Samuel Barkin offers a succinct and fair account of my book. He then challenges its fundamental premise: that we should focus more on epistemology than method in organizing the field of international relations. He has two principal objections. Any dichotomy – or other classification scheme – narrows the field by excluding research that does not neatly fit in its categories. My dichotomy includes approaches that do not really fit into either of my epistemologies.

To understand any field, we must introduce categories to define research programs, their goals, and methods. Categorization is valuable to the extent it tells us something useful about the field. My focus on positivism and interpretivism encourages practitioners to step back and look at their own research in comparison to others in their own and the other epistemology. How is it similar? Where does it differ? What can we learn about ourselves and our research by making these comparisons? Reflection of this kind can also improve our research by making us aware of our epistemological foundations and their problematic nature.

Some people – I hope Barkin does not count himself among them – assert that categories and comparisons are by their nature exclusionary. On the most superficial level this is true: something fits in a category, does not fit, or fits only in part. We cannot negotiate the world, let alone our research, in the absence of comparisons and the categories that make them possible. The same holds true for ethical judgments, which are equally dependent on categorization. Exclusionary or negative depictions are sometimes justified, as is true with research that does not meet the standards of its own tradition. When possible, we should try to use analytical categories in positive or at least neutral ways. We can do this by affirming value across categories. This is what I do with positivism and interpretivism. I gave historically rooted and, I hope, accurate accounts of their epistemological and substantive assumptions, reasons why people are drawn to them, and how they seek and claim to have produced knowledge. I am more favorable to interpretivism, but offer critiques of both epistemologies and approaches all or in part nested within them. I contend that scholars who work in either epistemological tradition often face similar problems and can learn from one another.

Barkin accuses me of including research approaches or programs that do not fit, or totally fit, in either epistemology and of excluding some that are outside of them. Barkin offers the examples of neorealism and rational choice. I contend that Waltz was unabashedly positivist in his search for a universal, parsimonious theory with predictive value. Rational choice is more complex. It is positivist in the sense that it seeks to explain and predict behavior and assumes that politics can be studied as a science. As biologists would be the first to acknowledge, typologies never map perfectly on to the world. They can tell us much about what we are studying. The discovery in the field of IR that approaches and research programs sometimes fit an epistemology only in part, or bridge epistemologies, says something important about them, the diversity of our field, and the nature of the epistemologies.

Barkin alleged that I delegitimize research – feminism, for example – that does not fit into either epistemology. I do nothing of the kind. I note that at the outset that my typology does not encompass everything and that some research traditions or programs fit my two epistemologies only in part. Much feminism is interpretivist, some less so. I do not use any feminist examples, because it is not associated with a distinctive research method. It is unfair of Barkin to criticize me for casting my epistemological nets broadly and at the same time criticizing me for including approaches that show only a partial fit. I offer broad readings of positivism and interpretivism to make them as inclusive as possible, something he should regard favorably.

Barkin maintains that my account of positivism is so broad that it ends up “a list of research programs.” I am very clear about the defining characteristics of positivism, devoting two chapters and parts of others to the subject. I describe and evaluate research programs in both epistemologies to illustrate the kinds of problems they encounter in practice. I show that interpretivism has more diverse intellectual roots than positivism; they include German Romanticism and historicism, neo-Kantianism, Max Weber, and Wittgenstein and the linguistic turn. Not surprisingly, interpretivism has generated more diverse research programs. By narrowing my definition of what fits as interpretivist I would be committing the very sin Barkin is so keen to avoid.

The comparison to King, Keohane, and Verba is unwarranted. They offer a parody of constructivism, and of positivism. They provide a very broad definition of the former and a remarkably narrow one of the latter, both with the goal of discrediting any approach that differs from theirs. My account is broad to make it inclusive. When I offer judgments, they are not made on the basis of some claim to know the right way to do things, but rather with reference to what the two epistemologies and associated approaches claim they can deliver.

Barkin also disapproves of my jumping from epistemological to empirical evaluations. However, they are closely connected. We can evaluate an epistemology on the basis of its internal logic and structure or how it enables useful discoveries about the world. I do both, and it is a legitimate exercise because the purpose of positivism and interpretivism is to generate knowledge. They also incorporate substantive assumptions about the social world.

I regret that Barkin does not engage my final chapters on reason, cause, mechanisms, and social science as an ethical practice. They are the most interesting, important, and original parts of the book.