Re: Greenbaum's word classes
Herb, Craig, and all,

When I first became acquainted with the form/structure class distinction back in the 60s, when I was introduced to structural grammar, I felt the first stirrings of anti-traditional-eight-parts-of-speech grammar.  It makes such logical sense to distinguish the four large open classes from the others, a distinction that makes use of our internal grammar computer.

I used to tell my students about a conversation I once had with my fifth-grade son many years ago (he's now 48:  "new grammar" isn't all that new!).  He was doing his grammar homework and asked me, "What part of speech is 'Wednesday'?"  His traditional "person, place, or thing" meaning-based definition wasn't working for him.  My response: "Can you make it plural?"  He knew right away that not only could he make it plural, but that his ability to do so meant that "Wednesday" is a noun.

I have told that story in various presentations through the years.  For many teachers,  those who have not gone beyond traditional grammar, my simple question comes as a revelation.  Nothing is more powerful than using the form of the four form classes to enable students to use their inner language expertise.

I actually use three criteria in my definitions:  form; signalers; function.

"A noun is a word that can be made plural and/or possessive; it can have certain derivational affixes, such as -tion, -ness, -ment, etc.   It is signaled by determiners.  It functions as the headword of a noun phrase."

"A verb is a word that has both present and past tense.  (I sometimes simply say, "a word that has both an -s and an -ing ending.  That works for all verbs.) It can have certain derivational affixes, such as -ize, -en, and -ify.  It is signaled by auxiliaries.  It functions as the headword of a verb phrase."

"An adjective is a word that can be made comparative (with -er or more) and superlative (with -est or most); it can have certain derivational affixes, such as -ous, -ful, and -able.  it can be signaled by qualifiers. It functions as a modifier of nouns.

I must admit that "Adverb" is fairly troublesome to define.  I often revert to meaning here.  But certainly many thousands of our adverbs can be identified by the -ly that is added to adjectives.  The -ly adverbs, which are adverbs of manner,  can be made comparative and superlative.  There are other adverbs without formal distinctions that add information of time, frequency,  place, etc.  Adverbs can also be qualified by words like very.  They function as modifiers of the verb and of the sentence as a whole.

My structure classes are very much Greenbaum's, except

(1) I do not include "interjection."  Among the traditional eight, interjection is the only one that is defined strictly according to emotion.  Yet most emotional words are nouns and verbs.   With the exception of  "wow" and "oh" and few others, this is not a word category.  In my book, I include interjection as a subcategory of sentence modifier (along with adverbs, prepositional phrases, clauses, etc.)

(2) I include the "qualifier" as a word class.   It's a small class, true.  But words like "very" and "rather" do indeed have a legitimate function.  In traditional grammar, they are included in the adverb category  ("modifiers of verbs, adjectives, and other adverbs").  But clearly, the qualifiers are separate from the adverbs that modify verbs.

(3)  I include "pronouns."   While it's true that many subclasses of pronouns (possessives, indefinites, demonstratives) function as determiners, the personal and reflexive pronouns have a legitimate function beyond that of determiner as stand-ins for nominals.  Pronouns are an especially important class, I believe, in helping students understand their subconscious grammar competence.  For example, students can easily find the dividing line between subject and predicate in most sentences  by substituting a pronoun for the subject--no matter how many modifiers that subject may have.  Another test of  pronoun power is the tag-question.  When a statement is turned into a tag question, the subject is automatically turned into a pronoun.

Her's an example:  "Another test of pronoun power is the tag-question, isn't it?"  (Note that the pronoun "it"  has taken the place of the complete subject .)

(4)  I don't include "numbers," although I do list "number" as a subclass of determiner, along with articles, the various pronouns, and possessive nouns (and possessive noun phrases).


Something I make clear in my discussion of word classes is that "determiner" and "qualifier" are not clearly word classes like the others, such as conjunctions and prepositions and modal auxiliaries; they can also be considered "functions."  I make that distinction because there are words we use as determiners and as qualifiers that are members of the open classes:  possessive nouns function as determiners (Craig's book; the dog's tail); adverbs function as qualifiers (exceptionally bright; devastatingly handsome).  In other words,  nether determiners nor qualifiers are really closed classes.

Fries, in his original description of word classes, calls all of them (he lists 15) other than the four form classes "function" classes.   I prefer the term "structure" class.  I think Greenbaum does too.)

That's my take on word classes.

I will save my lecture on the importance of the -al classes (nominal, adjectival, adverbial) for another posting.

Martha

 

 


 





Craig,

"Two sevens" is a good example of "seven" functioning as a noun, and
with a very specific meaning.  Not a problem since many words can belong
to more than one class.

Wasn't it Ursula Bellugi-Klima who did some early research on children's
command of morphological productivity?  She made up nonsense words, like
"wug", and put them in environments where the child would have to use
the word in the plural or some other inflected form.  Children did this
without prompting and with a high degree of accuracy, as I recall.  It's
been a long time since I've looked at that paper.

However, children show in all sorts of ways that they have a command of
regular vs. irregular morphology, as when they use verb forms like
"knowed".  I shouldn't think they would have much difficulty
understanding some simple morphology in the early grades.

Herb


Herb,
   I can think of occassions when a number would take a plural, as in
"He
threw two sevens in a row." (Maybe I'm confessing my social class.) But
you have me convinced.
   For verbs, adding endings may be the simplest and purest test. I
wonder
why we don't do that in early grades. I wonder if syntax is hard to
teach later precisely because we have avoided it for so long. Even at
the college level, I still have students who haven't been told the
difference between plural and possessive (or it hasn't taken; but it
takes so easily, I can't help suspecting it just isn't always taught.)
Are syntax and morphology hard to teach, or do we just avoid them?
   I have access to students in a K-4 charter school, so I may be trying
a
few things out.

Craig


Craig,
>
> I agree that the four open classes are the clearest, easiest to define
> notionally, and probably the best to present at lower levels.  As to
> auxiliaries, "have", "do" and "be" are examples of words that can be
in
> two different classes, auxiliary and verb, and their behavior differs
> depending on which class a particular example represents.  Multiple
> class membership and auxiliary are both items for later introduction,
I
> would think.
>
> As to the status of "numeral", number words do not inflect.
Adjectives
> do.  Number words can take derivational suffixes like "-th" and
"-some",
> which adjectives cannot take.  Notionally, they can't be comparative
or
> superlative.  Number words must occur initially in a noun phrase or
> between the determiner and any adjectives, so they are syntactically
> distinct as well.  They are like adjectives in that they are
> post-determiner, pronominal, and are attributive and can, to a degree,
> be predicative, as in the somewhat archaic "Now they were three",
> indicating number, not age.  Unlike nouns, they don't take plurals,
and
> when they are used as sole noun phrases, as in "I saw three on the
> table", they are usually elliptical ("three books").  So there are
both
> morphological and syntactic criteria for considering numerals a word
> class.  However, it's one I'd present rather late in K12.
>
> Like you, I'm uncomfortable with "adverb".  The distinctions among
> sentence-modifying, verb phrase modifying, and adjective-modifying
words
> are too great, from my perspective, to allow them to be properties of
> one word class.  I would prefer at least distinguishing intensifiers,
> like "very" and "sort of".  But it's also a practical pedagogical
> question, and I wouldn't oppose keeping the term as is.  I would,
later
> on in K12, want to distinguish carefully among types of adverb,
though.
>
> I'm not comfortable speculating on scope and sequence, since I don't
> teach K12 and have no expertise in K12 curriculum.  I would want such
> decisions made by people who know those areas well.
>
> Herb
>
> -----Original Message-----
> From: Assembly for the Teaching of English Grammar
> [mailto:[log in to unmask]] On Behalf Of Craig Hancock
> Sent: Thursday, August 10, 2006 8:42 PM
> To: [log in to unmask]
> Subject: Re: Greenbaum's word classes
>
> Herb,
>    I think this is a wonderful place to start. A number of people have
> suggested the four "open classes", and so maybe we can start with that
> as a consensus position. (These also carry over from traditional
> grammar.) Notional criteria seem the current way of going about it for
> early grades, so I wonder if people think morphology and syntax are a
> more mature perspective. I notice even with the NATE glossary (I don't
> have it with me) they tend to oversimplify in the early grades. They
> define "subject", for example, as "what carries out the action," which
> seems a terrible mistake to me.
>    It's hard to imagine getting far without prepositions and
> conjunctions.
> For auxiliaries, you need to determine whether "have", "do", and "be"
> verbs are verbs used as auxiliaries, which would mean it's a function
> label and not just a category label (as it would be for the modals, I
> presume.) What are the arguments for numeral as its own catgory and
not
> just noun or adjective?
>    I know we have talked a number of times on list about the category
> "adverb" being too large. Do we want to add "qualifier"? It shows up
> very early, I think, with "so" and "very". Unlike other "adverbs",
they
> can't head a phrase.     >
>    I like the idea of "typical" or "prototypical", especially for
> notional
> definitions. Even in early stages, I would opt for presenting language
> as very flexible.
>    These categories would have sub-categories, I assume. At what age
> would
> we assume that a typical child is ready for a full description?
>
> Craig
> In a previous posting, I mentioned Greenbaum's treatment of word
classes
>> in The Oxford English Grammar (OUP 1996).  I thought I'd summarize
> what
>> he lays out (pp. 90-95).
>>
>>
>>
>> He proposes four open classes (noun, verb, adjective, adverb) and
> seven
>> closed classes (auxiliary, conjunction, preposition, determiner,
>> pronoun, numeral, and interjection) and notes that many words belong
> to
>> more than one class.  In his treatment of the classes, he combines
>> determiner and pronoun into one section because there is a great deal
> of
>> overlap between them, even though there are words, like "the" and
> "she",
>> that are clearly one or the other.  (It's a good example of the fact
>> that category boundaries are fuzzy.)  In his two-page discussion of
> the
>> criteria that are used to determine word classes and their membership
> he
>> presents three types of criterion, notional, morphological, and
>> grammatical (syntactic), with the combination of morphological and
>> grammatical being the most useful where inflectional variants or
> affixal
>> characteristics are available.  For word classes that don't have
>> morphological variants, like prepositions and conjunctions, notional
> and
>> grammatical criteria work better.  He "notes that notional criteria
> are
>> often a useful entry to a recognition of a class."  He also touches
on
>> the notion "prototype", commenting that "some members of a class are
>> central (or prototypical), whereas others are more peripheral",
> pointing
>> out that "tall" is a central member of the adjective class because it
>> exhibits all the criteria of adjectives while "afraid" is peripheral
>> since it can only be predicative.  He points out also that members of
> a
>> class may contain more than one word, like "book review", "no one",
or
>> "in spite of", which are a compound noun, pronoun, and preposition,
>> respectively.
>>
>>
>>
>> I'm not suggesting that we simply adopt Greenbaum's description but
>> rather that it is a useful starting point for part of speech
> terminology
>> and concepts.  Clearly any such system must be analyzed in terms of
>> scope and sequence, deciding which criteria and which categories to
>> present when and in which order.  I'm also not suggesting that
>> terminology be limited to parts of speech.  Johanna's proposal is, I
>> think, an excellent place to start for more comprehensive
terminology.
>>
>>
>>
>> Herb
>>
>>
>> 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/