Bertrand M. Roehner and Tony Syme:
The book under review constitutes one of the most intriguing,
challenging, and at the same time, frustrating works I have come across
in the last few years. The authors make their focus the creation of a
``nearly'' scientific approach to comparative historical sociology using
examples drawn from the literature on the origin of revolutions and
strikes, war and territorial aggression, and the problem of wartime
logistics. But in another sense, the authors are interested in a
deeper question which should be of special interest to the readers of
EH.NET: How does one study history from a social scientific
perspective?
They want to know: What constitutes relevant and irrelevant historical
information? What does it mean to observe recurrent patterns in
similar historical events? How do we ``test'' or verify our beliefs and
claims? Economic historians have been especially concerned with these
problems and have developed a variety of partial answers and a number
of workable methodologies over the last few decades without coming up
with a systematic or definitive answer to the problem.
After reviewing the various methodologies in common use in historical
sociology, the authors make mostly common-sense decisions about the
best way to proceed. They try to break up historical events into
smaller, ``component'' parts. They compare and contrast these parts
across similar historical episodes, and they analyze these parts and
their interrelation using a mix of qualitative and quantitative
description. They seem to be indifferent to formal statistical
hypothesis testing, though I do not entirely understand why. Surely
Cliometrics, for all its limitations, explicitly deals with many
issues of importance to such sociological discussions. More
inexplicably, they criticize the use of individual choice methodology
as typically applied in economics and economic history, because
individual choice theory ignores ``group representation and collective
consciousness'' (p. 21). Yet when using what they consider a
``compelling argument'' against individual choice theory, they point to
a criticism of Jack Goldstone against Gordon Tullock’s positing of the
standard collective action problem. As we know, the standard problem
of collective action makes it unprofitable for individuals to
participate in a revolutionary movement if free-riding were
possible. Goldstone makes the point that individuals do not decide
to join, or not join, revolutionary movements as isolated
individuals. Instead, they decide ``as part of groups to which they
have prior commitments'' (Goldstone 1999). ``This leads us to analyze
the role of social constraints'' (Roehner and Syme, p. 21).
Yet this only seems to be an argument for expanding the set of
considerations that affect the payoffs to individual actors without
changing the desirability or viability of the individual choice
methodology as the appropriate tool of analysis. Perhaps a ``collective
consciousness'' technique (which I understand is neither universally
agreed upon, well-established, nor battle-tested even in sociological
circles) would do better than individual choice reasoning with
modified collective constraints, but it seems a bit premature to
simply assert the inadequacy of the individual choice methodology on
the basis of this simple critique. More important, I see little in
their actual historical cases and analysis which is not amenable to
individual choice reasoning. Indeed, they use individual actor models
throughout most of the book, albeit substituting the state as the
individual actor. But this is a tradition that is no more
controversial (albeit incompletely specified) in political economy
than is the habit of treating firms as single individuals in the
literature on industrial organization. These usages are certainly
well-known to the authors, one of whom holds a position as a lecturer
in economic history! Hence, this methodological discussion, no matter
how interesting, seems less than essential for the success or failure
of the bulk of the book.
Be that as it may, there is much pleasure and learning to be derived
from a careful study of the substantive chapters. Their chapter on
``General Strikes, Mushroom Strikes'' contains much insight in the
nature and timing of strikes and shows how a comparative perspective
builds on the more narrowly constructed national studies of strike
duration, which allow us to observe, for example, that strikes are
more frequent in the Spring and Summer, and less frequent in the
Autumn or Winter. An economist might point out that this would be a
good example of individual choice in action since it is both more
pleasant to be outside in the warmer months, and the opportunity cost
of striking or, even worse, of losing employment (due to a failed
strike) in winter months is greater than in the summer. The authors
themselves look to problems of partial unemployment and seasonal
disparities in work opportunities or demand for output (cf. coal in
winter) as important explanations. This sort of reasoning, at the
margin, is precisely what neoclassical microeconomic theory is all
about. Economics reasoning is not to be applied to the average. It
does not explain nor seek to explain why some people strike no matter
the individual cost to themselves. Rather, economists argue that at
the margin, the more costly it is to engage in an activity, the less
it will be undertaken particularly when dealing with aggregates of
separate actions. This is a common problem that pops up with some
regularity in social science critiques of economics. Economics is said
to be deficient in explaining why anyone bothers to vote in a large
election, or participate in a large-scale strike, but it makes no such
claims in the first place. And it is foolish to deny the usefulness of
marginal reasoning either in economic history, or for that matter, for
scientific analysis in general. Given their otherwise commonsensical
approach to modifying scientific methodology to the study of
sociology, the authors, and we, would have benefited from their
explicit incorporation of economic reasoning in their work or with
explicit confrontation on a case by case basis where such reasoning
fails.
Much of the book is taken up with the analysis of warfare, both in
the general subjects of ``Warfare for Territorial Expansion'' and ``The
Constraints of Logistics.'' There is too much to go over here in a
short review, as their discussions range from the Napoleonic Wars
through the First and Second World Wars. Much of their discussion
seems to match standard military history such as the fact that
submarine warfare in the Atlantic had striking parallels in both the
First and Second World Wars, suggesting that the internal logic of
submarine warfare tended to dominate over ephemeral
considerations. Yet these discussions only highlight the difficulty of
pinning down the authors’ hypotheses. They ask ``Would it be the same
again in a future conflict even though nuclear propulsion has
considerably altered the problem? We are convinced that the response
to this question must be sought not in an evaluation of the technology
of submersibles, but in a comparison of the economic and naval
potential of the belligerents'' (p. 263). (And surely it makes no sense
to posit the economic and naval potential against technology, since
the naval potential of the belligerents is surely determined in part
by the technology of the instruments of naval warfare.) After a few
rereadings of the subsequent analyses, I can only conclude that the
authors answer with a resounding ``Maybe.''
The authors, in my view both lose the opportunity to explore the
interaction between technological characteristics and military
capacity as well as the deeper question of which random events were
genuinely significant for the war, and which events were driven by the
logic of economic, military and political considerations For instance,
in the chapter on logistics, one would have liked to have seen a
discussion of the role of contingency in the Second World War, where
surely the success of the attack on France had much more to do with
risk-taking, Allied error, and sheer luck than the conduct of the War
in the East after, say Stalingrad, when individual battles ceased to
matter, and sheer numbers virtually guaranteed German defeat. The
issues they do tackle are stimulating but whether I agreed with them
or not, it was not always clear what ``test'' or comparison we should
look to when deciding whether a particular set of comparisons was
insightful or accurate.
So this is certainly an imperfect book, but in its verve and ambition,
one which nonetheless should be commended to the attention of EH.Net
readers.
John V. C. Nye specializes in French economic history and industrial
organization. His publications include ``Tax Britannica: Nineteenth
Century Tariffs and British National Income,'' (forthcoming in Public
Choice, 2004). He is a founding member of the International Society
for the New Institutional Economics (ISNIE).
Pattern and Repertoire in History.
Cambridge, MA: Harvard University Press, 2002.
xii + 413pp. $45 (cloth),
ISBN: 0-674-00739-5.
Subject(s): Labor and Employment History