It is perhaps true that any particular definite description of Donald Trump could turn out to be false. But all of them could not be (given that the character we know as Donald Trump is not a CGI, as per the example I gave earlier). — Janus

See the Thales example.

Hell, even the sentence "everything we know about Trump is false" is about Trump...

The other probelm is that if just start talking about somenone named Donald Trump outside of any pre-established context you might assume that i am speaking about the Donald Trump who is best known as the current POTUS, whereas as I might be talking about another Donald Trump who was born a woman but underwent a sex-change. — Janus

And in that case I would have misunderstood you. But that does not mean that "Trump" could not refer to Trump. Insofar as you have moved from description to context, we are not far from the same page. But what does not follow is that the context must involve a definite description.

Another example. Someone with no knowledge of US politics overhears a conversation about Trump. They ask "Who is this 'Trump' person?". Who is their question about, if not Trump. And this, despite their not having an available definite description.

But he did exist. — Banno

He may not have existed. In any case, if he did exist, and if everything else we know about him is false, then we have no idea who we are referring to. Just to say 'we refer to Thales' is nothing more than an empty tautology in that case, as far as I can tell. Also, would you not need to know at least approximately when and where he was born to make a purported reference to a purported Thales valid?

Hell, even the sentence "everything we know about Trump is false" is about Trump... — Banno

But it's not about a man named 'Trump' if such a man never existed.

You've let "know" creep in here. What's it doing? If you set up a counterfactual scenario involving a man named 'Donald Trump', then it is about the Donald Trump to whom you refer. That I don't know him does not change that. — Banno

It seems to me that at the very least I must know who I refer to, or else it is meaningless to say that I refer to someone. Would it be possible for me to know who I refer to, even if no one else does?

That's not difficult. There was a chap named Thales, who people told lies about. And this is a story about Thales, despite our not having definite description of him. — Banno

Thales could be a fictional character in that case. "There was a chap names Thales" is a description, and a definite one if there were no other men named Thales; I'm surprised you apparently can't see that.

Every description could not be false; it must mean at minimum that there was a man who was named Thales about whom many stories abound but nothing is known other than that he lived in some more or less definite area as some more or less definite time. — Janus


...and that contradicts your theory, because it is about Thales, and yet we have no definite description of him. — Banno

But it doesn't contradict what I have been saying, because that there was a man named Thales about whom many stories abound, and who lived in Ancient Greece and was a philosopher is a definite description that refers rigidly if there was no other man named Thales etc. If there was another then how would we know who we refer to if not by means of some other distinguishing description? And it's always the case that it is possible that historical figures are mythological; and that when we speak about Thales we are not referring to any actual previously existing person.

But crucially, not a definite description. It does not single him out, at least not without the circularity of "Thales" is the man named Thales. — Banno

As I've said repeatedly, it's not as if any one description can infallibly single out an individual, unless we posit that the person was the only one to whom the description applies, but then any such position cannot be infallible. Usually it's a constellation of descriptions, which are more or less definite depending on how precisely the time and locations involved in the description are specified.

If even those descriptions were false, then it could not mean anything to say that Thales had actually existed. — Janus


Again, you are asserting this without argument.

It seems, piecing it together, that you want to assert that if we know nothing about Thales, then we have no reason to think that he exists. But of course, we have no reason to think that he did not exist, despite our not being correct about anything we know about him

Same goes for Job, Noah, Jonah, and so on. That what we think we know about them might be false, simple does not imply that they do not exist.

And in the end, if you continue to insist that it does, you are just wrong. — Banno

If nothing we think about Thales is true, then what could it mean to say that Thales existed? In the case that nothing we think is true is true, and there was no man named Thales who lived at the time and in the place we think there was a man named Thales, because that description is false, who is it that we could be saying existed if even that most minimal description were false?

So. I'm not simply "insisting" I'm giving reasons for why I think it would be meaningless to say that a man existed about whom everything we think we know is wrong, even that he lived in a certain place and time and was called Thales. If you want to counter that assertion then you need to tell us why it would be meaningful to say such a man existed.

How do you know which Donald Trump I refer to in each case? I say you know because each implies a description. — Janus

Perhaps. But not all descriptions are definite descriptions. Context will be sufficient to differentiate the two without any definite description.

One of the issues hereabouts is a failure to differentiate clearly between descriptions and definite descriptions.

Yeah, I don't know what else to say, so I'll take a stab at showing you where our positions diverge. It is worth noting that that divergence is largely as a result of my participation here.

Definite descriptions are not a necessary part of successful reference within possible world scenarios. Definite descriptions are not a necessary part of successful reference in the actual world.

The examples of successful reference by false description are too numerous to deny. Those are actual examples. So, the above two statements follow from actual cases. That's worth mention as well.

So, if one holds that definite description is necessary for all reference, then one has an inherent problem taking account of actual examples of successful reference that do not include definite description. That's where you find yourself, and it's where I found myself as well... struggling as a result of attempting to use an inherently inadequate framework. It quite simply is found to be sorely lacking in explanatory power.

The problem is resolved by virtue of making the framework amenable to the facts(the actual cases of successful description by false definition).

Well done Banno. You proved my argument better than I could have.

You are right to point out the conflation of true/false descriptions. Definite ones must be true, if I understand correctly.

I don't even know what it could mean to say that the meaning of a name is given by an associated description. Names have references, not meanings. The references of names are determined by descriptions as a I showed in the examples above. — Janus

This.

You think that names do not have meanings, but that the reference is determined by a description.

Then that description gives the meaning of the name, doesn't it? That's the view of Russel, Searle and so on. Your view here is quite unclear.

Which example? There have been so many.

Perhaps. But not all descriptions are definite descriptions. Context will be sufficient to differentiate the two without any definite description.

One of the issues hereabouts is a failure to differentiate clearly between descriptions and definite descriptions. — Banno

As I've said from the start I don't think that an infallible distinction can be made between indefinite descriptions and definite descriptions, and my argument has been that descriptions are more or less definite depending on how precisely they are specified.

To say that Trump was POTUS is not a definite description. To say that he was POTUS at some specific time is a definite description and a true one if no other person was also president at that time. To say that he was called Trump and was president is also a true definite description if no other man named Trump has ever been president.

Anyway, I have provided this in good faith. I'm happy to be corrected if I have misunderstood something, but please don't just keep saying that I have not understood, and directing me to read Kripke again. Instead explain in you own words just where you think I am going wrong. If you can't or won't do that then that is basically the end of the discussion. — Janus

Here's the trouble. An hour spent responding to you. And now there are four more replies to deal with.

Philosophy is detailed. Seurat did not pain with a roller.

This.

You think that names do not have meanings, but that the reference is determined by a description.

Then that description gives the meaning of the name, doesn't it? That's the view of Russel, Searle and so on. Your view here is quite unclear.

Which example? There have been so many. — Banno

No, I don't believe that a description that determines the referent of a name "gives the meaning of the name" because there could be countless other descriptions that also determine the referent of that name.

So, I haven't been claiming that any description uniquely determines the referent of a name, although of course it could determine the unique referent of the name, which is a different thing.

In any case, if he did exist, and if everything else we know about him is false, then we have no idea who we are referring to. — Janus

Well, no, we are clearly referring to Thales. Who is it, about whom we know nothing? Thales.

Now I find that utterly convincing.

Edit: for the sake of keeping the discussion on track, let's follow this argument in more detail. What do you think?

Here's the trouble. An hour spent responding to you. And now there are four more replies to deal with.

Philosophy is detailed. Seurat did not pain with a roller. — Banno

What replies are you referring to? All the replies I have made are dealing with the same basic issue; if I have repeated myself it is only for emphasis; there is no need for you to reply to repeated points is you have addressed the point once. That would be the proper pointillist technique, each point should convey some new information, if it is to warrant a response.

But you are under no obligation to respond. If you don't find it interesting or you can't be bothered, then I'm happy to let it go. But I will not admit that you have adequately responded if I don't believe you have.

No, I odn't believe that a description that determines the referent of name "give the meaning of the name" because there could be countless other descriptions that also determine the referent of a name. So, I haven't been claiming that any description uniquely determines the referent of a name, although of course it could determine the unique referent of the name, which is a different thing. — Janus

The meaning of your paragraph remains oddly opaque.

Which Thales are we referring to then, if nothing we think we know about him is true? Or if there were a hundred Thales living in Greece at the time; which one are you referring to then? I'll find it "utterly convincing" when you can tell me which one without using any descriptions.

There were some typos and infelicities of expression. I've tidied it up. Here it is again:

No, I don't believe that a description that determines the referent of a name "gives the meaning of the name" because there could be countless other descriptions that also determine the referent of that name.

So, I haven't been claiming that any description uniquely determines the referent of a name, although of course it could determine the unique referent of the name, which is a different thing.

If there is anything in there you don't understand all you need to do is ask.

The very same one as you. The one these threads are about. That's all that is needed.

Well, even those are descriptions. And of course we are talking about a man called Thales, who is reputed to have lived in Miletus around 600 BC, and to have been a philosopher. Now, if there was such a man as fits that description then we are talking about an actual person, but if there was no man who fits that description then we are talking about a fictional character.

This seems unproblematic and perfectly in accordance with ordinary usage, as far as I can tell.

You are right to point out the conflation of true/false descriptions. Definite ones must be true, if I understand correctly. — creativesoul

I see no reason why any description, whether definite or not, could not be true or false.

I thought it was by definition that definite descriptions must be true.

There's a bit of conflation between what we're doing when referring to Thales by false description, and what is going on when we take account of that.

Janus seems focused upon what's necessary for us to take an account of successfully referring to Thales by false description, whereas Banno is just pointing to the fact that we're referring to Thales by false description.

Two different targets.

Well, even those are descriptions. — Janus

But not definite descriptions. They do not serve to single out one individual. But "Thales" might.

Hence the importance of differentiating a definite description from just any description.

So, I haven't been claiming that any description uniquely determines the referent of a name, although of course it could determine the unique referent of the name, which is a different thing. — Janus

So it might be worth your setting this out in more detail.

I can't see why. I can say something false about someone that could not apply to anyone else. Of course there must be something true in the description, but it need not be wholly true.

I mean the point is not that definite descriptions must be true, but that they are definite if they don't apply to anyone else.

So, as an example, if I said 'Donald Trump's penis is the smallest in the world', that may or may not be a true definite description, but it could not apply to anyone else (not named Donald Trump).

...the point is not that definite descriptions must be true, but that they are definite if they don't apply to anyone else... — Janus

Then what grounds your objections to the actual cases of successful reference by false description?

I don't see any real distinction. I don't think Banno admits that we refer to Thales by description, whether true or false, definite or indefinite, but claims that we refer to Thales merely by name. Of course, I admit that in a purely formal sense this is true, just as it is with fictional characters. But I see this formal truth as a trivial truth; it doesn't tell us anything interesting about our practices.

You missed the part where I said that here must be something true in the description.

Then that's not a case I'm talking about. There are numerous ones where there is nothing true in the description and yet successful reference happens anyway.

Gotta run. Catch you chaps later...

I think we are talking about two different things. I have said that a definite description does not need to be wholly true to refer uniquely to someone. I think the question about whether successful reference is possible by false description is related, and as I have already said, I am not convinced that it is possible without the person referring knowing something true about the person being referred to, even if that is nothing more than knowing what they look like

The very same one as you. The one these threads are about. That's all that is needed. — Banno

But not definite descriptions. They do not serve to single out one individual. But "Thales" might. — Banno

But they are definite descriptions if they determine which Thales we are referring to, which you seemed to be claiming that they do, when you used them to specify which Thales we have been referring to.

I have been thinking that it is not really the name 'Thales' which is a rigid designator, but rather something like 'that man'. Which man? 'The man called Thales who.....'

Interestingly, 'The man called Thales' is a description but not a definite one, until you specify exactly when and where he was first dubbed 'Thales', and/or add other identifying descriptions about him and his doings.

What if Donald Trump had not been POTUS?

What if Donald Trump had been born a woman? — Janus

I find that examples like these highlight the inadequacy of trying to analyse parts of speech rather than entire speech acts, and that approaching it correctly (according to me) provides both a defence and a criticism of Kripke's approach. A defence, because some of his examples of how he sees his theory working, that appear nonsensical to many, can be made sensible when analysed as part of a complete speech act, and a criticism because the examples he uses to attempt to demonstrate the inadequacy of descriptivism also rely on analysing a reference out of the context of the speech act in which it occurs.

Consider the first of the two above counterfactuals. We can remove all ambiguity by placing the question in the context of a speech act. For instance:

'I wonder what Donald Trump would have done if he had lost the 2016 presidential election. Would he have tried to incite civil unrest, claiming electoral fraud? Would he have gracefully departed the political scene? Would he have turned back to his business activities with renewed energy or would he have retired to live a life of seclusion and contemplation?

It is clear from the context provided by the full speech act that we are talking about the person that was the Republican candidate in the 2016 presidential election. The definite description 'the person that was the Republican candidate in the 2016 presidential election' picks out a single person in this world and is then used to contemplate possible worlds that split from this at some point between the Republican nomination of Trump in mid-2016 and the declaration of the election result in late 2016.

For the second one, a containing speech act might be:

'I wonder whether, if Donald Trump's parents' fourth child had been a girl, they would have continued to have children in the hope of having a second boy.'

Here, a definite description picks out the parents as those that are, in this world, the parents of Donald Trump. It then considers possible worlds that split from this at some time between the birth of that couple's third child and the naming of their fourth child. There is no doubt about which couple we are referring to, and the subject of the 'had been born' part is the fourth child of that couple.

I am confident that, if we reject analysis of references out of the context of a speech act as invalid, much of the metaphysical gymnastics that goes on in philosophy of language is shown to be just an entertaining diversion involving moving words around into new configurations.

The basic principle in such counterfactuals is to use the reference part of the speech act to pick a unique object in this world (the 2016 Republican candidate and Donald Trump's parents, in the above two examples) and then to consider possible worlds that split from this at some time after the entry of that object into that world. We then evaluate the question in relation to that object, or items that relate to it.

I am confident that this process can even work for the 'If Nixon were a golf ball' example but I'll leave it up to someone else to come up with a complete speech act in which that counterfactual is contemplated. If the speech act is complete enough to be understood by an ordinary person, it will point to a path for uniquely identifying the object it is concerned with.

PS Unless I've missed a post, the above discussion of references to Thales gets nowhere because it does not discuss speech acts that contain references to Thales. Discussion of references in the absence of their containing speech act cannot lead to any useful and valid conclusion. I am (fairly) confident that, if someone were to put forward an example complete speech act containing such a reference, it would be easily resolved in any of the situations discussed (1. Thales existed and did what we are told he did. 2. Thales existed but did only some of those things. 3. Thales existed but did none of those things. 4. There was no Thales).