logicandlanguage.net

[I could not have spoken properly and plain in any language. No chance for future either, because its main cause is lack of understanding and not just lack of articulation. So, sorry!]

(a1) “And I think it's a bit like suggesting that languages cannot contain singular terms on the grounds that there are (or might be) no objects.”
(a2) “Even if there are no objects, it still seems to me that there are expressions which (to speak loosely) try to refer to objects; things like names, indexicals like `I', definite descriptions and the like. Maybe they fail sometimes. Maybe they fail always. But that's still what they're about.”

1- About (a)s:
Apart from the meaning of “object” and its vagueness, I can’t really understand what does “there are (or might be) no objects” mean. I can’t imagine such a state of affair. Do you? For the sake of the argument let’s suppose we know what does it mean, now suppose that there haven’t been any objects during our lives and we haven’t encounter to one, then do you still say that our language can contain singular terms (without any basic changes in term ‘singular term’). Is it conceivable and/or possible at all? This is important mainly cuz it seems to be a stricter sense of “there are (or might be) no objects”. In what sense of ‘try’ there could be expressions which would try to ‘refer’ to objects? I can’t see how “things like names, indexicals like `I', definite descriptions and the like” can make it. [I tend to think the issue has similarities with what Putnam’s “Brains in a Vat”, of course, about different matter.]

2- Do you believe in rigid designators? If yes, then how do you see its relation to natural kinds? Anything fundamental more than what Putnam says?

3- I read that section of Greg’s dissertation entitled “Tarski’s objections to analyticity”. It was interesting. Has anyone anything else about Tarski ‘and’ analyticity?

4- Gillian! I have recently started philosophy. Do you think it worth spending my time on it? Are you happy with it? I don’t know … Thank you very much in advance.

I think Frege's response is not satisfactory. If it were, then the problem of empty names would disappear. But there is such a problem. Hence, Frege's response is not satisfactory. More specifically, if the semantic content of a name is its referent, then if it has no referent then it has no content. Trying to refer seems not to be enough.

I also do not see why Frege bothers himself with the problem of empty terms. After all, according to his semantics, the content of a name is its sense, not its referent.

Ignorant (but straightforward) question: who are the people who claim that there are no natural kind terms? (And is that meant to apply to all languages, or all natural languages, or our (best) scientific language(s), or...?)

Also, I'm not 100% certain that there's a fact of the matter/ a well-posed question about whether a term that tries to pick out exactly one (possibly extralinguistic) object but fails is really a genuine name. Such a term is certainly an individual constant (a fact which can be cashed out syntactically/ grammatically), but is it a full-blown name (taking 'name' as having an ineliminably semantic dimension)? It seems to me that some people would say yes, others no. Should a so-called doctor who never improves a patient's condition really be considered a doctor? I think some people would say yes, others no>.

Here's a more technical reason someone might say a term that tries to pick out an individual but fails is a name: allowing non-referring individual constants into a language "messes up" (to use the technical term) classical first-order logic in the two following ways. (We thereby get a natural characterization of the difference between pseudo-names, which induce these aberrations, and genuine names, which do not.)
(1) The rule of universal instantiation is no longer valid (that is, we cannot infer 'b is an F' from 'All x are F' when b does not refer to any object in the domain of discourse.
(2) If we reject the rule of universal instantiation, and allow our language to have non-referring singular terms, then our resulting logic will no longer be extensional. Specifically, co-extensive predicates are not interchangable salva veritate (see Karel Lambert's "Predication and Extensionality" Journal of Philosophical Logic 1974 for details and explanation).

Thus, if we say our first-order logic is only intended to deal with genuine names, not pseudo-names, then we can maintain classical logic. If I had to choose between altering classical f.-o. logic and altering my notion of 'name' such that 'Pegasus' is not a genuine name, I'd choose the latter. However, many people (all the free logicians) believe that the rule of universal instantiation is suspect, so they would likely be happy choosing the former.

Gregory! Thank you for your explanation. It was helpful for me.

1- We don’t know what does “a state of the matter in which there are no objects at all” mean (in its extremity) (call it Total No-Object State). If this is the case, then we shouldn’t use it and talk about it, simply because we just ‘think’ we are talking about something. For us it is senseless, isn’t it? (To me, it is similar to total-radical skepticism. In other words, its notion is inconsistent.) We know what it means that a thing or set of things don’t/wouldn’t exist. We ‘can’ talk about an empty domain of discourse and it makes sense to us. But all of these cases are Partial No-Object States. Extending Partial cases in order to grasp an understanding of the Total one seems to me completely misleading (False Projection). We can extend them but just for other Partial cases (and perhaps bigger ones) … [Oh! I’m going insane! H e l p!]

2- I tend to maintain classical logic and semantic too, because I think it is vastly superior to the alternatives in simplicity, power, past success, and integration with theories in other domains. But if I should understand “name” and “object” in their natural sense, (whatever they are,) and our current issue about the relation between these two notions in a general way and not restricted to a restricted language or domain, then I would still have problems in understanding the meaning of them.

3- (A poem)
She told me ‘Come’
She told me ‘Stay’
She told me ‘Smile’
She told me ‘Die’.
I Went
I Stayed
I Smiled
I Died.

Thanks for your comments, everyone.

Frank:

1. "What does it mean to say that there are no objects?"

Reasonable question. The scenario Frege has in mind is that advocated by idealists, according to which all that exists are our ideas of objects. Sometimes people hallucinate objects which are not really there, right? Think of idealism as saying that every object which we seem to perceive is really just a hallicination.

2. "suppose that there haven’t been any objects during our lives and we haven’t encounter to one, then do you still say that our language can contain singular terms (without any basic changes in term ‘singular term’)."

Yes. And so does Frege.

3. "In what sense of ‘try’ there could be expressions which would try to ‘refer’ to objects? "

Well, I haven't said too much about that, but it's a good question. I'm not sure what the answer is. For Frege, of course, it means that the speaker intended to refer to an object. But - for the reasons I gave in the post - I am not very happy with that answer.

4. "Do you believe in rigid designators? "

yers.

4b. "If yes, then how do you see its relation to natural kinds?"

Following Soames in Beyond Rigidity I think the important similarity between names and natural kind terms is non-descriptiveness, not rigidity. The criterion of membership of an extension of a natural kind term need not be known to speakers and does not form part of the content of a natural kind term. Similarly, the referent of a name is not semantically determined by any description which enters into the content of a name, nor need the way the referent is determined be known to competent speakers.

5. "I have recently started philosophy. Do you think it worth spending my time on it?"

Yes. But philosophy of language is a hard place to start. What have you studied so far?

Erhan:

1. "I think Frege's response is not satisfactory. If it were, then the problem of empty names would disappear."

That depends on exactly what you think the problem of empty names is. If you're a Millian (i.e. you think the content of a name is its referent), then the lack of any objects to act as referents would mean that names have no content (the problem of empty names) - even though, following the thought in the post - they are still names because they intend (or were intended) to refer to objects. It seems to me that there are really just two different problems here, and solving the one about how there can be names without objects doesn't solve the one about how there can be name-contents without objects.

Hi Greg:

1. "Ignorant (but straightforward) question: who are the people who claim that there are no natural kind terms? "

I'm thinking of some friendly philosophers of biology and an eminent logician with whom I once got into a conversation with in New Zealand. All concerned will remain nameless, but if anyone knows if the argument is published somewhere, I'd be interested in a reference.

1b. "(And is that meant to apply to all languages, or all natural languages, or our (best) scientific language(s), or...?)"

All languages, I take it. It's a point about the possibility of natural kind terms, and I don't see anything in the argument which would restrict the conclusion to a certain kind of language.

2. "Also, I'm not 100% certain that there's a fact of the matter/ a well-posed question about whether a term that tries to pick out exactly one (possibly extralinguistic) object but fails is really a genuine name. Such a term is certainly an individual constant (a fact which can be cashed out syntactically/ grammatically), but is it a full-blown name (taking 'name' as having an ineliminably semantic dimension)? It seems to me that some people would say yes, others no. Should a so-called doctor who never improves a patient's condition really be considered a doctor? I think some people would say yes, others no>."

Sure. I don't want to fight over the right to use the expression "name". If you'd rather say "individual constant" that's fine with me. "Name" seems pretty natural to me though.

3. Here's a more technical reason someone might say a term that tries to pick out an individual but fails is a name: allowing non-referring individual constants into a language "messes up" (to use the technical term) classical first-order logic in the two following ways....
(1) The rule of universal instantiation is no longer valid ...
(2) If we reject the rule of universal instantiation, and allow our language to have non-referring singular terms, then our resulting logic will no longer be extensional. Specifically, co-extensive predicates are not interchangable salva veritate."

Greg, if your language is rich enough to contain non-referring names, then yes, classical logic won't cover it. But look, natural languages just ARE much richer than the language of first order logic. They do contain non-referring names and non-extensional contexts. I'm as much a friend of classical logic as the next man, but I'm not going to make out that "Pegasus" is a referring term, or that co-extensional expressions can always be exchanged salva veritate in belief contexts, on the grounds that doing otherwise would complicate my theories! My theories of natural languages should be as complicated as the phenomena demand.

Hi Gillian,

Thank you for replying to comments.

Here is what you said in the previous post:

"It seems to me that there are really just two different problems here, and solving the one about how there can be names without objects doesn't solve the one about how there can be name-contents without objects".

So, if I am not mistaken, you say that the first problem which is about how there can be names without objects can be solved by the notion of "intending to refer". From this view, since intending to refer is essential to all names, the problem of how there can be names without objects disappears.

However, the problem about how there can be name-contents without objects (the problem of empty names) cannot be solved by that notion of "intending to refer". And, I agree with you, the problem of empty names cannot be solved by that notion, and this is what I said in my previous post. However, this is the real problem which is really troublesome, and the first problem is not really a problem since it says nothing about the semantics of names.

Hi Gillian! Thank you for your reply.

> “Think of idealism as saying that every object which we seem to perceive is really just a hallucination.”

Well … I am quite familiar with idealism as a philosophical doctrine which says, roughly, that reality is somehow mind-correlative or mind-coordinated – that the real objects constituting the “external world” are not independent of cognizing minds, but exist only as in some way correlative to mental operations.

As I mentioned before, even if we suppose that there is a clear doctrine called Total-Radical Idealism, at best, it seems to me that it will suffer from self-inconsistency. (‘Total’ refers to the greatest scope of things we are justified to call them ‘Objects’. And ‘Radical’ refers to ‘the highest degree of existence’ which Idealism wants to deny from Objects.)

I don’t have a problem with Partial-Radical versions of Idealism (, if there are any), at least now. It is true that “sometimes people hallucinate objects which are not really there”, but when one wants to generalize and extend it to all objects, and says that “think of idealism as saying that ‘every object’ which we seem to perceive is really just a hallucination [and/or illusion]”, then I begin to doubt about the meaningfulness of this claim. [Maybe I am just an idiot! Who knows?!]

I see Kant’s famous “Refutation of Idealism” as a try to show this self-inconsistency; one may consider it as unsound though. (Kant argues that our conception of ourselves as mind-endowed beings presupposes material objects …)

Another example would be what Putnam says in “Brains in a Vat” which was about a sort of radical skeptical claim and again I know that it has some difficulties. It seems to me that we can use his argument against radical Idealism.

> “What have you studied so far?”

My formal background has been in engineering. But within the realms of Analytic philosophy I haven’t read much. It’s just seven months that I’ve started reading. Epistemology, Phil of Logic, Phil of Lang, and Metaphysics are my main interests. During past few months I’ve read some introductions, and I’ve worked on some subjects such as Analyticity, A priority, Skepticism, Vagueness, Ontology/Epistemology distinction … I’m working on Analyticity and Implicit Definition now.
In Phil of Lang, I’ve read W. Lycan’s introduction published by Routledge and some entries of Blackwell’s Companion to Phil of Lang … What should I read now? I’d be very grateful if you could guide me.

> [Maybe a childish question:] Everybody! Philosophy for me is not just a field of interest or a field that I just want to become a specialist in it or make a living out of it. I am seeking Happiness and Wisdom (in any sense). If I could not find it in analytic philosophy, or/and it couldn’t help me in this direction, I will give it up most probably, although I am really into it now (I know that most would tell me that my way of thinking about philosophy is old-fashioned.). My problem is that I cannot see much further of this road? Could you please describe what is going on ‘that far’?

Hi Erhan,

I've interspersed my replies (in bold) through your text below:

So, Erhan wrote:

Here is what you [gillian] said in the previous post:

"It seems to me that there are really just two different problems here, and solving the one about how there can be names without objects doesn't solve the one about how there can be name-contents without objects".

So, if I am not mistaken, [writes Erhan] you say that the first problem which is about how there can be names without objects can be solved by the notion of "intending to refer".

[right, that's what the post is about, though the "intending" is, as I say, metaphorical, so it still needs work. - gillian]

From this view, since intending to refer is essential to all names, the problem of how there can be names without objects disappears.

However, the problem about how there can be name-contents without objects (the problem of empty names) cannot be solved by that notion of "intending to refer". And, I agree with you, the problem of empty names cannot be solved by that notion, and this is what I said in my previous post.

[well, in truth, you wrote "Frege's response is not satisfactory. If it were, then the problem of empty names would disappear." - gillian]

However, this is the real problem which is really troublesome, [well, perhaps, but it's not what this weblog entry is about - gillian]
and the first problem is not really a problem since it says nothing about the semantics of names. [I don't see how that argument works. Lots of problems are problems though they say nothing about the semantics of names. And of course, the main problem addressed in the post concerns natural kind terms. The idea of the post is to use the solution to the singular term version of the problem in the natural kind case - gillian]

I'm not sure how close the analogy between singular terms and natural kind terms is, and so I'm not sure I'm convinced the Frege strategy will work here. Even if the relevant part of our referential apparatus misfires (just because the world doesn't hold up its end of the bargain), I agree that it seems plausible to hold that nonetheless our language contains genuine singular terms - terms which have the role of referring to objects. It's just these terms wouldn't be able to discharge this role.

But things seem likely to be quite different if there are no natural kinds. As Wright points out in 'The Conceivability of Naturalism', if our best theory of the world told us that there was no underlying chemical structure in common between all the sample of the liquid that plays the water role on Earth, it seems plausible that we would regard our concept water as (what Wright calls) criterially-governed; that is, the canonical way of identifying whether a sample was water would be whether it met some appropriate cluster of descriptions.

Now say we discover that there are no natural kinds (in the broader sense you invoke). Would we say that our terms like water and (let's say) chair are terms whose role is to pick out natural kinds, and that unfortunately those terms hadn't been able to discharge that role all this time? I think Wright's suggestion seems the more natural thing to say. If that's right, then Frege's solution in the singular terms case doesn't look like it should apply here.