Wednesday, June 12, 2013

Hacking on Kuhn


The fourth edition of Thomas Kuhn's The Structure of Scientific Revolutions appeared in 2012, fifty years after its original appearance in 1962. This edition contains a very good introduction by Ian Hacking, himself a distinguished philosopher and philosopher of science. So it is very interesting to reread Kuhn's classic book with the commentary and intellectual frame that Hacking provides. (Here is an earlier post on Kuhn; link.)

Hacking's reading is somewhat deflationary, compared to the relativist and anti-rationalist interpretations that are sometimes offered of Kuhn's theories. Hacking sees a great deal of continuity between the Vienna Circle traditions of philosophy of science and Kuhn's own intellectual commitments about scientific rationality. (Hacking pursues this analogy even down to noting a parallel between Carnap's title Logical Syntax of Language and a similar description of Kuhn's later work, Logical Syntax of Scientific Language.)

According to Hacking, the scope of the concept of "paradigm" has been exaggerated by subsequent interpreters. Paradigms are not systems of thought or conceptual systems; they are not even discipline-specific sets of shared assumptions that don't get questioned in the ordinary pursuit of scientific knowledge. Instead, Hacking argues that Kuhn's intended meaning sticks fairly close to the classical meaning of the term, as an exemplar of something or other. He quotes Kuhn:
"The paradigm as shared example is the central element of what I now take to be the most novel and least understood aspect of this book."  (from the Postscript, kl 213)
So a good example of a paradigm in science is something like the Millikan oil drop experiment; it constituted a clear and admirable example of experimental design and implementation which helped to guide later experimentalists in the design of their own experiments.

Hacking also notes that Kuhn later allows for a local use and a global use of the concept, but he suggests that Kuhn did not wholly endorse the global use. Here is how Hacking paraphrases the global use, in the context of the things that hold a scientific research community together:
That's the global sense of the word, and it is constituted by various kinds of commitment and practices, among which he emphasizes symbolic generalizations, models, and exemplars. (Kl 318)
Hacking gently suggests that Kuhn under-values "normal science," because he shares a bias towards theory with many other philosophers of science of the preceding generation. But Hacking argues that later philosophers and historians of science, such as Peter Galison, have given more weight to the innovations associated with experimentation and instrumentation (kl 199), and the process of normal science is precisely the context in which innovations in these aspects of science are most likely to occur.

Hacking makes an interesting point about the scientific context in which Kuhn's ideas took shape. Physics, both classical and modern, set the standard for what was most exciting within the scientific enterprise in the 1950s and 1960s. But Hacking asks an interesting question: what if the examples of biology and the life sciences had been the backdrop against which Kuhn had formulated his theories? Molecular biology and the chemistry of DNA constituted a revolution in biology at roughly the time of the original publication of SSR. How valid are Kuhn's observations about scientific research and progress against that backdrop? Would the results have perhaps been quite different if he had concentrated on these examples?
Thus The Structure of Scientific Revolutions may be -- I do not say is -- more relevant to a past epoch in the history of science than it is to the sciences as they are practiced today. (Kl 98)
Hacking gives a very succinct summary of Kuhn's main theory of the course of science:
Here is the sequence. (1) normal science ...; (2) puzzle solving ...; (3) paradigm, a word which, when he used it, was uncommon, but which after Kuhn has become banal ... ; (4) anomaly; (5) crisis; and (6) revolution, establishing a new paradigm.  (Kl 114)
Hacking thoroughly rejects the most subjectivist aspects of many readings of Kuhn: the idea that scientists inhabiting different paradigms also literally inhabit different worlds. Hacking doesn't believe that Kuhn actually believes this, or even unambiguously asserts it.

There are many passages in Kuhn's original text that are worth pulling out again. Here is one, on the gap between observation and scientific beliefs:
Observation and experience can and must drastically restrict the range of admissible scientific belief, else there would be no science. But they cannot alone determine a particular body of such belief. An apparently arbitrary element, compounded of personal and historical accident, is always a formative ingredient of the beliefs espoused by a given scientific community at a given time. (4)
Effective research scarcely begins before a scientific community thinks it has acquired firm answers to questions like the following: What are the fundamental entities of which the universe is composed? How do these interact with each other and with the senses? What questions may legitimately be asked about such entities and what techniques employed in seeking solutions? (4)
These passages make it clear that Kuhn does in fact think that a scientific community possesses a set of unifying but contestable  beliefs -- what many of us now mean by a paradigm. And this seems more pervasive and comprehensive than Hacking's analysis would seem to allow.

I first read Kuhn as an undergraduate in 1969 or 1970, and I confess that my own understanding of his meaning concerning scientific knowledge and reasoning gravitated towards the more anti-objectivist reading that Hacking rejects. I understood paradigms as sets of semi-articulated assumptions about science, the world, and the instruments that hung together as a community-dependent worldview; a worldview that could not be directly empirically evaluated. And I understood incommensurablity to mean that scientists within these mental frameworks arrived at empirical judgments and theories that could not be strictly compared across communities, because their underlying conceptual structures were systematically different. I had read Quine on the indeterminacy of translation at roughly the same time, and I understood incommensurablity in analogy with indeterminacy across language communities. (Kuhn's preface to the book makes it clear that he too had read Quine, though in the 1950s and therefore prior to the publication of Word and Object (1960); kl 550.)

I also understood Kuhn to hold that standards of scientific reasoning were likewise dependent on the mental frameworks of the research communities -- with the result that some disagreements among physicists or biologists could not be resolved on the basis of standards of scientific reasoning or method. There was no "paradigm-independent" scientific method, no community-neutral standard of rational preferability.

It may be that Hacking is right, and that Kuhn never intended to support these radical claims about the limits of scientific rationality. But whether he did or not, the position is an intelligible one, and thinkers as diverse as Althusser and Feyerabend have advocated it. Frederick Jameson's title, The Prison-House of Language: A Critical Account of Structuralism and Russian Formalism, picturesquely captures the core idea.

Structure of Scientific Revolutions richly rewards a rereading fifty years after its original publication. And as is true of so many deeply original works, we are likely to find different things most striking today than we did on first reading decades ago.