The Authority of Karl Popper's Falsifiability Criterion

[ragerdude]

It seems that many people use Popper’s falsifiablilty criterion for a purpose which he did not intend.

His criterion does NOT determine when a claim is “true,” or “acceptable,” or “reasonable,” or even “debatable.” Rather, his criterion determines whether a claim is “scientific” or not; that is, whether it involves a genuinely empirical method. He very well admits that science often errs, and that non-scientific methods may yield the truth. Various disciplines have different methods. It seems inappropriate to judge claims in one field by the standards of another.

Example: The truths of mathematics are not empirically falsifiable. If I count 24 beans in one pile, and another 36 in another pile, and put them together, I expect 60 beans altogether. If I count 61 altogether, I will assume that I miscounted somewhere, rather than deny that 24 + 36 = 60. I cannot imagine any possible evidence that could convince me to deny that 24 + 36 = 60.

The truths of logic, math, philosophy, and theology are non-falsifiable. All that means is that they are not “scientific.” Yet, these approaches may be valid in their own right.

Is it pointless to debate an opponent who cannot (yet) describe what evidence would disprove his position? Not necessarily. That is expecting too much from one’s opponent: perfect knowledge of the nature of the controversy.

It is clear that the imagination of a single human is quite limited, and frequently has not considered all possibilities. Suppose I believe something, and I cannot imagine what the disproving evidence would look like. Then, it is my opponent’s task (if he so chooses) to make me aware of the both the possibility and then the actuality of disconfirming evidence.

Here is a good example: the addition of velocities under Galilean relativity, which is consistent with our everyday experience and also our intuitive notions of time and space. I suppose that a pre-Einsteinian would have difficulty imagining the possibility that the Galilean velocity transformation is wrong. In fact, I would have claimed that the Galilean transformation is a “natural law” and a “necessary truth.” I would not have been able to imagine how an experiment could disprove it. And yet, modern science has confirmed the Lorentz transformation and not the Galilean.

Thus, I don’t think an opponent must give his criteria for disproof before I attempt to disprove him. Perhaps when he sees the evidence, It will overcome his presuppositions.

Of course, various debaters have different standards as to what constitutes a “good debate.” I do not expect that everyone must agree with my debating preferences and debate a non-falsifiable view.

(Anyway, Popper’s falsifiability criterion is a philosophical claim, not a scientific one, and is not falsifiable.)

[Kimstu]

[PSA]

For those who wonder what the hell the OP is talking about, the famous “falsifiability criterion” was developed by philosopher of science Karl Popper in the 1930s:

Popper proposed falsification as a way of determining if a theory is scientific or not. If a theory is falsifiable, then it is scientific. If it is not, then it is not science. Popper uses this criterion of demarcation to draw a sharp line between scientific and unscientific theories.

For most philosophers of science and scientists, “falsifiability” has replaced “verifiability” as an empirical criterion for evaluating the scientific status of hypotheses.

[/PSA]

[David_Simmons]

@Kimstu

For most philosophers of science and scientists, “falsifiability” has replaced “verifiability” as an empirical criterion for evaluating the scientific status of hypotheses.

[/PSA]

Because a theory can definitely be shown to be false, whereas it can only be “verified” to a high probablility?

[Mr.Svinlesha]

Groovy.

ragerdude:

• It seems that many people use Popper’s falsifiablilty criterion for a purpose which he did not intend.

His criterion does NOT determine when a claim is “true,” or “acceptable,” or “reasonable,” or even “debatable.” Rather, his criterion determines whether a claim is “scientific” or not; that is, whether it involves a genuinely empirical method. He very well admits that science often errs, and that non-scientific methods may yield the truth.*

This reading of Popper is technically correct, IMO. There are, however, deeper epistemological implications to his argument.

Popper was really a “fan” of science. He was diplomatic in his dismissal of “non-scientific” theories, such as psychoanalysis, but in truth he really rejected them outright (or at least so I’ve been told). Because a “non-scientific” statement is “non-falsifiable,” we can never determine the truth content of the statement; so the criterion really condemns all non-falsifiable statements to a kind of epistemological no man’s land. Thus, (following Popper’s logic) falsifiable, scientific statements are our only means of “moving forward.” In other words, constructing and testing falsifiable statements about Nature (i.e., scientific research as Popper defines it) becomes the only avenue for the growth of knowledge. Non-testable statements about Nature might be true, but they’re also irrelevant, since we can never determine their truth or falsehood (until they can be reformulated into testable statements).

The whole thrust of the scientific project since the late 1600’s has been directed towards deriving “true” statements about Nature (correct me if I’m wrong, Kimstu). So while it’s technically true that Popper’s criterion creates a space for “non-scientific/non-falsifiable” statements of “truth,” it places such statements beyond the bounds of relevant discourse. If you believe that science is directed towards this goal, then for you the criterion is as much a metaphysical assertion of “truth” as it is a demarcation.

• Example: The truths of mathematics are not empirically falsifiable. If I count 24 beans in one pile, and another 36 in another pile, and put them together, I expect 60 beans altogether. If I count 61 altogether, I will assume that I miscounted somewhere, rather than deny that 24 + 36 = 60. I cannot imagine any possible evidence that could convince me to deny that 24 + 36 = 60.*

I understand your point and agree with you in principle, but this is a poorly chosen example. First off, your inability to imagine contradictory evidence that would convince you to deny that 24 + 36 = 60 doesn’t make the statement unfalsifiable. People believe in all kinds of crazy things despite the evidence of their senses.

But besides that, 24 + 36 = 60 is imminently falsifiable by means of mere counting. In addition, I’m not sure it’s the sort of statement Popper was really aiming at with his criterion.

• The truths of logic, math, philosophy, and theology are non-falsifiable. All that means is that they are not “scientific.” Yet, these approaches may be valid in their own right.*

I get your point but disagree with the way you lump “math, logic, philosophy and theology” into a single group. Perhaps “Christianity, psychoanalysis, astrology, and psychic healing,” would be better counter-examples?

If so, then I concur that this is a technically correct reading of Popper. However, I don’t think Popper would agree that the fields I listed above are “are valid in their own right.” For him, they wouldn’t be “valid,” or interesting, until they could be reduced to falsifiable observation statements, and at that point they would also become “scientific.”

David:

• Because a theory can definitely be shown to be false, whereas it can only be “verified” to a high probablility?*

No, because if one accepts the logic of Hume’s critique of inductive reasoning (which Popper did), then a theory can never be verified to any level of probability.

If you like, I’ll run the argument for you when I get a spare moment.

[flight]

Not such a good example.

@ragerdude

Here is a good example: the addition of velocities under Galilean relativity, which is consistent with our everyday experience and also our intuitive notions of time and space. I suppose that a pre-Einsteinian would have difficulty imagining the possibility that the Galilean velocity transformation is wrong. In fact, I would have claimed that the Galilean transformation is a “natural law” and a “necessary truth.” I would not have been able to imagine how an experiment could disprove it. And yet, modern science has confirmed the Lorentz transformation and not the Galilean.

It doesn’t matter if you could possibly imagine WHY it would be false, just that it could be shown to be false. I wouldn’t have been able to come up with a circumstance when they would not add in the Galilean manner, but I could easily see that someone might find some circumstance where he takes measuremements and shows that they do not. All you have to have is one repeatable experiment that disproves you conclusion and your theory, as it stands, is wrong.

So yes, it is scientific because it is falsifiable and always was.

[Kimstu]

DS: Because a theory can definitely be shown to be false, whereas it can only be “verified” to a high probablility?

Yup, I think that was Popper’s reasoning.

Mr. S: *The whole thrust of the scientific project since the late 1600’s has been directed towards deriving “true” statements about Nature (correct me if I’m wrong, Kimstu). *

You’re not wrong at all, IMHO. In fact, I’d say that the whole thrust of the scientific project since at least the second millennium BCE has been directed towards deriving “true” statements about Nature! It’s just that our ideas of acceptable truth criteria have been fluctuating a bit.

*I get your point but disagree with the way you lump “math, logic, philosophy and theology” into a single group. *

Hmmm, I think it makes sense, if we take the common quality of those disciplines to be that they’re methodologically supposed to draw conclusions by valid reasoning from premises, but cannot evaluate the truth of the original premises themselves by means of valid reasoning.

Admittedly, few people these days would agree that philosophy and theology have the same criteria for “valid reasoning” as mathematics and logic, but historically they were supposed to have.

[ragerdude]

I just realized that my point here was not one of epistemology per se, but of aesthetics: What constitutes, for me, a “good debate.” Some people may say to an opponent, “If you can’t tell me upfront what it would take to disprove your position, then debate with you is pointless.”

For me, however, I don’t usually mind debating an opponent who does not acknowledge his position to be falsifiable. My reasons are threefold: 1. Some fields are not empirical. 2. Even in empirical fields, I do not expect my opponent at first to be able to describe the disconfirming data. It may be beyond what he has even considered possible to test. 3. The challenge.

1. My mathematics example was intended to prove this point: Pure math is not an empirical science, but has its own self-contained rules for validity, and is a worthwhile field. The truths of mathematics are not empirically falsifiable. My example was inspired by the claim (of J. S. Mill, “A System of Logic”) that the truths of mathematics are not deductive truths, but merely very strongly confirmed inductive generalizations from experience.

Yet I cannot indicate what kind of physical evidence, if actually enountered, would disconfirm the (analytic) claim that 24 + 36 = 60. Likewize, I cannot indicate what kind of observation would disconfirm the hypothesis that “all bachelors are unmarried.”

But besides that, 24 + 36 = 60 is imminently falsifiable by means of mere counting.

Can you explain this further? I don’t see how an actual count can disprove the “necessary truth” that 24 + 36 = 60. We know that “all bachelors are unmarried” is true by definition, apart from the actual inspection of all bachelors.

2. Yes, Galilean relativity is empirically falsifiable by measurement. So are many claims about subatomic particles. But before the invention of particle accelerators, I never would have guessed that those claims could even be tested. “I never realized we could do that!” So I cannot always require my opponent, in his ignorance of technology, to even imagine the data that would disprove him. He may have no idea (yet) of what the data would look like. Anyway, some results of modern physics are so counterintuitive that it is impossible to imagine the possibility of disconfirming data even after you have collected that data!

3. I think a “good debate” involves a “good challenge.” It can be a satisfying experience to oppose a “non-falsifiable” position in such a way that the evidence is so overwhelming that it disproves what your opponent thought was certain. It is good to fight ignorance, especially the kind that is most stubborn and resistant to the truth. This is especially relevant in theology, where everyone thinks he is right and cannot acknowledge the possibility that he is wrong. It is a great victory for the truth indeed when one can overcome the most incorrigible false beliefs.

[Liberal]

Interesting thread, the topic of which was recently raised in another, concerning the merging of science and religion. I’m a big fan of Popper, and believe that he profoundly understood the essence of science. But it should be pointed out that he does have serious critics, and there are competing theories of science. I recently finished Stove’s Anything Goes. It is a scathing critique of four science philosophers in particular: Lakatos, Kuhn, Feyerabend, and unfortunately Popper. Lumping him in with the other three was a huge mistake, one which Stove even had to admit from time to time. He so often had to point out that Popper was an exception to his criticism that I began to wonder why he included Popper at all. In the end, the book was unsatisfactory, and Popper remains in my mind as the Dean of Science.