Are you familiar with Charmer's argument, mentioned above?

Good reply to

So in a way, the question for those of us with a Wittgenstienian bent is, can folk make up a game of metaphysics that can be played in a coherent fashion?

From our point of view, the ball is in the metaphysician's court, to show that there is a way to play the game that makes sense.

So, the question isn’t meaningful, it’s misguided. It treats certainty as something that needs to be justified, when in truth, certainty is what makes justification possible in the first place. — Sam26

Good post. Yep.

I know of two viable responses. The first is from Austin, and looks at how we use the word "real", noting that we contrast it with something that is not real. It's a real dollar bill, not a counterfeit, or it's real vanilla, not artificial, and so on. This works fine.

The other comes from David Chalmers, who agrees more or less with the Wittgensteinian argument that we usually don't use "real" in this way, but goes on to ask why we couldn't. He proposes a room in to which we can go, within which we can ask such questions, and discuss the consequences. Now that strikes me as quite a good response - and we could go down that path. I don't think it quite works, but I won't rule it out forthright. It's part of the thinking behind the renewal of metaphysics that spread out from Australia a few decades back.

A worthy topic. I don't now of any one here who could explicitly defend such a view.

If the only description of Homer is that he wrote the Odyssey, then this story just establishes that Homer is Kostas. — Ludwig V

Sort of. We might say Homer is the guy we think wrote the Odyssey. But turns out it was Kostas who wrote it. Now at stake is the difference between thinking of "Homer" as denoting exactly and only "the bloke who the Odyssey", and thinking of it as denoting Homer, that person. That's what this group of thought experiments target. And that in turn is the difference between the descriptive theory of reference and the idea of a rigid designator. If "Homer" and "Kostas" are rigid designators, then we can say that it was Kostas that wrote the Odyssey, and do so without fear of our system of reference collapsing. If we think in terms of the descriptive theory, and so "Homer" refers to "The guy who wrote the Odyssey", then "Homer" refers to Kostas.

There's the interim possibility, implicit in "the guy we think wrote the Odyssey", that reference is dependent on intent, that "Homer" denotes whomever I intend it to. There are all sorts of troubles with that, not the least being that it begs the question. How is it that what you intend to denote and what I intend to denote by using "Homer" happen to be the very same individual? Which is the very question we were seeking to answer.

There's the point, too, that we might well see that the descriptivist theory is inadequate and yet not have at hand another theory to replace it. We sometimes have to be comfortable to say "I don't know", and to see that doing so is better than trying to repair a defunct theory.

Does this work the other way round? I mean if "a" designates an object in all possible worlds in which that object exists, is it also true that that object is designated by "a" in all possible worlds in which "a" exists. Then is there a possible world in which that object exists, but the Roman alphabet was not invented? — Ludwig V

Sure. It's possible that you were named "Ebenezer" instead of "Ludwig". That would be a fact about you. That we in this world use "Ludwig" does not meant that folk in some other possible world could not refer to you using "Ebenezer". Or 以本尼泽尔, which the AI assures me means "stone of help", which is the meaning of "Ebenezer".

Similarly, if "a" necessarily designates a, can we conclude that a necessarily has the property of being designated by "a"? — Ludwig V

Being designated by "a" is not a property of a. So it can't be a necessary property of a.

It's not a property because that "a" designates a is not a formula within the system, but part of the interpretation, of the model.

Much of the apparent bumpiness here might be worked out by your looking at the formal system and how it functions. You seem to have. good intuitive grasp of the ideas involved.

Are you familiar with the idea of a family resemblance? How much success would you have if you set out to define your family by listing their attributes? Blonde hair and a hooked nose, maybe, except for cousin Philippa, with their less aquiline features and mousy hair. Or all descended from Grandpa Jerome, except the adopted twins. Supose that for whatever feature you choose, there are exceptions, or you include folk that you would not want included.

The idea is that we can talk about our family, despite not being able to give a strict and explicit definition that includes all and only those members we want; and this can be generalised to claim that for some terms there is no explicit definition that sets out all and only those things that are to be included. The other example is "game" - without resorting to mere stipulation, can we provide a rule that includes all and only those things that we have described as a "game"? Not all games involve winning, nor competition, nor amusement. And yet despite this we make good use of the word.

Point being that we do not need to be able to present a definition as a prerequisite for using the word.

We use the word "knowledge" quite adequately, and widely, and yet when we try to tie it down we end up in these interminable philosophical meanderings.

So, do we need to provide a definition of knowledge at all? Perhaps it would be better to just map out the different ways we use the word, as you have begun to do.

One thing we can do is to mark the difference between knowing an believing. We can believe something that is not true. We can't know something that is not true. If you thought you knew something, but it turns out you were mistaken, then you didn't know it at all.

Kaplan is the go-to for this sort of stuff.

That "I" designates me is not a property belonging to me. It's a grammatical function of the use of "I". It's a bit of semantics, not a bit of metaphysics.

It's a bit like the novices who come to the forum with what they take as a profound question - how is it that I am me and not you? They haven't noticed that even if they were me, they could ask the same question.

Otherwise it seems that you're just saying that knowing that p is equivalent to knowing how to assert p. Which would be such a cop-out. — Michael

To sincerely say "I know that P" is to assert that P, while it would be exceeding odd to assert that P while claiming not to know the P.

That's been the line all along: "Knowing that" is a form of "knowing how".

Far form being a cop out, it's the central point here. Bits of knowledge fit in with our form of life, and so with what we do. Our knowing that the sun will expand is a part of our broader knowledge of the nature of the sun, built from our looking around and interacting with the world. It's not an isolated factoid.

Notice that it's not knowing how to just say P that is salient, but how to assert P. That involves knowing how to make use of the speech act of making an assertion, along with all the paraphernalia of truth, justification, reference and so on.

Which is what is not captured by saying that knowledge is true information.

Not really sure how to make use of this information — Michael

You just did.

I am arguing that it would not be possible to overturn all the known descriptions at the same time. That is like trying to saw off the branch you are sitting on - success would be catastrophic. — Ludwig V

Yep. And yet, from the examples given, it seems that even when we saw off the branch, the reference succeeds. And the quest becomes, how can this be?

Yep.

,
,

q,v.

Nice summary of Kripke's view. — Richard B

Thank you.

It's curious that I don't think Kripke would disagree with what you have to say - and if he did, I'd be disagreeing with him!

He's not - at least here - proposing a radical skepticism. He does that elsewhere, and in a very different context.

Adding to that, I don't think he anywhere suggests that we might discover that everything we know about water might be wrong. That would be a long bow to pull, as you point out. Am I mistaken?

letters — Banno

In my reply to , I gave an account of the strategy and argument I think he is adopting in setting out these arguments. His target is not knowledge generally, and he is not advocating radical skepticism. He's arguing against a once-common view of reference, and extending that from proper names and individuals to kinds.

So that's why the odd examples, tigers and such. Let's look at a bit where he sums up what he is doing wioht tigers:

Now tigers, as I argue in the third lecture, cannot be defined simply in terms of their appearance; it is possible that there should have been a different species with all the external appearances of tigers but which had a different internal structure and therefore was not the species of
tigers. We may be misled into thinking otherwise by the fact that actually no such 'fool's tigers' exist, so that in practice external appearance is sufficient to identify the species. — N & N p.156

Note well "...and there fore was not the species of tigers". The conclusion isn't that there could be a tiger with different internal structure, but that if it had a different internal structure, it would not be a tiger.

And the conclusion with regard to discovering that some stuff we thought was water had a radically different structure to our water would similarly be, it's not water.

He's not in the end all that far from your own view.

His account is dependent on the idea of a chain from name to referent, a chain he says is "causal", but I think that's a stretch, since "causal' covers a multitude of other sins. I've my own thoughts.

Scribbles are just scribbles unless they refer to something. — Harry Hindu

We've been over this previously, and it's a bit of a side issue, but I don't agree with your theory that words are all proper names, that all they do is refer.

Another time.

For me, things are real if they possess causal power. — Harry Hindu

I don't find this very useful, since "causal power" is not as clear a concept as "real". Indeed, I doubt that the idea of causation can be made all that clear. But there is a clear use of "real", which I've explained previously - it is used in opposition to some other term, that carries the explanatory weight - it's real, and not a counterfeit, not an illusion, and so on.

It doesn't help us if we explain one unclear idea by using another idea that is even less clear.

In the formalisation, there are letters - "a", "b" and so on - that stand for individuals in possible worlds. The standard interpretation is that each rigidly designates that individual in each possible world in which that individual exists. So "a" designates a in world one, and also in world two, and so on. Now in each of these different worlds, a and b can have different properties - f(a) in one world, ~f(a) in another. And that there need be no properties that a has in all possible worlds.

The question Kripke and others were asking is, to what do these letters match in a natural language? And what are the consequences of that matching?

And the answer, speaking roughly, is that "a" and "b" are proper names for a and b and so on.

And the conclusion seems to be that there need be no properties that are had by a thing names, in every possible world in which it exists. Naming and Necessity is at it's core an attempt to fill out the consequences of this idea in a way that is consistent.

I think of it this way. We know that the formal system is consistent. We can look at a natural language such as English and match the bits of that language to the formal description, and perhaps in doing so learn how to treat modality in a natural language in a consistent fashion.

So we say that "a" and "b" are like proper names, and since "a" and "b" are rigid designators, we say that proper names are also rigid designators. And so it seems that since no property need be true of "a" in every possible world, no property need be true of a proper name in every possible world

But then we run into the problem that the then most popular theory of how proper names work is that the name matches a description. And a description is just a bunch of properties. So we have the problem that if proper names are rigid designators, then they are not descriptions.

There's something very odd about saying that we learn what some thing is, and then discover that what we have learnt about it is false. What is the "it" here? — Ludwig V

Yep, that's the issue.

Here's a rough solution. We might well learn how to use the name - what it designates - by using a description. But thereafter, we are not reliant on that description for the name to work. We learn who "Charles Mountbatten-Windsor" is by watching his coronation on TV, perhaps. But if it turned out that they had put an actor in to take his place, perhaps for security reasons, that would be something we learned about Charles Mountbatten-Windsor, despite his not being the chap on the TV.

It gets more complicated, of course, wich is why Naming and Necessity is a book, and not an essay.

Try a different example. Homer. I'm sure you know about him, and that there are good grounds for thinking that he never existed. But those stories exist; someone must have written them - or perhaps they are folk tales with no author in the sense that we apply the term. So our expectations when we learn the Homer wrote those epics are disappointed. But not everything that we learnt when we learnt the name is false. — Ludwig V

There's a few different ways this could pan out. We might supose that there was a bloke names Homer, and indeed he wrote the Odyssey. But possibly, it was Kostas, his acquaintance, who did the writing, and Homer stole the text and took the credit. Now if what we mean by "Homer" is just the person answering the description "the bloke who wrote the Odyssey", when we say "Homer", we'd be referring to Kostas.

Indeed, if the referent of "Homer" is fixed only by "the bloke who wrote the Odyssey", we could not coherently claim that homer did not write the Odyssey, becasue that would amount to saying that the bloke who wrote the Odyssey did not write the odyssey.

And what we can conclude is that, contrary to both Russell and Quine, proper names are not just shorthand for descriptions, but work even in the absence of a description. They do function s rigid designators.

Good posts on your part, by the away. Fine analytic stuff.

...potential energy... — frank

Potential energy is creative accounting for physicist. They invented it in order not to falsify the principle of the conservation of energy. :wink:

Suppose I know P, but I never act on it. How am I different from a person who knows P, but can't act on it? — frank

You're not.

Perhaps we can assume honesty, and you said that you know, while the person who doesn't know also says that they don't know. But saying you know amounts to acting on your knowledge...

So that question might not be as simple as it at first seems.

I found that a bit hard to follow, but it looks to be a galant attempt at elucidation and analysis.

The justified true belief account comes from Socrates in the Theaetetus, and even he wasn't happy with it.

You're on the right track, I think, in looking to the way we use the word "knowledge". But here's a puzzle for you: must there be one statable phrase that covers all our uses of "knowledge"? Could it be that we use the word in different ways, such that no fixed definition is both accurate and compete?

Moreover, will we say is the correct uses of "knowledge" are only those that conform to some stated definition?

At the least, that rules out any novel uses. Do we want to do that?

...the modal, which was the topic of another thread with Banno from which I've not yet recovered. — Hanover

:grin: Meta is in worse shape, thanks to you.

I think I agree, but I'm not sure what the conclusion was.

Yep.

Knowledge doesn't need to be about how; that's just one kind - practical knowledge. The input of one's own senses and internal functioning is another kind - direct internal knowledge. The second kind doesn't need further study, since it's already integrated: it's established in the material body as well as in the mind. Sensations are known without reference to language or concept. — Vera Mont

This is also good. Wittgenstein pointed out that we do not know we have a pain, we just have a pain - and here he is using "know" as justified true belief, and pointing out that it makes little sense to talk of justifying to oneself that one is in pain - since what counts as the evidence is just the pain itself.

And yet it also makes little sense to say that one is in pain yet doesn't know one is in pain.

This fits in with knowhow. One is said to know how to ride a bike once one rides a bike. The justification is the act.

The upshot? Folk sometimes supose that knowhow is an exceptional case of knowing that; that propositional knowledge is central. It's the other way around. Propositional knowledge is s special kind of knowhow.

Do you take the assessment of the truth value of a proposition as knowing-how knowledge — Hanover

Pretty much. Working out what is true and what isn't, is an activity, something we do. We look around, we do the calculation.

equivalent to juggling balls? — Hanover

...not so much...

Seems evaluating statements requires cognitive grasp of concepts — Hanover

"cognitive grasp of concepts..." You are said to grasp a concept if you can show that you understand it. You show that you grasp the concept of bike riding by riding a bike, or at least by recognising a bike rider.

To grasp a concept is being able to act in certain ways.

Good to see you doing some analysis.

We might require of a definition that it explicitly sets out what is and what isn't included in the definends. Sometimes we can do just that. For some definends, all we can do is set out a family resemblance, listing the things that are sometimes included, sometimes not, and understand that there may be exceptions.

Treating knowledge as strictly Justified True Belief will have as a consequence the contradictions that result in threads such as this.

Better then to look at the sort of things that are and are not included in knowledge, at how the word is used, rather than just stipulating a definition. Our use of words tends to exceed any such stipulation.

So sure, Longshanks knew what would happen, and yet couldn't act on that knowledge. It's the abnormality of his not being unable to act in this case that makes it startling and exceptional; and dramatic.

If you prefer, while knowledge doesn’t logically require the ability to act (as Longshanks shows), it normatively and ordinarily includes that capacity.

To know something and not be able to act on it is the exception, and a performative contradiction.

sweet, passive-aggressive Banno. — T Clark

No more than

...to tell the truth, I don’t really care about what it means to know how to do something. — T Clark

By ignoring knowhow you are protecting your ideas from critique, rather than willingly exposing them to analysis. As I said, that's a shame.

when I literally tried explaining the difference between changes occurring and time passage. — ArtM

Your explanations might not be as convincing as you suppose.

Let's go back to that first paragraph:

Imagine waking up tomorrow, realizing that thirty years of your life vanished, not forgotten, but as if they never existed at all. You jumped from infancy to adulthood in the blink of an eye, with no memories in between. This scenario sounds impossible, yet it’s exactly what occurs in situations like comas, alcohol-induced blackouts, or even during periods of deep, dreamless sleep. Here’s the profound question that emerges: if time is genuinely a fundamental dimension of our universe, why does it cease to exist the moment consciousness fades away? — ArtM

The hypothesis is that thirty years have passed, while you were unconscious. And that also time ceased to exist while you were unconscious. It's self-contradictory.

Welcome to philosophy.

Really, so if you were the only conscious being in the world, and you woke up after a Thirty years, you would be able to tell that Thirty years have passed? You must be different. — ArtM

Dude, learn to use a calendar.

I think you are enthralled with a pretty thought. It's not such a good one on reflection. If you go to sleep and thirty yers pass, then by that very supposition, time passes when you are unconscious.

Here’s the profound question that emerges: if time is genuinely a fundamental dimension of our universe, why does it cease to exist the moment consciousness fades away? — ArtM

Time didn't cease while you slept - it passed. Thirty years of it, demonstrably.

I already answered what you're saying several times, even during the hypothesis. — ArtM

I'm not seeing it. Perhaps it's not as clear as you think.

Is your claim that time does not pass when you sleep? No, becasue other folk are awake. So if we all go to sleep, no time passes? Why would time stop becasue we were all unconscious? It was there before the world was formed, so far as makes sense.

Time seems to pass during sleep only because there are still other conscious beings around observing and measuring it. — ArtM

The sun sets and rises, the stars turn, the frost settles in - You see the changes when you wake.

You do not have direct experience of other folk's consciousness. You only infer it. So why can;t you also infer that time passes while you are unconscious?

OK. That looks hopelessly confused. Time passes when you sleep. Time passed since the Cretaceous period. But I get the impression that you are inured to such things.

So I'll say again, that without consciousness, we would not be aware of time passing, is a very different thing to time being brought about by consciousness. You want the latter from the former.

Time passes when you are asleep. Therefore consciousness is not needed for time to pass.

How extraordinary, to have to point this out to an adult.

Actual knowledge can't be divorced from the whats, hows and whys of the physical world. — Vera Mont

Yep. Very much so. Knowledge is embedded in what we do, in ways well beyond the place of information.

That, without consciousness, we would not be aware of time passing, is a very different thing to time being brought about by consciousness.

There seems to be an error along these lines going on in this thread. It's a very common, popular misunderstanding.

Thanks for the clear response, Ludwig

Forgive my ignorance. That suggests that you have an independent definition of "extensional context". But I thought that intersubstitutability was the definition of an extensional context. ? — Ludwig V

No, that's right, this is the circularity I mentioned. It's an extensional context if substitution works. Being extensional and allowing substitution that preserves truth are the same thing. We are either in an extensional context, or not.

Are you possibly confusing "All the propositions that we think we know about tigers are false" with "Each of the propositions that we think we know about tigers may be false"? — Ludwig V

I don't believe so. The idea is that we learn what some thing is, name it, and then discover that everything we knew about it was false.

I couldn't locate the original Thales example - I think it was Kaplan - so I had ChatGPT reconstructed something similar:

Consider the historical figure Thales of Miletus. Tradition holds that he was the first philosopher, perhaps the first to suggest that water is the fundamental substance of all things. Yet, on closer inspection, our supposed knowledge of Thales collapses into uncertainty. Were these views really his? Were the anecdotes true? Or are they accretions of later doxography and myth-making?

As historical scrutiny deepens, it becomes clear that we know almost nothing about Thales with certainty. Yet this very realization—that we know nothing about him—is itself a fact about Thales. It is not a fact about someone else or a mythological construct; it concerns the very individual to whom the name "Thales" refers.

Therefore, paradoxically, our ignorance becomes a form of reference. The name "Thales" successfully picks out an individual in history, even though our beliefs about him may be largely mistaken or minimal. This supports the view that the name refers rigidly and directly, independent of any particular descriptive content we might associate with it. The denotation succeeds not despite our ignorance but is revealed in it. — ChatGPT

What this shows is that we don't manage to pick out Thales in virtue of what we know about Thales, a somewhat counterintuitive result. There's a bunch of such examples, from Kripke, Donnellan, Kaplan and others, that have pretty much undermined the so-called "descriptivist" account. The suggested replacement - the "causal" account - has about twice as many adherents on the PhilPapers survey, despite not being all that well articulated.

What this doesn't rule out is the sort of view that might be seen in a Wittgensteinian account, in which reference is an aspect of the more general language games in which we participate, or even a sub-game within those games. On such a view a reference may be counted as successful if we get on with what we are doing, regardless of how it managed to denote it's target. I think Malcolm's concerns were misplaced (@Richard B)

So now I'm wondering how reference is achieved. — Ludwig V

Good question. To my eye, it's clear that we sometimes do work out a reference from a description associated with it; it's just that we can show that this is not what happens in every case. Indeed, it should hardly be a surprise to learn that there is more than one way for a reference to succeed.

And even less reason to suppose that references are dependent on some sort of essence.

I've my own ideas about how to explain reference and such, (@J), but we might move on without a general theory of reference, if we agree that somehow it manages to work despite our not understanding quite how.

I'm not too keen on talk of essences, either. Whatever they are, they are peripheral to the issue of what is real and what isn't.

We have three or four differing views of the nature of essences here.

There's the older view in which to understand what something is just is to understand it's essence. That's perhaps what Tim is thinking here. On that account, being real and having an essence are pretty much the same thing.

There's the more recent analytic natural language view, from the later Wittgenstein through Malcolm and maybe @Richard B, and close to that taken more formally by Quine and friends, that there's not much more to essences than confusion.

Then there's Kripke's suggestion, that if we must think of essences we can think of them as the properties had by something in every possible world in which that thing exists. This has the benefit of being formalisable and reasonably clear while keeping to a minimum any metaphysical consequences.

Then you may be suggesting that we can be rid of essences by doing some sort of Bayesian analysis that allows us to conclude that tigers are real. Maybe.

But you and I might agree that essences have little to do with what is and isn't real.

Knowing how to use a faucet is not the same thing as knowing that any particular faucet is working... — Count Timothy von Icarus

Sure.

The point made is that in order to be said to know something, it's not usually enough to have the information; one also should be able to act on that information.

That seems to have little to do with being able to carry on a ritual without having faith in the accompanying theology.

So what do we conclude?

I don’t really care about what it means to know how to do something. At least not in the context of philosophy. — T Clark

A shame. Fine.

If you had someone who could set out, to whatever degree of detail you like, what is involved in riding a bike, and yet fell off every time they tried to ride, would you say that they know how to ride a bike?

They have all the information.

But they can't do it.

Hence knowledge is more than information.

When I get involved in a discussion such as this one, I usually make it explicitly clear the kind of knowledge I'm talking about - specifically excluding knowing how to do something. — T Clark

Yep. Not an uncommon move. Is it justified? Is there a difference in kind here? You know that there is water in the tap. You can show that you know this by saying "There is water in the tap", or by going and getting a glass of water. Going and getting a glass of of water is something you do. But so is saying "There is water in the tap".

Indeed, if we came across someone who said "I know that there is water in the tap", but became confused when asked to locate and turn the tap on in order to obtain a glass of water, we might well conclude that they said they knew but really didn't.

There seems to be a pretty good argument that "knowing that" is a type of "knowing how".

I'm not convinced that you can neatly slice knowing how from knowing that.

What do you say?

Was it @counterpunch, 4 years ago, who had a thing about this? I might have the wrong bloke.

Yep. Notice that reference remains intact despite the failure of each description. Hence reference is not achieved by using descriptions, nor by essences.

It's a poor example that Kripke chose, and a somewhat difficult idea to get across, but the point Kripke makes is at least in part quite right, if misunderstood.

You may be familiar with the theory that a name refers in virtue of an associated description, and the various arguments mounted against it after the advent of Kripke's semantics. It was found to be inadequate in certain regards, and few still adherence to it.

What Kripke is doing is pointing out that this applies to types as well as to individuals. It's a hypothetical, in line with the familiar Thales example and others. If we did find out that everything we knew about tigers were mistaken or in error, that would nevertheless be a discovery about tigers. It follows that "tiger" does not refer to tigers in virtue of some description that sets out their characteristics.

In the hypothetical, we had "established" what a tiger was on the basis of an "agree on definition and judgment as it is applied to our natural surroundings" that was, in the hypothetical, wrong. And yet we nevertheless still manage to pick out what is a tiger and what isn't. It follows that we do not pick out what is a tiger and what isn't, on the basis of supposed essential characteristics of tigers.

The upshot is agreement - we indeed do not identify tigers on the basis of some essence that exists in all possible worlds.

The argument is on p.120 of N&N, for those reading along.

Secret figures show Liberal party’s ageing membership in freefall in NSW and Victoria

Average age of 68.

“One of the biggest expenses we used to have [at our local branch] was on funeral wreaths.”

If I've understood, knows things he doesn't believe, while knows things that are not true.

And neither account can explain what it is to know how to ride a bike.

:grimace:

I didn't actually respond to your post.

What exactly is Kripke's value in calling them identity statements? — Richard B

It's not just Kripke, and it's about substitution.

In set theory, the Axiom of Extensionality is

$\forall A \, \forall B \, \left( \forall x \, (x \in A \leftrightarrow x \in B) \rightarrow A = B \right)$

Given two sets A and B, if they have the same elements then they are the same set. Now one way to treat this is as a definition of "=". It's also the definition of extensionality.

If we are instead talking about functions,
$\forall x \, (f(x) = g(x)) \Rightarrow f = g$

and for predicates
$\forall x \, (P(x) \leftrightarrow Q(x)) \Rightarrow P = Q$

and in arithmetic
$x = y \rightarrow f(x) = f(y)$

That is, x=y if they denote the same number.

Or generally,
$a = b \quad \iff \quad \forall \varphi \big( \varphi(a) \leftrightarrow \varphi(b) \big)$
In an extensional context, a=b iff for any string, substituting a for b does not change the value of that string.

Now in physics, extensional equivalence might best be thought of as when two sides of an equation are measured in different ways but always yield the same values in every case where the law holds, and the equation is not a definition, but an empirical or theoretical identification.

$F=ma$ and $v = \frac{d x}{d t}$ are definitions, so the extensionality is built in. But in Ohm’s Law, $V = IR$, the two sides of the "=" are measured in quite different ways, and yet their value is the same. Much the same for $\langle E_k \rangle = \frac{3}{2} k_B T$. What's suggested is that $F$ can be substituted for $ma$, $V$ for $IR$, and $\langle E_k \rangle$ for $\frac{3}{2} k_B T$.

There's a catch here, since the circumstances in which the substitution occurs must be carefully controlled. And indeed there is a benign circularity in that extensionality is defined in terms of substitution, and yet substitution is not permitted in cases that are extensional opaque.

If your "⇔" is understood as, in the appropriate circumstances, permitting the stuff on the left to be substituted for the stuff on the right, then it is extensional and does much the same job as "=".

The identity here isn't metaphysical; it's just substitution.

And that's pretty much Kripke's point. If a=b, then you can substitute a for b and get the very same result, provided that you are working in areas where extensionality works.

So if we discover that this lectern is made of wood, then in every possible world in which this lectern exists, it is made of wood. And if in some possible world the lectern before us is made of plastic, then it is a different lectern.

And if we discover that water is H₂O, then ☐(water=H₂O). And so on through the many different examples. Notice how each of these is a hypothetical – an "if... then...", in which the antecedent is found a posteriori - by looking around. But all that is being done here is ensuring that we keep our language consistent. Once we fix reference by accepting the antecedent, we are obligated to respect the modal consequences of that act of naming as expressed by the consequent.

Going back to the "heat is average energy of the molecules" example, Kripke fell over because heat is a sensation, and fails the test of being extensional. But temperature is extensional. That's why changing from heat to temperature works. And that also why his examples of pain and c-fibres are problematic - pain is not extensional.

So much of Malcom's - and your - criticism is valid. But Kripke wasn't entirely wrong, either.

If the whole ambit of philosophy is human experience and judgement then is it not always a matter of "what can (coherently and consistently) be said? — Janus

What can be said is a start. What can be shown might be more important. That's part of what is problematic about mysticism. If it is showing stuff rather than saying stuff, it's not actually false. But when it says stuff, it is almost invariably false.

So, the Op question reframed would be not "how do we know what is real?" but "how do we decide what counts as real?" — Janus

I still prefer "How do we use the word real?"

I am sir. — Moliere

Good to know.

I am sure our paths shall cross again about this topic. — Richard B

Sure. I still haven't responded to the points you made in your previous. Will do so later.


Perhaps. But at the very least philosophical theories ought be internally consistent, so there is a point to the process of working out what that looks like. If it doesn't matter what can be said then anything goes.