Napoleon III “was what I often think is a dangerous thing for a statesman to be — a student of history, and like most of those who study history, he learned from the mistakes of the past how to make new ones.”[i]
This quote is from A.J.P. Taylor, described in Wikipedia as “a British historian, journalist, broadcaster and scholar.” I was led to him by Francis Gavin, in his recent book “Thinking Historically, A Guide to Statecraft and Strategy.” © 2025, Yale University Press, 233 pages.
I know, I know: light summer reading. I blame McKinsey Global Publishing’s 2026 reading list[ii], which made the book sound interesting. It is.
This is not a book review. Rather, I want to speculate about how Gavin’s ideas have validity and application in the world of business. Because I’m an Ops person, I’m going to focus on failure analysis.
Let’s start with Gavin’s first identified pitfall of history, because I think it’s the one which can cause the most trouble: Overcertainty and Rigidity. “History misused to reinforce unwarranted certainty can result in an unwillingness to challenge or modify beliefs about the world.”[iii]
Said another way, “we’ve always done it this way, therefore this is the right way to do it.”
Gavin lays out a History Checklist of twelve Questions. All of them can be applied in business, but I’m going to focus on four.
- How did we get here? (Vertical history)
- What else is happening? (Horizontal history)
- Was this inevitable? (Outcome bias)
- Are we using the past correctly? (Historical malpractice)
And within Historical malpractice, “facile historical analogies,” which is the idea that as much we believe we fully understand the past event, this is almost never the case.
When something goes wrong, we use a variety of RCA techniques (Root Cause Analysis) to try to find the root cause. The Five Whys is one technique, you may be using another. I see Five Whys as a Vertical History approach, which is fine by itself.
What if we bring in some or all of the other Questions? Some folks will argue those kinds of questions only delay or confuse the RCA team, but I disagree.
Suppose we find that a part drawing had an error in the size of a hole, resulting in a hole that was slightly larger than intended. The mating shaft was the correct size, and the delta was very small, so the assembly line had no reason to think there was a problem. Both parts meet spec, right?
The assembly gets out into the world, and a wobble develops which causes all sorts of problems. (Go with me, it’s just an example.)
If we use only the Vertical History approach, we will fix the drawing, rework the field, and declare victory. Let’s go further, and look at what else was happening.
- Did a new document review process get implemented?
- Was an ECO review meeting cancelled / rescheduled the week the drawing was released?
- Has this happened before? If yes, did we do anything about it?
- Are we using an AI-assist evaluation tool? If yes, did it flag a problem?
- Is there something in our Institutional Knowledge which might help us understand why the assembly line did not, or would not, flag this as a problem?
Use the context of the problem to help you expand the scope of the RCA…within reason, of course.
Everyone does failure analysis a little bit differently. But I believe using Professor Gavin’s History Checklist will yield a stronger result, and urge you to try it.
[i] Quotations from Taylor, A. J. P. | WIST Quotations
[ii] https://www.mckinsey.com/featured-insights/annual-book-recommendations
[iii] Francis J. Gavin, Thinking Historically, a Guide to Statecraft and Strategy, page 53