Kuhn 5: First General C&Q Thread

[koganbot]

OK, start talking. This is what I'm calling the First General Comments And Questions (C&Q) Thread. The article under consideration is Thomas Kuhn's "What Are Scientific Revolutions?" which you can find your way to here, pp 13 to 32. I'm asking six questions but you can ask your own as well, and you don't have to answer mine 1, 2, 3, 4, 5, 6 separately in that order (though you can if you want). You should have a good idea how to answer the first three, however. Questions 4 through 6 are generated by the essay (at least in my mind), but they're not specifically asked and definitely not answered in it.

1. Kuhn starts the essay by distinguishing between "normal" and "revolutionary" scientific development, saying that the former is "cumulative" and the latter "noncumulative." He uses the metaphor of the brick: "normal science is what produces the bricks that scientific research is forever adding to the growing stockpile of scientific knowledge." Not sure he isn't mixing his metaphors here, "brick" and "stockpile," since I think he means the brick metaphor to give us a picture of bricks being layered atop one another to build an enduring structure.

What do you suppose he means by "cumulative" and "noncumulative" change? Surely he's not saying that in a science that's undergoing normal change there are no widely believed ideas that turn out to be in error. So what's the difference between a normal correction and a correction that leads to a revolution? What's the difference between a normal new idea and a revolutionary new idea? Is the idea that Pluto is not a major planet a revolutionary idea? How about the idea that the demise of the dinosaurs 65 million years ago was not drawn out but was one of the relatively sudden after-effects of a strike by one or more meteors?

2. What's a paradigm and what's a paradigm shift? Kuhn doesn't actually use the terms in this essay; but nonetheless, the revolutions he describes are paradigm shifts. What do you think he means by "paradigm" and "paradigm shift"? (If you click on the "thomas kuhn" tag up above you can find your way to some previous talk on the subject. He uses the term in both a narrow and a broad way, and it's good to be clear on the difference.)

3. What does Kuhn mean by "incommensurability"? This is another term that doesn't appear in the piece, but the concept gets well-described in it. In a different essay he uses the words "residue" and "loss" in association with "incommensurability." What's the residue? What's lost?

4. Is Kuhn's conception of "normal science" a good one? Are there really periods when a science undergoes no noncumulative adjustments, where all the basic terms are at ease with themselves and with each other?

5. Do any of the nonsciences* have equivalent periods, or is what Kuhn is saying is "normal" in normal science not normal elsewhere?

6. If competing paradigms are incommensurable, how does one choose between them? If Kuhn's model is right, an entire field can and sometimes does end up abandoning one paradigm as wrong and embracing another as right. How does it do so? Another way of putting the question - one that obviously doesn't just apply to the sciences - is: if different premises (and their related models and frameworks) generate different "facts," facts that support the premises, how do you go about testing your premises and, when there are competing, incompatible premises, how do you choose one set of premises over another? Is there a rational way of doing so, or is this really just a matter of taste? How would you test the contention that motions or changes must have endpoints, or the competing contention that motions or changes need not have endpoints?

*For example, math, psychology, music criticism, art, politics, situation comedies, girls night out, organized sports, etc. etc. etc.

Comments

[petronia.livejournal.com]

I'm going by my own feel for these words first, before analyzing Kuhn's text (which in any case I'm finding difficult to do within the parameters you laid out ^^;): to me a "paradigm" is a trick of the mind, where a "model" is external - I won't use the word objective... An open cube made of sticks is a model. A Necker Cube drawing is also a model, but the brain's interpretation of the Necker Cube's conflicting perspective is paradigmatic. The corner points either inward or outward; it's not merely difficult to hold both views at once but actively inpossible. The mind is trapped by its own design, a biological computer optimized to match input with preexisting stored patterns... IOW I almost feel like the usefulness of the word is more in describing the scientist than in the science, that if there were no mental conflict (i.e. in periods of cumulative change where what came before can be assumed) there would be no need for the concept at all. The battery example fits with my definition, I think, because it pretty much is an optical illusion. (A misnomer, as they have nothing to do with optics.)

[petronia.livejournal.com]

Also: this is how I would generalize the concept to social sciences as well, and to the arts. Psychology and economics certainly have/had paradigms... With art it is a little different, less axed on theories re: what is happening than on systems/hierarchies of value judgment, perhaps... i.e. I wouldn't want to call the rise of abstraction in the visual arts (say) a paradigm shift so much as a change in an entire nexus of beliefs concerning originality, intellectual property, tradition, progress...

[skyecaptain.livejournal.com]

Since I can jump ahead and speculate wild 'n' recklessly here, I should just state flat out that one reason my readings haven't been super close so far is because I keep thinking about how to or not to apply this stuff to not-science. (A teacher of mine once told me, "You always jump straight to the end. What must your girlfriend think?") Specifically to discussions of culture -- my feeling so far is that Kuhn's ideas of revolutionary change aren't applicable to matters of culture, because the nature of the change in cultural matters is always cumulative, and there are no systems of understanding (through language) that are "violated" in order to include some new idea. (I like Kuhn's use of "univocal" to describe our understanding of a given word or concept in only one way, so that the Ptolemaic planet and Copernican planet can't coexist as one word, "planet," with one meaning in the same sentence.)

Social ideas co-exist, if not peacefully, then at least in ways that aren't categorically incompatible. But I am encouraged by how wrong I have been so far and haven't given up hope that I may well be quite wrong now, too. I say "hope" because I would LIKE to be able to (1) understand Kuhn better and then (2) apply to stuff that, frankly, I like talking about anyway. But I'll try to keep Dick 'n' Janing it for the time being...

[petronia.livejournal.com]

I'm also a jump-straight-to-the-end-then-fill-in-the-middle type! That's why the start-with-basic-definitions approach is hard for me.

Mmm I see what you mean... I admit the examples I was considering when I wrote the above have nothing to do with progress/change, per se, but rather cultures/value systems coexisting in time if not space. (And that by analogy only with the process of discovery described by Kuhn, going from superficial "wrongness" - "Why are you being knowingly rude to me?" - to a realization of fundamental and systematic differences - "Oh, it's not that you don't respect my property, it's that we don't have the same definition of ownership or what causes it to be transferred.") But this was the idea I was groping at in my comments anyway: that a "paradigm" can only be recognized/defined insofar as it comes into conflict/contrast with another. Otherwise you're fish trying to describe water. Aristotlean physics only came into existence as the destruction was underway... I think the idea that social paradigms can coexist when scientific paradigms cannot is a paradigmatic belief in and of itself, i.e. that when one is able to hold both social models in one's head (and move between them) it makes them both somewhat/possibly true, whereas understanding both scientific models doesn't make them both true. But both the relativism of the former statement and the objectivism of the latter are modern ideas.**


** I'm likely to have sinned here in using words without being certain about their meanings, as koganbot has previously griped. XD;

[koganbot]

because the nature of the change in cultural matters is always cumulative, and there are no systems of understanding (through language) that are "violated" in order to include some new idea

I don't understand this, and I can't make sense of your use of "cumulative." I'd say that what Kuhn means by science not being cumulative is that something of significance is lost: old meanings of words, old ways of behaving and connecting ideas, old ideas of what is like what, etc. This is every bit as true in matters of culture. E.g., when you go out dancing, you don't become the vehicle for supernatural spirits called "moas," do you? So something is lost in the "translation" from the dance's African antecedents* to your own dance. Wiki: "Alternatively, a language is said to be extinct if, although it is known to have been spoken by people in the past, modern scholarship cannot reconstruct it to the point that it is possible to write in it or translate into it with confidence (say, a simple dialogue or a short tale written in a modern language)." That certainly seems to be a case of cultural loss. And Kuhn's got a section of one of his essays where he talks about concepts in French that have no equivalent in English, so something is lost in translation. And Native American culture wasn't absorbed fundamentally unchanged into modern North American culture, was it? Whole social forms were obliterated, belief systems, etc. And where the beliefs have survived, they're not necessarily compatible with Christianity and atheism and so forth. Just because ideas exist at the same time as others and that some - a few - people are willing to understand the ones they don't subscribe to doesn't make the ideas compatible. And though astrology still exists, its ideas have been "lost" by modern astronomy. And if you want to say, "Oh, but astronomy is a science, but astrology hasn't been lost to the culture," this doesn't make a point. Astronomy is part of culture too, but astrology isn't part or astronomy. I'm an atheist, and atheism isn't a science (or an ism, actually), but astrology isn't part of my beliefs. Unless your criterion for "cumulative" is "exists on the same planet at the same time," I don't see how culture gets to be any more cumulative than science (and even there, lots of culture is no longer part of the planet).

I think you're confusing (1) conflicting ideas may nonetheless coexist within a field or enterprise (which I think is probably always true in all fields and enterprises that aren't sciences), and (2) those conflicting ideas therefore aren't incompatible, since they co-exist (this is false), (3) nothing is lost as such fields and enterprises change over time, since conflicting ideas can coexist within those fields (this is false too, since the fact that some ideas coexist doesn't mean either that those particular ideas will continue to coexist or that there aren't other ideas that haven't coexisted or can no longer coexist, etc.).

*I'm not saying that all the antecedents are African, just that the dance has some African ancestry.

[koganbot]

Astronomy is part of culture too, but astrology isn't part or astronomy

Astronomy is part of culture too, but astrology isn't part OF astronomy

[koganbot]

I wouldn't want to call the rise of abstraction in the visual arts (say) a paradigm shift so much as a change in an entire nexus of beliefs concerning originality, intellectual property, tradition, progress...

Why wouldn't you call that a paradigm shift? This depends what you mean by "paradigm shift," of course. But look at what Kuhn says about Aristotle's concepts in physics. They seem to belong to a nexus as well. Get rid of one belief - motion as a change in quality - and the whole thing unravels.

Bear in mind that for Kuhn, "paradigm" as in a "model" or "exemplar" doesn't mean the same thing as "paradigm" as in a "disciplinary matrix," and it's a broad shift in the latter that he's calling a "paradigm shift" (though obviously in a scientific revolution the former will undergo change as well).

But then, bear this in mind too: Aristotelian physics is gone, is history, is done for. Whereas representational painting did and does exist in fine art painting at the same time as abstract painting does. Yet, e.g., though astrology still exists, it's no longer remotely a part of astronomy. But cookery and botany still not only coexist in near and cheerful proximity, cooks draw on knowledge gained from botany and biology, while still considering a tomato a vegetable and not a fruit for the purposes of cuisine.

(Not sure what point I'm making, except that certainly models exist in the nonsciences and nexuses exist in the nonsciences as well, and broad shifts exist too, and incomprehension across different discourses. And the shifts may be just as great and disruptive. But I don't think (though I don't want to say this dogmatically, since I don't know) that we get anything remotely like scientific revolutions, where one matrix/nexus has complete consensus, and concepts are fundamentally stable, then there's the period of revolution, with one nexus unraveling but another not fully formed, and competing incompatible ideas, etc., and then after the revolution we're back to consensus, stable concepts (but many of them different from what you had before), a new matrix, nexus, etc. but with enough precision of concepts and measurements so that when enough anomalies occur that can't be explained away they can set off a crisis and possibly provoke a new revolutionary period. [Of course, it's not necessarily true that we get this in the sciences, either, just because Kuhn says that we do.])

[petronia.livejournal.com]

I think you misunderstood my sentence, but only because it was constructed terribly. XD; It really should have been something like

I wouldn't want to call the rise of abstraction in the visual arts (say) a "paradigm shift", so much as I would reserve the term for a change in an entire nexus of beliefs concerning originality, intellectual property, tradition, progress - a coherent example of which is not currently coming to mind.

I am only groping toward what I really mean here. Something along the line of: the appearance of abstract painting in Western fine arts is only a symptom, what really changed was the system of values underlying the artists' work, i.e. the definition of what it means to "paint" (or create art) changed, and thus the activity and its goals naturally broadened to take in strategies unavailable under the definition that previously held sway.

[koganbot]

We're almost up to 49 comments, after which the fuckers at lj start collapsing responses so you have to click all over the place to see what anyone is saying, which I hate, so in a few minutes I'm going to be posting an all new freewheeling dance-in-the-sky Kuhn q&c thread so that we can continue the discussion there.

[koganbot]

You win an award for clairvoyance.

Necker Cube


Not quite Necker Cubes, but the same principle

[petronia.livejournal.com]

W00t! XD

[koganbot]

(take two)

but the brain's interpretation of the Necker Cube's conflicting perspective is paradigmatic

Yes! But this applies every bit as much to the open cube made of sticks, if that cube made of sticks is being used as a model - i.e., if it is being used as a representation of something else or a prototype that can be used to create something else. Which is to say, if you point to the cube made of sticks and say, "Use that as a model," with no elaboration, the person you're instructing will either have to already know what you mean, on the basis of past interaction with you or on the basis of the standard practice among her and your social set or her and your profession of what cubes of sticks are models for and how to apply them, or she will have to make an educated guess or a wild guess. How to use a model isn't written into the model's physical features. And even if the person you're talking to is a structural engineer and she knows that the cube of sticks is part of a scale model of some structure she's supposed to build - say the cube is to be a structure on the front lawn of The Institute - what materials to use and how to construct them for an object that's 20 feet by 20 feet rather than 3 inches by 3 inches is something she'll only know from training and experience.

I think you're creating an unnecessary problem for yourself in your distinction between "internal" and "external." Are social practices internal or external? - my point here being that using something as a model is almost always a social practice, the result of specific training with that particular model or something like it, or a skill you've picked up in the course of your life. It's a matter of culture, and I don't think it's useful to worry about whether culture is internal or external. It starts external, when you're a baby, but you have to internalize it; but you haven't successfully internalized it if you can't recreate it in your observable behavior, in the world.

(Even if, let's say, you're the first person to model your derivation of your law of physics on someone's derivation of his law of physics, this comes from your experience of doing derivations and of your knowledge of yet other people modeling their derivations on someone else's derivations, etc.)

[koganbot]

it's not merely difficult to hold both views at once but actively impossible

Kuhn has made this exact point, probably referring to a duck-rabbit or something of that sort rather than a Necker Cube, but the point is the same. An analogy to the Necker Cube would be this:

Let's say an Aristotelian points at an apple falling from a tree and says, "That is an example of motion; use that as a model and go point to other examples of motion." And let's say a Newtonian points at an apple falling from a tree and says the same thing. Each will expect different behavior, will see the apple falling as representing a different set or species of phenomena and expect you to see (or to learn to see) what he sees. The Aristotelian will want you to be able to identify an apple growing into an tree and a man being restored from sickness to health as examples of motion, similar to the apple falling towards its place in the center, whereas a Newtonian would not. And of course one can learn both, learn to see what an Aristotelian sees and learn to see what a Newtonian sees, just as one can learn to see the Necker Cube as tilted diagonally down from right to left or diagonally up from left to right, but one can't see the same cube doing both at the same time. And you can't apply the falling apple as a model in a Newtonian way and an Aristotelian way at the same time to the same phenomenon.

(At least I don't think you can. I don't want to be dogmatic about this. A botanist and his student can be in the produce section of a grocery store and the botanist can point to the tomato and explain why in botany the tomato is considered a "fruit," and can explain this in terms of the tomato's role in plant reproduction, and at the same time be musing to himself about the price of the tomato and how it seems to match last week's price and so the price of vegetables seems no longer to be rising - he looks at the broccoli and this seems to confirm his hypothesis - whereas the price of fruit (he's looking at oranges and bananas) does seem to continue to be going up. One can walk and chew gum at the same time. But I'd say here he is compartmentalizing his thoughts, applying one set of criteria with one part of his mind to one problem and another set of criteria with another part of his mind to another problem.)