This semester, at the initiative of a few undergraduate students, with me leading the reading, we organized a reading group on Kuhn’s *Structure*. By last week, we finally finished *Structure* (skipping the postscript), and starting this week the reading group has begun *Being and Time*.
This time around, leading *Structure* was quite a coincidence: just when Professor Wu went abroad, the undergraduate students in our department’s history and philosophy of science project were left without anyone to supervise them. These two students were especially eager to learn and spontaneously wanted to start a reading group, asking Professor Wu to find a teaching assistant to lead it. Professor Wu knew that I have always been happy to “recruit” younger scholars, so he naturally handed them over to me.
Reading *Structure* suited me perfectly as well. I have always thought that *Structure* is a very suitable book for those moving from complete outsiders to beginners in the history and philosophy of science. Of course, if one were reading on one’s own, books such as Chalmers’s *What Is This Thing Called Science?* or Chen Jiaying’s *Philosophical and Scientific Common Sense* might be somewhat better. But if it is a reading group format, *Structure* is the most suitable. That is why, back when I was running the Xindao Salon, I had also once imagined getting people together to read *Structure*; unfortunately, in the end the salon had fewer and fewer participants, and there was no stable membership, so nothing came of it.
The reason I say *Structure* is well suited to a reading group is, first of all, of course, that it is worth reading, and it is neither too difficult nor too simple—that is to say, even someone without much background can, with some effort, read it through, yet it is not something one can breeze through without effort. Added to that, it is after all a classic one cannot not read, so once you finish it you can feel a certain sense of accomplishment. Second, although *Structure*’s argument is plain enough, its views are rather subversive for contemporary students in their default configuration; this challenge to “common sense” is a shortcut to entering an academic state as quickly as possible. Finally, perhaps most important of all, *Structure* is quite suitable for riffing, or so at least it is for me. In fact, that is exactly how this reading group proceeded: basically, after every one or two passages, I would toss out a few comments. What is meant by riffing here, in Chinese terms, is playing the foil; in more academic terms, it is roughly “congratulatory explication.” Basically, the lead reader says a couple of lines and I add one, either to point out the key point, reveal the strangeness or profundity within it, or to interject in agreement, making the context feel more vivid.
Kuhn’s book is written quite vividly as well. Speaking of which, Teacher Tian has turned the writing course in his department into a Chinese language course, and I am not entirely satisfied with that—I feel it is too unacademic. Still, the three writing principles Teacher Tian emphasizes do make sense: the so-called “tell stories, give examples, make analogies.” In fact, academic writing is nothing more than these three things. “Tell stories” refers to the overall structure of a book or article, its logical thread, the rhythm of its exposition, and so on. Good works need to unfold step by step and speak in a measured, engaging way, but they also need to leave room for foreshadowing and design some passages that are “unexpected yet entirely reasonable.” “Give examples” refers to rich concrete instances, while “make analogies” includes all kinds of vivid comparisons or metaphors. In fact, even the most serious philosophical works still have moments of giving examples and making analogies; it is just that we generally tend to read texts by “summarizing and drawing conclusions,” and after reading them we produce ready-made, formulaic conclusions such as “the author has points one, two, three, four, five,” thereby missing the livelier things in the work. For example, when reading Aristotle’s philosophy of nature, merely noting the first, second, third, and fourth of the four causes is not very meaningful, whereas Aristotle’s technique of using artifacts as examples is perhaps more worthy of pondering (this is also Professor Wu’s line of thought).
And *The Structure of Scientific Revolutions* is precisely such an extraordinarily well-executed work of “telling stories, giving examples, and making analogies.” The plot is coherent and fluid, the narration unfolds in a gentle, unhurried manner, there are foreshadowings and callbacks, and there are abundant cases as well as vivid comparisons.
Kuhn is also not like some big-name figures who are wildly freewheeling and eclectic. Kuhn basically plays by the rules: he cites a fair amount, and his reasoning proceeds step by step. In other words, what Kuhn presents is basically a mode of writing that beginners can imitate. As for the Heidegger and McLuhan types, although beginners can certainly read them too, let’s take it slow when it comes to imitation.
What is meant by the structure of scientific revolutions is a model of scientific development such as “normal period (anomaly) — crisis — revolution — new normal.” This model can be compared to the “punctuated equilibrium” model in the history of science, and it opposes the traditional Whiggish, linear, cumulative model of history. Kuhn believed that only in normal periods, or periods governed by a stable “paradigm,” can science properly be called cumulative; around revolutions, however, paradigms shift and standards of evaluation shift as well. Different paradigms are incommensurable; in other words, there is no common standard by which to measure science under different paradigms. Therefore, a scientific revolution is not a simple cumulative development, but a subversion (only in retrospective textbook reconstructions does revolution also get cast as progress).
A paradigm is not a set of written doctrines, or a collection of laws or data, but rather the mindset, patterns of behavior, and so on of a scientific community—the “customs” that every novice must undergo through spoken and embodied instruction, through constant immersion, when entering that community. The stability of normal science corresponds to the closure of the scientific community: the more unified and specialized the community is, the more science resembles science; the more the community is full of disagreement and a mix of all sorts, the blurrier the boundary becomes between science and other traditions such as philosophy and religion.
The central point of *Structure* is expressed very clearly. Basically, after reading only a little over half the book, we can already more or less grasp Kuhn’s core insight. Yet the whole book never feels cumbersome, because Kuhn unfolds things carefully, step by step, from the various stages before and after the revolution through a large number of examples, often with hidden barbs. If you have not accepted Kuhn’s insight, these examples will keep unsettling you; if you already have accepted it, then you will be able to smile knowingly from time to time, making the reading quite enjoyable.
Besides the abundant examples, Kuhn’s analogies are also very important. A common analogy is to use the Gestalt-switch “duck-rabbit” figure to compare how the shift in worldview during paradigm change is holistic. The most important analogy, however, is to compare normal science to a “puzzle-solving game,” especially like a jigsaw puzzle. A jigsaw puzzle always has a fixed set of rules; every new experience is grasped from within a certain established perspective and integrated into the framework of knowledge according to fixed rules. Even if some pieces cannot be properly placed for the time being, the puzzle game does not question the legitimacy of the rules themselves, much less does it lightly smash the existing picture. Many times, puzzle-solving even presupposes knowing the answer in advance. Under normal science, scientists often know in advance, before conducting inquiry, what they are going to obtain, or at least know roughly what it will look like. But just as a jigsaw puzzle can still be challenging even when the result is known in advance, normal science can also be exciting.
However, when the “anomalous” pieces that cannot be properly placed accumulate more and more, normal science enters a state of “crisis.” There is no clear boundary between crisis and anomaly. At any moment in normal science, science faces anomalous phenomena; this is precisely the object that puzzle-solving activities need to work hard to crack. If there were no anomalies waiting to be solved, there would be no normal scientific activity either. But once anomalies accumulate to a certain extent, some researchers gradually lose patience, to the point that they try to re-position the fragments of knowledge in ways that the original rules did not allow—for example, allowing pieces to overlap, or assembling them on a spherical surface—and a previously unimaginable new picture slowly emerges. This new picture is often initially vague and indistinct. It may integrate many anomalous fragments that for a long time could not be placed, yet at the same time it makes some fragments that were once harmoniously integrated become nowhere to fit. Some details that had originally been ignored become important, while some pieces that had been handled with great seriousness may be judged to be disturbances that ought to be excluded. In short, it is very hard to find a simple and clear standard between the old way of assembling and the new way of assembling by which to measure which one is bound to win out.
The process by which scientists move from the old paradigm to the new one is compared to religious conversion. It is difficult to say which ironclad piece of evidence forced him to decide to convert; psychological, cultural, and faith-related factors all play a role. And the emergence of the new worldview is also like a jigsaw puzzle: evidence accumulates piece by piece, but there is no clear standard for when that accumulation suddenly makes the overall shape visible to people (for example, that it is an apple). Someone may see the clue after only three pieces, while someone else may still be vague even after thirty pieces are assembled. But in any case, once the new paradigm is established, the new world always replaces the old picture as a whole.
In addition, Kuhn’s introduction of the concept of scientific revolution is intended to draw an analogy with “political revolution”; I wrote about this in my article “What on Earth Is a Scientific Revolution?” back in the day.
Also, although Kuhn did not explicitly mention “punctuated equilibrium,” he did consciously compare scientific progress with biological evolution. The history of biological development described by Darwinian evolution is “directional but not goal-directed,” and scientific history is the same. Each revolution cannot simply be said to be a process of moving toward some predetermined goal (such as objective truth), yet science is still constantly advancing forward; this “forward” direction is determined through historical retrospection.
Finally, one point worth mentioning is that Kuhn’s theory is reflexive: he explicitly applies the theory of paradigm revolution to his own theory. He believes his theory is a revolution in relation to the traditional epistemological philosophy of science. He clearly recognizes that he cannot persuade traditional philosophers of science by appealing to standards recognized by both sides, but he believes that things considered anomalous and difficult to explain in traditional philosophy of science become, in Kuhn’s account, the basic definitions that go without saying.
Translated from the Chinese original with AI assistance. The original text is authoritative.
Leave a Reply