Which of These Quotes Violate Popper’s Epistemology? (Hint: None)

On Twitter, I’ve been putting quotes I’ve come across that use language that, in the past, I’ve seen Popperians call out or even go so far as to claim is bad philosophy. But each of these quotes came from Popper himself or one of his followers. I was curious how comfortable Popperians were with Popper’s own language. I was happy to find that, at least among the Popperians I’m linked with on Twitter, people were pretty comfortable with these quotes.

Popper was not a word essentialist so it’s not surprising that he didn’t mind utilizing language flexibly. Here are the quotes I used plus a few new ones:

Popper Endorsing Using Observations As Our Starting Point

There seems to be no reason why we should not make observational experience our provisional ‘starting-point’ — a starting-point, like common sense, not involving commitment to truth or certainty. As long as we are critically inclined it does not matter much where or how we start. (p. 72)

Karl Popper, Objective Knowledge, p. 72

Popper’s point here (in full context) is that since observations are theory-impregnated anyway, sure, go ahead and ‘start with an observation’ just like empiricist claim. Just realize that your senses are not a source of certain truth either.

Popper Endorsing the Concept of “Certainty”

There is a commonsense notion of certainty which means, briefly, ‘certain enough for practical purposes.’ (p. 78)

Objective Knowledge, p. 78

It is simply a fact (whatever philosophers may say) that we are commonsensically certain that the sun will rise… tomorrow.

Objective Knowledge, p. 97

Popper’s point is that when we use the word certainty, we don’t really mean 100% absolutely certain. We mean, for this practical purpose, I’m certain enough. He goes on to give examples of how the situation changing (someone’s life is in danger) might modify what we even mean by ‘certain.’

Popper Endorsing Justification

…we can, if we are lucky, rationally justify a preference for one theory out of a set of competing theories, for the time being… And our justification, though not a claim that the theory is true, can be the claim that there is every indication at this stage of the discussion that the theory is a better approximation to the truth than any competing theory so far proposed.

Objective Knowledge, p. 82

Popper is here clearly saying justification is only valid if you think of it as endorsing the current best theory available. Certainly, the word “justification” can mean just that, and thus be compatible with his epistemology.

Popper Endorsing the Language of Probability

I do not… object to calling… a ‘good’ (or ‘the best’) conjecture a ‘probable’ conjecture (or the most probable of the known conjectures)…

Objective Knowledge, p. 101

When people speak of a theory being ‘probable’ Popperians instantly attack them as being inductive or bayesian. But actually, Popper said that was fine. In context, he goes on to add:

…as long as the word ‘probability’ is not interpreted in the sense of the calculus of probability. For probability in the sense of the calculus of probability has in my opinion nothing whatever to do with the goodness of a hypothesis.

Objective Knowledge, p. 101

I would note that Deutsch tries to avoid the language of probability by instead saying things like ‘that theory doesn’t seem less plausible.’ But go look up the word ‘plausible’ and you’ll find the dictionary definition might turn out to be something like this: “(of an argument or statement) not seeming reasonable or probable; failing to convince.”

It’s hard to get away from the language of probability, in the English language. So it makes sense that Popper made a distinction between referring to a theory as “probable” in the sense of being “corroborated” and believing that said theory was “probable” in the sense of probability calculus. (i.e. that we had any way to assign probability values to it.) This is the essential mistake made by Bayesian Epistemology.

Popper Speaks of Evidence Supporting a Theory

[Suppose in a critical discussion hypothesis h1 is deemed negative because the] agreement of the discussion at time t is that the evidence refutes [hypothesis] h1, while [hypothesis h2] is positive, because the evidence supports h2.

Objective Knowledge, p. 83

To many Popperians today, there is no such thing as evidence “supporting” a theory. Not so for Popper. An attempt to falsify a theory that fails to do is considered “support” for the theory because it “corroborates” it. This is different than merely being “consistent” with the theory. For example: verifying that objects fall as 9.8 m/s is “consistent” with General Relativity but does not “support” the theory (because it is also consistent with Newton’s theory.) Edington’s eclipse experiment “supports” that theory.

This was probably the only quote that got real pushback on Twitter. (Though eventually there seemed to be agreement that it was fine so long as it meant the theory was becoming harder-to-vary.) But Popper is really quite consistent on this one. For example, to Popper a theory is “stronger” if it is “better tested” and it then is understood as having a higher “degree of corroboration.” (And in fact, a higher degree of “confirmation” depending on what you mean by that word.) Is it really that surprising he also spoke of evidence supporting a theory? There is nothing wrong with such wording. You just have to understand the intended meaning.

Donald Campell Refers to Knowledge Creation as “Inductive Achievement”

A blind-variation-and-selective… process is fundamental to all inductive achievement, to all genuine increases in knowledge…The many processes which shortcut… [the] process are in themselves inductive achievements originally by… variation and [selection].

Even in context (feel free to check me), Campbell is talking about any form of knowledge creation. He’s literally using “inductive achievement” as a synonym for any type of knowledge creation.

But Campbell goes on to explain in the footnotes:

The use of the phrase “inductive achievement” is for convenience in communicating and does not in the least imply advocacy of the Bacon-Hume-Mill explanation of those achievements nor disagreement with Popper’s brilliant criticisms of induction.

DONALD CAMPBELL, Evolutionary Epistemology, Rationality, and the Sociology of Knowledge, P. 56

Campbell, unlike modern Popperians, had no problem referring to knowledge-creation as “inductive achievement” because that was the word people used at the time.

Keep in mind that Popper strongly endorsed this article from Campbell. But this is also consistent with how Popper thought of the term “induction” as well.

I never quarrel about words, and I have of course no serious objection if you wish to call the method of critical discussion ‘induction.’


Popper goes on to explain this is a new definition for the term distinct from how it was used previously. But Campell was just taking Popper at his word and his footnote explanation reflects this.

Leave a Reply

Your email address will not be published. Required fields are marked *