View the H-Diplo Discussion Logs by month
View the Prior Message in H-Diplo's April 2006 logs by: [date] [author] [thread]
View the Next Message in H-Diplo's April 2006 logs by: [date] [author] [thread]
Visit the H-Diplo home page.
[X-Post H-German] H-NET BOOK REVIEW Published by H-German@h-net.msu.edu (April 2006) Daniel Ziblatt. _Structuring the State: The Formation of Italy and Germany and the Puzzle of Federalism_. Princeton: Princeton University Press, 2006. xiii + 220 pp. Map, figures, tables, notes, bibliography, index. $39.50 (cloth), ISBN 0-691-12167-2. Reviewed for H-German by Jonathan Sperber, Department of History, University of Missouri The Use and Abuse of Social Science for History and Life Italian and German unification are the two classic examples of the founding of nation-states in nineteenth-century Europe. In both cases, one particular state--the Kingdoms of Piedmont-Savoy and Prussia, respectively--took up the demands of the nationalist movement and conquered or otherwise coerced other states into a new, national entity. Using approaches of the universalizing social sciences, Harvard political scientist Daniel Ziblatt wishes to explain the origins, process and outcome of these two cases of national unification. In particular, he wishes to explore why the end result was a unitary nation-state in Italy and a federalist one in Germany. For political scientists, as the author notes, this result is paradoxical, since their theories see federalism as occurring when a central power is too weak to impose its will on outlying polities. Yet relatively weak Piedmont-Savoy created a unitary Italian state, while powerful Prussia created a federalist German one. Ziblatt divides his explanation of this development into three parts, analyzing the movement for national unity, exploring the attitudes of the rulers of the different states toward this movement and, finally, considering the process of national unification itself. The author is thus investigating a significant development of nineteenth-century European history, one noted by contemporaries and discussed and debated down to the present. His use of a universalizing, analytical and quantifying social science approach holds the promise of offering new insights into these developments. Even historians who are proponents of an idiographic rather than a nomothetic scholarship (to use Wilhelm Windelband's celebrated epistemological distinction) might agree with Ziblatt's contention that a study of the modalities of German and Italian national unification could yield results with broader relevance to the understanding of the creation and dissolution of nation-states and other polities in today's world. Unfortunately, the book fails to live up to its promise. Showing the weaknesses of universalizing social science much more than its strengths, the work is characterized by a poor grasp of existing scholarship; problematic analytical and quantifying procedures; the misuse of these very procedures; and the neglect of other significant factors (for those who prefer social science lingo, the omission of intervening variables) that exerted a significant influence on the outcome of the events the author is analyzing. Let me exemplify these points by discussing each of the three steps of Ziblatt's analysis; namely, his consideration of the origins, government sponsorship and outcome of the process of nation-state formation. Ziblatt understands the movement toward national unity in essentially economic terms. Merchants, industrialists and market-oriented large landowners sought an expanded and unified market, to be achieved in a united nation-state. In particular, he argues that large landowners and merchants of Piedmont and Lombardy, on the one hand, and merchants and manufacturers of the Rhineland, on the other, were the moving forces behind the demand for national unity. We can recognize this line of explanation as one historians made about a generation ago. In particular, Kent Greenfield and Theodore Hamerow are the godfathers of Ziblatt's argument. While their assertions may have been convincing in the 1960s, they appear less so today, in view of the strong support for nationalism among social groups other than capitalist entrepreneurs and in economically less-developed regions. Issues such as the role of intellectuals and professionals in nationalist movements, the importance of voluntary associations, the effects of the dissolution of the old regime society of orders, the influence of religious and confessional factors, the gendered dimension of nationalism--in short, just about everything that has been written on the subject over the last thirty or so years--do not appear in the book. The single most problematic feature of the work is Ziblatt's explanation of how the rulers of the individual German and Italian states responded to the nationalist movement. Here, his basic argument is that the more economically developed a state was, the more likely its government was to support a pro-nationalist policy. Economically-developed larger (in terms of the size of the budget) states--that is, Prussia and Piedmont--would be "initiators" of moves toward national unity, while affluent smaller ones, such as the Hanseatic cities of Hamburg and Bremen, would be supporters of it. Thus, among wealthy states, the larger the state, the stronger its support for national unity. By contrast, economically backward larger states--the kingdoms of Bavaria and the Two Sicilies are Ziblatt's prime examples--would be openly hostile to national unification, while backward smaller states, such as Württemberg or the Duchy of Modena, would be reluctant to go along with it. Among poorer states, the larger the state, the stronger its opposition to national unity. To prove this point systematically, Ziblatt goes through all the pre-unification German and Italian states, and assigns each a number from one to four, representing the attitudes of their ruling groups toward national unity. Prussia and Piedmont get fours, as the strongest supporters of nationalism, while Bavaria, the Two Sicilies and the Papal States get ones, as its greatest opponents, and the other states get twos or threes, representing intermediate attitudes. This is a deeply flawed procedure, for three distinct reasons. First of all, with regard to Germany, in Ziblatt's analysis support of nationalism means, entirely, support of _kleindeutsch_ nationalism. _Großdeutsch_ nationalism does not exist in his analytical world; the Habsburg monarchy does not count as part of Germany, and is hardly mentioned in the book. Most of the German states he describes as enemies of nationalism--Bavaria, Saxony, Württemberg, Hanover, Hessen-Kassel--were supporters of _großdeutsch_ policies. Since any potential German unification along _großdeutsch_ lines would have been even more federalist in nature than Bismarck's _kleindeutsch_ nation-state, and Ziblatt is trying to explain why Germany ended up as a federalist polity, this neglect of _großdeutsch_ policies seems analytically very inappropriate. Even accepting the author's neglect of different forms of nationalism, the very procedure of assigning numerical values to the German and Italian governments' policies is problematic. Ziblatt does not just use his numbers as illustrations--the four meaning that the Prussian government was a much stronger supporter of nationalism than the one for Bavaria--but employs them as actual quantitative values, as ratio variables, to use the technical term. He takes averages of the numbers to calculate the mean support for nationalism among different-sized states and uses a transformed version of the numerical values as dependent variables in regression equations. This procedure implies that Prussian government policy was four times as strongly supportive of national unity as Bavarian, and twice as much as the government of Württemberg, or that the Piedmontese government was four times as supportive as that of the Papal States and twice as supportive as that of the Grand Duchy of Tuscany. But there is simply no justification for these values. Why a ratio of four to two to one? Why not seven to five to three or fifty to forty to one? The technical description of this procedure is "introducing a metric," and it is often understood as a misuse of statistical methods. The problems with Ziblatt's numerical coding scheme do not end there, for his application of it is inconsistent in a way that supports his conclusions. For instance, consider his coding of Lombardy with three, as strongly supportive of national unity. The government of Lombardy, part of the Austrian Empire, was, naturally, very opposed to Italian national unification. (For some unknown reason, the other Habsburg Italian province, Venetia, does not appear in the author's analysis.) Since Lombardy was one of the economically most advanced parts of Italy, this would contradict the author's thesis that economically advanced states supported national unity. Hence, he makes an exception for Lombardy, coding it not by its government--as he does with all the other states in his analysis--but by the attitudes of its social elites (pp. 158-159), who were, indeed, supportive of national unification. But if he does that, then why not treat other German or Italian states in the same way? The Grand Duchy of Tuscany, scored with a two, counts as opposed to national unity because of its government's policies, but the Grand Duchy's elites were apparently supporters of national unity, because the _Società Nazionale_, the Italian counterpart to the _Nationalverein_, had its greatest membership density there (p. 60). Coding problems are particularly apparent in Ziblatt's analysis of the Duchy of Nassau. First, it is not clear why Nassau, whose government went to war with Prussia in 1866, and was conquered and annexed by Prussia for its pains, should get a three as a strong supporter of national unity, while the two Mecklenburgs, the only German states to fight with Prussia in the war against Austria, only get twos. But after looking at Ziblatt's list of German states, I noticed an even greater problem. One of them, the Free City of Frankfurt am Main, is missing. This is not because the author intended to eliminate the free cities; Hamburg and Bremen are present in his analysis and play a significant role in it. Rather, this omission of Frankfurt seems to emerge from Ziblatt's misunderstanding of the book by the German economist Harald Frank from which he took his figures for regional GDP. Frank gives regional GDP for the post-1867 Prussian _Regierungsbezirk_ Wiesbaden, which Ziblatt takes to be identical with the pre-1866 Duchy of Nassau, apparently unaware that the Prussian administrative unit also included the city of Frankfurt am Main. This oversight works against Ziblatt's analysis in two different ways. First, it leads him to overestimate grossly the affluence of Nassau, since a good part of the region's total product was due to the economic activity in the wealthy city of Frankfurt. Second, Ziblatt simply excludes the former independent city government of Frankfurt from his analysis of the policies of the German states toward national unity. Both omissions seriously undermine his argument that "rich but smaller states, such as Hamburg, Bremen and Lombardy, supported the national unification efforts of Prussia and Piedmont" (p. 30). Nassau was, supposedly, one of these rich, smaller states supportive of national unification, but its "wealth" comes from the author's misunderstanding of German administrative boundaries, leading him to assign the assets of Frankfurt to Nassau. By contrast, the wealthy small state, the free city of Frankfurt am Main, was a strong opponent of Prussian-led German national unification--and this opposition was characteristic of Frankfurt's government, most of its political elites and its population as a whole--once again directly contradicting Ziblatt's argument. It could be asserted that the questionable features of the author's coding of Lombardy and Nassau relate to just one case each for Germany and Italy and do not necessarily undermine his analysis. Unfortunately, Ziblatt has a very small number of cases in his analysis--seven for Italy and seventeen for Germany. Indeed, he divides the German states into two groups and performs two regression analyses with just ten and seven cases respectively. When one works with so few cases, misattributing even one of them calls the entire analysis into question. Consequently, these errors rather suggest that the author's analysis of the policy of the different German and Italian governments toward the national unity movement is more than a little flawed. Size and economic development are insufficient as explanatory variables. Many other factors need to be considered--regional location, confessional composition and political culture, to mention just a few. Of course, increasing the number of variables would make some of the author's quantitative analysis, regression equations with relatively few cases, impossible (Table 2.6), but such a strategy might also call into question other analyses, such as the one presented in Figure 3.1, purporting to show a relationship between regional GDP and membership in the _Nationalverein_. This lack of consideration of other variables also presents problems with the author's efforts to apply his conclusions about German and Italian national unification to Europe more generally. His list of federal and unitary states in contemporary Europe seems to overlook the obvious: namely that federalist polities only exist in central Europe, in Switzerland, Austria and Germany, while all other states (with the exception of Belgium, and that only since 1997) are unitary ones. Clearly, regional political culture must have played an important role in such an outcome, but the author seems reluctant to consider it. Finally, we come to the author's explanation of the differential outcomes of the national unity processes. In the most interesting part of his book, he argues that the existing political science theory on the origins of federalism has it exactly backwards. It is not the strength or weakness of the central state that matters, but the strength of the peripheral ones. In Italy, the Papal States, the Duchies and the Kingdom of the Two Sicilies were weak states; they collapsed as a result of the wars and revolutions of 1859-61, so that Piedmontese statesmen had no choice but to create a unitary nation-state, in spite of the federalist sympathies of Cavour. By contrast, the stronger German states were able to withstand the shock of the war of 1866, so that Bismarck was willing to use their administrative capacities to construct a German nation-state, rather than starting from scratch or using the Prussian bureaucracy for this purpose. This argument definitely has something to say for itself, although the primary evidence for the weakness of the Italian states and the strength of the German ones is their collapse or non-collapse during the military conflicts of the respective wars of unification. On other measures of strength, which the author employs, the weaknesses of some of the Italian states (at least relative to each other) is not so clear. The Kingdom of the Two Sicilies had an enormous number of state officials, far more than any other state in Italy--two-and-half times as many as Piedmont-Savoy did in 1860 (Table 4.5). Is that the sign of a weak state? Ziblatt does argue that the other Italian states collected noticeably less revenue per capita than Piedmont, so that Piedmont must have been the strongest Italian state. However, Piedmont's per capita GDP was also higher than that of the other Italian states (Table 4.1). The Piedmontese kingdom could extract about 1.68 times as much revenue per capita from its subjects as could the Grand Duchy of Tuscany, but the Piedmontese per capita GDP was 1.74 times that of Tuscany, so Piedmont was actually extracting a smaller proportion of its GDP in revenue. Parallel calculations show that the differences in per capita revenue collection between Piedmont-Savoy and the Two Sicilies also ran parallel to differences in their per capita GDPs. The author is rather reluctant to consider other factors that might have influenced the outcome of the two national unifications. One that immediately comes to mind is the very different processes of unification. Although begun with a war, Italian national unification also involved revolutionary uprisings in the Duchies, the northern part of the Papal States and, of course--following the expedition of Garibaldi's one thousand red-shirts--in the Kingdom of the Two Sicilies. By contrast, 1866 saw the military conquest of the German states by the Prussian army. Revolutionary events were distinctly lacking. Members of the _Nationalverein_ did not occupy the prince's palace in Kassel; Rudolf von Bennigsen did not lead one thousand gymnasts in an invasion of Bavaria. This very different turn of events is also a reflection of the lack of popularity of Prussia's course in 1866. At least before and during the war, German public opinion and the nationalist movement were generally opposed to Bismarck's policies. These distinct differences in the degree of upheaval, and in the coincidence between the nationalist movement and the policies of a "pro-nationalist" state ought to be considered alongside the coercive and administrative capacities of different states. The author does note the place of the Great Powers in the development of national unification, but perhaps does not fully consider how the Powers' interaction raises questions about his thesis on the relevance of the strength or weakness of central powers for state creation. The weak state Piedmont was relatively strong in the Italian peninsula, because the one potentially intervening power, the Habsburg monarchy, had been defeated by the French in 1859. By contrast, powerful Prussia was actually not so powerful in the central Europe of 1866; the threat of French intervention hung over both the war with Austria and its immediate aftermath. Bismarck's decisions, such as respecting the territorial integrity and sovereignty of the south German states, were strongly influenced by this diplomatic constellation. Bringing in the interactions of the European Powers actually supports the political science thesis the author is trying to refute, namely that federalist polities emerge from weaker central states, while unitary policies are the result of stronger ones. In general, this book leaves a distinctly disappointing impression. Although dealing in a potentially interesting and innovative way with an important topic of nineteenth-century European history, one also with broader ramifications, the work is characterized by a problematic methodology, itself employed in questionable fashion, and a neglect of significant factors (or, if one prefers, independent variables) affecting the outcome of the process being studied. Unfortunately, these problems seem to have escaped the notice of American social scientists. The book has been published by a leading academic press; its dust-jacket contains fervent accolades from prominent sociologists and political scientists; the dissertation on which it was based received not one, but two, prizes of the American Political Science Association. I suppose this does not matter very much to historians, who have stopped taking their cues from social scientists, preferring nowadays to import their theory from cultural and literary studies. In view of the important past contributions made to the study of European history by social scientists such as Charles Tilly, Stein Rokkan, Rainer Lepsius, Jürgen Falter or Seymour Martin Lipset, this lack Of attention to important empirical and methodological problems does seem like an odd state of affairs. Notes . Kent Greenfield, _Economics and Liberalism in the Risorgimento_ (Baltimore: Johns Hopkins University Press, 1965); Theodore Hamerow, _The Social Foundations of German Unification, 1858-1871_, 2 vols. (Princeton: Princeton University Press, 1969-72). . I tried using these different numbers for the degree of support of national unification and plugging them back into regression equations given in Table 2.6 and found I was able to reduce the R-squared by a factor of three and render the independent variable insignificant. . Harald Frank, _Regionale Entwicklungsdisparitäten im deutschen Industrialisierungsprozess 1849-1939_ (Münster: LIT Verlag, 1993), Appendix 8, p. xxx. Ziblatt's annoying habit of referring to nineteenth-century GDP figures in Reichsmark shows yet another misunderstanding of the work he cites (and of German history more generally), since the Reichsmark only became the German currency in 1924. Copyright (c) 2006 by H-Net, all rights reserved. H-Net permits the redistribution and reprinting of this work for nonprofit, educational purposes, with full and accurate attribution to the author, web location, date of publication, originating list, and H-Net: Humanities & Social Sciences Online. For other uses contact the Reviews editorial staff: firstname.lastname@example.org.