Resumptive pronoun of the week

« previous post | next post »

Jonathan O’Connell, “Feds, Trump attorneys wrangle over president’s D.C. hotel lease“, Washington Post 2/10/2017 [emphasis added:

Chaffetz told reporters this week that he was interested to learn how officials intended to grapple with the potentially awkward situation in which the Trump-led government intended to negotiate with a business controlled by the president’s family.

“His being both the landlord and the tenant is something that we’re curious what the GSA’s opinion of that is,” Chaffetz said.

The earliest version of this quote seems to have come from Kyle Cheney, “Chaffetz has no idea why Trump wants to see him“, Politico 2/7/2017, and it’s been reproduced in several other stories. But I haven’t been able to find a recording, and there’s no evidence that reporters’ shoddy quotation practices have improved, so despite the quotation marks, we have no way to know whether these are Chaffetz’s words or a reporter’s  paraphrase.

Whoever created the sentence, however, it offers a nice example of what linguists call a “resumptive pronoun“.

Here’s the idea.

In English relative clauses, there’s generally an implicit gap where the head fits in:

They had these eggs for breakfast.
These are the eggs that they had ___ for breakfast.

But there are some limits on the relationship between the head and the gap:

If they had these eggs for breakfast, we need to worry about food poisoning.
*These are the eggs that if they had ___ for breakfast, we need to worry about food poisoning.

Violations of those limits can be ameliorated by inserting a pronoun:

?*These are the eggs that if they had them for breakfast, we need to worry about food poisoning.

These limits were first explored by J.R. Ross in his 1967 dissertation, “Constraints on Variables in Syntax“.  And as John Lawler explains in a page on “Ross Constraints”,

Violations of Ross Constraints are very ungrammatical. Most people never encounter them. We appear to formulate our discourse to avoid them. Occasionally, we get in a bind and see one looming at the end of the clause, and have to do something quick. What we do is often illuminating about the relative importance of syntactic rules.

For instance, consider the following:

?That’s the book_i [that Bill married the woman_j [who_j illustrated it_i]].
*That’s the book_i [that Bill married the woman_j [who_j illustrated ___i]].

Neither sentence is terrifically grammatical, but the first seems more appropriate (and common as a type) than the second, though the last word in the first sentence still feels strange. The ordinary rule of relative clause formation operating on the last clause should result in its deletion at the end of the clause (and thus the sentence). However, it appears inside another relative, an island, and is thus safe from such “Movement” by the Complex NP Constraint.

Sentences like the first one are generated when, at the last minute, the speaker realizes what is going to result, and cancels the deletion, substituting an alternative relative-formation rule (called a Resumptive Pronoun in the trade), which merely pronominalizes the coreferential NP, instead of deleting it in the object position.

This is not the way English forms its relative clauses (though other languages use it frequently, e.g, Hebrew), and the sentence is thus ungrammatical. But this turns out to be a venial syntactic sin by comparison with a violation of a Ross constraint, which typically produces extreme ungrammaticality.

In my opinion, John exaggerates the badness of his example, “the book that Bill married the woman who illustrated”; and it’s a step too far to say that “most people never encounter” things like that.

So how about a version of the sentence attributed to Mr. Chaffetz without the resumptive that?

*His being both the landlord and the tenant is something that we’re curious what the GSA’s opinion of is.

That version certainly looks pretty bad on paper, though it can be performed in a way that makes it work better. And the difference between writing/reading and speaking/listening is crucial here.

We don’t really know — as far as I know — how all of this really works out in assorted kinds of spontaneous English-language speech. How frequent are relative clauses that violate various Ross constraints? How often are the problematic relations ameliorated by resumptive pronouns, compared to the frequency of resumptive pronouns in unproblematic cases? And how frequent are the problematic structures predicted to be, on the basis of the probabilities of the substructures that make them up? Are the actual frequencies really a lot lower than the predicted frequencies?

We’ve now got data and models that allow such questions to be answered, at least in principle.




  1. stephenl said,

    February 11, 2017 @ 9:43 am

    I feel that

    “These are the eggs that if they had ___ for breakfast, we need to worry about food poisoning.”

    could also be patched up, without needing a pronoun, as

    “These are the eggs that if they were had for breakfast, we need to worry about food poisoning.”

    I don’t know if that’s particularly interesting as an option, though.

  2. Breffni said,

    February 11, 2017 @ 10:01 am

    You certainly don’t find them very often, but I noted this one down years ago, a scripted utterance from an Irish travel programme:

    People here on the island are working really hard to change Ibiza’s reputation. Yes, it’s an island all about clubbing – and the clubbing is good, don’t get me wrong. But there’s a lot more to it than that. There’s sophistication, and there’s also a traditional island that when you go looking for, you’ll find.

    It’s the kind of island constraint violation that when you go looking for, you’ll find hard to find.

  3. Rodger C said,

    February 11, 2017 @ 11:32 am

    But surely the point is that constraints are rare on that island?

  4. Mark Meckes said,

    February 11, 2017 @ 11:51 am

    Although I agree that “His being both the landlord and the tenant is something that we’re curious what the GSA’s opinion of is” looks bad on paper, when I imagine hearing it spoken it sounds like perfectly ordinary colloquial speech.

    [(myl) Yes, I agree — though the results of “[imagining] hearing [something] spoken” are notoriously unreliable in cases of this kind.]

  5. Breffni said,

    February 11, 2017 @ 12:15 pm

    Rodger C: Indeed. I like to think the presenter was a linguistics graduate who saw an opportunity for an extremely elaborate in-joke.

  6. AntC said,

    February 11, 2017 @ 3:05 pm

    Yeah, this tricky stuff: relative clauses, recursion, embedding are things I wonder if it wouldn’t be better to speak a language that doesn’t have them.

  7. Guy said,

    February 11, 2017 @ 6:50 pm


    Your example still has a resumptive pronoun, the “they” after “if”. Though I agree your example has a high level of acceptability compared to most resumptive pronouns.

    Often the easiest solution when you want to say something that would violate an island constraint in some obvious phrasings is to use “such that” or (at least in my dialect) “where”. The “such that” solution is often awkwardly formal sounding, though.

    “I don’t want to implement a procedure where I have to get approval from 50 people if I want to modify it”.

    This also works in cases where there is no gap or resumptive pronoun, like in this sentence.

  8. philip said,

    February 12, 2017 @ 5:42 am

    The Irish travel reporter in Ibiza might have solved her problem with relative clauses if she had used ‘if’ instead of ‘when’. It sounds OK to me that way.

    “These are the eggs about which food poisoning is a worry if they had them for breakfast.”

    But all the patches and the original sentence are dodgy, as it seems to be a biscuit conditional, i.e. food poisoning is a worry whether or not they had them for breakfast?

  9. Brett said,

    February 12, 2017 @ 12:50 pm

    I can see that the quote about Ibiza seems to be violate a constraint, but I find it entirely grammatical.

  10. philip said,

    February 12, 2017 @ 2:46 pm

    Brett: transitive and intransitive verbs? Transitive verbs, like looking for, require an object:
    What are you doing?
    I am looking for it
    *I am looking for
    Intransitive verbs do not require an object:
    what are you doing?
    I’m reading
    but they can take an object if they want to:
    I’m reading a book.

  11. Guy said,

    February 12, 2017 @ 6:23 pm

    I don’t find the Ibiza quote acceptable with either “if” or “when”.

    Philip, many verbs with intransitive uses also have transitive uses, but it’s not correct that they always do. “He died” but *”he died a thing”.

    Also, “looking for it” is technically an intransitive construction (CGEL would call it complex-intransitive), “it” is not the object of “looking”, it is the object of the preposition “for”, the preposition phrase “for it” is a complement of “looking”, but not an object. This relationship is usually described by saying that “it” is an oblique of “looking”, but being an oblique is not the same as being an object.

  12. philip said,

    February 12, 2017 @ 7:09 pm

    Thanks Guy – I wasn’t really serious when I said they can ‘take an object if they want to’.

    He died a death.

    Yeah, the Ibiza thing is dodgy with if, but I am looking at it like this, with an extra comma:

    … and there’s also a traditional island that, if you go looking for, you’ll find.

    Which seems sort of all right to me.

  13. Bloix said,

    February 17, 2017 @ 10:18 am

    This is a tweet from Matt O’Brien, who works for the Washington Post:
    “Trump-22: Anyone who wouldn’t want to work in this White House is someone who it’d be good if they worked in this White House.”

RSS feed for comments on this post