Re: Grammar as patterns
Bob,

I know we’re rehashing some familiar positions here, or maybe we’re following rules, or a pre-existing pattern (a point which, I cannot help but point out, we can probably recognize even though the actual sentences used in the related older postings aren’t the same; and all this without positing a Universal Listserv Argument Grammar). Still... you’re implying that modern pattern-based approaches don’t use abstract grammatical categories, but they do (at least, in a sense that’s relevant here).

I don’t know offhand of any pattern grammar that posits that all patterns have specific individual words in all the slots. That wouldn’t be a pattern anyway; it’d be an instance. A category that can be realized as a range of lexical items that occur in specific configurations counts as a grammatical category, I’d think. It may not be only a grammatical category, but that’s a different — important, but different -- issue.

Here’s a different way to get at the point of disagreement, although it runs the risk of setting up a straw man. Suppose we have two strategies for deciding what to try to accomplish something specific in a language we know we don’t know well yet:

(1) Come up with a list off all possible configurations (or rules) that could be made with the categories you know so far, and randomly test-fire them.
(2) Take a couple of configurations (or rules) that you already know work for a related purpose, and start by test-firing one or two tweaked versions of one of them.

Strategy (1) is likely to give you tons of false hits; you won’t get what you want a good deal of the time — but it’s darn creative. If we assume (1), and If it turns out that what’s actually produced isn’t the kind of thing we’d get from farming all the a priori possibilities, it makes sense that something must be constraining those possibilities (and thus there’s a clear need for a UG). Strategy (2), in effect, uses caution, or maybe pragmatism, as a limiter. It’s a bit like deciding that if you’ve been using onions in a recipe, and you’re out of onions today, maybe leeks would work better than chocolate as a stand-in.  You won’t get noticed as a breakthrough chef, but your diner won’t go out of business.

Some approaches to grammar are based on supposing that children use strategy (1) and others assume that children use strategy (2). There were approaches based on a kind of strategy (0), which claimed that children didn’t do (1), but didn’t try to do things with language either — that their use of language was rather like Pavlov’s dogs’ use of drool. No one likes the strategy (0) approaches, really (unless they’ve been trained to). I don’t think we’re at a point where we can say children definitely use a particular strategy to construct language, but I’d argue we can say that we can’t rule out strategy (1) or strategy (2), and thus benefit from strands of research devoted to each.

Bill Spruiell
Dept. of English
Central Michigan University
 



Robert Yates wrote:

Colleagues,

Whether grammar is a set of rules or a set of  patterns (learned from the input we get) is a discussion that has occurred before on this list.

If I understand the following correctly, (Craig writes:)

"we are dealing with flexible, dynamic patterns sustained and reinforced by use"

then the claim is that we do not know very much about grammatical categories.  Such categories are the result of the "patterns" we are exposed to.  There are all kinds of examples I could cite to show how such a common sense idea is problematic, but let's consider two pairs of sentences.

Sentences 1 and 2 clearly have different meanings.

1) Bob needs someone to work for.
2)  Bob needs someone to work for him.

In 1, Bob wants to be the worker, and in 2, Bob is an employer.

What is the "pattern" we acquired that lead to those interpretations?  It is not just the presence or absence of the pronoun. Sentences 3 and 4 have the same meaning.

3) These are the letters Bob threw away without reading.
4) There are the letters Bob threw away without reading them.

Without making reference to abstract grammatical categories, I have no idea how to explain the meanings of sentences 1-4.

These sentences suggest there is something incomplete in a claim that our knowledge of language is based on patterns we perceive from the input.

Finally, Craig and I fundamentally agree on one point.

There are those who say there is little value in making
these conscious. I would disagree with that as well.

I could not agree more -- there is great value in making conscious the knowledge of language that we all have.

Bob Yates, University of Central Missouri

  
 



Craig Hancock <[log in to unmask]> <mailto:[log in to unmask]>  9/16/2010 7:47 AM >>>
        
 
 


Eduard,
    I agree that we are in rough agreement and apologize for making my
post seem like something else.
   A big question might be whether the "rules" are there before use (and
thus predetermine it to large extent) or whether we are dealing with
flexible, dynamic patterns sustained and reinforced by use. I would
embrace the latter, sometimes called "usage-based." Some people would
see grammatical forms as meaning-neutral (semantically and
pragmatically), with meanings added through the lexicon. It is also
possible to see that they are meaningful in their own right, deeply tied
to both cognition and discourse.
    Patterns are sustained to the extent that we find them highly
productive. From this view, form ENABLES rather than constrains. The
rules of prescriptive grammar tell us what we are not supposed to do.
But without the natural grammar, no substantial meaning is possible.
Frequency of a construct can also make us unaware of the contributions
it is making. There are those who say there is little value in making
these conscious. I would disagree with that as well.
   To me, the challenge has always been how to present views like this
on the list as perspective, not as argument. People like Bybee are doing
wonderful work along these lines, and it would be good for the list to
be aware of it.

Craig

Eduard Hanganu wrote:
  
 

Craig,

I have no problem with the way you express the matters because I don't
see too much of a difference between what I state and what you state.
True, some elements of a category (word class) are more central and
reflect better the basic characteristics of that class. Other elements
are borderline or peripheral, and their characteristics intersect with
or overlap the characteristics of peripheral or borderline elements of
another class. On the whole, though, there are "standard" elements of
word classes, and there are "peripheral" elements of such word
classes. Denial of such facts, though, is a denial of the empirical
evidence that concerns what I stated above.

Some people continue to believe that the Latin language structure is
artificially superimposed on the English language, but they forget
that language is a social phenomenon, and that we humans do
construct language structure implicitly or explicitly. This fact is
evident from information collected from humans who had never been
socialized in language. Those people don't speak a human language, and
if they are beyond the critical period of language acquisition they
are never able to acquire language, except for a few unstructured
rudiments.

If there is an "universal grammar" as Chomsky has been claiming for
more than five decades, no linguist or other kind of scholar has been
able to provide evidence for the claim. So, we remain with what is
observable: language is a human construct, and whether we
differentiate between acquisition and learning or not, the bare truth
is that without socialization in language no human will speak a human
language.

Eduard



----- Original Message -----
From: Craig Hancock <[log in to unmask]> <mailto:[log in to unmask]>
Date: Wednesday, September 15, 2010 19:16
Subject: Re: like
To: [log in to unmask]

    
 

Eduard,
    I would express it somewhat differently.
Frequency is often
self-reinforcing. Frequency makes something more accessible for use,
which in turn makes it more frequent. And so on.
    I just asked a friend how she likes her new
job (from teacher to
counselor), and she said "I'm liking it." It occured to me that she
might not have said that without the influence of the McDonald's ad.
Progressive is not common with stative verbs, but an ad campaign can
change that.
    Rather than intersection of word classes, it
might be more of an issue
of centrality. Some elements of the category are more central than
others, some more borderline or peripheral.
    You also have a tendency (from that cognitive
frame of reference) to
see far more lower level constructions. It's much more a
lexico-grammar than a set of abstract rules. (Pattern is closer than
rule.) A great deal of language includes set constructions, many of
them with their own more local patterns. So it could be that "like"
brings with it a unique kind of grammar.

Craig>

Geoff,
      
 

You probably did not have time to read "Frequency of Use and the
Organization of Language" by Joan Bybee, in which the author, after
decades of research, documents that language organizes itself,
        
 

and that
      
 

parts of speech or word classes are not an idiot's fantasy,
        
 

but one way in
      
 

which language acquires and shows structure. These word
        
 

classes are real,
      
 

and understanding them makes a great difference when one
        
 

learns a
      
 

language. That difference goes beyond boundaries, which are
        
 

nothing more
      
 

than points at which word classes intersect. To inflate the
        
 

importance of
      
 

these points of intersection to a generality (which is a
        
 

fallacy) shows
      
 

lack of understanding of the role of morphology and syntax in the
production and conveyance of meaning - the main functions of
        
 

language.>
      
 

Eduard

----- Original Message -----
From: Geoffrey Layton <[log in to unmask]> <mailto:[log in to unmask]>
Date: Wednesday, September 15, 2010 16:13
Subject: Re: like
To: [log in to unmask]

        
 

Craig - I know we've had this discussion before, but my reaction
is "what difference does it make what we call it?"  I don't
see how you can have anything except flexible boundaries, which
then leads to the more interesting question of the rhetorical
effect of "shading" into a verb - what happens to the meaning of
the sentence? Labeling the choices as preopositions, adjectives
or verbs really doesn't go very far to answer this question.

Geoff Layton

          
 

Craig,

My first reaction was that this use of "like" was adjectival,
            
 

but since you want a traditional treatment I checked the OED
Online and Merriam Webster Dictionary Online. Both treat as an
adjective, although MW doesn't have an example with BE.
          
 

Herb
            
 


I am curious about how traditional grammar handles "like"
            
 


 
in a
      
 


sentence like "One of these things is not like the others." (I
know; Sesame Street).
          
 
 
My instinct is to say "like the others" is prepositional
            
 
 
phrase, complement to "is", therefore referring back
(adjectivally?) to "One of these things." Would that be standard?
          
 
 
If it can be easily replaced by "resembles" (or "doesn't
            
 
 
resemble"), does that mean "be like" is shading into a verb like
status with "the others" as object? Are we OK with flexible
boundaries around our categories?
          
 
 

            
 
 
Craig
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/

        
 
 
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/

    
 
 

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/


  

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/