Karl Popper's falsifiability

[Led Zeppelin]

I read this recently in the criticism section of Historical Materialism wikipedia page:


Philosopher of science Karl Popper, in his Conjectures and Refutations, critiqued such claims of the explanatory power or valid application of historical materialism by arguing that it could explain or explain away any fact brought before it, making it unfalsifiable.

Anyone know some good Marxist responses to this?

[black magick hustla]

Actually, I am writing a paper about popper for my philosophy of science course. Long story short, science does not work by falsifiability, but by paradigms. physics, his fav science, has some shit that is not falsifiable at all (like dark matter). no wonder why someone who is not a scientist knows jackshit about science at all.

[JimFar]

One Marxist school that attempted to deal with, if not answer Popper were the Analytical Marxists. It is interesting to note Popper's influence on the Analytical
Marxist school, both positively and negatively. The late G.A. Cohen in his Karl Marx's Theory of History: A Defence made no mention at all of Popper, and yet his book reads to me as a kind of reply to Popper, since Cohen attempts to reformulate historical materialism (or at least historical materialism as understood by the Second International) as a rigorous empirical theory of history. William Shaw (in Marx's Theory of History) and Dan Little (in The Scientific Marx), on the other hand, do attempt to answer directly Popper's criticisms of Marxism, and they both draw upon Lakatos' critiques of Popper, in doing so.

An earlier Marxist work that attempts to answer Popper is The Open Philosophy and the Open Society by Maurice Cornforth, which is subtitled, by the way, "A Reply to Dr. Karl Poppers Refutation of Marxism."

Also, in connection with the Analytical Marxian school there is another book that people may wish to look at on this issue, the unjustly neglected book Analyzing Marx by Richard W. Miller. In that book he draws a distinction between the technological interpretation of historical materialism which was articulated and defended by many writers of the Second International (i.e. Kautsky, Plekhanov) and which was cast into an especially rigorous form by G.A. Cohen in his Karl Marx's Theory of History, and what he calls the mode of production interpretation which abjures the technological determinism and the economic determinism of the latter.


Miller draws a link between these two different interpretations of historical materialism and different philosophies of science. The technological interpretation, Miller links to positivist philosophies of science with their covering law models of scientific explanation and their presuppostion of Humean notions concerning causality. Here, Miller does not draw a very sharp distinction between positivism and Popperism. While Popper clearly did not see himself as being a positivist, he nevertheless, still had many notions in common with them. In Miller's view Popper's hypothetico-deductivism placed him within the positivist camp. In any case, Miller contends that the technological interpretation of historical materialism does represent the sort of theory that can be regarded as falsifiable from a strictly Popperian standpoint. Hence, it is scientific by Popper's criteria. The only thing that is wrong with it is that history has indeed (as Popper had contended) falsified it, and the other thing that is wrong with it, is that in Miller's view it represents a distorted interpretation of how Marx undertook the study of history and political economy.

The mode of production interpretation in Miller's view offers us a view that is closer to the spirit of Marx's actual methodology. But it is not falsifiable in the strict Popperian sense. One might then think that Miller would propose to throw away falsifiability as a criterion of demarcation between science and non-science but surprisingly enough he does not. Instead, he attempts to reconstruct the notion of falsifiability, drawing upon the work of Thomas Kuhn and Paul Feyerabend. He embraces their historicist approach to the philosophy of science and he develops a reconstructed version of the notion of falsifiability. The mode of production interpretation of historical materialism while perhaps not falsifiable in Popper's sense, is nevertheless falsifiable in Miller's sense and that justifies retaining the label of science for it.

Miller also BTW contends that the postivist (and Popperian) analysis of natural science is fundamentally flawed so that while the positivists were quite correct in seeking a unified science which would assimilate the social sciences into the natural sciences , they misunderstood the nature of natural science. For Miller, the antipositivists were correct in attacking positvism for trying to force social science into a narrow mold centering around the covering law model and deductive-nomological models of explanation and Humean causality, but the same flaws also applied to their analysis of natural science. In reality such an analysis, in Miller's view is not properly applicable to either natural science or social science.

[JazzRemington]

I read this recently in the criticism section of Historical Materialism wikipedia page:



Anyone know some good Marxist responses to this?


One of the repeated criticisms of Popper's falsifiability claim is that it, itself, cannot be falsified. Further, Samir Okasha, writing in his Philosophy of Science: A Very Short Introduction, gives an example of which Popper's criteria didn't work: the discovery of Neptune. The basic conditions of the discover were similar to what Popper criticizes historical materialism for: something was inconsistent with a sound theory and various proposals were made in an attempt to explain deficiencies in said theory. But Popper wouldn't probably consider Neptune's discovery as violating his criteria because Neptune was eventually discovered.......

[Calmwinds]

Does everyone really treat falsifiability the same? It seemed to me that there were several types, and most attempts to attach them to each other were artificial. They were

a narrow critique of a class of propositions

A)If a proposition cannot be falsified, it is in trouble(though not always bad, indeed Popper noted that several important propositions could NOT be falsified)

And the quasi-historicist part
B) That scientific history is a series of events of falsifying theories and then dealing with that. Etc

It seemed to me that I could always accept A without B.


One Marxist school that attempted to deal with, if not answer Popper were the Analytical Marxists. It is interesting to note Popper's influence on the Analytical
Marxist school, both positively and negatively. The late G.A. Cohen in his Karl Marx's Theory of History: A Defence made no mention at all of Popper, and yet his book reads to me as a kind of reply to Popper I feel the same.

So if I may say so, what historical materialism needs to deal with is the claim that it cannot be falsified(A) and not as a response to falsifiability as historicism(B).

[mikelepore]

It has always seemed to me -- when people make "all", "none", and "can't do" hypotheses, they can be falsified but not verified; when people make "some" and "can do" hypotheses, they can be verified but not falsified. Therefore Popper's falsifiability criterion amounts to an editorial opinion about what kinds of hypotheses are useful and worth making. I believe that both kinds are useful.

[yuon]

Just a quick note, I personally have thought about this issue a little.
I have come to the conclusion that nothing can be "falsified", as it requires just as much evidence as "proving" something.

It is impossible to prove something, because it might turn out tomorrow not to be correct (Hume wrote some good stuff on this idea). Equally, it is impossible to falsify something, because it might turn out tomorrow that your falsification was incorrect.

Another angle on the same idea is the lack of an infallible being. Because we all make mistakes, how can we know that we made a mistake in the theory (and the test showed the theory wrong), or a mistake in the test (and the results thus cannot show what we thought)?

To take the swans example. You can say, "all swans are white", until you get to Australia and see a black one. Now, according to Popper, you only need one black one swan to disprove the claim that all swans are white. True, if you could be sure that you actually saw a black swan! (Which, due to the nature of humans as fallible beings, is never 100%.) (It's the same line of reasoning Hume used on miracles. Perhaps you were mislead, hallucinating, lied to, etc. Why believe in miracles?)

So, I would say that one black swan can never disprove the claim that all swans are white. (Of course, science works on the basis of continued, over whelming evidence, from multiple people. So, it is possible to say that not all swans are white, because the "experiment" it has been repeated many many times and works. But, if it stopped working tomorrow (all the black swans turned white), it would take some time before the new reality became accepted. Just like some people claim that there are still Tasmanian Tigers around.)

[Luís Henrique]

You can say, "all swans are white", until you get to Australia and see a black one. Now, according to Popper, you only need one black one swan to disprove the claim that all swans are white. True, if you could be sure that you actually saw a black swan! (Which, due to the nature of humans as fallible beings, is never 100%.)
You catch a black swan, and see if it breeds with white swans. If so, it is a swan, and it is black. It is not a matter of "seeing" but of incorporating it into practice.

Luís Henrique

[YKTMX]

Interesting thing to note about Popper: not only did he not think Marxism was a science, he didn't think evolutionary theory was scientific.

Think about that. It's something critics of Marxism tend to forget when they use a Popperian critique.

[Parker]

hi all - my first post here.

My understanding of Popper's attitude towards Marx was that Popper thought that Marx's work was falsifiable and thus to Popper genuine science, but that history had in fact shown it to be falsified. Popper, iirc, pointed to rising living standards as disproving Marx's claim that the working class would face increasing immiseration.

FWIW, I think this is wrong. It is based on a misunderstanding of the analysis in the chapter on the General Law of Capitalist Accumulation in Capital Vol I.

However, I think that falsifiability in a general sense is a good principle to follow in constructing arguments. It's a misunderstood concept. I am not a follower of Popper, and I am not overly familiar with his work, but I like the general approach.

[Paul Cockshott]

Maurice Cornforths book Open Philosophy and the Open Society

[JimFar]

hi all - my first post here.

My understanding of Popper's attitude towards Marx was that Popper thought that Marx's work was falsifiable and thus to Popper genuine science, but that history had in fact shown it to be falsified. Popper, iirc, pointed to rising living standards as disproving Marx's claim that the working class would face increasing immiseration.

FWIW, I think this is wrong. It is based on a misunderstanding of the analysis in the chapter on the General Law of Capitalist Accumulation in Capital Vol I.

However, I think that falsifiability in a general sense is a good principle to follow in constructing arguments. It's a misunderstood concept. I am not a follower of Popper, and I am not overly familiar with his work, but I like the general approach.

That I think, accurately represents Popper's take on Marxism. In that respect he distinguished Marxism from psychoanalysis, which in his view was never a falsifiable theory from the get go. However, he argued that once Marxist theories were falsified, later Marxists began to introduce all sorts of ad hoc hypotheses which save the theories from refutation but at the cost of rendering the theory unfalsifiable, hence, not scientific.

I have earlier in this thread outlined the responses of William Shaw, Daniel Little and Richard Miller to this line of argument.

[Jean-Luc Lebris]

Jim,
What would you recommend as a source for the following,

"However, he argued that once Marxist theories were falsified, later Marxists began to introduce all sorts of ad hoc hypotheses which save the theories from refutation but at the cost of rendering the theory unfalsifiable, hence, not scientific."

I'd love to read about this.

-j

[JimFar]

See:

Science, Pseudo-Science, and Falsifiability
Karl Popper, 1962

http://www.kenrahn.com/jfk/critical_thinking/Science_pseudo_falsifiability.html

[Jean-Luc Lebris]

Thanks for that.

[mel]

That I think, accurately represents Popper's take on Marxism. In that respect he distinguished Marxism from psychoanalysis, which in his view was never a falsifiable theory from the get go. However, he argued that once Marxist theories were falsified, later Marxists began to introduce all sorts of ad hoc hypotheses which save the theories from refutation but at the cost of rendering the theory unfalsifiable, hence, not scientific.

I have earlier in this thread outlined the responses of William Shaw, Daniel Little and Richard Miller to this line of argument.

I don't have a source right now, but I can grab one in a couple of hours, but in the two-three essays of Popper's that I read, he lumped Marxism straight in with Psychoanalysis as unfalsifiable from the beginning. Do you think he just changed his mind on this, or is it more likely that I simply misread?

[Luís Henrique]

My understanding of Popper's attitude towards Marx was that Popper thought that Marx's work was falsifiable and thus to Popper genuine science, but that history had in fact shown it to be falsified. Popper, iirc, pointed to rising living standards as disproving Marx's claim that the working class would face increasing immiseration.

:lol: I remember that I "falsified" Newton's gravitational theory when I was eight, and my hydrogen baloon flew away instead of falling to the ground.


FWIW, I think this is wrong. It is based on a misunderstanding of the analysis in the chapter on the General Law of Capitalist Accumulation in Capital Vol I.This in change seems likely to become a Natural Law:


Whenever anyone discovers a fundamental flaw on Marx's reasoning, the way that person expresses this discovery is enough to show that such person did not actually read Marx - or at least was not paying attention when s/he did it.


However, I think that falsifiability in a general sense is a good principle to follow in constructing arguments. It's a misunderstood concept. I am not a follower of Popper, and I am not overly familiar with his work, but I like the general approach.I like it because it allows me to say:

"The difference between religion and science is that religion is always right, and science always wrong. This is also the reason we should prefer science to religion."

Luís Henrique

[Parker]

lol, I'm not quite sure what you are trying to say. In any case your "Law on Criticism of Marx" is unscientific, as there are indeed cases of where someone gets it wrong about Marx because they mis-interpreted him. For instance, I was reading yesterday Ludwig von Mises's book on socialism and it contains no end of appalling errors. It is a sure-fire case in point of demonstrating that Mises had got it wrong. [The same goes the other way: many Marxists haven't really bothered to acquaint themselves of their opponents' positions either.]

But point taken, and there are more or less honest ways of going about it.

For the record, I think we could show in quite a few ways how and if Marx;s theories could be disproved. If a high organic composition of capital did not tend to yield a lower rate of profit, for instance ...

[JimFar]

I don't have a source right now, but I can grab one in a couple of hours, but in the two-three essays of Popper's that I read, he lumped Marxism straight in with Psychoanalysis as unfalsifiable from the beginning. Do you think he just changed his mind on this, or is it more likely that I simply misread?

That, I don't know. He wrote lots of stuff on Marxism, psychoanalysis and falsificationism over the years. I am not aware of him changing his mind on Marxism and falsifiability but it is possible. The source that I provided earlier presented views that I have seen in many of his other writings including the autobiographical notes that appear in the Schilpp volume on Popper.

[Glenn Beck]

I read this recently in the criticism section of Historical Materialism wikipedia page:



Anyone know some good Marxist responses to this?

Yeah. Popper was butthurt that Marxists didn't curl up and die when the living conditions of the working class in imperialist countries failed to simply fall in a linear fashion and the stability of the capitalist system failed to do the same. So he decided to paint all later Marxist analysis of the contemporary situation explaining this as mere hand-waving from irrationalists who didn't want to accept that the march of history had objectively disproven their beloved theory.

The refutation of this is simply that Marx was not making Nostradamus-like predictions but explaining general tendencies of the system ceteris paribus. Well, ceteris weren't paribus, and later Marxists explained that within the bounds of Marx's analysis. Popper and many other liberal academics chose to ignore that, probably because they didn't see any need to keep up with Marxist scholarship.

[Dean]

I read this recently in the criticism section of Historical Materialism wikipedia page:



Anyone know some good Marxist responses to this?

I would say that it's a totally hypocritical statement as a criticism. His notion of "falsifiability as a definitive measure of scientific validity" is absolutely ridiculous, and is self-referential as a proof. In certain fields it can be a useful system, but it should never be used as a sole evidence to discount a theory.

If we were to apply Popperian logic to the sciences today, all forms of sociology would be thrown out, as would most literature, political and economic theory, nto to mention a number of emergent sciences, and even classics like quantum theory.

Popper was basically a fucking hack who gained notoriety with his science / philosophy ideology which was mostly conjecture.

Popper's kind of ideology tends to rely heavily on axioms and therefore tends to be hopelessly self-referential. His theory of science more or less degenerated into a slew of dogmatic mystical attitudes towards science which generally hold practitioners of the philosophy back.

[black magick hustla]

popper did think much of sociology/psychology was unscientific though, so he was well aware of the "conclusions" of his belief

[Calmwinds]

Can someone explain to me why the principle that something should be able to be proven false is a bad one? I clearly am not seing it. I am not the naive falsificationalist that says "All good propositions must be falsifiable", but it serves as a good rule of thumb[along with much more] for evaluating propositions.

[black magick hustla]

the reason is that virtually no science works like that at all. Read some kuhn about the issue. Scientists work within conceptual frameworks, and they routinely try to "fill the holes" with puzzle solving than to "falsify". the biggest example i can think about is dark matter. astronomers saw that the velocity of stars did not correlate with the expected amount of gravitational force, so they just added a fudge factor called dark matter to account for the excess of velocity, instead of trying to overthrow modern physics.

[mel]

Can someone explain to me why the principle that something should be able to be proven false is a bad one? I clearly am not seing it. I am not the naive falsificationalist that says "All good propositions must be falsifiable", but it serves as a good rule of thumb[along with much more] for evaluating propositions.

Well, Popper wasn't talking about "propositions" in general, and he also wasn't talking about a "rule of thumb".

Poppers position was ultimately an attempt to solve what is known as the "demarcation problem" (what is science and what is "pseudo-science"). For Popper, the way that Science works is that a Scientist makes a bold, universal statement (a hypothesis) and then experiments in an attempt to falsify that statement. If a proposition is not falsifiable, it's not scientific. In addition, by limiting science to universal, falsifiable statements, inductive logic was not necessary to evaluate the truth value of claims. I don't really want to get into the induction/deduction issue right now, but if you're interested, then let me know, and I can try to see if I have some notes or articles I can send you about it.

Back to Popper's view on science: There are a number of problems with his view of science. Firstly, as dada pointed out, that's simply not how science works. Scientists don't routinely go around trying their hardest to falsify their own theories. Secondly, there's a problem with the falsification model "Holism of Testing" (or the Duhem-Quine thesis), which basically argues that when you bring a scientific theory up for testing, and it is falsified, you don't know if it was the theory that was false, or one of the auxillary assumptions which it was based off of. It could be a problem with the instruments used to test, a problem with a previous theory that yours was based on, a problem with mathematics, or the logic on which it was based. In this way, no theory can be "falsified" in isolation, only grouped together with all of its other auxillary theories and assumptions.

The last major problem with Popper is that when attempts to falsify a theory are unsuccessful, we cannot actually use this as evidence of the theory's truth. It's no more likely to be true than it was when you started (this is a result of his solution to the problem of induction, by disallowing induction to enter the fold). This means that if there are two unfalsified theories, there is no good way to choose between them. For example, if there is one method for building a bridge that has been used for 50 years without a single collapse (that method has yet to be falsified) and one brand new method for building a bridge that has never been modeled or tested, there is no reason not to build a bridge using the brand new method than with the "tried and true" method, since the status of both is simply "not falsified". Eventually, he introduced something called "corroboration" (a theory is "corroborated" if it has been tested and unsuccessfully falsified) but within his framework he really had provided no reason that a "corroborated" theory is better than an "uncorroborated" one, since a "corroborated" theory was still no more likely to be true than it was before it had been tested.

So basically, the problems with popper are simply that

1) Science doesn't work that way
2) A theory can't actually be falsified, only the theory in combination with its entire set of auxiliary assumptions
3) No mechanism for theory choice

There's not really a problem with the idea that "a good proposition is probably falsifiable", but Popper didn't stop there.

Edit:

Also, if this is interesting to you, I do strongly recommend Thomas Kuhn's "The Structure of Scientific Revolutions". It's a short book, it's easy to read, and it's the source of the commonly used idea of a "paradigm shift". Obviously it's been hugely influential, and in some ways, it closely mirrors Marx's views on the progression of history. (A "paradigm", in this case a model for society, is found which seems like a great model, and is used as the basis for all later work. Problems develop in that model, but they are simply patched over, until it reaches a state of "crisis". A new model is found which is viewed as somehow superior to the previous model, and that model takes its place as the dominant model). It really is a very good book, and I'd encourage anybody interested in Philosophy of Science to pick it up.

[Calmwinds]

Yeah, I was just wondering if the dislike of falsification was towards "Decent propositions should be falsifiable" or his view on the march of science.

And yes, I have read Kuhn's book and I am indeed a Kuhnian. (Seems weird to say)

[mel]

Yeah, I was just wondering if the dislike of falsification was towards "Decent propositions should be falsifiable" or his view on the march of science.

And yes, I have read Kuhn's book and I am indeed a Kuhnian. (Seems weird to say)

I mean, ultimately it's more about the march of science, but I do think it's important to note the problems with falsifiability (not limited to the Duhem-Quine thesis) for most nontrivial claims. A "falsifiable" claim is also not inherently better than a falsifiable one depending on exactly what it is that you're concerned with. Emotional and ethical claims come to mind as things which can't generally be falsified but which can be highly relevant in some situations.