Poor tools to think with: the human space in digital diplomatics

The speed of change encapsulated in Moore’s Law and its consequences dictate that it is very easy for those who took the time to educate themselves in computing for the humanities to find that it has now moved far beyond them. Projects with cutting-edge methodologies blunt quickly at this pace. Nonetheless, it would seem that in some areas progress is being made more slowly, these being those where the techniques required are more those of cognition than of data man- agement. This paper uses the example of the speaker’s own doctoral project, a ‘straight’ socio-political historical enquiry that happened to employ electronic means of an untheorised kind to manage the diplomatic data on which it was founded, to explore the line between these zones. When I began my doctorate, which was an attempt to study the workings of power in a frontier area, the Marca Hispanica of the Carolingian Empire, my available computing power was a 486 DX-2-66 PC with 8 Mb of RAM and no database software. I was also sharing care of a toddler, while my home institution was in a different city a nd its computing staff were overstretched. Had my circumstances been otherwise, had I had good technical advice, had I known what I was doing, this paper would probably not exist. I know now that what I wanted was a Wiki-style database with pages for each document, each person, each place, linked to each other by HTML. What I made instead, and still have, was a number of Microsoft Word files peppered with embedded DDE and OLE links to each other. The number of ways in which this is not ideal would be a paper in themselves, but it has one surprising advantage, its subjectivity. Because there is no data schema, I can record what I like in these files, without having to deform the record into shapes determined by a computing structure. The chief disadvantage was of course that the data capture tools were my eye and my brain, which were not enough. By the time I reached the end of the project I had acquired enough database experience to be working from a reasonably elaborate database in Microsoft Access for certain parts of my sample (some 150 documents) which allowed me to be sure of working with all the data they contained. But this involved huge numbers of decisions as to what each document was about, what role the parties to it were playing, how their names should be spelt and so forth. Since one of the principal purposes of the database was to allow me to decide on the basis of the available information whether the people in the documents, who used no surnames and rarely specified relationships, were recurring, how could I avoid making those choices at data entry stage? This paper uses these questions as a basis for asking larger ones, about the judgements do we ask computers to make, about where human input is still required in digital diplomatics, and whether the advantages of human and computerised judgement can be enjoyed at the same time without diminishment of either?