Reconsiderations: Schweller's Deadly Imbalances

Introductory note:  
Randall Schweller's name is well known to those familiar with international-relations theory.  Schweller's first book, Deadly Imbalances: Tripolarity and Hitler's Strategy of World Conquest (1998), was based on his Columbia University dissertation.  His second book, Unanswered Threats (2006), also dealt with balance-of-power issues, focusing on instances of so-called underbalancing.  As mentioned before on this blog, in recent years Schweller's attitude toward the conventional versions of Realist international-relations theory has become much more critical (see his 2014 book Maxwell's Demon and the Golden Apple).  In this guest post, Peter T. looks back at Schweller's first book Deadly Imbalances, offering some thoughts on the book and on the enduring problem of the relation between theory and history.  I have added one sentence in brackets. -- LFC

-----

Schweller's theoretical argument starts from Waltz's classic Theory of International Politics (1979).  Schweller acknowledges the validity of criticisms that Waltz's theory, modelled explicitly on economic theories of the market, is "too abstract to generate useful hypotheses about specific foreign-policy behaviour."  By adding a number of other factors Schweller hopes to bring theory into a closer approximation to reality.  [By contrast, Waltz's view is that, within certain limits, a theory's "[e]xplanatory power...is gained by moving away from 'reality,' not by staying close to it." (Theory, p.7)]

The argument of Deadly Imbalances centers on the lead-up to and conduct of World War II.  Schweller argues that in the late 1930s, the world was effectively tripolar, with the U.S., the Soviet Union, and Germany the central players.  This requires that Schweller establish a more nuanced hierarchy among states than a simple division into Great Powers and others.  So from the classic IR world we go to a world of Poles, Lesser Great Powers, middle powers, and others.  Further, states are no longer simply status quo or revisionist in their aims, but status quo, revisionist or neutral (the U.S. up to 1941).  Revisionist powers can have limited or unlimited aims, and states' strategic choices are not just about balancing but also buck-passing, band-wagoning, binding, distancing or engagement.

Not all these strategies are realistic options for all states, but each state has sufficient choice that, in combination with their power status and aims, there are a myriad of potential outcomes.  And this is the problem with the approach: as Schweller adds new factors his theoretical base becomes less like an explanation and more like a description.  And, as the book proceeds, the theory gives way to what looks very like old-fashioned diplomatic-military history.  It is a long way, for instance, from calculations about the overall balance of power to Hitler's belief that shipping constraints would prevent effective U.S. intervention in the European war.

In this way the book falls between two stools.  The theory illuminates very little of the complex forces at play, while the exploration of the detail draws almost entirely on secondary sources and has dated rapidly.

A second issue is that Schweller's theory, like Waltz's, assumes that states have a clear and relatively accurate view of their own and others' power. Schweller goes into some detail on the strengths of the various Poles and Lesser Great Powers, using the Correlates of War (COW) project estimates of power.  For the period, these are a composite of industrial production, population, and military strength.  Yet, as Schweller details, all states made major errors in estimating their own and other's power.  And all considered not just these factors but many others -- national morale and political cohesion, geographic position, financial resources, allies and sympathisers, operational proficiency, military technology and more.

As an instance, the COW rankings put Great Britain well behind Germany in the '30s.  Yet Great Britain had an overwhelming advantage at sea, the backing of the dominions (most of whom disposed of considerable military and industrial resources), its position as the second financial centre of world trade, and the manpower and other resources of India.  Moreover, Great Britain's economy was more advanced than that of Germany (with its still large agricultural sector) and, of course, Britain had the advantage of being an island with free access to the Atlantic.  Britain worried about the German army and air force; Germany worried about British financial pressure, grip on overseas trade, ability to call on colonial and imperial resources and on U.S. support.  Germany had an immediate superiority, Britain an ultimate one.  These are not commensurate capabilities, to be summed into two numbers and compared.

Again, Germany's attack on the Soviet Union was predicated on the certainty that German operational proficiency and Soviet political weakness would more than compensate for superior Soviet numbers, the effects of distance, and German logistic shortfalls.  Soviet calculations were, of course, in the reverse direction.

The inability to exactly define or measure power is a central problem for IR theories.  A major reason that wars happen, as Geoffrey Blainey (in The Causes of War) pointed out, is that states are uncertain of their relative power.  Clausewitz likened battle to cash settlement in commerce -- the moment when true credit-worthiness is tested and revealed.  Likewise, war usually provides a moment of clarity about relative power.  In the absence of a recent test, all parties are left to manoeuvre in uncertainty.

And this uncertainty extends to defining who are the major players.  In the '30s the U.S. preponderance of industrial and financial power was widely acknowledged.  Yet how this translated into international influence given U.S. isolationism, its distance from Europe, and the small size of its army left a lot of room for error.  In some areas the U.S. was a major force, in others a minor player.  The same could be said of the Soviet Union, China, Japan and Italy.  It is this zone of uncertainty that gives rise to alterations in the ranks of the Great Powers -- the sources of their strength go unrecognised until revealed in some contest, rendering previous calculations and strategies moot.

So Schweller's Deadly Imbalances is an interesting, and not unrewarding, read. Its strength is his willingness to engage with the detail; its weakness the inability of abstract theory to explain that detail in any convincing way.  It is a general weakness of grand theory in this area of study.

-- Peter T.

On "turning points" and "critical junctures"

It is a truism that historical narratives impose a pattern on the past.  Narratives demand a structure, which the messiness of reality resists.  Hence historians' talk of "trends," "major developments," "main events," "phase transitions," "turning points."  Most writers deprived of recourse to these sorts of words and phrases would probably produce something either unreadable or confusing, or both.

Moreover, historians have long understood that metaphors -- tides, winds, take-offs, etc. -- can help make their narratives more vivid.  To take a not-quite-random example, E.J. Hobsbawm, in his chapter on the industrial revolution in The Age of Revolution (Mentor paperback ed., 1962, pp.45-6), borrowed the phrase "take-off" from W.W. Rostow.  Hobsbawm wrote:

...[S]ome time in the 1780s, and for the first time in human history, the shackles were taken off the productive power of human societies.... This is now technically known to the economists as the 'take-off into self-sustained growth'.... [C]areful enquiry has tended to lead most experts to pick on the 1780s...as the decisive decade, for it was then that, so far as we can tell, all the relevant statistical indices took that sudden, sharp, almost vertical turn upwards which marks the 'take-off.' The economy became, as it were, airborne.

***

The recognition that history has to be narrated and that most narratives involve the imposition of pattern and order has given rise to at least two sorts of academic controversies, or, to use a perhaps better word, conversations.  Both of these conversations themselves have a long history, but to simplify things we can restrict ourselves to their recent installments. 

One of these discussions has revolved around issues of objectivity and truth.  As recounted by Andrew Hartman in A War for the Soul of America: A History of the Culture Wars (Univ. of Chicago Press, 2015, pp.259-60), a divide emerged or re-emerged in the 1980s and '90s between those historians who, influenced by postmodernism, were inclined to emphasize the subjective dimension of historical narrative, and those who stressed, in the words of the authors of Telling the Truth about History (1994), "the need for the most objective possible explanations...."   Much could be said about all this, but it's not the focus of this post.

A second scholarly conversation, the one with which this post is concerned, involves the question of continuity and change.  Almost everyone agrees that these are not completely opposed categories.  There is no such thing as a completely static social system, and even those that appear to be static are subject to changes or variations in the course of reproducing themselves, as sociologist Wilbert Moore, among many others, pointed out (Social Change, Prentice-Hall, 1963, pp.11-16).  The anthropologist Marshall Sahlins put a very similar point this way: "Every actual use of cultural ideas is some reproduction of them, but every such reference is also a difference.  We know this anyhow, that things must preserve some identity through their changes, or else the world is a madhouse" (Islands of History, Univ. of Chicago Press, paperback 1987, p.153).      

The proposition that continuity and change are not cleanly opposed categories, that no change is ever total, doesn't resolve the issue, of course, but simply opens it.  Which changes are more or less important, and how does one decide?  John Lewis Gaddis has endorsed another scholar's suggestion that historians should look for "a point of no return," i.e., "the moment at which an equilibrium that once existed ceased to do so as a result of whatever it is we're trying to explain." (Gaddis, The Landscape of History, Oxford Univ. Press, 2002, p.99, citing Clayton Roberts, The Logic of Historical Explanation, 1996.)

Somewhat more helpfully perhaps, Paul Pierson, a political scientist, has emphasized in Politics in Time (Princeton Univ. Press, 2004) that the evolution of societies or institutions is often heavily influenced by relatively small events that happen early in a developmental path; that is, he stresses the "self-reinforcing" character of path-dependent processes.  Pierson cites (pp.52-3) as one example his sometime co-author Jacob Hacker's "analysis of the development of health-care policy in the United States...."  The failure to adopt national health insurance during the New Deal "generated powerful positive feedback, institutionalizing a set of private arrangements that made it much more difficult to make a transition to national health insurance at a later point in time" (emphasis in original).  In other words, the U.S.'s failure to adopt national health insurance in the 1930s was not a dramatic-appearing event, but it had long-term consequences for the direction of future policy.  This is the sort of 'self-reinforcing' effect that Pierson suggests occurs quite often.  

This point that "critical junctures" need not be large-scale, big-bang events -- that they can even be the failure of something to occur, rather than a positive occurrence -- is a significant one, as are, no doubt, some of the other arguments in Pierson's book (I haven't gone through all of it carefully).  But I suspect that, in the end, an element of subjective judgment is inescapably involved in deciding what counts as a critical juncture or a turning point, except for a few obvious instances that would garner wide agreement.  Some may find this a disappointing conclusion, but so be it.

Added later: See also Michael Bernhard, "Chronic Instability and the Limits of Path Dependence," Perspectives on Politics (Dec. 2015): 976-991.