History Is the Discipline That Decides Which Explanation Mattered

In his world history course, Yuval Noah Harari opens with the question that every field should ask and few do: what is this subject, and how is it different from its neighbours? His answer for history is worth restating, and then I want to add the part I think he leaves out.

History Is a Social Science

Harari's first move is deflationary. History studies the change of human societies over time, in every aspect: their economies, beliefs, institutions, arts, wars. Put that way it is simply one of the social sciences, alongside economics, psychology, sociology, political science and anthropology. It has no separate subject matter. It looks at the same people the other disciplines look at.

This is a useful correction to the way history is usually taught, as a store of dates and kings that sits apart from the analytical disciplines. It is not apart. It is one of them, with the same evidential problems and the same obligation to explain rather than merely narrate.

Two Things That Make It Different

If history has no subject of its own, what distinguishes it? Harari names two features.

Breadth. A historian explaining an event has to take account of economics, psychology, sociology, politics, anthropology and the history of ideas all at once, because the event was produced by all of them at once. No other social science asks this of its practitioners; an economist is allowed to hold the psychology fixed. The cost of that breadth is depth. History gives the shallowest explanation of any of the social sciences, in the specific sense that it cannot go as deep into any single mechanism as the specialist discipline can. That is not a flaw. It is the price of looking at the whole.

Change. The other disciplines emphasize continuity. An economist explaining a famine in the fifteenth century and a famine in the twentieth will reach for the same models, because the models are supposed to be about people in general. A historian's instinct runs the other way: the assumptions underneath those models, about what people want, what they know, what institutions exist to channel them, change over time, and a model that ignores the change is applying twentieth-century people to a fifteenth-century world. History does not claim that everything changes. It claims that change is the thing to watch for, where the others watch for the constant.

What Harari Leaves Out

Both points are right and I would add a third, which I think is the actual job.

The specialist disciplines each supply a mechanism. Economics can tell you how a trade collapse propagates. Psychology can tell you how a humiliated population behaves. Political science can tell you what an unstable coalition does under pressure. Each is genuinely explanatory within its domain, and each will, if asked, explain the whole event in its own terms, because that is what a discipline does.

Ask an economist how important the economic factor was in the outbreak of the Second World War and the answer will be, in some form, very. Ask a political scientist about the political factor and you will get the same answer about a different factor. Every discipline finds its own cause to be the important one, and none of them is lying. They are each looking through the one lens they own.

The job of history is to resolve which of the mechanisms actually did the work in this case. Not to supply another mechanism, but to weigh the ones on offer against each other and against the record, and say: here, the economic collapse mattered more than the ideology; there, the reverse. That is a question no specialist discipline can answer, because answering it requires standing outside all of them.

Why This Is Hard, and Why It Is Worth Doing

It is hard for the reason Harari gives: the historian is shallower in each mechanism than the specialist, and is nevertheless being asked to rank the specialists' explanations. It is like asking a general practitioner to adjudicate between the cardiologist and the neurologist. The GP knows less than either about their organ. The GP is also the only one in the room who is looking at the patient.

And it is worth doing because the weighting is the part that transfers. A mechanism, once understood, is understood forever; you do not need history to tell you how inflation works. What you need history for is the knowledge that in this kind of situation, the inflation mattered and the speeches did not, or the reverse, and that knowledge is exactly what a decision-maker facing the next such situation needs and cannot get from any single discipline.

The Failure Mode

When history forgets this job it degenerates in one of two directions. It becomes narrative, a well-told sequence of events with no claim about which causes carried the weight, which is entertainment. Or it captures itself to one of the specialist disciplines and becomes economic history, or the history of ideas, each of which has already decided in advance which lens is the right one.

Both are respectable. Neither is doing the thing only history can do. The reason to have a discipline with no mechanism of its own is that somebody has to referee the disciplines that have one, and the referee cannot also be a player.