Criteria For Selection

The term selection implies a procedure based on the general policy of the archive and certain criteria within the limits of the policy. What criteria can we establish without hampering future research and destroying recordings which, in a hundred years or more, could have become interesting or even indispensable? Are there methods to avoid disaster and to protect ourselves from blame by our successors? It is doubtful whether such criteria can be found but we should try to formulate a few points which can be applied without too much risk. Apart from obvious things like the discarding of dubbings or recordings of very bad quality, the following should be taken into account when an archive begins to define selection criteria.

  1. The specific qualities of the medium
    Sound archives are collecting music and spoken recordings or are concentrating on one of the many other fields. Spoken word can, of course, also be preserved in writing or in print. It is, however, not really possible to convey on paper variations in tone, laughter, sighs, chuckles, interruptions and intervals-in short, non-verbal expressions. This does not mean that one has to preserve every recording of spoken word. We should restrict ourselves to records, which contain medium-specific information. So many recordings of speeches by official persons, made entirely in accordance with the policy of their government, are in fact second-rate sources which do not add significantly to the knowledge stored in traditional archives of written and printed records.

    All of this means that we should concentrate on recordings made without previous preparation such as live-interviews, discussions and improvised talks; in other words, recordings, which enrich already existing, printed reports in the daily papers and official documents.

    Medium- specific qualities apply also to music recordings, since such recordings cannot be replaced by printed music in any way. Thus the first criterion will seldom apply to music, because it is by nature medium-specific and irreplaceable.

  2. The division of work between archives
    Most spoken word archives are in fact specialized institutions, concentrating on restricted fields, and usually there is only a small overlap with other institutions. If there is duplication, as is sometimes the case with broadcast sound archives and research archives outside the radio, it is there because radio archives are not able to provide a service outside their broadcasting institutions. However, the general policy of archives should be very clear about the limitations of their own collection as well as others and selection policy should take account of these limitations. This applies equally to spoken word and music.
  3. The length and completeness of recordings
    Selection has also to do with the length and completeness of recordings. This does not mean that only extensive and complete records are valuable, because a very short abstract from an early broadcast may be worth many long recordings of later date. In the case of spoken word it is particularly difficult to decide to what extent fragmentary recordings are useful. News broadcasts, for instance, which are transmitted by the dozen every day, usually consist of many comments and few authentic sounds. They are useless for research and for most educational applications. On the other hand complete recordings of live interviews belong to the more important part of every archives collection and must certainly not be eliminated because of a too strict selection policy. In the case of music, complete recordings are preferable in most cases.

The above-mentioned points give us something upon which to base a policy. In short: are our recordings adding to the traditional written media or are they worthwhile because of their specific qualities as sound records? Are they held elsewhere in the country or abroad, or are they too short and too fragmentary to provide useful information? Criteria along these lines will in general not impede any future research.

There are a few additional points, which are more risky, but can do little harm to our descendants in the world of sound archives.

4. Single records or complete collections
Most records, be they spoken word or music, belong to series or to collections brought together with a specific aim. In many cases records derive their importance from the mere fact that they belong to a collection, while single records without any relation to other recordings stand apart and may be less valuable. The recording of a well known Haydn Symphony by a certain orchestra under a particular director is of course different from the same symphony recorded as part of the complete series by Antal Dorati and the Philharmonia Hungarica.

5. The importance of the subject: estimation of value
Frequently spoken word recordings have been made because at the time people seemed to be interested in the subject. Radio broadcasts, in particular, tend to be of temporal value, fashionable or tied up with sudden bursts of sensational curiosity. Archivists should be able after some time to distinguish between temporal and more enduring subjects. There are risks in this approach, because any tape may contain the one and only recording which eventually proves to be of outstanding value but, as long as we deem selection to be necessary, the subject-criterion provides another weapon against pollution of our precious collections. Archivists of music recordings may easily find parallels within their field of interest.

6. The importance of the subject: social history
There is a tendency to apply social sciences and historical research to daily life, the life of the man in the street, the unemployed, workers in factories or minorities in great cities. Aside from the inevitable exaggeration of this movement, it is nowadays a matter of common understanding that historians and archivists have spent too much time on outstanding events and very important persons, and that they should change their course. While a lot of documents cover the dealings of the so-called establishment, the number of records related to the circumstances of living and the cultural interests of the public at large is relatively small. Selection should take care of this distinction and place less value on outstanding persons and more on social history, at the cost of our customary collections of voice portraits of VIP’s, who as a mater of fact are well prepared for eternal life anyway.

Further to this summary of the criteria of selection in general terms, it should be indicated that for each specific subject of research, whether music or one of the many fields of spoken word, the archivist may develop his own criteria within general parameters, dependent upon the policy of the archive and the point of view within that field research. However, it is not easy to use more specific criteria without grave risks of the wrong kind of perfectionism. General directives and a well-developed common sense are better remedies than so-called scientific criteria which, in practice, spoil much of the fun of collecting and do not really add to a well-balanced archive collection.