"* Λ 
 
 
 
 
 
*CL *• • » 
 
 
 <λ '•••* A° ,,, ~«* 
 
 V^ •ΐ£ί• «Ϊ6 4!»* »l*i- ^> ^ lii^C» 
 
 
 
 
 
 
 
 
 
 J* 4 A * 
 
 AT * 
 rV • 
 
 
 
 4.>* *Vi' > V 4 •Λ^> «k 4?* 
 
 
 
 
 
 
 
 
 
 
 %.. ••"• «* 
 
 >\.1X.% 
 
Critical Opinions of this Work. 
 
 JOHN BULL. 
 
 1 While the Author (the well-known Harrow Master) justly 
 apologises for the production of a new Greek grammar, he 
 fully justifies doing so, not so much because his colleagues 
 pressed him, as from the scholarlike and, above all, from the 
 intelligible manner in which he simplified his Greek Grammar 
 Rules into this Brief Greek Syntax, which bids fair to become 
 a standard work.' 
 
 EDUCATIONAL TIMES. 
 
 * Mr. Farrar's Greek Syntax differs in its method from all, 
 or nearly all, preceding Greek grammars ; partly in its freer, 
 larger, and more unhackneyed treatment of the subject, and 
 partly in its constant reference to the general principles of 
 comparative philology, and in its endeavour, wherever prac- 
 ticable, to illustrate the idioms of Greek by the similar idioms 
 or peculiarities of other languages, especially English. . . . The 
 whole of this Syntax is very well done. Mr. Farrar seems 
 to have a happy way of explaining an intricate subject ; and 
 we are sure that any fairly instructed youth will find no diffi- 
 culty in going through this volume without any aid from a 
 teacher. The Author has made his Greek Syntax indeed a 
 really readable work — something far beyond a compendium of 
 dry rules. He gives many apt quotations from some of our 
 ! best old English poets ; and illustrates, often very happily, not 
 ι a few peculiar constructions in Greek by reference to similar 
 i pages in other languages ... In freshness and interest, in 
 ! copiousness of illustration, and in its freedom from all 
 
Critical Opinions of this Work. 
 
 grammatical mysticism and pedantry, Mr. Farrars volume 
 surpasses all the Greek grammars -we have seen.' 
 
 MUSEUM. 
 1 Mr. Farrar has produced a book in every way admirable, 
 and calculated in no common degree to facilitate the study of 
 Greek, and to make that study profitable for the educing the ι 
 powers of the pupil. Mr. Farrar has shown by his previous 
 works that he was thoroughly acquainted with the subject 
 of comparative philology, and had taken a high place as an ! 
 original thinker and discoverer in that department. He has 
 applied his knowledge in this little work to the elucidation of 
 Greek Syntax. Perhaps the most striking feature in the book 
 is that Mr. Farrar grapples, in a fresh, independent way, with 
 every question of Greek Syntax that comes up. He knows 
 when he knows a thing with certainty, and he states what he 
 knows in remarkably clear and unmistakable language. He is 
 eqrally decided in knowing when a point is justly a matter of 
 doubt, and he is also equally distinct in stating where exactly 
 the doubt arises, and how it arises. This is a feature of the ' 
 utmost importance in a school-book. Most of the treatises 
 on Greek Syntax often leave the young student at a loss as to ; 
 what the meaning of the writer really is, and he is apt to go 
 away from the perusal of these treatises with vague, imperfect 
 ideas. This one feature of Mr. Farrar's work will recommend 
 it strongly to teachers. But there are many others which will 
 make it" exceedingly acceptable. Mr. Farrar carries his com- 
 parative philology into all portions of the work, and gives his 
 explanations of the formation of the tenses, of the derivations: 
 of particles, of the meaning of the various terms used in gram- 
 mars, and their history, and many other things only to be got 
 by much reading and research. He has also employed, to a 
 large extent, analogous examples from a variety of languages, 
 and he calls to his use not merely classical Greek, but the 
 Greek of the New Testament and modern Greek. In one 
 word he has made the study of Greek Syntax an interesting 
 study for boys, and he has done this at the same time that he 
 has amply satisfied all the demands of the present stage of 
 scholarship and of comparative philology.' 
 
GREEK SYNTAX. 
 
• Inter virtutes grammaticas habebitur aliqna nescire.' 
 
 Quinct. 
 
 • Non obstant has discipline per illas euntibus sed circa illas 
 haerentibus.' Id. 
 
A BRIEF 
 
 GREEK SYNTAX 
 
 AND • 
 
 HINTS ON GREEK ACCIDENCE : 
 
 WITH SOME REFERENCE TO 
 
 COMPARATIVE PHILOLOGY, AND WITH ILLUSTRATIONS FROM 
 
 VARIOUS MODERN LANGUAGES. 
 
 BY THE^J* 
 
 REV. FREDERIC W/^AKBAK, M.A., F.E.S. 
 
 \* 
 
 Honorary Chaplain to the Queen ; late Fellow of Trinity College, Cambridge; 
 
 Honorary Fellow of King's College, London ; one of the Masters at Harrow School : 
 
 Author of ' The Origin of Language,' ' Chapters on Language,' 
 
 'Families of Speech,' &c. 
 
 THIRD EDITION, REVISED AND ENLARGED. 
 
 LONDON: 
 LONGMANS, GREEN, AND CO. 
 
 1870. 
 
PA3C7 
 
 )270 
 
 irOTTISWOODE AliD CO., ITEW-STEEET SQUABS 
 A1TD ΡΑΕΙ/ΙΑΪΙΕΝΤ δΤΒΕΕΓΓ 
 
 GIFT 
 S ESTATE OF 
 UtfiLLIAW C. R 
 e X APRIL, 1940 
 
TO THE 
 
 EEV. H. MONTAGU BUTLER, D.D. 
 
 AND TO MY FRIENDS AND COLLEAGUES 
 
 THE ASSISTANT MASTERS OF HARROW SCHOOL 
 
 | gtbitate 
 
 ■WITH FEELINGS OF CORDIAL ESTEEM 
 
 "WHATEVER MAY BE FOUND WORTHY OF APPROBATION 
 
 IN THIS ATTEMPT TO RENDER THE STUDY OF GREEK GRAMMAR 
 
 BROADER, MORE INTERESTING, AND MORE FRUITFUL. 
 
PREFACE 
 
 TO 
 
 THE THIRD EDITION. 
 
 I have taken the opportunity offered me by the demand for a 
 third edition to revise this Syntax carefully, to add a con- 
 siderable number of illustrations, and to introduce some fresh 
 matter which struck me as likely to be curious, interesting, 
 or important. I have also corrected a few trifling blemishes 
 which have been pointed out by the kindness of friends or 
 reviewers. For the convenience of all who possess the pre- 
 vious edition, I have left the structure of the book and the 
 numbering of the sections undisturbed. 
 
 I trust that these improvements may secure for this Syntax 
 a continuance of the approval with which it has been generally 
 received. I have tried, even more than in the previous 
 editions, to illustrate many of the more remarkable idioms of 
 English Syntax by comparing them with similar idioms in the 
 classical and other languages. 
 
 April 1870. 
 
 Λ 3 
 
PREFACE 
 
 το 
 
 THE FIEST EDITION. 
 
 The publication of a new Greek Grammar -when there are 
 already so many in existence, is an act which requires justi- 
 fication ; and as it is also an act of some temerity, I will 
 briefly state the causes that induced me to undertake the 
 
 I observed from the comparison of a large number of 
 1 Grammar and Scholarship papers ' that the same questions, 
 — or questions involving the same points of scholarship, — 
 recurred with a remarkable frequency. As there is a Gram- 
 mar Examination every year at Harrow, I wished to draw up 
 for my own pupils a manual which should, in as clear a 
 manner as possible, give them some insight into these special 
 points. With the encouragement, and by the wish, of some 
 competent judges among the Harrow masters, I published in a 
 small compass my card of ' Greek Grammar Rules,' in which 
 I had attempted to fulfil this object; and in drawing up these 
 rules it appeared to me that many most valuable points 
 relating to them and to the general structure of the Greek 
 Language, had not hitherto found their way into any ordinary 
 schoolbook. I therefore thought that I could render a service 
 to the cause of Classical Philology, by amplifying my ' Greek 
 Grammar Rules' into a larger and fuller Syntax; and the 
 great favour with which the ' Rules ' were received, the 
 
PREFACE TO THE FIRST EDITION. IX 
 
 number of schools that adopted them, and the many eminent 
 scholars and teachers who wrote to me to express their appro- 
 bation of them, confirmed me in this belief. 
 
 I aimed above all things at making every point intelligible 
 by furnishing for every usage (as far as was possible) a 
 satisfactory reason ; and by thus trying to eliminate all mere 
 grammatical mysticism, I hoped that I should also render 
 grammar interesting to every boy who has any aptitude for 
 such studies, and is sufficiently advanced to understand them. 
 On the latter point I venture to lay some stress. I have 
 published elsewhere my reasons for believing that we com- 
 mence too soon the study of formal grammar, and that this 
 study, which is in itself a valuable and noble one, should be 
 reserved to a later age and for more matured capacities than is 
 at present thought necessary. I should never think of putting 
 this Grammar into the hands of boys who have no aptitude for 
 linguistic studies, or of any boys below the fifth or sixth forms 
 of our public schools; and. I have purposely avoided stating 
 rules or reasons under a form in which they could be learned 
 by rote. Taught in a parrot-like manner to crude minds, I 
 believe that grammar becomes bewildering and pernicious; 
 taught at a later age and in a more rational method, I believe 
 that it will be found to furnish a most valuable insight into 
 the logical and metaphysical laws -.which regulate the expres- 
 sion of human thought, and that it will always maintain its 
 ground as an important branch of knowledge, and a valuable 
 means of intellectual training. 
 
 All grammars must necessarily traverse a good deal of 
 common ground, but the careful perusal of a very few of the 
 following pages will prove, I trust, that this Syntax differs 
 in its method from all, or nearly all, that have preceded it ; 
 partly in the more free and informal manner of treatment, 
 partly in its perpetual reference to the general principles of 
 Comparative Philology, and partly in its constant endeavour 
 to leave no single idiom of Greek unillustrated by the similar 
 idioms or peculiarities of other ancient languages, of modern 
 languages, and of English. A good illustration often throws 
 over an idiom a flood of light unattainable by the most; 
 
 A3 
 
X PREFACE TO THE FIRST EDITION. 
 
 lengthy explanation ; and I feel great hopes that a student 
 who has gone carefully through the folio wiog pages, will, 
 — in addition to what he will have learnt about ancient 
 Greek, — have acquired some insight into the principles of 
 his own, and of other languages. Further than this, I shall 
 have failed in my endeavour if he do not also gain some 
 interest in observing the laws and great cyclical tendencies 
 of Language in general. The historical development of one 
 language bears a close analogy to the historical development 
 of a large majority of the rest; and this is the reason why I 
 have called such repeated attention to Modern Greek, and to 
 the traces in Hellenistic Greek of those tendencies which in 
 Modem Greek are still further developed, and carried to their 
 legitimate result. 
 
 I am not so sanguine as to hope that I have escaped errors. 
 He would be a bold man, who, even after years of study 
 should suppose that he had eliminated all the chances of error 
 in treating of a language which is so delicate, so exquisite, 
 and so perfect a medium for the expression of thought, as the 
 Greek language is felt to be by all who have studied it. For 
 myself, I may candidly confess that I have entered on the 
 task with the utmost diffidence. Some critics may doubtless 
 regard as erroneous, views which I may have deliberately 
 adopted, and which I believe that I could adequately defend ; 
 but independently of these I may doubtless have fallen into 
 positive mistakes, 
 
 1 quas ant incuria fadit, 
 Aut humana parum cavit natura.' 
 
 For the correction of any such errors I shall be grateful, and 
 I trust that they will neither be sufficiently numerous nor 
 sufficiently important to outweigh some other advantages. 
 My plan is necessarily, to a certain degree, tentative : if it meet 
 with any favour, the knowledge and the experience of others 
 may enable me in the future to introduce, from time to time, 
 considerable further improvements. I have given to it the 
 best thought and care at my command. With more leisure 
 I could doubtless have rendered it far more perfect; but I 
 
PREFACE TO THE FIRST EDITION. XI 
 
 hoped that the result might still be found commendable, how- 
 ever much I may have fallen short of even my own standard of 
 ideal perfection. The inability to reach the excellence which 
 would have been attainable under more favourable circum- 
 stances is no excuse for declining to attempt anything at all. 
 
 It is unnecessary to give a list of the large number of 
 grammars, monographs, and works of scholarship which I 
 have felt it a duty to consult in the composition of these pages. 
 I believe that I have not neglected any Greek grammar of 
 great importance; and special obligations will be found acknow- 
 ledged in their proper place. I have of course constantly 
 referred to the chief works on Comparative Grammar both 
 English and German, and to that immense repertory of Greek 
 scholarship, the Greek Grammar of Mr. Jelf. I have found 
 much that was most useful in Bernhardy, in Burnouf, in 
 Winer, in Madvig, in the Student's Greek Grammar of Dr. 
 Curtius edited by Dr. Smith, in Mr. Miller's Greek Syntax, 
 and in ' Die wichtigsten Regeln der Griechischen Syntax 1 by 
 Dr. Klein. There are however three authors to whom I am 
 under more peculiar and extensive obligations, viz., Mr. F. 
 Whalley Harper, Dr. Clyde, and Dr. Donaldson. Mr. Harper's 
 book en l The Power of the Greek Tenses ' has rendered me 
 most material assistance in treating that part of the subject. 
 The well-known works of Dr. Donaldson have been constantly 
 in my hands, even when I venture to dissent from the con- 
 clusions of that admirable scholar. The Greek Syntax of 
 Dr. Clyde, which is much less known in England than it 
 ought to be, is a most suggestive and valuable book, to which 
 I have been under constant obligations. I have often been 
 surprised by finding that it was unknown to English teachers 
 to whom I have mentioned it. If its arrangement had been 
 a little more convenient, and if it had seemed to be well- 
 adapted for school usage in our higher forms, I should not 
 have undertaken my present task. I am indebted to Dr. 
 Clyde's work for many hints and many illustrations, all or 
 most of which I believe that I have acknowledged in their 
 proper places. If in any instance (and especially in the treat- 
 ment of the Moods) I should have omitted to do so, I must 
 
Xll MEFACE TO THE FIRST EDITION. 
 
 content myself now with this more general reference to his 
 Syntax, and to the other admirable books which I have just 
 mentioned. I have gained more suggestions from the study of 
 them than it was always possible specifically to acknowledge.* 
 One pleasant task remains. I have to offer my warmest 
 thanks to the Eev. Dr. Collis, the distinguished Head Master 
 of Bromsgrove School, and to my friend and colleague 
 Ε. M. Young, Esq., Fellow of Trinity College, Cambridge, for 
 their kindness in helping me to get through the task of 
 correcting the proof sheets. Mr. Young was good enough 
 to correct for me the sheets of the earlier part of the book ; 
 Dr. Collis, though I am personally unknown to him, yet with 
 a kindness for which I hardly know how to express sufficient 
 gratitude, not only helped me to revise and correct the proofs 
 of the entire book, but constantly enriched them with many 
 acute and interesting suggestions, the result of his own ripe 
 learning and judgment. Should this Syntax succeed in rend- 
 ering the study of Greek Grammar more fruitful and more 
 interesting, some of its success will be due to the kind offices 
 of that well-known scholar. 
 
 F. W. Farrar. 
 Harrow : 
 
 March, 1867. 
 
 * I may observe that the same fact or rule is in some instances 
 intentionally repeated. 
 
CONTENTS, 
 
 INTEODUCTOEY. 
 
 PAGE 
 
 Τπε Greek Language 1 
 
 1, 2. The families of languages. 3. The Semitic family. 4, 5. 
 The Aryan family. 6. The classes of languages. 7-9. 
 Synthetic and analytic languages. 10, 11. The progress of 
 language from synthesis to analysis. 12. Eespective advan- 
 tages of synthesis and analysis. 13-15. Inflections not 
 arbitrary. 16. Eeasons for the study of Greek . . , 1-7 
 
 HINTS ON THE ACCIDENCE. 
 
 The Alphabet 8 
 
 1 . The Greek alphabet borrowed from Phoenicia. 2. The original 
 sixteen letters. 2 (bis). Epsilon, Omega, &c. 3. The di- 
 gamma, &c. 4. The Ionian letters. Archonship of Euclides. 
 5. San. 6. Koppa, yod. Origin of the alphabet. 
 
 Letters as Numerals 12 
 
 7. Numerical value of letters. 8. στοιχεία -γράμματα. 9. 
 Earliest Greek writing. 
 
 Pronunciation 13 
 
 10. Consonants. 11. Sound of vowels. Itacists and Etists. 
 
 Classification or Letters 14 
 
 12. Importance of the subject. 13. Labials, gutturals, dentals. 
 14. Pinal consonants. 15. Laws of euphony. 
 
 Vowels . . . . ' 16 
 
 16-19. Ecthlipsis, syngeresis, crasis, &c. 
 
 Dialects 18 
 
 20. The chief dialects : i. Ionic and Attic, ii. JEolic. iii. Doric, 
 iv. Hellenistic. 
 
XIV CONTENTS. 
 
 γ -agk 
 
 Parts of Speech « . . 20 
 
 21. All roots nominal or pronominal. 22. The eight parts of 
 speech. 
 Nouns . 21 
 
 23. The declensions. 
 
 Cases ib. 
 
 24. 'Casus.' 25. The five cases. 26. The nominative and voca- 
 tive. 27. The locative. 28. Origin of case-endings. 29. 
 Evanescence of case-distinctions. 
 
 Numbers 23 
 
 30. Named. 31. The dual number. 
 
 Genders ... 25 
 
 32, 33. Origin and history of genders. 34. General rules of 
 gender. 
 
 Declensions . 28 
 
 35. A declension, ormed by suffixes. 37. Heteroclites. 
 
 Adjectives 29 
 
 38. Adjectives not indispensable. 39. Their gender. 40. Ad- 
 jectival terminations. 41-43. Degrees of comparison. 44. 
 Intensive prefixes. 45. 'Ayadbs and kclkOs. 
 
 Pronouns 31 
 
 46, 47. Nature of the pronouns. 48-50. The third personal 
 pronoun. 51. Peculiarities of ου. 52-56. Eeflexive and 
 demonstrative pronouns. 57, 58. Possessive pronouns. 59. 
 Autos. 60, 61. "Οστ is, 8s. 
 Numerals ........... 35 
 
 62. Cardinals. 63. Ordinals. 64. Other numerals. 
 
 Adverbs . . 36 
 
 65-68. Nature and classes of adverbs. 
 
 Verbs 37 
 
 69-71. Nature and defin ition of the verb. 72,73. Person-end- 
 ings. 74, 75. Duals. 76. Voices. 77. Nature of the middle. 
 78. Deponent verbs. 79. Eeduplication. 80. Chief rules of 
 reduplicatiou. 81. Augment. 82. Chief kinds of augment. 
 83. Moods. 84, 85. Verbs in -μι. 86-94. Verbs in -ω. 
 Irregularities of verbs. 95-100. Classes of verbs: incep- 
 tives, desideratives, frequentatives, &c. 
 
 Compound Words 50 
 
 101, 102. Parathetic compounds. 103-105. Synthetic com- 
 pounds. 106. Inferior power of composition in Latin. 107- 
 109. Laws and irregularities of composition. 110. The 
 word ' telegram.' 
 
CONTENTS. XV 
 
 SYNTAX. 
 
 PA6E 
 
 1-3. Sentences and clauses 54 
 
 The Article .......... 55 
 
 4. Originally a demonstrative pronoun. 5, 6. Subsequent 
 traces of this. 7. It also served as a relative. 8. Its original 
 form. 9. Development of the article in other languages. 10, 
 11. It both specifies and generalises. 12. Its use with proper 
 names. 13. Anarthrous words. 14. Distinguishes the sub- 
 ject from the predicate. 15. Used instead of the possessive. 
 16-19. Its position. 20. The tertiary predicate. 21. Appa- 
 rent violations of the law. 22. Main usages of the article. 
 23. "With the infinitive. 24. Various phrases. 
 
 Concoed 64 
 
 25-27. Apparent violations of the concords. 28. Duals agree- 
 ing with plurals. 29. The Schema Pindaricum. 30. Whole 
 and part figure. 31, 32. Plural of excellence. 33. Use of 
 &ye, &c. 
 
 Cases 67 
 
 34. The case-endings were once separate words. 35. Varying 
 points of view from which the relations of objects may be 
 observed. 36. Gradual evanescence of case-meanings. 37. 
 Comparison of cases in Greek, Latin, and Sanskrit. 
 
 The Nominative . . . 68 
 
 38. The ' nominative absolute.' 39. Copulative verbs are fol- 
 lowed by a nominative. 
 
 The Vocative 69 
 
 40. The slightest of all the cases. 41. The nominative often 
 substituted for it. 42. Its origin. 
 
 The Genitive 70 
 
 43; The name due to a mistake. 44. Its three main uses. 45. 
 Genitives of ablation (causal, material, &c). 46. Genitives 
 of partition. 47. Genitives of relation. 48. Genitives of the 
 subject and of the object. 49. Double genitive. 50. Geni- 
 tive absolute. 51. Compared with the ablative absolute. 52. 
 Absolute cases in other languages. 
 
 The Dative 77 
 
 53. The dative expresses juxtaposition. 54. Hence it is used to 
 express accidents, instruments, &c. Datives of place, time, 
 manner, general reference, &c. 55. The ethic dative in 
 Greek and various other languages. 
 
XVI CONTENTS. 
 
 PAGE 
 
 The Accusative % . . 81 
 
 56. The accusative implies motion towards and extension over. 
 57. Accusatives of space, time, the cognate notion, &c. 58. 
 Double accusative. 59. Whole and part figure. 60. Accu- 
 sative after passive verbs. 61. Accusative in apposition to 
 the sentence. 62. Omission of the governing verb. 63. Anti- 
 ptosis. 64. ' Aceusativus de quo.' 65. The accusative used 
 absolutely. . 
 
 Contrasted Meanings of the Cases ..... 86 
 66. Fundamental conceptions of the cases. 67. Their absolute 
 use. 68. Contrasted instances. 
 
 Adjectives. 87 
 
 69. Illustrations of the chief idioms in the use of adjectives. 
 
 Comparatives 91 
 
 70. Illustrations of the chief idioms in the use of comparatives. 
 
 Superlatives . . .92 
 
 71. Qualitative superlative. 72. Inclusive use of superlatives. 
 73. Phrases used to strengthen superlatives. 
 
 Prepositions 94 
 
 74. The prepositions were originally local adverbs. 75. Their 
 meanings are modified by the cases with which they are used. 
 76. Due to the analysing tendency of language. 77• Spurious 
 prepositions. 78. The name ' preposition.' 79. The eighteen 
 prepositions. 80. Variation in the use of prepositions. 81. 
 Manner in which they alter the meanings of verbs. 
 
 Prepositions which govern the Genitive . . . .97 
 
 82. i. Αντί. ii. πρό. iii. e/c, e|. iv. από, ευθύ, μεταξύ, &c. 
 
 Prepositions with the Dative 98 
 
 83. i. Έν. ii. σΰν. 
 
 Prepositions w t ith the Accusative , , , , , ,99 
 
 84. i. Els. ii. as. 
 
 Prepositions with the Genitive and Accusative . , , ib. 
 
 85. i. Ata. ii. κατά. iii. υπέρ. 
 
 Prepositions with the Dative and Accusative . . .101 
 
 i. 'Ανά. Contrast of ava. and κατά. 
 Prepositions which govern Three Cases ib. 
 
 86. i. Άμφί. ii. περί. iii. επί. iv. μετά. v. παρά. vi. npos. 
 vii. υπό. 87. Passage of Philo-Judaeus. 
 
 Prepositions in Composition , .105 
 
 88. Prepositions in Composition. 
 
CONTENTS. XVli 
 
 PAGE 
 IDIOMS CONNECTED WITH PREPOSITIONS . . . . .105 
 
 89. ConstrvMio prcegnans. 90-92. Other idioms. 93. Variation 
 of prepositions in the same clause. 95. Various phrases. 
 
 Pronouns . 108 
 
 96. Personal pronouns used for emphasis. 97• Meanings of αυτός. 
 98. Possessive pronouns. 99. Use of reflexive pronouns. 
 100. Interchange of reflexive and reciprocal pronouns. 101. 
 Chief uses of demonstrative pronouns. 102. Chief uses of 
 relative pronouns. 103. Chief uses of indefinite pronouns. 
 104. Chief uses of distributive pronouns. 
 
 The Verb 115 
 
 105. The kinds of verbs. 106-110. The voices. 111. Four 
 chief uses of the middle voice. 112. Contrasted meanings of 
 the active and middle. 113. Special uses of the middle. 
 114. The middle voice in other languages. 
 
 Tenses 119 
 
 115-117. General remarks on the tenses. 118. Nine possible 
 tenses of the indicative. 119, 120. Tabulation of the tenses. 
 121. Important inferences from this scheme of the tenses. 
 122-128. Full explanation of the nature and importance of 
 the aorist. 129-131. Perpetual contrast of the aorist and 
 imperfect tenses. 132. The past-aorist sometimes used for 
 the (wanting) present-aorist. 133. Tabulation of the tenses 
 of the passive. 
 
 Chief Idiomatic Uses of the Tenses 130 
 
 134. ' Idioms.' 135. Dramatic use of the present and imperfect. 
 136. Used to express an attempt. 137. Potential use of the 
 imperfect without &v. 138. Use of the present with πάλαι; 
 illustrated from other languages. 139. Use of κλύω, &c. 
 140-143. Idiomatic uses of the imperfect. 
 
 The Future 133 
 
 144. Imperative use of the future. 145. The periphrastic future. 
 146. Four passive and middle forms. 147, 148. The future 
 perfect. 
 
 The Perfect 134 
 
 149, 150. Its use to express abiding results. 
 
 The Aorist ib. 
 
 151. The aorist as an historical tense. 152, 153. Its connection 
 in form and meaning with the future, 154. The gnomic 
 aorist. 
 
XV111 CONTENTS. 
 
 PAGB 
 
 The Pluperfect M . .136 
 
 155. Comparative neglect of the tense in Greek. Its use to 
 imply rapidity. 
 
 Moods %b, 
 
 156. Difficulty and importance of the subject. 157, 158. The 
 moods properly three in number. 159. Unsatisfactory no- 
 menclature of the moods. 
 
 The Indicative 137 
 
 160. Already treated of under 'The Tenses.' 
 
 The Impeeative 138 
 
 161-163. Slightness of tense-distinctions in the imperative. 
 164. Other modes of expressing command. 
 
 The Subjunctive and Optative 139 
 
 165. They are 'by-forms of the future and aorist.' 166. They 
 form one subjective mood. 167, 168. Consideration of their 
 tenses. 169. The tense-distinctions chiefly preserved in 
 oratio obliqua. 170. Possible origin of the aorist subjunctive. 
 171, 172. Only four tense-forms (the present and aorist sub- 
 junctive, and the present and aorist optative) in frequent 
 practical use. 173. The optative mood a refinement of lan- 
 guage. 174, 175. Its comparative unimportance and gra- 
 dual evanescence. 
 
 The Subjunctive in Simple Sentences 142 
 
 176. Used in Homer as a modified future. Its use in pro- 
 hibitions ; its deliberative, hortative, and elliptic use. 
 
 The Optative in Simple Sentences 143 
 
 177. 1. The optative not, in reality, a separate mood. 2. Its use 
 in wishes due to an ellipse. Its potential force. 3. Used 
 with ttv as a milder future, and 4. as a polite command. 5. 
 Its use to express a hopeless wish. 6. Its use to express 
 indefinite frequency. 7. The correspondence of optatives. 
 
 The Moods in Compound Sentences 146 
 
 178. The chief kinds of compound sentence. 
 
 Final Sentences . . .147 
 
 179. The infinitive and future participle not exactly final. 180. 
 "Os, Ζστπ with the future indicative after verbs of sending, 
 &c. 181-183. The moods with final conjunctions. 182. 
 Violations of the rule due to the dramatic tendency. 184. With 
 past tenses of the indicative the final conjunctions express an 
 unfulfilled result. 185. The subjunctive and optative used in 
 the same sentence to express the nearer and the more remote 
 result. 
 
CONTENTS!. XIX 
 
 PAGE 
 
 Relative Sentences . . ... . . . . .149 
 
 186-188. Use of the moods in relative sentences. 
 
 Oratio Obliqua 150 
 
 189. Exiles of the oratio obliqua. 190, 191. The optative and 
 subjunctive in oratio obliqua, and in indirect questions. 192. 
 The tenses in oratio obliqua. 193. The accusative and infini- 
 tive. 
 
 Conditional Sentences 151 
 
 194-196. Advantage of treating separately the protasis and 
 apodosis. 
 
 The Protasis 153 
 
 197. Etandeai/. 198. Four kinds of protasis. 199. Et with the 
 indicative to express possibility. 200. Έαν with the subjunc- 
 tive to express slight probability. 201. Ei with the opta- 
 tive to express complete uncertainty. 202. Et with past tenses 
 of the indicative (followed by &v with a past tense of the indi- 
 cative) to express impossibility. 203. Difficulty and vague- 
 ness of the English versions of conditional sentences. 
 
 The Apodosis 155 
 
 204. Variation of the apodosis. 
 
 Complete Conditional Sentences 156 
 
 205-208. Complete and regular conditional sentences, with 
 their English and Latin equivalents. 209, 210. Impossibility 
 of representing them accurately in idiomatic English. 211, 
 212. Influence of the dramatic tendency. 213. Instances of 
 the four classes of conditional sentences with regular and 
 varied apodoses. 
 
 Temporal Sentences 161 
 
 214. General rules of temporal sentences, with examples. 
 
 Special Uses of irpiv, ewr, &e, 162 
 
 215. i. irpiv Uv never used unless a negative conception precedes. 
 ii. irpiv with the optative in oratio obliqua, or with reference 
 to the thoughts of another, iii. Correspondence of optatives. 
 iv. Difference between irpiv, e»s, and ttpiv &v, tees &v. The 
 infinitive with ttpiv. General summary of the uses of irpiv. 
 
 The Infinitive 164 
 
 216. The infinitive not properly a mood» 217. Its connection 
 with the noun. 218. Its use in Greek and English more 
 extensive than in Latin. 219. Close analogy between the 
 use of the infinitive in Greek and English. 220. Its use to 
 express a consequence. 221. Qualified by various conjunc- 
 
XX CONTENTS. 
 
 PAGE 
 
 tions. 222. The opcxegctic infinitive. 223. Adverbial use 
 of the infinitive. 224. Used elliptically in commands, prayers, 
 &c. 225. Tenses of the infinitive. 226-228. The accusative 
 and infinitive. 229. The nominative and infinitive. 230. 
 The infinitive with other cases. 231. Future infinitive after 
 verbs of promising. 232-234. Declension of the infinitive 
 with the article in Greek and other languages. 
 
 The Participle 169 
 
 235. Affinities of the participle with the adjective. 236. The 
 Greeks φιλομέτοχυι. 237. Its two main uses. 238. It com- 
 pletes the notion of the verb. 239. Differences between the 
 infinitive and the participle after verbs of knowing, &c. 
 240. φθάσας, Καθών, avvaas. 241. The participle expresses 
 the accidents of the verbal notion. 242-245. Other uses of 
 the participle. 246. Adverbs used to define participles. 
 
 Verbals in -t4os 173 
 
 247- Verbal adjectives. 248. Used in the neuter plural. 249. 
 Verbal adjectives in -rbs and -rios. 
 
 *Av with the Moods . . . . . . . ib. 
 
 250. Meaning of fa>, Key. 251. Used with three moods. 252. 
 Used with three tenses of the indicative. 253. Potential use 
 of &v. 254. Frequentative use of &v. 255. Illustrated from 
 English usages. 256. Kev with the present and future indi- 
 cative. 257, 258. Special uses of av. 260, 261. When com- 
 bined with relatives and relative particles av takes the sub- 
 junctive. 262. Exceptions to this rule merely apparent. 
 263-267. *Av with the infinitive and participle. 267. i. The 
 verb belonging to av sometimes omitted, ii. av sometimes 
 omitted, iii. Sometimes repeated, or iv. misplaced, v. The 
 conjunction &v. vi. Elliptical use of &v. 
 
 The Pinal Conjunctions 179 
 
 268. &s, 8πα>5, 'ίνα. Eule for their use. 269. Irregularities in- 
 troduced by the dramatic tendency. 270. όπως with the 
 future indicative. 271. Its elliptical use. 272. Pinal con- 
 junctions with past tenses of the indicative. 273. I. Sum^ 
 mary of the uses of us. II. Summary of the uses of 'onus. 
 III. Summary of the uses of Ίνα. 
 
 The Negatives ..,.,...*. 182 
 274. Differences of ου and μή. 275. Distinctions between ου 
 and μ-ή. 276. Cases in which jut? is used. 277. μτ? after 
 verbs of fearing, &c. 278. Illustrations of this apparent 
 pleonasm. 
 
CONTENTS. XXI 
 
 PAGE 
 
 Ob 185 
 
 279. General uses of ob. 280. Its power of coalescing with 
 words. 281-283. Special uses of ου. 284, 285. Contrasted 
 uses of ου and μή. 286, 287. The accumulation of negatives. 
 288. Omission of negatives. 
 
 Ου μη 19L 
 
 289. Prohibitive and negative uses of oi> μή. 290, 291. Ex- 
 planation of them. 
 Μη ου 192 
 
 292. Use of μη ου after negative notions. 293, 394. Use of μη 
 with the infinitive. 
 
 Various Negative Phrases • • 194 
 
 295. Negative terms. 
 
 Particles . 135 
 
 296. Importance of the particles. 297-303. Various classes of 
 conjunctions. 304. Particles of emphasis. 
 
 Interjections . . . . . , , . . .SOI 
 
 305. Importance of the interjections. 
 
 Order of Words and Figures of Speech .... ib. 
 
 306, 307. Difference of order in synthetic and analytic lan- 
 guages. 308. Ehetorical inversions. 309. Sense-construc- 
 tions, a. Constructio praegnans. b. Zeugma, c. Syllepsis. 
 d. Comparatio compendiaria. /. Various forms of anakolu- 
 thon. g. Aposiopesis. 310. Htperbaton. a. Antiptosis. 
 δ. Chiasmus, c. Hysteron Proteron. d. Hypallage. 311. 
 Euphemism, a. Irony, b. Hypokorisma. c. Litotes, d. Anti- 
 phrasis. e. Ambiguity. 312. Pleonasm, a. Periphrasis, b. 
 Polyptoton. 313. Hendiadys. 314. Asyndeton. 315. 
 Paronomasia, a. Onomatopoeia, b. Alliteration, c. Oxy- 
 moron, d. Antithesis, e. Rhyme. /. Ehythm. 
 
A BRIEF 
 
 GREEK SYNTAX. 
 
 INTRODUCTORY. 
 THE GREEK LANGUAGE. 
 
 1. The Greek Language belongs to the Aryan or Indo- 
 European family of languages. 
 
 2. There are two great recognised Families of Language, 
 the Aryan and the Semitic. These languages are spoken by 
 the most advanced and civilised of human races. The other 
 languages of the world, which may be classed together under 
 the names Sporadic or Allophylian, have not vet been reduced 
 to any unity, but fall under a number of different divisions. 
 
 3. The Semitic languages are Hebrew, Phoenician, Cartha- 
 ginian, Aramaic (i.e. Syriac and Chaldee), and Arabic. The 
 name 'Semitic 1 is purely conventional, and they might con- 
 veniently be called, from their geographical limits, Syro- 
 Arabian. 
 
 4. The Aryan languages consist of eight main divisions, 
 which we may call the Sanskritic, Iranic, Hellenic, Italic, 
 Lithuanian, Sclavonic, Teutonic, and Celtic. The name Aryan 
 is derived from the title Arya, ' noble,' which was arrogated 
 to themselves by the first founders of the race. 
 
 5. The Aryan family of languages is the most perfect family 
 in the world, and Greek is the most perfect language in this 
 family ; it is ' the instinctive metaphysics of the most intelli- 
 gent of nations.' 
 
 6. Again, there are four different Classes of Languages, divided 
 according to their structure. 
 
 These morphological or structural divisions are : 
 
 i. Isolating languages, which have no proper grammar, and in which 
 the words suffer no change to express any shades of thought or varieties 
 
 Β 
 
2 A BEIEF GREEK SYNTAX. 
 
 of circumstance ; of these Chinese is the chief. Thus in Chinese the 
 prayer ' Our Father which art in heaven,' assumes the form ' Being 
 heaven me-another ( = our) Father who ; ' a style not unlike the natural 
 language of very young children. Isolating languages are perhaps the 
 oldest of all, and yet by that curious cyclical process which is observable 
 in language, many modern languages in the last stage of their history 
 resemble them. For instance, Chinese has never possessed cases or 
 inflections of any kind, and English has lost nearly all which it once 
 possessed ; or, as Dr. Latham expresses it, Chinese is aptotic, English 
 anaptotic. 
 
 ii. Agglutinating, like the Turkish, in which the material elements of 
 words (root or stem), and the formal elements (pronouns, indicating 
 space, position, &c), are juxtaposed in one word without undergoing any 
 modification. In these languages all compound words are separable, i.e. 
 the component parts are not fused together and altered in the process, 
 but are merely parathetic or joined mechanically, as in the English words 
 star-fish, railroad, clock-work, &c. 
 
 iii. Polysynthetic (also called holophrastic or incorporant), in which, 
 as in Basque, and in the aboriginal languages of America, each sentence 
 is one long compound word, and is an agglomeration of simple words 
 ' in a violent state of fusion and apocope,' e.g. in one of these languages 
 nicalchihua means • I build my house,' but neither ni ' I,' col * house,' or 
 chihua ' make,' can be employed as separate words.* 
 
 iv. Inflectional languages, in which, as in Greek and Latin, the mate- 
 rial elements (roots), and the formal elements (pronouns, &c, expressive 
 of various modifications), are united by synthesis into one inseparable 
 whole, and in which the inflections have so entirely lost their force as 
 separate words that their very origin is often undecipherable. 
 
 7. Greek presents the most perfect specimen of an inflectional 
 or synthetic language. 
 
 8. A language which gets rid of inflections as far as possible, 
 and substitutes separate words for each part of the conception, 
 is called an analytic language ; and next to Chinese (which 
 has never attained to synthesis at all) few languages are more 
 analytic than English. Thus in nouns we have only retained 
 one case-inflection, viz. the s which is a sign of the genitive ; 
 and in verbs only one inflection to express tense, the -d in past- 
 aorists, as I loved (= I love-did). Yet English continues 
 to be a thoroughly synthetic language, and it contains hun- 
 dreds of single words which in any isolating language would 
 require four or five separate words for their expression. 
 
 9. A synthetic language will express in one word what 
 requires many words for its expression in an analytic language, 
 as will be seen by an instance or two : e.g. 
 
 * Strange as this holophrasis may appear to us, there are distinct 
 traces of it both in Greek and Latin; see Origin of Language, p. 174. 
 
SYNTHESIS ΑΝί) ANALYSIS. 3 
 
 φιΚηθήσομαι, amabor, I shall be loved, Ich werde geliebt 
 
 werden. 
 πεφιλήσομαι, I shall have been loved, Ick werde geliebt 
 
 worden sein. 
 ετετιμημεθα, honorati eramus, we had been honoured. 
 λύσωμαι, que je rue sois delie. 
 λελνσοίμηΐ', may I have been unloosed ! que j'eusse du 
 
 etre delie ! 
 ωχετ-ο, abierat, il s'en etait alle. 
 
 Similarly the synthetic character of the Semitic languages 
 enables them to express by an affix or a suffix some modifica- 
 tion of meaning, which in modern languages would necessitate 
 one or more separate words for its enunciation ; e.g. to render 
 the one word ^ΓΟΙΠΠΊ. vehirJcabhteeka* we require at least 
 seven words, ' and I will cause thee to ride ; ' and yet in spite 
 of this the one Hebrew word expresses more than our seven, for 
 it implies that the person addressed is a male, so that in feet 
 to give the full meaning of that one word we should require 
 the nine words, i And I will cause thee, Ο man, to ride.' No 
 instance could illustrate more forcibly than this the difference 
 between Synthesis and Analysis in language. 
 
 10. The tendency of all languages, at least in historic times, 
 is from synthesis to analysis, e.g. from case-inflections to the 
 use of prepositions, and from tense-inflections to the use of 
 auxiliaries. This tendency may be seen by comparing any 
 modern language with its ancestor, e.g. Arabic with Hebrew, 
 Bengali with Sanskrit, Persian with Zend, Danish with Ice- 
 landic, German with Gothic, or English with Anglo-Saxon. 
 
 11. It may also be constantly illustrated by a comparison 
 of Modern with ancient Greek, for which reason Modern Greek 
 is often referred to in the following pages. But the simplest 
 way of studying the tendency is to compare Latin with any 
 of those six Romance languages (Italian, French, Spanish, 
 
 * Ancient Hebrew, says Herder, ' seeks like a child to say all at once.' 
 This reminds ns of the remark in Le Bourgeois Grentilhomme. Mons. 
 Jourdain : ' Tant de ckoses en deux mots ? Cov. Oui, la langue turque est 
 eomme cela, elle dit beaucoup en peu de paroles.' Gothe remarks of 
 French, ' Ο eine Nation ist zu beneiden, die so feine Schattirungen in 
 oinem Worte auszudriicken weiss ' (Wilhelm Meister) ; but the remark 
 is true in a far higher degree of Greek than of any other language ; e.g. 
 to represent fully in French the word avrinape^ayeiv, we should require 
 •'faire sortir une armee en face de l'ennemi, et la mener contre lui' — 
 thirteen words for one. See Burnouf, Metkode pour etudier la langue 
 qrecque, p. 165. 
 
 β 2 
 
4 A BRIEF GREEK SYNTAX. 
 
 Portuguese, Wallachian, and Engadinish) which have been 
 immediately derived from it ; e.g. amabo becomes in French 
 fatmerai, which is a corruption of the analytic expression Ego 
 amare habeo I have to love.* 
 
 12. The advantage of a synthetic language lies in its com- 
 pactness, precision, and beauty of form ; analytic languages 
 are clumsier, but they possibly admit of greater accuracy of 
 expression, and are less liable to misconception. What they 
 lose in euphony, force, and poetic concision, they gain in the 
 power of marking the nicest shades of thought. What they 
 lose in elasticity they gain in strength. If they are inferior 
 instruments for the imagination, they better serve the purposes 
 of reason. Splendid efflorescence is followed by ripe fruit. In 
 the tragedies of iEschylus and the odes of Pindar, marvellous 
 as is the power which crams every rigid phrase with the 
 fire of a hidden meaning, we yet feel that the form is cracking 
 under the spirit, or at least that there is a tension injurious to 
 the grace and beauty of the general effect. A language which 
 gets rid of its earlier inflections, — English for instance as 
 compared with Anglo-Saxon, — loses far less than might have 
 been supposed. 
 
 13. It is most important to observe that no inflection is 
 arbitrary ; it is now certain that every inflection is the frag- 
 ment of a once separable word, having its own distinct mean- 
 ing. Among all the richly-multitudinous forms assumed by 
 the Greek and Latin verbs, there is not one which does not 
 follow some definite and ascertainable law. The actual analysis 
 of the inflections has been carried to considerable perfection ; 
 but the derivations of many of them are as yet to a certain 
 extent disputable and uncertain. The wise warning of 
 Quinctilian is still required, 'Inter virtutes grammaticas 
 habebitur aliqua nescire.' 
 
 14. Parsing, — the hopeless stumbling-block of so many 
 young students, — loses its difficulty and repulsiveness, when it 
 is once understood that there is a definite recurrence of the 
 same forms in the same meaning, and that the distorted 
 shape assumed by some words is not due to arbitrary license 
 but to regular and well understood laws of phonetic corruption. 
 
 15. i. For instance, the wordf βουλενσαντο means 'they took 
 counsel for themselves ;' we express the same conception by 
 five words, and should require seven, but that we do possess 
 
 * For further remarks on Ibis subject see Origin of Language, 
 pp. 173-181. 
 
SYNTHESIS AND ANALYSIS. 5 
 
 an aorist* ('took') in English verbs, and also an inflection 
 1 s ' to express the plural ; but if we analyse the word 
 εβουλενσαντο we shall have to write it 
 
 ε-βυυλεν-σ -a-yT-o, 
 and shall find that it consists of six f parts, viz. : 
 
 1 . An augment ε (the fragment probably of the same root 
 which we find in the preposition ανά, expressing indefi- 
 nite ])ast time). 
 
 2. A root or stem, βουλεν. 
 
 3. A tense-letter, σ, here characteristic of the first aorist, 
 and derived from the root as to be. 
 
 4. A vowel, a, used as a tach between the tense-letter 
 and the person-inflection. 
 
 5. The relic of a pronoun, vr, characteristic of the third 
 person plural. Perhaps we ought to call this the 
 relics of two pronominal roots, ana, and the demon- 
 strative -ta [lie and he = they~].\ This termination 
 was slurred in pronunciation, as we see from the Latin 
 forms fuere, amavere, &c. 
 
 6. A voice letter, o, indicating the passive or middle. 
 
 ii. Similarly, ε-τε-τίμ-η-ντ-ο consists of six parts, the re- 
 duplication being used to mark the perfect, and the augment 
 to place this perfect event still farther back in the past. 
 
 iii. So too in Latin, such a word as amabantur is analysed 
 thus : ama-ba-nt-u-r = root + sign of the imperfect + sign of 
 the 3rd pers. plur. -f junction-vowel -f pronominal elements. 
 In this instance we know that ' ba ' is a fragment of the root 
 which we find in the auxiliary verb <bv, fu, &c, and the original 
 form may have been am-a-ba-nt-u-se. 
 
 iv. Again, take such a form as λυθήσομαι, ' I shall be 
 loosed ;' this, when analysed, is λυ-θ-η-συ-μαι, and consists, 
 no less than the English phrase, of five parts, viz. : 
 
 1. The root λυ-. 
 
 2. 61- the relic of the root dha, to do or make : this mean- 
 ing is preserved even in the Greek τίθημι,&δ τί κε θεΊμεν; 
 Sapph./r. 62. 
 
 * "When this aorist is formed qualitatively, i.e. by mere internal modi- 
 fication of the root as in take, took, (which is the ordinary Semitic 
 method,) it is called a strong aorist; when it is formed by the addition 
 of some extraneous word as love, love-did (= loved), it is called a weak 
 aorist. 
 
 f See Dwight's Modem Philology, ii. p. 274. 
 
 \ See A. Schleicher, Vergleichende Grammatik, § 276. 
 
6 A BRIEF GREEK SYNTAX. 
 
 " 3. η- the representative of the roofc^a =ire (ajui), to go. 
 
 4. σο- the future sign, which we find in ερο-μαι, eso (ero). 
 
 5. μαι, the first personal pronoun (in oblique case). 
 
 The whole conception therefore is synthetically built up of 
 the elements There will be (σο) a going (η) to make (Θ) me 
 (μαι) loose (\v)* Thus the two auxiliary verbs 'to go' ana 
 1 to be,' however much disguised, occur in every Greek and 
 Latin future. 
 
 15 (bis), i. Sometimes the original constituent elements are 
 greatly obliterated. 
 
 Take, for instance, the pluperfect επεπηγειν, or, to use the 
 more Attic form, Ιπεπήγη. This is resolvable into ε-πε-πηγ-εα, 
 i.e. augment + redupl. + root + auxiliary. This εα is really εσ- 
 αμ (cf. eram), which is the root εσ, a junction vowel a, and 
 the first personal pronoun. 
 
 ii. The traces of a previous form of the word are sometimes 
 unexpectedly preserved in the accentuation. Thus ελυον, in 
 the 1st person plural, is proparoxytone ; but in Doric the 
 3rd pers. plur. is accented ελνον. The reason of this is that 
 the 1st person was originally ελνομ (cf. inquawi, sum, and the 
 provincial Ich bim=Ich bin) ; but the 3rd pers. plural has 
 been softened from an original ελυοντ. 
 
 iii. It will be seen that this analysis of Greek inflections de- 
 pends entirely on the distinction between the material and 
 formal elements of words, i.e. between the stem or inflective 
 base (which the Hindoo grammarians call the atiga or body) 
 of a word, and the various affixes or suffixes, which indicate 
 its special meaning and relations. This distinction was un- 
 known or disregarded until the discovery of Sanskrit led to the 
 study of Indian works on grammar ; but it is a distinction of 
 extreme importance, and one which reduces grammatical con- 
 ceptions to an extreme simplicity. 
 
 The root of a word must be carefully distinguished from its 
 stem. 
 
 A root is the ultimate constituent sound of a word reduced 
 to its simplest form. It is in fact the core, or vocal skeleton of 
 a group of kindred words. In some languages, as in Chinese, 
 all words are also roots, and their mutual relations are only 
 indicated by position. 
 
 ' The Indian grammarians called a root dhatu, from dhd, to 
 nourish : dhatu means any primary or elementary substance, 
 and consequently shows that these grammarians looked on 
 
 * See A. Schleicher, Vergleichende Grammatik, § 300, 
 
SYNTHESIS AND ANALYSIS, 7 
 
 roots as the primary elements of words.'—Ferrar, Comp. Gram. 
 p. 178. 
 
 AH. roots are either verbal (i.e. predicative) or pronominal 
 (i.e. demonstrative). 
 
 The stem of a word is what remains of the word when its 
 inflections have been removed. It may be identical with the 
 root : e.g. 07r-, στιχ-, due-, are both the stems and the roots of 
 όμμα, στίζ, and dux. But more often the stem is the root 
 already modified and followed by various suffixes, as in στοϊγο-ς, 
 όπτίκύ-ς, ductili-S. Thus of πράγμα the root is upay-, but 
 the stem is πραγματ. The stem, says Bopp, may be con- 
 sidered as a sort of general case, never employed in an isolated 
 form, but which in a compound word takes the place of all 
 cases : e.g. τελες-φόρος, λογο-γράφος. Some stems are conson- 
 ant, some vocalic. 
 
 The inflections, or formal elements of a word, are those little 
 syllables — the relics of pronouns and auxiliary verbs — which 
 express the mutual relations of ideas, the various conditions of 
 time, space, and circumstance. Elastic in their form and fluid 
 in their meaning, they lend themselves to the expression of all 
 modifications in the sense, and add in a marvellous degree 
 to the clearness, wealth, and freedom of language (see Breal, 
 Bopp, Gram. Comp. II. xxviii.). 
 
 16. The reasons why we spend so long a time in acquiring 
 a mastery over the Greek language are manifold. We do so 
 partly because it is one of the most delicate and perfect in- 
 struments for the expression of thought which was ever 
 elaborated by the mind of man, and because it is therefore 
 admirably adapted, both by its points of resemblance to our 
 own and other modem languages, and by its points of dif- 
 ference from them, to give us the Idea or fundamental con- 
 ception of all Grammar; i.e. of those laws which regulate the 
 use of the forms by which Ave express our thoughts. Again, 
 Greek is the key to one of the most astonishing and splendid 
 regions of literature which are open for the intellect to explore, 
 — a literature which enshrines works not only of imperishable 
 interest, but also of imperishable importance (both directly 
 and historically) for the development of human thought. It 
 is the language in which the New Testament was first written, 
 and into which the Old Testament was first translated. It 
 was the language spoken by the greatest poets, the greatest 
 orators, the greatest historians, the profoundest philosophers, 
 that the world has ever seen. It was. the language of the 
 most ancient, the most eloquent, and in some respects the 
 
8 A BRIEF GREEK SYNTAX. 
 
 most important of the Christian fathers. It contains the 
 record of institutions and conceptions which, lie at the base 
 of modern civilisation, and at the same time it contains the 
 record, and presents the spectacle, of precisely those virtues 
 in which modern civilisation is most deficient. Nor is ir an 
 end only ; it is also a means. Even for those who never suc- 
 ceed in reaping all the advantages which it places within 
 their reach, it has been found to be in various nations and 
 ages* during many hundred years, one of the very best 
 instruments for the exercise and training of the mind. It 
 may have been studied irrationally, pedantically, and too 
 exclusively ; but though it is desirable that much should 
 be superadded, yet with Latin it will probably ever continue 
 to be, — what the great German poet Gothe breathed a wish 
 that it always should be, — the basis of all higher culture. 
 4 Greek,f the shrine of the genius of the old world, as universal 
 as our race, as individual as ourselves ; of infinite flexibility, 
 of indefatigable strength, with the complication and the dis- 
 tinctness of nature herself, to which nothing was vulgar, from 
 •which nothing was excluded ; speaking to the ear like Italian, 
 speaking to the mind like English ; with words like pictures, 
 with words like the gossamer film of the summer ; at once the 
 variety and the picturesqueness of Homer, the gloom and the 
 intensity of iEschylus ; not compressed to the closest • by 
 Thucydides, not fathomed to the bottom by Plato, not sound- 
 ing with all its thunders, nor lit up with all its ardours, even 
 under the Promethean touch of Demosthenes himself.' 
 
 THE ALPHABET. 
 
 1. The Greeks borrowed their alphabet from the Phoenicians. 
 It originally consisted of sixteen letters, w T hich were said to 
 have been introduced by Cadmus. Hence Ausonius calls 
 letters, ' Cadmi nigellas filias.' J The name Cadmus is probably 
 a mere mythical personification of the Hebrew word D"^ 
 Kedem ' the East.' § 
 
 * For the study of Greek formed one of the main branches in the 
 education of the young Eomans. 
 
 f Η. N. Coleridge, Introduction to the Greek Classic Poets. 
 
 \ Auson. Ep. iv. 7. It is sometimes stated that, according to Hesy- 
 chius, 4κφοίΐ'ίξαι may mean ' to read ' with a reference to Phoenician letters. 
 This is not the case. His gloss is 4κ<ροινί£αι, uvayvwffai, for which 
 Abresch doubtfully suggested avayvwvai ; but probably the word should 
 be αίματωσαι. 
 
 § This word also means ' the ancient.' See Ps. xliv. 2, &c. 
 
THE ALPHABET. Ό 
 
 2. These original sixteen letters, called τα Φοινίκη ϊα (Herod. 
 v. 58, 59), or τα άπο Κάδμου, or τα Πελασγικά, were probably 
 as follows : — 
 
 ΑΒΓΔ 
 Ε F ΗΘ 
 Ο Π 9 Τ 
 
 and the liquids Λ Μ Ν Σ. 
 
 In this list F is digamma ; 9 * s koppa ; Η is the sign of 
 the aspirate. 
 
 The arrangement of this alphabet is evidently systematic, 
 viz., α followed by three media?, ε followed by three aspirates, 
 ο followed by three termes ; and the four liquids (see Donald- 
 son, Neiv Cratylus, ch. v.). 
 
 The other letters of the Semitic alphabet were gradually 
 borrowed. The Semitic alphabets, however, differ from the 
 Aryan : i. in having no vowels ; ii. in being arranged in no 
 phonetic order. # 
 
 2 (bis). The names Epsilon, Upsilon, Omega, Omicron were 
 wholly unknown to the ancients, and were not introduced till 
 the vowel-sounds were confused. It is now known that -ψιλόν 
 is opposed not to δασύ (as smooth to aspirate) but to δίφθογγον. 
 Plat. Orat. 393 d.; Athen. x. 453 f. Υ and Ώ should bear the 
 same name as they do in English, unless ' Omega ' be retained 
 for its association. Ε was called el, Ο was called ου. Hence 
 Ο was a positive refusal. When Dionysius the Tyrant invited 
 Philoxenus to Syracuse, his only answer was a page of circles, 
 one within the other, (ο), εμφαίνων on πολλάκις και σφόδρα 
 αρνείται. Hence το Φιλοίένου ου became the proverb for any 
 emphatic negative. The Laced aemonians gave a similar answer 
 to Philip of Macedon. Plut. De Garrulit. c. 21; Auson. xxiv. 
 36, 37. 
 
 3. The digamma, or vau, F (βαϋ), and koppa, 9 (κόππα), 
 represent the Hebrew 1 vau, and p kooph. Although found 
 in some old inscriptions, they early fell out of use in Greek ; 
 but are retained in Latin under the forms of F and Q. The 
 digamma was replaced by υ and <f> ;* 9 by κ and χ. Η, which 
 
 * The digamma F was evidently in use when the Homeric poems were 
 composed ; but it had ceased to be employed as a written character when 
 they were first preserved in manuscripts ; hence such apparent hiatuses 
 as όσσα ϊοικς at the end of an hexameter line. The first grammarian 
 who called attention to it was the celebrated Apollonius Dyscolus in the 
 time of Hadrian. In many Greek words very early took its place, as 
 we see by finding Fa$os foT"Oa£os on old coins, and by a comparison of 
 
 β 3 
 
10 A BRIEF GKEEK SYNTAX. 
 
 was originally an aspirate, and continues to be so in the 
 Latin H, was adopted as a sign of the double ε. Palamedes 
 is the legendary inventor of ν, φ, and ψ*; Simonides and 
 Epicharmus are variously asserted to have added the two other 
 double letters £ and £, and the long vowels η and ω (Eurip. 
 Fr. Palam.-, Plin. Ν. H. vii. 26). 
 
 The entire Greek alphabet of twenty-four letters, as it now 
 stands, is said to have been first used by the Ionians of Asia 
 Minor, and hence is called τα 'Ιωνικά γράμματα. It was early 
 adopted by the Samians ; and it is very probable that 
 Herodotus, who often resided at Athens, and was a warm 
 friend of the poet Sophocles, first introduced it among the 
 educated Athenians. Hence (even before the archonship of 
 Euclides) when Euripides introduces a peasant who cannot 
 read, describing the written characters of the word θησενς, he 
 distinguishes between η and ε* The passage, which is a very 
 interesting one, is preserved by Athenasus (Deipn. x. 79, 80) 
 in his curious chapters on the Greek alphabet. 
 
 4. The Ionian letters were npt, however, formally adopted 
 by the Athenians, or used in public monuments, until the 
 archonship of Euclides, B.C. 403. Hence they are called -a 
 γράμματα τα άπ ΈνκΧείΰον άρχοντος. The alphabet of 
 
 οΊδα, oJkos, olvos with the Latin video, vicus, vinum ; in others v, as we 
 see by comparing 0c«nAefs (still pronounced vasilofs in Modern Greek) 
 with βασιλεύς, and by the absence of contraction in πλ4ω, ρ4ω, χ4ω, 
 which are the ultimate forms of TrXeFw, πλεύω (cf. aor. Ζπλευσα), &c. 
 The digamma was called iEolic, because it was retained latest in that 
 dialect ; and the traces of it abound in Latin, which resembles JEolic 
 more than any other form of Greek. It is represented in Latin by va- 
 rious letters, as b, p, f, and especially v. Thus irpaFos becomes pro^us, 
 SaFis daps, ΐορμίαι Formise, ώόν, cap, eWepos, %ov, ovum, ver, vesper, viola, 
 &c. It may however be considered probable that the r had a complex 
 sound, viz. the sound of a guttural combined with a labial, a fact which 
 is etyniologically of the utmost importance, since it accounts for many 
 otherwise impossible letter-changes in Greek words. See Garnett, Phi- 
 lolog. Essays, p. 241 seqq. The F is fully handled in Ferrar's Compara- 
 tive Grammar, pp. 87-90. He says it had nearly the sound of w, 
 quoting Dionysius of Halicarnassus, who defines it as ου συλλαβή h\ 
 στοιχύω γραφομενη. 
 
 * He describes the Η thus: 
 
 ■πρώτα μ\ν γραμμαϊ δυο 
 
 ταύτας Steipyet δ' iv μ4σαΐ5 άλλη μία. 
 and Ε thus; 
 
 ■f\v μ\ν tls ορθόν μία 
 
 λοξαϊ δ' έπ αυτής T/>e7s κατ€στηριγμ4ναι. 
 
 Similarly, Agathon in his Telephus. 
 
THE ALPHABET. 11 
 
 twenty-one letters (i.e. all except £, ψ, ω, the three which 
 were last adopted), is called τα 'Αττικά. 
 
 5. Besides the obsolete F and 9, the Greeks at one time had 
 a letter Σάν, the representative of the Hebrew Zain ; it was 
 ousted by ;, which properly was the representative of the 
 Hebrew Shin. Both Σαν and Κόππα were retained as marks 
 of the breed of horses ; a horse branded Σα ν was called 
 Σαμφόρας, 
 
 ουκ έλας ώ Σαμψύρα ; Arist. Eq. 603 ; cf. Νιώ. 122 ; 
 
 and was guaranteed as being of a particular breed. A horse 
 branded with Κό^ττα* was called Κοππατίας, and was sup- 
 posed to be of the Corinthian breed descended from the fabled 
 Pegasus. Hitzig, however, thinks that these two letters were 
 used in branding horses to represent the first and last letters 
 of BHp Kodesli l holy,' i.e. precious. 
 
 5 (his), i. Koppa (kooph = Q) was obviously valueless, as Κ 
 could easily supply its place. In Latin, where Κ was not an 
 indigenous letter, an irate grammarian called Q ' littera 
 mendica, supposititia, vere servilis, manca, et decrepita ; sine u 
 tanquam bacillo nihil potest, et cum u nihil valet amplius 
 quam &.' 
 
 ii. The letter yod, though obsolete in Greek, leaves repeated 
 traces of its presence. Thus άμείνων, κτείνω, στέλλω, κορΰσσω 
 are assimilations for αμενγων, τενγω, στελγω } κορυτγω; μάλλον 
 is for μαλγον, μέλαινα for ^ελακνα, τέρεινα for τερενγα. We 
 can often detect the original existence of this yod by referring 
 to the Latin ; e.g. famo is the Latin equivalent of φράσσω. 
 
 6. The discovery of the Alphabet, and its representation by signs, 
 must always rank among the very highest discoveries of human in- 
 genuity ; probably, however, the discovery was very gradual. 
 
 Writiug seems to have passed through three stages ; viz.: 
 
 1. The pictorial stage, in which, as in hieroglyphics, and the Mexican 
 picture writing, each object was represented by its picture, and abstract, 
 immaterial things by some picture which metaphorically indicated them. 
 
 2. These pictures were taken to stand not for the object itself, but for 
 the syllable which named the object ; e.g. a picture of the sun stood no 
 longer for the sun itself, but for the word, sound, or syllable which 
 meant sun (this in Egyptian is Ba, so that a picture of the sun would 
 stand in any word in which the syllable ra occurred). 
 
 3. The picture was taken for the letter with which the syllable it 
 represented commenced (so that in Egyptian a picture of the sun would 
 
 * We still find qoptvBos in inscriptions, &c, for Κόρινθος, and it is found 
 in the inscription on a helmet brought by Col. Leake from Olympia, 
 9otos μαποζστν = Kotos fi ςποιή]σςν, 
 
12 A BRIEF GREEK SYNTAX. 
 
 stand for r). We can still trace the pictorial origin of the Hebrew 
 alphabet, from which the Greek is derived. Thus aleph (alpha) means 
 ox, and is represented by ^, originally V• 
 
 Beth (beta) means house, and is represented by 3, onginally Λ, a tent, 
 and so on. To this day we can trace back our sign for the letter m to 
 the wavy line which was the conventional representation of water. See 
 Chapters on Language, p. 139. 
 
 LETTERS AS NUMERALS. 
 
 7. The letters of the alphabet from α to ω are used in 
 regular order to number the twenty-four books of Homer; 
 but, besides this, they had the following numerical values, 
 which should be remembered, because they not unfrequently 
 occur in Greek books. When used as numerals, the letters 
 are distinguished by a dash, as α', β', &c. 
 
 a' to f' stand respectively for 1, 2, 3, 4, 5. Then to make 
 up for the lost digamma the sign =r', called stau or stigma, was 
 used for 6. ζ' to i r stand respectively for 7, 8, 9, 10. Then 
 ia, φ', &c. for 11, 12, &c. *> is 20, κα' 21, φ' 22, &c. Then 
 λ' = 30, μ'^40, *'=50, £'=60, o'=70, π' = 80 ; but the next 
 letter p'=100. From this fact we see at once (as in the cor- 
 responding numerical gap for the lost digamma between 5 and 
 7 ) that a letter has been lost : this is the letter koppa 9, which 
 is accordingly retained as the sign of 90. 
 
 The remaining letters from σ' to ω' are used for the hun- 
 dreds from 200 to 800. For the number 900 the Greeks use 
 the obsolete sanpi ^ or sp, the reverse of φ or ps. 
 
 For the thousands the dash is placed beneath the letter to 
 the left; thus ,α = 1000, /( β = 2000, ,γ=3000, &c. 
 
 Thus 1865 would be expressed in Greek by ,αωξε'; and 
 10,976 by ,,JW. 
 
 8. The word Alphabet, which is comparatively late, is 
 derived from the first two letters α, β* The letters considered 
 as elementary sounds are called στοιχεία ; considered as 
 written signs γράμματα. 
 
 9. The earliest known piece of Greek writing (not later 
 than B.C. 600) is on a prize vase brought from Athens by Mr. 
 Burgon. It runs from right to left,"f and is — 
 
 IMS 14ΠΛΘΛ •ΙΛΘ3Μ3ΘΑ HOT 
 
 * The Latin elementa (perhaps — ol-e-mentu-m, from ol-ere) has been 
 by some derived from the three liquids, I, m, η ; and there is something 
 to be said for. this derivation, strange as it may appear. See Hitzig, 
 Die Erfindungdes Alphabetes, S. 13, 14. 
 
 t The modes of writing varied ; some inscriptions are found in which 
 
PRONUNCIATION. 13 
 
 or των Άθηνηθεν άθλων ειμί, ι Ι am one of the prizes from 
 Athens.' Here we see ο for ω, and ε for η. The shape of the 
 Λ is, however, more modern than the shape V which is re- 
 tained in the Latin L. 
 
 PRONUNCIATION. 
 
 10. The Greek consonants w ere probably pronounced much 
 as we pronounce them now, except that <f>, which we pro- 
 nounce as f* (compare φώρ far, φηγός ,/agus, ψράτωρ prater, 
 φάναι /ari, &c), was probably more often pronounced like 
 ph in haphazard. We know that the Macedonians pronounced 
 it like p, and talked of Πίλιττ-ος. But although graphically φ 
 was represented in Latin by ph, yet in all the words of the 
 original Aryan stock the Greek φ appears in Latin as / (e.g. 
 <ρέρω fero, φήμη fama, &c). That there was, however, a dis- 
 tinction between the two in sound appears from . Cicero's 
 ridicule of the Greek witness who could not pronounce Funda- 
 nius (Quinct. Just. Or. i. 4. 14). See Ferrar's Comp. Gram. 
 p. 108. 
 
 Zeta was probably pronounced like the s in maison. It was 
 a weak sibilant, which often has its origin in the obsolete yod. 
 Cf. Ζευς with Dyaus, and ζα with ha. 
 
 11. The school of Erasmus used to dispute with that of 
 Reuchlin whether the η should be pronounced like our i, as 
 in Modern Greek, or like our e. This is what is meant by the 
 quarrel between Itacists and Etacists, of which we hear so 
 much at the revival of letters. Neither were exactly right, 
 for η must have had the sound of (la, since it was used to 
 represent a sheep's bleat, as in the line of Cratinas : 
 
 ό ο' ι)\ίβως ώσττερ πρόβατον βή βή λέγων βαδίζει, 
 1 but the booby goes saying baa baa like a sheep.' 
 
 t was clearly pronounced as in French, for κοί, κοί, is a 
 pig's squeak, Arist. Acliam. 780 ; and ποί, the peewit's cry, 
 Av. 227. 
 
 the words are written from the top to the bottom, which is called 
 κιονηδόν ; others are -written first from right to left, and then from left 
 to right, as the ox turns in the furrow ; this style is called βουστροφτ\Β6ν. 
 (Pausan. Eliac. i. p. 338.) The amoves and κνρβειε of Solon are said to 
 have been written βυυπτροφηδόν (Hesych. s. v.), as is the famous Sigean 
 inscription. Originally none but capital letters were used, which is 
 called the Uncial style ; the ordinary cursive Greek letters are not 
 found in MSS - till the eighth or ninth century. 
 
 * Ph is the more frequent Latin equivalent of ψ, as in pkilosophia, &c. 
 
14 A BRIEF GREEK SYNTAX. 
 
 αύ must have been pronounced ' ow,' since bow-wow," a 
 dog's bark, is in Greek αν av (Aristoph. Vesp. 903) ; and 
 to bark is βανβάζειν, baubari. 
 
 ου must have been pronounced oo, as we see in the onomato- 
 poctic * word βυϋς (compare our childish rnoo-cow) ; and the 
 exclamation Ιοΰ for ugh ! 
 
 CLASSIFICATION OF LETTERS. 
 
 12. i. It is of the utmost importance to know and to re- 
 member the divisions of the letters ; a division which lies at 
 the root of all etymology. For, as a general rule, it is only 
 letters pronounced by the same organ that are etymologically 
 interchangeable (dentals with dentals, labials with labials, 
 &c). Whenever it appears to be otherwise, "j" we may gene- 
 rally assume that both letters existed in the original form. 
 Thus bis does not come from δις, but the b represents the w 
 in the Sanskrit dims ; nor is βα να derived from γυνή but 
 from the F in yFava. Similarly μέλας and κελαινος are the 
 same word, but the original form of the word was κμελας, 
 and the labial μ has not been interchanged with the guttural 
 κ. Similarly συν and cum are the same word, but the fact is 
 accounted for by the form ζνν=κσυν (cf. ναπ-νος and vap-ox 
 with the Lithuanian kvap-a-s). 
 
 Donaldson, who claims to have discovered this principle 
 (art. Philology, Enc. Brit. p. 539), calls it ' the law of diver- 
 gent articulations.' Older grammarians called it Metalepsis ; 
 e.g. Sanskr. paJctas -^πεπτός = coctus ; but ρ cannot pass into k, 
 so that Sanskrit differs from Greek in Inlaut, and from Latin in 
 Anlaut. But even in Quinctilian's time coquus was pro- 
 nounced quoquvs (Milt. Or. vi. 3. 47) ; and here we see the 
 origin of the divergent forms of the word, since qv = kp. 
 Similarly, by comparing vivus and ' quick ' (' quick and dead '), 
 we are led to an original form qviqvas. Cf. Gothic quivs. 
 See on this whole subject Curtius, Grundzuge d. Griech. Etym. 
 n. 36. 2a ; Corssen, Lat. Formenlehre, p. 28. 
 
 ii. The vowels (φωνήεντα) are α, ε, t, ν, ω. 
 
 iii. The consonants are divided into : i. semi-vowels (ημί- 
 
 * An onomatopoeia is a word formed in imitation of a sound. 
 
 f The digamma F was really and originally a compound of y or σ and 
 υ ; ' and from their combination, and from the different changes which 
 they separately and together admit of, arises that great variety of letters 
 which are traced to an original identity.' Donaldson, GJc. Gr. p. 10. 
 
CLASSIFICATION OF LETTEES. 
 
 15 
 
 φωνα) or liquids, which are λ, μ, ν, p, and the sibilant σ; ii. 
 double letters, £, £, φ; and iii. mutes (α^ωΐ'α), which do not 
 form a syllable, unless a yowel follows them, 
 iv. Mutes are divided into three classes, viz. : 
 
 Kough (aspiratce, ΰασεα), <j> χ θ. 
 Smooth (temtes, ψιλά), π /,•. τ. 
 Middle (media?, μέσα), β γ ό. 
 
 It is easy to remember the three aspirates, which at once recall the 
 three temtes ; the medics are the three first consonants, β, y, d. 
 
 13. Letters are also divided, according to the organs required 
 to pronounce them,* into 
 
 Labials, or lip-letters, π β φ μ. 
 
 Dentals, or teeth-letters, τ £ θ λ ν. 
 Gutturals, or throat-letters, κ γ χ. 
 
 In Hebrew grammar these letters are remembered by useful mnemonic 
 words ; e.g. the Labials by the word bumapk; the Dentals by d&thn&tk; 
 the G-utturals by g'ich&k. They are exhibited conveniently in the follow- 
 ing table, and should always be borne in mind. 
 
 
 Tenues 
 
 Mediae 
 
 Aspirat&e 
 
 Labials .... 
 
 π 
 
 β 
 
 Φ 
 
 Gutturals . . . 
 
 κ 
 
 Ύ 
 
 Χ 
 
 Dentals .... 
 
 τ 
 
 δ 
 
 θ 
 
 14. No Greek word (except ονϊ and εκ), ends in any con- 
 sonant except v, p, or ς (£, φ). Any other consonant at the 
 end of a word is rejected, as μέλι(τ), σώμα(τ), ήσαι (τ), &c. 
 Hence r has superseded μ in ετυπτον, and the first person 
 singular of other historical tenses. 
 
 15. Two laws of euphony are of constant recurrence : 
 
 i. When two letters of different organs (e.g. labial and 
 dental) come together, a tenuis only can precede a tenuis, a 
 medial a medial, and an aspirate an aspirate. 
 
 * This classification of letters is first found in Dionysius of Halicar- 
 nassus irept συνθέσεως ονομάτων. R was called by the Latins litera 
 canina — ' Irritata canis quod rr quam plurima dicit.' Lucil. S was 
 called litter a serpentina, and also solitarmm, because it stands alone. 
 
16 A BRIEF GREEK SYNTAX. 
 
 This is why we have 
 
 πλεχβείς, not πλεκθεις from πλέκω. » 
 τνφβείς, not τνπθεις from τυπτω. 
 εφθημερος, not επθημερος from επτά ημέραι. 
 νύχθ' δλην, not νι/κί)' ολην 
 λ£λ.τ<ίς•, not λεγΓος from λέγω ; 
 
 and so on. 
 
 The only exception admitted is in the case of the preposition e/c, as in 
 inSovvcu, ϊκθιΓιναι, ϊκβάλΧειν, &c. 
 
 ii. The Greeks dislike the concurrence of aspirates (when 
 not necessitated by the last rule, as is the case in τεθάφθαι, 
 εθρέφθηΐ', &c), and avoid it when possible. They had no ob- 
 jection to φβ, especially when the φ belongs to the root. 
 Bopp, i. 104, a. 
 
 Thus aspirates cannot be doubled, but the former is 
 changed into the corresponding tenuis, as in Βάκχος, Σαπφώ, 
 ΤΙιτβενς. 
 
 For the same reason, in reduplication, Ave have κεχώρηκα, 
 τίθημι, πεφνκα, for χεχώρηκα, θίβημι, &c; ετύβη)', σωθητι, for 
 έθνθην, σώθηθι, &c. And this accounts for such peculiarities 
 as θρίζ, τρίχας — τρέχω, βρέζω — ταχύ, Οάσσον — έχ ω > έ'£ ω > &C. 
 
 Exceptions are a. Some compounds, as ανθοφόρος, ορνιθοθήρας, &c. 
 
 b. The formative syllables -0tj and -θι are not changed, 
 as in πανταχόθεν, Κορινθόθι, ώρθώθην, τίθναθι ; or, 
 if any change is made, it is not, in the -07j of the 
 first aorist, but in the aspirate which follows it. 
 Thus Ave have τύφθητι, not τυπτηθι 
 
 C. αφή, υφαίνω, k0ev, ηχι. 
 
 Ν. Β. This dislike of concurrent aspirates, though found in Greek and 
 in Sanskrit, is not a peculiarity of the Aryan languages generally ; e.g. 
 in such Latin reduplications as /e/elli the f 's represent an original aspi- 
 rate. Ferrar's Comp. Grammar, p. 184. 
 
 Some interesting remarks on the peculiarities of the aspirate may be 
 found in Meissner's Palastra Gallica, p. 16. 
 
 VOWELS. 
 
 16. Attic Greek avoids hiatus, or the concurrence of vowels, 
 as much as possible, especially in verse. 
 
 17. The fusion or coalescence of vowels is called σνναλοιφη; 
 of which the varieties may be tabulated as follows : i. Ec- 
 thlipsis, or cutting off; ii, Crasis, or mixture of two words into 
 one ; iii. Synaeresis, or contraction of two syllables into one. 
 
VOWELS. 1 7 
 
 18. Synalcepha. 
 
 I 
 
 . • I . . J ... I . 
 
 i. Ecthlipsis or ii. Crasis or iii. Syna?resis or 
 
 Elision, as u(f ου Mixture, as Contraction, as 
 
 for άπο ου. και; for και εκ. τιμάτε for 7 ιμάητε. 
 
 i. Ecthlipsis. Elision and hiatus are often avoided by adding 
 a ν (called ν εψελκυστικον or παραγω-γικόν)* to various datives, 
 neuters, and 3rd persons. 
 
 The ι in -/, οτι, περί, and the datives in the 3rd declension 
 do not suffer elision in Attic. 
 
 ii. Crasis. The absorption of a short vowel at the beginning 
 of a word is called improper crasis ; as in ή "μη for η έμή, ΐ] 'γώ 
 for η ενώ. This is also called Prodelision. 
 
 The aspirate in a compound word may prevent crasis ; as 
 προέζω from προ and έ',ξω ; but προυχω from προ and έχω. 
 
 iii. Synceresis. The following of the least obvious contrac- 
 tions should be remembered : — 
 
 α.η=α, as τιμάητε =τιματε. 
 οη=ω, as ΰηλόητε =ΰηλώτε. 
 αε£ = ςι, as τιμάει —τιμψ. 
 οει=οί, as ΰηλόει =ΰηλοΊ. 
 ηο(=ω, as τιμάοιμεν =τιμωμεν. 
 α|7=9, as τιμάη =τφ$. 
 
 οη = οι, as ΰηλόη =ΰηλοϊ. 
 
 Besides this, there is an incipient crasis called Synizesis or 
 subsidence, by which two written syllables are pronounced as 
 one; thus in verse θεός is often a monosyllable, πόλεως a 
 dissyllable, &c. 
 
 * It must not, however, be supposed that this ν is a mere arbitrary 
 suffix. It may be laid down as a proved fact that in language nothing 
 is arbitrary. If the so-called ν ϊφζλκυστικόν is not purely a phonetie 
 necessity, it is the mutilated relic of some older termination. Schleicher 
 says, • Das bekannte ν ϊφζΧκυστιιών ist kein Best einer fruheren Sprach- 
 periode, sondern eine speciellegriechischejunge Erscheinung, z.B. e<pepe-v, 
 altind. und grundf. abharat ; in diesem Falle trat das ν also erst ein, 
 nachdem das urspriingliche auslautende t geschwunden war, und das 
 Sprachgefuhl sich gewohnt hatte, die Form als vocalisch schliessend zu 
 empfinden.' Vergl. Gram. § 149. (I have not thought it necessary to 
 preserve Schleicher's orthographic innovations.) 
 
 The ν ecpiXKvariKbv is in fact a kind of anusvarah or after-sound, as it 
 is called in Sanskrit grammar ; such as we find in τύμπανον, πίμπρ-ημι, 
 anguis (e%'s)> j8eV0os (jSaflos), &c, and twice over in such words as 
 λαμβάνω, μανθάνω, τνγχάιω, &c. 
 
18 A BEIEF GREEK SYNTAX. 
 
 19. While we are on the subject of these changes of form (meta- 
 plasms, as they are called), we may mention Apocope, the shortening of 
 a word, as δώ for δώμα ; Aphcercsis, the cutting off an initial sound, as 
 (ίβω for λε/βω; Metathesis, as θάρσος for θράσος; Syncope, as idolatry 
 for ειδωλολατρία, τράπεζα for τετραπβζα, &c. 
 
 DIALECTS. 
 
 20. Greek has three chief dialects, which may be tabulated 
 thus : — 
 
 Greek {<ρων)} 'Ελληνική). 
 
 J. I 
 
 i. li. iii. 
 
 Ionic rj 'las iEolic η AtoXis, Doric τ) Aoopis, 
 
 διάλεκτος of the lyric poets, of Pindar, Theocri- 
 
 Alcseus, Sappho, &c. tus, and the tragic 
 choruses. 
 
 J . ι . .1 
 
 Old Ionic or Epic, New Ionic of Attic η Άτθίς, 
 of Homer, Herodotus. of the tragedians, 
 
 Hesiod, &c. orators, historians, 
 
 philosophers, &c 
 
 τ) κοινή, or 'Ελληνιστική, 
 of the Septuagint, and 
 the New Testament. 
 
 i. The Old Ionic or Epic of Homer contains many forms 
 which afterwards became special in other dialects; hence 
 arose the common absurdityf of old Homeric commentators, 
 when they say that one form is Doric, another iEolic, &c, in 
 the same verse, as though Homer wrote in many different 
 dialects at once. 
 
 From its use in the soft regions of Asia Minor, and many 
 iEgaean islands, Ionic became pleasant and musical ; it rejects 
 aspirates (as όεκυμαι, ovnc), tolerates hiatus (as φιλίεαι), and 
 
 * Donaldson derives Awpiels from δα- and opos = Highlanders ; "iwvts 
 from riiov'ia = Coast-men (cf. "Αχαιοί Sea-men, AfyaAets Beach-men), 
 Αίολε?* from aio\os= Mixed men. Attica is Άκτικτ) the shore-land, 
 ακτή ' shore,' being derived from &•γνυμι ' I break.' 
 
 f The grandest instance of this is the remark of Herakleides on the 
 word ειλήλουθμεν, which he says is a mixture of four dialects, τίσσαρσι 
 πεποί-ηται διαλέκτου ! The ν is Attic ; the ο Boeotian ; the t Ionic ; and 
 the syncope iEolic ! Nothing can beat this ! (See Kleist, Be Philoxeni 
 Stvd, Etymol. p. 41.) 
 
TABLE OF DIALECTS. 19 
 
 avoids contraction (as τνφθέω, -έης, -εη) ; it uses η where the 
 Doric uses α (as ήμερη), ου for ο (as μοννος), ω for οη (as 
 'ίνωσα for ενόησα), εϋ for εο (as πλεΰτες for πΧίονες), &c* 
 
 The chief peculiarity of the Attic is its proneness to con- 
 tractions ; this may seem a strong contrast to its kindred 
 dialect the Ionic, but in point of fact the uncontracted vowels 
 of the Ionians spring from the rejection of intermediate con- 
 sonants, and the Attics only went one step farther by contract- 
 ing the vowels in order to avoid the resultant hiatus. 
 
 ii. The iEolic is chiefly interesting from the points of 
 resemblance which it offers to Latin. 
 
 a. Thus, like Latin, it has no dual;f such at any rate is the 
 case in Lesbian iEolic. 
 
 b. Like the Doric, it makes the first person plural in μες 
 (not μεν), the Latin mus, as ηνθομες venimws, τύπτομες verber- 
 amus ; and the third person plural in rri, like the Latin nt, 
 τΰπτοντι verberafti. 
 
 c. Nominatives in της it forms in tu, as ιππότά, αιχμητά, 
 like the Latin poeta, nauta, scriba, &c. 
 
 d. It makes but little use of the middle. 
 
 e. It accentuates, more frequently than other dialects, on the 
 penultimate or antepenultimate syllable. 
 
 iii. Doric was characterised by its πλατειασμός (brogue, 
 or broad sound), especially in the use of α for η, &3*φαμά, 
 τεθνακώς. This very breadth and richness of sound made it 
 better suited for songs and music (as the Scotch dialect among 
 us), and hence {among other reasons) its appearance in the 
 tragic choruses. 
 
 It puts α for ω, as τάν μονσάν for the gen. plur. 
 
 α for ε, as εγώ γα. 
 
 ε for εί, as τύπτες, μελίσΰεν (for μελίζειν). 
 
 κ for τ, as πόκα for ποτέ. 
 
 ν for λ, as r'jvdov, βεντιστος. 
 
 τ for σ, as τίθητι, φατι. 
 
 * Numerous Epic forms may be observed by attentively reading any 
 page of Homer, e.g. the infinitives in ςμεναι, the genitives in oio, the 
 dative plurals in fjai, &c; and new Ionic forms in any page of Hero- 
 dotus, as S>v for ουν, ivdavra for ϊντανθα, &C. 
 
 f The grammarian Theodosius (Bekker, Anecd. Grcsc. p. 1184) says 
 Ot AtoAets ουκ ϊχουσι δυϊκά, 'όθεν ovBh oi 'Ρωμαίοι, άποικοι fores των Αΐο\4ων. 
 The ' Cui est sermo noster simillimus ' of Quinctilian is well known. 
 (Instt. Or. i. 1-6.) But no genealogical connection between the two 
 must be dreamed of. The interesting question of the real relation of 
 Greek to Latin belongs to Comparative Philology. 
 
20 A BRIEF GREEK SYNTAX. 
 
 iv. The common dialect (>/ κοινή), often called Hellenistic 
 Greek, or Greek spoken by those who had acquired it as a 
 foreign language, owed its origin and dissemination to the 
 conquests of Alexander. It is a somewhat corrupt and loose 
 Attic, with an admixture of Macedonian and Alexandrian 
 words. It adopts various new forms, as φεΰσμα, νϊκος, νου- 
 θεσία, έκχύνειν, στηκω, ομνύω for ψεΰΰος, νίκη, νουθέτησις, 
 έκχέειν, 'ίστημι, υμνυμι ; it admits various poetical words, as 
 αυθεντείν l to lord it,' αλέκτωρ for άλεκτρυων, εσθω for εσθίω, 
 βρέχω 'to rain,' &c. ; it uses old words in new senses, as 
 σννίστημι ' I prove,' όφώνιον ' wages,' έρεΰγεσθαι eloqui, γεννή- 
 ματα l fruit,' λαλιά ' language ; ' and it frames new words and 
 new compounds,* as γρηγουώ, παιίιόβεν, καλοποιεΊν, αιματεκ- 
 χυσ'ια, ταπεινοφροσύνη, άκροβνστία, σκηνοπηγία, εΙΰωΧόθυτον. 
 Besides this, it ceases to employ the dual ; entirely abandons 
 the use of the optative in oratio obliqua ; uses the infinitive 
 instead of the future participle after verbs of going, sending, 
 &c; admits ει with the subjunctive, όταν and Ίνα with the 
 pres. ind. ; and, finally, shows a tendency to analysis, by using 
 prepositions f where the case-terminations would have been 
 originally sufficient to express the meaning, and by employing 
 the active with εαυτόν instead of the middle (ετάραξεν εαυτόν 
 = εταράζατο). 
 
 PARTS OF SPEECH (τα μέρη, τα στοιχεία, του λόγου). 
 
 21. It is probable that all words may be reduced to roots 
 which are either the bases of nouns, or are pronouns denoting 
 relations of place ; and indeed, at first, roots stood (as is still 
 the case in Chinese) for any or every part of speech. The 
 distinction between their functions is due to the advance of 
 Language. (See Chapters on Language, p. 197.) 
 
 22. A long time elapsed before men learned to analyse into 
 distinct classes these ' grammatical categories.' Plato (Crat. 
 § 88; Soph. p. 261) only recognises the noun and the verb. 
 Compare the remark of Jack Cade, ' It will be proved to thy 
 face that thou hast men about thee that usually talk of a noun 
 and a verb and such abominable words as no Christian ear can 
 endure to hear.' — Henry VI. , part ii. iv. 7. To these Aris- 
 totle adds conjunctions (συνίεσμα, συγ κατηγορήματα, see 
 
 * Many Latin words in Greek characters occur in the New Testament, 
 as Χζγζών, KiVTOvpiwv, σουΰάριον, σπεκουλάτωρ, κηνσος, &c. 
 
 t e.g. αποκρύπτςιν τι από rivos, iaOUiv άπ& rwv ψιχίων, προσκννίιν 
 Ινώπιόν tivqs, &e. 
 
PARTS OF SPEECH. 21 
 
 Quint. Instt. Orat. 1. iv. 12), and the article (Arist. Poet. 20). 
 The Stoics* and the Alexandrian grammarians finally adopted 
 the division into eight parts of speech, which the Romans 
 borrowed from them, only omitting the article and distinguish- 
 ing the interjection from the adverb. 
 
 NOUNS (Ονόματα). 
 
 23. The Greek noun has five cases, three numbers, and 
 three genders. There are usually said to be ten declensions 
 {κλίσεις), and it is true that all substantives, not anomalous, 
 may be classed under ten types. But there was originally 
 only one declension, and the various types alluded to, arise 
 from the gradual changes assumed by the inflections in course 
 of time under phonetic influences. In all more modern and 
 philosophical grammars (as, for instance, those of Curtius, 
 Donaldson, &c.) the declensions are more properly ranged 
 under three heads, viz. the vowel declension, which has two 
 divisions, i. the α declension, when the uninflected^ form of 
 the noun ends in α or η (ταμία-ς, κριτη-ς) and the fern, noun 
 in α or η ; ii. the ο declension, when the uninflected form of 
 the noun ends in o, as λόγο-ς ; J and iii. the consonant declen- 
 sion, when the uninflected form ends in a consonant, or (the 
 final consonant having been lost) in t or v. 
 
 There is no doubt that this is the better and truer arrange- 
 ment; in any case, however, the declension of a certain 
 number of typical nouns must be learnt by heart. A better 
 arrangement may enable the student to understand better, 
 and to master with more rapidity, the laws and genius of the 
 language, but there is no royal road by which labour in the 
 acquisition of the language can be avoided. 
 
 CASES {ΐίτωσεις). 
 
 24. Cases (πτώσεις, casus, fallings) were probably so called 
 because the nominative was regarded as the normal or upright 
 
 * For other tentative divisions of the Parts of Speech, see Burggraff, 
 Principes de Grammaire Generate, p. 176. They are all contained in the 
 Greek, line, rrpbs δ' ^ite rbv ϋνσ-ττινον <ίτι φρονέοντ' iKi -ησον, Ε. xxii. 59, 
 and in the Latin line, ' Vae tibi ridenti quia mox post gaudia flebis.' 
 
 f The stem or uninflected form must be carefully distinguished from 
 the nominative case. Thus προγμοτ- is the stem of the nominative 
 irpay^a, gen. νρά-γματ-ος ; and \oyo- of the uom. KOyos. 
 
 X This includes nouns like v6os, vovs, οστίον, \eos, &c, where the 
 uninflected form ends in oo or eo. 
 
22 A BHIEF GREEK SYNTAX. 
 
 form of the word, and the other cases as deflections from it 
 (πλάγια ι obliqui). The Sanskrit grammarians call a case 
 vibhakti, i division.' Hence also come the terms κλίσις, 
 declensio. 
 
 25. The cases are — 
 
 Nominative* (ευθεία or ορθή πτώσις casus rectus). 
 Genitive {γενική, κτητική, πατρική). 
 Dative (βοτική, εττισταλτική). 
 Accusative (αιτιατική). 
 Vocative {κλητική). 
 
 26. The nature and use of these cases will be briefly ex- 
 emplified farther on. We must however observe that neither 
 nominative nor vocative are properly cases, nor did the Stoics, 
 from whom the term is derived, ever call them so ; since 
 they are independent and, so to speak, upright forms of the 
 word, not resting or depending on other words. 
 
 27. Besides these cases there was originally a sixth locative 
 case, which is still retained as a distinct form in some nouns, as 
 Άθηνησι, Πλαταιάσι, Όλνμπίασι, &C. at Athens, Platea, Olym- 
 pia, &c; θνρασιν ' foris,' out of doors; ΜεγαροΊ, Πυθοϊ, Μαρα- 
 θώ νι, οίκοι (domi) at Megara, at Pytho, at Marathon, at home. 
 
 28. That the case-endings in Greek, as well as in all other languages, 
 are mere corruptions of words once separable, is certain ; and that in 
 Greek these words were pronominal in their nature (i.e. forms of pro- 
 nouns) may also be considered certain. (See Donaldson's GJc. Gram. 
 p. 80, G-arnett's Philolog. Essays, 217 seqq.) The case-endings, like the 
 pronouns from whence they spring, originally represented only concep- 
 tions of space (nearness, distance, presence, absence) ; but they were after- 
 wards extended to express relations of time, cause, &c. Bopp, Compar. 
 Gram. § 115. The etymology of inflections is of course difficult from 
 their antiquity, and the numerous contractions and other changes they 
 have undergone. Having bit upon these pronominal words as mere 
 formative elements, language naturally made them as mechanical as 
 possible. For the original sense oi: the pronominal roots is nearly iden- 
 tical, and many new meanings had to be given to them. 
 
 There are three pronominal elements π, ο, τ, or pa, qua, ta, which 
 mean primarily here, near, and there. 
 
 1. The first (π) under the forms πα or μα, signifies superposition, and 
 occurs in the first personal pronoun Que) and the first numeral (^eis, μία, 
 μ4ν, compare our ' number one ' = I). 
 
 2. The second (9 qua), under a great variety of different forms, sig- 
 
 * The first passage in which the names of the cases occur is in 
 Chrysippus περί των irivre πτώσεων (αρ. Diog. Laert. vii. 192). ττΧά -yiai 
 he 7ττώ<Γ€ίϊ eiVt yeviK^ [κα\ δοτζκ^] κα\ αΙτιατική, Lersch, Sprachphiloso- 
 phie, ii. 185. 
 
NUMBEES. 23 
 
 tides proximity, and occurs in the second personal pronoun, and in the 
 nominative and dative cases. 
 
 3. The third (τ) denotes distance, and, variously modified, is found in 
 the third personal pronoun, in negatives, in the genitive and the accusa- 
 tive cases. 
 
 To make this quite clear, and to follow these elements through their 
 various changes, would require an entire treatise; we may, however, at 
 once make the important observation that these three main relations of 
 derivation, proximity, and direction towards, are respectively expressed 
 by the genitive, dative, and accusative. 
 
 29. Language, as it advances, tends to discard cases, and indeed all 
 synthetic forms. The dative has disappeared from Modern G-reek. The 
 Komance languages have almost entirely discarded cases, using preposi- 
 tions instead, i.e. expressing the requisite shades of meaning analytically, 
 not synthetically. So too in English, where the s of the genitive is 
 almost the only remaining case, except the m of the old dative plural in 
 them, whom, seldom, whilom, &c. In some ruder languages (e.g. Basque, 
 Greenland, &c.) there are very many cases. 
 
 30. The numbers are singular (ενικός), dual (ΰυϊκός), and 
 plural (πληθυντικός). 
 
 NUMBERS (Αριθμοί). 
 
 How many numbers is there in nouns ? 
 
 Two ' Merry Wives of Windsor, iv. 1. 
 
 31. The dual number (in the possession of which the Greek 
 noun resembles the Sanskrit and Hebrew, but differs from 
 Latin and most modern languages) is a mere luxury of lan- 
 guage,* probably arising from the number of things which 
 are usually and necessarily spoken of in pairs.\ That there 
 
 * The dual survives in Lithuanian and Icelandic, and once existed in 
 the Vnglo-Saxon personal pronouns. In English we have the one dual 
 word twain, but even this is corrupted into twins. 
 
 f Another theory about the dual is that it was an older plural, origi- 
 nating in the primary notion of the Ego and the Non-ego, or in the fact 
 of there being two speakers, / and you, which stamps a character of 
 dualism on the very essence of speech. It is curious that nos and vos in 
 Latin are obviously connected not with νμ^, νμβις, but with the duals 
 νώ, σφώ. (Cf. vaifoepos noster.) Donaldson accepts the theory that the 
 dual is an older and weaker form of the plural, and mentions that some 
 considered the Latin forms dixere, &c. for dixerunt, &c. as duals. 
 (Quint, i. 5, § 42 ; New Crat. p. 396.) Schleicher (Compend. § 243) 
 thinks that the dual may have been originally a mere doubling of the 
 plural. Du Ponceau's jest that it must have been invented for lovers 
 and married people finds a curious illustration in certain dual-forms in 
 Australian dialects. Eor this and many other interesting facts about 
 dual and plural, see Geiger, Urspr. d. Sprache, § ix. 369-386. Lord 
 Monboddo's remarks (Orig. of Lang. i. 550) are a strange mixture of 
 shrewdness and error. 
 
2-4 A BRIEF GREEK SYNTAX. 
 
 is a slight distinction between the conceptions of duality and 
 plurality we may see at once from the fact that we cannot 
 use the word ' all ' of two, though we can of three things. For 
 instance, we could not say ' Two birds sat all together on a 
 tree.' Nothing but an instinctive feeling that such a form 
 corresponded to some external reality, could account for its 
 existence among people so utterly unlike each other as Green- 
 landers and New Zealanders on the one hand, and Attic 
 Greeks on the other.* It is however quite unnecessary to have 
 a separate inflectional form for so slight a difference of con- 
 ception, and as it is the tendency of advancing language to 
 get rid of its original superfluous exuberance, it is mainly in 
 dead languages and obsolete dialects that the dual exists. A 
 language may be too perfect in its synthetic forms, and so 
 tyrannise over the free motion of the intellect. Simplicity, 
 not complexity, is the triumph of language ; and an immense 
 wealth and multiplicity (divitias miseras !) of grammatical 
 forms f is mainly to be found in the most savage languages, 
 such as Kaffir, and the languages of the American aborigines. 
 Hence the dual, being unnecessary, early begins to evanesce, 
 and to be treated as quite subordinate to the plural.J It is 
 not found in iEolic, barely in Hellenistic Greek, and in Modern 
 Greek it has ceased to exist. § Long before it disappeared, the 
 sense of it as a grammatical form is so vague that it may 
 always be put with a plural verb ; and as in Hebrew Ave find 
 such collocations as JYID") WW ' lofty eyes,' where the noun is 
 dual, and the adjective plural, so in Plato we have εγελασάτην 
 
 * See on this whole subject the very interesting pamphlet of W. von 
 Humboldt, Ueber den Bualis, Berlin, 1828. He quotes from Lactantius 
 the remark, ' Ex quo intelligimus quantum dualis numerus, una et sim- 
 plici compage solidatus, ad rerum valeat perfeetionem.' Be Opif. Dei. 
 
 f The Abipones, a tribe in Paraguay, have two kinds of plurals, one 
 for two or three objects, and another ending in -ripi for larger numbers. 
 We may observe that as long as language is regarded as in itself an 
 end, it abounds in forms capable of expressing the minutest distinctions ; 
 but, as civilisation advances, language becomes more and more a mere 
 instrument, and therefore only retains those forms which are necessary 
 to produce immediate comprehension. 
 
 \ Another trace of this fact is that the masc. of the dual in the article, 
 and in αυτός, ovros, 4μός, &c, is in Attic put with fern, nouns ; as δύο 
 ripe loea (Plato), τούτω τω νμ4ρα, τω χεΤοε, &c. (Xen.). Observe, too, 
 that the dual has only two case-terminations; having only three even in 
 Sanskrit. (Meyer, Gedrangte Vergl. d. Gr. und Led. Bed. S. 54.) 
 
 § Chceroboscus wrongly argues from this fact, τά δυϊκα vaTepoyevrj 
 iffTiv νστίρον yap 4π(νοΊ]θησαν το δυϊκά. (Bekk. Anecd. Grcec, iii. 1184. 
 
GENDERS. 25 
 
 άμψω, βλέφαντες εϊς αλλήλους (Plato, Euthyd. 273 d) ; and 
 even in Homer we find such concords as οσσε φαεινά, and 
 βασιλήες .... πεπννμένω αμφω, Od. xviii. 64. No doubt, 
 however, the possession of a dual stamps on language some 
 of that beauty of form which is so remarkable in Greek ; and 
 the κρατερόφρονε γείνατο παΊΰε of Homer is more lively and 
 expressive than the ' Ambo conspicui, nive candidioribus 
 ambo Vectabantur equis' of Ovid. ' The strong logic of the 
 Italians,' says Mommsen, ' seems to have found no reason for 
 splitting the idea of moreness into two-ness and many-ness.' 
 Besides the words ambo, duo, and possibly octo, the only trace 
 of a dual in Latin is the neuter dual termination I in vigintl 
 (see Corssen, Krit. Nachtr. zur Latein. Formenl. S. 96). The 
 same is true of Pali. In Prakrit the dual disappears alto- 
 gether. 
 
 31 (bis), i. The Sanscrit plural as for masc. and fern, nouns 
 is an enlargement of s, the sign of the nominative singular, the 
 enlargement being a symbolic indication of plurality. The 
 neuter alike in the singular, dual, and plural is deprived of s, 
 which is reserved for genders which indicate persons. Bopp, 
 § 226. 
 
 ii. The method of forming numbers in other languages forms 
 a curious chapter of philology. In Chinese and other mono- 
 syllabic languages, plurality is expressed by the addition of 
 words meaning c another ' or ' crowd.' In Basque the plural 
 can only be expressed by suffixing the plural article, e.g. 
 gizon = man, gizonak = men (homme-les), ak being the plural 
 article ; ' mais il n'est pas possible a exprimer hommes,' Van 
 Eys, p. 14. See too Geiger, ubi supr. 
 
 GENDERS (γένη). 
 
 32. In the ancient, and in many modern languages, the 
 substantive expresses the gender (γένος), real or imaginary, 
 of the object which it names. There are usually, as in 
 Greek, three genders, masculine (άρσενικόν), feminine (θηλυκόν), 
 and neuter (ουΰέτερον),* but some languages (e.g.theHebrew)f 
 
 * Words like 'ίππος, άνθρωπος, &c, are common ; and -words which do 
 not change their gender, though applied to different sexes, are called 
 ίπίκοινα epicene ; e.g. Aristotle says, κα\ δ θήλυς 5e opevs έπΧηρώθη. Hist. 
 Anim. xxiv. The sophist Protagoras is said to have been the first to 
 call marked attention to the genders of words. See Aristoph. Nub. 669. 
 
 f Hence we have the fern, for the neut. in the LXX. version of Ps. cxix. 
 60, cxviii. 23. The names ohZfrepov, neutrum, ' neither of the two,' show 
 
 C 
 
26 Λ BRIEF GREEK SYNTAX. 
 
 use the feminine to express the neuter, to which we find, some- 
 thing analogous in the fact that, in Greek and Latin, feminine 
 names are often of a neuter form, as Πλόκω%>, Gly cerium,* just 
 as in German all diminutives in -chen and -lein are neuter 
 (das Madchen, das Frauleiri), even when they signify females. 
 The feminine is generally indicated by a weakening of the 
 masculine termination. 
 
 33. The attribution of any gender to inanimate things 
 only leads to endless confusion and anomaly, and a multipli- 
 cation of rules and exceptions, for the most part admitting of 
 no rational explanation, but due to the varying influences ol 
 fancy or caprice. It is the relic of a time when the imagina- 
 tion was much more active than now, and when the energetic 
 fancy of mankind attributed a life, analogous in some respects 
 to its own, to the whole external world ; and, as some would 
 express it, tinged everything with which it dealt with some 
 faint trace of its own subjectivity. The necessity of regarding 
 everything as partaking of life, and therefore as having some 
 gender, is a heritage of the childish-poetic stage of human in- 
 telligence, when")" language was regarded as an end as well as 
 a means, and when the mind felt an imperious necessity that 
 the forms of language should faithfully reflect the slightest 
 variations of conception. 
 
 The fancifulness of genders may be seen by comparing the 
 same word in different languages. Thus κ•αρδ<'α ' heart ' is 
 feminine ; but cor is neuter, and coeur masculine. In French 
 laleur is masculine, doulenr feminine ; and couleur though 
 derived from color is feminine, arbre though from arbor is 
 masculine. In most languages, for obvious reasons, the sun 
 is masc, the moon fern. ; but in Gothic, Anglo-Saxon, J and 
 
 how purely negative was the conception of the neuter gender; in San- 
 skrit it is called Jcliva, ' eunuch ; ' in Servian srednji, ' intermediate gender; ' 
 inDutch onzijdig, unsided, ' qui ne penche d'aucun cote.' — Du Meril, p. 356. 
 
 * It is a well-known rule in Greek that when women speak of them- 
 selves in the plural, they also use the masculine. 
 
 f See the author's Origin of Language, p. 45 ; Chapters on Language, 
 p. 212. There is really no more necessity for gender in nouns and adjec- 
 tives than there is in verbs which also express gender in Hebrew, Arabic, 
 and Berber. The American languages ar6 without it. 
 
 \ ' Mundilfori had two children, a son Mani, and a daughter Sol/ 
 — The prose Edda. See Latham, Engl. Lang. ii. 156. In Hebrew &Κ>ψ 
 sun is sometimes fern., ΠΤ moon is masc. But another word for moon 
 Π33? is fern. (cf. b μ^ν, η ashtyn. ' Dispicite .... masaUum Lunam.' 
 Tertul. Apol. 15. Forcellini, s. v. Lunws). 
 
ANOMALIES OF GENDERS. 27 
 
 German, it is the reverse, der Mond, die Sonne, and in Russian 
 the sun is neuter. Again, in German, a spoon is masc. (der 
 Loffel), a fork fern, {die Gabel), a knife neuter {das Messer) : 
 so too a jug is masc. {der Krug), a cup fern, {die Tasse), a 
 basin neuter {das Becken); wine is masc, milk fem., beer 
 neuter {der Wein, die Milch, das Bier) ; the beginning is 
 masc, the middle fem., and the end neuter {der Anfang, die 
 Mitte, das Ende). And to crown this capricious absurdity, 
 the word for wife, of all things in the world, is neuter {das 
 Weib !).* French has discarded the neuter gender ; and Eng- 
 lish (like Persian and Chinese) abandons genders altogether, 
 or only expresses them (when necessary) by a separate word, 
 except in the 3rd personal pronoun {he, she, it), and the rela- 
 tive {who, which). We may well congratulate ourselves, 
 therefore, that our language has been one of the very few 
 which have had the wisdom to disrobe itself of this useless 
 rag of antiquity, and to make all inanimate objects neuter, 
 except in the rare cases where they are personified for the 
 purposes of poetry (Prosopopoeia). 
 
 Many of these anomalies are accounted for by the fact that 
 sometimes the form of the word determines its gender, entirely 
 irrespective of its meaning, and sometimes the meaning irre- 
 spective of the form. Thus rivers and hills are generally 
 masc, but Α'ίτνη, "Οσσα, Αήθη, Στύζ, are fem., Avcaiov neut. 
 And in spite of their meaning μειράκιον, παώίον, άνΰράποΰον are 
 neuter ; while in spite of their form κύρίοπος and κάμινος are 
 feminine. 
 
 It is curious to observe that in Modern Greek the prevalence 
 of diminutive forms — (e.g. ώίΐί from όψίΐιον = snake, \papi from 
 d\[/apioj' = fish, and in the Tzaconian dialect, spoken about the 
 Gulf of Nauplia,"]" -φιουχαρον^α = butterfly, a diminutive of 
 ψνχι?, &c.) — is due partly to a desire to secure uniformity of 
 genders. 
 
 Rules of Gender. 
 
 34. The following are the general rules of gender : — 
 
 1. Names of male persons and animals, of rivers {6 ποταμός), 
 
 hills (ό λόφος), winds .(ο άνεμος), and months {δ μην), "are 
 
 masculine. 
 
 * Possibly because a wife was regarded as a chattel ; possibly, how- 
 ever, on the other hand, the neuter may here be a term of endearment, 
 as we speak of a child as ' a dear little thing.' 
 
 t See Le Dialecte tzaconien, par G-. Deville. Paris, 1866. 
 c2 
 
28 A BRIEF GREEK SYNTAX. 
 
 2. Names of female persons and animals, of trees, lands 
 (>/ γή), islands (»'/ νήσος), and cities (η πόλις^, are feminine; 
 also most abstract substantives, as ι/ ελπίς hope, η νίκη victory, 
 η αρετή virtue. 
 
 Exception. — A few trees and plants are masculine ; of which the com- 
 monest are φοίνιξ palm, ipivebs wild fig, Karros lotus, κΰτισο*, αμάρακοί, 
 αο~<ρόδ(\ος, k\\4fiopos. 
 
 3. Most diminutives, names of fruits, and names of things 
 regarded as mere material objects, especially if they are re- 
 garded collectively as forming a class, are neuter ; also all 
 infinitives used substantively, as τό ζην, life. Such phrases as 
 το άνθρωπος mean * the word " man." ' 
 
 The following common words, which are fern., though they 
 end in oc, should be remembered : — 
 
 i. Names of countries, islands, cities, plants. 
 
 ii. Names of earths and stones, as // ψάμμος sand, η πλίνθος 
 the brick, η \Ιή<ρος the pebble, // λίθος the gem. 
 
 iii. Different words for ' a way,' as οδός, κέλευθος, ατραπός, 
 αμαξιτός. 
 
 iv. Various receptacles, as γναθός jaw, κιβωτός chest, λ//ιός 
 wine-vat. 
 
 v. Adjectives used substantially, as η ήπειρος, χέρσος, 
 έρημος (sc. γη)* η κέρκος (ουρά), η διάλεκτος (φωνή). 
 
 A few other feminines in ος are difficult to class, as νόσος 
 disease, δρόσος dew, δοκός beam, ράβδος staff, βίβλος book. 
 
 The feminine also denotes a collection of things, as // "ίππος 
 cavalry, // κάμηλος a troop of camels ; in the case of animals 
 this is probably due to the fact that in a number of animals 
 the females largely predominate. 
 
 DECLENSIONS (Κλίσεις). 
 
 35. Besides the ordinary forms of declension, there are 
 traces of another declension formed by suffixes : -θεν for the 
 genitive, -0i for the locative, -δε for the accusative. These 
 terminations answer the questions πόθεν ; που ; ποΊ ; 
 
 Thus — ποϋ ; where ? ο'ίκοι at home, θύρασι at the doors, 
 ΤΙύθοι at Pytho, άλλοθι elsewhere. 
 πόθεν ; whence ? ο'ίκυθεν from home, θύραθεν from the 
 door, αυρανόθεν from heaven, ριζό- 
 θεν from the root (radicitus). 
 
 * Possibly η vrjaos (yrj) may be ' the floating land ' (veot). 
 
ADJECTIVES. 29 
 
 τοι ; whither? όίκαίε (domum) homewards, θνραζε 
 towards the door, Άθηιαζε to 
 Athens, νόλινΰε to the city, εραζε 
 to the earth. 
 
 36. Homer also uses -<pi for the gen. (or perhaps we should rather 
 say locative — Bopp, ii. 23, ed. Br£al) and dat. both sing, and plur. (evi- 
 dently analogous to the Sanskrit instrumental bht/as, bhis) ; of which 
 we find a trace in the Latin ibi (dat. of is), tibi, alicufo", sicuit, vobis, 
 nobis, and the dat. plurals in -bus. (Corssen, Latein. Formenl. S. 206.) 
 
 The derivation of this syllable bhi is unknown. Pott derives it from 
 abhi ' towards,' but this is probably itself a case of the pronoun a. See 
 Breal, Bopp, ii. 36. 
 
 Heteroclites, &C. 
 
 37. "Words that mix two declensions are called heteroclites, 
 as σκότος gen. σκότου and σκότους, Ύάρταρος plur. Τάρταρα, 
 σίτος pi. σϊτα. 
 
 ADJECTIVES (Επίθετα). 
 
 38. Adjectives, though highly convenient, are not indis- 
 pensable to a language. The fact that substantives are fre- 
 quently used adjectivally (e.g. mahogany table, door lock, 
 artillery officer, &c), and that their place can always be sup- 
 plied by a periphrasis of the noun and preposition (e.g. aurea 
 corona = une couronne oVor, multi homines =beaucoup d'hom- 
 mes, ein goldener Ring=ein King von Golde, &c), accounts for 
 the non-existence in many languages of adjectival forms which 
 occur in languages cognate to them. For instance, the Latin 
 tot, quot, quotus, pauci, &c, can only be rendered in French 
 by autant, tant, combien, peu, &c, with de. In Arabic, ' all 
 men,' ' no men,' ' some men,' &c, can only be expressed by 
 * the totality of men,' ' not one among men,' * a portion of 
 men,'* &c. In Greek, as in all languages, many adjectives 
 are used for nouns, especially in poetry ; as πέιτοζος the five- 
 pronged, i.e. the hand, φερέοικος the house-bearer, i.e. the 
 snail, άνόστεος the boneless, i.e. the cuttlefish, &c; and in 
 English, ' the deep,' ' the blue,' ' the true and the beautiful,' 
 &c. Milton uses many such adjectival substantives, e.g. ' the 
 palpable obscure,'' ' the vast abrupt,' &c\ Compare, ' till that 
 wicked be revealed,' 2 Thess. ii. 8 ; ' the silent of the night,' 
 
 * Silv. de Sacy, Gram. Gen. p. 54 ; Lobeck, Aglaopham. p. 845 ; Ede"le- 
 etand du Meril, Sur la formation de la languefrang. p. 54. 
 
 t In French many nouns have been formed from adjectives, e.g. 
 eanglier (porcus singularis), bouclier (scutum bucculatum), &c. 
 
30 A BRIEF GREEK SYNTAX. 
 
 2 Henri/ VI. i. 4 ; ' and mighty proud to humble weak doth 
 yield,' Spenser, F. Q. iii. 7. 
 
 39. As there Was no prima facie reason why the adjective 
 should so closely reflect the nature of the substantive with 
 which it is joined as to express its gender by a different in- 
 flection, we find many adjectives (especially those compounded 
 with Ιυσ-, εν-, α-) which have only two terminations, and do 
 not express the feminine by a separate termination ; nouns also 
 are often used in apposition with other nouns as though they 
 were adjectives of one termination ; as η μαινας γυνή, η πατρίς 
 •γη, &c. This is a gradual approximation to the English use 
 of the adjective, for in English also the adjective used to agree 
 with nouns, as, Ο younge Hughe, thinges espiritueles, water es 
 principales, &c. 
 
 40. The adjectival termination is, at any rate in very many 
 cases, derived from the pronominal suffix which forms the 
 genitive case of nouns ; e.g. Ιημον—ΰημό-σιο, which becomes 
 the adjective ΰημόσω-ς by adding a new case-ending. {New 
 Cratylus, p. 474.) In many languages genitives become 
 adjectival without any change at all ; e.g. in Finnish, kav-en 
 =of a stone, and stony; in Basque, guizon-aren-a=o/ man, 
 and human,* &c. 
 
 41. The three degrees of comparison are Positive (όνομα 
 άπλοϋν), Comparative (συγκριτικόν), and Superlative (νπερ- 
 θετικόν). 
 
 42. There are in Greek two modes of forming the compa- 
 rative and superlative, one by means of the terminations τερος, 
 τατος, and the other by ίων, ιστός ; τερος, τατος imply excess 
 (more, most) ; τερος indicates ' motion from ' (cf. Tpr&ter, subter, 
 propter), and τατος ' motion through a series of points,' since 
 τα denotes distance, and pa motion. (Donaldson.) 
 
 43. The comparative and superlative in -ιων, -ιστός (being 
 in fact mere strengthened forms of the adjectival termination 
 ως) are originally qualitative ; i.e. they do not so much imply 
 excess, as ' a considerable amount of, 1 like our termination -ish 
 in brack-ish, or our qualifying word ' somewhat,' meaning ' a 
 little too much,' as in ' somewhat bitter,' &c. 
 
 [N.B. The ι in ιων is long in Attic, short in Homer.] 
 
 44. It is clearly a defect both of Latin and Greek that they 
 use the same form to express two conceptions so distinct as 
 1 somewhat ' and < ??iore;' e.g. that ηΖίων according to the 
 
 * Garnett, Philol. Ess. p. 267. 
 
PRONOUNS. 
 
 31 
 
 context may either mean 'sweeter' or 'sweetish,' of which 
 the former is a comparison between relative qualities, and the 
 other & judgment about a positive quality.* There were how- 
 ever certain intensive prefixes which served the latter pur- 
 pose, such as the Epic intensive prefixes ζα-, ερι-, apt- (ζάπλου- 
 τος, έρικνΰής, άρίζηλος, &c), the comic prefixes Ιππο-, βου-\ 
 (Ίππόκρημνος, βούλιμος, βοΰπαις ' cf. our horse-laugh, horse- 
 mushroom, &c), and τρις-, παν-, which are used in all poets 
 and even in prose (πάγκαλος, παγγέλοιος, παμπόνηρος, τρισμα- 
 κάρως, &c. ; cf. our Almighty, &c, and the German prefix 
 oiler-, in allerliebst, &c). To express a less degree they used 
 the preposition υπό, as υπόλενκος subalbws, whitosA, υπογελαν 
 to smile. 
 
 45. 'Αγαθός good, and κακός bad, borrow several compara- 
 tives and superlatives from other forms ; but these comparatives 
 and superlatives are not absolutely synonymous. 
 
 'Αγαθός good, 
 
 Κακός bad, 
 
 άμείνων\ better ex- 
 ternally, 
 κρείττων stronger, 
 βελτίων morally better, 
 
 λώων preferable, 
 
 ψέρτεροςιηοΓβρΓοβίαΙ^ 
 κακίων baser, more 
 
 cowardly, 
 χειρών inferior, 
 
 Ν. Β. ύστερος, 
 πρώτος from προ 
 
 ησσων weaker, 
 
 ύστατος are derived 
 έσχατος from εζ. 
 
 άριστος (from" Αρης the 
 War-god). 
 
 κράτιστος (from κράτος). 
 
 βέλτιστος (Latin bonus, 
 comp. Ionic βεντ ι- 
 στός). 
 
 λωστος (from λάω to 
 choose). 
 
 φε'ρτατος. 
 
 κάκιστος. 
 
 χείριστος (from χειρ, 
 
 χείριος subject). 
 ηκιστα (adv.). 
 
 from υπό ; πρότερος^ 
 
 PRONOUNS (Άντωννμίαι). 
 
 46. A few words of explanation will perhaps throw some 
 light on the nature of pronouns. 
 
 * The kind of confusion thus introduced may be illustrated by this 
 passage : ' If that collar-bone of yours had not been all the harder, you 
 would have been,' &c. &c. — Tom Cringle's Log. ch. xvi. 
 
 f ^ίωθε yap η προσθήκη των τοιούτων ζφων rb μςγςθοε του ύποκ€ΐμ4νου 
 δηλονν. — Etym. Magn. 
 
 \ On these forms see Donaldson, New Crat. § 262. They are also 
 distinguished in Donaldson's grammar, and partially in Burnouf s, § 197. 
 
32 A BRIEF GREEK SYNTAX. 
 
 Language is a sort of drama, in which, as in the older 
 tragedies, there are only three characters (πρόσωπα),* who 
 have different roles to play. 
 
 These three characters are : 
 
 1. The speaker, εγώ 7". 
 
 2. The person to whom I speak, συ thou. 
 
 3. The person about whom the conversation is occupied, Ϊ 
 Ae; for which the Greeks have no precise or definite form, 
 but use demonstratives, ούτος, εκείνος, αυτός, 61ε, as will be 
 seen immediately. 
 
 47. The noun names, and specifies exactly, as Cassar, Lu- 
 cullus, the king, &c; the pronoun only indicates the part 
 which the speaker plays in the dialogue, and is therefore not 
 merely in the place of the noun. ' I ' may be any one in the 
 world, from the king to the peasant, but necessarily implies 
 some one who is speaking of himself; * thou ' may be any one, 
 but must mean the person addressed ; ' he ' may be any one, 
 from Adam to the child of yesterday, but must imply the 
 person spoken of. 
 
 48. ' I ' and ' thou ' are declinable in Greek, but have no 
 gender. The third person is expressed by various words 
 which are not only declinable, but also (as in English) express 
 gender, as αυτός ipse, οντος hie, 61ε hicce, εκείνος iste, ille. 
 
 49. The reason of this is that Ί ' ' thou ' suppose two inter- 
 locutors who are present, and who therefore need no further 
 specification, their gender being regarded as obvious; one 
 word, without gender, suffices for each. But the third person 
 is or may be absent, so that for clearness the gender must be 
 indicated (he, she, it) ; and this person may be more or less 
 near, as oh hicce, the person here, questo (pointing to him, 
 ΰεικτικώς) ; or close by me, cotesto (υυτος hie) ; or there, by you, 
 quello, εκείνος (ille, iste). 
 
 50. Greek however is far from being the only language 
 which has no distinct and separate form for the third personal 
 pronoun. Some languages have, for the third personal pronoun, 
 
 * πρόσωπον, persona, originally the mask worn by an actor in playing 
 his part; hence the remark of Rousseau in his cynical old age, ' Le mot 
 latin persona signifie un masque, nom tres-conrenable assurement a la 
 plupart des gens qui portent parmi nous celui de Personnes.' — Lettressur 
 la Botanique. Milton uses it in its classic sense : ' If it were an honour 
 to that person which he sustained.' — Hist of Engl. 
 1 Which was thy part, 
 And person, hadstthou known thyself aright.' — P. L. x. 155. 
 
PRONOUNS. 33 
 
 expressions which imply a person sitting, standing, lying down, 
 &c; others, as is partly the case in Greek, have pronouns 
 which represent the third person as being at nearer or further 
 distances from the speaker ; but many have not arrived so far 
 in the analysis of conceptions as to have any one word for 
 the abstract ' he.' (See W. v. Humboldt Ueber den Dualis, § 21, 
 and Ueber die Verwandtschaft der Ortsadverbien mit dem 
 Pronomen in einigen Sprachen.) 
 
 51. The uses of ου, which is given in grammars as the third 
 personal pronoun in Greek, are very liable to lead to confusion : 
 first of all it is defective, having lost its nominative ; and 
 secondly, in Attic Greek (though not in Ionic) it is not a 
 personal, but mainly a reflexive pronoun. 
 
 52. A reflexive pronoun is one which refers bach to the 
 subject of the sentence, or one which expresses that the object 
 of the sentence (i.e. the person spoken of) is also the subject 
 (or the person speaking) ; as έτυψα εμαυτόν, I struck my self; 
 εδίδασκεν τον εαυτόν παϊδα, he was teaching his own son. 
 
 53. The reflexive pronouns are ου of himself* εμαυτοϋ of 
 myself σεαυτον of thyself εαντοϋ of himself f It will be ob- 
 served that they have no nominatives. Why ? For the 
 obvious reason that in strict grammar they never serve as the 
 subject of a principal sentence, but as the complement to some 
 other word ; i.e. they are used when the subject of the verb 
 is also its object, as / strike myself Such a sentence as εγω 
 αυτός έπραξα τοϋτο is not strictly reflexive. The reason why 
 ου once had a nominative is because it was a demonstrative 
 pronoun ; but when its reflexive use prevailed the nom. be- 
 came obsolete.^; Similarly we have lost the custom of using 
 himself, myself as nominatives in English. 
 
 54. In Attic Greek, then, what is placed as the third personal 
 
 * The plurals of έμαυτοΰ, σταυτου, are τιμών αυτών, υμών αυτών, of 
 iavrov either kaυτώv, or σφών αυτών. 
 
 f The French language uses meme to form a reflexive for the first and 
 second personal pronouns ; as, Je me suis blesse moi-meme. Other lan- 
 guages use a periphrasis for this purpose ; e.g. in Hebrew and Arabic it 
 would be ' I have wounded my soul,' &c. Silvestre de Sacy, Gram. Gen. 
 p. 51. The simple pronouns are sometimes in poetry used reflexively in 
 English, as ' He sat him down at a pillar's base.' — Byron. ' I will lay 
 me down and sleep.' ' I gat me to my Lord right humbly.' ' But go, 
 shewe thee to prestis.' — "Wiclif s Bible. 
 
 \ "We have traces of the obsolete nominative ts or ί in ϊνα, Lat. is, 
 Engl, it ; and also in μίν, and viv ; a dative and accusative ίν are found 
 in fragments, 1, himself ox herself, is only found in objective sentences, 
 
 c3 
 
34 A BRIEF GREEK SYNTAX. 
 
 pronoun is not a personal pronoun at all, but reflexive ; and 
 as its nominative t is obsolete, it borrows αυτός instead ; thus : 
 
 αυτός, ή, ό, himself, herself, itself (obsolete t) ; 
 
 ου of himself, &c. ; 
 
 ol to himself, &c. (ol enclitic = to him) ; 
 
 ε himself, &c. ; 
 
 and so on, reflexively throughout ; but εαυτόν is used more 
 frequently than ε, as άπέκτεινεν εαυτόν, he killed himself. 
 
 55. As for the third personal pronoun, there is none in the 
 nominative, in Attic, but the demonstratives are used instead ; 
 but for the other cases, the oblique cases of αυτός (derived by 
 some from αυ roc again he ?) are used, so that we have really : 
 
 Norn, ούτος, εκείνος, oh used for ' he.' 
 
 Gen. αυτοϋ of him. 
 
 Dat. αυτω to him. 
 
 Ace. αυτόν or vtv him, &c. 
 
 56. For 'him,' 'her,' * it,' μ\ν is used in Ionic; in the 
 Tragcedians viv, and σψέ ; σφέ, and, sometimes, though rarely, 
 v\v are also used for αυτούς αυτας αυτά. The root σφέ, Doric 
 ψε, is seen in the Latin ipse. 
 
 Possessive Pronouns. 
 
 57. In most languages the possessive pronoun is either 
 directly formed from, or closely allied to, the genitive case of 
 the personal.* 
 
 58. Greek is richer than Latin in possessive pronouns 
 (κτητικαϊ αντωνυμίαι). Besides εμός mens, σος tuus, ημέτερος 
 noster, υμέτερος vester, it possesses σψος his, her, their, σφέτερος 
 their, and in Ionic νωίτερος belonging to us two, σφωίτερος be- 
 longing to you two. The Latin has no simple possessive adj. 
 of the third person (his, her, its, their), for suns is reflexive ; it 
 uses instead ejus, illorum, &c. (It is remarkable that the neuter 
 possessive pronoun of the third person ' its ' is quite modern 
 in English also, see Lev. xxv. 5, ed. 1611.) 
 
 as in a fragment of Sophocles, preserved by Apollonius Dyscolus (Be 
 Pronom. p. 70) : 
 
 7} μ\ν ws i βάσσον, η δ' ws ? τίκοι 
 ποοΒ' 
 ' One woman said that she (herself), the other that she (herself), bore the 
 swifter son.' ol•, oT, e are both demonstrative and reflexive in Ionic 
 and Epic. For the authorities on ί see Donaldson, New Crat. § 139. 
 * See G-arnett, Philol. Ess. p. 260. 
 
NUMERALS. 35 
 
 Αυτός. 
 
 59. i. Observe that αυτός means ipse, -self (reflexive) ; * 
 but αυτοΰ of him, αύ-ω to him, &c. {demonstrative'). 
 
 ii. 6 αυτός means ' the same.' 
 
 iii. Although αυτό is the neut. of αυτός, yet for ' the same ' 
 in the neuter, the Attic form is generally ταυτόν not ταϋτό. 
 
 "Οστις. 
 
 60. όστις, quisquis, is a compound of the relative and the 
 indefinite. Its declension in Attic is ότου, οτω, οτων, οτοις. 
 In the neut. plur. arret is the contraction of άτινα, and must 
 not be confused with άττα, which is used in Attic for the 
 neut. plur. τίνα quondam. 
 
 61. There is no relative pronoun (αναφορική αντωνυμία) in 
 Homer, for ος, ή, 6 in Homer is demonstrative ; to form a 
 relative he adds τ ε to ος, so that ' and he' is equivalent to 
 ' who' (qui=et is). Similarly in Hebrew ΠΪ. ' this,' is some- 
 times a relative (Ps. lxxiv. 2, &c), and in German l dew' 
 
 NUMERALS. 
 
 Cardinals. 
 
 62. i. Cardinals answer the question 'how many?' The 
 word is derived from cardo a hinge. 
 
 ii. The first four cardinals only are declinable, from their 
 being the most frequently used ; but after 200 they are 
 regular adjectives of three terminations, as διακόσιοι, αι, a. 
 
 Obs. 18 and 19 may be expressed either by όκτωκαίδεκα, 
 εννεακα'ιδεκα, or by δυοΊν, ενός δέοντες ε'ικοσιν. Similarly 28, 
 29 may be δυο'ιν, ενός δέοντες τριάκοντα, &c. ; and even 7000, 
 8000 may be τριακοσίων, διακοσίων δέοντα μύρια (Thuc. ii. 13). 
 This resembles the Latin duodeviginti, undeviginti, &c, and 
 our way of reckoning time (e.g. a quarter to eight=forty-uve 
 minutes past seven). 
 
 iii. 21, 22, &c, may be either ε'ικοσιν εις, είκοσι δύο or εΤς, 
 δύο και ε'ικοσιν just as in English it may be twenty-one, or 
 one and twenty ; the rule being that if the smaller number 
 precedes, the copula must be used. 
 
 iv. Distinguish between μύριοι 10,000, and μύριοι inde- 
 finitely numerous ; the regular number has the regular accent. 
 
 * Thus we have in Shakspeare, ' Myself have letters.' — Jul. Cces. 
 iv. 3. 'Were you sick, ourselves would wait upon you.' — Tennyson, 
 The Princess. But for obvious reasons the nominatives of reflexive pro- 
 nouns do not hold their grounds. See § 53. 
 
36 A BRIEF GREEK SYNTAX. 
 
 Ordinals. 
 
 63. i. Ordinals express the position or order ; and answer 
 the question ' which of the number ? ' 
 
 ii. Except δεύτερος, which has the form of the comparative, 
 they all take the superlative termination τος. They are all 
 declinable adjectives of three terminations. 
 
 iii. The student should distinguish carefully between the 
 decads and the hundreds; 30th, 4Cth, &c, are τριακοστός, 
 τεσσαρακοστός, &α; but 300th, 400th, &c, are τριακοσιοστός, 
 τεσσαρακοσιοστός, &C. 
 
 iv. 21st, 22nd, &c, may be expressed in three ways, viz. : 
 είς κάϊ εικοστός, πρώτος και εικοστός, or είκοστός πρώτος ; simi- 
 larly 32nd, &ο.=δύο και τριακοστός, δεύτερος και τριακοστός, or 
 τριακοστός δεύτερος ; and so on. 
 
 Other Numerals. 
 
 64. Both Greek and Latin are particularly rich in their 
 forms for numerals ; e.g. 
 
 Multiplicatives. απλούς, διπλούς, τριπλούς, κ.τ.λ. simplex, 
 duplex, &c, from which are derived our English 
 multiplicatives simple, double, triple, &c, referring 
 to size. 
 
 Proportionals, διπλάσιος, τριπλάσιος, κ.τ.λ. duplus, tri- 
 plus, &c, our twofold, threefold, &c, referring to 
 number.* 
 
 Numeral Adverbs, δίχα, τρίχα, τέτραχα, κ.τ.λ. in two, 
 three, four ways, &c, answering to multiplicatives. 
 άπα'ξ, δίς, τρις, κ.τ.λ. once, twice, thrice, &c, an- 
 swering to proportionals. 
 
 We have also δευτεραϊος, τριταίος, τεταρταΊος, κ.τ.λ. on the 
 2nd, 3rd, 4th day, &c. ; ποσταΊος ; on what day ? These are 
 only adjectival forms of the dative feminine of δεύτερος, κ.τ.λ. 
 
 ADVERBS (Επιρρήματα). 
 
 65. ' When some case of a declinable word — whether sub- 
 stantive, adjective, or pronoun — has fixed itself absolutely for 
 the expression of certain secondary predications, it is called 
 an adverb. The prepositions, conjunctions, and interjections, 
 which are generally considered as distinct parts of speech, are, 
 
 * This distinction is due to Ammonius (de Diff. p. 43), UnrXovs κατά 
 /teyeflos, διπλάσιος κατ' αριθμόν. (Donaldson.) 
 
VERBS. 37 
 
 in regard to their origin and primitive use, neither more nor 
 less than adverbs. Their right to a separate place in the 
 grammar of an inflected language depends on their syntactical 
 functions only. The preposition is an adverb of place, .... 
 the conjunction an adverb of manner, . .-. the interjection an 
 exclamatory adverb.' — Donaldson, Greek Gram. -p. 148. Hence, 
 in spite of Home Tooke's sneer, ' the old grammarian was 
 right, who said that when we know not what else to call a 
 part of speech, we may safely call it an adverb.' 
 
 66. Almost every adjective, and many participles, furnish 
 an adverb in -ως, a termination derived from the old ablative 
 case. The neuter accusative of adjectives both singular and 
 plural is often used adverbially. Adverbs derived from 
 adjectives are compared by taking the neuter sing, of the 
 adjective for the comparative, and the neuter plur. for the 
 superlative, as ή£έως, ήέιον, ήΐιστα. 
 
 67. Other adverbs coincide with the actual cases of nouns, 
 as κομιΰϊί exactly, δημοσία publicly, Ilia privately, tcotvrj in 
 common, στουορ zealously, σχολ /j leisurely* (i.e. vix, scarcely); 
 αρχήν at^/?rs£, ουκ αρχήν not at all (omnino non), άκμήν just, 
 or hardly, Ιωρεάν, πμοϊκα gratis, μακράν afar. 
 
 68. Others consist of a preposition and noun, as παραχρήμα 
 immediately, »ca6a7rf^justas, πρυνργου advantageously, εκποδών 
 out of the way, ενσχερω in order, &c. 
 
 N.B. i. Observe that ευΗύς is ' immediately,' and ευθύ (with 
 the gen.) ' straight towards.' Similarly άντϊκ•ρνς= outright, 
 ai'rijC|«i)=opposite. 
 
 ii. The ω-ς of Greek adverbs is the Sanskrit a-t (cf. ΰίόωσι 
 didati) ; thus όμώ-ς=ύιβ Sanskrit sama-t ' simili ;' t is the 
 case-ending of the Sanskrit ablative, and in some Greek 
 adverbs it is suppressed (e.g. ούτω), in others it becomes ς. 
 Compare the Latin adverbial ablatives raro, perpetuo, quomodo, 
 &c. For the proofs of this identification see Bopp, § 183. 
 
 VERBS (Ρήματα). 
 
 69. The nature of the verbf (ρήμα verbum, i.e. the word 
 par excellence) has been variously defined by different gram- 
 marians. All acknowledge its importance ; i Alteram est quod 
 loquimur,' says Quintilian, ' alteram de quo loquimur.' 
 
 * Compare Shakspeare's 'I'll trust by leisure him that mocks me 
 once.' Cf. Soph. Ο. T. 434. 
 
 f See Burggraff, Principes de Gram. gen. p. 345-349 ; Origin of 
 Language, p. 104 ; Du Meril, p. 56, 
 
38 " A BRIEF GREEK SYNTAX. 
 
 1. According to most ancient grammarians its distinctive 
 peculiarity is the expression of Time (ρήμα ΰ?εστι το προσ- 
 σημαϊνον χρόνον, Arist. De Interp. iii. 1). Hence the Ger- 
 mans call it Zeitwort time-word, and the Chinese ho-tseu 
 living word (just as Plato calls the verb and noun τα έμψυχο- 
 τατα μέρη του λόγου). But verbs which should express no 
 circumstance of time are quite conceivable, and actually exist 
 in some North American languages. 
 
 2. Others say that it necessarily expresses an Action, and 
 hence some Germans call it Thatigkeitsivort. Thus in Chinese 
 a hand added to a hieroglyphic shows that a verb is intended ; 
 for instance, a bent bow and a hand signify ■ to shoot an 
 arrow.' In Chinese also ' to be ' is ' to make ' (wet). Ob- 
 viously however many verbs imply inaction rather than 
 action. 
 
 3. In the Grammaire Generate of Port-Boy al the verb is 
 defined as l un mot qui signifie V affirmation J and this definition 
 may stand if we make affirmation include negation. 
 
 4. Humboldt and others say that the verb must involve the 
 abstract conception of existence, and so furnish the connection 
 between the subject and the attribute (die reine Synthesis des 
 Seins mit dem Begriff). This is only true if with Harris we 
 resolve every verb into a participle with the verb ' to be,' so 
 that, e.g. γράφω=έγώ (ειμί) γράφων. No analysis of the verb 
 however can succeed in reducing it into a participle coupled 
 with the verb to be. What is there participial in the root 
 γραφ ? ' A verb divested of its paraphernalia may become an 
 Irish participle, which is merely an abstract noun, but cer- 
 tainly not a Greek, Latin, or even an English one.' 
 
 5. Mr. Garnett, following out a hint in Dr. Prichard on the 
 Celtic language, first showed that verbs do not differ from 
 nouns by any inherent vitality ; they are simply nouns with a 
 pronominal affix. ' Motion or action is no more inherent in a 
 verbal root than the power of forging a horseshoe in a smith's 
 hammer. It requires an extensive moving power to make it 
 efficient, and so do the roots of verbs.' Their power of ex- 
 pressing action, motion, sensation, or their opposites, resides 
 only in the addition to them of the person or agent. In other 
 words, a verb is ex necessario a complex, and not a simple 
 term, and as such it could not have been a primary part of 
 speech. 
 
 70. A comparison of the English and Greek verb shows the 
 immense difference between an analytic and a synthetic Ian- 
 
 
PERSON-ENDINGS. 39 
 
 guage. The English verb has five forms (e.g. love, lovest, 
 loves, loved, loving) ; the Greek verb has about 1,200 forms. 
 
 71. The inflections by which a verb expresses its various 
 modifications are called its conjugation (συζυγία). 
 
 72. The endings or inflections by which the Greek expresses 
 the three persons in the singular are really the three per- 
 sonal pronouns (1, thou, he), although all trace of this fact has 
 been nearly obliterated in the course of time. Thus to take a 
 verb in -μι (those verbs being the oldest, and therefore the 
 least disguised in their person- endings), it is easy to see that 
 in ει-μι, εσ-σί, εσ-τί(ν), μι is connected with the stem με, -σι 
 with σε, and π with the article* τό. The passive terminations 
 -μαι, -σαι, and -rat show the same fact no less distinctly. The 
 termination ω looks as if it were connected with εγώ, JEolic 
 i&v ', but it is certain that the person-ending comes not from 
 the nominative but from objective cases of the pronouns, so that 
 ΰίδωμι^ would mean ' giving here, i.e. my giving,' and £t<Wt 
 ' giving there, i.e. his giving.' It is the object of Comparative 
 Grammar to analyse all inflections in a similar way, and to 
 show their original significance. At present however the 
 results are not all certain, and the explanation of them would 
 require a separate treatise, because each termination has to be 
 traced through a long series of phonetic changes ; and in 
 Sanskrit and Greek especially ' a vast number of articulations 
 have been sacrificed to euphony, the restoration of which is 
 often conjectural, and sometimes impracticable.' 
 
 * We shall see in the Syntax the close connection between the article 
 and the third personal pronoun. It is the same in German, where the 
 definite article der, die, das is constantly used as a pronoun ; and the 
 French article le is derived from ille, as is the Italian il, lo, and the 
 Spanish lo, la. In the third person plural the termination is due to 
 phonetic change ; e.g. τύπτονσι = τύπτοντι = yevherant. In "Welsh (which 
 is an Aryan language) the pronoun of the third person plural actually 
 ends in nt, uynt or htiint = they (cf. Introd. § 15, 5, p. 5). 
 
 f Only two Latin verbs, inquara and sum, retain a trace of the old 
 termination in μι. The first philologer to point out that the person-end- 
 ings were pronouns in oblique cases was Mr. Garnett, and he illustrated 
 the fact from Syriac, in which ith = existence, ithai-ch existences of thee 
 = thou art, ithai'-hun existences of them = they are. The same result 
 becomes very clear from a comparison of the Hungarian olvas-owi I read, 
 olvas -θίΖ thou readest, 6lva.s-ato/c ye read, &c, with olma-m my apple, 
 olma -ί thy apple, olma-tok your apple. See Garnett's Philolog. Essays, 
 p. 291 ; Dr. Latham, Led. on the Study of Language. Obviously, as 
 Bopp observes, the moment that language began to mark persons by 
 the addition of suffixes to the verb, those suffixes could not have been 
 anything but personal pronouns. 
 
40 A BRIEF GREEK SYNTAX. 
 
 73. Many grammars throw no light whatever on the ordinary 
 omission of a first person dual in the active. Thus we find for 
 the dual of the pres. act. 
 
 but for the dual of the pres. pass. 
 
 τνπτόμεθον, τύπτεσβον, τύπτεσθον, 
 
 with no explanation of the reason why we should have no 
 form for ' we two are striking,' and yet should have one for 
 1 we two are being struck.' The reason is that in the act. the 
 first pers. plural is always used for the first person dual. We 
 can only conjecture why no distinct form wa3 retained, or 
 why in the passive the aorist alone should have no first person 
 dual. 
 
 74. There is an ingenious theory on the subject of the dual 
 in the article ' Dual ' in the ' Penny Cyclopaedia.' Believing 
 that the dual is an older plural which was only colloquially 
 retained, the author points out how easily a termination in 
 ν might have been changed into one in c (compare τΰπτομεν 
 and τύπτομες verberamws ; shoon and shoes, eyne and eyes, 
 housen and houses, &c), and how easily this ς might be 
 dropped ; on this theory τυπτετον and τνπτετε, &c. might also 
 very easily have been phonetic varieties of the same form. 
 
 75. In many grammars both the second and third pers. dual 
 of the historical tenses (imperf., plupf, and aorists) are made 
 to end in ην : as in the impf. act. of τύπτω 
 
 ετνπτέτην, ετυπτέτην ; 
 
 but in other modern grammars (and even in that of E. Bur- 
 nouf) the second person dual even in historical tenses is made 
 to end in ov, so that we find 
 
 ετντττετον, ετνττίτην • 
 
 this latter is the more correct, for the Attics always prefer the 
 form in ov for the second person of the dual, if we may trust 
 the best MSS: 
 
 VOICES (Διαθέσεις). 
 
 76. 'The Greek verb has three voices — 
 
 1. Active (διάθεσις ενεργητική),* as τύπτω I am striking, 
 
 * The Stoics called the Active κατηγόρημα bpBhv ' upright' the Passive 
 vrniov ' supinum,' and the Neuter obhtTcpov 'neither of the two' Dionysius 
 
REFLEXIVE FORMS. 41 
 
 This may be either transitive (άλλοπαθης), i.e. the action 
 may pass on to some object, as Ζίδωμι αρτον I am giving bread. 
 
 Or intransitive (αΰτοπαθης), i.e. the action may stop with 
 the agent, as τρέχω I run. These verbs are also called neuter. 
 
 2. Passive (παθητική), as τΰπτομαι I am being stmck. 
 
 3. Middle (μέση), as τϋπτομαι I am striking myself. 
 
 In Sanskrit the Active Voice is called Parasmai-pada i fall- 
 ing on another ;' the Middle Atmane-pada ' self-affecting.' 
 
 77. The only tenses for which the Middle has any special 
 forms are the future and aorist.* What are usually called the 
 perf. and plupf. middle are not middle forms at all, but are 
 other forms of the perf. and plupf. act. The name perfect 
 middle for such forms as τετυπα ought to be finally discarded ; 
 the error of calling them so, rose from the instances in which 
 this second perfect has an intransitive meaning, as εγρήγορα 
 I am awake, πέποιθα I am confident, έ'αγα / am broken, 
 πέπηγα I stick fast, ερρωγα I burst forth, &c. But this is a 
 mere speciality of meaning. 
 
 78. Verbs which have an active meaning, but only a passive 
 or middle form, are called deponents (from depono I lay 
 aside). It is probable however that they have not laid aside 
 the active form, but never had one at all ; it is generally be- 
 lieved that the -μαι form of verbs is the oldest of all. For it 
 was most natural that verbs should be primarily regarded as 
 middle, i.e. as expressing direct reference to the subject (or 
 self). Hence the μαι forms often exist in Homer side by side 
 with the forms in ω. Reflexive forms are far more common 
 in other languages (e.g. French, Italian, German) than they 
 are in English. That the transitive form and meaning of 
 verbs was due to a later development of language is clear, 
 since, a3 we have seen, the cases represent adverbial additions 
 
 Thrax (p. 886) says that the two former names were suggested by a 
 metaphor from the position of athletes. On the derivation of the Latin 
 word ' supine,' Priscian remarks, ' Supina vero nominantur, quia a pas- 
 sivis participiis, quse quidem supina nominantur, nascuntur' (p. 811). 
 Lersch, Sprachphil. d. Alien, ii. 197 ; Burggraff, p. 357. 
 
 * This is just what we should expect from the close connection 
 between the passive and middle, of which the middle or reflexive form 
 was probably the earliest. "We have very few reflexive forms (I bethink 
 me, fear me, &c.) in English, but we represent many of the German, 
 Italian, and French reflexive verbs by passive or neuter verbs ; e.g. Ich 
 freue mich I rejoice ; si dice it is said ; se emplearon diez hombres ten 
 men were employed, &c. The gradual evanescence of the middle in 
 Greek is analogous to the disuse of many old reflexive verbs in French, 
 such as se monnr, se partir, &c. Pellissier, L,a Languefrang. p. 177. 
 
42 A BRIEF GREEK SYNTAX. 
 
 to the noun, and would therefore be originally independent of 
 all verbal government, so that it would have been needless for 
 the verb to have a transitive sense. Hence we find many 
 Greek verbs that fluctuate between a transitive and intransitive 
 meaning, as εχω ' I have ' and i I am,' άγω ' I lead ' and ' I 
 move,' αίρω ' I raise ' and ' I rise ' (e.g. of the sun, Soph. 
 Phil. 1315), εΧαΰνω Ί drive' and Ί ride,' πράσσω 'Ido' 
 and ' I fere.' The same is true in other languages ; e.g. in 
 Latin, vertere, mutare, &c; in German, Ziehen, brechen, 
 schmelzen, &c. ; in French, decliner, change?-, sortir, &c. ; in 
 English, to move, break, turn, &c. (Jelf, § 360). 
 
 EEDUPLICATION (^ΑναΙίπΧωσις). 
 
 79. i. Reduplication, i.e. a repetition of the root twice over, 
 was a very primitive process, found in all languages, and 
 adopted as the simplest known method of strengthening the 
 meaning of the word to which it is applied. 
 
 ii. Thus it is found in substantives both in Greek and Latin, 
 as βάρβαρος, παιπάΧη, βόμβος, marmor, murmur, turtur, 
 papilio, &c. 
 
 iii. And in verbs both in Greek and Latin, πέπηγα, λέλνκα, 
 &c, pepigi, tutudi, cucurri, tetigi, nemini, &c. 
 
 iv. It is by no means confined to the perfect and pluperfect. 
 Distinct traces of it appear in many presents, as μίμνω, πίπτω, 
 γιγνώσκω ; especially in the older verbs, viz. those in μι, as 
 ΰίΰωμι, τίθημι, (σ)'ίστημι, πίμπλημι, πίμπρημι, όνίνημι, sisto, 
 gigno, pipilo, titubo, &c; and in the paulo-post-futurum, 
 as τετύψομαι, ΧεΧΰσομαι, &C. 
 
 v. It is also frequently found in the aorist, as ήγαγον, 
 ήραρον. In Homer these reduplicated second aorists abound, 
 as πέπιθον, κέκΧυθι, άμπεπαΧώΐ', τεταρπόμην, ΧέΧαθον, πεώραΐον. 
 It will be seen that it always emphasises * the meaning of the 
 verb, and is therefore peculiarly adapted to represent repeated 
 or continued actions, such as vibration (άμπεπαΧων), thought 
 (πεφραΐον), careful attention {κεκΧνθι), scolding (ήνίπαπον), 
 &c. 
 
 vi. It is natural therefore that it should be mainly charac- 
 
 * Precisely on the same principle as in Hebrew, in Armorican, in 
 Hindoo, and in Modern Greek, an adjective is repeated to represent the 
 superlative, as E^Tf? ^«P ^oly °^ holies = holiest; μία ψηλή ψηλή 
 κρεμάθρα a very high gallows. The process is constantly resorted to in 
 common conversation, and is a regular idiom of Italian, e.g. 'Ella sen 
 va notando lenta Icnta,' Dante, = very slowly, &c, 
 
REDUPLICATION. 43 
 
 teristic of the primary tenses, and especially of the perfect. 
 (Besides such perfects as momordi in Latin, we find traces of 
 reduplication in many others, as feci (=fe-fici), jeci (je-jici), 
 veni (ve-veni), and many more.) 
 
 vii. Unlike the augment, which is a mere prasfix or ex- 
 traneous adjunct, the reduplication is regarded as an organic 
 part of the word, and therefore is retained through ail the 
 moods, while the augment is found in the indicative alone. 
 
 Chief Kules of Eeduplication. 
 
 80. 1. Words beginning with p, with yv, with double letters 
 ζ, ζ, ψ, with two mutes,* or with vowels, cannot take redupli- 
 cation, but substitute the augment for it. This is only for 
 the sake of euphony ; ρεηριφα, ψέψαλκα, &c, would sound in- 
 tolerable, and therefore ερριφα, εφαλκα, &c. are used instead. 
 
 2. Verbs beginning with an aspirate, use the tenuis in 
 reduplication, as τέθυκα, πεφίληκα. 
 
 3. Three verbs take εΐ instead of the reduplication, viz. : — 
 
 λαμβάνω, ε'ίληφα. 
 Χαγγάνω, ε'ίληχα. 
 μείρω, ε'ίμαρμαι. 
 
 We have also έίρηκα used as the perfect of ώημί. λέγω makes 
 both Χελεγμαι and ε'ίλεγμαι in composition. 
 
 4. Some verbs, beginning with a vowel, take what is called 
 the Attic reduplication, as 
 
 αγειρω, 
 
 αγηγερκα, 
 
 αγηγερμαι. 
 
 ακούω, 
 
 άκηκοα. 
 
 
 εγείρω, 
 
 εγηγερκα, 
 
 εγίιγερμαι. 
 
 εσθίω, 
 
 εΰήΰοκα, 
 
 εΰηΰεσμαι. 
 
 ελαυνω, 
 
 ελήλακα, 
 
 ελήλαμαι. 
 
 ερείΰω, 
 
 ερήρεικα, 
 
 ερήρεισμαι. 
 
 ομιυμι, 
 
 όμώμοκα, 
 
 όμώμοσμαι. 
 
 όρνσσω, 
 
 όρώρνχα, 
 
 όρώρυγμαι. 
 
 We also have ελήλνθα, ενήνοχα used as perfects of έρχομαι, 
 φέρω. 
 
 5. Verbs in ω with a reduplicated present, as βιβρώσκω, 
 γιγνώσκω, ΰιΰράσκω, μιμνήσκω, πιπράσκω, τιτρωσκω, drop the 
 reduplication in other tenses ; hence their futures are βρωσομαι, 
 γνωσομαι, ΰράσομαι, μνησω, &C.f 
 
 * Except κτάομαι, κίκτ-ημαι, μνάομαι, μέμνημαι, 
 f But διδάσκω fut. διδάξω, βιβάζω fut. βιβό,σω, 
 
44 A BRIEF GREEK SYNTAX. 
 
 AUGMENT (Αυί,ησις). 
 
 81. The Augment entirely differs from the Reduplication, 
 both in meaning and usage. 
 
 a. It is probably a fragment of the root which we also find 
 in ανά, signifying remoteness, and merely refers an action to 
 the past. It was originally ' a demonstrative particle, primarily 
 expressing remote place, and secondarily remote time',''* and 
 was no original part of the verbal root. 
 
 β. It properly belongs only to the historical tenses. 
 
 γ. It is dropped in all moods but the indicative, except 
 where it is used instead of reduplication. This is a trace of 
 its independent existence as having once been a separate 
 word. In the older Sanskrit, for instance, it is separable from 
 the verb, and (as in Homer) it may be omitted at pleasure."]• 
 This helps to account for the fact that Latin has lost all traces 
 of a syllabic augment. 
 
 82. Augment is of two kinds ; syllabic (συλλαβική), which 
 adds the syllable ε, and temporal (χροπκή), which only in- 
 creases the length of a vowel. 
 
 The chief peculiarities in augments are as follows : 
 
 1. In later Attic βούλομαι, δύνομαι, and (sometimes) μέλλω 
 make ήβουλόμην, ηΐυνάμην, η μέλλον. 
 
 2. The diphthongs ά and ου are not augmented ; J the other 
 diphthongs are augmented by giving the augment to the first 
 vowel of the diphthong, and subscribing the second if it be ι, 
 as α'ιρέω, γρουν, αυξάνω, ηυζανον. 
 
 3. Ten verbs beginning with ε take the augment ει. The 
 commonest of them are : 
 
 Ιάω I pe?init, ε'ίων. 
 
 ελίσσω I roll, ε'ίλισσον. 
 
 ελκύω I drag, ε'ίλκυον. 
 
 έπομαι I follow } ε'ιπόμην, 2nd aor. εσπόμην. 
 
 εργάζομαι I work, εϊργαζόμην. 
 
 έρπω I creep, εφπον. 
 
 εχω I have, είχον, 2nd aor. εσχον. 
 
 We have also εϊπον, and ειλον. 
 
 * Garnctt's Philolog. Essays, p. 206. He adduces analogous forms 
 from many other languages. Buttmann's conjecture that it is a mutila- 
 tion of the reduplicate prefix, and Bopp's that it is a relic of the nega- 
 tive prefix, are justly exploded. 
 
 f Max Muller, Sanskr. Gram. p. 144. 
 
 \ It is now generally belieyed that the diphthong ci can be aug- 
 mented. 
 
MOODS. 45 
 
 4. ρ is doubled after an augment, as ρίπτω, ερριπτον. 
 
 5. A few verbs take both the temporal and syllabic aug- 
 ment, as 
 
 οράω impf. ε ώρω ν pf. εωρακα 
 ανοίγω „ άνέωγον „ άνέωγα 
 
 οινοχέω „ έωνοχόονν. 
 
 Notice the pluperfects i&cetvl seemed, icf\veivl hoped, *4p7* lv I did. 
 
 6. In synthetic compounds, i.e. compounds where the two 
 parts are not separable, but are so fused together that they 
 cannot exist as two separate words, the augment is placed at 
 the beginning of the word, as in οικοίομεω ωκοΐόμησα, κάθημαι 
 εκαθημην, έπίσταμαι ήπιστάμην. 
 
 But where the compound is parathetic, i.e. where the two 
 parts are separable, and are merely juxtaposed, the augment 
 is put between them, as in προσφέρω, προσέψερον ; and this is 
 the case in most verbs compounded with prepositions. 
 
 7. The augment, which is constantly omitted in Homer, is 
 never omitted in Attic except in χρήι> for έχρήν* But there 
 are a few words, ' quibus augmentum non proponunt tragicif 
 e.g. άνωγα, καθεζόμην, καθήμην. Porson Pro?f. ad Hec. xvi. 
 (He adds καβεϋοον, but see Veitch, Greek Verbs, p. 300.) 
 
 MOODS (Εγκλίσεις). 
 
 83. The moods (modi) in Greek are : 1. The Indicative 
 (οριστική εγκλισις). 2. The Subjunctive (υποτακτική). 3. The 
 Optative (ευκτική). 4. The Imperative (προστακτική). Besides 
 these, there are: 5. the Infinitive (απαρέμφατος); and 6. the 
 Participle (μέτοχος) ; but the two latter, including the verbal 
 adjective in -τέος, are by modern grammarians usually treated 
 as verbal nouns, and not as moods. 
 
 Protagoras is said to have been the first to distinguish the 
 different moods of verbs.f 
 
 The first four of these moods are called personal, the latter 
 impersonal, as having less formal reference to a subject. 
 
 The nomenclature of the moods is far from perfect. ' The 
 indicative, i.e. mood of declaration, is continually used where 
 
 * Exclusive of prodelisions like those in (Ed. T. 1602, 1608, Hec. 387, 
 there are only a few instances of an omission of the augment in tragedy 
 at the beginning of lines in the speeches of messengers. And the 
 augment is sometimes omitted in the pluperfect — usually so in the New 
 Testament. See Winer's Gram. § xiii. 8. 
 
 f See the authorities quoted in Donaldson, Neiv Crat. p. 204, 2nd ed. 
 
46 A BRIEF GREEK SYNTAX. 
 
 no declaration is made, — in interrogatives for example, and in 
 conditionals. The optative has very many uses with which 
 the expression of a wish has no concern, and has moreover 
 quite as good a claim to the title of subjunctive.' (Harper.) 
 
 CLASSES OF VERBS. 
 Verbs in -μι. 
 
 84. There are two main classes of verbs, those in ω, and 
 those in μι. 
 
 The former (verbs in ω) are far the most numerous; 
 the latter are the oldest. That this is the case appears, 
 because : 
 
 1. The pronouns which formed all person-endings are least 
 obliterated, and most easily recognisable in verbs in μι (see 
 ante § 72) ; and besides, these person-endings are attached 
 directly to the stem, as εσ-μεν, ΰί-ΰο-τε, whereas the verbs in 
 ω require a connecting vowel, as λν-ο-μεν, τιμά-ο-μεν. 
 
 2. The verbs in μι contain the simplest roots, and involve 
 the most elementary notions, as ' being,' ' going,' ' giving,' 
 ' saying,' placing,' &c. 
 
 3. This form in μι is predominant in Sanskrit, and the 
 oldest languages of the Indo-European family.* 
 
 85. Observe that: a. This form of conjugation is only 
 found in a few tenses, — chiefly in the present, impf., and 2nd 
 aor.; but 
 
 β. Traces of a similar form of conjugation appear, especially 
 in the 2nd aorists, in many other verbs, as εβην I went, εΖραν 
 Iran, ετλην I endured, 'έφθην I anticipated, σχες hold ! the 
 imperative of εσχον, εάλων I was caught, h /νων I knew, the 
 imperative πΊθι drink, and others. 
 
 y. In Latin we find traces of it in inquam, suw?, and in the 
 endings of the 3rd person sing, (as, stat^arart, &c), and 
 3rd pers. plur. {dant = licoy~i), &c. 
 
 N.B. i. In the imperf. τίθημι and όίΰωμι follow the analogy 
 of verbs in ω, having ετίΰυνν, ετίθεις, ετίθει, and εΰίΰονν, εΰίΰονς, 
 έΰίΰου more usually than ετίθην, ης, η, and £<$icu)y, ως, ω. 
 
 π. "στη μι varies in its tenses between a transitive and in- 
 
 * The rarity of verbs in μι is no argument against this conclusion ; 
 for, when one form has been nearly superseded by another, the feeling 
 of analogy works so powerfully in language that the few remaining 
 specimens of the old form soon disappear; 'thns in Modern Greek even 
 δίδωμι, τίθημι have given way for δίδω, θ^τώ.' 
 
FORMS OF THE FUTtTBE. 47 
 
 transitive meaning : thus ϊστημι I place, 'ίστην I was placing, 
 στήσω I will place, έστησα I placed ; έστηνα I stand, είστηκειν I 
 was standing, εστην I stood. [Similarly from the present of 
 the German verb ich stehe we get our transitive verb to stay, 
 and from the perfect ich stand our intrans. verb to stand. 
 Don.] 
 
 iii. There are 3 aorists in κα, ίθηκα I placed (pf. τέθεικα), 
 εΰωκα I gave, ηχα I sent (pf. tlm). Whether these represent 
 an older, or merely a modified form of the aorist is uncertain.* 
 It is remarkable that they are used mainly in the singular, 
 the second aor. being more common in the plural. On the 
 varying use of first and second aorists, see the admirable 
 Greek Verbs of Mr. Veitch, p. 46. 
 
 Verbs in -ω. 
 
 86. The Dorians made the fut. mid. in οϋμαι, hence the fol- 
 lowing are called Doric futures : — 
 
 πίπτω fut. πεσοϋμαι 
 κλαίω fut. κλαυσυϋμαι (or ομαι)^ 
 7τλεω fut. πΧευσονμαι (or ομαι) 
 πνέω fut. πνενσοϋμαι (or ομαι) 
 ψενγω fut. φενζοϋμαι (or ομαι). 
 
 87. Contracted futures like κόμμα from κομίζω I convey, 
 σκεΐάζω I scatter, fut. σκεΐώ, τελέω I accomplish, fut. τελώ, 
 are called Attic futures. J 
 
 88. The following futures have no tense sign : — χέω I shall 
 pour, ερώ / shall say, εΕομαι and ψάγομαι I shall eat, πίομαι I 
 shall drink, νέομαι I shall return, ει μι I ivill go (compare the 
 English 'I am going (=1 shall go) next week.' In fact the 
 verb ' go ' involves a notion of futurity, § as when we are going 
 to do a thing ; and as in ' The first said unto him, I go, Sir, 
 and ivent not.' 
 
 * In tfveyica, the borrowed aor. of φέρω, the <r has been lost ; as also 
 in elira, εχςα, eaaeva, and tceas from καίω. 
 
 f In English in the same way we often have two forms coexisting, as 
 in swelled and swoll, chided and chode, hanged and hung, rang and 
 rung, &c, but the tendency always is to give different meanings to them 
 (i.e. to desynonymise them). We are more alive to these varieties of 
 form assumed by the same tense in Greek, because we have specimens 
 of their language extending over the space of hundreds of years. 
 
 i A few rare dialectic forms like κίνσω, ττεφύρσομαι, &c, are called 
 iEolic futures. 
 
 § So in Spanish ' Nosotros nos vamos maiiana, y ellos saicn el dia 
 despues,' we go to-morrow, and they leave the next day. Delmar's 
 Span. Gram. p. 139. See too Veitch, Greek Verbs, p. 200. 
 
48 A BRIEF GREEK SYNTAX. 
 
 89. There are fourteen verbs in which the fut. mid. has a 
 passive meaning, partly for metrical reasons, partly because the 
 fut. passive was not in use ;* such are 
 
 Χέζομαι I shall be said, 
 μισησομαι, στυγησομαι I shall be hated, 
 άλώσομαι I shall be taken, 
 ϋρΐ,ομαι I shall be ruled, 
 έάσομαι I shall be suffered. 
 οικήσομαι I shall be inhabited, 
 τιμήσομαι I shall be honoured, 
 άδικησομαι I shall be injured, 
 ζημιώσομαι I shall be punished. 
 
 90. The following verbs among others (especially denoting 
 some bodily activity) use the fut. mid. in an active meaning. 
 These verbs present an analogy to such verbs as se taire, s'en 
 aller, &c, which are similarly reflective in form but not in 
 sense. 
 
 (/Ζω, η.σομαι I shall sing. 
 
 ακούω, άκοΰσομαι I shall hear. 
 
 απολαύω, άπολαύσομαι I shall enjoy. 
 
 βαίνω, βήσομαι I shall go (Je wi'en irai). 
 
 γιγνώσκω, γνωσομαι I shall know. 
 
 γελάω, γελάσομαι I shall laugh {Je me rirai de). 
 
 ΖιΙράσκω, Ιράσομαι I shall run. 
 
 θαυμάζω, θαυμάσομαι I shall wonder (Je m'etonnerai). 
 
 θηηάω, θηράσομαι I shall hunt. 
 
 κλέπτω, κλέφομαι I shall steal. 
 
 σιγάω, σιγησομαι silebo, I shall be still (Je me tairai). 
 
 σιωπάω, σιωπησομαι tacebo, I shall hold my tongue. 
 
 σπουδάζω, σπουΐάσομαι I shall be busy (Je m'etudierai a)."}* 
 
 91. The presents ήκω I have come, οίχομαι I have gone, have 
 a perfect meaning. 
 
 The perfects άνωγα Ibid, εοικα I seem, κεκτημαι I possess, 
 
 * These verbs tend to prove the theory of the original identity of the 
 passive and middle ; and the evolution of the passive out of the middle, 
 as is actually the case in the Scandinavian languages. A similar argu- 
 ment might be deduced from the fact that several aorist.s middle have a 
 passive sense, and aorists passive a middle sense, as ΙαΧίχθ-ην I conversed, 
 ■ηρνήθην I denied, &c. (Clyde's Gk. Syntax, p. 57.) In the New Tes- 
 tament, άπ€κρίθην is constantly used in the sense of απίκρινάμην. 
 
 t A list of peculiarities like these, as well as of the commonest irre- 
 gular verbs, nouns, &c, has been drawn up by the author, in a little 
 card of three pages, for the use of the Harrow School. 
 
IRREGULAR VERBS. 49 
 
 οίδα / know, novi, μέμνημαι I remember, memini, and some 
 others, have a present meaning.* 
 
 92. The four verbs ζάω I live πεινάω I hunger, διψάω I 
 thirst, χράομαι I use, contract into η not into a; thus the 
 infinitives are ζην, πεινην, ΰιψήν, γρησθαι,^ being contracted 
 from older forms of the infinitive ζάεν, πεινάεν, &c. 
 
 93. When a verb has tenses derived from several stems 
 the reason is that originally several verbs were synonymous in 
 meaning. Language at an early stage abounds in synonyms ; 
 but at a later period cannot be burdened with this superfluous 
 exuberance, and either desynonymises the words (i.e. uses 
 them to express different shades of meaning) or drops them 
 altogether. Sometimes, as in the cases before us, it retains 
 only one tense of a verb, dropping all the others. Thus the 
 verbs φέρω, φημί, τρέχω, δράω, εσθίω, &c. borrow their tenses 
 from other obsolete roots conveying a similar meaning. 
 
 94. The irregular verbs are precisely those which the 
 learner will encounter most frequently ; he can hardly read 
 any page of Greek without finding some which are of constant 
 occurrence. In truth, the irregularity of verbs is often due 
 to their antiquity, and to the fact of their expressing con- 
 ceptions so common as to be most liable to phonetic corruption 
 from the wear and tear of language. Philologically speaking, 
 too, such verbs are generally the most interesting, since their 
 very peculiarities often reveal to us secrets respecting the 
 growth and structure of language at which we might otherwise 
 guess in vain. 
 
 95. Verbs in άω, έω, ενω, ώσσω, imply to be or to have that 
 which the name signifies, as κομάω I have long hair, φιλέω I 
 am a friend, φονεύω I am a murderer, νπνώσσω I am sleep- 
 ing. 
 
 96. Causatives usually end in όω, ίζω, όζω, ννω, αίνω, as 
 δουλυω I make a slave, πολεμίζω I make war, αρμόζω I fit, 
 ηΐννω / sweeten, σημαίνω I signify, κοιλαίνω I make hollow. 
 
 * ' Rien n'est plus facile que d'expliquer cette irregularite apparente ; 
 θν^σκω je meurs, τ4θνηκα j'ai souffert la mort ; done, je suis mort; 
 κτάομαι j'aequiers, κίκτημαι j'ai acquis; done, je possede.' — Burnouf, 
 Gr. Gram. § 254. 
 
 f The infinitive of these contract verbs should not have the iota sub- 
 script, as they have in many editions ; τα els αν απαρέμφατα ουκ ίχςι το i 
 •χροση/νγραμμένον ' Ζτι τα, els ν λήγοντα βήματα ουδέποτε έχςι προ του ν τ* 
 αν€κψώνητον. — Etym. Magn. See Viger, Idiot, p. 220. 
 
 D 
 
50 A BRIEF GREEK SYNTAX, 
 
 97. When a noun gives rise to several derivative forms they 
 differ in meaning, as 
 
 πολεμάω I make hostile,* πολεμάω I am at war, πολεμίζω 
 
 I make war. 
 πλουτέω I am rich, πλουτίζω I enrich. 
 ΰουλόω I enslave, ΰουλεύω I am a slave. 
 δρμέω I lie at anchor, δρμίζω I bring to anchor {ορμάω I 
 
 stir up, is from a different root). 
 ρίπτω jacio I throw, ριπτώ jacto I boast, ριπτάζω I throw 
 
 often. 
 
 98. Frequentatives usually end in άζω, [ζω, νζω, as στενάζω, 
 ώθίζω, ερπύζω. 
 
 99. Inceptives in σκω^ as ηβάσκω juvenesco, -γηράσκω senesco, 
 μεθνσκω I begin to make drunk, &c. 
 
 100. Desideratives in σείω, as γέλασε /ω I am inclined to 
 laugh, ΰρασείω I want to do, πολεμησείω I should like to go to 
 war, εργασείω I long to work ; cf. esurio, parturio, &c. 
 
 Obs. i. The inceptive form σκω has the same iterative mean- 
 ing as the Epic substitution of σκον for the augment, e.g. 
 ΰινεΰεσκε for εΰίνενε, -γοάασκεν for εγόα. 
 
 ii. The desiderative form σείω is probably ' an old future in 
 -σείω, of which the corresponding aorist is found in the so- 
 called iEolic aorist optative in σεια/ as τνψεια.| 
 
 COMPOUND WORDS. 
 
 101. There are two kinds of compounds, Synthetic and Para- 
 thetic. 
 
 It is a curious and interesting fact that in Aryan languages the de- 
 termining word always precedes ; in Semitic languages, where however 
 compounds of any kind are rare, the determining word is always suf- 
 fixed; e.g. compare Newtown, Neapolis with Carthage; ite»-Yakoub 
 with Jacobscw, &c. See Families of Speech, Lect. iii. 
 
 102. i. Parathetic compounds are formed by the mere 
 juxtaposition of two separate words, as ναυσικλντός famous 
 for ships {ναυσι κλντός), γαστρίμαργος greedy, κυνόσσημα the 
 dog's tomb, &c. 
 
 * Where a verb has two forms, one in 6ω and one in ea>, the former is 
 usually transitive, the latter neuter ; e.g. πολεμούν to make an enemy of, 
 ποΧεμέιν to be at war. 
 
 f Some verbs in ιάω have a quasi inceptive meaning, as ϊΚιγγιάω I 
 grow dizzy, κςλαινιάω I grow black, ώχριάω I grow pale, &c. • 
 
 f See New Cratylus, § 386. 
 
SYNTHETIC COMPOUNDS. 51 
 
 English is very rich in these parathetic compounds. Ben Jonson in 
 his quaint grammar (1640) says, 'in which kind of composition our 
 English tongue is above all other very handy and happy, joyning together 
 after a most eloquent manner sundry words of every kind of speech.' 
 But he confuses such parathetic compounds as mill-horse, lip-wise, cut- 
 purse, with such synthetic compounds as notwithstanding, nevertheless, 
 &c. One of his instances, twy-light, has since become the synthetic 
 twilight. 
 
 ii. The commonest class of parathetic compounds in Greek 
 is furnished by the junction of verbs with prepositions, hence 
 these compounds admit of tmesis, as κατά -κίονα μηρι εκηαν, or 
 εκ δε οι ηνίοχος πλήγη φρένας ; this tmesis is found, though 
 rarely, even in Attic, as εκ <Γ ήνσ (Soph. Tr. 565), εκ δε πηδή- 
 σας (Eur. Hec. 1172). See too Ant. 420, 427, 432. 
 
 Sometimes even, in Homer, the preposition follows, as 
 ενάριζον απ εντεα. 
 
 iii. Yet merely parathetic as the compound is, a verb is often entirely 
 altered in meaning by the preposition with which it is compounded ; 
 e.g. yiyvaffKw is I know, but avayiyvaaKw I read ; κaτayιyvώσκω I con- 
 demn, iTriyiyuacncou I decide, μεταγ^νώσκω I change my mind, avyyiy- 
 νώσκω I pardon. Hence such a sentence as 'Aveyvoos a\\' ουκ eyvws • et 
 yap eyvws ουκ αν κοτεγι/ωί, you read it but did. not understand; for 
 had you understood you would not have condemned. 
 
 So, too, ακούω I hear ; βπακούω I overhear ; υπακούω I answer the door; 
 Εισακούω I obey; παρακούω I mishear, &c. 
 
 103. Synthetic compounds consist of elements which are 
 not separable, but have been modified before being moulded 
 into one organic whole, as μεγαλύδοξος, παντομίσης. 
 
 104. i. Adjectives and nouns in composition usually assume 
 their crude form, as ποΧυπονς, μεγαλόπολις, and if any con- 
 necting vowel be needed, ο is generally used, as in πατροκτόνος, 
 φνσιολόγος. 
 
 ii. This ο is not contracted if the second part of the word 
 originally began with a digamma, as in μηνοεώής, όρθοεπής, 
 μενοεικης. 
 
 iii. Some synthetic compounds are however joined by the 
 letter η, as ζιφηφόρος, εΧαφηβόλος, άσπώηφόρος, θανατηφόρος, 
 στεφανηφόρος. This may possibly have arisen from a desire to 
 avoid the concurrence of short syllables, since side by side 
 with these forms we find ϊ,ιφοκτόνος, εΧαφοκτόνος, άσπιΰοφέρμων, 
 στεφανοποιός. 
 
 105. In these compounds both words are generally signifi- 
 cant, as in ζνγηφόρος. Sometimes however one half is 
 merely poetical and ornamental, as in μονόσκηπτρος θρόνος, 
 γέννα θηΧνσπορος, άνηρ όιόζωνος. And frequently one half of 
 
 d2 
 
52 A BEIEF GREEK SYNTAX. 
 
 the word has become superfluous, and lost all its meaning, the 
 entire compound being only accepted in some secondary sense, 
 as μονόψηφον ξίφος a single (-voting) sword, οιόφρων πέτρα a 
 lonely (-minded) rock, ιπποκόμος καμήλων a (horse-) groom of 
 camels, νέκταρ εωνοχόει, &c. So in Sanskrit aqwa-go-sht'lia 
 a horse cow-stall, and even go-go-sht'ha a cow-cow-stall.* 
 
 N.B. i. Notice that Xt0o/3oXoe=pelted; λιθοβύλος = pelting; 
 μητμόκΓονος=^.ΐ]\βά by his mother; μητροκτό νυς = matricide. 
 
 ii. Compounds of εργάζομαι, if they imply bodily action 
 only are oxytone, as λιθονργός, αμπελουργός ; but on the other 
 hand we have πανούργος, κακούργος, περίεργος, &c, implying 
 moral action. 
 
 106. Latin has to a great extent lost — perhaps by contact 
 with some aboriginal language — the rich power of composition 
 possessed by Sanskrit and by Greek. ' Faciliore ad duplicanda 
 verba Grseco sermone.' — Liv. xxvii. 11. Even in historical 
 times we can trace something of the loss. Virgil, for instance, 
 has no compound words to compare with the ' Ubi cerva silvi- 
 cultrix ubi aper nemorivagus ' of Catullus. 
 
 107. It is an important and almost invariable law in Greek 
 that a verb never occurs as a synthetic compound except as 
 derived from some other synthetic compound. i Verba non 
 possunt nisi per flexuram quandam cumahis orationis partibus 
 praeter praapositiones consociari,' observes Lobeck. In other 
 words, ' a verb, without losing its nature, can only be com- 
 pounded with a preposition. When any other word is to be 
 compounded with a verbal stem a noun is first formed of the 
 two, and then a verb is derived from the noun.' Hence such 
 words as λιθοβάλλω, ιπποτρέφω, ναυμάχομαι, ευτυγχάνειν, 
 μετριοπάσχειν, &c, would be simple monstrosities in Greek ; 
 the only admissible forms being λιθοβολέω (from the inter- 
 mediate substantive λιθοβόλος), Ιπποτροφέω (from Ίπποτρόφος), 
 ναυμαχέω (from ναΰμαχος), ευτυχεω (from ευτυχής), μετριοπα- 
 θεΊν (from μετριοπαθής). 
 
 108. Apparent violations of this rule are either wrong read- 
 ings or the result of carelessness, as in Euripides συνασοφεΊν, 
 Ζυσθνηακειν, σταΰιοδραμοϋμαι, κακοβονλευθεΊσα. The latter 
 however should be σταοωΰρομησω (Here. F, 863), κακοβουλη- 
 θεΊσα (Ion, 867), and were probably altered by some ignorant 
 copyist. 
 
 * See Pott, Zahlmethode, p. 127. I have collected many other in- 
 stances in my Chapters on Language, p. 217, and may add 'brass fire- 
 ironSy ' tin shoe-horns,' ' wooden mile-stones,' &c. 
 
< TELEGRAM.' 53 
 
 In the N. Test, we have ενΰοκεΊν to be well pleased ; and 
 καραΰοκε'ιν to expect earnestly is found in some writers. Even 
 Scaliger had seen that such a verb as εναγγέλλω is not Greek, 
 ' nam το εν και τα στερητικά μόρια componuntur non cum verbis 
 sed cum nominibus.' The careless violation of analogy in the 
 Ινσθνησκω of Euripides {Rhes. 791, El. 834) may be due to 
 the metrical impossibility of Ινσΰανατεω ; yet in any other 
 dramatist we should have been more surprised to find it.* 
 
 109. The same rule applies to abstract substantives. Com- 
 pounds like λιθοβυλή, ν αν μάχη, ενπράζις would be impossi- 
 bilities in Greek ; the substantive must receive a derivative 
 ending as λιθοβολία, νανμαχία, ενπραζία. 
 
 110. Hence the word i telegram ' is a monstrosity, — ' a spot 
 of barbarity impressed so deep on the English language that 
 criticism never can wash it away.' From the words τήλε and 
 γράψω might have been formed the substantive τηλέγραφος, 
 and then through the verb τηλεγραψέω the abstract substan- 
 tive telegraph eme.\ ' Telegram ' violates the laws of Greek 
 synthesis, and if it meant anything, could only mean ' a letter 
 at a distance.' It must be regarded as a convenient English 
 hybrid ; and unfortunately many English hybrids are by no 
 means convenient. It is said that we owe many of them, and 
 this among the number, to the French. 
 
 * New Cratylus, p. 624. For a list of other careless peculiarities of 
 Euripides, see Bernhardy, Griechische Syntax, s. 14. 
 
 t Cf. from ζφον and γράψω, ζωγράφος, ζωγραψίω, and then ζωγράφημα 
 a painting. Plat. PhU. 39 d. 
 
54 A BRIEF GREEK SYNTAX. 
 
 SYNTAX. 
 
 1. i. Syntax (συνταζις, constructio, arrangement) gives the 
 rules for expressing or arranging sentences. 
 
 ii. The syntax of a language is not elaborated till late. 
 There could not be said to be such a thing as Greek grammar 
 till the Alexandrian epoch. Suetonius tells us that the first 
 Greek grammar was brought to Eome by Crates Mallotes, the 
 ambassador of King Attalus, between the second and third 
 Punic wars. 
 
 iii. In the grammar of any language there must be a great 
 deal which is common to it with every other language, and 
 which must necessarily arise from the fundamental resemblance 
 between the intelligence of different races. The points in 
 which a language differs from others are called its idioms 
 (ΙΙιωματα or peculiarities). Some such idioms are isolated or 
 unproductive ; others form a starting-point for many similar 
 phrases, and may be called paradigmatic (see Craik, Engl, of 
 Shakespeare, p. 203). 
 
 2. When a sentence, however short, offers a complete sense, 
 it is called a proposition (αυτοτελής λόγος oratio), i.e. an ex- 
 pression of judgment. 
 
 3. A sentence must consist of three parts — 
 a. The subject, or thing spoken of. 
 
 β. The predicate, i.e. what is stated of the subject. 
 
 γ. The copula* some separate verb expressed or under- 
 stood, or some lingual contrivance to express the mental act 
 which connects the subject and predicate. 
 
 N.B. i. As both the copula and subject are often understood, or merely 
 implied in the termination of a verb, a sentence may be expressed in 
 Greek and Latin by a single word, as ΰει, βροντά, έσεισε, σαλπίζει, it rains, 
 it thunders, there is an earthquake, the trumpeter is blowing. In Eng- 
 lish and most modern languages, at least two words are required, since, 
 owing to the analysing tendency, we express the pronouns even when 
 they are unemphatic. 
 
 * The copula belongs however rather to logic than to syntax; in 
 Greek it is constantly omitted. Thus ayaQos δ ανηρ means • the man is 
 good,' but we in English must express the ' is, 1 to give any meaning. 
 On the supposed necessity of this copula, see Origin of Language, 
 j>, 104 seqq. 
 
THE ARTICLE. 55 
 
 ii. Most forms of the finite verb make a sentence, containing these 
 three parts ; e.g. τύιττω means ' I (subject) am (copula) striking (jpredi- 
 cate).' 
 
 iii. Whatever may be the length of a simple sentence (i.e. a sentence 
 that contains but one finite verb), it can always be reduced to these three 
 parts, all other words being accessory either to the subject or the predi- 
 cate ; e.g. The virtuous and happy old man lived in peace and prosperity ; 
 here ' the virtuous, &c. man ' is the subject, ' was ' is the copula, ' living 
 in,' &c, is the predicate. 
 
 iv. A compound sentence (i.e. a sentence that has more than one finite 
 verb in it) may contain many simple sentences which are called its 
 
 v. Clauses are either coordinate, i.e. of equal importance with the 
 main sentence, as ' Alexander conquered Darius, and died young ' (παρά- 
 ταξα) ; or subordinate, as ' Alexander collected an army that he might 
 conquer' (υπόταξα). 
 
 THE ARTICLE ("Αρθρον).* 
 
 4. The Article ό, >/, ro, was originally a demonstrative pro- 
 noun, which also served as a personal pronoun; as in 
 Homer — 
 
 φθίσει σε το σον μένος that courage of thine will ruin 
 
 thee.f 
 την εγώ ου λύσω her I will not set free. 
 Αητοΰς και Διό> νιος' 6 γαρ βασιΚή'ί χολωθείς κ.τ.λ. the 
 son of Leto and of Zeus ; for he angry with the king, 
 &c. 
 ώς εφατ' εΰΰεισεν Γ 6 γέρων' So said he; but he, the old 
 man, feared. 
 
 N.B. In this last, and in similar instances, 6 is not an 
 article, J but a pronoun in apposition, as in ' The Lord, He 
 is the God.' 
 
 4 My banks, they are furnished with bees.' — Shenstone. 
 
 * The word άρθρον in this sense is first found in Aristotle, Poet. xx. 
 It means 'a joint' or 'limb'; see Egger, Apollon. Dyscol. pp. 112, 118. 
 
 f The τί> in this and similar examples merely adds to the emphasis, 
 and is like the use of the Latin ' ille ' before possessive pronouns, as ille 
 tuus pater,' that father of yours ; it is retained in the Eomance lan- 
 guages, — as ' il mio cavallo,' &c. It is a constant Spanish idiom to use 
 the article in a demonstrative sense as a personal pronoun, as ' El que es 
 sabio ' he (lit. the) that is wise. 
 
 \ In some instances however this demonstrative is, even in Homer, 
 to all intents and purposes an article ; e.g. 11. vii. 412, xii. 289, το δε 
 τεΓχοί ΰπίρ παν Soihros opdpei, &c. Apollonius Dyscol. Synt. i. 31. But 
 these instances are not numerous : and on the other hand it is often 
 
56 A BKIEF GREEK SYNTAX. 
 
 5. Even when b, ή, το had developed into a definite article 
 (like our ' the '), it was used as a demonstrative ;* as 
 
 τον γαρ και γένος εσμέ ν, ι for we are also his offspring.' 
 
 — Aratus, quoted in Acts xvii. 28. 
 προ τοϋ, before this (German ehedem). 
 f/ το'ισιν η το'ις πόλεμον αίρεσθαι to take up war against 
 
 these or those, 
 οι εν άστει those in the city. 
 
 6. Especially with various particles, as μεν, Ιέ, και, &c. 
 έβλαψε με 6 ΰέϊνα το καϊ το ποιήσας so and so injured me 
 
 doing this and that (or doing such and such a thing). 
 και μοι καλεί τον καϊ τον now call me so and so. 
 oi μεν έθαΰμαζον, οι ΰέ έβόων some were in astonishment, 
 
 others were shouting. 
 
 7. This demonstrative pronoun (ό, η, το) also served origin- 
 ally for the relative {ος ή o),f with which it is most closely 
 connected. In fact ος τε not ος means ' who ' in Homer (et 
 is = qui); or, in other words, language originally states 
 co-ordinately what was afterwards made subordinate. 
 
 άλλα τα μεν πολίων έζεπράθομεν τα ΰέΰασται the things 
 which we sacked from the cities those things have been 
 divided.— II. i. 125. 
 
 (The example is a curious one because it is, I believe, the only instance 
 in which Homer puts the relative before the antecedent.) 
 
 This usage continued in Ionic, and even in Attic, as 
 
 ret μεν Ότάνης είπε . . . λελέχθω κάμοϊ ταντα the things 
 
 which Otanes said, &c. 
 £i7r\ju μάστιγι, τήν'Άρης φιλεϊ (iEsch. Ag. 642), with the 
 
 double scourge, which Ares loves. 
 
 It is even continued in Modern Greek, as τά <pepvei η 8>pa what an hour 
 brings. (Clyde.) 
 
 8. i. Possibly δς ή το was the original form of this demon- 
 strative, and the ς was dropped because (e.g.) ό(ς) αγαθός άνήρ 
 would not sound well ; just as in German we have der gute 
 
 omitted where an article is required, as rtfis δε μοι i)y Ιστηκε» eV &γρου 
 νοσφϊ ποληο$ far from the city. 
 
 ά\\οι μέν pa 0eot re καϊ avepes the rest of gods and men. 
 
 * Similarly, in Hebrew Π was originally demonstrative, and occa- 
 sionally retains its demonstrative force, as in DVH this day. 
 
 f ' Thus too in English the demonstrative that has come to be also a 
 relative.' — Clyde, Grr. Synt. p. 9. 
 
 
DEVELOPMENT OF THE ARTICLE. 57 
 
 Mann, not der gute(r) Mann, because the grammatical instinct 
 would have been offended by the conscious repetition of the 
 pronoun (which -was felt, though not recognised) in der guter. 
 See Breal, Bopp, ii. § 281. *Oc in Attic is still demonstrative 
 in the phrases και ος and he, ή δ' δς said he, &c. 
 
 ii. In fact the use of an article with the nominative is an 
 unconscious pleonasm, due to the obliteration of the nominatival 
 termination. The nominative termination is derived from sas 
 the Sanskrit article : many ages afterwards the Greeks used 
 this same article under the form ό to accompany and define 
 the nominative. This double process of obliteration and re- 
 production in language has already been illustrated in § 105. 
 See Breal, Bopp, ii. xxxvii. 
 
 9. We see then that the article, the demonstrative, and the 
 relative are merely developments of one and the same form.* 
 This is illustrated by the fact that — 
 
 a. There is no article in Latin in which hie and tile serve 
 the same purpose, when anything very definite is wanted. 
 * Noster sermo articulos non desiderata says Quinctilian (Instt. 
 Or. i. iv. 19). It must however be admitted that the article 
 if unnecessary is at any rate very convenient. So far from 
 being, as J. C. Scaliger called it, ' otiosum loquacissimce gentitt 
 instrumentumf it adds to language a most desirable precision. f 
 
 * In fact they are all three simply determinative adjectives. DuMeril, 
 Form, de la Languefrang. p. 60. 
 
 f Duclos cites, as instances of the precision attainable by the use of 
 articles, the sentences — 
 
 a. Charles est fits de Louis 
 β. — unfits — 
 y. — lefts — 
 
 Here a. expresses the general fact ; β. shows that Charles has brothers , 
 y. shows that Charles is an only son. Here then one may see both the 
 desirability of the article, and the absurdity of Scaliger's remark, 'Dis- 
 pleased with the. redundance of particles in the Greek, the Romans 
 extended their displeasure to the article, which they totally banished ! ' 
 Prof. Trithen observes that his arrogant dictum ' Articulus nobis nullus, 
 et Groecxs superfluus ' is much as if he had said ' There are no Alps in 
 England ; they exist in Switzerland, but they are superfluous.' {Trans, 
 of the Philolog. Soc. 1850, p. 11.) Moreover, colloquial Latin in all 
 probability did use the pronouns as definite articles, and the numeral as 
 an indefinite article ; hence such phrases as Terence's ■ Forte unani 
 aspicio adolescentulam.' — Andria, i. i. 91 ; cf. Plaut. Most. iv. 3. 9. This 
 is an instance of one of ' those instincts of clearness which anticipate 
 grammatical development.' For other methods by which the Latin 
 makes up for its want of an article, see Nagelsbach, Lateinische Stylistik, 
 §3. 
 
 P3 
 
58 A BRIEF GREEK SYNTAX. 
 
 β. The article has been developed by the Romance lan- 
 guages (i.e. those derived from Latin) out of the demonstrative 
 pronoun Me, as : 
 
 In French le, la, les. 
 
 In Italian il, lo, la, i, gli, 
 
 In Spanish el, la, los, las, 
 
 In Wallachian lu, a ; le, ί, 
 
 In Sanskrit the article did not exist, the demonstrative sas, sa, tat 
 being used instead (as in Latin); nor does it occur in Sclavonic and 
 Lithuanian. 
 
 y. The same three uses of the article (as article, demon- 
 strative, and relative) are found in German, as Der Mensch, 
 den (relative) tch befreundete, der (demonstrative) hafsgethan, 
 the man whom I befriended, he has done it (Clyde). The 
 demonstrative der has been applied as a definite article, just 
 as the Anglo-Saxon ' poet ' has become ' the.' Similarly, in 
 many languages, the indefinite article a or an (the Scotch ane) 
 has been developed out of the numeral one. An for one is 
 first found in Layamon's Brut, and at one time they seem to 
 have been used almost interchangeably, e.g. ' The Owl and 
 the Nightingale' (a.d. 1250) line 6, ' An hule and one night- 
 ingale.' Probably in later Greek the numeral was used in- 
 definitely, cf. Matt. xxi. 14, ΙΙων συκην μίαν επι της όΐον. 
 
 Chief Uses of the Article. 
 
 10. The Greek article (as in English) either (i.) specifies 
 and individualises, as — 
 
 ό βοϋς εσφάχθη the ox (which you know of) has been 
 killed ; 
 Or (ii.) generalises, i.e. represents an individual as belong- 
 ing to a class — 
 
 δ βονς ζωον -χρησιμωτατόν εστί the ox is a most useful 
 animal. 
 Both uses exist in modern languages. Thus, in German, Der Mensch 
 ist sterblich man is mortal ; in Spanish, El caballo es animal noble the 
 horse is a noble animal, &c. 
 
 * See Clyde's Gr. Syntax. In Wallachian (as in Basque) the article 
 is suffixed, just as ille may follow its word in Latin, as ochiul for ochiu il, 
 Musc'ei for Mused lei. (Du Meril, p. 362.) It has also formed the 
 articles aquestu, aquelu, from hie iste, hie ille. In the Eomance lan- 
 guages the article still constantly retains its demonstrative force, as in 
 Spanish, ' Mis libros y los que έΐ tiene,' my books and the which he has; 
 l Los de vuestra nacioD,' those of your nation; in French, 'Le roi le 
 veut,' the king wills it, de la sorte, a ^instant meme, &c. 
 
USES OF THE ARTICLE. 59 
 
 11. In the latter case we often use our indefinite article a, 
 an, as— 
 
 τα σημεία του αποστόλου the signs of an apostle. — 2 Cor. 
 
 xii. 12. 
 ουΰέ . . . τιθέασιν αυτόν υπό τον μόΖιον άλλα επί την 
 
 λυχνίαν they do not put it under a bushel but on a 
 
 candlestick. — Mt. v. 15. 
 del τον στρατιώτην τον άρχοντα Φοβεϊσθαι a soldier should 
 
 fear his general. 
 
 12. The article is only used with proper names* when they 
 have been previously mentioned, or to call special attention to 
 them, as 6 Σωκράτης ; but not generally if any designation is 
 added, as Σωκράτης 6 φιλόσοφος, θουκυΰίΰης 6 'Αθηναίος, Κροίσος 
 6 των Αύδων βασιλεύς. So in Southern Germany Der Johann 
 {the John, i.e. our servant John) soil das Pferd bringen is to 
 bring the horse. And we talk of the O'Donoghue, the Chis- 
 holm, &c. (Clyde.) In French this is common when names 
 are used familiarly, as ' la Taglioni,' &c. 
 
 Our non-usage of the article with proper names leads to the style of 
 deeds, &c, with their troublesome addition of ' the said,' 'the aforesaid,' 
 &c. ' This tedious repetition which clogs and encumbers the style of 
 our writs so much would be saved if we used the article in the way the 
 Greeks do, and the style would be as well-connected as it is without 
 such gouty joints, to use an expression of my Lord Shaftesbury's.' — 
 Monboddo, Orig. of Lang. ii. 57. 
 
 13. Words signifying objects of which only one exists, are 
 used as proper names, and need take no article, as βασιλεύς 
 the king of Persia,f εν άστει i in town,' εν άγορ$ at market, 
 επι θαλασσή at sea, νυκτός by night, &c. Hence ήλιος, γη, &c. 
 and the names of virtues and vices are often anarthrous. 
 
 14. The article distinguishes the subject from the predicate, 
 as: 
 
 βασιλεύς εγένετο το πτωγάριον the beggar became a king. 
 
 * Names of places are expressed very variously with the article, as 
 δ ποταμός δ Ευφράτης the river Euphrates ; η Αΐτι/η το ορός Mount JEtna ; 
 Πάρνης το opos Mount Parnes ; Ίϊικ€λία η vr\aos Sicily ; t\ πόλις οΐ Ταρσοί 
 the city of Tarsus, &c. The common order however is δ Ευφράτης ποταμός 
 the river Euphrates ; η Βολ/3ή λίμνη the lake Bolbe ; to Alya\4wv ορός 
 Mount JEgaleum ; η ®ςσπρωτ\ς γη the Thesprotian land ; η Αήλος νήσος 
 the isle of Delos. The substantive and proper name are really in apposi- 
 tion, and a similar collocation is not uncommon in English poetry, as 
 'This great Oxus stream,' 'famous London city,' &c. 
 
 f The king of Persia was called βασιλώς king, or β, δ μέγας, but not 
 δ β.) e.g. ol πρόγονοι ol βασιΚίκς. 
 
60 A BRIEF GREEK SYNTAX. 
 
 νυζ ή ημέρα iy ενετό day was turned into night. 
 θεός 7]v ο Λόγος the Word was God. 
 
 The .same rule holds in Hebrew and English. There is a strange 
 violation of it in Milton's 
 
 1 Light the day and darkness night he named,' 
 where Bentley reads * the Light, Day.' 
 
 15. Often Greek (like French) uses the article where we 
 use the possessive pronoun ;• as 
 
 άλγώ τήν κεφαλήν j'ai mal a la tete, my head aches. 
 6 βασιλεύς συν τω στρατενματι the king with his army, 
 έχει όζείς τους οφθαλμούς his eyes are sharp (compare the 
 
 French il a les yeux beaux, and the Italian egli ha la 
 
 vista acuta). 
 
 16. You may say in Greek either αγαθός άνήρ, or άνήρ 
 αγαθός for α good man ; but ' the good man ' (and every 
 similar collocation) must he in Greek in the same order as the 
 English : 
 
 6 αγαθός άνήρ the good man, 
 
 or which is equally correct but more pleonastic 6 άνήρ 6 
 αγαθός. 
 
 17. The attributive genitive follows the same order, as η Θεμιστοκλέους 
 αρετή or η αρετή Θεμιστοκλέους, δ Αθηναίων δήμος or δ δήμος δ Αθηναίων ; 
 and this holds true no matter how many intermediate words are inter- 
 posed, as in 
 
 τδ ttjs του βαίνοντος τέχνης έργον the work of the wool-carder's art. 
 η των τα. της πόλεως πράγματα πραττόντων αρετή the virtue of our 
 statesmen. 
 
 In phrases like ' my mother,' ' thy word,' the order is η έμη μήτηρ, or 
 f} μί]τηρ μου, δ σδς λόγος, or δ λόγος σου. 
 
 Ν.Β.— The attributive genitive must have the article, if the noun on 
 which it depends has it, unless there be some special reason to the con- 
 trary, as 
 
 η του γεωργού δόξα the husbandman's opinion. 
 τί> της αρετής κάλλος the beauty of virtue. 
 
 18. But if the adjective, when it occurs with a substantive 
 and article, is placed either first or last, it becomes & predicate ; 
 as: 
 
 αγαθός 6 άνήρ good (is) the man. 
 
 6 άνήρ αγαθός the man (is) good. 
 
 So in Chinese ngo-jin = & bad man; but jin-ngo = the man is bad. 
 
 * • The G-reek article here denotes that the subject has a definite kind 
 of property it is known to possess.' — Winer, m. § xviii. 2. 
 
USES OF THE ARTICLE. 61 
 
 19. This must be specially noticed in all the cases ; thus : 
 οι λόγοι ψενΰέΐς ελέχθησαν not ' the false words ' but ' the 
 
 words spoken ' were false. 
 6 μάντις τους λόγους ψενΰεΊς λέγει the words which the 
 prophet utters are false. 
 
 20. The last example is an instance of what Dr. Donaldson 
 calls a tertiary predicate, which assumes or anticipates the 
 existence of another predicate, and must therefore be often 
 rendered by a separate sentence, as : 
 
 όζύν έχει τον πέλεκνν the axe which he has is sharp. 
 αρχαία τα Ααβΰακιΰάν οίκων δρώμαι ιζήματα the woes of 
 
 the Labdacidae which I see are ancient. 
 ΰιπλά δ' ετισαν θάμάρτια the penalty which they paid was 
 
 two-fold. 
 ov γαρ βάναυσον την τέχνην έκτησάμην for the art which I 
 
 acquired is no mean one. 
 
 Notice the position of the adjective and article in the fol- 
 lowing sentences : 
 
 άφίεσαν την ΰοκον χαλαραΐς ταΐς άλύσεσι they let down 
 
 the beam with the chains loosened. 
 ένέπρησαν τάς σκηνάς έρημους they burned down the tents, 
 
 deserted as they were, 
 ζενχθη όζυχολος ττα~ις ό Δρύαντος the son of Dry as, because 
 
 he was keen in wrath, was bound. 
 κανταΰθ' 6 πάϊς ΰΰστηνος ovr όΰυρμάτων ελείττετ* ο'υΰεν and 
 
 thereupon the boy, unhappy as he was, was neither 
 
 lacking in lamentations, &c. 
 
 21. Sometimes the law of the position of the article appears 
 to be violated, as in 
 
 μηθ' 6 Χνμεων έμός nor he who is my outrager. — Soph. 
 
 Aj. 572. 
 Ζευς a 6 γεννητωρ έμος Zeus who is my father. — Eur. 
 
 Hipp. 683. 
 τώμττέχονον ποίησας εμον ράκος you've made my dress a 
 
 rag. — Theocr. xxvii. 58. 
 
 In all these instances probably the true reading is έμοί* 
 (New Crat. p. 487). Some editors however think that the 
 possessive is emphatic, and content themselves with the re- 
 mark, ' Articuli collocatio valde inusitata.^ 
 
 * Possibly however the e/*ta is added as an afterthought. 
 
62 A BRIEF GREEK SYNTAX. 
 
 22. The following examples will illustrate the chief pecu- 
 liarities of the article : 
 
 i. dig τον μηνός twice a month. 
 
 τρία ημιΰαρεικά τον μηνός τω στρατιώτη three half darics 
 
 a month to each soldier. 
 ΰραχμήν της ημέρας a drachma a day. 
 
 This is called the distributive use of the article ; Clyde 
 
 compares the German, Zweimal den Monat, and the Italian 
 
 due volte il mese ; so too in French, un franc la bouteille, &c. 
 
 ii. οντος 6 avy)p* this man. 
 
 έκε~ινος υ λόγος that argument. 
 
 ifts η γνώμη this opinion. 
 
 εκάστη η άρχ?) each kingdom ; or, which is equally correct 
 
 though less emphatic, ό άνηρ ούτος, η γνώμη ή$ε, &C. ; 
 
 but ό must never immediately precede οντος, εκείνος, 
 
 οΰε, έκαστος, εκάτερος ; preceding αντος it means ' the 
 
 same,' as : 
 
 6 αντος άνθρωπος the same man ;f (homo idem). 
 
 but 6 άνθρωπος αντος),•, -,• lf /, . λ 
 
 , χ \ „ a >the man himself ; (homo ipse). 
 
 αντος ο άνθρωπος ) ' ν r ' 
 
 iii. Notice the difference made by the article in the following 
 phrases : 
 
 τριάκοντα thirty, ol τριάκοντα the thirty (tyrants). 
 ενΰεκα eleven, οι ένδεκα the eleven (executioners). 
 ολίγοι few, οι ολίγοι the oligarchy. 
 πλείονς more, οι πλείονς the majority; sometimes = the 
 
 dead (cf. ί abut ad plures '). 
 πολλοί many, ol πολλοί most, the mob. 
 αλλοί others, ol άλλοι the rest. 
 πάντα Ιέκα ten of each, τα πάντα δέκα ten in all. 
 ΰνο μέρη two parts, τα δύο μέρη two thirds. 
 άλλη χωρά another land, η άλλη χωρά, the rest of the 
 
 land. 
 ανά πάσαν ημέραν every day, ανά πάσαν την ημέραν all 
 
 day long. 
 πάσα πόλις every city, πάσα η πόλις or η πάσα πόλις the 
 
 whole city. J 
 
 * When ούτος, inuvos, &c, are used without the article, they are in 
 apposition, as ταύτην έχει τέχνην he has this as an art ; τούτψ παραδεί- 
 γματι χρώμζνος using this as an example. 
 
 "j* αυτός, αυτή, ταύτό or ταυτόν, are used for δ αυτός, τ) αυτή, το αυτό. 
 
 \ The difference between ό 7r«s and ttbs δ is much the same as that 
 
 
USES OP THE ARTICLE. 63 
 
 ΰοϋλος έμός a slave of mine, 6 έμός ΰοϋλος that slave of 
 
 mine. 
 εσχατον το ορός the farthest part of the mountain, το 
 
 εσχατον 'όρος the farthest mountain. 
 η μέση πόλις the middle city, μέση η πόλις or // πόλις 
 
 μέση the middle of the city. 
 το μέσον τείχος the middle wall, μέσον το τείχος the 
 
 middle of the wall. 
 τόϊς άκροις ποσ\ν with the toes, άκροις τοΊς ποσιν on tiptoe. 
 βασιλεύων b Κϋρος Cyrus when he was king, Κϋρος 6 
 
 βασιλεύων Cyrus, who is king. 
 το καλόν the beautiful, τα καλά things beautiful. 
 
 23. The article can turn any infinitive into a substantive : 
 τλησομαι το κατθανεΊν I will endure to die. 
 
 το λέγειν speaking, του λέγειν of speaking, &c. 
 
 So our ' to ;' as 
 
 ' To err is human, to forgive divine ' 
 
 (like the Italian ilpeccare. Clyde) ; and even in oblique cases, 
 as Spenser's 
 
 1 For not to have been dipped in Lethe's stream 
 Could save the son of Thetis from to die.' 1 
 
 24. Observe the phrases ol πάνυ* the elite, 6 αεί κρατών the 
 l^ing for the time being, οι πάλαι the men of old, το σύμπαν 
 on the whole, τάλλα for the rest, τα πολλά for the most part, 
 ret μάλιστα in the highest degree, το επ έμοι for my part, ra 
 από τοϋΒε henceforth, το άρετη the word ' virtue.' 
 
 between ' the whole ' and ■ all the ;' i.e. the difference is almost inap- 
 preciable. We might say that δ πάς, like the Italian tutto, meant an 
 indivisible whole ; and that πα? δ, like ogni, was a distributive whole ; 
 — but in point of fact both orders are used in the same clause, as πάσι 
 to7s Kpirah καϊ rots Oearais πασι. — Ar. Av. 444. παε = omnis ; anavrss = 
 cuncti (i.e. conjuncti) ; σύμπαντα = universi, all by common consent ; 
 'o\os = tottcs. — Donaldson, Lot. Gfr. p. 79. 
 
 * This adjectival use of adverbs is not unknown in English ; e.g. 
 
 1 My sometime daughter.' — King Lear, Act i. sc. i. 
 • Mild innocence 
 A seldom comet is.' — Donne. 
 
 1 They hoped for a soon and prosperous issue.' — Sidney. ' The then Par- 
 liament voted,' &c. Even in Latin, though it has no definite article, we 
 find such phrases as • discessu turn meo,' by my then departure. Cic. Pis 
 ix. 21 ; ' ipsorum deorum sespe prsesentise,' the frequent presences of the 
 gods, &c. — Cic. Be Nat. Deor, n. lxvi. 166 ; Nagelsbach, Lat. Styl, § 75 
 
64 A BRIEF GREEK SYNTAX. 
 
 εν τοίς π ρώτ οι= omnium primi. — Thuc. i. 6. 
 
 εν τοίς πλε'ισται quite the most. , 
 
 το and το 2ε sometimes=therefore (at the beginning of 
 
 sentences). 
 το τών& ευνουν the good will of these ; cf. CEd. Col. 
 
 8, 579, &c, vide § 38. 
 
 N.B. Before we leave the article, it is worthy of notice that 
 in such phrases as ' the more they have, the more they desire,' 
 we use οσω, τοσοΰτω, and in Latin quo, eo. Here ' the ' in 
 English is not an article at all, but a corruption of the Ger- 
 man je. 
 
 CONCORD. 
 
 25. The rules for the three concords are the same in Greek 
 as in Latin. The numerous violations of them which are 
 given below are nearly all self-explaining, and arise from the 
 fact that the Greeks being an extremely quick race, often 
 allowed the sense to overrule the grammar, or substituted the 
 logic of thought to that of grammatical forms. They saw 
 through the form, and often disregarded it. This important 
 principle of construction is called the sense-figure, — σχήμα 
 προς το σημαινόμενο}', constructio ad sensum, or briefly κατά 
 σΰνεσιν. Hence all such expressions as the following : — 
 
 ο ογλος . . . επικατάρατοί είσιν the people . . . are ac- 
 cursed. — John vii. 49. 
 φίλε τέκνον dear child. 
 
 το μειράκιον εγένετο κάλος the boy grew up handsome. 
 Ύροίαν ελόντες . . . στόλος the host, after taking Troy. 
 φεύγει ες Κερκύραν ώς αυτών ευεργέτης* he flies to 
 
 Corcyra, as being their benefactor. 
 πόλιν επραθον ώλεσα ο" αυτούς I burnt the city, and slew 
 
 them (i.e. the inhabitants). 
 ες ΰε την Σπάρτην ώς ήγγέλθη . . . ε^οζεν αυτοίς when it 
 was announced at Sparta, they decided, &c. [compare 
 Gibbon's expression 'Each legion, to whom was al- 
 lotted,' &c.]. 
 τερπνόν τράπεζα πλήρης a full table is a good thing. 
 
 * Expressions like • The skip sailed, and they (i.e. the crew) were 
 brave,' or ' The city was in confusion, and they voted,' &c, are very 
 common in Greek, which very properly despised a pedantic accuracy of 
 grammatical structure, when the meaning could be quite as clearly 
 expressed with more brevity. In Thuc. i. 110 we find τρι -fipcts . . . ουκ 
 clSOTts. 
 
THE 'SENSE-CONSTRUCTION.' 65 
 
 ol παϊΰές είσιν άνιαρον boys are a bore. 
 
 αΐννατά έστιν άποφνγεΊν it is impossible (neut.plur.) to fly. 
 
 άμυντέα εστίν αντώ we must defend him. 
 
 Ζόζαν ταύτα when this had been decreed. 
 
 ΰοκεϊ μοι ορών it seems to me, seeing, &c. 
 
 εμα κήΰεα θυμοϋ the woes of my heart. 
 
 26. Neuter plurals take a verb singular, because mere 
 multeity or mass implies no plurality, or separation ^ of 
 agencies ; * in fact, the neut. plur. is an accusative or objective 
 case, things not animate being regarded as only capable of 
 being acted on. Hence τα ζώα τρέχει properly means ' as to 
 the animals there is running.' This is called the Attic figure 
 (σχήμα Άττικόν), and it exists also in Hebrew and Arabic. 
 
 27. But here also the sense also controls the form, when 
 requisite : 
 
 ret τέλη εξέπεμψαν the magistrates sent out. 
 
 σπάρτα λέλυνται the ropes have grown slack (i.e. one and 
 
 all of the ropes). 
 τοσάΰε έθνη εστράτευον so many nations were going to 
 
 war. 
 
 28. Duals agree with plurals, because the dual is a subor- 
 dinate plural, as εϊλετο δ' άλκιμα ΰοΰρε and he grasped two 
 stout spears. — Horn. 
 
 In άμφω τώ πόλπ both the cities (Thuc.) we have a masc. dual with a 
 fern, noun (τώ for τα), as is always the case in Attic Greek. 
 
 29. Sometimes by what is called the Pindaric or Boeotian 
 figure a singular verb is put with a plural noun, as μελιγάρυες 
 νμνοι υστέρων άρχαϊ λόγων τέλλεται. — Olymp. xi. 4. Honeyed 
 hymns becomes the origins of later songs. The exigences of 
 metre have even forced from Shakspeare this violation of 
 syntax, as 
 
 1 Hark ! hark ! the lark at heaven's gate sings, 
 And Phoebus 'gins arise, 
 His steeds to water at those springs 
 On chaliced flowers that lies. 1 
 
 * ' The neuter plural governing, as they call it, a singular verb, is one 
 of the many instances in Greek of the inward and metaphysic grammar 
 resisting successfully the tyranny of formal grammar. In truth, there 
 may be multeity in things, but there can only be plurality in persons. 
 Observe also that, in fact, a neuter noun in Greek has no real nominative 
 case, though it has a formal one — that is to say, the same word in the 
 accusative. The reason is, a thing has no subjectivity or nominative 
 case ; it exists only as an object in the accusative or oblique case.' 
 —Coleridge, Table Talk. 
 
66 A BRIEF GREEK SYNTAX. 
 
 Mr. Morris shows that lies is a plural form in some English 
 dialects, but similar phrases are common in Shakspeare, Bacon, 
 &c. ' Is this the fashions.' — 2 Henry VI. i. 2. ' There is 
 tears for his woe.' — Jul. Cms. iii. 2. 'There is none of 
 Hercules's followers,' &c. — Bacon, Adv. of Learn. ' Good 
 Things cometh from God,' is the title of one of the 
 Homilies. This idiom is confined in Attic to ειμί, used im- 
 personally at the beginning of sentences. 
 
 εστί γαρ εμοιγε και βωμοί I too have altars. 
 
 εστίν ο'ί=•ίν ιοί — sunt qui. 
 
 εστί 3' επτά στάοιοι εξ Άβνΰον it is seven stades from 
 Abydos. 
 
 ήν (Γ άμφίπλεκτοι κλίμακες there ivas wrestling tricks. — 
 Soph. Tr. 520. 
 We have the same idiom ; e.g. ' it is now a hundred years 
 since,' &c. Dr. Priestley defends the propriety of a singular 
 verb after 'there ' even when a plural follows.* Compare the 
 French il y a des homines ; and the German ' Es sind 
 Menschen? This construction is the rule in Turkish (Barker, 
 Turk. Gram. p. 83). 
 
 30. A singular and plural are often mixed | by what is 
 called ' the whole and part figure ' (σχήμα καθ* όλον και μέρος) ; 
 as 
 
 εμενον εν rrj εωυτου τάζει έκαστος they stayed, each in his 
 
 own rank. 
 ούτοι μεν άλλος άλλο λέγει they say, some one thing, 
 
 some another (cf. Matt, xviii. 35). 
 
 31. The plural of excellence (by which a person says ' we ') 
 often leads to a mixture of concords, J as 
 
 ην βάνω Θανοϋμεθα if I die, we will die. 
 So in Ovid : 
 
 'Et flesti et nostros vidisti flentis ocellos.' 
 
 * Such a construction apparently used not to be uncommon ; e.g. we 
 find in Dowsing's record of his desecration of Cove Hythe Church in 
 1643, ' There was four steps with a vault underneath. There was many 
 inscriptions to Jesus in capital letters,' &c. 
 
 t Karely a plural is put between two singxilars, as in 
 
 et de κ* "Αρης &ρχωσι μάχης η Φο?/3οϊ Άπόλλωψ. — Β. ν. 138. 
 
 This is called the σχήμα 'Αλκμανικόν (see Lesbonax, p. 179), from the 
 occurrence in Alcman of the phrase Κάστωρ re πώλων ταχέων δμητηρες 
 καϊ Πολυδεύκης Castor, tamers of swift steeds, and Pollux. Bernhardy, 
 Griech. St/nt. s. 421. . 
 
 X Compare in Hebrew p»"W ^ΓΡΚ 
 
CASES. 67 
 
 32. A woman using the plural also uses the masculine; 
 thus Electra says : 
 
 πεσονμε& ει γρη πατρι τιμωρούμενοι. — Soph. El. 391. 
 
 33. aye, (pipe, Ide, είπε, being merely interjectional, can 
 be put with plurals ; as 
 
 είπε μοι, τι πάσγετ\ ων\)ρες ; — Ar. Pax, 325. 
 
 CASES (Πτώσεις) * 
 
 34. The case-endings, which once were separate words 
 although in course of time they have got inseparably united to 
 the noun-stems, originally denoted the simplest and most 
 obvious relations, viz. those of place. From these relations, 
 which, as we have seen, were expressed by pronominal ele- 
 ments, the others were developed. There are some languages 
 in which the cases are expressed by entirely separate words ; 
 e.g. in Chinese the word tchi i bud ' is used for the genitive 
 case, as metaphorically indicating the ideas of dependence and 
 causality. 
 
 35. The relations of objects may be considered from• so 
 many points of view, that we must not be surprised to find 
 that the border-limits of the cases are by no means very de- 
 finite, and that different cases can be used to express nearly the 
 same conception. Thus ε£ αριστεράς (α dextra), εν άριστερξ., 
 ες άριστεράν (zur Rechten), επ αριστερά are all good Greek for 
 on the left ; and we can say equally well in English on the left, 
 at the left, and to the left. (Clyde.) The nominative and voca- 
 tive are generally treated as cases, but they are not really so, 
 because they express no objective relations. The word -κτωσις 
 casus in its original meaning (falling) is entirely inapplicable 
 to either of them. 
 
 * The word πτωσις ' case ' from πίπτειν is first found in this sense in 
 Aristotle, Categor. i. For a full account of it see Lersch, Sprachphilos. 
 der Alien, ii. 182 seqq. Indeclinable words are called άπτωτα. The 
 nominative was not regarded as a πτώσις, and hence in Aristotle it is 
 called simply όνομα; but each other case was considered ws από του 
 ονόματος πεπτωκυΊα ; they were called πτώσεις πλά-γιαι, obliqui cases ; and 
 also, by Chrysippus, ΰπτιαι. The number of cases differ greatly in 
 different languages. Many modern languages (e.g. French, Italian, &c.) 
 have lost them altogether ; Hebrew has two, Arabic three, German four, 
 Greek five, Latin six, Kussian seven, Sanskrit eight ; while some lan- 
 guages, like Basque and the American languages, have as many cases as 
 there are prepositions, or rather postpositions. See Burggraff, Princ. de 
 Gram. gen. p. 243. 
 
68 A BRIEF GREEK SYNTAX. 
 
 36. The metaphysical nicety with which the Greek cases 
 are employed rendered their use very difficult to foreigners. 
 This is one of the reasons why in the New Testament preposi- 
 tions are so often employed where they would be superfluous 
 in classic Greek, as in έιδοναι εκ, εσθίειν από, πυλεμε'ιν μετά, 
 &c. In Modern Greek the dative case (and the genitive 
 plural) have been entirely displaced by analytical phrases 
 (prepositions, &c.).* 
 
 37. Of the eight cases found in Sanskrit (which is pro- 
 bably the oldest language of the Aryan family) the Greek 
 retains but five, and the Latin six ; so that we have these 
 three tables : 
 
 Sanskrit. Greek. Latin. 
 
 1. Nominative. 1. Nominative. 1. Nominative. 
 
 2. Genitive. 2. Genitive. Ablative. 2. Genitive. 
 
 3. Dative. 3. Dative. Instrumental. 3. Dative. 
 
 Locative. 
 
 4. Instrumental. 4. Accusative. 4. Accusative. 
 
 5. Locative. 5. Vocative. 5. Vocative. 
 
 6. Accusative. 6. Ablative. Instru- 
 
 7. Vocative. mental. Locative. 
 
 8. Ablative. 
 
 From this table it appears that in Greek the accusative 
 alone of all the cases has preserved its exact original force. 
 The genitive and dative are mixed, or, as Pott calls them 
 (El. Forsch. i. 22), syncretistic cases, and cannot be reduced to 
 a single principle. Thus the gen. is also an ablative ; the dat. 
 is also an instrumental and locative. 
 
 The cases fall under two divisions, of which one consists of 
 the nom., accus., and vocative ; the other cases admit of fre- 
 quent interchanges. 
 
 On this view of the cases see Quinctilian (Instt. Orat. i. 4- 
 26), who points out the distinct traces of a locative in the 
 Latin (militiae, humi, domi, belli, ruri, ibi, ubi), just as we 
 have similar traces in the Greek οίκοι, &c. JEsch. has πέΰοι, 
 cf. μέσσοι (iEol.) ποι, οι. Simon., fr. 209, has εν Ισθμοί, 
 where the locative is defined by a preposition. The only 
 locative of the α declension is χαμαί. 
 
 Such forms as ουρανόθεν, θύραθεν, are ablatival. 
 
 NOMINATIVE (Πτώσις ορθή, ευθεία, ονομαστική). 
 
 38. By an example of the constructio ad sensum, the nomina- 
 tive is sometimes placed in independent apposition to the 
 
 * Deville, Dialecte traconien, p. 98. 
 
THE VOCATIVE. 69 
 
 notion of the sentence, though not to the form in which it is 
 expressed. This is called the nominative absolute, as 
 
 αΙΖως μ* 'έχει (=cu3ov/xcu) τάδε πράζας I am ashamed at 
 
 such conduct, 
 λόγοι δ' εν αλληλοισιν ερρόθονν κακοί, 
 φνλαζ ελέγχων φύλακα there was an angry dashing of 
 mutual reproaches, guard reviling guard. — Soph. .471%. 
 259: 
 Obs. Such phrases as ohdev deov where it was not necessary, ovSev 
 νρονηκομ avrois though it did not concern them, ςίρημένον although it 
 had heen said, δεδογμέι/ον after it had been resolved, δόξαν ταύτα when 
 this resolve had been taken, &c, have been sometimes regarded as no- 
 minatives absolute ; but this, as we shall afterwards see, is an error. 
 
 The nominative absolute, which is not unfrequent in English, especially 
 in poetry, is of a different kind from this ; e.g. ' And we being exceed- 
 ingly tossed with a tempest, the next day they lightened the ship.' — 
 Acts xxvii. 18. 
 
 These instances are not like the so-called Greek nominative absolute, 
 but like the genitive absolute. They have risen from the loss of case- 
 endings in English, exactly like the nom. absol. of Modern Greek. See 
 § 52 inf. 
 
 39. Copulative words (implying existence, seeming, being 
 called, chosen, &c.) take the same case after as before them 
 (as in English ' it is I,' &c.) ; as 
 
 καθέστηκε βασιλεύς he is appointed king, 
 θεός ώιομάζετο he was styled ' a god.' 
 
 So too άκονω in the sense / am called, as in εχθροί άκον- 
 ουσιν they are called enemies.* 
 
 N.B. Bopp connects the c, which is the common suffix of 
 the nominative, with the Sanskrit pronominal theme sa ' he,' 
 'that person there' (Co?np. Gram. § 134), from which root 
 the article is also derived. 
 
 THE VOCATIVE (Κλητική). 
 
 40. The vocative is the slightest of all cases, and has no 
 influence on the syntax. Hence in many languages it does 
 not exist at all ; even in Latin it is almost non-existent, for 
 the nominative is constantly used for it in the 2nd declension, 
 
 * So audio in Latin — ' Seu Jane libentius audis,' or whether you 
 prefer to be called Janus ; and in English, ' Do I hear ill of that side 
 too ? ' = Am I ill spoken of in that quarter also ? — (Ford.) 
 
 1 Or hear'st thou rather, pure etherial stream, 
 Whose fountain who shall tell ? ' — Par. Lost, iii. 6. 
 
 Cf. Spenser, F. Q. i. 23. 
 
70 A BRIEF GREEK SYNTAX. 
 
 in which alone it is found at all. Greek does not possess it in 
 neuter words, and even in some masculines, as ώ θεός, ώ φίλος, 
 ώ Ήέλιος ; and Buttmann observes further {Grain, p. 180), 
 that the nominative is used for it in all instances where its 
 occurrence would naturally be rare, e.g. ώ πους. 
 
 41. Hence too the nominative (especially with the article) 
 is often substituted for it, as 
 
 Ιημοβόρος βασιλεύς' επε) οντι^άνοισιν ανάσσεις people- 
 eating king ! since thou lordest it over weaklings. — 
 //. i. 231. 
 
 ώ οντος Α'ίας ho Ajax ! 
 
 συ 6 πρεσβύτατος you, the eldest. — Xen. Cyrop. iv. v. 17. 
 
 χαίρε 6 βασιλεύς των ΊουΙαίων hail, king of the Jews ! 
 
 Compare Degener ο populus. — Luc. ii. 11. Vos ο Pompilius sanguis. 
 — Hor. A. P. 293. 
 
 42. It is not unreasonable to conjecture that when a separate 
 form for the vocative exists, it is merely due to the change 
 produced in the nominative when used rapidly in calling or 
 addressing others ; in fact, that it is due like other phonetic 
 corruptions to what Prof. Muller calls ' muscular effeminacy.' 
 It usually contains the stem of the word, occasionally modified 
 by euphonic laws (Bopp, § 205). 
 
 THE GENITIVE {Γενική* 
 
 43. i. The name of this case is probably due to a simple 
 mistake. The Stoic grammarians called it π-ώσις γενική or 
 general case, because it expresses the genus or kind ; in fact, 
 there are many languages in which the genitive is directly 
 formed from the nominative by adding to it the adjectival 
 termination, and it is often a matter of indifference whether 
 we use an adjective or a genitive case, e.g. 'an aquatic bird* 
 is the same thing as ' a bird of the water.' 
 
 ii. The genitive termination is derived from dya or tya, the 
 pronominal root of the second person. Probably the termina- 
 tion was first used for adjectives (βημο-σιο-ς) before it was 
 adopted for the expression of genitival relations. 
 
 * Genitivus would have been a translation, not of yevucbs but of 
 γεννητικός. (See some valuable remarks on this point in Max iluller's 
 Lectures, i. 103-105.) Obviously, the Latin names of thiscase {genitivus, 
 patriots, possessivus, «fee.) cover but a very small part of its signification. 
 Some authors call it the wfance-case. The nomenclature of the cases is 
 very inadequate, though Priscian observes of it, 'Multas et diversas 
 unusquisque casus habet significationes, sed a notioribus et frequentior- 
 ibus acceperunt nominationem ' (lib. v. de Casu). 
 
THE GENITIVE. 71 
 
 44. All the multitudinous uses of the genitive are traceable 
 to its employment for the expression of three* main con- 
 ceptions; and these are so wide that they are often almost 
 interchangeable, — in fact, both ablation and partition fall in 
 reality under the head of relation. 
 
 1. Ablation, in which it is an ablative case, and corresponds 
 to the English ί from.'' 
 
 2. Partition, in which it implies ' some of? 
 
 3. Relation, in which it involves the notion of connection or 
 comparison, &c. The vagueness of this term is quite in accord- 
 ance with the essence of the genitive, of which the characteristic 
 suffixes in Greek are -oc, οι -o, derived from the Sanskrit 
 pronoun sya ; and of which the general function is ' to per- 
 sonify an object in attaching to it a secondary idea of local 
 relation' (Bopp, §§ 189, 194). 
 
 45. To the first head Ablation f belong the genitives of 
 cause, material, fulness, exclusion, motion from, perceptions, 
 both mental and physical (as derived from an object), &c. ; 
 a very little thought will show how these conceptions can be 
 arranged under this head, although some of them (e.g. full of, 
 made of, &c.) might be, from some points of view, equally 
 well arranged under the genitive of partition. The close 
 connection of the two classes of conceptions may be seen from 
 the possible interchanges of our i of and ί from,'' the German 
 von, the French de, and the Greek εξ and από. 
 
 Causal Genitives ; 
 
 κύματα παντοίων άνεμων waves caused by all kinds of 
 
 wind. 
 "Ηρας άλατείαι wanderings caused by Hera. 
 εάλωσαν προδοσίας they were condemned for treachery. 
 ενχωλής επιμέμφεται he blames me for a vow (unpaid). 
 χωόμενος γυναικός angry about the woman. 
 οϊμοι της τύχης % alas for my misfortune (Germ. Ο des 
 
 Leides ! and in vulgar French ' pauvre de moi '). 
 της μωρίας what folly ! 
 χρηστοϋ ανδρός excellent fellow! 
 
 * Donaldson, Gtr. Gram. p. 464 seqq. 
 
 f Although Greek has not a distinct ablative (άφαφβτική πτώσιε) like 
 the Latin, yet some Greek grammarians recognised the forms obpavoQev, 
 4μ4θεν as a sixth case. The name ablativus for the sixth case is believed / 
 to have been first used, if not invented, by Julius Ccesar, in his treatise 
 J)e Analogia, Lersch. ii. 231. 
 
 \ De is used after exclamations in Spanish, as Infeliz de mi ! ah poor 
 me ! Ay de mi hijo ! alas ! my poor son ί 
 
72 A BRIEF GREEK SYNTAX. 
 
 είτε τευ άγγελίης μετ ε μ ηλυθες ; didst thou visit me 
 
 for the sake of some message ? 
 τοϋ δ" εφυν εγώ from him I sprang. 
 κρατίστου πατρός τραφείς nurtured by a noble sire. 
 Σωκράτης 6 Σώφρον ίσκου Socrates the son of Sophron- 
 
 iscus. 
 
 Material;* 
 
 νόμισμα αργυρίου a coin of silver. 
 
 πώρινου λίθου ποιέειν τον vabv to build the temple of tuff. 
 Fulness, or Emptiness ; f 
 
 εκπωμα ο'ίνου a cup of wine. 
 
 άλις ΰε παίδων" but enough of sons! 
 
 πλήρης στεναγμών ουδέ δακρύων κενός full of groans, 
 nor void of tears. 
 
 1 Supplied of kernes and gallow-glasses.'— Mad: i. 2. 
 
 4 1 am provided of a torchbearer.' — Merch. of Ven. ii. 2. 
 Exclusion, or Separation ; J 
 
 άπέ-χομαι οίνου I abstain from wine. 
 
 λήγε χόλοιο cease from wrath (cf. Abstine irarum, 
 desine querelarum, &c, Hor.). 
 
 * It might be better perhaps to regard the genitive of material as 
 falling under the head of partition — something detached from the whole. 
 In Modern Greek it is expressed by από, as σπαθί avb £υλο a sword of 
 wood. 
 
 f So in English, ' empty of all good ' — Milton ; and in Italian, ' Dei 
 beni della fortuna abbondante.' — Boccaccio. With these we may range 
 genitives implying skill, ignorance, as μάχης e3 elSSre πάσης ; compare 
 * Pugnae sciens,' Hor. ; and Milton's ' Intelligent of seasons,' Par. Lost, 
 vii. 427 ; and 
 
 ' Yet oft his heart, divine of something ill, 
 Misgave him.' — Id. ix. 845 
 
 (' mens prsesaga futuri,' Claud.). Similarly in Italian, pratico, 'skilled 
 in,' takes a genitive, e.g. * pratichissimo di questa sorte d' antichita ' ; 
 and in Spanish, 'Dotado de ciencia,' gifted with learning; 'escasotZe 
 medios,' scanty in means. 
 
 X Here belong the genitives after compounds in a privative, as άφωνος 
 ipas, θ76υστΣ>5 κακών, απεπλος φαρέων Κςυκών, &παΐ5 τέκνων, &α, and the 
 Latin imitations ' Immodicus ira?,' Stat. Th. ii. 41 ; ' Immunis aratri,' 
 Ον. M. iii. 11 ; 'interritus leti,' Id. x. 616. "We have something like it 
 in English, as in Shakspeare's ' Un whipped of justice;' and Milton's 
 1 the teats Unsucked of lamb or kid ; ' and Keats' • Innumerable of hues 
 and splendid dyes ; ' and still more closely Sheridan, • The land-lord was 
 unfurnished of every kind of provisions.' — Life of Swift. It is probably 
 to an imitation of this idiom that we owe the much-abused line — ' Yet 
 virgin of Proserpina from Jove.' — Par. Lost, ix. 396. 
 
THE GENITIVE. 73 
 
 σφάλλομαι της έλπίοος I am balked of my hope. 
 
 ελεύθερος φόβυυ free from fear. 
 
 πλην y εμού except me. 
 
 άπηλλαγμαι της νόσου I am quit of the disease. 
 
 ημαρτον σκόπου I missed the mark. 
 
 Ίστασθε βάθρων get up from the steps. 
 
 άλλοθι γαίης elsewhere in the earth. 
 
 Motion from ; 
 
 γης οποίας ήλθον from what land I came. 
 Perceptions ; 
 
 οζονσι πίττης they smell of pitch. 
 
 ακούω του ΖιΙασκάλου I listen to the teacher. 
 
 και κωφού συνίημι I even understand the dumb. 
 
 46. Under the second head 'Genitive of Partition' fall those 
 which express time, possession, place, and all which can 
 possibly imply that the action affects Ά part of the object. 
 
 The following are all partitive genitives of one or other class ; 
 and with them may be compared such English expressions as 
 1 Oblong time,' Acts viii. 11 ; ' There be of them,' &c, Lev. 
 iv. 16: 
 
 και θέρους και χειμωνος both winter and summer.* 
 
 νυκτός by night, ημέρας by day. 
 
 συν σοι μετεϊχον τών "ίσων with thee I shared an equal 
 
 fortune. 
 συμβάλλεται ΰέ πολλά τοϋΐε Ιείματος many things con- 
 tribute to this terror. 
 εστίας μεσομφάλου εστηκεν ηΖη μήλα, JEsch. Ag. 1054, 
 already the victims stand on the central altarf (cf. 
 Soph. El. 900, έσχατης ορώ πυράς . . . βόστρυγον I see 
 on the mound's edge ... a curl). 
 τής γης ετεμον they laid waste some of the land. 
 κρητήρας επεστέψαντο ποτοΊο they crowned the goblets 
 with wine. J 
 
 * Comp. Italian, di notte ; French, de nuit ; German, Nachts, eines 
 Abends ; Spanish, de noche, &c. The English ' o' nights ' is probably ' on 
 nights.' See Morris, Specimen of Early English, p. Iv. 
 
 f The genitive of place is confined (mainly) to poetry, but is found in 
 the local adverbs ου, που, αυτού, &c. Cf. the German, Ich gehe des 
 
 % Buttmann, in his Lexilogus, shows that even the learned Virgil 
 misunderstood this genitive, and took it to mean ' they crowned (with 
 flowers) the goblets of wine ; ' hence his expressions ' Vina coronant,' 
 and ' Magnum cratera corona Induit implevitque mero.* 
 
 Ε 
 
74 A BRIEF GREEK SYNTAX. 
 
 βεβρωκως κρειων τε και αίματος battened on flesh and 
 
 gore.* 
 πάσσε <Γ άλυς and he sprinkled some salt over it. 
 γεΊρας νιφάμενος πολιης αλός washing his hands in the 
 
 foamy brine. 
 άλλ' εστί τοϋ λέγοντος but he is at the mercy of the 
 
 speaker. 
 •j"7roXic άνΰρυς εσ& ενός the state belongs to one man. 
 ουκ εστε εαυτών ye are not your own. 
 πολλής άνοιας εστί it is a matter of no slight folly. (Cf. 
 
 James iv. 1.) 
 ου παντός άν}ρος εις Κόρινθόν εσθ' 6 πλους it isn't every 
 
 man who can sail to Corinth. 
 ούτι μη λάχωσι τοϋΰε συμμάχου they shall certainly not 
 
 gain me as an ally. 
 ποΰών ελαβεν he grasped him by the feet. 
 κισσός δρυός εχεται the ivy clings to the oak. 
 γενείου αναμένος touching his beard. 
 εις τόο ημέρας to this day. — Eur. Phcen. 428. 
 εις τοΰτο κίνδυνου to such a pitch of danger. 
 
 47. Under the wide term of Genitives of Relation (which is 
 in point of fact merely a convenient term for such genitives 
 as do not obviously fall under the two other heads) are classed 
 those which express or involve comparison, J value, price, &c. 
 
 * Cf. the French 'manger de' and our 'eat of my venison,' or 'he 
 that drinketh of this water.' Similar is the Latin ' Implentur veteris 
 Bacchi, ping uisque f evince.' Many such idioms in Latin are mere imita- 
 tions of the Creek idiom, only admissible in the poetic style. They 
 abound in Silius Italicus, who has been called by Jani ' the great patron 
 of the genitive case.' 
 
 f The instances in which the possessive genitive sinks into a mere 
 epithet are few ; as in &ο~τρων ευφρόνη a night of stars, χιόνος πτέρνξ a 
 wing of enow, στολίί τρυφας a robe of luxury, τραύματα α'ίματος wounds 
 of blood. This is frequent in English poetry, as in Crabbe's 
 
 His cap of davkness on his head he placed. 
 His shoes of swiftness on his feet he braced. 
 His sword of sharpness in his hand he took, &c. 
 
 Cf. ' Nearer there grew no sticks of bigness.' — Fuller's Holy War. And 
 in Hebrew, as 'Ships of desire,' Job ix. 26 = pleasant ships, &c. 
 
 % Some may prefer to arrange the genitive of comparison under the 
 head of ablation, as in Latin ; in Modern Creek, comparison is expressed 
 by από, as δ καπνός due ελαφρότερος άπί> τον aipa. Sophocles, Mod. Gv. 
 Gram. p. 125. 'When two objects are compared, it is natural to say 
 that one is the better, &c. of the two, and it is an easy transition to say 
 that one is better of the other.' — Sir C. C. Lewis, Romance Languages, 
 
THE GENITIYE. 75 
 
 μείζων εστ\ του πατρός he is taller than his father. 
 διπλάσιος αυτός εωυτοϋ εγένετο became twice as great as 
 
 before. 
 οτε δεινότατος σαντοϋ ήσθα when you were at your best. 
 ουδενός δεύτερος second to none. 
 άλλα των δικαίων things other than what is just. 
 κρείσσον αγχόνης things worse than hanging. 
 δ~ια θεάων divine of goddesses.* 
 κρεισσόνων νικώμενοι conquered by superiors. 
 περιδόσθαι της κεφαλής to bet one's head. 
 άμείβειν χρύσεα χαλκείων to exchange golden for brazen. 
 κέκρισθε . . . μηδενός civ κέρδους τα κοινά των Έ,λλήνων 
 
 προέσθαι ye have determined that for no gain would 
 
 ye abandon the common interests of the Greeks. 
 πόσου τιμάται ; how much is it worth ? 
 τιμάται μοι υ άνηρ θανάτου he fixes my penalty at death. 
 θανάτου εδίωκε he brought a capital charge. 
 ώς είχον τάχους with all the speed they could. 
 ^χρημάτων ευ ηκοντες being well off for money. 
 μετρίως εχειν Φρενών to be fairly intelligent. 
 πώς έχεις της γνώμης; what do you think? 
 πώς αγώνος ηκομεν; how does the contest stand with us? 
 δ φόβος τών πολεμίων the fear of the enemy (i.e. which 
 
 they feel; subjective genitive). 
 This genitive of relation is common in English ; e.g. ' 'Tis pity of 
 him.' — Meas.for Meas. ii. 3. ' Eoses are fast flowers of their smells.' — 
 Bacon, Ess. 
 
 48. This last instance may also mean i the fear about the 
 enemy,' i.e. with respect to them. This is often called the 
 objective genitive. It may sometimes be regarded as causal; 
 but it usually belongs rather to the ablative meaning of the 
 genitive than to its meaning of relation. Other instances of 
 the so-called objective genitive are λΰσις θανάτου deliverance 
 
 p. 148. Compare the Italian pin ricco di me/ more rich than I ; ' meno 
 grande delta citta,' less large than the city, &c; 'in comparison of.' 
 Judg. viii. 2. 
 
 * Here the 87a is a quasi superlative ; compare Milton's ' Ο sovran, 
 virtuous, precious of all trees in Paradise.' — P. L. ix. 795. Virg. Mn. 
 iv. 576 : ' Sequimur te, sancte Deorum.' ' Ο prsestans animi juvenis.' 
 
 t Compare the Italian ' antico di sangue, nobile di costumi,' Boc- 
 caccio ; and the Spanish ' agudo de ingenio,' acute of intellect ; ' ancho de 
 boca,' wide of mouth, &c. Similar too are such genitives as ' holy and 
 humble men of heart,' ' Ancient o/'Days,' and in Chevy Chase — 
 ' For a better man of heart, nare of hande 
 "Was not in all the north countree.' 
 e2 
 
76 A BRIEF GREEK SYNTAX. 
 
 from death, άφορμη έργων a stimulus to deeds, άπόστασβς των 
 'Αθηναίων defection from the Athenians, πόθος vlov desire felt 
 by a son (subjective), or desire felt towards or in respect to a 
 son (objective). This possibility of a genitive being either 
 objective or subjective (amphibologia) leads occasionally to 
 uncertainty, e.g. ευαγγέλιο»' τον Χριστού may be either the 
 Gospel about Christ (objective), or which emanated from 
 Christ (subjective). The objective genitive is common in 
 Hebrew; and in Latin after injuria, metus, &c. Nor is it 
 unknown in English ; cf. Rom. x. 2, 'a zeal o/God.' Addison 
 has ' such of my readers as have a taste of \_=for~\ fine 
 writing.' Ή άγαπή τον θεον ΠΙΓΡ ΓΙ?Π^ amor Dei, Γ amore 
 di Dio, l'amour de Dieu, all involve the same ambiguity.* 
 
 49. Very frequently we find a double genitive after a word, 
 as Ζενς, οστ ανθρώπων ταμίης πολεμοιο τίτνκται Zeus who is 
 the arbiter of war for mortals. — II. iv. 84. For instances of 
 accumulated genitives see Rom. viii. 21, Rev. xvi. 19. 
 
 50. The Genitive Absolute properly falls under the causal 
 use of the genitive, as ορών τον χωρίον χαλεπού οντος τονς 
 τριηράρχονς . . . άποκνονντας seeing the captains hesitating 
 because the place was steep. It is therefore a genitive of 
 ablation, and so resembles the Latin ablative absolute. It is 
 used also however to express time and circumstance, as έμον 
 καθενΰοντος while I was sleeping, τοντων όντως εχόντων such 
 being the case, σαλπίζοντος while the trumpeter was blowing. 
 It derives its temporal and other meanings from the participle 
 with which it is joined. 
 
 51. This construction is less frequent than the ablative 
 absolute, because Greek possesses past participles active, and 
 Latin does not, e.g. ταντα ειπόντες άπ^μεν his dictis egredie- 
 bamur ; this could not be in Greek τοντων λεχθέντων, which 
 could only mean when this had been said by others. (Madvig ; 
 see too Nagelsbach, Lat. Stylistik, § 97.) 
 
 52. This genitive absolute is found in German, in such 
 phrases as l stehenden Fusses' 1 (Curtius). In Modern Greek 
 the nominative absolute has superseded it, as Άποθανόντας Ό 
 Σωκράτης b Πλάτωνας πήγε ες την Α'ίγνπτο Socrates being 
 dead Plato went away into Egypt. So too in English we use 
 the nominative absolute f where the Greek would require tha 
 
 * Crombie, Etym. and Synt. p. 34. 
 
 f The absolute objective case is much more rare in modern English, a j 
 
 1 him destroyed, 
 Or won to what may work his bitter loss.' — Milton. 
 
THE DATIVE. 77 
 
 genitive, and the Latin the ablative; as '/ being in the way, 
 the Lord led me,' Gen. xxiv. 27. But this nominative is due 
 to the loss of case-endings, i.e. it is not^ properly speaking, a 
 nominative, although in uninflected languages it has the same 
 form, e.g. 
 
 1 And by her side there sate a gentle paire 
 Of turtle doves, she sitting in an yvory chair.' — Spenser. 
 
 THE DATIVE (Αστική). 
 
 53. The fundamental conception of the dative case is juxta- 
 position. It corresponds both in the sing, and plur. to the 
 Sanskrit locative. The £, which is its characteristic suffix, 
 is used to indicate permanence in space and time, and is the 
 root of the demonstrative pronoun (Bopp, §§ 177, 201). 
 
 Hence the dative is diametrically opposed to the genitive, 
 of which the fundamental conception is* ablation. Thus the 
 dative is used with εν, συν, επί; the genitive with ε£, από. 
 
 a. The dative signifies proximity, the genitive separation ; as 
 ΙΙολυκράτει ώμίλησε he associated with Polycrates ; 
 
 but πάλιν τράπεϋ' νιος εοϊο he turned back f?*om his son. 
 
 b. The dative denotes addition, the genitive subtraction ; as 
 ΙίΙωμί σοι τα χρήματα I give the money to you, 
 
 but δέχομαι σου τα -χρήματα I receive the money from you. 
 
 c. The dative expresses equality or sameness, the genitive 
 comparison of things different ; as 
 
 ουτός εστίν ό αντος εκείνω this man is the same as that. 
 επιστήμη επιστήμης διάφορος one science differs from 
 another.* 
 
 53 (bis). It will be seen from the following remarks that 
 the dative is an eminently syncretistic case (see § 37), being 
 both dative, instrumental, locative, and comitative. 
 
 The him here is a dative ; the Anglo-Saxon having no ablative, used 
 instead the dative absolute ; e.g. vp-a-sprungenre sunnan, the sun 
 having risen. See Latham, The Engl. Language, ii. 437. So we find in 
 Wiclif s Bible (Matt, viii.), • and hym seen, thei preiden hym that he 
 shulde pass fro her coostis,' which becomes in Tyndale's Bible, ' when 
 they sawe him.' This dat. absolute is of constant occurrence in 
 Wiclif, ' And hem gadrid togidre, he seide to hem.' — Mark iii. 23 ; 
 vi. 20, &c. 
 
 * Donaldson's Gr. Gram., p. 486. Horace imitates this use of the 
 dative with idem — ' Invitum qui servat idem facit occidenti,' which 
 might be in Greek rahrh ποιέι τψ κπίνοντι. Burnouf, p. 257. 
 
78 A EKIEE GREEK SYNTAX. 
 
 54. Hence the dative expresses accidents, accessories, cir- 
 cumstances, instruments ; as 
 
 1. Place. We have already seen traces of the locative case 
 in the dative, in such phrases as Μαραθώ rt at Marathon, οίκοι 
 at home. Thus we find in the poets — 
 
 τΰζ ώμοισιν έχων having his bow on his shoulder. 
 
 αίθέρι ναίων dwelling in the sky. 
 
 μίμνει άγρω he is staying in the country. 
 
 But in prose, and even in poetry, the preposition εν is usually 
 added to express place. 
 
 2. Time. Though iv is not so frequent with the locative of 
 time, it may be used ; as 
 
 τί] τρίτη ημέρα on the third day. 
 
 rrj νουμηνία on the first of the month. 
 
 iv τω παρόντι in present circumstances. 
 
 3. The manner of a thing, i.e. limit, specification, accom- 
 paniment, resemblance ; as 
 
 βίο: εσιέναι to enter by force (so σπουΰτ}, oiyy, έργω, τω 
 
 οντι, ί£/α). 
 γίνει "Ελλην by race a Greek. 
 ναυσιν ϊσχυειν to be strong in ships. 
 
 κατεστρατοπείευσατο τω ττεζω he encamped with the foot. 
 το~ις κακοΊς ομιλών associating with the bad. 
 ΰούλω ϊοικας you are like a slave. 
 
 N.B. The dative of accompaniment is more usually expressed 
 by συν, except when αυτός is used ; as 
 
 τήλ' αυτϊ} πηληια κάρη βάλε he flung away the head 
 
 helmet and all. 
 μίαν ναϋν ελαβον αυτοίς ανΰράσιν they took one ship 
 crew and all. 
 
 And συν may be used even with αυτός, as άνόρουσεν Άχιλ- 
 Χεύς αυτή συν φόρμιγγι uprose Achilles, harp in hand. 
 
 4. Instruments of all kinds, as κάμνειν νόσω, ττατάσσειν 
 ράβΖω, ώθεΊν ταΊς χερσίν, πολεμώ προσκτάσθαι. 
 
 Hence with such verbs as χρήσθαι, αίσχυνεσθαι, λυπεϊσθαι, 
 τεκμαίρεσθαι, &c. 
 
 N.B. The English ' with ' is also both instrumental and 
 comitative, e.g. ' I went with him/ ' I cut with a knife.' 
 — Schleicher, Compend. p. 577. 
 
 5. Agents, as being in one point of view instruments; thus 
 
 
THE DATIVE. 79 
 
 after passive verbs we may have either νπο with the genitive, 
 or the dative ; as 
 
 προσπόλοις φυλάσσεται he is guarded by attendants. 
 ταϋτα Χέλεκται ημ'ιν these things have been said by ns* 
 
 (or b(f ημών}, 
 τί πεπρακται τοις άλλοις ; what has been done by the 
 others? (or νπο των άλλων, just as in Latin poetry, 
 1 Non intelligor ulli ' or ah ullo ; ' cui non sunt audita?,' 
 or a quo, &c). 
 6. General reference, advantage, and disadvantage. 
 Hence with such verbs as ΰίίωμι, νπισχνονμαι, πιστεύω, ειμί, 
 άρηγω, νπακυΰω,^ υπηρετώ, ηγούμαι, μάχομαι, πολεμώ, &C. } 
 after each verb it expresses the remote or indirect object. 
 εστί μοι I have.;}: 
 
 εγώ σιωπώ τώΰε ; am I to hold my tongue for this fellow? 
 τωΐε δ" οίχομαι as far as he is concerned, I am dead. 
 ΰέξατό οι σκηπτρον he received at his hand the sceptre. 
 W αριστερά εσπλέοντι to the left as one sails in. 
 άνάζιαι γαρ πασίν εστε Ιυστυ-χεϊν ye are unworthy of 
 misfortune in the judgment of all. — Soph. 0. C. 1446. § 
 
 * Burnouf compares the French ' 6 est bien dit a vous? 
 
 f The verb 'to obey' used to take a dative in English, no less than 
 in Greek and Latin ; e.g. ' That as a harp obeyeth to the hand.' — Chaucer, 
 Legend of Women. ' Yet to their general's voice they soon obeyed.' 
 — Milton, Par. Lost, i. 337. Comp. Spenser, F. Q. in. xi. 35. In fact, 
 verbs of advantage, disadvantage, &c. govern a dative in English no less 
 than in Greek and Latin, only in English the datival inflection has 
 disappeared. ' If you please ' is really as much a dative as ' si tibi 
 placet.' Cf. wethinks with Ζοκεΐ μοι, and the Anglo-Saxon poet (5e seolfum 
 mislicaft with % απαρέσκει σοι. The following are instances : ' Beleve 
 yee to the gospel,' Wiclif, Mk. i. 15 ; ■ thretenyde to hym,' id. v. 25; 
 f commaundith to unclene spirits,' id. 27 ; ' the wind and the see obey- 
 ghen to hym, iv. 40 ; 'pleside to Eroude,' vi. 22, &c. Even in our ver- 
 sion we read ' answered him to never a word.' — Matth. xxvii. 14. 
 
 \ Thus the dative as well as the genitive may be used to express pos- 
 session. In Hebrew ? ' to ' is used for possession, and the Gascon says 
 ' la fille a Mr. N.' instead of de. In Greek such a phrase as ή κεφαλή 
 τφ ανθρώπω for του ανθρώπου was called the σχήμα Κολοψώνιον. Lesbonax 
 περί οχημάτων, ρ. 181. The collocation is rather clumsy, but similar 
 phrases are common, as αναίρεσιν rdis νεκροΊε, Time. vi. 18; αναθήματα 
 Κροίσω, Hdt. ii. 113. 
 
 § Gf. άξιον yap Έλλάδί, Ar. Ach. 8 ; ημ'ιν δ' Άχιλλευς &£ios τιμής, 
 γύναι, Eur. Hcc. 313; and many other instances in Bernhardy, Gricch. 
 Synt. S. 78. Under this head fall such phrases as ol -πρεσβύτεροι outoTs 
 των ει /dai μόνων, Thuc. i. 6. αυτω is frequently used in this way in Thuc. 
 and Plato ; and sibi has a somewhat similar redundancy in some Latin 
 sentences. 
 
80 A BRIEF GREEK SYNTAX. 
 
 This is especially found with various participles ; as 
 
 ci σοι βυυλομενω εστί if you please (cf. Tac. Agric. 18, 
 
 1 Quibus bellum volentibus erat '). 
 συνελόντι είπεϊν to speak briefly. 
 εμοϊ Ιε κεν ασμενω ε'Ίη I should be glad of it. 
 θέλοντι κάμοϊ τοϋτ αν ήν I too should have wished for 
 
 this. 
 ως εμοί, or ως γ' εμοϊ κρίτη meo quidem judicio. 
 
 55. To this dative of reference belongs what is called the 
 ethic (i.e. emotional) dative ; the apparently superfluous intro- 
 duction of personal pronouns to show the speaker's or hearer's 
 interest in what is said ; as 
 
 μη μοί γε, μη μοι, μη ΰιασκανΰικίσης don't, dont't, I beg 
 
 of you, dose me with cabbage. 
 ώ μήτερ, ως καλός μοι 6 -πάππος bless me ! mother, how 
 
 handsome my grandfather is. — Xen. Cyr. i. 32. 
 άλλα μοι εσθίεμεν και πίνεμεν but eat, I pray you, and 
 
 drink, 
 οδ' έίμ εγώ σοι κείνος look you, I am that famous man. 
 
 N.B. a. The same use is found in Modern Greek, where 
 however the dative case has disappeared and resigned its 
 functions to the genitive, as συν τον ετίναζαν ενα κάλο ραβδί 
 they thrashed him soundly — / know you are pleased to hear it. 
 See Sophocles, Mod. Gr. Gram. p. 151. 
 
 β. This ethic dative is common in other languages ; as 
 
 ' At tibi repente . . . venit ad me Caninius ! lo you of a 
 sudden comes Caninius to me ! ' — Cic. 
 
 Quid mihi Celsus agit ? what is my Celsus doing ? — Hor. 
 
 Non mihi bellus homo es / don't think you a good- 
 looking person. 
 
 Es lief mir ein Hund iiber den Weg there ran me a dog 
 across the road.* 
 
 1 Afin qu'il fut plus frais et de meilleur debit 
 
 On lui lia les pieds, on vous le suspendit.' — Fenelon, 
 Fables, iii. 1. 
 
 y. It was extremely common in English, e.g. 
 
 * Look how this river comes me cranking in.' — Henry IV. 
 
 * ' Einen Apfel schiesst der Vater dir vom Baum auf hundert Schritte.' 
 My father shoots you an apple from a tree at a hundred yards. — Schiller, 
 Tell, 
 
THE ACCUSATIVE. 81 
 
 1 This scull lias lain you in the ground these three years.' 
 
 — Hamlet.* 
 4 Tour serpent of Egypt is lord now of your mud,' &c. — 
 
 Ant. and Chop. ii. 7. 
 
 It is not unknown even in modern writers ; e.g. in Taylor's 
 4 Philip von Artevelde ' we have 
 1 Mount me a messenger.' 
 
 I Gag me this graybeard.' 
 
 i And twinkling me his dagger in the sun.' 
 
 I I might eat four hoofs of an ox yet my stomach would 
 flap you, look you, and droop you, look you, like an 
 empty sail.' 
 
 This latter phrase, ' look you ' (or ' for you '), is the most 
 common modern substitute for the Ethic Dative. 
 
 THE ACCUSATIVE (Αίηαπιφ^ 
 
 56. i. The accusative is probably, next to the vocative, the 
 oldest of the cases, as is seen from the fact that its charac- 
 teristic suffix m appears even in the nominative of pronouns, 
 as aham iywv, tvam Boeot. τοΰν, ide?ft, &c. This suffix pro- 
 bably acted the part of an article, i.e. it called attention to the 
 word to which it was attached. See Ferrar, Comp. Gram. 
 p. 211. 
 
 ii. The ους of the accus. plur. is a relic of νς, which is 
 preserved in Gothic, vulfans, sununs, &c. (cf. τΰπτονσι = 
 τνπτοντι). It was preserved in the Cretan and Argive dialects, 
 τονς (Goth, thans) ; and in Borussian deiwans = deos (Breal, 
 Bopp, ii. 55 ; Ahrens, De Dialect, ii. § 14, 1). 
 
 56 (bis). The fundamental conception of the accusative is 
 
 * In the Taming of the Shrew, Act i. sc. 2, G-rumio affects to mis- 
 understand it. 
 
 4 Petr. Villain, I say, knock me here soundly. 
 Gram. Knock you here, sir ; why, sir, what am I, sir, that I should 
 
 knock you here, sir ? 
 Petr. Villain, I say, knock me at this gate 
 
 And rap me well, or I'll knock your knave's pate.' 
 
 t Varro renders this ' accusandei casus,' deriving it from αΐτιάσμαι 
 I accuse ; but more probably it comes from αιτία, a cause. Hence Pris- 
 cian calls it causativus. See Trendelenburg, Act. Soc. Grcec. 1836, i. 119 
 seqq.; Lersch, Sprachphil. d. Alten, ii. 186. The characteristic suffix of 
 the accusative is in Greek v, in Sanskrit and Latin m ; for its pronominal 
 origin, seo Bopp, § 156. 
 
 B3 
 
82 A ERIEF GREEK SYNTAX. 
 
 motion towards, and therefore also extension over space. It is 
 the case To ichich* and is therefore put after transitive^ verbs 
 to express the end of the motion or action ; as τύπτω avrbv I 
 strike him, i.e. the direction of my blow is towards him. It 
 also expresses the action itself, as τύπτω πληγή ν I strike a 
 blow. Three accusatives may occur after one verb, in each 
 of which this fundamental conception is discernible, as νύκτα 
 αγγέλους Άθ/|ΐ'«ς επεμπεν he was sending messengers all 
 night long towards Athens. (Compare ' docere aliquem phi- 
 losophiam aliquot annos.') 
 
 57. In accordance therefore with the idea of the case (motion 
 towards^ and extension over) it expresses 
 
 1. Space, as απέχει πεντήκοντα, σταΰίους it is fifty stades 
 distant. 
 
 2. Time, as τρεΊς μήνας εμεινεν he stayed three months. 
 
 3. Any notion cognate to, i.e. connected in meaning \ with 
 that of, the verb, even when the verb is neuter, as κακίστην 
 Ζουλείαν εΰοΰλενσε he served the worst slavery. 
 
 This cognate notion is capable of a very considerable ex- 
 tension, as in 
 
 στε~ιγε γύας go to the fields. — Eur. Med. 668. (Comp. Go 
 home ; but even this phrase has become analytic in the 
 American ' Go to home,' and the Cornish ' Is she to 
 home ? ') 
 ^πολλούς αγώνας εζιών going out for many contests. — ■" 
 Soph. Tr. 185. 
 
 * Donaldson connects the form δε in accusatives like ΟυΧυμπόνΰζ 
 ■with δύο, just as in English two, too, to, are different stages of the same 
 word. 
 
 .. f The particle eth which so often precedes the accusative in Hebrew 
 signifies towards. The same fact is well illustrated in Spanish, where, 
 by' a strong extension of the analytic tendency, the preposition a usually 
 precedes the accusative if it expresses a person ; e.g. • Amar a Dios,' to 
 love [to or towards] God ; ' Cain mato a Abel,' Cain killed Abel, &c. 
 
 \ This form of the cognate accusative {-πόΧΐμον ποΧ^μύν, &c.) is called 
 Figura etymologica. See Lobeck, Paralip. Gram. Grcec. dissert, viii. 
 
 § Cf. the Latin exsequias, suppetias, infitias ire ; and see Lobeck's 
 note to Soph. Aj. 290, and Curtius's Erlauterimgen, 163. Milton, who 
 has left few classical idioms unadapted, even ventures on the cognate 
 accusative after a neuter verb of motion : 
 
 ' Upborne with indefatigable wings 
 Over the vast abrupt, ere he arrive 
 The happy isle.' — Par. Lost, ii. 410. 
 
 And ' Whatever creeps the ground, Insect or worm.' — Id. vii. 475. Early 
 English admitted a wider use of the accusative than modern; e.g. w 
 
THE ACCUSATIVE. 83 
 
 ομνυμι τους θεούς I swear by the gods. 
 
 νικάν 'Ολυμπία to win in the Olympic games. 
 
 βλέπειν νάπν, ομφακας, ναύφρακτον to look mustard and 
 
 cress, sour grapes, a three-decker.* 
 γραφήν ΰιώκειν to bring an action. 
 Γ4 ϊ>ήτα ποίμναις την&επεμπίπτει βάσιν ; why did he thus 
 
 rush striding (=εμπεσων βαίνει) on the nocks? — Soph. 
 
 Aj. 42 (πόΰα and χέρα are frequently thus used).| 
 
 4. It defines or localises the action of the word to which it 
 is joined, i.e. in strict accordance with the idea of the case, it 
 expresses the extent affected by the word on which it depends. 
 
 άλγώ την κεφαλήν I have a headache. 
 
 τούτου μάλλον την ψύσιν ίστϊ its nature is rather of this 
 
 kind. — Arist. Meteor, iv. 4. 
 πνρίτης την τέγνην a smith by trade. 
 κάλος τα 'όμματα with beautiful eyes. 
 ΰεινοϊ μάχη ν skilled in battle. 
 ονοείς άπαντα σοφός no one is wise in everything. 
 
 These and similar instances used to be explained by the 
 ellipse of κατά ; the fact is however the very reverse, since the 
 case expresses these conceptions by its own natural force and 
 meaning, and when κατά is expressed it is due to the analysing 
 tendency of all language in its progress from its original con- 
 dition. The superfluous preposition only shows that the true 
 meaning of the case is a little worn out. 
 
 find in Wiclif s version of the Bible, ' Blessid be thei that hungren and 
 thirsten rightioisnesse ; ' and in Milton, ' I gazed the ample sky.' 
 
 * This is a favourite idiom of Aristophanes ; he even uses it with a 
 neuter participle, as κλέπτυν βλέπει he looks thievish ; and with an infi- 
 nitive, as τιμαν β\4πω. — Ach. 879. Theocritus has the exquisite ex- 
 pression tap δρόωσα looking spring. — Id. xiii. 45. So we talk of ' look- 
 ing daggers? ' a vinegar aspect.' 
 
 t αίσσω means I rush, yet Sophocles (Aj. 40) has irpos τι δυσλό-γιστον 
 ωδ" γξζν χέρα ; ' for what inexplicable cause did he thus rush (i.e. wield) 
 his hand ? ' This accusative describing the residt of the verbal notion 
 is common in English ; e.g. ' to walk a horse,' ' to dance a baby,' ' to boil 
 a kettle,' &c. Cf. Spenser, F. Q. i. i. 17. Such verbs are said to be 
 used factitively, and, as in Hebrew, all absolute verbs admit this cau- 
 sative use. (Ewald, Hebr. Gram. § 102, and Lobeck, ad Aj. 40.) Latin 
 uses the accusative in the same bold manner in apposition with the 
 notion contained in the verb, and expressing the extent affected by it, 
 as in ' pedibus plaudunt choreas,' Virg. Mn. vi. 664; 'Bacchanalia 
 vivunt,' Juy. &c. ι Comp. Par. Lost, i. 723, ' The ascending pile Stood 
 fixed her stately height.' See Abbott, Shaksp. Gram. p. 69. 
 
84 A BRIEF GREEK SYNTAX. 
 
 57 (bis). Curtius, &c, call this cognate accusative, the accus. 
 of the inner object. It is either, (i.) immediately cognate, as 
 μάχην εμάχοντο, or (ii.) indirectly cognate, as τύπτεται πλη- 
 γήν, or (iii.) it defines the verb, as νόσους κάμνει, or (iv.) 
 it gives the result of the verb, as άγγελίην έλθέΐν. Often 
 (especially in poetry) a neuter accus. specialises a verb 
 almost like an adverb ; e.g. μέγα ψεύδεται, παΐσον ϊιπλήν, 
 &c. — Curtius. 
 
 58. As some verbs may have two objects, a nearer and a 
 more remote, a person and a thing, an external object and 
 an internal, such verbs (especially those of asking, teaching, 
 clothing, depriving, doing good or ill to) may take a double 
 accusative.* 
 
 ε£ίΓ)α£α τον παΐΖα την μουσικην I taught the boy music. 
 Θηβαίους χρήματα φτησαν they asked the Thebans for 
 money. 
 
 59. In one large class of instances in which there is ap- 
 parently a double accusative, one of the two may be regarded 
 as being in apposition with the other, and defines it; this 
 is called the ' whole and part figure,' σχήμα καθ' όλον και 
 μέρος, as 
 
 μέθες με προς θεών χείρα by the gods, let go my hand 
 
 [lit. release me, that is my hand~\. 
 Ύρωας ΰέ τρόμος αίνος υπηλυθε γυ'ια έκαστος dread tremor 
 
 invaded each Trojan's limbs [lit. the Trojans, each one, 
 
 as to his limbs]. 
 
 60. The accusative of the thing still remains when the verb 
 itself is the passive, as 
 
 άφγρημαι τον "ιππον I have been robbed of my horse. 
 πεπίστευμαι το ευαγγέλιον I have been entrusted with the 
 gospel. 
 
 61. The accusative is sometimes put in apposition to the 
 sentence, as 
 
 ΈΧένην κτάνωμεν, Μενελέω Χύπην πικραν let us kill Helen, 
 
 a bitter grief to Menelaus. 
 ρίψει απο πύργου, Χυγρον οΧεθρον you will be flung from 
 
 a tower, a terrible death. 
 
 * In such instances one of the accusatives expresses the object directly 
 affected by the verb, and the other expresses some notion cognate to the 
 meaning of the vorb, 
 
DOUBLE ACCUSATIVE. 85 
 
 62. The verb on which an accusative depends is often 
 omitted,* as in 
 
 σε ΰή σε την νεύουσαν ες πείον κάρα (sc. λέγω). — Soph. 
 
 Ant. 441 (cf. Aj. 1228). μη τριβας If — Soph. Ant. 
 
 577. ονκ είς ολεθρον. — 0. It. 415. Finem inquit inter- 
 
 rogandi ! — Cic. 
 άλλα τις χρεία σ εμοϋ (sc. έχει) ; — Eur. Hec. 976. 
 
 63. Not unfrequently the nominative of a dependent clause 
 is anticipated by being made the accusative of a principal 
 clause, as 
 
 ταρβεϊν τον εν πράσσοντα μη σφαλφ ποτέ to dread the 
 prosperous man, lest he should slip. 
 This is called Antiptosis, and is also found in Latin, as 
 * Nosti Marcellwn quarn tardus sit.' You know Marcellus 
 
 how slow he is. — Cic. 
 4 Earn veretur, ne perierit.' He fears her lest she should 
 perish. — Plaut. 
 And in English, as 
 
 ; I know thee, who thou art.' — Luke iv. 34. 
 ' Conceal me what I am.' — Shakspeare, Twelfth Night, i. 2. 
 1 Didst thou not mark the king, what words he spake V — ■ 
 King Richard II. v. 4 (cf. id. iii. 3 ; Merchant of 
 Venice, iv. 1). 
 This may be called the accus. of the redundant object. 
 
 . 64. Sometimes this accusative is placed first in the sentence, 
 and is called by some the accusativus de quo, as 
 
 τους κρίτας α κερΰαίνουσι βυνλόμεσθ' υμ~ιν φράσαι the 
 judges, what they get, we want to tell you. — Ar. Nub. 
 1113. 
 Χαιρεφώντα άνηρετο ψύλλαν οπόσονς αλλοιτο τους αυτής 
 ποίας ; he asked Chaerephon — a flea, how many of its 
 own feet it jumped ? 
 
 So in Latin, Urbem quam statuo vestra est. — Virg. JEn. i. 
 577. Cf. Is. i. 7, ' Your land, strangers devour it in your 
 presence.' 
 
 65. i. The accusative is used absolutely,•) - chiefly in the case 
 
 * The verb thus omitted is often some subjective conception, like 
 ' knowing,' &c. ; e.g. rjjueXet ws ανδροψόνου, κα\ obShv hv πράγμα €t καΐ 
 αποθώοι. — Plat. Euthyph. 4. d. 
 
 f The accusative absolute, when the expression is not adverbial or 
 impersoDal, is very rare, as in τ4κν' et φαν4ντ' άελπτα μ-ηκύνω Koyou. 
 
8b A BRIEF GREEK SYNTAX. 
 
 of certain participles, as Ιόζαν ταϋτα on this decision, προσήκον 
 it being fit, είόν, -παρόν, whilst it is allowed, &c; and in certain 
 neuter adverbial expressions like -ίνα τρόπον ; how ? πρύφασιν 
 in pretext, ε μι) ν χάριν for my sake, αμφότερα both ways, το 
 λοιπόν for the future, &c. (Cf. the use of wc in Soph. (Ed. 
 Tyr. 101 ; (Ed. Col 407.) 
 
 It is less correct to regard Ζόζαν, &c. as nominatives abso- 
 lute, since, as we have seen already, neuters have, properly 
 speaking, no nominative. They are rather adverbial inde- 
 clinable expressions, in which however the accusatival con- 
 ception of duration may generally be detected. 
 
 ii. ο, ά, τοϋτυ, εκεηο (like the Latin Quod in adjurations, as 
 Quod per te lacrimas oro, &c), sometimes mean wherefore, 
 therefore with the same sense as δι' ο, as in Eur. Hec. 13, &c. ; 
 and in the phrase αυτά ταϋτα ήκω I have come for this very 
 purpose. See Phcen. 145, 263 ; Thuc. ii. 40, iii. 12, &c. 
 
 Contrasted Meanings of the Cases. 
 
 66. ' From this examination, the learner may derive brief 
 rules as to the meaning of the cases. 
 
 The genitive denotes motion from, and separation. 
 The dative „ rest in, and conjunction. 
 
 The accusative „ motion to, and approach.' — 
 
 Donaldson. 
 
 67. The so-called ' absolute ' use of the cases springs from 
 their simple meanings ; e.g. 
 
 The genitive absolute expresses time as a cause τοϋ έαρος 
 
 ελθόντος τα άνθη βάλλει when spring comes the flowers 
 
 bloom. 
 The dative absolute represents time considered as a point, 
 
 as περιιόντι τω ενιαντω at the return of the year. 
 The accusative absolute, duration in time, as ταϋτην την 
 
 νύκτα during this night. 
 
 68. A few instances in which the distinctions of the cases 
 are brought into prominence or contrast, are added. 
 
 νυκτός during the night ; noctu (part.). 
 
 νύκτα all night ; ζ noctem ; ' answering the question ' how 
 
 long?' 
 νυκτι in the night ; node ; answering the question ' when ? ' 
 ημέρας during the day (part.). 
 ημέραν throughout the day (duration). 
 ήμερη, in the day time (limit). 
 
ADJECTIVAL IDIOMS. 87 
 
 πέντε μνών worth five minae, as a price (relation). 
 πέντε μναϊς worth five minae, as an instrument, 
 πέντε μνάς five minae {extension over a certain value). 
 πόσου πωλείς ; at how much do you sell ? (cause). 
 πόσω ώνεΊίοτ (=with) how much do you buy (instrument). 
 πόσον δύναται ; how much is it worth ? (extension). 
 τέρπομαι τούτου I am delighted for this (cause). 
 
 „ τούτω I am delighted with this (instrument). 
 
 „ τοντο I am delighted at this (cognate notion = 
 τοϋτο χάρμα). 
 πάρα του βασιλέως from the king (motion). 
 πάρα τω βασιλεϊ with the king (rest). 
 πάρα τον βασιλέα to the king (approach). 
 προοραν του πολέμου to provide about the Avar. 
 
 „ τω πολέμω to provide for the war. 
 
 „ τον πόλεμον to foresee the Avar. 
 μεθίημί σε I dismiss you; μεθίεμαί σου I let go of you. 
 ελαβόν σε I caught you; ελαβόμην σου I seized hold of 
 
 you. 
 
 έχειν ri to possess a thing ; έχομαι βρετέων I cling to the 
 
 images. 
 ηψε βρόχους he fastened nooses; ήψατο του τείχους he 
 
 grasped the Avail. 
 ώρεζε την κύλικα he held out the cup ; ου παιΰος όρέζατυ 
 
 he yearned for his son. 
 
 ADJECTIVES. 
 
 . 69. The chief peculiarities in the use of adjectives will here 
 be given, and a line of explanation appended when required. 
 
 i. πολλά τε και κακά ελεγεν he uttered many reproaches. 
 συνεΐϋως αυτω πολλά και πονηρά being conscious of many 
 wicked deeds. 
 
 The Greek and Latin idioms require ( many and wicked,' &c. 
 
 ii. πτανόν ΙΊω-γμα πώλων winged pursuit of steeds, i.e. pur- 
 suit of winged steeds. 
 
 λευκοπηχεις κτύποι χερών white-armed clappings of hands, 
 i.e. clappings of white-armed hands. 
 
 ypalai οσσων πηγαΐ aged fountains of eyes, i.e. tears from 
 aged eyes. 
 
 πολιάς πόντου θινός of the hoary sea-beach, i.e. beach of 
 the hoary sea. ■ 
 
88 A BRIEF GKEEK SYNTAX. 
 
 Compare ' Sansfoye's dead dowry f i.e. the dowry of dead 
 Sansfoye. — Spenser, F. Q. I. iv. 51. 
 
 It will be seen from these instances that the adjective is 
 liable to a strange inversion* of order, agreeing with the 
 wrong word, or rather with the whole notion implied. This 
 is an instance of the constructio ad sensum, and is called 
 Hypallage. Bold as these inversions are they may be pa- 
 ralleled in English by such expressions as ■ his all-obeying 
 breath,' ' tearfalling pity,' ' the church-going bell.' Words- 
 worth's severe criticism of the latter expression was mis- 
 placed. (See next page.) 
 
 iii. Σκνθην ες όϊμον to the Scythian track (=Σκυθι κήν). 
 την 'Ελλάδα φωνήν εξέμαβον I learned the Greek tongue 
 (=' Ελληνική ν). 
 
 Here we see that substantives (especially the names of 
 countries) are sometimes used adjectivally, as in the Latin 
 Asia prata, Virg. G. i. 383 ; Aquae Baice, Prop. I. xi. 30 f ; 
 and our India rubber, Russia leather, China bowl, Turkey 
 carpet, &c. All such phrases, ' a labouring day,' ' a walking 
 stick,' ' a riding whip,' ' a fox-hunting country,' fall under the 
 same head : the two substantives are in apposition, and one 
 qualifies the other. A substantive in apposition often defines 
 another in an adjectival way, as άνηρ βασιλεύς, άνήρ ναύτης, 
 άνθρωπος γεωργός, &C. ; as in the Latin hostes turmse, Stat. 
 Th. xi. 22 ; Fahulm manes, Hor. Od. I. iv. 16 ; and our a sailor 
 man, a butcher fellow, a warrior host, &c. 
 
 iv. Νεστορέη πάρα νηϊ by the Nestorean ship (i.e. Nestor's). 
 Βερενε ικεία θυγάτηρ Bereniceian daughter (i.e. of Bere- 
 nice). 
 νόστιμον ημαρ returning day, i.e. day of return. 
 
 * In Latin we find ' Alexandra Phrygio sub pectore,' Lucret. i. 475, 
 and ' Nerneeeus hiatus Leonis,' id. 24. We have something like it in 
 Ossian, * The hunter's early eye.' Carlyle, in his French Revolution, 
 speaks of ' the housemaid with early broom.' 
 
 The genitive may be even involved in the epithet, as δξύχαρ ktvttos 
 a sharp clapping of hands. See Lobeck's Aj. p. 63, on epithets in 
 general. Often, by a kind of metonymy, the adjective represents the 
 general conception or result of the substantive, as ' pallida mors,' χλωρών 
 ticos, ' Rugosum piper et pallenth grana cumini,' Pers.; 'vulnera despe- 
 rantia' Plin. ; ' As messenger of Morpheus on them cast sweet slom- 
 bring deaw,' Spenser, F. Q. i. i. 30 ; 'the sleepy drench Of th&tforgetftd 
 lake.'— Milton, P. L. ii. 74, &c. 
 
 f See Jani's Art of Poetry, Engl. Tr. p. 44. 
 
ADJECTIVAL IDIOMS. 89 
 
 In all such instances the adjective is used for the genitive 
 
 of the noun ; as in Milton's 
 
 1 Above the flight of Pegasean wing.' — Par. Lost, vii. 4 ; 
 
 and in Tennyson's 
 
 1 A Nioheian daughter, one arm out 
 
 Appealing to the bolts of heaven.' — The Princess, 
 
 v. haira π ενόντα Ζειελινοί they in the evening were preparing 
 their meal. 
 σκοταϊος* ήλθε ν he came in the dark, 
 τεταρτάϊος άφίκετο he arrived on the fourth day. 
 ορκιός σοι λέγω I tell you on oath. 
 
 Hence observe that the Greek uses adjectives in many in- 
 stances in which we use prepositions with a substantive, and 
 that this is especially the case in expressions of time. Compare 
 the Latin 
 
 'iEneas se matutinus agebat' was bestirring himself in 
 
 the morning. 
 Hesterni Quirites citizens of yesterday. 
 Domesticus otior I am at ease in my home. 
 
 We have precisely the same idiom in English, as 
 
 1 Gently they laid them down as evening sheep.'' — Dryden. 
 * The nightly hunter lifting up his eyes,' &c. — Words- 
 worth. 
 1 The noonday nightingales.' — Shelley. 
 
 VI. Ζηλη η οικονομία ετι οτι κατά σπονΖήν εγενετυ it is still 
 
 evident on the face of it that the building was hurriedly 
 
 done. 
 Ζήλος εστίν ως τι Ζρασείων κακόν it is evident that he means 
 
 some mischief. 
 στέργων φανερός ήν ουΖενα it was obvious that he loved 
 
 no one. 
 
 The Greeks are much less fond than ourselves of the 
 impersonal^ construction; they substitute the personal con- 
 struction for it. (There is no true impersonal in Greek ; either 
 the nom. is merely understood, or the sentence is the nom.) 
 
 * Compare Milton's ■ As the wakeful bird Sings darkling* Clyde 
 compares Virgil's ' Ibant obscwi.' 
 
 t In fact, the constant use of ' it ' is a strange idiom, in which English 
 differs from most languages, ancient and modern ; e.g. It was they who 
 did it = €Ke?jO» εποίησαν, isti fecerunt, Eran ellos los que hicieron, etc. 
 
90 A BRIEF GREEK SYNTAX. 
 
 vii. των σων αοερκτων ομμάτων τητώμενος. — Soph. 0. C. 
 
 1200, robbed of thy blinded eyes, i.e. robbed of thine 
 
 eyes so that they are blind, 
 ενψημον ω τάλαινα κοίμησον στόμα, — iEsch. Ag. 1247, lull 
 
 thy tongue to silence, Ο hapless one. 
 εϊσυκε θερμά λόετρα θερμήνη till he warmed the baths hot. 
 
 This is what is called the proleptic or anticipative * use of 
 the adjective. It is found quite as strongly in Latin ; e.g. in 
 Virgil, 
 
 Suhmersas obrue puppes overwhelm the ships in the 
 
 depths. 
 Scuta latentia condunt they conceal the shields in hiding. 
 Spicula lucida tergunt they wipe their darts bright. 
 
 We also find it in English, | as 
 
 ' The Norman set his foot upon the conquered shore.'—• 
 
 Drayton. 
 1 Heat me these irons hot? — Shakspeare. 
 
 1 Who with our spleens 
 ' Would all themselves laugh mortal? — Id. 
 i And strikes him dead for thine and thee.' — Tennyson. 
 
 viii. By what is called antimeria the adjective is often used 
 .where the adverb would be more correct ; as in 
 
 Χϋσαν 3' άγορήν αιφηρήν ' they loosed the assembly quick? 
 θυαν ννμψαν άγαγες thou leddest a swift bride, i.e. 
 
 swiftly (Soph. Tr. 862. Lobeck on Aj. 249). 
 κρήνη άφθονος ρέουσα a fountain flowing abundantly. 
 άσμενος νμας ε'ώον I saw you gladly. 
 
 Similarly in Milton we find 
 
 1 Meanwhile inhabit lax (i.e. loosely), ye heavenly powers.' 
 
 — Par. L. vii. 161. 
 1 Thou didst it excellent? — Shaksp. Tarn, of Shrew, i. i. 89. 
 
 * Some call it the factitive adjective. For abundant instances, see 
 Lobeck, Paralip. Gram. Gr<sc. p. 531 seqq., and id. ad Aj. 517. The 
 neglect of this has led to strange errors. Thus, in Soph. Ant. 883, rby 
 ipbv πότμον αδάκρυτον ovSds στενάζει ' no one groans for my tearless 
 fate.' Valcknar, not observing that the αΖάκρυτον is proleptic of the 
 result, makes it = πολνΜκρυτον, adopting the purely fictitious alpha 
 intensivum. 
 
 f There is a fine and ghastly instance of prolepsis in Keats's Pot of 
 Basil, 
 
 * So those two brothers, and their murdered man, 
 Bode to fair Florence.' 
 
COMPAKATIVES. 91 
 
 Compare the Biblical expressions ' Open thy hand wide,'' 
 ' Cry shrill with thy voice,' &c. But in English these phrases 
 are often due to the obsolescence of the final adverbial -e; 
 e.g. righte= rightly, sothe = truly, &c. (Morris, Specimens of 
 Engl. p. lv.). 
 
 COMPARATIVES. 
 
 70. The following instances illustrate the chief idioms in 
 the use of comparatives : — 
 
 i. άγροικότερόν εστίν είπέϊν it is somewhat rude to say. 
 
 ajueivoV εστί κ.τ.λ. it is as well to, &c. 
 ii. ήν οι αΰελφεός νπομαργότερος he had a brother rather mad. 
 
 These instances merely express degree. The want of two 
 forms in Greek, one comparative, and one qualitative, has 
 already been pointed out. (See § 44, p. 30.) 
 
 ελαφρότεροι j) άφνειότεροι swifter than richer (i.e. rather 
 
 swift than rich).* 
 έποίησα ταχύτερα i) σοφώτερα more quickly than (more) 
 
 wisely. 
 
 Notice the two comparatives, like the Latin ' Suhtilius quam 
 verius.' 
 
 Phrases like the following are common with comparatives: — 
 
 iii. ανδρειότερος γίγνεται αυτός εαυτού he grows braver than 
 
 he ever was. 
 άμβλύτατα αντος εαυτού ορξ. he sees more dully than ever. 
 μείζον φορτίον τ) καθ' αυτόν a burden too great for him 
 
 (lit. greater than in proportion f to himself). 
 κακά μείζω y κατά ΰάκρυα or y ώστε ΰακρΰειν or y l•. woes 
 
 too big for tears. 
 μείζον y κατ άνθρωπον too great for man. 
 λόγου μείζον too big for words. 
 θανών αν ε'ίη μάλλον ευτυχέστερος he would be more 
 
 fortunate (literally ' more happier ') when dead. 
 
 Compare μάλλον ασσυν, Soph. Ant. 1210, Eur. Hip. 485; 
 Hec. 377. 
 
 * ' He was more of a knave than fool,' might be expressed in Greek, 
 μοχθηρότερος ήν ί) ανούστεροε. One way of hinting at a superlative is 
 et tls κάϊ &\Aos ' if any one ever was you are,' as elf ns καΐ &\\os 
 σώφρων e? you are the most temperate of men. 
 
 f πρό, αντί, and πάρα are often used after comparatives. (Cf. Virg. 
 Mn. i. 346, ' Pygmalion scelere ante alios immanior omnes.') 
 
92 A BRIEF GREEK SYNTAX. 
 
 This last phrase shows a tendency to that analytic mode of 
 expressing the comparative,* which began in the similar Latin 
 phrases ' magis certius,' ' magis dulcius,' &c. So in the Bible 
 i The Most Highest ; ' and in King Lear i I am sure my heart's 
 more richer than my tongue.' The gradually analytic tendency 
 in comparatives and superlatives may be seen from the fact 
 that we should no longer use such terms as grievousest, 
 famousest, artificialest, &c, which we find in Bacon, Shak- 
 speare, Milton, &c, or even the ' impudentest ' of Gray. Ben 
 Jonson calls this i a certain kind of English Atticism, imitating 
 the manner of the most ancientest and finest of the Grecians.' 
 
 iv. On the other hand μάλλον is sometimes omitted, as 
 θάνατον η βίον αιρούμενοι choosing death (rather) than life. 
 This is frequent in the New Test., as Mk. ix. 43 ; Lk. xv. 7, 
 xvii. 2; 1 Cor. xiv. 19; and in the LXX., as Ισχύει ούτος η 
 ημε~ις he is stronger than we. — Num. xxii. 6. So in Plaut. 
 Rud. iv. iv. 70, Tacita bona est mulier semper quam loquens; 
 Liv. vii. 8, Ipsorum quam Annibalis interest, &c. 
 
 v. Another peculiarity of μάλλον j) is, that ου is sometimes 
 inserted after it, as 
 
 ουδέν τι μάλλον επ ήμέας μάλλον η ου και επ υμεας, Hdt. 
 
 iv. 118, no whit more against us than against you. 
 πόλιν ολην όΊαφθεφαι μάλλον ΐι ου τους αιτίους, Thuc. Hi. 
 36, to destroy a whole city rather than the guilty. 
 
 [Donaldson compares the English vulgarism * rather nor ; ' 
 and Clyde the redundant negative after comparisons in Italian, 
 as Io scrivo piu che io non parlo I write more than I (lit. 
 don't) speak. Still closer is the Spanish parallel, El es mas 
 rico que no ella he is richer than she; mejor es el trabajo 
 que no la ociosidad labour is better than idleness.] 
 
 vi. The common Comparatio Compendiaria, or Brachylogy 
 of Comparison, should be noticed ; as πυραμίς μείζων πατρός 
 a pyramid larger than (that of) his father. Instances of it 
 will be found in the Syntaxis Ornata at the end. 
 
 SUPERLATIVES. 
 
 71. The superlative, like the comparative, sometimes merely 
 expresses degree, as σεΊο <Γ Άχιλλίν οΰτις άνηρ προπάροιθε 
 μακάρτατος no one, Achilles, was ever before so very happy as 
 you (Keiner war mehr so ganz glucklich als du). 
 
 * The analytic comparative begins to appear in later Latin ; e.g. 
 ' Plus tamen ecce meus, plus est formosus Iollas.' — Calpurn. 
 The instances from Plautus show that it always existed colloquially. 
 
SUPERLATIVES. 93 
 
 72. The Greeks had a peculiar idiom with superlatives. 
 Instead of saying ' more beautiful than all others,' they said 
 
 * most beautiful of all others, 1 as 
 
 Νιρενς ος κάλλιστος άνηρ υπο "ϊλιον ήλθεν 
 των αλλωτ Δαναών. — Horn. 77. ii. 673. 
 άζιολογώτατον των προγεγενημένων more worthy of narra- 
 tion than any that preceded it. 
 
 Milton boldly imitates this inclusive use of the superlative 
 in the lines 
 
 ' Adam the goodliest of all men since born 
 His sons, the fairest of her daughters Eve ; ' 
 
 where not only ignorant critics, but even Addison and Bentley, 
 have censured him for making Adam one of his own sons, 
 and Eve one of her own daughters ! For an explanation 
 of this idiom see supra § 47 note. Cf. Hor. Sat. I. i. 100 : 
 
 * fortissima Tyndaridarum ' braver than the Tyndarids ; i Diana 
 . . . comitum pulcherrima ' fairer than all her comrades. 
 
 Nor is Milton the only English writer who has adopted the 
 idiom. Shakspeare has ' This is the greatest error of all the 
 rest ' (Mids. Night's Dream, v. 1) ; and Sir Thomas Elyot c A 
 young woman, the fairest of all others,' &c. (The Governone). 
 
 73. The following are phrases to strengthen superlatives*: — 
 άνηρ εν τοΊς μάλιστα ενάντιος τω Ιημω especially opposed 
 
 to democracy. 
 £ΐς άνηρ πλείστον πόνον παρασ\ών giving more trouble 
 
 than any one. 
 πάγου οίον δεινότατου of the sharpest possible frost. 
 όπως άριστα in the best possible way. 
 όσον τάχιστα as speedily as possible. 
 ώς οίον τε βέλτιστον in the best possible manner. 
 ότι μάλιστα as much as possible. 
 
 N.B. i. In St. John (i. 15; xv. 18) πρώτος is used as a 
 comparative, — έμπροσθεν μου γέγονεν οτι πρώτος μου ι) ν. 
 
 ii. There is sometimes a reduplication of superlatives, espe- 
 cially in comic writers, as in the words ελαγιστότερος, πρώτισ- 
 τος, αυτότατος (Plaut. ijisissimus); μειζοτέρως, 3 John 4. 
 
 * One of the ways of expressing the superlative in Hebrew is by a 
 mere repetition of the word, as 'good good' = very good. We find a 
 trace of this in Heb. x. 37, crt yap μικράν ίσον 'όσον very, very soon. 
 There is something like it in ίσον ίσον στίλ-ην, Ar. Vesp. 213, A tiny 
 tiny drop =q\i&ntillum. (Winer, Gr. Ν. T. § 35.) By a similar prin- 
 ciple we find μ€7€0« μί -yas = μβγιστο* in Pausanias. 
 
94 A BRIEF GREEK SYNTAX. 
 
 PREPOSITIONS (Προθέσεις). 
 
 74. The prepositions (as we still see in Homer) were 
 originally mere local adverbs, i.e. like the case-endings, they 
 originally denoted relations of place, but their meaning was 
 gradually extended to express all kinds of metaphysical or 
 figurative relations. 
 
 75. Cases, without prepositions, are sufficient for languages 
 which are at their simplest stage. A reminiscence of the 
 previous existence of case-inflections often remains when the 
 inflections themselves have disappeared (e.g. Le fils VEm- 
 pereur, Ville Hadrien ; cf. Hotel-Dieu, Faubourg St.-Antoine, 
 Bar le Due, De part le roi, &c). But every language, as it 
 advances from synthesis to analysis, develops prepositions, and 
 uses them more and more to give precision to the obliterated 
 forms and more extended meanings of the case-terminations. 
 Moreover as the requirements of language become more and 
 more complicated, the quickness of the mind is naturally 
 diminished and encumbered. In fact, prepositions become 
 more and more necessary to distinctness and accuracy in 
 language,* and hence they are often used in prose where they 
 would be omitted in poetry. It should then be clearly 
 understood that it is the case which indicates the meaning of 
 the preposition, and not the preposition which gives the 
 meaning to the case. Each preposition has some one distinct 
 meaning of its own, varied by the cases with which it is used. 
 Its purpose is only to supplement and to define. Thus απυ 
 meaning 'from ' entirely coincides with the conception of 
 ablation, and hence is used with the genitive only ; iv denotes 
 
 1 position in,' and therefore coincides with the meaning of the 
 dative, and is joined with the dative only; elg indicates 
 motion towards, and therefore (naturally) is only joined with 
 the accusative. Παρά means ' alongside of,' and really retains 
 this sense with all three cases, παρά σοϋ •= from (alongside of) 
 you ; παρά σοι at alongside of you = with you ; 7ταρά σε to 
 alongside of you = to you. It is therefore not strictly accurate 
 to talk of prepositions governing cases ; since in point of fact 
 they merely define the exact sense in which the case is used. 
 It is the case which borrows the aid of the preposition, not 
 
 * See some excellent remarks on this subject in Burggraff, p. 268 
 seqq. As Mr. D'Arcy Thompson expresses it, modern languages have 
 all discarded (or nearly so) the tight affixes (or case-endings) of the 
 ancient languages for loose prefixes or prepositions. 
 
PREPOSITIONS. 95 
 
 the preposition which requires the case. It should be observed 
 also that where prepositions appear to change their meanings 
 with the cases which they define, it is really a difference in 
 the meaning not of the preposition but of the case. 
 
 76. We are not therefore surprised to find that prepositions 
 have nearly superseded cases in Modern Greek and in the 
 Eomance languages ; and Ave can see the tendency to use them 
 (which ended in the final evanescence of case-distinctions), 
 on the one hand in the New Testament where they abound ; 
 and on the other in the practice of the Emperor Augustus,* 
 who was observed to make great use of them in the endeavour 
 to speak as perspicuously as possible. Thus he preferred to 
 say or speak ' impendere in aliquam rem,' and ' includere in 
 carmine,' when most of his cotemporaries would have used 
 the phrases ' impendere alicui rei] and ' includere carmine,'' 
 or carrnini. In doing this he was only a little before his age ; 
 but the same tendency is fotmd often enough, as ' ad carni- 
 ficem dare,' Plaut. ; ' Fulgorem reverentur ab auro] Virg. ; 
 1 Genera de ulmo] Plin. ; ' Scribas ad me] Cic. Att. xi. 25 ; 
 1 OfFerre se ad mortem,' id. Tusc. Disp. i. 15. 
 
 76 {bis). The same remarks apply to our own language, as 
 will appear at once by a comparison of our English version of 
 the Bible, first with Tyndale's, then with Wiclif's, and then 
 with the Maeso-Gothic fragments of Ulphilas. 
 
 77. Several prepositions (called improper or spurious) are 
 also adverbs, as εγγύς, αμα, πόρρω, πέλας, -χάριν, &c, as in 
 English ' before,' i after] &c. This adverbial use of prepositions 
 is most frequent, as might have been expected, in the older 
 writers. 
 
 78. The name Προθέσεις proeverbia is due to their use in 
 composition with verbs, &c. When they stand alone many of 
 them may (especially in poetry) be placed after \ the words 
 
 * See Egger, Gram. Comp. p. 195. The very interesting passage in 
 Suetonius, which mentions this analysing phraseology of the careful 
 emperor, is as follows : Praecipuam curam duxit sensum animi quam 
 apertissime exprimere ; quod quo facilius exprimeret, aut necubi lectorem 
 vel auditorem obturbaret ac moraretur, nee praepositiones verbis adderc, 
 neque conjunctiones saepius iterare dubitavit, guee detractce afferunt aliguid 
 obscuritatis etsi gratiam augent! The passage might have been used to 
 describe the style of Lord JMacaulay, and the last clause hints at the 
 respective advantages of synthetic and analytic languages, the latter 
 gaining in accuracy what they lose in vivid conciseness. 
 
 f In many languages (e.g. Turkish) they are entirely postpositions ; 
 in Latin we have mecum, YObiscum, &c. ; in English whereiw, -wherewith, 
 &c. ; in German Demetwegen, &c. 
 
96 A BRIEF GREEK SYNTAX. 
 
 they govern. When this is the case, the accent is thrown 
 back by what is called anastrophe, as τέκνων περί, μάχ$ en, 
 &c* Δία and ανά are excepted from the law of anastrophe, 
 lest they should be confused with the accusative of Ζευς, and 
 the vocative of άναζ. 
 
 79. There are eighteen prepositions, of which four, από, 
 ε'£, αντί, προ, govern the genitive ; two, εν and συν, the dative ; 
 one, ά»ά, the dative and accusative; three, ΰιά, κατά, υπέρ, 
 the genitive and accusative ; and seven, άμψί, περί, επί, μετά, 
 παρά, προς, νπό, take three cases, the genitive, dative, and 
 accusative. Besides these there are the improper prepositions. 
 
 80. Examples will only be given where the meaning is 
 peculiar or not obvious; and those usages which are very 
 rare or quite abnormal, are omitted ; for completeness in treat- 
 ing of the prepositions cannot be combined with brevity. In 
 all languages the usages and phrases connected with preposi- 
 tions are too numerous to be briefly exhausted. For instance, 
 in English the same prepositions may even have opposite 
 meanings, as ' I fight with you,' which may either mean ' at 
 your side and for you,' or ' against you ' ; so in Latin we may 
 have ' pugnare cum hostibus,' and ' ire cum sociis ' ; and προς 
 τίνος may mean either against or for a person, according to 
 the context, &c. The reason of this is that even the com- 
 monest matters may be viewed under many aspects ; compare, 
 for instance, the phrases ' to talk about a thing, λέγειν περί 
 τίνος dicere de aliqua re, 2 Ί|Π uber etwas sprechem' ' Here 
 we and the Greeks regard the object spoken of as something 
 encompassed; the Latins as a whole of which part is supplied; 
 the Hebrew as a ground to stand on ; the Germans as a ground 
 to be gone over'' (Winer, Gram. Ν. T. ii. § 47). Besides, 
 when mental and metaphysical relations have to be figuratively 
 expressed by words and cases which originally had only a 
 local meaning, it is obvious that the metaphor must be of so 
 very general a character that the same relation may be ex- 
 pressed with equal propriety in several ways. It is generally 
 easy with a little thought and care to trace the metaphysical 
 meaning directly from the physical, but, as the explanation 
 
 * But otherwise πάρα, %πι, μβτα, πίρι, νπο, %νι (notice the accents), 
 stand for πάρςστι, ίπεση, &c, and &va for ανάστηθι stand up ! or for the 
 vocative of &.va\ {in Homer). A change of meaning is in all languages 
 naturally accompanied by a change of accent, or spelling; thus in 
 English ' sith ' is a causal particle, but since (sithens) is also a preposi- 
 tion and an adverb. 
 
PREPOSITIONS AND COMPOSITION. 97 
 
 would require an entire treatise, and as views differ on the 
 subject, this is best left to the student himself. 
 
 81. The student should accustom himself to notice the 
 manner in which the meaning of a verb alters according to 
 the prepositions with which it is compounded ; e.g. 
 
 ΰίΰωμι I give ; ε κΰιΰόναι to disembogue ; ενΰιΰόναι to 
 yield ; επιΰιΰόναι to increase ; παραδίδομαι to hand 
 down ; προΰιΰόναι to betray ; άποδίΰοσθαι to sell, &c. 
 
 τείχισμα a fort; Ιιατείχισμα a partition; έπιτείχισμα a 
 fort built in an enemy's country ; παρατείχισμα a cross- 
 wall ; προτείχισμα a bulwark ; περιτείχισμα a line of 
 circumvallation, &c. 
 
 "ιστημι I place ; συνίστημι I introduce ; εζίστημι I drive 
 mad ; καθιστή μ ι I establish. 
 
 ϊημι I send; άνιέναι to remit; εφιέναι to give up to; 
 μεθιέναι to relax, &c. 
 
 'έχω I have; άνέχειν to continue, to rise up; έζέχειν to 
 project; προσέχειν to attend; κατέχειν appellere, to 
 touch at a shore; νπερέχειν to excel; άντέχειν to 
 resist; έπέχειν to wait for; {ανεχε και πάρεχε 'bear 
 and forbear '). 
 
 σείω I shake ; προσειω I threaten, or entice by waving ; 
 επισείω I hark on, &c. 
 
 έρχομαι I come; κατέρχομαι I return from exile; μετέρ- 
 χομαι I go after, &c. 
 
 Prepositions which govern the Genitive. 
 
 82. These are: 
 
 i. αντί opposite to, contra] then instead of, for. (Compare 
 the words αντ α, άντην, αντικρύ, ενάντιος, ante.) 
 αντ εμοϋ instead oj me. 
 άν& ου on account of which. 
 άλλάττεσθαι άντι χρυσοΰ to change for gold. 
 χάριν άντι χάριτος grace for grace, i.e. unceasingly re- 
 newed. 
 
 ii. προ (pro3) before, both of time, place, and preference. 
 It is closely connected with, but slightly more general 
 than, αντί] hence αντ οφθαλμών = προ των οφθαλμών. 
 
 iii. εκ, εζ 'from out of extrinsecus. 
 
 εκ παίΐων from boyhood (cf. ' of a child,' Mk. ix. 21 ; 
 1 Being o/so young days,' Haml ii. 2). 
 
 F 
 
98 A BRIEF GREEK SYNTAX. 
 
 εκ often = after ^ as 
 
 γελάν εκ δακρύων to laugh after tears. 
 
 εκ δείπνων νπνος ηδύς sweet is sleep after dinner. 
 
 τυφλός εκ δεδόρκοτος blind after seeing. 
 
 εκ κυμάτων γαλή ν ορώ I see after storms a cairn. 
 
 Compare the Latin ex : e.g. Scriba ex quinqueviro ; ex 
 homine factus est Verres (Cic. Div., Υ err. 17 f.). 
 
 Our of is used in just the same way by Milton, as 
 
 ' I of brute, human, ye of human, gods.' — Par. Lost, , 
 
 ix. 712. 
 1 How cam'st thou speakable of mute V — Id. ix. 563. 
 1 Is of a king become a banished man.' — Shakspeare, ι 
 3 Henry VI. iii. 3. 
 
 iv. άπό (a, ab, ahs, off) l from 1 ; ά-ό means { from the out- 
 side,'' εξ from the inside of a thing; as ύπό Ταλιλαίας, 
 εκ πόλεως Ναζαρέτ. — Luke ii. 4. It expresses place, 
 time, and cause ; also sometimes the agent, as επράχθη 
 άπ αυτών ούδεν έργον άξιόλογον. 
 
 Besides these fom*, the spurious prepositions άνευ without, 
 άχρι, μέχρι until, μεταξύ between, ένεκα and εκατι for the sake 
 of, ευθύ straight towards, πλην except, τρόπον and δίκην like, 
 and χάριν for the sake of, govern a genitive. 
 
 N.B. ευθύς = immediately, ευθύ with the gen. = straight 
 towards ; μεταξύ by a curious ellipse sometimes omits one of 
 the two things between which another is placed, as μεταξύ 
 τών Ίνοϋς (Ar. Ach. 434) between those of Ino (and the ones 
 last mentioned). Compare our word ' twilight, 1 i.e. twixt light 
 (and darkness). Cf. Par. Lost, ix. 50, and Shilleto, Bern, de 
 F. Leg. § 181. μεταξύ δείπνων = ichilst dining. 
 
 Peepositions with the Dative, εν, συν. 
 
 83. i. εν (in with the ablative) of place and time ; also of 
 the instrument and manner, as 
 εν or σύν τάχει with speed. 
 εν όφθαλμο'ις ορών seeing with the eyes. 
 ήν εν τοΊς Ίεροσολύμοις (place), εν τω πάσχα (time), εν Trj 
 εορττ} (circumstance). — 2 Cor. vii. 16. 
 
 ii. Σύν (ξύν, cum) with. It implies a closer union than 
 μετά. See Soph. Ant. 115. πολλών μεθ* οπλών, σύν θ' Ιππο- 
 κόμοις κορύθεσσι (Donaldson), σύν τινι implies coherence; 
 μετά τίνος coexistence (Winer). 
 
Συν, sis, ώ$, δία. 99 
 
 Ν.Β. Συν is by no means coextensive with the English 
 1 with ; ' thus ' they fought with him,' would be not συν αυτω 
 but προς αυτόν. 
 
 With the Accusative, είς, ώς. 
 
 84. i. εις (in with accusative), into, of place. Also up to, 
 of time, as έτος εις έτος year by year, εϊς είκοσι μάλιστα up to 
 about twenty. Also of purpose, as είς τόΰε ήκομεν for this 
 purpose we have come. 
 
 είς into stands in the same relation to προς towards, as εξ 
 out of does to άπό away from. 
 
 ε\ς sometimes, in the tragedians, means c as regards ; ' ώς 
 ουτις ανδρών εις απαντ ευΰαιμονεϊ since no man is happy in 
 all respects (cf. Eur. jPharn. 619, 1645; Or. 529). 
 
 είς is often used with ellipses, as ες διδασκάλου into the 
 teacher's (house), ες "Αώου to (the realm of) Hades, &c. 
 
 ii. ώς * to ' only with persons, or words that involve per- 
 sons, as 
 
 επεμφεν αυτόν ώς βασιλέα he sent him to the king, 
 ώς τάσδε χείρας to these hands of mine. 
 
 Probably it is a merely elliptic expression for ώς προς, ώς 
 επί, &c, which we frequently find ; e.g. είς Φωκέας, ώς προς 
 συμμάχους. — Demosth. (cf. Acts xvii. 14). Constructions like 
 ώς "Αβυΰον % to Abydos,' are very rare. 
 
 With the Genitive and Accusative, διά, κατά, υπέρ. 
 
 85. i. διά through (connected with δύο ; δι' εκ = right 
 through ; cf. Engl, between with twain). 
 
 a. With genitive =per* 
 
 cY αγγέλων by means of messengers. 
 δια των οφθαλμών ορώμεν we see with our eyes. 
 δια χερών εχειν to have in hand. 
 δια φιλίας ιέναι to be on friendly terms.f 
 δια στόματος έχειν to talk about, 
 δια μακροϋ after a long interval. 
 
 δια δέκα επάλξεων πύργοι towers at intervals often battle- 
 ments. 
 
 * Aicl with the genitive is rarely used of the direct agent (which is 
 νπο or πάρα with the genitive) ; δι' ου is not ' by whom] but ' by whose 
 means,' per quern not a qiio. 
 
 f Cf. άγςιν δια φροντίδος curare, δια μνήμης mentionem facer e, δι 
 αϊδονς venerari, δι' (ύχης in votis habere, &c. 
 
 F 2 
 
100 A BRIEF GBEEK SYNTAX. 
 
 β. With the accusative, through or about (poet.), as ΰιά 
 δώματα. Also 011 account of —propter, as έ'χω γαρ άχω £ιά σε. 
 Thus ΰιά νήσου iivai would be to pass through an island ; 
 ΰιά νήσον Ιέναι would be in poetry to make a tour through an 
 island ; and we should say Ιια πείίον εμάχετο he was fighting 
 all about the plain, but δια πείίου εΰραμεν he ran through the 
 plain. 1 Cor. xi. 9, ουκ εκτίσθη άνηρ Ιιά την γυναίκα ' for the 
 sake of; id. vers. 12, δ άνήρ δια της γυναικός * by means of.' 
 ci ov τα πάντα και δι' ου τα πάντα, Ileb. ϋ. 11, for 
 
 whose sake, and by whose means all things exist. 
 dia σοϋ per te, by your means; αυτός δΓ εαυτού εποίικ 
 
 he was doing it by himself, sua unius opera, 
 δια σε propter te, because of you ; ει μη δι' αυτόν but for 
 
 him. 
 ΰια τούτων by means of these things, per ho?c. 
 ΰια ταύτα wherefore, propter ho2C. 
 
 N.B. Ιιά νυκτός and δια νύκτα differ very little ; the former 
 calls attention to the fact that a thing lasted till next morning, 
 the latter that it occupied all night long. 
 
 ii. κατά l down.' 
 
 a. "With genitive, down from; also against, as λέγειν κατά 
 τίνος to speak against any one. 
 
 β. With accusative, along, about, according to, in re- 
 ference to.* 
 
 κατά ρόον down stream. 
 
 κατά τον αυτόν χρόνον about the same time. 
 
 κατά γνώμην την ε μη ν according to my notion. 
 
 το κατά Μάρκον ευαγγέλιον the gospel according to Mark. 
 
 Compare the following : j κατά with the genitive, vertical 
 motion ; — >- κατά with the accusative, horizontal motion. 
 οι κατά χθονός the dead. 
 οι κατά χθόνα the living. 
 
 κατ Ουλύμποιο καρήνων down from the crest of Olympus. 
 κατά θάλασσαν εκορεύετο he went by sea. 
 
 iii. ύπερ over. 
 
 a. With the genitive, position over, super ; also on behalf 
 of,^ as in ύπερ σοϋ άπυκρινοΰμαι I will answer on your behalf. 
 
 * Hence both κα& εαυτόν, and δι εαυτού, mean ' by himself! seorsum ; 
 but the former implies ' in reference to,' the latter ' by means of.' 
 
 f Both ΰπ\ρ and προ with the genitive mean ' on behalf of,' because a 
 
Κατά, ανά. 101 
 
 β. With the accusative, over and beyond, ultra ; as ρίπτειν 
 νπερ roy Ιόμον to fling over the house. 
 
 With the Dative and Accusative. 
 
 Άΐ'ά 'up.' 
 
 a. With the dative, only in Epic and lyric poetry, on. 
 ευίει δ' ά^ά σκάπτω Διός αϊετος and the eagle slumbers on 
 the sceptre of Zeus. 
 β. With the accusative, up, throughout, &c. 
 ανά ρόον up stream. 
 avh πάν έτος quotannis. 
 άνα παν το έτος throughout the year. 
 
 N.B. i. 'Ανά, κατά, are probably the origin of the hypo- 
 thetical particles αν, κεν. 
 
 ii. They are used in constant contrast, as άνω κάτω up and 
 down, sursum deorsum ; άια κατά ultro citroque, άνέβη he 
 went inland, κατέβη he went to the sea, άνέΰυ it rose, κατέδυ 
 it set, άνανεύω I throw back the head in token of dissent, 
 κατανεϋω I nod assent. 
 
 iii. And yet, since up and down are but two ways of re- 
 garding motion along the same line, it is often indifferent 
 which of the two Ave use;* hence we find either /caret or ava 
 κράτος forcibly; κατά or άνα στρατον throughout the army; 
 Kara or άΐ'ά στόμ εχειν to talk about, κατά or ava τετταρας 
 by fours (also έπι τεττάρων), κατά or ανά πόλεις about the 
 cities. 
 
 With Genitive, Dative, or Accusative, Άμψί, περί, επί, 
 μετά, παρά, προς, νπο. 
 
 86. ί. άμφί (Lat. amb-, apud, German um). * It is mostly 
 confined to Ionic Greek | and to poetry, and it is the only pre- 
 
 champion in battle stood in both positions, as μη θνήσχ' ύπίρ τοί/δ' 
 ανδρός, ουδ' εγώ προ σου. — Ale. 690. (Donaldson.) 
 
 * We must not suppose because two prepositions are interchangeable, 
 even with different cases (as ivl τεττάρων and ava. τέτταρας) that they 
 mean the same thing. The explanation is that the same relation may be 
 regarded from two entirely different points of view. In German Auf die 
 Bedhigung and Unter der Bedingung both mean ' on the condition,' but 
 auf ' on ' is not = unter, ' under.' (Winer, iii. § xlvii.) 
 
 f In Later Greek (e.g. in Plutarch and Lucian), bya wild extension of 
 the dislike to all directness or personality of speech, οι &μφι Πλάτωνα 
 simply means Plato! In Herod, i. 62, ol άμφϊ Πςισίστρατον . . . άπικν4*ται 
 is due not to this phrase, but to anacoluthon. 
 
102 A BRIEF GREEK SYNTAX. 
 
 position which has disappeared in Modern Greek.' (Clyde.) 
 As usual, we may trace its comparative insignificance in the 
 fact that it never occurs in the New Testament except in* com- 
 position. 
 
 With all three cases it means around or about. 
 
 άμφϊ τον -χειμώνα about winter. 
 
 άμφι τους μνρίονς about ten thousand. 
 
 οι άμφϊ Πλάτωνα Plato and his school.* 
 
 It is not used with the dat. in Attic prose. 
 
 ii. περί around and about (Lat. per-, as adv. περί = very. 
 Compare our English phrase, ' good all round '). This becomes 
 the Gothic four-, the German ver-, the English for- ; e.g. for- 
 lorn = verloren = utterly lost, etc. 
 
 a. "With the genitive = de, about. Notice the phrases περί 
 εριΰος pros ira (Horn.), περί πολλού εστίν ημΐν it is of much 
 consequence to us. 
 
 β. With the dative,f around, of place, and concerning, as 
 θαρρε~ιν περί tlvl to be of good cheer about any one. 
 
 γ. With the accusative around, and in regard to, and about, 
 as περϊ τοντον τον -χρόνον about this time. 
 
 In these two prepositions the distinctions of meaning with 
 the different cases are not at all distinctly marked. Hence we 
 find in the same sentence ενφραίνειν θυμον άμψί τινι, and αμφί 
 τίνα, and in the same sentence of Herodotus, vii. 61, περϊ μεν 
 Trjai κεφαλ^σι εΐ-χον τιάρας . . . περϊ 3έ το σώμα κιθώνας. 
 And ' both are used with vague indications of time or number.' 
 — Donaldson. 
 
 iii. επϊ upon. It has various meanings, which can gene- 
 rally be deduced from its adverbial sense, and the meaning of 
 the case with which it is joined. Thus with the genitive it 
 implies partial superposition ; with the dative absolute super- 
 position, or rest upon ; and with the accusative motion with a 
 view to superposition (Donaldson).- 
 
 a. With the genitive — 
 
 εφ' 'ίππων όχεϊσθαι to ride on horseback. 
 
 πλεΊν επί Σάμου to sail towards Samos. 
 
 ε πι Ααρείον ky ενετό it happened in the time of Darius.J 
 
 i(f ημών in our 
 
 * See note f on preceding page. 
 
 f περϊ and inrb are never used wdth the dative in the New Testament. 
 
 I This temporal meaning of eVi is partly derived from the participles 
 
Έπ/, μετά, παρά, 103 
 
 β. With the dative— 
 
 επί τΐ] θαλασσή οϊκείν to live near the sea (i.e. upon the 
 
 shore). 
 επϊ τούτοις thereupon, or besides, 
 εφ' οϊς τε on condition that. 
 επί θήρα or επί θηραν έζιέΐ'αι to go a hunting. 
 επι τόκοις Ιανείζειν to lend on interest. 
 το επϊ σο\ as far as you can ; nearly = το επϊ σε quantum 
 
 in te est. 
 y. With the accusative, motion towards — 
 αναβαίνειν k<f Ίππον to mount on horseback. 
 στρατεΰεσβαι επϊ Ανΰους to go on an expedition against 
 
 the Lydians. 
 το επϊ σφάς tlvcu as for as depended on them.* 
 
 iv. Μετά with (connected with μέσος, German mii) implies 
 separable connection. 
 
 a. With the genitive = with, (Lat. cum) accompanied by 
 (but never our ' with ' in the sense of an instrument, as { with 
 a sword '). 
 
 β. With the dative = among (only in poetry). 
 
 γ. With the accusative = ' after] either in space or time ; 
 e.g. βη ()ε μετ Ήομενήα he went after (i.e. in quest of) Ido- 
 meneus; μετά ταντα after these things. 
 
 Our ί after 1 has the same two meanings, for we say (collo- 
 quially), ' To send after a person, a book,' &c. Succession in 
 place and time are constantly confused, as in the word ' inter- 
 val] used of time, but properly a space between two ramparts. 
 
 v. πάρα beside (apud). 
 
 a. With the genitive, from, ελθεΊν παρά τίνος = venir de 
 chez quelqu'un. 
 
 β. With the dative, near, ήν πάρα τω βασιλεΊ he was with 
 the king. 
 
 γ. With the accusative, towards. All its shades of meanings 
 with the accusative are derived from the notion of ' motion 
 near, or with a view to conjunction.' 
 
 Ιέναι πάρα νηας to go to the ships. 
 πάρα θΊ,να θαλάσσης along the sea beach. 
 
 •with which it is generally joined ; we use a very similar phrase when 
 we say 'upon this' = when this happened; ' Upon his coming to the 
 throne,' &c. 
 
 * In several of its meanings επί resembles the German auf, which is 
 used both of hills and plains ; as eV ep^ias = auf dem Felde. (Winer.) 
 
104 A BRIEF GREEK SYNTAX. 
 
 πάρα όλον τον βίο ν during one's whole life, 
 παρ' ελπίδα beyond expectation. 
 αμαρτωλοί παρά πάντας sinners beyond all. 
 7τοφα νόμον contrary to the law. 
 παρά ταΰτα besides these things, 
 παρά μικρόν within a little. 
 παρ ήμαρ from day to day. 
 
 The causal meaning of παρά, as in πάρα tj)v εαυτού αμέλειαν, 
 has been compared with our colloquial, ' it's all along of his 
 own neglect ; ' in this instance however ' all along ' possibly 
 means ' throughout ,' and of is the preposition denoting the 
 source of action. 
 
 παρά σον = apud me a te, i.e. from you ; παρά σοι = apud 
 te a me, i.e. with or by you ; παρά σε a me ita ut apud 
 te sit, i.e. towards you. It is however simpler to ex- 
 plain it as meaning from (alongside of) you, near 
 (alongside of) you, towards (alongside of) you. 
 
 vi. προς* (adversus), to. 
 
 a. With the genitive, on the side of, &c, προς μητρός on the 
 mother's side (cognati a matre versus me). 
 
 οι προς αίματος blood relations. 
 
 προς θεών by the gods. 
 
 ουδαμώς προς σου λέγεις you're not talking at all like 
 
 yourself, 
 προς τίνος λέγειν to speak for a person. 
 
 β. With the dative, at, to, besides. 
 
 γ. With the accusative, towards, with respect to; ουδέν 
 προς έμέ it's nothing to me ; προς βίαν, violently, &c. 
 
 προς τούτων in consequence of this (motive). 
 προς τούτοις in addition to this (juxtaposition). 
 προς ταΰτα therefore (with reference to this) ' so then.' 
 προς σε θεών αιτούμαι per te Deos oro : notice the posi- 
 tion of the pronoun. 
 
 See Eur. Phoen. 524; Jisch. P. V. 992. 
 
 * Since 'from ' and ' to ' may imply motion along the same line, only 
 regarded from two different points, we are not surprised to find in the 
 same sentence rbu μ\ν irpbs βορέω kareuna rbv 5c irpbs νότον one standing 
 from (i.e. towards) the north (as in Latin ' ah oriente ' = versus orientem), 
 the other towards the south. — Herod, ii. 121, 
 
'Τττο. 105 
 
 vii. νπο under. The physical meanings of νπο are very dis- 
 tinct ; thus 
 
 a. With the genitive =from under (motion from), 
 
 νπο πτερών σπάσας dragging from under the wings. 
 β. With the dative=(a£) under (position), 
 
 καλή νπο πλατανίστω under a fair plane tree. 
 γ. With the accusative = to under (motion to), 
 
 vir"I\iov ώρτο sped under (the walls of) Ilium. 
 
 νπο with the genitive is the commonest method of expressing 
 the agent after passive verbs, as 
 
 ίάλω νπο των 'Ελλήνων it was taken by the Greeks. 
 Notice the phrases, 
 
 νπο vi»KTct=sub noctem, about nightfall. 
 
 νπο σάλπιγγος πίνειν to the sound of the trumpet. 
 
 87. Donaldson quotes an interesting passage of Philo Judasus 
 (i. 162), in which he says that the efficient cause or agent 
 (ί/0' ov) in creation was God ; the material cause (έ'ξ ov) was sub- 
 stance (// νλη) ; the instrument (tV ov) was the Word ; the final 
 cause or reason for it (cV 6) is the goodness of God. 
 
 Prepositions in Composition. 
 
 88. In compounds, the use of the prepositions is generally 
 obvious ; but the following may be noticed. Sometimes από 
 has a negative force, as in άποψη μ ι nego, απαρέσκω displiceo ; 
 άΐ'ά resembles the Latin re- in άνατίθεμαι retracto, αναβάλλω 
 rejicio ; δια has a reciprocal force, as in διαμάχονται they fight 
 together; ut means besides, as επιγαμεΊν to marry a second 
 wife ; παρά=?ιιαΙβ, &c. as παραφρονεϊν to be mad, παρακρονειν 
 to cheat; νπο — secretly or slightly, as νπαγελάν subridere, 
 νπόλενκος whitish, νπεκπέμπειν to send out secretly. 
 
 Common Constructions with Prepositions. 
 
 89. i. The agility of intellect among the Greeks, and their 
 love of terseness, led them to a frequent use of what is called 
 the constructio prwgnans (one of the forms of the constructio 
 caret σννεσιν or ad sensum), by which they put a preposition 
 implying rest with a verb implying motion, or vice versa, so 
 that two clauses are compressed into one, as 
 
 εφάνη λίς . . . εις όΰόν a lion appeared into the road (i.e. 
 
 came into and appeared in). 
 οι εκ της αγοράς άπέφνγον those who were in the forum 
 
 fled from it. 
 
 f 3 
 
10G A BRIEF GREEK SYNTAX. 
 
 καθημεθ' άκρων εκ πάγων we sat (on and looked) from 
 
 the hill tops. 
 στασ kl Ουλνμποιο standing (on and looking)* from 
 
 Olympus. 
 προς το πϋρ καθήμενος sitting to the fire (i.e. going to and 
 
 sitting at). 
 Φίλιππος ΰε ευρέθη είς " Αζωτον Philip was found into 
 (=at) Azotus.* 
 li. So in Latin we find 
 
 In amicitia receptus. — Sail. 
 In aquam macerare. — Cat. 
 Eesponde ubi cadaver abjeceris. — Tac. 
 And in English, ' To place a thing in (=^io) his hands ;' l to 
 hang something from (==ο?ι) a peg;' ' where {—whither} are 
 you going ?' But our instances are fewer and far less strongly 
 marked, f Chaucer, however, has, ' Whan Scipio was come In 
 Africke? — Assembl. of Fowles (see Bible Word Book, p. 263). 
 
 90. In poetry, if there be two substantives the preposition 
 is often put with the last only, as 
 
 τ) ΝεΤλοι/ ?} Vt ΜέμΦιν. — Anacr. 
 
 i] αλός ?) επί γης. — Od. i. 247. 
 
 "ιθί ναούς, 'ίθι προς βωμούς. — Eur. Hec. 146. 
 It is the same in Latin, as 
 
 c Quas nemora, aut quos agor in specus ? ' — Hor. 
 'Baias et ad Ostia currant.' — Juv. 
 
 91. On the other hand, the preposition is omitted from the 
 second of two verbs, as 
 
 προβάτε βάτε.— (Ed. Col. 859. 
 
 κατήγεν, ήγεν, ήγεν, ες μέλαν πε^ον. — Eur. Bacch. 1018. 
 So, too, in Latin — 
 
 1 Retinete, tenete.' — Pacuvius in Niptris, Cic. 
 
 * In the New Testament this occurs all the more frequently from its 
 also being a Hebrew idiom, as 2 Κ•Ί2 ζΐσίρχ^σθαι iv. (Winer.) Compare 
 ' Ye shall be beaten into (els) the synagogues.' — Markxiii. 9. In Col. iv. 
 16, tV e/c AaoSiKeias ϊπιστοΧ^ν means the letter written to L. and sent 
 thence to you ; not ' from L.' as it has been erroneously taken by those 
 who were not aware of this construct™ pragnans. Winer, § lxvi. 6. 
 Cf. Ps. lxxxix. 39. 
 
 t The strongest instance I have found is in the ballad of Sir Patrick 
 Spens — 
 
 ' And lang lang may the ladies sit, 
 With their kaims into their hands ; ' 
 
 unless this be a Scoticism. 
 
INTERCHANGE OF PREPOSITIONS. 107 
 
 92. Two prepositions are often used with the same word for 
 the sake of greater distinctness, as 
 
 άμψΐ συΰνεκα, Soph. Phil. 554. 
 από βοής ένεκα, Thuc. viii. 92. 
 μη προς Ισχνός γάριν^, Eur. Med. 538. 
 
 And we find compounds such as υπεκπέμπειν, έζαποφθείρειν, 
 προπροβιάζεσθαι, &C. 
 
 Various Instances of the Use of Peepositions. 
 
 93. The prepositions are often varied in the same clause, 
 which shows how often the shades of difference between their 
 meaning are very slight ; as ούτε έπι γήν ούτε δια θαλάσσης, 
 Thuc. ; της επί την "Άττικην υδοϋ και της είς ΥΥελοπόννησον, 
 Demosth.; μη περί των δικαίων μηδ' υπέρ των εζω πραγμάτων 
 είναι σην βουλην, id.; εκ: τε της Κέρκυρας και άπό της ηπείρου, 
 Thuc. νϋ. 33 ; εκ πολέμου μεν . . . acf ησυχίας δε, Thuc. i. 124. 
 
 94. i. We find the same variety in the New Testament, as 
 ος δικαιώσει την περιτομήν εκ πίστεως (the source) και την 
 άκροβυστίαν δια της πίστεως (the means), Born. iii. 30. άπό 
 and εκ are synonymous in John xi. 1 ; Rev. ix. 18. 
 
 ii. We might say 
 
 Χρίστος υπέρ άσεβων απέθανε, Rom. v. 6, 8, xiv. 15; 
 or δούναι την -φυχην αυτοϋ λύτρον α ν τι πολλώ ν, Matt. χχ. 28 ] 
 or αίμα το περί πολλών έκχυνόμενον, Matt. xxvi. 28. 
 
 In all these passages we might use 'for' in English, but υπέρ 
 means in behalf of , αντί instead of (Joe), and περί on account 
 ofixs, as the cause. Yet the difference of meaning is so slight 
 that the readings often differ, as in Gal. i. 4. 
 
 iii. The variation of prepositions to present the thought 
 from all points of view is very common in St. Paul, as 
 
 απόστολος ουκ απ' ανθρώπων (as the source) ουδέ δι αν- 
 θρώπου (as the intermediate authority) αλλά δια Ίησοϋ 
 Χρίστου, Gal. i. 1. 
 εξ αυτοϋ {from him), και δι αυτοϋ (by his means) και είς 
 αυτόν (to him as their end) τα πάντα, Eom. xi. 36. 
 
 95. Notice the phrases, 
 
 i. καθ' ημέραν day by day, singulis diebus. 
 
 με& ημέραν in the day time, interdiu (properly after 
 day-dawn). 
 
108 A BRIEF GREEK SYNTAX. 
 
 παρ ημεραν during the day, per diem ; also = ήμερα 
 
 παρ 1 ημέραν from day to day, alternis diebus. 
 άνα πάσαν ημέραν daily, quotidie. 
 
 ii. κατίι το αυτό eodem tempore, 
 υπό το αυτό sub idem tempus. 
 
 iii. Ammianus (Anthol. xi. 231) says to Marcus — 
 
 θηριον εΐ κατά γράμμα και άνθρωπος ΰια γράμμα 
 
 [(Μ)αρκ(τ)ος']. 
 
 PRONOUNS. 
 
 96. The Personal Pronouns, being involved in the finite 
 verb, are only expressed when emphatic, as εγώ μεν διδάσκω, 
 συ δε παίζεις I am teaching, but you are playing.* As might 
 have been expected, they are more common in later than in 
 earlier stages of the language ; e.g. they abound in the New 
 Testament. 
 
 97. Αυτός when placed first is emphatic, as αυτόν ετνφεν he 
 struck him (and no one else), but ετυ-φεν αυτόν merely 'he 
 struck him ;' αυτός παρεγένου ; were you present in person! 
 
 αυτοί εσμεν we are (by) ourselves, i.e. alone. 
 τέταρτος, πέμπτος αυτός with three, four others, &c.f 
 αυτός εφη the master said it. 
 
 98. i. Possessive Pronouns are sometimes put for personal, 
 as 
 
 σος πόθος regret for you. 
 
 ες την ίμην άνάμνησιν in memory of me. — Luke xxii. 19. 
 
 τω υμετέρω ελέει the mercy shown to you. — Rom. xi. 31. 
 
 ii. They are placed after the article, as 6 σος υιός ; whereas 
 the genitives of the personal pronoun are placed after the 
 noun, as 6 υιός σον. J 
 
 iii. The attraction of a personal into a possessive pronoun, 
 as in 
 
 τάμα ΰυστήνου κακά the woes of me unhappy, 
 εμα κήΰεα θυμοϋ the cares of my mind, 
 
 * A pronoun is sometimes emphatically inserted in the latter of two 
 clauses, as ήτοι μανύε f) Zye αποπληκτ^ γενόμΐνος, Herod, ii. 173. 
 Nunc dextra ingeminans ictus nunc ille sinistra, Virg. 
 
 f Cf. II allait lui cinquieme. 
 
 \ In Soph. Aj. 572, b \υμ&ν 4μόί is at any rate a very rare expres- 
 sion for δ \υμ^ν οΰμός ; but probably the reading should be ίμοΐ. See 
 | 21. 
 
PEONOUKS. 109 
 
 is very common ; and is closely paralleled by the Latin ' meas 
 praBsentis preces,' ' nomen meum absentis.' It is also found 
 in German, as ' An meiner Scliwelle, des armen Mannes.' — 
 Schiller, Tell. 
 
 iv. The form αμο$=.ημ£τ epos is sometimes found in the tragedians. 
 When it stands for 4μος some would write it without the aspirate. 
 JBrunck says, άμο$ Doricum pro ημέτερος, αμος Atticum pro εμός. See 
 Eur. Hcl. 531 ; Iph. Aid. 1455 ; iEsch. Cho. 428. 
 
 v. As Greek has no possessive pronoun for the 3rd person 
 (' his,'' &c), αυτού is used for ' his, 1 εαυτού for ' his own ;' e.g. 
 μετεπέμψατο την εαυτού θυγατέρα καϊ τον 7τα7?α αυτής arces- 
 sivit suam filiam, ejusque filium. 
 
 1 His ' in English till Shakspeare's time meant also ' its,' just like the 
 Greek ούτου. See Craik, Engl, of Shaksp. p. 97 seqq. 
 
 vi. Σφέτερος is exclusively reflexive =their own. 
 
 The Reciprocal and Reflexive Pronouns. 
 
 99. The reflexive pronouns (those implying ' -self or 
 1 own ') give to Greek and Latin a clearness absolutely unat- 
 tainable in English ; e.g. if we say, ' he laid the wounded man 
 on his own bed,' it is impossible to mark in English whether 
 4 his own ' refers to the subject ' he ' or to the accusative ' the 
 man.'* In Greek and Latin, ες το εαυτού λέχος 'in suo lecto,' 
 would at once show clearly when the former was intended. 
 Similarly, such sentences as 'Quis profitetur suum esse dicere ?' 
 1 Suum CaBsari gladium restituit,' could only be rendered in 
 French or English, unequivocally, by a long periphrasis. See, 
 too, Eve's Gennan Syntax, p. 36. 
 
 N.B. i. ου, ε, are not found in Attic prose ; ol is rare in 
 the orators. 
 
 ii. The reflexive is often used when the thoughts of another 
 are referred to, as κελεύει £ε ol συμπέμ-ψαι άνδρας and bids 
 them to send him(self) men. 
 
 iii. The dramatic and graphic tendency of Greek writers is 
 
 * As a specimen of the utter confusion thus introduced into English, 
 take this sentence of Goldsmith: ' He (Philip) wrote to that distinguished 
 philosopher in terms the most polite and flattering, begging of him 
 (Aristotle) to undertake his (Alexander's) education, and to bestow upon 
 him (Al.) those useful lessons which his (P.'s) numerous avocations 
 would not allow Am (P.) to bestow.' See Dalglish, Engl. Gram. p. 116. 
 There are several inaccuracies in the common usage of the English 
 reflexives. See Latham, Engl. Gram. p. 150. 
 
110 A BRIEF GREEK SYNTAX. 
 
 generally sufficient to account for any apparent inaccuracy in 
 the use of the pronouns. 
 
 iv. There is no reciprocal pronoun in Latin ; its absence is 
 supplied by such phrases as inter se, invicem, alius alium, &c. 
 (See Nagelsbach, Lat. Sti/listik, § 89.) 
 
 Compare έτυψαν αλλήλους verberavit alius alium (ils s'entre- 
 frapperent, or ils se irapperent l'un l'autre). 
 
 The Greek αλλήλων is only a reduplication of άλλο-, and is therefore 
 a synthetic form for the quasi-parathetic alius alium. (For the η com- 
 pare εψηλα from ψάλλω, &c.) 
 
 100. Eeflexive pronouns are often substituted for re- 
 ciprocal, as 
 
 εΐουλωβησαν ουκ αμΰνοντίς σφισιν αυτοϊς they "were en- 
 slaved, not defending themselves (=one another). 
 
 ΰιελεγόμεθα ημ'ιν αυτοίς Ave conversed with ourselves 
 (i.e. with one another). 
 
 i.e. the reciprocity is extended into identity, just as in the 
 German ' Wir sehen wis wieder,' ' Ave see one another again,' 
 and in the French se battre, s' entendre, se disputer, &c. : ' les 
 republiques italiennes acharnees a se detruire.' So in Italian, 
 1 S' amano Γ un Γ altro,' they love each other. — Boccaccio. 
 In Spanish, se aman, they love one another. The case is 
 reversed in this sentence of the Spectator, ' The greatest 
 masters of critical learning differ among one another ' (reci- 
 procal, instead of 'among themselves,' reflexive).* 
 
 Demonstrative Pronouns. 
 
 101. i. οίε hicce, ούτος hie, ille,^ εκε~ινος iste ; compare the 
 Spanish este hie, ese ille, aquel iste ; and the Italian questo, 
 cotesto, auello. 
 
 ii. ό'£ε like questo is often used of the first person ; in the 
 tragedians άνηρ 6οε=εγώ. 
 
 iii. So 6£ε=εμός, Soph. Ant. 43, εΐ τον νεκρον συν τΐβε 
 κουφιεϊς γερ\ with my aid. 
 
 The avoidance of the personal pronoun as being too posi- 
 tive and self-assertive, leads to the most curious page in the 
 history of language ; e.g. the use of the first person plural by 
 
 * Dr. Latham has adduced many instances of reflexive pronouns 
 becoming reciprocal and vice versa. Philolog. Trans. 1844. So the 
 Hebrew Hithpahel or middle voice is often reciprocal, as hisktakshak, to 
 run to and fro among one another. Ewald, Hebr. Gram. § 243. 
 
 f 651 οιποσί &c. are still more emphatic forms. 
 
DEMONSTRATIVE PRONOUNS. Ill 
 
 royal personages, the editorial ' we,' &c; the invariable substi- 
 tution of the second person plural for the second person 
 singular, ' you ' for ' thou,' until in modern languages to 
 1 duzen ' or ' tutoyer ' a person is either a great familiarity or an 
 insult.* In Spanish, instead of thou and you, we have listed, 
 Ustedes (written Vmd.) which are contractions of Vuestra 
 Merced, &c. your honour. In German we have 526=' they,' 
 and in Italian ella i she,' agreeing with vostra signoria under- 
 stood. The use of a demonstrative (as ούτος, οίε for εγώ) is 
 carried to most extravagant lengths in Chinese, where a person 
 speaking of himself to a superior says, ' this thief,' or ' this 
 little dog,' ' this pigeon,' &c. Cf. p. 28. 
 
 iv. οίε also ushers a new character on the stage = ΰεΰρο or 
 <5δε. 
 
 αλλ' ηΚ όπαΰών εκ Ζόμων τις εργεται but Ιο ! one of the 
 attendants is coming hither from the palace. 
 
 v. ούτος often calls a person (cf. Heus tu !)j as 
 
 ώ ούτος ούτος Οιδίπους, τι μέλλομεν ; what ho ! GEdipus, 
 why are we lingering? — (Ed. C. 1627. 
 
 vi. καί ταϋτα=αηά that too ; και ταϋτα ΰή τοιαύτα so much 
 then for that. 
 
 vii. ravra and τοιαύτα usually refer to what goes before, 
 rah and τοιάΰε to what is coming ; as 
 
 ει μή ταντά εστίν, ουΖε τάΰε if it isn't that, neither is it 
 
 this.—PteA. Phced. 76 e. 
 όταν τοντο λέγωμεν, τόΰε λέγομεν when we say that, we 
 
 say as follows, 
 rouro μεν συ λέγεις, 7ταρ' ημών h' απάγγελλε τάΰε SO you 
 
 say, but announce our reply as follows. 
 δίά τήνΰε αϊτίαν for the following reason. 
 
 viii. εκείνος has the sense of i the famous,'' like the Latin 
 Me ;f as 
 
 * ' All that Lord Cobham did was at thy instigation, thou viper, for 
 I thou thee, thou traitor.' — Coke to Sir Walter Baleigh. ' If thou thou'st 
 him some thrice, it shall not be amiss.' — Twelfth Night. An extract 
 from the Journal of Gr. Fox might show that the change took place in 
 his lifetime (1624-90); but even Ben Jonson says, 'The second person 
 plural is used for reverence to a singular thing.' Compare too the rude 
 1 What trade art thou ? ' with the polite ' You, sir, what trade are you ? ' 
 — Julius Ccesar, i. 1. See De Vere, Studies in English, p. 242 seqq. 
 Guesses at Truth, i. 163-190, &c. 
 
 f Cf. Cic. Tusc. Quest, v. 103, 'Hie est ille Demosthenes.' 'Hcec ilia 
 Charybdis,' &c, Virg. Mn. iii. 558. 
 
112 A BRIEF GREEK SYNTAX. 
 
 oh' {Ιμ εγώ σοι κείνος look, I am that famous man. 
 τοϋτ εκείνο, κτάσΟ' εταίρους this is the well-known 
 proverb ' get friends.' 
 
 ix. αυτός = he himself; as 
 
 αυτός b ά ν η ρ the man himself, 
 but 6 αυτός άνηρ the same (or selfsame) man. 
 ταϋτα τα χρήματα these things. 
 ret αυτά χρήματα the same things. 
 
 x. The supposed distinction between αΰτως ' likewise ' and 
 αυτως ' in vain ' is a mere fiction of the grammarians. They 
 are one and the same word passing through various phases of 
 meaning.* 
 
 Relative Pronouns. 
 
 102. i. It has already been pointed out that ό'ς, ή, 6, was 
 originally a demonstrative, not a relative pronoun, and was 
 probably another form of o, >;, ro'.f Hence such phrases as 
 και ος and he, ή 3' δς said he, &c. 
 
 δς μεν πεινίρ ος ΰε μεθύει one man is hungry, another 
 
 drunken. — 1 Cor. xi. 21. 
 ov μεν kfoipav, ov hi απέκτειναν. — Matt. xxi. 35. 
 
 ϋ. δς = who (definite), όστις whoever, referring to a class 
 (indef.) ; οσπερ the very person who, referring to a distinct 
 person, as 
 
 εστίν ΰίκης οφθαλμός, ος τα πάν& bpq. there is an eye of 
 
 justice, which sees all things. 
 ώεύγειν μεν ούν χρη πόλεμον όστις εύ φρονεί nay rather, 
 
 any one who (quicunque) is wise should avoid war. 
 ημείς κτενοϋμεν ο'ίπερ εζεφύσαμεν I, the very person who 
 bore them, will slay them. 
 
 iii. But όστις does not always retain this indefinite sense ; 
 as η πόΧις ήτις εν ΑελφοΊς κτίζεται. 
 
 iv. The demonstrative is often pleonastic, or merely em- 
 phatic, after the relative, as 
 
 ων b μεν αυτών of which one of them. 
 
 * See Hermann, Annot. de Pronom. auras, § xv. In such phrases as 
 αυτή irpbs αΰτ^ν sola mecum, tois avrbs αυτού πήμασιν βαρύνεται, &c, the 
 aspirate shows that αυτήν, &c, are contractions for cases of the reflexive 
 εαυτόν, &c. 
 
 f Sanskrit offers a remarkable analogy to this dropping of the final s ; 
 see Monier Williams, SansJcr. Gram. § 67. 
 
EELATIVE PRONOUNS. 113 
 
 οις 'Ολύμπιοι Θεοί ΰο~ιέν ποτ αντοϊς, κ.τ.λ. 
 to whom may the Olympian gods grant in their own 
 persons, &c. 
 From the frequency of this idiom in Hebrew, we find it 
 constantly in the LXX. and N. T. See 1 Pet. ii. 24, &c. 
 
 This is precisely analogous to the English vulgarism ' which 
 it's a shame ; ' see especially Hdt. iv. 44, ' the Indus, which 
 it's the second river that,' &c. In Chaucer we find such ex- 
 pressions as ' Crist which that is to every wound triacle.' — 
 Man of Lawe's Tale. 
 
 v. όστις, οποίος, οποσος, όπως, οπού, &c* are used in 
 indirect (or repeated) questions and sentences, for τις ; ποίος ; 
 πώς; &c. Thus 
 
 τ'ις εποίησεν; who did it? ουκ olh' όστις 7jv I don't know 
 
 who it was. 
 οντάς τι ποιείς ; you sir, what are you doing ? on ποιώ ; 
 
 what, quotha? 
 πώς ΰή, φράσω εγώ. "Οπως ; ψήσει How then, / shall 
 say. How, quotha ? he will say, &c. 
 vi. The contemptuous use of ποίος, especially with the 
 article in repeated questions, should be noticed, as 
 
 ποίον τον μϋθον εειπες ; what manner of speech is this of 
 
 thine ! 
 Κ. οι πρέσβεις 61 πάρα βασιλέως. Δ. ποίον βασιλέως ; Her. 
 The ambassadors from the king. Die. Fine king for- 
 sooth !— Ar. Ach. 62; cf. 157, &c. 
 
 vii. Pronouns (and especially relatives) are peculiarly liable 
 to attraction, as 
 
 μέμνησθε ov δμωμόκατε remember the oath which you 
 
 swore. 
 χρώμαι οίς εχω βιβλίοις I use the books I have. 
 αντρον ας Μακράς κικλήσκομεν sl cave which we call Macrae. 
 
 In English, by a reverse process, the antecedent is sometimes 
 attracted into the case of the relative ; as l When him we 
 
 * These being mere luxuries, not necessaries of language, have for 
 the most part disappeared in the New Testament ; and, as usual, in 
 Modern Greek. When the question is not repeated out of any surprise, 
 irony, misapprehension, &c, then these forms are not used ; e.g. 
 
 Π. κα\ πωε iv &ρτρφ παΤδα ahv Χιπέίν %t\t]s ; 
 Kp. πό05 δ 1 ;— Ion, 958. 
 
 And how didst thou endure to leave thy child in the cave ? Cr, Ah ! 
 how indeed ! [' You may well asJc how.'] 
 
114 A BRIEF GREEK SYNTAX. 
 
 serve's away.' — Ant. and Chop. iii. 1 ; cf. Coriol. v. 5. This 
 resembles the Latin ' Eunuchum, quern dedisti nobis, quas 
 turbas/eci*.'— Ter. Eun. iv. 3. Cf. Virg. vEn. i. 573. *• 
 viii. Notice the phrases, 
 ουκ εσθ' οπον nowhere, 
 ουκ εστίν όπως nullo modo. 
 ουκ εσ& όπως ου most certainly. 
 
 δ δ' έζήλωσας ημάς quant a ce que vous nous portez envie, 
 ( as for your jealousy of us' (cf. quod in Latin). 
 ix. Notice the following pronominal adverbs : 
 
 ttws ; how ? quomodo ? πώς, somehow ; aliquo modo. 
 που ; where ? ubi ? πού, somewhere ; alicubi. 
 
 πτ} ; which way ? qua ? πή, some way ; aliqua. 
 πότε ; when ? quando ? ποτέ, at some time ; aliquando. 
 πόΐ; whither? quo ? ποί, somewhither; aliquo. 
 
 The forms forou, οπότε, &c, are used in indirect sentences ; πόΐ, πγ, are 
 the dative masculine and feminine of an obsolete pronoun πός (as $ 
 from 8s). 
 
 Indefinite Pronouns. 
 
 103. i. τίς ; = who ? Γίς enclitic = a, or α certain, 
 η τις η ουδείς scarcely any one. 
 τρεΊς τίνες some three, ' one or two.' 
 
 ii. The indefinite is sometimes politely put for the definite, 
 as we say 'some one shall smart for it'=you; κνίζω τίνα 
 I'm annoying some one == you. 
 
 iii. The indefinite τις resembles our ' one,' the German man, 
 the French on, as 
 
 τοντο ΰή τις άποκρίναιτ αν on pourrait repondre, cela ; 
 
 hoc juste responderis. 
 ποϊ τις τρέφεται ; whither shall one turn oneself? 
 
 iv. ό ΰε'ινα i a certain person,' f so and so/ some one whom 
 we do not know, or do not choose to name. 
 
 6 δείνα και 6 δείνα = ' John Doe and Richard Roe,' 'Brown, 
 Jones, and Robinson ; ' compare the Latin i Caius et 
 Sempronius.' 
 
 v. Observe the phrases, 
 
 τι παθών; from what cause? 
 
 τι μαθών ; on what inducement ? ^ = why ? 
 
 τι έχων ; with what reason ? 
 
 ri yap ; why then ? 'ίνα τί ; why ? 
 
 τι μην; of course ! why not ? 
 
VERBS. 115 
 
 Distributive Pronouns. 
 
 104. i. "Αλλος alms, another; έτερος* the other of two, 
 alter ; έκαστος unusquisque, εκάτερος uterque. 
 
 άλλοι = others ; ol άλλοι the rest, ccetert. 
 
 οι έτεροι the opposite party, pars altera ; ετερόφθαλμϋς 
 having lost one eye. 
 
 μετατίθεσθε . . . εις έτερον ευαγγέλιον, ο ουκ εστίν άλλο, 
 Gal. i. β, Ye are changed to a quite different Gospel, 
 which is not another of the same kind (Clyde). 
 
 ii. By a curious apposition of άλλος with its substantive, we 
 get the common Greek form of expression, ' sheep and other 
 camels ' = sheep, and other animals, viz. camels ; as 
 
 νπο των πολιτών καϊ των άλλων ξένων, Plat. Gorg. 473 C, 
 
 by the citizens and the rest, viz. foreigners. 
 ηγοντο ΰε και έτεροι ΰύο κακούργοι συν αυτω αναιρεθήναι, 
 Luke xxiii. 32, And two different persons, viz. male- 
 factors, were led to be crucified with him (not as in 
 the Eng. Ver. l two other malefactors '). 
 
 N.B. "Αλλο καί άλλο one thing after another. 
 
 άλλος άλλο λέγει one man says one thing, another another. 
 Cf. ' Alia ex aliis in fata vocamur,' JEn. iii. 496, We are 
 
 summoned into one destiny after another. 
 ' Alii alio intueri,' Liv. ix. v. 8. 
 
 It will be seen how much more awkward is the English 
 idiom. 
 
 THE VEKB. 
 
 105. i. The very name Verb (ρήμα verbum) implies that it 
 is the word, the most important word, in the sentence (see 
 §69). 
 
 ii. The forms of verbs may be tabulated thus : 
 
 Verbs. 
 
 Transitive. Intransitive. 
 
 I I I I 
 
 Active. Deponent. Neuter. Passive. 
 
 * erepos, Sanskrit antaras, Grerm. ander, &c. 
 
116 A BRIEF GREEK SYNTAX. 
 
 VOICES (διαθέσεις). 
 
 106. A Greek verb has three voices, active, passive, and 
 middle. 
 
 107. Active Voice. — We have already seen that the reason 
 why so many transitive verbs have also an intransitive mean- 
 ing, is that the latter is the older meaning out of which the 
 other was developed. 
 
 108. Deponent Verbs have only a middle form, and it is 
 probable that they were all originally reflexive. It is not 
 surprising that many deponents have also tenses of a passive 
 form (e.g. είεζάμην excepi, εοέχθην exceptus sum; εβιασάμην 
 coegi, εβιάσθην coactus sum, &c.) ; or that their tenses are 
 used in a passive sense,* as is so commonly the case with the 
 future middle (άρζομαι, τιμησομαι, ΰηΧώσομαι, Χέζομαι, κηρύ- 
 ζομαι, αλώσομαι, &C.). 
 
 109. i. Passive Voice. — The passive form implies that the 
 subject of the proposition is not the agent ; the agent is usually 
 expressed by υπό with the genitive, or, in verbs which imply 
 comparison, by the genitive alone ; also by εκ (poet.), and 
 πάρα. (more rarely by προς and από) with the genitive ; and, 
 especially after the perf. pass., by the dative case ; as εμοϊ 
 πέπρακται rovpyov the deed has been done by me. 
 
 ii. Even those verbs which govern a genitive or dative may 
 in Greek be used passively, and this genitive or dative may 
 become the subject of the passive verb; e.g. άποτέμνειν τίνος 
 την κεφαλήν, and in the passive οι στρατηγοί απυτμηΟεντες τας 
 κεφάλας ; πιστεύω τινί τι, and in the passive πεπ'ιστενμαί τι 
 I have been entrusted with something. 
 
 N.B. Notice the difference between the Greek and Latin 
 idiom in -φεύστης ου πιστεύεται mendaci non creditur. 
 
 110. Middle Voice. — The middle voice always refers to 
 self in some relation or other, which may be expressed a. by 
 the genitive, b. dative, c. accusative, or d. by a pronominal 
 adjective; as 
 
 α. άπωσάμενος pushing away from myself. 
 b. παρασκευάζομαι I prepare for myself. 
 
 * Just as, on the other hand, some passive forms are used in the 
 sense of neuters, as πορενθηναι to march, κοιμηθηναι to sleep, φοβηθηναι, 
 απαΚΧαγηναι, «fee. In later Greek, the middle is often used in a passive 
 sense. Such peculiarities cause no practical confusion ; in French the 
 reflexive verb is often passive, as in ' Votre heureux larcin ne se peut 
 plus celer.' — Kacine. 
 
THE MIDDLE VOICE. 117 
 
 C. άπάγζασθαι to hang oneself. 
 
 d. τύπτομαι την κεφαλήν I beat my own head. 
 
 In later Greek a reflexive pronoun with the active is often 
 used instead of the middle, as ζωννύειν εαυτόν, John xxi. 18 ; 
 and this reflexive pronoun is even added to the middle, as" 
 διεμζρίσαντο εαυτοΊς, John xix. 24. The gradual obsolescence 
 of the middle in the New Testament appears from its being 
 sometimes used indifferently with the active (cf. συγκαλεί, 
 Luke xv. 6, with συγκαλείται, id. 9). 
 
 111. There are four chief uses of the middle. 
 i. Simply reflexive, as λούομαι I wash myself. 
 
 ii. Causative, as παρατίθεμαι τράπεζαν I get a table spread 
 for me ; διδάσκομαι τον υ'ιον I get my son taught (do- 
 cendum euro). This is like the German reflexive 
 (sich) lassen. 
 
 iii. Indirect or appropriative, as παρασκευάζομαι τα επιτήδεια 
 apparo mihi commeatum ; κατεστρέφατο τον Μήδον he 
 subdued the Mede to himself; πράττομαι χρήματα I 
 get myself money. 
 
 iv. Eeciprocal, as τύπτονται they strike each other ; ώβτι- 
 οϋνται they jostle each other ; κελεύονται they exhort 
 each other ; διαμάχονται they fight each other. (Cf. 
 the Latin deponents convicior, cohortor, &c.) 
 
 Sometimes too a distinctly reflexive middle takes an accusa- 
 tive of the object affected by the state, as in Homer, έ'ιπερ αν 
 αύτον Σεύωνται τα-χεες τε κύνες even though swift dogs should 
 stir themselves in pursuit of him; εκόπτοντο αυτήν (Luke viii. 
 52) they beat then' breasts for her. Cf. Aristoph. Lys. 397. 
 
 112. Notice the difference of θε"ιναι νόμους of a despot ; 
 θεσθαι νόμους of a legislator who will himself be bound by the 
 laws he makes. 
 
 θεΊναι οΐκίαν to mortgage a house ; θέσθαι οϊκίαν to take 
 
 a house on mortgage. 
 Χϋσαι to set free ; λύσασθαι to ransom. 
 χρησαι to lend (or give an oracle) ; χρήσασθαι to borrow 
 
 (or consult an oracle). 
 δανείζω I lend; δανείζομαι I borrow. 
 λανθάνω I lie hid; λανθάνομαι I forget. 
 φοβέω I frighten ; φοβούμαι I fear. 
 παύω I make to cease ; παύομαι I cease. 
 αιρεω I take; αίροΰμαι I choose. 
 βουλεύω I counsel ; βουλεύομαι I consult. 
 
118 A BRIEF GREEK SYNTAX. 
 
 άποδίδωμι I restore ; αποδίδομαι I sell. 
 
 περιδίδωμι I give round ; περιδίδομαι I wager. 
 
 γράψω I enrol ; γράφομαι I indict. 
 
 φράζω I speak ; φράζομαι I think. 
 
 μισθώ I let ; μισθοϋμαι I hire. 
 
 πείθω I persuade ; πείθομαι I obey. 
 
 άρχω I rule ; άρχομαι I begin. 
 
 στέλλω I send ; στέλλομαι I set out. 
 
 γαμώ cfofco uxorem (of a man); γαμοϋμαι nubo (of a 
 
 woman). 
 σπένδω I pour a libation ; σπένδομαι I make a truce. 
 σκοπώ I look ; σκοπονμαι I look mentally, I consider. 
 ποιώ λόγον I compose a speech ; ποιούμαι λόγον I make 
 
 a speech. 
 πόλιτευω I am a citizen ; πολιτεύομαι I live as a citizen. 
 
 The last two instances are typical of many others. 
 
 113. The following passages will illustrate some uses of the 
 middle : 
 
 'Λνερα τις λιπόγυιον υπέρ νωτοιο λιπαυγης 
 
 ?)γε, πόδας χρησας, όμματα χρησάμενος (Anthol.) a blind 
 man was carrying on his back a lame man, lending his 
 feet, borrowing his eyes. 
 
 εκείνος ουκ έγημεν άλλ' εγημα~ο (Anacr. 84) he didn't 
 marry her, but she married him (of a henpecked hus- 
 band ; comp. Martial's ' uxori nubere nolo meae,' I don't 
 want my wife to marry me). 
 
 Toy τε άετον άνεσωσάμην και τον στρατοπεδάρχην 'έσωσα 
 (Dion Η. iv. 2088) I saved my eagle and saved the 
 tribune. 
 
 αϊτείτε κα\ ου λαμβάνετε, διότι κακώς αΐτεϊσθε (Jas. iv. 2) 
 Ye ask and receive not, because ye ask for yourselves 
 amiss. 
 
 114. It will be observed that the active form of verbs is 
 often used when the meaning is simply physical, the middle 
 when some action of the mind is involved ; compare, for in- 
 stance, ποιεΊν δώμα and ποιεισθαι άναβολήν, βρόχους άπτειν 
 and αφασθαι πέπλων (sc. in supplication), ώρεξε κύλικα and 
 παιδός ώρέίατο. 
 
 Ν.Β. i. The Hebrew middle voice (Hithpael) is closely analogous to 
 the Greek, and is similarly reflexive, indirect, and reciprocal. (Ewald, 
 Hebr. Gram. § 243.) 
 
 ii. The middle voice exists in Latin, though not developed to the same 
 extent as in Greek ; e.g. accingi, to gird oneself ; provolvi ad pedes t to 
 
NINE POSSIBLE TENSES. 119 
 
 roll oneself at a person's feet; misceri, to mix with others; mutari, to 
 change ; vertor, versor, volvor, plangor, circumfandor, &c. 
 
 iii. There is no middle voice in English ; in such sentences as ' the 
 book reads badly,' ' the doors open at six,' &c, the verbs are merely 
 transitives used intransitively. The same remark applies to many 
 Latin verbs, such as muto, &c. 
 
 iv. The name Middle is clearly defective, since it is as active as the 
 Active ; it is also a name of little meaning (see Clark, Comparat. Gram. 
 p. 182). 
 
 Tenses (χρόνοι). — Comparison of the Greek, Latin, and 
 English Verbs. 
 
 115. A tense (tempus χρόνος) is properly speaking a form 
 of the verb which by its termination (or inflection) expresses 
 time. 
 
 116. There are- two main classes of tenses, primary and 
 historical. 
 
 Since there are only three primary modes of regarding time, 
 viz. present, past, and future,* the three primary tenses are 
 
 1. Present (6 ένεστως χρόνος). 
 
 2. Perfect (or past, perfectum= finished) (6 παρακείμενος). 
 
 3. Future (ό μέλλων). 
 
 All the other tenses are called historical, f viz. aorist (αόριστος), 
 imperfect (παρατατικός), and pluperfect (νπερσυντελικός). 
 
 117. Observe that the 3rd pers. dual of the primary tenses 
 (and also of the subjunctive mood) ends in ov ; but the 3rd 
 pers. dual of the historical tenses (and of the optative mood) 
 ends in ην. 
 
 Besides this difference, simple reduplication belongs mainly 
 to the primary, and the pure augment only to the historical 
 tenses. 
 
 118. Since any action can only be regarded as either 
 1. present, 2. past, or 3. future; and since every action may 
 be a. finished, or perfect ; β. going on, i.e. unfinished, or 
 imperfect ; and γ. indefinite ; it is clear that any verb, to 
 be faultlessly synthetic, would provide nine tenses J in the 
 
 * Hence the inscription on the veil of the mystic Isis, ' I am that 
 \rhich is, hath been, and shall be.' — Plut. Tsid. ix. 
 
 f This distinction of primary and historic tenses applies mainly to 
 the indicative, and with far less precision to the other moods ; e.g. in 
 the imperative \4£ov is as much a primary tense as Aeye. 
 
 | The number of tenses varies greatly in different languages. In 
 Sanskrit there are six, in Hebrew only two, in French five, in English 
 
120 
 
 A BRIEF GREEK SYNTAX. 
 
 indicative mood, viz. three past tenses, three present tenses, and 
 three future tenses ; or, which is another way of expressing 
 the same thing, three tenses (past, present, and future) to ex- 
 press that an action is, was, or will be going on ; three (past, 
 present, future) to express that it is, has been, or will be 
 finished ; and three (past, present, future) to express that it is, 
 has been, or will be indefinite. [' Nulla dum temporis habitd 
 ratione, res quceque potest tripliciter significari, et ut futura, 
 et ut incfioata, et ut absoluta. Jam tempus in universum tri- 
 plex est, prceteritum, instans, futurum. 1 — Keizius.] 
 
 119. These tables may be tabulated thus, and a thorough 
 mastery of their classification is essential to a right under- 
 standing of tenses. It is easy to master, and when once 
 mastered, cannot well be forgotten :* 
 
 1 . Three present tenses — 
 
 Time, 
 a. Finished or per- *ζ 
 
 feet ... .J 
 β. Unfinished ογΊ 
 
 imperfect . ./ 
 7. Indefinite or~j 
 
 aorist . . .J 
 
 a. Finished or per- Ί 
 feet ... .J 
 β. Unfinished or\ 
 
 imperfect 
 7. Indefinite 
 aorist . 
 
 :}l 
 
 English. 
 I have (sc. now) dined 
 
 I am dining. 
 
 I dine. 
 
 Three past tenses- 
 I had dined. 
 I was dining. 
 I dined. 
 
 Greek and Latin. 
 f δίζς'ιπμηκα 
 \cmiavi. 
 f Senryco 
 \ coeno. 
 
 f [wanting both in 
 L Greek and Latin] .f 
 
 \ cainaveram. 
 f iSeiwovy 
 \ ccenabam. 
 
 {έδείπνησα 
 [wanting in Latin]. 
 
 two, &c. It will be observed that I confine the name tense to actual 
 inflected forms of the verb, and do not include in it compound tenses, 
 i.e. expressions formed by auxiliaries. 
 
 * Harris, in his celebrated Hermes, has the credit of originating (by 
 improvements on the hints of the Stoics and Varro) this very lucid and 
 philosophical view of the tenses. It is admirably developed in a useful 
 book of Mr. F. Whalley Harper's — Powers of the Greek Tenses. An 
 inferior but ingenious tabulation had been previously given in S. Clarke's 
 note on Horn. H. i. 37, which Wolf called the best note in his edition. 
 For a vast amount about the whole subject, see Herm. Schmidt, Boctrina 
 Temporum verbi Graci et Latini, 1836. It was partially, but indepen- 
 dently, elaborated by Reizius, Dissert, de temporibus et modis verbi. 
 Lips. 1766. Burnoufs classification, adopted by Donaldson and others, 
 appears to me much less accurate and philosophical. 
 
 f The unfinished present or present-imperfect, Scivvw, coeno, used 
 instead. 
 
SCHEME OF TENSES. 
 
 121 
 
 Time. 
 
 a. Finished or per- 
 fect . . 
 
 β. Unfinished 
 imperfect 
 
 7. Indefinite or"l 
 aorist . . ./ 
 
 :} 
 
 Three future tenses — 
 
 English. 
 I shall have dined. 
 
 I shall be dining. 
 I shall dine. 
 
 Greek and Latin. 
 f [wanting] 
 \ coenavero. 
 f [wanting both in 
 \_ Greek and Latin].* 
 
 \ canabo. 
 
 120. Or we may have the same scheme reversed, and as it 
 is very important that it should be understood, let us give 
 it in the reverse order, as follows : 
 
 Three finished or perfect tenses — 
 
 Time. 
 
 1. Present 
 
 2. Past . 
 
 3. Future 
 
 English. 
 1 have (now) dined 
 I had dined 
 I shall have dined 
 
 Greek. 
 ZeUeiirvnKa 
 
 [wanting] 
 
 Latin, 
 coenavi 
 coenaveram 
 caenavero 
 
 β. Three unfinished or imperfect tenses — 
 
 1. Present 
 
 2. Past . 
 
 3. Future 
 
 1. Present 
 
 2. Past . 
 
 3. Future 
 
 lam dining 
 I was dining 
 I shall be dining 
 
 δειπνω 
 
 iZsiiTvovv 
 
 [wanting] 
 
 coeno 
 
 cmiabam 
 
 [wanting] 
 
 y. Three indefinite or aorist tenses— 
 
 / dine 
 I dined 
 I shall dine 
 
 [wanting] 
 ϊδείττνησα 
 δςιπνησω 
 
 [wanting] 
 [wanting] 
 ccenabo 
 
 * "Έ,σομαι δείπνων (comp. New Testament, Matt. xxiv. 9 ; ίσεσθε μισού• 
 μ*νοι, Luke i. 20, v. 20) would be admissible for the future-imperfect 
 1 1 shall be dining; ' and this is an approach which the Greek verb makes 
 to the use of auxiliaries for the purpose of conjugation. But the instances 
 are not common, as νεποιτηκώς %σομαι I shall have done it. — Isoc. ir. αντιδ. 
 §317. ουκ€τ' €/c καλυμμάτων \ ίσται δβδορκώί. — iEsch. Ag. 1178. ye' /ραμ- 
 μένο* %σθα you were painted. Of course we find the auxiliary in the 
 moods of the perfect passive τετυμμίνος ω, &c. Another instance of this 
 tendency is the occasional resolution of a future into Θ4\ω or μέλλω with 
 the infinitive, an analytical proceeding which has ousted the synthetic 
 future from modern Greek ; as θα πο\εμώμ*ν we shall be fighting ; θα 
 %χω I shall have. Such forms as ατίμασα* έχει, Soph. ; ί}τ€ πάσχοντες 
 raSe, Eur., are not mere auxiliaries, but periphrases adopted to imply 
 continuance (cf. Ps. cxxii. 2 ; Heb. Matt. vii. 29) ; and the same remark 
 applies to the σχήμα Χαλκιζικον (or Oropism) of τυγχάνω, υπάρχω, &c, 
 with various participles (cf. Mark i. 4). 
 
 G 
 
122 
 
 A BRIEF GREEK SYNTAX. 
 
 Or the same arrangement might be tabulated as follows : 
 
 OBJECTIVE TENSES 
 (i.e. tenses of the Indicative, expressive of facts). 
 
 DEFINITE. 
 
 INDEFINITE, or AOEISTIC. 
 
 I 
 
 Present, Past. Future. 
 
 [wanting] ΐδίίπνησα δΐίπντησω 
 I dine. I dined. I shall dine 
 
 Fd^SHED, or 
 
 Perfect. 
 I 
 
 Unfinished, or 
 Imperfect. 
 
 I II III 
 
 Present. Past. Future. Present. Past. Future. 
 
 8eSeLnv7)K.a eSeSetirvrJKeiv [wanting] SeinvCi ίδείπνονν [wanting] 
 
 I have dined I had dined I shall have I am dining I was dining I shall be 
 
 (now) coenaveram dined coeno. ccenabam. dining Ισομαι 
 
 ccenavi. coenavero. Seirrvuv ccena- 
 
 turus sum. 
 
 121. This scheme of tenses suggests several important 
 remarks and inferences. 
 
 1. Observe that it offers us a means of comparing the 
 Greek, the Latin, and the English verb, and that taking the 
 word ' tense ' to mean an inflected verbal-form significant of 
 time, there are 
 
 In Greek six of the nine tenses ; 
 In Latin six „ „ 
 
 In English two „ ,, 
 
 The six Greek tenses are not however the same as the six Latin, 
 for Greek has a separate aorist (εΐείπνησα) which Latin has 
 not ; * and Latin has a future perfect (coenavero) which Greek 
 has not (except in rare forms like εστήζω, τεθνιβω). The 
 only tense which is wanting both in Greek and Latin is the 
 aorist-present or indefinite-present ( { 7" dine '), vohich strange 
 to say is one of the only two tenses which English possesses ; 
 
 * It lias been said that ' the superiority of the Greek verb to the 
 Latin, consists in the possession of another voice, another mood, another 
 tense, and a much greater variety of participles.' This judgment is by 
 no means correct. We shall see hereafter that Latin is not destitute 
 of a middle ; that the optative is no mood at all, but merely a name for 
 past tenses of the subjunctive, and that Latin has an optative ; that if 
 it has no separate form for the past-aorist (I dined, εδβίπνησα) it has 
 on the other hand in the active a future-perfect (coenavero, I shall have 
 dined), which Greek has not ; and that, although it has fewer parti- 
 ciples, it has gerundives and supines which are wanting to Greek. 
 
NOMENCLATURE OF TENSES. 123 
 
 the other English tense, the aorist-past or indefinite-past 
 (' I dined '), being also wanting in Latin, though it exists in 
 Greek (εΰε'ιπνησα). 
 
 The other so-called tenses of the English verb (I have dined, 
 I shall dine, &c.) are not properly speaking tenses at all, not 
 being formed by inflection, but by a mere use of the auxiliary, 
 which is much less neat and expressive than the synthetic or 
 inflectional forms of Greek and Latin. 
 
 2. Observe particularly that, whenever strictly and properly 
 used, 
 
 τύπτω is not ' I strike,' but ' I am striking.'* 
 τνπτομαι is not ' I am struck,' but ' I am being struck.' 
 
 In other words, they are unfinished (imperfect) tenses ; and if 
 the tenses were at all correctly named, τύπτω, τΰπτομαι would 
 not be called presents (as though there were only one present 
 in each voice, whereas as we have seen there are three) but 
 present-imperfects. Thus ΰείκννται ταϋτα is, ' these things are 
 being proved] but most boys would render it quite wrongly, 
 1 these things are proved,' which would be the rendering (not 
 of Ιάκννται but) of οέίεικται. Frequently indeed, just as the 
 Greeks have no present-aorist, and sometimes use the present- 
 imperfect for it (i.e. they say ΰειπνώ ' I am dining' when they 
 mean ' I dine '), so we translate their present-imperfect by our 
 present-aorist; thus 
 
 Στρ. πρώτοι' μεν 'ότι ΰρφ,ς άντίβολώ κάτειπέ μοι. 
 
 Σω/φ. άεροβατώ και περιφρονώ τον ήλων. 
 
 This has been racily rendered 
 
 Streps. First tell me, I implore, what are you doing ? 
 Socr. I tread the air and circumspect the sun. 
 
 But literally it is, ' I am treading the air,' &c, which is much 
 more vivid in Greek ; it would also be more vivid in English, 
 but for the intolerable awkwardness of the English periphrasis 
 ('lam' with the present-participle) for the Greek present- 
 imperfect. 
 
 The translators of our English Version have failed more 
 frequently from their partial knowledge of the force of the 
 tenses than from any other cause, and their neglect of the 
 continuous meaning of the present often loses us lessons of 
 profound significance ; e.g. in Col. iii. 6, oY α έρχεται ή op -γη 
 
 * So that in this respect Greek is the reverse of German, which has, 
 like the English, a present aorist {ich Use, I read), but no present imper- 
 fect, ' I am reading,' for which they must use ich Use jeU;t or eben. 
 
 g2 
 
1^4 A BRIEF GREEK SYNTAX. 
 
 του θευν επ\ τανς νιείς της απείθειας on which account the 
 wrath of God is ever coming upon, &c, i.e. by a process of 
 natural laws; Matt. xxv. 8, αί λαμπάδες ημών σβεννννται our 
 lamps are going out, are being quenched, not ' are gone out.' 
 
 3. Clearly then the present nomenclature of tenses is very 
 misleading unless we are specially careful to see through it, 
 and not suffer it to mislead us ; it is of course far too deeply 
 rooted to be superseded, but any one who has understood the 
 above tables will see that 
 
 The so-called present is a present-imperfect : 
 
 1 I am dining ; ' i.e. an action is going on, which is not 
 yet finished. 
 The so-called imperfect is a past-imperfect : 
 
 ' I was (at some past time) dining ' (and the action 
 was not finished). 
 The so-called perfect is a present-perfect : 
 
 ' 1 have (at this moment) dined.' 
 The so-called pluperfect is a past-perfect : 
 
 i I had (at some past time) dined,' or ' finished dining.' 
 The so-called aorists (1st and 2nd) are past-aorists : 
 
 i I (at some time or other not specified) dined.' The 
 Greek has no present-aorist, ' I dine.' 
 The so-called future is a future-aorist : 
 
 * I shall (at some time or other not specified) dine.' 
 
 4. It may be asked why in the above scheme no notice is 
 taken of the second aorist? Simply because the first and 
 second aorists, when both exist, are merely two different forms 
 to express the same* meaning. 
 
 122. The terms first and second aorist are misleading; 
 indeed the second aorist is always the older form of the two ; f 
 for the second aorist is formed directly from the stem, thus 
 preserving the simplest form of the verb, and its most un- 
 qualified meaning (e.g. ετνπον from τνπ), whereas the first 
 aorist is formed not only by the prasfix of an augment, but 
 
 * The same remark applies to the first and second perfect, except 
 that in this case it is disputed among grammarians "which of the two 
 forms is really the older. The grounds on which Donaldson decides in 
 favour of the second perfect being a younger and mutilated form, seem 
 to me wry unconvincing. (New Crat. p. 566.) 
 
 flew verbs have both the first and second aor. in use. The exist- 
 ence of two forms, one older and more recent, side by side, may be 
 paralleled by the English, as in clomb climbed, squoze squeezed, clave 
 cleft, &c. The archaic forms clomb, squoze, clave, &c, are analogous, tq 
 the Greek second aorist (so-called). 
 
NATURE OF THE AOMST. 125 
 
 also by the suffix of the letter σ (which is no doubt connected 
 with έσ-μι, εσ-τι), denoting futurity. 
 
 The reason why the first and second aorist have the same 
 meaning is because the second aorist (e.g. ετυπον) by simply 
 prefixing the augment to the pure stem of the verb, implies a 
 momentary action in the past. And the first aorist by pre- 
 fixing the augment (which indicates past time) and suffixing σ, 
 which indicates future time, implies an action which was future 
 and is past, i.e. an indefinite past action, which thus coincides 
 in meaning with the second aorist.* (Clyde, Gk. Syntax.) 
 
 123. The student should avoid rendering the aorist by 
 1 have,'' which is the sign of the present-perfect. It is indeed 
 true that the Greeks sometimes used the aorist indicative 
 where we use the perfect, and in this case we must substitute 
 our idiom for theirs ; but this does not obliterate the distinc- 
 tion between the aorist and perfect (see note f, next page). 
 
 124. Whatever difference there is in English between 
 I dined (e.g. ten years ago at Kome) 
 
 and 
 
 I have dined (this evening),! 
 the same difference exists in Greek between 
 
 εΰείπνησα sss I dined. 
 
 ΰεΰείπνησα = I have dined. 
 
 It is one of the main defects of the indicative of the Latin 
 verb, that it is obliged to use one form ccenavi for these two 
 very different meanings. In fact the existence of the aoristic 
 termination in such perfects as vixe, scrip-si, &c. shows clearly 
 that in Latin verbs there is sometimes a perfect, formed by 
 reduplication, and sometimes an aorist substituted for it. Thus 
 
 * Curtius calls the second the strong, and the first the weak aorist, 
 because the latter is formed by extraneous additions to the stem. Thus 
 in English • I took ' is a strong aorist, being formed from ■ I take ' by a 
 modification of the vowel (called by Pott a qualitative change, as in 
 Hebrew, and named by German philologists laut, and by the French 
 apophonie, as in sing, sang, sung) ; but ' I loved ' is a weak aorist, being 
 I love-d = I love-did, and thus being formed by an auxiliary. In fact 
 the strong aorist erxmov differs from the weak ετυψο hardly more than 
 λ€λυτ<Μ does from solutus est. 
 
 f Burnouf says, ' Le parfait exprime une action accomplie, mais dont 
 l'effet subsiste au moment ou Ton parle ; tandis que l'aoriste presente 
 Taction comme simplement passee ;' e.g. if I say 'he has lived well,' I 
 can only be speaking of some one yet alive, or just dead ; if I say ' he 
 lived well,' I may be referring to any one since the days of Adam. 
 
126 A BRIEF GREEK SYNTAX. 
 
 the Latin perfect has both meanings, but is more oflen an aorist 
 than a perfect. This accounts for the fact that veni ut videam 
 and veni ut rider em are both right ; the former meaning ' I 
 have come that I may see,' the latter ' I came that I might see.' 
 It is extremely probable that a slight difference in pronuncia- 
 tion may have helped to distinguish between the meanings.* 
 
 125. The aorist, which most English boys look upon as some 
 unknown Greek monster, ought to be the most familiar tense 
 of all, because the only tenses in their own language are aorists ; 
 * I dine ' (the present aorist), ' I dined ' (the past aorist). 
 
 126. The word aorist, which is first found in Dionysius 
 Thrax, simply means indefinite,•)• being derived from a not, 
 and ορίζω I limit (whence comes our word horizon, the bound- 
 ing line). A boy usually takes • I dine,' ' I strike,' &c, for pre- 
 sents, and ' I dined,' ' I struck,' &c, for perfects ; yet in answer 
 to the question ' what are you doing ? ' he would not dream 
 of using the aorist ' I dine,' but the present ' I am dining;' 
 nor when leaving the table would he say ' I dined,' but ' I 
 have dined.' 
 
 127. Thus it will be seen that the aorist, as the tense of 
 narration, the tense in which all history is written, is one of 
 the most necessary tenses of all ! Consequently it is more 
 important and more frequently used than the perfect, which 
 belongs to the present rather than to the past. Hence in 
 Modern Greek the aorist has almost superseded the perfect, 
 and the so-called Latin perfect is far more frequently aoristic 
 in sense. 
 
 128. Very rarely indeed we are compelled by the English 
 idiom to introduce the present-perfect (or perfect with ' have ') 
 in rendering the aorist (especially the aorist participle) ; \ but 
 
 * BurggrafF suggests that when the aorist meaning was intended, the 
 word may have been pronounced slightly more rapidly. {PHncipes de 
 Gram. Gen. p. 373.) 
 
 f It is the same word as ' infinitive,' which also means ' indefinite,' 
 being a form of the verb not limited to any subject. Curiously enough 
 the aorist is called in French 'le preteri defini' (e.g. fecrivis). The 
 reason is that it is definite with reference to some other action which 
 may be in the mind ; e.g. Ά Varrivee du messager j'ecrivis unelettre.' 
 Greek often uses it when no other term to mark time is employed ; but 
 French does not. E. Burnouf, Grammaire grecque, § 60. 
 
 I ' χρονικά i -πφρ-ηματα aoristo conjungi solent ; &ρτι έιτοίησα, πολλάκις 
 έθαύμασα, &c. ; unde naturam perfecti quodam modo induere videtur.' 
 Shilleto on Demosth. JDe Feds. Legat. § 228. Mr. Cope (Pre/, to his 
 edition of the Gorgias, p. xvi.) quotes %<pvyov κακόν, evpov άμαιον, the 
 
AORISTS AND IMPERFECTS. 127 
 
 the rule is, never translate the aorist by ' have.'' The past- 
 aorist must often be rendered by a present aorist, because the 
 Greek uses it in this sense, having, as we have seen, no special 
 form for the present aorist ; e.g. ( many things happen contrary 
 to experience,' would be in Greek πολλά πάρα γνώμην 'έπεσε, 
 
 129. Unless the student is alive to the true nature of the 
 aorist, and the fact that it is often used with imperfect tenses 
 to express the contrast between momentary and continuous 
 actions, he will miss half the beauty and picturesqueness of 
 the best Greek authors. 
 
 Take some instances : 
 
 ΚροΤσος "Αλυν ΰιαβάς μεγάλην αρχήν καταλύσει not 
 
 ' having crossed the Halys,' but ' Croesus on crossing 
 
 the Halys will ruin a great kingdom.' 
 παθών Ik τε νηπως εγνω i even a child learns by suffering, 1 
 
 not ' having suffered.' 
 γελάσας• είπε not ' having laughed,' but ' he exclaimed, 
 
 laughing,' or ' he burst out laughing, and said.' 
 
 130. In our English version of the Bible the aorist is often 
 wrongly rendered by have, and the picturesque difference 
 between aorists and imperfects lost ;* e.g. 
 
 Luke viii. 23 : 
 
 κατεβη λαΐλαψ . . . καϊ σννεπληροΰντο there came down a 
 gust of wind and they (not ' were filled,' but) began to 
 be filled. 
 
 Mark vii. 35 : 
 
 ελνθη 6 δεσμός της γλώσσης αυτόν, και ελάλει ορθώς the 
 string of his tongue was loosed, and he began to speak 
 plainly. 
 
 John vii. 14 : 
 
 άνέβη . . . και εΰίΰασκεν went up, and began to teach. 
 
 exultant cry of the newly-initiated, as an instance of the aorist where we 
 should use the perfect. All such cases prove, not any identity of mean- 
 ing between the tenses, but a different intellectual stand-point; the aorists 
 here (as in Modern Greek) express merely a finished past action, with no 
 reference to the time of completion. And the same is true of the gnomic 
 aorist (§ 154) ; e.g. in such a line as ' Qui ne sait se borner ne sut jamais 
 ecrire' (Boileau), either 'ne sait pas,' or 'n' a jamais su' would have done 
 equally well ; but this does not prove any identity between the tenses. 
 As we have no aorist participle or infinitive, we must, of course, some- 
 times use the auxiliary ' have ' in rendering those forms. 
 
 * German, like Latin, has no aorist; it therefore uses the imperfect 
 regularly in its place. 
 
128 A BRIEF GREEK SYNTAX. 
 
 John xii. 13 : 
 
 εζήλθον . . . και εκραζον went out, and kept crying., 
 John xiii. 27 : 
 
 υ ποιείς ποίησον do (at once) what you are about. 
 Acts xi. 6 : 
 
 άτενίσας κατενόονν και εϊδον gazing, I began to distinguish 
 
 (impf.), and saw (aor.), &c. 
 κρεΊσσον γαμήσαι η πυρυΰσθαι it is better to marry (once 
 for all) than to be burning. 
 
 In Matt. iii. 7, 8, ποιήσατε τους καρπούς is not 'bring 
 forth,' but ' have done bringing forth,' i.e. do it once for all. 
 See, too, John vii. 8, 24", xii. 6, xvii. 12. 
 
 131. In classical Greek take one or two further instances : 
 Nub. 233 : 
 
 έ"ιπερ βάλλει τους επίορκους πως ουχί Έιίμων ενεπρησε ; 
 1 If his way is to strike the perjured, why does he not 
 blast Simon ? ' 
 
 ol "Ελληνες επαίανιζον . . . και αμα τα δόρατα καθίεσαν' 
 ενταύθα ουκέτι είεζαντο οι πολέμιοι αλλ' εφευγον the 
 Greeks began the war song, and at the same moment 
 levelled their spears; whereon the enemy no loDger 
 awaited them, but began to fly. 
 
 Iph. Taur. 1306 : 
 
 άνωλόλνξε και Karjjfo ' She raised her voice, and began to 
 sing* 
 
 Plat. Parmen. 127 : 
 
 εβαΐΐζομεν και κατελάβομεν τον ' Αντιφώντα we were walk- 
 ing and overtook Antipho. 
 
 χαλεπό»' το ποιεϊν το ΰε κελεΰσαι ροίΰιον it is difficult to 
 carry out a thing, but to give the order is easy. 
 
 μη τύπτε do not be striking (a general prohibition) ; μ 17 
 τύψης do not strike (a special prohibition).* 
 
 εάν τις κάμνη των οίκετών should any of the servants be 
 - sick \_κάμη=8Ϊιοχάά fall sick] παρακαλείς Ιατρούς όπως 
 μη άποθάνη. 
 
 τούτον ημείς φοβώμεθα ; are we to be afraid of him ? τού- 
 τον ημείς ψοβησώμεθα ; are we to take alarm at him ? 
 
 * Donaldson points out that in John xx. 17, μ^ μου απτού is not 
 ' touch me not ' (which would be όίψρ), but ' do not be clinging to me ' — 
 a most important difference. 
 
USES OF THE TENSES. 129 
 
 132. Owing to the use of the past-aorist [e.g. εΐείπνησα] 
 to supply the absence of any present-aorist [Ί dine'] in 
 Greek, many past-aorists have permanently acquired a present 
 sense, as ηνεσα I praise, άπέπτυσα I hate, ε θαύμασα I wonder, 
 εΰεξάμην I accept, &c. For a list of such expressions see 
 Hermann in Vige?wm, 162. Dr. Clyde thinks that the usage 
 may have gained ground because a personal statement becomes 
 less obtrusive if put into a past tense (cf. odi, novi, &c). 
 
 133. The same scheme of tenses might of course be made 
 for the passive, the only difference being (which is curious) 
 that in the passive the Latin has not and the Greek has a 
 future-perfect. What anomaly it was which gave the Greek 
 a form for ' I shall have been struck,' and no form for * I shall 
 have struck' cannot be explained.* 
 
 In the passive, therefore, we have 
 
 Three finished tenses, or perfects. 
 Present. I have been struck . . τέτυμμαι verberatus sum. 
 Past. I had been struck . . ετετΰμμην verberatus eram. 
 Future. I shall have been struck τετύψομαι verberatusfuero. 
 
 Three unfinished tenses, or imperfects. 
 Present. I am being struck . . τύπτομαι verberor 
 Past. I was being struck . . ετυπτόμην verberabar. 
 Future. I shall be being struck . [wanting] [wanting]. 
 
 Three aorist tenses, or indefinites. 
 Present. I am struck .... [wanting] [wanting]. 
 
 (τέτυμμαι and verberatus sum 
 used instead). 
 Past. I was struck .... [wanting] [wanting]. 
 Future. I shall be struck . . . τνφθήσομαι verberabor. 
 
 To complete therefore our comparison of the indicatives of 
 the Greek, Latin, and English verb, we see that of the nine 
 possible tenses, in the passive, 
 
 Greek has six tenses, 
 
 Latin has three tenses only, and 
 
 English has no tenses. 
 
 The only passive form in English is that of the participle 
 ( { struck '=having been struck). 
 
 * One or two Greek verbs have an active future-perfect, as βστήξ», 
 τίθνίιξω. Deponents Have to make their future-perfect by the auxiliary, 
 as elpya^4vos ίσομαι. The comparative want of future-forms may be 
 due to the fact that men care to speak with less precision of the unknown 
 future than of the past. 
 
 α 3 
 
130 A BRIEF GREEK SYNTAX. 
 
 Chief Idiomatic Uses of the Tenses. 
 
 134. When a language has a peculiar form or motie of 
 expression this is called the idiom of the language (Ιδίωμα 
 from ϊΰιος l private,' ' peculiar ') ; and these idioms are what 
 specially need to be learned and remembered ; for the ordinary 
 meanings and uses present no difficulty. 
 
 The Present and Imperfect. 
 
 135. The present, used dramatically in narratives in order 
 to represent the events narrated as going on before the eyes, is 
 called the historical present;* and the imperfect is used in 
 the same way for the same reason ; as 
 
 και επιτηΖίς σε ουκ ηγειρον 'ίνα ώς ηΐιστα ΰιάγης I was not 
 awaking you on purpose, that you may be going on as 
 pleasantly as possible. 
 
 έρχεται προς τους μαθητας κα\ ευρίσκει αυτούς καθεύΰοντας 
 he cometh to the disciples and findeth them sleeping. 
 
 The historic present, in the sequence of tenses, is treated as 
 an historical tense, and is therefore followed by the optative. 
 
 136. Both the present and the imperfect are used to express 
 an attempt (conatus rei efficiendas) : 
 
 δίά ποΊον αυτών έργον λιθάζετέ με ; for which work of these 
 
 are you for stoning me ? — John x. 32. 
 Κύριε, συ μου νίπτεις τους πόνας ; Lord, dost Thou mean 
 
 to wash my feet ? — John xiii. 6. 
 
 * The historical present, seldom used except colloquially in English, 
 is very common in Grerman ; and tolerably so in French, as in the lines 
 of Eacine : 
 
 4 «Fai vu, seigneur, j'ai vu votre malheureux fils, 
 Traine par les cheraux que sa main a nourris. 
 Π veut les rappeler, et sa voix les effraie. 
 lis courent. Tout son corps n'est bientot qu'une plaie.' 
 
 Of English writers Carlyle uses it most frequently ; e.g. ' Ear down in 
 their vaults the seven prisoners hear muffled din as of earthquakes ; their 
 turnkeys answer vaguely," &c. In one passage of Milton, the historical 
 present is powerfully used for the future : 
 
 ' If from this hour 
 Within those hallowed limits thou appear, 
 Back to the infernal pit I drag thee chained 
 And seal thee so,' &c. — Par. Lost, iv. 965. 
 
 Comp. JEn. in. 367. So far as I am aware no such usage is found in 
 classical Greek. 
 
IMPERFECT TENSES. 131 
 
 εκαλονν αυτό . . . Ζαχαρίαν they wished to call him 
 
 Zacharias. — Luke i. 59. 
 δ ΰε Ιωάννης Ιιεκωλνεν αυτόν John tried to prevent liim. 
 
 — Matt. iii. 14. 
 εζανεχωρει τα εφημένα he tried to back out of his words. 
 
 — Thuc. iv. 28. 
 
 In all these instances ' Vere incipit actus, sed ob impedimenta caret 
 eventu.' — Schaefer, Eur. Phoen. 79. 
 
 The constant substitution in the New Testament of a parti- 
 ciple and auxiliary (e.g. ήν καωμένη, Luke xxiv. 32) shows 
 that when the continuance required to be emphasised, the 
 simple imperfect was no longer sufficient. 
 
 137. Hence the impf. alone is often, rhetorically, used 
 where the impf. with αν would have been more regular, as 
 
 τις μοι φύλαζ ήν εί συ συμφοράς τύχοις ; (Eur. JBacch. 612) 
 who were my guardian (= would have been) should 
 you have met with a misfortune ? 
 
 This suppression of av is very common in conditional sen- 
 tences, as 
 
 ουκ εϊχες εζουσίαν . . . εί μη you would not have had 
 
 power, unless, &c. — John xix. 11. 
 καλόν ήν αυτω εί ουκ εγεννήθη it were well for him if he 
 had never been born. — Matt. xxvi. 24. 
 
 A similar potential use of the impf. is not unknown in 
 Latin ; as 
 
 Eespublica poterat esse perpetua, si patriis viveretur in- 
 stitutis.— Cic. de R. P. iii. 29. 
 
 138. The present is used with πάλαι ' long ago,' &c. ; as 
 απ αρχής μετ εμοϋ έστε ye are (=have been) with me 
 
 from the beginning. — John xv. 27. 
 
 γή νοσεί πάλαι the land has long been sick. — Eur. 
 So in Latin : 
 
 Jampridem cupio Alexandrian! visere. — Cic. 
 And in German : 
 
 1 FiXnf Jahre trag' ich schon den gluh'nden Hass.' — 
 Schiller, Turandot. 
 And in French : 
 
 { II y a longtemps que je suis ici.* 
 { Je le regarde depuis longtemps.' 
 
132 A BRIEF GREEK SYNTAX. 
 
 And very rarely in English. Mr. Boyes quotes from Hey- 
 wood : 
 
 * 'Tis dinner-time at least an hour ago.' 1 
 And in Walpole's letters : 
 
 ' Lord Dalkeith is dead of small-pox in three days.' 
 Compare 
 
 1 'Tis now a nineteen years agone at least/ — Ben Jons., 
 
 Case is Altered. 
 { He is ready to cry all this day.' — Ibid., Silent Woman. 
 
 139. κλύω, ακούω, μανθάνω, γιγνώσκω (verbs of perception), 
 and those which indicate an abiding result (as ηκώ, φεύγω), 
 are used in the present where we use the perfect ; as 
 
 άρτι γιγνωσκεις τό^ε ; have you only just learnt this ? 
 απαγγέλλετε ΰτι ημείς νικώμεν βασιλέα answer that we 
 have conquered the king. 
 
 140. The imperfect expresses incompleteness, continuance, 
 and (especially with ay) repetition. Karely it is used as 
 giving a more emphatic meaning, where we should use the 
 present; as 
 
 ος κε θεοΊς έπιπείθηται μάλα τ' εκλνον αυτόν whosoever 
 
 obeys the gods, him they ever hear (cf. 77. i. 418). 
 'Tenrpus eraV (Hor. Od. i. 37) 'Tis full time. 
 
 141. εδ«, έχρην, εικός ήν, ώψελον imply dissatisfaction, and 
 a wish that something else had happened ; as 
 
 εικός i]V υμάς μη ^αλακώς, ώσπερ rvv, συμμαγεΊν you 
 
 ought not in all fairness to prove yourselves such 
 
 feeble allies as you do. 
 
 Here ' it was right ' means ' it would have been right,' and is 
 
 equivalent to εικός civ ήν, precisely as in these two English 
 
 sentences : 
 
 1 Was man like his maker ... I should be for allowing,' 
 
 &c. (Addison) [ = if man had been]. 
 1 It were well for the insurgents ... if the blood that was 
 now shed had been thought a sufficient expiation for 
 the offence ' (Goldsmith) [= it would have been well 
 αγαθόν αν ήν~].* 
 So in Latin : 
 
 * Si mihi omnes, ut erat gequum, faverent.' — Cic. de Div. 
 
 iii. 10. 
 
 * Compare 'Gold were as good as twenty orators' (= would be). 
 Observe however that ' were ' is the English $ubjunctive t 
 
THE FUTURE. 133 
 
 142. Notice the graceful and modest use of the imperfect in the 
 inscriptions used by old artists, Πολύκλειτο* iiroUt ; this implied how 
 far they felt themselves to fall short of ideal perfection, ' tamquam inchoatd 
 semper arte et imperfecta ' (Plin. Nat. Hist. i. 20), and it showed them 
 to be imbued with the highest spirit of art. 
 
 143. Sometimes the imperfect expresses what was but is not, as Eur. 
 Troad. 585, πριν ποτ' ήμεν we once were (but are no longer) ! Compare 
 Fuimus Troes, fuit Ilium, Virg. Mn. iii. 325. After the execution of 
 the Catilinarian conspirators, Cicero said of them, Vixerunt. 'Pro- 
 bablement a midifaurai vecu, pour parler le langage romain.' — Letter 
 of Charlotte Corday. There is a fine instance in Dante, Inf. x. 67, 
 
 Di subito drizzato grido : Come 
 Dicesti egli ebbe ? non \W egli ancora ? 
 
 The Future. 
 
 144. The future active answers to our shall and will, even 
 in its imperative use ; as 
 
 εΐ,εις άτρέμας ; will you keep quiet ? * 
 
 εσεσθε ovv νμεΊς τέλειοι be ye therefore perfect ! 
 
 145. The periphrases of μέλλω, θέλω, βονλομαι with the 
 infinitive are by no means ' periphrastic futures,' as they are 
 sometimes called, but differ from the simple future in mean- 
 ing, by emphasising the purpose or wish to do a thing. They 
 show however the dawnings of an aim at analytic precision 
 (see Herod, i. 109). 
 
 N.B. ποιηηω I will do, faciam ; μέλλω ποιησειν I am on 
 the point of doing (cf. the Italian sono per lasciarti I am on the 
 point of leaving you) ; μέλλω ποιεϊν I intend to do. 
 
 146. Few verbs have all the four -μαι forms of the future 
 in use (τυφθήσομαι, τυπησομαι I shall be struck, τύψομαι I 
 shall strike myself, τετυφομαι I shall have been struck). 
 
 147. The future-perfect")• (ό μετ ολίγον μέλλων, paullo- 
 post-futurum) , as its name implies, mingles the future and 
 the perfect both in form and meaning (as in English ' I shall 
 have been struck'). It also expresses rapidity, as 
 
 φράζε και πεπράζεται speak and it shall be done at once ; J 
 
 * Both in English and Latin the future is a polite substitute for the 
 imperative ; e.g. Valebis et salvebis = yale et salve! 
 
 'Tuinterea non cessabis.' — Cic. Epp. ad Fam. v. 12. 
 
 ' Inter cuncta leges et percunctabere doctos.' — Hor. Epp. i. xviii. 26. 
 
 f Being a mere luxury of language, it occurs but once in the New 
 Testament (Luke xix. 40), κζκράξονται, and there, only because the 
 simple future of κράζομαι is not used. The name Futurum exactum was 
 invented by Pomponius Lsetus (1497). 
 
 % Cf. Cicero, Fp. — ' Tu invita mulieres, ego accivero pueros.' 
 
134 A BRIEF GREEK SYNTAX. 
 
 and a continued result ; as 
 
 ουΐεϊς κατά σπουΰάς μετεγγραφήσεται, , 
 
 αλλ' ώσπερ ήν το πρώτον εγγεγράφεται 
 
 * No one shall be transferred to another list by favour, 
 
 but shall remain inscribed as he was at first.' 
 κληθησεται he shall be called; κεκλησεται he shall bear 
 
 the name. 
 
 148. Since μεμνημαι, κέκτημαι, &c. have the sense of pre- 
 sents, μεμνησομαι I shall remember, κεκτήσομαι 1 shall possess, 
 &c, are simple futures. 
 
 The Perfect. 
 
 149. The perfect corresponds to the English perfect with 
 'have'; it is a present--perfect, e.g. £ I have struck' means 
 ' I have now struck,' or ί I struck and the effect continues.'' * 
 Hence it is substituted for the aorist (which is the ordinary- 
 tense in which events are narrated) to describe past events of 
 which the result remains ; as 
 
 πενεστέρονς πεποίηκε και παΧλούς κινΰύνονς νπομενειν 
 ήνά-γκασε it has made us poorer (and we still are so), 
 and it compelled us to undergo many dangers. 
 
 150. This explains such meanings as κέκτημαι I possess, 
 τεθανμακα I ivonder, κέκλημαι I am called, ερρωμαι I am 
 strong,^ &c. ; and it is curiously paralleled by the German 
 idiom (see Clyde, Greek Syntax, p. 69). In the same way 
 such a phrase as " I have often wondered ' generally implies 
 that the effect still continues. For another view of these 
 perfects with a present sense, see p. 49, note *. 
 
 The Aorist. 
 
 151. The nature of this tense ought to be clear, from all 
 that has been said about it in the previous section. Its vitality 
 is accounted for by its importance. It is the regular tense of 
 narration, as it is in English, because it has no relation to the 
 present. Take any sentence from a history, such as ' "William 
 Kufus died from the wound inflicted by an arrow ' ; here 
 
 * This use of the perfect in Homer is very common ; e.g. in describing 
 a chariot he says, αμ<ρ\ 8e πέπλοι ΤΙέπτανται tapestries hang around it. 
 E. v. 195. (Exigua tantum ratione habita praeteriti temporis, quo stragula 
 ilia expansa ftierunt, sed prsesentis prsecipue, quo expansa sunt. — . 
 Schmidt, JDoctr. Temp. ii. 10.) 
 
 f Compare the Italian ho capito I understand. — Clyde. 
 
THE AORIST. 135 
 
 1 died ' is an aorist, κατέθανεν, and we could no more substitute 
 an imperfect (' was dying '), a perfect (' has died '), or a plu- 
 perfect (' had died ') in English than Ave could in Greek. 
 
 N.B. — The aorist with &v sometimes expresses iteration, as δπότε 
 τροσβλεψειε τίνα τότε μεν εΊνεν &ν, κ.τ.λ. ' whenever he saw any one, 
 then he would say,' etc.; and sometimes is equivalent to the Latin 
 pluperfect subj., as ουδέν hv επραξεν nihil fecisset. Thus ελεξεν tci> may 
 mean ' he used often to say,' or ' he would have said,' according to 
 context. 
 
 152. There is an obvious connection in form between the 
 aorists and the future,* as Ave see at a glance : 
 
 τύψω τυφθησομαι τυπήσομαι τύφομαι 
 
 έτυψα ετύφθην ετύπην ετυψάμην. 
 
 And there are one or two cases in which either future or 
 aorist is admissible ; f e.g. 
 
 άνηρ σοφός τας συμφοράς pyov ο'ίσει των άλλων a wise 
 
 man ιοϊΙΙ hear his misfortunes more easily than the rest 
 
 of mankind. 
 
 It would be just as good Greek to say ηνε-γκε hove, and just as 
 good English to use the present-aorist ' hears ' ; and we find 
 the aorist subj. in the same clause with the future ind. ; | as 
 ε\πωμεν ϊ) σιγώμεν ; ?/ τί λέζομεν ; are we to speak, or be 
 silent, or what shall we say ? 
 
 153. Obviously what has taken place (especially if it be 
 frequently) in the past, will probably recur in the future, § so 
 that either aorist or future may be used, for instance, in com- 
 parisons, and so far there is a connection between the tenses. 
 Further than this no theory has ever established what was 
 the historical connection between these tenses, except that 
 
 * Besides this, the first aor. subjunctive is τύψω, which is the same 
 form as the future active. In Latin there is no difference in form 
 between the future perfect and the perfect subjunctive (except in the 
 first person), and very little in meaning. See Koby's Lat. Gram. p. xv. 
 
 t Similarly in John xv. 6, εαν μη tls μείνη εν εμοί, εβλήθη £ξω, the 
 future βφΚησεται would have given the same sense. 
 
 \ In such a line as ov yap πω τοίους ΐδον avipas ουΖε Ιδωμαι never saw 
 I nor shall I see such men, the aor. subjunctive ϊδωμαι is practically a 
 future. 
 
 . § Burnouf's view that the future expresses 'posteriority relative to the 
 present moment, and the aorist, posteriority with reference to some other 
 (unspecified) time, does not seem to me free from objection ; e.g. his 
 explanation of the aorist in the line 'Je chantele heros qui regna sur 
 la France,' seems to me impossible on his own principles. 
 
!3ί» A BRIEF GREEK SYNTAX. 
 
 the σ of both aorist and future is derived from the auxiliary 
 verb ' as ' to be (εσμίν, εστί). 
 
 154. The aorist is used in proverbs, &c. (gnomic aorist), to 
 express what once happened, and has thereby established a 
 precedent for all time ; as 
 
 πολλά παρά γνώμην έπεσε many things fall out contrary 
 to expectation.* 
 In Eev. iv. 10 the future is used in this gnomic sense, as in Gaelic. 
 
 The Pluperfect. 
 
 155. This tense is comparatively neglected in Greek, f the 
 aorist being substituted for it in many instances where it 
 would be used in Latin, and even in English ; e.g. 
 
 ώς ηχούσαν τονς Χόγους . . . Ζιηπόρονν when they (had) 
 heard the words, they began to doubt. 
 
 Its chief idiomatic use is to express rapidity ; as 
 
 ovV άπίθησε 
 μνθω ' Άθηναίης' η δ' Ονλνμπόνΰε βεβήκει 
 nor did he disobey the order of Athene ; but she had 
 
 already vanished heavenwards. — II. i. 221. 
 Ore οι σύμμαχοι επλησίαζον, οι Αθηναίοι τους Πέρσας 
 ενενικηκεσαν when the allies were approaching, the 
 Athenians had already conquered the Persians. 
 
 MOODS (Εγκλίσεις). 
 
 156. In coming to treat of the moods, we have reached by 
 far the most difficult part of Greek syntax. The clumsy 
 analytic periphrases of our own and most modern languages 
 are quite inadequate to represeut the delicate accuracy and 
 beauty of those slight nuances of thought which the Greek 
 reflected in the synthetic and manifold forms of his verb. 
 One of the chief reasons for the study of Greek is the fact 
 that it presents us with the most perfect instrument for the 
 expression of thought. Our own language is singularly noble, 
 powerful, and splendid, but its points of excellence differ 
 entirely from those of Greek. 
 
 * The Latin aorist has a similar use, as ' Hinc apicem rapax Fortuna 
 cum stridore acuto Sustulit,' Hor. Od. i. 34, = solet tollere. Non tarn 
 praecipites bijugo certamine campum Corripuere. — Mn. v. 145. 
 
 f The form of the pluperfect in η (iyeypa<fn}, &c.) is older than that in 
 ttv, and more Attic, ea = e -σαμ =^ eram. 
 
THE MOODS. 137 
 
 But the study of Greek would not be valuable as a mental 
 discipline if it presented no difficulty. There is no royal road 
 to anything worth acquiring; χαλεπά τα καλά. Yet after a 
 thoughtful and careful study of the following pages, the 
 student ought at least to have some clear notions which will 
 serve as a guide to further study. 
 
 157. The moods express the aspects or modes under which 
 the action is regarded, and are three in number, viz. : 
 
 The indicative, which deals with facts, certainties, direct 
 questions, &c, i.e. it is the objective mood; and therefore the 
 tense-distinctions exist mainly in this mood. 
 
 The imperative, which deals with commands. 
 
 The subjunctive and optative, which deals with suppositions, 
 uncertainties, contingencies,* &c. The subjunctive connects 
 such modes of conception with the present or future; the 
 optative connects them with the past. The two together form 
 but one subjective mood. 
 
 158. The infinitive is no mood at all, since it represents the 
 verb absolutely, in no particular aspect, and with no relation 
 to any subject (εγκλισις απαρέμφατος). 
 
 159. It will be convenient to treat of the moods first as they 
 occur in simple sentences, and afterwards in compound. 
 
 But we may observe at once that the names of the moods 
 are as unsatisfactory as those of the tenses, f The indicative 
 mood, or mood of declaration, does not declare at all in inter- 
 rogative or conditional sentences. The optative, or wishing 
 mood, does indeed sometimes express a wish, but this is a very 
 small part of its meanings, and it is quite as much subjoined 
 as the so-called subjunctive, of which, as we shall see, it forms 
 a part. 
 
 The Indicative. 
 
 160. The indicative mood (εγκλισις οριστική) denotes an 
 actual, or (in the future tense) a certain state. In treating of 
 the separate tenses we have given all its most distinctive 
 
 * • Indicatives res per se, seu nude positas, conjunctivus autem res ex 
 mente agantis spectatas [yelut luminis radios vitro fractos) vel in cogita- 
 tionem inclusas notat.' — De Formis dictorum conditionalium, F. Ellendt. 
 Konigsb. 1827. The illustration is an exceedingly good one, but the 
 treatise itself is not very clear. 
 
 f See F. Whalley Harper On the Powers of the Greek Tenses, p. 137. 
 
138 Λ BRIEF GREEK SYNTAX. 
 
 The Imperative. 
 
 161. The imperative mood (προστακτική) commands,^ and, 
 with negatives, prohibits. As all commands must refer to the 
 future, we see that the temporal meanings of the indicative 
 tenses vanish in the imperative ; the distinctions between the 
 tenses in the imperative not being those of time. 
 
 162. μη πράττε don't be doing it (of continuous or recurring 
 actions, and of actions already begun = leave off doing 
 it!). 
 
 μη πράξης don't do it (of momentary or single actions). 
 \αβε τας μαρτυρίας καϊ arayiyi ωσκε take the depositions 
 
 (aor. imp.=an instantaneous act), and read them (pres. 
 
 imp.=a continued act). 
 
 163. The perfect imperative denotes the permanence of the 
 result; as 
 
 τέθναθί lie dead !=κε~ισο τεθνηκώς. 
 
 είς τον Τίυριφλεγέθοντα εμβίβλήσθω let him be flung (at 
 once, and for all) into Phlegethon ! 
 
 164. Other ways of expressing command are 
 a. By the infinitive ; as 
 
 μη δη μοι άπόπροθεν "ισγεμεν 'ίππους do not I pray you 
 rein the horses at a distance from me. 
 
 β. By the optative with &v ; as 
 
 χωροΊς αν είσω like our l perhaps you would go in? 
 γ. By the subjunctive ; as 
 
 ίωμεν let us go. 
 d. By various periphrases ; as 
 
 ο~ισ& olv ο Ιράσον] do then — know'st thou what?f i.e. 
 dost thou know what thou must do ? 
 
 οϊσθ' ώς ποίησον ; do — know'st thou how ? 
 
 όπως άνδρες εσεσθε see that ye he men (sub. οράτε). 
 
 * In Sanskrit the imperative has a first as well as a second and third 
 person. This is also the case in English, though only in poetry and in 
 the plural, as ' Leave we the theme.' ' Charge we the foe.' — New Crat. 
 p. 593. 
 
 f Mr. Boyes quotes a close parallel from Chaucer : 
 
 ♦ And deemith you, what ye shall do therefor ? 
 G-o thanketh now my lady there, quoth he.' 
 "We find the same idiom in Latin ; ' Tange, sed scin quomodo ? — Plaut. 
 Budent. ϊΐι.ν. 16. 
 
THE SUBJUNCTIVE AND OPTATIVE. 139 
 
 φέρε ΰή άναγνώ τας μαρτυρίας come now let me read you 
 the evidence. 
 
 μη Ιήτ αδικηθώ let me not be injured. — Soph. 0. C. 174; 
 cf. Tr. 802. ' Prima conjunctivi persona sic usurpatur 
 tit admonitio ad secundam spectet.' — Herm. 
 
 The Subjunctive (υποθετική) and Optative (ευκτική). 
 
 165. £ The subjunctive is a byforni of the future, the optative 
 a by form of the aorist.' * 
 
 We have already seen the points of connection between the 
 future and the subjunctive^ and in fact the notion of futurity 
 is essentially involved in the subjunctive, since that which is 
 contingent and dependent must necessarily be analogous to 
 what is future. Hence the student must not be misled by 
 such names as perfect subjunctive, &c. to suppose that the 
 forms of the subj. and opt. express time in the same way as 
 their cognate indicative tenses. 
 
 166. The subjunctive and optative are not two moods, but 
 one subjective mood,\ which expresses not facts and. realities, 
 but suppositions and contingencies ; the subjunctive forms are 
 the present or future tenses of this mood, and the optative 
 forms its past tenses. In other words, the optative is merely 
 the subjunctive of the past or historic tenses. It carries with 
 it a reference to the past. 
 
 Everything that we say about these moods will illustrate 
 and explain this fundamental fact, which the student is urged 
 to master and to keep steadily in mind throughout the follow- 
 ing observations. 
 
 167. The Greek subjective mood furnishes seven separate 
 forms, usually called tenses ; e. g. 
 
 pres. subj. Βειπνώ, aor. subj. ΰειπνήσω,ρβ?^ sub. ΰεΰειπνήκω, 
 pres. opt. Ιειπνοίην, aor. opt. Ιειπνησαιμι, perf. opt. δε- 
 ΰειπνήκοιμι, fut. opt. ΰειπνήσοιμι.§ 
 
 * ' The subjunctive and optative are by -forms of the future and aorist.' 
 — Don. p. 546. The connection is indicated by a similarity of form. 
 
 f We see it also in Latin, where dicam is both future indicative and 
 present subjunctive, the termination -m being a relic of the old -μι form 
 of verbs. In Gothic Ulphilas often renders Greek futures by the 
 subjunctive. 
 
 f In treating this part of the subject, I have on the whole received 
 more assistance from Mr. F. Whalley Harper and Dr. Clyde's Greek 
 Syntax, than from any other of the numerous treatises which I have 
 consulted. 
 
 § Some verbs have also second aorist optatives and subjunctives, but 
 
140 A BRIEF GREEK SYNTAX. 
 
 168. And Latin offers four, as ccenem, camarem, ccenaverim, 
 ccenavissem. 
 
 We shall find that on examination these forms evaporate 
 considerably ; but before discussing them let us try to under- 
 stand them in the form of a table. 
 
 We have already tabulated the actual and possible Indicative tenses ; 
 the table of the Subjective tenses should be compared with it, although 
 it will be seen immediately that these tense-forms are in reality evan- 
 escent, and in part illusory. 
 
 SUBJECTIVE TENSES. 
 (i.e. Tenses of the Subjective Mood, expressive of suppositions, &c.) 
 
 DEF 
 
 NITE. 
 
 
 INDEFINITE, or 
 1 
 
 AORISTIO. 
 
 
 1 1 
 
 Present. Past. 
 &ίίπνήσω. δΐίπνήσαιμι. 
 
 Future, 
 δΐίπνησοιμι.. 
 
 Perfect, or 
 Finished. 
 1 
 
 
 Imperfect, or 
 Unfinished. 
 1 
 
 
 Present, 
 δί&απντηκω. 
 
 SeS 
 
 1 
 Past, 
 •ίπνήκοιμι. 
 
 1 1 
 
 Future. Present. Past. Future. 
 [wanting] δειπνώ. δειπνοίην. [wanting] 
 
 On this table we have to remark — 1. That very little etress must bo 
 laid on the exactness of any direct English or Latin equivalents ; the 
 idiomatic uses of Greek being very strongly marked in the use of the 
 moods. Even the French equivalents, as λύσω que j'aie de!i£; Χύσοιμι 
 que j'eusse delie, are quite inadequate. 2. Observe however that the 
 English may is the best general representative of the Greek subjunctive, 
 might of the optative. 3. Two of the future forms are wanting; and the 
 other future form, although it occurs, is merely a chose de luxe, because 
 the whole mood involves futurity, so that the present forms serve instead. 
 4. The past tenses of the Latin subjunctive are equivalent to the Greek 
 optative. 
 
 169. Further: of the seven Greek forms, three are very 
 rarely used, viz. the perfect subj., the perf. optative, and the 
 future optative. We may in fact dismiss those three forms, 
 with the remark that the perfect forms are only used where 
 something is specially to be marked out as completed ; and 
 the future opt. only in oratio obliqua (or reported speech), 
 and that very rarely, to represent the future indicative. Thus, 
 in direct speech : 
 
 Σνέννεσις λέλοιπε ret άκρα Syennesis has left the heights ; 
 
 these being merely other forms of the same tense, are not noticed ; e.g. 
 in English no one regards hung and hanged as two separate tenses. 
 
TENSES OF THE SUBJECTIVE MOOD. 141 
 
 in reported speech : 
 
 ελε£εν οτι Συέννεσις λελοιπως ε'ίη τα άκρα he said that 
 Syennesis had left the heights. 
 
 Direct speech : 
 
 η 6ΰ6ς εσται προς βασιλέα our march will be to the great 
 king. 
 
 Keported speech : 
 
 έλεγες οτι f] οΰος εσοιτο προς βασιλέα he kept saying that 
 their march would be to the great king. 
 
 We may then draw this conclusion : the tenses of the optative 
 only retain a tense-meaning in oratio obliqua. 
 
 170. But it may be asked how come we to have an aorist 
 subjunctive δειπνήσω, if the subjunctive be merely the form 
 assumed by the primary tenses in the subjective mood ? for the 
 aorist is an historical and not a primary tense, and therefore 
 its form in the subjective mood ought to be only ΰειπνησαιμι. 
 
 The answer to this very natural objection appears to be 
 that the past aorist is necessarily sometimes used in Greek for 
 the present aorist (' I dined ' for ' I dine '), as we have seen 
 already (§ 126) ; and it is perhaps this use of the past aorist 
 so frequently as a present that accounts for the existence of 
 such a form as Ιειπνήσω. And in full accordance with this 
 hypothesis we find that the present and aorist forms of the 
 subjective mood are in many sentences used interchangeably 
 and almost indifferently. 
 
 171. We have then considerably reduced the importance 
 of the number of tenses in the subjective mood, by showing 
 that in practical use three of them at least are nearly eliminated. 
 Further than this, as we have just observed, the differences 
 between Ιειπνώ Ζειπνησω, and between Ζειπνοίην Ζειπνησαιμι, 
 are very slightly marked, and are not distinctions of time ; the 
 present forms merely imply that the result continues, the aorist 
 forms draw no attention to more than the momentary fact. 
 Thus we may say almost indifferently 
 
 σπουδάζω 'ίνα μανθάνω or μάθω. 
 έσπονΰαζον Ίνα μανθάνοιμι or μάθοιμι. 
 
 172. And since these are the only forms in constant use, 
 it will be seen that the subjective mood for all ordinary 
 practical purposes contains (as in Latin) but four tenses, viz. 
 a present and an aorist form which follow the primary tenses ; 
 
142 A BIttEF GREEK SYNTAX. 
 
 and a present and an aorist form which follow the historical 
 tenses.* « 
 
 173. Then, further, notice that this so-called optative mood 
 (which we have, as far as any frequent use is concerned, re- 
 duced to a present and an aorist form, differing but little from 
 each other in meaning, and used as the dependent and sub- 
 jective form of the historical tenses) was itself a refinement of 
 language but little needed; and therefore that it gradually- 
 fell into desuetude, and in Modern Greek nearly disappears, 
 the few forms in which it appears (such as μι) γένοιτο) being, 
 as Dr. Clyde says, ' merely the coffin of the dead optative.' 
 
 174. Even by Attic writers the distinction between subjunc- 
 tive and optative was (if Ave may believe the MSS. rather than 
 the editors) very negligently observed ; in the New Testament 
 and in later Greek writers the optative in final sentences (seeinf. 
 § 179) almost disappears ; f and it is very probable that in the 
 speech of the vulgar the optative hardly existed at all, being too 
 delicate in its distinctions for daily use. Possibly the very 
 existence of such a mood may have been practically disre- 
 garded by an Athenian cobbler. Observe too that whereas 
 (owing to the dramatic principle which led the Greeks to omit 
 the reference to the past, and to represent past things as still 
 going on before the eyes) the subjunctive is often used where 
 the optative would be more regular, the reverse of this is never 
 the case, i.e. we never find the optative for the subjunctive. 
 
 175. "We shall continue to use the names subjunctive and 
 optative, but it must not be forgotten that by optative we do 
 not mean a different mood from the subjunctive, but only a 
 name for those subjective forms which correspond to the 
 historical tenses of the indicative. 
 
 The Subjunctive in Simple Sentences. 
 
 176. 1. Used absolutely, the subjunctive in Homer differs 
 
 * It has already been pointed out that the third person dual of the 
 subjunctive (like that of primary tenses) ends in ov ; and of the optative 
 (like that of the historical tenses) in t\v. 
 
 f The past tenses of the French subjunctive (which correspond to the 
 Greek optative) are disappearing in the same -way. In English, the 
 whole subjunctive mood is very rapidly disappearing, and its evanescence 
 is much to be regretted ; by all our best writers it was, and still is, used 
 regularly after all causal and hypothetical conjunctions ; but in common 
 conversation it is now rarely heard. See some admirable remarks on 
 this subject in Craik's Engl, of Sha/csp. p. 104. 
 
THE OPTATIVE MOOD. 143 
 
 but little from a future,* as is also the case with the subjunctive 
 aorist after ov μή in strong negations ; as 
 
 ov μη ποιήσω I certainly won't do it ; ov μη φνγΐ}ς you 
 certainly will not escape. 
 
 2. It is used (in the aor. 2nd per. sing, and plur.) in 
 prohibitions; as 
 
 μη κλέψης don't steal (this or that). 
 
 3. Deliberatively (1st pers. sing, and plur.) ; as 
 
 irg. βώ ; whither am I to go ? πον στώ ; where am I to 
 
 stand ? 
 π φώ ; what am I to say ? f 
 
 4. Hortatively (1st pers. sing, and plur.) ; as 
 
 "ιωμεν let us go; έγκονώμεν let us exert ourselves; espe- 
 cially with φέρε, αγε, 'ίθι, ειπέ, &c. 
 
 5. It is often used elliptically after βοΰλει, θέλεις, κ.τ.λ.; as 
 Θέλετε θηρασωμεθα ; do you wish that we should hunt ? 
 
 —Eur. Bacch. 719. 
 θέλεις μείνωμεν αντον ; do you wish that we should re- 
 main on the spot ? — Soph. El. 80. Compare Ov. Met. 
 ix. 734, Vellem nulla for em. 
 
 6. In Plato and Demosthenes the subjunctive is often used 
 with αν=•εάν> ην. Thus: 
 
 αν σωφρονή. — Phced. 61 Β ; αν θεός εθέλη.— Ιά. 80 D. 
 
 [This is curiously analogous to the obsolete English ' an ' with 
 the subjunctive, ' an God be willing,' &c] 
 
 The Optative in Simple Sentences. 
 
 177. ( L'optatif n'est point reellement un mode a part ; c'est 
 une simple denomination sous laquelle on a range les temps 
 secondaires du subjonctif.' — Burnouf. 
 
 The distinctive sign of the optative is derived from ya to go. See 
 Max Miiller, Stratific. of Lang. p. 30. 
 
 1. The optative gains the credit of being a separate mood, 
 as well as its name (ϊγκλισις ευκτική), simply because when 
 u ed absolutely it often expresses a wish ; as 
 
 * e.g. in II. vi. 459, καί ποτ4 tis ςϊττρσι corresponds to ws ποτέ tis ipeei 
 a little further on. Cf. E. i. 262 ; Od. xvi. 437, vi. 201. 
 
 t Cf. ουκ 1ω ; shall I not go 1 which resembles the Latin quin with 
 the present indicative. Quin redimus ? — Plaut. Mencechm. n. i. 22. 
 
144 A BRIEF GREEK SYNTAX. 
 
 ω παι, γενοιο πατρός ευτυχέστερος, 
 
 τα δ' αλλ' όμοιος' και γένοι αν ου κακός. — Soph. Aj.* 550. 
 
 1 Boy, mayest thou (lit. mightest thou be) more fortunate 
 
 than thy father, but like him in all else, and then thou 
 
 wouldest be noble.' 
 ουτ" 1 αν Ιυνα'ιμην μητ επισταίμην λέγειν (Soph. Ant. 
 
 682) I could not, and may I never know how to say. 
 
 We express wishes by ■ mayst thou,' &c, using the subjunctive, which, 
 by referring to the present time, hints at the possibility of the thing 
 becoming realised ; the Greek, more accurately, uses a mood which refers 
 altogether to the past* and therefore can be regarded as a wish, and a 
 wish only. We however use ' might ' after ' would that ; ' and probably 
 the wishing-power of the optative is merely due to an ellipse f of one of 
 those frequent formulas which are used with it, as e?, ei yap, e?0e, ύφελον, 
 πώ$ &v, eft? &<pe\ov [which, in the case of impossible wishes, are used 
 with past tenses of the indicative, as eWe συ\ tStc συν^γ^νόμην would I 
 had then been with you !] In ZeC irarep, us Χαλύβων παν απόλοιτο yevos, 
 Callim. (Jupiter, ut Chalybum omne genus pereat, Cat. lxiii. 54), every 
 one would at once recognise an ellipse ; is there any less reason for the 
 ellipse, if &>s be omitted ? 
 
 N.B. — Μη is used (not oi>) in negative wishes, as Μη yivoiro would 
 that it might not be! God forbid! [μη ykvono utinam ne fiat! μη 
 yeviaOw jubeo ne fiat! μη γίνηται cavendum ne fiat!] 
 
 υμίν Se τοιούτο μ\ν ovShv οί»τ' ?jv μητ€ yevoiro τον λοιπού but in your 
 case nothing of the kind ever happened, and may it never happen 
 hereafter. 
 
 2. If it be correct to suppose that this votive force of the 
 opt. is merely due to an ellipse, the name £ optative'' becomes 
 more unfortunate than ever. No separate name for it is needed, 
 because, as we have seen, it consists merely of the past tenses 
 of the subjunctive ; but, if it must be named, potential would 
 perhaps be better, since it not only regularly expresses poten- 
 tiality (could, might, &c.) with av (which makes the possibi- 
 lity depend on conditions), but even without it, especially in 
 poetry. If this view be correct, the prevalence of av with 
 the optative was due to the analytic tendency of all advancing 
 language. This potential use of the optative without av would 
 not be so rare as it is, if the MSS. had not been repeatedly 
 altered by scholars who wished to square them with their own 
 views. The following are instances : 
 
 νεογνος ανθρώπων μάθοι a mere child might understand 
 it.— JEsch. Ag. 1163. 
 
 * Latin uses both subjunctive and optative, the former for possible 
 wishes, as Utinam dives fiam ; the latter for impossible, as Utinam Deus 
 essem. * The subjunctive gives a notion of the realisation of the proposed 
 end ; the optative represents it as a mere possibility.' — Jelf, § 809. 
 
 f Just as in the Italian volesse Iddio =plut a Dieu. — Clyde, 
 
THE OPTATITE MOOD. 145 
 
 Iv είκοσι πασι μάθοις νιν you might know him among a 
 score. — Mosch. 
 
 πείβοι av tl πείθοι, άπειθοίης δ' 'ίσως (iEsch. Ag. 1048) 
 comply (a mild imperative) if thou wouldst comply, 
 but perhaps 'thou wouldst not comply (sc. under any 
 circumstances). See Paley's notes to iEsch. Ag. 535, 
 1133, 1847 ; and Jelf, 426, 1. 
 
 το δ' ϊπος ουξερώ τά-χα 
 
 ήΰοιο με»', πώς δ' ουκ aV, άσχάλλοις ο' ίσος. — Soph. Ο. Τ. 
 936. 
 
 1 you might possibly rejoice at what I am about to say- 
 how should you not ? — but you might be grieved.' 
 
 Some however would understand the av (from the previous 
 clause) in the clause where it is not expressed ; as in Xen. 
 Hier. ii. 11 : 
 
 ου μόνον φιΧοϊ 1 (iv, άλλα καϊ ερωο. 
 
 3. With αν the optative is often used as a milder future, or 
 less positive assertion. This is due to the refinement and 
 sensitiveness of the Greek intellect, and their dislike of what 
 is blunt, and downright, and uncontingent ; as 
 
 ουκ αν άπέΚβοιμ' άλλα κόψω την θύραν I won't go away 
 
 but I'll knock at the door. 
 ουκ αν εγωγε θεοϊσιν επουρανίοισι μαχοίμην 1 will not fight 
 
 with heavenly gods. 
 ουκ αν φθάνοις λέγων ; quantocius dicas ! quin statim 
 
 loquere ? speak at once ! 
 ουκ olci' αν ει πείσαιμι I doubt whether I could persuade. 
 
 —Eur. Med. 941. 
 ουκ αν οίδ' εί Ιυναίμην I doubt whether I should be able. 
 
 —Plat. Tim. p. 26. 
 
 In the last two examples the cu> belongs to the optative, but 
 is merely transposed by a spurious hyperbaton ; as 
 
 ουκ οΙ(Γ ει = Ι doubt whether, πείσαιμ αΐ' = Ι could per- 
 suade him. 
 ουκ olb" εΐ —haud scio an. 
 
 4. In polite commands, the optative is often used with av 
 which points to a suppressed protasis ; as 
 
 χωροΊς αν είσω go in, please ! (literally, * you would go in 
 
 if it should please you.') 
 ερΰοι τις ην έκαστος είΰείη τεχνην = ϊΐβ SUtor ultra cre- 
 
 pidara, 
 
 Η 
 
146 A BRIEF GREEK SYNTAX. 
 
 άλλα ταντα μεν και φθάνω αν είποιεν (Herod, ix. 71) but 
 people might say this even out of envy (sc. ει είποιεν if 
 they were to say it). 
 
 5. It expresses a sort of hopeless wish (hopeless because the 
 optative throws it in connection with things past) ; as 
 
 ποτ τις φύγοι (Ar. Plut. 438) whither could one fly ? 
 but 
 
 ποί τις αν ψνγοι ' whither in the world' 
 
 is more common, and ποί τις φυγή. 
 
 6. The optative is often used in sentences which imply 
 iteration, or indefinite frequency ; * as 
 
 οπότε προσβΚέφειέ τίνα whenever he saw any one. 
 Ζεινότατον Ιε ήν η άθυμ'ια οπότε τις αίσθοιτο κάμνων but 
 most terrible was the despair whenever any one felt 
 that he was falling ill. 
 This is also the case in English where ( might ' is used to 
 express recurrence, as in Shelley : 
 
 1 The sweet nightingale 
 Ever sang more sweet as day might fail.' 
 
 7. What is called the correspondence of optatives should be 
 noticed, where the principal verb in the optative seems to 
 attract the dependent verb into the same mood ; as 
 
 γενοίμαν κ.τ.\. όπως προσείποιμεν Άθάτας (Soph. Aj. 
 1217) would that I were, &c, that we might address 
 Athens. 
 οΚοιο μηπω πριν πάθοιμι (Soph. Phil.) may you perish — 
 not till I have learnt. 
 N.B. It may be as well to repeat, that as an all but invariable 
 rule ει takes the optative, εάν, ην the subjunctive ; civ by itself 
 the optative. 
 
 The Moods in Compound Sentences. 
 
 178. Of the different kinds of possible sentences, those 
 which chiefly need elucidation are : 
 
 1. Final sentences ('in order that'). 
 
 2. Declarative sentences (oratio obliqua). 
 
 3. Conditional or hypothetic (' if,' &c, ' then,' &c). 
 
 4. Temporal (' when, until,' &c). 
 
 * Not that the mood of itself necessarily involves this concepth 
 Burggraff acutely remarks, 'L'emploi d'un temps dans telle ou te 
 circonstance et son emploi pour exprimer cette circonstance, sont det 
 choses differentes que les grammairiens ont souvent confondus.' — p. 412, 
 
 \ 
 
FINAL SENTENCES. 147 
 
 Final Sentences. 
 
 179. A final sentence is one which expresses a purpose, 
 motive, or end (finis). In English it is generally expressed 
 by ' to,' but never by the infinitive in Latin prose, and not 
 properly in Greek. 
 
 It may sometimes appear to be expressed by the infinitive ;* 
 as 
 
 %kQev άΙικεΊν or ως, ώστε ά^ικεΊν he came to do wrong. 
 στρατηγείν ηρη μένος chosen to be a general. 
 βή δ' ιέναι he started to go. 
 
 But here it is rather a fact or consequence which is indicated ; 
 and when the final sentence appears to be expressed by a 
 future participle it is really temporal ; as 
 
 ήλθεν άίικησων he came to do wrong. 
 έρχομαι φράσων I come to tell. 
 
 180. After verbs of sending, coming, &c, 6ς, όστις, are 
 used with the future indicative (whereas in similar Latin in- 
 stances qui requires the subjunctive) ; as 
 
 πέμπειν τινας . . . ο'ίτινες κατηγορησονσι των τα Φιλίππου 
 πραττόντων (Demosth. De F. Leg. § 349) to send some 
 to accuse Philip's faction. 
 
 κίφνκα προαπεστε'ιλατε όστις ημ'ιν σπείσεται (Id. § 189) 
 ye sent a herald before us to make a truce for us. 
 
 N.B. <fl Oc cum conjunctivo nunquam ponitur post verba 
 mittendi, veniendi, similia.' — Shilleto. 
 
 181. Sentences really final, or expressive of purpose, are 
 expressed by 'ίνα, όπως, ως in order that (always with μη not 
 ov in negative clauses) ; and the rule about them both in 
 Greek, Latin, and English is, that they are followed by the 
 subjunctive after primary tenses, and by the optative after the 
 historical tenses ; as 
 
 γράφω, γράφω, γέγραφα Ίνα μανθάνης or μάθης 
 scribo, scribam, scripsi (perfect) ut discas 
 I am writing, will write, have written that you MAY be 
 learning, or may learn ; 
 
 εγραώον, έγραφα, έγεγράφη ϊνα μανθάνοις or μάθοις 
 
 scribebam, scripsi (aorist), scripseram ut disceres 
 
 I was writing, wrote, had written that you might learn. 
 
 * But see Jelf, § 669, p. 300, and supra. 
 h2 
 
148 A BRIEF GREEK SYNTAX. 
 
 182. This rule is constantly violated in the New Testament, 
 and by later writers (e.g. Lucian), because the optative fell 
 out of general use. When it is violated by any Attic writer, 
 the reason is the same as that which leads to the use of the 
 imperfect tenses (historic present, &c), namely, a desire to be 
 graphic (προ ομμάτων ποιεΊν) by representing the event as 
 passing under the eyes ; e.g. 
 
 κτείνει με "χρυσού τον ταλαίπωρον 'χάριν 
 
 ξένος πατρώος, και κτανών ες οίΰμ' άλος 
 
 μεθηχ ϊν αυτός χρυσον εν Ζόμοις έ'χτ/. 
 
 * My father's friend slays me, unhappy that I am, for the 
 sake of gold, and after slaying, he flung me into the 
 sea- wave, that he may be having (=may keep, the 
 effect being represented as present and continuous) 
 the gold in his house.' — Eur. Hec. 
 
 183. i. The historic present is syntactically regarded as an 
 aorist, and may therefore be followed by the optative. 
 
 ii. The subjunctive and imperative, as they connect the 
 action with the future, are regarded as primary tenses, and 
 are therefore regularly followed by the subjunctive. 
 
 184. When the final particles ως, Ίνα, όπως are used with 
 past tenses of the indicative, they imply an impossible or un- 
 fulfilled result ; as 
 
 τί μ ου λαβών 
 εκτεινας ευθύς ως έδειξα μηποτε, κ.τ.λ. — Ο. Τ. 1393. 
 i why didst thou not seize and slay me instantly, that I 
 might never have shown,' &c. 
 
 ει ο ακουουσης ετ ην 
 
 πηγής δι' &των φραγμός, ουκ αν εσγόμην 
 
 το μη 'ποκλεΐσαι τουμον αθλιον ΰέμας, 
 
 ϊν 1\ν τυφλός τε και κλύων μηΰέν. — Ο. Τ. 1389. 
 
 *had there been besides any stoppage of the fount of 
 hearing, I had not restrained myself from closing up 
 my wretched frame, that I might have been both blind 
 and hearing nothing.' 
 
 εζητησεν αν με . . . 'ίνα μηΰεν δίκαιον λέγειν έΰόκουν he 
 would have sought me . . . that I might have appeared 
 to be saying nothing just. — Dem. 
 
 N.B. These passages are sometimes rendered i in which 
 case I should have,' &c. ; the negative μη shows that such a 
 rendering is incorrect. 
 
RELATIVE SENTENCES. 149 
 
 185. Sometimes in Thucydides and other writers the imme- 
 diate and certain result is in the subj., the remoter and less 
 certain in the opt. ; as 
 
 ■περί yap lie ποιμενι Χαών 
 μη τι πάθη μέγα £ε σφεας άποσφηλειε πόνοιο. 
 for he feared greatly for the shepherd of the people, lest 
 he may suffer harm, and might so greatly thwart them 
 in their toil.— 11. v. 567.* 
 παρανΊσχον φρυκτούς όπως ασαφή τα σημεία τοΊς πολεμιοις 
 ή και μη βοηθοΊεν they kept raising counter fire-signals, 
 that the signs may be unintelligible to the enemy, and 
 they might not come to the rescue. — Thuc. iii. 32. 
 
 Relative Sentences. 
 
 186. The rule about final clauses holds also in correlative 
 sentences ; as 
 
 ουκ εχω, ε£ω, εσχηκα οποί τραπωμαι. 
 ουκ εϊχον, εσχον οποί τραποίμην. 
 
 187. In relative sentences αν follows the relative when the 
 subjunctive is required ; as 
 
 ov civ 'ίΰη κολάζει he punishes whomsoever he sees ; 
 but 
 
 ov 'ίΰοι εκόλαζεν he kept punishing every one whom he 
 
 saw (i. e. as often as he saw them, — the opt. implying 
 
 iteration). 
 
 The reason of this is obvious ; it is here due to the futurity 
 
 involved in the subjunctive, which requires an av to qualify it. 
 
 188. And here we may add the important rule that ώς αν, 
 όπως αν, υς αν, όταν, επειΖάν, ει άν (εάν), &c, go regularly 
 with the subj. ; in the rare cases in which 6ς, όστις, ώς, όπως, 
 εΐ, followed by άν, occur with an optative, the av belongs, not 
 to them, but to the verb ; as 
 
 ουκ εστί τοϋτον όστις αν κατακτάνοι there is no one who 
 
 would KZZhim [not όστις av whoever, but όστις who av 
 
 κτάνοι would kill]. 
 ουκ εστίν ο,τι άν τις μείζον τούτου κακόν πάθοι there is no 
 
 evil which (ο,τι) one could suffer (αν πάθοϊ) greater 
 
 than this. 
 
 * See Arnold, ad Thuc. iii. 22. Other instances of this succession of 
 consequences, indicated first by the subjunctive and then by the optative, 
 are Thuc. viii. 17 ; Herod, ix. 51 ; Eur. Hec. 1120 ; El. 56 ; and in Latin, 
 Virg. Mn. i, 298. 
 
150 A BRIEF GREEK SYNTAX. 
 
 επιμέλονται ώς αν βέλτιστοι ειεν οι πολϊται tliey take 
 pains Ιιοιν (ώς) the citizens might be (av ehv) most 
 excellent. 
 
 οΰκ-οΐδά-γ'-εί φθαίης-αν I almost doubt whether you will 
 be in time (φθαίης av)=Vm afraid you won't. 
 
 N.B. The general rule is that the relative, when definite, 
 takes the indicative, as ους είΰεν those whom he saw ; when 
 indefinite the optative, as ους 'ίΰοι those whom he might see; 
 when combined with av, invariably the subjunctive, as ονς av 
 Uy whomsoever he may see. 
 
 ORATIO OBLIQUA. 
 
 • 
 189. In oratio obliqua (indirect assertion, reported speech), 
 when it is not expressed by the accus. and inf., the indicative 
 may be used with ώς or on, 
 
 i. when the exact words of another are quoted ; or 
 ii. when the statement is vouched for as a, fact ; or 
 iii. when some special emphasis attaches to one part of the 
 sentence; as 
 i. λέγει οτι 6 άνήρ θνητός εστί he says that l the man is 
 mortal.' 
 φας επ\ χώρην άζειν όθεν χρυσον ο'ισονται saying that 
 he will lead them against a country from which they 
 will (for a certainty) win gold, 
 ii. ελεγον οτι Κύρος μην τέθνηκεν, Άριαίος £ε πεφευγως 
 εν τω σταθμω ε'ίη, και λέγοι οτι περιμένειεν άν αυτούς 
 ει μελλοιεν ηκειν they said that Cyrus was dead [a 
 fact], and that Ariasus having fled was in his camp, 
 and that he said he would wait for them if they 
 intended to come [assertions which might be true 
 or not], 
 iii. εκέλευε της εωυτοϋ χώρης οΐκέειν οκου βοΰλονται (Herod, 
 i. 136) he bade them live in his own country where- 
 ever they prefer, 
 θαυμάζοντες οποί ποτέ τρέπονται οι "ΈΧληνες κα\ τ'ι εν 
 ν ω εχοιεν wondering whither the Greeks will turn 
 themselves, and what their purpose possibly could be. 
 
 In Latin, this opinion as to the truth or doubtfulness of what is 
 reported cannot be shown by the form of the sentence, because the 
 accusative and infinitive is their only form for indirect assertions ; * nor 
 
 * The reason of this is that Latin has no equivalent to the Greek 'ότι 
 with the indicative merely stating a fact ; ut is a final conjunction in 
 
CONDITIONAL SENTENCES. 151 
 
 can it be shown in English. But in German the distinction is just tho 
 same as in Greek, i.e. the indicative is used of certainties (Er sagt er ist 
 gefallen), the subjunctive of uncertainties (Er sagt er set gefallen). 
 
 190. The optative however is the ordinary mood for oratio 
 obliqua after historical tenses (including the historical pre- 
 sent); as 
 
 ηρετο εϊ αίσθάνοιτο he asked whether he felt it. 
 This use of the optative in oratio obliqua once existed in 
 English, e.g. Sir I. Newton, in a letter to Hadley, writes : 
 ' Since my writing this I am told how that Mr. Hooke should 
 make a great stir,' &c. 
 
 This subjunctive is only used irregularly when the reporter 
 involuntarily slips back into the oratio recta, generally from 
 some allusion to the future ; as 
 
 ελεγον, ως χρήν υμάς ευλαβε'ισθαι μη ι/π' εμον έζαπατηθήτε 
 I kept telling you that ' you ought to be on your guard 
 that you may not be deceived by me? 
 
 191. The same rule holds good of indirect interrogation.' 
 
 192. The tenses used are those which would be used in 
 oratio recta, or direct speech; thus the three assertions 'he 
 did it,' ' he has done it,' ' he will do it,' would be respectively 
 in oratio obliqua, ελεγον on 7Γ0ί//σειε, πεποιηκώς ε'ίη, ποιησοι. 
 
 193. The accusative and infinitive may always be used in 
 oratio obliqua ; as 
 
 ?')γγειλαΐ' τον Kvpov νικάν they announced that Cyrus 
 was victor. 
 
 Conditional Sentences. 
 
 194. Every complete conditional sentence consists of two 
 clauses, of which the clause which contains the condition 
 (' ι/ 1 ') is called the protasis, and the clause which expresses 
 the inference or consequence is called the apodosis. 
 
 195. Since, in these sentences, Greek is able to express 
 very numerous shades of thought (modified even by the pass- 
 ing emotion of the moment), which English does not, and 
 often cannot idiomatically (i.e. in accordance with the ordinary 
 use of the language) express; and since, in consequence of 
 this, the apodosis often places the statement in a slightly 
 
 Latin. The difference between 'ότι and ώ$ in declarative sentences is 
 slight, but of the two 'ότι implies rather ' the fact that,' and «y the asser- 
 tion that, 
 
152 A BRIEF GREEK STiNTAX. 
 
 different point of view from that on which the protasis is 
 framed, it will be convenient to treat the forms of protasis* and 
 apodosis separately, and then to give instances of them in 
 combination. 
 
 196. A categorical proposition declares that something 
 actually took place ; a conditional proposition only states a 
 connection between two events of which one depends on the 
 other. 
 
 The Protasis. 
 
 197. The common way of expressing the protasis is by ει 
 or εάν. 
 
 Ει,* ' if, 1 is derived by Donaldson from the dative of the 
 pronoun t, gen. ov. It would therefore mean ' on this condi- 
 tion.' It is joined with the indicative (generally the imperfect 
 or aorist), and the optative ; very rarely with the subjunctive. 
 
 'Εάν = ει ην, and may be compared with our pleonastic 
 1 anif \ it invariably takes the subjunctive. 
 
 198. The protasis may imply: I. Possibility, or mere 
 assumption (sumptio dati). II. Slight probability. III. Un- 
 certainty, or mere supposition. IV. Impossibility (sumptio 
 ficti) ; as in the following typical sentences to which the 
 English and Latin equivalents are appended : 
 
 199. I. Possibility (the condition being assumed) ; as 
 ει τι έχει if he has anything, si quid habet. 
 
 ει λέγει τοντο if he says this, si hoc dicit. 
 
 ει γενήσεται "f ταΰτα if this shall happen, si haec accident. 
 
 ει τοντο επεπράχει if he had done this (the result still 
 
 continuing) : this is a nuance of meaning which we 
 
 cannot express in English. 
 
 We see then that εί with the indicative implies a mere 
 assumption ; and is equivalent to our ' assuming that.' It is 
 purely neutral, and expresses no opinion either way. 
 
 N. B. In this sense ει may go with any tense of the indica- 
 
 * el also = δτι ' that;' for which it is a politer form, after verbs im- 
 plying disapprobation ; and verba affectuum generally (θαυμάζω, αγαπάω, 
 δανόν ian, &c). It also has the sense of num ? si ? whether ? in in- 
 direct questions. 
 
 f el, si, ' if,' with the future is comparatively rare in all three lan- 
 guages. Notice the difference between ei vet νίφη ecri, if it is raining 
 there are clouds, and el v<rei νικήσουν, if it rains (at some future time) 
 we shall win. 
 
THE PROTASIS. 153 
 
 tive ; it only indicates impossibility (or that a thing is not the 
 case) when it is folloiued by the indicative with ar, e. g. 
 
 ει ποτέ τοι •χαρίεντ 1 επι νηον έρεψα 
 
 . . . τοίε μοι Kpyrjvov εε'λδωρ. — II. ί. 39. 
 
 * If ever I reared for thee a beauteous fane . . . accom- 
 plish for me this my desire.' 
 
 εί τις και τότε ώργίζετό μοι . . . άναπειθέσθω (Thuc. νΐ. 
 89) if then any one was angry with me . . . let him 
 now change his opinion. 
 
 σοι ει πή άλλτ/ ΙέΙοκται λέγε if you have come to any 
 different conclusion, tell me. 
 
 200. II. Slight probability ; as 
 
 εάν τι εχτ} if he have * anything, si quid habeat. 
 εάν τοϋτο λε'γί? if he say this, si hoc dicat. 
 lav γένηται ταντα if this happen, si hsec accidant (or 
 acciderent). 
 Έαν is a compound of ει and ar, and calls attention to some 
 condition ; it is invariably joined to the subjunctive ; hence it 
 differs from I. because it must always refer to future time.f 
 
 201. III. Complete uncertainty (the condition being purely 
 imaginary) ; as 
 
 ει τι έχοι if he were (or, should be) having anything, si 
 quid habeat. 
 
 * The English subjtmctive, in this phrase, implies the same shade of 
 probability; whereas ' if he has,' like d £χ«, expresses no probability 
 whatever, but merely 'assuming that, then,' &c. Yet the difference 
 between the two is so slight that both may be used in the same clause. 
 (Herod, iii. 36.) 
 
 f El (as well as iav) may, very rarely, be joined even in good writers 
 with the subjunctive. (See Hermann, ad Soph. Aj. 491, de particula &v, 
 p. 96.) The distinction between the very rare el •γ4ντ\ται and the common 
 correct construction iav γένηται can hardly be expressed in English or 
 Latin, except by using ' forte ' ' perhaps ' in the latter case. Thus 
 we have — 
 
 1. €t yerfjaeTcu ταύτα assuming that this will happen (possibility). 
 
 2. iav yivTjrai ταύτα if perchance this happen (probability). 
 
 3. et 7«/7)ται ταύτα if this happen (apart from any conditions). 
 
 4. «ί γένοιτο ταύτα if this should happen (uncertainty). 
 
 It will be seen that the nuances of meaning here conveyed are too deli- 
 cate to be expressed except by periphrases in Latin or English, and barely 
 even by them ; in fact, even high authorities (e.g. Host) deny the exist- 
 ence of any perceptible difference between 1 and 3, and Liddell and 
 Scott between 2 and 3. Certainly, et with the subjunctive is rare and 
 archaic ; one would but rarely require to say ' if — leaving all conditions 
 out of sight — not implying the probability or even the possibility of the 
 supposition.' 
 
 h3 
 
154 A BRIEF GREEK SYNTAX. 
 
 ει τούτο λέγοι if he were (or, should be) saying thi3, si hoc \ 
 dicat. * 
 
 el γένοιτο ταντα if this were to (or, should) happen, si heec 
 accidant. 
 
 Both the English l were ' and the Greek optative strictly 
 belong to the past, but in these instances the supposition refers 
 to the present (if he were now to, &c). This form of protasis 
 might also be correctly rendered in English by i If he hadf 
 L if he said, 1 &c; but this, though more idiomatic, would not 
 be strictly correct or accurate. 
 
 Latin makes no distinction between this and II., using the 
 pres. subj. for both ; or else employing ' si quid haberet,' &c. 
 for both this and IV. 
 
 N.B. "When ei is used with the optative, the sense varies 
 with the tense ; e.g. 
 
 εϊ ταϋτα ποιοϊ if he should be doing this (now), 
 „ „ ποιήσοι if he should do this (hereafter), 
 „ „ ποιήσειε if he did this. 
 
 202. IV. Impossibility (the condition being denied). 
 
 α. ε'Ί τι είχε ν if he were (or had been) having, si quid 
 
 haberet. 
 β. ει τι εσχεν if he had had, si quid habuisset. 
 
 α. ει τοντο ελεγεν if he were (or had been) saying this, 
 
 si hoc diceret. 
 β. ει τοντο ελε'ξεν if he had said, si hoc dixisset. 
 
 α. ει εγίγνετο ταντα if this were (or had been) happening, 
 
 si haec acciderent. 
 β. ει εγένετο ταντα if this had happened, si haec acci- 
 
 dissent. 
 
 N.B. When these sentences are set in examination papers, 
 as is so frequently the case, the student should give an accurate 
 English translation, even at the expense of our ordinary idiom ; 
 and therefore ει τι εΤγεν εΐίΰον αν should not be rendered ' if 
 he had anything he would give it' (as in Arnold, Dr. Donald- 
 son, &c), but by these two formularies (either of which is 
 correct, and both of which should be given) : 
 a. i If he were having anything," 
 
 he would be giving it ' 
 or l. 'If he had been having any- j-si quid haberet, daret. 
 
 thing, he would have 
 
 been giving it ' 
 
THE APODOSIS. 155 
 
 This is a literal translation of the Greek which is required ; 
 but, no doubt, neither sentence is in idiomatic English, which 
 would require for 
 
 a. 'If he had anything, he would give it,' for 
 
 b. i If he had had anything, he would have given it ;' 
 
 which last would be expressed in Greek by el τι εσχεν, εδωκεν 
 α r. The very fact that a study of Greek enables us to appre- 
 ciate shades of thought so subtle as to be scarcely capable of 
 being expressed in our own language, adds to its value as an 
 educational instrument. 
 
 203. The reason why the student will constantly see dif- 
 ferent English forms used to render these expressions, is the 
 practical inaccuracy of the English language in neglecting all 
 these shades of thought. We have tried to use the most 
 accurate English equivalents ; but, practically, English en- 
 tirely neglects the distinction between continued and single 
 actions in conditional sentences ; and thus, though ε'ί τι ειγεν 
 means i if he ivere (or had been) having,' and ει τι εσγεν means 
 ' if he had had, and although these forms convey clearly 
 distinct meanings, yet ordinary English would use ' if he had 
 had' for all three. 
 
 Dr. Collis, in a letter to me, writes : ' We in English should 
 say, If you took that money, you are a thief. We do not 
 stop to weigh whether the stealing is a habit, or a repeated 
 single act, or in what degree of uncertainty, possibility, or 
 probability it may be predicated ; nor whether the result is 
 that with such or such a degree of contingency you will be, 
 or may be, or may be considered to be always, or in that one 
 particular instance a thief; we simply say, with a thump on 
 the table, You are a thief.' 
 
 N.T3. — Notice the use of eftte, el (like the Latin si) in wishes ; as e¥0e 
 τούτο iyiyvero utinam hoc fieret ; eftJe iyivero utinam factum esset ; 
 el yap yevono utinam fiat ! In unfulfilled wishes, eWe, el yap, are used 
 with the imperfect (of continuous) and aorist indicative (of single acts), 
 as eftP ήσθα δυνατά τούτο δραν would that you had been able to do this ; 
 eWe <re μηποτ' βίδόμην would. I had never seen you ! 
 
 The Apodosis. 
 
 204. The same Protasis may have different Apodoses ac- 
 cording to the meaning required. The commonest forms of 
 apodosis are 
 
 a. The imperative. 
 
 β. Some tense of the indicative. 
 
156 BRIEF GREEK SYNTAX. 
 
 γ. The optative with av tvhich is the commonest of all, 
 and may follow any protasis, because being more 
 polite and indirect the Greeks preferred it to the 
 indicative. 
 l•. When the now-fulfilment of the condition is implied, a 
 past tense of the indicative with av. 
 And here Ave again meet the distinction between the aorist 
 and the imperfect with av, which may indeed be unidiomati- 
 cally expressed in English, but which for the most part we 
 neglect; thus 
 
 άπέθνησκεν av means 'he would be dying,' or ' he would 
 
 have been dying ; ' 
 άπίθανεν av ' he would have died ; ' * 
 ετεθνηκει av he would have been dead. 
 
 COMPLETE CONDITIONAL SENTENCES. 
 
 205. I. Possibility, or mere assumption, with no expression 
 of uncertainty. 
 
 ε'ί τι έχει, ΰίΰωσι si quid habet, dat. If he has anything, 
 
 he gives it. 
 ει τοϋτο λέξεις, αμαρτήσει si hoc dices, errabis. If you 
 
 say this, you will be in the wrong. 
 
 206. Π. Slight probability. 
 
 εάν τι εχ)/, δώσει si quid habeat (or habebit), dabit. If he 
 
 have anything, he will give it. 
 εάν ταντα λέξρ, αμαρτάνει si hoc dicat, errat. If he say 
 
 this, he errs. 
 
 207. III. Uncertainty, or mere supposition. \ 
 
 ε'ί τι εχοι, ΰοίη αν J si quid habeat, det {rare in Latin). 
 
 If he were (or should be) having anything (sc. now), 
 
 he would give it. 
 ει ταντα λέγοι, άμαρτάνοι av si haec dicat, erret. If he 
 
 were (or should be) saying this, he would be erring. 
 
 * Some scholars maintain that απέθανζν &v may mean ' he would die,' 
 as well as ' he would have died ; ' but this is exceedingly questionable, 
 and therefore I have taken no notice of it. 
 
 f Or indefinite frequency ; as e5f που i^eXavyoi irepiriye τον Kvpov when- 
 ever he went out riding he used to take Cyrus about with him. 
 
 X This is the favourite apodosis, and is often put with one of the other 
 protases ; e.g. τ&μ iav OeKrjs ίττη Κλύων δέχεσθαι . . . 'Αλκήν λάβοπ &ν 
 (Soph. Ο. Τ. 216) if you be willing to listen to and obey my word . . . 
 you would gain help (where λάβοπ \ν is politely indefinite for λήψ€ί). 
 
CONDITIONAL SENTENCES. 157 
 
 208. IV. Impossibility, or the implied nonfulfilment of the 
 condition. 
 
 α. ε'ί τι ειχεί', εΐίΐου αν* si quid haberet, daret. If he 
 were (or had been) having anything (which is not the 
 case) he would be (or have been) giving it. 
 ει ταϋτα ελεγεν ημάρτανεν αν si hsec diceret, erraret. 
 If he were (or had been) saying this, he would be (or 
 have been) in the wrong. 
 
 β. ε"ι τι εσχεν, 'εΊωκεν αν si quid habuisset, dedisset. If 
 he had had anything, he would have given it. 
 ει ταντ" 1 ελεζεν ημαρτεν αν si hiec dixisset, errasset. If 
 he had said this, he would have been in the wrong. 
 
 209. It will be seen at once, as already stated, that the 
 chief difficulty in understanding the use of conditional sen- 
 tences arises from the fluctuating and uncertain use of the 
 English equivalents, since our ordinary idiom often prevents 
 us from representing the accurate meaning of the Greek ; yet 
 we may in English accurately render 
 
 I. by ' if 1 with the indicative.! 
 II. by ί if with the subjunctive. 
 ΠΙ. by ' if' with ' were to' or ' should.' 
 IV. β. by 'if with the pluperfect, and by 'would have* 
 in the apodosis. 
 
 210. The main difficulty is with IV. a. Many scholars 
 translate ε'ί τι είχετ, εΰίΰου αν by ' if he had anything, he 
 would give it ; ' others, declaring this to be inaccurate and 
 unphilosophical, render it ' if he (were, or) had been having 
 anything, he would (be, or) have been giving it.' It is clear 
 that in many sentences, such periphrases would be intolerable 
 in classical English, although they are correct, and discri- 
 minate well such sentences as 
 
 α. ει μη τότ 1 επόνονν, ουκ αν νυν εύφραινα μην had I not 
 then been toiling, I should not now have been rejoicing. 
 
 β. ει τοϋτ επο'ιει μέγα με ωφελεί αν if he had been acting 
 this, he would have been doing me a great service. 
 
 Clearly ει τοϋτ εποίει, and therefore the apodosis dependent 
 
 * Compare the French S'il avait, il donnerait. 
 
 f The protasis of every one of these four may be represented by 
 %χων τι ; and that of I. by & e%ei ; of II. by & αν εχρ ; of III. by & ίχοι ; 
 Of IV. a. by δ εΤχ«> ; of IV. β. by & ίσχςν. 
 
158 A BEIEF GREEK SYNTAX. 
 
 on it, sometimes refers to the present* sometimes to the 
 past; e.g. 
 
 εϊ top Φίλιππον τα δίκαια πράττοντα εωρων, σφόδρα αν 
 Οαυμαστον ηγοΰμην avroy if I but saw Philip acting 
 Avith justice, my opinion of him would be that he is 
 very admirable. 
 ούτος ει ϊ\ν προφήτης εγίνωσκεν άν if he were a prophet, 
 he would be aware. 
 
 211. The Greek love for dramatic imperfects, expressive of 
 continuous acts, going on as it were before the eyes, leads 
 them to a constant use of this form of the conditional sen- 
 tence; e.g. 
 
 ουκ αν ττροέλεγεν εϊ μη επίστευεν άΧηθεύσειν he would not 
 have been in the habit of saying so beforehand, had he 
 not been confident that he would be speaking truth. 
 
 ουκ αν ουν νήσων εκράτει, εϊ μη τι καΐ ναυτικον εϊγεν he 
 would never, then, have held sovereignty over the 
 islands, had he not been in possession of some fleet 
 also. 
 
 212. To sum up then what has been said about IV. a., the 
 context only can determine exactly whether in the particular 
 instance any such sentence as 
 
 ει ταΰτ εγίγνετο, άττέθνησκεν αν means 
 If these things were taking place, he would be dying ; 
 or, If these things had been taking place, he would have been 
 dying. 
 
 213. One or two instances of conditional sentences, both 
 Greek and Latin,f are added, in some of which the apodoses 
 are varied J from the regular construction. In the light of 
 
 * Dr. Donaldson cannot be right in making it refer to the present 
 only. (Gr. Gram. p. 540.) In the same way, 'Si quid haberet, daret,' 
 may mean either ' if he had been having anything, he would have been 
 giving it.' Vellem = έβουλόμην hv lit. I should have been wishing, or 
 ' I should be wishing,' sc. if it were, or had been, possible. In English 
 however we should use neither of these imperfects to express the con- 
 tinuous action, but merely ' I could have wished.' 
 
 f I borrow some of these from a difficult, but careful little treatise on 
 The Theory of Conditional Sentences, by Mr. E. Horton Smith (Mae- 
 millan). Many Latin instances are given by Jani in his Art of Poetry 
 (Engl. Tr.), p. 52. 
 
 \ Such a change in the apodosis of a sentence is regarded as an 
 inaccuracy in English (however frequently it may occur); e.g. such a 
 sentence as Steele's, ' If you please to employ your thoughts on that 
 subject, you would easily conceive,' &c, where * you mil,' &c, would 
 
CONDITIONAL SENTENCES. 159 
 
 what has gone before they will be easily understood by the 
 attentive student; their occasional irregularities are all due 
 to the triumph of the dramatic tendency over formal grammar. 
 
 I. Possibility (condition assumed). 
 
 Εΐ μ* εθέλεις πολεμίζειν," Αλλους μεν κάθ ίσον if you want 
 
 me to fight, make the rest sit down. — II. iii. 67. 
 ή καλόν, ην δ' εγώ, τέχνημα κέκτησαι, έίπερ κέκτησαι in 
 
 truth, said I, a fine contrivance you havfe acquired, if 
 
 you have but really acquired it. — Plat. Prot. p. 319 a. 
 εΐ μεν θεοΰ ήν, ουκ ήν, φήσομεν, αισχροκερίης if he was 
 
 the son of the god, he was not, we shall say, basely 
 
 avaricious. — Plat. Rep. 408 c. 
 Erras, si id credis, et me ignoras, Clinia, you are mistaken 
 
 if you think so, and don't know me, Clinia. — Ter. 
 
 Heaut. i. i. 53. 
 Si quod erat grande vas la3ti afFerebant, if there was any 
 
 large vessel, they would bring it to him with exultation. 
 
 — Cic. 77. Verr. iv. xxi. 47. 
 
 II. Slight probability. 
 
 Νέος αν πονήσ^ς γήρας εΕεις ευθαλες si juvenis labora- 
 
 veris, senectutem habebis jucundam. 
 και ην άρα μη προχωρήσφ 'ίσον εκάστω εχοντι άπελθεϊν, 
 
 πάλιν ττολεμησομεν and if by any chance things proceed 
 
 not smoothly for each side to separate on equal terms, 
 
 we will go to war again. — Thuc. iv. 59. 
 Nunquam labere, si te audies You will never slip, if you 
 
 listen to your own guidance. — Cic. ii. ad Fam. vii. 1. 
 Pol si istuc faxis (=feceris) haud sine poena feceris Faith 
 
 if you do so, you will not have done it with impunity. 
 
 — Plaut. Capt. in. v. 37. 
 
 have been more regular ; but in Greek, which submitted less tamely to 
 formal rules, and allowed more for the passing play of thought, such a 
 sentence would have been regarded as quite admissible. It is the same 
 in French, where one might have either ' Si vous aviez fait le contraire 
 il aurait mieux valu, il valait mieux, or il vaudrait mieux.' 
 
 I collect one or two English instances of conditional sentences with 
 varied apodoses from an excellent pamphlet by the Eev. E. Thring, ■ On 
 Common Mood Constructions.' They will show that Creek is not in 
 this respect one whit more irregular than our own language. 
 
 k III speak to it though hell itself should gape.' 
 
 1 Thou wrongst thyself, if thou shouldst choose to strike.' 
 
 ' If I answer not you might haply think 
 Tongue-tied ambition yielded.' 
 
 ' An I might live to see thee married once 
 I have my wish.' 
 
160 A BRIEF GREEK SYNTAX. 
 
 III. Uncertainty (condition imaginary). 
 
 ΣΤΡ. γυναίκα φαρμάκια ει πριάμενος θεττα\ην } 
 καθελοιμι την σελήνην, είτα £ε . . . . 
 .... »c£ra τηροίην έχων, . . . 
 ΣΩι. τι Ζήτα τοντό σ ώφελήσειέν σ } ; ΣΤΡ. ΰ,τι ; 
 εϊ μηκέτ' άνατέΧΧοι σεΧηνη μηΖαμοΰ 
 ουκ αν άποΰοίην τους τόκους. — Ar. Nub. 749.* 
 Str. If purchasing a Thessalian witch I should draw 
 down\ the moon {single act), .... and then 
 keep it in my own possession {continued act) . . . 
 Soc. Why, what good would that do you ? 
 Str. What good, quotha ? why if the moon should no 
 longer be rising {continued act) I should not pay 
 {single act) the interest on my debts. 
 
 IV. Impossibility (condition denied). 
 
 a. and β. (combined). Ιίλάτων προς τίνα των παίδων 
 
 Μεμαστίγωσο αν, εψη, ει μι) ώργιζόμην Plato exclaimed 
 
 to one of his slaves, ' You would have been flogged, 
 
 were I not in a passion.' 
 ει επείσθην ουκ αν αρρωστούν had I then taken your 
 
 advice I should not now have been suffering from 
 
 illness. 
 Si has inimicitias cavere poluisset, viveret had he been 
 
 able to avoid this enmity, he would now be living. — 
 
 Cic. p. Rose. vi. 17. 
 Si possiderem (regnum) ornatus esses ex tuis virtutibus 
 
 were I in possession of it, you would have been 
 
 decorated in accordance with your merits. — Ter. Adel. 
 
 ii. i. 214 
 μενοιμϋ αν' ηθεΧον 3' αν έκτος ων τυγεϊν (Soph. Aj. 88) 
 
 I suppose I must stay ; but I should have wished (lit. 
 
 been wishing) to be out of the way. [Here the 
 
 protasis ' had it been possible' ει δυνατόν ήν is (as often) 
 
 suppressed.] 
 
 * Several idioms occur in this instructive example ; e.g. the difference 
 of present (τηροίην, &c.) and aorist (καθ4\οιμι) tenses; the use of the 
 relative 'ό,τι in repeating a question, &c. 
 
 f ' His mother was a witch, and one so strong 
 
 She could controul the moon.' — Shaksp. Tempest. 
 
 ' While the labouring moon 
 Eclipses at their charms.' — Milton. 
 
 \ For other instructive Latin instances, see Mn. iv. 19, ii. 55, xi. 12 ; 
 Ov. Trist. v. v. 42, &c. 
 
TEMPORAL SENTENCES. 161 
 
 TEMPORAL SENTENCES. 
 
 214. In sentences which indicate time by means of any of 
 the particles of time, as οτε, εως, επε'ι, πριν, μέχρις, &c, the 
 general rule is that a. the Indicative is used when facts are 
 stated; β. the Subjunctive with av (as in δταν, επεώάν, &c.) 
 after primary tenses, when anything future and uncertain 
 is mentioned; and y. the Optative (without αν) in oratio 
 obliqua, and after historical tenses, frequently implying re- 
 currence; as 
 
 a. The indicative of facts. 
 
 επε\ Ιε φέγγος ηλίου κατεφθιτο but when the light of the 
 
 sun waned. 
 ουκ ήν αΧίζημ ουίεν πριν γ' εγώ σφισιν εΰειζα, κ.τ.Χ. there 
 was no remedy till I showed them, &c. — iEsch. P. V. 
 479. 
 πίνει εως εΰερμην' αυτόν άμφιβασα 0λό£ he drank till the 
 
 pervading flame warmed him. — Eur. Ale. 757. 
 εφυγον δτε ήλθον οι σύμμαχοι when the allies came, they 
 fled. 
 
 β. The subj. with av of things future and uncertain. 
 
 όταν α χρη ποίησης ευτυχήσεις whenever you do your 
 
 duty you will prosper, quum ofneia tua expleveris, 
 
 felix eris. 
 επειΰάν άπαντα άκουσητε, κρίνατε whenever you have 
 
 learnt all, judge. 
 
 γ. The opt. (generally without civ) after historical tenses, 
 often of indefinite frequency. 
 
 νπερψον είχεν όπότ εν ΐιστει ΰιατρίβοι he used to occupy 
 
 an upper-room as often as he was staying in town. 
 περιεμένομεν εως άνοιχθείη το Ζεσμωτηριον we used to 
 
 wait about, until the prison should be opened.* 
 ουκ ηβούλοντο μάχην ποιεΊσθαι πρ\ν οι σύμμαχοι παραγε- 
 νοιντο they did not wish to fight till the allies should 
 have come up. 
 
 * Sometimes, but rarely, av is added to eW, &c, with the optative, as 
 in Soph. Track. 684, σώζςιν (itciXevev) %ws αν αρτίχριστον άρμόσαιμί που 
 he bade me keep it until (should occasion arise) I might perchance use it 
 fresh-spread. Cf. Ax. Eq. 133. Hermann accounts for this anomaly by 
 saying that where πρ\ν &v, &c, would have the subjunctive in oratio 
 recta, the av may still be retained in oratio obliqua, although there tho 
 optative is substituted for the subjunctive. 
 
162 A BKIEF GREEK SYNTAX. 
 
 Special Uses of πριν, εως, &c. 
 
 215. Notice these facts about the uses of πριν l before? and 
 εως ' until? 
 
 i. πρ\ν αν is never used unless a negative, or something 
 equivalent * to a negative precedes, as 
 
 ου ποιήσω ταϋτα πριν αν κέλευσης non haec faciam, prius- 
 quam jubeas. 
 
 ii. πρ\ν is only used with the optative in oratio obliqua, or 
 when there is reference to the thoughts or words of another. 
 
 ουκ ηθεΧον ποιήσαι ταϋτα πριν κεΧευσειας antequam ju- 
 
 beres. 
 άπηγύρευε μηΖενα βάλΧειν πρϊν Κΰρος εμπΧησθείη he 
 
 forbade any one to shoot until Cyrus was satisfied 
 
 [referring to his own words]. 
 ουκ εθεΧεν φεΰγειν πρ\ν πειρησαιτ Άγ^ιΧηος he did not 
 
 wish to fly till he had made trial of Achilles [referring 
 
 to his thoughts]. 
 
 iii. Sometimes (as we have already noticed § 177, 7) an 
 optative after πρ\ν is due to the attraction of a previous opta- 
 tive, as 
 
 bXoio μηπω πριν μάθοιμι (Soph. Phil. 961) mayst thou 
 perish ! Yet no, not till I learn. 
 
 Here we should have expected the infinitive, but compare 
 0. T. 505. 
 
 iv. πριν, εως, with the subj. differs from πριν άν, εως αν, by 
 being only used in poetry when something certain to happen is 
 spoken of; e.g. an actually dying man should not say μίμνετε 
 'έως αν θάνω but μίμνετε εως θάνω. 
 
 μη στέναζε, πριν μάβης (Soph. Phil. 917) do not groan 
 till you have learnt (which will be the case imme- 
 diately) ; 
 
 but 
 
 εως δ' αν εκμάθης εχ' ελττ -ica till you have learnt (which 
 you may or may not do) keep hope. 
 
 * e.g. a question, or such words as άφρων, &c. In fact, πρϊν very 
 rarely occurs before the optative or subjunctive at all without a nega- 
 tive preceding. ( Jelf, § 848, obs. 8.) For a few trifling exceptions or 
 irregularities, see Shilleto, Dem, de F, Ley. § 235. 
 
uses of πριν. 163 
 
 Usually * however ay is added, because the Greeks disliked 
 talking of future certainties, and ' amant omnia dubitantius 
 loqui.' 
 
 v. We find a similar fact with ws, 'όπως, which (in Attic poets) are 
 used alone with the subjunctive of things certain, as αλλ 5 ws τόδ' etSrjs 
 εννεπω σαφεστερον but I tell you more plainly that you may know it 
 (which of course you will do, when I have told you) ; but σταθ&μεν 
 εκποδών, ws αν μάθω let us stand aside, that I may (sc. if possible) learn. 
 Thus we find them in the same passage, iEseh. Choeph. 983 — 
 εκτείνατ' αυτόν . . . . w s ϊδη πατήρ, 
 ονχ ούμ6ε αλλ' δ πάντ εποπτεύων τάδε 
 "HXios avaryva μ -qTpbs epya rrjs εμής ' 
 ws αν irapfj μοι μάρτυε εν δίκη ποτέ 
 ws τόνδ' εγώ μετηλθον ivti'iKws μόρον 
 τον μητρός. 
 
 Unfold it that .... the sun may see (which of course will be the case) 
 the unhallowed deeds of my mother, so that perchance he may here- 
 after be my witness (of the fact) that I justly wrought this fate of my 
 mother. 
 
 N.B. i. The infinitive with πριν may be substituted for 
 any other mood. 
 
 ii. πριν Ζειπνεϊν before dining, priusquam c&nem. 
 
 πρ\ν Ιειπνησαι before having dined, priusquam 
 
 camavero. 
 πρίν ΙεΙειπνηκεναι before having finished dinner, 
 priusquam a ccena surrexero. 
 
 iii. The following sentences will illustrate the com- 
 monest uses of πρίν. 
 εποίησα ταύτα πρϊν εκέλευσας anteO 
 
 quam jubebas , „ 
 
 . ,/A χ ~ - ν > or πριν σε κελευσαι. 
 
 ουκ ηοε\ον ποιησαι ταντα πριν ι Γ 
 
 κελεύσειας antequam juberes J 
 
 ποιήσω πρίν σε κεΧεϋσαι. 
 
 ου ποιήσω ταντα πρ\ν αν κέλευσης. 
 
 On these sentences we may observe: a. That πρ\ν may 
 
 * %ws &v, with the subjunctive present, often implies duration, — so 
 
 long as. 
 σιωπάτε 'έως αν καθεύδη as long as he continues sleeping, be still, 
 λε'γείν χρη εωε αν εώσιν οι Αθηναίοι, Plato, Thced. 85 A, one must 
 
 continue speaking as long as the Athenians permit. 
 
 It is easy to see that the av is here used because of the uncertainty of 
 tho duration alluded to ; but χρησμούς ενεγκε έ'ων καθεύδει, Ar. Eq. 110, 
 bring the oracles while he is asleep (where no ay is needed — his sleep 
 being a, fact), 
 
16-4 A BRIEF GREEK SYNTAX. 
 
 always go with the accusative and infinitive, except where a 
 negative statement is limited by a future contingency. /3* It 
 takes the indicative when certain facts are spoken of in the 
 past. γ. It takes the optative in oratio obliqua, and after 
 another optative, δ. It is rarely used at all, and with the sub- 
 junctive or optative never, unless a negative notion precedes. 
 
 The INFINITIVE (εγκλισις απαρέμφατος). 
 
 216. The Infinitive can hardly be considered as a mood ; 
 it is rather a noun expressive of action, and therefore it can 
 take the article. Hence some grammarians call it i the noun 
 of the verb ' ('όνομα τοϋ ρήματος). It resembles however the 
 verb in having tenses, in governing cases, in being used with 
 ay, and in being qualified by adverbs, not by adjectives, as 
 καλώς θνήσκειν, but καλός θάνατος. 
 
 217. The connection between the infinitive and the abstract 
 noun accounts for the fact that in many languages — for in- 
 stance in Arabic and in Modern Greek — there is no infinitive 
 mood. We shall see that in most languages infinitives with 
 the article may be used as substantives; e.g. in French le 
 savoir, le toucher, &c. 
 
 218. The uses of the infinitive in Greek are far more rich 
 and varied than its uses in Latin ; e.g. 
 
 τις Φίλιππον κωλύσει hvpo βαΐίζειν, quis Philippum 
 
 impediet quominus hue veniat ? 
 τοίς Αίγινήταις ΪΙοσαν θύρεαν ο\κεϊν dederunt Thyream 
 
 habitandam. 
 πάντες αιτούνται τον θεον τάγαθά ΰιΰόναι omnes homines 
 
 precantur Deum ut bona largiatur. 
 άκονσαι μαλθακά dulcia ad audiendum. 
 φοβερός bpav horribilis aspectu. 
 alia αποΐέζασθαι digna qua? quis accipiat. 
 
 219. Most of the idioms in which the Greek infinitive is 
 employed closely resemble those of English, as will be seen 
 by the following instances, in which the infinitive completes 
 or qualifies the meaning of various words ; as 
 
 ικανός ήν ειπείν he was able to speak. 
 θείειν ανέμοισιν δμοίη like the winds to run. 
 εστί πόα καθίζεσθαι there is grass to sit down upon. 
 μέγα και εσσομένοισι πνθεσθαι great even for posterity to 
 hear of. 
 
INFINITIVES, 
 
 165 
 
 δοκεϊς άμαρτεϊν you seem to have erred. 
 ονχ ηΰύ πολλούς εχθρούς ΐχειν it is not pleasant to have 
 many enemies. 
 For some good remarks on the English infinitive see Prof. 
 Whitney's Lectures, p. 119 ; Abbott, Shahsp. Gram. p. 81. 
 
 220. The Greek infinitive is even used, as in English (but 
 never in Latin prose*), to express a fact or consequence 
 almost resembling a purpose, where the Latin supine would 
 be used : 
 
 μανθάνειν ήκομεν we have come to learn. 
 
 Ξενοφών το ήμισυ rod στρατεύματος κατελιπε φυλαττειν 
 
 το 'στρατόπεϊον Xenophon left half the army to guard 
 
 the camp. 
 ηλθομεν προσκυνήσαι αυτφ we have come to worship him. 
 
 Matt. ii. 2. 
 
 221. It is often qualified by various conjunctions, ώστε, εφ 1 
 ω, &c, and by ?/ after comparatives ; as 
 
 ελπίΰα δε ΰη τίν εχομεν, ώστε μη θανεΊν; but what hope 
 
 then have we of escaping death ? 
 το γαρ νόσημα μείζον τ) φέρειν the disease is too great to 
 
 bear. 
 
 222. In such instances as χαλεπον ευρεϊν, ηΰύ ακοΰειν, θείειν 
 Άριστος, αζιος θαυμάζεσθαι, &c, the infinitive is called epexe- 
 getic, because it defines or limits the notion of the adjective 
 with which it is joined."]• This infinitive is not uncommon 
 after ΙίΙωμι. 
 
 223. It is used in various adverbial phrases, as 
 εκών είναι ' not if I can help it ' (after negatives). 
 εμοΧ ΖοκεΊν in my opinion. 
 
 όσον y εμ ει^έναι so far as I know. 
 ως ειπείν so to speak. 
 το vvv είναι at present, at all events. 
 κατά τοντο είναι in this respect. 
 ολίγου ΰεΊν almost, &c. 
 
 * Latin poets however allow themselves to use a similar idiom with 
 verbs of going, sending, coming ; as 
 
 ' Non nos . . . Libycos populare Penates 
 Venimus.' — Virg. i. 527. 
 'Vultisne eamus visere?' — Ter. Phorm. i. ii. 52; 'ibis frsenare 
 
 cohortes.' — Stat. Sylv. iv. iv. 61. 
 'Legati veniunt spcculari! — Liv. xlii. 25-8 ; Prop. i. 1-12, &c. 
 
 f The Latins copy the Greek epexegetic infinitive in such phrases as 
 
166 A BRIEF GREEK SYNTAX. 
 
 224. In commands,* prayers, laws, expressions of wonder, 
 &c., it is used elliptically, generally with a sententious or 
 dictatorial tone (Jebb, Soph. El. 9). 
 
 γαίρειν πο\\α τον άνδρα θυώνιχον good morning, Thyoni- 
 chus ! (sc. κελεύω ■χαίρειν). 
 
 τους θρψκας άπιέναι παρεϊναι δ' εις ενην the Thracians to 
 go away, and appear the day after to-morrow. 
 
 μη με δουλείας τυχεΊν (grant) that I may not be en- 
 slaved ! 
 
 γνμνόν δε σπείρειν γυμνον δε βοωτεΊν (Hes. Ορρ. 389) 
 nudus ara, sere nudus. 
 
 τούτον νβρίζεη>, άναπνεΊν δε' that this fellow should be 
 insolent, and that he should be alive ! 
 
 So in Latin : 
 
 1 Men' incepto desistere victam?' — Virg. JEn. i. 41. 
 1 Adeone hominem . . . infelicem esse ut ego sum.' — Ter. 
 Andr. I. v. 11. 
 
 225. After verbs of declaring, feeling, &c, the tenses of the 
 infinitive are used in their proper meaning ; as 
 
 ήνάγκασε τους μαθητάς εμβήναι εις το πλόων καϊ προ- 
 άγειν ai /τον he made the disciples embark on the ship 
 (single action), and go before him (continued action). 
 —Matt. xiv. 22-t 
 
 226. The subject of the infinitive is put in the accusative, 
 not in the nominative as in the case of a finite verb, as 
 
 6 Κϋρος ενίκησε, but 
 ηγγειλαν τον Κϋρον νικησαι. 
 
 Kevubs Ιδε7ν niveus videri, Hor. Od. iv. 2, and also the infinitive in appo- 
 sition to the meaning of the sentence; compare δώρ' αθανάτων 61a διδονσιν 
 εχειν, Theogn. 1164, with 'Ille suo moriens dat habere nepoti,' Mn. ix. 
 362, and Ιωκεν ανέμοιε φέρεσθαι with « dederatque comam diffundere 
 ventis,' Virg. Mn. i. 323. ' And give Mm to partake Full happiness 
 with me.' — P. L. ix. 818. ' Une seule remarque reste a faire! — 
 Chateaubriand. 
 
 * This use of the infinitive as an imperative is found in other lan- 
 guages. In Hebrew the infinitive and imperative are generally the 
 same in form. In Provencal ΝΌη temer Maria = fear not Mary. In 
 English military commands, ' Left division to march,' &c. 
 
 f The very frequent use of the infinitive with τον to express purpose 
 in the New Testament (e.g. εισήλθε τον μεΊναι σνν avTols, Lukexxiv. 29) 
 is neither an ellipse of 'ίνεκα, nor a Hebraism, but may be paralleled in 
 classical Greek (see Winer, Gram. Ν. T. § xliv.), and arose from the 
 meaning of the genitive. It is however used in a lax and extended 
 manner, especially by St. Luke. 
 
ACCUSATIVE AND INFINITIVE. 167 
 
 227. This use of the accusative and infinitive in good 
 classical English is very much more rare, although it is not 
 unknown; e.g. I hear you sing, I bid you go. — Clyde. It 
 is really due to what is called antiptosis, i.e. to that pro- 
 lepsis of the subject of the dependent clause, which has been 
 already explained in § 63 ; e.g. 
 
 ελεγον otl 6 Κύρος τέθνηκε they said that i Cyrus is 
 dead,' 
 
 may become 
 
 ελεγον τον Κΰρον otl τέθνηκε,* 
 which is the same as 
 
 ελεγον τον Kvpov τεθνηκεναι. — Curtius. 
 
 228. Instead of the accusative and infinitive after verbs 
 of declaring, otl may be used with the indicative where we 
 should use inverted commas to show that we are quoting a 
 person's exact words, as 
 
 they said * Cyrus is dead' ; 
 
 but where the narrator does not wish to vouch for the fact 
 
 stated, ώς with the optative is used, as 
 
 ΰιαβάλλει τον Kvpov προς τον άΰελφον ώς επιβονλεύοι 
 αντω he accused Cyrus to his brother, alleging that 
 he was plotting against him (compare the English 
 vulgarism ' saying as how '). 
 
 229. If the subject of the infinitive is the same as that of 
 the finite verb, the nominative and infinitive j* are used, as 
 
 εφη ουκ αϊτός άλλ' εκείνον στρατηγεΊν he said that not he 
 
 (himself), but that Nicias was general. 
 6 'Αλέξανδρος εφασκεν είναι Δίος ν'ιός Alexander alleged 
 that he was a son of Zeus. 
 [So too with participles ; as Ίσθι ανόητος ων know that you 
 are foolish.] 
 
 * And this construction with 'ότι being more precise, becomes more 
 frequent in later writers (e.g. in Hellenistic Greek). Accordingly, we 
 are (once more) not surprised to find that the infinitive has vanished 
 from Modern Greek, being replaced by να (='ίνα) and a finite verb ; just 
 as in French, que with a verb is often used where the infinitive would 
 have been used in Latin, because in later Latin quod or quia with the 
 finite verb is substituted for it. 
 
 f This is really a case of brachylogy, i.e. a shortened form of expres- 
 sion, for avrbs ουκ %φη ζαυτον στρατηγών. In Latin, and sometimes in 
 Greek, the full construction is used, as οίοιχαι ψαυτον αμαρτΰν credo me 
 errasse. 
 
168 A BBIEF GREEK SYNTAX. 
 
 It is the same in Latin ; as 
 
 1 Rettulit Ajax 
 Esse Jovis pronepos.' — Ο v. M. xiii. 141. 
 
 230. ' Predicative qualifications referring to a genitive or 
 dative may be in these cases.' — Clyde. 
 
 kUovTo αυτοϋ είναι προθύμου they besought him to be of 
 
 good cheer. 
 εζεστί μοι γενέσθαι ευδαίμονι licet mihi esse beato. 
 
 231. English differs from Greek and Latin in taking a 
 present instead of a future infinitive after verbs of promising, 
 &c. ; as 
 
 ελπίζω ευτυχησειν spero me heatum fore I hope to he 
 
 happy. 
 νπέσχετο Ιώσειν πέντε μι-ας proniisit se quinque minas 
 
 daturum he promised to give five mina?. 
 
 232. The infinitive with the article becomes a declinable 
 substantive, and may be used in any case (το τΰπτειν striking, 
 τοϋ τΰπτειν of striking, &c), thus answering to the Latin 
 gerund; as 
 
 Nom. το άμαρτάνειν ανθρώπους οντάς οΰΰέν θαυμαστον 
 
 4 to err is human.' 
 Gen. επιθυμία τοϋ πιε'ιν desiderium bibendi. 
 Dat. κεκράτηκε τψ πρότερος προς τους πολεμίους Ιεναι 
 
 he has conquered by going first against the 
 
 enemy. 
 
 233. Accus. αΰτο το άποθνήσκειν ουΰε \ς φοβείται no one fears 
 the mere dying. Even without the article the infinitive is 
 often substantival ; as 
 
 δεϊ λέγειν it is necessary to say. 
 
 ογί]σω σε πηΐάν I will stop your leaping. 
 
 ov θανε'ιν ερρυσάμην whom I saved from death. 
 
 234. This substantival use of the infinitive is common to 
 most languages ; e.g. it is found in Hebrew : 
 
 In Latin : Matris lallare recusas, you refuse your mother's 
 lullaby. — Persius. Multum interest inter dare et ac- 
 cipere. — Sen. Benef. v. 10. 
 In German : 
 
 Und ihr Leben ist immer ein ewiges Gehen und 
 
 Kommen, 
 Oder ein Heben und Tragen, Bereiten und SchafFen 
 fur Andre. — Goethe, Herm. und Dorothea. 
 
PARTICIPLES. 169 
 
 In French : II en a perdu le boire et le manger. 
 
 In Italian : Non era l'andar suo cosa mortale. — Petrarcli. 
 
 In Spanish : El mucho estudiar, too much study. 
 
 In English : 
 
 For not to have been dipped in Lethe's stream 
 Could save the son of Thetis from to die. — Spenser.* 
 
 The Participle (μετοχή). 
 
 235. The Participle f has affinities with the adjective, as the 
 infinitive has with the noun. Hence Voss calls the participles 
 mules, ' because they partake alike of the noun and the verb, as 
 the mule of the horse and the ass.' Its essential force is attri- 
 butive, and hence italways refers to some substantive expressed 
 or understood. The present participle in Sanskrit was origin- 
 ally an ablative (or genitive) of the verbal root ending in at ; 
 the nasal addition of η is non-essential, though it appears in 
 the Greek termination ων and the Latin ns. Thus the parti- 
 ciple would be analogous to our participial forms a (i.e. on) 
 hunting, a fishing, &c. We have already seen in the instance 
 of the adjective that it is a common practice in most languages 
 to form new declinable expressions by adding case-endings to 
 some oblique case of a noun ; e.g. in German the adjective 
 vorhandener is obviously formed by declining a dative case. 
 
 236. In the use of the participle, as in that of the infinitive, 
 English and Greek are more rich and varied than Latin or 
 German. In consequence of their frequent use of the parti- 
 ciple, one of the grammarians calls the Greeks φίΚημέτοχρι. 
 
 237. Like the infinitive, the participle may express 
 
 I. Either the necessary accessories of the verbal notion ; as 
 χαίρω τω πατρϊ ίλβόντι I rejoice at my father's arrival. 
 Or 
 
 Π. i It expresses notions of time, cause, manner, which are 
 the mere accidents of the verbal notion ; J as 
 
 * * Our English infinitive is the mutilated form of the dative of a 
 gerund. Rask says that the present infinitive is never used in Anglo- 
 Saxon with the particle to as in Modern English, though the gerund 
 always requires to. 1 — New Crat. p. 603. 
 
 f Μβτοχή iffri \e'|ts μςτίχουσα τί)ί των βημάτων καϊ τη$ των ονομάτων 
 ιδιότητος, Dionys. Thrax, § 19 ; i.e. it is so called from participating in 
 the nature both of verbs and nouns. 
 
 + Jelf, § 680. 
 
170 A BRIEF GREEK SYNTAX. 
 
 τελευτών είπε at last lie said. 
 
 ληϊζόμενοι ζώσι they live by plunder. 
 
 -χαίρων with impunity. 
 
 κλαίων to your sorrow, &c. 
 238. I. It completes the verbal notion by expressing the 
 exact circumstances under which the action took place ; as 
 
 ορώ άνθρωποι' τρέχοντα. 
 
 ακούω Σωκράτους λέγοντος. 
 
 In such cases it is really equivalent to a separate clause 
 
 introduced by οτι, and when the subject of both these clauses 
 
 is the same, the participle is attracted into the nominative, 
 
 e.g. ' / know that I am mortal,' becomes in Greek oUa θνητός 
 
 1 * 
 
 ων. 
 
 The verbs which take this construction are a. Verbs of 
 physical or mental perception, b. Verbs of emotion, c. Verbs 
 of pointing out. d. Verbs which express a state or condi- 
 tion ; as 
 
 α. αδύνατοι ορώμεν οντες περιγενέσθαι we see that we are 
 unable to conquer. 
 προς ανδρός τ\σθετ ήδικημένη she perceived that she 
 
 had been injured by her husband. 
 επειΰάν γνώσιν απιστοΰμενοι when they know that 
 they are distrusted. 
 
 b. ol θεοί -χαίρουσι τιμώμενοι the gods rejoice in being 
 
 honoured. 
 6 ΰέ φρεσϊ τέρπετ' άκονων he rejoiced in heart to 
 hear it. 
 
 c. κακός ων άλίσκεται he is convicted of being base. 
 δήλος εστίν ως τι Ιρασε'ιων κακόν it is evident that he 
 
 intends to do some mischief, f 
 στέργων δε φανερός μεν ϊ\ν ουδένα it was obvious that 
 he loved no one. 
 
 d. τις έτυχε παραγενόμενος ; who happened to be present ? 
 ουκ άνέξομαι ζώσα I will not endure to live. 
 παϋσαι λέγουσα cease saying. 
 
 ηρξαντο οίκοδομοϋντες they began building. 
 διατελεί με αγαπών he continues loving me. 
 
 * With σύνοώα, συγγιγνώσκω εμαιπφ ' I am conscious of,' the nomina- 
 tive or dative may be used, as avvotda 4μαυτω aotybs &v, or σοφψ ovri, 
 N.B. oiSa aya6bs t>v I know that I am good; but οΊμαι ayadbs ehai I 
 think that I am good. 
 
 ^ f Notice the personal construction of λέγομαι, SrjXos, <f>avepos, δίκαιο* 
 €t>t, unlike the English idiom ' it is evident that,' &c. 
 
PARTICIPLES. 1^1 
 
 We find the same idiom in Latin ; as 
 
 Sensit medios delapsus in hostes, lie perceived that he had 
 slipped into the midst of foes. — Virg. 2En. ii. 377 
 (=ζησθετο εμπεσων). Video deceptus ab illis, I see that 
 I have been deceived by them (αισθάνομαι έζηπατη• 
 μένος). 
 
 And it has been imitated by Milton (Par. Lost, ix. 792) : 
 4 She engorged without restraint, 
 And knew not eating death,' 
 
 i.e. that she was eating death. Cf. Oppian, Halieut. ii. 106 : 
 ovV ενόησαν εον σπενΰοντες ολεθρον. 
 
 239. With the infinitive some of these verbs express an 
 entirely different meaning ; e.g. 
 
 έπίσταμαι ποιών I know that I am doing it ; έπίσταμαι 
 
 ποιεϊν I know how to do it. 
 oI£ct αγαθός ων I know that I am good ; olla αγαθός 
 
 είναι I know how to be good. 
 μέμνησο άνθρωπος ων remember that you are mortal; 
 
 μεμνησθω άνηρ αγαθός είναι let him remember to be a 
 
 brave man. 
 φαίνομαι ων it is obvious that I am ; φαίνομαι είναι I 
 
 appear to be. 
 αισχΰνομαι λέγων I am ashamed though I say it ; αισχυ- 
 
 νοίμην αν ειπείν I should be ashamed to say. 
 άρχομαι ΰιΰάσκων I enter on the position of a teacher ; 
 
 άρχομαι ΰιΰάσκειν I begin to teach. 
 λε'ξας έχει he has declared ; έ'χω λέγειν I have something 
 
 to say. 
 
 240. Φθάνω and λανθάνω may have two constructions, as 4ποίηο•4 
 ψθάσα,ς (or avvcas) he did it beforehand or quickly ; σ.πο τείχβος §λτο 
 λαθών he leapt from the wall unnoticed ; or ίφθ-η τνζζος Ιων he was before- 
 hand going afoot, Ιλαθε ψειρών he escaped notice in his flight. It is 
 equally correct to say φθάσον ποιών or ποίησον φθάσαί. 
 
 241. II. The participle expresses the accidents of the verbal 
 notion, — time, cause, manner ; as 
 
 απερ κα\ αρχόμενος ειπον as I said at first. 
 λη'ιζόμενοι ζώσιν they live by plunder. 
 τ'ι μαθών, τι παθών ταντα εποιησας; cur haec fecisti? 
 ουκ εστίν άρχειν μη ΐίΰοντα μισθον one cannot rule if one 
 does not pay. 
 
 ι 2 
 
172 A BRIEF GREEK SYNTAX. 
 
 242. In this way the participle serves as a substitute for 
 the Latin gerund, as in 
 
 θρηνεΊν επωίας προς τομώντι πηματι to shriek charms 
 
 over a cutting wound, i.e. one that requires to he cut. 
 oruv τις ες πλέον πέση τον θέλοντος. 
 
 243. Participles tend to compact sentences together, and to 
 supersede that constant necessity for conjunctions which exists 
 in English, as 
 
 'Αλλ' άναστάντες καταψηφίσασθε But now rise and con- 
 demn me. 
 The sentences of the Greeks, it has been observed, were like 
 their earliest buildings, Cyclopean in structure, — dispensing, 
 as far as possible, with mortar. 
 
 244. "Εγων, φέρων, «γωι•,* λέγων, χρωμενος, άπιών, are 
 used where we use ' with,' as 
 
 ϊππον άγων ήλθεν, ζίφος φέρων προσηλασε, τέχνη χρώμενος 
 ενίκησεν. 
 "Έχων is sometimes colloquial and superfluous, as 
 
 τΐ ληρεΊς, φλυαρείς 'έχων ; why do you trifle so ? &c. 
 
 245. The uses of the genitive and accusative absolute (έμοϋ 
 διδάσκοντος while I am teaching, δέον it being my duty, &c.) 
 are explained under the heads of those cases. 
 
 246. Various adverbs are used to add distinctness to parti- 
 ciples, as 
 
 αμα φεύγοντες whilst flying. 
 
 μεταξύ δείπνων during dinner. 
 
 ευθύς ιδών on seeing (a person). 
 
 ατε παις ων inasmuch as he was a boy. 
 
 άχιύμενός περ though grieved. 
 
 καίπερ είδότες though knowing. 
 
 N.B. Notice the difference between such phrases as 
 
 κολακεύοντες άπατώσι they deceive by flattery, 
 and 
 
 οι κόλακεύοντες άπατώσι flatterers deceive ; 
 between 
 
 εποίησε βασιλεύων he did it during his reign, 
 and 
 
 ό βασιλεύων εποίησεν the reigning sovereign did it. 
 
 * "Ayeiv κα\ <pepeiv « to harry and carry,' the former of animate, the 
 latter of inanimate things. 
 
173 
 
 Verbals in -τέος. 
 
 247. Verbal adjectives are a kind of participles passive. 
 They are found in -τέος or -τός, and when derived from transi- 
 tive verbs may be used either 
 
 i. Personally, as 
 
 άσκητεα σοι εστίν η αρετή you must practise virtue ; 
 or 
 
 ii. Impersonally,* as 
 
 άσκητεον εστί σοι την άρετήν. 
 
 επώυμητεον εστί σοι της αρετής. 
 
 248. They are frequently used in the neuter plural, as 
 
 ους ου παραδοτέα τοΊς ' Άθηναίοις εστίν whom we must not 
 
 give up to the Athenians. 
 γυναικός ουδαμώς ησσητέα we must by no means be 
 
 worsted by a woman. 
 
 249. Verbal adjectives in -τος usually imply possibility ; 
 those in -τέος necessity ; as 
 
 λυτός one who is loosed, or able to be loosed ; λυτέος one 
 
 who is to be loosed. 
 ποιητόν what may be done ; ποιητέον what must be 
 
 done. 
 
 The Particle *Av with the Moods. 
 
 250. The very important particles av, and epic κέ, κά, are 
 supposed to be derived respectively from ανά and κατά, 
 * according fo,' and to be connected with the Latin an, and 
 quam. They always imply a verb and a condition^ but have 
 no exact equivalent in any language. Their chief use is to 
 articulate, analyse, give prominence or emphasis to the con- 
 ditionality of a notion. 
 
 * This resembles the use of the Latin participle in -dus, in such 
 phrases as 'pacemTrojanoaregepetendam,' Virg. 2E,n. xi. 230 {σΧτιγτέον 
 eipi]ut\v). Cf. Lucr. i. 111. Canes paucos et acres habendum. — Varro. 
 
 f The particles re, που, ίσως, αν express ascending degrees of uncer- 
 tainty; viz.: i. surely, ii. very likely, iii. possibly, iv. contingently, or 
 on certain conditions. The very existence of this unparalleled particle 
 shows how intensely the Greeks realised the conception of contingency, 
 and their general dislike to positive directness. On its derivation see 
 Pott, Etymolog. Forschungen, i. 420. In some of its usages (&v=iav) it 
 offers a curious fortuitous analogy to the now obsolete ' an,' which indeed 
 might often be used in rendering it, ' An,' and ' and,' in the sense of 
 
174 A BRIEF GREEK SYNTAX. 
 
 251. *Av is used with three moods, the indicative, optative, 
 and {when combined with other words) the subjunctive ; and 
 also with the infinitive and participle. But it is never found 
 with the imperative. 
 
 252. In the indicative, it is generally found with the im- 
 perfect (of continued acts), the aorist (of momentary acts), and 
 less frequently the pluperfect (of abiding results) ; but not 
 with the present and perfect, and very rarely (if ever) with 
 the future.* 
 
 253. Its potential meaning is always clear ; thus 
 απέθνησκεν he ivas dying ; 
 
 άπέθανεν he died; 
 έτεθνηκει he had died; 
 but 
 
 απέθνησκεν αν he would be, or have been, dying ; 
 άπέθανεν αν he would have died ; 
 έτεθνήκει αν he would have been dead ;f 
 
 i.e. in each case 'he would, if so and so had happened ;' and 
 
 1 if were once common, as ' an it please you,' ' an I should catch 
 you,' «fee. 
 
 ' What knowledge could we have of ancient things past, and historic 
 were not ? ' — Lord Berners, Preface to Eroissart. 
 
 ' To glut up, and you could, your wasting hunger.' — Sir John Cheeke. 
 
 See Craik, Engl, of ShaJcsp. p. 114; Abbott, ShaJcsp. Gram. p. 29. 
 
 * The best scholars (Hermann, Porson, <fec.) decide against av with 
 the future ; there is indeed no reason in the nature of things against such 
 an idiom (since what wiU be may be supposed to depend on conditions), 
 and fee is used freely with the future in Epic ; but as it is certain that a 
 people ' qui amant omnia dubitantius loqui,' would hare used this formula 
 if it had not grated against their sense of fitness, it is better to attribute 
 to carelessness or corrupt readings the few cases which do occur. 
 
 f The position of av is always nearest to the word which colours the 
 sentence. Sentences like ουκ ο?δ' αν εΐ πείσαιμι, Eur. Med. 941, Ale. 48, 
 vereor ut suadeam, 2" fear I shall not persuade, are mere instances of a 
 spurious hyperbaton, meaning ούκ-οΐδ'-εΐ = haud scio an, πε'ισαιμι-&ν, for 
 εαν in Attic is never resolved into el &V, and never takes the optative 
 (or the indicative), ουκ olb" αν εϊ δυναίμην=Ι fear I shall not be able 
 = <ροβουμαι-μ)) ου-οΰνωμαι. 
 
 It is true that in late Attic εαν is found with the optative (e.g. twice 
 in Lucian) ; in Thuc. iii. 44, the reading ήν re καϊ έχοντες τι συγγνώμη* 
 εϊεν is probably wrong, or else the expression is a mere solcecism, such 
 as is found even in the best writers. Thomas Magister lays down the 
 rule, ήν αε\ μετά των υποτακτικών πάρα to7s ακριβέστατοι, %ν is always 
 found with subjunctives in the most accurate writers ; and then alluding 
 to this passage of Thucydides as an exception, he adds αλλ' ου οε~ ζηλοίν 
 τύ «τ«| Ρηθ*ν isolated exceptions should not be imitated. 
 
USES OF αν. 175 
 
 αν always implies a protasis of this kind, even where such 
 protasis is not expressed. 
 
 τα γαρ τοιαύτα οΰτ εγίγνετο οΰτ άν εγ ενετό for such 
 things neither were taking place, nor could have taken 
 place (sc. on any conditions). 
 On αν with the imperfect see Mr. Jebb's Electra, 1. 323. 
 
 254. But, besides this potential usage, άν with the imperfect 
 is also used frequentative! y, to mean 'you did so as often as 
 such and such circumstances recurred ; ' and sometimes it can- 
 not be certainly known which of the two meanings is intended. 
 -»- ni;is ο,τι μάθοιμ εκάστοτε 
 
 έπεΧανθανόμην άν ευθύς υπό πλήθους ετών (Ar. Nub. 831) 
 but whatever I learnt on each several occasion, I used 
 to be forgetting directly in consequence of my old 
 age. 
 
 ως προτού 
 
 ουΰε\ς επρ'ιατ άν Βρεπανον ούΰε κολΧΰβου may be either 
 'since previously, no one used to be buying a sickle 
 even for a farthing,' or, ' no one would have been buy- 
 ing one,' i.e. if it had been for sale. 
 
 255. This double use of αν with the imperfect (potential 
 and frequentative) is closely paralleled by the English ' would,'' 
 which not only implies a condition, but also indefinite re- 
 currence j* as 
 
 ' Pleased with my admiration, and the fire 
 His words struck from me, the old man would shake 
 His years away,' &c. — Wordsworth. 
 
 256. In Epic κε is found both with the present and future 
 indicative ; but in Attic Greek, άν with these tenses is so ex- 
 tremely exceptional, that it must be regarded as due to mere 
 carelessness. 
 
 257. *Av becomes rarer in the New Testament and in later 
 Greek. 
 
 258. We have seen that the optative by itself has a potential 
 force ; and thus we find both 
 
 ποϊ τις φύγοι ; whither can one fly ? — Ar. Plut. 438 ; 
 and 
 
 ποϊ τις άν ώύγοι 'A whither could one fly ? — Eur. Or. 
 598. 
 
 * F. Whalley Harper On the Greek Tenses. 
 
 f In ποϊ tis <pvyrj • the subjunctive expresses a sort of hopeless delibe- 
 
176 A BRIEF GREEK SYNTAX. 
 
 But when the optative is potential in meaning, it is generally 
 accompanied by &v, as 
 
 τοϋτο γένοιτ av this might happen. 
 
 Hence it is used to soften the asperity i. of commands ; ii. of 
 inferences ; and iii. direct assertions ; as 
 
 i. χωροΊς αν ε'ισω you might go in=be so good as to 
 
 enter, 
 ii. ουκ άρα σωφροσύνη αν ε'ίη αιδώς it seems then that 
 
 sobriety and modesty could not be synonyms, 
 iii. υθλείς• άπερρ\ ού/ο αν ΙιΙαξαίμην σ ετι you talk non- 
 sense ; get away; I couldn't [=will not] teach 
 you any more, 
 συ μεν κομ'ιζοις αν σεαυτον rj θέλεις you then may 
 convey yourself where you like. — Soph. Ant. 444. 
 
 N.B. Expressions like the last being in form conditional 
 (though really polite imperatives), are negatived by oh, not . 
 by μή. 
 
 259. In negative sentences the omission of av with the optative makes 
 the negation stronger, by denying the potentiality absolutely and inde- 
 pendently of all conditions, as 
 
 rb yap 4p.<pv*s οΰτ' αϊθων αλώπηξ οϋτ' 4ρίβρομοι \4ovt€S διαλλάξαιντο 
 ήθος neither tawny fox nor loudly-roaring lions could change their 
 inborn nature. — Pind. 
 
 irws &v ; ris &v ; are used with the optative in wishes. 
 
 260. "A ν does not properly go with the subjunctive •* but it 
 often qualifies ει, ος, οίος, πριν, εως, &α, and often coalesces 
 with some other particle, as in εάν, όταν, έπειΰάν, &c; and 
 these combinations always take the subjunctive. In such 
 cases therefore av does not belong to the verb, but modifies the 
 particle* or relative ; thus ος who ; ος av whoever ; ίνα where ; 
 Ίνα av wheresoever ; οτε when ; όταν whensoever ; πριν ere ; 
 7rp\v av or ever, &c. 
 
 ration, ' whither is one to fly ?' N.B. You can say ττοΐ ris φνγτι ; because 
 this is equivalent to ττοΐ φύγω; whither am I to fly? but you cannot 
 say rat (pvyr) without the ris. 
 
 * As Hermann briefly states it, ' you cannot say \4yrj &v ; and in 
 phrases like is av λ4γρ, 'όταν λ4γρ, 4av \4yrj, Sec, the particle modifies, 
 not the verb but, the preceding relative. Not &v therefore, but its com- 
 bination with the preceding word, is correctly said to be construed with 
 the subjunctive ; for is av \4yy gives a meaning, and so does is Av who- 
 ever, but av λ4γρ combines into no meaning at all. Hence we always find 
 is av \4yrj, never is \4yrj &v.' The rule for beginners, says Dr. Donaldson, 
 is ' Relativa et particulce relative cum fa subjunctivum exigunt* 
 
USES OF αν. Ill 
 
 ος ποιεϊ he who does ; ος av voirj whosoever may do. 
 ους εΐΰεν those whom he saw ; ους αν 'ίΰη whomsoever he 
 
 ίνα where ; Ίνα αν wheresoever ; as πατρ\ς yap εστί πάσ ιν 
 αν π parry τις εν for every land is one's country where- 
 soever one fares well (ίνα αν aivra,js=ubicunque). 
 
 261. We get therefore this rule : Whenever an indefinite 
 sense is not required for ος, όστις, οτε, επεί, &c, the optative 
 is almost always used ; when an indefinite sense is required, 
 they are combined ivith av and followed by the subjunctive* 
 
 262. If however any such combination of a conjunction 
 with av is found in the same clause with the optative, the av 
 then belongs to the verb and not to the conjunction, as 
 
 έσθήτα oY fjv αν μάλιστα η ώρα είαλάμποι. dress such as 
 through it her beauty might best shine (αν-οΊα\άμποι)] 
 
 but if it had been (ϋιαλάμπτ) it would mean through whatever 
 dress (cY fjv av) her beauty may best shine. 
 
 So too 
 
 ουκ-εχω-οπως αν-άπιστοίην I know not how I-COllld- 
 possibly- disbelieve. 
 
 τις δ' ούτως ανους 
 ος νμέ κα-πρίαιτο. — Ar. Ach. 720. 
 
 Ν.Β. Compare 
 
 όσους ε'ιΰεν as many as he saw (on some past occasion). 
 ΰσονς 'ίοοι as many as he saw (i.e. ' from time to time') 
 
 (the optative being iterative = happened to see). 
 όσους av 'ihj as many as ever he sees. 
 
 263. *Av with the infinitive f and participle gives them a 
 potential or hypothetic meaning ; J as 
 
 * We hare already seen that os, 8re, «, &c, may be joined with the 
 subjunctive without av in those very rare cases in which it is intended to 
 exclude all notion of any possible condition. 
 
 f In Latin we cannot express the distinction between the aorist and 
 the present ; so that we get 
 
 ypa<peiv av = scripturum esse Ί , , ., _. , 
 
 yeypa<p4vai au = scripturum fuisse / =^ α Ψ αι »"•— Clyde. 
 
 \ In Thuc. iv. 24, we have rots *Αθηναίοΐ5 re ουκ av eivat 4φορμέΐν καΐ 
 του πορθμού κρατέΐν * In that case they thought that it would be im- 
 possible for the Athenians to lie at anchor there, and that they them- 
 selves would remain masters of the strait,' where the av with tlvat 
 implies th;\,t that result is slightly less probable than the other. 
 
 ι 3 
 
178 λ BRIEF GREEK SYNTAX. 
 
 Κϋρος ει εβίωσεν άριστος αν ΰοκεΐ άρχων γενέσθαι Cyrus, 
 had he lived, ivould I think have been a consummate 
 general (=οΤμαι on αν έγ ενετό). 
 
 δυνηθείς αν αυτός 'έχειν άπέΰωκεν though he might have 
 kept it, he gave it back (= έΰυνήθη άν). 
 
 264. Practically it is not used with the future infinitive or 
 participle. The few apparent cases in which this occurs are 
 so rare, that they must be due to carelessness. 
 
 265. Just as 
 
 ταϋτ αν εγίγνετο = these things would be taking place, or 
 would have been taking place ; 
 so 
 
 εφη ταϋτ αν γίγνεσθαι = he said that these things would 
 be, or would have been taking place. 
 
 And as 
 
 ταϋτ αν έγένετο = these things would have taken place ; 
 so 
 
 εφη ταϋτ αν γενέσθαι = he said that these things would 
 have taken place. 
 
 266. With the participles we have 
 
 τα γιγνόμενα the things which are taking place; τάαν 
 γιγνύμενα the things which would be (or, would have 
 been) taking place. 
 
 τα γενόμενα the things which took place ; ret αν γενόμενα 
 the things which would have taken place. 
 
 267. Demosthenes often uses the phrase 
 
 πολλά δ' αν έχων ειπείν though I should have plenty to 
 say, &c. 
 
 N.B. i. The verb belonging to av is often omitted, as in 
 Plato's phrases 
 
 πώς γαρ άν ; πώς ουκ άν ; 
 and in 
 
 τάχ άν, ώσπερ αν ει. 
 
 οι δ' οϊκέται ρέγκουσιν άλλ' ουκ αν προ τοϋ and the servants 
 are snoring, but they would not have been heretofore. 
 φέρε τι Ζητ άν ; come then what would you have done ? 
 
 ii. On the other hand av itself is sometimes omitted where 
 it can be easily understood, and this is the usual way of 
 explaining such phrases as 
 
THE FINAL CONJUNCTIONS. 179 
 
 πείθοι αν ει πε'ώοι, άπειθοίης δ' 'ίσως obey if thou wouldst 
 obey ; perhaps thou wouldst disobey, iEsch. Ag. 1049 
 (where however, as we have already seen, § 177, 2, άπει- 
 θοίης may be potential without av being understood). 
 
 iii. av is sometimes repeated with the optative, partly 
 for rhetorical effect,* and partly to emphasise two words in 
 the same conditional sentence, of which one is often the nega- 
 tive; as 
 
 φθάνοις 3' αν ουκ αν τοΊσΰε συγκρύπτων δέμας you could 
 not possibly be too soon in clothing your person with, 
 these arms (i.e. do it with all speed). 
 
 τω γαρ αν και μείζονι 
 Χεζαιμ αν η σοι ; for to whom in the world even greater 
 than thyself could I possibly say it ? — Soph. Ο, T. 772. 
 
 ουκ αν γενοίμην Ηρακλής αν I shouldn't at all like to be 
 Hercules (ich mag nicht etiva Hercules werden). 
 
 iv. αν is sometimes misplaced, by hyperbaton, as in 
 
 ουκ οίδ' ΐιν εΐ πείσαιμι I think it doubtful whether {ουκ 
 oW εΐ hand scio an) I could persuade {πείσαιμ av). 
 
 v. av as a conjunction means if= εάν, ijv, as is often the case 
 in Plato (but not in the poets). It may be distinguished from 
 the particle av by its standing first in the sentence, which the 
 particle av never does. This usage of av closely resembles 
 the obsolete English ' cm,' as 
 
 άι> θεός εθέλτι an God will. 
 
 vi. av may sometimes be rendered ' otherwise ' (pointing to 
 a suppressed clause), as 
 
 επιστευόμην υπο των ΑακεΙαιμονίων' ου γαρ αν με επεμ- 
 7τον I was trusted by the Lacedaemonians, otherwise 
 they would not have sent me. 
 
 THE FINAL CONJUNCTIONS. 
 
 268. Final Conjunctions are those which express an end or 
 purpose, viz. ως, ωπως, 'ίνα, and in Epic οφρα. 
 
 We have already seen that after primary tenses they 
 regularly take the subjunctive (where we use may), and after 
 historical tenses the optative {might). 
 
 * The first av is called by the grammarians ΰυνητιιών ' effective,' and 
 the second παραπληρωματικά 'complementary.' 
 
180 A BRIEF GREEK SYNTAX. 
 
 269. When this rule is violated, it is from a desire to be 
 graphic (προ ομμάτων ποιέΊν) ; as in the following sentence of 
 Lysias (de Cade Eratosth. ix. 2) : 
 
 επειδή δε το παιΰίον εγένετο ημ~ιν, η μήτηρ αυτό εθήλαζεν, 
 Ίνα δε μη, οπότε λουεσθαι ΰέοι, κινδυνευη κατά της 
 κλίμακυς καταβαίνουσα, εγώ μεν άνω Ζιητώμην, αί δε 
 γυναίκες κάτω . . . μετά δε το ΰεϊπνον το παώίον εβόα 
 και εΰυσκόλαινεν υπυ της θεραπαίνης επίτηδες λυπούμε- 
 νον Ίνα ταϋτα ποιγ . . . but when our boy was born, 
 the mother used to nurse it. But that she may not 
 run a risk by descending down the stairs whenever it 
 wanted washing, I used to live upstairs, and the women 
 below. And after dinner the child used to cry and 
 fret, being pinched on purpose by the nurse that he 
 may be doing so, &c. 
 
 It will here be seen at once that κινΰυνεύοι ' might run no 
 risk,' and ποωΊ might do so, would have been the regular con- 
 structions; and that the subjunctives are only dramatically 
 substituted for them, to represent the events as going on 
 before the hearer's eyes. 
 
 270. On similar principles όπως is constantly joined with 
 the future indicative ; * as 
 
 δε'δοιχ' 07τως μοι μή λίαν φανεί σοφή I fear that you will 
 
 seem too wise to me (cf. the vulgar English * I fear as 
 
 how'). 
 
 και το μεν καλώς έχον 
 όπως χρονίζον ευ μενεΊ βουλευτέον (jEsch. Ag. 846) and 
 
 we must take measures whereby all which now is well, 
 
 shall long continue so. 
 άλλ' όπως μή V τοΊς τρίβωσιν εγκάθηνταί που λίθοι see 
 
 that there are not stones lying anywhere in your cloaks. 
 
 — Ar. Ach. 343. 
 
 271. όπως with the future constantly means ' see that,' • take 
 care that,' ' I fear that,' &c. 
 
 όπως μή σαυτον οικτιείς ποτέ take care that you will not 
 have some day to pity yourself. — JEsch. P. V. 68. 
 
 νυν ουν όπως σώσεις μ' επει κάπώλεσας now then see that 
 you save me, since you too destroyed me. — Ar. Nub. 
 1177. 
 
 * This is less frequently the case with ίνα ; and when it is, ίνα may 
 always have its quasi-locaj meaning of where = in which case. 
 
THE FINAL CONJUNCTIONS. 181 
 
 272. With the past tenses of the indicative ώς, όπως, ινα 
 imply that something has not occurred, — -an impossible or un- 
 fulfilled result. It is often rendered c in which case,' but such 
 a rendering is unnecessary, and in the third of the following 
 examples would have required ουποτε not μήποτε. 
 
 ουκοΰν εχρήν σε Πήγασου ζεΰζαι πτερόν, 
 όπως επαίνου τοίς θεοΊς τραγικώτερος. — Αχ. Pax, 135. 
 Ought you not to have, &c, that you might have appeared 
 to the gods more tragic-looking ? 
 
 εΐ της άκουυύσης ετ ή ν 
 
 πηγής δι' ώτων φραγμός, ουκ αν εσγόμην 
 
 το μη 'ποκλ&σαι τουμόν άθλων δέμας, 
 
 iV ήν τυφλός τε και κλύων μηΖεν. — Soph. Ο. Τ. 1386. 
 
 If there had been any further means of stopping the 
 fount of hearing through the ears, I would not have 
 abstained from closing up my wretched frame, that I 
 might have been both blind and deaf. 
 
 τι μ? ου λαβών 
 'ίκτεινας ευθύς, ως εδει£α μήποτε 
 
 Ιμαυτόν άνθρώποισιν ένθεν ην γεγως ; — Soph. Ο. Τ. 1393. 
 
 Why didst thou not take, and slay me at once, that I 
 
 might ne'er have shown to men whence I was sprung ? 
 
 273. We may thus briefly sum up the uses of ώς, όπως, 
 ϊνα : 
 
 I. wc=as; [£c=thus; except when ώς follows the word 
 which it compares, as πατήρ ώς like a father.] 
 
 ώς is the adverb of δς η ο ; when a>s=as, ώς ay means * in 
 whatever way.' 
 
 a. It is used with superlatives, as 
 
 ώς τάχιστα quam celerrime as quickly as possible. 
 
 b. Like the Latin ut, ώς sometimes means when. 
 
 c. It is sometimes used declaratively for on quod when 
 
 we intend to express an assertion rather than a fact. 
 
 d. ώς as a final conjunction = in order that', ώς av* in 
 
 order that perhaps ; the former used, as we have 
 
 * In one or two instances only, &s av appears to mean ' so long as ; ' 
 e.g. Soph. Aj. 1096, 
 
 rod Se σου ψόφου 
 ουκ &v στραψείην ws &v γ -s olos περ el but I will not swerve because 
 of thy clamour, so long as thou art what thou art. (Comp. 
 Eur. Ion, 77, Hcc. 330 ) 
 
182 A BEIEF GREEK SYNTAX. 
 
 seen, when the result is certain ; the latter when 
 less certain (but only in poetry ; ώ ς αν is never used 
 of a purpose in Attic prose). 
 
 II. α. όπως how stands to πώς in the same relation as ΰστις 
 to τι;, &c, as has been already explained. 
 
 N. ecu πώς• Δ. ό'ττως ; Ν. How then? D. How quothal 
 πώς ; how ? ουκ oil•' όπως I don't know how. 
 
 When O7rwg=how, όπως av=howsoever ; as 
 
 αξιών αυτω τε εζέιναι ΰιαλέγεσβαι όπως βούλεται, κα'ι σοι 
 όπως αν αν συ βου\τ} claiming the right for himself to 
 discourse how he likes, and for you too however you 
 like.— Plat. Prot. 336 b. 
 
 b. Like the English hoiv, όπως comes to mean that, and in 
 many sentences either translation may be used.* 
 
 c. When 6πως=ίη order that, όπως av=in order that 
 perhaps. 
 
 III. a. ίνα— where ; as 
 
 ουχ οράς tv' εΐ κακοϋ ; see you not in what evil plight (lit. 
 where of evil) you are ? 
 ίνα ai'=wheresoever (sicufa', ubicunque). — Soph. (Ed. Col. 
 189. 
 
 b. As a final conjunction, iva=ivhereby, i.e. in order that. 
 But in this meaning it differs from ώς, όπως in two respects : 
 i. It is never combined with av. 
 ii. It is never found with the future indicative. 
 
 THE NEGATIVES. 
 
 274. The Greek language has two classes of negatives, ου 
 and its compounds ουδέ, ούτε, ουδείς, ουδαμώς, &c. ; μή"\ with 
 its compounds μηδέ, μήτε, μηδείς, μηδαμώς, &c. The differences 
 between them are simple and definite. 
 
 * 'How' and ' that' are interchanged throughout the whole of Cole- 
 ridge's beautiful poem of Genevieve ; and Johnson quotes as an instance 
 of this sense the following sentence, ' Thick clouds put us in some hope 
 of land, knowing how that part of the South Sea was utterly unknown,' 
 &c. — Bacon. [Harper, p. 117•] 
 
 f Naturally the subjective negation μη is too refined and luxurious for 
 some dialects of Modern Greek ; accordingly in Tzaconian we find only 
 the negatives 5ev ( = ούΒ4ν), and 6 ( = oi>). See Suidas, s.v. φιλόξενος, 
 Athen. Deipnos. xi. v. p. 466 ; Farrar, Chapters on Language, p. 91. 
 
THE NEGATIVES. 183 
 
 275. The main distinctions between ov and μή are as 
 follows : £ oh negat, μή vetat ; ov negat rem, μη conceptionem 
 quoque rei.' — Herm. In fact, as Madvig observes, ob is always 
 used when some specific rule does not require the use of μη. 
 
 i. ov denies, as 
 
 ovk εστί ταντα it is not SO. 
 μη forbids, as 
 
 μη κλέπτε do not steal. 
 
 ii. ov is objective and categorical, i.e. it negatives facts and 
 certainties. 
 
 μη is subjective and hypothetical, i.e. it negatives concep- 
 tions, thoughts, &c. 
 
 iii. ov is the negation of the judgment ; μη of wishes and 
 suppositions. 
 
 ov . . . ; expects the answer Yes; as άρα ov',=nonnel ov 
 μενέΐς ; quin manes 1 Won't you stop ?=stop ! 
 
 μη . . . ; expects the answer No ; άρα μή=μών ; {μη ovv)= 
 num ? μη τέθνηκεν 6 πατήρ ; I hope my father is not dead, num 
 mortuus est pater ? 
 
 Μ if.* 
 
 276. Mj7 is used 
 
 i. With the hypothetical participle, as 
 
 μή ΰρών if he does not do it. 
 ii. After εϊ, εάν, επειίάν, όταν, as 
 
 εί μή λέγεις unless you say. 
 iii. After final particles, 'ίνα, όπως, &c, as 
 
 παρακαλεί ιατρόν, όπως μή άποθάνη summon a physician 
 that he may not die. 
 
 iv. After all hypothetical, indefinite, or causal relatives, ος 
 άν, οποίος αν, &C. 
 
 v. In all wishes, as 
 
 μη γένοιτο God forbid ! 
 vi. In all prohibitions, as 
 
 μή κλέφης τοντο do not steal this. 
 
 Μηΰεις άγεωμέτρητος ει σ/τω let no one untrained in 
 geometry enter. 
 
 * In Hebrew ?X α£=μή, Ν? Ιο = ου. 
 
184 A BRIEF GEEEK SYNTAX. 
 
 vii. With the hortative and deliberative subjunctive, as 
 
 μή γράψωμεν let tlS not write. 
 
 μή άποκρίνωμαι ; am I not to answer you ? 
 
 viii. With the infinitive* (except after verbs declarandi et 
 sentiendi, because then the infmitive=the indicative with οτΐ)> 
 as 
 
 συ\ το μή σιγήσαι λοιπόν 7]v it remained for you not to be 
 silent. 
 
 ix. With questions which expect the answer no ; as 
 
 μή αρχιτέκτων βούλει γενέσθαι you don't want to become 
 an architect, do you ? 
 
 Hence μών,—μή olv ]=num? 
 
 It will be seen at once that every one of these uses of μή 
 springs from its character as a subjective or hypothetical 
 negative. 
 
 277. An apparently superfluous μή is found after verbs 
 which involve a negative notion, e.g. verbs of refusing, fear- 
 ing, f doubting, denying, hindering, &c, as 
 
 * &are when followed by the indicative requires ου, when by the infi- 
 nitive μή. Thus 
 
 outus &φρων η" ν Scrre j adeo stultus fuit ut I he was so foolish that 
 ουκ τ]βού\€το I noluerit, he did not wish 
 
 (expressing the fact). 
 
 outws άφρων fy c&trre I adeo stultus fuit ut 1 he was so foolish as 
 μη βούλεσθαι nollet, not to wish 
 
 (expressing the natural consequence). 
 
 The former construction is the more oratorical and picturesque. 
 
 Sometimes, when the negative belongs to a single word, ου with the in- 
 finitive follows &στ€, and sometimes by an apparent irregularity as in 
 Soph. El. 783. See Shilleto on Bern, de F. Leg. App. c. 
 
 t φοβούμαι μη =forsitan, ουκ alb" u = haud scio an, which signifies less 
 probability. Notice the distinction between the following, 
 
 δέδοικα μη iroifjs vereor ne facias, I fear that you maybe doing it. 
 
 — iroifotis — facturus sis, I fear that you will do it. 
 
 But for δίδοικα μη iroius, inoUis, 4ποίησα$, πζποίηκας I fear you are, were 
 doing, did, or have done it (where no doubt is expressed, and the δέδοικα 
 is merely due to courtesy), there is no exact Latin equivalent, since in 
 Latin the subjunctive must be used. See Shilleto, Demosth. de F. Leg. 
 App. a. Hearing a person soliloquise on the spelling of a word I might 
 eay δβ'δ. μη αμαρτάνηε, but if I saw him beginning to spell it wrong, I 
 should say δεδ. μη άμαρτάνζίς. — Jebb's Electro, 1. 581. 
 
THE NEGATIVES. 185 
 
 φοβούμαι μή αμφοτέρων ημαρτήκαμεν I fear we have missed 
 
 both. 
 ήναντιωβην αυτω μηΰεν ποιέϊν παρά τους νόμους j'empechai 
 
 qu'il ne fit rien contre les lois. 
 ουκ αν εζαρνος γενοιο μη ουκ εμος νιος είναι tu ne nieras 
 
 pas que tu ne sois mon fils. 
 μή λαβείν εζαρνούμενος denying that he received them. 
 
 278. In all these instances the μή is merely a repetition of 
 the negative implied in the verb ; e.g. 
 
 ήρνοϋντο μή πεπτωκέναι they made a denial to the effect 
 that ' they had not fallen.' 
 
 After verbs of fearing and considering /xj) = to, as 
 
 Ι'εΙοικα μή θάνη vereor ne moriatur, I fear lest he die, i.e. 
 that he will die. 
 
 This pleonastic negative is common in modern languages, 
 e.g. 
 In English : 
 
 1 First, he denied you had in him no right.' — Shakspeare, 
 
 Comedy of Errors, iv. 2.7. 
 ' If any of you know . . . just impediment why these two 
 
 should not be joined together.' — Prayer-book. 
 1 Can any man forbid water that these should not be 
 baptised . . . ? — Acts x. 47. 
 In French : 
 
 Je crains que sa maladie ne soit mortelle, I fear his disease 
 is fatal. 
 In Italian : 
 
 Guardarsi di noil credere, be on your guard against 
 believing. 
 In Spanish : 
 
 Temia no entrara, I feared he might come in. 
 
 Por poco no me caigo, haud multum abfuit quin caderem. 
 
 Ob. 
 
 279. ov is the proper negation of the indicative, and of all 
 forms that can be directly resolved into the indicative ; e.g. 
 in Homer of the subjunctive, where it scarcely differs from a 
 future (see § 176) ; of the optative in oratio obliqua (after 
 ore and ώς), where it merely represents the indicative of the 
 oratio recta ; and of the optative with άν, which is merely a 
 milder future or imperative. 
 
186 A BRIEF GREEK SYNTAX. 
 
 280. ov has a property, not possessed by μί], of coalescing 
 with single words, like the privative α ; as 
 
 τα ου καλά inJionesta ; οΰχ ηκιστα decidedly ; ου φημι 
 nego ; ούχ υπισχνοΰμαι I refuse ; ου στέργω I hate. 
 Hence such sentences as 
 
 εΐ τους βάνοντας ουκ ερς θάπτειν if you prevent the burial 
 of the dead, 
 or 
 
 εΐ Η τοι ου ΰώσει if he shall refuse it to you, 
 
 are no violations of the rule that μη should be used after con- 
 ditionals, because ουκ tw^veto, ου δώσω— recusabo ; and so of 
 all similar cases. Such expressions are due to the figure of 
 speech called litotes, by which less is said than is meant ; e.g. 
 ' Shall I praise you for these things ? I praise you not ' 
 =1 do anything but praise you.* 
 
 281. The same thing sometimes occurs where εΙ=οτι after 
 verbs of disapprobation, &c, an indirect form due to Attic 
 politeness ; as 
 
 θαυμάζω εΐ ταύτα ου ποιεϊς I wonder that you do not act 
 thus ; 
 
 but here μη is more usual [see Jelf, 804, 8]. 
 
 282. Similarly verbs declarandi et sentiendi may be followed 
 by ου with the infinitive, as 
 
 ομολογώ ου κατά Μέλητον και" Ανυτον είναι ρητωρ I confess 
 that I am not an orator after the fashion of Meletus and 
 Anytus. 
 
 283. ob is redundant after fj than generally in negative 
 sentences, as 
 
 πόλιν ολην ΰιαφθείραι μάλλον η ου τους αιτίους (Thuc. iii. 
 36) to destroy a whole city rather than the guilty ; 
 
 so in French : 
 
 On meprise ceux qui parlent autrement qu'ils ne pensent. 
 II n'ecrit pas mieux cette annee-ci qu'il ne faisait l'annee 
 passee.— Jelf, § 749, 3. 
 
 284. A few contrasted and mixed instances of oh and μη 
 will illustrate the principles here laid down, which are 
 sufficient to meet every case which occurs in good Greek. 
 
 * This is a common idiom in Hebrew with N? = ' anything but.' See 
 Hos. i. 9 : Ps. i. 4. 
 
THE NEGATIVES. 187 
 
 εΐ μη ταΰτά εστί, οι/δε τάδε (Plat. Phced. 76, ε) if that is 
 
 not true, neither is this. 
 μη θνήσχ υπέρ rovd' ανδρός, ουϋ" εγώ προ σου (Eur. Ale. 
 
 690) die not on my behalf, nor {will I die) for thee. 
 εγώ δ' όπως συ μη λέγεις ορθώς τάδε 
 
 ουκ αν Ζυναίμην μητ έπισταίμην λέγειν (Soph. Ant. 682) 
 but I could not say, and may I never know how to 
 
 say, that you are not right in what you say. 
 
 [μη λέγεις because it follows the indefinite relative όπως ; ουκ 
 αν Ζυναίμην because αν ΰυναίμην is a mild future ; μην επισταί- 
 μην because this is a wish.] 
 
 b πιστεύων εις αΰτον ου κρίνεται, 6 δε μη πιστεύων ηΰη 
 κέκριται, οτι μη πεπίστευκεν κ.τ.λ. (John iii. 18) he that 
 believeth on him is not condemned, but if any one 
 believeth not he has been condemned already, because 
 he hath not believed, &c. 
 
 [ού κρίνεται is a fact; 6 μη πιστεύων is an hypothesis=if any 
 one does not; οτι μη because this depends on the former 
 hypothesis.] 
 
 εζεστι κήνσον ΰοϋναι η ού; Ιώμεν η μη Ζώμην \ (Markxii. 
 14) is it lawful to give tribute, or (is it) not ? [direct 
 question with ου,] are we to give, or are we not to 
 give? [deliberative subjunctive with μη.'] 
 
 ουκ εστίν εν το~ις μη καλο"ίς βουλεύμασιν 
 
 ονδ' έλπ'ις.— Soph. Tr. 727. 
 
 there is not even hope in any plans if they be not 
 honourable. 
 
 b ου πιστεύων is qui non credit. 
 
 b μη πιστεύων si quis non credat. 
 
 b αληθής τα μη 'όντα ως ουκ οντά λέγει he who is true re- 
 presents whatever things are not \_μή = an indefinite con- 
 ception] as not-being (or as non-entities). 
 
 η ουκ εμπειρία the actual want of experience. 
 
 η μη εμπειρία want of experience if, or wherever it may 
 
 exist. 
 το ουκ αγαθόν that which is bad ; το μη αγαθόν whatever 
 
 may not be good. 
 
 ΐ>ς ου ποιεί ταΰτα qui non facit haec. 
 
 ος μη ποιεί ταΰτα qui haec non faciat, or si quis, &c. 
 
 α ουκ olca certain things which I do not know; α μη 
 olla whatever things I may not happen to know. 
 
188 A BRIEF GREEK SYNTAX. 
 
 προσπασσαλευσω . . . 'ίν ούτε φωνην ούτε του μορφή ν βροτώι 
 
 οφει. —Msch. Prom. 20. 
 I will nail thee to a spot where thou shalt never see, &c. 
 
 (of a definite place). 
 
 μέλλουσι yap σ εΐ τώνΰε μη λήξεις γόων 
 
 ενταύθα πέμψειν ένθα μη ποθ' ηλίου 
 
 φάος προσάψει. — Soph. Elect. 379. 
 
 for they are about to send thee, unless thou wilt cease 
 from these complaints, to some (unknown) region 
 where thou shalt never gaze on the sun's light. 
 
 ούτοι φίλα ra μη φίλ\ ω κόραι (Eur. Troad. 468) truly 
 things are not acts of friendship, if they he not pleasant, 
 maidens. 
 
 εζεστι yap μοι μη λέγειν α μη τελώ (iEsch. Eum. 859) 
 for it rests with me not to mention anything ivhich I 
 shall not carry out. 
 
 α μή φρονώ γαρ ουποτ* άζιώ λέγειν I never think fit to 
 speak anything which I do not think (ά ου φρονώ would 
 be any definite things). 
 
 285. Ov and μη are frequently combined in the same sen- 
 tence, as in the following examples : 
 
 ου σίγα', μηΰεν τώι -V ερείς κατά πτόλιν silence ! mention 
 none of these things throughout the city. — JEsch. 
 Sept. c. Theh. 250. 
 
 oh σίγ* ανεζει, μηΐε ΰείλίαν άρείς ; keep silent, and as- 
 sume not cowardice ! — Soph. Aj. 75. 
 
 ohy\ συγκλείσεις στόμα, 
 και μη μεθήσεις αύθις αίσχίστους λόγους ; 
 close thy mouth, and utter not again most disgraceful 
 words !* — Eur. Hipp. 498. 
 
 άλλ' εϊσιθ 1 ' ου σοι μι) μεθέψομαί ποτέ but enter ; I shall 
 certainly never follow after you. — Soph. El. 1052. 
 
 * Of the two very difficult lines — 
 
 iyu> δ' ου μη ποτ* 
 τ&μ &s αν είπα? μ^ τά <? 4κφ•ηνω κακά, Soph. Ο. Τ. 329, 
 
 one can only say ' Quot viri tot sententise.' Donaldson supposes that 
 μ% is repeated before the verb, because the ου μί) is separated from it. It 
 would then mean ' Never will I, for the sake of uttering my own pre- 
 dictions, never will I reveal thy woes.' — New Crat. p. 587. 
 
MULTIPLIED NEGATIVES. 189 
 
 These passages are usually and simply explained by under- 
 standing the ov before the following μη in the manner illus- 
 trated in § 290 infra. Some scholars however put the inter- 
 rogation after each clause of the sentence, and maintain that 
 μη with the future is admissible in prohibitions. We believe 
 that in point of theory this is correct, although the actual 
 instances are so few, that the idiom must never be imitated.* 
 
 286. Two negatives only destroy each other when they 
 belong to different predicates, as 
 
 ουδείς όστις ου γελάσεται there is no one who will not 
 laugh, i.e. every one will ; 
 
 otherwise they only strengthen the negation. In fact it may 
 be laid down as a rule that all men have a tendency to 
 strengthen negation by adding negative words to each accessory 
 of the sentence ; f as 
 
 μήποτ άσεβες μηδέν μηδέ ανόσιοι' μήτε ποιήσητε μήτε 
 βονλενσητε neither do, nor plan anything either im- 
 pious or unholy. — Xen. Cyr. viii. vii. 22. 
 ov ουκ 7jv ούΐίπω ουδείς κείμενος wherein never man had 
 
 yet been laid. — Luke xxiii. 52. 
 ακούει δ 1 ούδεν ουδείς οΰδένος no one obeys any one in 
 anything. 
 
 * Mij νυν μοι ν^μΐσ-^σ^τ 'Ολύμπια ΰώματ' (ίχοντςς. — II. xiv. 
 καί τά /uct τβυχτ; μητ' α-γωνάρχαι rives 
 θήσουσ* Άχα<οϊί.— Soph. Aj. 572. Cf. Ant. 84. 
 
 The other instance sometimes quoted (Eur. Med. 882, \£}>(is hs μτ$4ν, 
 κ.τ.λ.) is perhaps not to the point; but Elmsley's attempt to change as 
 many of such instances as possible into subjunctives, was one of those 
 premature applications of a priori reasoning which have done so much 
 to injure scholarship. Dawes' restriction of the use of ου μη with the 
 subjunctive to the second aorist only is another instance. 
 
 f ■ No sonne were he never so old of yeares might not marry.' — 
 Ascham, Scholemaster. ' Not nohow,' said the landlord, thinking that 
 where negatives were good, the more you heard of them the better. — 
 Felix Holt, ii. 198. Whatever may be said of the genius of the English 
 language, yet no one could have misunderstood the query of the London 
 citizen, ' Has nobody seen nothing of never a hat not their own ? ' The 
 addition of words like ypv in Greek, hilum in Latin (ne hilum, nihil), 
 pas and point in French, jamas and nada in Spanish, &c. is due to the 
 same tendency. ' And carfed not for God or man a point.' — Spenser, 
 F. Q. ii. 12. 
 
 Two negatives are often found in Hebrew also (1 K. x. 21 ; Zeph. ii. 2 ; 
 Is. v. 9, 'without no inhabitant,' &c). So we have ovdk πολλού 5c? 
 minime gentium, far from it, after negatives. 
 
190 A BRIEF GREEK SYNTAX. 
 
 287. Old German and Old English both agreed with Greek 
 in this idiom, and have only lost it from the influence of 
 Latin ; * thus we find in Chaucer — 
 
 ' He never yet no vilanie ne sayde 
 In all his life unto no manner wighte.' 
 ' His horse was good, but he ne was not gaie.' 
 ' There ne was none him like,' &c. 
 And even in Queen Elizabeth's time the idiom prevailed, for 
 we find her writing to King James, 
 
 * If I had meant it, I would never lay it on others' 
 shoulders, no more will I not damnify myself that 
 thought it not. 1 
 
 And, in the same letter — 
 
 ' but as not to disguise fits not the mind of a king.' 
 
 The latter instance is illogical though the meaning is clear ; 
 it shows how prevalent was the use of the double negative. 
 
 Hence Dr. Clyde correctly observes that ' I don't know 
 nothing'' is simply the relic of a once classical idiom ; and this 
 is true, it may be added, of many vulgarisms and colloquial 
 forms of speech. They are frequently relics of the old infantine 
 pleonastic condition of all languages at their commencement. 
 
 Hickes says that before the Conquest we often find as many 
 as four negatives combined : 
 
 i He is fre of hors that ner nade non' (=never had one). 
 — Hendyng's Proverbs (circ. 1300). 
 
 288. The first of two negatives is sometimes omitted ; as 
 Πάρις οντε πόλις neither Paris nor the city. — iEsch. Ag. 
 
 514. 
 λέγουσα μηΗ ϊρώσα. — Eur. Hec. 374. 
 
 * In Latin however the rule is sometimes broken; e.g. Nulla nee 
 exustas habitant aniinalia terras. — Tib. iv. i. 164. Absenti nemo ne 
 nocuisse velit ( = ne quis). — Prop. n. xix. 32. Cf. Luc. n. xix. 32, &c. 
 The Eomance languages hare not imitated the pedantic purism of Latin 
 in this matter. Thus in Latin nonnullus = someone, non nemo = some- 
 body; but in Italian ' Non dice nulla,' 'non v'e niuno,' are negatives. 
 So in Provenqal, ' Nuls horn non pot ben chantar sens amar ' is ' no man 
 can sing well without loving.' — Sir Gr. C. Lewis, Romance Languages, 
 p. 238. So in Spanish no lo sabe nadie nobody knows it ; no lo he visto 
 jamas I have never seen it. In fact in Latin the colloquial instinct was 
 often too strong for grammatical nicety. Thus in Plautus, Mil. Glor. v. 
 v. 18, we find ' Jura te non nociturum esse hominidiQ hac re neminij and 
 even Cicero has ( Verr. ii. 57) ' Non mihi praetermittendum videtur ne 
 illud quidem genus,' &c. See Jani, A. P. p. 236. 
 
Ου μή. 191 
 
 As in Milton — 
 
 ' Fearing God nor man ;' 
 and Shakspeare — 
 
 1 Tongue nor heart cannot conceive nor name thee.'— 
 Macb. ii. 3 ; 
 
 and in Carew — 
 
 ' Give Lucinda pearl nor stone ; ' 
 
 ' Gums nor spice bring from the East ; ' 
 
 and in Gifford — 
 
 ' Pallas nor Licinus had my estate.' 
 So too in Latin — 
 
 ' Qua fornace graves, qua non incude catena? V — Juv. 
 
 Ov μη. 
 
 289. i. ov μη -with the 2nd person of the future, is a strong 
 prohibition] as 
 
 ov μη ποιήσεις ; do not do it ! 
 
 ii. ov μη with the aorist subjunctive and with other persona 
 of the future, is a strong negation ; as 
 
 ov μη ποίησης you certainly shall not do it. 
 
 Instances of i. are 
 
 ov μη φλναρησεις έχων ; don't keep playing the fool.— 
 Ar. Man. 202. 
 
 ού μή προσοίσεις χεΤρα, βακχενσεις δ' Ιων, 
 μηο^ εζομόρζει μωρίαν την σην ιμοί; — Eur. JBaCch. 343. 
 put not forth thy hand, but go play the bacchanal, and 
 wipe not off thy folly on me. [The ov is understood 
 both before βακχενσεις and before ρ/δ' εξομόρξει,'} 
 
 cv μη προσο'ισεις χείρα, μή& άψει πέπλων ; put not forth 
 thy hand, nor grasp my robes! — Id. Hipp. 601. 
 
 290. These are usually explained by the interrogative; 
 thus 
 
 ov μη προσοίσεις ',=z will you not not -put -forth? 
 
 = will - you - not abstain - from - putting 
 
 forth ? 
 ass put not forth ! 
 
 Undoubtedly this explanation is open to the serious objection 
 
192 Λ BRIEF GREEK SYNTAX. 
 
 that it attributes to μη that power of coalescing with, and so 
 reversing, the meaning of a word which properly belongs to ov 
 only. It is far better to explain the idiom thus : 
 
 ov ποιήσεις ; — μη ; i.e. you will not do it — will you ? 
 =do not do it ! * 
 
 Instances of ii. are 
 
 ου σοι μη μεθέάομαί ποτέ I will never follow after thee. — 
 ^ Soph. EL 1052. 
 ov τι μη ληφθώ οολω I shall certainly not be caught by 
 
 craft. — iEsch. Sept. 38. 
 αλλ' ov μη οίος r' 77c but you certainly will not be able, 
 
 —Plat. Hep. 341 c. 
 
 291. These are usually explained by the ellipse of Ιεος or 
 Ζεινόν (' There is no fear lestj &c.'), which are often expressed, 
 as in Ar. Eccles. 646 : 
 
 οΰχι ΰέος μη σε φιλησ^ there's no fear of his kissing you. 
 So in Latin : 
 
 *Non metus officio ne te certasse priorem Pceniteat' — ■ 
 jEn. i. 548. 
 
 This is a simple explanation, and is certainly admissible. It 
 may however be doubted whether these idioms, arising from 
 the union of an objective and subjective negative, do not owe 
 their prevalence to that accumulation of negative words towards 
 which there is an instinctive tendency in all languages. 
 
 M») ov. 
 
 292. After negatives, verbs expressive of negative notions 
 (fear, doubt, shame, disapprobation, &c), and in indirect 
 questions, μη ov=ne non, or ut, is used.f The μη really 
 
 * I have never met with any formal explanation of this idiom which 
 satisfied me ; I feel convinced that these idioms are simply due to the 
 tendency to accumulate negatives for the sake of emphasis. 
 
 f Verbs of fearing in Attic poetry are also followed by 8w&js = vereor 
 ut, I fear that not ; and frirws ^=vereor ne, I fear that. δ45οικα Ihrws 
 %\θχι I fear that he will not come ; Μδοικα 8V«s μί) *λ0τ? I fear that he 
 will come ; as 
 
 δεδοίχ' 'όπως 
 μτ} 'κ rrjs σιωπής τησδ' ίναρρήξ^ι κακά. — Soph. Ο. R. 1047. Ί fear 
 that calamities will burst forth from this silence.' [Literally, ' I fear 
 how lest,' &c] Here again the French idiom resembles the Greek, ' Je 
 crains que vous ne m'abandonniez ' I fear you will abandon me ; ' Je 
 crains qu'elle soit heureuse ' Γ fear that she is not happy. — Clyde, 
 p. 185. 
 
Μη ού. 193 
 
 belongs to the previous words, and expresses that their general 
 result and effect is negative. 
 
 ΰέΰοικα-μη ουκ άποθάνη I fear he will not die, vereor ut 
 
 raoriatur. 
 ΖίΙοικα-μη ουκ ελθη I fear that he will not come, vereor 
 ut veniat. 
 
 άθρει μη ου τοΰτο y το αγαθόν consider whether this be 
 not ' the good.' 
 
 293. M>) ου with the infinitive often has the sense of quin, 
 quominus, after negatives, or quasi-negatives ; after verbs of 
 preventing, denying, &c; and after Ιεινόν, αισχρόν, αισχύνη, 
 έσ-ί, &c. ; e.g. 
 
 ούΰέν κωλύει μη ουκ αληθές είναι τοΰτο nihil impedit 
 quominus id verum sit, nothing hinders this from being 
 true. 
 τι εμττοΐων μη ουκ ατοθανεϊν έμέ ; quid impedit quominus 
 
 moriar ? what prevents me from dying ? 
 μη παρτ}ς το μη ού φράσαι do not omit saying it. 
 
 ούΰεν έλλε/ψω το μη ού 
 πάσαν πυθέσθαι τώνο' άληθειαν ττερι 
 nihil praetermittam quin verum cognoscam, I will leave 
 no stone unturned to discover the whole truth respect- 
 ing these matters. — Soph. Ir. 88. 
 
 πείσομαι γαρ ού 
 τοσούτον ούΰεν ώστε μη ού θανείν καλώς 
 for I shall suffer no penalty so great as to prevent my 
 dying nobly. — Soph. Ant. 96. 
 
 ούχ οίος τε el μι μη ού λέγειν ηοη possum quin dicam, I 
 cannot but say. 
 
 294. Μη ού with the participle follows negative expressions, 
 and means unless ; as 
 
 ΰυσάλγητος γαρ αν 
 είην τοιάνοε μη ού κατοικτείρων εΰραν 
 I should be ruthless [a negative motion] if I did not pity 
 
 such a suppliant posture. — Soph. 0. 2\ 12. 
 at τε πόλεις .... χαλεπαϊ λαβείν . ... μη ού χρόνω the 
 cities are difficult (=not easy) to take except by time. 
 — Dem. de F. Leg. § 135. 
 
194 A BRIEF GREEK SYNTAX. 
 
 VARIOUS NEGATIVE PHRASES. 
 
 295. Distinguish between ουπω, μήπω nondum, not yet. 
 
 ονκέτι, μηκίτι non amplius, no 
 
 longer. 
 ovre=nec, ονΰε=ηβ quidem. 
 ου τι =not a whit. 
 ονχ ori=7iot only. 
 μη ort=nedu?n, ne dicam, not to mention.* 
 
 These two phrases however, like ονχ όπως, οΰχ olov, often 
 mean * not only not ;' as 
 
 μη όπως όρχε'ισθαι αλλ' ουδ' όρβοϋσθαι δύνασθε you were 
 not only unable to dance, but even to stand upright ; 
 so too ούχ oiov, as 
 
 ονχ υ'ιον ώψελεΊν Sircar' αν, αλλά μήο' αντην σωζειν not 
 only unable to assist, but even to save herself. 
 
 i. ovk εσθ' οπως=ην.ΙΙο modo. 
 
 ονκ έ'σθ 1 όπως λέζαιμι τα ψενΰή καλά I could not possibly ,- 
 call lies honourable. — iEsch. Ag. 620. 
 
 ii. ουκ έ'σί)' οττως ov non fieri potest quin, it cannot be but' 
 that.— Soph. El. 1479 ; Ar. Eq. 426. 
 
 iii. όσον ov, μόνον ου all but, tantum non. 
 
 όσον ονκ ήΰη άπήλθεν he has only just gone, il ne fahV 
 que de partir. 
 
 iv. ov μην άλλα ' not but what, 1 'however.' 
 
 ov μην άλλα επεμεινεν υ Κϋρος μόλις πως not but what ι 
 with some difficulty Cyrus kept his seat. 
 
 v. μη πολλάκις in Plato means * lest perchance." 1 
 vi. ούτε μέγα οντε μικρόν nothing whatever (cf. 1 Kings xxxii.h 
 21, fight neither with small nor great, &c). 
 
 vii. ohlkv χείρον l it is just as well to.' 
 
 ονΰεν Βε χείρον ντομνησθήναι καϊ Ένπόλιΰος one may^ 
 just as well mention Eupolis also. 
 
 viii. ονίεν οίον there is nothing like (doing so and so) ; as 
 
 ονΰε ν γαρ οίον άκονειν αυτοΰ τον νόμου car il n'y a rien 
 de tel que d'entendre la loi meme. 
 
 * As &xpr}(TTou καϊ γυναιξί, μτ] 'ότι ανδράσι useless even to women, not: ; 
 to mention (or muckmore to) men ; so in Italian ■ i fortissimi uomini non 
 chele tenere donne' the bravest men, not to mention delicate ladies, &c. 
 Clyde, p. 175. 
 
THE PARTICLES. 195 
 
 PARTICLES. 
 
 Μη νεμίσα βαιοΊσι, χάρις βαιοίσιν όπηίεϊ. 
 
 296. A perfect knowledge of the particles in which Greek 
 abounds can only be obtained by extensive reading.* The 
 manner in which, especially in Homer, ' they sustain and 
 articulate the pulses of emotion' is in itself a fruitful and valu- 
 able study. By them alone we can perceive that Greek was 
 the language of a witty, refined, intellectual, sensitive, and 
 passionate people. It would be impossible in any book to 
 tabulate the delicate shades of meaning, the subtle intricate 
 touches of irony or pathos, the indescribable grace and power 
 which the particles lend to many of the grandest passages in 
 ancient literature. Indeed these can often be only felt at all 
 by a scholarlike appreciation of the entire context, and of the 
 circumstances which dictated the particular expression ; so 
 that in very many instances, not in Greek only but in German, 
 and in most languages to a greater or less degree, the force of 
 the particles cannot be accurately transferred into a foreign 
 version. In short they are often untranslatable, and can only 
 be approximately represented hy some look, gesture, emphasis, 
 or tone of the voice. Thus μεν and £ε, two of the commonest 
 Greek particles, correspond to the English ' on the one hand,' 
 'on the other hand ;' but to substitute these long and heavy 
 periphrases^ for them in all cases would be utterly unidiomatic, 
 and would not in any way represent their force and meaning 
 in Greek. 
 
 It would be out of the question to attempt here anything 
 approaching to a complete treatment of the conjunctions, which 
 Apollonius DyskolusJ and Priscian arrange logically under 
 no less than eighteen heads. All that Ave shall here attempt 
 will be to give one or two notes and suggestions, which can 
 be amplified by each student for himself. 
 
 * Hence even the New Testament, though it represents the spoken 
 Greek of its day, yet being Greek written hy foreigners, is comparatively 
 poor in the use of particles. 
 
 t The attempt to translate a particle exactly leads to curious results. 
 Dr. Cyril Jackson used always to render TpcDe's pa by ' the Trojans, God 
 help them!' and a former head-master of Eton always distinguished 
 between σοι Sir, to you, and roi at your service (Coleridge, Gk. Classic 
 Poets, p. 221). 
 
 { Egger, Apollon. Bysc. p. 209. On the other hand, Dionysius Thrax 
 only recognised eight classes of conjunctions. 
 
 k2 
 
19t) A BRIEF GREEK SYNTAX. 
 
 297. Copulative Conjunctions.* — icat=et, r£=que. In 
 poetry we have ήΐέ, <2e'=atque. Often κα\ is used to mean 
 also, even • and sometimes ' and yet,' as 
 
 συ Αιος εψνς . . . και ι αχ») ση άδικος and yet thy utterance 
 is unjust !— Eur. Hel. 1147 ; cf. Here. F. 296. 
 Occasionally και nearly means ' when,' as 
 
 ηΰη ηώς ΰιεώαινε κα\ επ ακρωτήριο) εγενόμεθα. — Herod. 
 
 νϋ. 217. 
 ήΰη τε ην οψε .... και οι Κορίνθιοι πρύμναν εκρούοντο. — 
 Time. i. 50. Cf. Soph. Ο. Τ. 717; Herod. Hi. 108; 
 iv. 139, 181 ; Hebr. viii. 8 ;f Luke xix. 43. 
 ku\ raira=and that too. 
 
 μικρά και ουΜν little or nothing (literally, ' and even 
 nothing '). 
 After 'ίσος, όμοιος, b αυτός, and words of likeness generally, 
 και=' as,' like the Latin similis et, ac; 'ίσα Ka\=a?que ac. 
 
 ουχ ομοίως πεποιηκασι και "Ομηρος they did not act in the 
 
 same way as Homer. — Plat. Ion, p. 500 r>. 
 ε'ί τις και άλλος more than any one (by litotes). 
 άλλως τε κα\ especially. 
 
 και ΰή well, suppose, or granted ; fac ita esse. 
 και with πώς, &c, often expresses surprise, &c. It is used too 
 in eager appeals, as 
 
 καί μοι 3ός την χείρα ' give me then your hand.' 
 i] και τοιαύτας τωο επιρροιζείς φυγάς ; dost thou too really, 
 &c— ^sch. Eum. 424. 
 It often seems to connect the speaker's first words with a long 
 train of his thoughts. One of Lord Lytton's tales begins with 
 the word ' and ' — ' And the stars sat each upon his ruby throne, 
 and looked with sleepless eyes upon the world.' — Pilgr. of the 
 Rhine. 
 
 i And, 1 says Ben Jonson (Engl. Gram. p. 82), * in the 
 beginning of a sentence serveth for a mark of admiration.' 
 ' What, quoth shee, and be ye wood ! 
 And wene ye for to doe good, 
 And for to have of that no fame?' 
 
 Chaucer, Man of Laive's Tale, 
 καί ει etiam si, even if; ει καί quamauam, even though 
 (wenn audi). 
 
 * The Hebrew 1 ' and ' means a hook, and resembles a book in shape. 
 
 f So in the Latin et : ' Nox media, et dominae mihi venit epistola 
 
 no.strse.' — Prop. in. xiv. 1. 
 
 
 < 
 
THJJ PAKT1CLES. 197 
 
 Negative clauses are coordinated (united together) by οΰτε nee, 
 ονΰε ?ie quidem, &c. 
 
 ούτε followed by τ ε= so far from . . . that. 
 
 298. Disjunctive Conjunctions. — η . . . η ; είτε . . . e'tre. 
 
 299. Adversative Conjunctions. — μεν ' indeed,' ' on the 
 one hand,' the old neuter from tic, μία, εν = ' one thing.' 
 
 δε l but,' ' on the other hand,' derived from δεις = δυο = 
 i two things.' μεν is always {regularly) followed by δε, or, 
 less accurately, by some other adversative particle, as αλλά* 
 αυ, μέΐ'τοι, &c. μην, δ//, are lengthened forms of μεν, δε'. 
 
 καίτοι = ' and yet,' ' although,' verum, sed tamen. καίπερ 
 1 although ' is used with the participle ; καίτοι with the finite 
 verb, as καίτοι αγαθός ijji>, καίπερ αγαθός ων. 
 όμως ι nevertheless,' nihilominus ; as 
 
 ήκουσα κάγω τηλόθεν μεν, άλλ' όμως I heard it from a 
 
 distance, indeed, but still I heard it.f — Eur. El. 753. 
 Oic. άλλ' εκκυκληβητ. Eur. άλλ' αδύνατον. Die. άλλ' 
 όμως. 
 
 Ό. Now do be wheeled out. N. Nay I can't. D. Nay 
 
 but do !— Ar. Ach. 401. 
 κάγω σ ικνοΰμαι, και γυνή περ ούσ* όμως and I too beseech 
 
 thee, though but a woman, still ! — Eur. Or. 671. 
 
 300. Conjunctions of Comparison. — ώς, ώσπερ, ώστε. Horn. 
 ΐ'ΐ'ότε. 
 
 ώς = as, ως thus ; but when ώς as follows its word it 
 receives an accent ; as λέων ως like a lion. 
 
 301. Temporal Conjunctions. — οτε, οπότε quando, quum. 
 Horn. ευτε. 
 
 επεί, επειοη, εως, εστε, άχρι, μέχρι, πριν, πάρος [see Tem- 
 poral Sentences, § 214 seqq.]. 
 
 αντίκα immediately, is used by Plato to mean l for instance? 
 
 302. Causal Conjunctions. — οτι, ΰώτι, ένεκα, γάρ, &c. 
 γαρ is derived from γ ε and άρα. γαρ in animated style often 
 points to a suppressed sentence. 
 
 πώς γαρ ου ; of course ! J 
 
 τί γάρ ; how so ? τι yap κακόν εποίησε ; why, what evil 
 hath He done ? 
 
 * 'Αλλά νη Δία = but some one will say, at cnim. 
 
 t Compare the position of tamen in ' Perfida, sed quanivis perfida, cara 
 tamen.' 
 
 J Cf. Ital. percke no 1 — certainly ! 
 
198 Λ BRIEF GREEJv «ΥιΝΤΑΧ. 
 
 ti γιιρ utinam. 
 ου γαρ άλλα liowevcr. 
 
 ή γαρ τέθνηκεν ούτος ; what ! is this man dead? 
 γαρ also may express indignation, as 
 
 Ατρείδη κύδιστε, ψιλοκτεανώτατε πάντων, 
 
 πώς γαρ τοι δώσουσι γέρας μεγάθυμοι Αχαιοί] — II. ί. 122. 
 
 'Ανδρες Έφεσιοι, τις γάρ εστίν άνθρωπος ος ου γιγνώσκει, 
 
 κτ.λ. (Acts χχ. 35), Ephesians ! why what person is 
 
 there who is not aware, &c. 
 
 Like the Latin nam, as 
 
 Nam quis te, juveninn confidentissime, nostras 
 
 Jussit adire domos? — Georg. iv. 445 (cf. Mn. ii. 373). 
 
 303. Ikferential Conjunctions. — "Αρα. (Ep. ap and pa) 
 often expresses surprise, emotion, like ' it seems,' ' after all? 
 &c. So that the Dean (see note f p. 195) was not so far wrong 
 when he translated Τρώες αρα i the Trojans, God help them ' 
 (New Crat. p. 335) ; as 
 
 ταϋτα άκουσας 6 Κϋρος επαίσατο άρα τον μηρον when 
 Cyrus heard this, he smote on his thigh. 
 υφ? ου 
 ψονεως άρ εζέπνευσας ; 
 by whose murderous blade after all you died. — Soph. 
 
 Aj. 1025. 
 ήλθεν ει άρα ευρησει τι εν abrrj he came if haply he might 
 
 find anything thereon. — Mark xi. 13. 
 ώ πα"ιδες, ως άρα εφλναροϋμεν boys, how we were trhiin 
 after all ! 
 This is like the Latin ergo, as in 
 
 1 Ergo Quintilium perpetuus sopor urget ' 
 so then the sleep that knows no end is weighing down 
 
 Quintilius! — Hor. Od. i. xxiv. 5. 
 dpa . , ■. ; = ne, 
 άρα oh . . . ; = nonne, 
 αρα μη . . . ; = num ? 
 ουν then, ουκονν not then, ουκουν therefore. In this sense the 
 ουκ becomes simply otiose (see § 103, and Herni. Vig. n. 261). 
 μεν ουν nay rather, immo. 
 τά& αν δικαίως ή ν, ύπερδίκως μεν ουν this would have been 
 
 justly done, nay more than justly. — iEsch. Ag. 1363. 
 εγώ ου φημί', ψημί μεν ουν εγωγε do I deny it? nay on 
 the contrary, / assert it. — Plat. 
 
THE PARTICLES. 199 
 
 Iu die Knights of Aristophanes when Kleon proposes that 
 Demos, the personified Great Public, should wipe his nose on 
 — but we must leave the line untranslated, Eq. 910 : 
 
 άπομυζάμενος, ώ Δήμ\ εμού προς την κεφαλήν άποφώ, 
 
 the sausage-seller feeling that he cannot beat that proposal, 
 cries out 
 
 εμοϋ μεν ουν, εμοϋ μεν ουν nay rather on mine, on mine ! 
 
 Particles of Emphasis. 
 
 304. Γε ' at least ' is used to modify various words ; as 
 ό'ς γε quippe qui, * seeing that he.' 
 εγωγε equidem, I for my part. 
 εί γε since, 
 γε μην however. 
 
 Often ironical, as 
 
 ευ γε κηΰευεις πόλιν good care you (forsooth) take of the 
 
 city ! 
 παΰσα/ γε do cease ! 
 
 The exclamation μη συ γε oh do not ! is often used with 
 great pathos by Euripides, as in 
 
 μη ΰήτα, θυμέ, μη συ γ' εργάση τά£ε. — Med. 1056. 
 βυΰλει . . . άσχημονήσαί τ εκ νέου βραχίονας 
 σπασβέϊσ, α πείσει' μη συ γ' ■ ου γαρ άζιον. — Hec. 405. 
 
 See too Ion, 439, 1334; Pham. 531 ; Iph. Aul. 1460. 
 
 που often expresses surprise, ούτι που { not, I presume ; ' 
 ου Ιηπου ' not, I suppose ; ' e.g. 
 
 ■πώς', ούτι που σω φασγάνω βίου στερείς', — Eur. Hel. 95 
 
 [cf. 475, 541, or 1510]. 
 ov τι που minantis et indignantis est, ου Ιηπου suspicantis. 
 — Stallbaum. 
 
 γοΰν at any rate. 
 Ιη ' certainly : ' 
 
 και τότε ΰή even then ; ούτω δ// then at last. 
 
 νυν οράτε fa) now of course you see. 
 
 μέγιστος ΰή far the greatest [compare αυτός ΰή i-dem, 
 πριν Ιη -pri-dem, αγε οη age dum~\. 
 
 Often like Ιηπου ' of course,' ' forsooth,' with a shade of sarcasm. 
 ml $ή often means fac ita esse; as 
 
200 A BRIEF GREEK SYNTAX. 
 
 «rat h) τεθνίισι• τίς με ΐέζεται 7τόλ<ς; well, suppose them 
 dead; what state then will receive me? — Eur. Med. 
 386; Hel. 1066. 
 Sometimes it implies quid turn? as in Hel. 101 ; El. 655. 
 
 βλέψον κάτω look downwards. 
 
 και ύή βΚέπω well, I am looking — what then ? 
 
 σχεδόν τι ' it may perhaps be said ' also expresses great 
 irony; as 
 
 σχεδόν τι μωρφ μωρίαν όφλισκάνω (Soph. Ant. 470) perhaps 
 it is a fool at whose hands I incur the charge of folly. 
 
 Ζήτα is a lengthened form of h) ; e.g. 
 
 όΊκτειρε ό' ημάς .... ο'Ίκτειρε Ζήτα but pity us — ay, do 
 
 pity us.— Eur. El. 678. 
 "ιω ϊω h~]r woe ! ay, woe ! — Soph. Ο. E. 541. 
 δήθεν ' naturally enough ; ' or, as they alleged, i scilicet,' 
 
 mostly in an ironical sense. — Hdt. i. 59 ; Thuc. i. 92. 
 ΰήπουθεν ' I should hope.' 
 
 μην i verily,' * truly,' vero, a lengthened form of μεν — 
 
 τιμήν, why not? of course; what then? 
 
 έπου μην do follow. 
 
 ύλλ' εστί μην οίκητός well, it certainly is inhabited. — Soph. 
 (Ed. Col. 29. 
 
 και μή enimvero, moreover. 
 μα a form of adjuration, generally in negative oaths, as 
 
 ου μα Δ/α no by Zeus ! 
 
 ου μα τόδε σκήπτρον never by this sceptre ! 
 
 περ a shortened form of περί', in its adverbial sense of 
 * exceedingly ' it increases the force of words, like per in Latin, 
 as ' pergratus, jerque jucundus.' 
 
 εάν περ even if. 
 
 αγαθός περ very good ; compare our colloquial expression 
 4 good all round,' and the French tres, which is derived 
 from trans, so that tres bon == thoroughly good ( = good 
 throughout). 
 
 Often it comes to mean i although, 1 as 
 
 γεννοΛος περ εων though noble, &c. 
 τοι ' ay,' as 
 
 σε τοι, σε κρίνω you, ay, you. — Soph. El. 1445. 
 Probably the τοι in τοιγαρ ' therefore ' is derived from τω since 
 it may begin a sentence, as in Soph. Tr. 1249 ; Ant, 594. 
 
ORDER OF WORDS. 201 
 
 INTEEJECTIONS. 
 
 305. Interjections being, as their name implies, passionate 
 exclamations thrown in to the sentence, are for the most part 
 unsyntactical. The Greeks did not even regard them as 
 forming separate parts of speech, but classed them with ad- 
 verbs. The Eoman grammarians first treated them separately. 
 Their claim to be separately considered, and their high lin- 
 guistic importance, I have vindicated elsewhere (Chapters on 
 Language, pp. 88-103). Their antiquity and their truthfulness 
 have justified grammarians so eminent as Scaliger and Destutt 
 de Tracy in regarding them as words par excellence. 
 
 ώ the sign of the vocative (άρθρον κλητικής πτώσεως) is an 
 interjection in all languages, and is in reality the same as ώ 
 the interjection (επίρρημα σχετλιασμού). 
 
 Interjections may be followed either by the causal genitive 
 (as οϊμοι των κακών) ; or, more rarely, by the accusative of 
 the object. 
 
 The tragedians often have interjections extra metrum ; i.e. 
 they do not take them into the scansion of the line. 
 
 ORDER OF WORDS, &c. 
 
 306. A sentence is arranged in the natural order when the 
 subject with all that belongs to it is placed first, and then the 
 predicate with all that belongs to it, the copula being either 
 expressed between the two, or understood, or involved in 
 some inflection. 
 
 307. Thus in all languages such a sentence as 
 
 Alexander conquered Darius 
 
 is expressed in the natural order (φυσική τάξις) ; and it would 
 usually be so expressed in Greek, as 
 
 υ 'Αλεξάνδρας ενίκησε τον Ααρείον. 
 
 But owing to the inflection of the accusative in Greek and 
 Latin, the order may be altered in those languages in every 
 possible way (πλαγιασμός), without any modification of the 
 sense, — the subject, the verb, or the accusative being placed 
 first, according as it is requisite to make any one of them em- 
 phatic ; whereas in English or French any variation of the 
 order destroys the sense, and if it were necessary to bring 
 Darius into prominence we should be obliged to adopt some 
 entirely different turn of sentence, as 
 
 Darius was conquered by Alexander, 
 κ 3 
 
202 A BRIEF GREEK SYNTAX. 
 
 308. We can indeed use a rhetorical inversion in English 
 poetry (though but rarely in prose), and often with the finest 
 effect; as 
 
 And over them triumphant Death his dart 
 
 Shook, but delayed to strike. — Milton. . 
 
 Under a coronet his flowing hair 
 
 In curls on either cheek played ; wings he wore, &c. — Id. 
 
 But our power of doing this is extremely limited, as must 
 always be the case in a flexionless language ; and it is impos- 
 sible to read a page of Demosthenes, or Cicero, or Virgil, 
 without seeing the immense rhetorical power which they are 
 able to command by a mere variation in the order of construc- 
 tion. It is almost impossible to render in an analytical lan- 
 guage the matchless force of such expressions as 
 
 kv ϊε φάει. καΐ Όλεσσον, 
 or, 
 
 Me, me, adsum qui feci, in me convertite ferrum, 
 Ο Eutuli ! 
 
 And although the rich and powerful vocabulary of English 
 renders it one of the noblest of all languages, yet in harmony, 
 precision, elasticity, variety, grace, and force, it must yield an 
 easy victory to the Greek. 
 
 309. We may here mention one or two of the figures, 
 rhetorical and idiomatic, which are of the most constant occur- 
 rence in Greek. It will be seen that many of them are due to 
 that agility and acuteness of the Greek intellect which enables 
 them readily to sacrifice the grammar of a sentence to its logic, 
 or in other words its form to its meaning. Hence arose the 
 many forms of the sense-figure (σχήμα προς το σημαιιόμειον 
 constructio ad sensum) ; e.g. 
 
 i. When the concord is only a concord of the sense,* as 
 
 φίλε τεκνον ; varium et mutabile semper Fcemina ; Δως 
 τέκος ητε μοι αιε/, &c. 
 
 ii. When the expression is shortened by the suppression of 
 a clause or word (Brachylogy, breviloquentia), as 
 
 Οειρα /Soar, SC. βοι'ιματα, τύπτομαι πολλάς, SC π\ηγάς. 
 
 * Cf. the Italian Corsevi le sorelle ; {each of) the sisters ran thither. — 
 Boccaccio. 
 
FIGURES OF SPEECH. 203 
 
 Of this there are several varieties, as 
 
 a. Constructio prosgnans, where two clauses are compressed 
 into one ; as 
 
 Φίλιππος ευρέθη εϊς'Άζωτον P. was carried to Azotus, and 
 found there. 
 
 b. Zeugma, where two nouns are joined to a verb, which 
 only suits one of them, but suggests the other verb, which may 
 often be even opposite in sense ; as 
 
 γάλα υμάς έπόπσα, ου βρωμά I gave you milk to drink, 
 not meat. — 1 Cor. iii. 2. 
 
 κωλυόιτων -γαμίΊΐ', άπέχεσθαι βρωμάτων preventing from 
 marriage, (ordering to) abstain from meat (where the 
 positive κε\ευοντω\• is understood out of the negative 
 κωλυόντων). — 1 Tim. iv. 3. 
 
 'See Pan with flocks, with fruits Pomona crowned' 
 (where from ' crowned' we must understand ' sur- 
 rounded' in the first clause). — Pope. 
 
 This figure of speech is very rare in English, and illustrates 
 more than any other the Greek quickness of apprehension. 
 
 c. Syllepsis, often confounded with Zeugma,* where the 
 same word is applied to different nouns but in a different 
 sense ; as 
 
 εΧεν c' Οινόμαου βίαν τταρθενον τε σΰνευνον he subdued 
 the mio;ht of CEnomaus, and [won] the virgin as his 
 bride.— Pind. 01. i. 88. 
 
 * Quas et aqua} subeunt et aura? ' under which the waves 
 and breezes flow. — Hor. 
 
 In English the chief instances are comic, as 
 
 1 This general is a greater taker of snuff as well as of 
 towns.' — Pope. 
 
 1 And there he left his second leg, 
 And the forty-second foot.' — Hood. 
 
 1 Miss Bolo went home in a flood of tears and a sedan- 
 chair.' — Dickens. 
 
 1 He flung his powerful frame into the saddle and his 
 great soul into the cause.' — Earl of Carlisle, Siege of 
 Vienna. 
 
 * On the distinction between the two, see Lobeck, ad Soph. Aj. p. 429 
 seqq. 
 
204 Λ BRIEF GREEK SYNTAX. ' 
 
 d. Comparatio Compendiaria,ox Brachylogy of comparison ; 
 as 
 
 κομαι Χαρίτεσσιν ομοΐαι hair like (that of) the Graces. — 
 
 II. xvii. 51. 
 είχε κέρατα ΰύο όμοια άρνίω he had two horns like (those 
 
 of) a ram. — Rev. xiii. 11. 
 πνραμις πατρός μείζων a pyramid loftier than (that of) 
 
 his father. 
 1 His ascent is not so easy as tliose who,'' &c. — Shakspeare, 
 
 Coriolanus, ii. 2. 
 
 e. Ellipsis, the omission of a word easily understood, as 
 
 εϊς $δον, ώς βαθύν εκοιμηθης sc. νπνον, ες κόρακας sc. ερρε, 
 
 ποτήριον ψνχροΰ sc. νΰατος, calida sc. aqua, &c. 
 'To whom thus Eve in few.'' — Milton. 
 
 This is common in all languages, as when we say a coach and 
 six (sc. horses), a bottle of port (sc. wine), to St. Paul's (sc. 
 church), he sat on the right (sc. hand), &c. 
 
 /. Anakoluthon, or non-sequence ; when the sentence begins 
 with one construction, and continues in another. This is very 
 common in Greek, which is a language eminently swayed by 
 emotion, and one in which the syllogism of passion often super- 
 sedes and transcends the syllogism of logic. It is found in 
 writers who adopt a naive, simple, childlike style, as Herodotus; 
 in those profound and powerful writers whose thoughts flow 
 more rapidly than their words, as Thucydides, Pindar, iEschy- 
 lus, and St. Paul ; and in those who, like Plato, adopt the 
 informal and easy style of common life.* 
 
 Sometimes, a., they are common sense -constructions ; some- 
 times, /3., rhetorical ; and sometimes, γ., merely due to care- 
 lessness or accident. 
 
 α. εΐοζε τοϊς ' Αποστόλοις . . . γράψαντες. — Acts χν. 22. f 
 
 β. Under this head fall the instances oioratio variata, where 
 for the avoidance of monotony, the phrase is altered, as 
 
 ζηλοΰτε τα πνευματικά μάλλον ΰε ίνα προψητενητε. — 1 Cor. 
 xiv. 1 ; 
 
 * See Jelf, § 901. 
 
 f Cf. απ•ηγγέ\η αντ$ λεγόντων, Luke viii. 20. and similar idioms in 
 the LXX. passim. 
 
FIGURES OF SPEECH. 205 
 
 and the frequent transition from oratio obliqua to oratio recta ; 
 as 
 
 παρηγγειλεν αυτω μηίενί εΙπεΊ,ν αλλ' άπελθών }ε~ιζον κ.τ.Χ. 
 
 he bade him to tell no one, but departing shew thyself, 
 
 &c. — Luke v. 14; cf. Acts xxiii. 22; Ps. lxxiv. 16 
 
 seqq.; Virg. JEn. viii. 291. 
 
 This is sometimes used with fine effect in poetry, as in Milton 
 
 {Par. Lost, iv. 721) : 
 
 ' Both turned, and under open sky adored 
 The God that made both sky, earth, air, and heaven . . . 
 And starry pole. Thou also madest the night, 
 Maker Omnipotent, and thou the day,' &c* 
 
 See Stebbing's Longinus, pp. 102, 103. 
 
 y. Careless anakolutha are found even in the best writers ; as 
 θεο>ρώ, on μετά ύβρεως .... μέλλειν εσεσθαι Toy πλουν. 
 
 Acts xxvii. 10. 
 ' Those who he thought true to his party.' — Clarendon. 
 The sun upon the calmest sea 
 Appears not half so bright as thee. — Prior. 
 
 g. Aposiopesis, the passionate suppression of the latter part 
 of a sentence ; as 
 
 κ α J' μεν ποίηση κάρπον . . . ε\ Ιε μηγε. — Luke xiii. 9 (for 
 other instances see Luke xix. 42 ; xxii. 42 ; Acts 
 xxiii. 9). Here, as Vainer finely observes, 'sorrow has 
 suppressed the apodosis.' 
 μη σύ y. — Eur. Ilec. 405. f 
 
 Quos ego — sed motos pnestat componere ventos. — Virg. 
 JEn. i. 135. 
 Compare the German "Warte, ich will dich . . . ! 
 ' Bertrand is — what I dare not name !' — Scott. 
 310. Among other figures of speech Ave may mention 
 
 ΗΥΡΕΕΒΑΤΟΝ,ί 
 verbi transgressio, the rhetorical misplacement of a word, as 
 
 ω και δεκάτην Αβραάμ εΰωκεν ει: των άκροθινίων, 6 πατρι- 
 άρχης to whom even Abraham gave a tithe of his first- 
 fruits, the patriarch. — Heb. vii. 4 ; cf. Mark xi. 10. 
 
 * For similar instances see Forbiger, Virg. Mn. ii. 182, iii. 185. 
 
 f See 11. i. 340. 
 
 % The word, which first occurs in Plato (Protag. p. 343 e) was pro- 
 bably borrowed from him by the scholiasts. See Weil, Be Vordre des 
 mots dans les langues anciennes, p. 8. 
 
206 A BRIEF GREEK SYNTAX. 
 
 This is not uncommon in Elizabethan English. 
 
 ' More than ten criers and six noise of trumpets.' — Ben 
 Jonson, Sejanus, v. 7. 
 
 Under this head we may range, 
 
 a. Antiptosis, the transposition of the subject from one 
 clause to another, as 
 
 uv ειίες άνδρα οντάς εστίν. Cf. Acts xxi. 16; Kom. vi. 
 
 - 17 ' 
 olV ην ϊθρεφεν Έρμιύνην μητηρ εμη. — Eur. Or. 1117. 
 
 Urbem quam statuo vestra est. — JEn. i. 572. 
 
 Him I accuse 
 The city gates by this hath entered. — Shaksp. Ant. and 
 
 Cleop. iii. 1. 
 1 And God saw the light that it was good.' — Gen. i. 4. 
 See p. 78. 
 
 b. Chiasmus, when words are arranged cross-wise like the 
 letter X, as 
 
 ψόνη βραχεία 
 
 μακράν Χΰττην τίκτει. 
 
 This is very common in Latin, where the arrangement 
 
 Eatio consentit, repugnat oratio (Cic. de Fin. iii. 3) 
 
 is more elegant and forcible than ratio consentit, oratio repug- 
 nat. Something like it is found in English, as 
 
 1 He hath fed the hungry — the rich he hath sent empty 
 away.' 
 
 ' Foreknowledge, will, and fate, 
 Fixed fate, free will, foreknowledge absolute? 
 
 Milton, Par. Lost, ii. 560. 
 
 c. Hysteron Proteron (πρωθυστερον) or Last-first, as 
 τας μεν άρα θρέφασα τεκηϋσά τε. — Od. χϋ. 134. 
 1 Moriamur et in media arma ruamus.' — Virg. JEn. ii. 353. 
 1 In Africam redire atque ex Italia decedere.' — Cic. Cat. 
 
 iv. x. 21. 
 ' Is your father well, the old man of whom ye spake, is 
 
 he yet alive ? ' — Gen. xliii. 47. 
 ' 1 die, I faint, I fail/ — Shelley. 
 
FIGUJEES OF SPEECH. 207 
 
 d. Hypallage, an attraction of the adjective to a substan- 
 tive with which it does not properly agree, or more generally 
 a change of case (Enallage, as dare classibus Austros, for 
 classem Austris). 
 
 qjkov ονόματος μητρωον motherly boast of a name=boast 
 of a mother's name. — Soph. Tr. 817. 
 
 Nee purpurarum sidere clarior 
 Delenit usus. — Hor. Od. m. i. 42. 
 
 { Holy and humble men of heart ' = men of holy and 
 humble hearts. Cf. Isaiah. 
 
 i With the innumerable sound 
 Of hymns and sacred songs.' — Par. Lost, iii. 147. 
 
 311. EUPHEMISM, 
 
 the principle of avoiding all strong or unpleasant forms of 
 expression. This tendency has exerted a most powerful in- 
 fluence over the Greek language,* and leads to the use of such 
 terms as εάν τι πάθη for ' if he die,' ενηθης for ' silly,' οίκημα 
 for ' prison,' &c. (See Abbott, ShaJcsp. Gram. p. 75, and some 
 remarkably beautiful lines of Faber, quoted in Heed's Led. on 
 Eng. Lit. p. 90.) We may range under this head 
 
 a. Irony (χλευασμός, very different from the Greek ιϊρωνεία 
 of which the style of Plato is so perfect an example), Persiflage 
 (γαριεντισμός), complimentary expressions (άστεϊσμός), &c, 
 which need no special illustration. 
 
 b. Hypokorisma, the use of exaggerated terms of endear- 
 ment, and the veiling over of that which is disagreeable or 
 vicious by sr^ecious glosses (see Chapters on Language, pp. 
 281, 282). 
 
 c. Litotes (smoothness), the suggestion of a strong notion 
 by the use of an over-weak form of speech, as 
 
 ου πάνυ = omnino non, ουχ ηκιστα = μάλιστα. "j" 
 
 ονΐέ κέ μ'ιν τις 
 γηθησειεν ί£ώι\ — II. 
 
 * In fact euphemism is woven into the very structure of Greek, and 
 explains many of its words and idioms. Hence au with the optative for a 
 polite imperative, and an indirect future ; the use of the optative as the 
 most indirect mood in wishes ; the use of the indefinite ns for a personal 
 pronoun (as in English ' one ' — ' it's enough to enrage one,' &c). See 
 Chapters on Language, p. 278. 
 
 f This particular use of the negative, as when we say of a poor man 
 ' he's not rich,' of a short man oi> μί -yas, &c. is called Meiosis. 
 
208 A BRIEF GREEK STKTAX. 
 
 Illaudati Busiridis aras. — Virg. Georg. iii. δ. 
 1 Shall I praise you for those things? I praise you not.' 1 
 4 Narcissa's nature tolerably mild 
 To make a wash would hardly stew a child.' — Pope. 
 
 d. Antiphrasis, the suggestion of a word by the use of its 
 opposite, as ευώνυμος and αριστερός for the ill-omened left. 
 
 e. Ambiguity, the use of a formula to dismiss an unpleasant 
 subject ; * as 
 
 δ γέγραφα γέγραφα what I have written I have written 
 (cf. ' If I perish, I perish ; ' ' If I be bereaved of my 
 children, I am bereaved,' &c.).— 0. T. 1376, &c. 
 
 He is that he is, I may not breathe my censure. — Othello. 
 Among other figures we may briefly mention 
 
 312. PLEONASM, 
 
 or the use of words apparently superfluous, as in 
 
 πό\εμον ττολεμεΊν, μεγέθει μέγας, πανύστατον ΰή κουποτ 
 αύθις αυ πάλιν, εψη λέγων, cursim currere, ' we have 
 seen with our eyes,' &c.f 
 
 This is an important tendency in language, and admits of a 
 very wide range of illustration, which cannot here be given. 
 Under this head we may range two out of many rhetorical 
 figures (such as Epanaphora, Anadiplosis, Palillogia, &c), e.g. 
 
 * Hauc forraulam et similes adhibent ii qui rem clarius exponere aut 
 nolunt, aut nequeunt. — Seidler. 
 
 f ' Pistol. He hears with his ears. 
 
 Sir Hugh. The tevil and his tarn ! what phrase is this, " He hears 
 with ear ? " Why it is affectations.' — Shaksp. Merry Wives of 
 Windsor, i. i. 
 Lobeck has treated the subject with his usual exhaustive learning, 
 Paralip. Gram. Grcec. 61 seqq. and Dissert. 8; and on Aj. v. 140, 866; 
 see too Id. pp. 181-185. It is a special characteristic of immaturity, 
 and therefore of children ; hence it is very common in colloquial usages, 
 and in infant literatures. One very common form of pleonasm, espe- 
 cially in the tragedians, is the repetition of a participle after the principal 
 verb ; e.g. Kreivei Κρέοντα κα\ κτανων &ρχ(ΐ χθονός. — Eur. Here. F. 33. 
 Cf. Hec.25, Pheen. 22, &c. There is an instance of pleonasm in Pope's 
 Odyssey, which Lord Macaulay used to call ' the very worst line in the 
 English language,' viz. : 
 
 1 To the rock he clung 
 And stuck adherent, and suspended hung I' 
 
 See Origin of Language, p. 168. 
 
FIGUltES OF SPEECH. 209 
 
 a. Periphrasis, or circumlocution ; a3 
 
 μέγα χρήμα σνάς,* βίη Ήρακλήος, σθέιος "Εκτορος, ιερή 
 ϊς Τηλεμάχου), κ.τ.λ. 
 
 Compare : 
 
 4 When once the service of the fort is gangrened.' — Shaksp. 
 4 The higli promotion of his Grace of Canterbury, 
 Who holds his state at door with pursuivants.' — Hen. 
 VIII. v. 2. 
 
 Milton — 
 
 1 where the might of Gabriel fought 
 And with fierce ensigns pierced the deep array 
 Of Moloch, furious king.' — Par. Lost, vi. 345. 
 
 and Gibbon — 
 
 * The youth and inexperience of the prince declined a 
 perilous encounter.' 
 
 and Schiller — 
 
 1 Zu Aachen in seiner Kaiserpracht, 
 Im alterthtimlichen Saale, 
 Sass Konig Rudolphs heilige Macht 
 Beim festlichen Kronungsmahle.' 
 
 Der Graf von Habshurg. 
 See Stebbing's Longinus, p. 108. 
 
 b. Polyptoton, the collocation of different cases or tenses 
 of the same word, as 
 
 ϊόσιν kukclv κακών κακυίς. — -ZEsch. Pers. 1035. 
 
 Clipeus clipeis, umbone repellitur umbo, 
 
 Euse minax ensis, pede pes, et cuspide cuspis. — Stat. 
 
 Dart follows dart, lance lance. — Byron. 
 
 Alive they shall not take him ; not they alive, him alive. 
 
 — Carlyle, French Rev. i. 282. 
 ' Both stricken strike, and beaten both do beat." 1 — Spenser, 
 
 F. Q. v. 7. 
 
 313. HENDIADYS, 
 
 the use of two nouns to convey one notion, as 
 
 βυτά καϊ λε'ιαν = plundered booty. — Soph. Aj. 145. 
 Pateris libamus et auro = with golden cups. — Virg. Georg. 
 ii. 192. 
 
 * See Bernliardy, Griech. Syntax, S. 52. 
 
210 A lililEF GREEK SYNTAX. 
 
 See Lobeck ad loc. p. 112. He distinguishes four kinds of 
 hendiadys : 
 
 1. Where the second word is explanatory, as 
 πνρι και στεροπα~ις ' with lightning flames.' 
 
 2. Where the dependent notion precedes, as 
 αίμα και σταλαγμον ' a drop of blood.' 
 
 3. Where two entire synonyms are united, as 
 
 λήγε βοών και παύε (compare ' I am a widow woman, 
 and my husband is dead,' 2 Sam. xiv. 5). 
 
 4. When words of similar origin are joined, as 
 στροβεΐ καΐ στρέφεται. 
 
 314. ASYNDETON, 
 
 the omission of conjunctions, as Abut, excessit, evasit, erupit. 
 There is a fine instance in Eur. Hipp. 352, expressive of the 
 most violent emotion. Many epithets are often thus joined 
 (πΰργωσις επιθέτων), as in Homer, II. xi. 32 :* 
 
 άμφιβρότην πολυΰαίΰαλον άσπιλα θοϋριν καλήν. 
 Thus we find in Shakspeare — 
 
 Unhouseled, unanointed, unanealed. 
 and Milton — 
 
 Among innumerable false, unmoved, 
 
 Unshaken, unseduced, unterrified, 
 
 His loyalty he kept, his love, his zeal. — P. L. v. 501. 
 
 315. PARONOMASIAS 
 
 the juxtaposition of words of similar sound, which is especially 
 frequent in proverbs, and proverbial expressions, as 
 
 παθήματα μαθήματα, bear and forbear, changes and 
 chances, giving and forgiving, &c. 
 
 In Eom. i. 29, 31 we have πορνείφ πονηρία, φθόνου φόνου, 
 ασύνετους άσυνθέτους. 
 
 1 Quam ferus et vere ferreus ille fuit.' — Tibullus. 
 
 i Fear the^rceness of the boy.' — Ben Jonson. 
 
 * In JEschylus we have six epithets to one noun, Ag. 155, 
 μίμνει φοβέρα παλίνορτος, 
 οικονόμος, δολία, μνάμωρ μηνις τεκνόποινος. 
 
 t This subject is treated at some length (being a very important one 
 in the history of language) in Chapters on Language, p. 265. 
 
FIGURES OF SPEECH. 211 
 
 I 
 
 I Such assonances form the staple ornament of Arabic prose (see 
 1 Families of Speech). They were very popular in euphuistic 
 i style : 
 
 1 Who can perswade where treason is above reason, and 
 might ruleth right, and it is had for lawfull whatsoever 
 is lustfully and commotioners are better than com- 
 missioners, and common woe is named common- 
 wealth ? ' — Sir John Cheeke. 
 
 Under this head fall the numerous plays on names and words* 
 j found in writers of every age and every language ; and under 
 \ the same general division fall such figures as, 
 
 a. Onomatopceia, the imitation of the sense by the sound; 
 whether in words, as τηνεΧλα the sound of a harpstring, 
 taratantara the blast of a trumpet, &c, or in lines, as 
 
 Ιεινη Ιε κλαγγη γένετ άργύρεοω βιο~ιο (of a twanged bow- 
 string). 
 
 πολλά £' αναντα, καταντά, πάραντά τε, Ζύγ^μίά τ i)\Qov 
 (of galloping horses). 
 
 Quamquam sunt sub aqua sub aqua maledicere tentant 
 (of the ovoaking of frogs). — Ovid. 
 
 Quadrupedante putrem sonitu quatit ungula campum. — 
 Virg. jEn. viii. 596. 
 
 1 Shocked like an iron-clanging anvil banged 
 With hammers.' — Tennyson, The Princess. 
 
 Und es wallet, und siedet, und brauset, und zischt, 
 
 Wie wen η Wasser mit Feuer sich mengt, 
 Bis zum Himmel spritzet der dampfende Gischt, &c. 
 
 Schiller, Der Tancher. 
 
 This figure abounds in the best poets of every age.f 
 
 * It is particularly common in Tennyson ; as 
 
 'Every soldier waits 
 Hungry for honour, angry for his king.' 
 
 ' the sea-wiud sang 
 Shrill, chill with flakes of foam.' 
 
 1 To break my chain, to shake my mane.' 
 
 t It is a principle of immense importance. See Origin of Language, 
 chap. iv. ; Chapters on Language, p. 168 and passim. 
 
212 A BRIEF GREEK SYNTAX. 
 
 b. Alliteration, as 
 
 Σώσος καϊ Σωσώ Σωτείρη την& ανέβηκαν 
 
 Σώσος μεν σωθείς Σωσώ ύ' on Σώσος εσώθη. — Simonides. 
 
 ' Ο Tite, tute, Tati, tibi tanta, tyranne, tulisti.' — Ennius. 
 ' Alliteration adds its artful aid ' very commonly in our own 
 poets, and is, as alternate alliteration, used very subtly in the 
 following examples : 
 
 Her c/ainty Jimbs did /ay. — Spenser. 
 
 His henvy- shotted Aamniock-s/uOud. — Tennyson. 
 
 c. Oxymoron is the juxtaposition of opposite words, as 
 γάμος άγαμος, χάρις άχαρις. 
 
 Funera ne-funera i living deaths ' (Catull. lxiv. 83), 
 splendide mendax, &c.,* insaniens sapientia, impietate 
 pia est (Ον.), strenua nos exercet inertia (Hor.). 
 
 ' His honour rooted in dishonour stood, 
 And faith unfaithful kept him. falsely true. 1 
 
 Tennyson's Idylls, p. 192. 
 
 4 Shall make the name of Danton famous infamous in 
 every land.' — Carlyle. 
 
 d. Antithesis, the contrast of opposite conceptions, as 
 
 Infelix Dido, nulli bene nupta marito, 
 
 Hoc fugiente peris, hoc pereunte fugis. — Auson. 
 
 κτασθαι μεν ώς χρωτο, χρησθαι Ι ε ώς τιμωτο to obtain that 
 he might use, and to use that he might be honoured. — 
 Ar. Met. iii. 9. 
 
 This sentence illustrates both antithesis, parisosis (balancing 
 of clauses), and paromoiosis (assimilation of endings). 
 
 The πάρα γράμμα σκώμμα or sudden pun, referable to anti- 
 thesis, is frequent in Aristophanes. A good example of this 
 σκώμμα is the verse 
 
 εκ κυμάτων γαρ αύθις αν γα\ην όρώ.^ 
 
 So in English, 
 
 1 Here the first < V oses of the year shall blow.' 
 
 * Hor. Od. in. xi. 35 ; cf. i. xxxiv. 2, ni. xvi. 28. 
 
 f The line in Euripides ( Or est. 279) ran γαλ-ην' = γαληνά ' calm ' — ' after 
 storm I see a calm,' but the actor did not pronounce so as to allow for 
 the elision, and it became a standing joke at Athens — 'out of the waves 
 I sec — a weasel ! ' 
 
FIGURES OF SPEECH. 213 
 
 The σκωμμα πάρα προσϊοα'αν corresponds in some measure to 
 the ' pleasantry by surprise ' of the (miscalled) Augustan a L 
 of English literature ; as 
 
 εστειχε δ' έχων νπο ποσσΧ . . . χίμετλα he was walkinf 
 having under his feet — chilblains. — Ar. Arist. Rhei. 
 iii. 6. 
 ' Where thou, great Anna, whom three realms obey, 
 Dost sometimes counsel take, and sometimes — tea ! ' 
 
 Pope. 
 
 e. Rhyme. The secret of the pleasurableness of Ehyme was 
 not unknown to the ancients, and it is found in many pas- 
 sages, an 
 
 'fiore εβιεα εισι μεΧισσάων αΖινάων, 
 
 ΐίέτρης εκ γΧαφυρης ahi νέον ερχομενάων. — Hom. II. ϋ.87. 
 
 Ccelum nitescere, arbores frondescere, 
 
 Vites lsetiflcae pampinis pubescere, 
 
 Kami baccarum ubertate incurvescere. 
 
 Ap. Cic. Tasc. Qitcpst. i. 69. 
 
 f. Ehythms. Occasionally an accidental verse, or a sentence 
 with the cadence of a verse, occurs in good writers, but this is 
 as much a defect as the blank-verse style of English prose. 
 
 πάσα Ιόσις αγαθή καϊ πάν ΰώρημα τέΧειον. — James i. 17. 
 και τροχιάς όρθάς ποιήσατε τόΊς ποσ\ν υμών. — Heb. xii. 13. 
 Auguriis patrum et prisca formidine sacram. — Tac. 
 
 Germ. 39. 
 Urbem Eomam a principio reges habuere. — T-dC.Ann. c. 1. 
 Cnaei Pompeii veteres fidosque clientes. — Sail. Cat. 19. 
 
 It will be readily understood that many figures of speech 
 are here designedly passed over as of secondary importance, 
 but the subject is one which will bear examination, and is 
 essential to the study of language as illustrating psychological 
 tendencies. 
 
 FINIS. 
 
 LONDON : PBIXTBD BY 
 
 SPOTTIS rVOODE AND CO., XEW-BTBEET' BQUAKI» 
 
 AND PAKLIAMENT STREET 
 
BY THE SAME AUTHOR. 
 
 Seventh Edition, in 8vo. price Is. 6d. 
 
 GREEK GRAMMAR RULES 
 
 DRAWN UP 
 
 FOR THE USE OF HARROW SCHOOL. 
 
 ' The Greek Grammar Rules drawn up for the use of Harrow Schoo 
 by a Harrow Tutor (the Rev. F. W. Farrar) are at once simple and 
 exhaustive. Seldom could a more apposite publication be found.' 
 
 John Bull. 
 
 'Mr. Farrar has managed to compress his Rules into the fewes 
 possible words, and at the same time to keep them free from ambi- 
 guities and technicalities. The examples seem well chosen ; the Editor 
 seems to have chosen the correct mean between too much a"d too little ; 
 and we think his tractate will be found an excellent companion and 
 guide to the ordinary Greek grammars in our public schools.' 
 
 Educational Times. 
 
 ' Mr. Farrar has hit upon an exceedingly happy idea in this little 
 book, and has carried it out with great skill. In teaching Latin or 
 Greek, the master's first concern should be to imprint the main inflexions 
 and the rules of syntax indelibly on the memory. Exceptions will be 
 easily remembered if the regular forms and laws are so thoroughly 
 learned that they cannot be forgotten, and the pupil can have no hesi- 
 tation in regard to them. If he is not absolutely and entirely master 
 of these regular forms the exceptions will perplex and confuse him. 
 And indeed the secret of success lies in selecting from the mass of 
 grammatical details just those points which form, as it were, the back- 
 bone of the grammar. Mr. Farrar' s work is a model of the kind of 
 book which should be thoroughly mastered. He gives as much of Greek 
 syntax as, if perfectly learned, will form a first-rate foundation. Nothing 
 essential is omitted. The Rules are arranged in a natural order, and 
 explanations are given which will rivet them on the memory. The 
 work bears traces, as might be expected, of a thorough knowledge of 
 comparative philology, and Mr. Farrar employs his rare knowledge of 
 English literature and modern languages to throw light on the Greek 
 idioms. The book deserves a hearty welcome from teachers and scholars.' 
 
 Museum. 
 
 London: LONGMANS, GREEN, and CO. 
 
WORKS BY THE SAME AUTHOR. 
 
 The Influence of Classical Studies on English Literature. 
 
 The Le Bas Prizo Essay. 1S-56. 
 
 The Christian Doctrine of the Atonement. 
 
 The Norrisian Prize Essay. 18-37. 
 
 Eric ; or, Little by Little. 
 
 Tenth Edition. 1857. 
 
 Julian Home. 
 
 Fourth Edition. 1857. 
 
 St Winifred's ; or, The World of School. 
 
 Fourth Edition. 1862. 
 
 The Origin of Language, based on Modern Researches. 
 
 I860. 
 
 Chapters on Language. 
 
 1865. 
 
 Greek Grammar Rules. 
 
 Seventh Edition. 1870. 
 
 The Tall of Man, and other Sermons, 
 
 Preached before the University of Cambridge, &c. 1868. 
 
 Seekers after God. 
 
 (Sunday Library. 1868.) 
 
 On Some Defects in Public School Education. 
 
 A Lecture delivered before the Royal Institution. 1867. 
 
 Essays on a Liberal Education, 
 
 By Various Writers. Edited by the Rev. F. W. F<u*RAR. Second 
 Edition. 1868. 
 
 Chapters on Language. 
 
 1865. 
 
 Families of Speech ; 
 
 Four Lectures delivered before the Pioyal Institution of Great Britain in 
 March 1869, and published by request. 
 
 W 44 l 
 

 
 
 ^ 
 
 
 ;-V% 
 
 : #ϊ7 •ν.• 
 
 •*»•* ^ 
 
 l°. *%£, <V* *JG Deacidified using the Bookkeeper proceg 
 V**V Γ i§l Neutralizing agent: Magnesium Oxide 
 
 Treatment Date: July 2006 
 
 «W\ -*§|£2i?• c^*^ °%/ Treatment Date: July 2006 
 
 * ** # **7^ % * 0^ ^ ° # *' PreservationTechnologie 
 
 om β *^ # * J^ . I. » • - *^r> A W0RLD LEADER IN PAPER PRESERVATIC 
 
 # _r*« *# *£ fO •.Lj*?*».* v3 111 Thomson Park Dnve 
 
 
 1 1 1 Thomson Park Dnve 
 Cranberry Township. PA 16066 
 (724)779-2111 
 
/.^V 
 
 
 
 »-o e -0 
 
 J• 4o A 
 
 4 
 
 
 ^ *•»ο 
 
 ν* 
 
 : x?V •: 
 
 i \f - 
 
 
 "ov^ :■ 
 
 
 v*.i^r.*« 
 
 **<* : 
 
 '.• -s^ 
 
 
 . .ο 4 A 0• 
 
 
 
 r* Λ^ 
 
 A * v ••Λ•».* 
 
 *♦ .tffefc V/ .•&&£>. %J 
 
 5?^ 
 
 , V*< 
 
 ^** νλ^/^\