The Birthing of an Historical Fact

Jacques Barzun and Henry F. Graff, in their book The Modern Research, discuss how an historian determined whether John Stuart Mill was the author of an anonymous letter that appeared in Le Globe in 1832, on the doctrines of the Saint-Simonians. Here is the evidence Hill Shine marshalled to decide the question (I quote from the book):

• First, research discloses two earlier allusion [by a Le Globe editor] to the effect that "one of the most powerful thinkers in London" intended to write a series of open letters on the new ideas [for the paper].
• Second, there exists a letter to that same editor announcing the visit of a third party who would bring him... "the work of your young friend M."...
• Third, it was within three days of receiving the piece of news just recorded that the newspaper published the open letter of "an Englishman" who signed himself "J."
• Fourth, a letter of May 30 from Mill to his Saint-Simonian friends refers to "my letter which appeared in Le Globe."
• Fifth, a footnote added to that private letter by the editor of Mill's correspondence, states that Mill's public letter appeared on April 18, 1832.

Barzun and Graff here write, "At this point, no further doubt is possible" (pp. 80-81, emphasis mine).

This is a great illustration of what I have been talking about re history:

• The historian starts not with facts upon which he puts his subjective interpretation, but with evidence from which he must derive the facts.
• The actual work of the historian qua historian consists in determining what actually occurred. The facts are the conclusion of his research, and not at all his starting point.
• His conclusions are as reliable as anything in any of the "hard sciences": not immune to revision based on further evidence, but fairly definitive given the evidence we currently possess.

(In fact, as Collingwood notes, if history is not reliable, the "hard sciences" cannot possibly be so, since all that we know of all earlier experiments in the hard sciences is itself historical knowledge.)