80IFLA

IFLA General Conference 1980

Division: General Research Libraries

Section: University Libraries and other General Research Libraries

4/UN/3E

UAP AND DECENTRALISED UNIVERSITY LIBRARIES

By

D.G. LAW

Sub-Librarian

Edinburgh University Library

EDINBURGH

United Kingdom

A steadily increasing flow of articles continues to appear on Universal Availability of Publications (UAP) and in particular on its relevance to inter-lending systems and to national libraries. The present paper does not therefore intend to repeat the concepts of and the programmes behind UAP or even to argue the need for it. University libraries are sophisticated organisations in what they acquire, how they acquire it and in the demands made of them. Exchange and inter-loan are routinely sought, if not always acquired. It might therefore be thought that universities have a great deal to gain from UAP, but little to contribute. It is therefore intended to examine one type of university library system - the decentralised system in large universities to review its advantages and disadvantages, to examine how it may be affected by UAP and how it may promote it. Naturally, this will concentrate on library systems in the developing countries, principally the United Kingdom, with which the author is familiar.

Decentralised libraries have a long history. We know that the Alexandrine Library had created a branch library as early as 235 B.C. It was housed near the Temple of Serapis and was known as the Daughter Library. This library was probably created because the main library was full; but branch libraries can also develop to meet the needs of users at a distance from the central building, or because a large inflexible central library is found unable to readily satisfy the legitimate requirements of small special interest groups. Branch libraries tend to be regarded as expensive luxuries, eating up resources which could better be spent on central services. Probably since the creation of the Daughter Library and certainly throughout this century university libraries have expended a great deal of effort in trying to eliminate branch libraries, so far with an almost total lack of success, but the idea persists. Ashworth¹ thinks the new universities of the United Kingdom "have been fortunate enough to have new and central library buildings" and that libraries with many sites "should work constantly to reduce their number”; Harrison Bryan² of Sydney University Library sees that as a threat and as a poor substitute for the central library. Rogers and Weber³ of Yale and Stanford respectively view centralisation as a tremendous advantage and a very meritorious principle. A. L. McNeal⁴ of the University of Miami thinks the arguments for centralisation have force and justification. Yet these branch libraries must have some value to users, or they would not exist. By adopting a negative attitude to decentralisation, libraries create an area of conflict. A policy which aims simply to close down as many small libraries as possible may be bureaucratically correct without being in the interest of all users. A more positive approach to decentralisation can help to further the cause of UAP. Liebaars⁵ is surely correct when he writes that “availability may in the end be due to large and national libraries but it begins in the small and local libraries.”

Level of decentralision

Schumacher⁶ argues that all large organisations are biased towards order at the expense of freedom. The larger an organisation, the more obvious and inescapable the need for order. This leads to centralisation and the imposition of an unyielding bureaucracy; but it denies creative freedom and the organisation becomes moribund and a desert of frustration. The burden of proof lies always with those who want to deprive a lower level of its function; they must prove that the higher level can actually fulfil that function much better. The higher level must not absorb the function of the lower level on the assumption that being higher it will automatically be wiser and fulfil them more efficiently. Schumacher’s ideal large organisation will consist of many semi-autonomous units. Its structure can be symbolised by a man holding a large number of balloons in his hand. Each of the balloons floats free, but the strings are firmly held by a central power. The centre will now more freely do the things which belong to it alone, because it alone can do them: directing, watching, urging or restraining as occasion requires. This is not the only organisational model and is not even necessarily generally applicable, but where a large multi-site library exists, it gives a more positive approach to management.

One important element of the model is the need for central control. Although scarcely contentious, it is always worth repeating the importance of the university librarian having authority over all libraries in any part of a university.

The greatest danger of decentralisation is perhaps that of going too far. Where a branch library is small and probably housed in a department, it will tend to be manned by junior or part-time staff and, perhaps worse, is regarded by users and even library staff as part of the department rather than an extension of the library. Soon such a library has adopted its own rules and regulations and very likely has a private classification scheme which gives every sign of having been drawn up over coffee by some of the academic staff, all of this completely at variance with the needs of the whole university community and often inimical to making publications available to users from outside the department. At the same time, that small library is the only face of the library which members of the department regularly see and its successes and failures and the quality of its staff will mould the user’s opinion of the whole library system. Yet, no matter how good they are, junior or clerical staff will never achieve as much as professional staff. A department which creates and then relies on its own small library cuts itself off from the wider world of knowledge. The small departmental library is therefore the enemy of UAP; it deprives other users of its stock and fails to meet all the needs of its own users.

Larger units with professional staff and an adequate range of bibliographic tools provide a much more satisfactory service and allow the library to adopt a more positive approach to UAP in the most distant parts of the university. Although it is generally supposed and has been remarked by Garside ⁷ that British university libraries commonly conform to the pattern of a single university library with one librarian and one library committee in control of policy, it has, in practice been quite common to have some more or less independent libraries in subjects which were traditionally self-contained, such as Law or Medicine. It is true that in theory these are simply branches of a single unitary library, but in practice they have operated more in the federal manner described in Schumacher’s model. A larger unit, such as a Faculty library, combines the advantages of decentralisation with the advantages of size. It is too big to be identified with any one department and so clearly forms part of the library system, but is not so big that it cannot react readily to users’ changing needs.

The parts of a decentralised library tend to be criticised for waste of one sort or another, but this arises in part through viewing them in isolation. The alternative, which is a very large central library, has to be considered and it can be argued that the waste is at worst only exaggerated by decentralisation and is certainly not created by it. It has been suggested that staff in a decentralised system are inflexible, that someone cannot be moved from the Law Library to the Medical Library when illness strikes. In a large centralised library it is equally unlikely that a cataloguer could take over the serials office or that subject specialists could move between subjects. Really the criticism is of inadequate training programmes rather than decentralisation. Another criticism is the increased time spent in communication; but communication is just as difficult amongst two or three hundred people in a ten floor building as amongst the same number of people in ten separate buildings. Decentralisation does have penalties such as duplication and increased transport costs, but if the alternative is always borne in mind it will be found that much of the criticism relates to the size of the organisation rather than the manner of distribution of its staff and stock.

The advantages are great. Firstly and most importantly, the user is much more likely to have confidence in such a library. It is large enough to be manned by professional staff who are the user's peers; it will have relevant stock and will have the resources to rapidly identify and acquire items not in stock; it may well be arranged by one of the specialised classification schemes more suited to the users’ needs than general schemes such as Dewey or Library of Congress. At the same time it will not be so large that an inflexible bureaucracy must refuse what seem (to the user) reasonable requests lest it create precedents which might affect other users. This responsiveness to the needs of users is vital to UAP, which is also promoted by such simple facts as that a larger library can readily be staffed for longer opening hours, so increasing availability.

The negative view is to consider branch libraries as needless and wasteful and to try to eliminate them. The positive approach is to accept that where the people are is where the library should be, but to ensure that only viable units are created.

Duplication

The University of London has an extremely large library system based on fifty-nine colleges and schools, many with duplicated teaching programmes. It was, at least until recently, an unco-ordinated library system which one would expect to exhibit as high a degree of duplication as might be found anywhere. Yet when this was examined, the figure proved to be 15%.⁸ If this is taken as a high figure it is reasonable to expect that in a library system with no overt duplication in the university’s teaching, the figure will be lower. The Universitaire Instelling Antwerpen (UIA) has a totally decentralised library with no central collection, but a network of departmental libraries. Such a system might also be expected to display an extreme of duplication and UIA therefore monitors it carefully. All recommendations are checked centrally. Initial duplicate recommendations run at 10 %, but these are sent back to the recommender, who often withdraws ⁹ When the largely decentralised Edinburgh University Library system recently surveyed the duplication of journals, which according to folklore was rampant and contagious , just under 10% of titles were found to be in this category. To suggest that duplication in a reasonably efficient decentralised library system will run at something under 10% is not to say that such systems cost 10% more to run. Whether by accident or design, large centralised libraries duplicate publications and that must be deducted from any figure before a true estimate can be given of the amount of duplication indirectly attributable to decentralisation.

Ashworth ¹⁰ rightly points out that there is other duplication . Machines such as photocopiers, typewriters, duplicating machines, microfiche readers and security systems are necessary but expensive and there will certainly be more of them than in a single unitary library. But how many more is open to question. It is interesting, for example, to look at a large centralised library and see how many typewriters exist in library departments outside the typing pool. Unless the most stringent of regimes is enforced, the departments of a large centralised library will rapidly and deviously acquire the machines they feel necessary to circumvent the delays inherent in a large bureaucratic system.

Duplication is an undeniable penalty of decentralisation, but its level and cost cannot be considered in isolation. Some of the duplication is simply a result of size and is equally present in very large centralised systems.

Processing

The decentralised library does have one difficult choice to make and that is whether to centralise or decentralise its technical processes, particularly acquisitions and cataloguing and classification. The benefits of centralisation are largely those of economies of scale and the efficiency of specialist staff. It is broadly true that at least with acquisitions and cataloguing, the subject of the publication is irrelevant and that the expertise lies in the efficient performance of the task. On the other hand, the separation of these functions from the service point causes misunderstandings and it is less easy to make the system responsive to the priorities which users may require. It is probably more efficient for the library to centralise, but more satisfying to the user to decentralise.

Automation makes the decision no easier, but does resolve some of the difficulties. If centralised, it speeds processes up so that urgent treatment is scarcely required and the decentralised branches should be satisfied. At the same time, access to central files from the periphery can give the illusion of local control. Alternatively, provided common policy and standards are laid down centrally, there is no reason why acquisitions and cataloguing cannot be performed locally, with the input being fed into a shared central database. This does not lead to the loss of the specialist knowledge of central functional units. Ashworth ¹¹ points out the vulnerability of local branches to illness and suggests that central technical units can, as a subsidiary function, act as a pool of staff to meet temporary emergencies. This is no doubt true, but the smaller units tend to command greater loyalty and there is a suspicion that they suffer less illness involving absence from work.

Collection building

An ideal for every library is that its collections should be built towards specific targets rather than haphazardly arrived at. Although it is now generally agreed that no library can be self-sufficient, this does not absolve librarians of the responsibility for producing the best and most effective collections they can, within the budgetary or other constraints imposed upon them. Now the programme of UAP contains within the concept of ‘availability’ two opposing tendencies which have created a dilemma for university librarians since the advent of mass higher education; these are the provision for recent and future needs.

Present needs are centred on a very small part of the collection. As Line¹² has pointed out, it is not enough to acquire a copy of a publication, process it, shelve it and assume that by the repetition of these acts UAP is achieved. Many studies now point to the fact that a reader will only find on the shelves some 40 - 50% of the items he finds in the catalogue of a library. Other studies at the British Library Lending Division have shown the heaviest demands to be concentrated on relatively few journals. What most users need is not access to an unending number of journal titles, but the prompt provision of a few of the best known journals. So, for present needs, judicious duplication of heavily used material can disproportionately increase the availability of publications and hence the value of the library to its present users ¹³.

But what is of value now, especially textbooks, will not necessarily be of value to the future user of the library. Trueswell¹⁴ sees one function of the university library as supporting research into the furthest depths of obscurity. We do not know what these future obscurities may be and we cannot provide for them all, but large libraries at least have an obligation to provide for the future. Scilken's¹⁵ comment that most librarians would rather provide for the future readers they will never see than the underserved readers they have at present is an uncomfortable home truth, but scoffs at the value of all material purchased for the future rather than just the wrong material purchased for the future. The best guideline to buying the right material is an acquisitions policy and the best guide for that is the present strengths of a library’s collections. We may not know how the branches of learning will grow or wither, but by clearly identifying our roots and strengthening them, we make the tree of knowledge most secure. If we have coherent collections they will be of greater value to scholars, while a successful programme of UAP will ensure that the fact that some areas of knowledge are not well represented in a library is no disadvantage.

A decentralised library inevitably increases the problems of collection building. An acquisitions policy, even a skeletal one, is of the greatest importance to prevent needless duplication and to ensure that the library as a whole is pursing common policies. Duplication is not a bad thing when it is a planned measure to increase the availability of publications: too often however, it happens by chance and is wasteful. Here it must be said that the UAP programme’s requirement for union catalogues applies just as much within institutions as between them. It is surprising how many advanced libraries lack a single complete union catalogue of their holdings, although this is vital in preventing duplication.

Inter-lending

UAP requires the development of effective inter-lending systems. Just as important for the well organised decentralised library is an efficient intra-library lending system. Where acquisitions policy has been defined to minimise duplication, each library in the system has to rely on other parts of that system to meet some of its needs. This again demonstrates the necessity of a union catalogue, coupled with an efficient delivery service to exploit the library's resources to the full. In some cases, whether because of size, security or inconvenience , it may be preferable to have the reader go to the book, rather than the reverse. In such cases, the decentralised library must strive to ensure that its rules and regulations apply universally throughout the system and do not unduly penalise the visitor, who should have the same status in all sections of the library.

The UAP programme stresses the need for national planning, and in several articles explaining the need for UAP, Maurice Line has emphasised the value of fairly comprehensive central collections coupled with a rapid and efficient inter-lending system. It is undoubtedly true that such collections in every country would represent an enormous step towards UAP. In the United Kingdom, the British Library Lending Division (BLLD) at Boston Spa approaches that ideal, but even BLLD , with its enviable reputation, modestly disclaims perfection. However, by concentrating on this level of provision, there is a danger of ignoring the value of local collections and particularly their possible role in the rapid provision of fairly mundane material.

Libraries tend to co-operate, when they co-operate at all, by type: university with university, public library with public library and so on. Yet university libraries can be surrounded by a wealth of material and the geographical sprawl of the decentralised library magnifies this. The British government recently funded the preparation of a Union list of serials held by and in the vicinity of Edinburgh University Library. The university library itself has a large number of units, but the feasibility study revealed an enormous undergrowth of learned societies, government supported research institutes, museums, botanic gardens, government departments and agencies, technical colleges and quangos, each with a library, often of a highly specialised nature and in some cases of surprising size. While this can never replace BLLD, this small local venture means that while Edinburgh University Library receives only 10,000 serials, it now has rapid local access to 25,000 serials. The bonus for a decentralised library such as Edinburgh is that it already has a van delivery service between the various parts of the library and it requires minimal additional outlay to include these non-university libraries in the van route if a journal is required. This venture illustrates the need for awareness of the riches on our own doorsteps.. We should not simply increase our dependence on central collections but should increase our knowledge of specialised collections and work to create arrangements which make them more universally accessible. Line ¹⁶ has pointed out that this may not be the most economic means of document supply, but cost-effectiveness must be set against the convenience to the user and tempered by his needs. Cost-effectiveness is one point to be considered and a very important one, but it should not be the sole arbiter of our decisions.

The decentralised library can also improve the availability of publications for its users by the movement of stock between parts of the library. The branch library can hold heavily used material and regularly transfer out-of-date or little used items to a central store as research and teaching interests change and develop. The university itself is never static. Departments expand and contract, appear and disappear, but most frequently of all they move to another part of the campus. A decentralised library can always keep the books users require within easy reach. In the past, a great drawback to such movements of stock has been the time-consuming volume of changes to catalogue entries. While this can never be entirely eliminated, automation offers us the prospect of reducing this workload thus allowing us to tailor collections much more closely to the users needs.

Access and Technology

The decentralised university library system can pose problems of access to publications. It is a well known fact that users are reluctant to move even a few hundred yards from their own territory and many would rather do without a publication than visit another library. This does not matter if they can be confident that their “own” library will acquire the publication for them by loan, exchange or purchase. In this lies the paradox of technology for the large decentralised system, although much of the following probably applies to any large library. Microform, on-line information services, telex, facsimile transmission, the electronic journal present libraries with a wider range of information than ever before and the ability to acquire it more rapidly than ever, while at the same time the cost of the hardware prevents its distribution to all sections of the library. Even the simple microform, which has made available an enormous and ever-growing range of old and rare books and manuscripts, becomes a daunting prospect when a library is faced with the cost of purchasing several dozen microfiche readers.

A long-standing source of delay in acquiring books is the fact that when an order is placed, a library rarely knows whether the bookseller has an item in stock. If the bookseller then has to go to the publisher or wholesaler, a delay of weeks can occur. Some British booksellers are now considering allowing large customers access on-line to their computerised stock records. This promises to produce a significant improvement in the speed of supply, and hence availability, when a library is able instantly to establish which supplier has a book in stock, before placing the order. But again the large decentralised library is at a disadvantage if it has chosen to decentralise its technical processes. The capital cost of providing a large number of computer terminals is high enough, but even more prohibitive are the expensive maintenance contracts based directly on the number of machines which the library possesses.

Schumacher ¹⁷ pointed out the need for cheap, easily available technology and the above examples illustrate his point. When we consider the failure of librarians to persuade industry to produce a cheap, portable, popularly acceptable microfiche reader, despite the immense potential market, it may be doubted whether we form a strong enough pressure group to promote the development of cheap technology, but this is perhaps, worth further debate.

Co-operation

Mention should be made of the university library’s national role in promoting UAP. In the United Kingdom, as in other countries, there are a number of schemes, both official and unofficial, which ensure coverage of the publications of various parts of the world or periods of time, in which universities are involved. Just as important is the university’s role as local watchdog. No national library, however meticulous, will acquire every item published, especially the many items not formally published through commercial channels. By the very nature of their role, the staff of a university tend to produce quantities of working papers, reports and draft documents; they serve on boards, committees, working parties and pressure groups; they travel the world in the course of their research, gathering this sometimes ephemeral but often important material. Unless the university library collects such items, if necessary for onward transmission to a more appropriate library it is unlikely that they will survive.

The National Library of Scotland. The eight Scottish universities and Edinburgh and Glasgow City Libraries have produced a large joint statement listing their particular collecting interests. This allows them to concentrate on these interests, to prevent needless competition and duplication, to ignore and refuse involvement in areas already better covered by one of the other participants, and, most importantly, allows them to guide collections of material or even single items to the most appropriate home. All of the libraries involved had acquisitions policies of a sort, at least for their special collections. By working together, each of the participants has been forced to refine these policies and has seen the benefits of co-operation rather than competition.

The incomplete nature of national collections while always open to improvement cannot be a matter for real complaint. It is a salutary and chastening experience for a university library to try to assemble and maintain a collection of all the official and unofficial publications of the university. It is also very necessary, as the national library is unlikely to have the local resources to track every item down. The decentralised library system has an advantage here. Its tentacles spread to all corners of the university and in a good library its staff are in daily contact with the members of departments and their affairs, thus smoothing the path for the work of acquisition. Every library should be aware that it has a duty to ensure the preservation of items it is uniquely placed to acquire.

Conclusion

The decentralised library can actively advance the cause of UAP through the responsive relationship which staff can form with readers in each section of the library. Close contact with readers enables library staff to exploit to the full collections of the whole library system from local to international level. The university is based on the frontiers of knowledge and this can make great demands on library staff, but Foskett ¹⁸ is surely right to argue that what characterises good service is that the librarian becomes an integral part of the research team. This demands a breadth and depth of knowledge of the subject and of the library’s strengths and weaknesses and is best achieved when the librarian lives with the community he serves.

A decentralised system can also benefit the undergraduate. The normal argument against this is that all fields of knowledge should be available to the student, and as the student will almost certainly confine himself to one library, a branch library deprives him of this wider knowledge. But a large central library can be very intimidating to the undergraduate and may well deter library use. This is one reason for the appearance of undergraduate collections, which surely qualify for the description of decentralised. In any case, the true function of the university is not merely to impart facts to an individual, although this is inevitably part of the learning process; the true function is to teach the student how to assemble evidence and then how to evaluate it and the library is central to the first part of that process The decentralised library is in a much better position to guide the young student into the complexities at the heart of its system and to open up worlds of knowledge in the way that a university should.

The decentralised library is an inevitable consequence of the decentralised university. Ranganathan¹⁹ said that wherever people habitually congregate, that is a site for a library. A large library, indeed any library, has no value of itself; it has no meaning unless it is used. The library is there above all to serve, to meet the needs of the community in which it is placed. Our readers may at times be wilful, or unhelpful, or wrong; we may at times have to protect them from themselves, but without them we have no purpose. We must not regard outlying units of the library as rivals to the main collections or as a weakness in the system. Provided such branch libraries form adequately staffed and economic units, they can do more to achieve UAP than any single collection of a million volumes which intimidates rather than invites.

References

1. 1. Ashworth, Wilfred. The multi-site dilemma. Journal of Librarianship January 1980, 1-13

2. 2. Bryan, Harrison. University Libraries in Britain: a new look. London, Bingley, 1976

3. 3. Rogers, R.D. & Weber, D.C. University library administration New York, H.W. Wilson Co., 1971

4. 4. McNeal, Archie L. Divisional organisation in the university library. In: The library in the university: The University of Tennessee Library Lectures 1949-1966. Hamden, Shoestring Press, 1967

5. 5. Liebaers, Herman. Universal Availability of Publications - a concept and a programme IFLA Journal 4(2) May 1978, 117

6. 6. Schumacher, E.F. Small is Beautiful. London, Blond & Briggs, 1973

7. 7. Garside, Kenneth. Co-operation in collection building in British universities. In: Developments in collection building in university libraries in Western Europe, edited by Willem R.H. Koops and Johannes Stellingwerff. Munich, Verlag Dokumentation,1977

8. 8. Urquhart, John A. & Schofield, Kenneth. Overlap of acquisitions in the University of London Libraries: a study and a methodology. Journal of Librarianship 4(1) January 1972, 32-47

9. 9. Vervliet, H.D.L. Acquisition policies and methods in a new graduate college in Belgium. In: Developments in collection building in university libraries in western Europe, edited by Willem R.H. Koops and Johannes Stellingwerff. Munich, Verlag Dokumentation,1977

10. 10. Ashworth, Wilfred. op. cit.

11. 11. Ashworth, Wilfred. op. cit.

12. 12. Line, Maurice B. Universal Availability of Publications. Unesco Bulletin for Libraries 31 (3) May-June 1977, 142-151

13. 13. Gore, D. Let them eat cake while reading catalogue cards: an essay on the availability problem. Library Journal 100 (2) January 15, 1975, 93-98

14. 14. Trusswell, Richard W. Basis for the no growth collection. In: Farewell to Alexandria: solutions to space, growth and performance problems of libraries, edited by Daniel Gore (Westport, Greenwood Press,1976)

15. 15. Scilken, Marvin. Solving space and performance problems in a public library. In: Farewell to Alexandria: solutions to space, growth and performance problems of libraries, edited by Daniel Gore (Westport, Greenwood Press,1976)

16. 16. Line, Maurice. op. cit.

17. 17. Line, Maurice. op. cit.

18. 18. Foskett, D. J. Beyond reductionism. Journal of Librarianship 2(2) April 1970, 139-143

19. 19. Ranganathan, S.R. The five laws of librarianship. 2^nd. Ed. Madras, Madras Library Association, 1957