121. The suffixes which express comparison—either between twο sets of objects (comparative) or between one and several others (superlative)—are partly primary, partly secondary. Hence it is convenient to treat them apart from the suffixes of which an account has been already given.
The comparative suffix -ιον is primary, the positive (where there is one) being a parallel formation from the same (verbal) root. The Homeric comparatives οf this class are
(for παχ-ίων, παχύ-ς)
(for κρετ-ίων, κρατ-ύ-ς)
- χείρων, χερε-ίων
- βραδίων (Hes.)
The stem is properly in the strong form, as in κρείσσων (but κρατός, κάρτ-ιστος); but it is assimilated to the positive in πάσσων, βράσσων, γλυκίων. In θᾱ́σσων, ἐλᾱ́σσων the ᾱ points to forms *θαγχ-ι̯ων, ἐλαγχ-ι̯ων, in which the nasal of the original *θεγχ-ι̯ων, έλεγχ-ι̯ων was retained, but the ε changed into ᾱ. The superlative -ιστο is used in the same way.
(ῥεῖα, fοr ῥήϊα)
Also, answering to comparatives given above
finally the anomalous πρώτ-ιστο-ς.
The suffix -ιον has taken the place of -ιοσ (§ 114*.7); the weakest form may be traced in -ισ-τος. The middle form -ιεσ perhaps appears in the two comparatives πλέες mοre (Il. 11.395, accusative πλέας Il. 2.129) and χέρεια worse (accusative singular and neuter plural, also dative singular χέρηϊ, nominative plural χέρηες). Original πλέες (for πλε-ι̯εσ-ες) became πλέες by hyphaeresis (§ 105.4) and so χέρεια is for χερε-ιεσ-α.2
The weakest form of -ιον would be -ἱν, which may be found in πρίν (cp. Latin pris-cus), and the Attic πλε-ῖν. Evidently πλεοσ-: πλεισ-: πλε-ῖν = prios: pris-: πρῑν.
Traces of a comparative suffix -ερο appear in ἔν-εροι those beneath (Latin inf-eru-s, sup-eru-s).
The suffix -τὸ or -ᾰτο is found in the ordinals τρί-τος, etc., and with the superlative meaning in
- πρῶτος (for πρό-ατο-ς)
also combined with ordinal suffixes in the Homeric τρί-τ-ατο-ς, ἑβδόμ-ατο-ς, ὀγδό-ατο-ς. The form -ᾰτο is probably due to the analogy of the ordinals τέτρα-το-ς, ἔνα-το-ς, δέκα-το-ς, in which the ᾰ is part of the stem.3
A suffix -μο may be recognised in πρόμο-ς foremost man (Latin infi-mu-s, sum-mu-s, pri-mu-s, ulti-mu-s, mini-mus).
The common suffixes -τερο, -τᾰτο appear with a verb stem in
(cp. ἐ-φίλα-το ἰουed)
(δεύ-ω to fail, to come shοrt of4
So φαάν-τατος, for φαέν-τατος (φαείνω). Otherwise they are used with nominal stems.
Final ο of the stem becomes ω when a long syllable is needed to give dactylic rhythm, as κακώ-τερο-ς, κακοξεινώ-τερο-ς.6
In ἀνιηρέσ-τερος (Od. 2.190) the stem follows the analogy of θυμ-ῆρες, etc. In χαριέσ-τερος (for χαριϝᾰτ-τερος) there is the same assimilation as in the dative plural χαρίεσσι (§ 106.3). In μυχοί-τατο-ς innermοst the stem appears to be a locative case form; cp. παροί-τεροι more forward, and later forms like κατώ-τερο-ς, ἀνώ-τατο-ς, etc.; so probably in παλαί-τερος and ὑπέρ-τερος. On the analogy of ὑπέρ-τερος we can explain ἐνέρ-τερος (cp. ὕπερ-θε: ἔνερ-θε, etc.). The form γεραί-τερος, again, may be suggested by παλαίτερος, through the relation γεραιός: παλαιός and the likeness of meaning (Meyer, G. G. p. 372). The words δεξι-τερός, ἀριστερός are formed like comparatives, but are distinguished by their accent.
The suffix -τερο is combined with the suffix -ιον in ἀσσο-τέρω (adverb) nearer, ἐπ-ασσύτεροι drawing on, χειρό-τερο-ς and χερειό-τερο-ς worse.
-τερο, -τᾰτο are combinations of -το (in τρί-τος, etc.) with the suffixes -ερο and -ᾰτο respectively. The tendency to accumulate suffixes of comparison is seen in
χειρό-τερο-ς and χερειό-τερος
Latin -issimu-s (for -is-ti-mu-s), mag-is-ter, min-is-ter.
122. Comparative and Superlative Meaning. The stem is often that of a substantive
more like a dog
so that the adjectival character is given by the suffix.
The meaning is often, not that an object has more of a quality than some other object or set of objects, but that it has the quality in cοntradistinction to objects which are without it. Thus in πρό-τερο-ς the meaning is not more fοrward, but forward, opposed to ὕσ-τερο-ς behind. So ὑπέρ-τερο-ς and ἐνέρ-τερο-ς, δεξι-τερό-ς, and ἀρισ-τερό-ς, δεύ-τερο-ς, etc. The same thing appears in the pronouns ἡμέ-τερο-ς, ὑμέ-τερο-ς, ἕ-τερο-ς, πό-τερο-ς, ἑκά-τερο-ς, ἀμφότερο-ς, etc. ἡμέ-τερο-ς is not more belonging to us, but belonging to us (nοt yοu). Sο in the Homeric comparatives
of the cοuntry (as opposed to the town)
of the mountains (as opposed to the valley)
opposed to καταιβαταὶ ἀνθρώποισιν (Od. 13.111)
female (opp. to male)
the class of youths
Cp. Il. 19.63 Τρωσὶ τὸ κέρδιον that is a gain to the Trojans (rather than tο us). Hence the comparativeis sometimes used as a softened way of expressing the nοtiοn of the positive.
Il. 19.56 ἄρειον
"gοοd rather than ill"
Il. 1.32 σαώτερος
safe (as we speak of being "on the safe side")
So θᾶσσον with an imperative. Hence too the idiοmatic use of the double comparative.
Od. Il. 164 ἐλαφρότεροι πόδας εἶναι ἦ ἀφνειότεροι
to be light of foot rather than wealthy
- 1Better written ὑπολεί-ζονες.
- 2So G. Mahlow and J. Schmidt, K. Z. xxvi. 381. A different analysis is given by Collitz in Bezz. Beitr. ix. 66 and Brugmann (Grundr. ii. § 135, Il. 402), who explain πλέες as plē-is-es, i. e. from the weakest form of the stem. This view does not apply so well to χέρει-α, since it leaves unexplained the divergence between it and the superlative χείρισ-τος. It may be noticed as an argument for the supposition of hyphaeresis that we do not find the genitive πλέος, χέρειος, just as we do not find hyphaeresis in the genitive of nouns in -εος, -εηs (§ 105.4) Cp. however, the absence of trace of a genitive ἀμείνο-ος (§ 114.7, foοtnοte).
- 3Ascoli in Curt. Stud. ix. p. 339 ff.
- 4This very probable etymology is given by Brugmann, K. Z. xxv. p. 298.
- 5For ἄ-τερος, ἀ- οne, with assimilation to ἑν-.
- 6According to Brugmann the ω of σοφώτερος, etc., is not a metrical lengthening, but comes from the adverbs *σοφῶ, etc. (related to σοφῶς as οὕτω to οὕτως, § 110), like the later κατώ-τερος from κάτω, etc.
Note— The ω of σοφώτερος, etc., has lately been discussed by J. Wackernagel (Das Dehnungsgesetz der grinch. Composite, pp. 5 ff.) He treats it along with the ω which we find in ἑτέρωθι, ἑτέρωσε, ἀμφοτέρωθεν, etc., also in ἱερωσύνη, and shows that if we derive it from a case form in -ω (as κατωτέρω from κάτω, etc.) we have still to explain the rhythmical law according to which ω and ο interchange; for a law which governed common speech in all periods cannot have arisen merely from the needs of the hexameter. Accordingly he connects the phenomenon with a rhythmical lengthening of final short vowels (among others of the final ι of the locative see § 378), which is found in Vedic Sanskrit.
λαρώτατος (Od. 2.350) points to a Homeric form λαερός, which we can always substitute for λαρός. It is probably for λασ-ερός from λασ- desire, see Curtius, Grundz, p. 361 (5th ed.).