electronic publications is essential. How far they go beyond building, maintaining and preserving collections to exploiting them will be a big issue. Doubtless the better national libraries will continue to explore possibilities.
 Line, M.B. (1989) National Library and Information Needs: Alternative Means of Fulfilment, with Special Reference to the Role of National Libraries, UNESCO, Paris.
 Henty, M. (1995) Resource sharing ideals and realities: the case of Australia's distributed national collection. Advances in Collection Development and Sharing, 1, 139-152.
 Line, M.B. (1996) National self-sufficiency in an electronic age, in Electronic Documents and Information: from Preservation to Access. 18th International Essen Symposium...1995 (Helal, A. and Weiss, J., Eds), Universitatsbibliothek Essen, Essen, pp.170-192.
 Line, M.B. ( 1995) Back to basics for national libraries? (Editorial.). Alexandria, 7(1), 1-2.
 Cornish, G.P. (1992) The Changing Role of the National Library in the New Information Environment. Alexandria, 4(2), 125-141.
 Melot, M., Ed. (1996) Nouvelles Alexandries: les Grands Chantiers de Bibliotheques dans le Monde. Editions du Cercle de la Librairie, Paris.
Maurice Line is a former Director General in the British Library and is now a freelance consultant