...it is true at all times that Gödel was born on April 28, 1906, for instance. Of course, the sentence now being used to express this truth uses the past tense whereas a sentence used to express it prior to April 28, 1906 would use the future tense. But both sentences express the very same truth and there is no time when what it is that they express isn't true. — Pierre-Normand

Well, I differ here wrt predictions being true at the time of utterance. Bt my lights, they are not able to be.

"Godel was born on April 28, 1906" is not a definite description though, is it? "Born on April 28, 1906..." is, right? If so, then this doesn't clear up what was in question to begin with.

Yes, it is true at all times that Gödel was born on April 28, 1906, for instance. — Pierre-Normand

Every true description of an entity at any time in the form at 'At that time the entity was X' is true at all times.

If we can target the search to the entity in order to find a definite description for it, we must not require a definite description beforehand to do the search — fdrake

But we always already have webs of description about any entity we refer to (unless the entity is present and we point to it), it is never simply 'the entity'. 'Which entity do you refer to?' 'The one that....' How can we search for a definite description of an entity unless we know which entity we are searching for a definite description of? How can we know which entity if the entity is not right there in front of us?

I haven't seen a single argument in this thread (or anywhere else) that demonstrates that we can refer to anything without relying on definite descriptions (or ostention). The fact that we might not have descriptions specifically in mind when we refer is not a cogent counter-argument against their indispensability, in my view.

Every true description of an entity at any time in the form at 'At that time the entity was X' is true at all times. — Janus

Yes, that's basically what I am saying.

Well, I differ here wrt predictions being true at the time of utterance. Bt my lights, they are not able to be.

"Godel was born on April 28, 1906" is not a definite description though, is it? "Born on April 28, 1906..." is, right? If so, then this doesn't clear up what was in question to begin with. — creativesoul

"Born on April 28, 1906..." is a predicate. According to descriptivism, proper names have the same sense (meaning) as definite descriptions written as "The ...". Russell proposed to analyse them as incomplete symbols that introduce quantificational structure into sentences in which they occur (as Wikipedia puts it). For instance, the sentence "The King of France is bald" can be analysed as the conjunction of three quantified statements that assert (1) the existence of an x who is a King of France, that (2) any y who is a King of France is x, and that (3) x is bald.

But when a substance falls under a definite description at a time, then it falls under it at all times (including the times when it doesn't exist yet or anymore!) That's what makes it a definite description, rather than a general description. — Pierre-Normand

Alright. It's becoming more and more obvious to me that we're working from entirely different conceptual schemes(linguistic frameworks). Most everyone involved here seems to have an academic background. I have no formal philosophical background. As a result, I will not be able to recount many historical debates, let alone be able to recount them in great detail by virtue of offering an adequate account of the belief system/conceptual scheme/worldview at work on either 'side' of many well-known academic level debates.

All debates have something at issue and folk expressing differing thought/belief about that issue.

What does successful reference require? What does our knowledge of all successful reference require? There is no single criterion for both. The latter will include some things that the former cannot include. The latter will include everything in the former. Our knowledge of successful reference requires successful reference.

All successful reference is something done using language. All language use requires shared meaning. All successful reference requires shared meaning. Whatever shared meaning requires, so too does language. Whatever language requires, so too does successful reference. Whatever shared meaning requires so too does successful reference.

Successful reference has a criterion. In the preceding paragraph is the beginnings of a rough outline. I am not claiming that it is - as it stands now - adequate. However, I think that it is universally applicable - as it stands now. I think that it is universally extant - as it stands now. We can flesh out more detail later.

Successful reference is itself a complex thing. I do not think that we can offer an exhaustive account that includes every thing that successful reference is existentially dependent upon. I do think that we can safely posit a number of them. I do think that such knowledge serves as more than adequate ground to warrant it's use as a standard of measure.

Shared meaning requires some things that exist in their entirety prior to becoming a part of shared meaning. Since all successful reference requires these things, any conception or theory of reference that that contradicts this knowledge of shared meaning, and thus this knowledge of all successful reference is just plain wrong.

We can acquire knowledge of complex things. Our knowledge of such things requires targeting the thing. We must pick it out and carefully consider it. This must be the case, otherwise we could never acquire knowledge about elemental parts. We could never know or say stuff in particular about some thing if we do not first isolate that thing. In doing so it becomes the focus of our attention. Names are very popular tools for doing so. They are not the only means for successful reference. We can point without naming. We can name without pointing. We cannot describe the thing in detail without doing one or the other. We cannot do any of this without shared meaning.

So, all accounts of elemental constituents requires first pointing or naming; picking out the individual thing. Our accounts of elemental constituents requires language. The existence of some elemental constituents does not require our knowledge of them. Both, the existence of elemental constituents, and our knowledge thereof requires shared meaning.

If some thing consists of other things and some of those things exist in their entirety prior to becoming a part of that some thing then that some thing requires all of the things that it consists in/of.

All successful reference includes and requires shared meaning. All successful reference is existentially dependent upon shared meaning in addition to all that shared meaning is existentially dependent upon.

Shared meaning requires some thing to be symbol/sign, some thing to become symbolized/significant, and a plurality of creatures capable of drawing mental correlations between these things. Correlation presupposes the existence of it's own content. <---------That is the presupposition of correspondence to fact/reality that is inherent to all thought/belief formation, including but not limited to statements thereof.

All successful reference presupposes correspondence to fact/reality. If it is the case - in this world - that all X's consist of known elemental constituents, and a possible scenario stipulates otherwise, it is false. Such possible world scenarios are to easy enough to imagine. All that that takes is making up a coherent story that talks about reference without description.

We can know some things about this world. We can imagine what it would take in order for those things to be different. If we know that something is a composite then we may also know what it is composed of. If we know what it is composed of, we could easily talk about that thing as if it is not composed of what we know it is.

Does our doing so successfully refer to to the thing? Surely. Can any of it be true? Surely not. Is that mode of reference somehow not existentially dependent upon any description whatsoever? As if we could do any of that without already having picked that thing out of this world by virtue of both description(s) and names?

I think not.

Does our doing so successfully refer to to the thing? Surely. Can any of it be true? Surely not. Is that mode of reference somehow not existentially dependent upon any description whatsoever? As if we could do any of that without already having picked that thing out of this world by virtue of both description(s) and names?

I think not. — creativesoul

There is much for me to agree with in your long post, and a few issues that I could quibble with, but I am unsure how it connects with the previous line of inquiry. The issue of time came about when @fdrake suggested that a definite description could apply to an individual at a time and cease to apply to it later on. And therefore, as he had seemed to imply, for a definite description to single out a persisting individual it would need to apply to an unchanging individual. I pointed out that definite descriptions typically single out an individual through ascribing some property (or set of properties) that uniquely apply to it at a specified time. When the time is thus specified (either explicitly or implicitly) in the definite description, then, it becomes irrelevant that the item doesn't have the property ascribed to it at other times.

If my definite description of an apple is something like "The green apple that sits on my kitchen counter on December 14, 2018..." then, this description still will pick up the same apple in the future when it has turned red. Hence, if the sentence "The green apple that sits on my kitchen counter on December 14, 2018 has a stem on December 14, 2018" is true on December 14, 2018, it will remain true, about the very same apple, after the apple has turned red, has had its stem removed, or even has ceased to exist.

After thinking it through a bit, while it's true that if a definite description applies to something at one point it will apply forever, this does nothing to vouchsafe whether the definite description can actually be used to disambiguate a reference when required. If we have a 20 Bobs, and I choose to disambiguate the Bob by providing a a predicate which makes the appropriate Bob the member of a singular extension; say, 'the Bob that was born in 1972', this does not imply that a competent user of English can use the predicate 'was born in 1972' to pick out the right Bob. Being able to use the predicate which induces the singular extension is an epistemic state, requiring that a someone not just know the meaning of the disambiguating predicate, but must also have access to information which facilitates checking whether this predicate obtains in each case.

When I refer using a proper name out in the wild, it is incredibly easy to provide a disambiguating expression in most circumstances. But there is no guarantee that the disambiguating expression will actually disambiguate, or perhaps even could in principle, even if that disambiguating expression has a singular extension. So, having the logical structure of a definite description is not sufficient to actually narrow down the options through its use.

Perhaps a clearer example is that we're in a hospital and there are two Bobs which are extremely difficult to tell apart, if someone were to ask us 'Which Bob is this?', we could respond with 'The Bob that was born on...' if they had a different time of birth, but this will not allow a recipient to know which Bob is which without access to appropriate information; in this case a birth certificate or other medical note.

I suppose what this highlights is that despite the logical structure of a definite description ensuring that it focusses on a single entity, that logical structure only works to disambiguate if the predicate in the definite description is chosen appropriately by the speakers. We have the same kind of behaviour with the word 'The', whereby applying Russel's substitution procedure does not have to produce something which actually allows us to disambiguate. With the Bob example, responding to which Bob is which, nothing changes in terms of the epistemic states or filtering whether we say 'Oh, a Bob which was born on ...' or 'Oh, the Bob which was born on...'.

There's a really big difference between the logical structure of definite reference and the pragmatics of disambiguation.

Edit: the same goes if proper names are rigid designators, that rigidity doesn't facilitate the formation of disambiguating expressions in any way.

After thinking it through a bit, while it's true that if a definite description applies to something at one point it will apply forever, — fdrake

Yes, this really was my only point.

this does nothing to vouchsafe whether the definite description can actually be used to disambiguate a reference when required.

I agree with this, and with the rest of your post. Of course, one of Kripke's main objections to descriptivism is that it fails account for our evaluation of counterfactual conditional statements where the individual talked about fails to fall (or non uniquely falls) under its description in the counterfactual antecedent. And that's because, unlike proper names, definite descriptions aren't rigid designators. (They still can be used for purpose of initial "reference fixing", as Kripke would say, but then the issue of what it is that contextually, or informationally, is being relied on for purpose of disambiguation only is partially addressed by Kripke's "causal theory of reference").

I agree with this and would add my observation that - in my experience - nearly all sentences that use proper names use them in a way that is time independent or is tied to a specific time, thereby anchoring the implied definite description to that time.

Examples:

1. Aristotle tutored Alexander the Great.

The DD implied by the Proper Name Aristotle must relate to properties that held around the time of Alexander the Great, whenever that was.

2. Hugh Grant is married to Liz Hurley.

Since this sentence uses the present tense, the definite descriptions it uses need to be true at the time of utterance. It doesn't matter if they were previously untrue or subsequently become untrue.

3. General Gaius Julius crossed the Rubicon with his army to challenge the leaders of the Roman republic.

This sentence is tied by historical reference to the epoch when a Roman general crossed the river Rubicon with his army to challenge the Roman government. Since there was only one such event in Earth history, that ties the time to somewhere near 50 BCE. So that's the time at which the DD used to crystallise the Proper Name Julius Caesar must be true.

4. The area of a circle is Pi times the square of the radius.

I want to say that this is time independent and that Pi is a Proper Name and a DD for it is something like 'the number that is the ratio of a circle's circumference to its radius'. But I wonder what others think about whether Pi is a Proper Name and/or whether that is its definite description.

Examples:

1. Aristotle tutored Alexander the Great.

The DD implied by the Proper Name Aristotle must relate to properties that held around the time of Alexander the Great, whenever that was. — andrewk

Are you purporting to defend a form of descriptivism, then? What if the individual who we name "Aristotle" had not become a philosopher, and had become a carpenter instead (and he hadn't been Alexander's teacher, etc.) Are we talking about someone who isn't Aristotle, in that counterfactual scenario? And if we're still talking about Aristotle having had a different career, how it is that "Aristotle" picks up its referent in the couterfactual scenario?

When the time is thus specified in the definite description, then, it becomes irrelevant that the item doesn't have the property ascribed to it at other times... — Pierre-Normand

Then it would only follow that the retention of that particular property is not necessary for us to pick it out at other times. Those particular properties are not elemental constituents.

If we offer a true report of elemental constituents, there is no need for time stamp. Time stamps are irrelevant in these cases, for when X is existentially dependent upon Y it is always so. When X is a composite of other things which exist in their entirety prior to becoming a part of X, then it is the case that X is existentially dependent upon all of those things.

That never changes regardless of what we say.

Then it would only follow that the retention of that particular property is not necessary for us to pick it out at other times. Those particular properties are not elemental constituents. — creativesoul

Are what you call "elemental constituents" something akin to essential properties? In that case, the item being descriptively referred to could not persist through the loss of those properties, but they may still not guarantee that the item is uniquely being described by them since other items of the same essential kind also would have those properties. The purpose of a definite description is to uniquely pick up an individual, not just to pick it up under a description that it will never (and could never) cease to satisfy.

Are you purporting to defend a form of descriptivism, then? What if the individual who we name "Aristotle" had not become a philosopher, and had become a carpenter instead (and he hadn't been Alexander's teacher, etc.) Are we talking about someone who isn't Aristotle, in that counterfactual scenario? And if we're still talking about Aristotle having had a different career, how it is that "Aristotle" picks up its referent in the couterfactual scenario? — Pierre-Normand

One has to be careful how one sets up counterfactuals, because they usually end up being nonsense, no matter what metaphysics or language philosophy one favours.

One way of describing a counterfactual about Aristotle would be as follows:

Imagine a world that was identical to ours in every salient respect up to five years after the birth of the person that in this world we call Aristotle. Since everything matches up to that point, we can pick out a person in the imaginary world that corresponds to our Aristotle by virtue of having exactly the same history up to age five, and we will call that person Aristotle-2 (although people in that imaginary world would call him Aristotle, as events at his naming were identical to those in the naming of our Aristotle). Now let us imagine that in that world Aristotle-2 became a carpenter and worked happily at that all his life until he died at the ripe old age of 83.

The picking out is done by matching the first five years of the lives of Aristotle and Aristotle-2, including the parents and other environmental features.

That works for me. It avoids using ill-specified notions like 'referent' or asking (IMHO) meaningless questions such as 'was Aristotle-2 Aristotle?'

That works for me. It avoids using ill-specified notions like 'referent' or asking (IMHO) meaningless questions such as 'was Aristotle-2 Aristotle?' — andrewk

But I didn't ask any meaningless question, neither did I postulate any imaginary worlds. Counterfactual conditionals are judgement forms that we routinely make use of when we reason practically. If I let the cat outdoors and it gets run over by a car, then it isn't nonsensical to judge that it (the very same cat) would not have been run over if (counterfactually) I hadn't let it out. And this is the sort of counterfactual consideration that we rely on when planning future actions or when we are gathering evidence for the existence of causal relations.

Kripke is very insistent in Naming and Necessity, while arguing against the 'telescope conception' of possible world identifications of particulars (ascribed to Lewis), that he isn't talking about counterpart 'worlds' populated with doppelgangers who only share our past histories (or our histories prior to a specified moment in time). When he's talking about 'possible worlds', he's only talking about the our world (i.e. the real world) as it could possibly have been if something or other had been different; just like I was talking about the cat (likely) not having died if, counterfactually, it had not been let outdoors.

Back to Aristotle, the question simply is: what is it that would entitles you to speculate about what would follow from Aristotle counterfactually having had a different career (while referring to him by his name) if his having had his actual carrer (i.e. a philosopher) makes up part of the definite description by means of which the name "Aristotle" refers to him? Surely, you're not saying that it is impossible or nonsensical to suppose that he might have had a different career. But if "Aristotle", by definition, refers to someone who has been a philosopher, then, by definition, Aristotle (where I am using this name descriptively) couldn't possibly not have been a philosopher. This seems not to match how we use proper names in the context of making ordinary and meaningful counterfactual conditional statements.

The difference between us seems to be in how we interpret people's use of counterfactual statements. I interpret them as meaning imagining a world that is identical to ours up to a point, and then at a certain time starts to differ. Another way that's quite similar is to imagine it being the same world and then branching at the critical time, which is the way many people think of Everett's many-worlds hypothesis. Perhaps that's closer to your conception. Recently I have been debating with myself whether there's actually any difference between the two though, beyond the words in which they are expressed.

I suggest that if we were to interview a non-philosopher that has just used a counterfactual, exploring with them the question of what they were trying to convey, we would arrive at that sort of interpretation. I think that phrases like 'the very same cat' are unhelpful and confusing in that context. They hinder rather than help understanding, as the tendency to use italics when referring to them, as if that somehow makes the meaning clear, indicates. If it really were clear, surely we could provide a better explanation than just italicising the reference.

I'm interested in what you said in the first para that 'this is the sort of counterfactual consideration that we rely on when planning future actions or when we are gathering evidence for the existence of causal relations'. I would exclude considering future actions from that because that is usually a case not of imagining the past being different, but rather imagining more than one different possible future, neither of which contradicts current knowledge. Those are not counterfactuals but rather considerations of future possibilities - I call them 'Hypotheticals'. By 'gathering evidence of causal relations' I assume you are referring to the attempt to develop scientific theories. I agree that counterfactuals can play a key role in that but it seems to me that they work perfectly well with my interpretation of counterfactual, and don't require a Kripkean interpretation.

My view on these issues is set out in somewhat more detail in an essay I wrote a couple of years ago: Hypotheticals, Counterfactuals and Probability. Those are still my views.

BTW you asked above if I was seeking to defend descriptivism. I think I probably am, but that doesn't mean I think it's the best theory. I see Wittgenstein's language game approach as the best explanation of language, including proper names. But despite its faults (which I think are different from those that Kripke claims) I think there's a lot of valuable insight in Russell's theory of descriptions, and I am unable to find any such value in Kripke's theory.

The purpose of a definite description is to uniquely pick up an individual, not just to pick it up under a description that it will never (and could never) cease to satisfy. — Pierre-Normand

Then definite descriptions do not always take account of elemental constituents.

Are what you call "elemental constituents" something akin to essential properties? In that case, the item referred to could not persist though the loss of those properties, but they may still not guarantee that the item is uniquely being described by them since other items of the same essential kind also would have those properties. — Pierre-Normand

This misses the point. Indeed, all of those particular items cannot exist without their elemental constituents.

We can state otherwise.

One has to be careful how one sets up counterfactuals, because they usually end up being nonsense, no matter what metaphysics or language philosophy one favours. — andrewk

This is really not a good sign. Counterfactuals are ordinary tools of reasoning, and in many cases their meanings, and even truth conditions, aren't difficult to figure out. I am suspicious of error-theoretic or revisionary accounts of them, as I think Kripke is right to be in exploiting modal intuitions.

The idea that when we suppose something were some other way, we are not really doing that but positing some 'counterpart' to it, is otiose and makes understanding the semantics of referential expressions difficult for no payoff.

I see Wittgenstein's language game approach as the best explanation of language, including proper names. — andrewk

Could you elaborate on this point?

I'm still pondering over how existential quantification works for counterfactuals.

Anyone?

Please forgive me if this has already been covered, but isn't insensitivity to change or counterfactual supposition precisely the behaviour we'd want out of our practices of referring? — fdrake

Yes.

It seems to me that the issue concerning the reidentification of a material particular (or substance) as being numerically the same at two moments in time, in spite of qualitative change, is orthogonal to the issue of the rigidity (or lack thereof) of the referring expressions that are being used to denote it in particular instances. In fdrake's example, the first part stipulates the existence of an individual (i.e. the apple) and assigns some properties to it. Thereafter, it seems to be assumed that this individual has conditions of persistence and individuation such that it can survive some qualitative changes while remaining numerically the same individual. The several occurrences of "it" all pick up the same individual just in virtue of them being used to refer anaphorically to whatever the first singular expression (i.e. the anaphoric antecedent "an apple...") was referring to. That would be true, it seems to me, regardless of the individuation conditions for apples, and regardless of the rigidity of the anaphoric antecedent. — Pierre-Normand

When we "re-identify a material particular as being numerically the same at two moments in time" are we not counting its various (changing) temporal manifestations as all being related by identity precisely insofar as they are all manifestations of the one differentiable ( both within itself and from all other processes) process?

Some features of the decaying apple will probably be identifiable across successive stages, but the comparison of the beginning stage with the end stage may yield no identifiable features in common other than a more or less similarity of size of shape, say.

We have the same name 'that apple' for the entity as it is at any particular time and also for the total transformation which is the life history of the entity. The entity at it most complete, and thus its completed identity, would seem to consist in its entire history. If we say that the exhaustive history of an entity is a set of descriptions each more or less definite (or all equally because completely definite if precise specifications of size, shape time and location are included) then any 'remainder', what we might tend to think of as the entity itself apart from all of its descriptions (relations) and transformations, would seem to be purely formal, really just an artifact of language.

And, if that is indeed so, you think it has been a good thing? — Janus

Yes.

The idea that when we suppose something were some other way, we are not really doing that but positing some 'counterpart' to it..... — Snakes Alive

'counterpart'

Quote signs play the same role as italics in word games like this. People reach for them when they are unable to express what they mean but want to believe and imply that the meaning is clear and obvious.

You appear to be trying to say that my interpretation of how people use counterfactuals is artificial and unnatural. Naturally I disagree. I have the same negative feelings about Kripke's interpretation as you have about mine. I feel that my approach is the most natural in the world. Although nobody can put themselves in another's head, my confident guess is that that's what counterfactual-using non-philosophers would say if they could be persuaded to spend half an hour discussing it.

If we can't get beyond italics and quotes to agree on some concrete definitions, this discussion will remain one about what 'feels natural' to different individuals. In my opinion, there's nothing wrong with that. It only becomes silly if people start to insist 'No, my way of looking at it is right and yours is wrong!'

Of course, it's always good to question things; but it doesn't necessarily follow that what has been thrown into question thereby becomes redundant. If the questioning leads to folk simplistically thinking that what has been questioned has become redundant, then I can't see how that could be a good thing. So, to return to the case under consideration, if people oversimplify the issues and come to think that Kripke has shown that descriptions have no part to play in establishing reference, then that would not seem to be a good thing.

Although nobody can put themselves in another's head, my confident guess is that that's what counterfactual-using non-philosophers would say if they could be persuaded to spend half an hour discussing it. — andrewk

Or less. Obviously the Nixon who lost isn't our Nixon. That's the whole point of supposing.

Wittgenstein's approach to language is laid out in his later work 'Philosophical Investigations'. It's a bit like reading James Joyce - difficult and open to differing interpretations. I have to confess that I failed to understand most of it. But dazzling insights still shine through here and there. The dazzling insights that I see as relevant here are

1. language is not about reference and meaning but about what the speaker is trying to achieve with her speech act

2. many (possibly most) speech acts must be considered as a whole in order to discover their intention. Dividing them up into tiny bits and asking questions about reference and meaning of little components is often a hindrance to understanding.

Yes, it is true at all times that Gödel was born on April 28, 1906, for instance. — Pierre-Normand

That is correct. And yet he might have been born on April 27.

Also, I'm unsettled when folk treat time as a special case. It's also true that he was born in Brno, and yet he might have been born down the road in Vienna.

This is not to object to what you have said, but to widen its breadth.

This is not to object to what you have said, but to widen its breadth. — Banno

Yes. That was a follow up on a tangential line of inquiry that had been initiated by @fdrake. Of course, I agree that what is actually true at all times of Gödel (and hence might figure in a definite description of him) isn't necessarily true of him.