Notice of ' Θησαυρὸς τῆς Ἑλληνικῆς Γλώσσης : Thesaurus Græcæ Linguæ; Ab H. STEPHANO constructus'...... 150 &c..... 156 THE CLASSICAL JOURNAL; N°. LXXVII. MARCH, 1829. ON COMPOUND WORDS IN THE THE ancient languages have a great advantage over modern languages by the facility with which they form their compounds. The Greek, especially, is inexhaustible in compounds; the Latin is not half so rich, and many of the Latin compounds are only found in the old poets. Even the German language possesses a far greater store of compounds than the Latin. The modern Greek still uses a number of them in familiar conversation. On examining the mode in which compounds are formed, we find that two originally independent words are united by means of a vowel, or sometimes without it, or by joining the two words in such a manner that the flexion of the first word can be preserved. Compounds are beautiful when two ideas are comprehended in one image, but much less beautiful if one idea is expressed by two words. Thus the Latin malus, vinea, lotrix, mulctra, and many others, are preferable to the English apple-tree, vineyard, washerwoman, milk-pail. The vowel by which compounds are formed is always a short vowel. (A.) in the Latin i: this is not the i of the casus, because we find it after all nouns of every declension and gender, even after such, which occur only in the plural. Instances of the first are: stilli-cidium, sylvi-cola, terri-cola, virgi-demia, (as vin-demia, a demendo,) causi-dicus, aquili-fer, stelli-fer, bacci-fer, &c. lani-ficus, aqui-folium, terri-gena, ali-ger, barbi-ger, lani-ger, &c. nugi-gerulus, herbi-gradus, aqui-legium, spici-legium, capri-mulgus, equi-mulgus, sagitti-potens. Instances of the second declension are: ligni-cida, galli-cinium, auri-fer, anni-fer, belli-fer, caduci-fer, &c. ; lucri- ficus, dei- ficus, (as dei-loquus, dei-para,) veli-ficus, auri-fodina, argenti-fodina, auVOL. XXXIX. CI. JI. NO. LXXVII. A 1i-fur, cocli-gena, armi-ger, belli-ger, armi-lustrium, hirci-pes, sonipes, lucri-peta, hirci-pilus, armi-potens, belli-potens, vini-potor, fœni-secium, lecti-sternium. Instances of the third declension are: muni-ceps, muri-ceps, parti-ceps, homi-cida, infanti-cida, lapi-cida, and lapidi-cida, (from the old lapis, gen. lapis, lapicida,) matri-cida, parri-cida (pro patricida), regi-cida, muri-cidus (Plautus, a coward, a rat-catcher), marti-cola, monti-cola, rupi-cola, ruri-cola, pietati-cultrix, juridicus, flori-fer, frondi-fer, frugi-fer, imbri-fer, &c. morti-ferus, arti-fex, carni-fex, &c.: honori-ficus, muni-ficus, cruci-fixus, melli-fluus, rori-fluus, æri-fodina, ossi-fragus, aquiloni-gena, serpenti-gena, soli-gena, marti-genus, crini-ger, flori-legium, hæredipeta, vesti-plicus, auri-scalpium, vesti-spicus, sangui-suga, su-ovitaurilia, monti-vagus, nocti-vagus, igni-vomus, carni-vorus. There are few instances of the fourth declension: algi-ficus, arci-potens, corni-cen, corni-ger, corni-pes, domi-duca, domi-porta (snail), flucti-cola, flucti-gena, fructi-fer, geli-cidium. Of the fifth declension I know no example, except some words of barbarous Latinity, as speci-ficus from species; perhaps fideicommissum, rei-vindicatio, may also be quoted. In compounds with adjectives the adjective may again belong to the first three declensions. Examples of the first and second declension: parvi-collis, unicaulis, uni-cornis, uni-jugis, miseri-cors, vani-dicus, veri-dicus, soli-ferreus, magni-ficus, largi-fluus, alieni-gena, tardi-gradus, magni-loquus, multi-loquus, versuti-loquus, vivi-para, fissi-pes, longi-pes, solidi-pes, tardi-pes, multi-scius, alti-sonus, rauci-sonus, soli-vagus, alti-volans. Example of the third declension: omni-gena, omni-medens, brevi-loquus, blandi-loquus, dulci-loquus, gravi-loquus, suaviloquus, gracili-pes, levi-pes, levi-somnus, &c. Sometimes the vowel which unites the two words together is wanting, if the second word begins with a vowel: celt-iberi, pusillanimis, sun-ambulus, somn-ambulus, noct-ambulo; sometimes also before consonants: man-ceps, man-cipium p. mani-ceps, (cf. municeps,) man-suetus p. mani-suetus, Gr. Xepo-ions, vin-demia p. vinidemia, (cf. virgi-demia,) sol-stitium p. soli-stitium, (cf. armi-stitium,) puer-pera p. pueri-pera, as pueri-cida, perhaps also lus-cinia; or is the derivation from lux and cano false, and is luscinia from luscus, luscinus, therefore luscinia, the blinded singing-bird? Avi sinks into au: nau-fragium p. navi-fragium, au-ceps, aucupium, p. aviceps, avi-cupium, au-gur p. avigur. The letters of derivation remain in the first word, as the examples of the third declension show; infant, infant-i-cida: hence it is clear that infans, pons, pietas, stand for infant-s, pont-s, pietat-s, as in the Greek ἔρωτος, ἔρωτς, ἔρως. Lapidi-cida, lapi-cida, prove a double form lapid-s, and lap-is: unless lapicida is a contraction. We must notice the falling-off of the in in the words homicidium, sangui-suga, not homini-cidium, sanguini-suga. There is no Latin compound, where the in is preserved; for the word semini-verbius is of latter formation; it enters, however, partially into derivations, as graminosus, seminosus, ominosus, &c. The i and u of the derivation gives way to the vowel of composition, as the examples of the fourth declension show; scarcely any compounds are formed with those that end in ia and ius: the long vowel of tibicen is formed by tibi-i-cen; but why is the i short in medi-dies, medi-terraneus, medi-tullium? The u sometimes absorbs the vowel of composition; thus we find arcu-potens, cornu-peta: manu-factus, manu-missus, manuscriptus, are improper compounds by means of the ablative case of the noun: manu-pretium stands for manus-pretium, and domuitio probably for domum-itio. Has the qu in hirqui-tallus, sterqui-linium, where we have also hirci-tallus, a u of derivation? This qui is not found in compounds of acus, arcus, pecus, specus, lacus, quercus, &c. The first word is sometimes obscure, e. g. aru-spex, or haruspex, cf. au-spex p. avi-spex, meri-dies from merus ? or p. medi-? Sometimes the second word, as in mani-festus. The composition produces new adjectives from substantives without the vowel of derivation; as levi-somnus, magn-animus, hirci-pilus, longi-pes, miseri-cors, and with the vowel of derivation in pusill-animis, parvi-collis, (cf. im-berbis, im-bellis, de-pilis, &c.) Can compounds of verbs be formed? Generally they can only be formed from compound nouns; consequently tergi-versor presupposes a tergi-versus: the composition with the participle, as altivolans, omni-medens, does not seem to establish a rule for the rest of the verb. Those which end in -fico (they are very numerous), ædi-fico, ludi-fico, lucri-fico, melli-fico, nidi-fico, turpi-fico, veli-fico, &c. are grounded on words with -fex; but those in -fero, -feror, as voci-fero, pesti-fero, (adject. in -fer,) magni-facio, flocci-facio, flocci-pendo, parvi-duco, parvi-facio, parvi-pendo, do not seem to be derived from nouns: they might be improper compounds with the genit. flocci, magni, parvi, if it were not for vili-pendo, which can hardly come from the gen. vilis. Compounds of verbs with verbs are all in -facio and -fio: calefacio, frige-facio, labe-facio, ol-facio p. ole-facio, pate-facio, putre-facio, nigre-facio, terre-facio, torre-facio, tume-facio, &c. where the e is not the vowel of composition, but the e of the second conjugation, although no such form as labeo, nigreo, exists. It may have existed. Cale-facio is not a compound of calere-facio, as it might appear, for the vowel e in cale-facio is short; and we have also the compounds cale-fio, tepe-fio, and where we cannot think of an infinitive. Why do verbs of the first, third, and fourth conjugations not form such compounds? Those in -eo are all intransitive. (B.) The Greek vowel for forming compounds of nouns of every gender and of every declension is o. There is an immense number of such compounds, and we shall only give a few examples. From the first declension: αραχνοϋφής, γεο-ειδής, usually γεωειδής, ἡμερο-φύλαξ, κορο-κόσμιον, μελισσοτρόφος, νικο-ποιός, νυμφο γενής, υλο-δρόμος. From the second declension: ἀρτο-θήκη, γερανο-μαχία, ἱππο-νόμος, μυλοκόπος, οἰκο-δόμος, ὀνο-ειθής, σκοτο-μήδης, ῥοδο-δάκτυλος, στρατο-λόγος, ταυρο-βόλος, τοπο-γράφος, τοξο-βόλος, ὑπνο-φανής, χρυσο-φανής. From the third declension : μελιτοειδής, μυομαχία, νυκτο-ειδής, ὀρνιθο-λόχος, πιτυο-κάμπτης, πνευματοποιός, ποδο-στράβη, ῥινο-βόλος, φοινικο-τρόφος, χειρο-μαντεία, χιονο-βόλος. And with adjectives of the first and second: αὐτο-κύριος, γυμνοπόδης, ἰσο-βαθής, καλο-διδάσκαλος, μεσο-φανής, ὁμοιοπαθής :-of the third : ἀρσενο-γενής, μελανο-κόμης, τερενο-πλόκαμος.-(With regard to the Greek declensions, we refer to Matthiæ's Gr. Grammar.) If sometimes an 7 and a appear instead of the o of composition, it appears to be derived in the first declension from the flexion. For instance: δαμαλη-φάγος, δαφνή-φαγος, with δαφνο-γηθής, δαφνο-φόρος (Lob. ad Phryn. 634638.), νυμφα-γενής, κορυφα-γενής, μιτρα-φόρος, with μιτρο-φόρος, μιτρο-χίτων, νυμφο-γενής, νυμφό-κλαυτος (Lob. p. 641.). But then also substantives of the first and second declension and adjectives have the η, θαλαμηπόλος, ομφα λη-τόμος, (Lob. p. 650 seq.) άνθη-φόρος, ἀχθη-φόρος, στεφη-πλόκος, βοτρυη-φόρος, βοη-νόμος (Lob. 679, 680.). How can we account for it? Are the dialects wavering between o, a, n, or is it poetical licence? The accent very often falls on the vowel of composition, especially if the second word stands in a passive signification, and where the relation of the first word to it can be expressed by a preposition; e. g. ἱππόνομος, also ἱππο-νόμος, λιθοβόλος, and λιθό-βολος, throwing stones, and thrown with stones. Cf. Buttman Gramm. § 106, 9, 3. In other cases no such distinction can be made ; e. g. ροδό-κολπος and ῥοδοδάκτυλος, μελανό-κομος and μελανο-κόμης, the different accentuation does not change the meaning of the words. The vowel of composition is wanting, 1. when the second word begins with a vowel, τοπ-άρχης, κυναλώπης, ποδώκης; but it appears sometimes before et and e, ἀγαθο-εργός, καλο-εργός, μονοειδής, κυνοειδής ; by contraction η and ov: 2. sometimes after v, μελαγ-χρής γ. μελανο-χρής, μελάμ-φυλλος p. μελανό-φυλλος : 3. as in the Latin ναυ-βάτης, ναύκληρος, and others. The letters of derivation remain, and especially the vowels and v in derivations with voc. pur.; e. g. πολιο-ειδής, εὐρυο-δίνης, ιχθυονόμος, ιχθυο-φάγος. The vowel of composition is oftener lost, as ιχθυ-φάγος, εὐρυ-δίνης, γλυκυ-δερκής, δορύ-μαχος, πλατύφυλλος, ταχυ-πόρος, μελι-ηδής, πτολιπόρθος, νυκτι-φανής. Remarkable also is |