Russell On Syntax of Sentences Using the word 'Existence': Philosophy Forums

Philosophy Forums » Philosophy of Language » Russell On Syntax of Sentences Using the word 'Existence' » Russell On Syntax of Sentences Using the word 'Existence'

Russell On Syntax of Sentences Using the word 'Existence'
Russell says the sentence 'Bob exists' is bad syntax.

Russell On Syntax of Sentences Using the word 'Existence'

park
Initiate

Usergroup: Members
Joined: Feb 25, 2007
Total Topics: 1
Total Posts: 2

Posted 02/25/07 - 04:14 PM: Subject: Russell On Syntax of Sentences Using the word 'Existence'	#1
What does Russell mean when he says that we can only apply the verb 'exists' to a description as in 'The author of Waverly exists' but to say that 'Scott exists' is bad syntax? (It is true that Scott is the author of Waverly.) this is important to him (Russell) because he goes on to say that this clears up more than two millenia of muddle-headedness about 'existence' beginning with Plato's Theaetetus. I'll be very greatful to anyone clearing this up for me. Thanks

Thoughtless
Professor

Usergroup: Members
Joined: Jul 12, 2005
Total Topics: 24
Total Posts: 636

Posted 02/27/07 - 09:41 AM:	#2
I haven't read Russell, but I'll take a stab at it. The sentence 'Scott exists' is not a statement about reality, because we don't know who or what Scott is--the term is undefined--so there is no way of knowing whether or not Scott exists. The statement, 'The author of Waverly exists', however, is a description of a particular hypothetical person: the person who wrote Waverly. Therefore, the statement that this particular person exists is truth apt. I don't know if that's what he meant or not, but that's my interpretation based on your representation of his position.
You down with OPP(Original Poster's Prerogative)?

Ron Harvey
banned

Usergroup: Members
Joined: Jun 22, 2006
Total Topics: 1
Total Posts: 272

Posted 02/27/07 - 10:43 AM:	#3
This is explained here: http://www.philosophy.ru/edu/ref/sci/russel.html The old loony seems to have been of the opinion that exists should mean is because. In order to exist some sort of description has to be attributable, without which we don't know whether or not anything does exist, so the expression would thus be meaningless.

Timothy
The Author of Waverly

Usergroup: Members
Joined: Dec 18, 2004
Location: Unemployment line
Total Topics: 74
Total Posts: 1935

Posted 02/27/07 - 12:17 PM:	#4
From predicate logic: " x exists..." is not a first order predicate, but a quantifier (Ex) that ranges over propositional functions such as Px, where x is a variable that can take the value of any constant and where P is a first-order predicate such as "____ is tall". "Scott" would be a value of a variable. Let's call it s. But (Es) is not a well formed formula; neither is (Ex)s. Since it is not well formed, it has incorrect syntax. Add a predicate or description, such as "The author of Waverly" (let's call it A), and you will have a well formed formula: (Ex)Ax (there's an x such that x is the author of Waverly) or the usual Ac (Scott is the author of Waverly).
""Physics investigates the essential nature of the world, and biology describes a local bump. Psychology, human psychology, describes a bump on the bump." W.V.O. Quine

park
Initiate

Usergroup: Members
Joined: Feb 25, 2007
Total Topics: 1
Total Posts: 2

Posted 02/28/07 - 04:06 PM:	#5
Thanks to everybody who responded. It cleared up a confusion for me.

teleplasm
Aspirant

Usergroup: Members
Joined: Mar 03, 2007
Total Topics: 1
Total Posts: 40

Posted 03/19/07 - 11:18 AM: Subject: Is this....	#6
.....wellformed?: (Ex)(x = Scott) If so, it means exactly what "Scott exists" means in everyday language.

Timothy
The Author of Waverly

Usergroup: Members
Joined: Dec 18, 2004
Location: Unemployment line
Total Topics: 74
Total Posts: 1935

Posted 03/19/07 - 11:39 AM:	#7
teleplasm wrote: Is this wellformed? (Ex)(x = Scott) If so, it means exactly what "Scott exists" means in everyday language. It is. "=" is a two place predicate, namely this: ___ is equal to ___ So your remark would mean: "There's an x such that is equal to something, namely Scott". In this case, the quantifier does range across a predicate, thus being a well formed formula. But I'm not so sure that it is what it means in everyday language. When one says "Scott exists", one is implying that there's something equal to Scott, namely Scott? Or that ther'es a predicate that is true of Scott? The paradox of analysis regarding ideal languages seems to say something along these lines: if THAT'S what you meant by "Scott exists", then why did you said it in such a strange way? For example, when you're talking about a broom, Are you talking about the stick and the brush, and their connection? If the analysis makes the meaning more complex or even changes it completely, what do you gain by doing it?
""Physics investigates the essential nature of the world, and biology describes a local bump. Psychology, human psychology, describes a bump on the bump." W.V.O. Quine

teleplasm
Aspirant

Usergroup: Members
Joined: Mar 03, 2007
Total Topics: 1
Total Posts: 40

Posted 03/20/07 - 03:37 AM: Subject: What was Russell claiming?	#8
Your post wandered into other matters, which it would not be appropriate to deal with here. But I can't see how one of "Scott exists" and "(Ex)(x = Scott) could be true and the other false. I'd like to make some observations on the original message. Could Russell have been claiming that "Scott exists" is meaningless? This is very implausible: we all know what it means. And if he was trying to argue that though not meaningless it can't be expressed in formal logical terms, then so much the worse for formal logical terms. But as I've tried to show, it can be expressed in the calculus of predicates. It's strange that such a great logician as Russell couldn't see this. Perhaps he was so carried away by the thrill of clearing away two millennia of muddleheadedness about existence that he didn't think it through.

Timothy
The Author of Waverly

Usergroup: Members
Joined: Dec 18, 2004
Location: Unemployment line
Total Topics: 74
Total Posts: 1935

Posted 03/20/07 - 07:29 AM:	#9
teleplasm wrote: Your post wandered into other matters, which it would not be appropriate to deal with here. Did it? Before you asked "Is this a well formed formula?". I replied: "Yes". You also said "If so, then this and this follows". I directed that remark and tried to stablish "Not neccesarely". And now you say I'm wandering around? In any case, that specific matter, wheter (Ex)(x = Scott) means exactly the same as "Scott exists" in ordinary language, is quite relevant to this thread. teleplasm wrote: But I can't see how one of "Scott exists" and "(Ex)(x = Scott) could be true and the other false. And who said such a thing? teleplasm wrote: Could Russell have been claiming that "Scott exists" is meaningless? This is very implausible: we all know what it means. I think he did claimed that. The fatc that we understand it nonetheless could be explained by saying, for example, that when we hear "Scott exists", we assume that there's a true first-order predicate of a constant named "Scott". Russell's point, I think, is that "Scott exists" is bad syntax when no predicate is bounded to the constant "Scott". Hence the literal translation into Predicate Logic yields a formula with bad syntax, thus meaningless: (Es) where s: Scott. A paraphrasis would be needed to avoid the bad syntax. Hence (Ex)(x=s). teleplasm wrote: And if he was trying to argue that though not meaningless it can't be expressed in formal logical terms, then so much the worse for formal logical terms. It can be expressed in logical terms. It just needs a paraphrasis. teleplasm wrote: But as I've tried to show, it can be expressed in the calculus of predicates. It's strange that such a great logician as Russell couldn't see this. He could see that. See Principia Mathematica, vol. 1, page 176. You can check it online from http://www.hti.umich.edu/cgi/b/bib/bibperm?q1=AAT... Here's a cite: By Russell & Whitehead: If x and y are identical, any property of x is a property of y On the paraphrasis you proposed: (Ex)(x=Scott) would mean that any true predicate of Scott is also true of x. Moreover, that x=Scott iff any property of x is a property of Scott (from elementary considerations of membership in set theory, two sets are equal iff they have the same memebers). Still, predicates are required to stablish the existence of a variable that is equal to a constant. What the remark "Scott exists" seems to prima facie suggest is that existence is in itself a predicate instead of a quantifier that bounds a variable over (at least) a true predicate. If this is what one suggest when uttering "Scott exists", then one is using the logical syntax poorly. If that's not what one suggests, but rather something like (Ex)(x=s), then one is using correctly syntax. The trick consists on understanding that (Ex)(x=s) is a paraphrasis of "Scott exists", not a literal translation into the logical language. The theory of descriptions endorsed by Russell on his On Denoting is chiefly a theory of paraphrasis.
""Physics investigates the essential nature of the world, and biology describes a local bump. Psychology, human psychology, describes a bump on the bump." W.V.O. Quine

teleplasm
Aspirant

Usergroup: Members
Joined: Mar 03, 2007
Total Topics: 1
Total Posts: 40

Posted 03/21/07 - 05:46 AM: Subject: The possibles	#10
Existence can be a predicate bound by a variable if the domain of x is extended to the possibles. Some possible things exist and others do not exist. This of course provides a referent for "the Golden Mountain" in the classic problem sentence "The Golden Mountain does not exist". Apart from the Neo-Scholastics, the only modern philosophers who have studied the possibles seem to be the followers of Meinong.

Timothy
The Author of Waverly

Usergroup: Members
Joined: Dec 18, 2004
Location: Unemployment line
Total Topics: 74
Total Posts: 1935

Posted 03/21/07 - 09:29 AM:	#11
teleplasm wrote: Existence can be a predicate bound by a variable if the domain of x is extended to the possibles. I dont' quite follow this. Let's asusme that our Domain is "everything". Let this include the "unactualized possibles" that you seem to be thinking on. Even in this case, an x from the domain would be said to exist or not to exist in virtue of an object-language true predicate. Or maybe we could just implement the modal notation of <> for possibility and [] for neccesity, along with Kripke's S5 system of modal logic. Then one could say that "some x possibly exists" with this: <>(Ex)Phi(x) where Phi is a n-place predicate and x ranges across "everything" Likewise, you could say "it's possible that an x does not exist" with <>~(Ex)Phi(x) teleplasm wrote: This of course provides a referent for "the Golden Mountain" in the classic problem sentence "The Golden Mountain does not exist". Ahh I see what you tried to do. You're providing the swollen-ontology solution to the old Platonic problem of the non-being. But why accept this solution when we can have another one that doens't commit us to talk about "possible beings", thus swelling our ontology and avoiding Occam's Razor? Russell's theory of Descriptions advanced on his On Denoting (I highly suggest you check it out here, because I think you haven't read it yet) provides us precisely with a more sophisticated solution were we don't have to embrace infinite worlds of unactualized possibles. Then the problem of "The golden Mountain that does not exist" is solved without accepting a bizarre, potential existence of a Godlen Mountain. The sense and meaning of such a phrase is thus saved without a swollen ontology. teleplasm wrote: Apart from the Neo-Scholastics, the only modern philosophers who have studied the possibles seem to be the followers of Meinong. I disagree. Leibniz studied it as well as David Lewis, Quine, and Kripke, just to mention the most famous ones (I'm unsure if Pascal did too). Check this thread to find some bibliography on the subject.
""Physics investigates the essential nature of the world, and biology describes a local bump. Psychology, human psychology, describes a bump on the bump." W.V.O. Quine

teleplasm
Aspirant

Usergroup: Members
Joined: Mar 03, 2007
Total Topics: 1
Total Posts: 40

Posted 03/21/07 - 11:27 AM:	#12
Timothy wrote: I dont' quite follow this. Let's asusme that our Domain is "everything". Let this include the "unactualized possibles" that you seem to be thinking on. Even in this case, an x from the domain would be said to exist or not to exist in virtue of an object-language true predicate. So what? The predicate in this case is "exists". And it can be quantified in this sort of way: (Sx)(ex) Where S is the quantifier "subsists" (I'm using Meinong's terminology here), and e is the predicate "exists". Timothy wrote: Or maybe we could just implement the modal notation of <> for possibility and [] for neccesity, along with Kripke's S5 system of modal logic. Then one could say that "some x possibly exists" with this: <>(Ex)Phi(x) where Phi is a n-place predicate and x ranges across "everything" Likewise, you could say "it's possible that an x does not exist" with <>~(Ex)Phi(x) This obviously doesn't serve your purpose, because "everything" would have to include the possibles, and give rise to what you call a "swollen ontology". Also, of course, Ex reads as nonsense in the case of merely possible objects. Sticking a modal operator in front of it doesn't change that. Timothy wrote: Ahh I see what you tried to do. You're providing the swollen-ontology solution to the old Platonic problem of the non-being. But why accept this solution when we can have another one that doens't commit us to talk about "possible beings", thus swelling our ontology and avoiding Occam's Razor? See my comments above. It's significant that opponents have fallen back to rhetorical phrases such as "swollen ontology", and appeals to Occam's Razor. Timothy wrote: Russell's theory of Descriptions advanced on his On Denoting (I highly suggest you check it out here, because I think you haven't read it yet) provides us precisely with a more sophisticated solution were we don't have to embrace infinite worlds of unactualized possibles. Then the problem of "The golden Mountain that does not exist" is solved without accepting a bizarre, potential existence of a Godlen Mountain. The sense and meaning of such a phrase is thus saved without a swollen ontology. Again, see my comments above. Timothy wrote: I disagree. Leibniz studied it as well as David Lewis, Quine, and Kripke, just to mention the most famous ones (I'm unsure if Pascal did too). Check this thread to find some bibliography on the subject. I'm delighted that this is the case!

Timothy
The Author of Waverly

Usergroup: Members
Joined: Dec 18, 2004
Location: Unemployment line
Total Topics: 74
Total Posts: 1935

Posted 03/22/07 - 09:34 AM:	#13
teleplasm wrote: So what? The predicate in this case is "exists". And it can be quantified in this sort of way: (Sx)(ex) Where S is the quantifier "subsists" (I'm using Meinong's terminology here), and e is the predicate "exists". I see then. I confess I know nothing of Meinong, so I'll grant you your point without any complain. But here we're deviating from Russell's idea of why "Bob existst" is bad syntax (it's a non-russellian approach) but interesting to this subject anyways in the sense that provides another account of the same issue but different from Russell's. teleplasm wrote: This obviously doesn't serve your purpose, because "everything" would have to include the possibles, and give rise to what you call a "swollen ontology". "Everything" could be understood as "all things that are equal to themselves", and in russellian analysis something must exist (be an actualized possibility, to put it in the language of possibles, even thou I don't think Russell said it as such) in order for it to be equal to itself. But then we may just adopt Meinong to give a different account; fair enough (I can't fruitfully dwell into that discussion given my ignorance).
""Physics investigates the essential nature of the world, and biology describes a local bump. Psychology, human psychology, describes a bump on the bump." W.V.O. Quine

Andreios
Student

Usergroup: Members
Joined: Sep 23, 2005
Total Topics: 1
Total Posts: 66

Posted 04/12/07 - 12:44 AM:	#14
Park, Maybe I'm beating a dead horse here, but I'll try to answer your question. In The Philosophy of Logical Atomism Lecture V, Russell explains, "When you take any propositional function and assert of it that it is possible, that it is sometimes true, that gives you the fundamental meaning of 'existence'... Existence is essentially a property of a propositional function. It means that that propositional function is true in at least one instance" (my italics). A propositional function is an open sentence like "x is red." For Russell, to say that red things exist is simply to say: the propositional function 'x is red' has at least one instance. In formal logic, this is expressed by prefacing the sentence with an existential quantifier '(Ex)'. Beware , however! Russell does not think '(Ex)' reads "There is an x..." He writes, "When I say, e.g. 'there is x such that x is a man', that is not the sort of phrase one would like to use. 'There is an x' is meaningless. What is 'an x' anyhow? There is not such a thing." This might come as a surprise, but it illustrates his point. '(Ex)' is a predicate that applies to propositional functions, not to names. That is why he does not like the phrase 'There is an x...' "Scott exists" is bad syntax because "Scott" is a name, not a propositional function. In fact, Russell proclaims, "As regards the actual things there are in the world, there is nothing at all you can say about them that in any way corresponds to existence." Things get tricky when existence is applied to definite descriptions. Remember that, according to Russell's theory of definite descriptions, definite descriptions are eliminable in every context. (They are incomplete symbols.) So, we have: G[the x: Fx] ===> (Ex)(Fx & (y)(Fy --> y = x) & Gx) When we apply existence to descriptions we have: E![the x: Fx] ===> (Ex)(Fx & (y)(Fy --> y = x)) So, let me answer your question: (1) sentences of the form "NAME exists" are bad syntax because "exists" is a predicate which only applies to propositional functions and not names, and (2) sentences of the form "THE F exists" are okay because definite descriptions are incomplete symbols -- names are complete symbols -- and aren't subject to the same worry. One caveat. Russell's remarks notwithstanding, "Scott exists" is not somehow logically defunct. "Scott" is not a logically proper name at all -- it is a disguised description! The only names, Russell thinks, are "this" and "that". "Scott" is really shorthand for some description such as "the author of waverly".

You don't have permission to post.

Please login or register.