The premise of the last three posts is that anachronism is back in fashion. By contrast, there is nothing but mustiness about the labels “internal” or “internalism.” Roughly speaking, internalist historians are interested in ideas and evidence rather than politics and institutions. A single article, let alone a single blog post, can't make internal history fashionable again. But at least it can say what a viable defense of internal history of science might look like. Here goes... Works not workers. We should try to avoid applying the label “internalist” to particular people and instead apply it to particular books and articles. This is partly for clarity's sake, since it is not uncommon for the same person to write both internalist and non-internalist works. But it is also to remind us that we should aim for variety not hybridity in the history of science (see below). Non-internalism not externalism. It is often pointed out that “externalism” is a bad label for works that are not internalist. The label conceals important distinctions such as that between explaining scientific theories in social or cultural terms, and simply describing the social and cultural side of science. The term also fails to capture the large number of “hybrid” works that are in some sense a mixture of the internal and the external. We can avoid these thorny matters by saying “non-internalist” where we used to say “externalist.” Unity not uniformity. Like non-internalism, internalism is a broad church. One author might study the technical details of a nineteenth-century optical theory, with little or no reference to outside events; another might study broad changes in the methods used in geology from 1700 and 1900, making passing reference to social and political developments. Internalism also has fuzzy borders. Is it internalist to write about Newton's application of the calculus to questions of biblical chronology, or to study Michael Faraday's metaphysics, or the experimental procedures of alchemists? We should not worry if we can't give definite answers to such questions. We live with historiographical distinctions that are no less fuzzy—try drawing a clean line between “macrohistory” and “microhistory.” And if it really was impossible to distinguish internalism from non-internalism then there would be no sense in opposing internalism, as many people have! Inquiry not intellection. Internal history of science is the victim of misleading pseudonyms. For example, internalists are sometimes described as giving “philosophical” or “epistemic” explanations of scientific beliefs. This is useful short-hand, but it implies that internalists are more concerned with a priori ideas about science than other historians are, and hence that internalists can never be as “naturalistic” as cultural historians of science. Certainly some internalists are interested in answering the questions posed by philosophers of science. But not all of them are, and in general they no more rely on present-day epistemology than social historians rely on present-day sociology. Two other unhelpful labels are “history of ideas” and “intellectual history.” Intellectual historians of science stress the role of religious, philosophical and metaphysical ideas in past science. Internal historians of science may share this interest (it depends how you define them). But internalists are also interested in experiments, instruments, and data-collection. They study scientific inquiry in general, not just intellection. Present value not past associations. It is often implied (although never stated outright) that the internal/external distinction should be abandoned because it is merely a product of Cold War polarities. This is a terrible argument. Its premise is false: internal history of science was practised long before the Cold War. And its inference is shaky: to reject internal history on the grounds that it is popular among anti-Marxists would be like rejecting the theory of natural selection because it is popular among social Darwinists. If there's really something wrong with internalism, its critics should not have to resort to guilt-by-association. Similarly, critics of internalism have to do more than point to the many internalist works that have been guilty of Whiggism or naïve inductivism or whatever. Critics have to show that those sins follow from the very nature of internalism. The reverse is also true: even if all non-internalist works published so far were sloppy and inaccurate, it does not follow that these ills are inherent in the genre. Continuity not catastrophe. Non-internalists are fond of pointing out that what counts as “internal” to science has changed over time. It is a short step to the conclusion that all internalist history is naïvely committed to present-day ideas about what science is. The non-internalists have a point. Internalists must recognise that their subject-matter—like that of historians of France or food or banking—has been defined in different ways at different times, and that at any given time there was disagreement about where its boundaries lay. But the problem is not as bad as it looks. The boundaries have not changed as much as non-internalists sometimes make out. Even if they have changed, the internalist is free to adopt the boundaries that prevailed in the time and place that interest her. Some of the best studies of the changing boundaries of science have been done by scholars who most people would count as internalists, such as Alistair Crombie and Larry Laudan. Finally, consider the histories of France, food and banking: no-one has suggested that these fields are catastrophic failures just because their subject-matter has changed over time. Quality not category. When we defend internalist works, should we focus on their quality as history? Or should we instead ask whether they count as the history of science as opposed to the history of something else? I suggest that we start with the former question because it tells us something about the scholarly value of internal history of science, rather than about which academic box it belongs to. Also, the second question tends to stack the deck in favour of internalist works, which are usually about science in a pretty obvious sense. Whichever of these two questions we answer, we should be sure to distinguish between them. Otherwise we end up with fruitless debates where internalists emphasise that they are writing about the history of science and the non-internalists reply that at least they are writing the history of science. History not philosophy. Internalist works are sometimes defended on the grounds that they help us to answer questions usually posed by philosophers of science, or because they are useful for science teachers, or simply because they are more interesting to scientists than non-internalist works. All this may be true, but it does not follow that internalist works are defensible as history. A historiographical defense of internalist works means showing that they can achieve all the normal virtues of historical writing, and avoid all the vices—or at least that they have just as good a chance of doing so as non-internalist works. Parity not priority. Internalists should give up arguing for the priority of their approach, as history, and instead argue for parity. (This leaves open the possibility that internal history is better as history of science). Parity is an obvious goal but it is rarely defended in print. Instead, some commentators complain that too much attention is given to one side of the internal/external distinction. Most other commentators urge us to transcend the distinction by writing “hybrid” works that collapse the distinction. Variety not hybridity. There's nothing wrong with hybrids (I write them myself). What I object to is the idea that that such works are inherently superior, as history, to works that are either strictly internalist or strictly externalist. The usual argument is that only hybrids do justice to the inextricability of social and epistemic factors in past science (or something along those lines). The obvious reply is that there are lots of other inextricable pairs that the historian should try to knit together—like theory and experiment, or mathematics and physics, or science in France and science in Germany—the list is endless. Given that no single work can hybridize everything, why should the science-society dyad be given special attention? To conclude, if we really want to “transcend” the internal/external distinction, we should stop measuring works according to where they stand with respect to that distinction. There are a variety of legitimate genres ranging from institutional history through trendy hybrids to hard-core internalism, and each of these genres can (and has) produced works of great care and creativity. Expand post.