The empiricist attitude towards the sciences can appear so very familiar and so obviously correct that it is hard to imagine what an alternative theory of science would even look like. What could be more mundane than the observation that scientists gather data and then reason, or argue, on the basis of those data for certain theories and hypotheses. Surely this is exactly what scientists do, and by doing this they achieve, we suppose, an ever more accurate and complete understanding of the world we inhabit. Admittedly, as we've now seen, even the most basic empiricist ideas face challenges. A naïve empiricist might suppose that our observations are always reliable. They aren't, but the problem of distinguishing reliable from unreliable data admits of no general solution, in part because learning to make the distinction is itself an ongoing process. The problem of induction provides further reminders of both the fallibility of scientific theories and the existence of hidden complexities when it comes to describing and evaluating methods, hypotheses and theories. Nevertheless, these problems don't obviously require the abandonment of empiricism ' a little refining might seem more appropriate.

In the first half of the twentieth century, the most influential philosophies of the sciences were empiricist philosophies. Although there were important differences in the details, these accounts shared a commitment to articulate and justify the scientific method, and somehow vanquish the problem of induction. Such articulation, it was assumed, would elucidate what scientists do and explain why sciences make progress. These empiricist philosophers relied on ideas from the philosophy of language, as well as developments in logic, and generally assumed that scientific evidence can be derived unambiguously from observation. They assumed that understanding the scientific method required no serious engagement with the history of science. These were ambitious, extensive and influential theories about how sciences work, but in 1962 Thomas Kuhn's The Structure of Scientific Revolutions appeared, and things weren't the same after that.

Kuhn's book revolutionized the ways in which scholars came to think about the sciences. It contained profound challenges to conventional wisdom on the nature of scientific rationality, scientific methods and scientific progress. Kuhn argued that we cannot responsibly ignore the history and sociology of science if our hope is to understand the nature of the sciences. Prevailing empiricist descriptions, Kuhn opined, fail to describe both what scientists now do and what past generations did.