It does address them, directly. You can falsify a belief that not P, and so prove that P, sure, but only by observing something contrary to the observations implied by not P — that is, bu falsification. Merely observing something implied by P cannot prove that P. That is confirmation and is just affirming the consequent.

Likewise, you can falsify a belief that P and so prove that not P, by observing something contrary to the implications of P; but you can’t just observe something implied by not P and take that is proof that not P.

Sure you can prove anything by observing something contrary to the implications of its negation, whether that thing is it’s even a negation of something else or not; but you can’t prove anything just by observing something consistent with its implications.

So you admit that such people do exist. Why then were you pressing me for proof of them? — Pfhorrest

Because it's important that you specify. Imagine we we're arguing about the nature of 'true believers', back and forth for pages, only to find out that 'true believers' for you were the Catholics, whereas I presumed you mean the Protestants. The specific people whom you consider as being either fideistic or nihilistic matters. These people aren't one or the other solely on your say so, nor even on theirs. If there's a fact of the matter about what their thinking methods are then it is independent of either your beliefs or their beliefs about that matter.

It would not be very epistemologically sound, or discursively fair, to approach someone espousing something you think is false and reply to that only with an analysis of the conditions that have caused them to come to believe that, as if presuming that they are a crazy person who can't think rationally, just because they've reached a different conclusion than you. If what they believe is actually true, then you'd be dodging the issue they're trying to talk about entirely.

When people do irrationally believe falsehoods (or meaningless nonsense), it is good to figure out what's causing them to do that, but first we need to assess whether what they believe is false, and whether they believe it on rational grounds. To do that, we need to determine what the rational grounds for believing things are... and that brings us back to epistemology again.

(And if they are believing falsehoods on rational grounds, then doing philosophy with them, i.e. having a rational argument, is the most epistemologically sound and discursively fair way of changing their mind anyway. Only once that fails, and we conclude that they are not thinking reasonably, should be begin concerning ourselves with the irrational causes of their nominal belief). — Pfhorrest

All of this is begging the question. It assumes that there is some method of doing this which is the very matter about which there is disagreement. You'd have to first resolve Van Inwagen's argument from epistemic peers (broadly, if you don't already know it - if one of your epistemic peers disagrees with you about a matter, then that proves it is possible for someone with your knowledge and skills to be wrong despite thorough application to the issue - if that's the case then how will you ever know it's not you who are wrong?)

Unless you think fideism or nihilism will get us anywhere (which it seems you don't), then whatever other method could possibly get us anywhere will be some subset of my method, because it's just the negation of those two things. — Pfhorrest

That you think it's just the negation of those two things is one of the matter under contention. No-one is contesting that your position at least vaguely holds together as a system, you've built a perfectly adequate castle. We're contesting it's lack of foundations, not the integrity of it's later structure. Your claim that it is the best (or even the only) epistemological method for establishing which beliefs are justified is not supported by pointing to it's internal consistency. It's like trying to win an argument about which is the best car by pointing to the fact that yours has wheel firmly attached. Well, they all do, that's not what's at issue.

We're on the surface of an infinitely deep ocean, with the infinite sky above us. Therefore we cannot stand on the bottom, because there is no bottom. And we cannot grab for the sky, because the sky isn't some solid ceiling above us; nor can we just stick our arms up and hope our imaginary friend Superjesus will save us from drowning or anything like that. If we try to do either of those things, stand on the bottom or hang on to something above us, we will surely drown. Therefore we have to do something other than those things: neither try to stand on the bottom nor hang on to anything above us.

In other words, we have to swim. I'm not specifying how to swim, nor saying that the specifics of how to swim are unimportant. I'm just pointing out that there is no bottom to stand on and hanging from the sky isn't an option either, so we've got to do something else directly involving the water we're immediately surrounded by instead. — Pfhorrest

I can do no better than @Srap Tasmaner in responding to this, so I won't bother repeating it, but rather jump straight to the issue

No, I started off with arguments for why we must start in the middle. I’m not repeating those arguments in full in every thread, but exploring the implications of that conclusion on each sub-field of philosophy, thread by thread. — Pfhorrest

This essentially repeats the same foundationalist error you're trying to eliminate. Your previous arguments are not somehow 'foundational' to these, knowledge is not built up like an inverted pyramid, there are no 'implications of that conclusion' that are not also implications on that conclusion.

It's an interesting exercise nonetheless. Watching you oppose foundationalism by claiming you're building up an argument from foundational principles, watching you oppose dogmatic fideism by dogmatically sticking to you unaltered beliefs despite nearly all of your epistemic peers disagreeing with you...

Could you give a short presentation of what your criticism of Phfforest's position is. His position, not so much how he has presented it. I can't quite get what is going on in your dialogue though I get the feeling I would be interested. What is wrong with his version of critical liberal epistemology?

All of this is begging the question. It assumes that there is some method of doing this — Isaac

Doing what? Differentiating more correct beliefs from less correct beliefs?

I’ve already given an argument for why we could only ever assume one way or another about that, and why we pragmatically ought to assume there is, instead of just assuming there’s not.

Also, per critical rationalism generally, saying “you can’t prove that” is not an argument against anything ever. You need an actual disproof, not just pointing at how someone is assuming something, became we’re all always assuming tons of things and that’s not wrong, only persisting contra counter-evidence is.

You'd have to first resolve Van Inwagen's argument from epistemic peers (broadly, if you don't already know it - if one of your epistemic peers disagrees with you about a matter, then that proves it is possible for someone with your knowledge and skills to be wrong despite thorough application to the issue - if that's the case then how will you ever know it's not you who are wrong?) — Isaac

The resolution to this is built right in to the system I’m advocating. There are always a range of possibilities that someone with the same knowledge and skills can rationally disagree within, on my account, because nobody can ever pin down one exact comprehensive truth. Within that range you can’t know you’re not wrong; “you might always be wrong” is a core upshot of my view. You can only ever know what possibilities are for sure wrong, never which are for sure not wrong.

That you think it's just the negation of those two things is one of the matter under contention. — Isaac

It’s literally just defined as such. Anything at all that is neither fideistic nor nihilistic is okay on my account. I think you think I’m advocating something much narrower or more specific than I am.

This essentially repeats the same foundationalist error you're trying to eliminate. Your previous arguments are not somehow 'foundational' to these — Isaac

Foundationalism starts with basic beliefs that are taken to be self-evident or indubitable. I don’t do that. I start with reductio arguments against certain broad classes of view — fideism and nihilism — showing how assuming that those are true leads to problems, and then just taking whatever else is left over, which is a really broad class of things. Exploring the implications of that on other things just means seeing what possibilities on those other subjects fall outside those bounds, into the already ruled-out realms of fideism or nihilism, and what remains still in the acceptable domain between them.

sticking to you unaltered beliefs despite nearly all of your epistemic peers disagreeing with you... — Isaac

It really doesn’t seem so much like disagreement as it does a strangely aggressive agreement aggravated by some sort of miscommunication.

It’s like I’m saying something is a quadrilateral, and others are vehemently opposed to that because it’s obviously a trapezoid, or a parallelogram, or a rectangle, or a rhombus, or a square... and I’m saying yeah, it’s all of those things, all of those things are quadrilaterals and also it’s possible to be all of those kinds of quadrilateral at the same time. Why do you think we’re disagreeing? My only point is that it can’t have fewer than 4 sides, and it can’t have more than 4 sides, so it must be a quadrilateral.

I agree that only non-categorical statements can be verified by observation. That is because categorical statements are always over-reaching. To say that all swans are white is an over-reach insofar as we cannot ever be sure we have seen every swan. To verify that not all swans are white we have only to observe one swan that is not white.

When it comes to scientific theories, the situation is not really like this at all; it is never so clear cut. When the logical positivists speak about verification, they were not stupid enough to be imaging that observing the predictions of an hypothesis logically prove it to be correct. They saw verification as provisional. Likewise, when predictions fail to be observed this does not logically disprove an hypothesis; there may always be other unknown factors in play. Fasification in science is as provisional as verification is. When predicted results are observed, all that is verified is that we might be onto something; when predicted results are not observed all that is verified is that we might not be onto something.

Foundationalism starts with basic beliefs that are taken to be self-evident or indubitable. I don’t do that. I start with reductio arguments against certain broad classes of view — fideism and nihilism — showing how assuming that those are true leads to problems, and then just taking whatever else is left over, which is a really broad class of things. Exploring the implications of that on other things just means seeing what possibilities on those other subjects fall outside those bounds, into the already ruled-out realms of fideism or nihilism, and what remains still in the acceptable domain between them. — Pfhorrest

We all start with basic assumptions taken to be self-evident. Your argument against fideism and nihilism; your assumption that they must lead to problems, is just such an assumption. If, as I suspect, your problem with those is that they either believe without evidence (fideism) or deny all evidence (nihilism) then you are assuming the provenance of inductive thinking; which is fine since all scientists do that. All scientists conduct their enquiries as if the laws of nature have been proven; the basic assumption of science is that nature is law-like and that we have discovered some of her laws and that our scientific knowledge is a great coherent system based on those natural laws.

I agree that only non-categorical statements can be verified by observation — Janus

I’m not talking about categorical or not, nor about particular atomic statement at all. I’m talking about how experiences in agreement with your (entire set of) beliefs don’t give justification to your beliefs over others, they don’t tell you anything about the relative merits of your beliefs compared to the alternatives; unlike how experiences counter to your beliefs give you justification for discarding your beliefs in favor of others.

But now that you mention non-categorical beliefs, there is an asymmetry there: you can never finish falsifying (at least empirically) a non-categorical belief, because the negation of “there are some
black swans” is “there are no black swans” and you can never be sure that you just haven’t seen whatever black swans there are YET.

They saw verification as provisional. — Janus

But there is no rational reason to think that the truth of the consequent of an implication gives even weak support to the antecedent. It’s not just less than certainly, it’s nothing at all.

Likewise, when predictions fail to be observed this does not logically disprove an hypothesis; there may always be other unknown factors in play. — Janus

This is only if you’re talking about one atomic statement, which we already went over extensively in this thread is not what we’re talking above. Falsification is falsification of the entire set of beliefs, and in that sense it is certain disproof, because you still have to change something about your beliefs or another.

When predicted results are observed, all that is verified is that we might be onto something — Janus

But we already might have been onto something, so nothing new is learned.

when predicted results are not observed all that is verified is that we might not be onto something. — Janus

No, rather, they we are definitely off track, somewhere or another, even if it's not perfectly clear where.

We all start with basic assumptions taken to be self-evident — Janus

That's just asserting foundationalism, and so begging the question.

Your argument against fideism and nihilism; your assumption that they must lead to problems, is just such an assumption — Janus

It is not an assumption, its an inference. You're familiar with reductio ad absurdum, no? You start with the hypothetical assumption of some premises, derive an absurdity from them, and conclude the actual rejection of those premises. That's what I do to both fideism and nihilism, and then proceed from their negations.

If, as I suspect, your problem with those is that they either believe without evidence (fideism) — Janus

That's not fideism, that's merely what I call "liberalism". Fideism is believing against evidence, flatly rejecting the possibility that some particular part of your beliefs could be wrong, and so rearranging the rest of your beliefs however possible to excuse away evidence against those protected parts.

or deny all evidence (nihilism) — Janus

That's not nihilism. I'm not sure exactly what you mean there. The counterpart to fideism in epistemology is what I call "cynicism", which I reject because it leads to nihilism (which is an ontological position, with a counterpart I call "transcendentalism", which in turn leads to fideism), which is basically an overzealous skepticism, saying "prove it or lose it" to every belief, including those offered in proof of those ones, and those offered in proof of those in turn, ad infinitum, resulting in the demand that you reject all beliefs, forever, i.e. nihilism.

Could you give a short presentation of what your criticism of Phfforest's position is. His position, not so much how he has presented it. I can't quite get what is going on in your dialogue though I get the feeling I would be interested. What is wrong with his version of critical liberal epistemology? — Coben

I'll do my best.

1. Beliefs are not propositions. Beliefs are states of mind equivalent to a tendency to act as if... As such it is a) not possible to have a belief which is contrary to the evidence of your senses (beliefs are formed by a neurological process of response to stimuli), and b) people's stated propositions are not necessarily reflective of their beliefs and it is a category error to develop an understanding of one based on experience of the other (just because people say their 'belief' is based on foundations, doesn't mean it is; just because people say they doubt everything, doesn't mean they do)

-- this leads to the more general criticism that there is no target of the normative claim, it's like telling people that they ought to breathe.

2. If you look at the graph where 'critical liberalism' is defined (on the other thread) you see it is based on avoiding extremes of two axes. One is 'willingness to change one's belief in the light of evidence to the contrary', the other is 'extent to which beliefs are accepted/justified without foundation or evidence proving their necessity'. Going not far enough in the first is 'fideism', going too far in the second is 'nihilism'. But both of these scales contain subjective judgements and are superlative. As such they are useless normatively, which is the intended realm of the original proposition. Given my definition in (1), above, I contend that no-one would hold their beliefs were impossible to change even in the light of overwhelming evidence to the contrary, and that no-one would hold themselves to have no beliefs at all because they can't be proven. Rather what we find are degrees of faith and degrees of doubt about some belief(s). The former is that the belief should be maintained so long as it is even remotely possible to do so. The latter is the complete absence of any preference. The reasons for that choice, the extent to which they're maintained and the methods by which they are, are all subjective.

-- this leads to the general criticism that there is no conclusion to the normative claim because all the important elements required to use it are missing from the claim. Like telling people they ought to consult a certain etiquette pamphlet at all times but neglecting to give them the pamphlet.

The argument against 1 seems to be that people do claim beliefs to be immutable, such as Reformed epistemology, but I find this uncompelling because in such systems reasons are given for why the belief should be held.

The argument against 2 is that this is just a preliminary stage and that objective measures of methodological questions will follow. I find this argument uncompelling (notwithstanding the fact that I anticipate such methods will prove just as subjective) because I find it to be foundationalist. It appears to cement each 'foundation' and then move on assuming the only direction of play is from these foundations forward to their consequences, whereas properly their consequeses should no less be considered reasons to reject/alter the prior conclusions.

I hope that's served to clarify things.

I’ve already given an argument for why we could only ever assume one way or another about that, and why we pragmatically ought to assume there is, instead of just assuming there’s not. — Pfhorrest

More foundationalism. If your conclusion is in any way problematic, that is just as much cause to question your premises as your premises are cause to reach your conclusion. You can't keep referring back to things you've 'shown' before as if those matters were unaffected by the issues here whilst simultaneously maintaining an opposition to foundationalism (or at least maintaining an understanding of Quine).

You can only ever know what possibilities are for sure wrong, never which are for sure not wrong. — Pfhorrest

Right. So how do resolve Van Inwagen's position about possibilities which are 'for sure wrong'?

It’s literally just defined as such. Anything at all that is neither fideistic nor nihilistic is okay on my account. I think you think I’m advocating something much narrower or more specific than I am. — Pfhorrest

Yes, but your definitions are subjective (see my reply to Coben if you want a summary of why), so this amounts to nothing more than "anything which I find to be the right balance is OK", which is virtually tautologous.

Foundationalism starts with basic beliefs that are taken to be self-evident or indubitable. I don’t do that. I start with reductio arguments against certain broad classes of view — fideism and nihilism — showing how assuming that those are true leads to problems — Pfhorrest

Are you suggesting that 'absurd' is some kind of objective measurement? Otherwise how is your belief that your reductio arguments show what you claim they show not then "basic beliefs that are taken to be self-evident or indubitable"? You literally claim (by introducing a reductio) that it is self evident that fideism and nihilism lead to absurd or repugnant consequences. Furthermore, by limiting discussion to the consequences of this conclusion, you're holding those beliefs to be immutable.

But there is no rational reason to think that the truth of the consequent of an implication gives even weak support to the antecedent. It’s not just less than certainly, it’s nothing at all. — Pfhorrest

This ought to be a clue that you've chosen the wrong way of formalizing the process, because confirmatory evidence just obviously does matter. If you've seen thousands of swans in your lifetime and they were all white, there's nothing at all irrational about believing that swans are probably all white, or believing defeasibly that they are all white.

Given your general approach, I'm just not at all clear why you're so attached to this mid-century Quine-Popper thing instead of going in for something more like formal epistemology. Have you considered following the Quantitative Way? (LessWrong, SlateStarCodex, Overcoming Bias, et al.) I have some reservations, but it's a much more defensible model of rationality than yours, and it seems like it would be right up your alley.

They saw verification as provisional. — Janus


But there is no rational reason to think that the truth of the consequent of an implication gives even weak support to the antecedent. It’s not just less than certainly, it’s nothing at all. — Pfhorrest

You seem to be doubling down, so I'm only going to address this one point which is really the crux of where I think you are going astray. In empirical matters confirmation is not an abstract logical entailment, but we are induced to think that belief in the invariances that we unfailingly observe is justified by lack of any observed counterexamples. To labour the point: it does not follow that those invariances are logically proven; science and everyday inference is not strictly logical like that.

Serendipitously, this morning I was reading a passage in Carl Sagan's The Demon Haunted World speaking about the popular belief that arose in the eighties or nineties that there is a giant sculpted human face on Mars:

"Even if those claims are extremely improbable-as I think they are- they are worth examining. Unlike the UFO phenomena, we have here the opportunity for a definitive experiment. This kind of hypothesis is falsifiable, a property that brings it well into the scientific arena. [.......] Even if it becomes plain to everyone that these Martian features are geological and not artificial, monumental faces in space (and allied wonders) will not go away."

Sagan writes "faslifiable", but then what he writes in the next quoted passage amounts to saying that the falsification of the belief that the features are artificial just is the confirmation that they are geological. How would we know that they are geological? On the basis of past experience and the knowledge accumulated therefrom, of course; in other words from inductive investigation and analysis. This is the way (or one of the main ways) that science works, and no amount of armchair philosophizing will change that.

Beliefs are not propositions. Beliefs are states of mind equivalent to a tendency to act as if... — Isaac

This is a purely semantic disagreement that has no bearing on the substance of anything.

(I'm curious, since you're a working psychologist, do you differentiate between sensations, perceptions, and beliefs? I learned of the sensation-perception distinction that I employ in my philosophy from psychology classes, and it sounds like the sense of "belief" you're using is more or less what I would term a "perception" instead, so I wonder if and how you differentiate "belief" from "perception", presuming you also differentiate "perception" from "sensation" as my old psych textbooks said to do).

it's like telling people that they ought to breathe. — Isaac

It's like arguing against arguments that you ought not breath, or claims by people that they don't breath, on the grounds of the clear trouble you'd get into if you didn't, and that no matter how much you may try not to, you're going to end up doing it anyway, or else dying, so stop saying that you don't or telling other people not to. Instead, embrace the fact that you can't help but breath (or else die), and focus on doing it as well as possible.

both of these scales contain subjective judgements — Isaac

Repeatedly asserting that doesn't make it so.

I find it to be foundationalist. It appears to cement each 'foundation' and then move on — Isaac

Foundationalism isn't just any deduction from prior premises, it's only holding some premises to be immune to questioning. But I start off with questioning certain possible premises, finding them to inevitably lead to problems, and then further exploring the area of possibilities remaining once those are excluded. Possibilities that fall into the area already excluded are not live possibilities anymore, and the simple fact that that area has already been excluded doesn't mean that the possibility of them is not open to question.

(ETA: I just wrote something relevant to this in a different thread:

...end ETA.)

properly their consequeses should no less be considered reasons to reject/alter the prior conclusions — Isaac

Later consequences can be considered reasons to modify prior assumptions made within the realm of remaining possibilities, but they cannot be reasons to say that previous reasons to exclude certain possibilities are not good reasons anymore. The old reasons that lead to the exclusion of those possibilities and the new reasons from the new problems found have to be considered in tandem to narrow down the range of remaining possibilities; neither old nor new reasons can justify breaking back out into a range previously excluded by the other.

Right. So how do resolve Van Inwagen's position about possibilities which are 'for sure wrong'? — Isaac

You'll have to clarify this, because I think I've already answered this question, and if you don't think I have I don't know what you're still asking.

Are you suggesting that 'absurd' is some kind of objective measurement? — Isaac

In the general use of the word, no, but in the technical sense used in a reductio argument, yes.

confirmatory evidence just obviously does matter — Srap Tasmaner

"I'm just obviously right" is not an argument. I don't doubt that many people act like it matters. But there's good reason to think that it doesn't, once you actually think about why it would or wouldn't.

Have you considered following the Quantitative Way? (LessWrong, SlateStarCodex, Overcoming Bias, et al.) — Srap Tasmaner

From my exposure to them, they're all about Bayes Theorem, which I mention in the OP as being continuous with my own approach:

But this does not imply that all beliefs not yet shown false are equal. Beliefs not yet shown false can still be more or less probable than others, as calculated by methods such as Bayes' theorem. Falsification itself can be considered just an extreme case of showing a belief to have zero probability: if you are frequently observing phenomena that your belief says should be improbable, then that suggests your belief is epistemically improbable (i.e. likely false), and if you ever observe something that your belief says should be impossible, then your belief is epistemically impossible (i.e. certainly false). — Pfhorrest

It kinda seems like a lot of the nominal disagreement with me maybe stems from missing things like this that were right there in the OP, to instead attack some preexisting straw concept of falsificationism that I don't adhere to.

we are induced to think that belief in the invariances that we unfailingly observe is justified by lack of any observed counterexamples — Janus

That sounds like you're saying we in fact tend to think that way, which I'm not disputing. I'm disputing that there is any reason why we ought to think that way rather than otherwise; anything that says that way is the right way to think.

Also, again, it doesn't settle anything between people who both think that way but come to different conclusions that way. You see one pattern, someone else sees a different pattern in the same data, and then you see something new that fits your pattern... but if it also fits their pattern, we've learned nothing new. You need to see something that doesn't fit at least one of the patterns to judge between them.

the falsification of the belief that the features are artificial just is the confirmation that they are geological — Janus

Since he seems to be using "geological" just to mean "not artificial", then sure. But that conclusion was not reached by the confirmationist process: it was not "if these were geological we would see X, we see X, therefore they are geological", which would be wholly fallacious. It was "if these were non-geological (artificial), we would see Y, we don't see Y but instead X, therefore these are not non-geological, or in other words they are geological (non-artificial)". It's the difference between those two processes that's the point here.

On the basis of past experience and the knowledge accumulated therefrom, of course; in other words from inductive investigation and analysis. This is the way (or one of the main ways) that science works, and no amount of armchair philosophizing will change that. — Janus

Nobody here is disputing that past experiences matter or that we accumulate knowledge from them, or advocating "armchair philosophizing" in place of empiricism. The issue at question is the process by which experiences are applied to our beliefs to develop knowledge. My position is that experiences that agree with your beliefs do not elevate them over any alternatives, unless those experiences are also counter to the alternatives, because it's only experiences that go counter to some beliefs that elevate the alternatives over them.

Also, again, it doesn't settle anything between people who both think that way but come to different conclusions that way. — Pfhorrest

The vast bulk of science based on thinking that way is (for all intents and purposes, although not absolutely of course) settled though, so I can't see the point of your objection.

Since he seems to be using "geological" just to mean "not artificial", then sure. But that conclusion was not reached by the confirmationist process: it was not "if these were geological we would see X, we see X, therefore they are geological", which would be wholly fallacious. It was "if these were non-geological (artificial), we would see Y, we don't see Y but instead X, therefore there are not non-geological, or in other words they are geological (not-artificial)". It's the difference between those two processes that's the point here. — Pfhorrest

There is no difference between relying on the observation of features to determine that something is geological or that it is artificial, or obversely, to determine that it is not geological or not artificial. The whole process relies on our ability to distinguish between geological and artificial structures. Such a determination is not absolutely certain in the deductive sense; it is always logically possible, however unlikely it might be, that we could be wrong.

Geological means not merely "not artificial". The point to what Sagan is saying is that we are able to distinguish between something that has evolved geologically and something which has been created deliberately. In the inductive terms of science "if these were geological we would see X, we see X, therefore they are geological" is not fallacious at all (although note that "X" is not always or even often some single feature, but a suite of features).

It would only be fallacious if you thought it to be deductively certain, but I doubt many intelligent scientists would think that. You seem to keep falling back into the same conflation between deduction and induction. Based on that conflation you might say we have no justification to think that way, but the point is that scientists routinely do think that way, and the justification is that it works, has worked, to produce the comprehensive and (mostly) coherent body of knowledge we call 'science'. I don't know how many times I (and others) will have to try to make this clear before you finally get it.

The vast bulk of science based on thinking that way is (for all intents and purposes, although not absolutely of course) settled though, so I can't see the point of your objection. — Janus

It has been settled by falsificationist methods: the alternatives have been shown false or at least much less likely or less parsimonious. Things are settled by showing the faults of their alternatives. That is “the scientific method” inasmuch as there is such a thing. You keep claiming otherwise and I’m hesitating to engage with that because it’s beside the philosophical point what people do actually do — they could still be wrong even if it’s the popular way — but in matter of fact science is not done confirmationally and hasn’t been for a long time, since the problems with that were first pointed out.

In the inductive terms of science "if these were geological we would see X, we see X, therefore they are geological" is not fallacious at all — Janus

If they were geological we would see X.

If they were artificial we would see X.

We see X.

Therefore... nothing. We’ve learned nothing.


To learn anything it needs to be:

If they were geological we would see X.

If they were artificial we would see not-X.

We see X.

Therefore they are not artificial.

Therefore if the only alternative to artificial is geological (which you’ve just denied) then we can conclude they are geological;

Else, we only know they’re not artificial somehow or another, not necessarily geological.

the point is that scientists routinely do think that way, and the justification is that it works, has worked, to produce the comprehensive and (mostly) coherent body of knowledge we call 'science'. I don't know how many times I (and others) will have to try to make this clear before you finally get it. — Janus

Nobody here is denying science. I am (and many others, in actual publications, not here on this forum, are) denying that science works the way you say it works. I don’t know how many time I will have to try to make this clear before you finally get it.

If they were geological we would see X.

If they were artificial we would also see X.

We see X.

Therefore... nothing. We’ve learned nothing.


To learn anything it needs to be:

If they were geological we would see X.

If they were artificial we would see not-X.

We see X.

Therefore they are not artificial.

Therefore if the only alternative to artificial is geological (which you’ve just denied) then we can conclude they are geological;

Else, we only know they’re not artificial somehow or another, not necessarily geological. — Pfhorrest

Firstly I did not say that we would see the same X in case they were geological and in case they were artificial, so your first part here is irrelevant.

Your second part is somewhat badly worded. It should be 'if they were artificial we would see X, and if they were geological we would not see X'. If they were artificial we would see tool or machine marks, if they were geological we would not see tool or machine marks. I haven't said that there are other alternatives in this case, so I think you need to read more carefully. The point is that tool marks confirm artificial and absence of tools marks confirms geological, leaving aside any other unimagined possibilities (are there any other empirical possibilities you can imagine?). Obviously there would be many other criteria to confirm geological origin, since geology is a well-developed science with a good body of experientially derived and inductively justified knowledge.

Nobody here is denying science. I am (and many others, in actual publications, not here on this forum, are) denying that science works the way you say it works. I don’t know how many time I will have to try to make this clear before you finally get it. — Pfhorrest

I have not been talking about "the way science works" I have said that scientists generally think inductively, and that this way of thinking and the hypotheses it generates have worked to develop the body of knowledge we call science. For example, in relation to what I said above about geology being a well-developed science with a good body of experientially derived and inductively justified knowledge,do you not see that geologists' understanding of, and faith in the geological processes that they identify via associated them with observed geological structures relies upon the wholly inductive assumption that the laws of nature don't change?

Anyway it has become obvious to me that you are heavily invested in your own ideas, regardless of the fact that I and others have shown them to be either trivially true (in the deductive context) or mistakenly applied (in the inductive context) so if you don't produce any new arguments I am going to leave you to it.

Firstly I did not say that we would see the same X in case they were geological and in case they were artificial, so your first part here is irrelevant. — Janus

I'm not saying you said that, I'm pointing that out as the problematic scenario that demonstrates why confirmationism doesn't work. If the predictions do not falsify one of the possibilities, then seeing those predictions pan out tells us nothing, it doesn't distinguish between alternative possibilities. It doesn't matter if you see what your theory predicted, unless other theories predicted otherwise; it's the falsification of them that tells us something. Seeing something your theory predicted can't distinguish between that theory and any other theory that would also predict the same thing, and so tells us nothing.

I haven't said that there are other alternatives in this case — Janus

You said 'Geological means not merely "not artificial".' That implies that you think there could be something not geological, without being artificial; or something not artificial, without being geological; i.e. there's (at least) a third option.

I have said that scientists generally think inductively, and that this way of thinking and the hypotheses it generates have worked to develop the body of knowledge we call science — Janus

And I've said that induction is perfectly fine as a way of generating hypotheses, but it doesn't help us pick between competing hypotheses. The latter is where science differs from guessing and intuition.

Anyway it has become obvious to me that you are heavily invested in your own ideas, regardless of the fact that I and others have shown them to be either trivially true (in the deductive context) or mistakenly applied (in the inductive context) so if you don't produce any new arguments I am going to leave you to it. — Janus

Fine with me, I'm tired of repeatedly trying to get through to people who are saying things I already agree with as though they contradict me and then ignoring the actual arguments to the contrary of their other assumptions. I'm eager to let this thread die and move on to something different.

Seeing something your theory predicted can't distinguish between that theory and any other theory that would also predict the same thing, and so tells us nothing. — Pfhorrest

I agree with that, and that's perfectly fine in principle, but how rare are such cases; where two scientific theories predict exactly the same things? Can you think of a single example

You said 'Geological means not merely "not artificial". — Pfhorrest

All I meant by that is that the science of geology is not generally concerned with artificiality, and that the understanding of geological processes is not generally measured against the understanding of artificial processes. It is true though, that there is a general distinction between man-made and naturally evolved phenomena in science. No one is much interested toady in the the other possibility; that God did it.

And I've said that induction is perfectly fine as a way of generating hypotheses, but it doesn't help us pick between competing hypotheses. The latter is where science differs from guessing and intuition. — Pfhorrest

And I've said it is the comprehensive and cohesive knowledge that is based on inductive thinking, assumptions, investigations and analyses that enables a choice between competing hypotheses. You haven't produced any plausible alternative to that.

agree with that, and that's perfectly fine in principle, but how rare are such cases; where two scientific theories predict exactly the same things? Can you think of a single example — Janus

They never predict all of the exact same things (else they would be exactly equivalent, different formulations of the same thing), but there is usually significant overlap. GR, QM, and Newtonian physics all agree about their predictions on the medium scale we humans are accustomed to. So pointing at a ballistic missile flying as Newton would predict isn't evidence in favor of Newton vs GR, because GR also makes that same prediction. To decide between them, you have to pick a prediction that they disagree on; and then you've ruled out whichever one loses, but not supported the remaining one in any way against any other theories that also make that same winning prediction.

And I've said it is the comprehensive and cohesive knowledge that is based on inductive thinking, assumptions, investigations and analyses that enables a choice between competing hypotheses — Janus

It's the investigations and analyses that do the heavy lifting there. Inductive thinking and assumptions give you your competing hypotheses. Analysis of those gives you the expected observations. Investigation, i.e. empirical observation, compared against those expectations tells us which hypotheses we can keep and which we have to throw away. But it's the "throwing away" part that makes progress: we can no better tell between any of the "keepers" based on investigations that let us keep them, all we can tell is whether they're okay to keep or whether they must be thrown away.

I'll do my best. — Isaac

I appreciate the effort.

1. Beliefs are not propositions. Beliefs are states of mind equivalent to a tendency to act as if... — Isaac

Would this mean then that animals have beliefs?

a) not possible to have a belief which is contrary to the evidence of your senses (beliefs are formed by a neurological process of response to stimuli), and — Isaac

Does this mean that one cannot come to believe things that are counterintuitive: relativity, for example, or that the earth actually revolves around the sun. If we take the latter case that we can find empirical evidence that this is the case, very few people actually do that. Or that color exist outside us.

people's stated propositions are not necessarily reflective of their beliefs and it is a category error to develop an understanding of one based on experience of the other (just because people say their 'belief' is based on foundations, doesn't mean it is; just because people say they doubt everything, doesn't mean they do) — Isaac

I agree with this. I do think that people can be mistaken about their beliefs. though I think that their other beliefs are propositional, just dissonent with what they want to belief or they have contradictory beliefs (just as one can have contradictory tendencies to act as if.

-- this leads to the more general criticism that there is no target of the normative claim, it's like telling people that they ought to breathe. — Isaac

What was his normative claim?

Given my definition in (1), above, I contend that no-one would hold their beliefs were impossible to change even in the light of overwhelming evidence to the contrary, — Isaac

If you belief in the Christian model of faith, you might well do that. At least one is encouraged to by some versions of that faith. Though one might be better off putting on that end of the spectrum 'beliefs that are not supported at and do not seem to fit current models in science, say, or perhaps in general' It could be without the latter part of that. IOW one could try to believe only those things that you can demonstrate or have been demonstrated by experts to be justified OR you could accept things without justification (at least conscious justification one has access to) and ignore counterarguments. i would say most people do this about something.

I'll start there.

No one would dispute that if theory A predicts X and theory B predicts ~X, and then we observe X, that we can claim progress by eliminating theory B. In practice, it's messier than that, in principle falsifying some theories is progress.

But what about theory A? Part of Popper's program, as I understand it, was focused not on direct falsification as the way to choose between hypotheses, but on distinguishing science from non-science by requiring falsifiability of theories. Theory A has been submitted to being falsified and it wasn't. I think most scientists would claim that their confidence in A was increased by this result. For instance, when GR predicted that light waves would bend passing near the sun, but Newtonian mechanics did not, and then we observed that, you want to say we can now discard Newton, but our confidence in Einstein should not increase.

But there are two points about this observation that I think merit attention: (1) it was a prediction, not a retrodiction; (2) what was predicted was unlikely to us, a surprise. Philosophers have definitions of "surprise", but in this case we could just say, observation for which no explanation is ready to hand. Einstein of course offers both prediction and explanation, a prediction no competing theory was offering, and reasons for that prediction. I think that's why everyone's confidence in GR was bolstered by the result. If we had made the observation by chance, we would be baffled.

Of course, our confidence increases but not to the point that we think GR is the absolute truth; we only conclude that it's closer to the truth than other theories, in the specific sense that new theories will have to retrodict what GR predicted, so we know future theories have to look a little like GR.

My question then is this: why shouldn't we count as progress a theory submitting itself to being falsified and passing? We already have the theory, so we don't learn anything new, agreed; but we do learn that the theory isn't crazy, and that it might, like QM, like evolution, survive many many more rounds of potentially being falsified and not be. Also: each successful non-falsification adds another requirement to future theories, another datum they must retrodict, so there is a similar pruning function, only it's for possible rather than actual theories. Surely that ought to count for something.

I agree that finding a theory to continue passing potential falsifications is a surprising thing — like “wow this is working out surprisingly well, good guess Einstein!” — and the more observations we have made, even if they haven’t falsified the actual theories we have, the more progress we’ve made, on account of the ranges of potential theories we’ve eliminated, the restrictions on remaining theories we’ve placed.

It’s something like the relativity of wrong, if you’re familiarity with that. Newtonian physics is wrong, but it’s not AS wrong as Aristotelian physics. I picture concentric rings, or like topographic lines, centered on whatever the complete truth is, where we draw a new innermost ring with every observation, and the further out in the rings a theory is, the “more wrong” it is. As we keep drawing more and more rings, and a theory continues to still be within the innermost ring, it is “elevated” RELATIVE to the others in the outer rings, even those we never enumerated in the now-outer rings.

But the thrust of falsificationism in this metaphor is that that relative elevation is in fact due to each new ring lowering the possibilities outside of it: the things still remaining within the inside ring merely retain their initial elevation. But that doesn’t mean no progress has been made, because the progress is in the drawing of the rings (and so weeding out the possibilities outside of them), not in actually elevating the things in the middle.

good guess Einstein! — Pfhorrest

You're being cute -- "guess" is wildly inappropriate and if your approach steadfastly refuses to see the difference between a guess and a real theory, what's the point of any of this?

Every observation we make will constrain our future theorizing, whether the observation was predicted or not, whether it falsified any existing theory or not. If progress is only a matter of constraining the range of theories that might be true, all we really need are the observations and no theories at all. But no one does research that way and it is doubtful anyone could. Why is that? One reason to bother with theory is to know what kinds of observations there are, which should count as the same kind of thing -- so not really adding constraints, or not much -- and which are genuinely different, and especially which would be surprising.

One reason to bother with theory is to know what kinds of observations there are, which should count as the same kind of thing -- so not really adding constraints, or not much -- and which are genuinely different, and especially which would be surprising. — Srap Tasmaner

Sure, I agree with that, and I’m not at all advocating that we go without theory in any way.

So assigning a theory the status of "falsified" or "not-yet-falsified" is not the only way to make progress.

That theory is useful in guiding what observations to make doesn't change that the usefulness of the observations in turn is in differentiating which possibilities they rule out and which they don't. Theory guides observation by giving us a target of something to try to rule out.

So to you the only value of GR was in making a prediction that, if observed, would allow us to rule out Newtonian mechanics, and that observation did nothing to confirm GR, or nothing we should care about, nothing we should allow to increase our confidence in GR. In particular, the explanatory framework that comes along with GR, responsible for the prediction, that's of no interest.

No, the value of a theory is mostly in its explanatory framework, the way it allows us to abstract patterns in many many observations into simple easy-to-use rules. Like I said earlier, we could (in principle, but not in practice) just keep an unorganized list of all our observations as our "theory" and never worry about it ever being falsified, but that wouldn't be useful. Theories give us something much more useful to work with than that, but at increased risk of epistemic error.

There's two things that seem to be getting conflated here: one of them is what makes a theory useful, and another is what makes observations useful. Theories are useful for the reasons stated above. Observations are useful for telling us which theories are wrong and which are maybe not wrong. In adopting theories instead of just keeping unorganized lists of observations, we save effort, but at the cost of sticking our necks out and make assumptions without adequate justification, which could therefore be wrong. In making observations targeted at checking where they might be wrong, we make sure that our theories are not wrong in at least that respect, and so are safe to use at least in that domain.

To decide between them, you have to pick a prediction that they disagree on; and then you've ruled out whichever one loses, but not supported the remaining one in any way against any other theories that also make that same winning prediction. — Pfhorrest

The above referring to NM, GR and QM: You don't need to decide between them. They work in different contexts. NM as methodology is not falsified by GR. The greater accuracy to finer scales of GR might lead you have less confidence in the metaphysics associated with NM, but that would only be because we now have greater confidence in the GR metaphysic.

It's the investigations and analyses that do the heavy lifting there. Inductive thinking and assumptions give you your competing hypotheses. Analysis of those gives you the expected observations. Investigation, i.e. empirical observation, compared against those expectations tells us which hypotheses we can keep and which we have to throw away. But it's the "throwing away" part that makes progress: we can no better tell between any of the "keepers" based on investigations that let us keep them, all we can tell is whether they're okay to keep or whether they must be thrown away. — Pfhorrest

Hypotheses and theories are neglected, when other ones that prove to be more accurate and thus give us more reason to adopt them are found. They are not "thrown away". NM is still perfectly workable for many precise operations. It is simply not as precise as GR. Here's a challenging question for your position: what exactly do you think it was about NM that was falsified by GR?

Also something you said I meant to address earlier was that faith is belief in the face of contrary evidence. I think this is completely wrong; I think people cannot believe something contrary to evidence that they accept as such. One person's evidence, except in the most mundane empirical matters, is not another's.

Take, for example, faith in Christianity, in God; there is no evidence that God doesn't exist, and that Christ is not God incarnate, and that his message is not a revelation. There is no evidence that God does exist, that Christ is God incarnate and that his message is a revelation, either. So, faith is belief in the absence of evidence, not belief against evidence.

When it comes to evolution, the fossil record, and the inductively derived theories concerning genetic mutation and natural selection, give us positive reason to believe that species have evolved, and that confidence gives us reason to think that the world could not have been created 6,000 years ago. On that basis creationism is ruled out, and most intelligent believers don't reject the evidence for evoiution that the fossil record, genetic theory and the theory of natural selection represent. Those that do continue to believe in creationism simply reject the evidence for evolution, which allows them to continue to believe in creationism.

The problem with your view is that on the basis of its logic there can be no evidence for anything, only evidence against. What you fail to see is that if there cannot be evidence for anything, then there cannot be evidence against anything either; they are two sides of the one inductive coin.

Here's a challenging question for your position: what exactly do you think it was about NM that was falsified by GR? — Janus

The whole picture of NM. Which is not to say every piece of the picture, but the conjunction of all of the pieces together. (The negation of a conjunction is not the negation of every conjunct, only of some of the conjuncts).

That NM is still good enough for some purposes is besides the point. We know for sure that NM is not completely correct.

(We also already know for sure that GR and QM are not completely correct, but they are still less wrong than NM, and we don't yet know what in turn is less wrong than them).

I think people cannot believe something contrary to evidence that they accept as such. — Janus

The bold part is important. Finding reasons to reject that some observation is evidence to the contrary of a belief is exactly the behavior that someone clinging to that belief in the face of evidence would do. (Yes, it's also something that's sometimes done by people who aren't doing that, and we can't know for certain based just on that behavior whether they are doing it or not; only the person doing it themselves can know, if even they do. As already gone over extensively with Isaac in this thread).

there is no evidence that God doesn't exist — Janus

Depending on what you mean by "God", there is, or sometimes it's not the kind of thing held susceptible to evidence at all, and so must be rejected on those grounds.

faith is belief in the absence of evidence, not belief against evidence — Janus

This is a semantics game again. I specify two different terms, "liberalism" and "fideism", exactly to avoid playing this semantics game. I'm not against believing in the absence of evidence, I am against believing against evidence. Call those whatever you like, names don't matter except for convenience.

The problem with your view is that on the basis of its logic there can be no evidence for anything, only evidence against. What you fail to see is that if there cannot be evidence for anything, then there cannot be evidence against anything either; they are two sides of the one inductive coin. — Janus

Repeating this over and over again despite my repeated refutations isn't going to make it true.

It's possible to prove something true by falsifying the predictions of it's negation. It's not possible to prove something true by confirming its own predictions. Different kinds of proof, not different kinds of statements. That is the important distinction. And neither of those is "induction"; that's something else entirely, that comes well before the stage of testing like that.

at the cost of sticking our necks out and make assumptions without adequate justification, which could therefore be wrong — Pfhorrest

Hmm. But for you, beliefs can't have and don't need justification. You describe things here as if relying on a theory incurs a risk because we overstep what we actually know, we project beyond what we have adequate justification for.

But isn't that all belief? Isn't all our knowledge only probable? Or are there beliefs you will countenance treating as certain? Is relying on theory truly different? Or is it the same because we are always relying on some theory without adequate justification?

Or is it the same because we are always relying on some theory without adequate justification? — Srap Tasmaner

Yeah, that's what I meant. ("Theory" and "belief" are being used as rough synonyms here, along with "model" or "hypothesis"). Only a (wholly impracticable) completely non-theoretical approach relying on nothing but the aggregate of our particular experiences, without extrapolating at all, would not be sticking our necks out like that, and nobody does or in practice can do that, and it would be horrendously inefficient to do so even if we could. It's entirely pragmatic to trade the risk of error for the ease of theory.