Fallible Justificationism

This is adapted from a Feb 2013 email. I explain why I don't think all justificationism is infallibilist. Although I'm discussing directly with Alan, this issue came up because I'm disagreeing with David Deutsch (DD). DD claims in The Beginning of Infinity that the problem with justificationism is infallibilism:

To this day, most courses in the philosophy of knowledge teach that knowledge is some form of justified, true belief, where ‘justified’ means designated as true (or at least ‘probable’) by reference to some authoritative source or touchstone of knowledge. Thus ‘how do we know . . . ?’ is transformed into ‘by what authority do we claim . . . ?’ The latter question is a chimera that may well have wasted more philosophers’ time and effort than any other idea. It converts the quest for truth into a quest for certainty (a feeling) or for endorsement (a social status). This misconception is called justificationism.

The opposing position – namely the recognition that there are no authoritative sources of knowledge, nor any reliable means of justifying ideas as being true or probable – is called fallibilism.

DD says fallibilism is the opposing position to justificationism and that justificationists are seeking a feeling of certainty. And when I criticized this, DD defended this view in discussion emails (rather than saying that's not what he meant or revising his view). DD thinks justificationism necessarily implies infallibilism. I disagree. I believe that some justificationism isn't infallibilist. (Note that DD has a very strong "all" type claim and I have a weak "not all" type claim. If only 99% of justificationism is infallibilist, then I'm right and DD is wrong. The debate isn't about what's common or typical.)

Alan Forrester wrote:

[Justification is] impossible. Knowledge can't be proven to be true since any argument that allegedly proves this has to start with premises and rules of inference that might be wrong. In addition, any alleged foundation for knowledge would be unexplained and arbitrary, so saying that an idea is a foundation is grossly irrational.

I replied:

But "justified" does not mean "proven true".

I agree that knowledge cannot be proven true, but how is that a complete argument that justification is impossible?

And Alan replied:

You're right, it's not a complete explanation.

Justified means shown to be true or probably true. I didn't cover the "probably true" part. The case in which something is claimed to be true is explicitly covered here. Showing that a statement X is probably true either means (1) showing that "statement X is probably true" is true, or it means that (2) X is conjectured to be probably true. (1) has exactly the same problem as the original theory.

In (2) X is admitted to be a conjecture and then the issue is that this conjecture is false, as argued by David in the chapter of BoI on choices. I don't label that as a justificationist position. It is mistaken but it is not exactly the same mistake as thinking that stuff can be proved true or probably true.

In parallel, Alan had also written:

If you kid yourself that your ideas can be guaranteed true or probably true, rather than admitting that any idea you hold could be wrong, then you are fooling yourself and will spend at least some of your time engaged in an empty ritual of "justification" rather than looking for better ideas.

I replied:

The basic theme here is a criticism of infallibilism. It criticizes guarantees and failure to admit one's ideas could be wrong.

I agree with this. But I do not agree that criticizing infallibilism is a good reply to someone advocating justificationism, not infallibilism. Because they are not the same thing. And he didn't say anything glaringly and specifically infallibilist (e.g. he never denied that any idea he has could turn out to be a mistake), but he did advocate justificationism, and the argument is about justification.

And Alan replied:

Justificationism is inherently infallibilist. If you can show that some idea is true or probably true, then when you do that you can't be mistaken about it being true or probably true, and so there's no point in looking for criticism of that idea.

My reply below responds to both of these issues.


Justificationism is not necessarily infallibilist. Justification does not mean guaranteeing ideas are true or probably true. The meaning is closer to: supporting some ideas as better than others with positive arguments.

This thing -- increasing the status of ideas in a positive way -- is what Popper calls justificationism and criticizes in Realism and the Aim of Science.

I'll give a quote from my own email from Jan 2013, which begins with a Popper quote, and then I'll continue my explanation below:

Realism and the Aim of Science, by Karl Popper, page 19:

The central problem of the philosophy of knowledge, at least since the Reformation, has been this. How can we adjudicate or evaluate the far-reaching claims of competing theories and beliefs? I shall call this our first problem. This problem has led, historically, to a second problem: How can we justify our theories or beliefs? And this second problem is, in turn, bound up with a number of other questions: What does a justification consist of? and, more especially: Is it possible to justify our theories or beliefs rationally: that is to say, by giving reasons -- 'positive reasons' (as I shall call them), such as an appeal to observation; reasons, that is, for holding them to be true, or at least 'probable' (in the sense of the probability calculus)? Clearly there is an unstated, and apparently innocuous, assumption which sponsors the transition from the first to the second question: namely, that one adjudicates among competing claims by determining which of them can be justified by positive reasons, and which cannot.

Now Bartley suggests that my approach solves the first problem, yet in doing so changes its structure completely. For I reject the second problem as irrelevant, and the usual answers to it as incorrect. And I also reject as incorrect the assumption that leads from the first to the second problem. I assert (differing, Bartley contends, from all previous rationalists except perhaps those who were driven into scepticism) that we cannot give any positive justification or any positive reason for our theories and our beliefs. That is to say, we cannot give any positive reasons for holding our theories to be true. Moreover, I assert that the belief we can give such reasons, and should seek for them is itself neither a rational nor a true belief, but one that can be shown to be without merit.

(I was just about to write the word 'baseless' where I have written 'without merit'. This provides a good example of just how much our language is influenced by the unconscious assumptions that are attacked within my own approach. It is assumed, without criticism, that only a view that lacks merit must be baseless -- without basis, in the sense of being unfounded, or unjustified, or unsupported. Whereas, on my view, all views -- good and bad -- are in this important sense baseless, unfounded, unjustified, unsupported.)

In so far as my approach involves all this, my solution of the central problem of justification -- as it has always been understood -- is as unambiguously negative as that of any irrationalist or sceptic.

If you want to understand this well, I suggest reading the whole chapter in the book. Please don't think this quote tells all.

Some takeaways:

  • Justificationism has to do with positive reasons.

  • Positive reasons and justification are a mistake. Popper rejects them.

  • The right approach to epistemology is negative, critical. With no compromises.

  • Lots of language is justificationist. It's easy to make such mistakes. What's important is to look
    out for mistakes and try to correct them. ("Solid", as DD recently used, was a similar mistake.)

  • Popper writes with too much fancy punctuation which makes it harder to read.

A key part of the issue is the problem situation:

How can we adjudicate or evaluate the far-reaching claims of competing theories and beliefs?

Justificationism is an answer to this problem. It answers: the theories and beliefs with more justification are better. Adjudicate in their favor.

This is not an inherently infallibilist answer. One could believe that his conception of which theories have how much justification is fallible, and still give this answer. One could believe that his adjudications are final, or one could believe that his adjudications could be overturned when new justifications are discovered. Infallibilism is not excluded nor required.


Looking at the big picture, there is the critical approach to evaluating ideas and the justificationist or "positive" approach.

In the Popperian critical approach, we use criticism to reject ideas. Criticism is the method of sorting out good and bad ideas. (Note that because this is the only approach that actually works, everyone does it whenever they think successfully, whether they realize it or not. It isn't optional.) The ideas which survive criticism are the winners.

In the justificationist approach, rather than refuting ideas with negative criticism, we build them up with positive arguments. Ideas are supported with supporting evidence and arguments. The ones we're able to support the most are the winners. (Note: this doesn't work, no successful thinking works this way.)

These two rival approaches are very different and very important. It's important to differentiate between them and to have words for them. This is why Popper named the justificationist approach, which had gone without a name because everyone took it for granted and didn't realize it had any rival or alternative approaches.

Both approaches are compatible with both infallibilism and fallibilism. They are metaphorically orthogonal to the issue of fallibility. In other words, fallibilism and justificationism are separate issues.

Fallibilism is about whether or not our evaluations of ideas should be subjected to revision and re-checking, or whether anything can be established with finality so that we no longer have to consider arguments on the topic, whether they be critical or justifying arguments.

All four combinations are possible:

Infallible critical approach: you believe that once socialist criticisms convince you capitalism is false, no new arguments could ever overturn that.

Infallible justificationist approach: you believe that once socialist arguments establish the greatness of socialism, then no new arguments could ever overturn that.

Fallible critical approach: you believe that although you currently consider socialist criticisms of capitalism compelling, new arguments could change your mind.

Fallible justificationist approach: you believe that although you currently consider socialist justifying arguments compelling (at establishing the greatness and high status of the socialism, and therefore its superiority to less justified rivals), you are open to the possibility that there is a better system which could be argued for even more strongly and justified even more and better than socialism.


BTW, there are some complicating factors.

Although there is an inherent asymmetry between positive and negative arguments (justifying and critical arguments), many arguments can be converted from one type to the other while retaining some of the knowledge.

For example, someone might argue that the single particle two slit experiment supports (justifies) the many-worlds interpretation of quantum physics. This can be converted into criticisms of rivals which are incompatible with the experiment. (You can convert the other way too, but the critical version is better.)

Another complicating factor is that justificationists typically do allow negative arguments. But they use them differently. They think negative arguments lower status. So you might have two strong positive arguments for an idea, but also one mild negative argument against it. This idea would then be evaluated as a little worse than a rival idea with two strong positive arguments but no negative arguments against it. But the idea with two strong positive arguments and one weak criticism would be evaluated above an idea with one weak positive argument and no criticism.

This is easier to express in numbers, but usually isn't. E.g. one argument might add 100 justification and another adds 50, and then a minor criticism subtracts 10 and a more serious criticism subtracts 50, for a final score of 90. Instead, people say things like "strong argument" and "weak argument" and it's ambiguous how many weak arguments add up to the same positive value as a strong argument.

In justification, arguments need strengths. Why? Because simply counting up how many arguments each idea has for it (and possibly subtracting the number of criticisms) is too open to abuse by using lots of unimportant arguments to get a high count. So arguments must be weighted by their importance.

If you try to avoid this entirely, then justificationism stops functioning as a solution to the problem of evaluating competing ideas. You would have many competing ideas, each with one or more argument on their side, and no way to adjudicate. To use justificationism, you have to have a way of deciding which ideas have more justificationism.

The critical approach, properly conceived, works differently than that. Arguments do not have strengths or weights, and nor do we count them up. How can that be? How can we adjudicate between competing ideas with out that? Because one criticism is decisive. What we seek are ideas we don't have any criticisms of. Those receive a good evaluation. Ideas we do have criticisms of receive a bad evaluation. (These evaluations are open to revision as we learn new things.) (Also there are only two possible evaluations in this system. The ideas we do have criticisms of, and the ideas we don't. If you don't do it that way, and you follow the logic of your approach consistently, you end up with all the problems of justificationism. Unless perhaps you have a new third approach.)


Elliot Temple | Permalink | Messages (0)