§ 1. I. THE CONJUGATIONS. The I.-Eur. Verb had two Conjugations, (1) the Thematic, in which the Person-suffixes were attached to the verb-root augmented by -ĕ- or -ŏ-, e. g. Gk. φέρο-μεν, φέρ-ε-τε; (2) the Athematic, in which this vowel, the Thematic Vowel, as it is called, was absent, e. g. Gk. ἴστӑ-μεν, ἴστӑ-τε. In the Thematic Conjugation the 1 Sg. Pres. Ind. Act had -ō, e. g. φέρ-ω, Lat. leg-ō; the Subjunctive changed the Thematic Vowel of the Indicative to -ē- (and -ō-), e.g. Gk. φέρ-η-τε (φέρ-ω-μεν); the Optative changed it to -oi-, e. g. Gk. φέρ-οι-μεν, φέρ-οι-τε. In the Athematic Conjugation the 1 Sg. Pres. Ind. Act. had -mĭ, e. g. Gk. ἴστη-μι the verb-stem was weakened in the Dual and Plural Act. and in all Numbers of the Middle, e.g. Gk. ἴστӑ-τον, ἴστӑ-μεν, ἴστӑ-μαι, beside Sg. ἱστη-; the Subjunctive (with strong stem) showed -ĕ- or -ŏ- between the root and the Person-suffixes, and the Optative (with weak stem) -yē- in Sg. Act., -ī- elsewhere, e.g. Gk. ἱστӑ-ίη-ν, ἱστӑ-ι-μεν (ἱσταȋ μεν.

We find early Derivative Verbs like I.-Eur. tr-ā- from the root ter-, pl-ē- from the root pel- (Lat. in-trā-re, im-plē-re), and later Derivatives from Nouns, e. g. Lat. curā-re from the Noun-stem curā-, forming the persons of their Present Tense sometimes thematically with the suffix -yŏ- (§ 15), sometimes athematically, e. g. O. Ind. trā-yӑ-tē and trā-ti 3 Sg.

§ 1. ADVERBS. Latin Adverbs are for the most part cases of Nouns, Adjectives (or Participles), and Pronouns, the cases most frequently found being the Accusative (cf. O. Ind. náktam, ‘by night,’ satyám, ‘truly,’ Acc. Sg. Neut., pratarām Acc. Sg. Fem., Gk. πρότερον Acc. Sg. Neut., μακράν Acc. Sg. Fem., κρύφα Acc. Pl. Neut.), Ablative (cf. O. Ind. dūrāt, ‘afar’) [also Instrumental (?), cf. O. Ind. dívā, ‘by day,’ Gk. ἄλλη], and Locative (cf. O. Ind. dūrē, ‘afar,’ Gk. οἴκοι, ἀναιμωτεί, έκοντί), and often retain case-forms which have become obsolete in the ordinary declension. Thus, -im, the original form of the Acc. Sg. suffix of I-stems (ch. vi. § 29), which in classical Latin was replaced by the -em of Consonant-stems, e. g. partem from the stem parti-, like militem from the stem milit-, is retained in Adverbs like partim, and in the Adverbial Accusatives of Verbal Noun I-stems, e. g. raptim from the stem rapti- (class, raptiōn-, ch. v. § 42), sensim from the stem sensi- (class, sensiōn,-), uni-versim (Osc. úíníveresím) &c.; nox, an old equivalent of noctu, seems to be an early byform of noctis Gen.; -ē (older -ēd), the suffix by which Adverbs derived from Adjective O-stems are formed in the Italic languages, is an Abl. Sg. suffix (parallel with -ō, older -ōd, ch. vi. § 33), which has been reserved for Adverbs alone.

§ 1. Nature of the Latin Accent. Was the Latin accent one of pitch or stress? Did the accented syllable in a Latin word differ from the other syllables in being uttered at a higher note than they were, or with a greater force? The two things are obviously quite distinct. For a syllable to be sounded at a high or low note is one thing, with energy or with gentleness is another, just as a musical note may be sounded strongly or gently (forte or piano), a thing quite different from its being a note high or low on the musical scale. Most languages do indeed combine in a greater or less degree pitch-accent with stress-accent. The accented syllable, if pronounced with more energy than the unaccented, is generally at the same time pronounced at a slightly higher (or lower) pitch. But, for all that, it is usually possible to say decidedly of one language: this language has a stress-accent; of another: this language has a pitch-accent. Our own language for example is clearly a language of stress-accent. It distinguishes its accented syllables by giving them greater energy of articulation than the unaccented; and it shows the usual characteristics of a language with stress-accentuation, namely, a slurring or Syncope of short syllables immediately following the accented syllable (e.g. ‘méd(i)cine’; cf. dám(o)sel,’ ‘fánt(a)sy’ and ‘fancy,’) and an obscuring or reduction of unaccented vowels (e.g. ‘father,’ where the e has the sound of the ‘obscure’ vowel of the word ‘but’; ‘savage,’ ‘minute,’ ‘orange’).

§ 1. I. DECLENSION OF NOUNS AND ADJECTIVES. The I.-Eur. Noun had three Genders, Masc., Fem., and Neut., three Numbers, Sing., Dual, and Plur., and at least eight Cases, Nom., Gen., Dat., Acc., Voc. (if the Voc. may be called a Case), Abl., Instrumental, and Locative (the Dat., Abl., Instr., Loc. are not always easy to discriminate in the Plural). The three Genders are retained in Latin, but of the Numbers the Dual has disappeared, though traces of it remain in the Numeral forms duō (§ 59), octō (lit. ‘two sets of four’?), and in the Pronoun form ambō (ch. vii. § 29). (On vī-gintī, see § 74.) Of the Cases the Voc. hardly survives except in O-stems (§ 31), and the Instr. has left only doubtful traces of itself in some Adverb forms (§ 36); the Locative became by the operation of the phonetic laws of the language indistinguishable in Ā-stems from the Gen. Sg., while in O-stems it seems to have ousted the Genitive (§ 17), and in Cons.-stems the Ablative (§ 33).

The I.-Eur. Cases were indicated sometimes by the addition of suffixes, e.g. -s for Nom. Sg. Masc., -m for Acc. Sg., sometimes by modification of the stem, e. g. *pӑtēr Nom. Sg. of stem *pӑtĕr-, ‘a father,’ sometimes by both, e. g. *patr-ŏs, *patr-ōm Gen. Sg., Pl. of the same stem.

If an alphabet is to express the sounds of a language properly, each nation must construct one for itself. But this ideal was not realized by the ancient languages of Italy. The Oscan and Umbrian stocks borrowed for the expression of their language the alphabet used by the Etruscans, who had themselves borrowed it at an earlier period from the Greeks; and so neither Oscans nor Umbrians were at first able to express in writing some common sounds of their language, such as d and o, which were wanting in the Etruscan speech (von Planta, Osk.-Umbr. Dial. i., p. 44). The Latin Alphabet, consisting in the later Republic of twenty-one letters, ABCDEFGHIKLMNOPQRSTVX, was borrowed from some Chalcidian colony (e.g. Cumae), to judge from the form of the letters, which more nearly resemble those of the Chalcidian inscriptions than of any other Greek stock. So few Latin inscriptions earlier than the second Punic War have been preserved, that it is difficult to trace each separate stage in the process of adapting the Greek alphabet to the exigencies of the Latin language.

§ 1. A. In the words ‘man,’ ‘father,’ the vowels which we are in the habit of classing roughly as ‘short a’ and ‘long a’ are really very different from each other, and would be phonetically expressed by two distinct symbols. In Sweet's Handbook of Phonetics, while the second is written a, the first is denoted by a combination of the letters a and e, viz. æ, a symbol which implies that the vowel has something of the nature of an E-sound. If we compare our pronunciation of the words ‘man,’ ‘hat,’ with the German of ‘Mann,’ ‘er hat,’ we see that the German vowel is the same as the a of English ‘father’ or German ‘Vater,’ while we might say that our ‘man,’ ‘hat,’ ‘bat,’ have in them something of the sound of ‘men,’ ‘bet,’ Seelmamx, who classifies the varieties of A as ‘normal a,’ ‘a inclined to an E-sound,’ and ‘a inclined to an O-sound’ (this last being something not quite so definitely an O-sound as the vowel of our words ‘all,’ ‘awe’), is of opinion that the Latin a had a leaning to e rather than to o, and goes so far as to give to Latin a of the Imperial age the œ-sound of English ‘man.’

§ 1. CONJUNCTIONS. As Prepositions are hardly separable from Adverbs of Locality, so Conjunctions are closely connected with pronominal Adverbs. These pronominal Adverbs, as we have seen (ch. ix. § 10), are not always capable of being referred to their proper case-form (e. g. ȋbī, ŭbī), owing to our imperfect knowledge of the declension of the I.-Eur. pronoun. Nor is it easy to find their cognates in the various I.-Eur. languages; so rapidly does the meaning of a Conjunction alter. Thus Latin ĕnim, which in the older literature is a particle of asseveration, ‘indeed,’ had by the classical period appropriated the sense of ‘for’; and in French, pas (Lat. passus) and point (Lat. punctum) have acquired a negative sense from their use in the phrases ne … pas, ne … point. A feature of I.-Eur. Conjunctions is their tendency to append other Conjunctions or conjunctive Particles (e. g. ὡς in Greek may append δή, περ, &c., ὡς δή, ὣς περ); and this habit puts another obstacle in the way of identifying cognate Conjunctions in different languages, for in one language they may appear extended by one particle, in another language by another. The exact form of these conjunctive Particles is also a difficult thing to ascertain; we often see parallel stems in -o, -i, -u, &c.

Since Corssen's great work (last edition, Leipzig, 1868–70), there has been no book devoted to a separate investigation by Comparative Philological methods of the Latin Language, its declensions, its conjugations, its formation of the various parts of speech, and the changes of its pronunciation and orthography, if we except the short summary (last edition, Nördlingen, 1889) written by Professor Stolz for the Iwan Müller Series of Handbooks of Classical Antiquity. And yet the additions to our knowledge of the subject since Corssen's time have been very great. Not only has the whole Science of Comparative Philology been, by the help of men like Johannes Schmidt, Osthoff, and Brugmann, set on a sounder basis, but a vast amount has been added to our knowledge of the Early Latin authors, especially Plautus, of the Umbrian, Oscan, and other dialects of ancient Italy, of Romance, and above all of the Celtic family of languages, a family closely united with the Italic group. The time has surely come for a new treatment of the subject, such as I venture to offer in the ten chapters of this volume.

I should have liked to have added to them a fuller discussion of the relation of Latin to the other languages of Italy. But I had already exceeded the generous limits allowed by the Delegates of the Press, and it seemed to me that until more evidence is forthcoming in the shape of dialectal inscriptions certainty can hardly be attained.

The legend in Homer.

§ 1. THE story of Orestes the avenger was complete in every essential particular before it came to the earliest of those three Attic dramatists, each of whom has stamped it so strongly with the impress of his own mind.

In the Iliad there is no hint that the house of Pelops lay under a curse which entailed a series of crimes. The sceptre made by Hephaestus for Zeus, and brought by Hermes to Pelops, is peacefully inherited by Atreus, Thyestes and Agamemnon. Yet the Iliad makes at least one contribution to the material which Aeschylus found ready to his hand. It is the figure of Agamemnon himself, with eyes and head like those of Zeus, in girth like Ares, in breast like Poseidon; ‘clad in flashing bronze, all glorious, and pre-eminent amid all.’ As Helen stands with Priam on the walls of Troy, and watches the Achaean warriors moving on the battle-field, she asks who this one may be:–‘There are others even taller by a head, but never did I behold a man so comely or so majestic (γεραρόν); he is like unto one that is a king.’ This is the royal Agamemnon, ὁ παντόσεμνος, who lives in the Aeschylean drama, and whose image reappears in later poetry.