Selection in practice (Mark Jones)

Generally broadcasting archives exist, not in an academic or research vacuum, but to provide in-house programme makers with sound recordings for use in future programmes. This has the advantage of giving the archivist a clear user profile and the considerable satisfaction of tangible and consistent use of his or her collection, luxuries that can be denied to other sound archivists. However, this seeming advantage can lead to a tendency to be both complacent and defensive about their working methods.

Broadcasting organisations obviously have an interest in claiming that they make high quality programmes. However, they can easily assume a defensive pose when questioned about what they preserve and what they throw away or recycle. Given the vast quantities of transmitted programme output, some kind of selection process is inevitable. But an organisation that prides itself on the quality of its programmes must answer its critics when it declines to preserve that material. I hope that within the BBC, the Sound Archive does its best to avoid complacency and a siege-mentality but it remains a familiar and perennial problem.

Recordings have been selected and preserved by the BBC since the early 1930s on a systematic basis1. In the first ten years of BBC radio, from its beginning in 1922, there had been a strong feeling, shared by broadcaster and audience alike, that the essence of the new medium was the simultaneous transmission and reception of a signal - live broadcasting. In fact, the BBC’s first Director General, John Reith, expressed the view that to transmit previously recorded material was tantamount to a fraud on the listening public. Given this atmosphere, an understanding of the potential value of collecting recorded broadcasts was slow to develop. The advent of the Empire Service and advances in recording techniques in the early 1930s made the formation of a permanent collection both desirable and practically possible.

Today, some 200,000 recordings later, the practical problems of selecting for the BBC Sound Archives are extensive. Essentially the problems tend to focus on one factor; the sheer size and scope of the Corporation’s broadcasting output. To put this output into perspective consider the extent of the BBC’s broadcasting operation. In strictly radio terms it is:

4 National Networks Radio 1,2,3 and 4
3 National Regions Radio Wales, Ulster and Scotland
8 Local Regions Birmingham, Manchester, Leeds, Plymouth, Norwich, Southampton and Bristol
28 Local Radio Stations  
World Service 24 hour English Language Service
40 Language Services  

The BBC transmits, through all broadcasting outlets, in excess of 150,000 hours of radio a year, roughly 3,000 hours a week. From this large and highly complex output the Sound Archives selection team, split into different subject areas, aims to make a series of realistic judgements based on a set of established criteria.

Put simply these criteria are:

  1. Is the recording likely to be of use in future broadcasts as primary source material? Does it illustrate a particular person, event, social attitude or change in speech or music?
  2. Does the recording possess significance in sound, over and above the information and/or style of the script?
  3. Does the library possess similar material and, if so, does the new recording increase the value of the existing collection by providing additional examples, improved performances, or better technical quality?
  4. Is the recording technically suitable for preservation? (Here a balance has to be made between the intrinsic value of the content and the technical quality of the recording.)
  5. Are there copyright, contractual or other restrictions on the use of the recording? If so, is the material of sufficient importance to merit preservation despite the difficulties limiting or preventing use, which may be temporary and removable at a later date?
  6. Should the recording be selected in whole or in part?

The material selected on this basis tends to fall into the following general broadcasting categories:

(a) Events (political, economic, social, sport, etc.)
(b) Voices and reminiscences of prominent contemporaries in all fields
(c) Social history (social conditions, work, leisure, folklore, etc.)
(d) Miscellaneous material for documentary, reminiscent and general interest programmes
(e) Linguistic (dialect and accent)
(f) Drama and entertainment programmes
(g) Music (works unlikely to be issued on commercial records, authentic folk and national music, outstanding performances and occasions)
(h) Sound effects, natural history and authentic documentary sound.

To cope with this range of output, selectors essentially use three methods to take in programme material; they are selection, sampling and acquisition.

Selection, in the first instance, is carried out by the daily scrutiny of the various scripts, publicity information, copyright and performance details, cast lists and credits of all kinds that are an essential part of the production process. In this way a large and amorphous output can be broken down into subject or programme areas and made more manageable. At this stage, listening to programmes on playback becomes a realistic possibility. Additionally, live news or current affairs programmes can be scrutinised after transmission from running orders and scripts, while the programmes themselves are recorded off transmission. Sampling is an appropriate way of dealing with the regular, day-to-day and essentially ephemeral range of broadcast programmes. The very familiarity of these programmes in the weekly pattern is the problem: they are easily overlooked and under-valued. As major programming, such as drama, features and live music, becomes more and more expensive, a formula of regular sampling has to be worked out if this output is to be regularly reflected in the archive.

Acquisition is an important third way of adding material to the collection. It is an ever-open-door policy of encouraging production staff to offer material of all kinds, often uncut tape of which, perhaps, only part has been transmitted. Similarly material is constantly offered by freelance broadcasters, ex-staff, external organisations and other sources. Often recordings can be acquired in this way thirty or forty years after their transmission date. This open door policy to material is a vital part of the whole process of encouraging contact and liaison between selection and other areas of programme production.

Of the three methods of adding to the BBC Sound Archives, selection remains the most demanding. Successful selection requires a constantly changing balance of judgement, a mixture of perspective and immediacy. Additionally, it is important to continually re-evaluate basic selection criteria and to re-examine user needs at the same time.

However, there remain a number of practical problems in the selection process, which are common to all broadcasting organisations. A major problem is one of providing information to accompany selected recordings. The more information that can be provided the more likely the recording will be re-used. Broadcasting recordings tend to be made in a hurry; under-researched, under-documented at the time they are made; and difficult to validate after the event. News and current affairs material is notoriously a problem area in this respect. With certain contractual conditions adding to the need for this kind of information, time-consuming research is a constant part of the selection process.

A selection policy can only work in practice if there is close contact between the programme-producing departments and the selection team. This contact is essential in providing pre-production information, post-production assessment and a whole range of discussion on common needs and aims. Without this kind of exchange, on both a formal and informal basis, archiving and production will always be dangerously apart. Within the BBC, archive material is re-used so often - on average 40,000 recordings are issued to programme makers each year - that producers must be involved in both the theory and practice of selection.

While any kind of selection process will always have its critics, a number of factors help to make the BBC’s radio selection policy a workable one:

i) 3,000 hours a week is a daunting and unmanageable total to cope with in terms of storage and information retrieval;
ii) Almost 50% of all BBC programme output on radio consists of commercial recordings of music, i.e. recordings that already exist in an easily accessible form;
iii) Programme output is repetitive. For example, on an average day BBC radio services probably transmit more than 1,000 news bulletins. These bulletins may differ in length and style, but the basic news components remain much the same. The selection role here is to take out of bulletins the core material - the actuality, eyewitness or statement for re-use;
iv) Broadcasting tends to feed on itself. Up to 15% of output may have been previously transmitted, repeated or recycled in some form or another;
v) As well as the central BBC Sound Archive there are a number of regional and local collections of sound recordings, which are available for use.
vi) New developments in information technology and in recording media mean that more recordings can be better preserved and more quickly accessed by existing members of staff. Developments of this kind have helped increase the BBC’s annual preservation figure by over 300% in three years, since 1981.

By and large, broadcasting archives tend to attract criticism because of their very fidelity to the broadcasting output they both use as source and cater for. Producers and researchers looking through the BBC’s archives for the authentic sound of ordinary people in the 1930s talking about their daily lives will be disappointed for this kind of material was not recorded or broadcast at the time. Similarly, in terms of current affairs and the coverage of certain events, it is inevitable that some are treated in greater depth than others. The archivist may regret the imbalance this inevitably causes, but it would surely be inappropriate to try and redress the situation and lose the sense of context that is finally the real strength and value of a broadcasting archive.

A programme archive must reflect the output of its broadcasting organisation, its tastes, styles and its content. In doing so it will be providing its future users, both from within and outside the organisation, with a faithful and valuable record of the past .

Mark Jones is the Sound Archives Librarian of the BBC, London.
This paper is a revised version of the one given to the IASA conference in Brussels, 1982.


1 [Note added in 2010] Digital storage now allows the BBC Sound Archives to retain 100% of all BBC TV and radio output.