6.
mvp-aristotle-comment-729
σύνδεσμος.
The traditional text of this and the following section (that on the
ἄρθρον) is confessedly corrupt.
Twining finds a good deal of it unintelligible (comp.
Steinthal,
Gesch. der Sprachwissen- schaft 1 p. 263
); and
Vahlen, besides bracketing
ἢ φωνὴ μέσου in
1457a 8-10
, has to assume two lacunas, one after
συντίθεσθαι in
1457a 2
, and another after
δηλοῖ
in
1457a 7
. At this point
Aristotle passes from the constituents of words (letters and syllables) to actual words; and he begins by noting this distinction: some words are
φωναὶ ἄσημοι, and others
φωναὶ σημαντικαί ; these latter being the nouns and verbs in language, which have a meaning of their own and stand for something even when uttered by themselves. The
σύνδεσμοι and
ἄρθρα however, are said to be
φωναὶ ἄσημοι; i. e. they convey no meaning when uttered by themselves.
If we may provisionally take them to represent in some way the little connexive words in language, the question arises as to the nature of the two kinds of connexives and the distinction
Aristotle wished to draw between them. It will be observed that he describes them by reference to their position and function in the complex
φωνὴ σημαντική, called a
λόγος a little further on in the chapter (
1457a 23
). It is necessary, therefore, to go back to his idea of
λόγος, and consider the various forms of
λόγοι which he either recognizes or may be presumed to have recognized.
λόγος in the present sense of the term has no exact English equivalent, though 'speech', 'discourse', or 'phrase' is sometimes conventionally made to stand for it. According to the definition in
1457a 23
it is a significant combination of significant words, e. g. of two (or more) nouns, or verbs (including what
Aristotle would call their
πτώσεις), or of a verb and a noun, or of a noun and a verb. Α
λόγος, in
Aristotle's sense of the term, does not necessarily involve predication.
The instance he gives in
1457a 25
of a non-predicative
λόγος is a definition, that of 'man', e.g.
ζῷον πεζὸν δίπουν - where the combination of the three nouns is equivalent to 'man', and may replace it as one of the terms of a proposition. This form of
λόγος is the
λόγος ὀνοματώδης of
An. post. 2. 10, 93 b 31
.
Under this same head we may include certain other forms of expression, e. g.
βαδίζει χαίρων (a combination of two verbs), and
οἰκοῦσι νῆσον (a verb+ a noun). A more complex instance of the same type would be a combination like
οἱ Ἀθηναῖοι καὶ Λακεδαιμόνιοι, where the terms are combined into one by the copulative, and
οἰκοῦσι περὶ δυσμάς, where the terms are hyphened together by the preposition
1
(comp.
Hartung,
Lehren der Alten über die Dichtkunst p. 284
). The other form of
λόγος is that which involves predication-the
λόγος ἀποφαντικός of the
De Interpretatione (4.17a seqq.).
If we may work out the hint in
De Interpretatione5, 17a 20
, we may divide the
λόγοι of this description into two kinds, the simple and the composite, and take the example in
1457a 28,
Κλέων βαδίζει
, as the type of the simple predicational
λόγος. It is not so easy to find an Aristotelian instance of the composite predicational
λόγος; but its nature is indicated in general terms in the passage in the
De Interpretatione: it consists of two or more simple predicational
λόγοι combined in some way or other into one single
λόγος.
How, then, and by means of what kinds of words is this combination effected? The
De Interpretatione ignores this question. We may perhaps supply the omission, however, by reflecting for a moment on the structure of complex sentences and observing the different relations of the parts or members to the whole. In some instances the simple
λόγοι ἀποφαντικοί are merely co-ordinated, and held together by copulative or similar conjunctions, as in
Κλέων βαδίζει καὶ Σωκράτης κάθηται, Κλέων μὲν βαδίζει Σωκράτης δὲ κάθηται, Κλέων βαδίζει ἀλλὰ Σωκράτης κάθηται (comp.
συνδέσμῳ εἷς in
1457a 30
); in others they are subordinated to one another, and form a sort of organic whole knit into one by words which indicate relation and interdependence. This last point in the theory of the
λόγος would have to be recognized in any grammatical analysis of the sentence and its construction. I venture to suggest that it may have been actually present in
Aristotle's mind at the time of framing his definitions of
σύνδεσμος and
ἄρθρον, and that it is to be found in the text, if it be permissible to restore it in some such way as this:—
σύνδεσμος δέ ἐστι φωνὴ ἄσημος, ἢ οὔτε κωλύει οὔτε ποιεῖ φωνὴν μίαν σημαντικὴν ἐκ πλειόνων φωνῶν πεφυκυῖαν συντίθεσθαι [καὶ ἐπὶ τῶν ἄκρων καὶ ἐπὶ τοῦ μέσου], ἣν μὴ ἁρμόττει ἐν ἀρχῇ λόγου τιθέναι καθ αὑτόν, οἷον μέν, δή, τοί, δέ. ἢ φωνὴ ἄσημος ἢ ἐκ 5 πλειόνων μὲν φωνῶν μιᾶς σημαντικῶν δὲ ποιεῖν πέφυκε μίαν σημαντικὴν φωνήν, (οἷον τὸ ἀμφί καὶ τὸ περί καὶ τὰ ἄλλα). ἄρθρον δ᾽ ἐστὶ φωνὴ ἄσημος, ἢ λόγου ἀρχὴν ἢ τέλος ἢ διορισμὸν δηλοῖ [οἷον τὸ ἀμφί καὶ τὸ περί καὶ τὰ ἄλλα. φωνὴ ἄσημος ἢ οὔτε κωλύει οὔτε ποιεῖ φωνὴν μίαν σημαντικὴν ἐκ πλειόνων φωνῶν], πεφυκυῖα τίθεσθαι 10 καὶ ἐπὶ τῶν ἄκρων καὶ ἐπὶ τοῦ μέσου.
The words bracketed after
συντίθεσθαι in l. 2 I suppose to be a repetition in the wrong place of those after
τίθεσθαι in l. 9; and the clause
ἢ φωνὴ...φωνῶν in ll. 8–9 to have come in through a repetition of the beginning of the definition of
σύνδεσμος in l. 1 (comp.
Düntzer,
Rettung d. aristotelischen Poetik p. 194
). It will be seen also that the illustration
οἷον τὸ ἀμφί τὰ ἄλλα is assumed to be out of place where it stands, and is transferred to l. 6, so as to form part of the second definition of
σύνδεσμος—a view which has been anticipated by
Hartung. Two kinds of
σύνδεσμοι, then, are recognized in the text as amended. The first is said to consist of the connexive words whose presence in a
λόγος does not affect the coherence of the terms. There is no difficulty in illustrating this point; if
Κλέων and
βαδίζει are combinable terms, the combination of them in the proposition
Κλέων βαδίζει is neither hindered nor helped by the insertion after the first word of a
μέν,
δέ, α
δή, οι
τοί. Little words of this description are what the grammarians called 'expletives'. Though they stand at the beginning of the
λόγος, they are not an integral part of it; they belong to it not
per se (
καθ' αὑτόν), but
per accidens, and only serve to indicate a relation to some other
λόγος before or after it. In contradistinction to these
Aristotle recognizes a second species of
σύνδεσμος (l. 4), one that is a real and constitutive part of the
λόγος into which it enters. This function, as I have endeavoured to show above, in the general survey of the possible forms of
λόγοι, is fulfilled by the prepositions and the copulative conjunctions.
Aristotle's
σύνδεσμος, therefore, whatever else it may signify, may be taken to include the preposition, the copulative conjunction, and also certain particles.
The sense of the word
ἄρθρον is more difficult to determine owing to the absence of examples to illustrate its meaning. The term must clearly denote a particular kind of connexive required for a particular kind of
λόγος. The kind of
λόγος
Aristotle had in mind may perhaps be inferred from the fact that the
ἄρθρον in it is said to indicate the beginning, or end, or dividing-point in its structure. Such a
λόγος may be assumed to be a composite proposition, and moreover one of a more complex form than that in
Κλέων βαδίζει καὶ Σωκράτης κάθηται and the like. The two constituent propositions in it are either intimately connected by being related in such a way that the one is the beginning and the other the end of the whole; or they are in some way divided or separated one from the other. With this we may compare the definition of the 'period' in
Rhet. 3. 9, 1409
a 35
λέγω δὲ περίοδον λέξιν ἔχουσαν ἀρχὴν καὶ τελευτὴν αὐτὴν καθ᾽ αὑτὴν καὶ μέγεθος εὐσύνοπτον
. As a conjectural interpretation of this statement it may be suggested (1) that the
ἄρθρα which mark the beginning of the proposition represent the conditional and causal conjunctions, together with the relative pronouns and adverbs; (2) that those marking the end are the final and illative conjunctions; and (3) those marking the separation or division in the proposition, the disjunctives. The kinds of proposition which involve the use of such words as these are ignored in
Aristotle's logical writings, but it seems to me that he may very well have had something to say of them in an analysis like that in the present chapter of the forms and constituents of ordinary speech.
I am not unaware of the lacunas that may be found in this interpretation of the Aristotelian
ἄρθρον. It must be taken for what it is worth, and with many allowances, as an attempt to recover the rough draft of an all but lost chapter of early grammar. For a discussion of the textual and other difficulties in these two sections I may refer to
Vahlen's
Beiträge, 3 p. 229 and 306
, and
Schömann's
Animadversiones ad veterum grammaticorum doctrinam de articulo (
Jahrb. f. class. Philol. Suppl.-Bd. V p. 5)
.
Both
σύνδεσμος and
ἄρθρον were terms taken by grammar from anatomy (see on
7, 1450
b 34
); the former is properly a 'ligament', and the latter a 'joint'. The language in the text as to the function of the
ἄρθρον or joint-word,
ἢ λόγου ἀρχὴν ἢ τέλος δηλοῖ, recalls that in the passage in
De anima 3. 10, 433
b 21
, which describes the physical joint as the means of locomotion in the animal:
τὸ δὲ κινοῦν ὀργανικῶς, ὅπου ἀρχὴ καὶ τελευτὴ τὸ αὐτό, οἷον ὁ γιγγλυμός· ἐνταῦθα γὰρ τὸ κυρτὸν καὶ κοῖλον τὸ μὲν τελευτὴ τὸ δ᾽ ἀρχή (διὸ τὸ μὲν ἠρεμεῖ τὸ δὲ kινεῖται), λόγῳ μὲν ἕτερα ὄντα μεγέθει δ᾽ ἀχώριστα
(comp.
J. of Phil. 17 p. 64
). The joint-word in grammar, therefore, would naturally imply a more structural and organic connexion than is to be found when the
λόγοι are simply strung together by
σύνδεσμοι.
I should add that the instances of
σύνδεσμοι in
Rhet. 3. 5, 1407
a 20
are
μέν and
δή, together with
γάρ,
τέ,
καί, and apparently
ἐπεί [?
ἔτι]; and that a distinction not unlike that in the text between the two kinds of
σύνδεσμοι is implied in
Probl. 16. 20, 919
a 22
ἐκ τῶν λόγων ἐνίων ἐξαιρεθέντων συνδέσμων οὐκ ἔστιν ὁ λόγος Ἑλληνικός, οἷον τὸ τέ καὶ τὸ καί, ἔνιοι δὲ οὐθὲν λυποῦσι διὰ τὸ τοῖς μὲν ἀναγκαῖον εἶναι χρῆσθαι πολλάκις, εἰ ἔσται λόγος, τοῖς δὲ μή
.
6.
mvp-aristotle-comment-730
1
κωλύει,
scil.
ποιεῖσθαι.
