Quick question

While reading an influential text on direct reference, I found the following claim:

In the framework I have just sketched, a proper name is a word which must be used in a certain way, even though it may happen to be used in other ways.

I don’t know how to interpret this in a non-inconsistent, or non-contradictory, way.  I wonder, more specifically, how the quoted text differs from the following:

According to the rules we have just stipulated, you must come home every night at 8, even though you may come at other times.

Any proposals?


15 Responses to Quick question

  1. It’s consistent as long as the two uses don’t exclude one another. Compare: You must use the oven to heat food, though you may use it to heat yourself. That’s possible, as long as you are warmed by the heat given off from the oven.

  2. Edú says:

    Thanks Brian,

    When you say that the two uses shouldn’t exclude each other: do you mean there is some kind of topic-change between what the must and the may are supposed to talk about? If so then wouldn’t it be better to say:

    You must use the oven to heat food not yourself. But, of course, while using it to heat food you might be warmed.

    Notice, then, that what the ‘might’ allows results from observing what the ‘must’ demands. I doubt that this is what happens in the problematic quote above.

    The quote is about non-descriptive and descriptive uses of names. Let me reformulate my question then. Can we make sense of:

    A proper name is a word that must be used non descriptively, even though it may happen to be used descriptively.

    Could we follow your strategy and say that this means:

    A PN must be used non descriptively, though of course while using it non descriptively it may be used descriptively.

    Does that make sense?

  3. David Gawthorne says:

    Could ‘must’ be being used to express normativity while ‘may’ expresses a simple possibility? I guess it depends on ‘the framework just sketched.’

  4. Edú says:

    Hi David,

    I think you’re right. It’s unclear how to interpret the ‘must’ and the ‘may’. The problematic quote comes with the following footnote:

    [X] once objected to this account that it is normative rather than descriptive […] I reject this charge. The normative element which plays a crucial role […] does so only because it is integral to ‘language as it is’.

    Now, I guess the crucial part is ‘being integral to language as it is’. It suggests that the ‘must’ should be interpreted as saying: given the way that natural language is, names must be used in way F.

    If so, then the claim that they ‘may’ be used in way notF seems strange. It would have to be external to the way that natural language is. (Doesn’t that just mean ‘that’s not the way that natural language is’?)

  5. alexandra plakias says:

    Hmmm, I’m not sure what ‘integral to language as it is’ means, but I think David’s nailed it. Even if the ‘must’ expresses a rule that’s ‘integral’ to the language, there can be cases where the rule isn’t followed, either because it admits of exceptions (“‘I’ must come before ‘E’, although it may come after if it follows a ‘C'”) or because people are bad at using names and fail to follow the requirement (“speakers must keep their talks to thirty minutes, although they may happen to go longer, in which case it is the chair’s responsibility to cut them off”). The ‘may’ doesn’t actually give you a permission to do anything; it just describes possible violations of the requirement. (This may seem to contradict the claim that it describes language ‘as it is’, but if we wanted to describe, say, logic or math or English grammar ‘as it is’, we definitely wouldn’t try to account for all our students’ errors.)

  6. Edú says:


    I think that’s right. I still think there’s something odd because no one believes that descriptive names (e.g., Evans’ ‘Julius’ or Le Verrier’s ‘Vulcan’) are mistakes. These cases seem to be more like an aspect of language as it is rather than exceptions.

    But, I agree. David’s is probably the most charitable, sense-making, reading.

  7. Rafal says:

    One and the same word can have many uses. So I would read the claim:

    a proper name is a word which must be used in a certain way, even though it may happen to be used in other ways


    A proper name is a word for which
    (1) it is necessary that in one of its uses, it is use in a certain way, and
    (2) it is not impossible that there are other uses that it has.

    So (I’m not sure what the context of the quote is), for instance, you could consistently say that a word, in order to be a proper name, has to have a use in which it is treated purely referentially, even though it might have some uses where it’s taken to have some descriptional content.

  8. Edú says:

    Hi Rafal,

    I’m a bit less convinced about what you say. Your reformulation in (1) and (2) seems odd:

    (1) it is necessary that in one of its uses, it is used in a certain way.

    I don’t know how to interpret that. Do you mean the same as (1a):

    (1a) it is necessary that when used non-descriptively, a proper name is used non-descriptively.

    or (1b):

    (1b) it is necessary that when used in one way, there is a certain way in which it is used.

    Either way (1a) and (1b) seem useless to me.

    Your second reinterpretation, after (1) and (2) seems less odd. If I understand it, you want to say that for ‘X’ to be a proper name it must, at least, have a non-descriptive use, though it might have others.

    The problem I see with this reading is that it presupposes a contrast between names and words for which it is simply “impossible” that they be used non-descriptively. Are there any such words?

    Furthermore, it seems that on this second interpretation names that remain descriptive, e.g., ‘Jack the Ripper’ or `Julius’ are never names.

  9. alexandra plakias says:

    why not think that the second contrast is this: there are some words which can be used descriptively or not. others must have a nondescriptive use but can also have an (optional) descriptive one in addition to that. names are of the latter kind. in other words, why think that the contrast is in the ‘impossibility’ of non-names being used descriptively rather than in the ‘mandatoriness’ (for lack of a better word) of proper names having non-descriptive uses? (do ‘Jack the Ripper’ and ‘Julius’ have no non-descriptive uses?)

  10. Edú says:

    Thanks Alex,

    I think I understand your account. But I’m not sure. The idea is to distinguish between two kinds of words: those that must be used non-descriptively (call them “musties”) and those that need not (call them “maybies”). But your way of getting there confuses me.

    You say that maybies can be used both descriptively and non-descriptively. Then you say that musties can also be used descriptively and non-descriptively. So far, there’s no distinction.

    Then you add the distinctive feature: musties are those for which the non-descriptive use is a must. What does it mean to say that musties have a “must” added? Is this added normativity merely in theory? Because in practice that “must” has no impact: musties maybe used both descriptively and non-descriptively.

  11. dtlocke says:

    I’m with Brian.

    Compare: A baseball bat is an object that MUST be used to hit baseballs, even though it may happen to be used in other ways.

    As it happens, I don’t think the first part of this sentence is true. But still, it makes perfect sense. The first part says that an object couldn’t be a baseball bat if it weren’t used to hit baseballs. The second part says that baseballs bats may *also* be used in other ways. These two claims are perfectly consistent.

    I suspect this is what your author is getting at with respect to proper names—something simply couldn’t be a proper name if it weren’t used in a certain way, but still, proper names may *also* be used in other ways.

  12. Edú says:

    Hi Dustin,

    I’m not sure how to take your reading. I agree that it is false about baseball bats, and that the ‘must’ and the ‘may be used’ are not inconsistent. But I suspect this is not what the author is getting at.

    The author is trying to say that descriptive names are also names, even though they are used in that strange ‘descriptive’ way. Now, consider your alternative reading for proper names:

    ‘something simply couldn’t be a proper name if it weren’t used in a non-descriptive way, but they may be used in other ways.’

    If this is right then descriptive names that are never used non-descriptively (e.g., ‘Julius’ and ‘Jack the Ripper’) are never names.

  13. dtlocke says:

    “If this is right then descriptive names that are never used non-descriptively (e.g., ‘Julius’ and ‘Jack the Ripper’) are never names.”

    It seems to me (based on just the quote you’ve given) that that is exactly what the author is saying.

    BTW, which article/book are you reading?

  14. dtlocke says:

    Hi again, Edu. The passage you’re quoting comes from Recanati (1993), no? If so, the next line is:

    “A genuine proper name is defined (normatively) by what it demands; Russell thus speaks of ‘the direct use which (a proper name) always wishes to have.'”

    (Russell’s framework is the framework referred to as ‘this framework’.)

    It seems then that David is right: the ‘must’ is the normative ‘must’. Thus, logical consistency demands that we interpret the ‘may’ in some non-normative sense, which pretty clearly seems to be what Recanati intended.

  15. Edú says:

    Hi Dustin,

    Yes, the passage comes from Recanati 1993. And yes, it seems like the most plausible reading is to take the ‘must’ as normative and the ‘may’ as logical.

    I think that’s a proper reading of what Recanati wants to defend. What I doubt is something else: whether that makes sense. In other words, I’m wondering if the idea itself is acceptable.

    What puzzles me, as you can see by reading the footnote that comes with the passage, Recanati does not want to put any ‘external-to-language’ kind of normativity. He wants it to be about ‘language as it is’.

    Now, I take it that all the actual uses of proper names are possible in virtue of the nature of natural language. This is why the distinction between normativity and possibility becomes problematic: Recanati wants that normativity to be part of the nature of language.

%d bloggers like this: