Is History a Science?

Most history books read like narratives. This battle happened after that battle. Because king A did not like king B, A invaded the country of B. And yet there seems to be a mechanism for history. Jared Diamond in Guns, Germs, and Steel [GGS] quotes approvingly the German statesman Bismarck as saying: "The statesman's task is to hear God's footsteps marching through history and to try to catch on His coattails as He marches past" (p. 420).

I will paraphrase that statement as “leaders do not really lead, they sense what is coming and/or what the people want and then the leaders express it by word and action.” For example, it is widely believed that L. B. Johnson supported civil rights because he thought that was necessary for someone from the South to get national acceptance.

Often it is hard to give proper credit to leaders. Alexander is called the Great but his father Phillip had prepared the Macedonian war machine and he was ready to invade Persia when he was assassinated.

Diamond (pp. 419-420) offers a teaser: Hitler was almost killed in a car accident in 1930 . If he had died how would that have changed history? Diamond thinks that it would but several historians have examined the impact of Hitler’s person on history and based on what I have read, the answer to the question is “not that much.” An anti-Semitic party was inevitable. The treaty of Versailles had made WW-II inevitable. After all Hitler had a lot of rivals for the leadership of the Nazi party and was even accused by an Austrian writer that he had plagiarized his work. German military leaders were eager to rearm. On the other hand Hitler read the French correctly when he re-militarized the Rhineland and was able to trick Chamberlain. Another Nazi leader may not have been as adroit. Of course such a leader may not have attacked Russia and Nazi Germany would have stayed undefeated in Europe.

Diamond [ibid] devotes a long epilogue on the challenges of studying history as a science. The main challenge is that complex dynamic systems are chaotic, namely they are determinist and at the same time unpredictable. (Diamond does not use the term "chaotic" but he describes historical process as being both deterministic and unpredictable.) The best known example of a chaotic system is the physical process responsible for the weather. We can make broad predictions but not precise ones. In the same way we can predict that a country that runs large budget deficits will eventually be unable to pay its bond-holders but we cannot say when exactly this is going to happen.

So history is a science, not as precise as particle physics but as messy as the study of the weather. I use the term "undercurrent" to refer to all these processes that affect historical events, even if the connection is not always obvious. This is analogous to the effects of the Gulf Stream on the weather of both eastern North America and western Europe.

[GGS] demonstrates that environmental conditions such as suitability of local flora and fauna for domestication and the attractiveness of hunting-gathering determine when a human society committed itself to food production (agriculture, domesticated animals). That in turn is a good predictor of development. Environmental and geographical factors explain why China (food production started around 7500 BCE) and the countries around the Mediterranean (food production started in various locations between 8500 BCE and 6000 BCE) had a head start. In contrast food production in the Americas did not start until 3500 BCE. As a result American Indians had no chance against the European invaders (or "discoverers").

However, those broad factors do not explain the relative performance of different "advanced" regions and why the role of the dominant power changes over time. A question that is frequently asked is why the American continents were "discovered" by Europeans rather than by Chinese explorers, even though China had more advanced ship building than Europe.

Another question is why Western Europe moved so far ahead of the rest of the world and in particular of the Middle East, even though until around 1200 the Middle East was more advanced. Gibbon makes an interesting observation: "The most civilized nations of modern Europe issued from the woods of Germany, and in the rude institutions of these barbarians we may still distinguish the original principles of our laws and manners" [EG, Chapter 9, p. 208]. This is a paradox: the backward people overcame the more advanced.

One factor that can be discounted is differences in the intelligence of people in different societies. [GGS] is emphatic on this point and I can add one more argument. Modern times have seen immigration by individuals in large scale and people from "backward" countries can be very successful once they move in an "advanced" country. One particularly striking phenomenon is that scientists and engineers that immigrated to the United States as adults seem to dominate American academia. A cursory look at the backgrounds of the tenure track faculty in my former academic department reveals that two thirds are immigrants who received at least one of their degrees elsewhere (mostly in India or China). Therefore some countries have fallen behind not because they lack smart people.

The "brain drain" factor is often brought up but the fact is that, at least for Greece, most of the successful Greek people left Greece out of frustration and they are not welcome back to Greece. I have heard similar stories from immigrants from other parts of the world.

When we look at the flow of history in modern times the environmental factors mentioned in [GGS] are no longer at play and the intelligence of the people in a country does not seem to be a factor because there are no significant differences from country to country. In an Afterward [GGS] points out the importance of institutions and also that "institutions are not born out of thin air." The last few sentences of that book point to the prosperity of countries like Japan and South Korea compared to the poverty of New Guinea and the Philippines as explainable by the much earlier start of food production in the first group. However, that explanation fails in the case of Europe and the Middle East. 3000 years ago Iraq and Egypt were far more advanced than England and Germany while the opposite is true now.

What we are left with are societal dynamics and the possibility that an organization that is beneficial at one time becomes detrimental later on. Those of you who remember their high school physics may find the following example illustrative. A pendulum moves between two high points A and B while passing through a center C. From A to C gravity acts as an accelerant while between C and B gravity slows the pendulum down. Think of the tendency to introduce control of the people by the leaders as an analogue to gravity. Introducing order in a disorderly society helps prosperity. But as control increases it becomes counterproductive.

[GGS] (and others) distinguish four levels of human organization: bands, tribes, chiefdoms, and states. Bands usually contain about 100 people, members of an extended family. At the other end states are entities in excess of 50,000 people with leaders, priests, bureaucrats and the like. However, lumping all states together misses a lot of distinctions that are very important in the history of the world during the last 3000 years. The big question is whether there is a scientific way to describe the evolution of a state. States differ significantly from each other in longevity. The Roman Empire lasted over 1000 years (1700 if we include the years of the Roman republic and the years from the battle of Manzikert in 1070 to the fall of Constantinople in 1453), the British state has lasted so far almost 1000 years (using the battle of Hastings in 1066 as its start), and the Ottoman Empire lasted a little over 600 years (from 1300 to the end of WW I). At the other end, the Soviet state lasted less than 100 years. Can we predict when a state is going to disintegrate? How important is its size?

A prime example of history as science can be found in another part of Gibbon's work [EG, Chapter 5, vol. 1, pp. 101-102] that deals with large states. Gibbon presents an argument about the size of the state needed for an oppressive regime. He starts by citing studies that show that it is hard to maintain a standing armed force that is more than 1% of the population of a state. He then points out that a single person cannot terrorize 100 others and even 100 armed men cannot terrorize 10,000 people. But 10,000 soldiers can terrorize a million people. That was roughly the size of the Praetorian Guard in Rome that took effective control of the state after the death of Commodus. If we accept Gibbon's argument we have to distinguish between small states of a few hundred thousand people and states of a million or more people. Of course not all large states become oppressive and we will search for the reasons. All modern states exceed that critical population size while many of the ancient states did not. One of the largest ancient states the Achaemenid Persian Empire is estimated to have had around 50 million people while the Roman Empire at its largest extent had over 70 million people. These estimates are debatable but even if we assume that the population was only a quarter of these estimates, it was still large enough to enable oppressive regimes.

The uncivilized barbarians that overran the Western Roman Empire came from small nomadic states that had developed no means of systematic oppression of their citizens. That may be one reason why the states of Western Europe were able not only to defeat the Roman armies, but also to advance much farther ahead than the Eastern Roman Empire.

A recent book by Diamond, Collapse [COLL], discusses the evolution of societies with focus on environmental issues. He (and other authors) suggest that the success of the "barbarians" against the Roman Empire may have been the results of environmental factors. This might be true but I think there were other factors at play, such the governance of the empire just mentioned. Later I will try to carry Gibbon's methodology to the later eras that are not covered by his monumental work. One reason to downplay environmental factors is the many examples of a population persisting through different rulers. The mighty Persian Empire was conquered by Alexander and the Hellenistic kingdoms flourished for three centuries in its former lands. Then came the turn of Roman rule, and then that of Ottoman rule, all in the same lands.

When we look for historical forces we must certainly include environmental factors. But we should also include societal dynamics, in particular the effects of central control as I mentioned in the pendulum example above.

[GGS] Jared Diamond Guns, Germs, and Steel, Norton, 1997-2005.
[COLL] Jared Diamond Collapse, Viking, 2005.
[EG] Edward Gibbon, The Decline and Fall of the Roman Empire, first published in 1788.Note: I use the 1978 reprint of the 1910 Everyman's Library (Dutton: New York) unabridged edition with comments by Oliphant Smeaton.

Back to the index