Sound archives have various origins. Broadcasting sound archives came naturally into being because of the primary need for developed storehouses of recordings for use in radio programmes. Other sound archives have developed within research or educational institutions which took up sound recordings as yet another source of information in their specialised fields (e.g. music, ethnomusicology, dialectology, political or social history). There are many and varied examples of such specialised archives, ranging from the Department of Sound Records of the Imperial War Museum in London to the Ethnomusicology Archive of the University of California in Los Angeles, and from the sound archive of the Netherlands Theater Institute in Amsterdam to the audio collection of the Indian Classical Music Foundation in Bombay. Many other archive$, however, developed inside institutions responsible for general collections, frequently of a national or regional character, which do not accentuate any specialised field. Thus centres like the Library of Congress in Washington DC or the Public Archives of British Columbia in Victoria have gradually built up extensive collections of sound recordings of spoken word and music alongside collections of books, documents and other media. A few sound archives have come into existence simply because their founders wanted to concentrate on sound recordings as such, regardless of any particular subject or regional interest, and independent of all other media. The British Institute of Recorded Sound in London is perhaps the best known archive of this kind.

Although the categories described above represent the main stream of sound archive activity other types of audio collections are also to be found. There are, for example, lending libraries which primarily specialise in the distribution of published material such as audio discs and cassettes and there are scientific institutes where sound recordings may form part of their monitoring or experimental data (e.g. recordings of the heartbeat made for medical purposes or bio-acoustic recordings used in the study of animal behaviour).

Whatever their origins, however, developments so far give the impression that in many countries sound archives - outside the realm of broadcasting – have been established as a consequence of momentary needs and certainly without much preliminary deliberation about an overall structure of sound archiving on a national scale. Even in areas where some effort has been made to consider whether the establishment of one or more sound archives would best meet the needs of the country as a whole, the outcome has seldom if ever been a structure based on clearly defined and elaborated possibilities and priorities. Some people may argue that this kind of 'structural' course is apt to fail and that a more ad hoc establishment of sound archives provides a stronger and more flexible approach to national needs than any other policy.

Whichever conclusion may be reached, however, it is important that the issue should be seriously and systematically considered. In many countries without any kind of sound archive organisation audio collections of various kinds are nonetheless coming into existence. The first requests for funds to cover their financial needs enter government offices or private foundations. Once this happens the administrators responsible for public or private money may, without adequate guidelines, grant or refuse resources according to the feelings of the moment. They should, however, consider the national need for sound archives and base their decisions on the outcome of such a study, while bearing in mind also the main existing models and options that are available to them.

In countries with as yet little or no sound archival activities there are two main ways of considering the subject, at least in theory. One is to look at sound archivism from the point of view of the medium itself, the other is to consider it from the point of view of the contents carried by the medium. The differences between the medium centred and the content centred approaches need further explanation before their respective advantages and disadvantages are examined.

The medium centred approach starts with the assumption that the preservation of sound recordings is so important and so specialised that the needs of the medium take priority over any other consideration. An archive based upon this assumption will tend to collect as many recordings as possible in order to preserve and describe them professionally and thus save them for posterity. Also such an archive does not usually give priority to research based upon the content of recordings and of other media carrying the same kind of information. The situation is reversed in the case of a content centred approach. Here the medium of recorded sound is just one instrument among many for research or education. It is the content of the recordings, which contributes to the total amount of information available for a given research or educational field, that is the centre of attention.

Anyone who is familiar with the field of sound recordings will immediately recognize that a strict differentiation between these two types of archives would be an over-simplification. In reality nearly every archive is a mixture of the two types with some accentuation of one or other of the two approaches. It is, for instance, obvious that no archive of the first type will totally neglect the contents of its recordings, while no archive of the second category will completely ignore the needs of the medium. However, this broad distinction is helpful for considering a national strategy for sound archives particularly in countries where as yet little or nothing has been achieved.

Bearing this distinction between medium and content centred archives in mind let us first consider the needs of countries where recordings, of folk music or oral history for example, are coming into existence without being collected, preserved, described and made available in a professional way. In such countries there is much to be said in favour of creating a sound archive on a national basis which concentrates its primary efforts on the acquisition, the preservation and the description of every kind of sound record. Such an archive may conveniently be part of a national library or a public record office of written documents. The national institution would then take professional care of the preservation of the sound recordings in its collections, in the same way as it would of its paper and other records.

Although its staff would not perhaps spend much time on research on the contents of the records themselves, such a national archive would - in addition to preservation - certainly concentrate its efforts on their description, so making them fully available for research and educational purposes. If a national collection of sound recordings can be organised in this way as part of a state institution, then the government may also be expected to take care of its funding. In large countries additional centres may also usefully be established as branches of the national archive so as to provide recordings and facilities on a regional basis.

However, the national sound archive model has its limitations which should also be seriously considered before any conclusions are drawn about the most effective method for the organisation of sound archives. Every national archive is likely to try to live up to its broad objective by acquiring as many recordings as possible and to function as an archival centre for the total acquisition of every kind of sound recording in the country. However, unless the archive is part of such an enormously resourced and highly differentiated institute as the Library of Congress, even national archives will nonetheless generally tend to concentrate on a restricted range of subjects and be forced to leave some others aside. Although they are institutions with a wider field of view than is usually found in specialised archives the range of a national archive is still limited. The aperture of its lens has a wide-angle but nevertheless is not able to cover 360º or even more than 180º of the whole field of knowledge. This is not a situation which can be changed just by raising more funds and employing more staff. It is a structural problem, encountered in every institution where people try to cover comprehensively all fields of knowledge.

A related problem is that national sound archives will be required to administer a wide variety of highly specialised recordings many of which need much more detailed treatment than conventional cataloguing in order to make them most accessible for scholars. In this respect they are to be compared with public record offices and like those archives of written documents they seldom have the range of expertise to succeed in really giving their attention equally and on an equally high level to every subject of interest covered by their collections.

What then of the alternative approach to sound archive organisation: the specialised, single subject and content centred archive? From the point of view of the researcher and the educationalist a sound archive specialising in, say, history may be a better solution than a national archive, especially when it is part of a larger institution whose collections also include books, periodicals, films, photographs, written documents and newspapers relating to the same field. In other words, where all media combine to give maximum service to the user, who wants to study his subject of research or to have access to material for the class-room regardless of the medium it may be found in. This so-called multi-media approach is probably only possible in specialised institutes with archives concentrating on one field of interest or, perhaps, in some very great national institution like the Library of Congress.

A specialised institute with collections of pictures and sound recordings next to collections of printed or written records must, however, be prepared to direct a relatively greater part of its total budget towards the audiovisual media, for the obvious reason that the acquisition and the preservation of such media are generally more expensive than is the case with written or printed records. Such financial discrimination in favour of audiovisual records is not always achieved. Institutes with low budgets may easily feel that audiovisual media do not warrant such a high expenditure and decide to make less funds available for their administration than is professionally necessary. This is a problem for multi-media institutes and it can only be corrected or forestalled by firm decisions concerning the allocation of available funds.

There are other problems which must be taken into consideration as far as the specialised, single subject, content centred archive is concerned. One concerns the financial situation as seen from the national level. In the case of rather small collections, dispersed over several institutes, each specialising in different fields of research or education, the total cost may be rather high by comparison with the funds necessary to run a single, centralised sound archive which keeps all types of material on its premises. One should, of course, never accept such statements without applying a detailed cost-profit analysis but the assumption seems to be credible enough. Another major problem is that even a widespread network of specialised archives may not be able to cover the entire field of sound recordings.

Two institutions respectively covering folklore and ethnology, for example, may not have any activity in the field of dialect nor any interest in its development and, as a result, a major gap may be left in the national holdings.
 

Having considered a few arguments in favour of both national and specialised sound archives, it might be interesting to take a quick look at three other organisational models. First, the Arkivet för Ljud och Bild in Stockholm is an example of a national archive for sound recordings, videotapes and films, partly based upon the fusion of a few already existing archival collections. This Swedish model, of one integral national archive for all audiovisual media, is also interesting because it presents an alternative to both the national sound archive philosophy and the concept of audiovisual archives as part of a multi-media national institution like the Library of Congress. In Stockholm, sound recordings and moving pictures are brought together to the exclusion of other media such as books and written documents. It should be remembered, however, that storage and particularly the preservation of audio and visual media are in many ways different from both the technical aspects and from the financial needs and there may be some disadvantages to combining them. Also, in such an arrangement, the sound archive will only profit as long as the budget is fairly divided between the media, with each safely secured against interference, and so long as the management does not pursue a biased policy favouring the visual media at the expense of the audio.

Secondly, in order to profit from a similar kind of centralisation of the preservation and storage of recordings, several institutes in the Netherlands have proposed to the Government the establishment of an organisation which would act as a central depot and a clearing-house for audiovisual media and would also fill an intermediary role between the Netherlands Broadcasting Foundation (NOS) and the various public organisations and groups interested in access to radio and television material. These plans are primarily intended as a solution to the problems concerning films and video-tapes but they will undoubtedly have far reaching effects in the field of sound recordings as well. The clearing house idea may perhaps prove to be a valuable kind of compromise between the rigid centralised structure of a national archive and the 'anarchy' of quite independent and divergent specialised archives, especially in countries where such archives already exist.

Thirdly, there is the unique approach of Austrian sound archives united in the 'Arbeitsgemeinschaft Öesterreichischer Schallarchive AGöS' (Association of Austrian Sound Archives). The AGöS foresees a few primary archives concentrating on the collection and preservation of original sound records. These archives should not be open to the general public, but a network of other institutions including libraries, mediathèques or audio-visual centres should act as distributors of copies of the sound records preserved in the primary archives. The AGöS considers it necessary to maintain content centred archives in this range of primary archives, also distributing their recordings through the network of regional institutions mentioned above. The system starts of course from the assumption that certain standardisations in the field of title description and distribution media have first been accomplished.

In reviewing the models considered in the preceding pages it will be seen that, despite the existence of several variants, the broad choice to be considered for the national organisation of sound archives lies between the establishment of a single centre or of several. In choosing between the unification or the multiplication of sound archives there are three other major factors that have to be assessed before the balance of advantage can be finally judged.

The first consideration, already touched on, applies mainly to countries in which specialised archives already exist. Once a collection of sound recordings is part of a specialised institute the creation of a single national collection is likely to become much more difficult to achieve. There will be a natural reluctance among such institutes to give up their sound archives despite any practical arguments in favour of a central or national solution. The effectiveness of such a solution then also has to be weighed against the value and efficiency of the service which existing archives are already providing. The inevitably disruptive effects of ending their independence obviously has to be more than balanced by the benefits that can be achieved through their amalgamation.

Secondly, an important part of the sound archival resources of most countries is represented by the output of their radio and television organisations. The national pattern chosen for sound archives may be significantly influenced by how these resources can best be organised in any particular state. Although broadcasting organisations commonly maintain their own archival collections, these are seldom open for educational and research purposes. Thus separate arrangements will generally be needed to provide public access to such material. Given the complications of copyright and contractual obligations, broadcasting organisations are often reluctant to provide copies of their recordings for use outside their own premises. They will certainly be even more hesitant if they have to deal with several institutes, each putting forward its own demands, than if they only have to collaborate with one. From this point of view a national sound archive can usefully function as the sole agent for broadcasting collections of sound recordings, providing a suitable access point for non-broadcasters while also centrally controlling and safeguarding the rights of the broadcasting agency and its contributors. A further advantage of such centralisation is that it ensures public availability of all broadcast recordings, including those which may happen to fall outside the orbit of existing specialised archives.

Thirdly, a similar case for a national sound archive can be made in respect of commercial discs and tapes published by recording companies. The reluctance of the recording industry to sustain a proliferation of centres holding copies of archival material in which commercial companies own rights is indeed now confirmed by its official policy. Thus its agent, the International Federation of Producers of Phonograms and Videograms (IFPI), has begun to promote the establishment of national archives in its member countries to serve as the only intermediary archives between the recording industry and the general public.

Such a centralised arrangement for broadcast and commercial recordings may not, however, always be the best one for researchers. Most national archives are - as mentioned before - in fact specialised in certain restricted areas of research. In acting as central intermediaries between broadcasting organisations or recording companies and researchers, however, they would also have to deal with many other fields of interest in which their staff may have no specialist knowledge. The BBC, for example, has understood this problem very well. Thus, copies from the large and valuable collection of recordings made during the Second World War have been made available to the Imperial War Museum, specialised as it is in that field, and not to the British Institute of Recorded Sound, which serves as a national sound archive but does in fact concentrate primarily on the field of music. As a matter of course any general archive might be expected to handle the BBC Second World War collections at a lower level of description and research than the specialised staff of the Imperial War Museum.

To conclude, the problems of and the models for the national organisation of sound archives are manifold and it would be unwise to pretend that there is but one solution for every country developing activities in this field. However, the establishment of a national sound archive is in many cases the best safety net for the recordings which every country is producing in ever greater amounts. This will ensure that all kinds of recordings will find their way into a professional preservation and description centre, where at least they may be saved for the future. Eventually specialised archives may also come into being, but even then a national archive may continue to fulfil this central function while also preserving those recordings which would not be collected elsewhere.