The first of these is inadequate research. Sixty pages of notes and eleven of bibliography may seem impressive, but the reading required to undergird so sweeping a reinterpretation of history as Bobbitt offers would have to be at least ten times as great and yet also more discriminating-a nearly superhuman demand, to be sure, but one that is inseparable from the undertaking. Apart from discussions of Bobbitt's specialty, constitutional law, the only topic among the many huge, controversial ones he discusses on which his reading is adequate is that on the 15th-18th century revolutions in military affairs.

Inadequate research contributes to other defects but cannot wholly account for them. Ungrounded generalizations, naked assertions, logical leaps, vague language, conceptual confusion, contradictions, arbitrary definitions, exaggerations and distortions, and major omissions of vital material abound. Then there is the problem of outright factual errors.

Factual errors are bound to occur in any book as broad and ambitious as this one. What counts is the level of their incidence and significance, and it is unacceptably high. Some errors do not directly affect the overall theses, but others demonstrate a general misconception of a problem important to the story. As to the former, for example, when Bobbitt writes that in 1798-99 France attacked and conquered Switzerland, the Papal States, Piedmont-Sardinia and Naples (which is like saying that the Soviet Union in 1947-48 attacked and conquered Poland, Czechoslovakia and Hungary), the factual error shows only that he misunderstands the nature of French expansion and the allied response in a crucial period of one of his epochal wars. But the statement that after 1815 the majority of Poles lived under Prussian and Austrian (rather than Russian) rule is not an incidental error. It shows that the author cannot have understood the Polish question either in its domestic or international aspects. It would be like trying to understand the American race problem if one believed that after 1865 the majority of freed blacks lived in the North.

Still other errors reveal the vagueness and vacuity of certain core concepts. When Bobbitt writes that the Congress of Vienna "recognized the state of Switzerland as a single state-nation", he demonstrates not only a misconception of what the Congress did, but the meaninglessness of his category of the state-nation. Similarly, when Bobbitt contends that Russia was able to withstand Napoleon's onslaught in 1812 through strategic retreat because "she [Russia] was not a territorial state, and her dynasty did not constitutionally depend on the support of the nobility and the army", he not only gives an absurd explanation for the strategic and tactical decisions made in 1812 and betrays an ignorance of major facts about Russian history, but again demonstrates how malleable and hence meaningless his concept of the "territorial state" is.

Finally, certain factual errors reveal misconceptions so basic that they raise questions as to the author's ability to deal with a whole complex of central problems and issues. Bobbitt describes European feelings about war and peace at the end of the 19th century as follows: "There seems to have been widespread agreement on two expectations: that science and technology would make war impossible and that international law would govern the relationships between states." This is like saying that the general late 19th-century view on the solar system was that the sun revolved around the earth. A misconception so fundamental and so contradictory to central, well-known facts established by a massive literature disqualifies Bobbitt from being taken seriously on the international order in prewar Europe or the origins of World War I, the start of the most important of his epochal wars.

Even more disturbing in certain ways than factual errors are instances of Bobbitt's twisting the clear meaning of evidence to fit his theory. Two examples: he praises Hedley Bull's "path-breaking" book, The Anarchical Society, for describing the international world in Hobbesian terms as without law, "a world of all against all and each one against every other one." Bull's book was indeed important-for precisely the opposite reason. His whole emphasis was on the international system as a society, anarchical solely in the sense of having no recognized lawgiver but otherwise involving real community, norms, rules and elements of law. Bobbitt similarly takes a statement by Bismarck defending his policy of pure Realpolitik in the interests of Prussia against his opponents' charge of lack of principle and makes it into something Bismarck then opposed: a manifesto for nationalism as "the authentic voice of the nation-state."

A Reach Too Far

Instances of Bobbitt's bad scholarship could be multiplied, but simply to indict the book in that fashion ignores the big picture: the book's theses and argument. Could they not be mainly right and very much worth thinking about, even if Bobbitt is not a good historian? A good question; but the answer is no. The broad scheme of war, peace and the course of history given here is equally flawed. It is, in fine, an imposing fortress of papier-maché.

The first problem is that nothing in Bobbitt's very grand scheme is ever proved. Nothing major in it-neither categories, concepts, fundamental assumptions, alleged links and causal connections nor definitions of critical terms-is rigorously analyzed, hypothesized and operationalized. Nor is any of it tested against contrary evidence and alternative views and shown to be more solidly based than competing schemes and interpretations. If one believes it, one does so essentially because it looks good, or because one wishes to believe it.

The second point is that under close examination the grand scheme falls apart. Central categories develop fissures and cracks and their contents leak out and mingle with others. Crucial concepts, when tested, prove tautological or simply empty. Vital causal connections and links between phenomena in different spheres prove nonexistent or unconvincing. Major generalizations central to it prove untenable. An example: One of Bobbitt's main contentions is that epochal wars and their peace settlements establish the dominant constitutional form of the state and the constitution of an era's society of states. The first of these, the Habsburg-Valois wars and the Peace of Augsburg, meet none of the requirements of the theory. These wars were not epochal but local and sporadic, and did not end in 1555 but continued underground until once more breaking out openly in the 1630s and ending in 1659. The Peace of Augsburg had nothing to do with them; it established only a temporary and unstable truce in Germany's religious conflicts, and led to no new constitutional order either in Germany or elsewhere.

Or take Bobbitt's concepts of different state constitutions-princely, kingly, territorial and so forth. Not only is each of these fuzzy in definition and the distinctions drawn between them arbitrary and artificial, but at least one and perhaps two (the state-nation and the territorial state) are figments of Bobbitt's imagination, corresponding to no historical reality whatsoever. Of the five great powers at the Vienna Congress that supposedly established and legitimated the dominant state-nation form, not one remotely fits his definition of it-and not one wanted or dared to do after 1815 what he says state-nations do, which is to put their people at the service of the state. As for the territorial state, its supposed legitimating principle-the more efficient use of the state's resources-would fit some states and not others in any era; and some of the central defining characteristics Bobbit ascribes to its era as opposed to others (e.g., the unimportance of dynastic succession, a coolly rational secularism in regard to religion, and the waging of only limited cabinet wars) are all flatly untrue.

Indeed, the whole notion that each different constitutional form of the state rests on a particular distinctive legitimating principle-so that nation-states try to maximize the welfare of all their citizens while market-states try to maximize opportunity-is unhistorical to the point of absurdity. Had Bobbitt done any serious study of his 18th-century territorial states, he would have seen that many of them, at least, saw their raison d'être in promoting the welfare of all their subjects as they saw it, and they tried to do so in part by maximizing their opportunities within the established order.

Alas, most instances of incoherence and conceptual confusion in The Shield of Achilles come down, ultimately, to the central one involving Bobbitt's notion of what drives the course of history. He vastly exaggerates and distorts the roles of war and peace settlements in constituting states and the international system. Yes, states to a considerable degree are made by and for war; but states exist and endure by and for governance. The constitutional changes that states have undergone over time (in a way far more uneven, contingent and complex than Bobbitt indicates) have had more to do overall with trying to meet challenges of governance than those of war, and those demands, while changing enormously in detail over time, remain essentially the same in kind: order, welfare and legitimacy. How rulers, governments and peoples conceive and try to meet these needs varies; their need to do so in order to endure remains constant.