More just that language isn't something which is institutional, — Moliere

I'm not saying anything about institutions.

Here are some common definitions of "consensus":

* "general or widespread agreement"
* "majority of opinion"
* "agreement of the people"

And "convention"

* "a way in which something is usually done"
* "a rule, method, or practice established by usage; custom""
* "general agreement or consent; accepted usage"

Maybe you're using "institution" in some looser manner than referring to "formal" institutions however.

Agreement is a kind of institutional action . . . There is a certain history to words which agreement is unable to overcome. People don't follow suit and decide to create language. Rather, we are born into a world with language, and it already means something, regardless of my intent. — Moliere

I couldn't disagree with these two paragraphs more strongly than I do. In my opinion, it's rather clear that you have this stuff factually wrong.

I'd say that this is what your theory is unable to explain -- it explains how it is we can take a sign to mean something, but it doesn't explain the factual side of language. — Moliere

The factual side of language certainly isn't that words have definitions (which you'd call "meanings") that can be different than usage. That goes against the factual evidence. Words are defined however people choose to define them. They can do something highly idiosyncratic there, or they can follow suit with how the vast majority of people are defining the term, or they can do anything in between. None of that is right or wrong, by the way. And typically, those definitions, that usage (of the vast majority that is) shifts over time.

If I know that "tomato" means a round, soft, red fruit that is eaten raw or cooked and that is often used in salads, sandwiches, sauces, etc., then it is true that "tomato" means a round, soft, red fruit that is eaten raw or cooked and that is often used in salads, sandwiches, sauces, etc. — Moliere

Haha--no, that isn't true. It's only true that that's the conventional definition.

This is similar to saying, for example, (a) "Vanilla ice cream tastes good" versus (b) "80% of the people in Des Moines feel that vanilla ice cream tastes good." (a) isn't true (or false). (b) is true or false. In other words, it's only true or false in the context of some particular persons feeling that vanilla ice cream tastes good or not. It's not (that particular) context-free true or false that vanilla ice cream tastes good.

So likewise, "The definition of tomato is such and such" isn't true or false. But "Most people define tomato as such and such" is true or false. It's only true in the context of some particular persons defining tomato in the way in question.

How about both? I'd call the former a descriptive definition, and the latter an ostensive definition. So they are two different definitions of the meaning, but we can both fairly say we know the meaning of the word, I think. — Moliere

So isn't it necessary to distinguish between meaning in sense #1, and meaning in sense #2 then? If we don't maintain this distinction there could be ambiguity or even equivocation. Meaning in the one sense lends itself to truth as correspondence, and in the other sense, coherence.

But then suppose while I was in California I were to say, "It is 5:00 PM". And my cousin, who lives on the East coast, were to also say "It is 5:00 PM" at the same time in a telephone conversation. Only one of these utterances is true, — Moliere

Two different propositions were expressed. An utterance is sounds or marks, generally... not really a good candidate for truth-bearer.

One of the reasons I like the focus on utterances is that it seems, at least, to be a nice and neat way to accept all the messiness of context without getting lost in the mud of possible contexts — Moliere

Propositions don't operate on "possible contexts." A sentence is uttered to express a proposition. Listeners either understand what proposition was expressed or they don't. No malarky about computer generated poetry here.

And whatever the ontology of propositions might be (and it is quite the poser), Scott Soames lays out a pretty good argument which concludes that the price of denying propositions is giving up the possibility of agreement. And those who are fine with that probably don't bother to be understood, I would imagine.

I couldn't disagree with these two paragraphs more strongly than I do. In my opinion, it's rather clear that you have this stuff factually wrong. — Terrapin Station

Cool.

The factual side of language certainly isn't that words have definitions (which you'd call "meanings") that can be different than usage. That goes against the factual evidence. Words are defined however people choose to define them. They can do something highly idiosyncratic there, or they can follow suit with how the vast majority of people are defining the term, or they can do anything in between. None of that is right or wrong, by the way. And typically, those definitions, that usage (of the vast majority that is) shifts over time. — Terrapin Station

Hold on there. A definition is not a meaning. A definition describes the meaning. When someone uses a word, they are not defining it. They're using it.

Haha--no, that isn't true. It's only true that that's the conventional definition — Terrapin Station

I understand that you disagree. But the best reason I seem to get for you is that your idea about meaning accounts for being able to use words idiosyncratically.

But if we can figure out what words mean by their usage, then I don't see an issue with new uses of words, and it seems that we have a way of understanding shared meanings, rather than having them be private.

I don't think so. Meaning isn't the same thing as definition, so there's no need to say there are different kinds of meanings just because we define a word in different ways.

I don't think so. Meaning isn't the same thing as definition, so there's no need to say there are different kinds of meanings just because we define a word in different ways. — Moliere

But these aren't two types of definition, one is a definition, the other a direct relating the word to an object. The latter is knowing what "apple" means by being able to point to an apple, it is not in any way defining "apple".

Two different propositions were expressed. An utterance is sounds or marks, generally... not really a good candidate for truth-bearer. — Mongrel

What were the two different propositions? To my understanding, propositions are generally taken as the content of certain expressions. No? Does the location in which we speak change the semantic content of an expression?

I may just misunderstand propositions. Because I would say that "It is 5:00 PM" is expresses the same proposition regardless of the speaker -- so your assertion that there are two different one's runs counter to my understanding. (which, as I've noted, isn't in any way professional. I'm certainly open to reading more. I'm just interested in the topic, so I'm talking)

Propositions don't operate on "possible contexts." A sentence is uttered to express a proposition. Listeners either understand what proposition was expressed or they don't. No malarky about computer generated poetry here. — Mongrel

Propositions don't, but language does. An understanding of meaning without somehow incorporating context doesn't strike me as terribly helpful because meaning changes so much with context.

Why not computer generated poetry? Isn't computer generated poetry just as much a part of language as declarative sentences?

That's the crazy part about language. It has meaning regardless of intent. If a computer generates a sentence, then we know what it means even though there wasn't even a speaker.

What's wrong with sounds or marks, vs. propositions? To me it seems that I know the former exist because I see them. But the latter strike me as convenient inventions that don't even account for language meaning, but only the meaning of very particular types of sentences which some philosophers have an interest in. Granted, these are the sorts of sentences we're usually interested in when talking about truth-apt sentences, and therefore truth, but still -- it seems to me that meaning is wider than truth, and truth is just one goal a sentence can accomplish.

While Saussure certainly believes meanings are in the head, I don't think that is necessary to take on board if we talk about language in terms of signs. A sign is composed of both a signifier and a signified. The marks can be the signifier. And truth is the property which a signifier has to some fact. The signified is the meaning which "comes along with", but given that semantic meanings of words are resilient to change -- whereas the marks in a given context aren't (whether "mark" be understood as phonic or visual), and in fact vary considerably with context -- it makes sense to assign truth to the mark rather than the meaning to account for the variety in contextual use.

It may be counter-intuitive to say that the marks we see bear truth -- but that doesn't mean it shouldn't be the case, no? I can see that because a mark can have different meanings that maybe it's a bit of a bait-and-switch move... but it seems to me that just because a mark can look different and mean the same, or vice-versa, that there's still good reason for attributing truth to the mark because it's the very thing which is in context.

But these aren't two types of definition, one is a definition, the other a direct relating the word to an object. The latter is knowing what "apple" means by being able to point to an apple, it is not in any way defining "apple". — Metaphysician Undercover

If someone uses the word "apple", then they are not defining it. They are demonstrating competence of the language, but they are putting the language to use.

If someone is teaching another the meaning of "apple", then they may choose, rather than speak the meaning, to demonstrate the meaning by showing the pupil an apple.


So being able to point to an apple, and pointing to an apple, is using "apple". Demonstrating to someone how to use "apple" is a way of defining apple. Telling someone what an apple is is another way of defining "apple". So if we are asking a clerk for an apple then we are using the word "apple", and if we are telling someone about "apple" then we are mentioning the word "apple".

The relationship between X and what it represents (some unknown quantity or other) in algebra is anything but arbitrary. If it was arbitrary we would never be able to determine what the unknown quantity is. Of course it doesn't matter whether it is 'X' or 'A', or whatever symbol you like, used to represent the unknown quantity, but that is not the point at issue.

Also this is not a case of arbitrary semantic meaning; where a conventionally determinate meaning is substituted by an arbitrary stipulated meaning. Any stipulated meaning is still determinate, insofar as it is stipulated. In the algebra case the meaning of 'X' is only determinable by a process of calculation; and this is not analogous to conventional semantic meaning; which is either already known or has to be looked up. There is no process whereby semantic meanings can be calculated.

If someone uses the word "apple", then they are not defining it. They are demonstrating competence of the language, but they are putting the language to use. — Moliere

OK, so back to my original question then. When a child learning language knows how to point to the proper object when the word "apple" is spoken, does that child know what "apple" means? If so, then that child knows what "apple" means without knowing any definition of apple.

At first, you said that this person does know what "apple" means, and that's why I said we need to distinguish between two senses of meaning. One being associated with the capacity to define the word, the other being associated with the capacity to relate the word to an object.

They know what apple means.

But maybe they don't know how to define a word, yet. They haven't reached the ability to begin thinking about their words in the same way that they think about their apples. So they know what "apple" means, but they may not know what "The word "apple"" means.

Because I would say that "It is 5:00 PM" is expresses the same proposition regardless of the speaker — Moliere

One doesn't even know what proposition is being expressed unless the context of utterance is known.

John said "2 is a prime number."
Bill said, pointing to the number 2 written on the blackboard "That is a prime number."

Different sentences, different utterances, same proposition.

An understanding of meaning without somehow incorporating context doesn't strike me as terribly helpful because meaning changes so much with context. — Moliere

Yep. As I said.. you identify the proposition expressed by an utterance by attending to context of utterance.

What's wrong with sounds or marks, vs. propositions? To me it seems that I know the former exist because I see them. But the latter strike me as convenient inventions that don't even account for language meaning, but only the meaning of very particular types of sentences which some philosophers have an interest in. Granted, these are the sorts of sentences we're usually interested in when talking about truth-apt sentences, and therefore truth, but still -- it seems to me that meaning is wider than truth, and truth is just one goal a sentence can accomplish. — Moliere

Sounds and marks won't work as primary truth-bearers in spite of their ready visibility. If you and I are in agreement, it's not sounds or marks we're agreeing to.

I get the objection to propositions based on ontological considerations, but as photographer would often say: reality is what we can't do without. Before you ditch propositions, recognize what you're saying you can do without. As I mentioned, it's communication itself that's undermined by that rejection.

The relationship between X and what it represents ( an unknown quantity) in algebra is anything but arbitrary. — John

But X does not represent an unknown quantity, it represents a defined object "Let X = ..." The task of the mathematician is to assign a numerical value to that defined object. The definition of X is completely arbitrary because we can fill the space with whatever random definition which we want to figure the numerical value for.

They know what apple means. — Moliere

So do you agree that we need to respect two distinct senses of "meaning"? One is associated with the capacity to relate words to objects, and the other is associated with the capacity to relate words to other words, form a definition.

But X does not represent an unknown quantity, it represents a defined object "Let X = ..." The task of the mathematician is to assign a numerical value to that defined object. The definition of X is completely arbitrary because we can fill the space with whatever random definition which we want to figure the numerical value for. — Metaphysician Undercover

Give me an example of the kind of "defined" object you think 'X' represents in algebra.

Let X = the number of times that I drove to the store last week. "The number of times that I drove to the store last week" is the defined object.

The whole point of defining something like that as 'X' is to facilitate calculating it in cases where you don't know what it is, but you do know other things relating to what you are defining which will enable you to calculate it, which is pointless if you already know what X equals. So the number of times you went to the store is an unknown quantity, plus you have some other information that can be formulated in terms of X such as to enable that unknown quantity to be calculated, or if you do know the number of times you went to the store, then there is no point defining it as 'X".

Think of this example of an algebraic formula: 2X +15= 25, you are not defining X, but '2X +15' and you are defining the latter as being equal to 25. This enables you to calculate the value of X, which prior to calculation is an unknown quantity.

Sure, it doesn't matter whether X is the number of times you drove to the store or the number of times you deceived your wife (and the latter may well be incalculable ;) ) but that arbitrariness is a separate (non) issue.

In any case all of this is a red herring for the reason that algebraic definitions are not equivalent to semantic definitions as i already pointed out.

One doesn't even know what proposition is being expressed unless the context of utterance is known.

John said "2 is a prime number."
Bill said, pointing to the number 2 written on the blackboard "That is a prime number."

Different sentences, different utterances, same proposition. — Mongrel

OK. Then yeah, I certainly misunderstood them.

Sounds and marks won't work as primary truth-bearers in spite of their ready visibility. If you and I are in agreement, it's not sounds or marks we're agreeing to. — Mongrel

That's fair. I agree.

I get the objection to propositions based on ontological considerations, but as photographer would often say: reality is what we can't do without. Before you ditch propositions, recognize what you're saying you can do without.

There's the ontological consideration, but also it seems an odd way to talk about meaning too. Propositions focus on such a very specific use of language, and it seems to me that if one were to base a belief about meaning on them that they'd just be over-generalizing and getting it wrong. Some meaningful sentences are true, but then if we just understood meaning then that would give a means to truth (at least, truth understood in this way -- the kind of truth which telling the truth relies upon)

As I mentioned, it's communication itself that's undermined by that rejection. — Mongrel

Cool. I'd have to read the dude you referenced, I think.

My immediate thought is that we could just take meaning for granted. It seems more plausible, to me at least, to believe that our expressions are meaningful rather than to rely upon a belief in propositions to say what it is that makes them meaningful. We don't have to know what it is that makes a sentence meaningful to know that it is meaningful, after all.

X represents a known and defined object, "the number of times that I went to the store last week". Clearly you're wrong to say that we don't know what it is. We know, according to that definition. It is the value of this defined object which is unknown. It is determined by placing it in relations to other things with known values.

In any case all of this is a red herring for the reason that algebraic definitions are not equivalent to semantic definitions as i already pointed out. — John

You are simply denying the semantic part, where the symbol X is given meaning, and focusing on the mathematical part, to maintain your assertion that we cannot give symbols arbitrary meaning.

Sure, it doesn't matter whether X is the number of times you drove to the store or the number of times you deceived your wife (and the latter may well be incalculable ;) ) but that arbitrariness is a separate (non) issue. — John

No it is not a separate issue, it is a clear indication, that contrary to what you claim, we can, and do, by definition, give a symbol an arbitrary meaning.

My immediate thought is that we could just take meaning for granted. It seems more plausible, to me at least, to believe that our expressions are meaningful rather than to rely upon a belief in propositions to say what it is that makes them meaningful. We don't have to know what it is that makes a sentence meaningful to know that it is meaningful, after all. — Moliere

Again, you're sort of acting like propositions are objects of superstition... as if they're fairies magically making meaning where there was none. They're just content. You can deny that there really is any content to speech and writing. You can go behaviorist.. we're all just quacking ducks. Meaning is quacking.

I don't have the alternative figured out. I just know Hamlet's to be or not to be soliloquy is not the same as duck noises. :)

So do you agree that we need to respect two distinct senses of "meaning"? One is associated with the capacity to relate words to objects, and the other is associated with the capacity to relate words to other words, form a definition. — Metaphysician Undercover

If that's the conclusion I should draw from what I said, I don't see it as of yet. Meanings aren't related to our capacities, so just because we have different capacities -- are able to do different things with words -- that does not then mean that words have two different senses of meaning. We can use "apple" and we can mention "apple", but that's just us using the same word in a different way.

In fact I'd be rather suspicion of a theory of meaning that treats words like tags that we can put on objects, to be honest. We do things with words, and one of the things we do with words is refer -- but that's no different than if I were to point at something.


"meaning" has many meanings, and if that's all you mean by sense then I don't see it as controversial. Maybe I don't get what you mean by sense.

No, that's not exactly what I'm going after either. Or, at least, if it is true, I don't believe that language is a series of barks chained to our reproductive worth to the species.

(EDIT: though, by the same token, I don't think language makes us special in comparison to other animals either.)

Mostly because words have meaning (or sentences, at least, do).


Maybe it's just the sound of proposing "propositions" that makes me think of them as magical fairies -- that's the connotation that I hear when hearing someone propose propositions.

"Proposition" sounds math-like and logical to me. And it's probably as close to philosophy of math as I ever really need to get. But you're right.. meaning and truth are different animals. Meaning is probably more of a social issue (holism, atomism, molecularism) with some logical and metaphysical stuff in the wings.

Logic is upfront when we're talking about truth. A truth realist will deny that the mechanics of meaning are ultimately significant with truth because a proposition can be true though it's never been expressed and no one knows it. This must be so. Otherwise there would be no detectives.

No. you are simply conflating mathematical and semantic meanings. Whether X represents the number of times you went to the shop, cleaned your teeth, fed your dog or whatever (your so-called "defined object") is absolutely irrelevant to the mathematical meaning of 'X', which is a variable; and a variable is just a symbol for a number we don't yet know.

Logic is upfront when we're talking about truth. A truth realist will deny that the mechanics of meaning are ultimately significant with truth because a proposition can be true though it's never been expressed and no one knows it. This must be so. Otherwise there would be no detectives. — Mongrel

See. . . that definitely strikes me as a semi-platonic entity then (it may not be strictly platonic, so that's why I say semi- in that it relates to some features of platonic philosophy). If a proposition can be true even though it's never expressed and no one knows it, then the proposition has a reality all of its own -- and propositions are even thought to be the vehicle through which we can translate to different languages, so whatever language we might be speaking would be inconsequential to the existence of propositions. Language wouldn't even need to exist for propositions to exist, in that case, as I see it.

That's just something that I'm incapable of believing in.

Truth I could at least see as semi-plausible as a platonic entity. Not saying I believe in it, but it's at least believable and something I could consider seriously.

But some kind of language-invariant meaning that's also true above and beyond usage just seems like a convenient just-so story to me.

Language, I can see, exists all on its own. It has an independent reality, of a sorts. It's pseudo-real, and exists in the same way as anything else we might posit exists. But then I would also say that "it is raining" means something different from "Es regnet".


Surely there's some other way of thinking about truth, and believing in detectives, than believing in the existence of propositions which are never expressed and never known.

Surely there's some other way of thinking about truth, and believing in detectives, than believing in the existence of propositions which are never expressed and never known. — Moliere

Sure. You can be deflationist.

Hold on there. A definition is not a meaning. A definition describes the meaning. When someone uses a word, they are not defining it. They're using it. — Moliere

Definitions are correlated to meanings. I wouldn't say they "describe" meanings. For one, in order for something to count as a description, one has to assign meaning to it, because descriptions are representations. In my view, usage does indeed determine definition. This includes everything from simply stating a definition to contextual usage, both re the context of other language and re behavioral contexts.

But if we can figure out what words mean by their usage — Moliere

You don't figure out what words mean. You observe usage and assign meaning.

Re the reason I'm saying that it's not true or false that tomato is defined as x context-independently, that is outside of how someone(s) happens to be defining/using x, and getting back to the thread topic, is that context-independently, there is no truthmaker for "tomato is defined as x." The reason that I'm not reading "defined as" as necessarily referring to the context of consensus or conventional usage I've already explained in detail, and you've already agreed with this; you've already agreed that "defined as" needs to be contextually qualified, that it can't necessarily be read as implying (non-institutional) consensus or conventional usage.

You don't figure out what words mean. You observe usage and assign meaning. — Terrapin Station

I think we're chasing our tail on this one. :D

Re the reason I'm saying that it's not true or false that tomato is defined as x context-independently, that is outside of how someone(s) happens to be defining/using x, and getting back to the thread topic, is that context-independently, there is no truthmaker for "tomato is defined as x." The reason that I'm not reading "defined as" as necessarily referring to the context of consensus or conventional usage I've already explained in detail, and you've already agreed with this; you've already agreed that "defined as" needs to be contextually qualified, that it can't necessarily be read as implying (non-institutional) consensus or conventional usage.

I'd say that anything that is true is not true context-independently. "context-independently" is an imaginary scenario by which we may be able to judge certain things as more or less subjective, but since it is imaginary it's the sort of standard which we can draw wherever our heart desires. It's more a way of creating a point of contrast for comparison than it is a reality.

"All Bachelors are Unmarried" is true only in the context of English. "War is War" is not a necessary truth, but is true in our world. "(A + B) + C = A + (B + C)" is true in basic arithmetic


For what is there a truthmaker, in your view? What is context-independent?

You're not forgetting that I'm talking about a specific context here, right? Namely, how particular persons are defining/using the term(s) in question. I specified this a number of times.

And in conjunction with this, you've agreed that "definition," unqualified, does not refer to a non-institutional consensus or conventional usage.

You're not forgetting that I'm talking about a specific context here, right? Namely, how particular persons are defining/using the term(s) in question. I specified this a number of times. — Terrapin Station

No. I figured what you were saying, though, was that because " there is no truthmaker for "tomato is defined as x."" that it follows that meaning is mental/private/subjective. (Hence why you were saying that we do not figure out what words mean, but rather assign meaning)

And in conjunction with this, you've agreed that "definition," unqualified, does not refer to a non-institutional consensus or conventional usage. — Terrapin Station

Yup.

But I don't think that implies that meaning is mental/private/subjective.

No. I figured what you were saying, though, was that because " there is no truthmaker for "tomato is defined as x."" that it follows that meaning is mental/private/subjective. — Moliere

That wasn't the idea there, although yeah, I'd say that the two facts are related. The idea is more that if you want to claim that there is a context-independent (per this specific context) truthmaker for "tomato is defined as x," then you should specify what the truthmaker is.