Page 1 of 1

The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Thu Dec 08, 2016 6:50 am
by Necromancer
Anybody reading The Structure of Scientific Revolutions by Thomas Kuhn?

Although it's intriguing to read about the great knowledge of science that's provided in this book, I feel that the book to some extent ignores two entities and these are:
* Libraries of scientific facts
* Libraries of scientific theories and the corresponding anomalies
From this I also feel that the SSR ignores people's possibility to choose the best from these theories and that the best also represents the latest for the informed scientist/layman/student. It is actually possible to evaluate scientific theories and to read about the anomalies from these theories. Then, who wants to turn to an older theory for real no matter how separate from society?

In the Section I: A role for history, p. 8, in the book, Kuhn admits he breaches the divide between "the context of discovery" and "the context of justification". This mixes the facts with the theories in his book, something I think necessitates his story on history of science.

Anybody who wants to share their views?

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Thu Dec 08, 2016 7:03 am
by surreptitious57
Necromancer wrote:
Anybody reading The Structure of Scientific Revolutions by Thomas Kuhn
No but definitely a book I have every intention of buying as soon as I can

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Thu Dec 08, 2016 9:45 am
by uwot
Necromancer wrote:Anybody reading The Structure of Scientific Revolutions by Thomas Kuhn?

Although it's intriguing to read about the great knowledge of science that's provided in this book, I feel that the book to some extent ignores two entities and these are:
* Libraries of scientific facts
* Libraries of scientific theories and the corresponding anomalies.
I'm not sure how far you are into the book, and I don't want to spoil the ending, but the thesis is that science is in effect a library of scientific facts. These facts are empirical observations, massive objects fall to the Earth, for instance. Scientific 'theories', more accurately hypotheses, are the inferences drawn from the facts to explain them. Kuhn used the word paradigm in subtly different, and sometimes inconsistent ways throughout the book, but the general idea is that a paradigm is essentially what people often mean by theory. A paradigm, such as the geometric model of the universe, is believed until it reaches a crisis, because of a build up of anomalous empirical observations, until the model is rejected in favour of a new paradigm, that the sun is the centre of the universe for example.
Necromancer wrote:From this I also feel that the SSR ignores people's possibility to choose the best from these theories and that the best also represents the latest for the informed scientist/layman/student.
Imre Lakatos criticised it for exactly that. He pointed out that not every scientist accepts a given paradigm and that there are different 'research programs'. The issue is called the Duhem-Quine thesis, which basically states that for any set of empirical facts, there is no limit to the theories/hypotheses/paradigms that can be invented to explain them. So yes, people do in fact choose from the best of these theories, but it doesn't follow that the latest is necessarily the best.
Necromancer wrote:It is actually possible to evaluate scientific theories and to read about the anomalies from these theories.
You might be interested in looking up Peter Lipton and IBE, inference to the best explanation. In effect he was defending the rationalist position. The big cheese in the empiricist camp is Bas van Fraasen.

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Thu Dec 08, 2016 1:37 pm
by Necromancer
uwot wrote:Imre Lakatos criticised it for exactly that. He pointed out that not every scientist accepts a given paradigm and that there are different 'research programs'. The issue is called the Duhem-Quine thesis, which basically states that for any set of empirical facts, there is no limit to the theories/hypotheses/paradigms that can be invented to explain them. So yes, people do in fact choose from the best of these theories, but it doesn't follow that the latest is necessarily the best.

You might be interested in looking up Peter Lipton and IBE, inference to the best explanation. In effect he was defending the rationalist position. The big cheese in the empiricist camp is Bas van Fraasen.
Thanks a lot. You have reminded me. I've been away for a while from all the theory, just following up on old projects.

Bas van Fraasen comes close and Peter Lipton may be on. I'm with Scientific Realism. Van Fraasen is a closet sceptic? Just old?

Great, uwot! :)

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Fri Dec 09, 2016 9:41 am
by uwot
Necromancer wrote:Bas van Fraasen comes close and Peter Lipton may be on. I'm with Scientific Realism. Van Fraasen is a closet sceptic? Just old?
In a way, the development of western philosophy (epistemology and metaphysics, at any rate) has been motivated by the friction between realists and anti-realists from the very start. Thales kick-started the whole enterprise by taking an anti-realist view of the gods and insisting that natural causes should be sought for the things that happen. (If you're interested in the ancient stuff, I wrote an article for the magazine a while ago: https://philosophynow.org/issues/104/Ph ... d_Branches ) Plato and Aristotle came along and totally dominated western philosophy for almost two millennia. Both were realists, in that they each thought that we could know 'the truth' about reality, but Plato believed that reality was very different to what we actually see (allegory of the cave), so we had to work out the truth rationally. Whereas Aristotle based his models on observation. They weren't quite a rationalist and empiricist in the modern sense, but the basic division is there.
Anyway, the Romans came along and usurped the Greeks, flattening Corinth to make the point, and for them the important thing about philosophy was the role it could play in politics, as it had been for Plato. So the Roman Catholic Church reinstated the realist interpretation of god, and a great deal of European medieval philosophy was devoted to making Platonism compatible with a middle eastern creation myth, and proving that god is real. Meanwhile, muslim scholars were trying to make Aristotelianism compatible with the Koran. Muslims conquered much of Spain, Christians went on Crusades, the ideas started to mix and Hey Presto! Renaissance. As with Thales' revolution, the key was rejection of received religious orthodoxy; the refusal to accept certain things as real. Galileo used the observations he made with his telescope to show that the official picture of the universe was untrue and was placed under house arrest by the Pope for his troubles. Shortly before this, on a small European outpost, Henry VIII had broken ranks with the Catholic Church, because the Pope wouldn't allow him a divorce, and made himself head of the Church of England. Freed from the dogma of Rome, Galileo's contemporary, Francis Bacon, could argue that truth was not to be found in books, but rather by studying nature. Inspired by his example, a group of British scientists formed an association with the motto Nullius in verba, take no one's word for it. King Charles II approved, gave them a charter and the association became the Royal Society, the aim of which was to establish the truth about reality by looking at it.
On the continent, Rene Descartes created the template for rationalism by discovering the one thing we do know with absolute certainty: there are experiences. He hoped that he could use this to logically deduce the truth about the world, and even though he was careful to include god in his argument, the idea that truth could be discovered independently of scripture was enough to upset the loonies in the Vatican and Descartes fled to Holland.
Without droning on too much, the Royal Society published Newton's Principia Mathematica. The continental's, Leibniz in particular, complained that there was no explanation for how gravity works. Newton published a second edition in which he said I know and I don't care (hypotheses non fingo, to be precise). Newton's brutal empiricism worked very well, and eventually most scientists accepted the basic premise that what matters is not what is 'real', but what works. In philosophical terms, this has been endorsed by pragmatists, positivists, instrumentalists and probably some other ists I can't remember off the top of my head. It doesn't follow that anyone subscribing to those is an anti-realist, but it's a one size fits all that they can all comfortably wear.
Lipton and Van Fraasen are just following in that tradition. Essentially Lipton is defending rationalism, the idea that the models we use in science relate to something real. For example, the idea of four dimensional spacetime that underpins general relativity, can be thought of as actually existing. Van Fraasen would argue, not that it is or isn't real, but we can't know whether it is, because, as per Duhem-Quine, any number of explanations could be true. As it happens, there are several rival hypotheses to 4D spacetime for the cause of gravity, notably string theory, loop-quantum gravity and modified Newtonian gravity. In a way, the battle is over (or at least a stalemate), because Lipton concedes that all we can infer is the the best, not only explanation. Even so, the fact that we might not know that our theories are true doesn't mean that they are false.
Well done if you have made it this far, and thank you for the compliments.

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Fri Dec 09, 2016 1:57 pm
by Necromancer
Thanks again for the reply, uwot!

On models, it is often said that models may not be entirely true, but approximately true. That is, there is something about them that widens our understanding of the real world, the reality itself. This is what I find with the latest theories, hypotheses, experiments... that they make it quite hard to identify the next theory, hypthesis, experiment that takes us closer to truth than we already are. This puts a hard limit on what can be presented as science (seriously) at all.

* Of truth, think of it as plausibility for the community of the informed scientist, layman and student. Just a tip (from Epistemology).

I think. :D

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Sat Dec 10, 2016 10:45 am
by uwot
Necromancer wrote:On models, it is often said that models may not be entirely true, but approximately true. That is, there is something about them that widens our understanding of the real world, the reality itself.
Well, it depends on what it is you want to understand. Basically this boils down to what the real world does, and what the real world is. You'll have to excuse my rather limited focus on physics and cosmology, but those are just my areas of interest. With that in mind mathematical physics, as prescribed by the Royal Society and exemplified by Newton's Principia, concentrates on what the world does; whereas philosophers are generally more interested in ontology: what the world is. Many physicists are indifferent, or even hostile to philosophy, because any theory about what the world is doesn't necessarily make any difference to what it does.
Consider the example of Ptolemy. The geocentric model was based on naked eye observations made from the surface of the planet, of points of light in the sky. As a model for predicting where those lights would be, it is surprisingly accurate: it is approximately true. The fact that the ontological model is complete nonsense doesn't alter that. In other words, Ptolemy is very good for describing what the world does, assuming you are stuck on Earth without a telescope, but hopeless in terms of what the world is.
Bit archaic, perhaps; so think of Newton. His inverse square law is very accurate for describing the motions of the planets; it is more 'true' than Ptolemy, but all it says ontologically is that there is a force. That in itself is no big deal; everyone already knew that something was keeping our feet on the ground. (The big deal at the time was that it put Aristotle firmly to bed, in that the same force that acted on Earth kept the planets in place, rather than them being stuck on ethereal spheres, as Aristotle argued.)
Even more accurate are Einstein's field equations. Unlike Newton though, Einstein based his equations on an ontological model, i.e. that there exists a substance called 'spacetime' which is warped by the presence of matter/energy. It is tempting to believe that because the mathematical model works so well, that the ontological model must be true. In essence a realist will argue that it may well be. An anti-realist doesn't necessarily disagree, they just believe that there is no way of knowing.
Necromancer wrote:This is what I find with the latest theories, hypotheses, experiments... that they make it quite hard to identify the next theory, hypthesis, experiment that takes us closer to truth than we already are. This puts a hard limit on what can be presented as science (seriously) at all.
If you break it down, there is mathematical science, and philosophical science. In terms of what comes next in mathematical science, that depends on what measurements are made of what the real world does, and whether someone can devise a formula that describes that behaviour better than any current formula. In the meantime, Newtonian physics is good enough to put men on the moon (thanks to ArisingUK for pointing this out to me); it is quite literally Rocket Science. As for what the world is, that, as Duhem-Quine suggests, could be anything.
Necromancer wrote:* Of truth, think of it as plausibility for the community of the informed scientist, layman and student. Just a tip (from Epistemology).
That is certainly the consensus among the staff in my department, particularly those from EU after Brexit, but then the native English speakers come from a much stronger empiricist tradition. The problem is reminding those communities that what is truth for them, isn't necessarily truth for other communities. In science this just leads to pointless, but entertaining squabbles between fundamentalist fanatics. In religion, it's more serious.
For clarity, there are mathematical realists who, to put it crudely, believe that there is some formula out there that perfectly describes what the world does. Then there are philosophical realists who believe there is some hypothesis that perfectly describes what the world is. The anti-realist position is not that these descriptions don't exist, but that they are underdetermined. It's the problem of induction: even if we hit upon a perfect description, there is no way of knowing that a future observation won't undermine it.

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Sun Dec 11, 2016 3:12 am
by Melchior
I think Kuhn over-emphasizes the 'crisis' aspect. The book is very disappointing.

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Sun Dec 11, 2016 1:23 pm
by uwot
Melchior wrote:I think Kuhn over-emphasizes the 'crisis' aspect. The book is very disappointing.
Well, with 54 years of hindsight, it's a bit dated, but at the time it was revolutionary and made a very strong case that science isn't the gradual accumulation of facts, as was assumed. But you are right, although some of the examples Kuhn chose to illustrate his thesis fit his template reasonably well, the Copernican revolution, which was foundational in Kuhn's thinking is a good example, it is generally accepted that scientific progress isn't always so dramatic.

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Tue Dec 13, 2016 3:21 am
by Melchior
uwot wrote:
Melchior wrote:I think Kuhn over-emphasizes the 'crisis' aspect. The book is very disappointing.
Well, with 54 years of hindsight, it's a bit dated, but at the time it was revolutionary and made a very strong case that science isn't the gradual accumulation of facts, as was assumed. But you are right, although some of the examples Kuhn chose to illustrate his thesis fit his template reasonably well, the Copernican revolution, which was foundational in Kuhn's thinking is a good example, it is generally accepted that scientific progress isn't always so dramatic.
Yep. I found it childish, even when I read it back in about 1969.

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Wed Dec 14, 2016 3:04 pm
by Necromancer
uwot wrote:...but at the time it was revolutionary and made a very strong case that science isn't the gradual accumulation of facts, as was assumed.
For the part on Astronomy, the accumulation of facts run back to ancient China and more. Check out these 2 links:
https://en.wikipedia.org/wiki/Chinese_astronomy

https://en.wikipedia.org/wiki/History_of_astronomy
- On China, "Detailed records of astronomical observations were kept from about the 6th century BC, until the introduction of Western astronomy and the telescope in the 17th century. Chinese astronomers were able to precisely predict eclipses."

:)

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Fri Dec 16, 2016 8:01 am
by uwot
Thank you for the links, Necromancer. I'm not sure what point you are trying to make. On the one hand there was the gradual accumulation of facts, which Kuhn called normal science:
Necromancer wrote:- On China, "Detailed records of astronomical observations were kept from about the 6th century BC...
As the first link illustrates, based on these observations, the Chinese developed three different cosmological models. But then:
Necromancer wrote:...until the introduction of Western astronomy and the telescope in the 17th century..."
Which had the same effect as Galileo's observations had on the geocentric paradigm in renaissance Europe.

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Fri Dec 16, 2016 7:31 pm
by Necromancer
The point is:
Necromancer wrote:For the part on Astronomy, the accumulation of facts [astronomical observations] run back to ancient China and more.
Besides,
The SSR, 3rd ed., writes, p. 76
As in manufacture so in science - retooling is an extravagance to be reserved for the occasion that demands it. The significance of crises is the indication they provide that an occasion for retooling has arrived.
To this, one should be aware that the tools/apparatuses are principally "cumulative" even if some archaic stuff isn't existing on our planet as such.

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Sat Dec 17, 2016 11:08 am
by uwot
Necromancer wrote:The point is:
Necromancer wrote:For the part on Astronomy, the accumulation of facts [astronomical observations] run back to ancient China and more.
Yes, but Kuhn isn't simply talking about facts. Facts, such as astronomical observations, are just facts. You can check them, have them peer reviewed, and if everyone agrees that some astronomical observation has been made, it is taken as a fact. People analyse and interpret those facts to generate mathematical and metaphysical models; it is those models which Kuhn calls paradigms.
Necromancer wrote:Besides,
The SSR, 3rd ed., writes, p. 76
As in manufacture so in science - retooling is an extravagance to be reserved for the occasion that demands it. The significance of crises is the indication they provide that an occasion for retooling has arrived.
To this, one should be aware that the tools/apparatuses are principally "cumulative" even if some archaic stuff isn't existing on our planet as such.
The facts generally are cumulative; it is the analyses and models based on those facts that sometimes have to change, because some facts undermine or, as Popper insisted, flatly refute them.

Re: The Structure of Scientific Revolutions by Thomas Kuhn

Posted: Sat Dec 17, 2016 11:48 am
by Necromancer
uwot wrote:Yes, but Kuhn isn't simply talking about facts. Facts, such as astronomical observations, are just facts. You can check them, have them peer reviewed, and if everyone agrees that some astronomical observation has been made, it is taken as a fact. People analyse and interpret those facts to generate mathematical and metaphysical models; it is those models which Kuhn calls paradigms.
The facts generally are cumulative; it is the analyses and models based on those facts that sometimes have to change, because some facts undermine or, as Popper insisted, flatly refute them.
All good. This is how I see it too. :)