1/2 idea No. 8: Critique of ‘what-if?’ histories/Markov Chains

(I am sharing my possible research ideas, see my tweet here. Most of them remain only 1/2 or 1/4 ideas, so if any of them seem particularly promising or interesting let me know @jon_agar or jonathan.agar@ucl.ac.uk!)

This idea came from a state of grumpiness. In particular, it was a response to a growing willingness among historians of science to entertain counterfactual – ‘what if?’ – histories. Examples include Peter Bowler’s Darwin Deleted: Imagining a World without Darwin (2013) and Gregory Radick’s BSHS Presidential Address ‘Experimenting with the scientific past’. Both are actually thoughtful and rather good. Hence I think my reaction was due to unfair grumpiness.

Nevertheless, before I was a historian I was trained in mathematics, and there are both simple and complicated ways of thinking critically about counterfactual histories.

The simple point is that any counterfactual sequence of historical reasoning has to be a sequence of probabilities. One way of picturing this is as a decision tree. At each node there’s a probability of taking a new path. Now take even a small sequence, well within the kinds of sequential narratives we find in history of science, say sixteen nodes. Even if the chance of taking a counterfactual path at each node was 4 times out of 5 – pretty good individual odds I think – then the chance of the final outcome would be (4/5) squared four times – less than 3%. Any realistically long counterfactual sequence results in an outcome that is deeply unlikely.

The complicated point is that there might be fields of mathematics (Markov Chains being one, although I now have reasons to doubt their applicability) which might help model historical processes, ones pictured as chains of probabilities, in perhaps interesting and useful ways. I am well aware that the very thought would make many historians run for the hills.