[Previous] Refinements of inductive inference by Popperian machines | Home | [Next] Asperger's Syndrome


To a Popperian, absolute certainty and partial certainty are very similar things. To Popper's opponents, they are also very similar things: some people who previously advocated certainty backed off to partial certainty, to try to dodge some criticism without changing their views substantially.

What the hell *is* partial certainty? It doesn't even make sense. One attempt to make sense out of it is to say it's a matter of probability. We can have a theory that is 90% likely to be true, or 99%, or whatever. No attempts of this sort have ever actually worked. You run into problems like how do you establish what the probability of some theory is?

The other main kind of partial certainty is you say some theory is well "supported". Or confirmed. Or verified. Or justified. If anyone tells you they aren't, you just say you didn't mean to provide any guarantees or certainty, they are just *pretty good* theories, pretty well established, pretty reliable. This is all crap. No one really has any coherent idea of what it means. All they do is take philosophy and a ways of thinking that were designed to work with Knowledge meaning Justified, True Belief, and don't change anything at all except saying it's not *absolutely* certainty it's just really quite compelling. So it goes like this. I point out what they are saying might be false. They say, "Whatever. I didn't claim it was perfect. It's pretty good though." So I say, "No, look, it could be false. All the evidence you have is consistent with it being false. How the hell do you know it's pretty good?" So then they say we have to act on the limited knowledge we have, which all points to their theory. And I say, no look, the limited knowledge we have is consistent with infinitely many theories, and you can't just pick out the one you liked in advance and insist we have to act on that one just because we don't have perfect information. Definitely by now, sometimes earlier, they stop speaking coherently, so that's that.

Why are absolute and partial certainty so similar? When one retreats from the stronger form he keeps in mind the same questions and problems. He's still on the same quest. What he wants is to justify and establish Knowledge. He yearns for certainty. He asks questions like how do we know, how can we prove, what would be strong support, which theory has more support, and so on.

Here's another way to put it: there are people who think they Know, and people who accept we are fallible. People after partial certainty are in the Know camp. All they are doing is wearing a disguise because they can't deal with the fallibilist criticisms of the Know position when the position is layed out clearly. So they just won't lay it out clearly anymore.

Aristotle claimed he Knew, and Socrates claimed he did not know. Before Aristotle, philosophers differentiated between Divine Knowledge (episteme) and human opinion (doxa). Human opinion means guesses or conjectural knowledge. It's fallible, useful, and capable of improvement. Divine knowledge is the perfect truth which only the Gods have access to. Aristotle didn't like that. He wanted full blown Knowledge for himself. He invented induction trying to get it. He came up with the idea that definition statements are infallible trying to get it. He destroyed the old distinction, which had been very wise, and claimed the authority to dictate justified, true beliefs. Since then, it went from divine knowledge to just knowledge, and everyone tries to get it, with science, with authority, with anylytic philosophy, with induction, with statistics, with scientism, and so on. And human opinion no longer has a single word of its own, and is ridiculed. No one wants it. They don't want guesses, they want to Know. And if they can't have any guarantees, they want to say they Almost Know. They have Partial Certainty. They'd give anything not to be fallible humans with mere conjectures.

Elliot Temple on July 13, 2009

Messages (11)

There are things you can know for certain or you couldn't have written this post.

Anonymous at 11:13 PM on July 13, 2009 | #1838 | reply | quote

Can you give an argument and an example?

Elliot at 5:09 PM on July 14, 2009 | #1844 | reply | quote

"I exist." You know that for certain. You know for certain you live in the US.

You know for certain that you can write in English. You know for certain you published your post in the Internet.

What you can't know for certain is if I would reply to you. You can't even know how likely it would be. There is no way to calculate that because it solely depends on me, if I want to do it or not.

The question to me seems to be: why can we know some things for certain, and what things are these, and why can't we know other things for certain, at least not immediately, and what things are these?

Anonymous at 2:19 AM on July 15, 2009 | #1845 | reply | quote

I don't think you're using "certain" as exactly as I am. For example, I might not live in the US; I could be on a holodeck in space. That is possible, therefore we can't contradict it with certainty.

Elliot at 8:25 AM on July 15, 2009 | #1849 | reply | quote

There is a curious tension mainstream philosophy (and academia more broadly).

On the one had, dogmatism is eschewed, and open-mindedness is celebrated. Those with too much conviction, too much certainty, and too much zeal are berated as afflicted by a religious faith. On the other hand, the entire rational and scientific enterprise is envisioned as a quest for ever more certainty; although many have rejected absolute certainty as an obtainable goal, it is still the ideal which they strive toward.

It seems to me that the mainstream's objection to dogmatism is not because it is an undesirable goal; the objection is that the conviction of most--if not all dogmatists--is unjustified. The dogmatist, of course, rarely considers his views unjustified, for he has reached the pinnacle, the ideal, for which everyone else aspires--there is no place for a critical rationalist in his world.

Lee Kelly at 9:25 PM on July 15, 2009 | #1851 | reply | quote


What do you think would improve the situation?

Elliot at 9:27 PM on July 15, 2009 | #1852 | reply | quote


I don't know. A lot of these issues are taught and argued under the name of epistemology, but the demand for justification is ultimately an ethical doctrine, and it has permeated almost every level of society. To criticise and reject justificationism is not just mistaken, but actually immoral.

The underlying prescription of justificationism is that one ought not to believe that which one has not justified. Although different schools of philosophy disagree about what kind of experience justifies a belief, each reduces knowledge to beliefs which have the "right" causes--hence the apparent need for induction.

People do generalise from particulars, and, in a sense, it can be said that particulars cause generalisations. But this does not mean there is a branch of logic called induction, because even a bump to the head can cause a generalisation. By seperating logic and psychology, critical rationalists effectively opt out of most of the problems and discussions of mainstream philosophy. There seems to be a great desire to turn the physiological and psychological causes of beliefs into the premises of an argument.

Anyway, by not trying to justify your beliefs, you are being irresponsible, irrational, arbitrary, and possibly dangerous.

Lee Kelly at 11:56 PM on July 15, 2009 | #1853 | reply | quote

I think there are at least two possible things people might mean when they talk about certainty and they are often conflated.

1. A fleeting feeling that an idea is correct.

2. *Justification* for the above feeling.

I mean certainly I often feel certain I am right about some things, even though I clearly I could be wrong.

This is peripheral to what you wrote above, and so not intended as criticism.

I think there is an issue where someone might need to act on a theory, but they don't have confidence. Basically, they don't want to have to take responsibility for relying on the theory. So they prefer to cite how many other people have *tested* the theory, and thereby regard it as true. This has the effect of diffusing the responsibility or placing it elsewhere.

Maybe this is reasonable, given certain circumstances? I'm not sure.

I tend to think that the larger an institution is, the more bureaucratic it becomes, and the more concerned it becomes with avoiding responsibility. (Not just government, but also large corporations which have to respond to shareholders.)

On the other hand, I tend to think that small entrepreneurial firms don't have this burden, and so can take more risks.

This is why I would expect that on a level playing field, small new firms would have an advantage over large older firms. I guess I'm putting things too simplistically, however.

Of course, in neither case is there ever any epistemological certainty.

matt at anarchyjapan.com at 4:04 PM on July 19, 2009 | #1857 | reply | quote

Something I'd like to understand

Forgive me if I'm just having a slow day but don't atheists have certainty?

otherwise you'd be an agnostic right?

Jon at 3:48 PM on October 12, 2009 | #1917 | reply | quote

I have conjectural knowledge that there is no god. An agnostic doesn't.

Your question is about justificationism and fallibilism which is a completely separate issue from atheism and I don't think my stance on it should effect my terminology around atheism.

Elliot at 3:50 PM on October 12, 2009 | #1918 | reply | quote

Lord or Lady?

I've just stumbled on this site and I like its cool tone. Althought I'm here accidentally it's not totally in that I'm looking for ammo to shoot down in flames an old friend who gets on my goat every time she says righteously ,'but you're always making generalisations, and I never do. I hate them.' And your site sounds as if I might be on the track.

Generally I try not to confuse summer with a single swallow, and I avoid like the plague, making generalisations about the Chinese because there are so many of them, but I might be inclined to generalise about the people of Luxembourg, since they're comparatively few,even though I've only ever had a single experience with them and it was in a butcher's shop where they were very rude and I decided that I can't stand them. On the other hand I know quite a few Chinese and I notice that they do seem to have quite a lot in common in their behaviour but I try to confine myself to 'some Chinese are like this'. Have I got a thinking problem?

Am I to write a book here? at 3:47 AM on February 2, 2010 | #1972 | reply | quote

Want to discuss this? Join my forum.

(Due to multi-year, sustained harassment from David Deutsch and his fans, commenting here requires an account. Accounts are not publicly available. Discussion info.)