Would have, could have, must have: Professional historian at work! Do not attempt on your own.

Perhaps because 'tis the season to mark papers, I am especially on the lookout for leaps of logic, sweeping generalizations, and other unfounded arguments.

Also, because I am teaching my course on debates in history in the spring. There, I try to get students to understand several fundamental truths:  persuasive argument depends on the marshaling of evidence, and evidence demands interpretation. I try to get them to see that the facts are essential, but do not "speak for themselves." The science of history is in a very real sense therefore the history of evolving interpretations, a dialogue with our colleagues and predecessors. For that reason, too, we generally say, we still find it useful to read and engage with even the historians of the distant past, whose views we no longer share. This is commonly cited as one difference between the human and natural sciences. The laboratory scientist generally derives no practical benefit from reading antiquated and refuted theories.

We examine several cases in which two historians look at the same facts and arrive at very different conclusions. A classic modern example is Natalie Davis's celebrated Return of Martin Guerre (subject of the popular film of the same name) involving marriage and mystery in sixteenth-century France. A young woman's socially awkward and sexually inept husband abandons her. How could she not know that the strapping, virile fellow who returns years later, claiming his role and rights, is in fact someone else? Oh: and this was serious stuff back then. The new husband was eventually found guilty of imposture and executed. The wife was found innocent, on the grounds that she was deceived. For Davis, she must have known, and the two must have colluded in order to fashion new identities and a romantic marriage influenced by new Protestant ideals. The judge was sympathetic and covered up the truth. For Robert Finlay, by contrast, the sources say that the wife was duped, and that is that: Davis imposes modern feminist values and a postmodernist reading on the sovereign sources. Davis's book is full of phrases to the effect that one of the characters must have or could have or might have thought or done this or that. For Finlay, this is wild speculation. For Davis, Finlay has a mindnumbingly simplistic approach to sources, assuming that they can be taken at face value, that we do not need to look beneath the surface or take into account the implicit subjectivity or even duplicity of the author. The reader will judge.

Still, even as we attempt to teach this complexity in interpretation, we teach students to err on the side of caution in their own writing. Davis and Finlay are professionals, intimately familiar with the nature of the documents and the society. They can assemble extensive evidence in support of every point. Davis's footnotes on even minor contextual matters sometimes threaten to become miniature essays in their own right. Davis brought to her adventurous interpretations decades of experience that students by definition lack. They tend to see sources in a much more limited way, and to use them accordingly. They think in terms of quoting or discussing individual sources in detail rather than synthesizing and assembling them in service of a larger point. Above all, because they necessarily lack the contextual knowledge, they are much more prone to outrun their evidence. It's a little bit like driver ed: the student may understand the theory and moreover know all the correct individual things to do, but putting them together in a real-life situation without adequate experience can be dangerous: you don't take her onto the streets of Boston or New York on the first day.

I therefore always recall my father's teaching me a Latin phrase from Terence that he learned in Gymnasium: "Quod licet Iovi non licet Bovi": what is permitted to Jove [Jupiter] is not permitted to an ox. In other words, not everyone has equal rights.

Here's an example of a reviewer grappling with this issue in a modern scholarly journal:

By the beginning of 1833, 'his views must have assumed their full shape' (p. 166). (One winces at the 'must have', a phrase to which Vereté was prone but one which if used by an aspiring postgraduate would bring deserved coals of fire on his or her head). Vereté was very good at discovering documents but he was inclined to take them at face value and sometimes failed to ask why these arguments were used, how far those who used them believed in them and how far statesmen acted on them rather than on other arguments.

from M. E. Yapp, review of From Palmerston to Balfour: Collected Essays of Mayir Vereté, Middle Eastern Studies 29 no. 2 (April 1993): 358-59

None of this, or course, is to say that facts don't matter.  As we are accustomed to saying, everyone has a right to his or her own opinions, but not his or her own facts. John Adams famously said in his defense of the British soldiers accused in the Boston Massacre:

Facts are stubborn things; and whatever may be our wishes, our inclinations, or the dictates of our passion, they cannot alter the state of facts and evidence.

And, writing during the time of the Vietnam War and the controversy over the Pentagon Papers, Hannah Arendt (not someone I generally quote if I can help it) cites a conversation between the French Premier and a German politician over war guilt in the First World War:

'What, in your opinion,' Clemenceau was asked, 'will future historians think of this troublesome and controversial isue?' He replied, 'This I don't know. But I know for certain that they will not say Belgium invaded Germany.'

Nuff said.