About Rationally Speaking

Rationally Speaking is a blog maintained by Prof. Massimo Pigliucci, a philosopher at the City University of New York. The blog reflects the Enlightenment figure Marquis de Condorcet's idea of what a public intellectual (yes, we know, that's such a bad word) ought to be: someone who devotes himself to "the tracking down of prejudices in the hiding places where priests, the schools, the government, and all long-established institutions had gathered and protected them." You're welcome. Please notice that the contents of this blog can be reprinted under the standard Creative Commons license.

Monday, February 01, 2010

Podcast Teaser: Can history be a science?

Well, we did it! The first episode of Rationally Speaking, the podcast, is out and available both directly from our New York City Skeptics-sponsored web site and from the iTunes store. The second episode will come out in time for Valentine’s Day, and it will focus on the science and philosophy of love. For our third podcast we will have our first live guest, Prof. Peter Turchin from the University of Connecticut. Peter is a biologist by training, with interests ranging from theoretical ecology to population biology to biostatistics. In particular, much of his work has focused on what determines population cycles, a problem to which he has applied an array of statistical and conceptual tools, including chaos theory.
More recently though, Peter has made headlines in Nature magazine for work at the borderlines between science and of all things, history. He has published three books on the topic: Historical Dynamics: Why States Rise and Fall (Princeton University Press, 2003), War and Peace and War: The Life Cycles of Imperial Nations (Pi Press, 2006), and Secular Cycles (co-authored with S.A. Nefedov, Princeton University Press, 2009).
In War and Peace and War Peter argues that empires (his examples include the Roman and Russian empires as well as the United States) form and collapse because of continuous cycles of cooperation and conflict. Initially, humans band together and cooperate in the face of common enemies or other challenges, but then increasing prosperity is not equally distributed, with rich people becoming disproportionately rich. This causes conflict, and the consequent breakdown of cooperation, which in turn leads to collapse.
Of course, Peter is not the only scientist to have turned to history in an attempt to make that field more scientific, Jared Diamond’s Guns, Germs, and Steel and Collapse immediately come to mind (interestingly, unlike Turchin, Diamond argues that the collapse of civilization is the result of environmental disasters, not of internally generated social disruptions). And naturally, many historians vehemently object to what they perceive as a crude scientistic attempt at interdisciplinary colonization.
Which raises the question of the podcast, and about which Julia and I would like to hear your opinion: can history be studied and understood in a scientific manner? Are there patterns and logic(s) to history, or is it, as the saying goes, just one damn thing after another? For that matter, how do we determine where the demarcation line — as Karl Popper famously called it — between science and non-science falls, and why?


  1. I got into a very active debate on this one lately, as I tried to advocate that history-as-science should be doable in principle with standards equally as rigorous as the historical science of evolutionary biology, but that history just hadn't matured and gone through the equivalent purging process as the modern evolutionary synthesis yet. Then a colleague put me straight on why this would never be, with the comment

    "The difference between history and evolutionary biology is that the organisms studied in history make active attempts to manipulate the data."

  2. This comment has been removed by the author.

  3. I take a pretty broad and vaguely defined view of what constitutes science because it seems that any time one tries to apply precise definitions, it inevitably leads to absurdities. That is, very legitimate fields of scientific inquiry don't meet the definition, as when Popper claimed that natural selection wasn't scientific since it was tautological (although I believe he later recanted). From that perspective, I think we'd have to say that history can be studied scientifically, if for no other reason than that by not counting history as (potentially) science, we would have to say that many other fields (historical biogeography, paleontology, some evolutionary biology and ecology, etc.) are not science either. And since people in those fields work in science departments, they must be scientists, right? (Kind of kidding) I obviously I am not offering the most rigorous defense of why some studies of history are scientific, but it seems sensible to me to let science be a big tent, inviting of all who embrace testing their ideas about the natural world with data.

  4. "Are there patterns and logic(s) to history, or is it, as the saying goes, just one damn thing after another?"

    Oh you know full well it's a spectrum! :-) But of course we can't deal with history in a rigid scientific manner. I'm a mere undergrad, but prima facie...that seems impossible.

    One can argue till one is blue in the face about it, but if we all pretty much agree that Hume has made a case for a majority of our knowledge base about the present being predicated upon induction, what hope has the past?

    I'm definitely interested in seeing opposing viewpoints though.

  5. Yes, Jared Diamond comes to my mind also. As does Karl Marx, by the way.

    It may also be the case that several possible reasons can lead to the collapse of civilizations, internal ones and external ones, just like an organism can die from cancer, hunger or predation. (The parallels would of course be collapse of the social structure, environmental disaster and Mongol hordes, respectively.) And they can also fall together, such as an animal already weakened by an illness being killed by a predator. It is always helpful to be able to think in terms of multi-causal systems.

    In general, I would argue the same as I did in a previous discussion here on this blog: all that exists is subject to science; but that what science can tell us about an issue might not be all that would be interesting to learn about it.

  6. I think that the discipline in historical studies consists of justifying your assertions with primary sources, and for that reason history is pretty much immersed in contingency. This is not to say that occasional regularities may not be observed, but even when they are they are cut across by so many one-time-only circumstances that one has to be careful generalizing. Having said that I refer back to my first sentence. The historian is bound to justify his statements one-by-one, and to expose them to peer review, so to that extent history may be regarded as science.

  7. First of all, let's be clear. We are talking specifically about human history, yes? Moreover, we are primarily focused on populations, as opposed to individuals.

    Most of the difficulty with examining history scientifically comes from the dual facts that initial conditions cannot be controlled and experiments cannot be repeated. However, it's not a hopeless endeavor. In GG&S Diamond did a very, very good job of planting a scientific flag on history's shores by pairing sound anthropological data about Polynesia with the archaeological, geological, and biological records. He also drew upon the historical record in some small measure, but only inasmuch as it confirmed the others. Hence, no question-begging on that front. The thrust of his approach was to treat the human colonization of thousands of islands across Polynesia within a relatively short time frame by a single parent population with a stable culture as a series of experiments with negligibly differing initial conditions. By examining the cultural characteristics that resulted in the light of the geological and biological conditions they were subject to, he was able to draw very strong correlations between these factors. He showed that in many instances, other than some low-level linguistic similarities and the like, after a couple dozen generations the resulting cultures bore little resemblance to the parent culture they stemmed from, but in many instances bore striking similarities to other cultural children with whom they had no contact, but shared geological and biological resources.

    Hypothesizing on this basis that correlation might be a primary causative factor, he then compiled a list of rules governing which effects certain types of geological and biological conditions, singly and/or collectively, might induce which cultural characteristics. Having built his model, he proceeded to test it by identifying other localities around the world which were subject to similar sets of specific conditions, and examining the human cultures that existed there in light of them. He found repeatedly that the broad trends he had identified in one place could be seen elsewhere as well, and thus was able to further decouple cultural particulars from cultural precedents, given a similar amount of time for adaptation to current conditions. In this way, Diamond very effectively used a scientific approach to delineate broad trends about human societies from a sufficiently large sample size.

    The tricky part, then, is finding other historical patterns about which we have sufficient scientific evidence to conduct similar analyses. Provided, however, we are able to do so (and there is ample reason to believe there is), it is not absurd to suppose that we can establish some strong ground rules and build a more sophisticated predictive model that may be applied effectively even when examining more "messy" data. At that level, provided we are able to establish sound filtering rules about what does not contribute significantly to the shape of a culture via empirical testing, deducing more specific rules then becomes a matter of cleaning the data, i.e. signal processing.

    Of course, we would quickly run up against chaotic factors that defy predictive models, but this is no barrier to studying turbulent fluid dynamics scientifically, so neither should it stymie studying history scientifically.

  8. Ugh, I'm not happy about my wording on a few fronts. Bad editing.

    Last sentence of 2nd paragraph, reprise:

    He showed that in many instances, other than some low-level linguistic similarities and the like, after a couple dozen generations the resulting cultures bore little resemblance to the parent culture they stemmed from, but bore striking similarities to other cultural siblings with whom they had no contact but which had similar geological and biological conditions.

    1st sentence of 3rd paragraph, reprise:

    Hypothesizing on this basis that correlation might denote a primary causative factor, he then compiled a list of rules governing which effects certain types of geological and biological conditions, singly and/or collectively might induce.

    Last sentence of last paragraph, reprise:

    Of course, we would quickly run up against chaotic factors that defy predictive models, but this does not halt studying turbulent fluid dynamics scientifically or developing better models as a result, so neither should it stymie studying history scientifically.

    Again, ugh. Sorry for the poor wording and the clunky fix.

  9. Another great topic ! You might be able to accomplish the unthinkable and replace the SGU as my favorite podcast !

    On to the topic

    In the post you link to you contrast people with subatomic particles.I happen to be a physicist and i am also a human so i happen to know a little something about both and i think you happened to hit on a excellent analogy but in order to use it , you ll have to turn it on its head first...

    First of all the world of subatomic particles is nothing but tidy.Its a mess and as you might have heard ,non deterministic ! That is the essence of the analogy and to me , the real question at hand.

    Arguing that the study of history cannot be a scientific endeavor is essentially arguing that individuals , groups of people , local societies and ultimately civilizations operate in a non deterministic fashion.This is obviously not the case.Gross patterns can be identified even by non experts such as my self.A civilization and its culture are an "emergent property" of a group of people interacting with each other.The outcome is determined by the nature of the individuals that compose the group but its practically impossible to predict its form and properties just by studying the properties of the individual.The opposite is also true.

    To make the analogy with physics.

    Everyday objects are made out of subatomic particles and they have a certain variety of forms and properties which we describe in certain terms.Over the previous century we advanced to the point where we can study the particles themselves and we ve discovered that not only they dont vary as much as the macroscopic objects which they form but also that they cant even be described by the terms that we use in everyday life and by which we describe"real world" objects and most importantly , their behavior is fundamentally non deterministic ! Even though that has practically be proven to be a fact , what is also known is that the macroscopic properties that we can measure and even feel with our own senses such as phase,color,conductivity etc are consistent , measurable and can be derived (deterministically) by the very different properties of the subatomic particles by which they are composed.Another implication of quantum physics is that macroscopic objects are obligated to act deterministically by the same rules that obligate subatomic particles to act in the exact opposite way!

    I am hoping by now the analogy has been made clear.You cant predict the projection of an electron's spin on an axis just like you cant predict whether X will study this or that subject or will vote for this or that party.But its irrelevant.His or her actions will have little effect on the culture.The civilization is an entity of its own with unique properties and rules that govern its progression and evolution.Its just an issue of identifying the variables that need to be measured and the quantities that need to be defined.

    Many people would object to my view probably by saying that this is just an analogy and nothing more.The opposite might very well be true.Humans arent electrons.One man might make a difference , a single idea can change the world etc.All these things are true!It might very well be that human behavior is non deterministic and that cultures arent.It may be that cultures and people are both non deterministic.But again its irrelevant.If a quantum quantity cannot possibly be predicted and the only thing we have is the probabilities then so be it ! The knowledge of these probabilities allows us to make transistors and allows me to make this post ! What if we knew its homologous probability of say a nation going to war with another one or this or that moral concept being adopted or rejected ? Who knows how empowering and enlightening this could be ?Surely you cant predict transistors from quantum mechanics ! (or can you... ;)

    Continued in part II !

  10. Another objection could be that , yes surely societies are made of people who are macroscopic objects and are therefore deterministic and their interactions create the culture and cause all the events that we want to describe or predict but its just too complicated ! Chaotic behavior , non linearities , prohibitively large amounts of data to be collected and processed make this a futile cause.Maybe.But the same can be said for the brain.Unthinkably large number of neurons and connections that interact sometimes in very complex ways.Even the emergent property of the brain (the mind) is too complex to study so we resort to statistical and phenomenological techniques of describing it.But still who could argue that the study of neurons ,neuronal networks, and attempts to chart and simulate the connections are not worthwhile ? We already have some results.Who knows what form the "final" model will have ? How predictive or how detailed or how "deterministic" it will be ? How is history any different?Maybe we cant know much but undoubtedly we can know much more.Who knows what kinds of events can be predicted , or with what error , or how far into the future ? Maybe theres much that cant ever be known but undoubtedly theres much that can.

    I also believe that history isnt being studied scientifically for that exact reason.People think that its not possible.This is just a cultural meme(How would memetics play into this scientific study of science? Interesting question).It related to the idea that emotion and science dont mix.That there are well defined domains and that science and history belong to different ones.We have to do some consciousness raising.I cant wait to hear Peter Turchin talking about this.

    Sorry for my rant but i dont think you can address issues like that in "tweets"

    Keep up the good work !

  11. I'll say this much: Inasmuch as reason (as opposed to raw emotion) played a role in my parting with religious fundamentalism, the natural sciences played a minor role in relation to history (or, more precisely, the modern historical-critical method of studying religious texts and their claims).

  12. This comment has been removed by the author.

  13. Considering that I hope to study History at graduate school this topic is rather close to my interests. I've always believed that science can contribute immensely to historical inquiry - especially if we adapt approaches similar to those described by perspicio (and implemented by Diamond). The possibility of using scientific methodology to develop models that can help delineate broad trends in civilizations across history is nothing short of exciting. I'd like to hear some of the objections to such interdisciplinary endeavors, however...

  14. "can history be studied and understood in a scientific manner?"

    I guess that depends on what you mean by "scientific manner?"

    Historiography has certainly added a more scientific bent to itself in the last decade than it ever had before. You see much more statistical analysis today than you would have ever seen in the 1970's (see "Beyond the Founders: New Approaches to the Political History of the Early American Republic" as an example.)

    It must be recognized though that history will never reach the level of testability that classical physics, chemistry or molecular biology has reached. You can't create a duplicate Runnymede and interview King John. But that, I think, can be taken in stride. Cosmologists run into the same (at least very similar) problem, they have models that work for the data they do have available and sometimes must extrapolate them beyond where I'm sure they would like to. The cosmologists have learned to live with this, why can't the historians?

    As far as professionals in both science and history taking umbrage to the idea, I don't understand why. Back when I was in school, they were both members of "the knowledge arts." Has one or the other group withdrawn from this classification?

  15. This may be an aside but Roy asks, "Unintentional consequences of purposely leaving Africa?"

    We left Africa purposely? Really? At the time it was not "Africa" and I cannot imagine one of our very early predecessors saying, "Hey, let's get out of here. Italy is so much nicer."

  16. Yea, assuming the fixity of the past, the subject of history, it seems like it should be capable of being studied in a science-like manner.

    It seems analagous to trying to discern ancestral lineage of human beings, or any species for that matter. Do we consider that a "science"? It is certainly informed by science; carbon dating, for instance. The more pieces of that history we discover, fossils, the better we can build accurate theories of the lineage.

    A more general history of societies and peoples brings in problems that some have hit on above, but it seems we can definitely form more and less acccurate theories of history, just like anthropologists and others do for "scientific" and evolutionary histories.

    Furthermore, history is going to have similar problems that the social sciences and psychology (especially in its study of the interaction between an individual and the everchanging environment) have today in trying to make their pursuits "scientific." History has similar problems with the added difficulty of not relying on current subjects and peoples.

  17. This comment has been removed by the author.

  18. I also remembered that structuralism strived for understanding structures and patterns in history. I guess I am thinking of Pierre Vernant, Fernand Braudel, Foucault, and maybe Levi-Strauss.

    I've read some of Pierre Vernant's works who tried to understand Greek societies and individuals through their mythical framework. Also he tried to explain how rationality was a product of these myths. He argued for necessary structures annd patterns that led to the Greeks developing rationality. It seems like one could develop a scientific approach through this system, for instance, testing scientifcically to understand how rationality arises in an individual, and then using such psychological analysis to understand how certain strands of rationality arose in Greek societies, thus, retelling an historical development. I found Vernant's attempts strained though and incomplete. Although a novel approach with good insight into Greek society, I do question whether it and similar approaches (or any approaches) could lead to a scientific historical analysis. Similar problems also arise in Foucault's early historical works.

    Maybe many of the problems that structuralism ran into are similar to the problems that one would have in developing a more scientific account of history.

  19. Ok, I have chimed in on this topic before. I don't have much time but I want to remind people of some things.

    First, Diamond is a secondary source, parascitic and often mis-representative of those who did the primary work. He is often wrong and an over-simplifier of the topics he addresses.

    Scientific study of history is old news as far as I am concerned, as it has been done by archaeologists since at least the mid 1960s with the begginings of "processual archaeology".

    Perhaps those who are expecting a more "scientific history" are not respecting some of the results, which is there is a lot of contingency in human history.

  20. Is history just "one damn thing after another"? Well, as Nassim Taleb might say, it is mainly one (completely unpredictable) Black Swan after another - it's the
    Black Swans that have the major impact on the future. Historians are pretty good at finding plausible after-the-fact explanations of why things happened (that's basically their job!) but when some of them venture to extrapolate to the future, they nearly always miss the big factors, the Black Swans, such as WWI, the collapse of the Soviet Union, the housing bubble, 9/11, etc. etc. Those Black Swans are the events that really change the course of history.

  21. Sheldon, I agree with you re: Diamond as a secondary source. (Not that my agreement means anything, or needs to. He *is* a secondary source.) Nor would I go so far as to say that every conclusion he drew in GG&S was 100% accurate. (And neither does he.) As for oversimplification, he pretty explicitly states that very fact himself. But as far as scientific methodology goes, broadly speaking he did a very good job of demonstrating its efficacy in and pertinence to the study of history.

    It's also true that many scientists have made positive strides in related endeavors, often in much more elaborate depth, both before and after his two famous books. That's a good thing, right?

  22. How about cultural history and evolutionary psychology? They are both relatively young fields (with a lot of kinks to work out) and they seem like a good fit.

  23. Perspicio,

    Yes, Diamond does a lot of things well, he is a popularizer, and there is nothing wrong with that.

    What set me off, and this has happened before, is that Massimo seems to refer to him as some great trailblazer of the science of human history when he isn't. He just gets the credit here because he is who Massimo is familiar with.

    (I do cut Massimo slack because he has two Ph.D. in different fields after all, and so is master of those, and so he can't know everything).

    What Diamond says has been said before by those closer to the data. But the more recent primary literature paints a much more complex, variable and rich history. So Diamond gets a lot wrong. I am specifically thinking about the treatment of the U.S. Southwest in "Collapse". But I both know and suspect that the same is true of his other explanation/stories.

  24. Since I was being naively philosophical in my last post, I'll bend toward the empirical.

    So those who are pro the assertion that human history can be studied in a 'scientific manner' the same way biology, or physics can be, are perfectly comfortable with zero observation? Zilch.

    And who was it that practically gave Jared Diamond the award for marrying correlation and causation in his book? "Hypothesizing on this basis that correlation might be a primary causative factor..." At least you said he was hypothesizing. But your phrasing doesn't have enough ambiguity for you to wiggle around the endorsement of his work as concrete.

    Maybe the question is not whether human history can be studied scientifically, but whether our tendency toward interpretation has all of a sudden stopped with the modern era.

    Do you really think artifacts (better), or primary documents as a whole (worse) don't bear the weight of human perspective?

    Surely language and DNA are our best tools for studying human history and not haphazardly guessing at events outside of direct observation (aside from Pompeii, or some other direct "interaction" with catastrophe).

  25. phiwilli:

    I think you are completely missing the point of history-from-a-scientific-perspective.

    Firstly, making predictions about the future is nowhere in the job description of a historian, just like an evolutionary biologist is not tasked to speculate how organisms will look like in 100 mio years, or what novel organs they will evolve.

    Secondly, the issue is not which idiot king tries to conquer whom, but much larger scale questions like why did the Inka empire not have muskets in 1500 C.E., while the Europeans had, or why do civilizations collapse.

    And seriously, the housing bubble? You vastly overestimate the historical significance of our current daily affairs, it seems. This will at most be a curious footnote for the historians of 2500 C.E. while they try to figure out the factors behind the rise of religious fundamentalism, the transformation of societies through electronic communications, or the failure of the world to act on global warming, population growth and species extinction.

  26. Loved the podcast,

    To Massimo: Perhaps you could elaborate more on what you referred to as your "Moral Obligation". I assume that you were pointing towards using some sort of utilitarianism as the means of determining why you would present someone with the "Red Pill" as you said, I think it would be a great topic to go over.

    As for history, I think that there should be a separation of the facts (that which occurred, i.e. Roman Empire collapse, Russian Empire collapse) and the interpretation of the facts (what factors lead to the occurrences).

    I do think that it would be difficult to apply science to History though, just because I could imagine that the most parsimonious historical interpretation may not always necessarily be the correct answer, sometimes unlikely events may be the cause of historical facts. (assuming that parsimonious explanations don't always work in history)

  27. This comment has been removed by the author.

  28. Mintman,

    I agree with most of your comments, but to explain myself more clearly: I assumed (perhaps wrongly) that Massimo's query, "are there patterns and logic to history?" implied that whatever patterns or logic there might be will be continued into the future, so they would be implicitly predictive (not just retrodictive). And I was expressing skepticism about that, because, as I see it, major trends and patterns in the future are too susceptible to Black Swans such as wars, pandemics, earthquakes, asteroid impacts, economic disasters, etc.

    So I agree with you that prediction is not what historians do (or should do).

    As for Massimo's query about Popper's demarcation (what is unfalsifiable isn't scientific), seems to me that what historians do clearly falls on the science side of Popper's line; their claims, either about what happened or about why it happened, are falsifiable - by possible findings concerning documents, artifacts, monuments, et al. This in no way demeans those who write "narrative histories," eg. Shelby Foote's "The Civil War: A Narrative." Good story, fascinating reading - and his claims are falsifiable (but few of them have been falsified, although maybe some have).

  29. I haven't gotten a chance to listen to the podcast yet, but since I don't see it mentioned, I'll highly recommend this issue of Skeptic Magazine with the special section: "Can History Be Science?"


    Most notable I think is Shermer's piece: "History at the Crossroads".

    Here's some quotes from that piece.

    ~"The NSF (National Science Foundation) panel turned him down flat. Testing historical hypotheses? Using science to analyze history? Was he (Frank Sulloway) joking.* They rendered their judgment in no uncertain terms:

    "One of the most pervasive issues discussed by the panelists was the approach the Principal Investigator was taking toward history. Many panelists thought that applying a heavy-duty statistical analysis to history is naive, inappropriate, and even peculiar. Is it really the case that generalizations in history should be tested with statistics, rather than be tested through a detailed examination of the sources? Some noted that it seemed as if the Principal Investigator was going back to 19th-century beliefs that history is a science which could uncover laws. Panelists were opposed to such a narrow view of history."

    Sulloway was shocked. How could a panel of scientists think that using statistics to test hypotheses is “naive, inappropriate, and even peculiar”? Then he remembered, these were not scientists. They were historians and philosophers. “Besides being an odd response to receive from the National Science Foundation,” Sulloway recalled with some dismay, “where the principal criterion of grant evaluation is supposed to be ‘scientific merit’, this panel’s criticisms confuse a method of research (hypothesis testing) with a theory of history. Testing is what makes an approach scientific, not the particular viewpoint that is endorsed.” And to emphasize the point, Sulloway pointed out that “Even the claim that history can be studied scientifically can only be evaluated properly through hypothesis testing” (1996,p. 539).

    ..."To resolve the apparent paradox we need to understand the history of the profession of history, how and why historians came to reject science, and how Sulloway and others are working to create a science of history. Can history be a science.* I submit that it cannot afford not to be." ~

  30. Mintman: "And seriously, the housing bubble? You vastly overestimate the historical significance of our current daily affairs, it seems. This will at most be a curious footnote for the historians of 2500 C.E. while they try to figure out the factors behind the rise of religious fundamentalism, the transformation of societies through electronic communications, or the failure of the world to act on global warming, population growth and species extinction."

    So what? Both scientists and historians both study the behavior of "insignificant" events all the time because they can make generalizations about them. The housing bubble is informative precisely because it resembles other bubbles; the behavior of a single electron is interesting because it resembles the behavior of other electrons. Furthermore, I think there's a good chance that the housing bubble will be much more than a footnote, because there will probably be more recorded evidence about it than any major bubble in history until now.

    I'd also like to hear more about how history, as currently practiced, is supposed to be "unscientific." From the above description of his work, I don't see how Diamond's research was that different from what most historians do.

    Frankly I don't really see that much difference between science and history in the end, apart from one key point made at the very beginning of the thread: "The difference between history and evolutionary biology is that the organisms studied in history make active attempts to manipulate the data." Every historical study must face the fact that much if not all of the evidence available about many events -- especially those of large historical interest -- was generated by biased individuals. It's like trying to study an organism that systematically edits or destroys its own fossil record. It doesn't mean we can't draw some conclusions about the past, but it does complicate things greatly.

  31. This comment has been removed by the author.

  32. Well, I decided to read Guns, Germs and Steel, and I'm almost finished. I find the book quite impressive. Of course it relies on certain kinds of data- archaeological, linguistic, genetic, radiometric etc. Nothing wrong with that. These data are the kind of data that lend themselves to bulk manipulation, and are good for developing and testing hypotheses; and to the extent that they become available, history will become more scientific. But in areas of history where the researcher is forced to rely on documents, it will be business as usual... primary sources, and trying to interpret the reasons behind them. This goes back to the idea of ultimate causes, proximate causes etc. I don't think there is necessarily any conflict, it's just that the new kinds of usable data can now be brought to bear on human concerns, and this should be welcome.

  33. Scott:

    Nonetheless, the generalities of bubbles seem to be more of an issue for economists.


    Except that the difference is with the biologists who can't think out of the orthodox box that tells them their average organisms can't manipulate their destinies.

    As a botanist, I am pretty sure that my plants do not have the cognitive capabilities that would be required to call it "manipulating". And no biologist I know would be unaware of the attempts animals make at manipulating their environment, it is simply that their ability to do so also is under selective pressure.

  34. I am a practicing historian. I work in the field of environmental history, and I've readily used scientific papers in my own work.

    I look forward to listening to the podcast (I'm in my office at the moment), but I have a few comments based on what's posted here.

    A) As mentioned above, Marxist historians attempted to do something similar and failed miserably.

    B) Also mentioned above: Stochasticity is the central problem here. There are a lot of factors involved in any historical event. Yes, that's true for a lot of "scientific" fields. But, considering the limitations of understanding, say, modern economics, I don't necessarily buy the claim that we could understand the economics of the latter decades of the Roman Empire with our much more limited data (and economics was probably just one aspect of the collapse). We're talking about something as complex as climate with severe, impossible-to-break (barring the invention of time travel) limitations on our ability to collect data.

    C) It depends on the questions that historians are asking. Sure, models have a lot to say about big, world historical questions like the rise and fall of empires, but those aren't the only valuable questions that historians ask. I fail to see what scientific study would contribute to our understanding of something like the nature of female friendship in late Victorian England.

    So, I doubt Diamond's claim that history could be made a "science," but I certainly welcome the attempts to do so (though most of what was valuable in Guns, Germs, and Steel was recapitulation of the works of historians McNeill and Crosby).

  35. This comment has been removed by the author.

  36. This comment has been removed by the author.

  37. My Opinion:

    The strength of a Historical approach is that it posits that chronology matters. That is, a good way to understand how we got to where we are today, is to know what happened yesterday.

    When people come up with grand theories (more often, small hypothesis) to explain why things happened the way they did, they are usually practicing another discipline such as sociology, political science, economics, psychology, etc and using historical data as a basis. That is fine, but it should not be confused with good Historical research.

    Properly documenting what happened is the most important part of what a Historian does. In doing so, scientific techniques is beneficial in support of that goal. This could include using statistical techniques, textual analysis, scientific analysis of historical artifacts, etc.

  38. Mintman,

    "Nonetheless, the generalities of bubbles seem to be more of an issue for economists."

    This is where we get into pesky disciplinary squabbles. First of all, at the risk of stating the obvious, there are historians of economics. Second, don't you think historians might want to generalize about the social or otherwise extra-economic effects of bubbles? It seems to me that history has an inescapable interdisciplinary component; history (as a discipline in itself) is valuable precisely because it allows us to follow causal chains between phenomena of interest to multiple disciplines, but that no other single discipline studies in tandem.

  39. Er... "http://www.youtube.com/watch?v=Wdr4aAz7pP8"

    Pathetic, Roy. Mind to post something that at least makes a valid point to what you're saying?

  40. This comment has been removed by the author.

  41. Re-reading - er, re-skimming - all this after several months, it's apparent that most of the comments focus on generalizations (scientifically supported) as what does, or would, make history scientific. That seems to neglect other ways to be scientific about history, eg. when investigating individual people or events, or limited time periods. Archaeologists use lots of scientific tests (carbon dating, chemical analyses of bones, artifacts, dating by eclipse records, etc. etc. - or watch the History Channel or PBS's History Detectives). "Scientific" history needn't focus only on identifying general patterns!

  42. I have a request for clarification on what I suppose is a rather minor point, but still important to my understanding.

    In Malthusian dynamics the population eventually declines through emisseration, but eventually reaches stability as the population becomes supportable again. Under demographic structural theory what is the factor that eventually stabilizes the problem of elite overproduction?

    I understand that elites create factions and the factions fight civil wars. But don't those civil wars result in proportionately more deaths of peasants and fewer elite deaths? Doesn't that make the problem even worse?

    Or does this mean that the problem just remains and the civil wars between the factions continue until the elites at the top of the social pyramid are eliminated, and *that's* when a society regains its stability?


Note: Only a member of this blog may post a comment.