Brett,
     After I emailed you, I looked back at the arguments and realized my 
question was easily addressed. If a preposition doesn't need to take an 
"object", then "down" could be thought of as a preposition from the 
start. Easy to do. Like "away."
    I like your point that these are categories, not definitions. It is 
the understanding of how language works that matters, so one payoff to 
this sort of discussion is that we get to notice things that we might 
not otherwise pay attention to.
    I would agree with most of the observations, though I have always 
had trouble with the idea that adverbial structures can't act as copular 
or predicate complements. "I put the ladder in the shed" I would analyze 
as having an adverbial complement, as I would "the ladder is in the 
shed." We can also say "The time is now" or "The meeting right now is 
important," where "right now" acts as complement to "be" and adjunct to 
the noun.  Is "now" a preposition as well? The argument begins to fall 
apart when we take the category of adverb and pull out of it anything 
that doesn't fit the new test. If it can complement "be" it is not an 
adverb. Adverbs can't complement "be." It seems circular.
    We also have what I would see as another problem. We have some words 
that act like adverbs, but never act like propositions in the old sense 
(as taking noun phrase head.) We have some (like /before/) that act as 
adverbs, take NP heads, and act as heads of clauses. We have others that 
can head a subordinate clause, but can't take NP heads and can't act 
alone as adverbs. So we have, in the same category (of preposition) some 
words that have nothing in common with each other. To me, that's the 
biggest stretch. It's like saying that some words are both nouns and 
verbs, some are adjectives and nouns, some are all three, so we should 
treat this as one category. The overlap is very real (and useful), but 
hardly seems to me a compelling argument for redrawing the 
classification lines. It seems easier to say that some words quite 
easily cross borders.
      I agree that adverb sometimes acts like trash can for words that 
don't fit other categories. Some words may be more central to the 
territory than others. I would certainly want to hold on to "away" and 
"now", though that may just be that I have seen them as adverbs so long 
that it's hard to let that go.
    The other quarrel I would have, which is my quarrel with formal 
grammar for the most part, is that we often seem to think of 
classification as the primary goal. The language is not neat and clean, 
and often there are functional ways to explain fluidity that  I just 
find more productive.
    But I may very well be falling short of understanding the full 
position. In that respect, I am happy to be corrected.

Craig

On 10/8/2010 12:50 PM, Brett Reynolds wrote:
>  On 2010-10-07, at 12:10 PM, Craig wrote:
>
> > I don't have firsthand exposure to Jesperson's arguments, so I'm
> > curious about ways in which they are both simpler and more robust
> > than traditional accounts.
>
>  He explains them in The Philosophy of Grammar beginning on p. 87.
> 
<http://books.google.ca/books?id=1WcXVIgc2bUC&lpg=PP1&dq=the%20philosophy%20of%20grammar&pg=PA87#v=onepage&q&f=false>
>
>
>  To summarize and add argument from the CGEL:
>
>  -The principle of Occam's razor dictates that "entities must not be
>  multiplied beyond necessity," or in Newton's words, "to the same
>  natural effects we must, so far as possible, assign the same causes."
>  Under this principle, it is better to assign words like 'before' to a
>  single part of speech than to three parts of speech unless we cannot
>  reasonably incorporate them in the same category.
>
>  -The etymology of the word 'preposition' has mislead us into thinking
>  that prepositions come only 'pre' a noun. It's important to remember
>  that labels are not definitions.
>
>  -Why should prepositions be limited to taking nouns as complements?
>  Nouns, verbs, and adjectives all take complements of different kinds
>  (including no complement, noun complement, adj complement, and
>  different kinds of clauses). The traditional analysis needs to
>  explain this discrepancy.
>
>  -In fact, traditional grammar, tacitly acknowledges the possibility
>  that prepositions take adjective complements (e.g., It is seen AS
>  possible) or preposition complements (e.g., FROM behind the counter),
>  without allowing for this in its descriptive/explanatory framework.
>
>  -The meaning of words like 'before' typically don't change when
>  they're 'adverbs' or 'subordinating conjunctions'.
>
>  -Adverbs can't function as predicate complements. (e.g., *He is
>  quickly.) But words like 'before' can, even when they have no
>  complement. Traditional grammar needs an ad hoc rule to deal with
>  this. Jespersen's conception doesn't.
>
>  -Prepositions can mostly modified by 'just' and 'right' where adverbs
>  can't, except words like 'before', even when they have no
>  complement.
>
>  In short, it's simpler because it requires less explanation to deal
>  with a variety of phenomena, because it reduces the heterogeny of the
>  adverb category and almost entirely does away with the subordinating
>  conjunction category (leaving only the subordinators: 'that' as in
>  "It's important that they are on time", 'for' as in "It's important
>  for them to be on time", 'to' as in "it's time to go", 'how' as in "I
>  know how it's done", 'if' as in "I wonder if it's true", and
>  'whether' as in "I wonder whether it's true"), and because it
>  requires students to make fewer choices (i.e., 'before' is always a
>  preposition). It's more robust because it has fewer exceptions.
>
> > I think about a word like "down." The dictionaries seem to be
> > saying that its adverbial meanings came first, the preposition
> > meanings later.
>
>  I'm not sure what you mean by the difference between an adverbial
>  meaning and prepositional meaning. The OED definitions for the adverb
>  entry and the preposition entry are almost identical. Do you mean its
>  use without an object came first followed later by its use with a
>  object? If so, that has little bearing on Jespersen's argument. It
>  doesn't need an object to be a preposition.
>
>  In fact, some of the earliest uses of 'down' are as predicate
>  complements: 1340 HAMPOLE Pr. Consc. 1602 Þus es þis world turned up
>  þat es doune. (My understanding of Old English is extremely limited,
>  but I believe this translates to "Thus is this world turned up that
>  is down.") As I pointed out above, adverbs do not function as
>  predicate complements. This is evidence that 'down' was a preposition
>  from the beginning, even if it didn't originally license objects.
>
>  Best, Brett
>
>  ----------------------- Brett Reynolds English Language Centre Humber
>  College Institute of Technology and Advanced Learning Toronto,
>  Ontario, Canada [log in to unmask]
>
>  To join or leave this LISTSERV list, please visit the list's web
>  interface at: http://listserv.muohio.edu/archives/ateg.html and
>  select "Join or leave the list"
>
>  Visit ATEG's web site at http://ateg.org/
>
>



To join or leave this LISTSERV list, please visit the list's web interface at:
     http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"

Visit ATEG's web site at http://ateg.org/