Digging into a compound

« previous post | next post »

I recently stumbled upon the following sentence (here, for the curious):

After Windows 7 comes out in October, will Microsoft somehow force us XP users to stop using it?

I am able to figure out from context that the "it" at the end of this sentence is supposed to refer to "(Windows) XP", but no matter how many times I read the sentence, I stumble on that intention. The problem, as I see it, is that there's no noun phrase in this sentence that refers solely to "XP", and the pronoun "it" must be coreferent with a noun phrase. I explain a little more below the fold.

First, here's a relatively detailed syntactic tree for the sentence, with the intended-but-ungrammatical coreference between "it" and "XP" (and the nodes above these terminal items) indicated with identical subscripts. (Click on the image for a larger, more readable version.)

As you can see from the labels in the tree, the noun phrases (NPs) in this sentence are "Windows 7", "October", "Microsoft", and "us XP users" — as well as the "it" in question, of course. Others with more syntax bona fides than me may disagree with my classification of "us" in "us XP users" as a determiner, but I'm fairly certain that we can all agree that "XP users" is a compound noun, consisting of two nouns, "XP" and "users". There's no noun phrase corresponding to "XP", and thus the only thing that the "it" can grammatically corefer with is "Windows 7" (the most obvious reading that I inevitably stumble to), "October" (infelicitous but still grammatical), and "Microsoft" (more felicitous). (Of course, the "it" can't corefer with "us XP users" because the former is third person singular and the latter is first person plural.)

The mere fact that it is ungrammatical for the "it" to corefer with "XP" in this sentence should be enough to show that the "XP" part of the sentence is not a noun phrase, but we can also apply further, fairly standard constituency tests to "XP" to convince ourselves. Let's start with the grammatical clause "Microsoft will force us XP users to switch to Windows 7" and apply some movement tests. (Note that the last of these is technically a substitution + movement test.) These tests clearly show that movement of the compound-internal noun "XP" alone is ungrammatical, but movement of the entire noun phrase "us XP users" is fine.

Constituency movement tests
("*" indicates ungrammaticality; "[__]" indicates the source of movement)

  • Fronting:
    *XP, Microsoft will force us [__] users to switch to Windows 7.
    (cf. Us XP users, Microsoft will force [__] to switch to Windows 7.)
  • Clefting:
    *It is XP that Microsoft will force us [__] users to switch to Windows 7.
    (cf. It is us XP users that Microsoft will force [__] to switch to Windows 7.)
  • Pseudo-clefting:
    *XP is what Microsoft will force us [__] users to switch to Windows 7.
    (cf. Us XP users are who Microsoft will force [__] to switch to Windows 7.)
  • Passivization:
    *XP will be forced us [__] users to switch to Windows 7 by Microsoft.
    (cf. Us XP users will be forced [__] to switch to Windows 7 by Microsoft.)
  • Wh-questioning:
    *What will Microsoft force us [__] users to switch to Windows 7?
    (cf. Who will Microsoft force [__] to switch to Windows 7?)

But all this now raises a question: why do the ungrammatical sentences here sound so much worse than the original example that I'm claiming is also ungrammatical? Not being an expert on such matters, I'll merely point to some possible avenues for investigating such questions. On the one hand, it could be that the original example sounds a bit better than the ungrammatical examples above because there are at least two grammatical and felicitous ways to interpret the original example; that is, the intended interpretation could be strictly ungrammatical but the simple availability of the other options makes it sound relatively OK. Alternatively, it may be that violation of the rules for pronominal coreference is less consequential than violation of movement rules. (See here for another type of example of this type.) Both of these possibilities might also explain why someone wrote the original example, but it's of course also possible that the writer's correctness conditions simply differ from mine (and yours, if you agree with my judgment of the example).


  1. Coby Lubliner said,

    August 3, 2009 @ 5:10 pm

    Would you have a problem if "XP users" were replaced by "users of XP"?

    [(EB) No, I wouldn't — and, as Lee Morgan notes below, that would be because "XP" would be an NP in this case. ]

  2. Electric Dragon said,

    August 3, 2009 @ 5:18 pm

    What about "Will Microsoft somehow force us XP users to stop using it after Windows 7 comes out in October?"
    That sounds perfectly ok to me and still has the "it" referring to the XP part of "XP users".

    [(EB) You're right, that's much better, and I should have thought of manipulating the original example that way. As GKP points out further below, it seems that coreference with "XP" becomes easier given that there are fewer NPs for "it" to refer to strewn along the way. ]

  3. Lee Morgan said,

    August 3, 2009 @ 5:36 pm

    In "users of XP", XP corresponds to an NP so there is no grammatical issue with the sentence.

  4. JAK said,

    August 3, 2009 @ 6:01 pm

    If I may digress a little, most of the time ungrammatical sentences make me pause as I read them as they break the flow somehow. But this example reads like a perfectly good sentence to me until you pointed the error and I re-read the sentence to confirm.

    [(EB) As I suggest, perhaps too obliquely, in the original post, "it may be that violation of the rules for pronominal coreference is less consequential than violation of movement rules". ]

    Maybe it is because of the familiarity with the subject or maybe I am used to these sentences from the countless IT blogs that I come across everyday.

  5. Rubrick said,

    August 3, 2009 @ 6:42 pm

    I think it's the logical tie between "users" and "using" that helps make this work. In fact, I think the other possible (but wrong) referents actually make things worse. If the sentence had been simply "Will Microsoft make XP users stop using it?", I'll bet it would have passed unremarked ("Microsoft" being too unlikely a referrent to have much brain-grab).

  6. James D said,

    August 3, 2009 @ 6:53 pm

    It looks to me as if the problem lies in the analysis rather than the sentence. The sentence reads naturally; the analysis tries to force unnatural meanings that feel rather too much like Latin.

    [(EB) Latin?!? ]

    I'd try comparing a simpler sentence:

    "Force us XP users to stop using it!"

    There's absolutely no doubt as to what "it" is there.

    [(EB) Yes, you're right. But the mere fact that you had to come up with another example to prove your point shows that there's something up with the original example. My analysis may be wrong, but my reaction is not (even if others don't share it). ]

  7. John Cowan said,

    August 3, 2009 @ 7:04 pm

    I agree that XP is not an NP in this sentence, but I think your conclusion is unwarranted. The usagists have discussed such "missing antecedents" extensively. Follett's Modern American Usage (1965), for example, has a lot of examples of this sort of thing, all of which he dislikes (supposedly problematic pronouns in italics):

    They thought they could browbeat the lawyer, but she refused to take it seriously.

    Don't spit on the floor [railroad station sign]. Spit on the wall and watch it run down [appended graffiti].

    In the old days, no shark fishing took place, and they were caught only occasionally.

    Washington State is the biggest per capita small-boat market in the nation. Many of those are built in Anacortes.

    An Oregon draft bill to limit the use of a pesticide there has drawn opposition from farmers.

    She has loved everything Scottish since her childhood friendship with a man of that practical, plain-spoken people.

    I submit on the basis of this evidence that there simply does not exist any absolute requirement that it and related pronouns corefer with an overt NP. At most it is what Geoff Pullum calls a politeness constraint, like putting the second person before the first in coordination.

    [No; it was Jim Hurford who proposed that putting the first person pronoun last in a coordination is a politeness constraint. But I have used the term "etiquette" in talking about dangling modifiers, and I agree that this looks like very much the same sort of thing: the examples above make it clear that people do this too much for us to say plausibly that their unconscious knowledge of the syntax of their native language bans it. Rather, they commit a mild (and usually quite understandable) discourtesy when they force a hearer or reader to dig deep into some NP to try and ferret out an antecedent for a pronoun, ignoring syntactically better-placed NPs along the way, to get a coherent meaning. Eric basically has the constraints right, but John is right that we ought to start looking at phenomena like this in terms of more labile and negotiable desiderata: it's not about a black/white distinction between what the grammar permits and what it absolutely blocks. —GKP]

    [(EB) Right, and this was my intent with the "it may be that violation of the rules for pronominal coreference is less consequential than violation of movement rules" part of the original post. ]

    And if clearly acceptable sentences are rejected by some grammar as ungrammatical, so much the worse for that grammar.

  8. Neal Goldfarb said,

    August 3, 2009 @ 7:23 pm

    I don't see a problem with it as written. The it in stop using it needs an NP to corefer to. An NP is by definition an XP. QED.

    [(EB) For the nonlinguists struggling to understand Neal's point here: in discussions of phrasal syntax, "XP" refers to any kind of phrase (noun phrase, prepositional phrase, verb phrase … X phrase). Ha-ha. ]

  9. Doug said,

    August 3, 2009 @ 7:28 pm

    I stumble on the XP sentence in the same way; it seems ungrammatical to me too. However, I can think of a number of sentences in which a pronoun referring to part of a compound* sounds perfectly fine:

    1. I'd love to go mammoth-hunting, but unfortunately they're extinct.

    2. "It's baseball season!" — "Who cares? It's a dull game."

    3. Zealous Obama-haters continue to insist he wasn't born in the United States.

    What all this means, I don't know.

    *If these are not compounds, or aren't compounds in the same way as "XP users", forgive me. I'm a non-linguist.

    [(EB) These examples are spot on, and you're right that they sound better. Refer to Electric Dragon's comment further above. ]

  10. Ed said,

    August 3, 2009 @ 7:58 pm

    I'm a non-linguist, with a casual interest, and three of the six sentences offered by John above are off-putting to me. The shark-fishing and Washington state sentences sound totally wrong, and the everything Scottish is a little weird to me also.

    Those first two I would say sound ungrammatical for the reason discussed in the original post; the last one, I don't think is a lack of a coreferent noun phrase. I think that "everything Scottish" is the intended referent, but I think where the sentence falls down is when it substitutes a part for a whole; that is, "that practical, plain-spoken people" is only a subset of the group "everything Scottish". There's probably a technical term for this type of agreement, but I don't know what it is. Different from a completely missing noun phrase to me.

  11. John Roth said,

    August 3, 2009 @ 8:53 pm

    It's definitely a bad phrasing as far as my ability to understand it.

    I think a different kind of analysis might be called for. My copy of Miriam-Webster's Concise Dictionary of English Usage suggests that there are three factors that go into determining the referent of a pronoun: grammatical correspondence, notional correspondence and proximity.

    In that sentence, XP is buried inside of a noun phrase (us XP users), while Windows 7 isn't. If we hoist XP out of the noun phrase, both grammatical correspondence and proximity would accept it, and notional correspondence would accept either Windows 7 or XP.

    So I'm obviously not hoisting it, and a bit of introspection shows that's the case. The initial parse simply doesn't find 'XP' as a possible referent, so it ties 'it' to 'Windows 7,' and that fails reasonableness checks. A close reading shows what is meant, but that's a conscious slice, dice and reassemble process.

    John Roth

    [(EB) This is a perfectly reasonable alternative to my analysis in the post. However, as pointed out by Coby Lubliner in the very first comment, changing "us XP users" to "(us) users of XP" improves the sentence dramatically (for me, at least), meaning that the NP vs. N category distinction must be relevant in some way. Add that to your analysis, and you've basically got mine. ]

  12. Jonathan Rubinstein said,

    August 3, 2009 @ 9:17 pm

    It is ungrammatical of course and infelicitous. After all it is Microsoft. And they will force us all to abandon XP simply by declining to support it. Now that sentence raises a host of issues does it not, but all users of computers who read it understand precisely what it means.

  13. John Lawler said,

    August 3, 2009 @ 10:21 pm

    There is a literature on this, btw.
    Go to Google Scholar and enter "anaphoric islands".
    There are some strange facts.

    Bill's parents are dead and he misses them.
    *Bill is an orphan and he misses them.
    Every Kentuckian I know wants to be buried there.
    *Every Hoosier I know wants to be buried there.
    I always thought the '1812 Overture' was about the American war that year.

  14. risma said,

    August 3, 2009 @ 10:43 pm

    I stumble over the sentence every time too.

    I'm curious: what's different between those of you who accept the sentence as is and those of us who just can't? I'm a math/computer geeky person, always considered myself stronger on analysis than synthesis; how about you?

  15. Ellen said,

    August 4, 2009 @ 12:00 am

    Rereading the sentence now, it sounds fine to me. But when I first read it this morning, it seemed to be asking if Microsoft will ask people to stop using a product that hasn't come out yet. I had to reanalyze to get what was meant. But, having done that, it's easy now to read it and read what was meant.

  16. Jean-Sébastien Girard said,

    August 4, 2009 @ 12:36 am

    I think the test fails not because "XP" is not a noun phrase, but rather because if you apply a structure that _does_ have it as a noun phrase (i.e. replacing "XP users" with "users of XP"), those tests cannot be applied because of some constrains whose name escape me at the moment (my only–I'm in translation–structural grammar class was something like five years ago).

    [(EB) You're right, and that would be the A-over-A constraint. Thanks for the correction. ]

  17. Rob Van Dam said,

    August 4, 2009 @ 2:48 am

    Despite having done antecedent tagging in a large corpus, I don't think I had previously encountered the rule that a pronoun must reference an NP. The sentence did not give me pause initially by I can see how the structure might cause some hesitation. I think one possible explanation would be to either

    1) weaken the NP coreference rule to state that a pronoun can reference any constituent phrase that 'could be' an NP

    2) consider that some NP constituents get garden-path elevated to an NP-like status based on a needed semantic interpretation (here because no other syntactic valid reference functions well).

    I'm not sure not if #1 and #2 are all that different actually. Notice the very different interpretation of

    "After Windows 7 comes out in October, will Microsoft somehow force us XP users to start using it?"

    There is no syntactic difference and yet the reference has shifted.

    On the other hand, it may just be that this counts as a type of exophora. By mentioning XP, the author has established that it exists in the realm of objects relevant to the discourse, much the same way that pointing allows me to say things like "Can you hand it/that to me?" without needing to explicit declare the object in question.

  18. Chris Lance said,

    August 4, 2009 @ 4:24 am

    "What will Microsoft force us users to switch to Windows 7?"

    That sounds just like one of those catechism questions that James Joyce put into Ulysses—pushing the boundaries of grammatical language, but not quite going beyond them.

  19. NW said,

    August 4, 2009 @ 4:51 am

    I'm guessing it's a combination of pragmatics and parsing efficiency. The initial PP sets a full NP as the topic. This is the only antecedent candidate that makes perfect sense grammatically and semantically. The parser chooses that. In the absence of any such (if say 'October' was the only NP available), the parser would have to abandon its hypothesis and look for referents inside nominals. I once (ruthlessly) edited a sentence where the antecedent was the country "Italy" inside the attributive adjective 'Italian'.

  20. Stephen Jones said,

    August 4, 2009 @ 5:48 am

    It's perfectly felicitous, and as it didn't and doesn't cause a double take, probably quite grammatical.

    [(EB) What Ellen said. ]

    It's obvious that a pronoun can refer back to part of a noun phrase.

    'So are all us beer drinkers going to be deprived of it?"

    "Will those amused by Pullum's petulance, be hearing more of him?"

    I can't see how these phrases are any different from the one in the example.

    [(EB) See Matthew R.'s comment below, and for more detail, see Arnold Zwicky's. ]

  21. Paul O'Brien said,

    August 4, 2009 @ 6:01 am

    I agree, I don't see the problem in having a pronoun refer back to a noun used as a modifier. Perhaps the stumbling block here is that the sentence is ambiguous, so you have to reject a couple of other possible meanings in order to get to the right one. But it doesn't read oddly to me at all.

    [(EB) This would not explain the difference between the original sentence and Cory Lubliner's suggested rephrasing (changing "us XP users" to "(us) users of XP") in the very first comment above. ]

  22. Faldone said,

    August 4, 2009 @ 6:23 am

    I think that "it" refers to "Microsoft". The only question is whether Microsoft will be forcing users to switch to Mac OS or Linux.

  23. HunterT said,

    August 4, 2009 @ 9:13 am

    There is nothing about this sentence that is inherently ungrammatical, though I can see why people who study coreference for a living would imagine it to be.

    [(EB) I don't study coreference for a living. What are you implying? ]

    I didn't do a doubletake on first (or subsequent) readings, but it does make me wonder how grammatical judgements are influenced by the judgers' awareness of grammatical "rules." (Crossover territory between prescriptivists and syntacticians?)

    [(EB) Follow the last link in my post to find out how you're very wrong here. ]

  24. Matthew R. said,

    August 4, 2009 @ 9:13 am

    "Will those amused by Pullum's petulance, be hearing more of him?"

    Well this one's different from the others, because 'Pullum' is a full NP.

  25. Andrew said,

    August 4, 2009 @ 9:22 am

    A couple of examples from alleged student papers on Scottish history:

    'Six border chieftains were hanged in Edinburgh, and they were a lot quieter after that.'

    'James I was captured by the English on the high seas, and kept there for twenty years.'

    The intended referents are of course 'the borders' and 'England'. I think the reason these sentences stand out is that there in in each case another possible referent which gives a meaningful, though clearly absurd, reading: 'the chieftains' and 'the high seas'. So it looks as if we are less likely to accept pronouns with missing antecedents if there is another possible antecedent present.

  26. Brett said,

    August 4, 2009 @ 9:45 am

    I am afraid that in the original post, I see shades of the erroneous arguments used by the authors of the various writing guides which the Language Loggers so frequently decry. These arguments seem to take this form:

    Sentences like X are poorly written.
    They are poorly written because they contain the construction Y.
    Therefore, construction Y is ungrammatical.

    (The style guide version of this argument is not usually made explicit. The guides may also contain a slightly weaker conclusion, that the construction Y merely should not be used.)

    [(EB) There's a world of difference. Please follow the last link in the post. ]

    In this case, the example sentence is clearly infelicitous. The infelicity arises from the way the pronoun "it" refers to "XP," which is being used attributively. However, there is now ample evidence in the comments that pronouns can have attributives (and even more exotic objects) as their referents; a noun phrase is not required. Yet it is interesting that the role played by "XP" has a strong effect on how difficult it is to pick it out as the referent.

    [(EB) This last sentence (which I've underlined) pretty much summarizes the intent behind my post. ]

    At the same time, it is posible to have perfectly good noun phrase referents for pronouns and still produce something incomprehensible. Witness, "While Bert and Bill were fighting, he poked him in the eye." This is perfectly grammatical, just very badly written. The same is true of the original example, although what makes that sentence badly written is itself a linguistically interesting point.

  27. Zwicky Arnold said,

    August 4, 2009 @ 10:10 am

    Following up on John Lawler's comment: there's now a considerable literature arguing that constraints on anaphoric reference to parts of words are primarily pragmatic rather than structural (as NW suggests above). Much of this work is by Gregory Ward, in collaboration with various others (most recently Andy Kehler). See, for instance:

    Ward, Gregory, Richard Sproat, and Gail McKoon. 1991. A pragmatic analysis of so-called anaphoric islands. Language, 67:439–474.

    [(EB) Thanks for the explicit reference to Ward et al., Arnold — as Dan Scherlis's comment below notes, I has supplied a relevant link in my post. ]

    There are many different subtypes of anaphoric island structures; noun-noun compounds (like XP user) are one common subtype. Pronouns that pick up their reference from the first noun in such compounds (as in Eric's example) are acceptable to varying degrees, depending on features of the context and ease of processing. (I have examples from across the spectrum.)

    Stephen Jones:

    It's obvious that a pronoun can refer back to part of a noun phrase.

    'So are all us beer drinkers going to be deprived of it?"

    "Will those amused by Pullum's petulance, be hearing more of him?"

    The "structural" version of the Anaphoric Island Constraint requires that anaphors must refer back to *NPs*. Beer in beer drinkers is not an NP, just an N, so the structural version of the AIC says that the beer drinkers sentence is unacceptable, but (like Steve Jones) I have no problem with it; this is the sort of example that motivates giving up the AIC as a strict structural constraint.

    [(EB) Though it's better than the example discussed in the post, I *do* still have a slight problem with the beer drinkers sentence, as well as with the improvement suggested earlier by Electric Dragon. ]

    Pullum in Pullum's petulance is a different story. It is an N, yes, but it's also an NP — a (possessive-marked) NP functioning as a determiner (cf. our colleague in our colleague's petulance). So the AIC says nothing about it one way or the other. There is, however, something I've labeled the Possessive Antecedent Constraint, a proscriptional fiction that claims that possessives cannot be antecedents for pronouns; the PAP has been treated at some length in Language Log postings from October 2003; see the detailed handout here.

  28. John said,

    August 4, 2009 @ 10:38 am

    I think the problem is compound: some may object to (or have difficulty with) non-constituent references, and this is made worse by the availability of "Windows 7" as the referent, which is more clearly the case if we change the verb. Consider:

    After Windows 7 comes out in October, will Microsoft somehow force us XP users to start using it?

    Here, "it" seems to refer to "Windows 7" and not "XP". So not only is "Windows 7" available as the referent, the meaning of the verb influences the choice, due to the implicature (or is it presupposition?) of "start" vs. "stop".

    On the other hand, as Electric Dragon points out, moving the clause "after Windows 7 comes out in October" to the end makes it easier to infer "XP" as the referent. I think it also makes it harder to infer "Windows 7" as the referent. Consider:

    (?) Will Microsoft somehow force us XP users to start using it, after Windows 7 comes out in October?

    So I think it's the interaction between (1) a difficult antecedent, (2) a more available antecedent, and (3) implicature, which makes the sentence difficult.

    [(EB) A perfect summary of my now-updated thoughts on the matter. Thanks, John! ]

  29. Troy S. said,

    August 4, 2009 @ 10:50 am

    As an aside, I keep seeing the word "felicitous" used as if it is a technical term. From context, it seems roughly equivalent to "idiomatic."
    Is that right?

    [(EB) No, it's roughly equivalent to "pragmatically appropriate". For example, if I ask "Can you pass the salt?" at the dinner table and you reply "yes" without actually passing it to me, that would be an infelicitous reply. ]

  30. Dan Scherlis said,

    August 4, 2009 @ 11:01 am

    Ah! Thanks, Prof. Zwicky, for the pointer to the Ward paper. It's online, here.

    Following Eric Baković's link to "another example," I had followed on to Greg Ward's site, and thus was already thinking that his focus discourse-accessibility was likely the key here.

    Perhaps Ward's work helps explain the relative felicity of "beer drinkers will be deprived of it" over (perhaps) "anaphora scholars are obsessed with it." Or maybe the Kentuckian/Hoosier example above. (I'm not sure. I just found the full-text URL, and am looking forward to reading it.)

  31. Ellen said,

    August 4, 2009 @ 11:12 am

    Stephen Jones, you claim, "It's perfectly felicitous, and as it didn't and doesn't cause a double take, probably quite grammatical.". It didn't and doesn't cause a double take for who. People here have already attested that it does for them cause a double take.

  32. elinar said,

    August 4, 2009 @ 11:22 am

    To me, this is another example clearly showing that grammaticality cannot be distinguished from acceptability; competence from performance, or syntax from semantics and pragmatics. It makes only sense to talk about “violation of co-reference rules” if we assume that there is such thing as Autonomous, Rule-based Grammar, and that it plays an important role in language use.

    What is the difference between a linguist claiming that a sentence may sound fine, but is actually ungrammatical (and here is a syntax tree to prove it!), and a language maven arguing that “It’s me” or “Who did you see” may be idiomatic, but are ‘technically’ incorrect? Both the linguist and the maven appear to be saying that it is the Grammar, not the speakers, that determines whether something is grammatical. (The trouble with language users is, of course, that they all have their own views about correctness, which makes life difficult for language theorists.)

    [(EB) Please follow the last link in my post, read it, and see if you still feel this way afterwards. ]

  33. Dan Scherlis said,

    August 4, 2009 @ 11:25 am

    > the relative felicity of "beer drinkers will be deprived of it" over (perhaps) "anaphora scholars are obsessed with it."

    What was I thinking? Of course anaphora scholarship is as common as beer drinking amongst this community. Apologies to all.

    Maybe better to suggest "paper manufacturers are obsessed with it" is less felicitous for discourse-accessibility reasons than the "beer drinkers" example.

  34. NW said,

    August 4, 2009 @ 12:00 pm

    The difference between us linguists and prescriptivists is that when you point out a counterexample, we say, 'Oh yeah, look at that, our rule is wrong,' and use that to try and find out what the rules actually are.

    [(EB) Indeed. ]

  35. elinar said,

    August 5, 2009 @ 5:43 am


    Thanks for the link, but I already know about this ‘correctness conditions’ stuff. I didn’t mean to imply that people don’t make grammatical errors; I meant that the kind of rule-based/autonomous/modular grammar that many linguists seem to subscribe to cannot account for real language use.

    So, I still maintain that there is no such thing as ‘pure’ grammaticality (or ‘real’ competence) that is independent from acceptability (pragmatics or performance). For one thing, speakers, including the commenters in this thread, often disagree whether something is grammatical, acceptable or felicitous. Simply arguing that we all have slightly different correctness conditions is a cop-out in my view – an implicit admission that there are no ‘fixed’ rules or constraints, only contextualised rules of thumb, or tendencies. Whether we are formulating correctness conditions for an individual or for a language, we can’t ignore the ‘fluidity’ of these rules.

  36. elinar said,

    August 5, 2009 @ 5:45 am

    NW: “The difference between us linguists and prescriptivists is that when you point out a counterexample, we say, 'Oh yeah, look at that, our rule is wrong,' and use that to try and find out what the rules actually are”.
    But it’s not really as simple as that, is it? As far as I know (and please correct me if I’m wrong), at least some linguists (the ones who believe in the grammaticality/acceptability contrast) are happy to dismiss lay speakers’ grammaticality judgements as ‘mere performance data’. In other words, speakers’ judgements cannot determine the grammaticality of a sentence; only the linguist (presumably because of their special training) is able to decide whether (un)acceptability is due to grammatical principles or extra-grammatical factors, such as parsing strategies, real-world knowledge, or prescriptive norms.

    So a linguist can continue to claim that a particular sentence violates Constraint X, even if some speakers judge it as grammatical/acceptable. These judgements can be ignored because linguistically naïve speakers have this bad habit of making use of the linguistic and extra-linguistic context, instead of focusing on pure grammaticality.

  37. Bill Stone said,

    August 5, 2009 @ 6:18 pm

    Come on you guys! We all know that the sentence 'After Windows 7 comes out in October, will Microsoft somehow force us XP users to stop using it?' is badly phrased.

    Syntactically, the antecedent of 'it' may well be ' Windows 7', but semantically it's 'XP', hence the possible ambiguity.

    A better version might be: 'Will Microsoft somehow force XP users to upgrade to Windows 7 when it comes out in October?'

  38. Stephen Jones said,

    August 7, 2009 @ 1:33 pm

    The first possible referent for 'it' is XP so proximity rules in its favour. Secondly it is semantic nonsense to suggest that you can stop using something you're not using and that hasn't even come out yet.

    Take this sentence:
    The old man screamed and the policeman, seeing the mugger legging it down the street, shot him.</i.

    Nobody is seriously suggesting that 'him' can refer to 'the old man'. There is no ambiguity.

  39. Stephen Jones said,

    August 7, 2009 @ 1:39 pm

    and this is made worse by the availability of "Windows 7″ as the referent, which is more clearly the case if we change the verb.

    'Windows 7' is not available as a referent for the reasons I've given in the last post. To produce another different sentence where it is is irrelevant.

  40. Jesse Weinstein said,

    October 8, 2009 @ 3:59 am

    Since no one has mentioned it yet, I'll say that I originally read the sentence as wondering if Windows 7, like Vista, will be so bad that the people currently using XP (i.e. those who refused to upgrade to Vista) will also find Windows 7 to be unusable.

  41. David Walker said,

    September 22, 2010 @ 11:50 am

    To ignore the grammar and briefly talk about the question: No, Microsoft won't force us to stop using "it" (Windows XP). Windows XP with SP3 will end its "extended support phase" in August 2014. After that time, Microsoft will no longer offer security patches. But you can keep using XP until you die (or pass over into another plane of existence). Details at http://support.microsoft.com/lifecycle/?LN=en-gb&C2=1173.

RSS feed for comments on this post