Epistemology 2

[guesswest]

'I know a priori that I have the concept water, I know a priori that if I have the concept water then water exists. Therefore, I know a priori that water exists.' What, if anything, does this argument establish?

[Trantor]

It establishes you cannot construct a logical argument?
Seriously, assuming the truth of your two assumptions (axioms)the conclusion that follows is "Water exists" not "I know a priori that water exists".
(I assume the statement "Iknow apriori" is your way of declaring axioms.)
(I wait to be corrected on terminology by a logician!)

[madnak]

That you think too much?

Personally I have a very strong belief in human imagination. So I think the second concept is flawed. I don't believe that everything we know is based on previous perceptions or on concrete realities. I think the human brain can come up with some really trippy unprecedented stuff.

Assuming the argument is valid, it doesn't establish much. Because you need to define water for it to be meaningful. But I'm not sure if you can define something that you know a priori. I don't believe the imagination is restricted to things that have been experienced, but I do believe the use of language is based on things that have been experienced and labeled.

Your a priori concept of water, I think, is very different from my a priori concept of water. Even though we're talking about the "same" compound. When you think "what is water?" it seems obvious. But it's really not. In fact, I'd bet you have multiple concepts of "water" that you have to regularly synchronize.

[guesswest]

Well firstly I didn't construct the argument - as I mentioned on a previous thread I'm throwing out a few old exam questions for discussion. I believe the quote is Putnam but I'm not 100% sure.

As for your objection Trantor - I do believe the statement is at least internally consistent. Yes it follows 'water exists' but that's just a less specific variant of 'I know a priori that water exists'.

The immediate problem I see is that I could sub water for popcorn and this statement would still be coherent (it'd just be wrong), so coherence isn't the issue in terms of it's value to epistemology .

Edit: (scratch that)

[cliff]

That every formal system is limited by its axioms? This is why mathematicians, for instance, will generally say that a theorem is true or false in a system (defined by its axioms) rather then in any deeper sense. This is also why logic and philosophy in general can never give a "true" statement about reality, rather they can establish claims which are consistent with whatever initial axioms they have defined (which you may choose to believe are "true", but not without evoking some form of faith).

[Bork]

It doesn't establish anything because it has a false premise. (it also could be argued that its invalid)

Its not true that if you have the concept of water that water exists.

I have the concept of werewolves, god, etc... They don't exist.

[guesswest]

Putnam is an externalist and I believe that's the context of this statement - tho I'll confess to not knowing much more than that. So the idea would be that we get our conceptions from external things, therefore if we have a conception it must reference something that exists - even if it's an invalid rearrangement (eg a unicorn, because we have horses and we have horns). That'd cover werewolves also. It may or may not cover god, but that's a whole seperate issue and a huge tangent.

I'm not sure if I've got that right, or where to go with it if I have.

[bearly]

that you're pretty confused about a few things, or that you use some words in a rather odd way...........b

[Philo]

[ QUOTE ]
Putnam is an externalist and I believe that's the context of this statement - tho I'll confess to not knowing much more than that. So the idea would be that we get our conceptions from external things, therefore if we have a conception it must reference something that exists - even if it's an invalid rearrangement (eg a unicorn, because we have horses and we have horns). That'd cover werewolves also. It may or may not cover god, but that's a whole seperate issue and a huge tangent.

I'm not sure if I've got that right, or where to go with it if I have.

[/ QUOTE ]

What is an externalist, exactly?

[guesswest]

Well in the context of epistemology an internalist will maintain that everything we need to know to gauge the veracity of things, and thus to obtain knowledge, is available within our consciousness. So an externalist opposes that, but it's not polarized - they'll maintain that some component part of what we need to justify belief exists outside of consciousness.

Putnam more specifically though is a 'semantic externalist', and I don't fully understand what this means or how it's distinguished from garden variety externalism. Was hoping someone could help me get a handle on that.

Edit: The principle thing that's confusing me here is that the initial statement seems to be attacking externalism as oppose to defending it - by demonstrating a necessary absurdity.

[madnak]

I'm an internalist and not familiar with Putnam, but a quick search yielded this (http://en.wikipedia.org/wiki/Twin_Earth). Does that help?

[guesswest]

It does - ty /images/graemlins/smile.gif

[Philo]

[ QUOTE ]
Well in the context of epistemology an internalist will maintain that everything we need to know to gauge the veracity of things, and thus to obtain knowledge, is available within our consciousness. So an externalist opposes that, but it's not polarized - they'll maintain that some component part of what we need to justify belief exists outside of consciousness.

Putnam more specifically though is a 'semantic externalist', and I don't fully understand what this means or how it's distinguished from garden variety externalism. Was hoping someone could help me get a handle on that.

Edit: The principle thing that's confusing me here is that the initial statement seems to be attacking externalism as oppose to defending it - by demonstrating a necessary absurdity.

[/ QUOTE ]

Semantic externalism is the view that the environment sometimes plays a necessary role in determining the contents of our thoughts. In Putnam's Twin-Earth thought experiment he argued that the term "water" when used on earth would mean H20, but that if someone were transported secretly to "Twin-Earth" where everything is exactly like earth except that the clear liquid there that people refer to as water is not H20 but rather XYZ, then eventually the person from earth would be referring to that stuff XYZ when he used the term "water", even if he never found out that he had been transported to Twin Earth. The slogan associated with this view is that "meanings aren't in the head," and although it's not a consensus view among contemporary philosophers, it's a generally accepted or at least pretty popular view.

The original argument you mentioned about knowing a priori that I have the concept water, etc., is a kind of anti-skeptical argument (anti-skeptical with respect to the existence of the external world) that one might try to employ on the basis of semantic externalism, but there have been philosophers working on such an argument for quite some time (since around the time Putnam's piece was published which was 1975 I believe), and it's unclear that a very strong anti-skeptical argument can be built from the thesis of semantic externalism.

[Lestat]

[ QUOTE ]
'I know a priori that I have the concept water, I know a priori that if I have the concept water then water exists. Therefore, I know a priori that water exists.' What, if anything, does this argument establish?

[/ QUOTE ]

How do you know that your concept of water is correct? And if your concept of water is incorrect, then you now know a priori that water doesn't necessarily have to exist the way you think it does.

[guesswest]

[ QUOTE ]
[ QUOTE ]
Well in the context of epistemology an internalist will maintain that everything we need to know to gauge the veracity of things, and thus to obtain knowledge, is available within our consciousness. So an externalist opposes that, but it's not polarized - they'll maintain that some component part of what we need to justify belief exists outside of consciousness.

Putnam more specifically though is a 'semantic externalist', and I don't fully understand what this means or how it's distinguished from garden variety externalism. Was hoping someone could help me get a handle on that.

Edit: The principle thing that's confusing me here is that the initial statement seems to be attacking externalism as oppose to defending it - by demonstrating a necessary absurdity.

[/ QUOTE ]

Semantic externalism is the view that the environment sometimes plays a necessary role in determining the contents of our thoughts. In Putnam's Twin-Earth thought experiment he argued that the term "water" when used on earth would mean H20, but that if someone were transported secretly to "Twin-Earth" where everything is exactly like earth except that the clear liquid there that people refer to as water is not H20 but rather XYZ, then eventually the person from earth would be referring to that stuff XYZ when he used the term "water", even if he never found out that he had been transported to Twin Earth. The slogan associated with this view is that "meanings aren't in the head," and although it's not a consensus view among contemporary philosophers, it's a generally accepted or at least pretty popular view.

The original argument you mentioned about knowing a priori that I have the concept water, etc., is a kind of anti-skeptical argument (anti-skeptical with respect to the existence of the external world) that one might try to employ on the basis of semantic externalism, but there have been philosophers working on such an argument for quite some time (since around the time Putnam's piece was published which was 1975 I believe), and it's unclear that a very strong anti-skeptical argument can be built from the thesis of semantic externalism.

[/ QUOTE ]

That's an awesome explanation, thanks.

One thing that jumps out at me. Isn't it just the case that water then isn't an exact synonym for H20? That it's a looser classification? I'm assuming this thought experiment requires that, for whatever reason, the subject and their twin can't identify the chemical composition of water. So the result would be that 'water' only designates substances that meet certain criteria they both share. Is that a valid criticism or have I misunderstood his position?

[guesswest]

[ QUOTE ]
[ QUOTE ]
'I know a priori that I have the concept water, I know a priori that if I have the concept water then water exists. Therefore, I know a priori that water exists.' What, if anything, does this argument establish?

[/ QUOTE ]

How do you know that your concept of water is correct? And if your concept of water is incorrect, then you now know a priori that water doesn't necessarily have to exist the way you think it does.

[/ QUOTE ]

It's not my argument and I'm not sure I fully understand it's implications. But I don't think the 'water' part is important here, only that it establishes that whatever thing is referenced exists externally, whether it's identified by the correct word or not.

[bearly]

i think most folks (if i remember correctly) who fought this battle in the 30's and 40's---russell, dewey, et al---would like you to say "never give more than a 'formally true' statement". same thing, but they were picky. the immense interest in the "philosophy of language" starting in the early 60's was fueled by the consequences of the point you made above. despite the cult status, the "gospel according to st. wittgenstein" really did set the tone for the next 40 yrs. of modern acedemic philosophy (oh heck, we gotta throw in 'philosophy of mind')............b

[bearly]

i am not a logician---believe me. during my particular decade or so of academic work, the statement "i know, a priori, water exists" would have puzzled even the logical types, except in the trivially true case that the statement could stand as an assumption, an hypothesis ............b

[Philo]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
Well in the context of epistemology an internalist will maintain that everything we need to know to gauge the veracity of things, and thus to obtain knowledge, is available within our consciousness. So an externalist opposes that, but it's not polarized - they'll maintain that some component part of what we need to justify belief exists outside of consciousness.

Putnam more specifically though is a 'semantic externalist', and I don't fully understand what this means or how it's distinguished from garden variety externalism. Was hoping someone could help me get a handle on that.

Edit: The principle thing that's confusing me here is that the initial statement seems to be attacking externalism as oppose to defending it - by demonstrating a necessary absurdity.

[/ QUOTE ]

Semantic externalism is the view that the environment sometimes plays a necessary role in determining the contents of our thoughts. In Putnam's Twin-Earth thought experiment he argued that the term "water" when used on earth would mean H20, but that if someone were transported secretly to "Twin-Earth" where everything is exactly like earth except that the clear liquid there that people refer to as water is not H20 but rather XYZ, then eventually the person from earth would be referring to that stuff XYZ when he used the term "water", even if he never found out that he had been transported to Twin Earth. The slogan associated with this view is that "meanings aren't in the head," and although it's not a consensus view among contemporary philosophers, it's a generally accepted or at least pretty popular view.

The original argument you mentioned about knowing a priori that I have the concept water, etc., is a kind of anti-skeptical argument (anti-skeptical with respect to the existence of the external world) that one might try to employ on the basis of semantic externalism, but there have been philosophers working on such an argument for quite some time (since around the time Putnam's piece was published which was 1975 I believe), and it's unclear that a very strong anti-skeptical argument can be built from the thesis of semantic externalism.

[/ QUOTE ]

That's an awesome explanation, thanks.

One thing that jumps out at me. Isn't it just the case that water then isn't an exact synonym for H20? That it's a looser classification? I'm assuming this thought experiment requires that, for whatever reason, the subject and their twin can't identify the chemical composition of water. So the result would be that 'water' only designates substances that meet certain criteria they both share. Is that a valid criticism or have I misunderstood his position?

[/ QUOTE ]

It's still unclear to me whether or not you understand the position.

The reason that meanings aren't in the head, so to speak, is because the meanings of certain terms are determined in part by the causal relations the thinker bears to his environment. This is especially true of natural kind terms like "water" or "gold" or "tiger" where the term refers to a natural kind (as opposed to an artifact like a table or chair)--so that the existence of the stuff in the environment is what leads causally to the person's use of the term, and it is that causal relation which determines the referent and so the meaning of the term.

It seems to me that "water" and "H20" refer to the same thing--that water just is that stuff with chemical composition H20.

[guesswest]

Yes, I think I follow the position now.

By water not being an exact synonym I meant something like this: water is not as specific as H20, so can refer to H20 and this XYZ subtance, so probably encompasses all waterlike substances. In Putnam's model if you said that this XYZ substance would become known as H20 that wouldn't make sense. It'd possibly make sense if there was no hydrogen, oxygen or any of the components of XYZ anywhere else in this twin world, since you could just reterm XYZ as hydrogen and oxygen - but that isn't the case here. So I'd say Putmans thought experiment only demonstrates that 'water' can mean 'xyz' as well as 'h20', not that the meaning of 'water' itself is changed. However, it may well be that there are other thought experiments out there that demonstrate the intended proof without that failing, it seems intuitively likely to me.

Is that coherent?

[Scotch78]

[ QUOTE ]
What, if anything, does this argument establish?

[/ QUOTE ]

The existence of unicorns.

Scott

[Philo]

[ QUOTE ]
Yes, I think I follow the position now.

By water not being an exact synonym I meant something like this: water is not as specific as H20, so can refer to H20 and this XYZ subtance, so probably encompasses all waterlike substances. In Putnam's model if you said that this XYZ substance would become known as H20 that wouldn't make sense. It'd possibly make sense if there was no hydrogen, oxygen or any of the components of XYZ anywhere else in this twin world, since you could just reterm XYZ as hydrogen and oxygen - but that isn't the case here. So I'd say Putmans thought experiment only demonstrates that 'water' can mean 'xyz' as well as 'h20', not that the meaning of 'water' itself is changed. However, it may well be that there are other thought experiments out there that demonstrate the intended proof without that failing, it seems intuitively likely to me.

Is that coherent?

[/ QUOTE ]

The point of Putnam's thought experiment is to show that what *you* mean by "water" is not just a matter of what you have in your head psychologically. What you mean by "water" is partly determined by the causal relations you bear to your environment.

The form of the word is just a matter of convention (although there is an etymology to the word of course)--we could just as well be talking about what you mean by "worber"--and in that sense XYZ could come to be referred to as "H20" since letters and numbers are just symbols devoid of meaning until we supply them with a meaning. It's the meaning of the words, and so derivatively the contents of our thoughts (thoughts are being treated as represented linguistically), that Putnam is showing something about.

So, in one sense "water" can refer to anything--it's just the concatenation of five symbols, and what we use the symbols for is partly a matter of convention. In another sense though, "water" does just mean the same thing as H20 (I would say).

What does "water" refer to that you think is not H20?

Btw, there is a convention in philosophy called the "use/mention" distinction, which is why I sometimes put quote marks around a word and sometimes not.

Here's an example: If I *use* the word "Plato" then I might say something like: Plato wrote The Republic. To use a word is to talk about whatever the word refers to (in this case it refers to the Greek philosopher Plato).

If I am just *mentioning* the word Plato then it is the word itself I am talking about, as in: "Plato" has five letters.

Using this convention helps avoid conflating use with mention.

[guesswest]

I'm aware that I should be using quotes more rigorously, it's just lazy typing, apologies for my lack of clarity.

I did a bit of reading on this last night after I got offline. The term I'm looking for is 'rigid designator' - I'm suggesting that 'water' is not a rigid designator of 'h20'. I believe that would, if true, remove the external->internal causal relationship. To that end, the true meaning of 'water' would be something like 'waterlike', encompassing both 'H20' and 'XYZ'. So in this particular thought experiment it wouldn't be a case of meaning changing, rather a new material being placed under the classification 'water'. In exactly the same way we could find a new car model, a 'volkswagen xyz' on a twin earth and call it a 'volkswagen'.

[Philo]

[ QUOTE ]
I'm aware that I should be using quotes more rigorously, it's just lazy typing, apologies for my lack of clarity.

I did a bit of reading on this last night after I got offline. The term I'm looking for is 'rigid designator' - I'm suggesting that 'water' is not a rigid designator of 'h20'. I believe that would, if true, remove the external->internal causal relationship. To that end, the true meaning of 'water' would be something like 'waterlike', encompassing both 'H20' and 'XYZ'. So in this particular thought experiment it wouldn't be a case of meaning changing, rather a new material being placed under the classification 'water'. In exactly the same way we could find a new car model, a 'volkswagen xyz' on a twin earth and call it a 'volkswagen'.

[/ QUOTE ]

If "water" is not a rigid designator that does not undermine the thesis of semantic externalism. Putnam himself is not a proponent of Kripke's views on rigid designators, though both Kripke and Putnam ascribe to a causal theory of reference.

If you want to undermine semantic externalism you will probably want to reject the causal theory of reference, since if reference is determined by the causal relations a thinker bears to his environment, then what sorts of things there are in a thinker's environment will determine in part the reference of natural kind terms like "water", which again means that thoughts aren't just in the head.

In point of actual fact, I don't think "water" means the same thing as "waterlike"--if you are in a restaurant and you ask the waiter for a glass of water, and he brings you a drink that is clear and looks like water (so is 'waterlike'), but that is not water which becomes clear when you taste it, you're going to tell him that he didn't bring you what you asked for, right?

[guesswest]

In response to that I'd just say that it not tasting the same means it's insufficiently 'waterlike'.

But it's not really my intention to undermine semantic externalism, I find it very compelling. I just sense that the twin-earth thought experiment may be flawed. Putnam apparently does too, but for other reasons.

It was actually the quote on the wiki entry mad linked me to that persuaded me that it was a viable position, pasted beneath - I believe this idea demonstrates the same thing without that flaw:


Putnam pointed out, for instance, that he has no knowledge that could distinguish elm trees from beech trees. He has precisely the same concept of one as of the other: 'a deciduous tree growing in North America'. Yet when Putnam makes a statement containing the word 'elm', we take him to be referring successfully to elms. If he makes a claim about a property of elm trees, it will be considered true or false, depending upon whether that property applies to those trees which are in fact elms. There is nothing "in the head" that could fix his reference thus; rather, he concluded, his linguistic community, containing some speakers who did know the difference between the two trees, insured that when he said 'elm', he referred to elms.


If anyone can see an error in that line of thought, or something else I'm missing, I'd love to hear it.

[Philo]

[ QUOTE ]
In response to that I'd just say that it not tasting the same means it's insufficiently 'waterlike'.

But it's not really my intent to undermine semantic externalism, I find it very compelling. I just sense that the twin-earth thought experiment may be flawed. Putnam apparently does too, but for other reasons.

It was actually the quote on the wiki entry mad linked me to that persuaded me that it was a viable position, pasted beneath - I believe this idea demonstrates that same thing without that flaw:


Putnam pointed out, for instance, that he has no knowledge that could distinguish elm trees from beech trees. He has precisely the same concept of one as of the other: 'a deciduous tree growing in North America'. Yet when Putnam makes a statement containing the word 'elm', we take him to be referring successfully to elms. If he makes a claim about a property of elm trees, it will be considered true or false, depending upon whether that property applies to those trees which are in fact elms. There is nothing "in the head" that could fix his reference thus; rather, he concluded, his linguistic community, containing some speakers who did know the difference between the two trees, insured that when he said 'elm', he referred to elms.


If anyone can see an error in that line of thought, or something else I'm missing, I'd love to hear it.

[/ QUOTE ]

Why do you say that Putnam thinks it's flawed?

How is the elm example any different from the water example? An analogous case there would be to say that "Elm tree" just means "Elm-tree-like," so if what I say about Elm trees isn't actually true of Elm trees (those trees that are Elms and not just Elm-like), wouldn't someone naturally say that I got it wrong?

'Water' is normally taken to be a natural kind term (while "waterlike" is not), just like 'Elm tree'. So if what you are saying about the meaning of the term 'water' is correct then it would seem to apply to "Elm tree" as well.

And if what you're saying about "water" is correct, then Putnam's causal theory of reference for natural kind terms must be wrong. Since his causal theory of reference for natural kind terms is what supports the twin-earth thought experiment, what you're saying about "water" does undermine the twin-earth thought experiment, i.e., semantic externalism.

[guesswest]

Putnam I believe subsequently accepted Burge's criticism that the twins in the TE thought experiment would in fact not have identical psychological states, specifically that one's concept of water would be H20 whereas the others would be XYZ. I'm not suggesting that's specifically related to this line of enquiry though.

I don't believe 'waterlike' is the equivalent of 'elm-tree-like' here. I also don't believe my contention that 'water' isn't a rigid designator negates causal theory of reference.

Say technology advanced in such a way that we found a whole new level of understanding the composition of water, and we found there were in fact two different kinds of water (here on regular earth). This wouldn't make our current classification 'water' incorrect, it's just make it non-specific - we'd probably come up with names like 'water a' and 'water b'. They'd both still be 'water'.

If the word 'h20' was used in place of the word 'water' throughout the TE experiment it wouldn't make any sense. We'd just be incorrectly identifying the substance, not changing the meaning of the word, because hydrogen and oxygen necessarily exist elsewhere in this otherwise exactly replicated world (and xyz would be composed of neither).

[Philo]

Yes, Burge disagreed with Putnam's supposition that the thinker on earth and the thinker on TE would have the same thought when they thought something like "Water tastes good", b/c Burge thought the concepts differed on earth and TE, and this so b/c the referents differed--one was H20 and the other was XYZ. Neither Burge nor Putnam thinks this difference undermines the main point of the TE thought experiment, however.

I agree that the causal theory of reference would not be undermined if we determined that "water" was not a rigid designator, but why do you think that "waterlike" is not analogous to "Elm-like"?

I see where you have gotten these ideas, but note under the "Objections" section that "Some philosophers consider this a sufficient argument to debunk the thought experiment altogether." So, these philosophers do think that "water" must be understood to be a rigid designator if the TE thought experiment is to provide plausible support for semantic externalism.