No CrossRef data available.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 05 January 2009
‘Wir wollen ja mehr oder anderes als die Thatsachen classificieren; wir wollen sie erklären’ is the cry of a neglected linguist of the last century which it would not be idle to echo today, especially in the matter of syntactical oddities in the Greek and Latin authors. The immensely valuable work of those who have collected and classified instances of ‘aberrant’ syntax in the classical languages has the inherent disadvantage that lists and categories are made up according to the resultant form of the aberration rather than the underlying mental process which can itself often be analysed with profit. The result is a division of the phenomena into anacoluthon, contamination, pleonasm, ellipse, constructio ad sensum, and so on, a convenient but somewhat barren method, lessening the possibility of explaining and illuminating one example from another. Indeed to do thus cuts across the scientific method, putting together forms or uses which result from different thought- or speech-habits and separating many which owe their origin to identical or similar tendencies but have moved on somewhat divergent paths. Löfstedt, though he deals with the oddities of usage under conventional headings, pays no little regard to the psychological aspect and admits that a clear-cut division into conventional classes is not always free from considerable difficulty and is not particularly important for the scholar who would explain these things on a psychological basis.* Contamination in particular has been used rather to excess. The much quoted English oddity ‘I am friends with him’ is consistently stated to be a contamination of ‘I am friendly with him’ and ‘we are friendsür Völkerpsychologie und Sprachwissenschaft, 1860, 1.97.
page 1 note 2 Especially Löfstedt, , Syntactica, 2. 135 ff.Google Scholar; Glotta, 4. 253 ff.; Schmalz, , Glotta, 5. 209 ff.Google Scholar; Schwyzer, , Griech. Gramm. 2. 402 ff., 703 ff.Google Scholar
page 1 note 3 ‘Aberration’ is merely a convenient term, it being useful for recording purposes to assume the fiction that language is logical rather than psychological.
page 1 note 4 Syntactica, 2. 156, 167 n.
page 1 note 5 Paul, , Prinzipien der Sprachgeschichte5, 172, § 121. Cf. Lindsay, Syntax of Plautus, 2–3, for the same attitude to constructio ad sensum.Google Scholar
page 1 note 6 Even recently by Professor S. Potter, Our Language, 98.
page 2 note 1 Passio b. Georgii 15 (apud Schmalz, Glotta, 5. 209).
page 2 note 2 Where the importance of the person or thing referred to causes the speaker to suspend as unnecessarily restrictive the ‘rule’ that he must know at the outset the grammatical form it is to assume. It is not necessary to distinguish between nom. pend. (∊ἰπὠν δ' ἀκοὐσας ϑ' αἲματος μντρòς πἐρι φοĩβος μ' ἒσωσ∊ μαρτυρ
μν Eur. IT 964 f.) and nom. abs. (μἐλλων δἐ πἑμπ∊ιν μ' Oἰδíπου κλ∊νòς λóνος … ἐν τ
δ' ἐπ∊στρἀττ∊υσαν 'Aργ∊ĩοi πóλιν Eur. Ph. 283, 285) as the importance of the original nominative remains, whether subjected to a formal reprise or not. The ace. pend. is rarer. In all such cases the nominative or accusative precedes: otherwise, the process is one of mental equation or simplification (see below, p. 136, n. 6). Yet both types are commonly listed together (e.g. Schwyzer, Gr. Gr. 2. 403).
page 2 note 3 Cic. ad Q. fr. 2. 13. 2. Cf. Terence, Pho. 481, Ad. 648.
page 3 note 1 Contamination is the traditional term for mixing or blending of constructions; it is incorrect to ascribe any such meaning to contaminate, but useful to keep the modern word.
page 3 note 2 Thuc. 4. 18. 4; Plaut. Capt. 852 (cf. ibid. 1008, Cic. Phil. 5. 6, de dam. 1. 1; Livy, 2. 60. 5); Virgil, G. 4. 457; Eur. Hec. 225, and often.
page 3 note 3 Cognate is the use whereby a neuter relative pronoun is subsequently understood in a different case. Cf. Hor. Sat. 2. 6. 72–3; Lucr. 6. 45.
page 4 note 1 Cf. Thuc. 1. 136. 1, 3. 79. 3. Bogholm, English Speech from an Historical Point of View, 336, quotes from a letter to The Times Literary Supplement: ‘I read Shakespeare only for pleasure, and as if they were written only fifty years ago.’
page 4 note 2 Thuc. 2. 45. 2; Soph. Phil. 1126.
page 5 note 1 Hdt. 4. 132, Δαρ∊íου ἡ λνωμἠ ἠν … ∊ἰκἀℑων, is of this type and not to be listed with nom pend., or called simply ‘anacoluthon’ (Schwyzer, Gr. Gr. 2. 403, 705 with other similar exx.). For a complete equation of such a formula with a simple verb cf. rex Iuba non est visum dare, Bell. Afr. 25. 1. For a contrary case cf. Flav. Vopiscus, Vita Probi 22.1, ‘conferenti mihi cum aliis imperatoribus principem Probum … intellego …’, etc. At Thuc. 1. 62. 3 ἢν ἡ λνώμη = ἒδοξ∊.
page 5 note 2 The accusative of the neuter pronoun is expected with τυλἁνω, but for the rarer genitive cf. Eur. Hec. 42. (In Soph. Phil. 1315 ἐϕιἑμαι may well induce the genitive.)
page 6 note 1 This is even clearer at 108 ff., neque reintroducing the negative force temporarily at 111. Cf. Aesch. Ag. 366 ff., PI. Rep. 366 d 1–2.
page 6 note 2 ‘Analogy of opposites’ is too facile a formula: e.g. in Sanskrit only negative verbs of ‘ceasing’ use the present participle construction of asti, tiṣṭhati.
page 6 note 3 For a rare exception cf. Oscan ‘nep fefacid pod pis dat eizac egmad mins deiuaid dolud malud’, Tab. Bant, 10, as if ne faciat > prohibeat.
page 7 note 1 Thuc. 2. 40. 1, Pl Apol. 39 a 6.
page 7 note 2 A pure equation of two forms is visible at Pl Rep. 335 a 6 κ∊λ∊ὑ∊ις δἠ ἡμᾱς προσϑ∊ĩναι τ δικαí ἢς τὀ πρτον ἐλἑγομ∊ν, where προαϑ∊ῑναι τ δικαíῳ is taken to equal ἂλλο τι ποι∊ĩασϑαι υὀ δíκαιον, or μ∊ĩℑον ποιĩσαϑαι τò δíκαιον. The conjecture [ἢ] is not required.
page 7 note 3 Thuc. 2. 93. 3. Also Soph. Tr. 630f.; Xen. An. 6. 1. 28; Lys. 13. 51.
page 7 note 4 Pl. Phil. 13 a 4; cf. Xen. Cyrop. 2. 3. 6, and for a ‘delayed action’ example see Aesch. Pers. 115 ff. Cf. Eng. ‘It looks as if … ’ plus a future.
page 7 note 5 Plin. Ep. 6. 21. 1. Normally emended by editors. The qui may have occasioned ‘signal error’ (cf. Löfstedt, Synt. ii. 164, n. 2), with its form common to singular and plural.
page 7 note 6 To which should be added the partitive apposition types quoted by Schwyzer, Or. Or. 2. 403–4. A similar mental paraphrase causes what seems to be a hanging ace. participle following the main clause at Pl. Phaedr. 275 a 2 ff. (τν μαϑóντων λἠϑην μἐν ἐν ψυχαῑς παρέξ∊ι > τούς μαϑóντας ἒπιλἡαμονας ποιἡσ∊ι), and at Aesch. Pers. 913–14, Cho. 410—11 (in each case the main clause is equal to ϕóβος ἒχ∊ι μ∊). Cf. p. 137, n. 1. (At Soph. El. 479 f. the text is highly doubtat Xen. Lac. Pol. 5. 7 the correcion to ∊ἰδóτ∊ς is certain.)
page 7 note 7 v. Delbrück, Vergl. Syntax der IG-Sprachen, 3. 285 ff. Common in OE., but to be distinguished from explanatory partial apposition like eamus tu in ius (cf. Soph. Phil. 645).
page 8 note 1 Plaut. MG. 1253, Persa 70; Demosth. 19. 81. Cf. Aesch. Sept. 290, Supp. 566, Eur. 860, Ba. 1288.
page 8 note 2 Cf. Pl. Rep. 337 a 6. At Thuc. 7. 47. 2 the genitive abs. is taken as a causal clause, no doubt for variety. At Thuc. 4. 33. 2, χωρíων τ∊ χαλ∊πóτηι καì … τραχἐων ὂντων, a contrary process is visible, unless there is here a stylistic zeugma of two genitive uses in the noun.
page 8 note 3 Eur. Phoen. 1460 f. Very frequent with persons and abstract nouns; cf. Hdt. 9. 42; Thuc. 8. 66. 5; Pl. Rep. 345 e 7; Xen. Mem. 1. 2. 20.
page 9 note 1 This explanation, suggested by Kühner-Gerth, 2. 96, is rejected by Schwyzer, Gr. Gr. 2. 402 ff., who sees in the sentence ηὒχ∊το πρòς τούς ϑ∊ούς ἁπλς τἀλαϑἀ διδóναι, ώς τούς ϑ∊ούς κάλλιατα ∊ἰδóτας (Xen.Mem. 1. 3. 2) a contamination of two constructions after verbs of knowing: ώς τν ϑ∊ν ∊ἰδóτων X τοὐς ϑ∊οὐς ∊ἰδóτας and compares Soph. Phil. 253 ὡς μηδἐν ∊ἰδóτ' ἲσϑι μ' 
ν ἀνιατορ∊ĩς. But such a case, with the subject of the participle already present in the accusative, is not typical, and the equation is everywhere with a verb of thinking or hoping, rather than knowing. At Xen. Cyrop. 2. 4. 21, οἰ πολὲμιοι … προύκíνασαν τò στĩϕος, ὡς αυαομένους … (sc. τοὺς ἀμϕì Kũρν), the subject of the participle is already in mind, though unexpressed (see above,Content equation).
page 9 note 2 A term usually restricted in practice to changes of oblique cases.
page 9 note 3 As if paralleled by Virgil, Aen. 8. 526, Tyrrhenus tubae … clangor and similar cases.