Watermelons are long to grow

Laurence Horn laurence.horn at YALE.EDU
Sat Aug 7 16:23:05 UTC 2010

On first look, I had the same thought:  Interesting extension of
tough movement to a member of a previously unqualified class, I agree.

But when I googled "long to grow" the first (of many) hits were all
of the form "How long to grow...?", which I assume is an elliptical
form of "How long does it take to grow...?"  From this, it's
conceivable that one might* reconstruct "How long is it to grow X?" >
"It's Y long to grow X" > "X is (Y) long to grow" without ever
getting "Growing X is (Y) long" (compare "Growing melons {takes/*is}
(too) long").  In this case, "X is long to grow" wouldn't be related
to "Growing X is long", which for me is really worse, i.e. less
conceivable, than "X is/are long to grow", thus departing from your
intuition about (3a,b) below.  Granted, there's just one hit for "X
are long to grow", where X = "big beautiful oaks".

(* a suggestion rendered perhaps implausible by the virtual absence
of g-hits for "How long is it to grow", but I won't let that stop me)

Note that in general tough movement constructions have counterparts with "too":

Watermelons are (too) expensive to buy.
Calculus may be (too) hard to learn.
Melons are (too) time-consuming to grow.

but this one doesn't really work:

Melons are (too) long to grow.

(except on the irrelevant reading, where it's their actual length
that's involved:  they're too big to grow in this little plot)

So maybe, ahem, a blend:  "Melons are time-consuming to grow" + "how
long to grow melons"

There's also the fact that there's an "active" or unaccusative
reading--"The melons grew large"--but I don't think that messes
anything up here.

And FWIW, there's a construction I've always puzzled at:  "X takes
getting used to" (where "time" is swallowed up the way it sort of is
in your construction).


At 3:55 PM +0100 8/7/10, Damien Hall wrote:
>A friend recently remarked
>1) 'Watermelons are dear to buy and long to grow.'
>I have no problem, of course, with 'dear' (=BrE for 'expensive') in a
>tough-movement construction like this, but 'long' struck me as strange. It
>can be made to work, sort of, if you do some of the usual paraphrasing:
>{2) a) 'Melons are dear to buy' (OK) > b) 'Buying melons is dear' (OK)
>3) a) 'Melons are long to grow' > b) 'Growing melons is long'
>The two sentences in 3) are as OK as each other; if you can accept 3a), you
>might be able to accept 3b) - and there's no reason why many speakers of
>English couldn't accept 3b), it seems to me, since it's surely the same
>construction as
>4) 'Time is short!' (meaning: 'Let's do it now before it's too late!')
>Yet, probably because of the chain of reasoning that it's taken to convince
>myself that 'Melons are long to grow' is formally acceptable and 'should'
>sound right, 'Melons are long to grow' still sounds to me like an
>innovative extension of a construction I know, as opposed to something
>natural. I think it might well be to do with the different semantics of
>'dear', 'tough', 'easy', difficult', etc (the words that usually undergo
>this sort of tough-movement transformation) on the one hand, and 'long'
>(for time) on the other; but there my expertise stops. I can't comment on
>the semantics at all.
>How does this sound to others, and can you explain it? It's hard to Google
>(pardon the pun), of course, to see how common it is. The commonness of
>'time is short' is no gauge of the acceptability of this construction, as
>it's a fixed phrase which you might use even if you never used this
>construction anywhere else; If you Google 'time is long' you get pages of
>false hits for things like 'the time is long overdue'; etc.
>Damien Hall
>University of York
>Department of Language and Linguistic Science
>YO10 5DD
>Tel. (office) +44 (0)1904 432665
>     (mobile) +44 (0)771 853 5634
>Fax  +44 (0)1904 432673
>DISCLAIMER: http://www.york.ac.uk/docs/disclaimer/email.htm
>The American Dialect Society - http://www.americandialect.org

The American Dialect Society - http://www.americandialect.org

More information about the Ads-l mailing list