{"id":1912,"date":"2009-11-21T14:29:54","date_gmt":"2009-11-21T18:29:54","guid":{"rendered":"http:\/\/languagelog.ldc.upenn.edu\/nll\/?p=1912"},"modified":"2009-11-24T05:43:15","modified_gmt":"2009-11-24T09:43:15","slug":"questions-and-conditionals","status":"publish","type":"post","link":"https:\/\/languagelog.ldc.upenn.edu\/nll\/?p=1912","title":{"rendered":"Questions and conditionals"},"content":{"rendered":"

Decades ago, when I was little, I read this joke in Mad Magazine<\/I>:

Do your feet smell? Does your nose run? You may be built upside-down.<\/I> <\/P><\/BLOCKQUOTE>

I giggled for a short time — just a couple of days, I think — at the surprising coincidence of the two verb senses, and the double pun, and then got on with whatever boys in short pants do during those parts of the day that are not taken up with giggling. But I see now that there is something linguistically interesting about the joke: the two questions convey the effect of a conditional. So the content of the joke could be phrased (though for some reason much less amusingly) like this:

If your feet smell and your nose runs then you may be built upside-down.<\/I><\/P><\/BLOCKQUOTE>

This similarity of effect between interrogative clauses and conditional clauses has a connection to the historical reason for an identity of form between the words introducing the interrogative subordinate clause in (1) and the conditional clause in (2).

    (1) I don't know if the car will start.<\/I>
    (2) We won't go if the car won't start.<\/I><\/P> The two if<\/I>s share an etymology, but they have grown apart.
\n

Traditional grammars do not recognize this. They would say we have the same item in (1) and (2): a \"subordinating conjunction\". But the analysis that Huddleston and I set out in The Cambridge Grammar of the English Language<\/I><\/A> dispenses entirely with the term \"subordinating conjunction\", on the grounds that it is a label for a rag-bag of items that should have been analyzed in two different ways. A small number of the items in question are subordinators<\/B>: meaningless markers such as that<\/I> and whether<\/I> that appear at the beginnings of subordinate clauses. The rest are prepositions<\/B> that happen to take clauses rather than noun phrases as their complements: words like though<\/I> and because<\/I>. These have substantial semantic content, taking clauses and converting them into adjuncts expressing concession, cause, reason, etc.

The word if<\/I> is unusual in that it is actually two different items, one a subordinator and the other a preposition. The if<\/I> in (1) is a subordinator, with no more meaning than whether<\/I> has; but the if<\/I> in (2) is a preposition which forms an adjunct phrase expressing a precondition for the the truth of the matrix clause. In the contemporary language there are many syntactic differences between them. I recently set it as an exercise for my students to demonstrate these differences. Let me share with you the ten syntactic arguments with which I supported the view that the two items are grammatically distinct.

  1. The Subordinator if<\/I> can be replaced by whether<\/I> (which is slightly more formal), but the Preposition can't:

    I don't know if it's true.<\/I>    [subordinator if<\/I>]
    I don't know whether it's true.<\/I>

    Have some cake if you want.<\/I>    [preposition if<\/I>]
    *Have some cake whether you want.<\/I>

  2. The Preposition if<\/I> can be replaced by other prepositions such as after<\/I>, but the Subordinator can't:

    I'll jump in if you do.<\/I>    [preposition if<\/I>]
    I'll jump in after you do.<\/I>

    I don't know if I'll jump in.<\/I>    [subordinator if<\/I>]
    *I don't know after I'll jump in.<\/I>

  3. An if<\/I>-PP is an adjunct<\/B>, and adjunct PPs can be preposed very readily. Content clauses, on the other hand, are by no means so readily preposed. Sometimes an interrogative content clause can be preposed: we can get Whether this is true I cannot not say<\/I>; but clauses beginning with interrogative if<\/I> seem not to prepose (?If this is true I cannot not say<\/I> doesn't seem to mean I cannot say whether this is true<\/I>), and certainly the complement of wonder<\/I> does not seem to prepose at all; so there are fairly clear contrasts like this:

    If I learned how to walk on water, they would probably still ignore me.<\/I>    [preposition if<\/I>]
    *If I could learn to walk on water, I wonder.<\/I>    [subordinator if<\/I>]

    If you want you can have some cake.<\/I>    [preposition if<\/I>]
    *If there is any cake left I don't know.<\/I>    [subordinator if<\/I>]

  4. You can add or not<\/I> to a closed interrogative content clause, but not to a conditional phrase, so only the second of these is grammatical:

    *If I learned how to walk on water or not, they would probably still ignore me.<\/I>
    I wonder if I could learn to walk on water or not.<\/I>    [subordinator if<\/I>]

  5. The irrealis form of be<\/I><\/B> (namely, 1st and 3rd singular were<\/I>) occurs (at least in formal styles) in the complement of conditional if<\/I>, but not in interrogative content clauses:

    I would eat worms if I were<\/U> a bird.<\/I>    [preposition if<\/I>]
    Bad: *I wonder if I were<\/U> a bird in some previous life.<\/I>    [subordinator if<\/I>]

    He wouldn't treat you like that if he were<\/U> in love with you.<\/I>    [preposition if<\/I>]
    *I never knew if he were<\/U> going to come home at night.<\/I>    [subordinator if<\/I>]

  6. The Preposition, but not the Subordinator, can be modified by placing the focusing modifier adverb even<\/I> before it:

    They would probably still ignore me even if I learned how to walk on water.<\/I>    [preposition if<\/I>]
    They would probably still ignore me if I learned how to walk on water.<\/I>
    We need to know if he is on our side or not.<\/I>    [subordinator if<\/I>]
    *We need to know even if he is on our side or not.<\/I>

  7. The focusing modifier adverb only<\/I> can also occur before Preposition if<\/I> as a modifier but not before the Subordinator. It may be possible to interpret only<\/I> as modifying the main clause in the latter case (I wonder only if we can trust him<\/I> can be read as saying \"The only thing I wonder about is whether we can trust him<\/I>); but we can form a coordination of conditional if<\/I> with another if<\/I> that is premodified with only<\/I>, to get a familiar expression of the logical biconditional, and this clearly fails with the Subordinator, giving a reliable test:

    I will do it if and only if we can get agreement from the committee.<\/I>    [preposition if<\/I>]
    *I wonder if and only if we can get agreement from the committee.<\/I>    [subordinator if<\/I>]

  8. The focusing modifier only<\/I> also occurs in an idiomatic combination after the Preposition if<\/I>, and this too is specific to the Preposition:

    I could get some sleep if only the baby would stop crying.<\/I>    [preposition if<\/I>]
    *I wonder if only the baby could somehow be stopped from crying.<\/I>    [subordinator if<\/I>]

  9. Conditional if<\/I> takes reduced clause complements as well as full clause complements, but this is not true of the Subordinator:

    I will open the emergency door if instructed to do so.<\/I>    [preposition if<\/I>]
    *I wonder if instructed to do so.<\/I>    [subordinator if<\/I>]

  10. There is an idiom if that<\/I> meaning something like \"even if that much is true\". It is limited to the Preposition if<\/I> and cannot occur with the Subordinator:

    There were about six people in the audience, if that.<\/I>    [preposition if<\/I>]
    *Someone told me there were six people in the audience, but I wonder if that.<\/I>    [subordinator if<\/I>] <\/OL>

    These ten largely independent arguments provide overwhelming evidence for the difference between the two if<\/I>s. Those syntactic differences are there despite the fact that the two versions of the joke with which I began (the funny original version and the oddly unfunny conditional version) have very much the same semantic effect. A conditional has an effect very similar to that of posing a question that is crucial to the evaluation of the matrix clause: If the cap fits, wear it<\/I> has an effect very similar to the sequence Does the cap fit? Then wear it<\/I>.

    There are absolutely no dictionaries anywhere that treat if<\/I> as a preposition. They all disagree with what I have just said. But I'm not here to give you just the truths that dictionary makers have so far felt they can stomach. I'm here to give you the whole truth. All printed dictionaries of English are wrong.<\/B> Sorry, but that's the way it is. Deal with it. <\/p>\n","protected":false},"excerpt":{"rendered":"

    Decades ago, when I was little, I read this joke in Mad Magazine: Do your feet smell? Does your nose run? You may be built upside-down. I giggled for a short time — just a couple of days, I think — at the surprising coincidence of the two verb senses, and the double pun, and […]<\/p>\n","protected":false},"author":3,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_exactmetrics_skip_tracking":false,"_exactmetrics_sitenote_active":false,"_exactmetrics_sitenote_note":"","_exactmetrics_sitenote_category":0,"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[19,20],"tags":[],"class_list":["post-1912","post","type-post","status-publish","format-standard","hentry","category-semantics","category-syntax"],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/languagelog.ldc.upenn.edu\/nll\/index.php?rest_route=\/wp\/v2\/posts\/1912","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/languagelog.ldc.upenn.edu\/nll\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/languagelog.ldc.upenn.edu\/nll\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/languagelog.ldc.upenn.edu\/nll\/index.php?rest_route=\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/languagelog.ldc.upenn.edu\/nll\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1912"}],"version-history":[{"count":0,"href":"https:\/\/languagelog.ldc.upenn.edu\/nll\/index.php?rest_route=\/wp\/v2\/posts\/1912\/revisions"}],"wp:attachment":[{"href":"https:\/\/languagelog.ldc.upenn.edu\/nll\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1912"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/languagelog.ldc.upenn.edu\/nll\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1912"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/languagelog.ldc.upenn.edu\/nll\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1912"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}