Demonstrative pronouns and the predicative position

[Swth\r]

Many grammarians say that the pronouns οὗτος, ὅδε, ἐκεῖνος are used as predicatives. I certainly understand the typical criterion for such a thesis, but are they really predicatives? If they distinguish the noun from other nouns, how can they be predicatives? οὗτος ὁ ἀνήρ = this man (and not another man), and the same with the other pronouns mentioned. What is your opinion?

[modus.irrealis]

Personally I think "predicate position" is just bad terminology that's been generalized from things like σοφὸς ὁ ἀνήρ where σοφός is predicative. It's probably due to the fact that in traditional grammar, demonstratives are treated as a kind of adjective, but they really belong to a different class, namely determiners. I suspect a more modern approach would treat these things differently and simple see the demonstrative as part of the noun phrase. But Ancient Greek does have some weird things like ἄκρον τὸ ὄρος -- I don't if any other languages have something similar with adjectives of this meaning and I don't know how these would be handled.

But I would say οὗτος can be the predicate in something like οὗτός εἰμὶ ἐγώ.

[Swth\r]

Personally I think "predicate position" is just bad terminology that's been generalized from things like σοφὸς ὁ ἀνήρ where σοφός is predicative. It's probably due to the fact that in traditional grammar, demonstratives are treated as a kind of adjective, but they really belong to a different class, namely determiners. I suspect a more modern approach would treat these things differently and simple see the demonstrative as part of the noun phrase. But Ancient Greek does have some weird things like ἄκρον τὸ ὄρος -- I don't if any other languages have something similar with adjectives of this meaning and I don't know how these would be handled.

But I would say οὗτος can be the predicate in something like οὗτός εἰμὶ ἐγώ.

In Greek all these appear because of the special dynamic of the article to make almost enything to be an adjective or a noun. The example ἄκρον τὸ ὄρος is not an exception, but a rule, if the author/speaker wants to put stress on a attribute not permanent and distinguishing a noun from other instances of itself. So τὸ ἄκρον ὄρος means "the mountain at the end", and not another mountain, whether τὸ ὄρος ἄκρον says that the mountain is concearned by means of its end/edge, "the end of the mountain", and not the middle, or something else or it. So it is with any adjective and most of the genitive constructions and the adverbial/prepositional markers: αἱ νῆες τῶν Ἀθηναίων - αἱ τῶν Ἀθηναίων νῆες: "The ships of the Athenians (and not their infantry, or anything else)" - "the Athenian ships (and not the Spartan nor the barbarian or anyone's else ships)". αἱ καιναὶ νῆες - αἱ νῆες καιναί: "the new ships (and not the old ones)" - "the ships (that are) new (and not old or anything else) in this moment". The same with participles like "ἧκεν ἔχων τοὺς ἱππέας τεθωρακισμένους" -"ἧκεν ἔχων τοὺς τεθωρακισμένους ἱππέας", etc.

As "determiners", if am not wrong, are described some grammatic categories by generative-trasformational grammar since the early 90's at least. The article itself (among things like e.g. numerals, or, in deeper structure, categories like "gender", "number" "case"...) is also included in this general class:
ἧκεν ἔχων τοὺς ἱππέας ---> [ἧκεν [ἔχων [τοὺς [ἱππέας]]]]. In the following examples: ἧκεν ἔχων τοὺς ἱππέας τεθωρακισμένους ---> [ἧκεν [ἔχων [τοὺς [ἱππέας]] τεθωρακισμένους ]], ἧκεν ἔχων τοὺς τεθωρακισμένους ἱππέας ---> [ἧκεν [ἔχων [τοὺς [τεθωρακισμένους [ἱππέας]]]]] the second one has the participle in this position, the first not.
But when the pronoun is used, it is never used this way. It is placed always without the article. E.g.: ὁ ἱππεύς οὗτος, and neither * ὁ οὗτος (ὁ) ἱππεύς, nor * οὗτος ἱππεύς. (A few exceptions only for the last case, especially with proper names.)

I would appreciate it if you could tell me what exactly you mean when saying "determiners".

So my question(s) may be reasked as: Why are these pronouns always in this postition? Is it a short of peculiarity or the meaning has something that we are perhaps missing?

[modus.irrealis]

So it is with any adjective and most of the genitive constructions and the adverbial/prepositional markers: αἱ νῆες τῶν Ἀθηναίων - αἱ τῶν Ἀθηναίων νῆες: "The ships of the Athenians (and not their infantry, or anything else)" - "the Athenian ships (and not the Spartan nor the barbarian or anyone's else ships)". αἱ καιναὶ νῆες - αἱ νῆες καιναί: "the new ships (and not the old ones)" - "the ships (that are) new (and not old or anything else) in this moment".

Does this really work with any adjective? I would have thought αἱ νῆες καιναί could only mean "the ships are new", and the distinction you're making would be between αἱ καιναὶ νῆες and αἱ νῆες αἱ καιναί. But I didn't mean that they were an exception within Greek -- only that I don't think I've ever come across adjectives like ἄκρος or μέσος being used this way in other languages.

I would appreciate it if you could tell me what exactly you mean when saying "determiners".

I'm not sure what you want me to say. "Determiners" are a class of words that basically limit what a noun phrase can refer to and usually includes things like article, demonstratives, interrogative adjectives (but it can depend on the language).

So my question(s) may be reasked as: Why are these pronouns always in this postition? Is it a short of peculiarity or the meaning has something that we are perhaps missing?

I would say you've already answered that, in that demonstratives pick out one thing out of many and don't assign a quality and that the position of ὁ ______ ἀνήρ is more or less limited to phrases that function as attributes, and demonstratives, being determiners, can't have this function. There's also the related question of why Greek uses the article together with the demonstrative (in most cases as you mentioned), since that seems to be rare across languages. I don't know what the answer is, but Greek (at least since Classical times and more so as you approach the present) is one of the languages that makes heaviest uses of the definite article.

[Swth\r]

So it is with any adjective and most of the genitive constructions and the adverbial/prepositional markers: αἱ νῆες τῶν Ἀθηναίων - αἱ τῶν Ἀθηναίων νῆες: "The ships of the Athenians (and not their infantry, or anything else)" - "the Athenian ships (and not the Spartan nor the barbarian or anyone's else ships)". αἱ καιναὶ νῆες - αἱ νῆες καιναί: "the new ships (and not the old ones)" - "the ships (that are) new (and not old or anything else) in this moment".

Does this really work with any adjective? I would have thought αἱ νῆες καιναί could only mean "the ships are new", and the distinction you're making would be between αἱ καιναὶ νῆες and αἱ νῆες αἱ καιναί. But I didn't mean that they were an exception within Greek -- only that I don't think I've ever come across adjectives like ἄκρος or μέσος being used this way in other languages.

It is not the same.
(αἱ) νῆες αἱ καιναί : the idea expessed by the substantive is contrasted with that of anothe saubstantive (implied or usually present in the phrase).
αἱ καιναὶ νῆες:the emphasis is on the attributive.
αἱ νῆες καιναί: the attributive is not contrasted with another object, but with itself.

See Raphael Kuehner for more details, paragraph 245 (i think it is the oldest edition, the one translated from german to english by Edwards-Taylor, in 1844). I think that also Rutherford's "First Greek Grammar: Syntax, 1912", says something like this but I cannot find the passage by the moment...

I'm not sure what you want me to say. "Determiners" are a class of words that basically limit what a noun phrase can refer to and usually includes things like article, demonstratives, interrogative adjectives (but it can depend on the language).

Ok, Now I have got this.

So my question(s) may be reasked as: Why are these pronouns always in this postition? Is it a short of peculiarity or the meaning has something that we are perhaps missing?

I would say you've already answered that, in that demonstratives pick out one thing out of many and don't assign a quality and that the position of ὁ ______ ἀνήρ is more or less limited to phrases that function as attributes, and demonstratives, being determiners, can't have this function. There's also the related question of why Greek uses the article together with the demonstrative (in most cases as you mentioned), since that seems to be rare across languages. I don't know what the answer is, but Greek (at least since Classical times and more so as you approach the present) is one of the languages that makes heaviest uses of the definite article.

Ok, so the "predicate position" is somehow missleading and problematic as far as some demonstrative pronouns are concearned. The position is similar to that of the "predicative adjective", but the meaning is not.

We should aso consider that other demostratives are not treatedthe same way: τοιούτους τοὺς λόγους is only used when the pronoun is pure predicate and not "determiner", the same with τοσούτους τοὺς λόγους. It normally goes like τοιούτους/τοιούσδε λόγους, τοσούτους/τοσούσδε λόγους, τους τοιούτους/τοιούσδε λόγους (but τους τοσούτους/τοσούσδε λόγους ?), τηλικούτους λόγους etc. I suppose this happens because those pronoununs have within themselfs some quality-quantity meaning.

[modus.irrealis]

I never did respond to this, did I?

It is not the same.
(αἱ) νῆες αἱ καιναί : the idea expessed by the substantive is contrasted with that of anothe saubstantive (implied or usually present in the phrase).
αἱ καιναὶ νῆες:the emphasis is on the attributive.
αἱ νῆες καιναί: the attributive is not contrasted with another object, but with itself.

See Raphael Kuehner for more details, paragraph 245 (i think it is the oldest edition, the one translated from german to english by Edwards-Taylor, in 1844). I think that also Rutherford's "First Greek Grammar: Syntax, 1912", says something like this but I cannot find the passage by the moment...

I need to rethink about the other stuff but here, I don't see why one needs to consider αἱ νῆες καιναί, i.e., I don't see why this forms a constituent within the sentence. What I mean is that in this case the adjective seems to be connected through the verb to the noun. Could αἱ νῆες καιναί for example be used as the answer to a question?

It seems very different than the case with οὗτος, where αὗται αἱ νῆες could easily be an answer as far as I can tell.

[Swth\r]

I never did respond to this, did I?

Nevermind!

What I mean is that in this case the adjective seems to be connected through the verb to the noun. Could αἱ νῆες καιναί for example be used as the answer to a question?

It seems very different than the case with οὗτος, where αὗται αἱ νῆες could easily be an answer as far as I can tell.

Correct observation! Indeed this construction needs a verb, or any verbal form, because the attribute lasts as long as the verbal notion lasts. The attribute given to the noun is not permanent, like (τὰς) καινὰς ναῡς. However I cannot understand your point made with the use of a question to show how this works.

E.g. in "καταλαμβάνουσι διηρπασμένας τὰς ἁμάξας μεστάς αλεύρων" (It is from "Anabasis", but I don't remember the exact cite), how would you expect the question to be like?

[modus.irrealis]

Correct observation! Indeed this construction needs a verb, or any verbal form, because the attribute lasts as long as the verbal notion lasts. The attribute given to the noun is not permanent, like (τὰς) καινὰς ναῡς. However I cannot understand your point made with the use of a question to show how this works.

E.g. in "καταλαμβάνουσι διηρπασμένας τὰς ἁμάξας μεστάς αλεύρων" (It is from "Anabasis", but I don't remember the exact cite), how would you expect the question to be like?

Any question at all. My understanding is that one of the tests used in linguistics to determine whether a group of words form a constituent is to see if that group of words can be an answer to a question, although it's only a necessary but not sufficient test, so only if it fails the test does it tell you anything. It's kind of hard without any native speakers of Ancient Greek but my own feeling is that τὰς ναῦς καινάς is not possible as an answer to any question (I switched to the accusative to avoid the fact that in the nominative it's a full sentence).

But I'm trying to remember what my point was here, though...

We should aso consider that other demostratives are not treatedthe same way: τοιούτους τοὺς λόγους is only used when the pronoun is pure predicate and not "determiner", the same with τοσούτους τοὺς λόγους. It normally goes like τοιούτους/τοιούσδε λόγους, τοσούτους/τοσούσδε λόγους, τους τοιούτους/τοιούσδε λόγους (but τους τοσούτους/τοσούσδε λόγους ?), τηλικούτους λόγους etc. I suppose this happens because those pronoununs have within themselfs some quality-quantity meaning.

I agree with this. τοιοῦτος by its meaning should function like any qualitative adjective and τοσοῦτος like a quantifier/numeral.

[Swth\r]

In my opinion it is impossible to form any question whose answer would be, for example, "τὰς ναῦς κενὰς", because those two words do not unify a solid phrase, as the attributive adjective with the noun in many cases (like μεγάλη πόλις => μεγαλόπολις.)We can also see this in the fact that often an attributive adjective stands as a noun - usually the noun omitted as supplied by context or as a general notion; so close the connection between those two is. The same seems to happen with the (the so called "predicative") adjectives "πᾶς,ἅπας, σύμπας, ὅλος, μέσος, ἄκρος, ἔσχατος, μόνος, ἥμισυς, ἀμφότεροι,ἕκαστος, ἑκάτερος" ,the demonstrative pronouns "οὗτος, ὅδε, ἐκεῖνος" and "αὐτός". We can ask who? where? and get answers by phrases like "ἡ πόλις ὅλη", "αὐτοὶ οἱ Ἀθηναῖοι" or "ὁ ποταμὸς μέσος" etc.

I suppose we could say -as Smyth describes in 1169- that the construction adjective-article-noun/article-noun-adjective "may be the equivalent of a complex clause" (he cites in 1168-1169 "τὰς τριήρεις ἀφείλκυσαν κενάς", "παρ' ἑκόντων τῶν ξυμμάχων τὴν ἡγεμονίαν ἔλαβον", "πόσον τι ἄγοι τὸ στράτευμα;" and more). The same in the sentence "Καταλαμβάνουσι τὰς ἁμάξας μεστὰς ἀλεύρων διηρπασμένας"; we may add/implie, e.g. the participle "οὔσας" between τὰς ἁμάξας and μεστάς. So "τὰς ἁμάξας (οὔσας) μεστὰς ἀλεύρων" = the carriages (by then) being full of flavour. But I think this is only a trick to explain the case, not an explanation of this construction... And discussion could go on further considering cases like "φέρονται οἱ λίθοι πολλοί", "τοὺς φρουροὺς ὑποσπόνδους ἀφεῖσαν", "ἄσμενοι ἔπεμπον τὴν ὠφελίαν", "μετεώρους παρετάξαντο τὰς ναῦς", "ὃς ἀφικόμενος τριταῖος" and more...

EDIT: Dear "mode.irrealis" having given a second thought on the matter, I have to ask you... If the adfective doesn't go with the noun in such phrases, with what word of the sentence then does it go?

[NateD26]

There's also the related question of why Greek uses the article together with the demonstrative (in most cases as you mentioned), since that seems to be rare across languages. I don't know what the answer is, but Greek (at least since Classical times and more so as you approach the present) is one of the languages that makes heaviest uses of the definite article.

hey, modus.irrealis. have you found an answer to this question? it keeps bugging me.

[modus.irrealis]

I suppose we could say -as Smyth describes in 1169- that the construction adjective-article-noun/article-noun-adjective "may be the equivalent of a complex clause" (he cites in 1168-1169 "τὰς τριήρεις ἀφείλκυσαν κενάς", "παρ' ἑκόντων τῶν ξυμμάχων τὴν ἡγεμονίαν ἔλαβον", "πόσον τι ἄγοι τὸ στράτευμα;" and more). The same in the sentence "Καταλαμβάνουσι τὰς ἁμάξας μεστὰς ἀλεύρων διηρπασμένας"; we may add/implie, e.g. the participle "οὔσας" between τὰς ἁμάξας and μεστάς. So "τὰς ἁμάξας (οὔσας) μεστὰς ἀλεύρων" = the carriages (by then) being full of flavour. But I think this is only a trick to explain the case, not an explanation of this construction...

I agree, also because I don't think you can add a participle in all cases, with πρῶτος for example. In fact I would say that the relation is the other way: the circumstantial use of the participle is derived from the predicate use of the adjective, because the participle has a number of adjective-like properties. But I think the predicate adjective is equivalent to a participle clause in the sense that it fits into the same syntactic slot so to speak.

EDIT: Dear "mode.irrealis" having given a second thought on the matter, I have to ask you... If the adfective doesn't go with the noun in such phrases, with what word of the sentence then does it go?

I guess it really depends on what you mean by "go with". In one sense it obviously goes with the noun because it agrees with it in number and gender, but I don't think it goes with it in the sense of forming a separate constituent within the sentence.

[modus.irrealis]

hey, modus.irrealis. have you found an answer to this question? it keeps bugging me.

Unfortunately, no. The only thing I've come across is that there is a pattern as to what the article is used with, something like languages first use it with specific things, then abstract ideas, then possessive pronouns, demonstratives, personal names, until eventually it loses its definite meaning altogether. English is near the beginning of the process, Romance languages are somewhere in the middle, Greek is at the end, and if I remember correctly, Aramaic is an example of a language that's gone through the whole process and what used to be the definite article became a mere noun ending. But I haven't seen any explanation for what drives this pattern.

[Swth\r]

Dear "mode.irrealis" having given a second thought on the matter, I have to ask you... If the adfective doesn't go with the noun in such phrases, with what word of the sentence then does it go?

I guess it really depends on what you mean by "go with". In one sense it obviously goes with the noun because it agrees with it in number and gender, but I don't think it goes with it in the sense of forming a separate constituent within the sentence.

By"Goes with" I meant a syntactic construcion of any kind, a syntactic relationship.

[modus.irrealis]

Then I would say no, although any answer really requires some kind of grammatical framework in which syntactic relationship is well-defined, and I'm not familiar enough with linguistics to provide one of those. I would just say that a predicate adjective has the same sort of relationship that an adverb or a subordinate clause has and that it doesn't have a special relation to the noun. Do you think there's a relation with the noun in this case? In something like φέρονται οἱ λίθοι πολλοί it seems misleading to say that οἱ λίθοι πολλοί is a way of saying "(the) many stones". (By the way, this seems like a cases where you couldn't use εἰμί -- it doesn't seem possible to say φέρονται οἱ λίθοι ὄντες πολλοί without a change in meaning.)

[Swth\r]

Yes, there is no way we can use a participle of ΕΙΜΙ here. I think the meaning is "the stones are many as long as they are being carried". Or we could translate the adjective as "in a big ammoumt". It has more adverbial meaning.
But it is indeed a interesting matter (to me at least).

We could compare the situation to the fact (as somewhere H.W.Smyth points out) that circumstantial participles take the "predicative" position (=with no article). Perhaps the use of adjectives this way has to something to do with that...