word order (was Re: Random DWJ discovery of the day

deborah.dwj at suberic.net deborah.dwj at suberic.net
Mon Apr 4 18:00:34 EDT 2005

More information than you ever wanted to know about word order,
otherwise known as "what happens when Deborah was the best chick for a
doctoral student in linguistics, and decides to forward him a question
from the list".

Kiss my ass, I bought a boat, and I'm going out to sea.  -Lyle Lovett

---------- Forwarded message ----------
Date: Mon, 4 Apr 2005 16:54:21 -0400 (EDT)
From: Lance Nathan <tahnan at MIT.EDU>
To: deborah <deborah at suberic.net>
Subject: Re: word order (was Re: Random DWJ discovery of the day) (fwd)

On Mon, 4 Apr 2005, deborah wrote:

> just curious -- do you know offhand if what's going on here is idiom or
> actually something grammatical (in your sense of grammar, not my sense)?

Short answer: it's grammar, and it's quite complicated.  Casual poking at
the web tells me that linguistic work has been done on the subject, but
I'm not sure that anyone came to anything especially conclusive.

Longer answer: first, for those unfamiliar with the two senses of grammar
Deborah refers to, "her" sense is the rules of formal writing one learns
in school, such as not splitting infinitives or not ending sentences with
prepositions.  "My" sense is the instinctive collection of facts about
English that every native speaker has, the purely descriptive facts
that don't need teaching.

So, when Charlie asked,

> > Then again: have I infringed a rule of English if I write 'The old little
> > man entered the brown big house?' It feels 'unnatural', but if it's wrong it
> > seems to be wrong in a different way from a sentence like 'Mat on the cat
> > sat the.' Is it that in English, at least, rules involving a semantic
> > element are more gainsayable than ones based on syntax alone?

There's no doubt that there's a difference between syntactic rules and
semantic rules.  There's even a neurological difference--it's possible to
detect a difference in the electric energy generated by the brain when a
reader encounters a syntactic error or a semantic one.

But it's not the case that one or the other is more gainsayable.  For
instance, there are syntactic rules of grammar that can be violated.  In
English, one asks questions about multiple facts by leaving the object
question word in place.  For instance, faced with a potluck buffet, one
might ask, "Who brought what?" but not "What did who bring?"  But that's
not 100% inviolable, because one *can* ask "Which dish did which person

Similarly, some semantic rules *are* ungainsayable: "I inquired that it
was raining" is just no good semantically, because you can't combine a
verb meaning "inquire" with a statement, only with a question (compare "I
inquired whether it was raining", which has roughly the same syntax).

And of course, "the old little man..." *can* be used in a perfectly
natural way: if I'm telling you a story about two little men, one of them
young and one of them old, I might say at one point, "The old little man
went home, while the young one went swimming."

And when Judith asks,

> My first response to Charlie's example is, well, it feels "unnatural" or
> wrong, because we're simply not used to this order of words: "little old"
> (or "tired old" or "big bad" etc) is such a common expression that any
> syntactical change to it does sound unnatural. So then I wondered, why is
> this the natural order of these particular words? Is there a non-arbitrary
> reason for the order of adjectival categories Charlie cites...

There's certainly a pattern.  Many adjectives have no relative ordering:
"he's an interesting, intelligent speaker" vs. "he's an intelligent,
interesting speaker".  Some depend very much on meaning for their
ordering: a "former important senator" is someone who was a senator, and
who was important when he was, whereas an "important former senator" is
someone who was a senator, but who is important now.

But there is a certain natural ordering for many adjectives.  It's true to
a point that it's a matter of what we're "used to", and some phrases to
become practical idioms, like "little old man".  But that can only go
so far, because language learners generalize widely from the narrow range
of examples they hear.  (You don't have to be used to any combination of
the words in "My antelope dislikes kiwi sorbet" to know what it means.)
So even when you hear "he's a middle-aged tall man", there's a feeling
that "tall" should come before "middle-aged".

http://web2.uvcs.uvic.ca/elc/studyzone/410/grammar/adjord.htm gives a
rough ordering for adjectives, based on meaning.  But this is of course
rough.  And there's definitely a sense in which adjectives that are more
"inherent"  properties of the noun come closer to the noun than those that
are less inherent.  For example, if I want to talk about the stars that,
while generally visible in Boston, happen to be unseeable tonight due to
viewing conditions, I might mention the "unseeable visibile stars", but
not the "visible unseeable stars"--the stars' general visiblity is more an
inherent property than their accidental invisibility.  But this, too, is
not a 100% hard-and-fast rule (in what sense, for instance, is being "old"
more inherent than being "little"?  Or being "English" as opposed to being

A more complete ordering, recently proposed in the linguistic literature,
and one that seems to be universal, is:


(where "determiner" is an article such as "the" or "every"; and I'm not
positive what he means by "evidential", but possibly he means "possible"
or "definite").

Of course, even if this rule is somehow universal, I'd hesitate to say
that it derives from "some way our brain works or something--some natural
order outside of language"; after all, the way our brain works determines
how language works.  I'm not sure at all what it tells us about the brain,
or even what it tells us about language, but there's something there,


| Lance Nathan, Graduate student, MIT | You will not find in semantics  |
| Dept. of Linguistics and Philosophy | any remedy for decayed teeth or |
| web:   http://www.mit.edu/~tahnan/  | illusions of grandeur or class  |
| email: tahnan at mit.edu               | conflict.      --Alfred Tarski  |

To unsubscribe, email dwj-request at suberic.net with the body "unsubscribe".
Visit the archives at http://suberic.net/dwj/list/

More information about the Dwj mailing list