In my last post on Wheat and Tares, I wrote somewhat glowingly of Popper’s epistemology based on Conjecture and Refutation. In a post on Millennial Star I even went so far as to explain why I felt there were some touch points between conjecture and refutation and the Gospel. To summarize, Popper believes all knowledge of all types growths through a process of having problems, conjecturing solutions to those problems, then refuting those conjectures based on the discovery of new problems. Through this process we ‘evolve’ our explanations and they improve over time. The end result is increasing verisimilitude – i.e. closeness to reality – of our knowledge. (I note here that this produces increasing verisimilitude without use of induction.)
Now I will consider the strongest challenger to Popper’s epistemology as elucidated by Thomas S. Kuhn, author of The Structure of Scientific Revolutions. Popper and Kuhn are often considered to be two dynamically opposed views of scientific growth that are in struggle for the heart and soul of science. (See, for example, this book here. I have not read it and don’t intend to.) In actuality, Kuhn and Popper have far more in common than they have different from each other. But Kuhn’s view of science does ultimately pose a threat to the very concept of Scientific Realism and proposes, in it’s place, a Positivist view of the world as our ultimate reality. [1]
A Summary of Kuhn’s View of Science
Let’s begin with an attempt to summarize Kuhn’s view of science. I should first warn the reader that what you might think you know about Kuhn may not be accurate. Kuhn is largely misrepresented by his critics. (Always a good sign that he’s on to something.)
Kuhn’s core view of science is that science is broken up into two phases. Plateaus where what he calls “normal science” takes place and then periods of dynamic change where existing scientific theories and explanations are in upheaval.
During periods of “normal science” Kuhn argues that the goal of science is not to refute the theories of science, but rather to find ways to explain existing “anomalies” in terms of the accepted scientific explanation of the day. Kuhn used the term “paradigm” (popularizing it outside of it’s original use only in languages) to describe the currently accepted theories. For the periods of upheaval he uses the term “revolution” to intentional invoke a analogy to political revolutions.
At least, in so far as this goes, this seems rather incontrovertible. For example, I have a friend that is a Professor in Physics. When I asked him about his research, he was looking into why the measurements in certain atomic constants didn’t quite match the theory. He hoped to find a way to get a more precise measurement and therefore match theory and observation more closely. This is a wonderful example of what Kuhn calls “normal science.” He is not trying to falsify any theory (though he might end up falsifying his own hypotheses for how to fit observation with theory) but rather trying to fit the theory better to reality.
But Kuhn’s ideas undermine certain long held (and still strongly held) beliefs about science. The first thing that goes out the window is the idea that scientists are primarily trying to disprove their theories. This simply is not true and has never been true. Kuhn argues that Scientists are so deeply committed to the reality of their paradigms that it is difficult and even painful for them to give it up. In fact, many scientists – particularly the oldest ones – sometimes fail to ever do so even after their entire profession has abandoned the old paradigm. [2]
Kuhn therefore paints a picture of scientists being “apologists” for their respective paradigm. In fact, Kuhn argues, scientists are constantly faced with mismatches between observation and theory. This starts from the moment of the theory being created and continues right up until it is abandoned. [3] Therefore, the old story that scientists are so open minded that they are ready to abandon their theories if it doesn’t match observation is not true either. On the contrary, a scientists primary job is to figure out how to fit observation to a theory. [4] So it turns out that apologetics is one of the most important aspects of true science.
Notice how close this view of Kuhn’s matches the way the Rejectionist community portrays “apologists”:
[counterinstances] can at best help to create a crisis or, more accurately, to reinforce one that is already very much in existence. By themselves they cannot and will not falsify that philosophical theory, for its defenders will do what we have already seen scientists doing when confronted by anomaly. They will devise numerous articulations and ad hoc modifications of their theory in order to eliminate any apparent conflict. (The Structure of Scientific Revolutions, p. 78)
In fact, the religious analogy can, in Kuhn’s view, be taken even further. Kuhn believes that, outside of Theology, science is the only human endeavor where “orthodoxy” plays such an important role. In non-scientific fields, having competing paradigms is the norm. But within fields that we modernly consider science, one of the defining traits is that they have all reached the point of having one and only one agreed upon paradigm. If they did not, we would tautologically not consider them to be science. This ‘orthodoxy’ is (in Kuhn’s view) as rigorously defended by the “apologetic” practice of “normal science” as are religious theologies.
However, we should note here that the above quote from Kuhn does not mean that counter instances or “anomalies” cannot eventually result in abandonment of old theories in favor of new ones. In fact, he explicitly states that counterinstances are what eventually lead to crisis and therefore revolution.
In fact, isn’t this very similar to Popper’s view of conjecture and refutation? In both Kuhn’s and Popper’s views, observation does eventually play a role in science by creating problems that “normal science” simply cannot come up with a way to accommodate through more “apologetic” means. The end result is eventually “revolution” and the abandonment of an old paradigm in favor of a new and better one that solves more problems then the previous one. In this way both Kuhn and Popper agree that growth of scientific knowledge is real and even agree on the broad outline of how it takes place.
Even Popper fully accepts that Kuhn is right that scientists are not objective. Popper (like Kuhn) actually argues that this is a good thing.
…my guess is that should individual scientists ever become ‘objective and rational’ in the sense of ‘impartial and detached’, then we should indeed find the revolutionary progress of science barred by an impenetrable obstacle. (Myth of the Framework, p. 22)
The reason this is true is because if scientists were “objective” then the conjecture and refutation process would not function properly across the community.
Since the method of science is that of critical discussion, it is of great importance that the theories criticized should be tenaciously defended. For only in this way can we learn their real power. And only if criticism meets resistance can we learn the full force of a critical argument. (Myth of the Framework, p. 94)
So far Kuhn is not at odds with Popper, despite popular portrayals to the contrary. That the two most advanced schools of scientific epistemology agree on so much ought to give pause to all scientists that reject this more realistic view of science and scientists. It is not true that scientists are primarily trying to disprove their theories and it is not true that science teaches a person to be objective or even that scientific progress requires such objectivity.
Perhaps a bit more surprising is that neither Kuhn nor Popper favor a view of science where science is strictly cumulative (i.e. nothing in the past gets invalidated – not that we aren’t gaining knowledge) in growth of knowledge. Kuhn argues using the historical record to back up his view:
Cumulative acquisition of unanticipated novelties proves to be an almost non-existent exception to the rule of scientific development. The man who takes historic fact seriously must suspect that science does not tend toward the ideal that our image of its cumulativeness has suggested. (The Structure of Scientific Revolutions, p. 96)
Later, we’ll look at Popper’s acceptance, though modification on this point. But Popper does not deny that scientific knowledge is not cumulative. Rather, one scientific paradigm must give way to another. The new paradigm must largely disprove the old one and therefore destroy it. It is a creative but destructive process. Worse yet, the new paradigm often fails to give answers to questions the old paradigm did. Kuhn uses the example of how Aristotle’s physics explained why particles of matter attracted each other whereas Newton’s theory did not.
Must a theory of motion explain the cause of the attractive forces between particles of matter or may it simply note the existence of such forces? Newton’s dynamics was widely rejected because, unlike both Aristotle’s and Descarte’s theories, it implied the latter answer to the question. When Newton’s theory had been accepted, the question was therefore banished from science. That question, however, was one that general relatively may proudly claim to have solved. (The Structure of Scientific Revolutions, p. 148)
Aristotle’s theory of attraction was that there are four elements – earth, air, fire, water – and they have a natural affinity for each other and that is why, for example, some things fall to the earth and some float. They are finding their natural element that they are primarily made out of.
Before you laugh at this theory, wait for my post where I point out that this is a much better theory that we moderns seem capable of giving it credit for. But for the moment, just accept the key point. Newton’s theories were deemed a step back in some ways because it failed to answer questions that Aristotle’s theories did seem to answer. Therefore the growth of scientific knowledge was not strictly cumulative.
It was not until Einstein’s General Relativity that the accepted paradigm once again answered why particles of matter attract. An indirect triumph for Aristotle’s theories in that it turned out that this was a valid question after all and Newton’s theories were wrong to ignore it.
Kuhn’s Challenge to Popper
But both Kuhn and Popper agree that their two view are, at least somewhat, at odds with each other. Kuhn points out that if this view of science is correct (and remember Popper agrees with the facts so far) then Popper’s view of “Conjecture and Refutation” is (in Kuhn’s view) possibly flawed because, in fact, there is no such thing as “falsification” of theories per se (though there might be of hypotheses meant to support a theory) except between paradigms. Here, I will let Kuhn do his own arguing:
A very different approach to this whole network of problems has been developed by Karl R. Popper who denies the existence of any verification procedures at all. Instead he emphasizes the importance of falsification, i.e., of the test that, because its outcome is negative, necessitates the rejection of an established theory. Clearly, the role thus attributed to falsification is much like the one this essay assigns to anomalous experiences, i.e., to experiences that, by evoking crisis, prepare the way for a new theory. Nevertheless, anomalous experiences may not be identified with falsifying ones. Indeed, I doubt that the latter exist. As has repeatedly been emphasized before, no theory ever solves all the puzzles with which it is confronted at a given time; nor are the solutions already achieved often perfect. On the contrary, it is just the incompleteness of the imperfection of the existing data-theory fit that, at any time, define many of the puzzles that characterize normal science. If any and every failure to fit were ground for theory rejection, all theories ought to be rejected at all times. (The Structure of Scientific Revolutions, p. 146)
Popper’s anomalous experience is important to science because it evokes competitors for an existing paradigm. But falsification, though it surely occurs, does not happen with, or simply because of, the emergence of an anomaly or falsifying instance. Instead, it is a subsequent and separate process that might equally well be called verification since it consists in the triumph of a new paradigm over the old one. (The Structure of Scientific Revolutions, p. 147)
Therefore the real difference between Popper and Kuhn (in Kuhn’s mind anyhow) is that Popper disbelieves in “verification” and Kuhn points out (correctly in my opinion) that “falsification” only takes place once a crisis is reached and a revolution is underway. At this point we are always comparing theories and explanations. Therefore “falsification” through observation is identical to “acceptance through verification.” Popper is thus wrong on one point. Verification does, in a certain sense, exist.
Note, however, that Kuhn does not deny that “falsification” not exist at all. Therefore, even this quote above is not a direct attack on Popper’s epistemology because Kuhn does accept falsification in the case of comparison between paradigms. (Though Kuhn prefers to think of it more straightforwardly as verification rather than falsification.)
Kuhn’s View of Science and Organic Evolution
Kuhn then takes his argument even further. Having now essentially denied every popular basis for justification in belief the growth and progress of scientific knowledge, he then proposes (not unlike Popper before him) that science is actually very much like organic evolution. But, unlike Popper, Kuhn latches on to the non-teleological (i.e. non-purposeful) nature of evolution as his main thesis.
For many men the abolition of the teleological kind of evolution was the most significant and least palatable of Darwin’s suggestions. The Origin of the Species recognized no goal set either by God or nature. Instead, natural selection, operating in the given environment and with the actual organisms presently at hand, was responsible for the gradual but stead emergence of more elaborate, further articulated, and vast more specialized organisms.
What could ‘evolution,’ development,’ and ‘progress’ mean in the absence of a specified goal? To many people, such terms suddenly seemed self-contradictory. (The Structure of Scientific Revolutions, p. 172)
Kuhn therefore suggests that scientific progress is a non-teleological process, just like organic evolution. That is to say, the growth of scientific knowledge (which Kuhn does believe in) is not a growth towards some underlying reality, but merely a growth of solved problems that we humans found interesting.
This, then, is the primary difference between Kuhn and Popper – and it’s a huge difference in my opinion: Kuhn is a Positivist and Popper is a Scientific Realist. Popper believes that there is an underlying reality that we can grow closer to whereas Kuhn does not believe this is necessary to explain scientific progress.
Notes
[1] Supporting Kuhn’s views are luminaries such as Stephen Hawking, who recently wrote a book called The Grand Design that in part challenges the Scientific Realist’s view of science as well as (though he denies it) challenging the existence of God or at least the need for God in a scientific world view. I originally meant to include Hawking’s thoughts within this post, but eventually realized it was too long. So I’m going to briefly cover Hawking’s defense of Positivism in a future post since I think his view is worthy of consideration even though I reject it.
[2] I have noted elsewhere that this view of Kuhn’s isn’t entirely correct either. While is it true that there are many historical examples of scientific revolutions that failed to bring in older adherents to past paradigms, we also have the example of the standard model that quickly and easily converted all scientists despite the messiness of it’s model. Yet, Kuhn’s point is still often valid and that is enough for this discussion.
[3] Kuhn argues: “To be accepted as a paradigm, a theory must seem better than its competitors, but it need not, and in fact never does, explain all the facts with which it can be confronted.” (The Structure of Scientific Revolutions, p. 18-19)
Then even more strongly:
The same point can be made at least equally effectively in reverse: there is no such thing as research without counterinstances. For what is it that differentiates normal science from science in a crisis state? Not, surely, that the former confronts no counterinstances. On the contrary, what we previously called the puzzles that constitute normal science exist only because no paradigm that provides a basis for scientific research ever completely resolves all its problems. The very few that have ever seemed to do so (e.g. geometric optics) have shortly ceased to yield research problems at all and have instead become tools for engineering. (The Structure of Scientific Revolutions, p. 79)
[4] Kuhn says: “No part of the aim of normal science is to call forth new sorts of phenomena; indeed those that will not fit the box are often not seen at all.” (The Structure of Scientific Revolutions, p. 24)
Kuhn points to a not-so-well-known psychology experiment that is noteworthy. In this experiment, subjects are shown playing cards and are then asked to tell which suites they just saw. Unbeknownst to the subjects, the playing cards had been modified to contain non-existent suites, such as “a red six of spades and a black four of hearts.” Because the subjects had expectations built around a paradigm of what cards do and don’t exist, initially none of the subjects could even see the anomalous suites. With complete confidence. “The black four of hearts might, for example, be identified as the four of either spades or hearts.” Eventually, with increased exposure, some of the subjects started to see the anomalous suites. But some of the subjects not only never did, but started to not be able to tell what normal suites looked like any more. (The Structure of Scientific Revolutions, p. 63) Kuhn goes on to give numerous examples of how scientific theories form paradigms in which our expectations are built. Scientific theories therefore dictate not only what experiments are to be performed, but also how scientists will interpret the observational data.
Wheat & Tares 
I think the Reformation also fits this paradigm notion. The anomalies were the corruption within the church becoming more and more obvious, until the apologists for Catholicism could not prevent Protestantism from taking hold. The Counter-Reformation saved Catholicism by splitting with the Protestants on WHICH QUESTIONS were more important to answer — authoritative institutions or authoritative doctrine, perhaps?
If I may make a second point, Kuhn’s approach explains what happened to some extent with climate science. Today’s scientific establishment WERE the young rebels of the 70’s who overturned the notion that a return to the ice age was the great climate threat. Now, they defend the orthodoxy because it is THEIR reputations on the line and there beloved intellectual children threatened.
As to the sociology of science, I think Kuhn is rightly considered a genius. Contemporary climate “science” could hardly provide a better case study. However, as to the underlying reality, he is far worse than Popper.
Kuhn has this concept of “incommensurability” which basically denies that any scientific theory is better or worse than the one that preceded it. Or in other words, there is no such thing as scientific progress, scientific knowledge is an oxymoron, not even something like verisimilitude.
For example, from Kuhn’s point of view, Newton’s law and Quantum Mechanics are strictly incompatible – they don’t measure the same thing, at any scale. This must come as a big surprise to physicists and engineers who use them interchangeably at scales ranging from the microscopic to the astronomical. It is a mathematical fact that they are commensurable at classical scales, Kuhn’s denial to the contrary notwithstanding.
When you have a theory as well verified as Newtonian mechanics, nothing is going to replace it that isn’t commensurable. The entire idea is absurd.
Bruce, the problem is that falsification per se is not what is responsible for increasing verisimilitude. The credit for that goes entirely to the scientists who use a process of abduction to come up with a superior theory – superior both in terms of explaining existing data and predicting what future data is going to be.
Furthermore, as Kuhn suggests, falsification in science is rarely a binary state. More often like a trickle of anomalies and inconsistencies. It is also arguably improper to speak of falsification of any theory that hasn’t been experimentally verified to a substantial degree.
There isn’t the slightest experimental verification for anthropogenic global warming, for example, nor is there virtually any conceivable set of facts that could persuade its advocates that it is not occurring.
Of course “verification” in this context does not mean mere consistency with experimental facts, it means both radically superior consistency than all available alternatives and predictive power for future results.
MarkD, I agree with your point on falsification.
And I mostly agree with you on this quote:
I think Kuhn was wrong about incommensurability. However, Kuhn does not deny the existence of scientific progress. He does, however, disagree with verisimilitude.
And I do not accept that scientific progress comes from abduction per se, though I can see why you’d think it does.
At issues here is that you seem to think that abduction is the only way to make future predictions. And most people do think of it that way. But can we really infer the future from the past? No, not really.
The real basis for future prediction is, I think, explanation and algorithmic explanation in particular.
Abduction is the way viable explanations are developed, and viable algorithmic explanations in particular.
If you give me a dozen data points, I can come up with an arbitrary number of functions that include each one of them. Each one of those functions is an “explanation”. By long experience, however, some are much more likely to be consistent with future data points than others.
Simple functions (relative to the problem set) are much more likely to correspond to future data points than complex ones. William of Ockham was one of the smartest people that ever lived, but today he borders on a household name for that dictum alone, which we usually call “Occam’s Razor”.
When someone claims it isn’t probable that the sun will come up in the morning, what they are really claiming is that a simple explanation about uniformity of natural laws isn’t probable relative to the idea that the sun comes up each morning by statistical accident.
In that sense Popper is the anti-Ockham, and Kuhn is worse. Without a basis in probability – any kind of probability – the idea of verisimilitude is incoherent. The idea that scientists solve any problems without actual predictive power equally so.
“Without a basis in probability – any kind of probability – the idea of verisimilitude is incoherent”
This is what I just can’t agree with.
If we are talking about statistics, I could probably see your point.
But if we are talking about defining the laws of physics, I do not see what probability has to do with it.
Now if you want to make the case that probability plays a role in coming up with the laws — that is to say coming up wih the conjectures — I’d then agree with you. I do think induction (and abduction) plays such a role.
But the idea that we can say it is probable that the sun will come up tomorrow (or improbable) is, by itself, meaningless jibber.
Now let’s change that around a bit. Let’s say that we know the sun will come up tomorrow because the earth turns and that — baring a currently unknown catastrophe that stops the earth turning or blows up the sun — we are justified in saying the sun will rise tomorrow.
Yes, this is a fair statement. But probability has nothing to do with it.
Might we not then say “it’s probable that the sun will come up tomorrow because there probable isn’t unknown catastrophe?”
Well, yes, we would say such a thing. But what does it mean? Does it mean that God will then roll a giant die and if it comes up 6 he’s going to introduce a catastrophe that wasn’t going to happen otherwise? Of course not.
What we mean when we say “it is probable that the sun will come up tomorrow” is actually just “we don’t know for sure, but we have no reason to believe otherwise.”
The real truth is that the sun either will or won’t come up tomorrow because there either is or isn’t a catastrophe that is about to happen. It’s 100% either way with no probability involved at all.
So for me to make sense of what you are saying, it seems to me I first need to assume that you are using ‘probability’ in this way. Granting it’s a common use of the word, my point is that it’s actually go nothing to do with mathematical probability.
“In that sense Popper is the anti-Ockham”
And this is entirely untrue. Probably true for Kuhn to some degree though.
Mark D:
I read the article you sent me by David Samra.
First, excellent article.
But it seems to me that it undermines your point in favor of mine. He is in no way supporting the idea of induction as the basis for justification of scientific growth. He’s quite firmly against it.
His argument is against Popper’s hostility to synthetic a priori knowledge because Popper (according to Samra anyhow) mistakes synthetic a priori knowledge for induction.
Samra whole point is that induction is wrong, but that synthetic a priori knowledge is not induction.
e.g. “These problems do pose serious challenges to Popper’s theory, but they are not grave enough to reinstantiate induction.”
Personally, I agree with Samra that Popper over reaches in claiming that Kant was merely disguising induction. And I agree with Samra that falsification alone can’t justify belief in scientific knowledge.
However, I personally think Samra is playing a giant semantics game. Belief in synthetic a priori knowledge is not necessary to make sense of Popper.
I would assume Popper would reply that what Samra is calling “synthetic a priori knowledge” is actually just a conjecture that is not refuted yet.
But that being the case, it’s justifiable to believe in it because it’s going to have the highest verisilitude compared to the next best explanation.
Therefore Popper does not need synthetic a priori knowledge, contrary to what Samra is saying.
But then I think Samra would replay (and does reply) that I’m merely being to stringent in my understanding of what synthetic a priori knowledge is and that it’s really just (his words) “an approximation to the truth” based on animal knowledge built into our DNA.
But now Popper and Samra are just saying the same thing in different words. Popper refues to call it synthetic a priori knowledge because it’s refutable in principle. (Perhaps aliens show up that uses different laws of computation superior to ours) and Samra believes it’s effectively synthetic a priori knowledge so we should just call it that.
I’m fine either way. They’re both just saying the same thing and talking past each other.