This document was prepared for the project team in its early days. It was intended as a contribution to defining the methods that would be used for gathering data and to initiating a sort of quality assurance scheme. In fact it has been the main source of reference for contributors, the project team and its managers. The main conclusions the team drew from this paper were that: a) there was a clear need for a shared understanding about the evidential material it would work with; and b) it was neither helpful nor desirable to create a comprehensive set of objective rules that would govern the project’s actions – as is stated several times below, the evidence often needs to be judged on a case-by-case basis.

The paper is made public on this site partly because it might be helpful to users and potential contributors to the database, and also because doing so accords with the project team’s open access strategy.

The LED database will consist of text extracts that record the experience of listening to music. Material for the database is not limited according to the type of music or the period in which the experience occurred, but a protocol must be agreed concerning the quality and veracity of evidence. This paper outlines the protocol the team will follow and, implicitly, a set of guidelines for dealing with different types of evidence.

General guidelines

The database is devoted to evidence of the experience of listening to music. It is primarily intended to record personal experiences of listening to music which were not written with the intention of influencing public opinion, as well as ‘surrogate’, third-party accounts of the experience of listening in which commentators observe the effect of music on others, especially those whose voices may not otherwise be heard because they are, for example, illiterate (see ‘Records of the effect of music on others’ below). The texts may originate from anyone’s experiences, whether a professional musician, an amateur, or someone with no musical expertise.

It is difficult to offer absolutely precise and definitive guidelines on the sources that can be included in the database, but it may be helpful to think of the possible types of source as a continuum. At one end of the possibilities will be, for example, diary sources that were only ever intended for the eyes of the writer. These are undoubtedly important sources for the project. At the other end of possibilities lie professional performance reviews written for publication and mass consumption, in order to pass opinion on particular events. These sources will not be included. Between these extremes lie many possible types of source. For example, private correspondence may contain references to performances that are incidental, or it may contain references that pass opinion on performances. Both are likely to be acceptable for the database. However, published correspondence in a journal whose sole purpose is to have an impact on public opinion will not be included.

How is evidence judged to be authentic or genuine?

• The measure of whether any given piece of evidence is genuine or contrived for effect is difficult to define. For example, some utterances made through media outlets might be contrived for self-advertisement. As a general rule, the evidence should be taken at face value: it is the task of those who use it to assess its veracity and usefulness for any given purpose.
• Documentary material must be authentic. It may not be possible to fully test the authenticity of all material, but where there is any cause for doubt (for example, if the evidence occurs in a secondary source), reasonable steps must be taken to authenticate it. The database will acknowledge any doubt about the veracity of the material to be recorded so that end-users can be alerted to make their own investigations.

Who are the end-users and how should they be taken into account?

• The database will be available globally and will be open-access. Professional researchers will be attuned to the pitfalls of using any type of source, and it is up to them to revisit the evidence and test it for their purposes. Less experienced researchers will rely on the information in a less questioning way, so the guidelines given here need to be followed consistently, taking into account that the layout of the database (and the way it is presented more generally – for instance, its introduction and explanatory exemplars) can accommodate riders to any given evidential extract.

How is material for the database limited?

• Material must come into existence independently of the project. For example, the project team will not solicit material through interviews or the acquisition of written submissions. Although volunteers will be used to gather data (‘crowd sourcing’), the project will not create data.
• Evidence is not restricted to the experience of listening in Britain, but the project is devoted to material already existing in the English language, including material that has already been translated into English. The reasons for this are entirely pragmatic. A later stage of the project may look at foreign-language material in collaboration with other institutions, but there is no provision in the present scheme for translation or editing of foreign-language texts.
• Evidence that first appeared in published form presents challenges and can not be subject to a general rule; advice on different types of publications is given below. However music reviews – critiques of composition and performance – are excluded. This is because the quantity of such material is so vast and also because such writings have a premeditated quality that the project needs to avoid.

Records of the effect of music on others

While sources that record the effect of music on others can be seen as one step removed from the ‘intimate experience’ of listening, without such evidence the database would not contain testimony of the response of crowds or groups of people, but perhaps more importantly, it would be largely devoid of evidence emanating from the illiterate, especially in earlier times – for example, colonial accounts of the impact of music on indigenous peoples. Again, a case-by-case approach needs to be taken, but the key question to ask is whether a source reveals something important about the experience of music without which the database would be significantly poorer.

How does one judge the quality of evidence?

The quality of evidence must be assessed on a case-by-case basis, but many of the normal rules for dealing with historical sources can be applied, especially as the evidence (with some exceptions that are outlined below) consists of first-hand accounts of the experience of music. However, some guidelines might be helpful when making judgements about quality:

• In most cases, judgements should be possible about a mass of material rather than individual segments. For example, a judgement might be made about the content of an entire database (such as British and Irish Women’s Letters and Diaries or the British State Papers available through British History Online).
• As has already been said, most evidence can be taken at face value, with any equivocations recorded as necessary.

What follows are guidelines that might be drawn on when judging different types of source. This is just a start – further guidelines should be added to the list (so that everyone is working to the same understanding) as problems arise and are solved.

Guidelines on sources

Personal correspondence and diaries

These should be regarded as high-level sources because they are likely to contain expressions of personal and intimate experiences of listening.

Published books and articles

These have to be treated on a case-by-case basis. Works of writers such as Croyat and Burney are extremely important because they offer genuine evidence; some other writings are less obviously important.

Autobiographies

Autobiographies, whether published or unpublished, can be regarded as high-level evidence even though the subsequent use of such material may require caution. Autobiographies typically record experiences at a considerable temporal distance, but this is something that end-users can take account of.

Biographies

These are secondary sources and should be treated as such. Usually the evidence gained from such sources should be restricted to text that is directly quoted from the subject of the biography.

Journalism

While musical criticism is explicitly excluded from the database, there is not a general exclusion of journalism. For example, some of the most graphic records of the experience of listening to music in fields of conflict appear in the published writings of war correspondents (the reports by Russell from the Crimean War are a case in point). Of course, such evidence must be taken primarily as the experience of the journalist, but such writings often contain evidence of the apparent experience of other listeners.

Oral evidence

This should be regarded as high-level evidence if the manner of its acquisition is likely to draw a genuine rather than contrived response. Evidence gained through the standard protocols used by oral historians is especially valuable.

Official papers

Official papers such as Parliamentary Papers might yield evidence, and these sources are easily available online. Evidence might be found in Hansard, for example, and ships’ logs (which are official documents of the Royal Navy) might include relevant information.

Works of fiction

A decision has been taken not to include material taken from works of fiction.

Works of reference

OED is rich in quotations, as is ODNB; otherwise, it is difficult to see how such publications would be useful for the project.

Copyright and openness

It is not the purpose of this page to deal with issues of copyright and openness, please follow the link to our policy. However, copyright issues will apply to our use of sources, and advice has been taken on this and on the level of protection we wish to apply to both the content of the database and the software developed for it at the Knowledge Media Institute (KMI) of the Open University.

Prof Trevor Herbert