Going and heading

« previous post | next post »

In a recent comment, Amy asked:

If […] "go mad" is a modern formation that is perfectly grammatical, why would Mr Pullum label "head dagenham" as "…a little beyond the syntactic fringe"? What's the difference?

The Dagenham business is in Geoff Pullum's post "Beyond Barking", 6/24/2008; and the assertion about "go mad" is here. And I'm afraid that Amy's question doesn't have a very impressive answer, because this isn't, as far as I can tell, something that can (or should) be deduced from the fundamental axioms of grammar and logic. Essentially, it's just a fact about the verbs go and head.

We can put this answer into a broader context by observing that "go mad" is a fairly productive pattern, commonly attested not only with mad, but also with dozens of other words, such as bad, ballistic, bankrupt, crazy, native, sour, straight, wrong. If you interpret Dagenham to mean "spectacularly crazy", as a jocular implication of the fact that it's one stop beyond Barking on a certain rail line (and thus "beyond barking (mad)") then you could presumably say that someone has "gone Dagenham". There are even syntactic precedents in expressions like "go Hollywood" or "go Nashville".

In contrast, head (as Geoff pointed out) generally requires a prepositional phrase or similar direction word: you can "head out of town" or "head home" or "head south", but you can't "head crazy" or "head Hollywood".

There's no logical reason for this — it's perfectly clear what it would mean to "head crazy" or "head Hollywood", but the English language happens not to use head that way. This would be a reasonable extension to the language, so it's not a shock to find a few relevant examples out there on the web, like this one (though it's apparently not from a native English speaker). But to a pretty good approximation, expressions like "head crazy" just aren't used.

We could explore the history of the words and constructions in question — and discover that phrases like "go mad" are not particularly "modern" — and we could compare the analogous constructions across languages; but in the end, Geoff wrote one thing about "head Dagenham" and I wrote something else about "go missing" because that's just the way the English language is.



16 Comments

  1. Pavel Iosad said,

    October 26, 2008 @ 9:48 am

    I'm not really into cognitive linguistics, but I'm pretty sure this would be their cup of tea; something like "since 'go' has a lot of generalized readings, it is more readily available for reuse in a more abstract context" sounds, to me, like something a paper within that framework might say on the matter. But then, I'm a (OT) phonologist…

  2. Karen said,

    October 26, 2008 @ 11:23 am

    This is something that's harder to explain than it should be. That different verbs with much the same meaning take very different arguments seems unreasonable, and perhaps it is. An example I use is "stop" and "cease" – while you can "cease to smell the roses" and "cease smelling the roses" you can only "stop smelling". You can, of course "stop to smell the roses" but it means something completely different from "cease to smell". Unreasonable? Perhaps, as I said. But then, language isn't a construct of reason.

  3. Philip Spaelti said,

    October 26, 2008 @ 12:02 pm

    Indeed, your post immediately got me thinking about English's cousin German. German has the same "geht X" construction with the deterioration meaning, but when trying to transfer the English examples only some work. So "geht bankrott" parallels "go bankrupt" and "geht falsch" is similar to "go wrong", but many others don't work. On the other hand German has some that don't work in English. So "geht kaputt" (=become broken, of devices and machines) but while English has "go broke" it doesn't have "*go broken". German has "geht mir auf die Nerven" but in English it is "_get_ on my nerves" not "*go on my nerves".

  4. James C. said,

    October 26, 2008 @ 4:39 pm

    @Karen: I think you're wrong about the argument restriction on "smell". Perhaps it's just my idiolect, but I find "stop smelling X" to be perfectly acceptable, as in "stop smelling the roses and get back to work".

  5. Lance said,

    October 26, 2008 @ 7:46 pm

    I think you misread her, James C: she meant "only 'stop smelling'" as opposed to "stop to smell". It's the "…ing / to" alternation she's interested in. ("Leave off" works like "stop", as does "quit"…this seems to be something idiosyncratic about "cease", unless other synonyms accept "to VP" as their objects.)

    Language is, alas, quirky. Polly Jacobson pointed out to me, when I was working on Noun vs. Clause complements, that some verbs can't take clausal complements even though they can take NPs like "the fact that"; the one that comes to mind is "reflect". So it's OK to say "My theory reflects the fact that German is just like English", but not "My theory reflects that German is just like English." The two objects ("that X" and "the fact that X") ought to be pretty much synonymous—try making either one the object of "remember", for instance—and yet "reflect" is compatible with one but not the other. I don't know that there's any good reason for this; it's just something that happens, as far as I understand it. (Of course, as a linguist and as a pattern-seeking human being, I hate it. Ah well.)

  6. Mark Liberman said,

    October 26, 2008 @ 8:39 pm

    Lance: Language is … quirky.

    Indeed. And it's interesting (and perhaps less studied than it should be) that this quirkiness extends not just to the choice of syntactic complement structures, but also to the choice of lexical items in constructions like "go <adjective>".

  7. Alif Baa said,

    October 26, 2008 @ 9:49 pm

    And as far as language quirkiness, is there any fundamental reason why you can say "go home" but not "*go house" (whereas both go to my home and go to my house work).

    Maybe it has something to do with home being in a sense more conceptual or less physically well defined? Note that you can also use "go" in this sense with upstairs/downstairs, downtown, cardinal directions, etc.

    The nature of this distinction was a question my sister couldn't answer when asked by her English class in Spain. Is there a simple answer that I'm missing, or is it just the way it is?

  8. Philip Spaelti said,

    October 27, 2008 @ 4:03 am

    The word "home" is a preposition (as well as a noun). This has been so for a long time, as in German "heim" behaves the same way. "Upstairs", "Inside", etc. are prepositions too. In these cases it is slightly more obvious, as they derive from expressions with a more familiar preposition.

    In defense of the logic which allows us to call "home" a preposition, see for example the following radio interview with Geoff Pullum. I thought there was a post on this somewhere on Language Log itself, but I am not clever enough to find it.

  9. Matt Wildman said,

    October 27, 2008 @ 6:50 am

    I think the original 'go mad' *'go/head Dagenham' problem arises from the fact that 'go' is polysemous and we are looking at two separate senses of the verb. In 'go mad' the verb has a copular function; it is linking the subject with a predicate complement (and adjective or noun phrase which describes it) and so can be replaced by other copular verbs ('seem, become, be' etc.)
    In the other sense it is simply a lexical verb expressing movement to a destination and, in Standard English at least, it requires a prepostion phrase as its complement (eg. 'to Dagenham'). However you can hear instances of *'I'm going London on Thursday' and *'Wanna go pub tonight?' more and more so preposition phrases after 'go' may be dying out!

  10. Lugubert said,

    October 27, 2008 @ 7:57 am

    Several years ago, when I first read Geoff Pullum's interpretation of "head bush", I had a different reaction. I almost physically felt a case of a zero morpheme locomotive case.

  11. Aaron Davies said,

    October 27, 2008 @ 11:27 am

    Having noticed the sad shortcomings of the commentariat here, I must make the obvious allusion:

    "Romanes eunt domus."

    "People called Romanes they go the house?"

  12. Jorge said,

    October 27, 2008 @ 12:55 pm

    I like "go postal".

  13. Robert Coren said,

    October 27, 2008 @ 1:11 pm

    and discover that phrases like "go mad" are not particularly "modern"

    …led me to wonder whether, in fact, the phrase appears in Shakespeare, which I suspect it does, but can't think of an instance off the top of my head. But that made me think of the (now somewhat archaic?) "run mad", which leans more in the direction of *"head mad". I can't, offhand, think of other parallel uses of "run".

  14. Catanea said,

    October 27, 2008 @ 2:01 pm

    Can anyone shed light on "going spare", which seems to mean the same as "going mad", but I can't figure out why….??

  15. Irene said,

    October 27, 2008 @ 2:44 pm

    Why in American English do we say "go home", "go to school" and "go to the hospital"?

  16. marie-lucie said,

    October 28, 2008 @ 5:07 pm

    If you say "go to school", you are describing a habitual activity (eg "All my children go to school now" – even if they each go to a different school), as opposed to "go to the school" which a parent does in order to pick up a child or talk to a teacher: the school here is a specific place. Similarly with "go to church" which devout persons do on a regular basis (regardless of which denomination they belong to or which building the service is held at) vs. "go to the church" which refers to a specific place where you might go once for a specific purpose. "Go to jail" means you have been convicted and will be "in jail" for a period of time, but your friends and relatives might "go to the jail" to visit you and be allowed "in the jail" for a limited time. After your release you will hopefully "stay out of jail", but a troublesome visitor could be told to "stay out of the jail". I don't know whether medical personnel or patients with a chronic condition "go to hospital", but the ordinary person would only "go to the hospital" for an appointment or to visit a patient.

RSS feed for comments on this post