Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Epistemic Perspective : Meanings ARE in the head

  1. Jan 22, 2008 #1
    Constructive criticism please:

    In 1975, Hillary Putnam published a paper entitled “The Meaning of ‘Meaning’”, in which he proposed an interesting “thought experiment” (see Putnam, H; “The Meaning of 'Meaning'” In Philosophical Papers, Vol. 2: Mind, Language and Reality. Cambridge University Press (1975)). He begins by supposing that elsewhere in the universe there is a planet exactly like earth in virtually all respects, which we refer to as ‘Twin Earth’. On Twin Earth there is a Twin equivalent of every person and thing here on Earth. The one difference between the two planets is that there is no water (ie no H2O) on Twin Earth. In its place there is a liquid that is superficially identical, but is chemically different, being composed not of H2O, but rather of some more complicated formula which we abbreviate as ‘XYZ’. The Twin Earthlings who refer to their language as ‘English’ call XYZ ‘water’. Finally, we set the date of our thought experiment to be several centuries ago, when the residents of Earth and Twin Earth would have no means of knowing that the liquids they called ‘water’ were H2O and XYZ respectively. The experience of people on Earth with water, and that of those on Twin Earth with XYZ would be identical.

    Now Putnam poses the question: When an earthling, say Oscar, and his twin on Twin Earth say 'water' do they mean the same thing?

    According to Putnam, although Oscar and Twin Oscar are in the same physical mental states (i.e., mental states interpreted in a physicalist way), Oscar points at and therefore means H2O, whereas Twin Oscar points at and therefore means XYZ. Nothing in their heads would "tell" us the difference, that is, allow us to distinguish between these two different meanings. Therefore meanings are not in the head. Instead, we rely and depend on the world to give and assign meanings. From this, Putnam concludes that “meanings just ain’t in the head”.

    The fallacy in Putnam’s argument lies in the fact that he confuses two different epistemic perspectives (one perspective which is the one of the less privileged actors Oscar and Twin Oscar on the stage, who know nothing about H2O or XYZ but know only “water”; the other perspective, the more “privileged” perspective, is the one of the stage manager, who knows that some actors have access to H2O, whereas some have access to XYZ), and he assumes that the “meaning” embodied within one perspective must equate precisely to the “meaning” embodied within the other perspective.

    Meaning derives from understanding and knowledge. If I understand that water comes in two “types” (H2O and XYZ) then of course I must distinguish these two types within my semantics if I am to avoid confusion in meaning. This is the perspective of the privileged observer.

    However, if I understand only that water is a colourless, odourless, tasteless liquid which at times falls from the sky (and I have no knowledge of the fact that water is either H2O or XYZ), then my “meaning” when I talk of “water” does not embody this difference in types of water, my “meaning” does not distinguish between these two types. This is the perspective (the meaning) of the less privileged observer.

    In short – if my knowledge and understanding differ from yours, then it is quite possible that the meanings I ascribe to certain words (my semantics) are also different to the meanings you ascribe to the same words (your semantics). It is a mistake, therefore, to assume (as Putnam seems to want to do) that there is some unique “meaning” within words (unique semantics) which is not dependent on subjective knowledge and understanding.

    When Oscar points at H2O and says “water”, he has no idea that he is pointing to H2O. He knows, and when he utters “water” he means, simply “that colourless, odourless, tasteless liquid which at times falls from the sky”. Similarly, when Twin Oscar points at XYZ and says “water”, he has no idea that he is pointing to XYZ. He knows, and when he utters “water” he means, simply “that colourless, odourless, tasteless liquid which at times falls from the sky”. To Oscar and Twin Oscar, the word “water” means the same thing. Putnam cannot understand this, because he seems to believe there is some underlying absolute semantics which is independent of observer perspective, and he insists on confusing his privileged perspective (his knowledge and understanding that water comes as both H2O and as XYZ) with the more limited perspectives of Oscar and Twin Oscar.

    Once the relative nature of our epistemology is accepted, the problem is solved, and meanings are once again firmly rooted where they belong - in the head.
  2. jcsd
  3. Jan 27, 2008 #2


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi movingfinger,
    Nice writeup. I wish more people here were interested and conversant in the nuts and bolts of philosophy.

    I’m not sure, but I wonder if you’re interpreting Putnam correctly. He goes through this argument at length in his book, “Representation and Reality”. This particular thought experiment begins on page 30. I’d be glad to email you that portion of the book. Not sure if posting a portion of a book would violate any copywrite laws though.

    Perhaps a better reference is the Stanford Encyclopedia of Technology.

    What I’m wondering is if the term “meaning” as you are using it and as Putnam is using it is very slightly different. I think Putnam is using it to indicate that there is a reality which is external to us, and you are using it to indicate that there is something in our minds that correlates to reality.

    Also, if you have a copy of the paper, I’d be interested in reading it. I don’t see anywhere that it is available on the net, or is there? Also, posting a copywrite paper would be illegal, so if it’s not available on the net, may I send you my email?
  4. Jan 30, 2008 #3
    What you're attacking here is called essentialist metaphysics. It used to be hugely popular, then it was hugely unpopular, now it's hugely popular again. Swings and roundabouts.

    The question here is "what's in a name?" Does a name refer to the superficial, describable properties of something? Does it refer to what we can see (i.e. something to do with our epistemic situation)? Is it true that "water" literally means "stuff that ripples and is transparent and is in the sea etc. etc."? You think so. And a few decades ago everyone would have gone along with it. It's usually called the Frege/Russell theory of naming and your piece is a very good espousal of it.

    Then Saul Kripke (then aged 30) changed everyone's mind with his book "Naming and Necessity". I strongly recommend it. It's fantastic. Putnam here is basically allied with Kripke.

    Kripke thinks that a name like "water" is a rigid designator. It refers to the same thing in all possible worlds, no matter how different those worlds are. The laws can all be different, the superficial, describable properties can all be different, but the thing called water is still water. And H2O is just a rigid designator attached to the same thing. So water is H2O by necessity, in every possible world. It couldn't be otherwise. If it's not H2O, it's not water.

    I suspect you will take a lot of convincing about this. One question to ponder over is: "If Bill Clinton had died aged 2, would he still have been Bill Clinton?"
    Last edited: Jan 30, 2008
  5. Jan 31, 2008 #4
    Thanks for your contribution.

    Yes, I will take a lot of convincing. I would be interested to know whether you can see anything wrong with my argument (the OP)? If there is nothing wrong with my argument, I cannot see how the conclusion can be avoided - ie meanings are in the head.

    I accept that one can "define" water and H2O to be the same thing (in which case it becomes a matter of convenience). But that is then a matter of subjective definition (ie all in the head). There is nothing in the external world which constrains "water" and "H2O" to be the same thing. Indeed, on Twin Earth (in the OP) they are NOT the same thing.

    Sorry, but imho the question seems trivial. It is like asking "would the Statue of Liberty still have been the Statue of Liberty if it had been destroyed two years after being erected?". Of course the answer is yes.

    What does this have to do with the OP?
    Last edited: Jan 31, 2008
  6. Jan 31, 2008 #5

    Perhaps. Maybe there is indeed more than one meaning of "meaning" (as discussed here: http://www.philosophy.ubc.ca/prolegom/backissues/papers/Sajadi.htm) - which then reinforces one of my beliefs that there is no point debating something unless one clearly defines one's terms first. I will add a definition of "meaning" to my paper (but need to think carefully first).

    I'll put this up so that someone can shoot it down :
    Suggested definition of "Meaning": The "Meaning" of a word is defined as the combination of that word's intension and its extension.

    Sorry, don't know of any electronic version available on the net - but Putnam's work is available to buy from Amazon http://www.amazon.com/Philosophical-Papers-Language-Reality-Hilary/dp/0521295513
    Last edited: Jan 31, 2008
  7. Jan 31, 2008 #6
    To an essentialist it's all wrong, starting with "Meaning derives from understanding and knowledge." That's not to say it's a bad argument - you make a very plausible claim that has stimulated lots of debate.

    Here's an example: if I'm at a party and say to my friend "look, the man with the glass of wine over there..." it seems like the meaning of that statement is not just the description I have given. What if he is in fact holding a glass of lemonade? What if I'm myopic and he's a she? Did I in fact refer to someone different without realising it? Intuitively no - I'm still referring to the same person, even though my description was all wrong. I can know nothing about the properties of an object but still refer to it.

    If my friend looks where I looked and says "no, it's a woman with a glass of lemonade," my friend is picking out the same referent. So the man with the glass of wine (i.e. what I named my referent) is identical to the woman with the glass of lemonade (i.e. what my friend named her referent), and it couldn't possibly have been otherwise. This is analogous to the more contentious water/H2O situation.

    Kripke runs through several examples (some far more convincing than that one). For example, her knocks down Wittgenstein's famous claim that the standard metre in Paris is "neither one metre long nor not one metre long" because "one metre" just means the length of the standard metre. Kripke thinks that's bull**** and so do I.

    It's a very good book and essential (haha) reading in this field.
    Last edited: Jan 31, 2008
  8. Jan 31, 2008 #7
    While I think your criticism is certainly well thought out, and I actually agree it is a mistake many make. I think the meaning of Putnam's thought experiment is that 'meaning is community driven', or 'agreed upon'. It therefore demands more than strictly 'subjective' meaning, in the sense one might attribute to rationalism (Descartes subjectivism sets the standard). Classic rationalism says that all truth can be derived by reason, or thinking. So if one applies this strictly, as some might, and probably have, then meaning becomes limited to the individual. Of course this ignores the fact we develop meaning through interaction, and we accept official meanings, which are defined "external" to us. Most aren't solipsists.

    Mainly I think this is one of those teapot issues, which philosophers like to niggle into tempests.
    Last edited: Jan 31, 2008
  9. Jan 31, 2008 #8
    I would be interested to know how an essentialist derives meaning?

    Actually I think what you have described reinforces MY argument rather than Putnam's (or Kripke's). Clearly when I say "man with the glass of wine" I don't actually mean "man" and I don't mean "wine" - what I actually mean is "the person (whom I think is a man) with a glass of liquid (which I think is wine)". There is no "rigid designator", there is simply a mistaken belief associated with the collection of objects which (I believe that) I am referring to with the words that I use.

    Oh, and by the way, its not in fact analogous to the twin earth example. The twin earth example has one word (water) with two referents (H2O and XYZ) - which leads to ambiguity in the meaning of the word water; whereas the above example has two words (man and woman) with one referent (the person over there) - which is nothing to do with ambiguity in meaning but is simply due to a mistaken belief.

    and yes, I'm still unconvinced. :wink:
    Last edited: Jan 31, 2008
  10. Jan 31, 2008 #9
    Intension matters. The property green is what I mean by the predicate "green".

    Extension doesn't matter. "All green things" is not what I mean by the predicate "green".

    For a statement like "that thing is green" we at least need a grasp of the intension of green and the reference of "that thing".
  11. Jan 31, 2008 #10
    Interesting. In the Twin Earth experiment, I would argue that the intension of "water" (as far as Oscar & twin Oscar are concerned) is "that colourless, odourless liquid which at times falls from the sky", whereas the extension of "water" includes H2O (on earth) and XYZ (on twin earth). If you are correct in your assertion that meaning encompasses intension but does not encompass extension, then the meaning of "water" (as far as Oscar & twin Oscar are concerned) is simply "that colourless, odourless liquid which at times falls from the sky".

    If I say, pointing to a field of grass, "that thing is green", it does not follow that ONLY grass is green. Similarly, if I say, pointing to a bucket of H2O, "that thing is water", it does not follow that ONLY H2O is water. The Twin Earth thought experiment simply illustrates under what conditions something else, apart from H2O, may be considered to be water.
    Last edited: Jan 31, 2008
  12. Jan 31, 2008 #11
    I agree up to a point, but no amount of "consensus" between individuals (each with a subjective viewpoint) can turn subjective meaning into objective meaning - and believing that such consensus does result in objectivity is (I believe) the mistake made by Putnam. Even a communally agreed meaning is still subjective - but now the subjective result of consensus between a group of individuals rather than just one individual.
  13. Jan 31, 2008 #12
    The meaning of a predicate is its intension. Water in the Putnam example is a name not a predicate. I accept that there are subtle cases like "the stuff in that glass is water" where water is sort of being used as a predicate - a predicate that means something like "belonging to the natural kind named water". But Putnam wants to use it - quite accurately - as a name of a natural kind, the name of a particular kind of substance. Like "iron", "oxygen", "H2O".

    If I say "water is blue" I am attaching a predicate to a name. I am saying that the natural kind water has the attribute of blueness. If I say "H2O is water" I'm making an identity claim involving two names. I am not saying that H2O has the attribute of waterness. Everyone can agree on this much.

    The question is then whether the name connotes anything at all, or just denotes. In other words, are names definite descriptions or rigid designators?

    You're quite right that if I grant that "water" connotes "colourless, odourless liquid which at times falls from the sky", I have to concede that the name is a definite description. But of course I don't concede that at all. I don't think names do connote anything.

    So your strategy here is to say that names are definite descriptions, and that when we appear to get the definite description all wrong we in fact don't - we have a correct definite description in our heads but it comes out wrong. That's a good comeback, but there is still a problem.

    Say we've named this object "person holding glass of liquid" (that I really can't see very well). I say to my friend "If he'd dropped that glass then, it would've wrecked the carpet." Who am I referring to in that sentence? If "person holding glass of liquid" drops his glass, he can't still be "person holding glass of liquid", so I must be referring to someone else or nothing. Except I'm not - I can still refer to the same object, even though the definite description no longer picks out that object.

    Similarly for Bill Clinton: the best definite description of Bill Clinton I could give would be something like "the man who was president at such and such a time and is married to Hillary and is the father of Chelsea and looks like such and such... etc. etc." Yet I can imagine scenarios in which he possesses none of those attributes at all. I can meaningfully say things like "What if Bill Clinton was born in China?" "What if he'd died aged 2?" "What if he was an Egyptain pharoah?"! Our language doesn't fit the descriptivist picture.
    Last edited: Jan 31, 2008
  14. Jan 31, 2008 #13
    Privileged and less privileged are good uses, but I would go even further and say that in the OP example, it is simply the amount of 'resolution' that the observer sees in the situation.
    By resolution I basically mean how many words, and how many details, one can use to describe something, and how this can distinguish between objects/things that appear to be the same thing without that detail.

    So my conclusion would be that in regards to the OP example, the two twin people are in fact talking about the same thing, if we were to compare their two descriptions of water, but if we studied the chemical compounds of each type of water, we would find the difference. So basically two people see something that are in many ways identical, and it would be logical to say that they are depending on the information available to them..
    No objective meaning needed here, the people have a word for something, but if we look deeper they are not truly identical (the water.)
  15. Feb 1, 2008 #14
    Shortly after I was born, my mother pointed to something I was looking at in the fruit bowl and said "Apple."

    She could have said "Ivory Tower", and from that day on, between my mother and I, an Apple would have been an Ivory Tower.

    The meaning is in my head, the word provides common ground for alluding to the meaning in my head. Common ground is only required in community as has already been said.

    If the twins in the original question where given the benefit of the authors increased knowledge about h2o versus xyz, and they believed it, then they might agree that only one of them should use the social meaning pointer "Water" and the other should choose another name. Resolution of meaning has also been mentioned.

    They are of course at liberty to continue using the same pointer because now they actually know what meaning is being refered to.

    I think the original argument made the point very well, meaning is in the head.
  16. Feb 2, 2008 #15
    I think we need to take a step back and define intension. In my book intension means "any property or quality connoted by a word, phrase or other symbol. In the case of a word, it is often implied by its definition." This applies to all words, not just to "predicates". A noun such as "water" also has an intension. As far as Oscar is concerned, the intension of "water" includes things like "a colourless, odourless liquid which at times falls from the sky", but H2O certainly does not figure anywhere within Oscar's intension of "water".

    Now, it seems that you (and Putnam?) believe there is more to the meaning of a certain nouns than simply their intension (whereas this is not true of predicates). Am I correct?

    We need to agree what you mean by "connotation" and "denotation". According to Wikipedia (http://en.wikipedia.org/wiki/Denotation), these terms are (in philosophy) interchangeable with intension and extension respectively.

    Thus, you are perhaps suggesting that the meaning of all predicates (and most nouns?) is given by their intension, whereas the meaning of certain nouns (nouns that you call "names") is given not by their intension, but instead by their extension?

    What you call the "definite description" in this case changes over time. You seem to be assuming that the intension of any idea is fixed and unchangeable, but there is no reason why this need always be true. If the object "the person over there holding the glass of liquid" drops the glass, then the object becomes "the person over there who was until recently holding the glass of liquid". The intension has changed.

    All you are doing here is imagining "other possible worlds" in which the intension (hence the meaning) of the words "Bill Clinton" would be different. For example "if he'd died aged 2" then he would not have been the father of Chelsea (ie your above suggested description of him would no longer be accurate). Our language still perfectly fits the descriptivist picture.
    Last edited: Feb 2, 2008
  17. Feb 2, 2008 #16
    I clearly said that I don't think names connote anything. Their meaning is thus what you would call their extension. What you call the "intension" of water comprises the superficial properties Oscar in fact used to fix the reference of "water", but those properties do not determine the extension or meaning. Oscar is able to acknowledge that something might not have any of those superficial properties but could turn out to have been water all along, and he can acknowledge that something could have all the superficial properties of water but turn out never to have been water at all (the case of XYZ).

    This, at least, is Kripke's intuition. Some people disagree. But merely stating the opposite intuition isn't compelling and doesn't refute the alternative viewpoint - so you don't refute Putnam by starting from the opposite intuition.

    More Bill Clinton... You say if Bill Clinton had died aged two, the meaning of Bill Clinton would have been different to how it is in the actual world. But how, then, do we match up the referents of Bill Clinton in the actual world and Bill Clinton in the other possible world? What is the content of a statement like "If Bill Clinton had died aged 2, he'd never have become president." Your theory suggests it's meaningless, because Bill Clinton (when I use the name) can only refer to someone who did become president!

    Don't say "it's meaningful because both are called 'Bill Clinton' in their respective worlds". I can imagine a situation in which Bill Clinton was actually called "Fred Clinton". If Bill Clinton had been called "Fred Clinton" and had died aged 2, he wouldn't have become president. By your theory this claim is meaningless too - Bill Clinton can't be called "Fred Clinton" by definition and can't not have been president by definition. But it isn't meaningless. You can comprehend what I mean. You do it by stipulating the presence of Bill Clinton in the possible imaginary scenario, then you change what people call him, then you kill him off, and then you imagine the consequences.

    In fact, you can change every superficial property of Bill Clinton in a hypothetical scenario and still refer to him! (This is the intuition Kripke is playing on.)

    Note that when I said "you will take a lot of convincing" I didn't mean to say "I will convince you". And indeed there's nothing wrong with holding on to different intuitions about naming, provided you get the alternative viewpoint (which is the more popular view in analytic philosophy right now). The best thing to do is to read Naming and Necessity. It's very good and not very long.
    Last edited: Feb 2, 2008
  18. Feb 2, 2008 #17
    (my emphasis) Interesting. Thus (according to you), the extension of a name gives us the meaning; whereas the intension of all other words gives us the meaning, and intension does NOT determine extension. Now, according to Putnam, intension does indeed determine extension. So whilst on the surface you seem to be arguing in favour of Putnam's analysis, you are actually taking a very different position to his.

    (Putnam actually argued that there are two "assumptions" commonly held about meaning - the first assumption is that, since grasping a concept is a mental act, understanding a word comes down to being in a certain psychological state;the second assumption about meaning is that the same intension always carries with it the same extension, ie intension determines extension. Putnam argues that no concept of meaning can actually satisfy both of these assumptions simultaneously, and he rejects the former and retains the latter - hence he arrives at his famous conclusion "meanings just ain't in the head").

    No, he cannot - unless he is prepared to redefine his concept of water (ie the meaning of the term "water" as far as he is concerned).

    That depends on how you define "to refute". My dictionary tells me that it means both "to prove false/wrong" and "to deny, to challenge". In the latter sense, I do indeed refute Putnam's argument.

    Why do they need to be matched up? The whole point is that they are two different worlds, there is no requirement or necessity to match up anything between these worlds. Indeed, there are an infinite number of possible worlds where Bill Clinton never exists at all - how do you match these up with the real world?

    Your statement is paradoxical, because it assumes what it denies. The precise wording you have used assumes that Bill Clinton is or was a president, and at the same time suggests that under certain conditions he might NOT have become a president. A paradox. Either Bill Clinton is/was a president, or he was not. Change the wording slightly, and it becomes meaningful. It is NOT meaningless to say that "If X dies before the age of 2, X cannot become president" (substitute any name you like, including Bill Clinton, for X).

    Thus to you the statement "there is a possible world in which Bill Clinton dies at the age of 2" is either meaningless, or false?

    I didn't.

    In the actual world, I agree. But there are infinite number of possible worlds where Bill Clinton never becomes president.

    I "get" the alternative viewpoint, I just think its misguided or wrongheaded. For an alternative viewpoint, one may care to read the following refutation of Kripke's intuitions in Naming & Necessity : http://www.andrewboucher.com/papers/nan.pdf

    (I really don't see what relevance the "popularity" of a viewpoint has - it used to be popular to believe the earth was at the centre of the solar system)
    Last edited: Feb 2, 2008
  19. Feb 2, 2008 #18
    Putnam has already rejected the intuitions with which you "refute" him before writing the paper. So to respond to that paper you should do more than just blindly assert the intuitions Putnam has rejected. Your argument starts with "meaning derives from knowledge and understanding", so Putnam wouldn't concede your starting premise.

    I have been unclear then. The reference is the extension of the natural kind. The superficial properties fix the reference. According to Putnam the natural kind in question has its own intension which the speaker may well not know. I guess I agree with this too. So the intension and extension do contribute to the meaning, but the speaker may not know either. That's my understanding of Putnam's position. It sounds plausible enough to me.

    Putnam's notion of "intension" has nothing to do with superficial properties. Shininess, blueness and such do not feature in the intension of "water". What you call the intension is not what Putnam calls the intension. I admit that he thinks "essential properties" do constitute some kind of intension. This intension is in the objects, so to speak - we may not know it when using the word. This is why I prefer to say "names don't connote anything," where the connotation is something the speaker knows when using the word. But I am being vague here. My aim is not that you should change your position because of these points, but that you should take Kripke/Putnam seriously.

    In these possible worlds, how do you determine the referent of Bill Clinton? He doesn't have the superficial properties that supposedly determine the reference of the name (according to the description theory).
    Last edited: Feb 2, 2008
  20. Feb 2, 2008 #19
    More stuff.

    That's emphatically not the case according to Putnam. I agree that this is ultimately a battle of intuitions. So your challenge is to produce examples showing that your intuition is plausible. All you've done is said "No! I don't agree with that intuition!". This is not really an argument.

    Trans-world identity does matter for deciding whether a statement is necessary or contingent. That's what Naming and Necessity is about. Am I selling it to you yet? You always come back with loads of reasons why this book you haven't read yet must be a load of crap.
  21. Feb 2, 2008 #20
    I can see that there is a heady debate over points of view going on here. So I offer this as an example for your consideration. The example challanges the position of naming and identity.

    You may know it, since it comes from our old friends the ancient greeks.

    The Ship of Theseus:
    Over a period of years, in the course of maintenance a ship has its planks replaced one by one. Call this ship A. However, the planks are retained and themselves reconstituted into a ship. Call this ship B. At the end of this process, there are two ships. Which one is the original ship of Theseus.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: Epistemic Perspective : Meanings ARE in the head
  1. Epistemic Determinism (Replies: 20)