Counterfactual reasoning and analogical reasoning

Analogical reasoning is fine, if done well. It is normal for people to compare events with their own experiences or with other things that have happened in the past. Hopefully the analogy fits and doesn't lead

A problem with contrafactuals too far is that historians tend to freeze all other contingencies and let the history diverge wildly. In reality, things tend to 'repair' themselves. Losers in a battle work harder to get even, while winners have to move deeper into enemy territory and disperse to hold the conquered territory.

For example, in the US Civil War the capture of the CSA army at Ft. Donelson didn't end the war...instead the CSA scrambled and put out a new army that lost a close battle at Shiloh. But contrafactual histories of Gettysburg, where and CSA win would likely be a narrow one, assume that the war would end right there, rather than the US army falling back a dozen miles or so (where they had a defensive line mapped out) and fighting again. So in reality, a CS win at Gettysburg would likely end much like a CS loss did, with Lee eventually retiring to VA for the fall.

I would like to refer everyone to Professor Margaret MacMillan's use of analogous reasoning in her article here. It is indeed very interesting to note how analogy is used.

After further research, I came up with the following idea that analogous reasoning identifies a certain 'abstraction' from which common events are related. Historians then apply the same abstractions to newer events. I was pretty disheartened to see that there is no rigorous defence of analogous history, though. Anyhow, thanks for the numerous perspectives, and I hope you can share more ideas.

Edit: I also saw a recent post 'pattern for world wars?' which I have to clarify, is not my objective in discussing Prof MacMillan's paper here. I am more interested in the application of historical analogy, rather than coming up with a general framework to subsume all world wars under (that would just be speculative!)

Unfortunately, historians are not very scientific, so the kind of reasoning you envision is more or less absent in historical research. History, as practiced today, is more about collating facts from literary sources, rather than reasoning about those facts. Not that there haven't been plenty of people who have attempted theories of historical reasoning. For example...

Karl Marx (and Engels) are perhaps the best known historical logicians. They created a method they called "dialectical materialism" which proposed a sort of logic of history. It was popular in the former Soviet Union.

One of the best known theorists on historical reasoning is Karl Popper. He was a philosopher, not a historian. As a general rule, theories on historical reasoning tend to be constructed by philosophers like Popper, not historians.

Plenty of other such theorists can be found. For example, a man named Lewis published the book "A Treatise on the Methods of Observation and Reasoning in Politics" in 1852 and you can find many other such books. Historians ignore such books.

I took many history courses in college and not once on any of the reading lists was any kind of analytical or methodological work included, nor did any of the lectures ever include any attempt to teach the students how to reason about historical fact. You can assume all American history professors were brought up in the same tradition.

In terms of recent publications you can refer to Martha Howell's "From Reliable Sources: An Introduction to Historical Methods" which sketches out the basics of how historians work and various techniques for ascertaining "the truth", whatever that means.