Work for Songs for George Thomson is now in a critical phase, as the team pulls together the masses of research carried out over the last few years. This seems like an appropriate moment to reflect on one of the foundational processes that underpins editing practice: collation.
It is easy as a casual reader to forget the contingency of the text that we meet on the page – what source or sources has it been lifted from? What layers of mediation has it gone through? These are political questions in the world of textual editing, which has evolved over time through a number of dominant approaches. An older disciplinary consensus tended to favour the production of an ‘ideal text’, with editors making selective amendments, mediating between a number of sources to create a satisfactory iteration. Currently the fashion is for accurately reproduced manuscript sources (see my earlier blog on the philosophy of diplomatic transcription) and the idea of a ‘social text’ – which captures a moment in the history of the text, as it was experienced by a particular readership. The latter approach is acutely relevant here, as we are focussing on the songs as they first appeared in Thomson’s publications.
However, despite this apparently simple method, in order to be rigorous we need to provide our reader with complementary information on other important versions of the songs. In a field as complex as Burns’s, this is where things get complicated. Beyond Thomson’s own first publications, the major alternative sources are as follows: Burns’s own manuscript copy, often contained in letters to Thomson; other significant contemporary publications, including those by James Currie and R. H. Cromek, and James Johnson’s Scots Musical Museum; and (most labyrinthine of all) Thomson’s own subsequent, altered versions.
Where applicable, each of these sources must be carefully collated against our ‘base-text’ (in this case Thomson), noting any and all variants. Variants may be as superficial as minor changes in punctuation – though even these significantly impact on sense – ranging to a radically different imagining of a song. The priority must always be to see sources in person, yet for the supportive work we have a significant advantage over previous eras: where they often relied on photostats and transcriptions, we have access to high-quality digitisations.
Collation, then, is very much the bread-and-butter of textual editing. It is a task that requires patience, concentration and a good eye. Though an apparently innocuous job, removed from the higher theoretical practice of literary scholarship, it is a key mechanism in producing a reliable, informed and thoroughly scholarly text for the next generation of readers. Careless or imperfect collations can and have introduced stubborn errata into the canon – the only cure being a new, methodical return to the original sources.