One of the many strengths of Professor Sørensen's thorough critical response to my book is that it underlines the ways in which I tried to challenge what passes for conventional wisdom among IR scholars about realism. I readily confess to an interest in reviving heterodox thinking.

Sørensen endorses the mainstream view of classical realism by discounting my emphasis on its neglected reformist and cosmopolitan traits, which allegedly “did not exert any considerable influence” on its empirical analysis. For the classical Realists, however, the empirical and normative were typically interconnected. Not surprisingly, Morgenthau's Politics Among Nations devoted two chapters to world government and its prospects. So why then have many readers missed this key commitment to far-reaching global reform among major classical realists? Because IR students and scholars do not, in fact, typically read any but a narrow snippet of realist writings. Indeed, Morgenthau considered politics a struggle for power, and he believed states should pursue national interests. Yet a careful perusal of his writings, and not just of the brief textbook selections on which we too often rely, suggests that his cosmopolitan sympathies were indispensable to his understanding of what such realist ideas entailed under the conditions of what we now describe as globalization. My little book offers substantial textual evidence for this claim, as does an excellent study by Campbell Craig (Glimmer of a New Leviathan: Total War in the Realism of Niebuhr, Morgenthau, and Waltz, 2003).

Sørensen also seems taken aback by the occasionally-leftist views about capitalism found among some classical realists, which similarly clash with received textbook notions. Like Sørensen, Niebuhr and others thought that we needed to identify “the most effective state-market matrix.” In contrast, they posited—accurately, in my view—that only a critical and hard-headed assessment of capitalism's pathologies might help us identify that matrix. Where it arguably has been achieved (e.g., in Scandinavia), it only transpired because of powerful labor and socialist movements whose skepticism about capitalism at least partly humanized it.

Sørensen then repeats the commonplace, though hardly uncontested, view that “nuclear weapons have over time tended to lead to more caution, more prudence, more stability in the military relationships between nuclear powers.” Yet a great deal of scholarship follows classical realism in underlining nuclear weaponry's limitations as a source of political stability, and, like the realists, it emphasizes potential perils. In contradistinction to recent neorealism, whose overly self-satisfied views of nuclear weapons Sørensen seems to endorse, the realists thought that the nuclear revolution posed fundamental moral questions which no political science worth its name should circumvent. As I try to explain in my book, I believe they were right to do so.

My little volume never aimed to provide a full-fledged defense of world government. So Sørensen is absolutely right to note that much more needs to be said in its defense than I offer via my retrieval of classical realism. Nonetheless, I hoped that recalling some of its unfairly-neglected and occasionally-surprising reformist insights might play a positive role in generating a fruitful debate between IR and political theory. Our dialogue points to all the impediments that stand in the way of such an exchange.