Bruce,

This will be a highly theoretical post -- warning to those who are fed
up with them!

Prototype and schema are very well-known terms in cognitive and
functional linguistics, and certain subfields of psycholinguistics and
psychology. The work of Eleanor Rosch made prototypes famous, and the
work of Schank & Abelson, Rumelhart (and others) made schemas famous.

As to syntax and Cog. Gr., as I said, syntactic patterns are symbolic
units in CG; in other words, syntactic rules are meaningful units of
language. Their meanings are quite schematic (underspecified) but they
are concepts nonetheless.

I suggest you not judge the adequacy of CG as a theory of syntax until
you have read about it in detail. It has accountings for every example
you give.

furniture (singular in form and syntax and plural in signification and
concept)

This is a mass or collective noun. These nouns are generally singular in
form but plural in concept. The grammatical singularity of nouns like
this is due to a construal which foregrounds the unity of the member
items over their diversity. The users of a speech community decide what
counts as a mass vs. a count noun. In Arabic, for ex., the word for
'olive' is a mass noun. In order to say 'one olive', you have to use a
singulative suffix. Some  languages have plural words for furniture
(French and German, if I remember correctly).

news (plural in form but singular in syntax and concept)

It would be interesting to look up the origin  of this word and see if
it was ever grammatically plural. I don't have time to do so at the
moment. Whether something gets to be grammatically singular or plural
depends on the conceptual construal (what is foregrounded and
backgrounded, along with other things) of the concept. As we know, some
collective or semantically plural concepts (group; government; couple)
are grammatically singular in Amer. English but plural in British Engl.
This means that the Americans and British construe the concepts
differently; Brits foreground the plurality of the membership; Americans
foreground the unity of the group--what features unite them. Construal
is a major tool of CG for describing the semantics of an expression.

Queen Mary (the ship is neuter in concept but refered to with feminine
pronominal forms)
This is a cultural tradition. Cats are also usually 'she'. This might be
based on a metaphor, alive or dead, or some kind of stereotyping (with
respect to cats). There is a certain amount of language that is driven
merely by convention. In CG, we try to keep this explanation to a minimum.

the sun (common noun in form and use but proper conceptually)

We use 'the' because when we do, we know we are talking about the sun of
our solar system, not some other sun. Therefore we are signalling the
sun we all know about--hence it is definite. There are, of course, other
suns in the universe; each solar system has one. This is
discourse-definite. Another explanation is needed for cultures without
knowledge of other suns, but I don't have the wherewithal to look it up
or develop it right now.

resemble (verb in form but adjectival conceptually, "be like")

This is a temporal, but imperfective relation in CG terms (roughly
speaking,  a state verb). Temporal means that the development of the
situation (similarity between subject and complement) over time is
foregrounded in the semantics of the verb. It is imperfective in the
sense that it features no change of state over time. We can easily make
it perfective (and hence accepting of  progressive aspect) by imagining
an increase or decrease in resemblance over time: "As she grows, Nadia
is resembling her mother more and more" or "less and less".

be (verb in form but almost empty conceptually)

This emptiness is what we call schematicity in CG: a highly skeletal
semantic structure. All it does is relate the subject and the
complement, supplying only tense/aspect.  The fact that you can do
without the tense-aspect marking is attested in the numerous langauges
that have the zero copula (e.g. African-American English "he my
teacher"). Tense/aspect is either supplied by context or assumed to be
general present.

In CG, verbs are "processes": Their semantics feature the evolution (or
lack thereof) of a relationship or situation over time; time is
foregrounded in their semantics. They are relational because they relate
the subject  either to the situation itself (intransitives) or to a
complement. Nouns are atemporal things. Thing is a technical term in CG,
with a semantic definition. Basically it is an entity that is being
signalled as discrete from other entities; of course there is much more
to it than that. Adjectives and many other categories of word
(preposition, e.g.) are atemporal relations: they feature a relation
between two entities, but a timeline is not  foregrounded in their
semantics (more detail is required in cases like "soon", which does
involve relative time).

Cognitive grammar typically appears incoherent to people trained in any
kind of generative tradition. People who approach it who have no
linguistics background usually find it very sensible. There is now a
quite large collection of work using CG to account for all kinds of
phenomena, including the usual main defenses of formal syntax, such as
binding. It uses the same standards for testing its accountings as
formal linguistics: grammaticality judgments; non-occurrence of
logically possible structures; and appearance in corpora (valued a
little more in these schools, which view the competence/performance
distinction differently).

I once taught both formal and cognitive syntax in a grad class at the U
of Montana. In doing so, I often found that the formal accountings
stopped just short of relating the syntax to the semantics. It was very
enlightening. One of the first objections to CG among formalists is that
it looks, at first glance, like anything goes. This is not true. But
construal is flexible, and how different languages code different
concepts and therefore have them function differently in syntax is the
proof of this.

By the way, at both the Master's and the Ph.D. levels, I was trained in
formal, cognitive, and American functional syntax (mainly the first
two). The formal theories I  learned over those ten years are now
outdated (my last course was in the mid-late '80s), but I certainly had
plenty of opportunity to compare the two. I also did enough generative
phonology to have specialized in it for my dissertation (I decided not
to). I ceased following that at the dawn of Optimality Theory (which, by
the way, is functionalism in disguise--it's a formalization of
preference laws).

Major names in cognitive semantics, cognitive linguistics, and Cognitive
Grammar are Leonard Talmy, Ronald W. Langacker, George Lakoff,  Gilles
Fauconnier, and Joan Bybee. These are among the founders of the field. I
have left some important others out, I'm sure. Sadly, my teaching load
at Cal Poly and my focus on grammar pedagogy have deprived me of time to
keep up even with this, my chosen specialization.

***************************************************
Johanna Rubba, Associate Professor, Linguistics
English Department, Cal Poly State University
San Luis Obispo, CA 93407
Tel. 805-756-2184 ~ Dept. phone 805-756-2596
Dept. fax: 805-756-6374 ~  E-mail: [log in to unmask]
URL: http://www.cla.calpoly.edu/~jrubba
***************************************************

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/