Friday, July 13, 2012

One is the Loneliest Number and is Natural but Odd

There is an interesting post at the "Maverick Philosopher" on van Inwagen's argument for the univocity of 'exists'. I think Vallicella goes way too easy on the first premise of the argument:

Number-words are univocal in sense: they mean the same regardless of the sorts of object they are used to count.

This seems to me to need more precision to avoid a common misunderstanding of what it is to be univocal or equivocal; words are univocal or equivocal in use, not in the abstract. Taken strictly as stated, this premise would require us to say that number-words are never reasonably used in a metaphorical or in an ironic sense. But this is certainly not true. I can count ironically, e.g., if I were making fun of somebody for miscounting something. Those will not be univocal with ordinary counting uses, which are really being presupposed here. More seriously, however: while it involves a count in some form, and the use of the word can clearly be linked to ordinary counting uses, it is simply not at all obvious that 'one' in "One is the loneliest number" is univocal with 'one' in "One is the first odd natural number". Actually, it's pretty obvious that it isn't. We know, for instance, that the two do not have the same scope: you can have one item in a couple, but not in the sense in which it is the loneliest number.

The rest of the post is quite good, though.


  1. I have to take issue with your first point against Dr. Vallicella.  It is highly implausible that "two" has different senses in ironic and non-ironic uses: otherwise, there would be virtually no univocal terms, since every word can be used in an ironic sentence.  Wouldn't it make more sense to chalk the difference in ironic use up to pragmatics, rather than semantics?

  2. Good to see you're still reading.

    I think what you are calling implausible is in fact the only possibility, because I think you are making the common mistake. This is quite true: Every word can be used ironically, therefore in fact words cannot be univocal in themselves. Why would one think otherwise? It's uses of words, not words themselves, that are univocal. The same is true of equivocation, though. There is no such thing as an equivocal word in the abstract. Whether something is univocal or equivocal depends on how it is used, and therefore we cannot say that a word is equivocal or univocal unless we know the contexts in which the words are used. 'Two' obviously doesn't have exactly the same meaning if I use it  to indicate three objects as I make a joke about someone's skills as it does if I use it to count in the ordinary way in an ordinary context; irony is not merely stating flat-out falsehoods about something, it is using words to mean what is in some way the inverse of what they mean in most relevant contexts.

    It clearly has to be something like this. Whether something is univocal or equivocal depends entirely on its comparison to something else: in particular, two uses of a word are being compared, and saying that the word is univocal, or (what comes to the same thing) used univocally in that context is to say that both uses share the same meaning. If a word were the perfect hapax legoumenon, being used once only in the history of the world, ever, it would make no sense to ask whether its meaning were univocal or equivocal; there would be nothing to compare it to so that we could say it has the same meaning or a different meaning.

    To put the same argument in a different way: if terms could be univocal in the abstract or in themselves, there would be words that nobody could equivocate with, because they would be simply univocal. But this is obviously not possible. Therefore no terms are univocal or equivocal simply in themselves, but only in their uses in particular contexts (as compared with other uses).

    And this fits with how we actually deal with the danger of equivocation: we look at how the words are used in actual contexts. We don't classify them into groups of univocal words and equivocal words.

    Since the pragmatic/semantic distinction can only be made in terms of variances and invariances in meaning, it can only be made after we have determined what the variances and invariances in meaning are -- that is to say, it is downstream from the question of whether we have univocal or equivocal meaning.

  3. Good to still be reading!

    Very well, I concede that only two or more uses of a term can be equivocal or univocal; I still don't grant that sarcastic and literal uses of "two" have different meanings.  If someone were to correct me after ironically saying "There are obviously two pens on the table" (there are in fact three), telling me that my statement was false, it would be inappropriate for me to reply "No, my statement was ironic, and thus true, though the same sentence uttered in another context might be false."  Whatever the proper response, it would not involve suggesting my comment to be true or my critic's reply to be false: my statement was obviously false, and my critic was obviously correct.  This suggests that the truth conditions for "There are two pens", and thus the meaning of "There are two pens", and thus the meaning of "two", are invariant across literal and ironic uses of the word or sentence. So, as far as I can tell, irony just is "stating flat-out falsehoods about something."

  4. I'm not sure I follow your example; of course it would be appropriate to say that -- it just would usually be a waste of time, for the same reason it is usually a waste of time to go into detail explaining punch lines. If someone honestly can't figure out the irony, or if they are not privy to the information recognizing it, there's no going to be much point to explaining. But if you give a literal paraphrase of your ironic statement, "This is definitely not two but what so-and-so calls 'two'" would be a perfectly fine literal paraphrase of ironic "This is two" given the context. And if that's how you would put its meaning in other words, its actual meaning has to be something like that. The most natural way of taking that, in turn, is as showing that in this particular context 'two' simply means the opposite of what it would usually mean, because I'm deliberately using it that way.

    Irony is obviously not stating flat-out falsehoods about something because nobody treats it in the same way that they treat statements of flat-out falsehood; indeed, people go out of their way to make a distinction between the two.

    The only evidential difference between irony and literal sense is that the latter is common to more contexts; there's nothing about the words that shows that the latter has any more claim to be considered the meaning than the former. Literal meaning doesn't magically attach to words; it's just what the words are usually used to mean. But we can use words to mean anything, and often be understood: thus all figurative speech, which are just ways of rationally deviating from the more common meaning, give the word a new, although related, meaning. Irony is simply the most extreme example, or rather (since irony can often be more subtle than in our example)  the kind of figurative meaning that is capable of being the most extreme, because in ironic contexts you can sometimes convey your meaning by saying what would usually mean the opposite.  But there is absolutely no reason to think that the literal meaning has any sort of privilege over this, beyond its being common to more contexts. And we find regularly that irony shades into literal sense and vice versa. 'Specious' originally meant beautiful, and now means 'good only in appearance'; it's an ironic usage that spread until it became the literal usage.

    So I would say the problem with your argument is that it assumes that literal sense has a special privilege over figurative sense, beyond being more common, to such an extent that figurative sense is not actually any sense at all. This is an arbitrary assumption. When we actually look at how we mean things with words, we find that there is no such hard and fast boundary, and that we can mean anything with words -- which makes sense, because they are arbitrary -- and the only real constraint is our grasp of what is expected and unexpected in using them to mean things.

  5. I do assume that literal meaning has a privileged position.  Specifically, I think that "literal" is a redundant adjective, like "real" or "simpliciter"; thus, the literal meaning of a word, phrase, or sentence just is the meaning of a word, phrase, or sentence.  This comports well with the way we usually discuss meaning: if asked for the meaning of "The President is Russian-American", we give its literal meaning, and similarly for the meaning of the word "anaesthetic".  This is exactly what we would expect if literal meaning were just meaning simpliciter.

    You seem to suggest, to the contrary (and correct me if I'm reading you wrong), that literal meaning is simply what is most often meant by a word, phrase, or sentence.  But this interpretation is subject to counterexample: "as big as the sun" is probably used most frequently as an hyperbole and "a pitiless block of ice" as a metaphor, but their literal meanings (of the same size as the sun and a slab of frozen water exhibiting no pity, respectively) differ from what is meant by these more common uses.  So literal meaning is not just the same as what is meant most commonly, whereas reducing literal meaning to meaning simpliciter would explain why literal meaning is usually aligned with what is most commonly meant by a word, phrase, or sentence.

    Going back to the "There are two pens on the table" scenario, I would revise my previous description somewhat.  It isn't just that it is seemingly never appropriate (by my lights) when challenged in such a situation to say, "No, my statement was true, since it was ironic": it moreover would seem appropriate to say, "Of course the statement was false: it was ironic."  So, on the most straightforward interpretation of these intuitions of acceptability and unacceptability, it is wrong to call the statement true and correct, albeit somewhat unusual, to call it false, which is to say that the statement would be a falsehood.

    As for whether we treat irony differently than we do outright falsehood, it seems to me that neither of us can decide the matter one way or another without begging the question.  If irony is straightforward falsehood, then there is a species of flat-out falsehood treated the way irony is.  We can't appeal to whether we treat flat-out falsehood differently than irony unless we're already decided on whether they differ, which is precisely the point at issue.

    Finally, while I agree that words have meaning, in some sense, arbitrarily, I disagree "that we can mean anything with words," or at least that words can mean anything.  If this were so, then it would be in vain ever to ask for the meaning of a word or sentence: either the question would be elliptical for something like "What is most commonly meant by this sentence or word?" (which seems like an ad hoc stipulation), or the correct answer would be, properly speaking, "Anything at all."  But we can give the meaning of a word or sentence, literally and without implicit addenda to the question, without being uninformative and giving answers like "Anything."

  6. Hi, Leo,

    Your counterexamples aren't counterexamples, because they are obviously composite, both to eye and ear: a pitiless block of ice is a phrase  whose subordinate words are still recognizable, and which have obviously more common uses that have nothing to do with their operation in that one particular phrase. Thus a literal sense and a figurative sense can still be distinguished, and easily so, in terms of what is usual and expected. Obviously 'block of ice' is more often going to mean block of ice than it is a state of character. Contrast this with the etymology that people once gave for 'sincere' : sincere in this etymology meant 'without wax' (i.e., not touched up cosmetically to hide defects) in Latin, but, of course, nobody meant this when they used the word, because there was no general expectation of still using the 'cere' part in the way Latin 'cera' was used. People tend to question the etymology today, but it's not as if it were any sort of implausible account; this thing is always happening. The literal-privileged account can't make any sense of actual language dynamics: it cannot explain dead metaphors, for instance, despite the fact that all languages are littered with them.

    Of course, it would seem appropriate to say "Of course the statement was false: it was ironic"; the most natural explanation of irony to someone who doesn't grasp the irony is to refer to the literal sense. To what else would you refer in order to explain it to them? And the statement is false in the literal sense of the words. We could equally well say, "Of course the statement was false, if taken literally: it was ironic, however". And that would mean exactly what your version of the response means. Therefore your conclusion does not follow. (It also has to already assume that the literal sense has a special privilege beyond being more common and thus more expected.)

    On the distinction with flat-out falsehood: No, people explicitly distinguish irony from falsehood quite often. It isn't a hypothesis to save the phenomena, the distinction is one of the phenomena of irony. The most natural explanation is that they are, in fact, distinct; if there's another explanation, you've not given it.

    Of course we can mean anything with words: what do you think people do with codes? If I have a codebook in which 'the butler' means 'the assassin' and 'at the door' means 'is in the building', then when I say, "The butler is at the door" the meaning is, whatever it may look like, the same as the literal sense of "The assassin is in the building."

    Your argument in the last paragraph does not make any sense to me. The meaning of the word is whatever we use it to mean. When we ask about words in isolation, we are asking what the word would most often be expected to mean -- that's why we can look it up in a dictionary, because dictionaries explicitly and obviously track the most common uses of words. But for a word in actual use, we would either ask the person what they meant or try to piece it together from clues.

    The bottom line is that your position (1) cannot explain actual language dynamics, such as the change of the metaphorical into the literal and back again, despite the fact that this is an extremely common phenomenon; and (2) does not make sense of many of the ways people use figuratjve speech, such as distinguishing irony from falsehood. Besides that (3) it involves assumptions that seem either purely arbitrary or based on excessively limited examples. It's a common sort of position -- Davidson's atrociously bad but very influential theory of metaphor makes precisely the assumptions you do, in addition to others that make his account even worse.

  7. It occurs to me that this very old post, while not dealing directly with the issue at hand, does get into issues that are relevant:

  8. I'm not clear on what you object to in my counterexamples.  Even if their "literal sense and a figurative sense can still be distinguished, and easily so, in terms of what is usual and expected," their literal sense isn't just the same as what they are usually used to mean, which is what I took you to be suggesting.  If you aren't taking literal meaning meaning to be identical with what is most commonly meant by a word, phrase, or sentence, however, I'm unclear on what you do take literal sense to amount to.

    I'm also failing to see why my argument from what is acceptable in explaining irony is invalid: even if it's true that my original phrasing of the ironic speaker's self-explanation means the same as your paraphrase, the point still stands that it is intuitively acceptable to call ironic statements false, and not clearly acceptable to call them true, which seemingly tends to support my view over yours.  If I'm missing something, though, I'm happy to be corrected.

    As for flat-out falsehood, I grant that irony is unusual species of falsehood in that in ideal circumstances it is understood by all interlocutors that the ironic statement is false, whereas elsewhere it is usually unknown, or intended to be unknown, by some party the false statement is a falsehood.  Does that match the distinction you are trying to draw (or think that we normally draw) between irony and flat-out falsehood?

    A distinction that has guided me throughout this conversation, and possibly a distinction you would reject, is between word or sentence meaning (meaning*) and user meaning or meaning by (meaning**).  We can certainly mean anything by a word — or anything else, for that matter.  If, for example, I have a certain involuntary and idiosyncratic reaction to foul smells, and at a theatre performance with you I mimic that reaction while looking in your direction, I presumably mean by my action that I dislike the performance.  But that is not the sense of meaning according to which we can ask for the meaning of the word "clandestine", or what "The President is supercilious" means: for one, when asking for meaning**, we generally must specify an utterer, whereas we needn't for meaning*.  This second sense of meaning is what we appeal to when we ask what a word or sentence means, and also what sets the truth or satisfaction conditions for a sentence or word, since we obviously do not ask whether my mimicked reaction at the theatre is true or false.  And my claim is that in this second sense of meaning, the sense in which we ask for the meaning of "clandestine", the meaning of a word is the same in both its literal and ironic use.  Does this make sense?

  9. Hi, Leo,

    Words, phrases, and sentences are very different, and there is no reason to think that the literal sense is determined for each in the same way, so I was protesting your switching around among them, particularly since we were talking about words. It's quite clear, for instance, that literal sense is primarily attributed to words -- the literal sense of phrases or sentences is quite simply the sense they have if the words composing them are taken literally. (Figurative language is arguably more complicated, because it simply involves some substitution for literal sense in some way.) After all, how else can we figure out what a sentence literally means, except by taken its words literally? -- The one complication here is that words can grow together, as in the verb 'atone', and that there may be intermediate cases where the sense is ceasing to be figurative and becoming literal. But that there would be cases like this is exactly what one would expect given that there is no absolute division between literal and figurative, and that whether a sense is literal or figurative depends on normal expectations and not on the sense itself, which is the view I've been arguing for. And as I've noted, you have no explanation of this kind of language change. Currently 'pitiless block of ice' has not grown together -- as a matter of practical phrase-making, people don't generall think of 'pitiless block of ice' as a unit, and are perfectly capable of seeing that 'pitiless' and 'block of ice' are detachable, and always expect people to know the meaning of the phrase not holistically but by knowing what 'pitiless' and 'block of ice' mean. All show that the expectation is that 'block of ice' will normally mean a block of ice, not a person or state of character, regardless of its adjectives. That is, the phrase does not function as a single unit of meaning but as a composite. But if the whole phrase is used often enough that people come to treat it as a single unit, then it will start to shift as the figurative sense becomes a 'frozen figure of speech' or 'dead metaphor', and thence becomes the literal sense. It's happened before. (In fact, it's easy to come up with examples: flower bed, atone, broadcast, understand -- they were all originally figurative phrases that began to be treated as single units of meaning with a literal sense that was exactly the same as their original figurative sense. A related change is one like the collapse from the figurative expression, 'riding a hobby-horse' to 'hobby-horse' to 'hobby', whose literal sense is exactly what the figurative sense of 'hobby-horse' was in the original figurative phrase.  And precisely one of my points above was that I can account for these sorts of changes. You can't.)

    If my literal paraphrase, "<span>Of course the statement was false, if taken literally: it was ironic, however" is just as good as your paraphrase, "</span><span>Of course the statement was false: it was ironic", and, indeed, doesn't actually do anything but specify it, then it is not true that irony is falsehood; it is at most true that the ironic statement would be false if it were taken literally and not ironically. But, of course, it's an ironic statement, not a literal one. (It's also not possible to draw the conclusion "it is not clearly acceptable to call them true" from this kind of example, which tells us only about one way in which we would call them false, namely if they were taken in some sense other than that intended. But surely it's not at all surprise that something can be false if taken in a sense other than the sense in which they were intended; that can happen with the most unexceptionable statements.)</span>


  10. ...cont'd

    I don't think all forms of irony involve caring at all whether other people take it as false or not: whether or not that's of interest simply depends on the context.  In fact, I think in many cases people are ironic precisely in the hope that people will be stupid enough to take it literally and not ironically -- they are ironic in the hope that people will misunderstand their meaning. And sometimes that will mean not taking it in the sense in which it is true. But on your sense, there is no sense in which ironic statements could possibly be misunderstood: their meaning would just have to be whatever they happened to say, taken literally.

    Further, irony cannot merely be a kind of falsehood because not all irony is false when taken in the literal sense; it can be accidentally true even if taken literally. This is a general feature of figurative language. If I tell someone, "You live in a glass house, so you shouldn't throw stones," it may be true that they live in a glass house, but this true-in-the-literal-sense is incidental to the actual, metaphorical sense. So with irony: irony requires some inversion of expected sense, but it does not require an inversion of sense that would affect the truth value. Take the old story about the Dutch Calvinist woman who was hiding Jews from the Nazis, and finds her house being searched by the SS. One of them sternly asks her if she is hiding Jews anywhere, and she says tartly, "Of course, I always hide Jews under my kitchen table." He's seen her kitchen table, which is a bare little wooden thing, so he roughs her up, and then they all leave. After which she goes to her kitchen, moves her kitchen table, and pulls up the boards under where her kitchen table had been to get the Jews out so they can find somewhere else to hide in case the Nazis come back. It was a true in the literal sense, but that's incidental to the meaning.

    Another of the phenomena of irony that your position does not capture, besides the fact that ironic uses can become literal uses without themselves changing (as noted in an above comment with 'specious'), and besides the fact that ironic statements are often distinguished by competent language users from outright falsehoods, and besides the fact that ironic statements can be true by accident without ceasing to be ironic, is that everyone treats irony as a kind of sense just as they treat literal sense as a kind of sense. Everyone recognizes that statements can sometimes be taken in an ironic sense rather than a literal sense, or that you can misunderstand what is meant because it has to be taken in a different sense. The most natural explanation for this is that ironic sense is, in fact, a different sense from the literal, in some way; but you need an error theory for why everyone talks as if ironic sense were its own kind of sense.

    I have no idea what your meaning* is. When we ask the meaning of the word clandestine, we are asking what people usually or normally or generally mean by it. In fact I can (and often do) use exactly those words. And what else would we be asking? There aren't any meanings magically disconnected from agents capable of meaning things, and using things in order to mean things.

  11. Ah -- came up with a possibly better example of an ironic statement that isn't intended as a falsehood. In Hume's Dialogues Concerning Natural Religion, Cleanthes realizes that Philo's affirmation of the claim that the nature of God is unknowable or incomprehensible is ironic but Demea does not. Cleanthes, of course, is right. But what makes it ironic is not that Philo thinks that God's nature is not unknowable, but that Philo and Demea are taking such terms in different -- and mutually exclusive -- senses. (And Philo and Cleanthes know it, while Demea does not.)


Please understand that this weblog runs on a third-party comment system, not on Blogger's comment system. If you have come by way of a mobile device and can see this message, you may have landed on the Blogger comment page, or the third party commenting system has not yet completely loaded; your comments will only be shown on this page and not on the page most people will see, and it is much more likely that your comment will be missed.