Are you asking about certainty? That's easy to dispose of, although the answer in brief is not very satisfying: in the world there ain't no certainty. (Abstract certainty, e.g., 2+2=... is a different different discussion: it seems certain....)

Probability? Sure, why not. The math is (seems) certain, but the underlying assumptions, maybe not so much. (See frequentism v. Bayesianism.)

Both probability and certainty as abstract ideas become increasingly difficult to pin down, the closer you try to get to them. As tools, on the other hand, they're perfectly good. Is the woman in the robe who just cooked your breakfast your mother? Are you sure? Are you really sure, no shadow of a doubt possible?

...where can we look for a relevant kind of probability? Or is the idea of a probable belief just so much nonsense.... — PossibleAaran

It's useless to look for a tool if you don't know what the job is. Especially for a tool that does everything: that tool doesn't exist. Also, belief is almost always a sign that both certainty and probability are out the window.

Aristotle covered this (from Nicomachean Ethics, but I can't find the quote I'm looking for): it's a mistake to look for more clarity than a subject matter will allow.

It's useless to look for a tool if you don't know what the job is. Especially for a tool that does everything: that tool doesn't exist. Also, belief is almost always a sign that both certainty and probability are out the window — tim wood

The "tool" I am looking for is one which distinguishes mere guess work from credible belief. A credible belief is one which is probably true, but in what sense of "probably"? We can talk of distinguishing speculation from credible theory if you don't like talk of belief. Can't we say anything more exact than "a credible theory is one which is probable. Probable in some sense or other, but not statistical and not subjective"? That's not particularly illuminating.

Aristotle covered this (from Nicomachean Ethics, but I can't find the quote I'm looking for): it's a mistake to look for more clarity than a subject matter will allow. — tim wood

I know the quote you want, and I agree with it. But it also makes sense to look for clarity wherever possible, if only to stave off the impression that the unclear ideas are really empty words.

This is a good question and one I have thought about quite a bit. Probability has a mathematical definition, but it is instrumental, not fundamental, and is of no use in everyday life, or to philosophers.

Where I've ended up at the moment is to use the twin notions of confidence and surprise to make sense of probability. If I say a possible event has a high probability I mean I am very confident it will happen, and that I will not be at all surprised if it does happen, and fairly surprised if it doesn't.

The probability number, between 0 and 1, measures just how confident I am, and just how surprised I would be if it doesn't happen.

'Credibility' is a little more complex, as it can have a specific technical meaning, related to the volume of data on which a belief is based. There is a branch of Actuarial Science known as Credibility Theory, that is rooted in Bayesian analysis. But in the end credibility boils down to another measure of confidence.

The "tool" I am looking for is one which distinguishes mere guess work from credible belief. A credible belief is one which is probably true, but in what sense of "probably"? We can talk of distinguishing speculation from credible theory if you don't like talk of belief. Can't we say anything more exact than "a credible theory is one which is probable. Probable in some sense or other, but not statistical and not subjective"? That's not particularly illuminating. — PossibleAaran

The test is use, the criteria being, does it work? No belief, no theory, no probable.

I don't mean to be reductionist, but your question seems not yet entirely developed.

It's a tedious exercise, but you might try defining your terms. You may find in your effort that you don't find a definition, rather you establish it - are forced to establish it - and likely will find your questions dissolve in your efforts, the definitions you create, and the resultant understanding. In short, you're looking for something; in a sense you're looking to be told, and that's not quite how it works.

It is common to say that my belief is still credible even though not certain because it is "probable": — PossibleAaran

What does it mean that a belief is "credible?" Doesn't it mean there is justification for it? What does being probable mean? It's a statement of conditional truth, an expression of the uncertainty and conditionality of knowledge. Aren't you just recapitulating Descartes? I think I probably am therefore I am.

Maybe I've misunderstood you.

Hi

If I say a possible event has a high probability I mean I am very confident it will happen, and that I will not be at all surprised if it does happen, and fairly surprised if it doesn't. — andrewk

credibility boils down to another measure of confidence. — andrewk

Thanks for the interesting thoughts. I began by trying to distinguish mere guess work type beliefs from credible beliefs. Contrast the belief that God created humans from nothing, as they currently are, with the belief that humans evolved from prior species over time. On your approach, the only difference between the two is our confidence. We are (I am?) more confident that the evolutionary story is true than the creation story. Evolution is credible just in the sense that we are (I am?) more confident that it is true. Of course, the creationist is more confident that the creation story is true and so the creation story is credible in the sense that he is more confident that it is true. Is this not quite a sceptical result? That no belief is more credible than any other except in the very weak sense that we believe some things and don't believe others? Perhaps you are happy with this.

The test is use, the criteria being, does it work? — tim wood

Is the suggestion here that a credible theory is just one that is useful?

you might try defining your terms — tim wood

Which terms do you want defined? I am trying to figure out whether it makes any sense to think of some beliefs as more credible than others. In what sense is any belief or theory ever more credible than another?

What does it mean that a belief is "credible?" Doesn't it mean there is justification for it? — T Clark

We can say that a belief is more credible - more worthy of our ascent - than another when there is justification for it, but what do you mean by "Justification"? You don't necessarily have to give a definition here, although you could do that. But you could instead illustrate the notion with some examples. That would be helpful.

Aren't you just recapitulating Descartes? I think I probably am therefore I am. — T Clark

I'm not making the cogito argument at all, and as far as I know, Descartes didn't write on probability.

We can say that a belief is more credible - more worthy of our ascent - than another when there is justification for it, but what do you mean by "Justification"? You don't necessarily have to give a definition here, although you could do that. But you could instead illustrate the notion with some examples. That would be helpful. — PossibleAaran

So something that's credible is "worthy of our ascent?" Does that mean "true?" "Probably true." This is getting a bit circular.

Here's a definition of "justification" from the web - The action of showing something to be right or reasonable. In this context, justification is an answer to the question "How do you know that?" Example - I say "John stole the money." You say "How do you know that?" I say "He told me he did it." It doesn't mean it's definitely true, only that I have good reason to believe it.

I'm not making the cogito argument at all, and as far as I know, Descartes didn't write on probability. — PossibleAaran

You were writing about doubt. Ultimately, that brings us back to Descartes. I was being clever.

Or is the idea of a probable belief just so much nonsense put forward in a desperate attempt to stave off scepticisim? — PossibleAaran

That is what I believe. But I do not think the strategy of casting beliefs in terms of probability works.

Bayesian analysis of inference is bunk, as far as I'm concerned. It's just dressed up mathematical hoolabaloo -- it makes someone feel good about inference, while at the same time not really looking like inference in practice. One can use Bayesian analysis, but that's very different from inference -- one uses Bayesian analysis because inference has already taken place, and one desires to have some sort of mathematical model after that fact.

Beliefs are true, false, or neither. We are justified in holding a belief, or we are not. There is no percentage roll that makes a belief more or less justified. It simply is, or isn't.

So something that's credible is "worthy of our ascent?" Does that mean "true?" "Probably true." This is getting a bit circular. — T Clark

I am still not sure that you understand what I am trying to ask. It is usual to distinguish between beliefs which are, in some sense credible, and beliefs which aren't. Compare the creationist account of human beings with the evolutionary account; or compare me believing that there are exactly 3 million and two blades of grass in Birmingham city centre with your belief that you are looking at a computer screen (I haven't counter, and no one has told me about the grass). In some sense the latter beliefs of each pair are superior to the former; more credible. In what sense exactly? Your remark above accuses me of circularly defining "credible", but I'm not defining credible at all. I am asking someone to tell me the difference between these sorts of belief, and just stipulating that we call that difference "credibility".

Here's a definition of "justification" from the web - The action of showing something to be right or reasonable. In this context, justification is an answer to the question "How do you know that?" Example - I say "John stole the money." You say "How do you know that?" I say "He told me he did it." It doesn't mean it's definitely true, only that I have good reason to believe it. — T Clark

Sure, so the difference between the belief pairs above is that in each case, one item of the pair has "good reason to believe it" and the other doesn't. Is that what you are suggesting?

We are justified in holding a belief, or we are not. There is no percentage roll that makes a belief more or less justified. It simply is, or isn't. — Moliere

Interesting post Moliere. I am tempted to think you are right about probability. So you hold that the difference between, for example, me believing that there are exactly 3 million and two blades of grass in Birmingham city centre and you believing that you are looking at a computer screen (I haven't counter, and no one has told me about the grass), is that you are justified and I am not. Alright, what do you mean by justified?

Alright, what do you mean by justified? — PossibleAaran

I think there is a set of principles we can follow that makes justification work. And epistemology is the debate about which principles we choose. I remember that you and I -- we -- agree that seeming is one such principle. So if something seems that it is so, then we are justified in holding a belief. Intersubjective seemings -- as clunky as it sounds -- is another such principle. To go to science, experimental setup and communication of results from said setup is another justification. If it is widely accepted by others then we are justified in holding a belief.

There is a kind of hierarchy, I think, between the principles. Seeming is less than intersubjective seeming is less than experimental setup is less than widely accepted experimental setups. Maybe the hierarchy of justifications is another principle in the set?

I'm not entirely certain. Justification is the beast, IMO, of JTB theory of knowledge. (not that JTB is the end-all be-all. Far from it. It's just a good starting point that's nice to hash out)

↪tim wood
The test is use, the criteria being, does it work?
— tim wood
Is the suggestion here that a credible theory is just one that is useful?

you might try defining your terms
— tim wood
Which terms do you want defined? I am trying to figure out whether it makes any sense to think of some beliefs as more credible than others. In what sense is any belief or theory ever more credible than another? — PossibleAaran

You seem to scant evidence, experience, and boundaries established by sense, test, and logic. And it's not clear to me you know what "credible" means.

From your OP you list statistical probability and subjective probability; i.e., the maths of proportions, and plausible belief. Then your question:

If neither of these species of probability will do the trick, where can we look for a relevant kind of probability? Or is the idea of a probable belief just so much nonsense put forward in a desperate attempt to stave off scepticisim? — PossibleAaran

Trick for what, exactly? And, each - all - of the significant words in your post are trouble and ultimately themselves destructive of sense unless kept on a tight rein; i.e., well-defined.

Can you recast your questions? I cannot find the substance of them, and thereby cannot sensibly respond to them without first telling you either what to think, or what you're thinking, and I don't care to do either.

Is this not quite a sceptical result? That no belief is more credible than any other except in the very weak sense that we believe some things and don't believe others? Perhaps you are happy with this. — PossibleAaran

To use 'credible' in the sense it is used in this quote is to use it in an absolute sense. I see credibility as relative to the person who gives or withholds credence. It is like beauty in that way. I can see no more hope of coming up with a workable notion of mind-independent credibility than of beauty. After all, 'credible' relates to belief, which has no meaning at all in the absence of a mind.

But I don't think that leaves us in a state of sceptical helplessness. I can say that the Creationist story is not credible to me, while Darwin's story is. Further, I can observe that the Creationist story is not credible to most people that have had the benefit of a scientific education.

Another thing we can do is try to analyse what lies behind our confidence in a belief. The first step in that is often that we have to trace it back, working backwards through an implicit deductive process, to find the beliefs from which it was deduced, and then the beliefs from which they were deduced and so on. Continuing in this way, we eventually come to a set of fundamental beliefs that are accepted without proof - things like the principle of induction, the general reliability of my memory and the presence of other minds. I doubt there is anything in my fundamental set of beliefs that any but a tiny minority of the world's population would reject, and I don't think Creationists would reject them either.

But when we do the same with a Creationist, we will generally find some additional fundamental beliefs that I do not hold. Some of them are likely to be about the God they believe in. When those beliefs are put together with the ones we share, they end up having to make strange-seeming deductions in order to keep their system consistent. For example, they deduce that God must have planted fossils in the ground a few thousand years ago, even though the organisms of which they appear to be fossilised remains never lived.

There is nothing wrong with such a belief, but I reject it on aesthetic grounds. I like simplicity and clarity, and such an account takes us very far from such a standard. It raises the question 'Why on Earth would a God do such a thing?'. Now sure any worldview will have plenty of unanswered questions and mysteries, but that approach seems to be just multiplying the mysteries for the sake of it. I like my mysteries to seem natural and inevitable, not contrived - again an aesthetic judgement.

So for me, simplicity and clarity seem to form an important part of what influences the degree to which I find a belief 'credible'.

Statistical probability is a proportion of some reference class. So if it is statistically probable that I die before age 90 this is because the proportion of men (suitably similar to me) who have died before age 90 is quite high. Perhaps 7/10 men suitably like me die before age 90. That is what it means to say something is statistically likely. But what sense does it make to say that my belief that Plato taught Aristotle is statistically likely to be true? Statistically likely given what reference class? — PossibleAaran

You have identified two ways of probable belief. Statistical, which could be measured to some degree, and practical, which relies on facts -- were Plato and Aristotle both alive in the same era, for instance. If we are sincere about making and articulating a belief, we know that our beliefs must obey some form of reason. And note that I say some form of reason, because I don't necessarily mean epistemic. Your belief that Plato taught Aristotle must have been triggered by what you've read, heard, or learned from others. The same way your belief that you would die before the age of 90 was triggered by knowing some statistics on aging and men.