the attributive and predicate positions

Carl W. Conrad cwconrad at
Tue Jan 23 06:46:49 EST 2001

I really regret how long this is getting and I think I might try to cut out
some of the middle part so as to highlight my primary response to Moon:

At 11:27 PM -0500 1/22/01, Moon-Ryul Jung wrote:
>> >I learned several foreign languages, but never heard of
>> >"predicate position" in relation to noun phrases.
>> >I wonder why understanding Greek noun phrases needs such a notion?
>> I've learned several languages too, although not enough to become a
>> linguist, but (1) I don't know any other language that uses the article in
>> anything like the way Greek does; (2) I don't know any other language that
>> consistently omits the verb in a "noun sentence". That is why some term
>> referring to what is called traditionally "predicate position" seems
>> helpful to me--at least as a pedagogical device for teaching beginning
>> Greek--to point to the semantic difference made by positioning an adjective
>> INSIDE an article, where it is the attribute of a noun which it qualifies,
>> and positioning the adject OUTSIDE of the article-noun group, where it
>> serves as a predicate word;
>So far so good. The notion of predicate position is useful to
>differentiate the two roles of an adjective: attributive and predicative.
>As Iver noted in his response to my post, Greek is among the languages
>adjectives are used as predicates without linking verbs. They could be
>called "adjective-verbs" when used as predicates. My mother tongue, Korea,
>is such a language as well. But Korean is different from Greek in that
>attributive adjectives are differentiated from predicative adjectives
>by having different endings. (Korean is an "inflectional" language). In
>this difference is marked by the position of adjectives. So, we need the
>notion of attributive and predicative positions in Greek.
>(3) then the fact that PAS and demonstrative
>> pronouns must also take this position normally or otherwise the semantic
>> value of PAS and similar quantitative words is altered is added on.
>My original question was whether it is recommendable to apply the notion
>of "predicate position" to these cases, that is, NP's
>where the modifiers are OUTSIDE OF the article-noun configurations.
>Yes, PAS and similar quantiative words can be placed OUTSIDE OF
>the article-noun configurations. But it seems confusing to call that
>"predicate position", because the word at that position is not a prediate
>in the case of noun phrases.
>For that matter, even in English "all" assumes that position. Only
>difference seems that PAS can precede or follow the article-noun
>configuration in
>Greek. But that is simply because Greek is a language of free word order.
>As Iver argued convincingly, PAS has two semantic meanings, one as a
>quantifier ("all") and the other as an attritutive or descriptive
>("whole"). PAS as a descriptive adjective is placed within the
>configuration and PAS as a quantifier is placed outside of the
>configuration. This situation is different from ordinary adjectives.
>In this case, the position determines the function (either as an
>adjective or predicative adjective). But in the case of PAS, the function
>seems to determine the position. Moreover, both positions, i.e.
>within and outside of the article-noun configurations, are RESTRICTIVE in
>that they help determine the referent of the noun phrase.
>So, the crude taxomony of positions (to be used pedagogically at least) I
>would like to suggest is:
>Predicative Position - the position of adjectives, nouns, and adverbials
>used as predicates in sentences.
>Restrictive Position - the position of adjectives, nouns, and adverbials
>used as modifiers in noun phrase.
>- Attributive Position - the position of descriptive adjectives, nouns,
>adverbials, i.e. within the article-noun
>- Quantitative Positionn - the position of demonstratives and
>i.e. outside of the article-noun configuration.

Am I to take it that these last two, "Attributive Position" and
"Quantitative Position" are SUBCATEGORIES of "Restrictive Position"? Where
do demonstratives like hOUTOS, hODE, and EKEINOS go? What really disturbs
me about this is the same thing that disturbs me about Wallace's grammar:
the multiplication of categories for the sake of making distinctions that
are perhaps semantically valuable because of information they convey about
translating the word into a target language even if they describe the same
actual construction or, in this case, word-order phenomenon. The problem
here is comparable to the problem of the word "tense": we know very well
that outside of the indicative, the present and the aorist and the perfect
do not refer to TIME at all; does somebody have a better word to apply to
the Present, Aorist, and Perfect Greek verbal systems that liberates us
from the inadequacy of our terminology and still allows us to teach
beginners without making the 5,000 odd facts about Greek become more like
10,000? I reallize I'm exaggerating here, of course, rhetorically. There
couldn't possibly be 5,000 things about the Greek language that a
first-year Greek student must master, could there? And that's a rhetorical
question, isn't it? Or is it?


Carl W. Conrad
Department of Classics/Washington University
One Brookings Drive/St. Louis, MO, USA 63130/(314) 935-4018
Home: 7222 Colgate Ave./St. Louis, MO 63130/(314) 726-5649
cwconrad at

More information about the B-Greek mailing list