A historical novelist on working with sources, circa 1829

With that attention which is due to the Public, I have consulted a vast variety of books that are descriptive of the period, and most of the old chroniclers and black-letter volumes that treat on the subjects of my narrative.  In recurring to the suitable references, I found the most extraordinary discrepancies in many of their details.  However, as I was composing a romance, and not a history, I deemed it unnecessary to attempt to reconcile discordant accounts, conceiving that any one authority was a sufficient basis for the support of my relations.  Where I met with that, I was content to act upon its single evidence.¹

I've long since been disenchanted with attempts to fold historical fiction into historiography per se, but this is the most striking example of why that I've come across in quite some time. Miss Crumpe wants to cite her authorities and ignore them too: what matters is not the facts, but merely that some historical text recorded that such-and-such happened...whether or not it actually did.  Talk about the triumph of textuality! As it happens, Crumpe very much wants to produce a particular attitude to the past and its potential significance for the present, but she clearly thought that such a project required narrative probability, not accuracy.  The novelist's convenience trumps the historian's argumentation...        

¹ Miss [M. A. T.] Crumpe, Geraldine of Desmond, or Ireland in the Reign of Elizabeth.  An Historical Romance, 3 vols. (London: Henry Colburn, 1829), I:xxi.