How to record archival documents at volume
Please reference as: Liz Stanley (2017) ‘How to record archival documents at volume http://www.whiteswritingwhiteness.ed.ac.uk/HowTo/How-to-record-documents-at-volume/, and provide the paragraph number as appropriate when quoting.
1. Why record at volume
1.1. For many people working on archival or other kinds of documents, reading is an intensive that involves giving detailed attention to making sense of the content of a document, and equally detailed attention to its structural features including its placing of writer and intended readers. This takes time, care and focus.
1.2. This is not the kind of reading that this ‘how to’ discussion is concerned with, although reading in detail like this, in reading particularly important documents in an analytical and interpretive way, is the topic of a later ‘how to’ contribution. Instead, the concern here is with outlining a different approach to reading, which is reading at volume in order to gain a broad sense of many documents, who wrote them, for what reasons, and what the main contents are concerned with.
1.3 WWW research is concerned with a ‘big question’, how social change regarding race matters in South Africa occurred over the lengthy period from 1770s to the 1970s with respect to how white people represented this to others in their letter-writing. As a result, in this project it is important to look at how many people did this, not just a small number, because only by looking at different kinds of people, living and working in different circumstances, in different parts of the country, and at different points in time across this 200 year period, can defensible generalisations about the processes of change be made.
1.4 What this means in terms of archival work is that each research visit entails looking at many letters and other documents that collections contain and recording basic information about all of these, rather than homing in on just a few supposedly choice letters. This is for a number of reasons. Firstly, the ‘choice’ letters or other documents usually become apparent only later in a project’s life, rather than being self-evident at the start. Secondly, such things take much of their meaning from the documents that surround them and which are usually more quotidian and mundane, and so it is important also to record these. And thirdly, only by recording at volume can the significance of things which appear to be ‘choice’ be established. After all, something may stand out to a researcher with present-day relevances in mind, but have had little real significance in context.
2. How to
2.1 But the question remains, how to do this, how to record at volume? The key things are, (a) how to know what to record, and how this should be identified? (b) what information should be recorded, and in what kind of way?
2.2 The approach taken here is to not worry too much about supposed importance while in the thick of working on a collection in an archive, for what is important or not usually becomes apparent only afterwards. Instead to concentrate on recording sufficient information to enable things to be identified and retrieved if they later take on significance. To aid this, documents are skim-read and basic information about all of them is recorded, with a process of reviewing built-in after each day’s work, looking through all daily entries to build up a picture of what has been done and what it adds up to, and also in a preliminary way to think about anything that now seems more important than others.
2.3 This is a matter of imputing meaning to a document or documents.
A letter or other kind of document takes meaning from the mind being cast on it and the questions being asked of it. And, while a capacious memory helps in this, the thorough and systematic recording of information is essential. The main components are as follows, with a screenshot of the first page of the database recording form that WWW research uses shown here.
2.4 A collection – For each collection worked on, it is important to record its archival referencing information, including its title, its retrieval number, whether there is an inventory, why it has been archived where it has. Also to record who contributes to its contents, its earliest and latest dated items, the kinds of things overall it contains, what its main relevances to the research being carried out are. NB. Some of this information can be recorded at the outset, while some is best done at the end.
2.5 Meta-data, document basics – For each document, record its specific reference number, its date of writing, who it is from, who it is to, where it was written and (if available) where it was sent from and to, and the broad reason why it was written. Be systematic and consistent about recording such things, for this is what is known as ‘meta-data’, and it contains the information needed for referencing a document in a report, thesis or publication and also for retrieving the document from the archival system on a later occasion. NB. It helps to set up a simple form with suitable headings in a word-processing file and to copy and paste this many times, so that the same information is recorded for everything read. An ordinary spreadsheet is not really suitable for this, as additional information will also be recorded (see below) and this may be too lengthy, and so the alternative to a word-processing file is likely to be a database.
2.6 Content – remember that the purpose is to skim-read and to record information about a large number of items, and so comments about content should be succinct unless a particular document stands out. If it does, record the reason for this. For each item, skim-read quickly and then record two or three sentences summarising its tone or ‘voice’, structure, different sections, its overall message or point, its audience (both the named addressee and perhaps more widely). NB. Initially, this is likely to take some time. However, it is a matter of practice, the aim should be to do it speedily, and with perseverance and practice this will happen.
2.7 Intertextualities – Intertextuality is a component of content, but is important enough to warrant a separate comment. It identifies references of an explicit but also an implicit kind which a document makes to other letters or documents and can provide important clues to a network of people, organisations, events, beliefs or understandings which inform the particular item being read. NB. The term ‘document’ should be interpreted broadly here, recognising that it stands for different kinds of entities and not simply written ones.
2.8 Project specifics – The points made above are concerned with recording things that are project-transferable – whatever the research is about, ‘meta-data’ and broad content should be recorded. But in addition, all projects will have specific aspects to them which also need to be briefly recorded. For WWW, this includes: (a) the people mentioned and the role of this in providing a narrative of figurational connections, (b) the uses of ‘naughty words’, words which now have considerable opprobrium in racial terms but many of which started out in a more ‘innocent’ way, (c) labour, land and boundaries. NB. There is also the important ‘anything else that seems important and catches the eye’ which should be briefly noted as well. Again, this is for future reference.
2.9 Transcription and digital photographs – Sometimes, the importance of a document in project terms is immediately apparent and when this happen its contents will need to be recorded in more detail. If an item really is this important, then the best way to do this is by making a transcription, that is, an extract that has been recorded verbatim, mistakes and all. If the passage in question is short, this is best done in archival situ. If it is long or even an entire document, then a better use of time is to digitally photograph it if this is permitted, so the transcript can be made later. HEALTH WARNING! This should only be done sparingly. Digital photographing in a wholesale way defers making sense of what is in a collection, as explained in earlier ‘how to’ discussions, and is very difficult to recover from. NB. Where digital photographs are taken, a file structure should be set up beforehand which corresponds to the part of the collection these come from, so that the JPEGs or other file forms can be downloaded in a way that enables them to be easily retrieved later.
2.10 The points outlined above might seem to add up to a lot of things that will take a long time to do. However, doing this in a systematic and consistent way means that actually it can be done reasonably swiftly. What supports this is to use either a recording form with suitable headings in a Word file, or in the case of WWW we use a database form written in the ‘File Maker’ format into which these things are quickly recorded (see here the screenshot above). NB. WWW does not just digitally photograph material. Everything is skim-read in the archive, with digital photography reserved for what is particularly important. This ensures that the knowledge of a collection and its contents is built up incrementally while working on it and while in the archive it is located in, so any cross-connections can be followed up while still in the field.
3. The review stage
3.1 As discussed in a previous ‘how to’, a productive use of archival time is to work in the intensive way characterised as ‘extreme archiving’, by using each working day as a period of concentrated focused immersion in document after document. At the same time, it is important to gain perspective on the work that has been carried out and this is most productively done in an incremental, day on day, way.
3.2 At the end of a day of ‘extreme archiving’, there are usually in excess of a hundred short records, and also some notebook entries about puzzles and possibilities that have arisen over the course of the day. Going through these after the working day is finished enables sense to be made of who is who, why they write to particular people as they do, what the elliptical references in how they write are about, in project terms what are the more and the less significant things being thrown up, and also interesting leads that need to be followed up on and any documents that can helpfully be gone back to. Understanding then becomes iterative and incremental. And also if anything has been missed, it can become number 1 on the agenda for the next day’s work.
4. The incremental and the documentary method of interpretation
4.1 The term ‘documentary method of interpretation’ is Harold Garfinkel’s and signifies a fundamental of ethnomethodological thinking regarding how sense is made of the social world (see his (1967) Studies in Ethnomethodology). The analogy of the jigsaw puzzle helps explain. A small number of jigsaw pieces is taken as indexical of something a bit bigger, a pattern or picture, but in a trial and error way – some red bits, so perhaps they are part of a sunset; no, add some more pieces and it’s a red ball. And so with the larger puzzle that is a set of letters, an organisation’s papers and so on. There isn’t just one thread, one picture, that can be followed through the composing documents but a number. Eventually, incrementally, however, the picture becomes clearer.
4.2 For Garfinkel, the ‘document’ here is not a literal one. His concern is wider – how do we know what’s going on, how do we find out? what clues do we use to make sense of things? But the method fits working with literal documents such as words on paper or screen as well. What is in these clues? What do they add up to? What should be taken most note of? Anything that doesn’t fit the emerging pattern?
4.3 Understanding, then, is a matter of incremental additions to knowledge and consequent revisions to existing frameworks where relevant. So working in a focused day to day way supported by a daily process of review and revision is a highly effective means of understanding a great deal of material in a finite period of time.
5. Alternatives, cons and pros
5.1 One strategy for archival research is to use the secondary literature to identify particular items of interest and then build an archival investigation around these pre-decided relevancies. This might seem a very efficient use of time, but what it does is to lock a researcher within the confines of what they and some other colleagues already know. Another common strategy is to go through items in archive boxes by turning these over and looking for something that ‘catches the eye’, with any productive results sometimes referred to as produced by serendipity. The result is something like finding a needle in a haystack, and while the needle may be interesting the character of its location is also important but will be largely ignored, and this can sometimes lead to erroneous conclusions being drawn.
5.2 ‘Extreme archiving’ has been conceived and practised as an alternative to these and their weaknesses. Its critics might see its thoroughness as wasting time on much that is not very relevant to a project being carried out. There are three important reprisals to this. The first is that it uncovers a good deal of material that is in fact relevant, but which was not known about before. The second is that what is ‘important’ is a matter of incremental understanding, as already explained. And the third is that this approach enables any conclusions and generalisations made to be confidently asserted as meaningful and important, because made in full realisation of how they fit into or depart from the broader context.
Last updated: 22 December 2017