If I combine this:

In short, when a truth occurs, it occurs ontically—and that which ontically is is not subject to the possibility of being wrong, i.e. fallibility. But we can only appraise what ontically is epistemologically, which will always be to some extent fallible. — javra

And this:

When differentiating the ontological from the epistemological, ontically occurring truths (which are absolutely certain and not possible to be wrong) do occur all the time. But our epistemic appraisals of what are and are not ontic truths (the latter, again, do occur) will be fallible to some measure. — javra

Is there a contradiction?

Consider this proposition as if it were itself a truth:

<Ontological truths (which are absolutely certain and not possible to be wrong) do occur all the time.>

Is this "truth" an "ontological truth" or an "epistemological truth"? Because if it is an "epistemological truth," then it is not certain, and if it is an "ontological truth," then your appraisal is not fallible. This is why I'm not sure the way you are dividing up this territory is ultimately coherent. You are speaking as if your knowledge-claims about ontological truths are themselves ontological truths and not epistemological truths, and your theory seems to preclude this.

* The monkey wrench is logical and other putatively analytical truths. [...] There's something odd about asking whether "If A, then not (~A)" is a belief, or how we might justify it. But I'll leave that for others. — J

A prime example of this (and it does regard what can well be considered hinge propositions) are those who take dialetheism to be true. We thereby now have an inconsistency between the principle of noncontradiction being true and dialetheism being true. And this inconsistency as to which in fact conforms, or else corresponds, to the actual states of affairs can only be resolved via optimal justifications. Yes, maybe for now these are lacking, but, short of aggressions of each camp toward the other such that “might makes right”, what other avenue is available to us toward discerning what is true in respect to this aspect of ontology (what might possibly be termed the ontology of valid reasoning or of valid logic … or, maybe more esoterically, of logos)?

Is there a contradiction? — Leontiskos

Not as far as I know.

Consider this proposition as if it were itself a truth:

<Ontological truths (which are absolutely certain and not possible to be wrong) do occur all the time.>

Is this "truth" an "ontological truth" or an "epistemological truth"? Because if it is an "epistemological truth," then it is not certain, and if it is an "ontological truth," then your appraisal is not fallible. This is why I'm not sure the way you are dividing up this territory is ultimately coherent. — Leontiskos

To be clear, I'm not here writing a formal philosophical thesis but a forum post intended to address a specific issue. That mentioned:

The truth of the proposition here quoted would of course of itself be an epistemic truth. One which I so far find thoroughly justifiable: To keep things short, I so far find that there can be no epistemic truth in the absence of an ontically occuring truth it aspires to express. Can you, or anyone else, cogently justify the occurence of an epistemic truth that does not claim to be or else intend to conform to an ontic truth?

If not, then it remains cogently justifiable that ontically occuring truths do occur. Conversely, it then becomes unjustifiable that ontically occurring truths do not occur. (The "all the time" part I'll cut off for now, for it would require a great deal of further justification.)

The truth of the proposition here quoted would of course of itself be an epistemic truth. One which I so far find thoroughly justifiable: To keep things short, I so far find that there can be no epistemic truth in the absence of an ontically occuring truth it aspires to express. Can you, or anyone else, cogently justify the occurence of an epistemic truth that does not claim to be or else intend to conform to an ontic truth?

If not, then it remains cogently justifiable that ontically occuring truths do occur. Conversely, it then becomes unjustifiable that ontically occurring truths do not occur. — javra

That's a fair argument. It is similar to a comment asked me about, and which could perhaps be folded into this thread:

If there is no pole of knowledge then I don't see how one [inference to the best explanation] can be better than another (because no [inference to the best explanation] can better approach that pole). — Leontiskos

You seem to be saying that "epistemic truths" presuppose the existence of "ontological truths"; we all believe ourselves to be uttering "epistemic truths"; therefore we are all presupposing the existence of "ontological truths"; and because of this the belief in "ontological truths" is justified.

I think that's a good account on the "game of pool" approach, but I would prefer an account that provides for knowledge of at least some "ontological truths," rather than mere justified belief. Or in other words, if we take up your idea of fallibilism via Janus' conditional:

So, if we know p could be false, then we don't know that it's true, but we may well believe that it's true. — Janus

Then on the premise that we know that every p (epistemological truth) could be false, we cannot know any p.

has forwarded a theory where all (or almost all - this is contentious) beliefs are inferences to the best explanation, and are thus probabilistic.

On all of these conceptions certain knowledge is impossible, and yet knowledge is traditionally understood to be certain.


(It should again be noted that none of this has anything special to do with JTB. The one who thinks JTB does not understand JTB.)

You seem to be saying that "epistemic truths" presuppose the existence of "ontological truths"; we all believe ourselves to be uttering "epistemic truths"; therefore we are all presupposing the existence of "ontological truths"; and because of this the belief in "ontological truths" is justified.

I think that's a good account on the "game of pool" approach, but I would prefer an account that provides for knowledge of at least some "ontological truths," rather than mere justified belief. — Leontiskos

Here is a different approach to the same conclusion:

Can it be in any way validly justified that no ontologically occurring truths occur? If one believes that this is the case, what does one intend to express by the proposition of “no ontically occurring truths occur” if this proposition is not meant to conform/correspond to the actual states of affairs of the world and, thereby, of itself be an ontic truth? Thereby contradicting the very proposition made. Therefore, there is no justifiable alternative to the proposition that ontic truths occur.

As to providing knowledge of some "ontological truths", this, again, is what our ability to honestly and cogently justify offers us the possibility of. It just that our JTB knowledge will not, by a fallibilist account, be infallible. (Fallibiilty does not equate to being wrong.)

Then on the premise that we know that every p (epistemological truth) could be false, we cannot know any p. — Leontiskos

Remember that the JTB model of knowledge was presented by an Ancient Skeptic. If one presumes knowledge to be infallible, then this quote holds. If one presumes knowledge to be fallible, then it does not.

On all of these conceptions certain knowledge is impossible, and knowledge is traditionally understood to be certain. — Leontiskos

By everything I've so far stated, there then can occur ontically true beliefs which we can justify at will. These then will be instances of ontic knowledge, which is certain. Because we can only hold epistemic appraisals of what is ontically true, though, everything we uphold as knowledge will be epistemic knowledge, rather than ontic knowledge - which, as with epistemic truth, is less than "completely assured, fixed, and invariable."

I'll be back tomorrow.

↪Relativist
has forwarded a theory where all (or almost all - this is contentious) beliefs are inferences to the best explanation, and are thus probabilistic. — Leontiskos

That's correct, but I don't claim that most such inferences are rigorous. It's often just what seems most likely to the person making the judgement at the time he makes it.
Example of non-rigorous IBE: a Presidential candidate* loses the election; based on his belief that he was overwhelmingly more popular than his rival, and he judges that this is best explained as the election being stolen from him by illegal means. (* Any resemblance to any person, living or dead, is purely coincidental.)

More rigor enters into the analysis when one open-mindedly considers additional evidence that was previously unavailable or overlooked, often in the face of being challenged on the initial judgement.

__________________________
Regarding "probablistic" - do not mistake this with orthodox Bayesian epistemology, which depends on the absurd assumption that we can attach a consistent set of epistemic probabilities to every statement we claim to believe. Rather, I embrace Mark Kaplan's* "modest Bayesianism", which makes the modest claim that we can attach a relative confidence level to SOME pairs (or small sets) of statements of belief. If there's a reasonable basis for the ranking.

* source of this theory: Mark Kaplan's article "Decision Theory and Epistemology", in the Oxford Handbook of Epistemology.

That bird looks sad. Is he a cousin of the fly in the fly-bottle? (I also notice that he could leave the cage anytime he wanted to.)

Here is a different approach to the same conclusion:

Can it be in any way validly justified that no ontologically occurring truths occur? If one believes that this is the case, what does one intend to express by the proposition of “no ontically occurring truths occur” if this proposition is not meant to conform/correspond to the actual states of affairs of the world and, thereby, of itself be an ontic truth? Thereby contradicting the very proposition made. Therefore, there is no justifiable alternative to the proposition that ontic truths occur. — javra

That's also an intelligible argument, but I think it's weaker than the other one. This is because it seems to commit the error of applying the LEM to justification, so to speak. It seems to say, "If you don't think one is justified in affirming the existence of 'ontological truths', then you must have some justification for affirming their non-existence." I don't think this works because I think that someone could reject all particular justifications on offer in either direction, even if they do not affirm the truth that there can be no justification in either direction. One example of this approach would be the ancient skepticism that you mentioned, but in a more general sense I think an agnostic stance that does not affirm either of the two "ontological truths" is coherent.

As to providing knowledge of some "ontological truths", this, again, is what our ability to honestly and cogently justify offers us the possibility of. It just that our JTB knowledge will not, by a fallibilist account, be infallible. (Fallibiilty does not equate to being wrong.) — javra

But if "fallibility" means that we cannot be certain, then the same problem arises.

The words "infallible" and "fallible" are often used by "fallibilists" but never by "infallibilists," which makes me think they involve contentious presuppositions. I would say that what is at stake is the certainty of knowledge, not the infallibility of knowledge. Actually the object of in/fallibility is a faculty or power, not a piece of knowledge, which is another reason I don't find those terms helpful. It is understandable that one would use them, but given that no one accounts themselves an "infallibilist" the distinction's usefulness is questionable. More simply, I see no reason why someone who affirms the certainty of some knowledge must be a so-called "infallibilist."

Remember that the JTB model of knowledge was presented by an Ancient Skeptic. If one presumes knowledge to be infallible, then this quote holds. If one presumes knowledge to be fallible, then it does not. — javra

...and since no one presumes knowledge to be infallible, and yet pretty much everyone holds that knowledge is certain, @Janus' conditional must apply more generally than you allow.

By everything I've so far stated, there then can occur ontically true beliefs which we can justify at will. These then will be instances of ontic knowledge, which is certain. Because we can only hold epistemic appraisals of what is ontically true, though, everything we uphold as knowledge will be epistemic knowledge, rather than ontic knowledge - which, as with epistemic truth, is less than "completely assured, fixed, and invariable." — javra

Well you're walking a tightrope with these sentences. For one thing, I would want to ask what it means for "ontic knowledge" to "occur" or be "certain." And why is it "certain" that there are instances of ontic knowledge? Couldn't someone of your persuasion hold that there are no ontically true beliefs, even though every epistemological belief aims at ontic knowledge?

I would take the more traditional approach and agree with a great deal of what you have said, but add that sometimes "epistemological knowledge" and "ontic knowledge" coincide, and can be known to coincide.

I'll be back tomorrow. — javra

Okay, and if I don't manage to post tomorrow then I will be out for a few days.

Regarding "probablistic" - do not mistake this with orthodox Bayesian epistemology, which depends on the absurd assumption that we can attach a consistent set of epistemic probabilities to every statement we claim to believe. Rather, I embrace Mark Kaplan's* "modest Bayesianism", which makes the modest claim that we can attach a relative confidence level to SOME pairs (or small sets) of statements of belief. If there's a reasonable basis for the ranking. — Relativist

Okay good, and therefore let me try to answer one of your questions from a different thread here, in part because I will be out for a few days:

If there is no pole of knowledge then I don't see how one IBE can be better than another (because no IBE can better approach that pole).

Similarly, if we know what ice is then we have a pole and a limit for the coldness of water. If we don't know what ice is, then the coldness of water is purely relative, and there is nothing to measure against. I would argue that knowledge is prior to IBE, and that IBE is parasitic upon knowledge. Thus if you make IBEs the only option, then there is nothing on which an IBE can be parasitic upon or subordinate to, and this undermines IBEs themselves. — Leontiskos

Does your "tentpole" comment refer to the mere fact that knowledge exists, are you suggesting IBEs that aren't based on knowledge are all equivalent, or something else entirely? — Relativist

Not a tentpole, but a pole such as the North Pole. If some things are more North and some things are less North, then something must be most North.

So if knowledge is probabilistic, then it would seem to be asymptotic towards "100% probable." If someone doesn't know what it means to be 100% probable, then they cannot know what it means to be probabilistic at all. And if you admit IBEs or probabilistic knowledge without admitting traditional knowledge, then it looks like you have no pole to orient your IBE.

The phrase itself, "inference to the best explanation," presupposes the idea of an explanation, and an explanation is not merely probabilistic.

The general idea here—which will apply to a large number of the epistemological theories on offer—is that if we abandon the possibility or notion of certain knowledge, then the replacement form of uncertain knowledge will cease to make sense. If all knowledge is uncertain, then no knowledge is uncertain, so to speak. And if there is to be a spectrum of certainties, then one must account for what makes the more certain knowledge more certain and what makes the less certain knowledge less certain, and this accounting will itself reinstate the traditional view of knowledge that one was trying evade in the first place.

An attempt to analyse truth in terms of knowledge using a definition of knowledge in terms of truth with of course be circular.

The problem I think you see is of your own creation.