MASTER NEGA TIVE NO. 91-80211 MICROFILMED 1 992 COLUMBIA UNIVERSITY LIBRARIES/NEW YORK as part of the . „ "Foundations of Western Civilization Preservation Project Funded by the .„.^„ NATIONAL ENDOWMENT FOR THE HUMANITIES Reproductions may not be made without permission from Columbia University Library COPYRIGHT STATEMENT The copyright law of the United States - Title 17, United States Code ~ concerns the making of photocopies or other reproductions of copyrighted material... Columbia University Library reserves the right to refuse to accept a copy order if, in its judgement, fulfillment of the order would mvolve violation of the copyright law. AUTHOR: WEISE, OSKAR TITLE: LANGUAGE AND CHARACTER OF THE PLACE: LONDON DA TE : 1909 Restrictions on Use: COLUMBIA UNIVERSITY LIBRARIES PRESERVATION DEPARTMENT Master Negative # BIBLIOGRAPHIC MICROFORM TARGET Original Material as Filmed - Existing Bibliographic Record 877 V1P77 W432 Weise, Oskar, 1851-1933. Language and character of the Roman people, tr. from tlie German of Oscar Weise, with additional notes and references for English readers, by H. A. Strong ... and A. Y. Campbell ... London, K. Paul, Trench, Triibner & CO., limited, 1909. ' CffI X. 259p. 22i' Copy in Classics, 1909 • 1. l^atin lanj^age. 2. L^tin literature— Hist. & crit 3. IJational charac- teristics, Roman. I. Strong, Herbert Augustus, tr. ii. Campbell, Archi- bald Young, joint tr. Library of Congress t^->-{, -m9^^^PA20S7.W4 10-11035 TECHNICAL MICROFORM DATA FILM SIZE: ZS/Y) >n REDUCTION RATIO: Ij IMAGE PLACEMENT: lA Ql^ IB IIB DATE FILMED CtlJJjL^J INITIALS J2^j_^_. HLMEDBY: RESEARCH PUBLICATIONS. INC WOODBRIDGE. CT D Association for Information and Image Management 1100 Wayne Avenue, Suite 1100 Silver Spring, Maryland 20910 301/587-8202 / * Centimeter 12 3 4 mi|imlimjim]n I Inches 12 13 14 iiiill liiiililii III 15 mm III! 1.0 I.I 1.25 145 ■ 5.0 ■ 3.6 2.8 1^ I 7,1 ■ eo 1.4 2.5 2.2 2.0 1.8 1.6 MRNUFfiCTURED TO, RUM STflNDflRDS BY fiPPLIED IMRGE. INC. V T f lARLES KNAPP LIBRARY 1937 Columbia (Bnitif rsiftp intl)pCttpofJlrttigork LIBRARY It * A LANGUAGE AND CHARACTER OF THE ROMAN PEOPLE Language and Character of the Roman People • « ■ TRANSLATED FROM THE GERMAN OF ; ' •. . '^ iOSCAR WEISE < • • • • « • « « ( I •_ WITH ADDITIONAL NOTES AND" REFERENCES FOR ENGLISH READERS > * * < < » BY H. A. STRONG, M.A., LL.D. PROFESSOR OF LATIN, LIVERPOOL UNIVERSITY AND A. Y. CAMPBELL, B.A. ASSISTANT LECTURER IN CLASSICS, LIVERPOOL UNIVERSITY LONDON KEGAN PAUL, TRENCH, TRUBNER & CO. LIMITED Drvden House, Gerrard Street, W, 1909 % t * * • • CHISWICK press: CHARLES WHITTINGHAM AND CO. TOOKS COURT, CHANCERY LANE, LONDON. FROM THE AUTHOR^S PREFACE TO THE FIRST EDITION THE knowledge of any language must neces- sarily remain superficial, unless the student of the language in question has a clear conception of the various forms which make up its construction. The ordinary grammars give us little light on this point. School text-books regard such information as beside the mark, and, unfortunately, scientific works are content with a few scanty precepts. It is, however, to be regretted that our methods of teach- ing language should alone lag in the wake of other studies, and refuse to follow the spirit of the nine- teenth century, probing and noting every fact and tracing them in their historical development. It passes comprehension why teachers cannot dispense with the routine methods of exercising their pupils' memory at the expense of their intelligence. They might surely choose some way of stimulating the thought and reflection of their pupils. This small treatise may serve, it is hoped, as a stepping-stone to this end. f> '1 n 'In*" *r I FROM THE PREFACE TO THE SECOND EDITION THE Second Edition of this work might, strictly speaking, be called the third. For the French work, based upon my own, by Ferdinand Antoine, Professor of Classic Philology in the University of Toulouse (1896), contains a large number of im- provements and additions, which, at his request, I placed at his service. The new edition differs in many respects from Antoine's translation. A fifth chapter has been added on the Latinity of Cicero and Caesar respectively, so that, after passing in review the style of Poetry and that of the popular dialect, I might do justice to Classic Prose as well: an Index has been added, and a collection has been appended of researches and treatises which have appeared during the last few years in German lit- erature. VI PREFACE TO THE THIRD EDITION THE Third Edition differs from the second by the addition of a chapter on the Civilization and Vocabulary of the Romans: it contains also a large number of additions and amplifications most of which are to be found in the notes. I am indebted to M. Graziatos, Director of the Gymnasium at Argostoli in Cephallenia, for some suggestions : his translation into modern Greek appears contempor- aneously with this edition. ElSENBERG, S.A., 1905. Vll PREFACE BY THE TRANSLATORS IT is hoped that this translation of the suggestive work of Professor Weise may prove useful to Classical Students in Britain and America. We have endeavoured to render it so by adding refer- ences to English works on the subjects dealt with in the text and notes, and by a few additions and suggestions, particularly with regard to the ety- mology of certain words, for which we are mainly indebted to the full and scholarly work of Professor Walde of Innsbruck. The notes at the end will be found to contain many valuable references to the literature published in Germany in recent treatises dealing with the subject matter of the text. To the Bibliography at the end of the Appendix should be added the valuable work of Mr. Duff, " A Literary History of Rome," Fisher Unwin, 1909. H. A. Strong. A. Y. Campbell. VIU CONTENTS I. The Latin Language and the Roman Character (§§ 1-33) .... The Roman character interpreted by the follow- ing traits in the Latin language: (i) Word-formation; wealth of vocabulary; usage of particular words (§§ 3-8). (2) Metaphors (§§ 9-1 1); original signification of words (§§ 12-18); proverbial expres- sions (§ 19); "winged words" (§ 20); puns (§ 21). (3) Periods and style (§§ 22-29). (4) Inflection (§§ 30 sqq.) and system of sounds (§ 32); summary (§ 33). n. Style and Development of Culture (§§ 34-59) ( 1 ) The oldest Roman literature ( Appius Claudius, Cato) before and after its contact with Hel- lenic influence (Plautus, Terence, Ennius, Pacuvius, Accius, M. Antonius, L. Licinius, Crassus, etc. (§§ 41-46). (2) Prose and Poetry at the end of the Republic and at the commencement of the imperial rtgime (§§ 47-53)- (3) Period of the degeneration of style: Renais- sance and Rococo (§§ 54-59). ix PAGE 33 CONTENTS III. The Language of the Poets (§§ 60-88) (i) The Law of Beauty in Poetry (§§ 60-66). (2) The claims of Perspicuity (§§ 67-76). (3) The Language of Poetry and its truth to Nature (§§ 77 ^??-)- ^ . (4) The license allowed to Poetry as contrasted with that allowed to Prose (§§ 79-8»)- J IV. The Language of the People (§§ 89-1 1 1) (i) Tendency to spare trouble (§§ 89-98)- (2) Tendency to clearness of meaning (^S 99" (3) Tendency to reflect thought and character in language (§§ 108-111). V. The Classical Language of Caesar and * Cicero (§§ 112-130) (i) Care in selection of words (§§ 112-118). 2 Principles of these writers in the adoption of inflexional forms (§§ 119-120). (3) Syntactical peculiarities (§§ 121-124)- (4) Stylistic peculiarities (§§ i25-i3o)- Appendix ••••■' Roman culture as mirrored in the Latin Voca- ^ bulary. Notes ..•••'* Index .•••*** PAGE 105 144 181 211 225 255 LANGUAGE AND CHARACTER OF THE ROMAN PEOPLE THE mental activities of any given individual fall roughly into two categories — those of cog- nition and those of emotion. The proportions in which these psychical elements are mingled are responsible for the great differences in the mental endowment of mankind: in some in- dividuals we see the feelings developed at the expense of the intellect, while in others the intellect preponderates at the expense of the feelings. In some cases the understanding and the will, in other cases the emotions and the heart assert their pre- dominance. And as it is with the individual, so it is with nations as a whole. Few, indeed, are the indi- viduals, and few the nations that nature has evenly favoured with all mental endowments. Amone the nations of antiquity, however, the Greeks stand pre-eminent in respect of this general endowment, while in the Romans, reason and will power were unmistakably developed at the expense of the other mental faculties. *' The taste of the Romans," says Herder, *' was for History, or for solemn legal oratory, in a word for Action:' Thus Sallust says (Cat. 8, 5) : B 2 LANGUAGE AND CHARACTER "Optimus quisque facere quam dicere malebaC* a^d Livy puts these words into the mouth of Mucus Scaevola: " Et facere et pati fortia Romanum est^ The most striking traits in the character of the Romans were their stately and impressive demean- our their unflinching perseverance and constancy, their firm and imperturbable courage: or to cite Cicero's expressions, their^r^z^/^see note at end) continentia, and animi magnitudo (Tusc. i, i, 2). l ne beau ideal of a genuine Roman of the old stock is summed up in the old-world io.^n\^ vtrfortu a que siremms{C^toap. Festum,p. 201, A Cell, xvu, 13, 3) which, at a later period in the time of the Scip.os. was under Greek influence restated (as we find it on the tomb of Barbatus) in the form fortis vir sapiensque. The valour of Roman citizens qualified them in an eminent degree for soldiers, their intelligence and practical understanding made them statesmen and lawyers, their calm and unruffled common sense and their clear apprehension fitted them for oratory of every kind. The words applied by Cato the elder to the Gauls. " Duas potissimum res Gallia sequitur. rem militarem et argute loqui "O. ho^ good in a measure of his own countrymen. It was to the special capacity of her sons for war and politics that Rome owed her rise from an unimportant state to a world-power of the first order. 2 As the mental endowments of the Romans were severely practical, and such as inclined them to take a sober view of the circumstances of life, we cannot be surprised to find that they had no special taste I OF THE ROMAN PEOPLE 3 for either Art or Science. Their imagination could not soar to the height of either. Vergil confesses as much in his melancholy reflexions contained in the lines (Aen. vi, 847 sqq)\ Excudent alii spirantia mollius aera, Credo equidem, vivos ducent de marmore voltus, Orabunt causas melius caelique meatus Describent radio et surgentia sidera dicent, and Cicero confesses " Doctrina Graecia nos et omni litterarum genere superavit: (Tusc. i, i, 3), nay, he actually goes so far as to say " Nos, qui rudes harum rerum sumus" (Verr. ii, 87). In like man- ner the greatest Roman epic poet confesses that even as a she-bear brings forth awkward and mis-shapen cubs, which she has to lick into shape, even so are the offspring of his brain raw and imperfect, and he can only impart to them the features they should wear by long and toilsome labour. The inhabitants of Latium care to occupy themselves with such pur- suits only as far as may serve some practical advan- tage, more especially the good of the state; for, from a Roman point of view, as Tacitus says (Dial. 5): "ad utilitatem vitae omnia consilia factaque diri- genda." We cannot wonder that the unremunerative arts are designated by the significant appellations studia leviora (Cic. De Or. i, 49, 212, De Sen. 14, 50), studia minora (Cic. Brut. 18, 70), artes leviores (Cic. Brut, i, 3), or artes mediocres (Cic. De Or. i, 2, 6), and that it was only by a slow process, and after a long struggle, that under the influence of the Hellenic spirit they were enabled to attain a higher level and to claim more respect. U I I ', 4 LANGUAGE AND CHARACTER What is more, the Romans possessed in a very moderate degree the gift of sympathizing with the beauties of nature and penetrating into her secrets. The joy of wood and field, of rambles on wide moor- lands, of scaling lofty mountains, of all, in short, that has charms for chivalrous races like the Celts and the Greeks, has no voice for them, and while the Greeks enliven their heaven and their earth with a throng of gods of fair form and dazzling beauty, the Romans cannot rise above the idea of endowing certain abstract powers of nature with divine attri- butes. They are unable to create myths, or to people seas, rivers, mountains, and moorlands with the fair figures of graceful nymphs. 3. Now let us consider how these national charac- teristics of the Romans have stamped their features on the Latin language {'). It has long been recognized that the vocabulary of Latin is poorer than that of Greece {% and it is equally certain that a large portion of this vocabulary had to be recruited from foreign countries. Now when a nation borrows a large num- ber of words from a foreign tongue, it proves itself to have been deeply susceptible to the influence of the nation from whom it borrows ; it proves, more- over, that the borrowing nation possesses a less active mental activity and less power of imagination. It is notorious that while the number of Greek interlopers into Latin may be reckoned by the thousand {% the Greek language, in spite of the mighty tide of Orientalism which flooded all Hellas, can point to scarcely a few hundred words of Asiatic origin. The OF THE ROMAN PEOPLE 5 imaginative disciples of the Phoenicians have im- pressed the stamp of the Greek spirit on most of the gains for which they are indebted to their Eastern neighbours. They have suited their borrowings to their needs and have renamed them in their own style. Thus we could hardly guess from language that the potter's wheel (rpo^og from Tpix^i]/)y that frank- incense (6uo? from 9 Jf ii/) and thegourd (ttettcov from Trea-a-siv) are natives of Asia, or that voayoe, from u? (the hyena), the ichneumon (from Ix^usvsiv, to track, i.e., crocodile's eggs), and ^pofxocg, the dromedary (from ^poc[As7y, to run) are words of foreign origin (^). 4. The Roman methods were very different. With them the traces of such creative linguistic activity are small indeed. It is true that they made some efforts in this direction ; for instance, they invented some names of their own coinage for the pomegran- ate (malum granatuin), the arbutus, the litter (lecticd), letters of the alphabet {littera), the cloister {porticus), the amulet {amuletum from amoliri, a translation of (puAaxTiipioi',* see, too, Weise's essay in the '* Zeitschrift fiir Volkerpsychologie und Sprachwissenschaft," xiii, 244). But they never advanced very far in this direction ; indeed, in many cases they actually gave up genuine Latin expressions already in use in favour of foreign ones, as in the case of elephas for bos Luca, and the chestnut (nux mollusca or calva), etc. In cases where the origin and the derivation of a * More probably from amoliri, as an averter of evil; and if so, a genuine Latin word. See Walde, " Etymologisches Worter- buch," s,tK 6 LANGUAGE AND CHARACTER Greek word were obvious to their apprehension, they certainly rose to the height of translating it, more particularly from the middle of the first century B.C. There are many departments in which their efforts in this sense were perfectly successful (^), but their proceeding was, as a rule, to avail themselves of the Greek expressions for art. Can we therefore be surprised at Cicero thus expressing himself (De Nat. Deor. i, 4, 8): ^' Complures enim Graecis in- stitutionibus eruditi ea, quae didicerant, cum civi- bus suis communicare non poterant, quod ilia, quae a Graecis accepissent, Latine dici posse diffiderent," and (De Fin. iii, 15, 51): " Quod nobis in hac inopi lingua non conceditur"; or that Seneca (Ep. 6 i), thus laments: ''Quanta nobis verborum paupertas, immo egestas sit, nunquam magis quam hodierno die intellexi. Mille res inciderunt, cum forte de Platone loqueremur, quae nomina desiderarent nee haberent, quaedam vero, cum habuissent, fastidio nostro perdidissent '7 5. Another cogent reason for the large scale on which the Romans borrowed foreign words is to be found in the incapacity of their own tongue for the manufacture of compounds, a peculiarity which has descended also to its Romance daughters ('). The poet Lucretius (De Rerum Natura, i, 830) dwells on this fault in his own tongue in the following words : Nunc et Anaxagorae scrutemur homoeomerian Quam Graii memorant nee nostra dicere lingua Concedit nobis patriae sermonis egestas, Sed tamen ipsam rem facile est exponere verbis i OF THE ROMAN PEOPLE 7 and Livy makes a similar remark when referring to the word androgynus\ he writes in terms significant indeed, but intended to spare the national self-respect: **quos androgynos vulgus ut pleraque faciliore ad duplicanda verba Graeco sermone appellat." Cicero expresses himself in the same sense (De Fin. iii, 4, 15): **Equidem soleo etiam quod uno Graeci, si aliter non possum, idem pluribus verbis exponere. Et tamen puto concedi nobis oportere, ut Graeco verbo utamur, si quando minus occurrat Latinum, ne hoc ephippiis et acratophoris potius quam proeg- menis et apoproegmenis concedatur"; and Gellius writes in the same spirit (Noct. Att. xi, 16, i) when touching on the topic of the borrowing and transla- tion of Greek words like TroAuTrpay/xoo-Ji/Ti, 7roAi;a9po^), fennel: indeed many which look like genuine Latin words are merely literal translations of the Greek, like ranunculus, from (iocrpocx^ou. Again, while Greeks and Germans alike, to aid their designations of remarkable products of nature, especiallyin the case of plants and trees, borrow the * See Geiger, "Lectures and Dissertations" (1880), on colour sense. Both Romans and Greeks confounded blue and violet, especially with gray and brown. The Romance languages found no word for Mue in Latin, and were obliged to borrow one from the Germans; cf. fi/eu and old Italian h'avo, from Mm^, which itself originally meant /?/ack. t Probably connected with English dale in bale-fire, and with Russian ^/'^//>, white. II OF THE ROMAN PEOPLE 9 names of the most striking domestic animals, the Romans lack all sense for such comparisons. They have therefore no words which can challeng-e com- o parison with the Greek iTnroa-iXiuoi^, Iwirovpig, (^6vyXu(r(rogy or with the English horse radish, horse chestnut,* or the German Rosskastanie or Ochsenzunee, etc. : for words like Equisetum (horse's tail) betray at once that they are mere importations from Greece. Further, we find in Greek literature many more graceful adjectives which testify to a keen observa- tion of nature on the part of those who used them. In Homer all is light and colour: epithets such as shining, glittering, radiant, and again picturesque touches, like trailing-footed, crumpled-horned oxen meet us at every turn and become to our fancy an indispensable accessory to the Homeric poems. The Roman imagination, on the other hand, receives such faint impressions from nature that it is unable to impart them in any high degree to its poetry. Latin again lays in many cases a greater stress on - number and magnitude, where we commonly empha- size the quality or effect of a substantive. Thus the word fnagnus is combined with the following words : argumentum (a convincing proof), exemplum (a strik- ing example), suspicio (a strong suspicion), p^^eces (fervent prayers), vox (a loud voice), hiems (a violent storm), occasio (a lucky chance), coniunctio (a close alliance), usus (a lively intercourse), officium (a sacred duty). The adjectives which we attach to such words are less vague and general, and denote rather some * Cf. mare's tail; also such words as ladies-fingers, catkins, larkspur, henbane, cowslip, oxHp, etc. lo LANGUAGE AND CHARACTER quality which, as it were, individualizes the substan- tive in each case. Again, how poor is Latin in such words as the particles which serve to express differ- ent shades of our mental attitude, and to bring into bold relief the object of our thoughts! We have only to compare such Greek words as av, ap«, r«, ra, h, etc., which from Homer down serve to enliven and adorn the language of the Greeks, with the very meagre resources provided by Roman literature, and we shall find that the Greek language is far more flexible, and far more capable of expressing the finer nuances of thought than its Italian sister. 7. We find greater activity in the process of word- creation in Latin in places where the peculiar Roman characteristics most assert themselves. C. Abel, in his " Sprachwissenschaftliche Abhandlungen," p. 25, says with perfect truth: " A nation which possesses many words for any conception, be it material or spiritual, must be much concerned in the object of that conception, must have dwelt on it, developed it, and refined upon it." * . • 1. Examples are not far to seek : to bear pain with patience was not merely a trait of Stoicism, but an essentially Roman characteristic. From Mucius Scae- ' vola, who thrust his right hand into the burning fire before the eyes of the Tuscan King Porsena, Roman history has furnished us with plenty of instances of this national virtue. The Romans accepted bodily * See Heine, Reisebilder: Reise von Miinchen nach Genua. " The Arab has a thousand words for a sword, the Frenchman for love, the Englishman for hanging, the German for drmking. OF THE ROMAN PEOPLE T T pain without a murmur of complaint : the most that pain could wring from them was a short cry, the reflex of their agony. Hence it comes to pass that the interjections expressive of painful feelings are more numerous than those of any other nature, and they bear a more national and truly Roman stamp than interjections expressive of joy, which latter, it may be remarked, are mostly borrowed from Greece. To the distinctively Roman utterances belong Oy keu, eheu, pro, vae, ah, hei, ohe, au : while among those borrowed from the Greek we may mention io, euoe^ euax, eu, euge, eia. Again, the Roman has a large number of expressions for slaves : without slaves his life was impossible: he required their services at every turn and for every purpose : thus servus is to the Roman a slave looked at as a social inferior: famulus, as one of xSx^familia or household (Oscan fama, a house): mancipium, as a marketable com- modity: verna, as born in the household : puer, with reference to his age: minister and ancilla, with reference to his or her capacity for service. But it would take us too far to ransack the entire vocabulary of the Latin tongue for instances of this kind: two more may suffice. We are purposely setting aside the peculiar department of knowledge which the Roman from the earliest times proudly proclaimed his own, that of Law and Politics, or Statecraft. The terms in which these two sciences express them- selves permeate the whole Latin language, and can- not here be referred to more particularly. But it may be interesting to cite in favour of what we have advanced a few facts referring to the words which 12 LANGUAGE AND CHARACTER carry the signification of relationship, and to articles of food. 8. The Romans had a warmer feeling and sym- pathy for family and its ties than the Greeks. The entire contents of a single household were regarded as a single large unity, ruled by the paterfamilias, duly organized, and each member knowing accurately his position in respect to the rest; in fact, the family was in its constitution an exact counterpart of the Roman State. They reverenced and venerated their forefathers : the virtue of such reverence was called pietas : it was one of their chief delights to compose genealogical trees, and they loved to connect the origin of their own gens with the Fall of Troy and the arrival of Aeneas in Italy. Thus we need not be surprised to find that they had a rich store of names expressive of family relationships. We speak of uncles and aunts; the Romans mark the difference such between maternal and paternal relatives ; avun- culus ^nd patruus; matertera and amita\ their lineage extends back from avus, abavus, proavus to iritavus: patruelis denotes the brother's child, consobrinus the child of the sister. They actually possess a word to denote the relationship of two women married to two brothers : ianitrices,^ The favourite animal food of the Romans was pork. Pliny tells us that they knew no less than fifty different ways of preparing it for the table (Nat. Hist, viii, 209; cf. Friedlander, '* Sittengeschichte,'* iii, 28). The very term caro suilla, a diminutive * So glos is a husband's sister. I OF THE ROMAN PEOPLE 13 form, shows the weakness of the Romans for their national dish.* In the ancient compound suovetaur^ ilia ( = sus + ovis + taurus) it is the sow that takes precedence of the sheep and the ox. It is therefore natural to expect to meet in Latin with more terms to express "swine" than any other animal. Besides sus we find porcus.porca, verres, aper, scrofa, maialis, nefrens. In Roman farces the swine appears as a constant object of diversion : the writer of Atellanes, Pomponius, named no less than four pieces after this animal : Porcetra (a young sow which has once far- rowed); Maialis (a fat hog); Verres aegrotus (the sick boar) ; and Ve^-res salvus {the boar convalescent). We may regret that it was not usual in Roman times to christen the chief actors in the national farces with the name of one of the national tastes or fail- ings. In that case he would probably have been called some name like Jack Porker, as the Germans call their chief figure in their farces Hans Wurst, the French Jean Potage, and the English Jack Pudding. Besides, the weakness for this dish gave rise to a number of popular proverbs. The German talks of roasted pigeons flying into his mouth: the Greek makes roasted fieldfares (oVra* ydx^oci) per- form the same kind ofiice : the Roman people uses coctiporci in a similar sense (cf. Petron. 45, 4). Indeed Cato, quoted by Cicero (De Sen. 15, 56) declares that peasants call their gardens, *'a second flitch of bacon" ; "jam hortum ipsum agricolae succidiam alteram appellant." To act harshly and * It is noteworthy that one of the reasons which made the Jews unpopular at Rome was their aversion to pork. \ .> ,%«■ ki ,4 LANGUAGE AND CHARACTER without reflection is expressed by the proverb " apros immittere liquidis fontibus" ; to kill two birds with one stone, ^*duos apros capere"; I shoot the game and another eats it, *' ego semper apros occido, sed alter semper utitur pulpamento." All of these are convincing proofs that the ** animal propter convivia natum " was the delicacy most prized by the Roman palate.* 9. Metaphors are one of the main factors m the development of language, and they accurately reflect the spirit of the nation which employs them. We may therefore expect to find in the metaphorical expressions of the Romans a faithful mirror of their popular beliefs and predilections. It is only natural that when the speaker casts about for a fit com- parison, he should seize on the subject of his predilec- tion : t and mankind is only too prone to extend his own circumstances and qualities to the external world. Hence it happens that in the similitudes he * Cf. Cels., lib. iii, 9, " Protinus suillam assam et vinum homini dabant." . •• o t Mr Keble in his " Praelectiones Academicae, OxoniL i»44, p ICO describes the Homeric metaphors and similes. They will be found to show that Homer was a keen observer of nature II. viii, 553; iii. ^o; conversant with the sea iv 274; with agricultural occupations, xxi, 343; ^'h 45 M xui, 701, etc. ine metaphors in Aeschylus are very often taken from the customs of animals wild and tame. Cf. Agam. 11, tdtd., 35; Eu^n. i ; Suppl. 354, ef saepe. Pindar's are mostly taken from Public Games, cf. Isthm. 5, i- The metaphors in Lucretius indicate a great love of nature (De Rerum Natura, iv, i). The English reader may consult Minto^s « Manual of Prose Literature, p. 15 ; he also gives the sources whence the greatest English writers draw their figures of speech. OF THE ROMAN PEOPLE 15 employs he presents us with a view of his intellectual impulses, his feelings, his emotions. In Lessing^s works the commonest metaphors are those taken from combat, and this harmonizes with the writer's fondness for disputes and feuds.* If certain meta- phors are found to colour a language not merely in special periods, but in all its stages; when, in fact, they are the common property of all the writers and speakers in that language, we are justified in con- cluding that they comprise the favourite conceptions of an entire people. And it is indeed true that agriculture and military life, the two main columns on which the Roman state rested, are called, in Latin, to do service as metaphors with surprising frequency. 10. When we find a nation insisting on its members being addressed in their civil and polit- ical capacity as '' Quirites,'' i.e., "Spearmen" or '* warriors," and investing its politically emancipated citizens and its armed reserves with a similar name; when, in short, we find military service and military privileges regarded as identical with civic service and civic privileges, we cannot be surprised to find that such a nation scatters military metaphors broad- cast through its literature. It has been well said by D. Wollner (*' Landauer Programm," 1886), *' When the Romans have to express any circumstance in which two opposing forces meet, they immediately * English readers will remember that few writers can be said to have shown their complete philosophy by their choice of meta- phors so much as Omar Khayyam. Iff' i6 LANGUAGE AND CHARACTER employ some metaphor which indicates their warlike propensity"; and S. von Raumer lays stress on the fact that of all metaphors those which have reference to war are the commonest (- Die Metapher bei Lukrez/' Erlangen, 1893, P- i^i). Indeed war is the very life and soul of the Roman. 1 hus Uio Cassius (xxxviii) makes Caesar at Vesontio address his soldiers inclined to mutiny from terror of Ario- vistus and his Germans, "Oxav oZ. xiy^ tk, or. ou xP»^ .oX./x.r. .>.., oJJe. a'AAo ^..1. ^ or. oJ XP^ ttAout.., ou vp.^ mpc. ipx^^^. oCk IX^vUpov,, oJ 'F^f^cclov, a.a. ; and Livy (xxii, 12, 4) puts these words mto the mouth of Hannibal: '* Victos tandem illos Martios animos Romanis"; while Cicero (Tusc. ii, 16, 37) says: -Nam scutum, gladium, galeam in onere nostn milites non plus numerant quam humeros, lacertos, manus; arma enim membra militis esse dicunt" Expressions like s/>o/mre are of ancient date : it signifies strictly to strip a conquered foe of his arms : then, generally, to despoil. Intervallum means strictly the open space within the mound or breastwork of a camp, the space be- tween two palisades {inter vallos) and then comes to be used of any interval. Praemium {prae and emere, to get or take betore another) means in the first place profit derived from booty (cf. 2\sopraeda) and then, generally speaking, reward or recompense. . Princeps originally = qui primuni capit, he who is the first to seize booty (cf. particeps = partem capiens) : then the first or most prominent in rank. Excellere applies in the first instance to the shoot- OF THE ROMAN PEOPLE 17 ing of weapons over a mark, and so means " to sur- pass " generally. In the case of these words the original significa- tion has almost entirely disappeared. There are other words used in a tropical sense, in which the metaphor is more apparent: for instance, sub hasta vendere^ which means to sell at auction, but which refers to the custom of selling captured foes beneath the spear (German subhastiren). Substantives again like tiro [bonus homo semper tiro est), tirocinium, commilito, acies, telum, arx, stipendium, signifer, militia, bel- lum, castra, clipeus, etc., are frequently employed metaphorically. Fabius was nicknamed 5^^/^;;^; Mar- cellus, gladius Romanorum : the discoverer of a trick is in Plautus often called ** General": '*to outwit" is military strategy or a siege; the object of the trick is an enemy's town, more especially Troy. Novius says to a wordy poetaster, '* Ut sol crescit, cerea castra crebro catapulta impulit," and Cicero calls the lex Aelia et Fufia, **propugnacula tran- quillitatis." Varro begins his treatise on agriculture with the words, "Annus octogesimus admonet me, ut sarcinas colligam, antequam proficiscar e vita," and in Pliny the Elder we find the tropical use of such words as excubare, infestare, rebellare, occupare, quite an ordinary occurrence (J. Miiller, *' Der Stil des alteren Plinius," Innsbruck, 1883, p. 119). Ovid makes the morning star (Met. ii, 115) who occu- pies the last rank in the army of the stars ('* qua- rum agmina claudit Lucifer ") leave, last of the soldiers, his post in Heaven (** novissimus caeli sta- tione exit"). Our proverb, "to make a mountain c i8 LANGUAGE AND CHARACTER out of a mole-hill" is rendered in Latin by **arcem ex cloaca facere." To risk much for nothing is "hastis trium nummorum causa sub falas subire*'; to burn one's boats, " abicere hastam, scutum"; to abscond safely, " tecto latere, abscedere " : all these are proverbial expressions drawn from military life.* II. Agriculture and cattle-breeding are as fruitful a field for metaphors as the last. The inclination of the Latins was for agriculture, and they carry its stamp. Horace calls his countrymen (Carm. iii, 6): '• rusticorum mascula militum proles, Sabellis docta ligonibus versare glebas." ** Roman life depended wholly on agriculture, and maintained its moral force as long as this branch of social activity existed in its simple purity.'* The pursuit of agriculture remained even in the period of refined luxury the ideal life of the noblest and most honoured Romans, the life most respected after that of the statesman and the soldier, so that Horace can reckon the man happy, " qui procul neg- otiis Ut prisca gens mortalium Paterna rura bobus exercet suis" (Epod. ii, i sqq.\ cf Verg. Georg. *'divini gloria ruris"). The plough was used to draw the furrow round the enclave of a town about to be founded, to mark the circumference of the future walls, and the division xnlo gentes, and indeed the constitution itself was based during republican times on the possession of land. It thus happens * Cf. Macrob. Sat. ii, 8, " Congrediendum igitur et tamquam in acie quadam cum vini licentia cominus decernendum." OF THE ROMAN PEOPLE 19 that Latin displays a large store of expressions borrowed from agriculture and used in a new sense. Delirare, lit. to go out of the furrow : then to act like a madman (cf. delirious). Tribtilare, to thrash with a tribulum: then to plague. Praevaricari, to plough in crooked lines: then said of a counsel who plays into the hands of an opponent. Emolumentum, what is ground out {e-molere): then gain or profit. Calamitas, a plague, destructive to crops, such as fire or hailstorm : then calamity [the derivation from calamus is doubtful. See Walde, s,v^. Adoria, glory in war [connected by popular ety- mology with ador^ spelt].* Rivalisy a rival, connected by popular etymology with rivus as if it were *'the neighbour on the bank." AcervMS [possibly] from acus, aceris, chaff. Saeculum [probably] " sowing season " (cf. saison from satid) : then a century (cf. Saeturnus, Sa- turnus from the root of severe), Cohors, the hedge of a field or garden: then a cohort. Manipubts, an arm-filling bundle. Inanis, empty [possibly] from acnay a measure of land (with in privativum), * Copiae^ plenty, is specially applied to troops or forces^ and copiae marinae is used for the fish supply. Cf. Macrob. Sat. ii, X, ad init. 20 LANGUAGE AND CHARACTER Felix, originally "fruit bearing" (cf. fe-cundus, fetus, fenus) : then " happy." * Who would think of connecting pecunia with pecus [cf fee], egregius with grex, septentrio with the three oxen for treading out corn, as the people called the seven stars in the constellation of the Bear? In the language of the poets we find in common use such expressions as vada carina sulcare, cerea prata sulcare, aeqiwr arare, librum exarare, proelia severe, barbam metere, viam carpere \Jiorani carpere\ polus sidera pascit, uber glebae, mare mugit, etc. Then we find proverbial expressions like arare bove et asino to manage awkwardly; arare litus (to plough the sands); adhuc tua messis in herba est, 'tis too soon to begin : and 'Axpaya? is by popular ety- mology converted into Agri-gentum. Similarly measures of space like jugerum from jugum, actus from agere [in quo boves aguntur, cum aratur, cum impetu justo,—?Xm. i8, 59], vorsus from vertere (the turning of the plough), and such words as cam- pus, fios, ager, seges, fructus, trisulcus, give material for many metaphors: eg,, Cicero calls Clodius segetem (field or soil) ac materiem gloriae Milonis. 12. The signification of Latin words affords us a profound appreciation of the moral and intellectual views of the Romans; indeed, it is not too much to assert that in no other way are they so faithfully mirrored. Their wishes, their sentiments, their thoughts and their poetry all stand revealed through ♦ Clfructus, ctiltura, peculmm, evincere, protelare. OF THE ROMAN PEOPLE 21 I this medium. Language, as we know, never ex- presses any notion in its entirety. A curt denom- ination cannot possibly denote all the characteristics or qualities of any subject:* but only the most striking, or those which appear so to the speaker or writer. Lessing's maxim was true to life when he stated that the poet should not bring into promin- ence more than a single feature of a subject at one time. " The etymological meaning of a word never exhausts the full meaning; it is impossible that it should do so : all elements of language are merely representative [and not full pictures] (Steinthal, " Klassif." 281). And it is precisely for this reason that personal views and personal feelings are no small factor in the growth and spread of words. It may happen that one people may hold one feature as the essential characteristic of the word : another people may hold another feature as more truly so. Thus it is that etymology enables us to realize every corner of the intellectual storehouse of any given people. It is no doubt true that by its aid we are able to catch merely the earliest phase of the mean- ing of any given word; we can only state with absolute certainty the sense attached to the word by those who coined it, and what they considered the principal characteristic of the object denoted. But if we study the semasiology, the development of the signification of any given word, the restric- tions and expansions of its meaning caused by the feelings and impressions which have attached them- selves to it in the course of its existence, we shall * See Whitney, p. 409 sqq. 22 LANGUAGE AND CHARACTER be enlightened as to many psychological processes in the human mind, and shall obtain many a glimpse into the spirit of those who used the word as well as of those who coined it. The essential lesson for the Roman student [discipulus) to learn {discere) was discipline {disciplina).^ This word corresponds in form to the Greek juae»/xaTix>i, but how far apart have the words drifted! The father of a Roman family rules his household with autocratic rigour: and just as the father's authority over his son is unquestioned (as it is in- deed over his son's kin) so is that of the patromis over the cliens\ that of the patricii over the plebeii\ that of the /^/r^^ "elders" over other citizens: the idea of paternal authority is felt throughout. The very name for " country " is " fatherland " {patria): that for mother-tongue, patrius sermo. We call our language the "mother-tongue" — and think with more sentiment of the loving care with which she taught us to lisp our first sounds. It is significant that whereas Homer introduces his hero Odysseus by the epithet riQ<;, Vergil presents us his ^neas with the title oi pater, 13. Woman is in the Roman's parlance, mulier (probably connected with mollis), the soft creature who needs men's protection : he calls a boy puer, but employs the diminutive puella for a girl : so ancuSy and ancilla. The Germans, according to Tacitus, in * Discipulus is derived by Walde from dis-capio, I receive men and teach, its opposite being praecipere, to undertake something with pupils. OF THE ROMAN PEOPLE 23 the remotest times regarded the woman as " sanc- tum aliquid et providum": the word weib [wife] denotes something inspired [so Kluge; Skeat says it is thought to come from a root signifying to tremble] : hence the awe and veneration with which the priest- esses were regarded. At a later period the Germans exalt their woman into Frau (O.H.G., frouwa, M.H.G., vrouwe), i.e., house-mistress, wife of the house-master: this word is connected with the Gothic fratija, Lord, and with the H.G. Fro7t (seen in Frondienst, Frohnleichnamsfest, and fr'dnen [to labour for a master]). Comparing the mental atti- tude of Roman and German toward the gentler sex, we find that in Latin the wordfratres denotes brother and sister, and sponsus and sponsa are used for two spouses. The Germans use the terms Geschwister and Brautpaar respectively, denoting, it must be admitted, a greater feeling of reverence towards the ewig weibliche. On the other hand it must be con- ceded that language seems to indicate that woman stood higher in the estimation of the Roman than of the Greek. The Greeks say r£>ci/a xal yuvatxe?, the Romans say, conjuges liberique^ when they would express what they hold dearest, and mulieres puerigue, when they would dwell on their helpless- ness: and in this they agree with our method of expression. Love is to the Roman more an impulse of the intellect than of the heart. Diligere signifies in the first instance simply to discriminate.^ The idea of * Or it may be from the same root as dXcy^, "to trouble one- self about." See Walde, s.v. I iilij 24 LANGUAGE AND CHARACTER dutiful affection attending on certain situations con- ditioned by relationships or other outside circum- stances is genuinely Roman : caritas denotes affection for one's own flesh and blood or for a friend (cf. chariti) : pietas dutiful respect towards the gods or parents, and to the mother country as the lasting benefactor of each man : studium denotes an affec- tion based on political or personal obligations, and aiming at merely worldly ends.^ Here we have the picture of the Roman, his life and his love: he took full advantage of the closest natural relations, but he respected them as well, and he utilized them for his own purpose, while regarding them with honest goodwill. He turned his affections to the quarter whence came his needs, and he held it his sacred duty to requite those who aided him (C. Abel, " Sprachwissenschaftliche Abhandlungen," Leipzig, 1885, p. 88 sqq,\ Even the love which flows from the depths of his heart, the love which the Latins call amor, was regarded by the Roman people not from its spiritual side: amor was to the Roman a malady, a consuming fire, a fatal wound.f With the exception perhaps of Tibullus, the poets seized on the strongest possible expressions, which indeed they could not heighten, to express the power of such love (cf. Weidner on Verg. Aen. i, 660). How different is the Teutonic conception! Luther betrays a profound knowledge of his own mother- tongue when he says in a letter on interpretation : * Adfectus is the nearest Latin word for an emotional love, t This is most noticeable in the well-known passage of Lucre- tius (De Rerum Natura, iv), the most Roman of poets. OF THE ROMAN PEOPLE 25 *' I hardly know whether it be possible to translate the word lieb, dear, into Latin or any other language so as to express its tenderness, so that it may call such a responsive echo from our hearts as it does in German." The tenderness of love in the case of Teutonic nations depends on faith and mutual con- fidence : hence it is that such words as the German Glaube {ge-loube) and Liebe, and in Gothic indeed the word lubains (hope) come from the same stem. The Teutonic conception of love is that it rests on the emotions : and our emotional nature, irradiated and warmed by the quickening sun of Christianity, is a flower which never came to its perfection on Roman soil. Indeed neither Latin nor the Romance languages possess any expression which exactly renders Gemiif : and the derivatives of animus point rather to a source of wrath and passion than to one of what the Germans call Gemiitlichkeit (a term lacking in English as well). 14. Again, the conception of marriage in Latin is based on no deeper insight into nature. Betrothal (nuptiae) is simply ''taking the veil '' {nubere alicui, to veil oneself before the bridegroom) : ^ or again it is a matrimonium or ''mothering," ix,, an arrange- ment for the continuation of the race : or again a common sacrifice of a cake of spelt (confarreatio from far). In the eyes of the German, marriage is a lasting contract, a legal agreement and bond be- tween husband and wife, voidable only by death * Nubere is derived, however, by Walde from a root snu, signi- fying in Slavonic (O. Bulgarian) " to love." ■I'l 26 LANGUAGE AND CHARACTER {ewig and Eke^ originally Swa, are from the same stem as aevum [cf. to wed, from A.S. weddian, to pledge]). The Teutonic conception of marriage is a Hoch Zeit, a sublime and glorious day, or an event depending on mutual confidence; a betrothal. His consort stands so high in his estimation that he regards her as entitled to the same rights and privi- leges as himself, and calls her in fact his '* Ehehalfte " (cf. our "better-half").* The Roman regarded school not as a place for intellectual exertion, but as a '* sport " {ludus). Ac- cordingly we are not surprised to find that Latin takes over the Greek word a-^oAn', leisure, and em- ploys it in the signification of school^ nor that it attaches to the word otium the connotation of intel- lectual occupation : such occupation serves as a re- freshing rest after effort. It is significant, too, that Cicero represents most of his dialogues as spoken in the holidays (cf De Or. ii, 13; i, 102; and Seyffert-Muller on " Laelius," p. 93) [and Wilkins edit, of De Or., p. 6]. Literary activity in primitive times hardly goes beyond letter-writing: litterae signifies in the first instance what is committed to writing, especially a letter: and only at a later stage science in general. In the Greek language the words iroiuk, TTpdrTHv, and ayjtj/, to act or do, which have de- veloped a vague and colourless meaning, manifest in the substantives derived from them three essential characteristics of the Greek popular character — TTo/ncTK, Trp^^^g (Homer), cHyuu: the taste for poetry and art in general, for trade and for contests. The * Uxor is now supposed to mean "the woman carried home." OF THE ROMAN PEOPLE 27 Roman derivatives answering to the Greek assume a political, or, at any rate, a practical signification, far removed from any notions of literary or artistic taste. The religious side of the Roman character comes out in such ancient derivations as ag-ere, as axamen^a, and indigitamenta [but both these words are now connected with aio\ acta diurna would be a better instance]. 15. The Roman holds pleasures to be mere temptations {deliciae and delectare from delicereY and we may gather his ideas of dancing from Cicero^s utterance: "Nemo fere saltat sobrius, nisi forte insanit^t (Pro Mur. 13). An honest man may indeed allow himself to thaw a trifle over his meals: convivitim is, according to the Roman, "a living together" in the literal sense: a favourable opportunity for exchange of thought, not for a carouse, as with Teutonic nations, with whom it might be more correctly described as a convimum, just as the Greeks call it a symposium, Cicero is fully justified in putting into the mouth of the elder Cato the words : " Bene maiores accubitionem epu- larum amicorum, quia vitae coniunctionem haberet, convivium nominaverunt melius quam Graeci, qui hoc idem tum compotationem, tum concenationem vocant, ut quod in eo genere minimum est, id maxime probare videantur." Of the good gifts of life Glory is the noblest. * More probably connected with laqueiis, a snare. t Macrobius, Sat. ii, 10, devotes a chapter to proving that the ancient Romans saw no harm in dancing or singmg. y" 28 LANGUAGE AND CHARACTER Hence to be ignored {ignominid) is the greatest dishonour. " Apex autem senectutis auctoritas est " (Cic. De Sen. 17, 60). Virtus is the essence of all that shows man in his best and noblest light; and it brings into special prominence his bravery ("melius est virtute jus; nam saepe virtutem mali nanciscuntur," Enn. Fr. 223, v). Later it comes to mean uprightness in general. The Romance languages have adopted only the latter meaning (Fr. la vertM, It. virtu, Span, virtud). The Greek, on the other hand, held moderation, or o-wippoo-uk^, as the highest virtue. The maxim \*.r\Sky olyocv was attri- buted to one of the seven wise men; and it appeared in the Pronaos of the temple of Apollo at Delphi side by side with the caution '* know thyself" (yvuit (TiOCVTOy), The corresponding Roman word [modestid] has received its colouring from Roman ideas and has come to signify political loyalty, while in its military usage it means a sense of discipline. Bonus denotes in a legal sense a man of honour, and in a political sense a patriot; /ortis unites in old Latin the two meanings of brave and noble (cf. Plaut. Trin. v, 2, 9, and O. Hey, " Semasiolog. Studien," Leipzig, 1 89 1, p. 1 14); mot/is has a more or less depreciatory connotation, for constancy and rigour are the qual- ities prized. The Romans call an impudent person a novelty, or, as we should say, 2l freak {insolens). It is significant too that Cicero renders the Greek word KaXov, beautiful, or morally good, by honestum, hon- ourable (cf. Cic. De Off. ed, Heine, p. 23). The Greeks, then, look at morality from an aes- OF THE ROMAN PEOPLE 29 thetic point of view, and thereby betray their artistic appreciation of this virtue. The Romans, on the other hand, think first and foremost of the impres- sion likely to be made on others by moral actions, and they show their full sense of the value of such impression : Honour brings honourable posts (*' Hon- estum fert honores"). The temples of Virtus and of Honos stood side by side in Rome, and in fact after the victory over the Cimbri the two deities were united in one temple. 16. The pride of self-consciousness is manifested in the Roman denomination of the Mediterranean as mare nostrum. And indeed the sea which had once been swayed by Phoenicians, Greeks, Carth- aginians, and Etrurians, and which had assumed Greek names even for the portions adjoining Italy (such as Tyrrhenean and Ionic Sea), had passed into Roman possession from Cyprus to the pillars of Hercules. The British, who rule the Ocean and despise other European nations, express the Latin ego by I, always expressed in capital letters: can we then grudo^e the Roman this mark of his self-com- placency ? The pious Israelite in sign of greeting cries "Peace be with thee!" the merry Greek shouts Xar/)£, rejoice! The Roman regards health and strength as the prime necessities of life: hence his greeting is vale! and salve! " Bide ye strong and bide ye healthy ! " Names of measures of length, which in Greek are often taken from recreation grounds and from 30 LANGUAGE AND CHARACTER sports (cf. ardiiov, SoX^x"^, Irririxov, etc.), are formed in a much less imaginative fashion by the Romans, who reproduce in language merely the number of the feet {duo millia, i.e., passuum, etc.). Similarly Roman coins are named according to the sum of the asses which they contain : thus sestertius = sem-is-tertius, lit. the third half of an as, i.c, 2\ asses, denarius = deni asses. We may contrast with these names such Greek words as i(^oXo^ [probably = copper nails used as money] cTpaxi^r, ht. a handful, xaAakTov, a weight. 1 7. We may now consider the methods used by the Roman to denominate the months of year. Many of them he simply denotes numerically, as September, October, November', and we know that in addition to these there were originally a Quintilis and a Sextilis, whose names were changed in honour of Julius Caesar and Augustus. A significant con- trast to such names is seen in the old German and Greek nomenclatures. The Roman method reminds us of the American s method of simply numbering the streets of his towns instead of naming them after distinguished persons or accidents of situation. We need not be surprised at finding that the Romans apply the same numerical method of nomen- clature to their system of proper names, with the result that many of these remind us of a numbered exhibition catalogue, e.g., Quintus. Sextus, Septimus, Octavus, Decimus, and again the names Sextms, Sep- timius, Octavius, Nonius, Decitis. There are yet other conclusions to be drawn from a scrutiny of Roman proper names. Originally they OF THE ROMAN PEOPLE 31 were appellatives, and they bring out some marked taste on the part of the parents who conferred the name, more commonly a wish expressed by these for the future of their child. And, indeed, it seems quite natural that parents should wish to endow their off- spring with some name expressive of the quality which would be most useful to them throughout their whole life. Our Teutonic ancestors, who combined passionate love for quarrels and fighting with a deeply religious spirit, manifested in the names which they gave their children the feelings which animated their own hearts: hence a large proportion of German [and of English] names re- call memories of war cries and bellicose sounds; others again reveal what our forefathers regarded as the ideals of life, such as Prudence, Force, Wealth, Constancy, Courage, and Daring [cf. in English such names as Wise, Good, Strong, Richard, Steel, Dare, etc.]. Greek names likewise denote such noble and sublime qualities as youth may fitly imitate; they contain ideas of Glory, Valour, skill in wielding weapons, or again, of political influence: most of these end in -xArf?, i.e., xAfo?, glory, or begin with Kauto-, KAfo-. Names of this kind are comparatively rare among the Romans : on the other hand their taste for agriculture and for cattle-breeding comes out strongly in their nomenclature. Pliny the Elder has remarked (Nat. Hist, xviii, 3) that Fabius means Beanman, Lentulus, Lentilman, Piso and Cicero, Peaman (from pistim and cicer respectively), and the gentile names of the Porcii, Asinii, Vitellii, Caninii, Caprarii, Ovidii, Ovinii (cf. also Taurus, ^ 32 LANGUAGE AND CHARACTER AseUio, Bubulcus, etc.), all seem taken from the names of domestic animals.* Besides this, there are in Latin many more proper names derived from bodily peculiarities, such as infirmities of any kind, or the colour of the hair, than we find to be the case with either Greeks or Teutons. A whole series of gentes or clans bears the names of colours: Albtt, Rufii, Rutilii, Flavii, Livii, Caesii, Fulvii. Nigtdtt, etc.: then there are proper names like Plancus (Broadfoot), Plotus, Pedo, Peducaeus (Flat foot), Scaurus, Varus, Varro, Valgius (Crooked leg), Clau- dius (Lame), Flaccus{S\3.ck), Sulla from sura, surula (Small calf), Capita (Great head), Pronto (Great hxo^),Mento (Chin-man), iVa;^^ (Nosey), Silo (Snub nose), Labeo (Big lip), Bucco (Big mouth), Dentio (Big tooth), Barbo (Big beard), Balbzis (Stutterer), Turpio (Ugly man),t Lurco (Glutton), Strabo, Paetus (Squinter), Calviis and Glabrio (Bald head), Crisptis (Curly head), Crassus (Thickman), Tubero (Crook- back), Naevius (Warty), Stolo% (Dullman), etc. [so too Brutus] (cf. Horace, Sat. i, 3, 44)- Such names as these (and more might be added) show the delight manifested by the Romans in marking and pillorying bodily defects, and how they loved twitting each other and holding each other up to ridicule. All the proper names cited above are, in * Macrobius, Sat. i, 6, ad fin-, explains the origin of the Roman names Scrofa and Asina. t Plautus, Most. 4, 2, I, coins a nickname i?«ftb— rope-man, i.e., gallow's-bird— a parody on such names. + Stolo is properly " a stock," and Varro (R. R. 1, 2, 9) plays upon his name. OF THE ROMAN PEOPLE 33 fact, nicknames, and exemplify the *' Italum ace- tum" (Horace, Sat. i, 7, 32). The Romans were, in fact, at once coloni and ** clowns," like the English of old.* We conceive a higher idea of the Roman imagina- tion as evinced in its nomenclature when we turn to the list of stately agnomina conferred on victorious generals, AfricanuSy Asiaticus, Numantmus, Numi- dicus, Creticus, etc. The names were, of course, taken from the name of the country in which they had gained their renown. They testify at once to the deep gratitude borne by the Romans to those who had succeeded in bringing great wars to a happy conclusion, and to the pride and respect with which they uttered the names of such heroes. And this custom harmonizes with the Roman habit of select- ing the most impressive method possible of celebrat- ing great occasions in Roman national life, methods which could not fail to strike the imagination of the beholders, such as the triumphal processions, and the ceremonies observed in a declaration of war. With the Greeks, whose highest ambition was to win an olive crown in the Olympic games, we find nothing of the kind. Modern civilized nations have, however, in many cases copied the Roman usage: cf. Blucher von Wahlstatt, York von Wartenburg, Lannes, due de Montebello, Massena, due de Rivoli, Diebitsch Sabalkansky, Pasjewitsch Eriwansky [Lord Napier of Magdala, Lord Dufferin of Ava, Lord Kitchener of Khartoum]. * Coloni and clowns are not etymologically connected; the latter word is probably of Scandinavian origin. See Skeat, s.v. D n D 4 LANGUAGE AND CHARACTER 1 8. What do we learn from the names of the gods ? For they, too, throw light on the thoughts and genius of the people who worship them. Myth- ology is the product of popular imagination ; it is closely bound up with the spirit of the people, and thus serves as an index to their profoundest thoughts. In the legends of the Hellenic deities do we not see mirrored the brightness of the Greek skies, and the graceful charm that was the prerogative of the Hellenic nation? In old German myths do we not see a reproduction of the seriousness and the melan- choly of Northern races? But besides these general traits, our interest is further challenged by the changes undergone by separate ancient deities, as modified by the character of each nation among whom their cult has prevailed. It is highly charac- teristic of the mental attitude of our Teutonic fore- fathers that they should have taken the highest deity of the Indo-Germanic primitive epoch— the ZfuV of the Greeks and J ovis-pater= Jupiter of the Romans, and, under the name of Tin connected him with war, and made him their war god. Side by side with him, the Franks first, and shortly after them the other German tribes, revered Wotan, the wind god, the representative of the cloud-covered Heavens, and of the raging storms (O.H.G. Wuotan is connected with N.H.G. ;^«^ = wrath). Thus the dispenser of the radiant light which spread over Italy and Greece had to give place to the god of the northern cloudy sky: but at his side sat his sister and spouse, Freia, the loving and kindly mother of the gods. Thus the names of Tiu and of Freia liVf OF THE ROMAN PEOPLE 35 i\ represent the twofold aspect of the Teutonic nature: the mood for battle and the mood for profundity and earnestness; the two moods which we have already observed to be denoted by their personal names. In primitive times the Romans held their most important deities to be the agricultural god Saturnus, protector of crops {sala) and the war god Mars [Sabine Marners]. A number of Italian names of tribes and places were taken from Mars : such as the Marsi, Marrucini, Mamertini, Marruvium, etc. The first month in the Roman year, the mensis Martins, takes its name from this god : and Mars is dignified with the same honourable title as Jove himself (Marspiter), in fact, his name is in common metonymic use for bellum, as in the phrases aequo Marie, suo Marte, etc. Originally, like most of the Aryan deities, a god of light* he was metamor- phosed into a war god by a warlike people. Saturn was not identified with Kpovo? until the influence of Greek culture began to make itself felt in Italy. After this identification he is revered as the father of Jupiter, now raised again to the highest seat of power: Mars, on the other hand, appears as his son, just as Tiu appears as the son of Wotan and Freia. Saturnus owns Ops as his consort, the goddess of agricultural prosperity and agricultural industry (cf optisc\ whence too the Osci = Opsci, rural workers, take their name). Side by side with these we find * Cf. ^apiLLaipio. Very probably, however, the name Mars is con- nected with iidpva^ai. See Walde, s.v. Mars. t Opus is, however, probably unconnected with ops: opus = Sanskrit dpnas, wealth; ops = Sanskrit dpas, worth. ill 36 LANGUAGE AND CHARACTER in earliest antiquity a numerous company of other agricultural deities, such as Ceres, the deity who presides over Cerealia; Flora, the flower goddess; Maia, deity of the Spring to whom the barrow-pig {Maialis) was consecrated and sacrificed,* Tellus, the god of the fruit-bearing earth, Faunus (the favouring deity, ixomfavere\ the protector of herds worshipped as the wolf-scarer under the name of Lupercus {lupos arcens\ Pales, the tutelary deity of shepherds and cattle, Terminus, the god of bound- aries, and Pomona, who produces fruit in its season. But when we look for ancient Roman gods of the sea and of rivers, we look in vain. The sea and all its wonders have no attraction for the Romans, and hence it comes to pass that the deities of the river and sea are of Grecian or Etrurian origin, or at any rate they have taken their rise under the influence of these nations. Neptunus, the ancient Roman cloud-god, suffered his transformation under Hel- lenic influences, t The Greeks, then, by the aid of their lively ima- gination and their refined aesthetic sense, created tangible and palpable images of their own deities. The soberer imagination of the Romans contented itself with mere abstractions, and their creations were lifeless by comparison. \ On the other hand, ♦ Maia and maialis are, however, only connected by popular etymology. See Macrob. Sat. i, 12. t Macrobius, Sat. i, 17, tells us that Neptune was called both i^oerfveo^v and a^^aXc'a>v^at once earth's shaker and pacifier- epithets more appropriate to a deity of the sky than of the sea^ X Cf. Macrob. Sat. i, 7- ''Antevorta et Postvoria apud Ro- manos coluntur." So Forrima, Ov. Fast, i, 633. OF THE ROMAN PEOPLE Z7 % they feel it their bounden duty, owing to their con- scientious scrupulousness in religious observances, to set apart special divinities to preside over every possible manifestation of human activity. The countryman, on first ploughing up the soil, invoked the Vervactor: at his second ploughing, th^Redara- tor: on drawing the furrows, xh^Imporcitor: on sow- ing, the Insitor: on commencing to cross plough, the Obarator: to harrow, the Occator: to weed, the Sarritor: to trench, the Subruncator: to mow, the Messor: to bind the sheaves, the Connector: to store in granaries, the Conditor, and so on. The Roman people impressed even on their deified virtues and qualities that practical character which appears in their moral views. 19. Another important criterion of the connection between language and national character consists in proverbial expressions and '* winged words." Goethe said : " Proverbs mark nations, but these nations must have been their home." And it is a fact that none can appreciate the close relationship between a nation's humour and its proverbs but one who has had his finger on that nation's pulse, and is suffi- ciently familiar with its thoughts and feelings. Proverbs touch every side of popular humour : they disclose to us its attitude towards the animal world, to nature, and to all objects which recall primitive times and the childish simplicity of view of primitive people. They give us a purview of a nation's pro- cess in culture, and enable us to realize how it judges of its neighbours and of its progenitors. Thus it is HI 38 LANGUAGE AND CHARACTER interesting to observe how prone the Romans were to hold up to ridicule the prominent characteristics and the disagreeable traits of foreign nationalities, with whom commercial or other dealings brought them into contact. It is equally instructive to note how eager these same Romans were to magnify the glorious deeds of their own ancestors.* If we take proverbial phrases (for such they have become) like Punica fides (treachery), Gallorum credulitas, Cam- panorum arrogatttia,\ we are able to recognize not merely that the bad qualities referred to were be- lieved by the Romans to be inherent in those na- tions.t but further that the nations thus stigmatized were from the earliest times strangers to the Romans, and were regarded by them with little sympathy. In contrast to such phrases stands " more Romano " or " Latine loqui " (cf. Cic. Phil. 8, 6, and Wolfflin's "Archiv.," iii, 376a). This phrase signifies to speak out truly and plainly, and it is not hard to parallel in modern times. In German and English alike, if we desire to insist on an unpalatable truth, we commonly say, " to speak in good plain German " or " English " as the case may be.§ Again, we know that the Greeks used to drink out of larger wine-cups than the Romans. Accord- * Cf Macrob. Sat. ii, x, " Vetustas quidem nobis semper, si sapimus, adoranda est. lUa quippe saecula sunt, quae hoc impe- rium vel sanguine vel sudore pepererunt." t Cf. Cic. In Pisonem, 11, "buccae dignae Capua." t Cf. Macrob. Sat. i, Introd. " ' Sum ' inquit, ' homo Romanus, natus in Latio, et eloquium Graecum a nobis ahenissimum est § Other such popular maxims were, "crassa Minerva"; "hoc age "; "leges bonae ex malis moribus procreantur." OF THE ROMAN PEOPLE 39 ingly we find not merely the expressions Graeco more bibere, pergraecari {i.e., " maioribus poculis bi- bere "), but we have the Latin expression for " be- tween the lip and the cup"— "Inter os et offam" (Gell. xiii, 18, i), as contrasted with the Greek iroxxoi, (A-tTxli TrlXf. jtu'x.xoc xal x"''^"^ "^"P""- ^"^ where the Germans say of a person of whom nothing can ever be made, " All the hops and malt in the world can make nothing of him," the Romans say " operam et oleum perdidi,"* a metaphor taken from the gladiatorial schools. Moreover, in the spirit with which the Romans mention disastrous episodes in their country's history, and the names of their national heroes, we may note a great difference from that of the Greeks. The Romans take such events as th^pugna Osculana, Cannensis, etc., as stepping- stones ia their history, and for their national heroes tliey adopt Romulus f and Remus, Camillus, [Cethegus], Curius Dentatus, Fabricius, the rigorous moralist. Cato, and Fabius Maximus, the hero who - Cunctxndo restituit rem." The Hellenes prefer to cite the names of those of their countrymen who have distinguished themselves in science and art, as stock examples of those whom it is their delight to re- member with honour. Perhaps it has also some significance that, among all the Roman gods and heroes, none enters so frequently into proverbial expressions as the puissant figure of Hercules {Ex * "Tis labour lost," or, as the vulgar proverb has it, "You cannot make a silk purse out of a sow*s ear." t Cf. Macrob. Sat. ii, 17 "(Romuli) vita virtutes nunquam deseruit." I 40 LANGUAGE AND CHARACTER pede Herculem, mekercule, Herculi quaestum con- terere, Plaut. Most. 4, 2, 68]. 20. And now for the "winged words" or dicta which have passed into maxims. We may for our purpose disregard all such as owe their origin to Greek cul- ture and passed from the mouths of educated and influential families into the speech of the people, such as Circaeum poculum, Alcinoi dapes, etc. [" Epicuri porcus"].* There remains a large remnant of regular Roman dicta which were the genuine output of Roman feeling simply because they were the expression of the heart of the people. This holds true not merely of the characteristic utterances of old Cato, and of the still more ancient Appms Claudius, but of many epigrammatic sayings of later authors. What phrase reflects more accurately the genuine view of a Roman than the well-known " Fortes fortuna adiuvat"? And hence it comes to pass that no phrase in all Roman literature occurs, with its variants, so frequently as this. From ' Ennius and Terence down to Lucan and C audian, we find Roman authors ringing the same changes.! And could any words more truly reflect the com- placent haughtiness of the Roman character than the exclamation of Atreus in Accius (203 Ribbeck), " Oderint, dum metuant! " We are not surprised to find that it is so often harped on and cited. We * Cf. "Non cuivis homini contingit adire Corinthum," Hor. Ep. i, 17, 36. , , „• J t Cf. " Romani audendo, et fallendo, et bella .ix belhs serendo, magni facti," Tac. Hist, ii, 71. OF THE ROMAN PEOPLE 41 i meet with it no less than three times in Cicero (Phil, i, 14, 34; De Off. i, 28-97; Pro Sestio, 48, 102); Caligula frequently quoted it, as we are told ' by his biographer (Suet. Calig. 30) ; and we may gather from the pages of the gentle, nay almost Christian, Seneca, that the phrase had, even in later times, not lost its power of fanning the glow of martial ardour in Roman hearts. That philo- sopher employs it several times (De Ira, i, 20, 4; De dementia, i, 12, 3, and ii, 2, 2), and he couples it with the remark: " Illud mecum considero multas voces magnas sed detestabiles in vitam humanam pervenisse celebresque vulgo ferri, ut illam : oderint, dum metuant." Again, Cicero's remark "Silent leges inter arma " notoriously passed into the com- mon stock of the entire nation: Quintilian (v, 14, 1 7) and Lucan (i, 277) refer to it in their works. 21. We can hardly be surprised that a nation in whom intelligence was so strongly developed as it was in the Romans should have manifested a great predilection for playing upon words. This tendency shows itself at every period of Roman literature, more particularly in the comic poets and orators, but also in the epic and lyric poets. Plautus, Cicero, and Ovid are inexhaustible in their store of puns.* Each writer seizes on any occasion for introducing such: indeed, not infrequently, the same pun is employed to satiety. We may remember the * Macrobius has six chapters on Roman jokes and puns. Sat. ii, cap. I sgq. JuUa, the daughter of Augustus, was particularly noted for her smart sayings. i II 42 LANGUAGE AND CHARACTER laborious frequency of play on the word Verves in Cicero's Verrine orations * and his tedious juggling with the double meanings of such names as Brutus, Balbus, Lepidus, etc., in his letters to Atticus [cf. ** aureum nomen Chrysogoni "]. It may be, too, that many verbal quips occur in literature which have escaped our notice from insufficient knowledge of historical occurrences, f Cicero, in his orations, strains after this method of producing effect, that he may tickle the jaded ears of his audience. In this respect he forms a decided contrast to the Greek orator Demosthenes, with whom verbal echoes like paVa? — p>Ou/xnk (01. i, 13) are comparatively rare. Besides this we hear from Plutarch and Quintilian that Cicero employed witticisms in his ordinary con- versation to an even greater extent than in his writings (cf Herwig, ** Das Wortspiel in Cicero's Reden," Attendorn, 1889). 22. The syntax of a language, no less than the signification of the words, carries the mark of the spirit of the people. A masculine and vigorous tone characterizes the construction of Latin sen- tences — an energizing breath of logical consecution — which marks the Latin language as a fit vehicle for oratory, more particularly for speeches spoken by the accusing counsel, and for the historian of cam- paigns, but as a less suitable medium for lyrical expression. No one was more conscious of this * Cf. In Verr. iii, § 46, ad fin. t Tibullus calls his first love Delia, from B^Xoe, her real name being Fiania. See Postgate's "Tibullus," Introduction, p. xx. OF THE ROMAN PEOPLE 43 defect than the Romans themselves. Quintilian, for example, declares his conviction that it is impossible for Latin writers to attain to Hellenic grace and attractiveness (xii, 10, 36). ^' Non possumus esse tam graciles, simus fortiores: subtilitate vincimur, valeamus pondere." And if it be granted that Cicero succeeded in rendering the Latin language more flexible, by modelling it on the Greek, it must also be noted that such transformation was only partially possible : a complete revolution in the genius of the language would only have been pos- sible by an absolutely new creation and a radical revolution in the genius of the people. Cicero's followers, too, lag behind their master in grace of style. The truth was, that in order to ensure the growth of the new graft by which Cicero wished to improve the stock of the mother tongue, one neces- sary condition was absent: the Roman remained always a Roman, and could never belie his nature: '^ Naturam expellas furca, tamen usque recurret.'' 23. The first feature that strikes us in the ar- rangement of the Latin sentence is the energy and dectsion, the virility and the dignity which radiate from its very form. There is hardly any trace of affectation or literary refinement. The periods sue- ceed each other with dignity and in well marked cadence— spirited and irresistible like the Roman legionary. Their entire colouring recalls to us the picture of his weather-beaten face, and their stately march reminds us of his proud and masterful bearing. In fact, this well-matched pair, warrior and language, 111 11 11 J 44 LANGUAGE AND CHARACTER have stepped forth from their home in the full con- sciousness of victory, and have overcome the v^orld between them. Where pathos is demanded, the style of the Romans corresponds with their love of rhetorical colouring. In consonance with their love of oratory, expressions are unnaturally inflated in places where, according to our taste, simplicity and precision would have been preferred. We cannot then be surprised to find that the language employed often produces the effect of artificial measurement rather than that of simple and unconstrained movement, nor that the phrase, ''poets and prose-writers," should be represented in Latin by ** poetae et ora- tores." The superlative degree plays an important role in the Latin language, not merely in addresses like "viri nobilissimi, amplissimi, ornatissimi," but also when placed in apposition to proper names, e,g., " Corinthus, urbs opulentissima."^ Not infrequently we find the " Futurum exactum " taking the place of the simple future. The standard-bearer of the tenth legion, on the occasion of Caesar s landing in Britain, exclaims (Caes. B. G. iv, 25): " Desilite miUtes, nisi vultis aquilam hostibus prodere: ego certe meum reipublicae atque imperatori officium praestitero." The plural is employed instead of the singular to express emphatically and distinctly the strength of any emotion.f This is particularly remarkable in the case of abstract words. * Cf. "vir fortissimus, Piso Aquitanus," Cic. Verr. 4, 16. t Cf. inimicitiae, repeated acts of unfriendliness; so furiae, ^tc. Cf. also such uses as esuriiiones^ etc., stccitates diudi domes ticae for ti- OF THE ROMAN PEOPLE 45 24. Another distinctive Roman trait reflected in Latin style is the careful and strict principle of subordination. That force of will, which is so pro- minent in the Roman character, gives rise to a certain stiffness and inflexibility which we admire in T. Manlius Torquatus, and many others of his countrymen ; hence the uncompromising discipline, the stout soldierly spirit, and the unequalled obedi- ence to orders, which characterize the Romans. It is not without significance that Cicero employs the word velle to express the views and opinions of his ancestors with respect to what they deemed the welfare of the State {e.g.y Cic. De Off", iii, 31, iii; Pro Lege Man. 11, 39). Here again the Roman attitude is in strong contrast with the Greek, an attitude which Mommsen characterizes as follows: " The Greek sacrificed the whole to the individual : the nation to the commune: the commune to the individual burgess. The Greek's first proceeding, dictated by his religious views, was to create human beings out of his gods; he then proceeded to deny their existence : the Roman kept his son in the awe of the father: the citizen in the awe of the ruler, and kept every one in the fear of the gods. To the Roman the State was all in all, and the only lofty idea not proscribed to him was the enlargement of the State. The will of the all-powerful capital decided the destinies of the provinces : every one in the Empire who desired a wide culture, a political post, or fame and distinction, turned his gaze on tudines, cases of heroism in civil life, Cic. De Off. i, 7^: <^o^- scientiae^y^iioks of conscience; spumae, masses of foam. 46 LANGUAGE AND CHARACTER Rome. Centralization was carried out as completely under the sway of Rome as under our neighbours, the French. In the same way, it was Rome which was responsible for the formation of the literary lan- guage : the capital of the empire was also the central point of literary activity." The principle of subordination runs through all the arrangement of sentences and words in classical Latin, and is applied much more widely and com- pletely than in any other of the Indo-Germanic languages.* Even the Latin poets are not averse to long sentences, ^.^., Lucretius, i, 930-50; andCatullus, in the commencement of his poem on Berenice, employs a lengthy and unbroken period. The Roman writer likes to make his main thought stand out in relief by duly subordinating the less important clauses of the sentence ; and this not in- frequently in cases where the Greek, the German, and the Englishman would prefer to employ co- ordinate sentences. In the place of such particles as - indeed . . . but," ''and so," "and hence," and the Greek (xlv,.. (Te, we find, as a rule, subordinate clauses, denoting time, cause, concession. F. A. Krummacher has engaged -in some rather recondite speculations on the'words " and," '* but," as used by the Hebrews and Greeks respectively, and has endeavoured to show the relations of these words to the intellectual life of those two nations. No doubt he pushes these speculations too far, and he reads into these two little words more than they really contain; but it can hardly be disputed that they are characteristic * Cf. Zielinski, " Our Debt to Antiquity," lecture i, ad pi. OF THE ROMAN PEOPLE 47 of the genius of the two races. The emotional and sensitive Hebrew thought as a child and acted as a child, and his language, with its quaint and naive expressions, was the language of childhood. The imaginative Greek ,Vield it his first duty to render his language plastic and the mirror of his thouorhts. But the Latin.iis of another cast. In every trait of that languac^'e we catch the tendency to subordina- tion. Tho/method of connecting sentences by means of relative clauses (and this method occurs no less than t^iree hundred and eighty times in Caesar's *' De Belljo Gallico" and ''De Bello Civili") gives ex- pj"G/ssion to this tendency, and the Latin disposal of it^i moods does so in a yet higher degree. The Ljatin usage contrasts with that of German, Greek, ^fnd English, in the fact that it has developed Tadually, in place of the Indicative usual in asser- :ions, the dependent method of speech (conjunctive) Isimply with the idea of bringing the subordination of such dependent clauses more into prominence, and to show by this method that the subordinate clause represents the thought of the speaker, who is regarded as the subject. In sentences denoting sequence, and in sentences with the historic, or causal, or concessive cum, which in older stages of Latinity are not uncommonly em- ployed to denote a fact as having actually occurred, in ordinary Latin, the dependent form of the sen- tence has come to be the usual type. This usage is still seen in the case of qMoniam = quom iam = cum iam : as, indeed, it is still seen after the Greek con- junctions wo-Tf and Wti^ and after such German copu- 48 LANGUAGE AND CHARACTER lative words as so dass and ah (cf. English '* so that"). These conjunctions can all be used to intro- duce actual facts. The same holds good of indirect interrogative sentences. Indoed, after Livy's time, this usage took a wider range and spread even to such words as priusquant and to dtmi, quamquam, etc.; strictly speaking, words introducing simple narration, without any clear reason appearing in the sentences for the point of such usage. Again, cum iterative is, before Livy's time, seldom co'mnected with the conjunctive mood, but by him it Ss fre- quently so connected. Cf. xxi, 28, 10; xxxiiti, 3, 10. 25. This unmistakable note of discipline and suib- ordination manifests itself in the orderly way I'n which the Romans carry out the sequence of thei r tenses, all dependent tenses being subordinated tc ) the main clause: and it again comes out in the pre-^^ ference shown by Latin for dependent speech {pratio y obliqua\ in which sentence after sentence, and clause \ after clause, are set under the strict rigimeoi a single ^ governing verb {dixit, respondit, etc.), as soldiers \ under that of a general. Here, again, we have a con- | trast between Latin and Greek. Just as soldiers in a regiment keep their eyes fixed on their commander, all the pronouns in oratio obliqua which have re- ference to the speaker look back to him. Add to this the marked and energetic accent which doubt- less aided to invest Latin with its virile and almost defiant qualities, and we shall understand what Heine said : " The language of the Romans can never belie \ OF THE ROMAN PEOPLE 49 its origin." It is the language for commanders in the field, for administrators in their decretals, the legal language for usurers, the language for the in- scriptions on the adamantine Roman people (Heine, "Gesammelte Werke," v, 144). Again, in the management of the Latin accent, the law of subordination is well marked. In classical Latin we must suppose that the main stress-accent fell on the verb. As the verb was in most cases shifted to the end of the sentence this accent too gradually passed to the end of the sentence, and the series of unaccented or weakly accented words pre- pared the way for the accented or stressed expres- sion, as effectually as the lictors who preceded him prepared the way for consul or dictator. 26. Another sign of the practical turn of mind, and clear mental vision of the Romans, is found in their marked preference for concrete expression. The Germans (and, in a lesser degree, the English) prefer to soar in abstractions. The Roman, on the contrary, is a realist: he prefers to take a positive and actual instance to a general conception. We have only to think of such expressions as '' urbe capta," after the taking of the city; " prudentis est," one needs prudence; " alicui hortanti parere," to obey some one's exhortations: "verum dicere," to speak the truth; " ex aliquo quaerere quid sentiat," to ask some one's opinion; " clamor admirantium," a shout of admiration — and we shall find ample confirmation of this statement (cf. also such expressions as "inter- fectus Caesar " for " the murder of Caesar "; '' stans £ 50 LANGUAGE AND CHARACTER Caesar "—Lucan, for "Caesar brought to a halt"). The liveliness of representation and of feeling which comes out of such Greek constructions as '- These diminutives have more or less divested themselves of their nature, and their diminished vitality is shown in the neuter gender. Latin, on the contrary, exhibits neither the same freedom in its treatment of gender, nor the same delicacy of discrimination, for it passes on the gender of the original noun to its derivative diminutive, as /tier, libelbis\ silva, silvtda. 32. Finally, we must briefly examine the vowel conditions of Latin. W. von Humboldt long since I 6o LANGUAGE AND CHARACTER insisted on the fact that the vowel system in any language must stand in close relationship to the trend of the national taste of those who speak it, reflecting, as it does, the mental power of the human organism in its entirety.* This principle comes out very clearly in a comparison of the modern North and South European languages. In the German and, more particularly, in the Slavic sound-system, the consonants play a much more prominent part than in the Romance languages, which, however, are distinguished by greater variety in their vowel sounds. Thanks to this cause, Italian, for instance, is endowed with its incomparable grace and delicacy (cf. Byron, ** Beppo," 44). The language viewed as the artistic creation of an entire people reflects the fact that the Italians possess a remarkable sense of form, a sense which stands adequately revealed in other directions, such as the fine arts, painting and music, poetry and architecture. No one can deny that the northern nations stand in this respect far behind their southern neigh- bours. Latin holds a middle position between the rich vowel system and liquid sweetness of the Italian, and the consonantal agglomerations of the Russian language. In its position with regard to these it resembles rather the German written language than the Greek, and indeed it shares with German certain peculiarities in its sound-changes. Friedrich August Wolf said long since: ** The Latin language is far * See Byrne, " Principles and Structure of Language " vol. i, p. 12. OF THE ROMAN PEOPLE 61 from possessing the harmony of the Greek. It is military, stern, and stately. Its numerous consonants, and the paucity of its vowels, give it a hard repel- lent look, and are indeed characteristic of the nation"; and Fr. Scerbo gives as his opinion {'' Caratteristiche del Greco e del Latino," Firenze, 1893, P- 0* ** II vocalismo greco e piu ricco, piii delicato e vario, ritraente piu la freschezza e lagilita dei suoni primitivi; il vocalismo latino ci appare meno armonioso e snello od integro ed un po piu incerto." Lastly, W. von Humboldt gives as his opinion ('* Uber die Verschiedenheit des mensch- lichen Sprachbaus, herausgegeben v. Pott," ii, 232, Berlin, 1876) that '* in the language of the Romans no luxuriant variety, no freedom of imagination, has been wasted in the formation of sounds; the virile, earnest sense of that people which regarded rather the truth of things as they are, and craved only so much of thingfs intellectual as consorted with such truth, had no room for any such luxuriance or any such free upgrowth of sounds." Just as the Greeks were the masters of the Romans in sculpture, archi- tecture, painting, and music, in short, in all arts, so they display in their language, full as it is of the magic of their harmony, more feeling for formal beauty, and for pleasing and melodious tone effects. Hence it is that the Greeks possess such a strong taste for assonance and the correspondence of vowel sounds, while the ear and heart of the Romans were far more open to impressions from consonantal alliteration. This alliteration gives the verse a char- acteristic ring of its own, rather than a melody; it 62 LANGUAGE AND CHARACTER renders it not indeed more graceful, but stronger and more forceful.* The Romans, like our own ancestors, ranked character above beauty, essence above form. Old Roman versification — the Sa- turnian, for example — was full of alliteration,! and there are many old formulas, depending on this trick of language, which have maintained themselves through all the life of the Latin tongue; such as "purus putus," **sane sarteque " J [" locutio ex auguralibus sumpta "]. Such go to confirm the idea that the Romans regarded alliteration as an ancient national trait of the technique of their poetry; and thus it is that Vergil in his '* yEneid," that sustained eulogium of the national virtues of the Romans, has employed it to so large an extent. 33. We have exhausted our remarks on this sub- ject. We trust that it has been made plain that Latin contrasts with Greek in many essential points, and that this contrast depends for the most part on the difference in the national character of the two races. As they differ in thought and in action, so do they markedly differ in diction and in style. If it were necessary to cite in support of our conten- * It is well known that the most salient feature in Anglo-Saxon literature was its regular alliteration, and this holds good generally of the old Northern or Icelandic. Cf. Marsh, "Student's English Literature," p. 389 sqq., who gives many instances of its use by modern English poets. It may be worth noticing that Milton and the Classic School of poets generally avoid alliteration altogether; cf. "Alliterative Poems," Morris and Skeat, Part II, xiii. t E.g.^ " eorum sectam sequontur multi mortales " (Naevius). X Cf. our kith and kin, health and happiness, etc. OF THE ROMAN PEOPLE 63 tion that where the contrary cause holds good the contrary effect follows, we might easily show that peoples which share many prominent traits of char- acter, manifest also a great resemblance in their speech. We might cite as an instance of this the intellectual relationship between the Spartans and the Romans. Of all the Grecian stock, none was in this respect so nearly related to the Romans as the Spartans. Both nations were alike adepts at manual labour, and proud of their powers. Both were strict disciplinarians; both were weak in cavalry, and both alike had an aversion to a sea-faring life. On the other hand, both had a genius for jurisprudence and political activity ; in both we find two characteristics strongly brought out — great reverence for old age, and the lofty position assigned to woman. On the other hand, in artistic capability and in scientific attainments, both nations alike stand behind the other Greek races. We find, accordingly, in the languages of the two nations a number of similar traits : a lack of flexibility in the formation of com- pounds, a poverty of words, a stiff and formal rhythm, a logical acuteness, an endeavour after pregnancy of utterance (Cic. Ad Fam. ii, 25, 2), a taste for brief and neat witticisms (O. Muller, *' Dorians," ii, 385 sqq), especially for puns, a taste which comes from a fortunate trait of whimsical humour common to both: we also find in both less mobility in their vowel sounds, and a greater ad- herence to the old traditional form of the termina- tions of verbs. The traits and features of the language on which 64 LANGUAGE AND CHARACTER, ETC. we have touched are, in themselves no doubt, unim- portant enough, but " straws show how the wind blows." Just as we are enabled to understand the real character of a man through the trifling in- cidents of daily life, so the tiny stones that we have set together form, in their entirety, a faithful mosaic of old Roman action and deeds, poetry and thought. They thus permit us to appreciate more than super- ficially the salient traits of the Roman character; and, what is more, they enable us to take an intelli- gent view of the monuments of Roman art and of outstanding events in Roman history. Rlickert then is right when he says '' the science of language is the subject which of all the circle of the Sciences affords us the most satisfactory revelations about human thought and methods of apprehension.'' it I II ROMAN STYLE AND THE DEVELOPMENT OF ROMAN CULTURE 34 T ANGUAGE is the most faithful companion of -■— ' man on his earthly pilgrimage. The impres- sions of his journey stamp ineffaceable marks in the shape of language, like the annual rings in trees, and thus these recorded impressions indicate to future generations many facts in their past history.* How their ancestors lived and suffered, what they thought and what they felt, their aims and ambitions, all this is revealed by language in eloquent accents to those who can understand. Thus it is that in language we have a real history, and more especially a history of civilization or of culture. The views and prejudices of his time are apt to fasten indelibly on each individual. It falls to none but to a few privi- leged souls to free themselves more or less per- fectly from such prejudices. But even this chosen few, whose names are written in gold on the pages of history, cannot fully escape the influence of the * Cf. Geiger, " Language and its importance in the History of the Development of the Human Race," Triibner, 1880; but more especially Marsh, "Student's English Language," pp. 155 s^^. On /the influence of words on thought see Max Miiller, " Lectures on the Science of Language," vol. ii, pp. 622 s^^. F 66 LANGUAGE AND CHARACTER moral outlook of their age. The mental products even of such giants as these bear, to some degree at least, the markof their epoch. They think they are the propelling force, but they are in fact propelled them- selves : they would fain strike out a new path for the moral course of their age to pursue: but they are forced to run on confined in the old rut of the spirit of their time. Use and wont is a tyrant, whether in things intellectual or things material. Just as the architecture of any given age reflects the concep- tions of the generation of its builders, so the style of different authors sharply and clearly exhibits the traits of contemporary thought. It is no uninterest- ing task to follow the reciprocal relations between style and moral outlook by watching the develop- ment of the Latin language for several centuries. 35. QuintiliansaysofEnnius: ''Enniumsicutsacros vetustate lucos adoremus, in quibus grandia et an- tiqua robora jamnontantam habent speciem,quantam relio^ionem." His remark holds true of all the more ancient Roman literature. Indeed, the style of the old Roman writers does resemble the oak in its tough exterior. Simple, downright, and straightfor- ward was the life and character of the '* prisci Latini," and their expression is accordingly. Affectation and tricks of style are completely absent from their writ- ings, and there is no symptom of straining after effect. Whether they speak, or whether they write, they do either with a definite purpose, and they have little regard to form. Of all the maxims given by the old Cato, the genuine pattern of a genuine Roman, to OF THE ROMAN PEOPLE 67 his son, at his start in life, perhaps the most telling is that contained in the well-known saying: -Rem tene, verba sequentur." Those words of blame are indeed unsparing which the aged Consul Appius ad- dressed to the Senators; and they were totally un- fettered by what are to-day known as parliamentary conventions. "Whither has your sense, so sound and firm of old, senselessly strayed from the straight path?" (Ennius a^. Cic. De Sen. ix, 16). Thus it was that he controlled his people, a king among assembled kings, and the honourable counsellor gave way to the words of the most honourable speaker. Thus was it that he celebrated the highest triumph to be obtained by oratory, an oratory energetic and forceful, yet far enough removed from any artificial claims to embellishment, nor indeed was there any public at that time capable of welcoming and critic- izing any such claims. 36. It is almost impossible to think of two more wide contrasts than those which we witness in the early stages of Greek and Roman literature respect- ively. The oldest Greek work is the Homeric poems : the oldest Roman work, the Laws of the Twelve Tables. Differing as these do in the matter which formed the object of instruction of the youth of Greece and Rome, their linguistic contrast is not less striking. Liveliness and perspicuity on the one hand are confronted with sobriety on the other. To take another point of contrast, the old Roman heroes make dry speeches: the words of the old Homeric heroes run glibly from their mouths, fresh as morn- 68 LANGUAGE AND CHARACTER ing dew. It is not without reason that the aged Nestor is described as a Xiyv^ xyopnTYig, a clear-voiced orator: not without reason that the utterances of the Trojan graybeards are likened to the tuneful song of the Cicada, so loved by antiquity (II. iii, 151). Modu- lation and emphasis must at that time have produced the effect afterwards produced by the artificial struc- ture of (rhythmical) periods. In consonance with this we find in Homer the speeches introduced by words like aJJ'ai/, (pcaviTvy (pUyyB(r6oci, etc., which fitly represent the full sounding melody of the old recita- tions: while in Latin the correlative words lo^uz, dicere, fari, have no such delicate connotation. 37. True to the maxim " Naturalia non sunt tur- pia,'* the simple apprehension of primitive Rome took no offence at what was natural. ** To the pure all things are pure," and thus Sisenna, and after him Ennius and Plautus [Livy, Cicero, etc.], say without any misgiving, ** concubia nocte." [Thus again veftter is commonly used for appetitel\ In fact the practice was to adopt the Stoic principle: " suo quamque rem nomine appellare: nihil esse obscoenum, nihil turpe dictu*' (Cic. AdFam. ix, 22). 38. In these old times the difference between the diction of poetry and prose was not yet very marked. The cadence of the old Arval song and that of the **Carmina Saliorum " consorted well with the slow and measured march of the Saturnian measure, as did that of the trampling paces of the Roman legions in Naevius' *' Bellum Punicum." Alliteration and OF THE ROMAN PEOPLE 69 word-repetition, the main factors in poetical tech- nique, were not unknown to prose. Alliteration, which pervades both the Hymns and the tables of the Laws, lent energy and strength to the language, forcing the thews and sinews of its structure to stand out in bold relief, especially in the arrangement of its consonants. This particular device is a very old Indo-Ger- manic method of emphasizing and quickening lan- guage, especially in compressed style. In magic formulae the threefold repetition of a word plays a great part,* and in the popular songs of Germany the refrain is a regular feature. Thus in the song of the Arval brothers every sentence is pronounced with a like number of words, from Ems, Lases, iuvate, down to the concluding word triumpe. Ornamental adjectives are conspicuously lacking to the poetry of that age. The writers have no ap- prehension of tenderer feelings, finer thoughts, or captivating pictures. The structure of the sentence is forceful and compressed, reminding us of the Indian Vedas; but it is clumsy and without grace. The Latin Odyssey of Livius Andronicus is com- pared by Cicero to a stiff piece of wood-carving by Daedalus ; and it is true that the most ancient style of Latin poetry contrasts as strongly with its Greek model as an awkward wooden statue with a master- piece in marble. The prose of Cato, again, in the beginning of the third century b.c, is straightfor- ward and simple, lacking grace and art. We find * See the collection of Triads by Kuno Meyer in the Royal Irish Academy, Todd Lecture Series, vol. xiii, pp. 12 sqq. ^o LANGUAGE AND CHARACTER pleasure in his brief but pregnant sentences: they reflect so completely the character of the man, and indeed of the Romans generally; and they were noticed with approval by Cicero (De Or. ii, 12, 53) and by Sallust (Fr. i, 2) for their ** magna verborum gravitas et sententiarum." No one could have writ- ten more strongly, no one more energetically. The structure of his periods and of his rhythm have not gone beyond the first stages of literature. The sen- tences know no subordination : they are set paratac- tically. The language of feeling and sensibility does not, like that of careful and reflective intelligence, move in lengthy periods, artificially divided and balanced. Each several expression stands apart and is complete in itself; it is blunt enough to serve its purpose: it needs no rounding off, no gradation due to the orderly arrangement of a scrutinizing intelli- gence. 39. The expression is often obscure through the frequent change of subject. Asyndeton, too, which meets us in ancient formulae such as "velitis jubeatis," " patres conscripti," etc., is very common in Cato.* In Fragment 108 he says : '* multa me dehortata sunt hue prodire: anni, aestas, vox, vires, senectus"; Fr. I o I : " exercitum suum pransum, paratum, cohortatum eduxit foras atque instruxit." One may compare with this utterance passages from the old poets, as, for instance, that of N^vius [Bell. Pun. lib.iv,ii, Muller]: " The Roman goes to Malta— he burns the whole * We find asyndeton mounting to climax in later writers, e.g., Pliny, Ep. 9, 22, "in litteris veteres aemulatur exprimit reddit." OF THE ROMAN PEOPLE 71 island, the coast — ravages, lays waste, plunders — foes, property." He has recourse to the " Figura ety- mologica/' ^.^. Fr. 105 : *'cognobilioremcognitionem," and Orat., p. 73, 10: *' vecticulariam vitam vivere" [to live from hand to mouth, lit., to live the life of one who uses a vectisy a robber s instrument].* Then certain turns in his sentences recur frequently, remind- ing us of the '* versus iterati " of the Homeric Epos: e,g,y he uses the three adjectives magnus.ptilcher, and pisculentus in speaking of the Ebro (Fr. no) and also of the Nar (Fr. 97). At the same time he has no objection to massingiwords on words in order to obtain a particular effect: e.g,, in Fr. 95a, a sentence of his *' Oratio Rhodiensis " is reported, in which he brings out several conceptions in this way. ** Scio solere plerisque hominibus rebus secundis atque prolixis atque prosperis animum excellere atque superbiam atque ferociam augescere atque crescere." This peculiarity was noticed even by A. Gellius (Noct. Att. xiii, 25, 13). The passage cited shows also the predilection of the author for the emphatic word atque [ = and what is more], and it is Cato's way to employ such emphatic particles {e.g., verum, eniniy vero). Further, he is at pains to interlard his diction with such archaic words as tuburchinabundus, *' greedily swallowing," and lurchinabundus, " devouring " (cf. Ouintilian, i, 6, 42). Fronto calls these expressions * Asyndeton was common in later writers in animated narra- tion of events happening contemporaneously, as Liv. 3, 37, 7, "hi ferre agere plebem," "These worried and harassed the com- mons. )> 72 LANGUAGE AND CHARACTER ** iligneae nuces " ; they invest Cato's style with a primitive and archaic air. It is no wonder, then, as he so ostentatiously avoids any attempt to copy Greek rhetoric by any graces of style, that Cicero calls his speeches horridulae (Orat. 45, 152). 40. From what has been said we may gather that Cato did not seek to impress either his readers or his hearers with rhetorical embellishments or ortho- dox methods of emphasizing his statements, but relied on the force and vigour of their contents.* He wrote on agriculture and the right conduct of life, and sketched the outline of speeches made by himself, thereby responding to the needs and re- quirements of his time. And his Latinity was in the main the Latinity of his contemporaries : it was the lapidary style of the old inscriptions, unadorned by art and plain to a degree, but full of energy and of old-world strength. ** A good man, my son Mar- cus, can command his speech" (*' vir bonus dicendi peritus"; cf Quintilian, xii, 1,1) were the words of Cato to his son. He meant that a Roman had no need of Greek rhetoric to speak well. 41. For Greek rhetoric had at that time taken deep root in Rome, and had fallen upon no unfertile soil. Indeed, the influence of the Greek spirit had * See Macrob. Sat. i, Praef. for Cato's scomma against Albinus. *' Ne tu, inquit, Aule, nimium nugator es, cum maluisti culpam deprecari, quam culpa vacare. Nam petere veniam solemus aut cum imprudentes erravimus, aut cum noxam imperio compellentes admisimus. Te, inquit, oro, quis perpulit ut id committeres, quod priusquam faceres, peteres uti ignosceretur?" OF THE ROMAN PEOPLE n shown itself increasingly strong since the time of Livius Andronicus, that is to say since the Tarentine war; for it was in the agony of this war that Roman literature entered on its life. Pliny the Elder said several centuries later: '' Ingeniorum Graeciae flatu impellimur," and this saying had been already veri- fied. The contact of the Romans with the Greek colonial settlements of Lower Italy, brought about long before by commerce, entered now, thanks to this war, on a new stage of closer intercourse. The movement which began in the Tarentine war con- tinued in the Punic war. '' Bello Punico secundo Musa pinnato gradu Intulit se bellicosam in Romuli gentem feram" (Porcinus Licinus apudQ^., Noct. Att. xvii, 21). And hence we find Ennius in the beginning of his annals invoking the Greek Muses, and not the native Latin Camenae, to inspire his song. The symptoms of the influence of this Hel- lenic culture were notably manifested in all the departments of life, in art and in science, in trade and in commerce : and they were not slow to mani- fest themselves likewise in literature. For the liter- ature of Rome was — as is commonly the case in the early development of letters — exposed to great varia- tions not merely in its scheme of sounds and flexions, but in that of its periods as well; and if it be true that the idiom of the Capital had for many centuries differed from that of the adjacent communes — as, for instance, from that of Praeneste — still the language in any case needed a ripening and a confirming process. A simple example may serve to show our meaning. When Ennius in his ** Annals'' cites the 74 LANGUAGE AND CHARACTER names of Rome's twelve supreme deities in the two Hexameters: Juno, Vesta, Minerva, Ceres, Diana, Venus, Mars, Mercurius, Jovi(s), Neptunus, Vulcanus, Apollo, we can gather from the varying rhythmical power allotted to the final s, the doubtful pronunciation of the sibilant in his day. In the rest of the words ^ maintains its normal power of making position : but mjovis it is not recognized at all,* so that the word has to be scanned as a pyrrich. Even in Lucretius the traces of this shifting value of many terminal sounds may be seen : indeed, the elision of final m before the following vowel has maintained its position triumphantly through all periods of Roman poetry. We may, however, gather from the words of Cicero that the recos^nition of the existence of these final consonants grew stronger with time. He says (Orat. 48, 161) that the elision of the s in '' omnibu(s) prin- ceps" is "iam subrusticum," but he adds **olimautem politius." 42. Naturally enough this process was very gradual in its development. ** Language is the offspring of need and the foster child of social feeling : its growth and its enrichment are the effect of time : its beauti- fication is the work of taste, and we must look to the union of all the Muses for its perfection. The written language of a great nation which rises by slow degrees (and this merely by imitating other stages of culture alien to its own) from the mere level * Cf. "tempus fert" (Plautus), "magis stetisse" (Terence). OF THE ROMAN PEOPLE 75 of Nature, passing through every stage of barbarism — such written language requires a series of centuries before it can attain even a moderate degree of perfection. Such development presupposes the con- currence of numerous favourable circumstances. But of all these circumstances one especially must be insisted on: that the learned class of any nation, and chief of these the writers of genius and taste, the poets, orators, historians and popular philoso- phers, always contribute most to its enrichment, development and refinement" (Wieland, 1872). Luther, we know, had, as a Central German, a specially keen ear for dialectic peculiarities, seeing that he listened simultaneously to the dialects of Higher and Lower Germany, and thus seemed chosen by Providence to ensure for the written lan- guage of the High German chancelleries, by his translation of the Bible, the wide distribution which it enjoyed. In the same way the written language of Rome was influenced by Greeks, half-Greeks, Oscans, Umbrians, and Celts, for these had to learn Latin and to adapt it to their circumstances. 43. As a matter of fact, the taste of the old Roman poets was far from refined, and the Public made no great demands, because it was destitute of all pro- found aesthetic culture. Their very poems aimed at exciting interest rather by their subject than by their grace. The appeal of Ennius to the Muses in the Proemium to the " Annals," *' Musae, quae ped- ibus magnum pulsatis Olympum," might be as effect- ually addressed to trampling steeds : and the childish 76 LANGUAGE AND CHARACTER pleasure which the poet finds in imitating by onoma- topoeia the braying of martial trumpets, " At tuba terribili sonitu taratantara dixit" (Ann. 452, v), draws from us an involuntary smile. The mutilation of such words as gaudium into gau (" laetificum gau," Ann. 451) is harsh and violent, and the junction of such forms as quicqua7ny qiiisqtiam^ cuiquam in one drama (Trag. 448) is unclassical not to say un- Roman. The violent separation of words into two parts depends upon a complete misapprehension of so-called Tmesis, which completely jars with the genius of the Latin language. Such are *'cere — comminuit — brum " (Ann. 586) and " Massili — port- abant iuvenes ad litora — tanas" (Ann. 605). In their vocabulary the writers of this time were not delicate. Expressions which were at a later date banned and barred from classical usage, and con- signed to the language of the people, are, at this time, regular and normal : indeed, whole groups of words bearing the popular stamp, such as adverbs in -iter formed from adjectives of the second declension, substantives in -ela^ -monium, -tudo, are remarkably prevalent. Sometimes we alight on whole Greek sentences: and the anomaly of the formation of hybrid compounds of Greek and Latin, such as ikermo-potarey ante-logiumy ra-pacida, is seen in its infancy ; indeed it may be paralleled by the recast- ing of several Greek dramas to make one Roman play, and by the mixture of Greek and Roman local colour that we often see in the process. Again, most of the old Roman poets acted and wrote as though each one of them was equally able to compose OF THE ROMAN PEOPLE n comedies or tragedies, for following the example of the Greeks, Plautus was the first to adopt a system- atic separation of the two. Thus the characteristic of this period is its lack of finish. The writers are animated with the best intentions, but the words of Horace, "Versate diu, quid ferre recusent. Quid valeant humeri," had not yet been written — still less his maxim "nonum prematur in annum" — and we must remember the proverb ''ultra posse nemo obli- gatur." The Romans of that time thought differ- ently on this subject; indeed, even in later times there were found persons to admire and patronize those /r^V^/, and casci viri, who, misled by a prejudice for old Roman simplicity and nawet^, possessed, from our point of view, but little critical faculty. For instance, the inscription on the tomb of the poet Naevius, most likely composed by Varro, celebrates in Saturnian verse the high merits of the poet: ''Were it seemly that immortals should weep for mortals, the divine Camenae would weep for the poet Naevius. And so, since he has been made over to the place of death, Rome has forgotten to speak in the Latin tongue," i.e. in old Latin, in national Roman: and Aelius Stilo's view was that if the Muses had wished to speak in Latin, they would unquestionably have chosen the diction of Plautus : a verdict which did not quite meet with the approba- tion of Quintilian. He, the arch-Ciceronian, fully appreciated the difference between the language of a classical author with its fine gradations and ex- quisite style, and the straightforward diction of a Plautus, created to suit the demands of the popular 78 LANGUAGE AND CHARACTER taste. What Quintilian says about Accius and Pacu- vius, " Nitor et summa in excolendis operibus manus magis videri potest temporibus quam ipsis defuisse," is far more true of their predecessors. 44. Yet, with all their defects, those works con- tained the germ of a new phase of life in both litera- ture and language. Ennius was the first to take a bold and decisive step in the direction of progress — Ennius, who in Lucretius' judgment was the first to bear the evergreen chaplet from Helicon. The Saturnian verse had to give way to the Greek Hexa- meter, />., the accentual rhythm to the quantitative.* Syllables, after a long period of uncertainty and fluctuation, now for the first time received a certain fixed quantity, and terminal sounds acquired a ereater steadiness. The stiff lapidary form was gradually given up, and the *' broken-winded con- geries of lanky limbs " was replaced by flesh and blood and more pleasing harmonies, and, following the Hellenic model, the vocabulary was enriched by a stately train of newly-minted compounds. Foreign words were added. Till now Greek expressions had * There are, however, two main theories as to the character of Saturnian verse, the quantitative and the accentual. Those who hold the former theory regard the Saturnian verse as a verse of six feet with an anacrusis, and a break after the fourth, or more rarely after the third thesis; cf. The Queen was in her parlour, eating bread and h6ney. According to the second theory, this verse was an accentual one, no regard being paid to quantity. Lindsay holds that the first hemistich has three accents and the second two, as dabiint malum Metelli || Na^vio poetae. See Lane's Lat. Gr. § 2553. OF THE ROMAN PEOPLE 79 found their way into Latin merely as the result of long commercial intercourse, but now came the re- ception of other expressions due to distinct literary influence. Such were daedalus ((Ta/J^aAo?), malacus {{xocXocKog, mollis), cufuatilis (from kZ^o,^ sea-green), dia dearum {^^'io. ^iocoou)y pelagus {irkxocyoq), termo (repfjLuv, terminus), ephebus (e^'^Po?), poema {iroUfAa,), poeta (TTotnW?), pontus (TTOkTo?), cantpsare {y.(x.^irTiiy)y etc. Such words, found in great numbers in the oldest Latin poets, especially in Ennius, are a proof of the influence of Greek poetry. It became more and more the custom to enliven style by epithets coined after Greek models, and conforming to the exigences of the Hexameter. Similes and metaphors, formerly rare in verse, appear with increased frequency, though these in many cases were either translated or imitated directly from Homer. Side by side with metaphors taken from agriculture and war, tropes taken from the sea and the chase played a great part. The comparison of the people restlessly stirring to and fro in the assembly, with the sea, seemed to the Roman Senate in the year 189 B.C. new and striking (Polybius, xxi, 31, cf. xi, 29, 9, Hultsch), but by the time of Cicero it was trite, and in Livy's day hack- neyed. It was possibly at this time that the transition oi percontari i^xom contus = yioyro<;) properly **to sound with a steering pole" [and allied by popular ety- mology with percunctor] was applied to research in general, and such expressions appear as '* verborum fluctus — animus fluctuat" (Plaut. Merc, v, 2, 49), '' praeda undat" (Enn. Trag. 520), " iacturam facere," to jettison and then to lose. Of like character are 8o LANGUAGE AND CHARACTER indagare, to track out, properly to hunt into a net (cf. vestigium, a track), a meaning which was subse- quently transferred to all possible acts of pursuit: indeed, Ennius introduces the simile: "Sicut si quando vincHs venatica velox Apta solet canis forte feram sei nare sagaci Sensit voce sua nictit ululatque ibi acute," etc. While admitting that much of the metaphorical colouring of the early Latin poets was due to Hel- lenic influence, we must nevertheless remember that at the time to which we are now referring, ocean travel and the chase were fairly popular in Rome, other- wise the poets would scarcely have adopted so freely metaphors taken from such pursuits. It is a gener- ally admitted truth, that a nation's metaphors and similes reflect the contemporary culture of that nation. The language of Homer gives us information relative to the manners and customs of Homeric times. " The poet borrows the majority of his similes from elementary natural phenomena, the occupations of simple uncivilized men, hunters, the fishermen, cattle- herds, rustics, smiths, carpenters, tanners, etc." The ship seldom occurs in these early tropes because ocean travel was at that time but little developed. But we are able to follow the progress of the Greeks in sea-faring matters by the metaphors in use in their later poets. In Pindar we find already seven- teen such metaphors : in Aeschylus thirty, in Sophocles eleven, in Euripides no less than thirty-six. Thus Pecz ("Beitrage zur vergleichenden Tropik der Poesie," i Teil, Berlin, 1886) is perfectly right in maintaining that in the metaphors of Aeschylus we OF THE ROMAN PEOPLE 81 see reflected the times of the Persian war : in those of Sophocles the age of Pericles, and in those of Euripides the period of Demagogy. Thus the figures of speech of old Latin poetry teach us that the Romans of that time, after an intermission of cen- turies, entered into maritime commerce with spirit and energy, and that, after the Punic Wars, it was the custom to devote much time to the chase, after the Oriental fashion. 45. Generally speaking, poetry and poets alike stood in that age in no great repute. Cato says " Poeticae artis honos non erat." In the circle of the Scipios we meet with the most aesthetic taste and the highest scientific culture. This great family found pleasure in appreciating the poets of their own circle, not, it must be admitted, without an eye to their own advantage. In fact, just as they began by forming a cohors praetoria in order to increase, in the eyes of subject nations, the prestige of the greatest power in the world, they were eager to encourage in every way the singers of their exploits, and prob- ably also to influence the language of their race. At all events A. Gellius (Noct. Att. ii, 20, 5) states the tradition that *' Scipionem omnium aetatis suae pur- issime locutum." The scribae, to whom had been assigned till now a chamber intended for meetings situated on the Aventine in the Plebeian quarter, were summoned from their dark corner and invited to bask in the glory of their Imperatores. Cato brought Ennius with him to Rome, and M. Fulvius Nobilior, on his Aetolian campaign, kept him in his G 82 LANGUAGE AND CHARACTER entourage-, others followed his example, and It be- came the fashion, especially after the days of Afri- canus the younger, for generals to take poets in their escort. The influence of the Scipionic circle is particularly noticeable in Terence. The language of the plays of Terence is purer and more refined, and generally more correct, than that of Plautus. But how little such virtues of style were prized at that time is plain from the judgments of his mistrustful colleagues, who called his " oratio " tenuis, and his "scriptura" levis (pale and expressionless) in com- parison with that of Caecilius. It was with Terence, too, that rhetoric began to force its way even more and more into poetry : indeed, rhetoric raised itself in no long time to a power of the first rank, and spread itself gradually over all Roman literature. In the first half of the second century before Christ, the impulse given to literature by Greek philosophers and rhetors in Rome was so great, that all the efforts of the "national" party in Rome to stay the current proved unavailing.* The Epi- cureans Alcaeus and Philiscus, who were exiled in 173 B.C., and especially the grammarian Crates of Mallus [who in 157 B.C. was sent by Attalus as an ambassador to Rome, where he introduced for the first time the study of grammar], and further, the historian Polybius, with all the numerous other Achaeans who were detained for years in captivity at Rome as hostages, and lastly, the Athenian Embassy * For the effects of Greek culture on Roman thought, see Mayor's "Ancient Philosophy" (Cambridge University Press), pp. 209 sqq. OF THE ROMAN PEOPLE 83 sent to Rome under the superintendence of the Academic philosopher, Carneades, made such a strong impression on the youth of Rome, that henceforth grammatical and rhetorical studies entered into the daily necessities of a Roman's life. They were greedily caught at by every one, '* quasi diuturnam sitim explere cupiens" (Cic. De Sen. viii, 26). For rhetoric aided the Roman taste for lucidity of thought and logical definiteness of repre- sentation. Soon Latin rhetoricians, too, opened their schools. Thus it came to pass that in Pacuvius and Ennius the results of rhetorical studies were even more apparent than in their predecessors. The anti- theses and the parallelism observable in the construc- tion of the sentences, and the better rounded and fuller periods of their style, stand out in sharp con- trast to that of the ordinary language of the day, which contains many vulgarisms. 46. But the orators reaped the main advantage of the new rhetoric. The art of persuasion alike in the Senate, in the popular assembly, and in the Law Courts, had been practised from the earliest times, and it is sufficiently remarkable that the first prose work published by a Roman author contains a speech of the blind Censor, Appius Claudius. The opportunity now presented itself of learning the principles of a correct training, and these prin- ciples were eagerly hailed as offering a greater chance of success in oratory. Especially did M. An- tonius and L. Licinius distinguish themselves — the only orators whom Cicero considers worthy (as he 84 LANGUAGE AND CHARACTER says in his De Or.) to serve as the interpreters of his ideas and reflections on rhetoric. According to him (De Or. i, 26, 178) they especially avoided the " barbaries forensis," and endeavoured to employ a correct language. That they attained to this correct- ness by means of strict academical training appears, not merely from the surviving fragments of their speeches, but, at least in the case of Crassus, from a hexameter of Lucilius (/V. inc, xxxiii). '* Crassum habeo generum, ne rhetoricoteros tu sis " [quoted by Cic. De Or. iii, n, 171]; and when the same poet says: "Crassi pater huius panaethi " = splendidi, it seems probable that he refers to the same orator. Thus oratorical grace of form may be dated from Licinius Crassus. His style of expression was care- fully chosen and lucid, clever, and sparkling with wit. He aimed also at pregnancy of exposition, and strictly limited his periods. He employed, too, parallelism in the division of his sentences, which materially contributes to clearness of style. In con- trast to him, M. Antonius,'^ in his quality of zealous disciple of the great master, Cato, strove to attain a simpler and less ornate styleof expression. Buthehad the art of marshalling every clause in every sentence so that each fell into its appropriate place, with the result that his periods resembled a skilfully arranged army in battle array. Considerations not of beauty, * A full account of the oratory of Crassus and of M. Antonius is given in Prof. Wilkins' " De Oratore," p. 12. Antonius always tried to avoid the appearance of undue elaboration, though, as a matter of fact, he prepared his speeches very carefully. Brut. 37, 139- OF THE ROMAN PEOPLE 85 but of utility directed his impulses. Sallust, too, followed close on Cato's footsteps and set himself deliberately to seek out archaic forms. But in Hortensius the turgid style of Asiatic oratory seemed to gain new ground. 47. Not till the time of Cicero did the elegance and grace of Hellenic form ally itself with Roman earnestness and dignity. Cicero was the first who enabled the Latin language to become what fate intended it should become, the means whereby classic culture — in fact, it may be said all the culture of antiquity — became known to the north- ern barbarians. Thus Velleius Paterculus has good grounds for his assertion (i, 17, 3): "At oratio ac vis forensis perfectumque prosae eloquentiae decus, pace P. Crassi Scipionisque et Laeli et Gracchorum et Fanni et Servi Galbae dixerim, ita universa sub principium operis sui erupit Tullio, ut delectari ante eum paucissimis, mirari neminem possis." On this ground, too, Tacitus, Dial. c. 18, was justified in maintaining *' Mutari cum temporibus formas quoque et genera dicendi; sic Catoni seni comparatus C. Gracchus plenior et uberior, sic Graccho politior et ornatior Crassus, sic utroque distinctior et uberior et altior Cicero." Of him it may be said more truly than of any Roman that he was hivlq xkyay — as far as any Roman could merit this high praise; he, more than any other orator, was a supreme master of language. Doubtless Cicero's efforts were not always received with favour: opposition to them manifested itself from more than one quarter. Thus 86 LANGUAGE AND CHARACTER he was opposed by the Roman Atticists under the guidance of C. LIcinius Calvus, and these criticized him sharply and bitterly (Quint, xii, lo, 12).* Be- sides this, he was met by a set of critics of inflexible and defiant national pride, who actually piqued themselves on speaking in the highest degree in-- Msitate and mquinate, affecting as they did to believe that correctly and unusually were convertible terms. The improvement of style was proceeding rapidly, and was not to be checked: but still there were men of the old school who would have nothing to do with the new Hellenism, and expressed their decided preference for the old Roman style. Of course their efforts were fruitless. On the other hand, Cicero had no lack of admirers and disciples. Caesar gave him the most remarkable testimony when he wrote on dedicating to the orator his work, " De Analogia": '* You have discovered all the treasures of oratory, and have been the first to employ them. Thereby you have laid the Roman people under a mighty obliga- tion, and you honour your fatherland. You have gained the brightest glory, and a triumph which is to be preferred to the triumph of the greatest gen- erals : for it is a nobler thing to enlarge the bound- aries of the intelligence than those of the empire." 48. In any case it is true that with Cicero the ** parens facundiae Latinarumque litterarum" (Plin. Nat. Hist, vii, 30), oratorical and philosophical prose had attained its high water mark. No one, * [Cf. also Cic. Orator, § 76 sqq. : and especially Tac. Dial. 18.] x\ OF THE ROMAN PEOPLE 87 either of his predecessors or of his successors, has approached him in lucidity and appropriate expres- sion, in delicate exposition, in rhythm, in harmony, in the just accentuation of syllables, and in the careful balancing of his sentences, and of their periods. The orator might easily have been betrayed into a too implicit trust in his own oratory : and in this confid- ence he might have set himself to conquer even in a bad cause, and deliberately have tried to deceive the people: and certainly Cicero has not altogether escaped this danger. His style, together with his vacillating political views, worked deleteriously on his character. What Cato once dreaded for the young men of Rome, on the occasion of the visit of the three Greek philosophers— that they might be tempted to rate the glory of words higher than that of deeds, and that in the glamour of Greek dialectics they might find it hard to see the truth (Plut. Cato, 22; Plin. Nat. Hist, vii, 31, in)— followed as a natural sequel of the new methods of rhetorical training. 49. We have been engaged on the features of style which reflected and forwarded the improve- ment and the ennoblement of the Latin tongue; it is now time to turn to those which reflect more particularly the influence of growing culture on language. We get a good idea of these from the figures of speech, and especially from the metaphors, employed by Cicero. Side by side with the old and favourite figures borrowed from agriculture, war, and jurisprudence, we find a series of new metaphorical 88 LANGUAGE AND CHARACTER expressions: the technical terms of horse-racing and gladiatorial shows came into fashion. Then the stage, medicine, arts, and sciences contributed their colour- ing, and references to Greek literature, and especially to Homer, became common. The language, too, was enriched by the study, the translation, and the editing of philosophical writings and other scientific Greek works : hence new terminations were formed, and the number of abstract terms was materially increased: conceptions of species, too, the lack of which had not made itself felt in a primitive stage of culture, were more defined. Still this process was but slow: for instance, the word ''pardon" had to be expressed by "ignoscendi ratio" (Cic. Rose. Am. i, 3): for ** being," to U, even Seneca had no expression: he wrote (Ep. 58, 6): "to U dico 'quod est'; cogor verbum pro vocabulo ponere " ; at a later period essentia and ens were formed after Greek analogy. 50. The number of borrowed words multiplied in all branches of life, and more especially in intellectual conceptions. However successful Cicero's authority, and his endeavour to call into being a philosophical terminology, might be deemed, and however much encouragement he received in his efforts to supersede Greek artistic expressions by those of Latin origin, still, as a rule, the Greek word was taken over in its simplicity. Even such a genial poet as Lucretius, who solved the difficult problem of representing a philosophical system in verse, had to confess (i, 11^ sqq)\ OF THE ROMAN PEOPLE 89 Nee me animi fallit Graiorum obscura reperta difficile illustrare Latinis versibus esse, multa iiovis verbis praesertim cum sit agendum propter egestatem linguae et rerum novitatem ; and again, in iii, 259 sqq. : Rationem reddere aventem abstrahit invitum patrii sermonis egestas. and others found themselves in the same difficulty. Hence the number of foreign words in Latin in- creased amazingly, and Roman writers grew more and more to employ Greek as a neat auxiliary to round off their phrases, much as the Germans, es- pecially since the time of Louis XIV, employed French: only with this difference, that the Germans kept their poetry as far as possible free from foreign elements, while in Rome the poets, more than any other class of writers, had recourse to them. The Germans feel a profound conviction that poetry, as the expression of man's deepest feelings, of all that moves and stirs his heart most powerfully, must be before all things national', the Romans, on the other hand, acted on the principle that the ear of the hearers must be captivated by melodious harmonies and pleasing form: " Non satis est pulchra esse poemata, dulcia sunto" [Horace, Ars Poet. 99]. 51. A further sign of growing refinement in culture appeared in the endeavour manifested by the authors of the period to avoid or to veil words and ideas which suo^orested coarseness. Writers in the first instance abstained from using such words, replacing them by harmless, colourless 90 LANGUAGE AND CHARACTER expressions. But through frequent usage, even these came to be more and more connected with increas- ingly unpleasant associations, so that they in their turn began to be banned, and finally disappeared from use in cultured circles. Several vulgar words which had been in general use in Latin literature went out of usage and were employed by satirists only, and merely for the purpose of emphasizing the dark side of Roman civilization. Cicero writes in a letter to Paetus (Ad Fam. ix, 22): ** Ego servo et servabo — sic enim assuevi — Platonis verecundiam. Itaque tectis verbis ea ad te scripsi, quae apertissimis Stoici. Sed illi etiam crepitus aiunt aeque liberos ac ructus esse oportere," and in a similar spirit he says (De Or. iii, 164): *' Fugienda est omnis turpitudo . . . nolo dici morte Africani castratam esse rempublicam; nolo stercus curiae dici Glauciam." It thus appears that literary men knew the coarse terms, but avoided mentioning them, and preferred to cloak them with a decent veil. As a counterfoil to this process it was unavoidable that at this period perfectly innocent words and ideas received in some cases an ironical connotation, and were degraded into expressions of contempt. For the civil war, so long protracted, and especially the degrading influence of the delatores, had spoiled the character of the people. The period of childish art- lessness, self-complacency, and simplicity, had passed away. Malice and evil of every kind had become so much a matter of course, that it became an involuntary factor in the pessimistic colouring given to the signi- fication of words. Thucydides mentions (iii, 82) the OF THE ROMAN PEOPLE 91 influence of the Peloponnesian war on the language of Greece : " Proper shame is now termed sheer stupidity : shamelessness, on the other hand, is called manliness: voluptuousness passes for good tone: haughtiness for good education : lawlessness for free- dom : honourable dealing is dubbed hypocrisy, and dishonesty, good fortune." Sallust has a similar utter- ance with regard to his era. He puts into the mouth of the younger Cato, the tribune designate, the words (Catil. 52, 11): "Hie mihi quisquam mansuetudinem et misericordiam nominat ! Jam pridem equidem nos vera vocabula rerum amisimus : quia bona aliena largiri liberalitas, malarum rerum audacia fortitudo vocatur, eo respublica in extremo sita est?" And he represents Licinius as uttering the same thought (Hist. fr. iii, 82, 13, Kritz): *'Quod ego vos moneo quaesoque, ut animum advortatis neu nomina rerum adignaviam mutantes* otiumproservitio appelletis?'' 52. As soon as Augustus mounted the Imperial throne, a new chapter of Roman literature was opened. Poetry now rose to the zenith of its brilli- ancy. Rome was warmed into new life by the gentle air of peace : the rays from the sun of His Imperial Majesty sent a glow through men s hearts and ex- panded them. A spring-tide of song succeeded, such as Latium had never before witnessed : wine, woman and song were celebrated by singers of genius. And the ruler earnestly wished his people to devote themselves with increasing interest to art and science : he wished to divert their thoughts from poHtics. * Cf. too Hor. " at vos virtutes ipsas invertitis," etc., Sat. i, 3, 55. 92 LANGUAGE AND CHARACTER Hence, in conjunction with Maecenas, he made it his object to give poetry the greatest possible encourage- ment : he drew the most celebrated poets of the day into his circle and honoured them with his society : he expended vast sums on shows and spectacles, more especially on pantomimes and mimic naval battles. Oratory, which had hitherto won its laurels in the Forum, found itself, under the depressing influence of political restraint, now relegated to the schools of declamation : more than ever young and old flocked to the rhetoricians' schools to take their part in the Controversiae and Suasoriae which were held in these institutions, to learn the method whereby a given theme is treated from every side with all kinds of subtleties and refinements of argument. Unques- tionably poetry was the gainer by this method " That firm and sure technique of arrangement and representation, that plastic of the word, which gives the stamp of classicism even to mediocre writings, dates from this school, through which every poet passed " (Ribbeck, **Geschichte der romischen Dich- tung," ii, p. 7). But since it is true that such rhetorical methods belong rather to prose than to poetry, we cannot help feeling, even while perusing the most important productions of that time, that they were to some extent the creations of sober intelliofence : we often feel the lack of the warm breath of inspira- tion which comes directly from the heart, and goes straight to the heart in turn. And we are supported in our view by the inclination of the Roman poets to masquerade in the guise of superior erudition. Fol- lowing the precedent of the Alexandrian poets, whom OF THE ROMAN PEOPLE 93 it was the fashion to take more and more as models, it became increasingly the fashion to unpack the treasury of knowledge before the patient hearers or readers of the poets of this day. It is a genuine pleasure to Ovid to recite in his Metamorphoses the names of all the rivers and mountains which had to suffer the heat of the sun on the occasion of Phaethon's wild drive. Propertius, in his elegies, overwhelms us with references to Greek mythology; * Horace, too, likes to make a brave show with his Greek names; and Vergil not unfrequently breaks the calm flow of epic poetry by learned reflections, or again, by such phrases as that of Aen. vi, 173: ''si credere dignum est." Such phrases are not the natural language of poetry, which, as Schiller has well remarked, has to make its way not through the cold region of the intelligence, and ought not to summon erudition as interpreter, but, as it springs from the heart, so to the heart should it appeal. Besides this, no one hesitated to grovel before the mighty emperor with the utmost self-abasement, and, indeed, to pay him homage with almost oriental servility. 53. The thorough education which Augustus had enjoyed, had given him a fine appreciation of form : the brilliancy of contemporary literature rendered him unsympathetic to the simplicity and roughness of the old literature of Rome. He reproved his step- * See Postgate, "Propertius, Select Elegies," cap. v. "The ambition of Propertius was to be the ' Roman Callimachus ' " (v. I, 64). I 94 LANGUAGE AND CHARACTER son, the future Emperor Tiberius, for his taste for archaisms, and actually spoke of the " foetores recon- ditorum verborum" (Sueton. Aug. 86). With the sole exception of Vergil, who made it his object to attain the solemn dignity and earnest note of antiquity, scarcely one of the Augustan poets permitted him- self the scanty licence allowed by Horace in his " Ars Poetica," with respect to antique precedents (lines 48 sqq,). On the other hand, authors never ceased their endeavours to render their language pliable and flexible after Greek models, whence Tacitus speaks of ''calamistri Maecenatii " (Or. c. 25). In some cases Greek constructions were simply taken over, as: **gaudet potitus": in other cases genuine old Roman constructions were employed more freely than before, and made to follow Greek analogy; these constructions were used with words of similar signification. We may instance the objec- tive genitive after the adjective (as in the case of diveSy which follows the construction oi plenus, and is influenced by such Greek constructions as irXodcnoq Tivoc;): and again the simple infinitive [used instead of ut with the subjunctive] after impellere, which is made to follow the analogy of itibere, but was in- fluenced by l-KiT^k-Kuy : such constructions were much favoured. Again, following the example given by Greek poets, certain figures of speech came into general vogue, cg,^ the aVo xoii/ou,* the usage of which increased to such an extent that we find it in * Cf. Horace, Odes, i, 3, 6, and ii, xi, 11, "Quid aeternis mi- norem consiliis animum fatigas," and ii, xvii, 22, ''impio tutela Saturno." OF THE ROMAN PEOPLE 95 Catullus nine times, in Tibullus twenty-three, in Propertius fifty-seven, in Horace a hundred and eighty-eight ( Aken. " De figurae a-rrl xoti/ou usu apud Catullum,Tibullum, Propertium. Schweriner Progr.** 1884; " Zeitschrift fur Gymnasialw." xxxi, 337 seq,^. But Greek inflexional forms also took root in Latin; this usage was remarked in Accius, and criti- cized in his case, but was afterwards regarded as not unusual. In older Latin, writers adopting foreign words had been careful to give them a Latin stamp, and with this view had Latinized their terminations: but now an opposite tendency set in. Greek case- forms were held to be more melodious and graceful than those of Latin, and more suitable for the higher flights of Lyric poetry; thus they came into more constant use. Propertius is full of them, Horace employs them more sparingly. In the Satires he writes Europam and Penelopam : in the Odes Europen and Penelopen. More particularly in the case of proper names the Greek form is maintained, and thus we commonly meet with formations of the first declension in e, eSy en, and a7i : in the second in os and on : besides these we find accusatives of the third in iny yn, a, and as\ genitives in os, and dative plurals in sm. With this Censorinus' ^ remark tallies (De Die Nat. c. 24) ** Stella quam Plautus Vesperuginem, Ennius Vesperam, Vergilius Hesperon appellat." 54. Prose could not but follow in the wake of poetry; but its progress marked decadence. The language of prose should stand midway between the * Circ, A.D. 238. If ^ 96 LANGUAGE AND CHARACTER diction of the people and that of poetry, and should maintain itself at an equal distance from each ; if it approaches either extreme too closely, it loses its balance. Old Latin prose writers inclined too much to the vulgar style. Silver Latinity fell into the other extreme ; under Vergil's influence it simulated originality by the poetical colouring of its style. Tacitus admits in the Dialogue about illustrious orators (20) *'Exigitur iam ab oratore etiam poeticus decor," and Quintilian enlarges on his precepts by adding *'A corruptissimo quoque poetarum figuras ac translationes mutuamur'*; generally speaking the principle, ** Historia quasi solutum carmen," was challenged.^ But the declamations so popular at that time **necessariadeserunt,dumspeciosasectantur" (Seneca, Controv. 9, praef. 2). If the periods of the ancient writers may be compared to temples constructed *'rudi caemento et informibus tegulis," the periods of these later writers resemble more nearly such as "marmore nitent et auro radiantur" (Tacitus, he, €itat,\ Doubtless it may be objected that prose writing in Germany was mainly brought to perfection by poets, but these were at the same time masters of a good prose style. Indeed, it is open to discussion whether Lessing and Goethe, the former thanks to his shrewd insight, the latter owing to his realistic appreciation of all his surroundings, were not intended by Nature for prose writers, and for holding the mirror up to * Cicero's views on the language of poetry may be seen in the Orator, 20, § 66 sqq. OF THE ROMAN PEOPLE 97 Nature with marvellous exactitude. Should it, how- ever, be maintained that a good prose writer must perforce be a poet, this were to mistake the essence of prose, as indeed the writers of the Silver Latinity actually did. Still these were the children of their age; they were obliged, if they counted on any response to their writings, to reckon with the spirit of that age. 55. The Romans of that epoch were sunk in luxury and debauchery. With evil morals, evil words found their way into the language, " Tuncque pri- mum" (says Tacitus, Ann. vi, i) "ignota antea vocabula reperta sunt sellariorum et spintriarum ex foeditate loci ac multiplici patentia." In Cicero's time, perhaps, too much obvious attention was paid to masking indecencies. But now speakers and readers went so far as to suspect improprieties as lurking behind good, honest, innocent expressions. No doubt Sallust used the phrases " ductare exer- citus" and "patrare bellum " without any sinister connotation; but ordinary modesty had by Quin- tilian's time sunk so much in common estimation, that these expressions conveyed to the minds of readers or hearers some unpleasant or sinister significance. Expressions, harmless in themselves, were thus classed as improper, because the genera- tion of readers was morally depraved. The genera- tion was called xaxo>aT<,K, and exemplified the dictum of Quintilian, viii, 3: "Si mala consuetudine in ob- scoenum intellectum sermo detortus est." The graceful old custom of beginning letters with H 98 LANGUAGE AND CHARACTER the formula **si vales, bene est; ego valeo," which had begun even in Cicero's age to fall into disuse, now completely ceased. Hence Seneca could say (Ep. 15) : " Mos antiquus fuit usque ad meam serva- tus aetatem primis epistulae verbis adicere : si vales bene est." And Pliny (Ep. i, 1 1, i) confirms this with the words : " Scribe solum illud, unde incipere priores solebant: *si vales, bene est, ego valeo.' Hoc mihi sufficit; est enim maximum." As with the beginning of letters, so was it with the opening of speeches. In olden times the custom was to open every speech with an invocation to the gods. Servius on Vergil (Aen. ii, 301) says: ** Maiores nullam orationem nisi invocatis numinibus inchoabant sicut omnes orationes Catonis et Gracchi; nam generale caput in omnibus legimus." But by Cicero s time this pleasant old custom had com- pletely died out : there is no trace of any such thing in Cicero's speeches; nay, he actually treats with derision (Servius,/!?). Again they disliked lengthy words of four syllables .(Horace's sesquz- pedalia verba) at the end of the Hexameter, which the Greeks preferred as giving the verse a soft and melodious ending ("gracili mollem pede claudere versum," Verg. Cir. 20). It was for this reason, too, that they had such a strong objection to spondaic io8 LANGUAGE AND CHARACTER lines in the penultimate foot of the Hexameter {versus spondiaci) which, as we know, fell mostly on quadrisyllable words (cf. Quintilian, ix, 4, 65). Though Ennius, and following him Lucretius, em- ployed soft verse terminations like naturai, we may look in vain for such in classical Latin; the only ex- ception to this rule is to be found in the fact that writers of the latter period allowed certain excep- tions in the case of Greek words (cf. **lenissimus Onchesmites," Cic. Ad Att. vii, 2). 63. As in the case of the Hexameter, so the lyric metres which made their way more freely into Roman poetry had to yield to the levelling influence of the Roman linguistic spirit. Thus, for instance, it is notorious that Horace in his Alcaics and Sapphics re- places, where the verse admits, a trochee or an iam- bus by a spondee, just as in his Odes he has carried through the long syllable in the anacrusis; these are mere tricks of style, aiming at bringing the metres which took their origin on foreign soil into harmony with the peculiarities of the Latin tongue. 64. It was, however, the sense of beauty which dictated not merely the new shape of the metres, but also the choice of words. There is indeed no doubt that the tone and the expression of the Satires and Epistles approach much more nearly the lan- guage of the people than the more refined diction of the Odes and Elegies, and that many words are admitted into the former which are banned by the latter. But speaking generally we must admit that OF THE ROMAN PEOPLE 109 the poet has not merely in his expressions, but par- ticularly in his choice of words, kept the ideal of beauty before his eyes. His one irrefragable law is to avoid sullying his style with common words. The motto on his ensign is " Odi profanum vulgus (verb- orum) et arceo.""^ Hackneyed and vulgar expressions, far from set- ting off poetry, rob it of its charm and therefore are in place only when the poet wishes to attain a certain definite end.f Vulgar expressions like agaso, bala- tro, catipOy nebulo, popino certainly occur in Horace, but in his more or less popular works, the Satires J and the Epistles; the portals of lyric poetry are closed to them; we may look for them in vain in the Odes and Epodes. A genuinely inspired poet, in whom the true poetic fire burns bright and clear, will permeate his diction with harmony, stateliness, and purity; and noble as his mind and intellect will be the words which issue from his mouth : Audebit, quaecumque parum splendoris habebunt Et sine pondere erunt et honore indigna ferentur Verba movere loco, quamvis invita recedant Et versentur adhuc intra penetralia Vestae. (Hon Ep. ii, 2, iii.) * Cf. Mackail, " Latin Literature," p. 1 14. " In his measured epithets, his curious fondness for a number of very simple and abstract words, and the studious simplicity of effect in his most elaborately designed lyrics, he reminds one of the method of Greek bas-reliefs or ... of the sculptured work of Mino of Fiesole." t Such as characterization, or, again, it may be bathos. X The first book of the Satires shows, to quote Mr. Mackail, " a vein of artistic vulgarity " which is wanting in his later work. no LANGUAGE AND CHARACTER 65. In every Literature there occur a large num- ber of expressions which are exclusively, or almost exclusively, confined to poetic use. These expres- sions were either the actual creations of the poets, as many ornamental adjectives certainly were; or else they came in course of time to be specially favoured for the purposes of poetry, and were thereby maintained as living factors in the language of the poets, while they disappeared from the popular lan- guage; such are for instance latices and lympha, for "water." It were an interesting task to trace accur- ately the conceptions to which different nations, in their poetical vocabulary, apply such special words; such a quest would throw many an interesting side- light on national peculiarities. It is characteristic of German that the words Maid and Ross are con- trasted with Mdcicken and Pferd\ * we recognize in this distinction a testimony to the high admiration for woman and for the noblest of the brute creation entertained by Teutonic peoples. It is not less sig- nificant that the Hebrew in his poetic style possesses special words to express the name of God, The lifework of Israel lay, in fact, in religion; the main current of the Semitic spirit set not towards the world with its manifold external phenomena, but looked beyond this, to the Godhead itself. Thus again, the Roman possesses two words for the sword, the '^xos'SAZ gladiiis and poetical ensis. It would thus appear that ideas which appeal most to the popular imagination tend to lose by time the definiteness of * Much as " wench " and '* nag " may be contrasted with " girl " and " horse " in EngUsh. OF THE ROMAN PEOPLE III their meaning, and are the first to suffer from the differentiation between the diction of poetry and prose. It frequently happens that the difference between poetical and prose diction consists merely in the employment of a different suffix as pauperies =pau' pertas\ iuventa = inventus \ contagiuin — contagio\ oblivium = oblivio\ Graii = Graeci] rabidtis = rabi- ostis'y or again it may be in a newly formed plural such as sibtla = sibili] which last, as it could find no place in a Hexameter, may be due to metrical exi- gences. 66. Besides this, foreign influences must be taken into account. As the German looks on everything which comes from '* near here " as less valuable than what comes from a distance, the Roman resembles him in the preference shown by Latin authors for Greek snippets rather than for good old Latin words. For instance, the names Tartarus and carbasus, whose usage instead of inferi and velum is reserved almost exclusively for the language of poetry, hail from Greece. Besides, Greek expressions fell in most cases more agreeably on the ears than sounds of home origin. Indeed, Ouintilian expressly re- marks (xii, 10, 33): **Tanto est sermo Graecus Latino iucundior ut nostri poetae, quoties dulce carmen esse voluerunt illorum id nominibus exor- narent." How could the harmony of the words diota, barbitos, philyra, amystis, and the varied lights and shades of their liquid vowels, escape the notice of a writer like Horace? 112 LANGUAGE AND CHARACTER 67. The second main requirement of the poet is Vividness and Perspicuity.* "For poetical repre- sentation, keen and sharp-cut outlines and subtlety of reasoning are of less account than the impression produced on the mind of the reader and the fascina- tion produced by figurative expression. The poet appeals in the first instance to the heart; his crea- tions appeal to the feelings rather than to the under- standing; and from the feelings they challenge a lively response. The prose writer, on the other hand, appeals first and foremost to intelligence; his productions challenge careful and well-considered reflection. It follows that the prose writer must choose words appropriate to the subject, such as represent the subject of the discourse in proper perspective; — he must express himself clearly and logically, for his object is to produce conviction. The poet, on the other hand, must write gracefully and suit his style to his subject. He must write with liveliness and observation, and the form of his dis- course must be graceful and must appeal to the heart, for his aim is to give pleasure. f But, we may ask, how does a poet attain this vividness and per- spicuity? It may be that he brings objects directly before our view by means of picturesque expression * The German term Anschatilichkeit has no exact English equivalent. It means the property of standing out boldly before the eye or mind of the reader, so that he cannot fail to visualize the conception. t Poetry should be " simple, sensuous, and passionate" (Milton). By "sensuous" is meant that which appeals readily to the senses, and hence poetry prefers picturesque images to the enumeration of dry facts. (Abbot and Seeley, p. 56.) OF THE ROMAN PEOPLE 113 or action dramatically quickened into life, or it may be by means of rhetorical exaggeration and the effects produced by contrast." 68. Forcible pictures are gained in poetry by the use of picturesque side-touches. Much that in prose would be omitted as superfluous is often an indispensable element for the poet. Thus we find, e.g. in Vergil (Aen. i, 614), "ore locuta est''; i, 94, *' voce refert " ; i, 579, ** animum arrecti," and in other places words like nianu, ocmHs, etc., which appear for the sense of the passage superfluous. We may add to this the ornamental adjectives characteristic of poetry, which resemble dewdrops sparkling like diamonds under the sun's rays. They lend a mar- vellous charm to poetic language and appeal power- fully to the imagination, for by bringing out the most marked characteristics of different objects they force them on our attention in the most striking way."^ If they are new and original they produce a greater effect still. In this respect it must be admitted that the Roman poets are somewhat unfortunate; they frequently mutilate what they have found in their old Greek models, and thus it is that they often fall short of the fine observation and grace of the cor- responding Greek expression. How commonplace and ineffective appears the rendering of irrsposig by ce/er; of vriig ccfxcpieXio-a-oci by curvae naves] of xaAAtppow TTOToifxu by fiumine pulchro\ of fiko(r/., words which in the second half of the compound began with a short syllable, and were mainly derivatives of verbs with a short stem-syllable, as for instance inag7iiloqtms\ — in such a case we can see that a dactyl is produced by the process of composition, when a trochaic precedes it as the first member of the compound. 69. In cases where the poet finds that a mere epithet fails to touch our fancy he likes to avail himself of a fuller presentation of the idea, e,g,^ of ii6 LANGUAGE AND CHARACTER the figure called ** Distributio " or division of the parts of the statement. Thus when Vergil wishes to insist on the fact that something in his mind will last for ever, he expresses himself (Aen. i, 567) in these words : In freta dum fluvii current, dum montibus umbrae Lustrabunt convexa, polus dum sidera pascet, Semper honos nomenque tuum laudesque manebunt, and Ovid (Met. xv, 871) repeats the same thought: ** lamque opus exegi, quod nee lovis ira nee ignis Nee poterit ferrum nee edax abolere vetustas." * As we know from Lessing's '' Laocoon," it was a fine artistic touch of Homer's to translate the description of objects into action, in fact to change co-existence into sequence. Writers like Goethe in the same spirit (e.^:, in the description of the host in " Hermann und Dorothea ") followed this example. Now the old Romans were completely lacking in apprehension of this fine trait of epic technique, and although they read Homer as well as we do, they devoted their utmost efforts to dry descriptions of objects. What a feeble repro- duction of the famous picture of Achilles' shield is the corresponding episode of the eighth book of the *'Aeneid" (607-731) with its ever-recurring ''here is" and "there is 'M Homer presents us with the picture of Hephaestus, and we see by the aid of his master-hand the shield Cf.: Thou mass of honour, thou King Richard's tomb, And not King Richard: thou most beauteous inn, etc. Shakespeare, K. Rich. II. OF THE ROMAN PEOPLE 117 ordered by Thetis composed and welded together. Vergil tells us how one picture after another is seen on the work of art he is describing. It is worth while to compare the representation given by the same poet of the door of the Temple of Apollo at Cumae (Aen. vi, 20 s^^.), and the description of the pictures in the temple of Juno at Carthage (Aen. i, 465 s^g.), or the sketch of the sun-god given by Ovid (Met. ii, I s^^.): and we shall speedily be in a position to judge how inferior were the Romans to the Greeks in such pictures. 70. Effective expression is, however, sometimes secured by figures of speech. At one time the poet appeals to the imagination of the reader or hearer by putting a part for the whole: 2^^ ptippis, carina, or it may be velurii, for an entire ship. In this case he appeals to the reader to widen by his own efforts the conception presented to him.* Sometimes again the poet causes the hearer to apprehend, say, the idea of an elephant under an elephant s tooth, while the oak tree shrinks in his description down to an oak leaf. He gives us the ash for the spear, the gold for the golden vessels, or /lamina for heat, lux for day; that is to say he changes the agent and the object acted upon. Just as Schiller speaks of stones as feeling, of nature as devout, of flight as hurrying, so the Roman poet endows ears and arms with * This very effort produces a sense of surprise on the mind of the reader, and a series of new impressions is part of the tech- nique of poetry in general. See Herbert Spencer's " Philosophy of Rhetoric." I lli ii8 LANGUAGE AND CHARACTER feeling and receptivity when he sings ** Auriculae gaudent praenomine " or " brachia gaudentia loris." Inanimate nature assumes life before his mind's eye: he breathes the breath of life into all that surrounds him.* 71. It is true that even in their treatment of these figures of speech, the Romans must be ranked far behind their kinsmen the Greeks. They are most independent and most original in the employment of Synecdoche,! a form of trope employed by all poets more frequently than any other. There was nothing more to do in this case than to interchange two conceptions which, as a rule, stand spatially con- nected, and thus suggest each other. Those next in frequency are metonymy and antonomasia, the tropes most applicable to attributes and apposition. In these the relation of the conceptions to each other is somewhat harder to gather, as it does not present itself immediately to the mind. Now the employment of metonymy must be admitted to be a little mono- tonous, and the frequent recurrence of Mars for bellum, of Ceres iov frimienhim, of Liber or B ace htcs for vinnvi, of Vulcanus for ignis, of Phoebits for Sol, of Neretis for mare, and of all the rest of the deities who have to be marshalled in procession whenever their products are mentioned, is not calculated to * The English reader may consult Blair's " Lectures on Rhe- toric "(xiv), *'on the origin and nature of figurative language," and Campbell's " Rhetoric," and Whateley's " Rhetoric." t See Bain's "English Composition," p. 22, for numerous in- stances of these figures of speech. OF THE ROMAN PEOPLE 119 give us an exalted idea of the imagination of the Roman poets.* The violence of the change is still more felt in the case of antonomasia. The Greek patronymics are extremely useful in such cases \e.g,y Pelides = Achilles)', we may set against these satus, editus, status {e.g., Maia 7iatus^ Merctuntis), genus {lapeti genus = Prometkeus), senex Pylius = Nestor, filius Anckisae = Aeneas, fratres Helenae = Castor and Pollux, etc. 72. The invention of the Roman poets appears poorer than ever in its attempts at metaphors and similes. Setting aside the cases of such transfer- ences of signification as it shares with prose, it possesses but a scanty stock of metaphors; certainly such cases as Aen. vi, i sqq., where they are regu- larly packed together, must be considered rarities; much that we find in the poets of the Augustan age takes its origin from the Alexandrine poets. The similes, too, are in many cases borrowed direcdy from the Greek, and Father Homer, above all others, has been ransacked for the purpose: e.g., passages like Aen. i, 589 sqq., and i, 498, point straight to Odyssey, vi, 232 sqq., and vi, 102 sqq. But we cannot describe the imitation as particularly happy : it appears rather artificial and forced. How far more graceful is the comparison of Nausicaa sporting cheerfully in the circle of her playmates, with Artemis and her train of hunters and huntresses, than that of Dido, who, mid a circle of men, pro- ceeds to the temple of Justice, with the Huntress- * See Bain, p. 20, "Figures of contiguity." Ill;' .* i'l I20 LANGUAGE AND CHARACTER Goddess! When the Roman poets stand on their own ground they do not shrink from repeating them- selves. The comparison of human activity with the restless activity of the bees, which we find in Aen. i, 430 sqq,, is repeated by the poet almost word for word from Georg. iv, 162, 169. Certain similes, as, for instance, where a hard heart is likened to a rock, or to iron, occur quite frequently. As early as Ennius we meet with (Fr. loi) ''quasi ferrum aut lapis durat," and (Fr. 174) ^'lapideo corde": pos- sibly after the pattern of the Greek tragedians (Eurip. Medea, 29, 1279; Andr. 537). Ovid offers similar examples: Met. ix, 613; vii, 32; xiv, 712. Heroid. 7, i^\ Trist. i, 8, 41; iii, 11, 3; iv, 12, 31. In like manner we mark the recurrence of a com- parison of an unfeeling person with some monster of the sea such as Scylla or Charybdis, or with some beast such as a lion or a tiger; such are frequently met with (cf. Catull. 60, i, 64, 154. Ovid, Met. viii, 120; ix, 613; vii, 32). Besides this the poets fall not unfrequently into the fault of heaping simile on simile in a single passage; and they not seldom run the risk of wearying their readers by citing strings of examples. Tl, In one class of figures ot speech the Romans surpass their Greek masters, namely, in allegory, and in the personification of emotions such as Terror, Desire, Wrath. Such personifications are much in favour with authors. Indeed, Herder goes so far as to assign to Horace as a special virtue his personification of abstract, and especially OF THE ROMAN PEOPLE 121 of moral, ideals: e,g,, Odes, iii, i, 14 '' Necessitas sortitur" ("this is a master trait of his genius, and one of the ornaments of his odes "). But surely such personifications were not peculiar to Horace: other writers afford in a greater or less degree examples of the same use of this figure. In Tibul- lus Spes, Pax, Mors, Poena, etc., in Ovid Cura, Alitor, etc., appear as personified beings : and the more closely we scrutinize Roman literature from its origin downwards, the more we find the pro- pensity developed for dead abstraction and cold allegory. Doubt, Hunger, Age, Illness, etc., find full play in Silius, Italicus, Claudius, and his con- temporaries. The Italian too often peoples his Pantheon with bloodless and colourless figures, and similar figures compose a good portion of his poetry. From what has been said it is evident that the Roman poets were not endowed with the vigor- ous imagination or the versatility and cleverness of the Greeks, but that they devoted themselves to the purely intellectual mental processes of reflection and abstraction. Greek poetry is a delightful garden provided wath an abundance of Flora's choicest products, with many-hued and joyous nymphs sport- ing around. Roman poetry resembles rather a well- tended, tastefully laid out, and carefully parcelled vegetable garden. 74. If Plastic* in language serves to bring an * There is no word in English which exactly renders the Ger- man Plastik. Perhaps the nearest is visualization. It means the II 122 LANGUAGE AND CHARACTER object nearer to our view, figures of augmentation and contrast are employed by the poets to magnify such object, and to render it more sensible to our view. Repetitio (Anaphora), Epizeuxis or Epana- lepsis, Gradatio (Climax), Litotes, Hendiadys, Pleo- nasm, Hyperbole, Polysyndeton, Antithesis, Chias- mus, Oxymoron, and many other figures, all tend to the same end. Where the prose writer says *' ubi secuit, in membra redegit,'* it is open to the poet, in order to bring out the speedy sequence of the actions described, to use the pleonastic expression, *' secuit sectamque in membra redegit " (Ovid, Met. i, 33). Again, Vergil, with characteristically epic re- dundance, writes " cavae cavernae" (Aen. ii, 53), " rursus relegens" (Aen. ii, 690), etc."^ This kind of pleonasm is not, it is true, specifically Latin, but it is a prominent characteristic of Roman poetry, and can easily be explained as that of a people who have from the earliest times busied themselves with the study of jurisprudence, and who have accord- ingly accustomed themselves to exact and lucid methods of expression. The same purpose of " raising" is served by the frequent use of concrete nouns in the plural instead of in the singular, which is very common with parts of the human body, such as colla, corda, pectora, etc. ; objects serving for traffic, such as ctcrrtis, arcus, iuga, carinae, and designations of localities, such as litora^ power of presenting an image so that it shall stand out in just perspective and bold relief. * Cf. the Jigura etymologica so often met with in Plautus; e.g,^ "Venus venusta." OF THE ROMAN PEOPLE 123 rttra, sedes, tecta (cf. P. Maas, '' Studien zum poet- ischen Plural," Wolfflin's '' Archiv f. lat. Lex." xii, 479 ^9Q^\ Ed. Hailer, *' Beitrage zur Erklarung des poet. Plur. bei den rom. Elegikern, Freisinger Progr.," 1902, and above, § 27), and similarly the employments of inille, centum, etc., for a number however small. Ordinary mortals may find it neces- sary to reckon with accuracy the sum of certain figures; the singer does not worry himself about such prosaic trifles. He prefers to speak in round num- bers in order to increase his impressiveness: mille lacer spargcre locis is the prophecy uttered to Pen- theus in Ovid's "■ Metamorphoses" (iii, 522), and to the rainbow a thousand hues are ascribed (Aen. iv, 701). No scholar will take exception to such ex- aggerations on the score of his more exact informa- tion, for the store of colours in the broken sunrays can hardly be expressed in a single word more grace- fully than it is here. The figure called Litotes was a very favourite one with the classical poets : it occurs frequently in formulae which have passed from generation to generation; e.g., *'non dissimulator amoris," Ov. Met. V, 61; ''cura non levis," Hor. Carm. i, 14, 18 (cf. C. Weymann, '' Studien liber die Figur der Litotes," Jahrb. f. Phil., Supplem. xv, 1887, pp. 453- 556). The Hyperbole is more effective still; we find it in Vergil employed on a far more extensive scale than in Homer. Sometimes the number or the size is exaggerated, as in the case of mountains, rocks, trees, vessels; sometimes the qualities of human beinors or of beasts — their streno^th or swiftness — If 11 !|l 124 LANGUAGE AND CHARACTER sometimes the power of emotion (cf. R. Hunziger, **DieFigur der Hyperbel in den Gedichten Vergils/' Berlin, 1896). 75. We meet very frequently with allusions to natural monstrosities. Such allusions depend on the Roman taste for strong contrasts and their effects. This taste appears strongly developed as early as the Alexandrian poets, and in composers of idylls, like Theocritus; but it occurs also, though more rarely, in Archilochus (Fr. 76); Euripides (Medea, 410, etc.). The Romans must have borrowed from these models, as is clear from their frequently using identical phraseology. Thus Naevius (Bell. Pun. fr. inc. 1 1) says : " prius locusta pariet Lucam bovem.'* In Plautus we read amongst other passages, Poen. iii> 5» 31 • " lupo agnum eripere postulant," Asin. 99: *'iubeas me piscari in aere," and Asin. 79: ''nudo detrahere vestimenta '* : in Lucretius (v, 128 [and 878]) : " sicut in aethere non arbor, non aequore salso Nubes esse queunt neque piscesvivere in arvis Nee cruor in lignis neque saxis succus inesse.* This con- ception appears again and again in varying forms. The other figures of speech also had become part and parcel of the stock phraseology of the Roman people. Their genius for rhetoric and their forensic training alike rendered such figures indispensable adjuncts even to their poetry. A striking turn for * Cf. also for such pictures Hor. Ars Poetica, 1-5, which seems itself to have been borrowed from Plato's "Phaedr." p. 246 (Jowett's translation). Cf. also Vergil's picture of Scylla, Aen. iii, 426, and of the Triton, Aen. x, 211. OF THE ROMAN PEOPLE 125 oratorical and declamatory pathos manifested itself even in the Roman poets of the first rank, and only too often hollow phrases and empty verbiage took the place of warm and genuine feeling: they strove to mask their shallow thoughts and their lack of profundity by pompous pretentiousness and mere- tricious ornament. 76. Of course different writers have their own peculiarities : Vergil and Propertius display a marked tendency towards parallelism,"^ resembling that found in Hebrew poetry, and they thus enable us to ap- proach an idea from different sides : no one surpasses Propertius in rhetorical questions and in the figure of Apostrophe : the Hendiadys, of which we meet but a single example in Propertius (iii, 4, 9), meets us often in the poems of Vergil. We find the figure QCTTo xoii/ou more frequently in Horace than in other poets. The effective dismemberment of a conception into its parts, or of an occurrence into its separate stages is a characteristic of the technique of Tibul- lus : t Ovid — not to mention the comic poets — is fond of plays upon words. J Naturally all these rhetorical accessories give the * Cf. Postgate's "Propertius, Select Elegies," p. Ixxi. An in- stance is: "sive illam Hesperiis sive illam ostendet Eois, Uret et Eoos, uret et Hesperios." t A peculiarity of Tibullus is that an epithet which belongs to each of a group of nouns is sometimes expressed once only, and then with the last noun, as i, i, 32, "messes et bona vina date," Le., " messes bonas et vina bona." See Postgate, ad loc. X Cf. Met. xiii, " Non oblita animorum, annorum oblita suorum " = " forgetting her age but not her rage," as Simmons renders it. Other instances are Tristia, i, 16 j ii, 16; and iv, 5, 7. Ill ii 126 LANGUAGE AND CHARACTER language the appearance of artificiality. The expres- sion seems too often cold and affected: the verve that springs from the heart in Greek poetry is felt to be lacking. Just as the Romans fell below the Greeks in their power of creating life-like figures out of blocks of marble, so did they miss the secret of drawing living harmonies from language. 77. The third main law of the diction of poetry is Naturalness. The poet may be as childishly simple as Homer, or he may awake pathos as Horace has done in his Odes: in no case should his language suggest the result of deep thought or appeal to the intellect alone. No one in periods of high emotion thinks of speaking in orderly and artificially grouped periods : and in the language of the poet, the logic which marshals facts, the care which disposes them, the intellect which weighs them and calculates their consequences, should remain unseen. The tendency to employ simple and uncomplicated constructions corresponds with the effort after easiness of com- prehension and plainness of expression. The lan- guage of poets moves by preference in main sentences (cf. Aen. i, 402: ''Dixit et avertens rosea cervice refulsit"; i, 438: ** Aeneas ait et fastigia suspicit urbis "). The free use of adjectives {e.^:, Aen. i, 208 : "curisque ingentibus aeger*' = '*quamquam curis in- gentibus aeger erat '') * and the preference for par- * For a more striking instance still see Lucan "Pharsalia," ii, 231 sqq. : *' Neuter civilia bella moveret Contentus quo Sulla fuit ": "Neither Caesar nor Pompey would begin Civil War // they were content with what contented Sulla." OF THE ROMAN PEOPLE 127 ticiples instead of subordinate sentences lends their phrases an impressive terseness: clumsy gerundial constructions are avoided where possible, and final sentences have a tendency to be replaced by an in- finitive. Co-ordination in their sentences is some- times used instead of subordination ; the connections of the sentences therefore resemble a long-drawn chain, in which link is joined to link: while the rhe- torical and historical periods remind us rather of a closely welded ring which fastens all parts, great or small, in orderly and precise sequence in a single and well-arranged whole. Where the prose-writer would say *'ubi corripuere, ruunt," Vergil says (Aen. v, 145): **corripuere ruuntque" (cf. ix, 410: "dixerat et . . . conicit ") : and instead of " cum inversum " we often find ecce {e.g., "certum est dare lintea retro; ecce au- tem,'' Aen. iii, 686).* Sometimes we meet with a simple parataxis as *'iam Lucifer surgebat: cessi," Aen. ii, 801 sqq, (cf. alsovii, 621; viii, 83; ix, 432). A lengthy period of oratio obliqua is suitable enough for the historian, but for the poet it is too ponderous. 78. Though it is, generally speaking, true that the Roman poets have held by the principles men- tioned, there are still many passages in their works which might seem to support the contrary view. Too frequently they succumb to the innate weak- ness of the Roman writers, the habit of moralizing (cf. Aen. iii, 496; iv, 14). The Odes of Horace leave the impression of being constructed to order * Where we should expect some such expression as "When suddenly." Ill 128 LANGUAGE AND CHARACTER from a turner^s workshop. Thus we find these are written with due regard paid to the method of jom- incr sentences In prose: even such conjunctions as are usual in the case of syllogisms, such as ergo and quodsi* are not rarely found in these composi- tions. . . Dovetailing his sentences again is a characteristic trait of Horace: we often find all kinds of paren- thetical insertions just as in the artistically con- structed periods of the historian, so that the poet seems to have written his Strophes rather for the eye than the ear. Of all the Augustan poets Horace stands in his language nearest to the prose-writers. In Vergil's poetry we often find long periods, especially in the speeches of the persons intro- duced as actors; and the elegiac poets have uninter- mittently striven, in order to meet the requirements of the distich, to render their language more and more flexible Propertius was the first to achieve a fair success in closing the thought with the close of the pentameter, t * Cf Lucretius, who abounds with such conjunctions as igitur quandoquidem^ proinde. He, at any rate, never strives to conceal "the logic which marshals facts"; and he is wont to recapitulate the results of long passages in a few short lines-a ^^etoncal trait^ His scrupulous endeavour to be circumstantial, causmg him to repeat such phrases as ut docui, quod quoniam docut, sometimes reminds one of a legal document; another aspect in which he is typically Ro^^^^^^^^^ "Select Elegies,'' chap. iv. "Propertius^ general superiority in vigour and variety to Tibullus appears m their versification. That of Tibullus is hardly ever impressive, and is apt sometimes to become monotonous. Both in hexameter OF THE ROMAN PEOPLE 129 79. The fourth and last quality peculiar to poetry consists in its greater freedom from the restrictions which rule the composition of prose. In the first place the poet enjoys a greater licence in the position of his words than the prose-writer. In the case of modern languages this holds true in certain cases only, but in ancient languages, in which the reten- tion of the full terminations aided quick appre- hension of the meaning, and in which the close relationship of the several clauses could, without trouble, be discovered, the greatest licence prevailed. To emphasize very strongly two connected concep- tions, the poets not uncommonly inserted words so that the adjectival attribute formed the com- mencement and the substantive the end of the verse: indeed, they even postponed the subject, when particular stress was to be laid upon it, to the end of the sentence, and at the same time to the beginning of a verse. For instance in Ovid's Met. ii, 818, the three words ste77ms isto pacto are parted by the words introducing the oratio recta so that the verse runs: '' Steinus'' ait ^' pacto''' velox Cyllenius ^^istoT Again, by placing monosyllabic words at the end of the hexameter the impression of contrast is insisted upon, or some artificial aim attained, e.g,, ** parturiunt montes, nascetur ridiculus mus." ^ In short the poet, by the freedom granted him in the arrangement of his words, has the privi- and pentameter Propertius shows a freer structure than Tibullus, and, we need not add, than Ovid." * "Procumbit humi bos." Cf. also Verg. Aen. i, 105, "Prae- ruptus aquae mons." K I30 LANGUAGE AND CHARACTER lege of a method whereby he attains marvellous effects, assuming that he is capable of employing them artistically. 80. The poet enjoys one further privilege; he can employ archaisms much more widely than the prose- writer; he can overleap many barriers which from a linguistic point of view divide different ages. The historian, the orator, the prose-writer generally, are fairly circumscribed by these. Just as at the present day we seek by every means to maintain and to protect against wind and weather ruins hoary with age and rich in story, which rise from the smiling landscape silent witnesses of ages long past and gone: so do poets, more or less in their degree, aim at rescuing from oblivion the waifs and strays of language which linger in the diction of the ancient singers. The diction of poets is conser- vative : it cherishes and loves antique forms from a feeling of piety and discipline, especially as such forms generally possess a fuller and more power- ful sonority and lend a romantic flavour to the vehicle of verse. Many obsolete words, many forms which in prose are superannuated, and have passed out of use, are again introduced to language from an older period and restored to life. Klopstock has the merit of having introduced anew into the Ger- man language, under English influence, words from older stages of German, as Halle, Hain, Elf, Heim, Harm\2Si^ Uhland has quickened words like Gadem, balcony, Fej^gc, ferryman, pirschen, to stalk game, Wat, garment, Bracke, hound, fahen=fangen, to OF THE ROMAN PEOPLE 131 catch, lobesam, \di\idi2h\e,gemacksam, comfortable, into new, though artificial, life. 81. Horace, true to his maxim: ''Multarenascentur, quae jam cecidere " (Ars Poet. 70), recalls old words like altercari (Sat. ii, 7, 57) and indecorare (Od. iv, 4, 36). Again, words like divus, civicus, and hosticuSy which occur again and again in the Augustan poets, had, except in certain combinations such as divi manes, corona civica, in hostico, almost fallen into disuse. Well-known words meet us again with meanings which had long disappeared from the living language: ^ such as templum (Aen. iv, 484) = refMsvog: aptus, Aen. iv, 482 = fitted on to, armed: quiescere in Aen. iv, 523, and in other places, is used inchoatively after the analogy of other verbs in -sco'. orare stands in its original meaning ''to speak," Aen. x, 96; vii, 446, etc. Old forms of words, too, are saved from total disappearance by the language of the poet. In German the use of certain such words is allowed to poets, but not to prose-writers. Such words are zuruckey gesckwinde, I/er2e=M.H,G, zerucke, (ge) * For several instances of such words see Heerdegen, " Ueber historische Entwicklung lateinischer Wortbedeutungen," Erlangen, 1 88 1. He shows, Part III, p. 18, that the use of orare was already in Plautus' time an archaism, and that the way in which it came to mean to " beg " or " pray " was the fact that orare in the sense of "to speak "^ was commonly joined with Jus and aequum. Cf. Livy, 39, 40, 12, "ipse pro se oraverit scripseritque," referring to Cato the Elder. ^ The older sense remains, of course, in orator. I 132 LANGUAGE AND CHARACTER swinde, herze\ they prefer such forms as wob and ward to webte and warde, and they employ the forms Bande and Lande side by side with Bander and Lander."^ In the same way it is a favourite device with Horace to use words which in their form affect an archaic look : such are cupressus, intMnius, oplumus, proxumtis, lacrumosus, formonsus, thensaurus, lavere, sectarier, gnaius, mi= mihiy caldior = calidior, surpite = surripiiey surrexe = surrexisse \ and Vergil com- mits himself to such forms as olli = Hit, quis = quibus, impete = impeiu, faxo, accestis, accingier, fervere, ceu^ asty etc., all with the view of investing his epic with an old-world colouring. Then the poets use simple instead of compound forms, to excite the imagina- tion of the reader, who has accordingly to puzzle out for himself what is ordinarily expressed by the preposition: thus words like piare, solari, tabere, temnere, linquere, suescere, tender e, etc., have main- tained themselves. [In late Latin grammarians, such as the one who calls himself Vergilius Maro, we actually find forms V^'t sidera = considera^^ Active verbs again appear taking the place of the more ordinary deponents: thus populant (Aen. iv, 403) : on which Servius re- marks " Populant antique dixit ; nam hoc verbum apud ueteres activum fuit, nunc autem deponens * See Abbot and Seeley, "The Diction of Poetry," p. 55. "The antique and venerable associations which connect them- selves with everything that is ancient, contain in themselves suffi- cient reason why archaic words should linger in elevated poetry. From such considerations as these Spenser employed throughout the whole of his ' Faery Queene ' a diction which was almost as archaic to his contemporaries as it is to us." OF THE ROMAN PEOPLE 133 est '' ; and who can deny that the old imperfect forms mollibat (Ovid, Met. viii, 199), nutribat (Aen. vii, 485), lenibat (Aen. vi, 468), and that such forms as saecla, vincla^ oracla, with m slurred and omitted, are stronger and more effective, and hence more fitted to heroic Epic poetry than the corresponding forms common in prose: saecula, vincula, oracula^ Who can deny that the genitive in -urn in the first and second declension, the accusative in -is in the third declension, and the perfect form in -re instead of -runt give the language a more stately stamp ? 82. Often considerations of metre come into play. In German^ the unrelenting bond of rhyme has pro- tected and preserved many an old formation which would otherwise have fallen into oblivion. Roman poetry, however, which makes but spare use of rhyme, has preserved many forms from the fact that they fitted into the strict framework of the dactylic metre. Thus we find that in many cases the long vowels are maintained in verbal and nominal ter- minations. For the same reason Vergil forms the genitive plurals of participial and other noun forms in -ns exclusively in -um instead of -turn, as moder- antum, legentum ; and thus, under the stress of the demands of metre, he selects the old consonantal stems. But the demands of metre suggested other expedients as well : for as Cicero has said. Or. 202 : '* Poetae in numeris quasi necessitati parerecoguntur." Cf. Quintil. i, 6, 2; viii, 6, 17. Vowels again are shortened, lengthened, or suppressed; for instance * And in a less degree in English, e.g^ abideth, guideth. 134 LANGUAGE AND CHARACTER we read in Vergil constiterunt (Aen. iii, 68 1) for constiterunt, relligio iox religio (Aen. xii, 182), aspris for asperis (Aen. ii, 379). The forms imperator and imperare are brought into verse by Ennius [Juvenal and others] by the employment of induperator and induperare\ by Accius and Lucretius by that of imperitare: for magnitudo Accius writes magnitas, Lucretius maximitas [and Auct. Carm. de Phoenic. 1 45, magniHes] : for beneficia Catullus writes bene/acta ; for eloquentia Horace (Ars Poet. 2 1 7) writes eloquium as did Vergil (Aen. xi, 383) ; for supervacaneumWox2s:^ (Od. ii, 20, 24) writes supervacuum (so again Ars Poet. 337; Epist. i, 15, 3); so for the oblique cases of adulter those of moeclms are substituted (cf " Archiv fur lat. Lexicogr." xii, 435)- Then there are certain typical and standing phrases or collocations of words which are handed down from generation to genera- tion, and become current coin : indeed, it may well be doubted whether the Roman poets have not plundered the stores of their predecessors more effectually than those of any other country. Forms found in Ennius like Caernla caeli recur in Lucretius, Vergil, and others, and frequently in the very part of the verse which they occupied in the original. Thus the words " haec ubi dicta dedit," which Vergil borrowed from Ennius, open a verse in Vergil, and have even passed into Livy s prose, in which they open a sentence (xxii, 50). Thus Statins (Silv. iii, I, 15) takes over Vergil's formula **Cernere erat" (rii/ \h^iy) and uses it at the opening of a verse: thus again there occurs an Epic formula conditioned by the metre, in the case of the perfect participle OF THE ROMAN PEOPLE 135 passive coupled with the oblique case of a word of two syllables occurring at the end of a verse, as "dilecta sorori" (Verg. Aen. iv, 31), ''obsessacolono" (Tibul. iv, I, 139), *' Exterrita somno," ''concita cursu," etc. Thus Valerius Flaccus, after the model of Aen. vi, 273, '^primis in faucibus Orci," writes "primis stant faucibus Orci," i, 784; after Aen. viii, 25, '^summique ferit laquearia tecti," v, 243 "per summi fulgor laquearia tecti" (cf A. Gruneberg, "De Valerio Flacco imitatore," Berlin, 1893, P- 52» sqq). Few nations feel the influence of tradition and imitation so strongly as the Romans; in few is individualism so feebly developed. 83. Finally we have to range under this head syntactical archaisms. As such are to be counted the use of the simple accusative and ablative in answer to the questions whither and whence, in the case of words which are not names of towns ; and again the dative of the direction whither [" It clamor caelo " Verg.], which has maintained itself in the language of the poets, especially in the case of such common conceptions as Heaven, Orcus, earth, sea, Olym- pus, etc. 84. On the other hand innovations, or neologisms, appear in the language of the poets. These new turns given to language manifest themselves in the formation, the signification, and the syntax alike of words. We remember that Horace proudly claims the right of the poet to enrich his native language, " Ego cur, acquirere pauca Si possum, invideor, 136 LANGUAGE AND CHARACTER cum lingua Catonis et Enni Sermonem patrium ditaverit et nova rerum Nomina protulerit? Licuit semperque licebit Signatum praesente nota producere nomen" (Ars Poet. 55), and '* adsciscet nova, quae genitor produxeritusus" (Ep. ii, 2, 119). And without a doubt most of the Latin poets have availed them- selves of this right. As has been remarked before, the Latin writer felt the lack of compound adjectives, those almost indispensable auxiliaries to the poet for the embellishment of his diction. Hence since Ennius it was the aim of poets to supply this need as best they could. It is not unlikely that altitonans was a word coined by Ennius, arcitenens by Nae- vius, magnisonus by Accius, frugiferens by Lucre- tius, suaveolens by Catullus, blandiloquens by Lab- erius, auricomans by Vergil, centimanMs by Horace, racemifer by Ovid : these words appear for the first time in their respective works. But it would take us too far were we to attempt to submit all such ex- pressions to close scrutiny: so we content ourselves with pointing out a list of such similar formations as Ennius himself offers us. We find besides ^//Z- /^^^«^ mentioned above: velivolis A. 381, saxifragis A. 564, altisonus A. 561, bellicrepa A. 105, caeli" colum A. 483, doctiloqtti A. 568, dulciferae A. 71, Jlammiferam Tr. 50, nwrtiferum Tr. 363, opiferam Tr. 165, lanigerum Sat. 42, belligerantes A. 201, altivolans A. 84, bellipotentes^ Sapientipotentes A. 188, omnipotens Tr. 202, bipatentibus A. 62, blandilo^ qMentia Tr. 305, signitenentibus Tr. 132, velivolan- tibus Tr. 89. On the other hand, we must not pass over the final portions of the composite words most OF THE ROMAN PEOPLE 137 frequently in use, especially as they lend the words their typical stamp, and set the stereotyped patterns of word-formation to which the following genera- tions of Roman poets conform : for the latter made it their object not so much to find new derivative syllables, as to connect these with new word-stems. The following are the principal of these: sonus, loquusy volus, genus, fragus, comus, ficus, dtcus, sequus, rapus, capus, legus, fugus, petus, paruSy graduSy spicus, vagus, premus, vomus, iugus, terus, creptiSy fer, ger, canens, potens, parens, volans, manus, color, modus, etc. The poets of the Augustan and of the post- Augustan periods followed the precedent of the older poets, so that a large number of new formations arose, modelled on the pattern of those already in use. Thus, to quote a single example, Latin literature displays about 1 70 compounds ending in fer, and about 80 in ger, of which the following make their first appearance in the Aeneid: calli-, com-, fait-, fumi-, legi', niali-, olivi',paci', somni-, sopori''fer\ alt-, turri-ger; while Ovid shows 29 new formations in fer, and 9 in ger, which the following words seem to be employed by him alone : aerifer, alifer, artm- difer, bipemiifer, caducifer, chimaerifer, corymbifer, cupressifer, gramifer, herbifer, papyrifer, populifer, racemifer, sacrifer, securifer^ taedifer, tridentifer, turrifer\ bicorniger, penatiger, tridentiger. 85. Composition was not, however, the only pro- cess whereby new words were created; derivation played its part as well. In this process also Cicero 138 LANGUAGE AND CHARACTER assigns greater liberty to the poets than to the orators. He writes, Or. 20, 68 : " Ego autem, etiamsi quorundam grandis et ornata vox est poetarum, tamen in ea . . . licentiam statuo maiorem esse quam in nobis faciendorum iungendorumque ver- borum." Thus Horace forms from cinctus the ad- jective cincttihis (Ars Poet. 50), from iuvenis the verb iuvenari (Ars Poet. 246), from ampulla, am^ pMllari{Ep. i, 3, 14); Vergil among others gestamen, affatusy latrator, nimbosMS, fumetis, cristatus, crinalis, stridulus^ sternax,acervare\ Ovid is particularly fond of coining new adjectives in -alis, -abili:, -ens, -osus, and verbal substantives in us of the fourth declen- sion, as well as in -amen and -imen, which lend themselves better to the exigences of metre than those in -atio and 4tio, e,g., pacalis, agitabilisy dubita- bilis, narratus, simulamen: Martial has celebrator, dormitor, esuritor, panariolum, Greek terminations, too, are attached to Latin stems, and in this way hybrid stems were created as Sciptddes (Lucr. iii, 1032, etc.), Memmtddae (Lucr. i, 26), Stoicidae (Juv. ii, 65). 86. Further, the poet possesses an inexhaustible source of novelties in the domain of word-significa- tion. In this process he may give free rein to his fancy: he may exhibit his poetical genius in the most brilliant way, " Dixeris egregie, notum si callida verbum Reddiderit iunctura novum" (Ars Poet. 47). Horace himself gives in the same Ars Poetica, verse 49, an example of this maxim in the use of indicium. Again, such terms as corripere OF THE ROMAN PEOPLE 139 viam (Verg. Aen. i, 418) are new, as are exigere "to beg news'* (Aen. i, 309), 7nemorare (Aen. i, 631), resequi{Ov. Met. vi, 36) ''to answer": the important question of metaphors, too, comes into considera- tion in this connection. 87. Side by side with these changes, the syntax of the poets was enriched by a larger number of new constructions. They often seem purposely to make a new departure from the methods of prose- writers: otherwise what reason was there for changing the moods followed respectively by qua7n- quam and quamvis, and connecting the latter with the indicative, and the former with the conjunctive? For what other possible reason could Catullus, Ti- bullus, Propertius, and Horace have purposely avoided tttrum . . . ait, and have substituted an , . . an, ne . . . ne, or Vergil have written sen . . . sen (Aen. i, 287, etc.), requirunt . . . sett vivere credent, sive extrema pati} In most cases such novel methods of expression are analogical formations after ancient Roman or Greek models, though it is often hard to trace the exact source of the thought that in- spired the innovation. It was once the fashion to explain these new phenomena in language as due solely to Greek influence; at present there is an inclination to fall into the opposite mistake, of referring these wherever possible to old Roman methods of speech. Probably the right path lies mid-way. There can be no doubt that the Greek language in many cases gave the impulse, and there can be no doubt either that this impulse I40 LANGUAGE AND CHARACTER was followed more readily when old Roman forms of language were at hand to support it: in other words when the "feeling for language" was not outraged. The fact that verbs expressive of willing admitted a simple infinitive to follow them [as in Romance] is explained by the analogy of iubere, vetare, and other verbs which admitted of such constructions from the earliest times : at the same time it is probable that Greek influence was a factor in this construction. Less doubtful still is it that Greek influence was at work in combinations like maior videri ** more stately to behold/' niveus videri, "white to look at" = /x£/^ajv, Afuxo? I<^fV6ai [cf **vultus nimium lubricus adspici/' Hon Od. i, 19, 8]: "cernere erat/' e.g., in Aen. vii, 596, reminds us of h \$i!v: "quem virum aut heroa lyra vel acri Tibia sumis celebrare Clio?" of alpntr^oci, $i$ovxi, with the infinitive : " Pelidae cedere nescii/' **puer dignus cantari/' remind us of Uuvoq and i^Loq with the infinitive. More manifest still is foreign influence in places like Catullus, iv, 2: " phaselus ille . . . ait fuisse navium celerrimus," or Vergil, Aen. iv, 305 : " dissimulare sperasti "; in these cases the true Latin feeling for language would lead us to expect the accusative and infinitive. In the same way constructions like " sensit delapsus " (Aen. ii, 377), or " gaudent scribentes " in Horace, Ep. ii, 2, 107, remind us of Greek constructions like x^'^?^ ocKov(ro(,g : but more than all the infinitive of the perfect used in the sense of the present in- finitive — as in Propertius, i, i, 15: **ergo velocem potuit domuisse puellam " : and in Tibullus, i, 10, 6 1 : OF THE ROMAN PEOPLE 141 " sit satis . . . rescindere vestem, Sit satis ornatus dissoluisse comae," cf i, 1, 29, 45, and M. Haupt in Belger's " Biographie," p. 233. The treatment of cases is not unlike that of the treatment of moods. The so-called Greek accusative and dative, which belongs chiefly to Roman poetry, and, as its name indicates, was referred exclusively to a Greek origin, existed even in old Latin. This fact could not but encourage later poets to employ on a larger scale the construction which was so much favoured in Greek. Hence Ovid employs this dative more commonly than the ablative with ad, and it occurs in Silius Italicus about a hundred and fifty times as against twenty cases of the ablative employed with ad. On the other hand certain phrases appear to be direct copies of Greek idioms: such are desinere querelarum (Hor. Od. ii, 9, 18), desisiere pugnae (Aen. x, 44i) = a(piVTa(r9ai' xiko?, solvere operum (Hor. Od. iii, 17, i6) = a7roAu£ii/ Tii/e?, mirari laborum (Aen. xi, 126) = OaujU^a^ai/ livd Tti/o?, and again regnare populorum (Hor. Od. iii, 30, 12), and cupere alicuius in Plautus (Mil. 964) may be formed on the analogy of apx^n/, \'TrA\}\t.iiy\ though the construction regent, cupidum esse, may have sug- gested them: cf eius videndi ctipidtis in Terence, Hec. iii, 3, 12. And when Horace, in the passage quoted from the Ars Poetica in § 86, in speaking of the enrichment of language by the poet [Ars Poet. 56] writes invideor for mihi invidetur, it is obvious that he is copying the Greek (pOovoj^ai (from (p^Qvily Tin). [Of course the exigences of metre had here to be considered. Cf too the construction of 142 LANGUAGE AND CHARACTER imperor, Epp. i, 5, 21, and in Verg. Aen. ii, 247, " non unquam credita Teucris "]. 88. These, then, are the main features of the diction of the Roman poets as exhibited in their works. They convince us that these poets have worked with plenty of goodwill and honest effort, but that their strength was no match for that of the Greeks, and their language again could not com- pare in elegance with that of their Hellenic teachers. The assertion may fairly be made about the Roman poets which Lessing, at the end of his ** Ham- burgische Dramaturgie," makes about himself: viz., that these poets have no eye for the living source which by its innate power springs upwards with rich, fresh, and clear rays: but that they find themselves constrained to squeeze their outpourings from them- selves by dint of water-pipes and pressure. Even the most honoured bards of the grand era of Au- gustus were in the main gifted with talent rather than genius. While Horace says : " Graiis ingenium, Graiis dedit ore rotundo Musa loqui,*' it must be admitted that the nation which called even the set form in which war was declared a '' Carmen " was by nature rather intended for prose than verse, and that it has indeed attained to a high pitch of eloquence in oratory. But it is not alone in the bent of the Latin national poet^ that we have to look for the faultiness of their expression, but in the essence of the Latin language itself. This language was a hard metal, only to be worked by dint of much toil and pains, and it justified the complaint OF THE ROMAN PEOPLE 143 made in ''the legend of Pilate" regarding the Ger- man language * : its toughness renders it an unfit instrument for poetry, but it must be treated like steel which is hammered on the anvil till soft; it ' requires toil and labour to render it malleable. * For *' the Legend of Pilatus," see " Die geistliche Dichtung des Mittelalters," Zweiter Teil, " Die Legenden und die Deutsche Ordensdichtung," bearbeitet von Prof. Dr. Paul Piper, Berlin, Spemann, p. 24. IV The Language of the Latin People 89 ON L Y a few years ago the conviction was pre- valent in Germany that the language of the people had, by a process of mutilation and decay, developed out of the language of the educated classes. This view is at the present day superseded, mainly owing to the works of Klaus Groth, who has shown by irresistible proofs that dialect is not a caricature of cultivated language, but is in fact the marble block out of which the language of culture is hewn. The views of scholars have come to a similar conclusion with regard to the popular dialect of the Romans. The conviction is forced upon us that the relation- ship of daughter and mother, by which it was cus- tomary to illustrate that of vulgar to cultured Latin, is in this case inapplicable. Vulgar Latin cannot indeed have taken its rise by the simple process of vulgarizing the idiom of the better educated classes; rather are both idioms to be regarded as the child- ren of a common mother, viz., Old Latin.* They * "What we call Vulgar Latin is the speech of the middle classes as it grew out of Early Classic Latin. It is not an inde- pendent offshoot of old Latin; it continues the Classic, not the primitive, vowel system. Neither is it the dialect of the slums or of the fields; grammarians tell us of not a few urban and rustic 144 LANGUAGE AND CHARACTER, ETC. 145 thus are related collaterally, and neither preceded the other, but they lived side by side. At the same time it remains doubtful whether they were locally separate, i.e,, whether one idiom was prevalent in Rome, while another was spoken in the Latin dis- tricts (cf. B. Maurenbrecher, " Jahrbiicher f. Philol.,*' 1892, p. 204), or whether we are justified in suppos- ing with Schuchardt that the degree of education professed by the speaker or writer was responsible for his linguistic usage. Between the two extremes — the written language and the popular — stands the language of conversation, for which we may regard Cicero's Letters and the Epistles and Satires of Horace as our main authorities. Just as Quintilian, the erudite professor of rhetoric, loved to discard the stately stiffness of the language of the professorial chair (xii, 10, 40) and employed that of the sermo cotidianus {consuetudo), so does Cicero express him- self "Quid tibi ego videor in epistulis? Nonne plebeio sermone agere tecum ? Epistulas vero coti- dianis verbis texere solemus" (Ad Fam. ix, 21, i). One of the most characteristic examples of this familiar conversational language is to be found in Cicero's letter to Atticus (i, 16), with its loose con- nection of sentences, its terse and sketchy style : its ellipses, puns, and proverbial turns, its exaggera- tions and its frequent emphatic asseverations. How- vulgarisms that are not perpetuated in the Romance tongues. It is distinct from the consciously polite utterance of cultivated society, from the brogue of the country, and from the slang of the lowest quarters of the city, though affected by all of these." — (Grardgent, "Vulgar Latin," § 3; cf. also Olcott, "Studies in the wore formation of Latin inscriptions," Rome, 1898, p. xi, § i.) L \ 146 LANGUAGE AND CHARACTER ever, the most important sources of vulgar Latin are the writings of the Patristic Fathers, the Ro- mances, the Comedies; and also writers on special subjects, like Vitruvius and the later Jurists, the writings of Petronius,^ the " Bellum Hispaniense," the ** Bellum Africum," etc. 90. At the time when Roman literature came into life, the popular dialect had already suffered con- siderable losses in respect of its sounds. The ter- minal sounds of words were particularly exposed to such atrophy : the d of the singular in the termina- tions dd, od, ed, id, etc., had fallen away, m, s, and / were in process of disappearing {see Corssen, *' Voca- lismus," i, 294) : Vowels were abbreviated or cast off, in medial syllables they were syncopated, or again were inserted to avoid harsh sounds. All these changes owed their origin mainly to the conditions of the pitch accent. For the more strongly the accentuated syllable was uttered, the less power of articulation remained for the unaccented syllable which followed it, and this was accordingly more or less mutilated. » Other readjustments followed : m and n, when they preceded their kindred labial or dental sounds, lost their ancient force and were sometimes not pro- nounced at all, sometimes pronounced less forcibly. In the same way the contraction of diphthongs into simple sounds was noticeable. The sounds ez, eu, on, ai oi had already, in the '' Prisca Latinitas," shrunk to t, A ae, and oe, but now ae sunk to i, and au to S {e.g., sodes = si audes). It is to this sound-change that OF THE ROMAN PEOPLE 147 the gens Plotia and gens Clodia, which hived off from the Plautii and the Claudii, owe the form of their name. In certain cases this weakening process has spread even to the classic language, e.g., in ex- plodo compared with applaudo, and in lotus with lautus. The uncertainty which even educated Ro- mans attached to the pronunciation of the au in Cicero's time is well shown by the illegitimate in- trusion of this sound in the word origa (from oreae, i,e., kabenae-. the bridle which drags at the mouth). For even assuming that the form auriga owes its form to popular etymology which refers it to aureus or to auris, still when used by the educated Ro- man it makes on us the same impression as the form Kausche7^ for Koscher in the mouth of the half- educated German [there was the same tendency to pronounce osculunty ausculuTn], Both changes are referable to the efforts made to avoid plebeian pro- nunciation, and to ignorance of etymology. 91. From the beginning of the first century on- wards the confusion spread ever wider and identified the pronunciation of v and b (hence the French avoir = habere), of ^ and x (hence O. Fr. sa^nii, velvet = E5a>*To;/, from %l and ^urro?, six-threaded stuff), of i and e, of u and 0, while ct, pt, so, in medial syl- lables, are often reduced to tt and ss, and in the case of words commencing with s and consonant, the opening sound, or anlaut, was preserved by the sub- stitution of an inserted vowel (hence French itait, O. Fr. estait = stabat, and ipie = espee = spatha). In most of these changes it is obvious that a dis- 148 LANGUAGE AND CHARACTER position to ease pronunciation and a desire to spare trouble assert themselves. The masses like to save their breath; they are shy of long words, and where they meet with sound groups hard to pronounce, which they cannot manage to employ off-hand and with ease, they simplify them, and thus suit them to their articulation. 92. This trait is also markedly prominent in verbal inflexion. Ordinary persons are not prone to prolonged reflexion : they do not trouble to master the variety and multiplicity of inflexional forms; they are averse to a multitude of nominal and verbal endings. They are content with the differentiation of the word-stem comprising the meaning of the word, and they drop the terminations as soon as possible; these are, after all, of merely secondary importance. Nowhere has analogy such large and wide play as in the language of the people; no- where is the tendency towards a certain definite uniform model so marked. Thus the strong (?>., consonantal) conjugation has suffered considerable losses at the cost of the denominatival in -are, -ere, and 'ire. Not merely is the future in most of the verbs formed in -abo, -ebo, and -iboy but many verbs pass wholesale into the vowel conjugation: instead q{ fodere, consternere, spernere, we find fodare, con- sternare, spernare: the form nioriri so common in Plautus ( = Fr. mouriry^ for the classical form mori, has even found its way into the Metamorphoses of Ovid, xiv, 215. Reduplication, so seldom found in * Ital. morire. OF THE ROMAN PEOPLE 149 classical Latin, almost disappears, so that curri takes the place of cucurri. 93. Similar phenomena are also to be seen in the case of the declensions. A large number of con- sonantal stems have passed into the first or second declension by the addition of a or 0. This holds good of foreign words also, e,g., Crotona = Croton, Troezena = Troezen, Hellada = Hellas, lampada = lampaSy aulona = aulon, onycha = onyx, as well as of genuine Latin words; e.g,, of Cassida = Cassis, retium = rete, etc. The Greek neuters in -ma, and neuter -s stems in -us were treated more simply still by analogy with the termination of the nominative case; they were treated as feminine nouns of the first, or sometimes as masculine nouns of the second; diadema, diademae\ plasma, plasmae] tempus, tempi \ corpus, corpi] hence we get Italian plural forms like tempi, e.g., in the proverb tempi passati. In other cases the genitive case gives the impulse to the change: hence we find nominatives like lactis and falcis substituted for lac and falx. A remarkable uniformity established itself in the proper names belonging to the masculine as well as the feminine p-ender : most of these assumed the metaplastic forms o ... in 'tis and -nis] more particularly nomina propria m -es, -as, 'is, -os, -e, and -a\ Agatkoclenis (nom. Agatkocles) ; Niceronis (nom. Niceros) ; Hermionetis (nom. Hermione)', Felicianetis (nom. Feliciana). Irregular case-forms, such as those in -ius and -i — the genitive and dative of the pronominal second declension— were for the most part discarded and ISO LANGUAGE AND CHARACTER replaced by "regular" forms; e,g., totae = totius, nullo = nulli. Generally speaking, exceptions in every form were banned from use : thus the mascu- lines of the third declension in -isy such zsjinis and pulvisy became feminine under the influence of this termination, which is mainly characteristic of femi- nines: hence the French la fin and la poudre — neuters crumbled away in large numbers : they were mostly converted into substantives of the masculine or feminine gender, a circumstance which has led to the almost complete disappearance of the neuter in the Romance languages. It is easy to understand that this concentration of the genders was greatly helped by the disappearance of the terminal con- sonants; if 'tis and -tmi in the second declension were pronounced in the same way, it was not a difficult process to reduce the words of the second declension to uniformity in gender also; in which process the stronger masculine gained the day. 94. As in its inflexions, so in its word formations, vulgar Latin exhibits a strong tendency to uniformity. Thus the adverbial termination, -iter, which in classi- cal Latin is almost exclusively employed for deriva- tives of adjectives of the third declension, spreads to those of the second declension, as aequiter, amoeniter, amiciter (oi, Osthoff, ** Archiv fur Lexiko- graphie," iv, 455 and 99), Neue, '* Formenlehre," ii, 2, The following terminations were much favoured: -monta, -monium [tristimonia, fniserimonium), -ma {collina, calcinay lapsina)^ -nienitim {lustramentum. OF THE ROMAN PEOPLE 151 odoramentum, decoramentum), -ela {fugela, luela), -ntia (nascentia, crescentia, resonantia)\ and again, personal names ending in -o and -onis are in constant use, such as agaso, balatro, caupo. Adjectives in 'His, -bilis, -eus, -aster are as plentiful as leaves in Vallombrosa, cf. Wolfflin, " Archiv fur Lexiko- graphie," xii, 419): -idus is also a very common termination : and we find many so-called factitive words, especially such as come from adjectives in ficus, such as magnificare and pacificare, and -idus, such as frigidare, candidare. Inchoative verbs are also extraordinarily popular in vulgar Latin (see K. Sittl. " De Latinae Linguae Verbi Inchoativis," " Archiv fiir Lexikogr." i, 465-532); and these have multiplied with interest in the Romance languages, and notably in Italian. Verbs in -illare are also favourites (cf. A. Funck, loc. citat. N. 68, 223 sqq), as are desideratives in -urio, which it may be noted are avoided by Quintilian, Tacitus, the younger Pliny, and also by Livy (who has only the form parturio) ; but such forms occur with great frequency in comedy, satire, letters, in Petronius, Martial, and Apuleius, while they have almost disappeared from the Romance languages (vide loc. citat,, i, 408 sqq.). Finally there are certain verbs derived from super- latives like approximare, tdtimare, infimare, which seem to be a special characteristic of African Latin (vide loc. citat., ii, 355 sqq.). It may be argued that these features of vulgar Latin seem to imply a cer- tain monotony and uniformity; still, we cannot over- look the fact that the luxuriant prodigality and the ultimate triumph of such new formations are evidence 152 LANGUAGE AND CHARACTER of a vitality and propelling force of language quite foreign to the genius of classical Latin which, like other literary dialects, remains artificially barri- caded against outside influences. At the same time, the terminations mentioned testify that vulgar Latin prefers strength and weight to weakness and lack of energy: tristimonia is fuller toned and more effective than tristitia, miserimonium than miseriUy duriter than dure. It is also worth while remark- ing that these forms, like others, seem to have de- veloped differently in different localities, e.g,, the abstract-suffix -itia {-ezza) was much used in Italy, while Spain prefers -ura, and France, at least in early times, -tas {sanU = sanitatem), (Cf Meyer- Lubke, **Archiv flir Lexikogr/' viii, 313-338, especially p. 336).* 95. We may naturally expect that the syntax of vulgar Latin will in its turn afford plenty of ex- amples of a tendency towards uniformity in the shaping of constructions. The vulgar dialect mani- fests a clear effort to simplify the existing relations of a complex sentence. The ablative absolute gains ground at the expense of the verb with the con- junctive particle, and, in the place of the accusative and infinitive, sentences with quod appear with in- creasing frequency. As early as Petronius and the author of the " Bellum Hispanicum " we find traces of this change: at a later period it manifests itself very strongly in * See Olcott, "Word Formation," pp. 75, 80, and Grandgent, " Introduction to Vulgar Latin," p. 20. OF THE ROMAN PEOPLE 153 the writings of the African Fathers, especially of Tertullian : of the poets, Commodian was the first to adopt it. In French quod (in the shape of que^ **that") has almost completely displaced the old construction (cf. G. Mayen, " De Particulis Quod &c. . . . positis," Kiel, 1889, and - Archiv fur Lexikogr." viii, 148). It also happens that verba sentiendi et declarandi are parenthetically inserted or merely connected paratactically, according to the usage in modern languages, " You are ill. I fear," *' Tu es malade, je le crois." As early as Plautus, and also among other old Latin authors, we find this usage attached to the following words obsecro = amabo (cf. Lindskog, " Quaestiones de Parataxi et Hypotaxi apud Priscos Latinos," Lund, 1897, pp. 7 sqq?). The so-called dubitative subjunctive gives place more and more to the indicative : " cui dono hunc librum?" takes the place of the classically regular '' cui donem ? " Many impersonal verbs are treated like personal ones : paenites stands instead of te paenitet. In the speech of the educated, where the words alter, quisque, unuSy uterque are employed, the substantive is commonly attached in the same case ; so in the lingua vulgaris with maxima pars (homines), etc. As early as Cato we meet with accusatives like id genus, hoc genus, omne genus, in- stead of an attributive genitive with a substantive, e,g., "librl huius generis," "libri eius modi" (see Schmalz '' Lat. Syntax," in J. Muller s '' Handbuch," ii, 274). ' 96. Even in the matter of word signification, the / li* 154 LANGUAGE AND CHARACTER tendency to consult convenience is clearly to be seen. There are certain wide receptacles into which everything possible is packed. Such receptacles are words of quite ordinary signification, which are in every one's mouth, and which come ready to hand at a moment's notice. Such is the word machen in German. Whoever wants to travel to Berlin macht (is making) for it: commercial travellers make (are dealing) in cigars : a common greeting is " Was machst du?" For ''to open " and **to shut," the German idiom is " to make open " and " to make shut": for to blame "to make lower," for to split wood '*to make wood," etc.^ Similarly in Latin, fdcere in the vulgar idiom signifies (i) aestimare: (2) to travel, se facere Romam: (3) as a medical term curare: again (4) cacare and (5) coire\ (6) sero facit — the French il se fait tard: f (7) nunquam facit tale frigus (L. Augustin, serm. 25, 3) — Una jamais fait aussi froid. But it is particularly used in connection with an infinitive, e.g,, stomachari me fecistiX or in connecting words like lique-facere so as to form factitive words, in which Latin is some- what defective. From Lucretius to Ovid this usage is rare, but in Tertullian, Cyprian, and their contemporaries, it is very common (cf. Ph. Thielmann, ''Facere mit Infinitiv"; *'Archiv fur Lexikogr."iii, iiyff.; Deecke, * Cf. the uses of the English "to do" in "How do you do?" "do you see?" "to do up," "to do honour to," "to do away with," etc. t ^o facit se hora quinta, Bechtel, 126, quoted by Grandgent, §114. X Cf. "ecce Pater fecit Filium nasci de vergine," ib., § 117. OF THE ROMAN PEOPLE 155 ''Facere undf eri in ihrer Komposition mit anderen Verben," Strassburg, 1873). The same tendency is manifest in the treatment of substantives. Many conceptions occurring in the daily life of the ordinary man form the starting-point of new terms which are, in fact, simply adjectival attributes used as substan- tives. Thus there were different kinds of vesteSy such as a/6ay nigra, dalmatica : and each of these epithets was used as a new substantive. As the connection in which these words were used excluded any pos- sible misunderstanding, and as, in addition, the meaning and gender of the adjective indicate the way in which the word is intended to be understood, the substantive was for convenience sake merely dropped. In this way arose the numerous ellipses in which the vulgar idiom delights, e.g.Jerina.por- cina {caro), tertiana, quartana {febris), dectima (pars). 97. Finally we have to remember the borrowed words in Latin, for in these the popular desire for convenience and ease appears in a very marked way. The educated portion of a nation frequently imitates with elaborate conscientiousness the pro- nunciation of a foreign word introduced into their language, and faithfully reproduces all its sounds. Not so the masses: they follow the promptings of their own mind. For the plain man, no peculiar sanctity attaches to these strange words ; no law of the Medes and Persians forbids his remodelling them or changing them at his caprice. In their sounds and combinations of sounds no two Ian- 156 LANGUAGE AND CHARACTER guages are exactly similar; sounds assume different characters to suit each nation's idiosyncrasy. Hence it IS often a matter of difficulty for the borrowing nation to reproduce the borrowed expressions in their correct pronunciation. But the people have no great scruple in shaping anew, to suit the require- ments of their own language, what occurs to them as harsh; in some cases by dropping certain sounds, in others by modifying unmanageable sounds into more familiar ones. It follows as a matter of course that those words suffer the most mutilation in which the phonetic differences of the two idioms are most marked. ** All languages,'* says Jacob Grimm in the Introduction to his German Dictionary, p. xxvi, *' if they are in a natural and healthy state, possess an innate tendency to exclude foreign elements, and if these persist in intruding, to oust them again, or else to identify them with native ele- ments. No single language is capable of giving ex- pression to all possible sounds, and all languages reject such as are unnecessary, finding them a mere incumbrance. If by any chance a foreign word falls into the current of a language, it is tossed and pitched till it takes the same hue, and, in defiance of its alien stock, looks like a native product." 98. The terminations of words like Ulixes = '0$\jYiyiovy a cleft) suggested a connection with regius, ''Royal town": percontari, from contus a pole, to explore the depth of water, was perverted into percu7tctari, and connected with ctmctus: and if palma, the palm, is borrowed from the Phoenician tamar, tomer,^ with anlaut as \n pavo = Tcx,m,^ the notion of the flat hand contained mpalma may have contributed to this result. " The game of Troy," so popular in Rome from Sulla's time down to that of Nero, which seems to have derived its name from the word troare or truare = o-aArjo-ai [properly to move with a trua or trowel], was in the time of Augustus fancifully connected with the town Troja, whence the Julian dynasty drew their origin. ^i The name of the aborigines of Italy is probably * Or Padmar; cf. Palmyra, Tadmor. t Both Oriental loanwords. X Cf. the derivation of the French truie^ a sow, from Troja = the pregnant sow, suggesting the Trojan house full of armed warriors. M / i62 LANGUAGE AND CHARACTER a mere transformation of a word less understood, Aurunci = AusoniciJ^ The construction of the TuUianum, the well-known subterranean state prison of the Romans, was ascribed by the Roman legend to Servius TuUius. As a matter of fact the word comes from Tullius, a spring or source, and it indicates, originally, the spring of water in that prison. The quarter of Rome called Argiletum, mentioned in Aen. viii, 345, was commonly alleged by the ancients to have received its title from the fact that a certain Argos had found his death there {Argi'letum)\ but there is no doubt whatever that it takes its name from the clay pans in the vicinity, Argiletum from argilla,\ We are expressly told that the names of the towns Nequi- num and Epidamnus, owing to the ill-omened sug- gestions of neguam and darnnum, were changed into Narnia (Nar-town) and Dyrrhachium. The myth of the nursing of the twins Romulus and Remus by a wolf is to be explained not by the fact that the wolf was sacred to Mars, but solely by the similarity of the two words rmna, rumis, rumen (udder), and Rumo, the oldest name of the Tiber and of the city of Rome (7??^;;/^ = stream, cf. pay; Roma = Stream town), with Romulus = son of Stream- town. By this means the origin of the stubborn spirit and the unbridled strength of the Roman people are at once symbolically denoted .J * Fredegar renders the German proper name Wintrio by Quintio. t Cf. the name Tuileries. X Diez thinks that the mid-Latin cecinus, a swan, got its name OF THE ROMAN PEOPLE 163 loi. To the tendency towards clearness and ease of apprehension we may further ascribe many other properties of vulgar Latin. The masses prefer in- direct expressions and high-sounding, even strongly exaggerated words. The man of the people loves to fill his mouth with such expressions (cf. J. Egli, '' Die Hyperbel in den Komodien des Plautus und in Ciceros Briefen an Atticus, drei Gymnasial-pro- gramme von Zug," 1891-1893). Every kind of exaggeration in language, such as pleonasms, ad- verbial expressions, derivations, intensives, and com- position of words with particles of augmentative force, enter into his utterances : coepi, with the infinitive, replaced the so-called ingressive Aorist, as " clamare coepit," "he burst into a cry"; for simul and minquam they preferred to say mio tempore, and ntcllo tempore ; also instead of noctti and mane, nocturfio and matutino tempore\ for 7i07i, melius was often preferred, e.g., *' is nullus venit." For emere, the word cofnparare (Italian comprare) came into use as early as Plautus, and adcaptare (French acketer) at a later period; instead of disce7'e they preferred to say apprehendere and imparare. The periphrastic phrases with dare 3.nd /acere cum adiectivOy in place of the simple verb, were favourite methods of expression. A tendency to pleonasm is also manifested by the usage of fui, ftieram, fuero, for sum, eram, ero in the passive composite moods; and in the con- nection of the present participle with esse, e.g., amans from ctcer, with reference to the excrescences on its bill. See Palmer, p. 238. ili i64 LANGUAGE AND CHARACTER est instead of simple verb, amat. The regular addition of the personal pronouns ego, tu, nos, vos, to the verbs, even in unemphatic positions, gives a greater fullness to the language; while the strength- ening genitives,^^v////^w, lociy locortim, terrartim,Qtc., where places are defined, as in ubi gentinm, lend greater force to the language employed. Needless to say, such drastic expressions as/^r abeas, instead of the simple word abi, the more circumstantial ncscio qtiis for aliqnis, and the more emphatic tanien' etsi^etsi are in perfect accord with the tendency of the ordinary man to express himself with em- phasis. 1 02. The strength of exaggeration manifests itself with peculiar frequency in Latin negations. It is in- credible how many changes it is possible to ring on this theme; how many variations the fancy of the common people can bring into play. In classical Latin, as we all know, two negatives cancel each other, or, it may be, result in making an affirmation stronger; but in popular Latin, as indeed in the common German idiom, in old English, and in Greek, the multiplication of negatives is conceived solely as a method of strengthening an affirmation. And is there any possible object so insignificant as not to have been utilized for the purpose of denoting absolute nothingness ? The German, to emphasize his negations, can say '• not a hair," *' not a farthing," *' not a rush," ** not a copper," ** not an idea," " not a bean," '' not a try," ** not a trace." The Frenchman can say ne . . .pas, OF THE ROMAN PEOPLE 165 "not a step"; ne . , , point, **not a point" {punc- turn), and n(^ant (non ens), etc.^ Thus we cannot object to the Roman if, besides nihil^ne-hilum, " not a thread," he employs ideas such as no7i nauci, fiocci,pili, assis, tertmcii, hettae, etc., dihcr facer e, in the sense of " valuing at so much." Besides these, we read in Plautus, Ciccum non interduim, Rud. 580; granum tritici, Stich. IV, i, 52; phima, Most. II, i, 60; nux. Mil. II, iii, 45; digitus, Aul. I, i, 17; triobolus, Rud. V, iii, 11, all employed in this sense. We meet with the repetition of one and the same substantive (especially with the relative pronoun) in all periods of the popular dialect of Plautus, down to that of late Latin, especially with locus, dies, and res. And when an English peasant says " Your father, he was my friend," why should it not be permitted to the Roman peasant to say: " Pater tuus is erat patruelis meus," or "pone aedem Castoris ibi sunt homines"? 103. Frequentative and intensive verbs in vulgar Latin often take the place of their primitives. For instance, agitare, pulsare (Fr. pousser), iactare (Fr. Jeter), cantare (Fr. chanter), quassare (Fr. casser) are used where classic writers are commonly content to employ the simple verbs agere, pellere, iacere, etc., just as in German, where similar idioms are con- fined mainly or exclusively to the language of the people, such as lungern (to loaf), rankern (to plot), drdngeln (to press — as we should say, to squash), * Also ne mie^non mica. In English we say not a bit, not a rap, not a scrap, not at all, not a fig, etc. i66 LANGUAGE AND CHARACTER etc.=^ Just as these verbal components were weakened in their signification, so did the comparative and the superlative frequently subside into simple positives ; hence, in order to express degrees of comparison, the addition of suffixes denoting a higher degree or the prefixing of augmentative adverbs was found neces- sary. These peculiarities made their earliest appear- ance in the cases of superlatives in -mus. In this way such forms arose as proximior, postremior, minimissimus, postremissimus, praeclarissimus, per- paticissimi [cf. our Most Highest].! On the other hand, instead of comparison made by means of suf- fixes, we find the custom of using periphrases with adverbs such as valde, bene, plane, satis, adeo, tarn, sane, vehementer, fortiter, abtmde, nimium, affatim, multum. Such combinations as turpiter makvolus, insanum magmis, immaniter arrogans, crudehter inimims, are characteristic of these pleonasms.J Again, both methods of gradation are found zoxi- \Mtzt^A,e,g.,maxime dignissimtis, magis utt/tor {Colum, viii, 5, 5); and we must notice such exaggerations as immortaliter gaudeo (Cic. Ad Quintum Fr. iii, i, 9), immortales gratias (Plane, in Cic. Ad Fam. x, 11), and pleonastic combinations such as mox deinde (Colum. ii, I, 5), adfuodtim nimius {loc. ciL iv, 21, 2).§ * Cf. such English expressions as to pitch away, to chuck, to smash, etc. t This usage was very common in Elizabethan English; cf. Abbot's " Shakesperean Grammar," § 11. X With which we might compare such English conversational exaggerations as awfully pretty, dreadfully ugly, terribly small, etc. § Plautus has " mollior magis," "more tenderer," Aul. 422. '■' - -^ < II \ OF THE ROMAN PEOPLE 167 The same principle is answerable for the substitution of compounds containing one or more prepositions for simple verbs; e.^-.y sufflare (Fr. souffler) ior flare. The latter had come to be felt as too unexpressive and feeble; and the common people clamoured for a stronger speech than the educated classes. Words compounded with con and ad gained a wide popu- larity; e.g,, condignus, condensus, condormire, -com- placerey conflare (Fr. gonfler), concastigare (fourld in Plautus), assimilis, adaeque^ accredere, etc. Similarly monosyllabic prepositions were ousted by compound ones, as abante = Fr. avant, desub = Fr. dessous. This explains the fact that in Romance languages short substantive stems either disappear as in the case of res, OS, mus, zus, sus, ver, or were lengthentid by means of suffixes, as spes (It. speme), vas (It vasello), lex (It. legge), dux (It. doge), nox (It. notte). 104. Further we must remark that the desire for clearness entailed the substitution of cases * joined with prepositions for the simple cases ; only this does not arise from any weakening of signification; its cause is to be sought in the gradual weathering off and disappearance of the terminations, and in the shedding of final consonants, such as m, s, d^ /, etc. By this process the relations between the governing word and the mere inflexion fell into such confusion that finally in order to arrange relations of syntax and to promote definiteness in meaning, the usage of defining words to take the place of inflexions was * Cf. Grandgent, pp. 46 sqq, ^ I i68, LANGUAGE AND CHARACTER deemed necessary. This is specially true o{ de=^ French de,"^ which took the place of the genitive, and indeed, even in old Latin was often substituted for the partitive: of ^^(Fr. a) to express the dative; of in to denote incidence of time, and of per and cum to denote the means whereby, in other words, the agent. Thus Muller, in the Pfalzburger Programm for 1888, shows that in Sidonius Apollinaris ex occurs much less frequently than de\ cf. too Clairin, " Du gdnitif latin et de la preposition de," Paris, 1880. We have thus approached the question of the syntax of vulgar Latin which we have still to con- sider briefly, as well as that of style. The language of the people is, like many of its ways, harsh and brus(;ue, but is instinct with lively feeling, and is simple and easy to understand. It knows nothing of artistic combinations of periods; one thought connects itself to another with the greatest naiveti. The heart, and not the understanding, is the chief factor in the arrangement of the periods employed by the common man. ^ His sentences are set paratactically, and in the simplest form. • All complications of language are as far as possible avoided. Hence his style leaves the impression of brevity and abruptness; thoughp and sentences alike seem to be running away, and not unfrequently to proceed by a series of jumps, disregarding all logical continuity. And, what is more, even the semblance of connection is frequently enough very faint. The word and plays a great part in the paratactic arrangement of ♦ Cf. Grandgent, p. 43. I ) I OF THE ROMAN PEOPLE 169 the sentences of the uneducated, though not unfre- quently even this word is omitted. Oratio obliqtia is replaced by direct speech. If a peasant relates what another has confided in him, he brings his interlo- cutor before us as speaking the words in his own person, so that we have him as in actual life before our eyes, and can hear the words fall from his lips. This artificial form is precisely that used by the brothers Grimm in the composition of their Fairy Tales.* 105. Abstract ideas and high-flown phrases are unpopular with the man in the street. Not that he is lacking in the power of expressing conceptions remote from reality, and lying beyond the domain of the senses, but rather that he fails to give what we may call objective expression to his inner life and ideas. His thoughts have so naturally come to be identified with himself that he never comes to reflect upon the nature of these thoughts at all. He lives through the various circumstances of life with his inner consciousness, without caring to appeal to any exterior agency as accountable for their exist- ence. He looks on life as bounded and conditioned by his surroundings, and forms his views from those surroundings: and thus it comes to pass that he lends life and picturesqueness to his language by the formation of numerous metaphors taken from the various phenomena forced upon his attention by his surroundings. *'The journalist or closet philosopher," says Schroder in his treatise on the journalese style * This artifice is characteristic of Defoe's works. I70 LANGUAGE AND CHARACTER (no doubt with some exaggeration) " never suspects that the groom and the dairymaid employ in a single year more tropes and figures of speech than he will find in all the literary masterpieces of the world." And Biese actually asserts (" Das Naturgefiihl bei Griechen und Romern," ii, 20) that in Roman comedy figures of speech and similes are rare! Cer- tainly we do not find any long drawn-out similes such as characterize epic poetry, but we light on very many picturesque expressions and pointed parallels drawn from the sphere of daily life, as is the way of the people. In fact, the nearer the people stand to an object, and the better they feel that they know it, the more frequently and with the greater pleasure do they drag it into their language by the aid of metaphor. Things which they have loved or known force themselves on their attention, such as parts of the body, domestic animals, implements, trees, the heavens, the stars, and, again, common actions and processes, habits which have become second nature. Even Cicero was struck by the fact that the language of the people possesses a large store of metaphors; for he says. Or. 24, 81 : 'Hralatio qua frequentissime sermo omnis utitur non modo urbanorum, sed etiam rusticorum, siqui- dem est eorum: *gemmare vites, sitire agros, laetas esse segetes, luxuriosa frumenta' " ; and again, De Or. iii, 38, 135, he expresses himself in the same way: ** nam gemmare vites, luxuriam esse in herbis, laetas segetes etiam rustici dicunt " (cf. Quint, viii, 6, 6). The Roman people loved expressions quaint and forceful, such as testa (a potsherd), used for a heady OF THE ROMAN PEOPLE 171 tHe; spatha, a stirring spoon = ipde for a broad- sword (Tac. Ann. xii, 35); bucca, "puff cheek" for "mouth," (Fr. bouche)\ iugulare for interficere, pro- perly" to cut the throat"; calctdare (from calculus, a little stone which was used for reckoning) = com- putare, to reckon ; or phrases like " corium concidere alicui," " to cut about any one's leather," that is to say, to hide any one (applied strictly to wild beasts), and " sub manus succidere," originally a technical term of joiners and potters, which even in Plautus occurs with the metaphorical meaning of " to leave the hands of the maker" (cf. too O. Ribbeck. "Gesch. der rom. Dichtung," i^, 123). But the Roman loved above all things metaphors taken from military life and from the science of law. Each of these depart- ments lay near to his heart, and appealed so strik- ingly at once to his taste and his powers that he lived for them and in them. 106. This whimsical trait of the people reveals itself, moreover, in their method of naming the objects of daily life. Thus we have numerous plants and beasts for which the Roman countryman pos- sesses native names, or, it may be, names given them by himself in the course of time, while the classical language mostly took over the correspond- ing terms from the Greek. These expressions are pretty and picturesque, simple and easily intelligible. For example, the onion {caepa), from the fact of its possessing a single bulb, was called by the country- men umo=^¥r, oignon\ the almond tree {amygdala) was called nucicla = nucicula, properly "little nut''; 172 LANGUAGE AND CHARACTER the sycamore {sycaminos\ celsa\ the centaury {cen- taurea), fel terrae. In the same way he bestows upon the chamaepitys the name of abiga (the re- pellent) ; he calls the sneeze- wort (abrotommi) by the name of veratrtim, on account of its supposed power, if strewn upon the head, to sharpen men's wits.* He calls the heliotrope verrttcaria, wart-weed ; the rhamnus, our Christ's thorn, he calls sentis ursina, or bear thorn; the strychnos, tiva lupina, wolfs grape ; the giraffe {camelopardalis) becomes the ovis fera\ the elephant (elephantus) bos Ltica\ the hippopotamus, bos Aegyptius\ the ostrich, passer marimis (Plaut. Pers. H, ii, 17), the leech {hirudo) sanguisuga, the blood-sucker, etc. 107. And there is a strong contrast between the vulgar dialect and polite diction in the use of re- flexive verbs, tenses, and the Figtira etymologica. All these peculiarities mark the language of the people, and are intended to promote clearness. How seldom does Caesar employ such expressions as se fledere, se effimderey se 7novere, instead of flecti, efftmdi, nioveri, and how commonly do we meet with such expressions in the lingua rustical Involuntarily we call up such German expressions as *'der Rock nutzt sich bald ab," f or such French expressions as - Paris ne s'est pas fait en un jour; les spectacles se donnent." etc. Further, if we find that the mfinitive present instead of the infinitive future (a usage em- * Connecting it with verus. t Or such English phrases as "do move yourself off," "pull yourself up," etc. OF THE ROMAN PEOPLE 173 ployed by Caesar only occasionally. Bell. Gall. ii,32, 3; iv, 21, 5; 22, I ; vi, 9, 7; in order to express the immediate completion of any action) occurs far from unfrequently in vulgar Latin, this is surely a testi- mony to the lively and rapid thoughts of the people, who can with such facility transpose the future into the present. But the Figura etyfjiologica has ever been a favourite in popular Latin from the time of Plautus down to Apuleius, Tertullian, and St. Augus- tine. Thus as early as Plautus we find quite com- monly such expressions as '* vitam vivere " (e.g. Merc. 473), "servitutem servire" [e.g. Capt. 391), "messem metere" [eg. Epidic. 701), "obsonium obsonare" (e.g. Stich. 410), " statuam statuere " (e.g, Asin. 712); these and similar expressions are found equally in all later writers; indeed, new expressions are constantly being coined upon these models, eg,, "laudes laudare" (Fronto), "questus queri" (Statius), " vigiliam vigilare"(Gellius), "indumentum induere," " somnia somniare," " sortem sortiri " (all in Vul- gate). (Cf too Landgraf, '^ De Figuris etymo- logicis Linguae Latinae," Acta semin. philol. Erlang. ii, 1-70, and Fr. Leiffholdt, '* Etymologische Figuren im Romanischen," Erlangen, 1884.) 108. We have still to touch on the third main characteristic of the vulgar tongue, namely, the greater role assigned to the emotions than to the intellect. The educated man speaks only after mature reflection. We remember that Tallyrand goes so far as to remark sarcastically, " La parole a ete donnee a I'homme pour deguiser ses pensees! " X74 LANGUAGE AND CHARACTER The people, however, wear their heart always upon their sleeve. They repudiate all disguises and reveal all their feelings, all their thoughts, all their emotions, not merely in their looks and gestures, but in their utterances and words as well. Hence it comes to pass that in these words, unconsciously and even against their wishes, they reveal their sympathies and their antipathies alike. One instance of this tendency is their strong predilection for diminutives in sign of cordiality, and of the share taken by their sympathies in the moulding of their speech. Such diminutives are more particularly employed to denote sympathy and affection, as amicuhiSy the dear, or, it may be, the poor friend; lectulus, the dear, or the comfortable, bed ; or the use of the dimmutive may be, though it is a rarer case, a sign of dislike, as in the case of Asellus, the stupid, stubborn ass ; spemla, the faint hope; z^^r^^/a^?, disagreeable remarks. These diminutives have, however, become so completely identified with the thoughts of the people, and so little are they felt as real diminutives, that further diminutives were actually coined from these, as for instance from asellus, asellulus] from auricula, auri- cilla\ from cistula, cistella and cistellulaJ^ There are certain diminutive forms, adjectival and verbal, which bear a specifically popular stamp, such ^s ptilckellus, formosulus, tacitulus, misellus, liquidiusctilus, nitidms- cuius, minusculus, mamsculus. Parenthetical phrases such as the following imply ease and intimacy on the part of the speaker — narro tibi, *' I only tell * Os (mouth) gives osculum (which also has the particular sense of " kiss ") and oscillum. OF THE ROMAN PEOPLE 175 you"; mihi ausculta,'^\\^ds vA\2X\%2.y''\ a^nabo te = quaeso, '* tell me, love." ^ Again the vulgar tongue is rich in ethical datives, which even with Plautus had become favourites, and it is likewise partial to formulae of assertion and to interjectional expressions which meet us in all stages of the Latin language. The Roman comedies are throughout marked by expressions of assurance such as medmsfidius, hercle, pol, edepol, ecastor, nae, and again by particles be- tokening affection or encouragement as attat, attatae, babae, bombax; or of joy, as to, euax, euoe, euan, etc. Everywhere the feeling of the moment, or what we may call the subjective influence, forces itself upon our notice. For what are interjections but flashes of feeling shot straight and sudden from the heart? 109. But the deep sympathy of the people with the persons and objects of which it speaks displays itself equally in its very opposite trait: viz., in the reticence and painful anxiety which it manifests to avoid uttering certain words. The people recognize no words as tabooed; no words which the un- written law of polite society forbids them to men- tion : they are unaware that they must avoid this or that unconventional expression. But, religiously or superstitiously disposed as they are, they feel appre- hension and awe when called on to utter the name of the divinity which directs their destinies. The words spoken to the pious Israelites of old with such distinctness and emphasis, " Thou shalt not take the name of the Lord thy God in vain," * Only used by women. N 176 LANGUAGE AND CHARACTER were unconsciously in the mind of every Roman. Just as the Hebrews, from religious scruples, in- stead of pronouncing the m.m^ Jehovah, or Jahweh, used the word Adonai i-e,, ''the Lord"; and as the Greeks called the goddesses of vengeance, the ter- rible furies, the Eumenides, i.e., '' the kindly beings," and Xif^i^ocl '' the honourable goddesses " : just as the Germans mutilate the names of God, of Jesus, and of the Devil in the most varied ways, e.g:, Jesses, Potz (-Gotts), Deiker, Deichsel = Teufel,* so did the Romans mangle and disguise many expressions be- cause their natural reverence prevented them from unnecessarily repeating with their lips what they deemed in their hearts to be holy. One might cite as a proof of this assertion the formulae of asseveration hercle^pol, edepol\ but there are many other words of the same kind. The terrible goddesses of Fate who spared no one were called " Parcae,'* *'the Sparing ones," to win their goodwill.! The ancients were particularly unwilling to pro- nounce words which recall in any way the notion of death, because they were in terror of hastening its approach. For instance, the word morbus, which is etymologically connected with 7nors, was from the earliest times avoided, and was replaced by infir- mitas, languor, valetudo, vitimn, passio, etc. (cf. Wolfflin, " Sitzungsberichte der Bayr. Akad," 1880, pp. 387 sqq). * And as in England we have such words as deuce, marry, and zounds in older English. t This word is, however, connected with parcere by popular etymology only; it is more probably connected with /aw. OF THE ROMAN PEOPLE "^n For the word mors was substituted fatum, quies, finis, abitio, etc. : for the word ** to die," hinc migrare, discedei^e, t^^ansire, dor mire, oppete^^e, de- siderari, obire, vixisse. In the same way ior ftmestus and fatalis the ancients preferred to employ euphe- mistically infaustus and infortunatus,* and the Ger- man expression, '*geh zum Henker'' ("go to the deuce ") finds its equivalent in the Greek £pp' iU xopxxocg, and in the Latin '' i ad Graecum Pi " on account of the shape of the Greek letter tt, which resem.bled a gallows. 1 10. Another instance of the bent of the peasant mind is seen in the joy of the countryman exhibited at the time of harvest and on other festive occasions. This gaiety, coupled with the native Roman predilec- tion for banter, led to the introduction of such popu- lar amusements as the " Fescennines," the " Satura,'* the " Mimes," and the " Atellanae." Especially re- markable is the Roman fondness for verbal wit and puns.f This characteristic pervades the whole of Roman comedy ; on this more than on any other factor depends the great effect which the plays of the genial Plautus aimed at producing. But the taste is everywhere manifest, for the p/ebs of the capital found its delight in such jests no less than the peasant. The wit of the soldiers presumed even to play upon the sacred person of the Emperor. They * Cf. Velleius, 2, cap. 93, " Si quid accidisset Caesari," i.g.y " si mortuus esset." t Macrobius, Sat. ii, 4, gives many instances of Roman jokes. N 178 LANGUAGE AND CHARACTER twisted the name of Tiberius Claudius Nero malici- ously into '' Biberius Caldius Mero," with a play upon 6z6ere, calidum, and merum (Suet. Tiber. 42). The Emperor Macrinus, who treated his servants to a flogging for the slightest offence, gained as his reward the nickname of " Macellinus" or ^'butcher's boy" (cf. lul. Capitol, in vita Gordiani jun. c. 19). Not unlike the banter of the soldiers is the comic transformation of the word discipUna into dis- plicina, as if from displicere, referred to by different grammarians (Priscian, ii, ii4» 3'» Donat. 392, 20; Consent, p. 16); in the same way the mutilation of popina into propina shows popular influence (cf. Isid. XV, 2, 42; Rossi, Inscr. i, 1055). Trans- gulare for strmigtdare (cf. Schuchardt, ** Vokalismus des Vulgarlateins," iii, 12) testifies to the humour of the people, and it seems worth noticing that the saintly Cyprian was once styled '' Coprianus," with a play upon the Greek word xoVpo?, which naturally enough called down the righteous wrath of Lactan- tius (Inst. Div. V, i, 27): '* Audivi ego quendam hominem sacrilegum, qui eum (Cyprianum) immu- tata una littera Coprianum vocaret, quasi elegans ino-enium et melioribus rebus aptum ad aniles fabulas contulisset."* III. In conclusion, to review briefly the whole subject, we find the close connection between the spirit and the language of the people fully proved. Of the four phases of spiritual and intellectual life * Levir, a husband's brother, was conceived of as '' laeviis vir." See Walde, p. 333. OF THE ROMAN PEOPLE 179 imagination and natural feeling are more strongly emphasized than understanding and will power. The scanty endowment of the masses in respect of the last mentioned factors explains their tendency to seek ease of expression, which tendency manifests itself in the first place on the material side, as, for instance, in ease of pronunciation, and secondly, on the intellectual side, as revealed by their reduction to simplicity of inflexional forms of syntax and of word signification. On the other hand their power of imagination conduces to lucidity of speech, and their undisguised and lively feeling renders their language so homely, so winning, and so sane. The vulgar tongue, then, contrasts with classical prose by its effacement of intelligence and will. In the large play allowed to popular fancy and popular genius it approaches poetic diction ; both alike lay the greatest weight on liveliness of style, on pictur- esque lucidity in form, and on warmth of feeling. In both cases we find sentences loosely attached and loosely constructed, in both a predilection for figura- tive expressions, for alliteration, and for the use of frequentatives. Even in vocabulary there are some singular re- semblances. Just as in the German language the words *' kosen. Maid, Born " are at once poetical and vulgar, in the same way such vulgar Latin expres- sions diS factcndtis, facundia, amd focus ( = French /^^^, fire) are not found in Cicero and Caesar, though they appear in the Odes of Horace, and in the Elegies of Propertius. Of course, the linguistic methods by which the people attain their ends are widely dif- i8o LANGUAGE AND CHARACTER, ETC. ferent from those of the poet. Rough folk-lore has rough methods— the more refined style of the poet walks in tenderer ways. The former loves the real, the latter the ideal. In the vulgar tongue change in language sets in unconsciously, in that of the poet with full consciousness. In the former case it affects an entire class; but in the latter case the degree of its influence is proportionate to the genius of the author. V The Classical Language of Caesar and Cicero 1X2 WHILE the vulgar tongue resembles a meadow which flourishes and blossoms almost without man's aid, the artificial language resembles a garden fenced in by the hand of man, and demanding un- remitting attention, if it is to produce good fruit. Now one of the most important tasks of the gardener is to rid his garden of weeds, and in the same way it is imperative for a classical language to banish all words and all verbal forms to which objection may be taken from any quarter. It follows that classic writers must in the first place be on their guard against the introduction of obsolete and foreign terms, and in the next place against the creation of new and startling figures; lastly, they must do their best to get quit of every element which in the eyes of the educated classes must appear vulgar or com- mon. Caesar and Cicero have acted according to these maxims, and if we disregard the Letters, which strike a more familiar note, they have closely scrutin- ized their choice of words. That great statesman and general who brought Gaul under the Roman yoke, was renowned not merely as an orator and a historian, but also as an accomplished grammarian. i8i Ill 182 LANGUAGE AND CHARACTER As to the latter quality, in the two books, *' De Analogia," dedicated to Cicero, he has imposed the strict demand, " Habe semper in memoria atque pectore, ut tanquam scopulum sic fugias inauditum et insolens verbum" (Gellius, i, 10, 4), for which reason this document is described by Pronto as " libri scrtipulosissimi'' and the precepts therein laid down were followed by himself most con- scientiously; indeed, he goes so far, that of several expressions connected in meaning he will employ one only, because he deems the use of several words superfluous when a single one would suffice. For, as Caesar tells us (Brutus, 72, 253), he regarded the *' verborum delectus" as the " originem eloquentiae," and it is to this careful selection of words that he owes ** mira sermonis elegantia cuius proprie studio- sus fuit" (Quintilian, x, i, 114). It is for this reason that he is careful to avoid the words fltivms and amnis, while the word flti- 7nen occurs in his writings more than two hundred times. For the same reason he writes non posse for neqtdre, hand scire and non scire for nescir^e: he writes timere and diligere, not mettierc and antare\ interest , but not refert\ nudare and privare, but not orbare. In the same spirit he discards quamquam, licet, etiamsi, and quamvis (the latter word only in the Bell. Gall, iv, 2, 5, and there connected ^xxki pauci) in favour of etsi\ quia (only Bell. Civ. iii, 30, 4) for quod donee, and quamdiu (only Bell. Gall, i, 17, 6) for dum] igitur {only Bell. Gall, i, 85) for itaqne. The word quomodo occurs nowhere in his writings, and tanquam once only, in a fragment quoted by Gellius OF THE ROMAN PEOPLE 183 (Noct. Att. i, 10, 4); porro only Bell. Gall, v, 27, 4; haud only Bell. Gall, v, 54, 5; he renders our word "before" almost exclusively by priusquam, only twice by antequam (Bell. Civ. i, 2, 2, and ui, II, i); the word causa, ''for the sake of," appears one hundred and fifty times; gratia in the same sense, twice only (Bell. Gall, vii, 43, 2, and Bell. Civ. ii, 7, 3); frtistra occurs ten times; nequiquam twice only (Bell. Gall, ii, 27, 5, and Bell. Civ. i, i, 4); ^/M lare, '' to name," occurs forty-six times, but nominare and vocare only once (Bell. Gall, vii, TZ, 9> and V, 21, 3), whilst in Cicero's speeches appellare occurs some seventy-five times, and the other two verbs are used some thirty times in all. 113. We also miss in Caesars "Commentaries" many expressions which are found in other historical writers of that time. Although he speaks so often of his enemies' defeat, still he never uses the word clades, and if we compare the speeches which Sallust, in his work on Catiline's conspiracy, puts into his mouth, with his own writings, we discover that ex- pressions commonly recur in Sallust's version which are far removed from Caesar's usage. For instance, such words as divitiae (found forty times in Sallust's account), lubido ox libido (thirty-five times), i^;2^^;/^rar^ (twenty times), miseriae (fifteen times), stretiuus (fourteen XAm^s), prof ec to (fifteen times), etc., are no- where found in " Bellum GalHcum " and the " Bellum civile" (cf., too, Schnorr v. Carolsfeld, " Uber die Reden bei Sallust," Leipzig, 1888, bes. pp. 34 ;^^?-)- A large number of words not seldom met with in i84 LANGUAGE AND CHARACTER Cicero's speeches are not employed by Caesar, it may be from their rhetorical character or from some other reason, such as nedum, dummodo, nisi forte, qtiippe qui, utpote qui [praesertitn qui only in Bell. Gall. V, 47, 4), dubito an {kaud scio an only in v, 54, 4), tantum abest ut . . , ut, scquitur, restat, proximum est, reliquum est, extrenium est, piget, miseret, taedet. Further, a large quantity of grammatical peculiarities referred to at length in our school grammars, do not occur at all in Caesar, as, for instance, the use of supplicarc\ 7naledicere,obtrectare,operam ^^r^ followed by the dative, or se praestare and se p7^aebere\ the latter is only found in a letter in Cicero's collection. Ad Att. ix, 7, I, followed by the accusative ^nd parum followed by the genitive. 114. It must not, however, be assumed that the classical "stylist" sinks into monotony, nay, he rather displays, on fit occasions, such change in style and matter as suits his purpose, following the maxim variatio delectat. For certain transactions, especially in matters appertaining to war and the operations of war, he sometimes employs as many as three or four different expressions for the same idea. True it is that, in his method of producing variety, he by no means equals a Livy or a Tacitus, but he certainly escapes wrecking himself on the rock of monotony. For instance, as a variation of the word castris {milites continere), he writes also in castris, intra vallum, intra munhnenta. And as a variant for fortunam temptare he writes also experiri and pericli- iari. For the phrase, to draw a sword, he uses OF THE ROMAN PEOPLE 185 sometimes gladios stringere, sometimes destringere or educere-, to finish the war is with him belltmt conficere or finire. For *' to surpass " he employs super are, vincere, or praestare. On the other hand, he hardly ever uses antecellere, excellere, praecedere, antecedere, praecur7^ere, etc. Newly coined words are hardly to be found in his writings, and even Greek ones he employs very sparingly, although he was favourably disposed to Greek culture and Greek customs, and as Suetonius assures us (Julius Caesar, chap. 67) gave vent to the utterance, " what matters it if my soldiers use perfumes so long as they fight well." We must not set down to his account old expressions like scopulus and epistula, which had by his time admittedly assumed a genuine Latin stamp. However, such military technical terms as catapulta, ballista, sco7pio for siege-work, phalanga to express a roller for the launching of ships, haipago for a bill-hook to bring down walls, 2S\.di phalanx to denote a military parade, could hardly escape employment even by such a purist, as there was no Latin substitute at his dis- posal. The case was the same, and for the same reason, with tetrarches, theatruin^ ephippium, scaphay mackinatio', it is, however, noticeable that when he employs the word malacia to express a dead calm, he finds it necessary to add the explanatory sub- stantive, ac tranquillitas (Bell. Gall, iii, 15, 3). 115. Finally, Caesar has kept his narration as free as possible from vulgarisms, and he has avoided many words which are common both in old Latin i86 LANGUAGE AND CHARACTER and in later writers. Instances are: the intensive expressions oppido and actutum, and the preposition absque in the sense oi sine, the substantives /r^^^//^, obsequela, edulitini, ambtilacmm, the adjectives dis- cordiosus, extimus, which Sallust, for instance, has extracted from some ancient source. Moreover, he has not employed many of the frequentatives so popular in the vulgar tongue, in contrast to the last mentioned writer and to Livy; and if he (Bell. Gall. V, 27, i) says ventitare consuerunt or (v, 7, 8) saepe clamitans, these expressions are redundant only in appearance, as the meaning in the former case is, " they are wont to pass to and fro," and in the latter case, '' often crying aloud." On the other hand there is a trace of vulgar Latin in the employment of captiviis to qualify objects such as naves (Bellum Civile, ii, 5, i ; cf. Bell. Alex. 42, 4; 47> 2), and again in the use of sexennis and semestris for sex annortim and sex mensium (Bell. Civ. iii, 20, 5, and i, 9, 2), in the phrase in fiigant dare (Bell. Gall, iv, 26, 5, and V, 51, 5; cf. Bell. Afr. 78, 8, and Ph. Thielmann " Das verbum dare im Lateinischen," Leipzig, 1882, p. 105), in albente caelo (Bell. Civ. i, 68, i ; cf. Sisenn. Fr. 103 P., also Bell. Afr. ii, i, 80). Again the fol- lowing usages seem characteristic of vulgar Latin: that of consimilis (Bell. Gall, ii, 1 1, i ; v, 12, 3; vi, 21, i), convallis (Bell. Gall, iii, 20,4; v, 32, 2; regu- larly used in Bell. Afr. and Hisp. for vallis), con- fieri (Bell. Gall, vii, 58, 2), adatigere (V^^W. Civ. iii, 58, 14). 116. Precisely the same principles are illustrated OF THE ROMAN PEOPLE 187 by Cicero, theoretically in his rhetorical writings, but practically in his speeches, and in his philosophi- cal treatises. We are able to gather his own views about the admission of obsolete expressions from the passage De Orat. iii, 38, 153, where he says: ** Prisca fere ac vetusta ab usu cotidiani sermonis jam diu intermissa poetarum licentiae liberiora quam nostrae (i,e, oratorum)," and similarly in another pas- sage (Or. 24, 80) '' Sed etiam inusitata ac prisca sunt in propriis, nisi quod raro utimur." Still Cicero the orator is not so strict as Caesar the historian, for Cicero maintains that a certain mystical charm attaches to his style (''grandior atque antiquior oratio saepe videri solet," cf. De Orat. loc, cital.) if he, as occasion requires, employs old-fashioned terms such as tempestas used in the sense of tempics, ox pone in the place ol post. This holds true especially if the archaic expression is drawn from a poetical work, for *' raro habet etiam in oratione poeticum aliquod verbum dignitatem." The case is different in such philosophical writings as the '' Cato Maior," in which the employment of old-fashioned tricks of speech is intended to invest the language with an antique appearance, just as Goethe's method is intended to do in Hans Sachs' " Poetische Sendung." Still Cicero with all his toleration confines himself within the narrowest possible limits : he writes quasi in the sense of quemadmodum (19, 71), and, to sum up here briefly his use of inflexional forms and syntactical peculiarities, he employs meditatus (20, 74), dimensus (17, 59) in a passive sense (2, 4; adepti is now taken as the true reading, not adeptam), audaciter for audac- i88 LANGUAGE AND CHARACTER ier (20, 72), qicam viam ingrediendum sit, for quae via ingredienda sit (2, 6). 117. In his admission of foreign words Cicero is equally moderate, especially in his speeches, in which, apart from the borrowed words which had already acquired rights of citizenship in the Latin language, very few Greek terms are to be found. Cicero expressly advises (Or. 49, 164): '' Quare bonitate potius nostrorum verborum utamur quam splendore Graecorum." In accordance with this maxim, he introduces one, and only one, foreign word into the speech *' Pro Quinctio," viz., ephemeris (18, 57); in "Pro Ligario " none at all, for barbarus can hardly be considered as such. Most of these Greek words, as we might naturally expect, are found in the Verrine Orations, where so many Greek objects of art are mentioned. In his philosophical works, however, Cicero could not absolutely dispense with foreign appellatives, the less so because the entire material for these pro- ceeded from Hellenic sources, and technical ex- pressions had to be employed for which no Latin words had been coined. So the orator is justified in saying (De Fin. iii, 2, 5): **Quamquam ea verba quibus instituto veterum utimur pro Latinis ut ipsa philosophia, ut rhetorica, didactica, grammatica, geometria,musica,quamquam Latine ea dici poterant, tamen, quoniam usu recepta sunt, nostra ducamus." In cases where the new terms which he employs are not fairly incorporated into the language, he adds some expression like *' as they are called," as OF THE ROMAN PEOPLE 189 in De Nat. Deor. 21, 53: ''Qui theologi nomi- nantur," or as in Paradoxa, 4: " Mira ista paradoxa quae appellant maxime videntur esse Socratica," and in the same place, " Ea, quae dicuntur in scholis thetica." But in general his aim is to create satis- factory Latin substitutes, sometimes by simply transposing them, sometimes again by new forma- tions, in accordance with his purpose expressed in the Tusculan Orations (i, 8, 15): *' Dicam, si potero, Latine; scis enim me Graece in Latino sermone non plus solere quam in Graeco Latine." Thus it happens that in many of his philosophical writings the number of foreign terms is very limited, as, for instance, in the Timaeus. Often he feels con- strained to plead some excuse for his novel experi- ments, as De Nat. Deor. 8, 18 : " Stoicorum Trpovoiai/, quam Latine licet providentiam dicere "; Acad, i, 40: *' Quam illi (pai/rao-iai/ (appellant), nos visum appel- lemus licet"; De Fin. iii, 6, 21 : '* Quod cum posi- tum sit in eo, quod ofxoXoylocv Stoici appellant, nos appellemus convenientiam, si placet." He expresses himself again differently (De Fin. iii, 16, 53): ** Quod enim illi ol$ioc(popov dicunt, id mihi ita occurrit, ut indifferens dicerem," or Top. 8, 35: "Quam Graeci iTVfxoXoyUv vocant, id est verbum ex verbo veriloquium, nos autem novitatem verbi non satis apti fugientes hoc genus notationem appellemus." In other places he adds some qualification intro- duced by quidant or quasi, as, for instance, when he translates the Greek word -KOiOTy^c; (De Nat. Deor. ii, 94; Acad, i, 24 ff.): *' Id corpus et quasi quali- tatem quandam nominabant." He seldom contents I90 LANGUAGE AND CHARACTER himself with the simple conjunctive of demand as De Fin. iv, 27, 74: '* Haec paradoxa illi (dicunt), nos admirabilia dicamus "; similarly we do not often find him setting the Latin expression before the Greek, as De Leg. ii, 13, 32: " Divinatio, quam Graeci |uakTi)c^'i/ appellant," or De Div. 60, 124: "Con- venientia naturae, quam vocant (ru/A7ra0£tak Graeci. On the other hand, he sometimes prefers, in cases when he cannot translate quite literally, to use a circumlocution, as De Fin. iii, 4, 14: ** Equidem soleo etiam quod uno Graeci, si aliter non possum, idem pluribus verbis exponere," and De optimo gen. die. 14: ** Ne converti ut interpres, sed ut orator, sententiis isdem et earum formis tamquam figuris, verbis ad nostram consuetudinem aptis; in quibus non verbum pro verbo necesse habui reddere, sed genus omne verborum vimque servavi." Hence he sometimes leaves us the choice between two forms, as in Timaeus 4 : *' Quae Graece aj/aXoyta, Latine — audendum enim est, quoniam haec primum a nobis novantur — comparatio proportiove dici potest." 118. These examples may serve to show how careful Cicero was, alike in his new formations of words, and in his employment of expressions in novel and metaphorical tenses. This characteristic is rendered clearer by a comparison with other passages, in which no question arises of the mere transference of foreign names to express his ideas. In De Or. iii, 41, 165, he lays down the following precept for the orator: " Si vereare, ne paulo durior OF THE ROMAN PEOPLE 191 translatio esse videatur, moUienda est proposito verbo (ut ita dicam)." Besides, in his De Nat. Deor. i, 34, 95, speaking of the words beatitas 2SiA beatitudOy he adds: '* Utrumque omnino durum" (possibly thinking of the recurrence of t in the medial syl- lables) *' sed non mollienda nobis sunt.*' Lastly, a passage in his correspondence with Tiro (Ad Fam. xvi, 17, i) shows us how carefully he strove to maintain his principles in this respect. In that place hespeaksof theadverbyf^(^/2V^r in thephrase "fideliter inservire valetudini " to observe the laws of health. He remarks that the strict application of this word is to matters of duty : at the same time there are many occasions for transference of meaning *' verbo mierationes sunt in alienum multae," for the word *' faithful " or ** genuine " might be applied equally to education, to a house, to art, and even to agri- culture, so that the metaphorical meaning of the word is hardly felt (verecundus). 119. The same principles which guided their choice of words were observed by Caesar and Cicero in their use of inflexional forms. In this case, too, they took care to avoid formations which were either antiquated or vulgar. In Sallust, genitives like senatly tumulti, and in the case of vowels, such forms as lubido and optumus are quite common; again, in Nepos, such forms as lacrumo, tiltumuSy face=fac, parserat = pepercerat and other similar formations occur, bearing an archaic stamp, and actually deemed worthy of an apology by the writer, on the 192 LANGUAGE AND CHARACTER ground that the style of history was still rude and incorrMptum when compared with the rhetorical and philosophical style ennobled by Cicero. But in Caesar's Commentaries, and in the speeches and treatises of Cicero, little can be found which bears the slightest trace of such archaisms. The perfect forms ending in -re instead of -rmit are rarely found in either, and in Caesar especially the proofs of their existence are far from certain. In the ** Bell. Gall." the form -erunt occurs some three hundred and ninety or four hundred times, while, according to one class of MSS., a formation in -ere occurs once only {vertere, iii, 21, i); the rest show three cases of the same form (i, 32, 3; ii, n, 6; vi, 8, 6), while in the " Bell. Civ.," two cases alone, viz., stistimiere, i, 5^ 5. and accessere, iii, 63, 6, have been at all creditably attested. Further, >r^w is hardly ever substituted for essem in these two authors, and verbs joined with reflexive pronouns instead of the reflexive passive, as se flectere=flecti, seldom meet us in either: participles perfect of deponents are used only in the case of a few words in a passive sense (cf. emeritus, pacttis, partitus, and the forms mentioned above, dimensus, meditattis, and adeptus). Such middle voice formations as raius = arbitratus, pertaesMS. " annoyed ^X.:' perosus, -hating" are care- fully avoided. In the case of declensions it is re- markable that, according to Gellius, Caesar preferred the older genitive form acie to aciei, and that Cicero wrote on several occasions senati (Divin. in Caec. 5, 19; Phil, iii, 15. 38; De Har. Resp. 8, 14, and in some passages of his letters). OF THE ROMAN PEOPLE 193 1 20. Greek terminations are but rarely admitted by either writer, and then only in the case of Greek names like Salamis (ace. Salamina, Tusc. i, 46), or in that of appellatives, as in Bell. Gall, i, 52, 5, phala7tges, ox phalanga. Whether Caesar, like Tacitus and other historians, formed the accusative plural of Gallic and other foreign tribes in -as {e.g,, in Tacitus we find Brigantas, Nemetas, Siluras, Van- gionas) cannot, in view of the uncertainty of MS. tradition, be definitely ascertained. As against thirty accusatives in -es occurring in Bell. Gall., the termina- tion -as can claim to have been used in two places only in both classes of MSS. (i, 26, 6, Lingonas\ and iii, 7, 4, Curiosolitas — besides this three times in a, seven times in jS : cf. further Bell. Civ. 35, 4, Sallyas) : but even here it is possible that the termination may be due to the negligence of the scribe. Pronominal forms, like the datives singular alterae (Bell. Gall, v, 27, 5), nullo (Bell. Gall, vi, 13, i; Bell. Civ. ii, 7, i), altera (Cic. De Nat. Deor. ii, 66 Schn., Baiter reads alteri), and the genitives nulli (Cic. Rose. Com. 48), aliae (Cic. De Div. ii, 30), especially in the passages from Caesar, are quite uncertain readings : finally the adverbial ending -iter, connected with adjectives of the second declension, appears very seldom. In Cicero's speeches and philosophical writings we always read dure [not duriter\ large, kumaize, etc., andyfr;;^^, too, occurs more frequently \h2sifir1niter, which is found Rep. i, 69 and vi, 2\ only in the Letters do we meet this suffix with any frequency [Jiumaniter in Ad Fam. vii, 9 ; Ad Att. i, 2 ; Ad Quint. Fr. li, 11: o 194 LANGUAGE AND CHARACTER cf. too' Helmuth, **Acta Sem. Phil.," Erlang., i, 114); but in Caesar such formations are rarely found, e.g., firmiter, iv, 26, i, and largiter, i, 18, 6; duriter and humaniter are not represented. 121. Nor was more licence permitted in the syn- tax than in the inflexions. Collocations, such as that of eo with the factitive genitive {eo te^neritatis pro- gressus est = ad earn temeritatem progressus est), which are common alike to old Latin, to the popular dia- lect, and to silver Latinity, are not found either in Cicero or in Caesar: the same observation holds good with respect to the distributive genitive after the neuter singular or neuter plural of an adjective. The genitive of definition, especially in the case of place-names, a construction avoided by Caesar, is not certainly proved to be used by Cicero (for in oppido Antiochiae, Ad Att. v, 18, i, Heraeus reads in oppido Antiockid). The present participle fol- lowed by the genitive of a substantive occurs in Caesar in one passage only. Bell. Civ. i, 69, 3 {/ugiens laboris), and Cicero employs this in his philosophical works only (see, however, De Imp. Cn. Pomp. 3, 7; Pro Plane. 5, 13). The genitive of the gerund and of the gerundive after relative adjec- tives, a construction which is not found in Plautus at all, occurs but rarely in the case of either of the authors mentioned (cf. Bell. Gall, i, 2, 4, Cupidus bet- landi] v, 6, 3, insuetus navigandi)\ while, on the other hand, it frequently occurs in Livy and Tacitus. OF THE ROMAN PEOPLE 195 122. The dative of the gerund employed after adjectives, a favourite construction with later writers, is only used by the historian after the word par, by the philosopher and orator only after accommodatus: but the modal usage of the ablative form of the gerund, which became more and more frequent after Livy s time, so that it has actually passed into the Romance languages, is unknown to both alike. The accusative, commonly called the accusative of re- spect, dependent on a perfect participle passive, in such phrases z.s pulvere caput conspersus, found in different authors since Plautus and Ennius, seems to have been unknown to both Caesar and Cicero. The so-called Greek accusative can only be attri- buted to Caesar if we count as such the expression maximam partem, Bell. Gall, iv, i, 8, and it is scarcely found in Cicero. Caesar, in Bell. Civ. iii, 88, 2, and there only, connects a singular subject to which is attached an attribute with the preposition cum with the plural of the verb {Ciciliensis legio con- iuncta cum cohortibus Hispanis in dextro cornu posi- tae erant) : Cicero, too, is chary of such use, e.g., Phil. 12, 27; Fam. xiv, 7, 2 (cf. too Lehmann on Cicero's Epistles, p. 222); but this construction is found often enough in Cato, Sallust, Livy, and others. The use of the plural of a verb after a collective substantive as subject seems foreign to Caesar, though the different editions of his works afford a few examples of such construction (cf. Meusel im Jahresberichte des philolog. Vereins zu Berlin, xx, 1894, P- 263): it is also to be remarked that we do not find many instances of the supine in -um fol- 196 LANGUAGE AND CHARACTER lowed by an accusative, as in Bell. Gall, i, ii, 2, auxilium rogatum\ 31, 9, auxilium postulatum\ vii, 5, 2, subsidium rogatum). A noun connected with a preposition governing another noun [as, for instance, '•Every man in England"] is seldom found in classical prose without a participle to support it, as, for instance, '' oratio pro Milone habita," ** epistulae ad Atticum datae.'* [Exceptions: (i) Verbal sub- stantives, as " Rogatio de bello," " adventus in Gal- liam." (2) Substantives expressing mental emotion or service rendered, e.g., " pietas erga parentes : bene- ficium in Siculos." Moreover, Caesar us^sfa^na de and nuntius de without a supporting participle : and Livy has '• pugna ad Trebiam."] And when such participle is not added, the noun and preposition are commonly inserted between the attribute and the substantive, as Bell. Gall, v, 13, i, " omnes ex Gallia naves," and in the same section 4, '^Certis ex aqua mensuris": Bell. Civ. ii, 37, 2 ('* Caesaris in Hispania res secundae"). 123. The following peculiarities in the use of pre- positions seem worth mentioning. Ante is not used to indicate preference, intra is seldom used of time (Bell. Gall, vi, 21, 5); ad in a modal sense, except- ing in Cicero's juvenile writings, only occurs in the standing formula quemadmodum. The negative particle kaud, a favourite word of Livy and Tacitus, occurs in Caesar only in the phrase '' haud scio an," Bell. Gall, v, 54, 4, and in Cicero only when set in con- nection with single adjectives and adverbs, as kaud facile, haud sane. OF THE ROMAN PEOPLE 197 Sometimes we find great discrepancies in the lin- guistic usages of the two great classic prose writers. For instance, the so-called Greek dative (as in " mihi consultum ac provisum est," Cic. Catil. 1 2, 26) occurs very frequently in Cicero, while in Caesar two examples only are to be found, viz. Bell. Gall. vii, 20, and Bell. Civ. i, 6 (cf., too, Tillmann, '' Acta sem. phil.," Erlang. ii, 79 sqq.). Again, we meet with the temporal use of sub not unfrequently in the Com- mentaries, and also with the word propritis used in the sense of relation (nearer), which Cicero does not use. Again, in Caesar we meet with frequent cases in which the same substantive occurs connected with both a ''subjective" and an *' objective" geni- tive; for instance, Bell. Gall, i, 30, "veteres Helve- tiorum iniuriae populi Romani " ['* The ancient wrongs done by the Helvetii to the Roman people "]. This construction is more rarely found in Cicero, e.g., De Off. i, 43 (cf. too Andresen on Cicero s Epistles, p. 186). 124. Syntactic combinations, resembling those found in the popular dialect, meet us in Caesar s Commentaries oftener than in Cicero's speeches and philosophical works. One of these is the em- ployment of the reflexive pronoun — instead of the determinatival — not referring to the subject of the same sentence, as Bell. Gall, vi, 9, 2, "quarum (causarum) erat altera, quod auxilia contra se mise- rant " [because they had sent forces against him] : so again Bell. Civ. iii, 53, 5: '*quem Caesar, ut erat de se meritus, ad primipilum se transducere pronun- 198 LANGUAGE AND CHARACTER tiavit " [in which case for se we should expect ilium]. Cf. Cicero, Verr. 49, 128, and Pro Roscio Amerino, 2, 6, both, it should be remarked, productions of the author's youthful style. Conversely, Caesar uses the determinative instead of the reflexive (as in Bell. Civ. i, 2, 2>7^ ^b eo = a se: i, 35, 4; iii, 75, 2, and several times in the " Bellum Gallicum": cf Cic. Verr. i, 86). Then he employs constructions of attraction as Bell. Gall, i, 39, 6: "se rem frumen- tariam ut satis commode supportari posset, timere dicebant'* (and Cic. Ad Att. 14, 21, " nosti virum quam tectus ") : the connection o{ multus with sin- gular words like dies and nox {e.g.y multo die, Bell. Gall, i, 22, 4 : ad mullam noctem, i, 26, 3): persuasum mihi habeOy Bell. Gall, iii, 2, 5 : insertion of the word credo Bell. Civ. ii, 31 : quo maiorem, credo, licentiam habeant: the construction of {prae) optare with the infinitive (Bell. Gall, i, 25, 4) found also in Hirtius (Bell. Gall, viii, 9, 2): the double accusative in the case of velle aliquem aligtiid (Bell. Gall, i, 32, 2 : cf Terence And. 536, and Phorm. 151). The union of a sentence containing a condition with an imperative is exceptional in Caesar, as in Bell. Gall, iv, 25: " Desilite, nisi vultis aquilam prodere " (although such sentences are characteristic of popular language yet they are occasionally admitted into Cicero's speeches) : as is also the employment of the conjunc- tive in iteratival sentences as Bell. Gall, v, 35: **sin autem locum tenere vellent, nee virtuti locus relin- quebatur neque tela vitare poterant." There are also other respects in which differences in the linguistic usages of these two authors are OF THE ROMAN PEOPLE 199 noticeable. In Caesar, si is common after words of expecting and attempting: this construction is more commonly found in Cicero's letters [thus '^conabor an possim" would be the regular construction]. Nomie, in indirect questions, is known to Cicero only, as is the use of the logical perfect in a gnomic sense [as " multi cum obesse vellent, profuerunt, et cum prodesse obfuerunt," De Nat. Deor. 3, 70], and the dependent conjunctive of unrealized conditions in the periphrastic conjugation in -urns fuerim {e.g,. Pro Mil. 33; Verr. ii, 108; Phil, ix, i), [as **quaero nonne tibi faciendum idem sit," De Fin. 3, 13]. Caesar says "confidere alicui," but "aliqua re," Cicero more frequently puts the thing in the dative. The figiira etyrnologica, as it is called, is not found in the Commentaries (for expressions like "tridui viam progressi," Bell. Gall. 4, 5, contain simply an accusative of space). In Cicero, on the other hand, this figure is far from uncommon. In certain cases where the participle belonged to the construction with the accusative and infinitive, Caesar seems to have introduced a new construction by not placing this participle, where it would naturally fall, in the accusative, but by adopting it as the subject of the main sentence: and this construction is known to Sallust and Livy. An instance of this is found in Bell. Gall, v, 39, 4, '' Hanc adepti victoriam in per- petuum se fore victores confidebant " (in which pas- sage some editors read adeptos). Of the supines in 'U, Caesar admits, besides natu, only factu (iv, 30, 2) and aspectu (v, 14, 2). Cicero, on the other hand, exhibits no less than twenty-four different formations 200 LANGUAGE AND CHARACTER from the same supine, as audiht, dictu, memoratu, visu, cognitu, intelledUy scitu\ at the same time he favours far less than Caesar the construction with the gerundive, which occurs no less than sixty-five times in the '* Bellum Gallicum " (seventeen of these are in the eighth book) and thirty-nine times in the ** Bellum Civile." Further, the use of the ablative absolute to express emphasis, in the place of the mere connecting participle, is relatively more fre- quent in the Commentaries than in the speeches and treatises of Cicero, ^.^., Bell. Gall, iii, 14,4; iv, 12, i; 21, 6; V, 4, 3. 125. The greatest contrast, however, between these two classic authors manifests itself in their respective styles and their individual peculiarities. In Caesar we are frequently called to notice the occurrence of the so-called present of narration, which meets us in nearly every section, and the frequent use of the historical infinitive. This latter occurs at least ten times (six in the Bell. Gall, i, 0, I; 32, 3; ii» 30» 3; iii> 4. 2; v, 6, 4; 33, i), though not nearly as often as in Sallust: for the latter writer, besides many instances of present tenses of narration, has employed four hundred and fifty-two such infinitives. Polysyndeton is in Caesar exceptional (as in Bell. Gall, iv, 24, simul et , . , ei . . . ei) : but asyndeton quite usual ; this figure ex- presses either haste (as Bell. Gall, i, 7, 20, 22, etc.), or serves to mark a contrast (as Bell. Gall, i, i, 18; vii, 50, 76, and Bell. Civ. iii, 36, 8), or it may be to exaggerate such contrast (as Bell. Gall, i, 32, 39), OF THE ROMAN PEOPLE 201 or to emphasize it (as Bell. Gall, i, 5, 20; vii, 59, 77), or to mark a sequence (as in Bell. Gall, viii, 25, 45), or a further explanation (as in Bell. Gall, iv, 27; v, 30; vi, 28). Anaphora, too, and chiasmus, are favour- ite figures of this writer; chiasmus occurs in Bell. Gall, ii, 10; vi, 12, 16; vii, i, 42, 47, 63, 66, 80: anaphora in v, 6; vi, 21, 25, 26, 32, 34, 35, 36; vii, 20, 28, 32, 2,Z. 38, 52, 59» 66; Bell. Civ. ii, 2>7^ 6, etc. So often do these figures occur in Roman writers that they are called by Nagelsbach* " the forces that regulate the organism of the Latin sentence." 126. It is to his effort for clearness that we must ascribe the marked pleonastic traits of Caesar's style. The redundancy of his expressions is sometimes visible in grammatical, sometimes in rhetorical, pecu- liarities. To the former we must set down such cases as "postridie eius diei " (Bell. Gall, i, 23, and six times besides), "pridie eius diei " (Bell. Gall. 47, 2, and Bell. Civ. i, 14, 3): also the repetition of the substantive in relative sentences, probably after the model of the ancient Curial style, especially in the case of res, lex, pons, locus, dies, iter (e,g,, Bell. Gall, i, 6, 4; 16, 5; 49, i; iii, 3, i; iv, 7; v, 2; vii, 72, i), propterea ^zwd used in the sense of simple ^tiod (Bell. Gall, ii, 4, 4; iii, 21, 3, etc.; fourteen times in the first book alone); '* permittere, ut liceat," i.e., **alicuius voluntate" (Bell. Gall, i, 7,3; 30,4; 35,3; 39, 3, etc.) ; r/^r^^j occurring in connection with com- pounds with re, as se recipere (Bell. Gall, v, 34, 4), * iyo^ and to denote a general in armies other than Roman. In ancient Rome the name of Praetors was assigned to the pair of highest officials chosen yearly, who repre- sented the kings until the title of consules came in at the time of the Decemvirs. From them the porta praetoria, or main front entrance of the Roman camp, took its name " Gate of the General," in contra- distinction to the porta decumana, or postern camp entrance where the tenth cohorts (cohortes demmae) had their quarters: the tent of the general, too, bears their name {Praetorium), as does also his body-guard (cohors praetoria). It was not till a later date that the title of praetor was transferred to judicial magistrates [388 Urb. Cond.]. The qtmestor takes his name from quaerere, " to inquire into," be- cause under the kings and during the early years of the Republic, he was the magistrate charged with the investigation of criminal offences, or President of the Court of penal judicature, and, only in a minor capacity, Chancellor of the public exchequer. In the classic period he was exclusively charged with the public purse. The aedilis was the officer in charge of buildings in general (" aedilis qui aedes sacras et privatas procuraret," Paul, ap, Fest. 13, 7). He took his title from aedes, which in the singular denotes a hearth [cf. Gk. arSw], afterwards the cell of a temple, OF THE ROMAN PEOPLE 213 and finally the sacred building itself: but in the p lural it signifies a house with reference to all its rooms [in English we say **to search the house" — in German, from the more frequent use of flats, Vdurch alle Zimmer suchen"]. The tribuni were originally headsmen of Tribes : afterwards they came t:o be the magistrates charged with protecting the Plebs: the raised seat which they occupied in virtue of their high office was called tribunale (i.e., sugges- lum): whence comes our *' Tribunal." The tribes in question were the three original ones called Ramnes, Tities and Luceres, from which the free burgesses of Rome sprang. But after the new dis- tribution ascribed to Servius Tullius, they came to mean the [local] divisions of the Roman people into four city and twenty-six (later thirty-one) rural tribes. The word tribuere means properly to tax these tribes: and the word tidbutttm denoted the tax so imposed, as Varro expressly tells us (Ling. Lat. V, 8): "tributum dictum a tribubus tributim exigebatur." The auguresy whose business it was to mark the flight of birds, take their name from avis, a bird, and from the root gar found in garrire, to chatter. [The latter statement is uncertain, see Walde *' Etym. Wbch." p. 55], just as the auspices take their names from avis, and specere, to look. The name haru- spices is connected with hariolus, *'a diviner," and with hira, hilla, ** entrails." Pontifices are said to take their name from the building of the plank bridges which it was their duty to maintain in good order [but see Walde, p. 480, who derives the word 214 LANGUAGE AND CHARACTER from a Sabine form punlzs connected with guinpmre, *' to purify "]. i:\iefetiales are said to take their name from the duty laid on them of uttering {profiteri) the solemn declarations of war or peace. [This word is more* probably connected with facio and Sf^K, see Walde, p. 220; the meaning being "the statute maker."] The Salii, or priests of Mars, were called *' leapers '* because the worship which they superintended was associated with processions of jesting leapers, just- as we see in the Luxemburg town of Echternach at the present day. The Quirites, i,e., the Romans in their quality of free citizens, take their title from the spear with which they were armed; curis being the Sabine word for spear: thus the name means the spear-men. [More probably inhabitants of the Sabine town of Cures: Walde assumes that even this explanation rests on popular etymology, though probably the same root is seen in O/r- and Qtiir- alike.] The advocatus, or *' person summoned to aid/' was in the time of the Republic a friend conversant with law, on whose aid a suitor could rely, and who supported his case by his personal presence : it was not till Quintilian's time under the Empire that the word received the sense it bears to-day of counsel specially employed to conduct a cause. Candidahis was the epithet applied to an applicant for a magis- terial post, from the bright white (candida) toga which he wore when he went round to solicit votes: famulus is the slave regarded as a member of the household (O^c^n fama, a house), cf. oiyf-k-vnt; from oTxof : OF THE ROMAN PEOPLE 215 persona, connected with personare, denotes originally the player's mask through which the player spoke : then it comes to denote the actor, and, finally, " a person " generally. The same change of meaning is observable in the case of Trpoa-uwou [this is, however, doubtful, Walde connects persona with zona]. Coquus in old Latin denotes both cook and baker, because the same person performed both functions. Such at least is the testimony of Paulus ap, Festum, 58, 14: "Coquum et pistorem apud antiquos eundem fuisse accepimus": pistor (ixonx pinsere, to crush) is, strictly speaking, the name given to a person who crushes or bruises corn in a mortar, and then applied to one who grinds it small in a handmill. But as the grinder served as baker also, the word received the common acceptation of baker. There were certainly special bakers in Rome as early as the year 171 B.C., for Pliny tells us, Nat. Hist, xviii, 107: ** Pistores Romae non fuere ad Persicum usque bellum " {i.e., till the war against Perseus or Perses of Macedonia) " annos . . . gentium." The staple kind of corn in ancient times was spelt (Lat. far) : the meal ground from this was called farzna: but since 300 B.C. wheat, which was probably newly imported from Egypt about that time, took its name frumentum from frui\ the word originally signified what can be enjoyed generally, cf. Fr. fro- ment, corn; Ital. formento: or again it was called triticum, properly what is thrashed out {teritur) [cf. Spanish trogo : granum has undergone the same change of meaning]. Phrygio even as early as Plautus time has the signification of an embroiderer 2i6 LANGUAGE AND CHARACTER from the skill which the Phrygians always showed in this art [cf. such names as Cadurci and in French rouennerie\, Hastati was the name originally given to the combatants in a legion armed with the hasta, the spear; principes to the first rank, triarii to the third. Now the principes originally fought in the front line of the Roman legion : at a later period, however, the arrangements were changed, and the hastati were set before them [and were armed with the pnlum and gladius\. We can gather what kind of wood went to the making of the spear-shaft from the meaning attached to the words fraxinus and cornus, which, besides their ordinary significations of ash and cornel, denote the javelins, the shafts of which were made of these kinds of wood. We can con- fidently assert that besides the metal helmet which bore the name of cassis, a leathern one was in use, from the word galea = yoi.Xny weasel-skin [cf. xuytu]. Marius introduced the custom of employing as signals in the battlefield representations of wild beasts, such as eagles, wolves, horses, etc., attached to poles, so that the word signum came to be used generally, as we should say, for the Roman Flag. [The aquila was the signum of the entire legion: each maniple in the legion had its own special standard: see Plin. Nat. Hist, x, i6: " Romanis eam legionibus dicavit, equi aprique singulos or- dines anteibant."] The ruthlessness displayed in ancient warfare is well illustrated by the first mean- ing oi populari and depopulari, which signify "to dispeople " from populus, people (cf. kopfen, to be- OF THE ROMAN PEOPLE 217 ' head, from kopf, a head, and schlemmeny to wallow in luxury, gormandize, from schlamm, mud). [This is doubtful. Walde is inclined to connect these words with the root oi pello, pepuliT^ It is also to be remarked that spoliare, to rob or despoil, means properly to take the hide off an animal, and hence to strip a fallen foe of his armour. By tela, from tendo [in the sense of to aim at], we are to understand weapons of attack, by arma, weapons of defence {arcere : thus arma = arcmd). [These two words prob- ably come from different roots. Arceo = Qf\i, dpxiu: arma from the root ar found in apaptVxw.] Praediumy a landed property, is originally con- nected \^it\i praeSy a surety: it is thus regarded as a possession of value which may be pledged as a caution : hence the expression of Cicero and Livy, *' praedibus et praediis cavere populo," " to guarantee the safety of the State by sureties and by mort- gaging property." [This derivation is also called in question by Walde, who suggests a possible etymo- logy in prae(s)dium, i,e,, a property situated [sedeo) near a town]. Hortus, like cohors (etymologically connected with Goth, gards, a house : Gk. x^V'^^O ^^» strictly speaking, merely a court or a fenced-in en- closure. We may hence infer that in the earliest times the sites of the different houses were sur- rounded by an enclosed space which may have been planted with pot-herbs. If we compare culmus (German Halm, a stalk) with culmen, and 0^0(^0^, reed, with lpQ6aA|Uov cxxoij/w, avxfx- xoi/^a* Trapao-;;^^!/ -rly lauroiJ, was known to the ancient Romans. Writing was originally a mere process of scratch- ing marks on wooden tablets: hence scribere = (rxa/3i(pa(reat, " to Scratch in""^ (compare the English word ** write " with German ritzen), [The late gram- marian Virgilius Maro actually uses the word carax- ar^ = ;)^apa(ro-fii/, for to write: cf. also the English use of "characters" in script.] Liber, ''book," originally means " bark," and codex means properly a block or stump. Satura {lanx) was strictly speak- ing a pot-pourri of different ingredients [especially for the use of gladiators]: hence it was applied to a ** medley "in literature, as may be seen in the Satires of Lucilius. Ecule2is = equuleus, or ** little horse," was an instument of torture: a horse with a back full of sharp points, on which victims had to sit : ficatum, komJicMs, "fig," denotes the goose's liver artificially fattened with figs: hence Ital. fegato, and Yr.foie, ** liver": calculare, our word '' to calcu- * Cf. Geiger, " On the Development of the human race," chap, iv, on the origin of writing. The northern phrase was risia runir, to scratch Runes. OF THE ROMAN PEOPLE 221 late," comes from calculus, a small stone, such as were used for reckoning: caro, flesh, is identical with the Umbrian word karu andtheOscan^(^r^m = partis [allied to xs/pw] : it signifies strictly a portion cut off, and must have applied to the portion assigned to each guest at meal time. Stipplicium, punishment by death, is connected with supplicare, to ask on bended knees, because the convicted criminals re- ceived their death stroke in that posture [this deri- vation is disputed, the latter half of the word being referred to the same stem as placare]: funus, a funeral procession or ceremony, is connected with Oo/j/r), a sacrificial banquet which was a part of the ceremony in question [this derivation, too, is dis- puted: Walde connects the word with a root which appears in Gothic as gaunon, to utter wait- ings, keening]; nuere is to nod the head in token of acquiescence: hence numen, the deity who vouchsafes assent: the reverse is abnuere, to toss the head back, a sign among the Romans of dis- sent, which we signify by shaking our heads. Sublatus is used in the sense of natus: for the newly born infant was laid before the feet of the father and not recognized as his child till he raised it from the ground: cordatus means not merely "heartened'* but also '* clever," as in the line of Ennius quoted by Cicero, De Rep. i, 3: pensum, a task in general, is strictly speaking the quantity of wool " weighed out " to the female slaves to spin [and pendo itself is properly to hang on to a weight]. The three periods in the month from which the 222 LANGUAGE AND CHARACTER Romans, reckoning backwards, dated, are the Cal- endae, Nonae, and Idus, Calendae takes its name from calo, to call out [cf. our word '^to hail"], be- cause according to Varro, Ling. Lat. vi, 27/' primi dies nominati Calendae quod his diebus calantur qum- tanae an septimanae sint futurae." From the same stem we have calator and calebra, the place for call- ing out public announcements on the Roman Capitol, where the Pontifex minor publicly announced the various days in the month on which festivals were cele- brated or the law-courts opened : also nomenclator, the slave whose business it was to prompt his master by telling him the names of the acquaintances whom he met in the street. The name Nonae, or ninth, was applied to the fifth or to the seventh day of the month, because it was the ninth day before the Ides. The Idus signified the thirteenth or fifteenth day of the month, from a word meaning " to divide," which Macrobius, Saturn, i, 15, 17, assures us existed in Etruscan [Varro, Ling. Lat. vi, 28, makes the same assertion: but it is rejected by Walde, who connects it with isce Ir. = *eid—skiom = mensis lunaris]. The Ides divided the month into two halves. Nundinum ^novem dies is a space of eight days [reckoning in- clusively], hence trinundinmn, a space of twenty- four days, and nundinae, the closing day of the period which answered to our week, the market day, on which the peasants brought their produce to town. Bimus, two years old, and trimiis, three years old, are contracted from bitrimus and Iritrimus : the latter portion of these two words is connected with htems, and points to a time when reckoning by winters was OF THE ROMAN PEOPLE 223 common [cf. in some English dialects twinter = ^. beast two years old]. In Italy the Decemvirs, following no doubt the example of the Greeks, attached a definite value to copper, and thus created coin. The as, assis, very possibly comes from the same root as asser, a rod or stave (cf. voinis as against vomer), so that the Roman as may be paralleled by the Greek o|3oAo? = h^iXoq, a spit [Walde takes the derivation to be from an Umbrian root, ar- = Lat. ad-, denoting to settle or arrange, so that as, assis = ad-tt, "statutory unity"]. The eeneral name for a coin was 7zti7mmis, a word borrowed from the Greek kj/xo^, with the signification of " a statute," or statutory uniform standard of coin- age. In later times num^nus was confined to the meaninor of sestertius: and since Ovid's time the word moneta was employed from Juno Moneta [con- nected with moned] : for it was in her temple that after the introduction of silver coinage (269-268 B.C.) a buildinof was erected for a mint. The word ses- tertius = semis — tertius as [i-e,, two are understood and the third is an as like the German expression dritte- lialb = t\NO and a half]. Denarius is from deni= ten as: solidus (our " solid ") denotes in the first instance a gold coin of the value of twenty-five Denarii, which at a later date decreased in value: in modern times it denotes a copper coin = Fr. sou, and Ital. soldo. As in ancient times unminted copper was in use {aes rude, raudus), which was converted into copper pieces or bars, the use of scales was needed : and hence we find mpendo the double meaning of " to weigh " and " to pay," and the old formula to express a legal pur- 224 LANGUAGE AND CHARACTER, ETC. chase, "per aes at libram." For the bar-form which we noticed when speaking of the as the expression stipendium {stipipendium) is a good parallel. It is derived from stips, stipis, a stem or trunk. [The original meaning of slips is uncertain : it may mean an ear of corn, and thus a payment in corn; cf. Walde, p. 596.] A passus (from pandere, to stretch out) is the span of the arms when stretched out horizontally, from the end of one set of fingers to the other: thus it is neither a foot nor a step, but a double step = 4| of a German foot : for quinque millia passtmm = 24,000 feet or a German mile. \Passus does really signify a man's step, i.e., two and a half English feet; cf Pliny, ii, 23: "Stadium centum viginti quinque nostros efficit passus, hoc est pedes sexcentos viginti quinque." There were two kinds of passus : the passus minor, and the passus mator which consisted of two of the passus minores.'] NOTES (i) § I. Cf. " Archiv fur Lexicographic," vii, 333 sqg.\ and Isidore (Orig. ix, 2, 105): " Romanos graves, Graecos leves"; Livy, xxx, 7, 6: " Romana in adversis rebus con- stantia " : assiduus is from sedere^ and perhaps sedulus [this word is more probably derived from se dolo = sine dolo\ cf. Walde, s.v\ Livy, xxii, 14, i, praises the " insita Romanis industria": and Varro's dictum (i, 2, 2) tallies with this view : " Romanus sedendo vincit " [cf Cicero, "Pro Flacco,* for the Roman view of the Greek character]. [Cf Dr. Voigt in Iwan Mliller's " Handbuch," iv, 2, pp. 288 sqq. " The Roman is distinguished by a deep sense of religion, and his religion afforded him the means of satisfying its claims, while the believer experienced the need of ensuring the aid of the gods in every circumstance of life, by offerings, prayers, and invocations. He manifests a certain lack of imagination which imparts to his religious observances an earnest and sober character. On the other hand, the Roman is distinguished by intellectual aptitude and a good understanding, by presence of mind, and by an inclination to indulge in ready wit and racy banter: by self-control, self-confidence, and courage, and likewise by an inflexible will: he possesses a strong and even exag- gerated love and pride in his country which takes the form sometimes of undue self-esteem, sometimes of enlightened patriotism. Add to this strength, activity, and perseverance, straightforwardness and love of truth, pride in the honour of his name; add that he was conscientious in the discharge of his duties, and inexorable in his claim that his neighbour should requite him in kind. For the rest the Roman was industrious and a careful householder, simple, temperate, Q 226 NOTES and modest: he cultivated gravitas, or decorum in his own outward appearance and actions he ^^'^^^^^ ^^^^^^ and took care that his own authority should be respected. he was sensitive to the claims of friendship, and was him- self steadfast in his friendships." Cf. A. Novent, " De Mon- bus Romanis." Leod, 1829; C. L. Roth "Z"/ ^heone und inneren Geschichte der romischen Satire," Stuttgart, 1848, § 6-10; Bernhardy, Rom. Litt., § i, 4; Teuffel, Rom. Litt, S I ; Voigt, xii ; Taf., § 5.] • x> \ (2) 5 2. The reading argute loqui is not certain ; roly- bius, ii, 17, 10, transcribing Cato, says: Ta ^.^e^.«« ««' ^a V ' • ~ rf aUo VV Soltau "Prolegomena zu XOLTO. yiufiyiav adKzn. L.I. aiso vv . ouiidu, •■ •■ i, einer romischen Chronologic," p. 7°- (■I) 8 V The following works may be consulted on the relations between the language of a nation and its char- acter: Wedewer, " tjber die Wichtigkeit und Bedeutung der Sprache fur das tiefere Verstandniss des Volkschar- acters;- Frankfurt aM., 1859; Fr. Stehlich, " Die Sprache in ihrer Beziehung zum National -character, Casseler Programm, i88i ; J. Stocklein, " Beobachtungen uber den Zusammenhang zwischen Sprache und Volkscharacter Blatter fur das Bayr. Gymnasialschulwesen, xxx, 335-357. A Lefevre, " Les races et les langues," Paris, 1893; J. Le- coultre, " Du g6nie de la langue frangaise ^o">f j;^ Y^Hn de la langue latine," Neuchatel, 1 894 ; Lindsay, The Latin Language." Cambridge, ,895; and Weise's '• Abhandlung uber deutsche Sprache und Deutsche Volksartm den Sam- melwerke von Hans Meyer, Deutsches Volkstum, 2 Aufl Leipzig pp. 213-260. [Cf also "General Principles of the St"uctu;e' of Language," by James Byrne. M.A.. London, Triibner, 1885.] • o • (A) % V Cicero's assertion, De Nat Deorum, 1, 4, ^ is mere self-complacency and exaggerated patriotism. ';Qao in genere tantum profecisse videmur, ut a Graecis ne verborum quidem copia vinceremur"; and agam where he CDeFin i 3 10) says: " Saepe discerni Latmam Imguam non modo' non inopem, ut vulgo putarent, sed locupletiorem etiam esse quam Graecam." NOTES 227 (5) § 3- See Weise's treatise on the Greek loan-words in Latin (Leipzig, 1882, and G. A. Saalfeld, " Thesaurus Italo- graecus," Vienna, 1884). (6) § 3. Other such words are: ypafjL/xa, a letter; raAavrov, a talent; juo^oi/, a mulberry; 9r?^Tavo;, a plane tree; Kspdriovy fenugreek ; Ato; ^(zxavo^, edible chestnut ; /xsyag arpouOog, ostrich ; ^ov^ahoq, an antelope; pivoxspccg, rhinoceros; wiiyapyog, a Libyan antelope [or the great sea eagle] ; KpoKo^sixogy crocodile; KepoTTidmogy ape; jtaTu^xsTrujVy [African] buffalo; upoXoyioVy sun- dial ; TpiYipvg, three-decker, etc. (7) § 4. Cf Leo Tob, " De grammaticis vocabulis apud Latinos," Paris, 1893; L. Jeep, "Zur Geschichte der Lehre von den Redeteilen bei den lateinischen Grammatikern," Leipzig, 1893 5 B- Linderbauer, " De Verborum mutuatorum et peregrinorum apud Ciceronem usu et compensatione." Pars posterior. Programm von Metten bei Straubing, 1893. (8) § 5. It is admitted, even by Hehn, the great admirer of the Romance languages, that they are deficient in the power of forming compounds (Italien 3 Aufl., p. 201). For this question as applied to Latin, see P. Udolph, " De Latinae linguae vocabulis compositis," Breslauer Disserta- tion, 1865; G. V. Muyden, " De Vocabulorum in lingua Latina compositione," Halle, 1858; F. Seitz, " De adjectivis Latinorum poetarum compositis," Bonner Dissertation, 1878; F. Stolz, "Die lateinische Nominal-composition," 1877. [For compounds in French, see Darmesteter, His. Fr. Gr., § 272 sqq^ (9) § 7- " ^^ ^^y circumstance has made a particularly deep impression on the spirit of the people, this spirit is tempted to forge new expressions to meet the occasion: and to disclose ever new features in that spirit with a manifold redundancy of words. Every characteristic attri- bute which struck the fancy of a new observer yielded a new name " (O. Kares, " Jahrbucher fur Phil," 1884, ii, 595). [See H. Heine's witty application of this thought: " Reise von Munchen nach Genua, kap. iv, ad init." (Hamburg, Hoffman, 1871), and Whitney, "Language and the Study of Language," p. 123, Triibner and Co., 1870.] 228 NOTES NOTES 229 III (oa) 5 8 [So the name Palatium and Mons Palatinus seem to answer to the deities of the domestic hearth, Pales and Palatua.] .,..., -o (10) iJ 9. W. Cosack, "Bild und Gleichniss m ihrer Be- deutung fur Lessing's Stil," Danziger Progr., 1869, and Immisch, " Jahrbucher fur Philol," 1887, pp. 393 sqq^ (lO S 9. Reisig remarks in his " Vorlesungen uber lateinische Sprachwissenschaft," § I73: "We can com- monly learn certain characteristic traits of a nation by the Figures of speech which it employs, particularly in the case of certain special objects of its taste " ; and Jak. Bauer remarks in the Ansbacher Programm of 1 889, p. 33 : ^^^ peculiarities of a nation are in no way more clearly mir- rored than in its metaphors." See Brinkmann, " Die Meta- phern, Studien uber den Geist der modernen Sprachen, Bonn 1878; R. Thomas, " Zur historischen Entwickelung der Metapher im Griechischen," Erlanger Disputation, 1891 ; H. Bliimmer, " Studien zur Geschichte der Metapher im Griechischen," Leipzig, 1891; Burmester, "Uber den Einfluss der Metapher auf die Entwickelung der Sprache, Barmer Programm, 1863; A. Darmesteter, "La vie des mots," Paris, 1887, pp. 96 sqq. (12) § 10. Cf.the Sabine word curis-. Lange and Momm- sen, Rom. Geschich. 7 Aufl. i, 69. (13) § 10. The great influence exercised by the ideas ol the Romans on their national proverbs is brought out by Wolfflin, " Sitzungsberichte der Munchener Akademie, 1888 pp. 197 sqq. It is also worth noticing that the phrase- ology connected with bellum has had a large development; e.g., bellum indicere, movere,concitare,conflare,parare,ducere, trahere, componere, conficere, finire, etc. Again, some old Roman names were taken immediately from words con- nected with war, as Duilius^Duellitis, from duellum = bellum, Metellus = viercenarius, paid soldier (after Festus, P 147) [from metere: Duellius and Bellius are referred by Walde rather to bellus = bonus, Pliny the Younger (vi, 12) refers to a law court as his arena]. (14) I 10. Cf. Ribbeck, " Geschichte der romischen Dich- tung," i^ 123, and Kampmann, "Res Militares Plauti," Breslau, 1839. (15) § II. References to Law and matters of Law are very common among the Romans. Cf H. Demelius, " Plautinische Studien," Zeitschrift flir Rechtsgeschichte, i (1862), 351-372; ii, 177-238; J. Mispoulet, "Revue de Philol." xii, I sqq. Jubere is properly " to hold as right " (Jus Qfivai) [the connection of iubeo and ius is doubtful, V. Walde s.v.], arbitrari " to appoint an umpire," etc. (16) § 13. The Germans [and English] seem to speak more sympathetically than the Latins, as may be seen in such expressions as " our poet," " our author," " our volume," etc., as compared with the more colourless hie poeta, hie scriptor, hie liber, (17) § 14. It is improbable that the word ludus^Lydius, implying an influence exercised by Lydia on Etruria [Walde connects it with a root from which comes Anglo- Saxon gleo, our " glee "]. (18) § 14. These two words are referred by others to ajo, I say, agjo \agio\. (19) § 17. More details are given in Lohmeyer's " Zeitschr. des allgemein. deutsch. Sprachvereins," iv, i, 5 sqq., and in W. Wackernagel's " Schweitzerisches Museum," i, i, 69- 119. (20) § 17. Fick, "Die griechischen Personennamen," Gottingen, 1874, cites about 300 names of both classes. (21) § 19. Cf. the excellent collection of pertinent pro- verbs by Otto, " Archiv fur Lexicographic," iii, 355 sqq,, and W. V. Wyss, " Die Sprichworter bei den romischen Komikern," Zurich, 1889 bes. pp. 12 and 47, and for the Greek proverbs, the Programme of Martin, Plauen, 1889. (22) § 20. Also in the form "audentes fortuna iuvat" (Verg. Aen. x, 284), or the same sentiment abbreviated, "sed fortes fortuna" (Cic. De Fin. iii, 4, 16); or "fortibus est fortuna viris data" (Enn.); "Fortuna fortes metuit, ignavos premit" (Senec); "audendum est, fortes adjuvat ipse deus" (Ovid); "dimidium facti qui coepit habet; sapere aude" (Hon); "omnia deficiant, animus tamen 2 30 NOTES NOTES 231 I III omnia vincit; ille etiam vires corpus habere facit" (Ovid); or in various other forms. Cf. Buchmann, " Geflugelte worte," 21 Aufl., pp. 3^3 m- [Cic De Off. i, 23, " Fortis animi est non perturbari in adversis "]. (23) § 23. Cf. too Tac. Ann. iii, 12, " id solum Germanico super leges praestiterimus," besides the frequent use of videro. Also the Conjunctive Perfect, which occurs so often in sentences of commanding and willing, and m doubting questions and in the Potential seems always to imply that the speaker is under a strong emotion and full of decision, and especially that he lays stress on the speedy termination of the action spoken of (cf. H. C. Elmer, " Studies in Latin Moods and Tenses," Ithaca, N. York, 1898 and Weise's review of this treatise in the "Berliner Philol. Wochenschrift," 1898, No. 38, Sp. 1173 m^ We should mention in this place the rhetorical Pluperfect employed more especially by the historians, instead of the Perfect, when they are intent on calling attention to what is to follow, and pronounce their judgment from the point of view of the following action. (Cf on this point H. Blase, " Gesch. des Plusquamperf. im Latein," Giessen, 1894, pp. 38 sqq.\ also Schmalz' note on Catil. 18,6; Madvig, § 338; Kuhner, " Ausfuhrliche Gramm.," § 35, 3-) (24) § 24. Cmn concessive or adversative is used by Plautus with the Indicative only: it is found in Terence sometimes with the Indicative, sometimes with the Sub- junctive: quippe is, even in Sallust, always constructed with the Indicative. A. Dittmar is hardly right in attempt- ing to prove, as he does in his " Studien zur lat. Modus- lehre," that the Conjunctive, wherever occurring, has a polemic character, and is the expression of some mental excitement; and that its use is thus explicable as ex- pressive of doubt, irritation at contradiction, or some other emotion (cf Weise's review of this treatise in the " Literar- isches Centralblatt," 1897, Sp. 1464^^^- • and in the "Berlmer Philol. Wochenschrift," 1897, Sp. iSQi)- W. G. Hale views the matter from another standpoint in his treatise, " The Cum constructions," Ithaca, N. York, 1887 and 1889; his opinion is that the Conjunctive in ^ww-sentences is just as little conditioned by causal or adversative considerations as by the fact that the relation of subordination or the sub- jective nature of the speaker's point of view exerts its influence. He expresses himself thus: "The conjunctive rw;«-sentence expresses the situation actually in being at the time of the occurrence of the main action. The in- dicative <;//;;/-sentence expresses the time or date at which the main action occurs." The former, then, answers the question: ''How stood matters when the main action occurred?" The latter answers the question: " What was the date of the main action? " The following treatises on this question are also worthy of attention: E. Hoffmann, "Die Konstruktion der latein- ischen Zeitpartikeln," Wien, 1873, and "Das Modus- gesetz im lateinischen Zeitsatz," Wien, 1 891, which assume that the use of the two moods respectively in Latin time- sentences depends on the difference between absolute and relative time: M. Wetzel, "Das Recht in dem Streite zwischen Hale und E. Hoffmann uber die tempora und Modi in latein. Temporal satzen," Paderborn, 1892; Stegmann, "Jahrb. f. Philol.," Bd. 142, pp. 454-474; Heynacher, " Wochenschrift f. klass. Philologie," 1890, pp. 739 sqq., and Lubbert, " Die Syntax von quoin und die Entdeckung der relativen Tempora im Latein," Breslau, 1870. (25) § 29. The same holds good of other iterative clauses in Livy with quantum, quod, utcmnque, etc. Cf. O. Riemann, " Etude sur la langue et la grammaire de Tite-Live," 2 Aufl., Paris, 188S, pp. 294 .f^^. .. (26) § 29. H. Ziemer says in his treatise, " Uber das psychologische Moment in der Bildung syntactischer Sprachformen," Programm von Colberg, 1879, p. 8: "There can be no dispute that the Latin language during its course of eight hundred years, if we may judge from its documentary evidence, has undergone fewer changes than other tongues, such as the German [and the English], in a like space of time." We must also agree with G. Curtius, 232 NOTES NOTES 233 who lays stress on the larger capacity of Greek for ex- pression generally, and especially for the admirable subtlety displayed in the combination of its sentences. In Greek we findamorecopious dialectic literature than in Latin ;thequick and subtle mind of the Greek developed a vast redundancy of forms, and we find prevailing over the entire language undeniable traces of the activity of the " Psychological moment." The Latin language, on the other hand, mani- fests in its development, as disclosed to us, greater con- sistency, greater simplicity, and much less freedom; in its syntactic forms it follows more closely the Laws of Logic. (27) § 30. Lubbock ("Origin of Civilization," p. 403) declares that in the Brazilian dialect Tupi, out of a thousand words sixty-six are reduplicated; among the Hottentots, seventy-five; in the Tonga dialect one hun- dred and sixty-six, in the Maori one hundred and sixty- nine. Cf. also Deecke, " De reduplicato Latinae linguae praeterito," p. 19, and C. Jacoby, " Die reduplication im Lateinischen," Danziger Programm, 1878. [On reduplica- tion in the Polynesian dialects, see Whitney, " Language and the Study of Language," p. 338: the languages of the Australian aborigines are also largely characterized by reduplication. The repetition of the root, either complete, or by " reduplication," i.e., the repetition of its initial part, was made to indicate symbolically the completion of the action signified by the root, and furnished another tense, a perfect: e.g., from the root da = give, Sanskrit ddddu, Greek ^f^ojxa, Latin dedi, from dha, put, make, Greek ii^zina, O.H.G. teta, A.S. dide, our did. Whitney, p. 267 ; see also pp. 338 sqq. for reduplication in the Polynesian languages.] (28) § 30. E.g., " Super unus eram," Verg. Aen. ii, 567 ; " inque cruentatus," Ovid, Met. xii, 492 ; " Hac Troiana tenus," Aen. vi, 62 ; " quae me cumque vocant," Aen. i, 60. [" Inque salutatam linquo," Aen. v, 28, and Ennius, " Cere— comminuit— brum ! "] Lucret. "inque pediri," "conque globata," "ordia prima "; Cicero, Sest. 68 : "quod indicium cumque subierat"; Ad Attic, v, 18: " faciam tamen satis." Seyfifert-Muller on Laelius, p. 49; Schmalz, " Jahrbucher f. klass. PhiloL," 1892, ii, 364; M. Bonnet, "Le Latin de Gr^goire de Tours," p. 480. (29) § 32. The Greek is extremely sensitive as to the terminations of his words, in which he allows hardly any double letters, except such as those of which ^, p and v are one factor [such words as Tyrins are now held to be pre-Hellenic]: in the beginnings of his words, however, he admits a comprehensive number of consonantal collocations: the Roman, on the other hand, shows a greater sensitive- ness as to the beginnings of his words, in which he avoids such combinations as cm, dm, tm, sm : en, dn,pn, inn : et,pt: bd, gd: ps, X, tl, scl {stl), all of which come naturally to the Greek. On the other hand, in the endings of his words the Roman admits of a series not merely of simple consonants, but also of consonantal combinations, e.g., nt, rt, st. It: rs, ms, ns: nc. It seems as if the Roman were bent on verify- ing in his treatment of the vowels what Hanno says in Livy (Book xxi, 10, 7) about the national character of his enemies: "Quo lenius agunt, segnius incipiunt, eo, cum coeperint, vereor ne perseverantius saeviant." Cf. too Benary, " Kuhn's Zeitschrift f. vergleich. Sprachwissensch.," i, 51 sqq., and R. Kretschmer, loc. cit., xxxi, 412 sqq, Latin writers, again, seem not to feel the same objection to the hiatus which characterizes Greek writers. (30) § 32. Alliteration played a great part in the tech- nique of old Latin verse. Cf. S. Preuss, " De Bimembris Dissoluti apud Scriptores Romanos Usu Solemni," Eden- koben, 1881 ; and W. Ebrard, " Die Alliteration in der lat. Sprache," Bayreuth, 1882. This also holds true of the early stage of the German language: cf. Heine, " Die Alliteration im Munde des deutschen Volks," Anklam, 1882 [and was common in Early English, where it was an essential part of the rhythm. Cf. Morris and Skeat's " Specimens of Early English," pp. 151 sqq. The Deluge and the vision of William concerning Piers the Plowman.] (31) § 34. Cf. W. Stehlich, "Die Sprache in ihrem Verhaltnis zur Geschichte," Leipzig, 1892; and Weise's 234 NOTES treatise on the German mother tongue. 5 Aufl., Leipzig, 1904, pp. 87-104. J. Grimm, " Kleine Schriften," 1, 290, says: " Our language is at the same time our history," and W. V. Humboldt, in his treatise on the Kawi language, says: "Language is intimately bound up with the develop- ment of mankind : it accompanies it at every step of its progress as if its retrogression, and the state of civilization of those who speak it, is recognizable by its aid [of. Whitney, pp. 383 ^^^-l _ (32) § 40. Cf too O. Altenburg's essay " De sermone pedestri Italorum vetustissimo," Leipzig, 1898: "In the records of old Italian popular language, such as ancient specimens of law, the writings of Cato, the municipal laws of Bantia, the Eugubine tables, we meet with the same absence of form and grace in style. The thought presses ever to the front: the expression is of secondary import- ance. This is the peculiar mark of the style of the old Latin prayers, so grand in their very simplicity, of the Laws, and of Cato's precepts with their somewhat pedantic tone and character." (33) § 40. " M. Catonis quae exstant," ed. Jordan, p. 77' " Et hoc puta vatem dixisse, quandoque ista gens suas litteras dabit, omnia corrumpet." (34) § 43- Old Latin words found in Ennius, but not in later Latin literature, have been collected by A. Reichardt, -Jahrbucher fur PhiloL," 1889, pp. 81 sqq.: the old Latin words in Plautus by H. A. Koch, Rhein. Mus. xxv, 617, and S Bugge, " Neue Jahrb. f. Phil.," 1872, 91 s^^- For hybrid formations, see Tuchhandler, " De Vocabuhs Graecis in Linguam Latinam translatis," Berliner Dissert., 1876, p. 64; and Weise's essays in Bezzenberger's " Beitrage zur Kunde der Indog. Sprachen," ix, 90 sgq., and Philologus, N. F. i, 45-52. (35). Cf. Schulze, " De Archaismis Sallustianis," Halle, 1871 ; Briinnert, " De Sallustio imitatore Catonis, Sisennae aliorumque veterum historicorum Romanorum," Berlin, 1864. , - . (36). Cf. more especially § 44- In Cato alone we find no NOTES 235 less than five similes drawn from the theatre: 2, 6; 10, 48; 18,64; 19, 70; 23, 86. (37) § 50. Cicero (De Fin. iii, 2, 5) pronounces this opinion as to the borrowing of Greek terminations: " Quodsi in lingua concessum est, ut doctissimi homines de rebus non pervagatis Jnusitatis verbis uterentur, quanto id nobis magis concedendum, quia ea nunc primum audemus attingere? " (38) § 51. Cf. Herder, " Samtliche Werke," ii, 11, 258, der Cottaschen Ausgabe von 1862: "Such names were rejected by a general code of honour as improper: the objects denoted by such names, however, are not regarded as improper: nor indeed is there any diminution in the desire to find some way of indicating these objects, inno- cent as they are, and to do this gracefully. This is the origin of the polite words d double entente of modern society. Two or three expressions were ostracized from the standard language of respectability and consigned to the populace. But twenty periphrases, fifty "flowers of speech," and a hundred expressions a double entente were accepted in their place. These pass unperceived save by the subtlest minds. And this was called " the modest and simple language of the century " [cf. hoovufAog : for apio-Tspo;]. (39) § 53- He was reproached that it was his way, "a prisca consuetudine movere et ad formas Graecas verborum magis revocare." For the Greek inflexional terminations in Roman poets cf. L. Sniehotta, " De vocum graecarum apud Latinos poetas ab Ennio usque ad Ovidi tempora usu." Breslauer philolog. Abhandlungen ix, 2, Breslau, 1908; A.Thiel,"Juvenalisgraecissans," Breslau, 1901, pp. i43 •^^^• (40) § 54. Even Livy borrows much from the Augustan poets, especially from Vergil, and not merely single words but entire phrases like "haec ubi dicta dedit," xxii, 50, 10; "nubes iaculorum," xxi, 55; cf Aen. x, 8o8,"nubes belli." Cf. also Decolle, " Reste elegischer Poesie im Livius," Ber- liner philolog. Wochenschrift, 1892, Sp. 835, and Stacey, « Die Entwickelung des Livianischen Stils," Archiv f lat. Lexicogr. x, 17 sqg. (1898). Of 319 words created by Vergil 236 NOTES NOTES 237 no less than 57 are repeated by Tacitus. The articles by A. Czyczkiewicz, " De Taciti sermonis proprietatibus^ prae- cipue quae ad poeticum dicendi genus pertineant." The articles by Brody, 1890 and 1891, and those by Dosson, "Etude sur Quinte Curce," Paris, 1887, 278 sqq., are also worth consulting. (41) § 55. He never employs them in letters addressed to Tiro and Atticus, and in those to Terentia he does so out of "imple politeness: on the other hand he employs them regularly in official documents and in replies to people who had themselves employed them. (42) § 57. The " Argonautica " of Valerius Flaccus con- tain III similes: on the numerous metaphors found in Tacitus cf. Drager, " Einleitung zu Tac. Ann." pp. 30 sqq. ; A. Stitz, "Die Metapher bei Tacitus," Krems, 1883, 1884; Joh. Kitt, "De Translationibus Taciteis," Konitz, 1884; on the Personifications of Tacitus see the work of F. Meyer, Gottingen, 1884. (43) § 57- There are striking resemblances, too, m the writing, and in the architecture of both periods. The shapes of the letters are in both periods curved and spread out: in architecture during the empire under the Claudian dynasty vanity and luxury caused gigantic buildings to be raised, while in the time of the Antonines the buildings were overladen with ornamentation, just as in Germany during the prevalence of the Barocco style. (44) § 58. Quintilian, too, recommends the employment of archaic words (i, 6), provided that they be only used occasionally, and not too ostentatiously paraded. Cf., too Gell. Noct. Att. i, 10. (45) § 5S. The style of Tacitus and its historical develop- ment is treated by E. Wolfflin, " Philologus," xxv, 92 sqq.\ xxvi, 92 sqq.\ xxvii, 113 sqq. Cf, too, E. Norden, " Die antike Kunstprosa," Leipzig, 1898; ii, pp. 321 sqq, (Tacitus) : also Gontrelle, " Grammaire et style de Tac," Pans, 1874; E. Wolff, " Die Sprache des Tacitus," Frankfurt a M, 1879; Drager, " Uber Syntax und Stil des Tacitus," 3 Aufl. Leipzig, 1882; Constans, " Etude sur la langue de Tacite," Paris, 1893; E. Kucera, " Uber die taciteische Inconcin- nitat," Olmiitz, 1882; C. Clemm, "de breviloquentiae Taci- teae quibusdam generibus," Leipzig, 1881; R. Schmidt, " De ellipsi Tac," Dramburg, 1871. (46) § 60. H. Corvinus in the " Zeitscrift fur gymnasial- wesen" (1890), p. 319, says: " In poetry as contrasted with the dull sobriety of prose, ordinary subjects of apprehen- sion, the creations of poetic fancy, seem actually trans- figured : they bear the same relation to the conceptions of Prose as the image mirrored on the blue water-surface bears to the stiff object mirrored, standing out in bold relief under the sober light of day. Just as the mirrored image attracts our gaze with its supernal charm, so does the ever enchanting profundity of the poet's words attract and captivate the soul of his hearers." On the poetical Lan- guage of Rome cf J. Golling, "Syntax der lateinischen Dichtersprache," Wien, 1892; Kone, " Uber die Sprache der romischen Epiker," Munster, 1840; L. Miiller, " Q. Ennius, eine Einleitung in das Studium der rom. Poesie," Peters- burg, 1884; R. Stern, " Grundriss einer Grammatik fiir rom. Dichter," Arnsberg, 185 1 ; C. G. Jacob, " Quaestiones epicae siva symbolae ad grammaticam latinam poeticam," Oued- linburg, 1839. (47) § 62. Just so C. Humbert in a treatise on the laws of French verse has shown that the spirit of the French language, and also the French national character, exhibits itself in the poetry of the French nation: particularly in the stress accent, the dislike of the massing of consonants and of hiatus ; and Herder says : " Poetry is the very Proteus of the nations : it changes its form according to their language, their customs, their habits, their temper- ament and their climate; yes, and even according to their accent." [Cf Tobler vers frangais, Paris, 1885.] (48) § 62. In Ennius, out of 519 verses, 31 end in words of four syllables, in which the first two syllables are short, so that this peculiarity is found in that poet on the average once in every 17 verses, in Lucretius once in every 36 verses, and henceforward it becomes rarer and 238 NOTES rarer. In Catullus the proportion is i to 134; in Horace's Epistles I to 197 (Satires, i. 83); in Vergil i to 261 ; in Ovid I to 1,500; cf. also W. Meyer," Zur Gesch. d. Griech. und lat Hexameters," Munchen, 1884; C. F. Hultgren. "Die Technik der rom. Dichter im Epischen und eleg. Versmasse," Jahrbucher fur Phil. 1873, 745 m'\ Lorey, « Die Schwierigkeiten der Anwendung der Griech. Metrums auf die lat. Sprache," Hameln, 1874. (49) § 64. Cf., too, Fisch, " Programm des Andreas-Real- evmnasiums zu Berlin," 1888, p. 23. (50) %^7^ Cf CFreytag," Technik des Dramas,^ p. 275. J H von Kirchmann is of the same opinion in his intro- duction to the study of philosophical works, p. 27: "The Sciences concern themselves merely with general concep- tions of things: the Fine Arts on the other hand aim at the representation of a particular object ; it may be a monument of architecture, of a statue, of a picture, or a piece of music. Poetry likewise creates some such special object or unity in the imagination of the poet: but since the latter, in order to impart an appreciation of his picture to others, can only employ conceptions of general application, it follows that he never perfectly achieves his end,^ and the picture given by poetry hovers between the general concept and the in- dividual unity. This explains the fact that the poets, in the construction of their language, aim at individualizing their subjects, and making them stand out in bold relief, while the thinkers are constant in their endeavours so to develop their language that it may serve to express general con- ceptions and lofty ideas." ^ (51) §70- O" ^^^ Figures of Speech in the Roman poets, and more particularly on Synecdoche and Metonymy, see E. Lindskog, " In tropos scriptorum Latinorum studia, Upsala, 1903. ^ \ c^A \ (•52) 5 73. This tendency appears very strongly, Ud. 1, 36, where he speaks of " Erycina ridens, Quam locus cir- cumvolat et Cupido," etc. Cf, too, Od. iii, 24, 4; 1, 3S, 17; * For this view cf Nietzsche, " Origin of Tragedy," § 6 ad fin. NOTES 239 if, 17, 22; 17, IS; ii, 2, 7; iii, 2, 32; iv, 5, 24; iv, 14, 4; Tibull. i, 9, 4; Propertius, iv, 22, 20; and G. Dannehl, " De Tropis I : De translationis, metonymiae, synechdoches apud poetas Augustei aevi usu," Hallische Dissert, 1868. (53) § 75- Cf Catull. 61, 202; Prop, iii, 15, 31; 32,49; 1, 15, 29; ii, 3, 5; Verg. Eel. i, 59; i", 9i; iv, 91; Hor. Ep. 16, 31 ; Ov. Ex Pont, ii, 6, 37; iv, 5, 41 ; Met. xiii, 324; xi, 315; Trist. i, 8, i; iv, i, 57; Ars Am. i, 748. On late Roman poets like Claudian and Nemesianus consult Biese, "Naturgefuhl bei den Romern," p. 143: on the same subject in German see Weise's " Abhandlung in der Zeit- schrift fiir hochdeutsche Mundarten," iii, 47 sqq, (54) § 75- Even descriptions and sketches are affected by the influence of all powerful rhetoric, and Seneca has good grounds for jesting (Apocolocynthosis, 2, 3) at the poets, because, as he says: " acquiescunt oneri poetae, non contenti ortus et occasus describere, ut etiam medium diem inquietent." The places in which all the tricks and devices of rhetorical technique are most fully displayed, are the] speeches placed by the Epic poets in the mouths of their heroes, and the Dialogue of the Drama. The representation of the struggles which agitated Dido in the Aeneid, or Medea in Ovid, may well be paralleled with the regular Suasoriae. Aspiring orators modelled them- selves on these and other masterpieces, indeed Vergil, during the Empire, was always regarded as the Classic aid to rhetorical studies, and employed as such. Cf., too, H. Peter, " Rhetorik und Poesie im klassisch. Altertum.'* Jahrbucher fur das klass. Altertum, 1898, i, 637 sqq. (55) § 76- The first figure is found in his works 102 times, in the second book of 1,362 verses 62 times, while TibuUus employs them in the first two books, i,e., in 1,352 verses, only 24 times: the last occurs in Propertius more frequently than in all the rest of the Roman poets together, e.g,, i, I, 19, 20, 39, etc. ^ (56) § 'jj. The historical development of this infinitival construction is traced by Drager, and also by Schmalz in Iwan Miiller's Handbuch, ii, pp. 3^9 •^^^- (§ 217 -S"^^- of his 240 NOTES Latin Syntax). Special notice is taken of the works of Ennius Lucretius, and the Augustan poems. (57) § 78. Cf., too, Heerdegen, " Untersuchungen zur lateinischen Semasiologie," ii, p. 64. ^ _ M) 5 78. E.g. Aen. ix, 98-103; vi, 45i; vni, 213, 407, xi 309 In Propertius, too, we meet with long periods, eg i II 9.18; iii, 14, i-io: Tibullus avoids them. On Lucretius and Catullus see above. § 24. rcQ^ 5 80 On these and other characteristics of the Ian- guage of poetry consult '' Phil. Wegener, Neuhaldenslebener Frogramm," 1889, pp. 18 sqq. (60) §80. Cf. Hor. Ep. II, 2, 115: Obscurata diu populo bonus eruet atque Proferet in lucem speciosa vocabula rerum, Quae priscis memorata Catonibus atque Cethegis Nunc situs informis premit et deserta vetustas. Goethe, who in his "Iphigeni." and his "Tasso"uses the best literary German, employs in others of his poem^ strikincx archaisms: as, for instance, in his" Gotz, in some parts of "Faust," in the legend of the horseshoe and n Hans Sachs' " Poetische Sendung." He has complete y succeeded in his design of reproducing the old-wodd colour, and in suiting the language of the P-sonages^^ introduces to the old-fashioned times in which they - and speak. In this free intermixture of words and forms o? wofds of different ages Poetry found a rich recompense for the retrenchment of vocabulary imposed upon it by the exigences of metre. For many words could not accom- modate themselves to the rhythm, and these had accord^ insly to be excluded from the poet's use and replaced by others. Thus Homer could not employ words like .oxs/x., rhe uses ^&. .x,.xoWov, etc.): in the same way v^fuperare was useless for the purpose of the dactylic Poet^ ^nd had to be replaced by refrekendere E"7^^^" j/^^^^^^^^ use feroda and tenacta, but not the words in^ -fas which correspond thereto, because they would not suit the verse For qmUHordecimVevg,. Aen. i, 71, ^^ses dis septent', for NOTES 241 explicari we have in the same poem explicui-. for capiti- ^aj ;(Aen. ii, 219), in spite of the fact that the word is connected with cervicibuSy we have the singular capite. Celeriter with its four short syllables was replaced by celer^ cituSy etc. In the case of forms like deeraty deindey arjetey parjetibus (Aen. ii, 442), semjanimisy alteriuSy etc., recourse was had to synizesis, dissolution of semi-vowels or shorten- ing of vowels. (61) § 82. Herewith should be mentioned the essay of Reichardt, " Jahrbiicher fiir Philologie," 1889, i, pp. 797 sqq.y on the archaisms in Vergil: cf., too, Wotke, "Wiener Studien," viii, 1 31-148. (62) § ^2. E. Appel, " De Genere Neutro intereunte in Lingua Latina," Munchener Dissert., Erlangen, 1883, remarks that the poets who employ hexameters prefer to write gaudiuy incendiay conviviay etc., rather than gaudiufUy incendiuniy conviviujn with elision. In Ovid the Plural occurs more than 50 times, the Singular not once. Even syntactic innovations follow under the stress of metre. As " cruribus tenus " would not come into hexameter verse, Vergil writes (Georg. iii, 53): " Et crurum tenus a mento palearia pendent," and Catullus, for the same reason, writes for " nutricibus tenus," " nutricum tenus," 64, 18. This was remarked even by the ancients: and thus we read in the Corp. Gloss, v, 248, 19: " tenus praepositionem Vergilius necessitate metri genetivo pluralis inuxit." Cf., too, E. Wolfflin, " Hexameter und Silberne Prosa," Archiv f. lat. Lexicogr., xi, 503 sqq, (63) § 82. Ribbeck, "History of Roman Poetry," ii, 339: " Certain beginnings and endings of verses, the choice and position of certain words, certain similes and figures of speech, were for definite purposes bequeathed by one poet to another, and their use became traditional." Cf., too, A. Zingerle, "Ovid und sein Verhaltniss zu den Vorgangern und den gleichzeitigen romischen Dichtern," Innsbruck, 1 869- 1 87 1 ; Schmalz, " Zeitschr. fiir Gymnasial wesen," 1 890, 718 sqq, (64) § 83. E.g. Vergil, Aen. iv, 451: "it clamor caelo." R 242 NOTES Ovid, Met. ii, 580: " tendebam brachia caelo." Vergil, Georg. iv, 562: " viamque affectat Olympo." Hon Od. 1, 28, 10: "Oreo demissus," Prop, i, 15, 29: "nulla prius vasto labentur flumina ponto." Aen. vi, 126: " facilis descensus Averno." (65) § 84. The new words formed by Ovid have been collected by Drager in the Auricher Programm, 1888, p. 1 7- He calculates their number at 392, including 153 which occur in his writings alone, and 139 *^«^ £i>/x£v« like repostor, novatrix, renovamen. According to E. Linse, " De P. Ovidio Nasone Verborum inventore," Leipziger Dissert., 1891, the number of these new words is 487. Besides these, the following works are worth consulting: H. Ploen, « De copiae verborum differentiis inter varia poesis Ro- manae antiquioris genera intercedentibus,'' Strassburg, 1883, with interesting collections of words in -tudo, -tas, •ntia\ Deipser, " Uber die Bildung und Bedeutung der lat. Adjectiva auf-/^r und -ger;' Bromberg, 1886; Seitz, " De Fixis Poetarum Latinorum epithetis," Elberfeld, 1890; Ladewig, "De Vergilio verborum Novatore," Neustreliz, 1870; E. Stephani, " De Martiale verborum novatore," Breslau, 1889; A. Rothmaler, " De Horatio verborum in- ventore," Berlin, 1862; C. Zangemeister, " De Horatii vocibus singularibus," Berlin, 1862 ; R Teuffel, " De Catulli, TibuUi, Propertii vocibus singularibus," Freiburg im Breis- gau, 1872; W.Schneider," De Propertio sermonis novatore et amplificatore," Strassburg, 1888; G. Bordello, " De linguae Latinae nominibus -men et -mento ope formatis,' Grossglogau, 1879; W. Wilbertz, " De adjectivis poetarum Latinorum usque ad Catullum compositis," Marburg, 1884. (66) § 86. Cic. Orator, 202: " Poetae transferunt verba cum crebrius tum audacius." De Or. iii, 43, 170^ "Trans- latum verbum maxime tamquam stellis quibusdam notat et illuminat orationem." On the metaphorical use oUurvus, uncus, etc., introduced by Horace, cf A. Moller, " Archiv f. Lexicogr.," iii, 117 sqq., and Preuss, "Die metaphorische Kunst Vergils in der Aeneide," Graudenz, 1894; R- Brau- NOTES 243 miiller, "Uber Tropen und Figuren in Vergils Aneide," Berlin, 1877 and 1882; P. Langen, "Die Metapher im Latein von Plautus bis Terenz," Jahrbiicher fiir Philol., 1882, pp. 673 sqq., 753 sqq.] S. von Raumer," Die Metapher bei Lukrez," Erlangen, 1893; L. Geuther, " Uber den Ge- brauch der Metapher bei Juvenal," Wittenberg, 1878. (^7) § ^7' The latest critical inquiry into the whole question is from the pen of J. Shafler, Amberg, 1884, p. 95. Other papers on the same subject appear in the "Zeitschrift fiir Gymnasialwesen," 1886, p. 23: see also M. Brenous, " Etudes sur les hellenismes dans la syntaxe Latine," Paris, 1895; Piger, " Die sogenannten Gracismen im Gebrauche des lat. Accus.," Iglau, 1879; Engelhardt, "Passive Verba mit dem Accus. und der sogenannte Accusativus graecus bei den latein. Epikern," Bromberg, 1879; G. Landgraf, " Der Ace. der Beziehung nach Adj., Subst. und pass. Verben," Archiv fiir Lexikogr., x, 209- 224; H. Tillman," De dativo verbis passivis linguae latinae, subjecto, qui vocatur Graecus," Acta semin. philol., Er- lang., ii (1881), 71-140; H. Dittel, "De infinitivi apud Horatium usu," Ried, 1881; G. Overholthaus, " Syntaxis Catullianae capita H Diss.," Gottingen, 1875; G. V. Bucht, "De Usu Infinitivi apud Ovidium," Upsala, 1875; E. Trill- haas, "Der Infinitiv bei Ovid," Erlangen, 1877; v. Steltzer, " Uber den Gebrauch des Infinitivs bei Vergil," Nordhausen, 1875; C. Wagener,"Der Infin. nach Adj. bei Horaz," Neue philol. Rundschau, 1902, pp. 1-9. In 65 places in Horace adjectives are found connected with an infinitive, and of these 32 are followed by a genitive case. {6S) § 88. The judgment pronounced by Cicero on the productions of Lucretius holds good in a greater or less degree of all the Roman poets (Ad Quintum fratrem, ii, II): "Non multis luminibus ingenii, multae tamen artis." Their imaginative powers were by no means extensive: their greatest success lay in Elegiac poems and in Satire, which suited the particular bent of their genius. There were very few who could say with Ovid : " Quidquid tenta- bam dicere, versus erat" [or with Pope: " I lisped in num- 244 NOTES bers, for the numbers came." Quintilian's proud boast may be remembered here : " Satira tota nostra est "]. (69) § 89. Cf. K. Sittl, " Jahresbericht liber das Vulgar- und Spatlatein," 1884- 1890, im Jahresber. uber die Fortschr. d. klass. Altertumswissensch., Ixviii, 226-286; P. Monceaux, " Le Latin d'apres les dernieres publications," Revue des deux Mondes, 1891, 15 JuH, 429-448; M. Bonnet, " Le Latin de Gr^goire de Tours," Paris, 1890, 783 PP-, numerous papers in Wolfflin's " Archiv f. Lexikogr.," Leipzig, 1844, sqq,\ Wolfflin, " Uber die Latinitat des Africaners Cassius Felix," Munchen, 1880; H. Ronsch, " Itala und Vulgata," 2 Aufl., Marburg, 1875; H. Hoppe, " Syntax und Stil des Tertullian," Leipzig, 1903; H. Glasener, " Grammatik des Laktanz," Musee Beige, 190x3, pp. 26 sqq., 223 sqq., 1901, pp. 65 sqq,, 293 sqq, ; W. Kalb, - Roms Juristen nach ihrer Sprache dargestellt," Leipzig, 1890; F. Polle, "Wie denkt das Volk liber die Sprache?" 3 Aufl., Leipzig, 1904; O. Rebling, " Versuch einer Charakteristik der rom. Um- gangssprache," 2 Abdruck, Kiel, 1883; P. Meyer, " De Ciceronis in Epistulis ad Atticum Sermone," Bayreuth, 1887; A. Skinner, " De eo, quo Cicero in epistulis usus sit sermone," Oppeln, 1879, sqq.\ R. Klein, "Uber Ciceros Briefstil." Chemnitz, 189S ; Koffmane, " Geschichte des Kirchenlateins," 1879. (70) § 89. Sittl assumes that there were three kinds of well-defined non-classical Latin: (i), the language of the peasantry {rusticitas\ and (2) the language spoken (not written) by the educated classes {sermo cotidianus, con- suetudo), (3) a dialect standing midway between the two spoken by the inhabitants of the small towns {pppidamim dicendi genus) ; see his lecture read before the Gorlitzer Philologenversammlung, 1889, and cf. " Jahrbucher fur Phil.," 1890, ii, p. 142. He will not allow either inscriptions or writers to be the authorities on Vulgar Latin, but he regards the Romance languages alone as such. The litera- ture commonly quoted as authoritative on the subject is, according to him, composed neither in refined nor in popular Latin, but merely in bad Latin. He probably goes too far NOTES 245 in this assertion. No doubt it is true that no literary record can give us sufficient data on which to found a satisfactory theory of the pronunciation and accentuation of Vulgar Latin, but the form taken by the words, and the syntax, and the peculiarities in the formation and signification of the words, are shown by the traits of agreement in all the authorities mentioned, to mark a dialect contrasting with the language of the refined classes : we may therefore fairly call this dialect " the vulgar tongue." (71) § 90. This holds good more particularly of the so- called Svarabhakti, i.e., the insertion of a vowel sound before or after r, /, n : this sound developed itself from the pitch accent of these liquids in cases where a consonant preceded or followed them, e.g.^ Terebo7iius= Trebonius : Militiades — Miltiades, Cf J. Schmidt, " Zur Geschichte des indoger- manischen Vokalismus," ii, 342-370; Corssen, " Vokalismus usf ," ii, 384 sqq. (72) § 90. For this section reference may be made to H. Schuchardt, " Der Vokalismus des Vulgarlateins," Leipzig, 1 866- 1 868; E. Seelmann, "Die Aussprache des Lateins nach physiologisch-historischen Grundsatzen," Heilbronn, 1885; K. Sittl, "Die lokalen Verschiedenheiten der latein. Sprache," Erlangen, 1882; Diez, " Grammatik der romanischen Sprachen," pp. 170 sqq.\ G. Landgraf, " Historische Gramm. der lat. Sprache," Leipzig, 1903. [Cf also Grandgent, " Vulgar Latin," Heath, Boston, 1907; Lindsay's Latin Grammar.] (73) §93. Cf. Schuchardt, "Vokalismus des Vulgarla- teins," i, 34, 232; O. Sievers," Quaestiones onomatologicae " in Ritschl's "Acta societatis philol. Lipsiensis," ii, 55-I04; M. Bonnet, " Le Latin de Gregoire de Tours," pp. 349 sqq.\ Bucheler-Windekilde, " Grundriss der lat. Deklination," Bonn, 1879; F. Neue, " Formenlehre der lat. Sprache," 2 Aufl., Berlin, 1875-1877. (74) § 93- Cf , too, the treatise by E. Appel, " De Genere Neutro Intereunte in Lingua Latina," Erlangen, 1883; W. Meyer, " Das Schicksal des lat. Neutrums im Roman- ischen," Halle, 1883; and H. Suchier,"Der Untergang der 246 NOTES geschlechtlosen Substantivform," Archiv f. Lexicogr., iii, 161 sqq. Forms like la riponse, la merveille, are to be ex- plained by the fact that plurals like responsa and mirabilia were treated as nouns singular of the first declension- [See the whole question well and fully treated by Darmes- teter, " Historical French Grammar," 1899, pp. 225-231. See also Grandgent, § 352: " In late Latin this collective plural in -a came to be taken for a feminine singular": cf " Ne forte et mihi haec eveniat," Ronsch, " Itala und Vulgata," 1869.] The Patristic Fathers actually preferred sometimes to change the classical genders ; cf. St. Jerome, who on Ezekiel 40 writes that he purposely substitutes cubitus for cubitum, to be better understood by his readers. (75) § 94. Ott, " Jahrbiicher fiir Philol. und Padagog.," 1874, pp. 781 sqq.\ Ronsch," Itala und Vulgata," pp. 22-257; H. Ulrich, " De Vitruvii copia verborum," Frankenthal, 1883, und Schwabach, 1885; Stunkel, " De Varroniana verborum formatione," Strassburg, 1875; R- Fisch, " Die lat. nomina personalia auf -o, -onis, ein Beitrag zur Kenntnis des Vulgarlateins," Berlin, 1890. (76) § 96. J. N. Ott, "Rottweiler Programm," 1874, reckons in all no less than 208 such substantives which may be omitted at will : cf., too, Drager, " Historische Syntax der lat. Sprache," i, 47 sqq., and T. C. Rolfe, " Archiv fur Lex.," x, 229 sqq. on the Ellipse o{ Ars. {77) § 99- More details are given in Andresen, " Uber deutsche Volksetymologie," pp. 17 sqq., and in Weise's essays on the " Charakteristik der Volksetymologie," in the " Zeitschrift fur Volkerpsychologie und Sprachwissen- schaft," Bd. xii, pp. 203 sqq., and in Bezzenberger's " Bei- trage zur kunde der indogermanischen Sprachen," Bd. v, pp. 68 sqq. : also Weise's treatise on the Greek words in Latin, Leipzig, 1882, pp. 67-75. [See, too, "English Folk etymology," by Smythe Palmer (Bell and Sons, 1882). " The violent dislike to the use of a word entirely new to us, and of which we do not understand the source, is a matter of daily experience; and the tendency to give a meaning to adopted words by so changing them as to NOTES 247 remove their seemingly arbitrary character has exercised a permanent and appreciable influence on every language" (Farrar, "Origin of Language," p. 56, quoted by Palmer, p. x).] (78) § 102. Cf.,too,J.Grimm, Gramm.,iii,726.r^^.; Diez, « Gramm. d. latein. Sprache," iii, 431 ^ i99» 217. Lucan, 8, 40, 41. Lucretilis, 8. Lucretius, 6, 46, 78, 88, 108, 124, 134. Lupercus, 36. Maecenas, 92. Maeniana, 218. Maenius, C, 218. Maia, 36. Manes y 219. Marcellus, 17. Marius, 216. Mars, 35. Martial, 151. Metaphors, 14, 119. from Agriculture and War, 15. tropical sense of words in, 1 7. from Sea and Chase, 79. of Homer, 80. of Aeschylus, Euripides, Pin- dar, Sophocles, 80. of Cicero, 87. of the people, 170. Metonymy, 118. Mimes, 177. Mommsen, 45. Months, Roman denomination of, 30. Mucius Scaevola, 2, 10. Naevius, 70, 124. Negations, Latin, 164. Neologisms, 135. Nepos, 191, 212. Neptunus, 36. Nero, 100. Nerva, 10 1. Nobilior, M. Fulvius, 81. Nonae^ 222. Novius, 17. Numen, 221. Nundtnum, 222. Odyssey, Latin, 69. Ops, 35. Oratio Obliqua, Latin, 48. 258 INDEX Oratio Rhodiensis, 71. Osci, 35. Ovid, 41, 93, 117, 120, 121, 125, 137, 217, Oxymoron, 122. Pacuvius, 78. Pales, 36. Parallelism, 125. Parcae, 176. Parenthesis, 99. Patristic Fathers, 146. Pecz, 80. Persona, 215. Petronius, 146, 151, 152. Philiscus, 82. Phrygio, 215. Pistor, 215. " Plastic " in language, 121. Plautus, 77, 124, 153, 165, 173, 177, 195. Pleonasm, 122, 163. Rhetorical, 202. Pliny, 12, 73, 98, loi. Polybius, 82. Polysyndeton, 122, 200. Pomona, 36. Pomponius, 13. Pontifices, 213. Praedium, 217. Praemium, 16. Praeneste, 73. Praetor, 212. Princeps, 16. PrincipeSy 216. Proper Names, Roman, 30. Propertius, 93, 95, 125, 128, 179. Prose, Decadence of, 95. Proverbs, 37. popular, 13. Puns, 41, 63, 125, 177. Quaestor, 212. Quintilian, 41, 42, 43, 66, 71, 77, 78) 97) loi, 102, 103, III, 145, 151, 182, 209. Quirites, 15, 214. Quonianiy 47. Raumer, S. von, 16. Reformation, The, 100. Regium, 161. Rhetoric, Schools of, opened, 83. Roman Literature, contrasted with Greek, 67. Roman poets, old, 75. St. Augustine, 173. Saiii, 214. Sallust, I, 85, 91, 97, 186, 191, 195, 199- Satura, 177. Saturnian measure, 68. Saturnus, 35. Schiller, 93, 114, 117, 204. Schroder, 169. Scipios, Circle of the, 81. Scribae, 81. Semitic character, no. Seneca, 6, 41, Z^, 98. Servius, 98, 132. Servius Tullius, 213. Sidonius Apollinaris, 168. Silver Latinity, 96, 97, 104, 194. Similes, 79, 119. Sinister, 219. Slaves, Roman expressions for II. Spoliare, 16. INDEX 259 Statius, 134. Stoicism, 10. Suasoriae, 92. Suetonius, 185, 208. Superlative, Latin, 44. Supplicium, 221. Sus, numerous synonyms of, Synecdoche, 118. Tacitus, 3, 22, 96, loi, 151, 192, 194. Tallyrand, 173. Tarentine War, 73. Tellus, 36. Templum, 218. Terence, 82. Terminus, 36. Tertullian, 153, 154, 173. Theocritus, 124. Thucydides, 90. Tiberius, Emperor, 94. Tibullus, 24, 95, 121. Timaeus, 189. Tiu, 34. Tmesis, 58, 76. Torquatus, T. Manhus, 45. Trajan, loi. Triarii, 216. 13- 184, Tribuni, 213. Tullianum, 162. Tusculan Orations, 189. Uhland, 130. Varro, 77, 218, 222. Vedas, Indian, 69. Velleius Paterculus, 85. Venerium, 219. Vergil, 3, 18, 22, 62, 93, 94, 96, 116, 125, 128, 132, 133, 134, 220. Verrine Orations, 188. Virtus, 28. Virtus, Temple of, 29. Vitruvius, 146. Vowels, Latin, 59. Vulgar Latin, 144, 186. syntax of, 152, 168. Wieland, 75, 114. Willman, O., 55. Winckelmann, 104. "Winged Words," 40. Wollner, D., 15. Wotan, 34. • ••*••• a •• ^ " •• • •••• b* • f> • • • • • • • •• •• •••• • • •• •• ••"' ••- » • • •• • a*,^* • » • » • « • CHISWICK PRESS : PRINTED BY CHARLES WHITTINGHAM AND CO. TOOKS COURT, CHANCERY LANE, LONDON. I I • • / COLUMBIA UNIVERSITY LIBRARIES 010692797 i\ C 00 :s k ICW