Monday, December 31, 2007

An Objectivist critique of Popper examined

Nicholas Dykes' critique of Popper, linked by the poster Ian, while not as bad as most such Rand-inspired criticisms, does have its share of serious problems.

Dykes begins his critique by complaining about Popper's tone of assurance.
One thing which is quite certain is that Popper wrote with absolute assurance of his own rectitude, as I think the quotations in this paper reveal. For all his belittlement of knowledge and certainty, I have never read anyone who wrote so many books all imbued with such conscious certainty and authority— the authority of one who knows.
Dykes appears to be annoyed by the fact that Popper never prefaces all his remarks with the phrase I suppose or I conjecture. Of course, it would be very tedious to proceed in this way. Dykes also takes Popper to task for declaring "I am not a belief philosopher. I do not believe in belief" while at the same time refusing, in other places in his work, to stop using the phrase I believe to state one of his positions. Here Dykes shows himself deaf to the ambiguity of language. Popper's phrase "I don't believe in belief" plays on two senses of the word belief to make a point. In the first sense of the term Popper is using it to describe certain belief, in the second, the sort of conjectural belief Popper supported. In other words, Popper is simply saying that he doesn't believe (in the conjectural sense of the term) in certainty.

Dykes unquestioningly accepts the Platonic view that equates knowledge with certainty, asserting that all denials of certainty are "self-contradictory," because "in the absence of certain knowledge one is either forced into a position involving some kind of unfounded conviction, belief or faith, or into scepticism." Says who? Dykes here make use of an oft-repeated Objectivist fallacy: the either-or fallacy, where we are given the choice between the Objectivist position and several unappetizing alternatives. But who says that the only alternative between certain knowledge is unjustified belief and skepticism? What about probable knowledge? What about degrees of reliability? If one knowledge claim can be regarded as superior to another, isn't that good enough for practical purposes?

Some of Dykes criticisms demonstrate a lack of familiarity with Popper's ideas. He accuses Popper, for instance, of refusing "to have anything to do with definitions." This is an exaggeration. Popper accepted scientific, or nominalistic, definitions; he simply has no use for essentialist definitions — another matter entirely. I can find no evidence that Dykes understands this distinction, or has any idea why essentialist definitions, and the scholastic mythology that has grown-up around them, deserve the criticism and scorn Popper directed at them.

Dykes essay particularly flounders when he attempts to explode Popper's critical rationalism by associating it with the views of Kant and Hume. He repeats the Objectivist canard that Hume's "whole argument" is in conflict with the Law of Identity. Alas, this misses the point entirely. Hume's argument is not directed at the Law of Identity, but at our knowledge of specific identities. How do we know which attributes of an object are constant and which are not? This is not a question which the Law of Identity can answer. Reminding ourselves that objects have identities is nothing to the purpose if we don't know, or can't be sure, what those identities are.

Dykes is equally clueless when it comes to Kant. Because Popper believed that all observations are "theory impregnated," Dykes assumes that Popper regards all theories as prior to experience, and even wonders whether Popper is "asking us to accept that the heliocentric theory came before observation of perturbations in planetary orbits?" Again, however, Dykes has missed the point. When Popper asserts that observations are "theory-impregnated" he is merely noting how hopeless it would be for an empty mind, bereft of any presuppositions or "theories," to make sense of observations. As Kant put it, "Thoughts without content are empty, intuitions [i.e., sense experiences] without concepts are blind." Regardless of whatever confusions and pendantries Kant may have stumbled into trying to elucidate this paradox in the Critique of Pure Reason, the principle itself constitutes one of the seminal insights of epistemology, easily corroborated by common experience and scientific investigation. I wonder if Dykes has any familiarity, let alone understanding, of it. Or does he believe that facts can be understood without any prior theories at all? Hardly a plausible position, if so. After all, how can a blank mind ever make neither heads nor tails of the bewildering complexity of sense experience, without at least some prior heuristic to guide it? It can't. So some measure of theory does appear to precede fact. How, then, on a realist framework, are we to account for this? Popper at least deserves credit for taking this issue seriously and trying to provide a non-idealist, non-Kantian solution to it. Since Dykes does not even appear to grasp that this is what Popper is attempting to do, his criticism is worthless. One cannot effectively criticize what one doesn't understand.

And that seems to be the over-riding problem of Dykes critique. His understanding of the problems Popper attempts to solve is minimal, at best. His attachment to Randian categories of interpretion has rendered him hopelessly naive in the face of problems originally posed by Hume and Kant and later elucidated by Pierce, Santayana, Popper and Polanyi. His outlook is still trapped in the scholasticism of Plato and Aristotle.


Jay said...


What about Dyke's contention that Popper did not apply his exacting trial-and-error scrutiny to his own beliefs? I am admittedly new to Popper but this seems like a legitimate charge, if true.

Wells said...


Popper was human, all to human. Dyke criticism in that area doesn't mean much. Simply because nobody ever applies enough scrutiny to their own beliefs. That's why you have peer review, and publishing stuff so that other people can tell you that you are wrong.

Besides, even if Karl Popper did not apply his exacting trial-and-error scrutiny to his own beliefs. All that really means philosophically is that some young Popperian, fresh out of philosophy school can, instead of being unemployed, fix what Karl Popper broke.

Jay said...

Certainly he was human, however, it remains a serious charge because so much of his mantra is that we need to be constantly on the lookout for things that refute our conjectures. Dykes seems to be saying: "What of your own?"

Specifically, the most interesting question I saw raised was: if everything is just a conjecture, isn't that itself a mere conjecture that we can't be 100% certain of?

Dragonfly said...

I can't speak for Popper, as I don't know what he exactly said, but perhaps this link might give you an explanation:

gregnyquist said...


I would question whether Dykes' charge against Popper is really true. The charge seems to be based merely on the fact that Popper was rather aggressive and insistent about his own beliefs and that he could be a bully in debate. But that doesn't mean that Popper hadn't criticized his own beliefs in private, or that when confronted with valid criticisms of one of this theories, he didn't acknowledge the criticism and attempt to fix the theory. In the 1970s, when Popper's definition of verisimilitude was refuted, Popper acknowledged the fact and moved on. Popper also never insisted on being critical of oneself; he merely insisted on being open to criticism from others. He recognized the difficulty of self-criticism (perhaps from understanding his own difficulties in that line), and even went so far as to suggest that thinkers should aggressively defend their theories in debate and only abandon them when they were refuted by others. Popper placed great emphasis on discussion, on the social aspect of knowledge formation. He recognized that human beings are better at noticing the flaws in the ideas of other than in their own ideas.

Anonymous said...

Bottom line, it doesn't matter if Popper practiced what he preached. However, the reason why he didn't perform a systematic refutation of his philosophies as he did with Plato's and Marx's is because, well, he agreed with himself. It's absurd to suggest that he shouldn't.

That's not to say Popper didn't look for weaknesses in his arguments or consider opposing views. For example, consider this lecture Popper gave on his Three Worlds idea:

"Many of my philosophical friends, especially those who are materialists or physicalists, are strongly opposed to all this. They say that my way of talking is seriously misleading... They say that in speaking of world 3 objects, I am guilty of hypostatization... I therefore wish to explain to you the strong objections to my views about world 3 objects raised by my philosophical friends, the monists as well as the dualists. Let me first explain what a materialist or physicalist monist would say..."

The whole lecture, too, is peppered with "my thesis" and "I suggest." He's not speaking "with the authority of one who knows;" he's putting forth a conjecture.

-- Ian.

(Incidentally, thanks for the great post, Greg.)

Jay said...


That Wiki would be great for the Rand critics who dismiss the Objectivist ethics by pointing to Rand's personal life.

My point wasn't to ad hominem Popper, but to suggest that more critical thinking might have exposed holes in his ideas. It only seems logical that if all knowledge is mere conjecture, then so is that idea itself.

Wells said...


This whole argument is recursive, but since I've been called a master of recursion way back when I was in college, I'll try to tackle it.

We'll assume that all knowledge is statements.

Now we have a statement 'All statements are mere conjectures.' You naturally contend that this statement should apply to itself. If it does, than it is true, and if it doesn't than it is false. However if the statement applies to itself it is false because the universal qualifier (All) doesn't go with things like 'mere conjectures.' because it means that the statement is an absolute statement.
Remember this paragraph, I'll get back to it.

But first let's apply the statement how it was meant to be applied; to other claims of knowledge. I'll put examples that I believe 'Brains of organisms can be mathematically described as Turing machines.' and 'Quantum Physics describes reality.' Karl Popper would argue that I should be rather skeptical about both of these claims.
He's probably right, even if my claims are true. So I'll propose a few new statements 'The statement 'Brains of organisms can be mathematically described as Turing machines.' is a mere conjecture.' and 'The statement 'Quantum Physics describes reality.' is a mere conjecture.'
I could continue in this vein with everything I believe (and I probably believe an awful lot), as well as statements about everything I believe. These statements would have the form 'The statement X is a mere conjecture.' ('The statement 'The statement 'Brains of organisms can be mathematically described as Turing machines.' is a mere conjecture' is a mere conjecture.'!!!). But you wanted philosophical argument, so I won't waste your time.

Now we can go back to the first substantive paragraph, the one I asked you to remember. I can accept the statement 'All statements are mere conjectures.' as true, or I can reject it as false.
If it is false, then I would have to reject at least one statement of the form 'The statement X is a mere conjecture.' to make it false. The question is which statement? I can't tell, therefore it appears to be true.
If I accept the statement 'All Statements are mere conjectures' as true, it appears I have a different problem, I've accepted an absolute statement, so I would have to reject it as false, but in addition I would have to reject some other statement of the form 'The statement X is a mere conjecture.' to make it false. I can't do both because I don't know which other statement to reject, so I'm stuck with doing neither.

Ellen Stuttle said...

Jay wrote:

My point wasn't to ad hominem Popper, but to suggest that more critical thinking might have exposed holes in his ideas. It only seems logical that if all knowledge is mere conjecture, then so is that idea itself.

Popper is talking about universal generalizations about matters of fact, not about tautological statements or the conclusions of sound deductive reasoning. It's deductively demonstrable that universal statements about matters of fact can't be demonstrated.

Did you look at the post Dragonfly linked?

Here's the URL again:

I think he did a good brief job of explaining the issue there. Others of us also explained it on the same thread.


Jay said...


Just looked at that link. The way I see it, this is an argument over words that neither side is going to change their minds on. To Popperians, something we have every reason to believe is true is a conjecture, albeit a strong one. To Objectivists, we are contextually certain of it.

Interestingly, both sides invoke the lack of human omniscience in their arguments; Popperians say we can't empirically observe every example of something because of it, and Objectivists say the very notion of certainty is contextual because of it. Does the disagreement really go any deeper than that?

Ellen Stuttle said...

Jay, you asked (post next above):

Does the disagreement really go any deeper than that?

Interesting question. I think that the substantive disagreement doesn't go deeper. Problem is, though, that the O'ists want to accuse Popper of trying to destroy science, and blah blah, when in fact their notion of "contextual certainty" concedes that we never do know that a universalized generalization will hold up to further testing. The O'ists condemn Popper, while acknowledging, using a different way of saying it, that he was right.

An additional problem is the peculiarness of "contextually certain." I think it's weasel language: it gives the appearance of having established certainty, while the content of the thesis hasn't done that.


Daniel Barnes said...

>To Popperians, something we have every reason to believe is true is a conjecture, albeit a strong one. To Objectivists, we are contextually certain of it.

Just dropping briefly in passing.

This is something I have belaboured for some time. You're right, Jay, as Ellen also points out - there is no substantive difference between the two positions. The problem is that Objectivists don't know this. Further, having been conditioned by Rand's wild rhetorical attacks on all forms of skepticism they would find the idea that underneath it all it is logically equivalent to that which she attacked not only incredible, but even offensive.

Just to make the comparison perfectly clear, we might summarise "contextual certainty" (as Fred Seddon correctly did in his otherwise shambolic review of Greg's book) as follows: "We may know P, but P may be false." In other words, it's just another way of saying knowledge is tentative, and might be overthrown.

We Popperians would have no problem in accepting this comparison, and welcoming Objectivists as brothers in arms.

But would Objectivists do likewise? Personally I think we'll be snowboarding in hell before that happens...;-)

Daniel Barnes said...

>Specifically, the most interesting question I saw raised was: if everything is just a conjecture, isn't that itself a mere conjecture that we can't be 100% certain of?

Yes. Exactly! Perhaps we will find irrefutably certain knowledge one day. Hence this claim is not self-contradictory. That Dykes thinks this is some kind of logical problem for skeptics merely shows his enthusiasm for regurgitating fallacious Randian boilerplate.

Drew Zi said...

It seems to me there is a fundemental difference. No one in their right might would ever conclude that human's are not fallible or could be wrong in the future about anything. but to say that at heart Rands theory and Popper's theory are the same is to try to legitimize Rands epistemology, which is basically inductivism at heart (and even inductivists accept human fallibility and scientific uncertainty). The basic difference (and it is a substanial one) in Popper was that all theory is guess work, there is no one "method" for coming up with useful ideas or theories, useful ideas can come out of anything. Rand believed there was one method, which is her "concept formation" theory.

Daniel Barnes said...

Drew Zi,

The situation is highly confused in classic Randian fashion. Basically there's 1) what she thought her epistemology entailed, and 2) what it actually entailed. Cut through her verbalism and linguistic three card monte and you - charitably - can make a case that she ends up in a Popperian skepticism. In fact a Rand scholar, Fred Seddon, has tried to argue this. She of course would have hated the idea...

Drew Zi said...

I have read quite a lot into Ayn Rand (especcially Peikoffs popularizations) and I still feel they are inductivist,and ot say they are similar to Popper's is to either seriously simplify Critical Rationalism, or it is too see in Rand's writings nuances that aren't there. I have heard nothing by any rand popularized that come close to what popper was saying, and I also think they implicitly subscribe to naive realism. I will have to read this book by sneddon. Don't Objectivists think that Popper is an irrationalist? Then again, the might, because they seems to have no idea what he is talking about.