SYNTAX

                                OF THE

                      MOODS and TENSES

                                 IN NEW TESTAMENT GREEK

 

 

                                                            By

                                    ERNEST DE WITT BURTON

                              President of the University of Chicago

                                                       1923-25

 

 

 

 

 

 

                        THE UNIVERSITY OF CHICAGO PRESS

                                    CHICAGO, ILLINOIS

 

 


 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

                                                COPYRIGHT 1900

                                          By ERNEST D. BURTON

                                                All Rights Reserved

                                      Published in pamphlet form 1888

                                       Second Edition September 1898

                                             Third Edition June 1898

                                    Second Impression September 1900

                                          Third Impression April 1903

                                       Fourth Impression October 1906

                                       Fifth Impression November 1909

                                         Sixth Impression October 1912

                                       Seventh Impression October 1916

                                      Eighth Impression November 1923

 

 

 

 

 

 

                                            Composed and Printed By

                                    The University of Chicago Press

                                                Chicago, Illinois, U.S.A.


 

 

 

                        PREFACE TO THE SECOND EDITION.

 

            THE first edition of this work appeared as a pamphlet in

1888. In issuing this revised and enlarged edition, it seems

desirable to state somewhat more fully than was done in the

former preface the purpose which it is hoped the book will

serve. Classified according to its intent, it belongs among the

aids to the interpretation of the New Testament. It is de-

signed to assist English-speaking students in the task of

translating the Greek New Testament into English forms of

hought and expression. The work has not been undertaken

under the impression that grammar is an end in itself, or that

a knowledge of it is the sole qualification for successful in-

terpretation, but in the conviction that grammar is one of

the indispensable auxiliaries of interpretation. The book is

written, therefore, in the interest not of historical but of

exegetical grammar, not of philology as such, but of philology

as an auxiliary of interpretation. If it has any value for

historical grammar, this is incidental. Its main purpose is

to contribute to the interpretation of the New Testament by

the exposition of the functions of the verb in New Testament

Greek, so far as those functions are expressed by the dis-

tinctions of mood and tense.

            The student of the New Testament who would interpret it

with accuracy and clearness must possess -along with other

qualifications for his work -a knowledge of the distinctions

of thought which are marked by the different moods and

tenses of the Greek verb. If he would acquire facility in the

work of interpretation, he must have an easy familiarity with

the leading uses of each mood and tense. It is not enough


vi                                             PREFACE.

 

that he have at hand for reference an encyclopedic treatise on

the subject. He must acquire, as a personal mental posses-

sion, a knowledge of the leading functions of the several

forms of the Greek verb, and of the forms which express

those functions in English. For this purpose he needs a book

which, availing itself of the assured results of comparative

and historical grammar, and applying to the interpretation of

the Greek verb the principles of grammar and logic, the laws

both of Greek and of English speech, shall enumerate the

various functions of each mood and tense, exhibit in some

degree their relative importance, and define each clearly.

The definitions should be scientifically accurate, but they

should at the same time be constructed with reference to the

point of view of the interpreter. For the English-speaking

student English usage must be constantly considered and

must frequently be defined and compared with Greek usage.

If such a book does not solve all the problems of New

Testament grammar, it should, by its treatment of those -which

it discusses, illustrate to the student the right method of

investigation and so suggest the course which he must pursue

in solving for himself those problems which the book leaves

unsolved. My aim has been to provide a book fulfilling these

conditions.

            The aim of the book has determined the method of its con-

struction. The usages which are of most frequent occurrence,

or otherwise of especial importance, have been emphasized by

being set in the largest type, with a title in bold-faced type.

The table of contents also has been so constructed as to make

prominent a conspectus of the leading uses. It may be well to

require of students who use the book as a text-book that they

be able to name and define these leading usages of each mood

and tense; if they also commit to memory one of the Greek

examples under each of these prominent usages, they will do

still better.

            The matter printed in smaller type consists partly of fuller

exposition of the usages defined in the more prominently


                                                PREFACE.                                        vii

 

printed sections, partly of enumeration and definition of the

less frequent usages. The portions in smallest type are

chiefly discussions of the rarer or more difficult usages. They

are an addition to the text-book proper, and are intended to

give the work, to a limited extent, the character of a book of

reference. The occasional discussions of English usage would

of course have no place in a work on Greek grammar pure

and simple, but to the end which this book is intended to

serve they are as really germane as any discussions of the

force of a Greek tense. One often fails to apprehend accu-

rately a thought expressed in Greek quite as much through

inexact knowledge of one's own language as through ignorance

of Greek usage.

            As concerns the extent to which I have used the work of

others, little need be added to the testimony which the pages

of the book themselves bear. While gathering information

or suggestion from all accessible sources, I have aimed to

make no statement concerning New Testament usage which I

have not myself proved by personal examination of the pas-

sages. Respecting classical usage and pre-classical origins, I

have relied upon those authorities which are recognized as

most trustworthy.

            On a subsequent page is added a list of books and authors

referred to by abbreviations in the body of the book. To all

of the works there enumerated, as well as to those mentione:d

by full title in the body of the book, I am under obligation for

assistance or suggestion. It is a pleasure also to acknowledge

the valuable assistance privately given by various friends.

Prominent among these, though not completing the list, are

Professor W. G. Hale of the University of Chicago, Profes-

sors M. L. D'Ooge and W. W. Beman of the University of

Michigan, my brother, Professor Henry F. Burton of the

University of Rochester, and Professor George W. Gilmore

of Brooklyn, N.Y. But I am chiefiy indebted to Professor

William Arnold Stevens of the Rochester Theological Semi-

nary, under whose instructions I first became interested in the


viii                                          PREFACE.

 

subject of this book, and to whom my obligations in many

directions are larger than can be acknowledged here.

            In quoting examples from the New Testament I have fol-

lowed the Greek text of Westcott and Rort as that which

perhaps most nearly represents the original text, but have

intended to note any important variations of Tischendorf's

eighth edition or of Tregelles in a matter affecting the point

under discussion. The word text designates the preferred

reading of the editor referred to, as distinguished from the

marginal reading. In the English translation of the examples

I have preferred to follow the Revised Version of 1881 rather

than to construct entirely independent translations. Yet in

not a few passages it has seemed necessary to depart from

this standard either because the revisers followed a Greek text

different from that of Westcott and Hort, or because their

translation obscured the value of the passage as an illustration

of the grammatical principle under discussion, or occasionally

because I was unwilling even to seem to approve what I

regarded as unquestionably an error of translation.

            While I have given all diligence to make the book correct

in statement and in type, I dare not hope that it has altogether

escaped either typographical errors or those of a more serious

character. I shall welcome most cordially criticisms, sugges-

tions, or corrections from any teacher or student into whose

hands the book may fall.

 

                                                                                    ERNEST D. BURTON

 

NOTE TO THE THIRD EDITION.--It having become necessary to send the

plates of this book to the press again, I have availed myself of the opportunity

to correct such errors, typographical and other, as "have come to my attention,

and to make a few alterations of statement which use of the book has convinced

me are desirable. The chief changes are in §§ 67 Rem. 1,98, 120, 137,142-145, 153, 189, 195, 198, 200 Rem., 202, 225, 235, 236, 318,325-328, 344 Rem. 2, 352

Rem., 406, 407,485.

            CHICAGO, June, 1898.                                                                  E. D. B.


 

                                                CONTENTS.

 

                                           INTRODUCTORY.

SECTION                                                                                                                  PAGE

            1. Form and Function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .            1

            2. The Interpreter's Relation to Grammar. . . . . . . . . . . . . . . . . .            2-5

            3, 4. The four Moods and the seven Tenses . . . . . . . . . . . . . . . .           5

 

                                                THE TENSES.

            5. Two-fold Function of the Tenses. . . . . . . . . . . . . . . . . . . . . . .          6

 

                        TENSES OF THE INDICATIVE MOOD.

            6, 7. General Definition of the Tenses of the Indicative . . . . . . . .        6, 7

 

                                    The Present Indicative.

 

8-10. PROGRESSIVE PRESENT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       7, 8

            11. Conative Present. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       8

            12. GENERAL OR GNOMIC PRESENT . . . . . . . . . . . . . . . . . . .      8

            13. AORISTIC PRESENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      9

            14. HISTORICAL PRESENT. . . . . . . . . . . . . . . . . . . . . . . . . . . . .      9

            15. PRESENT FOR THE FUTURE . . . . . . . . . . . . . . . . . . . . . . . .      9, 10

            16. Present of ἥκω, πάρειμι, etc. . . . . . . . . . . . . . . . . . . . . . . . . . .        10

            17. PRESENT OF PAST ACTION STILL IN PROGRESS. . . . . .     10

            18. Similar use of the Aorist. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     11

            19. Present in Indirect Discourse . . . . . . . . . . . . . . . . . . . . . . . . . . .     11

            20. Periphrastic Form of the Present. . . . . . . . . . . . . . . . . . . . . . . . .    11

 

                                    The Imperfect Indicative.

21, 22. PROGRESSIVE IMPERFECT. . . . . . . . . . . . . . . . . . . . . . . . . . . . .   12

            23. Conative Imperfect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   12

            24. IMPERFECT OF REPEATED ACTION. . . . . . . . . . . . . . . . . . .  12

25-27. Minor uses of Secondary Tenses . . . . . . . . . . . . . . . . . . . . . . . . . . . .   13

28, 29. Imperfect translated by English Perfect and Pluperfect  . . . . . . . . . . 13, 14

30-32. Imperfect of Verbs denoting obligation, etc. . . . . . . . . . . . . . . . . . . .  14, 15

            33. Imperfect of Verbs of wishing . . . . . . . . . . . . . . . . . . . . . . . . . . .   15, 16

            34. Periphrastic Form of the Imperfect. . . . . . . . . . . . . . . . . . . . . . . . 16

                                                            ix


x                                              CONTENTS.

 

                                         The Aorist Indicative.

SECTION                                                                                                                  PAGE

            35. Fundamental. Idea of the Aorist . . . . . . . . . . . . . . . . . . . . . . .         16, 17

            36. Additional uses of the Aorist Indicative . . . . . . . . . . . . . . . . .        17, 18

            37. Functions of the Aorist distinguished . . . . . . . . . . . . . . . . . . .        18, 19

            38-40. HISTORICAL AORIST. . . . . . . . . . . . . . . . . . . . . . . . . . .        19, 20

            41. INCEPTIVE AORIST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       20, 21

            42. RESULTATIVE AORIST . . . . . . . . . . . . . . . . . . . . . . . . . . . .       21

            43. GNOMIC AORIST. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       21

            44. EPISTOLARY AORIST. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      21

            45. DRAMATIC AORIST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       22

            46. Aorist for the (English) Perfect. . . . . . . . . . . . . . . . . . . . . . . . .       22

            47. Use of the Aorists ἀπέθανον, ἐξέστην, ἔγνων  . . . . . . . . . . .           22

            48. Aorist for the (English) Pluperfect . . . . . . . . . . . . . . . . . . . . . .      22, 23

            49. Aorist Indicative in Indirect Discourse . . . . . . . . . . . . . . . . . . .      23

            50. Aorist used proleptically . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      23

            51. Minor uses of the Aorist. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     23

52-55. English Equivalents of the Greek Aorist Indica-

                        tive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      23-30

56, 57. Distinction between the Aorist and the Imperfect. . . . . . . . . . . . .       30, 31

 

                                    The Future Indicative.

58-66. PREDICTIVE FUTURE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      31-35

            59. Aoristic Future. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      31

            60. Progressive Future. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      32

61, 62. Relation of Aoristic and Progressive Future. . . . . . . . . . . . . . . . . .      32, 38

63, 64. Types of Aoristic Future. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      33, 34

            65. Predictive Future as assertive or promissory. . . . . . . . . . . . . . .      34, 35

            66. Predictive Future with οὐ μή. . . . . . . . . . . . . . . . . . . . . . . . . . .      35

67, 68. IMPERATIVE FUTURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      35

            69.GNOMIC FUTURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     35

            70. DELIBERATIVE FUTURE. . . . . . . . . . . . . . . . . . . . . . . . . . . .     35

            71. Periphrastic Form of the Future . . . . . . . . . . . . . . . . . . . . . . . . .     36

72, 73. MiAAtI1 with the Infinitive. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     36, 37

 

                                    The Perfect Indicative.

            74. PERFECT OF COMPLETED ACTION. . . . . . . . . . . . . . . . . . .    37

75, 76. PERFECT OF EXISTING STATE. . . . . . . . . . . . . . . . . . . . . . . . . .   37,38

            77. Intensive Perfect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   38

            78. Historical Perfect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   38, 39

            79. Gnomic Perfect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   39




                                                CONTENTS.                                                xi

 

SECTION                                                                                                                  PAGE

            80. Aoristic Perfect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .               39

            81. Perfect Indicative in Indirect Discourse. . . . . . . . . . . . . . .              39

            82. Perfect Indicative translated by English Past. . . . . . . . . . .              39, 40

            83. Perfect used proleptically  . . . . . . . . . . . . . . . . . . . . . . . . .              40

            84. Periphrastic Form of the Perfect. . . . . . . . . . . . . . . . . . . . .              40

            85. Definition of the term "complete”  . . . . . . . . . . . . . . . . . . .             40, 41

86-88. Aorist and Perfect compared. . . . . . . . . . . . . . . . . . . . . . . . . . .             41-44

 

                                    The Pluperfect.

            89. PLUPERFECT OF COMPLETED ACTION. . . . . . . . . . .             44

            90. PLUPERFECT OF EXISTING STATE . . . . . . . . . . . . . . .            44, 45

            91. Periphrastic Form of the Pluperfect. ...45

            92. Pluperfect and Aorist siInilarly translated. . . . . . . . . . . . . .             45

           

                                    The Future Perfect.

            93. Simple Future Perfect. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .            45

            94. Periphrastic Future Perfect. . . . . . . . . . . . . . . . . . . . . . . . . .            45

           

                        TENSES OF THE DEPENDENT MOODS.

            95. General Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .             46

96, 97. PRESENT OF THE DEPENDENT MOODS. . . . . . . . . . . . . .           46

            98. AORIST OF THE DEPENDENT MOODS. . . . . . . . . . . . .           46, 47

99,100. FUTURE OF THE DEPENDENT MOODS . . . . . . . . . . . . . .           47,48

101-103. PERFECT OF THE DEPENDENT MOODS. . . . . . . . . . . . .          48,49

104-109. Tenses of the Infinitive after Prepositions. . . . . . . . . . . . . . . .          49-51

110-114. Tenses of the Dependent Moods in Indirect

                        Discourse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .          51-53

 

                        TENSES OF THE PARTICIPLE.

115-118. General Principles. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .         53, 54

 

                                    The Present Participle.

            119. PRESENT PARTICIPLE OF SIMULTANEOUS ACTION.       54,55

120-122. PRESENT PARTICIPLE OF IDENTICAL ACTION. . . . . . .         55,56

123-126. GENERAL PRESENT PARTICIPLE . . . . . . . . . . . . . . . . . . .        56-58

            127. PRESENT PARTICIPLE FOR THE IMPERFECT 58

128-131. Minor uses of the Present Participle . . . . . . . . . . . . . . . . . . . . .        58, 59


xii                                                       CONTENTS.

 

                                                     The Aorist Participle.

SECTION                                                                                                                  PAGE

132, 133. General Force of the Aorist Participle. . . . . . . . . . . . . . . . . . .         59-63

134-138. AORIST PARTICIPLE OF ANTECEDENT ACTION. . . . . .        63,64

139-141. AORIST P.ARTICIPLE OF IDENTICAL ACTION. . . . . . . .         64,65

142-145. AORIST PARTICIPLE OF SUBSEQUENT ACTION. . . . . .         65-67

            146. AORIST PARTICIPLE WITH THE OBJECT OF A VERB OF

                        PERCEPTION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .         67

            147. Aorist Participle with λανθάνω. . . . . . . . . . . . . . . . . . . . . .           67

148,149. Exceptional uses of the Aorist Participle. . . . . . . . . . . . . . . . . .        67, 68

150,151. Equivalence of the Aorist Participle . . . . . . . . . . . . . . . . . . . . .        68-70

 

                                    The Future Participle.

            152. GENERAL FORCE OF THE FUTURE PARTICIPLE. . . . .       70, 71

            153. Μέλλων with the Infinitive, denoting inten-

                        tion, etc. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      71

 

                                    The Perfect Participle.

154,155. GENERAL FORCE OF THE PERFECT PARTICIPLE . . . . . .      71, 72

            156. Perfect Participle used as a Pluperfect. . . . . . . . . . . . . . . . . . .      72

 

 

                                                THE MOODS.

                                    MOODS IN PRINCIPAL CLAUSES.

 

                                                The Indicative Mood.

            157. GENERAL FORCE OF THE INDICATIVE. 73

158,169. INDICATIVE IN QUALIFIED ASSERTIONS. 73,74

 

                                                The Subjunctive Mood.

160,161. HORTATORY SUBJUNCTIVE. 74,75

162-167. PROHIBITORY SUBJUNCTIVE. 75,76

168-171. DELIBERATIVE SUBJUNCTIVE. 76-78

172, 173. SUBJUNCTIVE IN NEGATIVE ASSERTIONS. 78

 

                                                The Optative Mood.

            174. Infrequency of the Optative in later Greek.. 79 '

175-177. OPTATIVE OF WISHING. 79

178, 179. POTENTIA.LOPTATIVE. " 79,80


                                                            CONTENTS.                                                xiii

 

                                                The Imperative Mood

SECTION                                                                                                                  PAGE

            180. IMPERATIVE IN COMMANDS AND EXHORTATIONS.       80

            181. IMPERATIVE IN ENTREATIES AND PETITIONS. . . . .         80

182, 183.IMPERATIVE TO EXPRESS CONSENT OR AN HYPOTHESIS 80, 81

            184. Tenses of the Imperative in Commands and Pro-

                        hibitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .        81

 

                        FINITE MOODS IN SUBORDINATE CLAUSES.

 

185-187. Subordinate Clauses Classified. 81-83

 

                        Moods in Clauses Introduced by Pinal Particles.

188,189. Classification and General Usage. . . . . . . . . . . . . . . . . . . . . . . .       83, 84

190-196. New Testament Use of Final Particles. . . . . . . . . . . . . . . . . . . .        84, 85

197-199. PURE FINAL CLAUSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       85, 86

200-204. OBJECT CLAUSES AFTER VERBS OF EXHORTING, etc. .      87, 88

205-210. OBJECT CLAUSES AFTER VERBS OF STRIVING, etc.             88-90

211-214. SUBJECT, PREDICATE, AND APPOSITIVE CLAUSES INTRO-

            DUCED BY  i!na. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      90, 91

215-217. COMPLEMENTARY AND EPEXEGETIC CLAUSES INTRO-

            DUCED BY i!na. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     91, 92

218-223. CLAUSES OF CONCEIVED RESULT INTRODUCED BY i!na  92-95

224-227. OBJECT CLAUSES AFTER VERBS OF FEAR AND DANGER 95,96

 

                                    Moods in Clauses of Cause.

            228. Definition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      97

229, 230. Moods and Tenses in Causal Clauses. . . . . . . . . . . . . . . . . . . . .      97

231,232. Independent Causal Sentences. . . . . . . . . . . . . . . . . . . . . . . . . . .      98

            233, Other Methods of Expressing Cause . . . . . . . . . . . . . . . . . . . .     98

            234. Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     99

            235. Distinction between Indicative and Infinitive in Con-

                        secutive Clauses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

            236. Indicative with ὥστε . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99, 100

            237. Independent Consecutive Sentences. . . . . . . . . . . . . . . . . . . . . . 100

 

                        Moods in Conditional Sentences.

238-241. Definition and Classification. . . . . . . . . . . . . . . . . . . . . . . . . . .100, 101

242-247. SIMPLE PRESENT OR PAST PARTICULAR SUPPOSITION 102,103


xiv                                                       CONTENTS.

 

SECTION                                                                                                           PAGE

248, 249. SUPPOSITION CONTRARY TO FACT. . . . . . . . . . . . . .      103,104

250. FUTURE SUPPOSITION WITH MORE PROBABILITY . . . .      104

251-256. Variant Forms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      104, 105

            257. Particular and General Suppositions referring

                        to the Future . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      106

            258. Present and Future Suppositions in Indirect

            Discourse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      106

259. FUTURE SUPPOSITION WITH LESS PROBABILITY   . . . . .      106, 107

260, 261. PRESENT GENERAL SUPPOSITION. 107 108

262, 263. Third and Fifth Classes compared. . . . . . . . . . . . . . . . . . . . .       108

264. First and Fifth Classes compared. . . . . . . . . . . . . . . . . . . . . . . . . .       109

265. [PAST GENERAL SUPPOSITION]. . . . . . . . . . . . . . . . . . . . . .        109

266-277. Peculiarities of Conditional Sentences. . . . . . . . . . . . . . . . . .       109-112

Moods in Concessive Sentences.

            278. Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       112, 113

279-282. Ἐι καί and καὶ εἰ in Concessive Clauses. . . . . . . . . . . . . . . .        113, 114

            283. General Usage of Moods and Tenses in Con-

                        cessive Clauses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       114

284. CONCESSIVE CLAUSES OF THE FIRST CLASS. . . . . . . . . . .       114

285. CONCESSIVE CLAUSES REFERRING TO THE FUTURE. . .       114, 115

286. Concessive Clauses of the Fourth Class. . . . . . . . . . . . . . . . . . . . .      115

287. Concessive Clauses of the Fifth Class . . . . . . . . . . . . . . . . . . . . . .      115

288. Concessive Particles in English. . . . . . . . . . . . . . . . . . . . . . . . . . .       115, 116

 

                                    Moods in Relative Clauses.

289-291. Definition and Classification.. . . . . . . . . . . . . . . . . . . . . . . . .       116, 117

 

                        I. DEFINITE RELATIVE CLAUSES.

            292. Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       117, 118

            293. Moods in Definite Relative Clauses . . . . . . . . . . . . . . . . . .      118

            294. Definite Relative Clauses implying cause, result,

                        or concession. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      118

295. Restrictive and Explanatory Relative Clauses. . . . . . . . . . . . . . . . .      119

 

            II. CONDITIONAL RELATIVE SENTENCES.

296-300. Definition and Classification. . . . . . . . . . . . . . . . . . . . . . . . . .       119-121

301. SIMPLE PRESENT OR PAST PARTICULAR SUPPOSITION .       121

302. [SUPPOSITION CONTRARY TO FACT] . . . . . . . . . . . . . . . . . . . . .  121


                                                CONTENTS.                                                            xv

 

SECTION                                                                                                            PAGE

303-305. FUTURE SUPPOSITION WITH MORE PROBABILITY.      121, 122

306-309. Variant Forms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . .      122,123

310. Particular and General Suppositions referring

            to the future. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     123

311. [FUTURE SUPPOSITION WITH LESS PROBABILITY]. . . . .      123

312-314. PRESENT GENERAL SUPPOSITION . . . . . . . . . . . . . . . .      123,124

315. PAST GENERAL SUPPOSITION. . . . . . . . . . . . . . . . . . . . . . . .      124, 125

316. Clauses conditional in form, but definite in sense . . . . . . . . . . . .       125

 

                        III. RELATIVE CLAUSES EXPRESSING PURPOSE.

317. RELATIVE CLAUSES OF PURE PURPOSE. . . . . . . . . . . . . . .      125

318-320. Complementary Relative Clauses. . . . . . . . . . . . . . . . . . . . . .     125, 126

 

IV. RELATIVE CLAUSES INTRODUCED BY WORDS MEANING

                        UNTIL, WHILE, AND BEFORE.

321. Definition of  ἕως . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      126, 127

322, 323. CLAUSES INTRODUCED BY ἕως AND REFERRING TO THE

                        FUTURE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       127

324-326. CLAUSES INTRODUCED BY ἕως AND REFERRING TO

            WHAT WAS IN PAST TIME A FUTURE CONTINGENCY . .    127, 128

327. CLAUSES INTRODUCED BY ἕως (UNTIL), AND REFER-

                        RING TO A PAST FACT. . . . . . . . . . . . . . . . . . . . . . . . . .    128

328, 329. CLAUSES INTRODUCED BY e!wj (WHILE), AND REFER-

            RING TO A CONTEMPORANEOUS EVENT. . . . . . . . . . . . . . .    128

330.  !Ewj followed by οὖ or ὅτου . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   128, 129

331, 332. Clauses introduced by ἄχρι, ἄχρι οὗ etc. . . . . . . . . . . . . . . . .   129

333. Clauses introduced by πρίν . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .    129

 

                                    Moods in Indirect Discourse.

334-340. Definition and Classification. . . . . . . . . . . . . . . . . . . . . . . . . . . .   130-132

341, 342. Classical Usage in Indirect Discourse. . . . . . . . . . . . . . . . . . . . .   132

343-346. New Testament Usage in Indirect Discourse. . . . . . . . . . . . . . . .  132-134

347. Single dependent Clauses in Indirect Discourse . . . . . . . . . . . . . . . . .   134

348. Imperfect for Present, and Pluperfect for Per-

            fect in Indirect Discourse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   134, 135

349, 350. Relative Pronouns in Indirect Discourse. . . . . . . . . . . . . . . . . .   135

351-356. Indirect Discourse in English and in Greek

 

                                    Construction after Καὶ ἐγένετο

357-360. Three Forms of the Idiom. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .    142,143


xvi                                           CONTENTS.

 

                                                THE INFINITIVE.

SECTION                                                                                                            PAGE

361-363. Origin, and Classification of Uses. . . . . . . . . . . . . . . . . . .           143-145

 

                                    The Infinitive without the Article.

364, 365. IMPERATIVE INFINITIVE. . . . . . . . . . . . . . . . . . . . . . .          146

366, 367. INFINITIVE OF PURPOSE . . . . . . . . . . . . . . . . . . . . . . .          146

368. INFINITIVE AS AN INDIRECT OBJECT. . . . . . . . . . . . . . . .          147

 

369-371. INFINITIVE OF RESULT. . . . . . . . . . . . . . . . . . . . . . . . .           147-150

372-374. Exceptional usages. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .           150

375. INFINITIVE DEFINING CONTENT OF A PREVIOUS VERB

            OR NOUN. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .          150, 151

376, 377. INFINITIVE LIMITING ADJECTIVES AND ADVERBS.      151

378, 379. INFINITIVE LIMITING NOUNS. . . . . . . . . . . . . . . . . . . .        151, 152

380-382. INFINITIVE AFTER πρίν or πρίν     . . . . . . . . . . . . . . . . .        152

383. INFINITIVE USED ABSOLUTELY. . . . . . . . . . . . . . . . . . . . . .        153

384, 385. INFINITIVE AS SUBJECT. . . .. . . . . . . . . . . . . . . . . . . . . .        153

386. INFINITIVE AS APPOSITIVE . . . . . . . . . . . . . . . . . . . . . . . . . .        153

387-389. INFINITIVE AS OBJECT. . . . . . . . . . . . . . . . . . . . . . . . . . .       153, 154

390. Infinitive in Indirect Discourse. . . . . . . . . . . . . . . . . . . . . . . . . . . .      154, 155

391. Infinitive after verbs of hoping, promising, swear-

            ing, commanding, etc. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .        155

 

                                    The Infinitive with the Article.

392. General Use of Infinitive with the Article. . . . . . . . . . . . . . . . . . .       155, 156

393. INFINITIVE WITH τό AS SUBJECT. . . . . . . . . . . . . . . . . . . . .       156

394. INFINITIVE WITH τό AS OBJECT. . . . . . . . . . . . . . . . . . . . . .       156

395. INFINITIVE WITH THE ARTICLE, IN APPOSITION. . . . . . .       156, 157

396. INFINITIVE WITH τῷ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       157

397. INFINITIVE OF PURPOSE WITH τοῦ  . . . . . . . . . . . . . . . . . . .       157

398. INFINITIVE OF RESULT WITH τοῦ . . . . . . . . . . . . . . . . . . . . .      157, 158

399. INFINITIVE WITH τοῦ  AFTER ADJECTIVES. . . . . . . . . . . . .       158

400. INFINITIVE WITH τοῦ  AFTER NOUNS. . . . . . . . . . . . . . . . . .       158

401. INFINITIVE WITH τοῦ  AFTER VERBS THAT TAKE THE

            GENITIVE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       158, 159

402, 403. Various constructions after Verbs of hindering  . . . . . . . . . .       159

404, 405. INFINITIVE WITH τοῦ  AS SUBJECT OR OBJECT. . . . .       159, 160

406-417. INFINITIVE WITH THE ARTICLE GOVERNED BY PREPO-

            SITIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       160-163


                                                CONTENTS.                                                xvii

 

                                                THE PARTICIPLE.

SECTION                                                                                                      PAGE

418. General Nature of the Participle. . . . . . . . . . . . . . . . . . . . . .          163

419. Classification respecting logical force . . . . . . . . . . . . . . . . .         163, 164

 

                                                The Adjective Participle.

420, 421. Definition and Classification. . . . . . . . . . . . . . . . . . . . .        164

422. RESTRICTIVE ATTRIBUTIVE PARTICIPLE. . . . . . . . . .        164, 165

423. Restrictive Attributive Participle with Subject

            omitted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       165

424. Noun without the article limited by a Participle

            with the article. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       165

425.. Neuter Participle with the article equivalent to an

            abstract Noun. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       166

426. EXPLANATORY ATTRIBUTIVE PARTICIPLE . . . . . . . .       166

427. Order of words with Attributive Participle

            limiting a Noun with the article. . . . . . . . . . . . . . . . . . . . . .      166, 167

428. Attlibutive Participle conveying a subsidiary idea

            of cause, purpose, etc. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .       167

429, 430. PREDICATIVE ADJECTIVE PARTICIPLE. . . . . . . . .      167

431. Predicative Participle used to form periphrastic

            tenses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     168

432, 433. Participles in Predicate in various construc-

            tions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .     168, 169

 

                                    The Adverbial Participle.

434. Definition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  . . . . .     169

435. ADVERBIAL PARTICIPLE OF TIME. . . . . . . . . . . . . . . . . .    169

436. ADVERBIAL PARTICIPLE OF CONDITION.  . . . . . . . . . .     169

437,438. ADVERBIAL PARTICIPLE OF CONCESSION. .  . . . .     170

439. ADVERBIAL PARTICIPLE OF CAUSE. . . . . . . . . . . . . . . .     170

440, 441. Participle of Cause with ὠς . . . . . . . . . . . . . . . . . . . . . . .      170, 171

442. ADVERBIAL PARTICIPLE OF PURPOSE . . . . . . . . . . . . . .  171

443. ADVERBIAL PARTICIPLE OF MEANS. . . . . . . . . . . . . . . .   171

444. ADVERBIAL PARTICIPLE OF MANNER. . . . . . . . . . . . . .     171

445, 446. Ὡς with the Participle denoting Manner . . . . . . . . . . . . .      172

447. Participle of Manner or Means denoting same

            action as that of the principal Verb. . . . . . . . . . . . . . . . . . . . .    172, 173

448. Intensive Participle-Hebraistic . . . . . . . . . . . . . . . . . . . . . . . . .    173


xviii                                        CONTENTS.

 

SECTION                                                                                                      PAGE

 

449, 450. ADVERBUL PARTICIPLE OF ATTENDANT CIRCUM.

                        STANCE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .        173,174

451. More than one adverbial relation implied by

                        the same Participle. . . . . . . . . . . . . . . . . . . . . . . . . .        174

452-454. Genitive Absolute. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .        174, 175

455. Position of Adverbial Participle. . . . . . . . . . . . . . . . . . . . . . .        175

 

                                    The Substantive Participle.

456. Definition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      175

457. SUBSTANTIVE PARTICIPLE AS SUBJECT. . . . . . . . . . . .     175

458,459. SUBSTANTIVE PARTICIPLE AS OBJECT. . . . . . . . . .    176

460. Substantive Participle in Indirect Discourse. . . . . . . . . . . . . . .     176

461. SUBSTANTIVE PARTICIPLE AS A LIMITING GENITIVE   176

462. Position of Substantive Participle. . . . . . . . . . . . . . . . . . . . . . .     177

463. Substantive Participle distinguished from Ad.

jective Participle used substantively. . . . . . . . . . . . . . . . . . . . . . . . .     177

 

                        THE USE OF NEGATIVES WITH VERBS.

464. General Usage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .      178

 

                        NEGATIVES WITH THE INDICATIVE.

465. Negatives in Independent declaratory Sentences. . . . . . . . . . . .   178

466. Negatives with a Prohibitory Future. . . . . . . . . . . . . . . . . . . . . .   179

467. Negatives in Questions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   179

468. Μὴ οὐ in Rhetorical Questions. . . . . . . . . . . . . . . . . . . . . . . . .     179

469, 470. Negatives in Conditional and Conditional Relative

            Clauses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .   179, 180

471. Εἰ μή in the sense of except. . . . . . . . . . . . . . . . . . . . . . . . . .         180

472. Οὐ after μή as a conjunction. . . . . . . . . . . . . . . . . . . . . . . . .         181

473. Negatives in Indirect Discourse. . . . . . . . . . . . . . . . . . . . . . .        181

474. Negatives in Causal Clauses and in simple Relative

            Clauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .        181

 

            NEGATIVES WITH THE SUBJUNCTIVE, OPTATIVE,

                                    AND IMPERATIVE.

475. Negatives with the Subjunctive. . . . . . . . . . . . . . . . . . . . . . . . .    181, 182

476,477. Negatives with the Optative. . . . . . . . . . . . . . . . . . . . . . . .    182

479. Negatives with the Imperative. . . . . . . . . . . . . . . . . . . . . . . . . .    182, 183

 

 

           
                                                CONTENTS.                                                xix

 

            NEGATIVES WITH THE INFINITIVE AND PARTICIPLE.

SECTION                                                                                                       PAGE

480. General Usage of Negatives with the Infinitive. . . . . . . . . .          183

481. Negatives with a limitation of an Infinitive or of its

            subject. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .          183,184

482. Compound of οὐ with an Infinitive dependent on a

            principal verb limited by ou] . . . . . . . . . . . . . . . . . . . . . . .         184

483. Redundant μή with Infinitive after verbs of hinder-

            ing, denying, etc. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .         184

484. Negative with Infinitive dependent on a verb itself

            egatived by οὐ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .             184

485. General Usage of Negatives with the Participle. . . . . . . .              184, 185

 

                        SUCCESSIVE AND DOUBLE NEGATIVES.

486. Two simple Negatives, or a compound Negative fol-

            lowed by a simple Negative. . . . . . . . . . . . . . . . . . . . . . . .         185

487, 488. Double Negative οὐ μή. . . . . . . . . . . . . . . . . . . . . . . . .         185, 186

489. Negative followed by similar compound Negative or

            double Negative. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .         186




                                    LIST OF WORKS AND AUTHORS

 

                                    REFERRED TO BY ABBREVIATION.

 

A.J.P. . . . . .   American Journal of Philology.

Alf.  . . . . . .    Henry Alford, The Greek New Testament. 4 vols. Lon-

                                    don.

A. V.  . . . . .   Authorized Version of the New Testament.

B. . . . . . . . .   Alexander Buttmann, A Grammar of the New Testament

                                    Greek. Translated by J. H. Thayer. Andover, 1873.

Bib. Sac. . . .  Bibliotheca Sacra.

Br. . . . . . . . . Karl Brugmann, Griechische Grammatik, in Iwan Mut-

                                    ler's Handbuch der klassischen Altertumswissenschaft,

                                    vol. II. Second Edition. Miinchen, 1890.

Cl. Rev. . . . .  Classical Review.

Del. . . . . . . . B. Delbruck, Syntaktische Forschungen. Halle, 1871-

                                    1888.

Ev. Pet.. . . .   Apocryphal Gospel of Peter. (Verses according to the

                                    edition of Harnack, Leipzig, 1893.)

G. . . . . . . . . W. W. Goodwin, A Greek Grammar. Revised Edition.

                                    Boston, 1892.

Gild. . . . . . .  Basil L. Gildersleeve, various papers in A.J.P. and

                                    T.A.P.A.

G.MT. . . . . . W. W. Goodwin, Syntax of the Moods and Tenses of the

                                    Greek Verb. Revised and enlarged. Boston, 1889.

Gr. . . . . . . . . Thomas Sheldon Green, A Treatise on the Grammar of

                                    the New Testament. New Edition. London, 1862.

HA. . . . . . . . James Hadley, A Greek Grammar for Schools and Col-

                                    leges. Revised by F. D. Allen. New York, 1884.

Hr. . . . . . . . . W. R. Harper, Elements of Hebrew Syntax. New York,

                                    1888.

J. . . . . . . . .    J. W. E. Jelf, A Grammar of the Greek Language. Third

                                    Edition. 2 vols. Oxford and London, 1861.

J.B.L. . . . . . . Journal of the Society of Biblical Literature and Exegesis.

K. . . . . . . . . . Raphael Kuhner, Grammatik der Griechischen Sprache.

                                    Hanover, 1869-1872.

Ka. . . . . . . . . E. Kautzsch, Grammatik des Biblisch-Aramaischen.

                                    Leipzig, 1884.

                                                xxi


xxii                 LIST OF WORKS AND AUTHORS.

 

L. and S. . . . Liddell and Scott, Greek-English Lexicon, etc. Seventh

                                    Edition. New York, 1882.

Ltft. . . . . . . .  J. B. Lightfoot, Commentaries on Galatians, on Philip-

                                    pians, and on Colossians and Philemon.

Mart. Polyc. . Martyrium Polycarpi. (See any edition of the Apostolic

                                    Fathers. )

Meist. . . . . .   K. Heisterhans, Grammatik der Attischen Inschriften.

                                    Berlin, 1885.

Mey. . . . . . .  H. A. W. Meyer, Kommentar iiber das Neue Testament.

                                    Gottingen, 1867-1876. English Translation, Edinburgh,

                                    1873-1880.

Ps. Sol. . . . . The Psalms of Solomon. (Recent edition by Ryle and

                                    James, Cambridge, 1891.)

R. V. . . . . . .  The New Testament in the Revised Version of 1881.

S. . . . . . . . . . W. H. Simcox, The Language of the New Testament.

                                    London and New York, 1889.

Th. . . . . . . . . J. H. Thayer, A Greek-English Lexicon of the New

                                    Testament: Being Grimm's Wilke's Clavis Novi Testa-

                                    menti, translated, revised, and enlarged. New York,

                                    1886.

Tisch. . . . . .   Constantinus Tischendorf, Novum Testamentum Graece.

                                    Eighth Edition. 2 vols. Leipzig, 1869-72.

Treg. . . . . . .  S. P. Tregelles, The Greek New Testament. London,

                                    1857-79.

T .A.P .A.. .   Transactions of the American Philological Association.

W. . . . . . . . .  G. B. Winer. See WE and WT.

WH. . . . . . . . Westcott and Hort, The New Testament in the Original

                                    Greek, the text revised by B. F. Westcott and F. J. A.

                                    Hort. 2 vols. Cambridge and New York, 1881.

WT . . . . . . .  G. B. Winer, A Treatise on the Grammar of New Testa-

                                    ment Greek. Translated by W. F. Moulton. Third

                                    Edition. Edinburgh, 1882.

WT. . . . . . . . G. B. Winer, A Grammar of the Idiom of the New Tes-

                                    tament. Seventh Edition, enlarged and improved by

                                    Gottlieb Liinemann. Revised and authorized Trans-

                                    lation by J. H. Thayer. Andover, 1869.

WS. . . . . . . . G. B. Winer's Grammatik des neutestamentlichen Sprach-

                                    idioms, Achte Auflage, neu bearbeitet von D. Paul Wilh.

                                    Schmiedel, Gottingen, 1894- (in process of publication).

 

For classical and Scripture writers the ordinary abbreviations are used.

References to the Old Testament are to the Septuagint Version, unless

otherwise indicated.


 

 

                                                SYNTAX

                                                 OF THE

            MOODS AND TENSES IN NEW TESTAMENT GREEK.

 

                                          INTRODUCTORY

 

            1. FORM AND FUNCTION. The following pages deal with

the various functions of the various verb-forms of the Greek

of the New Testament, so far as respects their mood and

tense. It is important that the nature of the relation between

form and function be clearly held in mind. It is by no means

the case that each form has but one function, and that each

function can be discharged by but one form. Forms of various

origin may be associated together under one name and perform

the same function, or group of functions. Compare, e.g., the

Aorist Active Infinitives, λῦσαι and εἰπεῖν : these forms are of

quite diverse origin; in function they have become entirely

assimilated. The same is true of the Aorist Active Indicatives,

ἔδειξα and ἔστην. Forms also which still have different names,

and usually perform different functions, may have certain

functions in common. Compare the Aorist Subjunctive and

the Future Indicative in clauses of purpose (197, 198). On

the other hand, and to an even greater extent, we find that a

given form, or a given group of forms bearing a common name,

performs various distinct functions. Observe, e.g., the various

functions of the Aorist Indicative (38-48).

                                                            1


2                                  INTRODUCTORY.

 

            The name of a given form, or group of forms, is usually

derived from some prominent function of the form or group.

Thus the term Aorist reflects the fact that the forms thus

designated most frequently represent an action indefinitely

without reference to its progress. The name Present suggests

that the forms thus designated denote present time, which is

true, however, of the smaller part only of those that bear the

name, and of none of them invariably. The name Optative

again reminds us that one function of the forms so named is

to express a wish. While, therefore, the names of the forms

were originally intended to designate their respective func-

tions, they cannot now be regarded as descriptive of the actual

functions, but must be taken as conventional, and to a con-

siderable extent arbitrary, names of the forms. The functions

must be learned, not from the names, but from observation of

the actual usage.

 

2. THE INTERPRETER'S RELATION TO GRAMMAR. Both the

grammarian as such and the interpreter deal with grammar, but

from very different points of view. The distinction between

these points of view should be clearly recognized by the in-

terpreter. It may be conveniently represented by the terms

historical grammar and exegetical grammar. Historical gram-

mar deals with the development of both form and function

through the various periods of the history of the language,

and does this in purely objective fashion. Exegetical grammar,

on the other hand, takes the forms as it finds them, and defines

the functions which at a given period each form discharged,

and does this from the point of view of the interpreter, for

the purpose of enabling him to reproduce the thought con-

veyed by the form. To investigate the process by which the

several forms were built up, to determine the earliest function

of each such form, to show how out of this earliest function

 


                                    INTRODUCTORY.                                     3

 

others were developed, and how forms of different origin, and

presumably at first of different function, became associated,

discharging the same function and eventually coming to bear

the same name -all this belongs to historical grammar. To

reproduce in the mind of the interpreter, and to express as

nearly as may be in his own tongue, the exact thought

which a given form was in the period in question capable of

expressing -this is the task of exegetical grammar. Histori-

cal grammar views its problem wholly from the point of view

of the language under investigation, without reference to the

language of the grammarian. Exegetical grammar is neces-

sarily concerned both with the language under investigation

and with that in which the interpreter thinks and speaks,

since its problem is to aid in reproducing in the latter tongue

thought expressed in the former.

            The results of historical grammar are of the greatest interest

and value to exegetical grammar. Our interpretation of the

phenomena of language in its later periods can hardly fail to

be affected by a knowledge of the earlier history. Strictly

speaking, however, it is with the results only of the processes

of historical grammar that the interpreter is concerned. If

the paradigm has been rightly constructed, so that forms of

diverse origin perhaps, but completely assimilated in function,

bear a common name, exegetical grammar is concerned only to

know what are the functions which each group of forms bear-

ing a common name is capable of discharging. Thus, the

diversity of origin of the two Aorists, ἔλυσα and ἔλιπον, does

not immediately concern the interpreter, if it is an assured

result of historical grammar that these two forms are com-

pletely assimilated in function. N or does it concern him that

the ai at the end of the Infinitives, δεῖξαι and ἰέναι, is the mark

of the Dative case, and that the earliest use of such infinitives

was as a verbal noun in the Dative case, except as this fact


4                                  INTRODUCTORY.

 

of historical grammar aids him in the interpretation of the

phenomena of that period of the language with which he is

dealing. The one question of exegetical grammar to which

all other questions are subsidiary is, What function did this

form, or group of forms, discharge at the period with which

we are dealing? What, e.g., in the New Testament, are the

functions of the Present Indicative? What are the uses of

the Aorist Subjunctive?

            For practical convenience forms are grouped together, and

the significance of each of the distinctions made by inflection

discussed by itself. The present work confines itself to the

discussion of mood and tense, and discusses these as far as

possible separately. Its question therefore is, What in the

New Testament are the functions of each tense and of each

mood? These various functions must be defined first of all

from the point of view of the Greek language itself. Since,

however, the interpreter whom in the present instance it is

sought to serve thinks in English, and seeks to express in

English the thought of the Greek, reference must be had

also to the functions of the English forms as related to

those of the Greek forms. Since, moreover, distinctions of

function in the two languages do not always correspond,

that is, since what in Greek is one function of a given form

may be in English subdivided into several functions per-

formed by several forms, it becomes necessary not only to

enumerate and define the functions of a given form purely

from the point of view of Greek, but to subdivide the one

Greek function into those several functions which in English

are recognized and marked by the employment of different

forms. An enumeration of the uses of a given Greek tense

made for the use of an English interpreter may therefore

properly include certain titles which would not occur in a

list made for one to whom Greek was the language of


                                    INTRODUCTORY.                                      5

 

ordinary speech and thought. The Aorist for the English

Perfect, and the Aorist for the English Pluperfect (46, 48)

furnish a pertinent illustration. The interests of the English

interpreter require that they be clearly recognized. Fidelity

to Greek usage requires that they be recognized as, strictly

speaking, true Historical Aorists.

            3. The Greek verb has four moods,-- the Indicative, the

Subjunctive, the Optative, and the Imperative. With these

are associated in the study of Syntax the Infinitive, which is,

strictly speaking, a verbal noun, and the Participle, which is

a verbal adjective.

            The Subjunctive, Optative, Imperative, and Infinitive are

often called dependent .moods.

 

            REM. The term dependent is not strictly applicable to these moods,

    and least of all to the Imperative, which almost always stands as a prin-

    cipal verb. It has, however, become an established term, and is retained

    as a matter of convenience.

 

4. There are seven tenses in the Greek, -- the Present,

Imperfect, Aorist, Future, Perfect, Pluperfect, and Future

Perfect.

            Those tenses which denote present or future time are called

Primary tenses. Those tenses which denote past time are

called Secondary tenses. Since the time denoted by a tense

varies with the particular use of the tense, no fixed line of

division can be drawn between the two classes of tenses. In

the Indicative the Present and Perfect are usually, and the

Future and Future Perfect are always, Primary tenses; the

Imperfect, Aorist, and Pluperfect are usually Secondary

tenses.

 


 

 

 

 

                        THE TENSES.

 

5. The action denoted by a verb may be defined by the tense

of the verb

            (a) As respects its progress. Thus it may be represented

as in progress, or as completed, or indefinitely, i.e. as a simple

event without reference to progress or completion.

            (b) As respects its time, as past, present, or future.

The tenses of the Indicative mood in general define the

action of the verb in both these respects.

            The tenses of the other moods in general define the action

of the verb only as respects its progress. HA. 821; G. 1249.

           

            REM. The chief function of a Greek tense is thus not to denote time,

but progress. This latter function belongs to the tense-forms of all the

moods, the former to those of the Indicative only.

 

                        TENSES OF THE INDICATIVE MOOD.

 

            6. The significance of the tenses of the Indicative mood

may be stated in general as follows: --

            As respects progress: The Present and Imperfect denote

action in progress; the Perfect, Pluperfect, and Future Perfect

denote completed action; the Aorist represents the action

indefinitely as an event or single fact; the Future is used

either of action in progress like the Present, or indefinitely

like the Aorist.

            As respects time: The Present and Perfect denote present

time; the Imperfect, Aorist, and Pluperfect denote past time;

the Future and Future Perfect denote future time.

                                                6


                        THE PRESENT INDICATIVE                               7

 

            7. The tenses of the Indicative in general denote time rela-

tive to that of speaking. Most exceptions to this rule are

apparent or rhetorical rather than real and grammatical. In

indirect discourse the point of view, as respects time, of the

original speaking or thinking is retained. Cf. 351. Of two

verbs of past time, one may refer to an action antecedent to

the other, but this fact of antecedence is implied in the con-

text, not expressed in the tense. Cf. 29 and 48. By prolepsis

also a verb of past time may refer to or include events to take

place after the time of speaking, but before a point of future

time spoken of in the context. Cf. 50. In conditional sen-

tences of the second form, the tenses are properly timeless.

Cf. 248. See Br. 154 (p. 180).

 

                        THE PRESENT INDICATIVE.

 

            8. The Progressive Present. The Present Indicative

is used of action in progress in present time. EA. 824;

G. 1250, 1.

Matt. 25:8; αἰ λαμπάδες ἡμῶν σβέννυνται, our lamps are going out

Gal. 1:6;  θαυμάζω ὅτι οὕτως ταχέως μετατίθεστε ἀπὸ τοῦ καλέσαν-
τος ὑμᾶς, I marvel that ye are so quickly removing from him that called

you.

            9. The most constant characteristic of the Present Indica-

tive is that it denotes action in progress. It probably had

originally no reference to present time (see Br. 156). But

since, in the" historical periods of the language, action in

progress in past time is expressed by the Imperfect, and the

Future is used both as a progressive and as an aoristic tense

for future time, it results that the Present Indicative is chiefly

used to express action in progress in present time. Hence

in deciding upon the significance of any given instance of the

Present Indicative in the New Testament as well as in classi-


8                                              THE TENSES.

 

cal Greek, the interpreter may consider that there is, at least

in the majority of words, a certain presumption in favor of

the Progressive Present rather than any of the other uses

mentioned below.

            10. The Progressive Present in Greek is not always best

translated by what is commonly called in English the "Pro-

gressive Form." Some English verbs themselves suggest

action in progress, and do not, except when there is strong

emphasis on the progressive idea, use the progressive form.

Thus the verb (θαυμάζω), in Gal. 1:6, is a Progressive Present,

but is best translated I marvel, the verb itself sufficiently sug-

gesting the idea of action in progress.

 

            11. THE CONATIVE PRESENT. The Present Indicative is

occasionally used of action attempted, but not accomplished.

H.A. 825; G. 1255. This use is, however, not to be re-

garded as a distinct function of the tense. The Conative

Present is merely a species of the Progressive Present. A

verb which of itself suggests effort, when used in a tense

which implies action in progress, and hence incomplete, natu-

rally suggests the idea of attempt. All the verb-forms of the

Present system are equally, with the Present, capable of

expressing attempted action, since they all denote action in

progress. John 10:32, λιθάζετε, and Gal. 5:4, δικαιοῦσθε, illus-

trate this usage in the Present. Similar is the use of the

Present in Rom. 2:4, ἄγει, leadeth, i.e. such is its tendency.

            For examples of the Imperfect see 23. Respecting the

resultative force of such verbs in the Aorist see 42.

 

            12. The General or Gnomic Present. The Present

Indicative is used to express customary actions and general

truths. EA. 824, a; G. 1253, 1291.

Matt. 7:17; πᾶν δένδρον ἀγαθὸν καρποὺς ποιεῖ, every good tree

bringeth forth good fruit.


                                    THE PRESENT INDICATIVE.                            9

 

2 Cor. 9:7; ἱλαρὸν γὰρ δότην ἀγαπᾷ θεός,  for God loveth a cheerful

            giver.

 

            13. The Aoristic Present. The Present Indicative is

sometimes used of an action or event coincident in time

with the act of speaking, and conceived of as a simple

event. Most frequently the action denoted by the verb

is identical with the act of speaking itself, or takes place

in that act.

 

Acts 16:18; παραγγέλλω σοι ἐν ὀνόματι Ἰησοῦ Χριστοῦ, I command

            thee in the name of Jesus Christ. See also Mark 2:5, ἀφίενται; Acts

            9:34, ἰᾶται; 26:1, ἐπιτρέπεται; Gal. 1:11, γνωρίζω, and the numer-

            ous instances of λέγω in the gospels.

 

            REM. This usage is a distinct departure from the prevailing use of

the Present tense to denote action in progress (cf. 9). There being in the

Indicative no tense which represents an event as a simple fact without at

the same time assigning it either to the past or the future, the Present is

used for those instances (rare as compared with the cases of the Pro-

gressive Present), in which an action of present time is conceived of

without reference to its progress.

 

            14. The Historical Present. The Present Indicative

is used to describe vividly a past event in the presence of

which the speaker conceives himself to be.. EA. 828;

G. 1252.

 

Mark 11:27; καὶ ἔρχονται πάλιν εἰς Ἰεροσόλυμα,  and they come again

            to Jerusalem. See also Luke 8:49, ἔρχεται; John 18:28, ἄγουσιν.

            This use is very frequent in the gospels.

           

            15. The Present for the Future. In a similar way

the Present Indicative may be used to describe vividly a

future event.

 

Mark 9:31; υἱὸς τοῦ ἀνθρώπου παραδίδοται εἰς χεῖρας ἀνθρώπων, the

            Son of man is delivered into the hands of men. See also Matt. 26:18,

            ποιῶ; 27:63,  ἐγείρομαι; Luke 3:9, ἐκκόπτεται.


10                                            THE TENSES.

 

REM. The term "Present for Future" is sometimes objected to, but

without good reason. The arguments of Buttmann, pp. 203 f., and Winer,

WT. pp. 265 ff.; WH. pp. 331 ff., are valid only against the theory of an

arbitrary interchange of tenses. It is indeed not to be Supposed that

Greek writers confused the Present and the Future tenses, or used them

indiscriminately. But that the form which customarily denoted an act

in progress at the time of speaking was sometimes, for the sake of vivid-

ness, used with reference to a fact still in the future, is recognized by all

grammarians. See; e.g., J. 397; K. 382, 5; G.MT.32. The whole force

of the idiom is derived from the unusualness of the tense employed.

 

16. The Present form ἥκω means I have come ( John 2: 4 ;

4: 47; etc. ). Similarly πάρειμι (I am present) sometimes means

I have arrived (Acts 17: 6; etc.). This, however, is not a

Present for the Perfect of the same verb, but a Present

equivalent to the Perfect of another verb. The use of ἀκύω

meaning I am informed (cf. similar use of English hear, see,

learn) is more nearly a proper Present for Perfect (1 Cor.

11 : 18; 2 Thess. 3 : 11). Such use of the Present belongs to

a very few verbs. HA. 827; G. 1256.

 

17. The Present of past Action still in Progress.

The Present Indicative, accompanied by an adverbial

expression denoting duration and referring to past time,

is sometimes used in Greek, as in German, to describe

an action which, beginning in past time, is still in prog-

ress at the time of speaking. English idiom requires

the use of the Perfect in such cases. RA. 826; G. 1258.

Acts 15:21  Μωυσῃς γὰρ ἐκ γενεῶν ἀρχαίων κατὰ πόλιν τοὺς κηρύσ-

Σοντας αὐτὸν ἔχει, for Moses from generations of old has had in every

city them that preached him. See also Luke 13:7, ἔρχομαι 15:29,

δουλεύω; John 5:6, ἔχει; 2 Tim. 3:15, οἶδας. This Present is

almost always incorrectly rendered in R. V.

 

REM. Cf. Br. 156, "Das Prasens in Verbindung mit πάρος, πάλαι,

ποτέ wurde seit Homer gebraucht, um eine Handlung auszudriicken, die

sich durch die Vergangenheit bis zur Zeit des Sprechens hinzieht." In

the New Testament examples definite expressions of past time occur in

place of the adverbs πάρος, etc.


THE PRESENT INDICATIVE.                            11

 

18. The Aorist Indicative, limited by an expression mean-

ing up to this time, may also be used of acts. beginning in past

time and continuing to the time of speaking. Matt. 27:8;

28:15. Cf. 46, and 52.

 

19. Verbs in indirect discourse retain the point of View, as

respects time, of the original statement; a Progressive Present

in indirect discourse accordingly denotes action going on at

the time, not of the quotation of the words, but of the original

utterance of them. English usage in indirect discourse is

different, and from this difference it results that a Greek

Present Indicative standing in indirect discourse after a verb

of past time must often be rendered by a verb of past time.

These cases, however, involve no special use of the Greek

tense, and should not be confused with those of the Historical

Present. Cf. 351-356.

 

20. PERIPHRASTIC FORM OF THE PRESENT. One of the

clearly marked peculiarities of the Greek of the New Testa-

ment is the frequency with which periphrastic forms composed

of a Present or Perfect Participle (Luke 23:19 is quite excep-

tional in its use of the Aorist Participle; cf. Ev. Pet. 23),

and the Present, Imperfect, or Future Indicative, or the

Present Subjunctive, Imperative, Infinitive, and even parti-

ciple, of the verb εἰμί (rarely also ὑπάρχω), are used instead

of the usual simple forms. Cf. 431, and see the full dis-

cussion with examples in B. pp. 308-313, and the list (not

quite complete) in S. pp. 131ff.

Instances of the periphrastic Present Indicative are, how-

ever, few. The clear instances belong under the head of the

General Present.

Matt. 27:33; εἰς τόπον λεγόμενον Γολγοθά, ἐστιν Κρανίου Τόπος
           
λεγόμενος, unto a place called Golgotha, which is called Place of a

Skull. See also Matt. l:23; Mark 5:41; 2Cor.2:17; 9:12.

 


12                                            THE TENSES.

 

THE IMPERFECT INDICATIVE.

 

21. The Progressive Imperfect. The Imperfect is

used of action in progress in past time. HA. 829;

G. 1250, 2.

Mark 12:41; καὶ πολλοὶ πλούσιοι ἔβαλλον πολλά, and many that were

rich were casting in much.

Luke 1:66; καὶ γὰρ χεὶρ κυρίου ἦν μετ' αὐτοῦ, for the hand of the Lord

was with him.

John 11:36;  ἴδε πῶς ἐφίλει αὐτόν, behold how he loved him.

 

22. The statement respecting the translation of the Pro.

gressive Present (cf. 10), applies to the Imperfect also.

Notice the third example above, and see also Luke 2 : 51,

his mother kept [διετήρει] all these things in her heart; in Luke

24:32, A. V., did not our heart burn within us, is better than

R. V., was not our heart burning within us. Though the verb

is a periphrastic Imperfect, καιομένη, the English form

did burn sufficiently suggests action in progress to render it

adequately.

 

23. THE CONATIVE IMPERFECT. The Progressive Imperfect

is sometimes used of action attempted, but not accomplished.

Cf.11. HA.832; G.1255.

Matt. 3:14; δὲ διεκώλυεν αὐτόν, but he would have hindered him.

See also Luke 1:59, ἐκάλουν; 15:16, ἐδίδου; Acts 7:26, συνήλ-

λασσεν; 26:11, ἠνάγκαζον.

 

24. The Imperfect of Repeated Action. The Imper-

fect is used of customary or repeated action in past time.

HA. 830; G. 1253, 2

Acts 3:2;  ὅν ἐτίθουν καθ' ἡμέραν πρὸς τὴν θύραν τοῦ ἱεροῦ, whom they

used to lay daily at the gate of the temple.

 


                        THE IMPERFECT INDICATIVE                          13

 

25. For the use of the Imperfect, Aorist, or Pluperfect in

a condition contrary to fact, or its apodosis, see 248, 249.

 

26. The Imperfect and Aorist with ἄν are used in classical

Greek to denote a customary past action taking place under

certain circumstances. In the New Testament this usage

never occurs in principal clauses. The use of the Imperfect

and Aorist with ἄν in conditional relative clauses is possibly

a remnant of the usage. Cf. 315.

27. The Imperfect and Aorist are used in a clause express-

ing an unattained wish having reference to the present or past.

The Imperfect denotes action in progress. The Aorist repre-

sents the action indefinitely as a simple event. Either tense

may refer to either present or past time. All the New Testa-

ment instances seem to refer to present time.

 

Rev. 3:15; ὄφελον ψυχρὸς ἦς ζεστός, I would that thou wert cold

or hot. See also 1 Cor. 4:8 (Aor.); 2 Cor. 11:1 (Imperf.).

 

REM. 1. In classical Greek unattainable wishes are expressed by εἴθε

or εἰ γάρ with the Indicative (HA. 871; G. 1511) or ὤφελον with the

Infinitive. In Callimachus, 260 B.C., w@felon is found with the Indicative

(L. & S., ὀφείλω II. 3. fin.). In the New Testament εἰ γάρ (in this

sense) and εἴθε do not occur, but ὄφελον, shortened form of ὤφελον, is

used (as an uninflected particle) with the Imperfect and Aorist Indica-

tive. WM. p. 377; WT. p. 301, N. 2.

REM. 2. In Gal. 5:12 ὄφελον is followed by the Future, but the wish

is probably not conceived of as unattainable.

 

28. When an Imperfect refers to an action not separated

from the time of speaking by a recognized interval, it is

best translated into English by the Perfect, using preferably

the progressive form, unless the verb itself suggests action

ill progress.


14                                            THE TENSES.

 

John 2: 7; ἣν εἴχετε ἀπ' ἀρχῆς, which ye have had from the beginning.

See also Luke 2:49; Rom. 15:22; Rev. 3:2 (cited by Weymouth

in Theological Monthly, IV. 42, who also quotes examples from clas-

sical authors). Cf. 52.

29. When an action denoted by an Imperfect evidently pre-

ceded an event already mentioned, such Imperfect is sometimes

best translated into English by the Pluperfect. From the

point of view of Greek, however, this, like the preceding

usage, is an ordinary Progressive Imperfect or Imperfect of

Repeated Action. Cf. 52.

Matt. 14:14;  ἔλεγεν γὰρ Ἰωάνης αὐτῷ, Οὐκ ἔξεστίν σοι ἔχειν αὐτήν,
            for John had been saying to him, It is not lawful for you to have her.

See also Luke 8:27; Acts 9:39.

 

30. The Imperfect of verbs denoting obligation or possi-

bility, when used to affirm that a certain thing should or

could have been done, i.e. was required or possible under the

circumstances related, is a true affirmative Imperfect. It is

incorrect in this case to speak of an omitted ἄν, since though

it is frequently the case that the necessary or possible deed

did not take place, the past necessity or possibility was actual,

not hypothetical or "contrary to fact." Here belong Matt.

18:33; 23:23; 25:27; Acts 24:19; 26:32; 27:21; 2 Cor.

2:3, etc.

The Imperfect is also used of a past necessity or obligation

when the necessary deed did take place. Here also, of course,

the Imperfect has its usual force. Luke 13: 16; 24: 26 ;

John 4:4; Acts 1:16; 17:3.

31. Buttmann, pp. 216 f., 225 f., describes correctly the class of cases

in which the past obligation or possibility was actual, but in which the

required or possible deed did not take place, but wrongly includes in

his list several passages in which not only the fact but the obligation

or ability is hypothetical. Such are John 9:33; 1 Cor. 5:10; Heb. 9:26,

which are to be explained in accordance with 249. The distinction

 


THE IMPERFECT INDICATIVE.                                   15

 

between these two classes of cases is not always easily marked in English

translation, since the English forms could, should, etc., are used both

for actual and for hypothetical obligation or ability. Of. He could have

gone, if he had been well, and He could have gone, but did not wish

to go.

32. Through a dimming of the distinction between the

ideas of present and past obligation (which has occurred also

in English in the case of the word ought), the Imperfect with-

out Łtv is sometimes used to express a present obligation. The

Infinitive after such an Imperfect is always in the Present

tense. In accordance with this usage we are probably to ex-

plain Acts 22: 22; Eph. 5: 4; Col. 3 :18; cf. Ltft. on Col.

loc. cit. and G.MT. 416.

On these several uses of the Imperfect of verbs of obliga-

tion, etc., see G.MT. 413-423.

 

33. The Imperfect of verbs of wishing, without ἄν, is best

explained as a true Progressive Imperfect, describing a desire

which the speaker for a time felt, without affirming that he

actually cherishes it at the time of his present utterance.

This is especially clear in Philem. 13, 14, where the apostle

states in one clause what his desire--his personal prefer-

ence--was (ἐβουλόμην), and in the next his actual decision

(ἠθέλησα), as over against his preference. The reason for

describing the desire as past is not always, however, that

it has been put aside. Failure to realize the desire, or the

perception that it cannot be realized, or reluctance to express

a positive and deliberate choice may lead the speaker ,to use

the Imperfect rather than the Present. Similarly we some-

times say in colloquial English, I was wishing that such a

thing might happen, or even more commonly, I have sometimes

wished. Nearly the same meaning may be conveyed in Eng-

lish by the more usual potential form, I should like, I would


16                                            THE TENSES.

 

that, or I could wish. In Acts 25 : 22 the use of the Imperfect

ἐβουλόμην rather than a Present softens the request for polite-

ness' sake, and may well be rendered I should like. In Gal.

4:20 it is probably the impossibility of realizing the wish

that leads to the use of the Imperfect, and ἤθελον παρεῖναι
may be rendered, I would that I were present. In Rom. 9:3

ηὐχόμην may have been chosen because the apostle shrank

from expressing a deliberate choice in regard to so solemn.

a matter, or because he thought of it as beyond the control

or influence of his wish. I could pray expresses the meaning

with approximate accuracy. In all these cases, however, what

is strictly stated in the Greek is merely the past existence of a

state of desire; the context alone implies what the present

state of mind is. Of. G.MT. 425.

 

34. PERIPHRASTIC FORM OF THE IMPERFECT. Periphras-

tic Imperfects, formed by adding a Present Participle to the

Imperfect of the verb εἰμί, are frequent in the New Testament,

especially in the historical books. The large majority of

these forms denote continued action.

 

Mark 10:32; καὶ ἦν προάγων αὐτοῦς Ἰησοῦς, and Jesus was going

before them. So also Luke 1:10, 22; John 13:23; and probably

            Mark 2:18. In a few instances repeated action is referred to, as

            Luke 5:16; 19:47; Gal. 1:23. Cf. 431.

 

THE AORIST INDICATIVE.

 

35. The constant characteristic of the Aorist tense in all

of its moods, including the participle, is that it represents the

action denoted by it indefinitely; i.e. simply as an event,

neither on the one hand picturing it in progress, nor on the

other affirming the existence of its result. The name indefi-

nite as thus understood is therefore applicable to the tense in

all of its uses.


                        THE AORIST INDICATIVE                                  17

 

As respects the point of view from which the action is

looked at, however, we may distinguish three functions of the

tense common to all of its moods.

First, it may be used to describe an action or event in its

entirety. This use of the tense, since it is by far the most

frequent, may be called by pre-eminence the Indefinite Aorist.

In the Indicative it may be called the Historical Aorist. The

Aorist of any verb may be used in this sense; thus εἰπεῖν,

to say; διακονήσαι, to serve.

Secondly, it may be used to denote the inception of a

state. The Aorist thus used may be called the Inceptive

Aorist. It belongs to verbs which in the Present and Imper-

fect denote the continuance of a state; thus σιγᾷν, to be silent;

σιγῆσαι, to become silent.

Thirdly, it may be used to denote the success of an effort.

The Aorist thus used may be called the Resultative Aorist.

It belongs to verbs which in the Present and Imperfect denote

effort or attempt; thus κωλύειν, to hinder, obstruct; κωλῦσαι, to

prevent.

The genetic relation of these three functions of the Aorist

tense has not been satisfactorily defined. In the Greek, both of

the classical and the New Testament periods, however, they ap-

pear side by side as co-ordinate uses. Br.159; Del. IV.,pp.l00f.

 

REM. Respecting the force of the Indefinite Aorist, compare Brug-

mann's statement concerning the Aorist forms: "Am haufigsten wurden

diese Formen so gebraucht, dass man sich die Handlung in einen unge-

teilten Denkakt ganz und vollstandig, in sich abgeschlossen, absolut vor-

stellen sollte. Das Factum wurde einfach constatiert ohne Rucksicht

auf Zeitdauer." Br. 159.

 

36. In addition to these uses which belong to the Aorist in

all its moods, the Aorist Indicative has three uses, instances

of which are comparatively infrequent. These are the Gnomic

Aorist, the Epistolary Aorist, and the Dramatic Aorist.


18                                            THE TENSES.

 

The Aorist for the Perfect and the Aorist for the Pluper4

fect are, as explained below (52), not distinct functions of the

Aorist, but merely special cases of the Historical, Inceptive,

or Resultative Aorist.

 

37. The distinction between the Indefinite, the Inceptive,

and the Resultative functions of the Aorist is often ignored,

or its legitimacy denied. It is true that there are cases in

which it is not possible to decide certainly whether a given

verb refers to the inception of an action only, or to its entire

extent, and others in which there is a similar difficulty in

deciding whether the reference is to the action as a whole or

to its result only. It is true also that the genetic relation of

these three uses of the tense is not a matter of entire cer-

tainty, and that it is possible that, historically speaking, they

are but varying types of one usage. Especially must it be

regarded as doubtful whether the Resultative Aorist is any-

thing else than the Indefinite Aorist of verbs denoting effort.

The matter of importance to the interpreter, however, is

that, whatever the genesis of the fact, of the Aorists of the

New Testament some denote a past act in its undivided

entirety, others denote merely or chiefly the inception of an

action, and others still affirm as a past fact the accomplish-

ment of an act attempted.  These distinctions, which from the

exegetical point of view it is often Important to mark, are

conveniently indicated by the terms indefinite, inceptive, and

resultative. With reference to the validity of this distinction,

see Br. 159.

The Inceptive Aorist is illustrated in Acts 15 : 13, and after

they had become silent [μετὰ τὸ σιγῆσαι] James answered. It

is evident that the Infinitive must refer to the becoming

silent, not to the whole period of silence, since in the latter

case James must have been silent while the others were silent,


THE AORIST INDICATIVE.                                19

 

and have begun to speak when their silence had ended. In

2 Cor. 8: 9, we must read not being rich he was poor, but being

rich he became poor; ἐρτώχευσεν is manifestly inceptive. So

also in Luke 2:44, supposing him to be in the company, they

went a day's journey, it was not the holding of the opinion that

he was in the company that preceded the day's journey, but

the forming of it, and the participle νομίσαντες is inceptive.

Contrast Acts 16:27. See other examples under 41.

Illustrations of the resultative sense are less numerous and

less clear. In Acts 7:36, however, this man led them forth,

having wrought wonders and signs in Egypt and in the Red Sea,

and in the wilderness forty years, the verb ἐξήγαγεν seems to

refer only to the result, since the signs wrought in the Red

Sea and the wilderness would otherwise have been represented

as accompanying the bringing out, and instead of ποιήσας we

should have had ποιῶν. See also 42.1

38. The Historical Aorist. The Aorist Indicative is

most frequently used to express a past event viewed in its

entirety, simply as an event or a single fact. It has no

reference to the progress of the event, or to any existing

result of it. HA. 836; G. 1250, 5.

John 1:11; εἰς τὰ ἴδια ἦλθεν, καὶ οἱ ἴδιοι αὐτὸν οὐ παρέλαβον, he came

unto his own and they that were his own received him not.

 

39. Since any past event without reference to its duration

or complexity may be conceived of as a single fact, the His-

torical Aorist may be used to describe

(a) A momentary action.

Acts 5:5; ἐξέψυξεν, he gave up the ghost.

Matt. 8: 3; καὶ ἐκτείνας τὴν χεῖρα ἥψατο αὐτοῦ, and having stretched

forth his hand he touched him.

 

1 Cf. Mart. Polyc. 8 : 2, 3, where both ἔπειθον, were persuading, and

ἀποτυχόντες τοῦ πεῖσαι, failing to persuade, refer to the same event.


20                                            THE TENSES.

 

(b) An extended act or state, however prolonged in time, if

viewed as constituting a single fact without reference to its

progress.

Acts 28:30; ἐνέμεινεν δὲ διετίαν ὅλην ἐν ἰδίῳ μισθώματι, and he abode

two whole years in his own hired dwelling.

Eph. 2:4; διὰ τὴν πολλὴν ἀγάπην αύτοῦ ἣν ἠγάπησεν ἡμᾶς, because

of his great love wherewith he loved us.

 

(c) A series or aggregate of acts viewed as constituting a

single fact.

Matt. 22:28; πάντες γὰρ ἔσχον αὐτήν, for they all had her.

2 Cor. 11:25; τρὶς ἐναυάγησα, thrice I suffered shipwreck.

 

40. These three uses of the Historical Aorist may for con-

venience be designated as the Momentary Aorist, the Compre-

hensive Aorist, and the Collective Aorist. But it should be

clearly observed that these terms do not mark distinctions in

the functions of the tense. An Historical Aorist, whatever the

nature of the fact affirmed, affirms it simply as a past fact.

The writer mayor may not have in mind that the act was

single and momentary, or extended, or a series of acts, but the

tense does not express or suggest the distinction. The pur-

pose of the subdivision into momentary, comprehensive, and

collective is not to define the force of the tense-form, but to

discriminate more precisely the nature of the facts to which

it is applied as shown by the context or the circumstances.

Cf. G.MT. 56.

REM. The term Historical Aorist is applied to the use of the Aorist

here described only by pre-eminence. In strictness the Inceptive and

Resultative Aorists are also Historical. Compare what is said concerning

the term Indefinite under 35.

 

41. The Inceptive Aorist. The Aorist of a verb whose

Present denotes a state or condition, commonly denotes

the beginning of that state. HA. 841; G. 1260.


THE AORIST INDICATIVE.                                21

 

2 Cor. 8:9; δι' ὑμᾶς ἐπτώχευσεν πλούσιος ὤν, though he was rich, for

your sakes he became poor. See also Luke 15:32; John 4:52 ;

Acts 7:60; Rom. 14:9.

 

REM. The Aorist of such verbs is not, however, necessarily inceptive.

The same form may be in one sentence inceptive and in another historical

Cf. Luke 9:36 with Acts 15:12, the verb e]si<ghsa being in the former

historical, in the latter probably inceptive.

 

42. The Resultative Aorist. The Aorist of a verb

whose Present implies effort or intention, commonly de-

notes the success of the effort. Cf. 11, 23. Br. 159.

Acts 27:43; δὲ ἑκατοντάρχης . . .  ἐκώλυσεν αὐτοὺς τοῦ βουλήματος,

but the centurion. . . prevented them from their purpose. See also

Matt. 27:20; Acts 7:36.

 

43. The Gnomic Aorist. The Aorist is used in prov-

erbs and comparisons where the English commonly uses a

General Present. HA. 840; G. 1292; G. MT. 154-161;

B. pp. 201 ff.; WM. pp. 346 f.; WT. p. 277; Br. 160.

1 Pet. 1:24; ἐξηράνθη χόρτος, καὶ τὸ ἄνθος ἐξέπεσεν, the grass wither-

eth and the flower falleth. See also Luke 7:35; John 15:6; Jas.

1:11, 24.

REM. Winer's contention (WT. p. 277; WM. p. 346) that the

Gnomic Aorist does not occur in the New Testament does not seem

defensible. The passages cited above are entirely similar to the classical

examples of this ancient and well-established idiom.

 

44. The Epistolary Aorist. The writer of a letter

sometimes puts himself in the place of his reader and de-

scribes as past that which is to himself present, but which

will be past to his reader. HA. 838.

Eph. 6:22; ὃν ἔπεμψα πρὸς ὑμᾶς εἰς αὐτὸ τοῦτο, whom I send to you for

this very purpose. See also Acts 23:30; 1 Cor. 5:11; Phil. 2:28;

Col. 4:8; Philem. 11.


22                                            THE TENSES.

 

45. The Dramatic Aorist. The Aorist Indicative is

sometimes used of a state of mind just reached, or of an

act expressive of it. The effect is to give to the statement

greater vividness than is given by the more usual Present.

HA. 842; G.MT. 60; K 386, 9; Br. 160.

Luke 16:4; ἔγνων τί ποιήσω, I know [lit. I knew, or I perceived] what

I shall do.

 

REM. This usage is in classical Greek mainly poetical and is found

chiefly in dialogue. It is sometimes called "Aoristus tragicus." Brug-

mann thus describes it: "Nicht selten wurde der Aorist yon dem

gebraucht, was soeben eingetreten ist, besonders von einer Stimmung,

die soeben uber einen gekommen ist, oder yon einem Urteil, das man

sich soeben gebildet hat." See numerous examples in K. 386, 9.

 

46. THE AORIST FOR THE (English) PERFECT. The Aorist

is frequently used in Greek where the English idiom requires

a Perfect. G.MT. 58; H.A. 837; B. pp. 197, 198.

Lk19:9; σήμερον σωτηρία τῷ οἴκῳ τούτῳ εγἐνετο, to-day is salvation

come to this house.

Matt. 5:21;  ἠκούσατε ὅτι ἐρρέθη τοῖς ἀρχαίοις, ye have heard that it was

said to them of old time.

Ph. 4:11;  ἐγὼ γὰρ ἔμαθον ἐν οἷς εἰμὶ αὐτάρκης εἶναι,  for I have learned

in whatsoever state I am therein to be content. See also under 52.

 

47. The Aorist Indicative of a few verbs is used in the New

Testament to denote a present state, the result of a past act,

hence with the proper force of a Greek Perfect. Of. 75, 86.

So the Aorists ἀπέθανον (cf. Mark 5:35 with Luke 8:49, and

see John 8:52 et al.), ἐξέστην (Mark 3:21; 2 Cor. 5:13), and

possibly ἔγνων (John 7:26; cf. 1 Macc. 6:13). All these

Aorists may also be used as simple historical Aorists.

 

48. THE AORIST FOR THE (English) PLUPERFECT. The

Aorist Indicative is frequently used in narrative passages of

a past event which precedes another past event mentioned


THE AORIST INDICATIVE.                    23

 

or implied in the context. In English it is common in such

a case to indicate the real. order of the events by the use

of a Pluperfect for the earlier event. Of. 52, 53. H...4.. 837;

G.MT. 58; B. pp. 199 f.

John 19:30; ὅτε οὖν ἔλαβεν τὸ ὄξος Ἰησοῦς εἶπεν, Τετέλεσται, when

therefore Jesus had received the vinegar, he said, It is finished.

Matt. 14:3; γὰρ Ἡρῴδης κρατήσας τὸν Ἰωάνην ἔδησεν, for Herod

having laid hold on John had bound him. See also Matt. 27:31;

Mark 8:14; Luke 8:27; John 12:17; 13:12.

 

REM. It has been much disputed whether ἀπέστειλεν in John 18:24

is to be assigned to this head. The valid objection to this is not in any

inappropriateness of the Aorist tense to express an event antecedent to

one already mentioned)--the Aorist is the only form that can be used if

the event is thought of simply as an event (cf. Mey. ad loc., contra)--

but in the presence of οὖν, which is, in John especially, so constantly

continuative, and in the absence of any intimation in the context that

the events are related out of their chronological order.

 

49. From the general principles of indirect discourse in

English and in Greek it results that an Aorist Indicative in

indirect discourse after a verb of past time must usually be

rendered into English by a Pluperfect. Cf. 353. These cases

form a class entirely distinct from those that are included

above under the term Aorist for the English Pluperfect.

 

50. Both the Aorist and the Perfect are sometimes used

proleptically, but this is rather a rhetorical figure than a gram-

matical idiom. WM. pp. 341, 345, 347; WT. pp. 273, 277, 278.

 

1 Cor. 7:28; ἐὰν δὲ καὶ γαμήσῃς, οὐχ ἥμαρτες, but even if thou shalt

marry, thou hast not sinned. See also John 15: 8; J as. 2: 10.

 

51. For the Aorist in a condition contrary to fact, see 248.

For the Aorist expressing an unattained wish, see 27.

 

52. ENGLISH EQUIVALENTS OF THE GREEK AORIST INDIC-

ATIVE. It should be observed that the Aorist for the Perfect

and the Aorist for the Pluperfect are not variations from the

 


24                                            THE TENSES.

 

normal use of the Greek Aorist. Viewed strictly from the

point of view of Greek Grammar, these Aorists are simply

Historical, Inceptive, or Resultative Aorists. The necessity for

mentioning them arises merely from the difference between

the English and the Greek idiom.

The Greek Aorist corresponds to the English simple Past

(or Imperfect or Preterite, loved, heard, etc.) more nearly than

to any other English tense. But it is not the precise equiva-

lent of the English Past; nor is the Greek Perfect the precise

equivalent of the English Perfect; nor the Greek Pluperfect

of the English Pluperfect. This will appear distinctly if we

place side by side the definitions of the tenses which in gen-

eral correspond in the two languages.

 

The English Perfect is used                  The Greek Perfect is used

of any past action between                          to represent an action as

which and the time of speak-                      standing complete, i.e. as hav-

ing the speaker does not in-                        ing an existing result, at the

tend distinctly to interpose an                     time of speaking.

interval.1

The English Pluperfect is                     The Greek Pluperfect is

used to mark the fact that the                      used to represent an action as

event expressed by it preceded                   standing complete, i.e. as hav-

another past event indicated by                  ing an existing result, at a

the context, and this whether                      point of past time indicated

the earlier event is thought of                     by the context.

as completed at the time of

the later event, or only indefi-

nitely as a simple occurrence

preceding the later event!

 

1 The English Perfect and Pluperfect by their auxiliaries have and had

distinctly suggest completed action in the proper sense, viz. the posses-

sion of a thing in the condition indicated by the participle, and substan-


THE AORIST INDICATIVE.                                25

 

The English Past is used of                              The Greek Aorist is used of

any past action between which                   any past event which is con,

and the moment of speaking                       ceived of simply as an event

an interval is thought of as                          (or as entered upon, or as ac-

f existing. It affirms nothing                       complished), regardless alike

respecting existing result.                            of the existence or non-exist.

ence of an interval between

itself and the moment of

speaking, and of the question.

whether it precedes or not

some other past action. It

affirms nothing respecting ex-

isting result.

 

            It is evident from this comparison that the English Perfect

has a larger range of use than the Greek Perfect.

 

tially this is the meaning often conveyed by these tenses. Thus, I have

learned my lesson, differs but little in meaning from I have my lesson

learned. But this is by no means the only use which may be made of

these tenses in modern English. They have, in fact, ceased to be Perfect

tenses in any proper sense of that word. Compare, e.g., the Pasts and

Perfects in the following examples: The army arrived. The army has

arrived. Many men fought for their country. Many men have fought

for their country. He often visited Rome. He has often visited Rome.

Only in the first example is existing result suggested by the Perfect tense.

In each pair the distinguishing mark between the two sentences is that

while the Perfect tense places the event in the past time without defining

whether or not an interval has elapsed since the event, the Past tense

places it in the past time and suggests an interval.

Similarly, the English Pluperfect affirms only the antecedence of its

event to the other past event, leaving it to the context or the nature of

the fact to show whether at the past time referred to there were existing

results or not. Thus in the sentence, I showed him the work which I had

done, it is implied that the results of the doing remained at the time of

the showing. But in the sentence, He did not recognize the persons whom

he had previously seen, it is not implied that any result of the seeing

remained at the time of the non-recognition.


26                                            THE TENSES.

 

Thus a past event between which and the time of speaking

no interval is distinctly thought of may be expressed by the

English Perfect, whether the result of the event is thought of

as existing or not; but it can be expressed by the Greek Per-

fect only in case such result is thought of. So also the Eng-

lish Pluperfect has a wider range than the Greek Pluperfect.

For while the Greek can use its Pluperfect for an event

which preceded another past event only in case the result

of the earlier event is thought of as existing at the time

of the later event, the English freely uses its Pluperfect

for all such doubly past events, without reference to the

existence of the result of the earlier event at the time of

the later one.

On the other hand, the Greek Aorist has a wider range

than the English Past, since it performs precisely those func-

tions which the Greek Perfect and Pluperfect refuse, but

which in modern English are performed not by the Past but

by the Perfect and Pluperfect. The Greek Aorist, therefore,

in its ordinary use not only covers the ground of the English

Past, but overlaps in part upon that of the English Perfect

and Pluperfect. Hence arise the so-called Aorist for Perfect

and Aorist for Pluperfect.

If the attempt be made to define more exactly the extent

of this overlapping, it will appear that a simple past event

which is conceived of without reference to an existing result,

and between which and the time of speaking the speaker does

not wish distinctly to suggest an interval,--the interval may

be ever so long, in fact,--will be expressed in Greek by

the Aorist, because the result is not thought of, and in Eng-

lish by the Perfect, because the interval is not thought "of.

Cases of this kind arise, e.g., when the event is said to con-

tinue up to the time of speaking, so that there is actually no

interval [Matt. 27:8; διὸ ἐκλήθη ἀργὸς ἐκεῖνος Ἀγρὸς Αἵματος


THE AORIST INDICATIVE.                                27

 

ἕως τῆς σήμερον, therefore that field has been called Field of Blood

until this day. See also Matt. 28:15; John 16:24]; or when the

event is so recent as to make the thought. of an interval seem

unnatural [Luke 5:26; εἴδαμεν παράδοξα σήμερον, we have seen

strange things to-day. See also Mark 14:41; Acts 7:52, nu?n

. . . ἐγένεσθε]; or when the time of the event is entirely

indefinite [Matt. 19:4; οὐκ ἀνέγνωτε, have ye not read?  See

also Rev. 11:12; exx. are frequent in the New Testament];

or when the verb refers to a' series of events which extends

approximately or quite to the time of speaking [Matt. 5:21;

ἠκούσατε ὅτι ἐρρέθη τοῖς ἀρχαίοις, ye have heard that it was said

to the ancients; the reference is doubtless to the frequent

occasions on which they had heard such teachings in the

synagogue. See also 1 Esdr. 4 : 26, 27].

Instances of the Greek Aorist for the English Pluperfect

arise when a past event which is conceived of simply as an

event without reference to existing result is mentioned out

of its chronological order, or is expressed in a subordinate

clause. The Greek employs the Aorist, leaving the context

to suggest the order; the English usually suggests the order

by the use of a Pluperfect. See exx. under 48. Of. Beet, The

Greek Aorist as used in the New Testament, in Expositor, XI.

191-201, 296-308, 312-385; Weymouth, The Rendering into

English of the Greek Aorist and Perfect, in Theological

Monthly, IV. 33-41,162-180.

 

53. In many cases in which the Greek Aorist is used of

an event antecedent to another past event already referred to,

English idiom permits a simple Past. A Pluperfect is strictly

required only when the precedence in time is somewhat promi-

nent. The Revisers of 1881 have used the Pluperfect spar-

ingly in such cases. It might better have been used also in

Matt. 9:25; Mark 8:14; John 12:18 (had heard).


28                                THE TESES.

 

54. An Aorist which is equivalent to an English Perfect

or Pluperfect may be either an historical, or an inceptive, or

a Resultative Aorist. If historical, it may be either momentary,

comprehensive, or collective.

 

In Luke 15:32, ἔζησεν, and in 1 Cor. 4:8, ἐπλουτήσατε, are inceptive

Aorists which may be properly rendered by the English Perfect; probably

also ἐβασίλευσας, in Rev. 11:17, should be rendered, thou hast become

king.

In Rom. 3:23, h!marton is evidently intended to sum up the aggregate

of the evil deeds of men, of which the apostle has been speaking in the

preceding paragraphs (1:18 -3:20). It is therefore a collective historical

Aorist. But since that series οf evil deeds extends even to the moment

of speaking, as is indeed directly affirmed in the πάντες, it is impos-

sible to think of an interval between the fact stated and this statement

of it. It must therefore be expressed in English by the Perfect tense, and

be classed with Matt. 5:21 as a collective Aorist for (English) Perfect.

Of similar force is the same form in Rom. 2:12. From the point of view

from which the apostle is speaking, the sin of each offender is simply a

past fact, and the sin of all a series or aggregate of facts together consti-

tuting a past fact. But inasmuch as this series is not separated from the

time of speaking, we must, as in 3:23, employ an English Perfect in

translation. This is upon the supposition that the verb ἥμαρτον takes its

point of view from the time of speaking, and the apostle accordingly

speaks here only of sin then past, leaving it to be inferred that the same

principle would apply to subsequent sin. It is possible, however, that

by a sort of prolepsis ἥμαρτον is uttered from the point of view of the

future judgment [κριθήσονται], and refers to all sin that will then be past.

In this case the Future Perfect, shall have sinned, may be used in trans-

lation, or again the Perfect, common in subordinate clauses in English as

an abbreviation of the Future Perfect. Whether the same form in Rom.

5:12 shall be rendered in the same way or by the English Past depends

upon whether it is, like the other cases, a collective Aorist, representing

a series of acts between which and the time of speaking no interval is

interposed, or refers to a deed or deeds in the remote past in which the

"all" in some way participated. So far as the tense-form is concerned

there is no presumption in favor of one or the other of these inter-

pretations, both uses of the tense being equally legitimate. The nature

of the argument or the author's thought, as learned from sources

outside the sentence itself, must furnish the main evidence by which

to decide.


THE AORIST INDICATIVE.                                29

 

55. The Aorist εὐδόκησα in Matt. 3:17; 17:5; Mark 1:11; Luke

3:22; 2 Pet. 1:11, may be explained --(a) as a Historical Aorist having

reference to a specific event as its basis. I was well pleased with thee,

e.g. for receiving baptism. If all the instances were in connection with the

baptism, this would be the most natural explanation. But for those that

occur in connection with the account of the transfiguration this explana-

tion fails, and is probably therefore not the true explanation of any of the

instances. (b) as a comprehensive Historical Aorist covering the period

of Christ's preincarnate existence. Cf. John 17:5, 24; see W. N. Clarke,

Com. on Mark 1:11. If the passages were in the fourth gospel, and

especially if they contained some such phrase as πρὸ καταβολῆς κόσμου,

this explanation would have much in its favor. The absence of such

limiting phrase, and the fact that the passages are in the synoptic gospels

are opposed to this explanation. (c) as a comprehensive Historical Aorist,

having the force of an English Perfect, and referring to the period of

Christ's earthly existence up to the time of speaking. But against this

is the absence of any adverbial phrase meaning up to this time, which

usually accompanies an Aorist verb used in this sense. Cf. 18 and 52.

(d) as an Aorist which has by usage come to have the meaning which is

strictly appropriate to the Perfect, I became well pleased with thee, and

I am [accordingly] well pleased with thee. Cf. 47. There are a few pas-

sages of the Septuagint that seem at first sight to favor this explanation.

SeePs. 101:15; Jer. 2:19; Mal. 2:17. Cf. also Matt. 12:18; Luke 12:32.

The force of this evidence is, however, greatly diminished by the fact

that all these instances are capable of being explained without resort to so

unusual a use of the Aorist, that both in the Septuagint and in the New

Testament there is in use a regular Present form of this verb, and that

the Aorist in the majority of cases clearly denotes past time. (e) as an

Inceptive Aorist referring to some indefinite, imagined point of past time

at which God is represented as becoming well pleased with Jesus. But

since this point is not thought of as definitely fixed, English idiom requires

a Perfect tense. Cf. 52 (p. 27), 54. It may be described, therefore, as an

Inceptive Aorist equivalent to an English Perfect, and may be rendered,

I have become well pleased. This, however, can only be a vivid way of

saying, I am well pleased. If then this view is correct, the rendering

of the English versions is a free but substantially correct paraphrase.

A true Perfect would affirm the present state of pleasure and imply the

past becoming pleased. The Aorist affirms the becoming pleased and

leaves the present pleasure to be suggested. This explanation, therefore,

differs from the preceding (d) in that it does not suppose the Aorist

of this verb to have acquired the power of expressing an existing result,

but judges the existing result to be only suggested by the affirmation


30                                            THE TENSES.

 

of the past fact. This is rhetorical figure, on the way to become gram-

matical idiom, but not yet become such. Manifestly similar is the use

of προσεδέξατο in Isa. 42:1, and of εὐδόκησεν in Matt. 12: 18. Indeed, if

Matt. 12:18 represents a current translation of Isa. 42:1, our present

passages were probably affected in form by this current rendering of the

Isaiah passage. Similar also are ἐκάθισαν in Matt. 23:2, and ἔμαθον in

Phil. 4:11. In neither case is there any clearly established usage of the

Aorist for Greek perfect; in neither is there apparent any reference

to a definite point of past time; in both the real fact intended to be

suggested is the present state.

 

56. THE DISTINCTION BETWEEN THE AORIST AND THE

IMPERFECT. The difference between an Historical Aorist

and an Imperfect of action in progress or repeated being one

not of the nature of the fact but of the speaker's conception

of the fact, it is evident that the same fact may be expressed

by either tense or by both. This is illustrated in Mark 12:41

and 44, where, with strict appropriateness in both cases, Mark

writes in v. 41, πολλοὶ πλούσιοι ἔβαλλον πολλά, and in v. 44

records Jesus as stating the same fact in the words πάντες . . .

ἔβαλον. The former describes the scene in progress, the latter

merely states the fact.

 

57. From the nature of the distinction between the Imper-

fect and Aorist, it also results that the difference in thought

represented by the choice of one form rather than the other

is sometimes almost imperceptible. Cf., e.g., Mark 3:7 and

5:24; Luke 2:18 and 4:22. Some verbs use one of the two

tenses almost or quite to the exclusion of the other. The

form ἔλεγον is used in classical Greek without emphasis on

the thought of the saying as in progress or repeated, and in the

New Testament the Aorist of this verb does not occur. A dis-

tinction between the Imperfect ἔλεγον and the Aorist εἶπον is

scarcely to be drawn in the New Testament. Cf. G.MT. 56,

57, especially the following: "In all these cases tbe funda-

mental distinction of the tenses, which was inherent in the


THE FUTURE INDICATIVE.                              31

 

form, remained; only it happened that either of the two dis-

tinct forms expressed the meaning which was here needed

equally well. It must not be thought, from these occasional

examples, that the Greeks of any period were not fully alive

to the distinction of the two tenses and could not use it with

skill and nicety."

This approximation of the Aorist and Imperfect, it should

be noted, occurs only in the case of the Historical Aorist (38).

The Inceptive and Resultative Aorists are clearly distinguished

in force from the Imperfect.

 

THE FUTURE INDICATIVE.

 

58. The Predictive Future. The Future Indicative is

most frequently used to affirm that an action is to take

place in future time. Since it does not mark the distinc-

!tion between action in progress and action conceived of

indefinitely without reference to its progress, it may be

either aoristic or progressive. HA. 843; G. 1250, 6;

G.MT. 63, 65; Br. 163.

59. THE AORISTIC FUTURE conceives of an action simply

as an event, and affirms that it will take place in future time.

It may be indefinite, inceptive, or resultative. As indefinite

it may be momentary, comprehensive, or collective. Of. 35, 39.

1 Cor. 15:51, 52;  πάντες οὐ κοιμηθησόμεθα, πάντες δὲ ἀλλαγησόμεθα,
           
ἐν ἀτόμῳ, ἐν ῥιπῇ ὀφθαλμοῦ we shall not all sleep [indefinite com-

prehensive]; or; we shall not all fall asleep [inceptive], but we shall

all be changed, in a moment, in the twinkling of an eye [indefinite

momentary].

John 14:26; ἐκεῖνος ὑμᾶς διδάξει πάντα καὶ ὑπομνήσει ὑμᾶς πάντα
           
εἶπον ὑμῖν ἐγώ, he will teach you all things and bring to your remem-

brance all things that I said unto you [indefinite collective].

Luke 1:33; καὶ βασιλεύσει ἐπὶ τὸν οἶκον Ἰακὼβ εἰς τοὺς αἰῶνας, and he

shall reign over the house of Jacob forever [indefinite comprehensive].

Luke 16:31; οὺδ' ἐάν τις ἐκ νεκρῶν ἀναστῇ πεισθήσονται, neither will

they be persuaded if one rise from the dead [resultative].


32                                            THE TENSES.

 

60. THE PROGRESSIVE FUTURE affirms that an action will

be in progress in future time. HA. 843; G. 1250, 6.

Phil. 1:18; καὶ ἐν τούτῳ χαίρω· ἀλλὰ καὶ χαρήσομαι, and therein I

rejoice, yea, and will [continue to] rejoice. See also Rom. 6:2;

Phil. 1:6; Rev. 9:6.

 

61. It may be doubted whether any of the distinctions indi-

cated by the subdivisions of the Predictive Future are justi-

fied from the point of view of pure grammar. It is probable,

rather, that the tense in all these cases makes precisely the

same affirmation respecting the event, viz. that it will take

place; and that it is the context only that conveys the dis-

tinctions referred to. These distinctions, however, are real

distinctions either of fact or of thought, and such, moreover,

that the writer must in most cases have had them in mind

when speaking of the facts. From the exegetical point of

view, therefore, the distinctions are both justified and neces-

sary, since they represent differences of thought in the mind

of the writer to be interpreted. The terms employed above

are convenient terms to represent these distinctions of thought,

and it is to the interpreter a matter of secondary importance

whether the distinction in question is by his writer immedi-

ately connected with the tense of the verb.

 

62. Since the Aoristic Future is less definite respecting

progress than the' Progressive Future, the latter predicting

the act as continuing, the former making no assertion, it is

evident that any instance of the Predictive Future not clearly

progressive must be accounted as aoristic. If the writer did

not conceive the act or event as continuing, he left it in his

own mind and for the reader undefined as respects progress,

hence aoristic. Whether he left it thus undefined in his mind

must of course be determined, if at all, from the context, there

being no difference of form between a Progressive and an


THE FUTURE INDICATIVE.                              33

 

Aoristic Future. It should be noticed that it is not enough

to show that an act will be in fact continued, in order to count

the verb which predicts it a Progressive Future; it must ap-

pear that the writer thought of it as continuing. Every

Future form is therefore by presumption aoristic. It can

be accounted progressive only on evidence that the writer

thought of t4e act as continued.

 

REM. There is one exception to this principle. In verbs of effort a

Progressive Future is naturally like other Progressive forms, a conative

tense. An Aoristic Future of such a verb is like the Aorist, a resultative

tense. Since the latter is the larger meaning, the context must give the

evidence of this larger meaning, and such evidence failing, it cannot be

considered established that the verb is resultative. The verb in John 12:

32 furnishes an interesting and important illustration. .Since the verb

denotes effort, the Future will naturally be accounted conative if it is

judged to be progressive, and resultative if it is taken as aoristic. In the

latter case the meanwg will be, I will by my attraction bring all men to

me. In the former case the words will mean, I will exert on all men an

attractive influence.

 

63. To decide whether a given Aoristic Future merely pre-

dicts the fact, or refers to the inception of the action, or has

reference to it as a thing accomplished, must again be deter-

mined by the context or the meaning of the word. The dis-

tinction between the indefinite and the resultative senses will

often be very difficult to make, and indeed the difference

of thought will be but slight. Here also it results from the

nature of the distinction between the indefinite use and the

other two, inceptive and resultative, that any instance of

the Aoristic Future not clearly inceptive or resultative must

be accounted indefinite. In other words, if the writer did not

define the action to his own mind as inceptive or resultative,

he left it indefinite, a mere fact.

 

64. The distinction between momentary, comprehensive,

and collective is in respect to the Future tense, as in respect


34                                            THE TENSES.

 

to the Aorist, a distinction which primarily has reference to

the facts referred to and only secondarily to the writer's con-

ception of the facts. There may easily occur instances which

will defy classification at this point. A writer may predict

an event not only without at the moment thinking whether

it is to be a single deed or a series of deeds, a momentary or

an extended action, but even without knowing. Thus the

sentence, He will destroy his enemies, may be uttered by one

who has confidence that the person referred to will in some .

way destroy his enemies, without at all knowing whether he

will destroy them one by one, or all at once, and whether by

some long-continued process, or by one exterminating blow.

In such cases the verb can only be accounted as an Aoristic

Future, incapable of further classification.

 

65. From a different point of view from that of the above

classification, the instances of the Predictive Future might be

classified as (a) assertive, and (b) promissory. The distinc-

tion between the assertion that an event will take place and

the promise that it shall take place is difficult to make,

requiring delicate discrimination, but is often important for

purposes of interpretation. It is in general not indicated in

Greek, and its representation in English is complicated by the

varied uses of the auxiliary verbs shall and will. In general

it may be said that in principal clauses shall is in the first

person simply assertive, will is promissory; in the second and

third person will is assertive, shall is promissory, imperative,

or solemnly predictive.

R. V. employs shall almost constantly in the second and

third person, in most cases probably intending it as solemnly

predictive.

Matt. 10:42; ἀμὴν λέγω ὑμῖν, οὐ μὴ ἀπολέσῃ τὸν μισθὸν αὐτοῦ, verily

I say unto you, he shall by no means lose his reward.


THE FUTURE INDICATIVE.                              35

 

Mark 11:31; ἐὰν εἴπωμεν Ἐξ οὐρανοῦ, ἐπεῖ, if we say, From heaven, he

will say.

Luke 22:61; Πρὶν ἀλέκτορα φωνῆσαι σήμερον ἀπαρνήσῃ, before

the cock crow this day, thou shalt deny me thrice. See also Matt. 11:

28, 29; 12:31; John 16:7, 13.

 

66. A Predictive Future is sometimes made emphatically

negative by the use of the negative οὐ μή, Matt. 16:22; 26:

35; Mark 14 : 31 (Tisch. Subjunctive); cf. 172.

 

67. The Imperative Future. The second person of the

Future Indicative is often used as an Imperative. HA. 844;

G. 1265.

Jas. 2:8; ἀγαπήσεις τὸν πλησίον σου ὡς σεαυτόν, thou shalt love thy

neighbor as thyself.

 

REM. 1: This idiom as it occurs in the New Testament shows clearly

the influence of the Septuagint. It occurs most frequently in prohibi-

tions, its negative being, as also commonly in classical Greek, not μή but

ou]. G.MT. 69, 70; B. p. 267; WM. pp.. 396 f.; WT. pp. 315 f.

 

REM. 2. In Matt. 15: 6 the verb τιμήσει has the negative οὐ μή. Some

interpreters take this as a Predictive Future, but the thought requires the

Imperative sense, and in view of the frequent use of οὐ μή with the Future

in an imperative sense in the Septuagint, and its occasional use in classi-

cal Greek, the possibility of it can hardly be denied. WM. p. 636 f., n. 4;

G.MT.297.

 

68. One or two probable instances of the Imperative Future

in the third person occur, though perhaps no entirely certain

case.  Matt. 4:4, οὐκ ἐπ'ἄρτῳ μόνῳ ζήσεται ἄνθρωπος, is prob-

ably to be so regarded, though the Hebrew of the passage

quoted (Deut. 8:3) is apparently Gnomic rather than Imper-

ative. On Matt. 15:6, see 67, Rem. 2. See also Matt. 20:

26, 27.


36                                            THE TENSES.

 

69. The Gnomic Future. The Future Indicative may

be used to state what will customarily happen when occa-

sion offers.

Rom. 5:7; μόλις γὰρ "πὲρ δικαίου τις ἀποθανεῖται, for scarcely for a

righteous man will one die. See also Gen. 44:15; Rom. 7:3, χρημα-

τίσει. Observe the Gnomic Presents both before and after.

 

70. The Deliberative Future. The Future Indicative

is sometimes used in questions of deliberation, asking not

what will happen, but what can or ought to be done.

Such questions may be real questions asking information,

or rhetorical questions taking the place of a direct asser-

tion. Cf. 169.

Luke 22:49; εἰ πατάξομεν ἐν μαχαίρῃ, shall we smite with the sword

John 6:68; κύριε, πρὸς τίνα ἀπελευσόμεθα, Lord, to whom shall we go?

 

71. PERIPHRASTIC FORM: OF THE FUTURE. A Future tense

composed of a Present Participle and the Future of the verb

εἰμί is found occasionally in the New Testament. The force

is that of a Progressive Future, with the thought of continu-

ance or customariness somewhat emphasized.

Luke 5:10; ἀνθρώπους ἔσῃ ζωγρῶν, thou shalt catch men, i.e. shalt be a

catcher of men.

Luke 21:24;  Ἰερουσαλὴν ἔσται πατουμένη, Jerusalem shall [continue

to] be trodden under foot.

 

72. Μέλλω with the Infinitive is also used with a force

akin to that of the Future Indicative. It is usually employed

of an action which one intends to do, or of that which is

certain, destined to take place.

Matt. 2:13;  μέλλει γὰρ Ἡρῴδης ζητεῖν τὸ παιδίον τοῦ ἀπολέσαι αὐτό,

            for Herod will seek the young child to destroy it.

Luke 9:44;  γὰρ υἱὸς τοῦ ἀνθρώπου μέλλει παραδίδοσθαι εἰς χείρας
           
τῶν ἀνθρώπων, for the Son of man is to be delivered up into the hands of

men. See also Matt. 16:27; 20:22; Acts 5:35; 20:38; Rom. 8:13.


THE PERFECT INDICATIVE.                             37

 

 

73. By the use of the Imperfect of  me<llw with the Infinitive

it is affirmed that at a past point of time an action was about

to take place or was intended or destined to occur.

John 7:39; τοῦτο δὲ εἶπεν περὶ τοῦ πνεύματος οὗ ἔμελλον λαμβάνειν οἱ
           
πιστεύσαντες εἰς αὐτόν, but this spake he of the Spirit which they

that believed on him were to receive. See also Luke 7:2; John 6:71.

 

THB PERFECT INDICATIVE.

 

74. The Perfect of Completed Action. In its most

frequent use the Perfect Indicative represents an action as

standing at the time of speaking complete. The reference

of the tense is thus double; it implies a past action and

affirms an existing result. HA. 847; G. 1250, 3.

Acts 5:28;  πεπληρώκατε τὴν Ἰερουσαλὴμ τπης διδαχῆς ὑμῶν, ye have

filled Jerusalem with your teaching.

Romans 5:5;  ὁτι ἀγάπη τοῦ θεοῦ ἐκκέχυται ἐν ταῖς καρδίαις ἡμπων,

because the love of God has been poured forth in our hearts.

2 Tim. 4:7; τὸν καλὸν ἀγῶνα ἠγώνισμαι, τὸν δρόμον τετέλεκα, τὴν
           
πίστιν τετήρηκα, I have fought the good fight, I have finished the

course, I have kept the faith.

 

REM. On the use of the term complete as a grammatical term, see

85. On the distinction between the Perfect and the Aorist, see 86.

 

75. The Perfect of Existing State. The Perfect is

sometimes used when the attention is directed wholly to

the present resulting state, the past action of which it is

the result being left out of thought. This usage occurs

most frequently in a few verbs which use the Perfect in

this sense only. HA. 849; G.1263.

Matt. 27:43; πέποιθεν ἐπὶ τὸν θεόν, he trusteth on God.

1 Cor. 11:2; ἐπαινῶ δὲ ὑμμᾶς, ὅτι πάντα μοῦ μέμνησθε, now praise you

that ye remember me in all things.

Luke 24:46; οὕτως γέγραπται, thus it is written, i.e. stands written.

See also Rev. 19:13.


38                                THE TENSES.

 

76. There is no sharp line of distinction between the Perfect

of Completed Action and the Perfect of Existing State. To

the latter head are to be assigned those instances in which the

past act is practically dropped from thought, and the attention

turned wholly to the existing result; while under the former

head are to be placed those instances in which it is evident

that the writer had in mind both the past act and the present

result.

77. THE INTENSIVE PERFECT. The Perfect is sometimes

used in classical Greek as an emphatic or intensive Present.

It is possible that under this head should be placed certain

Perfects of the New Testament more commonly assigned to

one of the preceding uses. Thus πέποιθα a practically expresses

the thought of πείθομαι intensified.  Πεπίστευκα is also clearly

a stronger way of saying πιστεύω. John 6:69; πεπιστεύκαμεν
καὶ ἐγνώκαμεν ὅτι σὺ εἶ ἅγιος τοῦ θεοῦ, we have believed and know

that thou art the Holy One of God. See also 2 Cor. 1:10.

Whether this usage is in the New Testament a survival of the

ancient intensive use of the Perfect, regarded by some gram.

marians as an original function of the tense (Del. IV. 94 ff.,

Br. 162), or a later development from the Perfect of com.

pleted action, affirming the present existence of the result of

a past act, need not, for the purpose of the interpreter, be

decided.

78. Of the Historical Perfect in the sense of a Perfect

which expresses a past completed action, the result of which

the speaker conceives himself to be witnessing (as in the case

of the Historical Present he conceives himself to be witness-

ing the action itself), there is no certain New Testament

instance. Possible instances are Matt. 13: 46; Luke 9:36;

2 Cor. 12:17; Jas. l:24. Cf. Br. 162. This idiom is perhaps

rather rhetorical than strictly grammatical.


THE PERFECT INDICATIVE                              39

 

Κέκραγεν in John 1:15 is a Perfect expressing a past fact

vividly conceived of as if present to the speaker. But since

the Perfect of the verb had already in classical Greek come to

be recognized as functionally a Present, it is from the point

of view of the current usage a Historical Present rather than

a Historical Perfect. Cf. L. and S. s.v.

 

79. The Perfect in 1 Cor. 7:39, de<detai, and in 1 John 2:5, τετελείω-

ται, is probably Gnomic, referring to a state that is wont to exist. If

ἀπελήλυθεν in Jas. 1:24 is Gnomic, it is with nearly the force of a Gnomic

Present or Aorist. G.MT. 154, 155.

 

80. THE AORISTIC PERFECT. The Perfect Indicative is

sometimes used in the New Testament of a simple past fact

where it is scarcely possible to suppose that the thought of

existing result was in the writer's mind. See more fully

under 88.

2Cor. 2:13;  οὐκ ἔσχηκα ἄνεσιν τῷ πνεύματί μου τῷ μὴ εὑρεῖν με Τίτον,

I had no relief for my spirit because I found not Titus.

Rev 8:5;  καὶ εἴληφεν ἄγγελος τὸν λιβανωτόν, καὶ ἐγέμισεν αὐτόν, and

the angel took the censer, and filled it. See also Matt. 25:6; 2 Cor.

1:9; 7:5; 11:25; Heb.11:28; Rev. 7:14; 19:3.

 

81. The Perfect Indicative in indirect discourse after a

verb of past time is regularly rendered into English by a

Pluperfect. This involves, however, no special use of the

tense, but results from the regular difference between English

and Greek in the matter of indirect discourse. Cf. 353.

 

82. When the Perfect Indicative is used of a past event

which is by reason of the context necessarily thought of as

separated from the moment of speaking by an interval, it is

impossible to render it into English adequately. English

idiom forbids the use of the Perfect because of the interval

(present in thought as well 'as existing in fact) between the

act and the time of speaking, while the English Past tense


40                                            THE TENSES.

 

fails to express the idea of existing result which the Greek

Perfect conveys. In most of these cases R.V. has attempted

to preserve the sense of the Greek at the expense of the Eng-

lish idiom.

Acts 7:35; τοῦτον θεὸς καὶ ἄρχοντα καὶ λυτρωτὴν ἀπέσταλκεν σὺν
           
χειρὶ ἀγγέλου τοῦ ὀφθέντος αὐτῷ ἐν τῇ βάτῳ, him did God send

[R. V. hath God sent] to be both a ruler and a deliverer with the hand

of the angel which appeared to him in the bush. See also instances

cited by Weymouth in Theological Monthly, IV. 168 f.; Rom. 16:7,

who also were [γέγοναν, R. V. have been] in Christ before me; John

6:25, R. V. correctly, when camest γέγονας] thou here? Heb. 7:

6, 9; 8:5.

 

These cases should not be confused with those treated under

80. Here the Greek tense has its normal force, though it can-

not be well rendered by its usual English equivalent. There

the use of the Greek tense is somewhat abnormal.

 

83. For the Perfect used proleptically, see 50.

 

84. PERIPHRASTIC FORM OF THE PERFECT. Periphrastic

Perfects, formed by adding a Perfect Participle to the

Present of the verb εἰμί are frequent in the New Testament,

about forty instances occurring. In function these forms

more frequently denote existing state, though clear instances

of the Perfect denoting completed action occur. The former

use is illustrated in Luke 20:6; John 2:17; Acts 2:13;

25:10; 2 Cor. 4:3, etc.; the latter in Luke 23:15; Acts

26:26; Heb. 4:2, etc. Cf. 431.

 

85. It is important to observe that the term "complete"

or "completed" as a grammatical term does not mean ended,

but accomplished, i.e. brought to its appropriate result, which

result remains at the time denoted by the verb. "The Perfect,

although it implies the performance of the action in past time.

yet states only that it stands completed at the present time."


THE PERFECT INDICATIVE                               41

 

G. MT. 44. "Das Perf. hatte zwei altuberkommene Funktio-

nen. Einerseits hatte es intensi ven, beziehentlich iterativen

Sinn. . . . Anderseits bezeichnete es die Handlung im Zustand

des Vollendet- und Fertigseins." Br. 162.

An action which has ceased may be expressed in Greek by

the Aorist or the Imperfect quite as well as by the Perfect,

provided only the action is thought of apart from any existing

result of it. These tenses are indeed more frequently used

of actions which are complete in the sense of having come to

an end than is the Perfect. See, e.g., Gal. 4:8; τότιε μὲν . . .

ἐδουλεύσατε τοῖς φύσει μὴ οὖσι θεοῖς, at that time. . . ye were in

bondage to them which by nature are no gods; and 2 Cor. 7, 8;

οὐ μεταμέλομαι· εἰ καὶ μετεμελόμην, I do not regret it, although

I did regret [was regretting] it. The Perfect, on the other

hand, affirms the existence of the normal result of the action,

and this even though the action itself is still in progress.

See, e.g., the Perfect τετήρηκα, in 2 Tim. 4:7, quoted under 74.

 

86. Since the Aorist and the Perfect both involve reference

to a past event, the Perfect affirming the existence of the

result of the event, and the Aorist affirming the event itself,

without either affirming or denying the existence of the result,

it is evident that whenever the result of the past action does

still exist, either tense may be used, according as the writer

wishes either to affirm the result or merely the event. In

many cases the reason of the choice of one tense rather than

the other is very evident and the distinction clearly marked,

even when in accordance with the principle of 82 both tenses

must be translated by an English Past. See, e.g., 1 Cor. 15:4;

ὅτι ἐτάφη, καὶ ὅτι ἐγήγερται τῇ ἡμέρᾳ τῇ τρίτῃ, that he was buried,

and thai he was raised on the third day. The burial is simply

a past event. Of the resurrection there is an existing result,

prominently before the mind.

 


42                                THE TENSES.

 

But there are naturally other cases in which, though each

tense retains its own proper force, the two approximate very

closely, and are used side by side of what seem to be quite

coordinate facts. Instances of this approximation of the two

tenses are especially frequent in the writings of John. See

John 5:36, 38; 1 John 1:1; 4:9, 10; cf. also Acts 6:11

and 15:24.

 

87. It might be supposed that the Resultative Aorist would

be especially near in force to the Perfect. The distinction is,

however, clearly marked. The Resultative Aorist affirms that

an action attempted in past time was accomplished, saying

nothing about the present result. The Perfect, on the other

hand, belongs to all classes of verbs, not merely to those that

imply attempt, and affirms the existence of the result of the

past action, the occurrence of which it implies.

 

88. It should be observed that the aoristic use of the Per-

fect (80) is a distinct departure from the strict and proper

sense of the tense in Greek. The beginnings of this departure

are to be seen in classical Greek (G.MT. 46), and in Greek

writers of a time later than the New Testament the tendency

was still further developed, until the sense of difference between

the tenses was lost.

Meantime there grew up a new form of the Perfect, made

as is the English Perfect, of an auxiliary denoting possession

(in Greek ἔχω, as in English have) and a participle. This

periphrastic Perfect, traces of which appear even in classical

times (G.MT. 47), at length entirely displaced the simple

Perfect for the expression of completed action, and the process

by which the Perfect had become an Aorist in meaning and

been succeeded in office as a Perfect tense by another form

was complete. See Jebb in Vincent and Dickson, Modern Greek,

pp. 326-330. In the New Testament we see the earlier stages


THE PERFECT INDICATIVE.                             43

 

of this process. The Perfect is still, with very few exceptions,

a true Perfect, but it has begun to be an Aorist. In Latin this

process was already complete so far as the assimilation of the

Perfect and the Aorist was concerned; the new Perfect had

not yet appeared. In modern English we see the process at a

point midway between that represented by the Greek of the

New Testament and that which appears in the Latin of about

the same time. Modern German represents about the same

stage as modern English, but a little further advanced.

It should be borne in mind that in determining whether a

given Perfect form is a true Perfect in sense or not, the

proper English translation is no certain criterion, since the

functions of the Perfect tense in the two languages differ so

widely. Cf. 52. The Perfect πεποίηκα in 2 Cor. 11:25 seems

evidently aoristic; that it "goes quite naturally into Eng-

lish" (S. p. 104) does not at all show that it has the usual

force of a Greek Perfect. Many Aorists even go quite natu-

rally and correctly into English Perfects. Cf. 46. The Per-

fects in Luke 9:36; 2 Cor. 12:11; Heb. 1:13 (προσέσχηκεν);

9:18; 11:28; Rev. 3:3; 5:1 are probably also Aoristic

Perfects, though it is possible that in all these cases the

thought of an existing result is more or less clearly in mind

and gives occasion to the use of the Perfect tense. The

Perfect pe<praken in Matt. 13:46 must be either aoristic or

historical, probably the former (see Sophocles, Glossary, etc.,

82, 4). The evidence seems to show clearly that Matthew

regularly used ge<gona in the sense of an Aorist; some of the

instances cannot, without violence, be otherwise explained, and

all are naturally so explained. Mark's use of the word is pos-

sibly the same, but the evidence is not decisive. All other

writers of the New Testament use the form as a true Perfect.

Still other cases should perhaps be explained as Aoristic

Perfects, but for the reasons mentioned in 86 it is impossible


44                                THE TENSES.

 

to decide with certainty. While there is clear evidence that

the Perfect tense was in the New Testament sometimes an

Aorist in force, yet it is to be observed that the New Testa-

ment writers had perfect command of the distinction between

the Aorist and the Perfect. The instances of the Perfect in

the sense of the Aorist are confined almost entirely to a few

forms, ἐσχηκα, εἴληφα, ἑώρακα, εἴρηκα, and γέγονα, and the use of

each of these forms in the sense of an Aorist mainly to one

or more writers whose use of it is apparently almost a per-

sonal idiosyncrasy. Thus the aoristic use of γέγονα belongs

to Matt; of εἴληφα to John in Rev.; of  ἔσχηκα to Paul; but

see also Heb. 7:13. The idiom is therefore confined within

narrow limits in the New Testament. Cf. Ev. Pet. 23, 31.

2 Cor. 12:9 and 1 John 1:10 are probably true Perfects of

Completed Action, the latter case being explained by v. 8.

John1:18; 5:37; 8:33; and Heb. 10:9 also probably con-

vey the thought of existing result, though the use of an adverb

of past time serves to give more prominence to the past action

than is usually given by a Perfect tense.

 

                                    THE PLUPERFECT.

 

89. The Pluperfect of Completed Action. The Plu-

perfect is used of an action which was complete at a point

of past time implied in the context. HA. 847; G. 1250, 4.

Acts 9:21; καὶ ὧδε εἰς τοῦτο ἐληλύθει, and he had come hither for this

            intent.

John 9:22;  ἤδη γὰρ συνετέθειντο οἱ Ἰουδαῖοι, for the Jews had agreed

already. See also Luke 8: 2; Acts 7: 44; 19: 32.

 

90. The Pluperfect of Existing State. Verbs which

in the Perfect denote a present state, in the Pluperfect

denote a past state. HA. 849, c; G. 1263.


THE PLUPERFECT.                                    45

 

Luke 4:41; ᾒδεισαν τὸν Χριστὸν αὐτὸν εἶναι, they knew that he was the

Christ. See also John 18:16, 18;  Acts 1:10.

 

91. PERIPHRASTIC FORM OF THE PLUPERFECT. A peri-

phrastic Pluperfect formed by adding the Perfect Participle

to the Imperfect of the verb εἰμί is somewhat frequent in the

New Testament. In classical Greek this was already the only

form in the third person plural of liquid and mute verbs, and

an occasional form elsewhere. In the New Testament these

periphrastic forms are frequently, but not at all uniformly,

Pluperfects of existing state; about one-third of the whole

number of instances belong to the class of Pluperfects denot-

ing completed action, referring to the past act as well as the

existing result. Cf. G.MT. 45.

Matt. 26:43;  ἦσαν γὰρ αὐτῶν οἱ ὀφθαλμοὶ βεβαρημένοι, for their eyes

were heavy, lit. weighed down.

Luke 2:26; καὶ ἦν αὐτῷ κεχρηματισμένον ὑπὸ τοῦ πνεύματος τοῦ ἁγίου,

and it had been revealed to him by the Holy Spirit.

 

92. The ambiguity of the English sometimes renders it

impossible to distinguish in translation between a Pluperfect

of Existing State and an Historical Aorist. Thus in Acts 4:21

and 31 we must, in both cases read were gathered, though the

verb in the former case is an Aorist and refers to an act, and

in the latter a Perfect and refers to a state. Cf. also the two

verbs in Luke 15:24.

 

93. The simple Future Perfect does not occur in the New

Testament. Respecting Luke 19:40, see B. p. 61; and the

lexicons s.v.

 

94. A periphrastic Future Perfect, expressing a future

state, occurs in Matt. 16:19; 18:18 ; Luke 12:52 ; Heb.

2:13.


46                                THE TENSES.

 

TENSES OF THE DEPENDENT MOODS.

 

95. The tenses of the dependent moods have in general no

reference to time, but characterize the action of the verb in

respect to its progress only, representing it as in progress,

or completed, or indefinitely, simply as an event. H.A. 851;

G. 1212,1213; G.MT.85.

 

96. The Present of the Dependent Moods is used to

represent an action as in progress or as repeated. It may

be altogether timeless, the action being thought of without

reference to the time of its occurrence; or its time, as

past, present, or future, may be involved in the function

of the mood, or may be indicated by the context.

Phil. 3:1; τὰ αὐτὰ γράφειν ὑμῖν ἐμοὶ μὲν οὐκ ὀκνηρόν, to be writing the

same things to you, to me indeed is not irksome.

Matt. 5:23; ἐὰν οὖν προσφέρῃς τὸ δῶρον σου ἐπὶ τὸ θυσιαστήριον, if

            therefore thou shalt be offering thy gift at the altar.

Mk. 12:33l καὶ τὸ ἀγπᾷν αὐτὸν ἐξ ὅλης καρδίας . . . περισσότερόν
           
ἐστιν πάντων τῶν ὁλοκαυτωμάτων καὶ θυσιῶν, and to love him with

all the heart. . . is much more than all whole burnt offerings and

sacrifices.

 

97. PERIPHRASTIC FORM OF THE PRESENT. A periphras-

tic Present Infinitive, formed by adding a Present Participle

to the Present Infinitive of εἰμί, and a periphrastic Present

Imperative, formed by adding a Present Participle to the

Present Imperative of εἰμί, occur rarely in the New Testament.

Luke 9:18; 11:1; Matt. 5:25; Luke 19:11. Cf. 20, and

431.

 

98. The Aorist of the Dependent Moods represents

the action expressed by the verb as a simple event or fact,


OF THE DEPENDENT MOODS.                         47

 

without reference either to its progress or to the existence

of its result. As in the Indicative the verb may be indefi-

nite, inceptive or resultative (cf. 35), and when indefinite

may refer to a momentary or extended action or to a

series of events (cf. 39).

The time of the action, if indicated at all, is shown, not

by the tense, but by some fact outside of it.

 

An Aorist Subjunctive after ἐάν, ὅταν, ἕως etc. is sometimes properly

translated by a Perfect or Future Perfect, but only because the context

shows that the action is to precede that of the principal verb. In the

great majority of cases a Present Subjunctive or a Future is the best

translation. See examples under 250, 285, 303, 322.

 

Luke 9:54; εἰπωμεν πῦρ καταβῆναι, shall we bid fire to come down?

John 15:9;  μείνατε ἐν τῇ ἀγάπῃ τῇ ἐμῇ, abide ye in my love.

Luke 17:4; καὶ ἐὰν ἑπτάκις τῆς ἡμέρας ἁμαρτήσῃ εἰς σὲ . . . ἀφήσεις
           
αὐτῷ,  and if he sin against thee seven times in the day. . . thou shalt

forgive him.

Acts 15:13; μετὰ δὲ τὸ σιγῆσαι αὐτοὺς, ἀπεκρίθη Ἰάκωβος, and after

they had become silent, James answered.

Acts 11:17; ἐγὼ τίς ἤμην δυνατὸς κωλῦσαι τὸν θεόν, who was I that I

could withstand God?

 

REM. Compare the Presents and Aorists in the following examples:

Matt. 6:11; τὸν ἄρτον ἡμῶν τὸν ἐπιούσιον δὸς ἡμῖν σήμερον, give us

this day our daily bread.

Luke 11:3;  τὸν ἄρτον ἡμῶν τὸν ἐπιούσιον δίδου ἡμῖν τὸ καθ' ἡμέραν,

give us day by day our daily bread.

Acts 18:9;  μὴ φοβοῦ, ἀλλὰ λάλει καῖ μὴ σιωπήσῃς, be not in fear, but

[continue to] speak and hold not thy peace.

Matt. 5:17; οὐκ ἦλθον καταλῦσαι ἀλλὰ πληρῶσαι, I came not to destroy,

but to fulfil.

John 9:4;  ἡμᾶς δεῖ ἐργάζεσθαι τὰ ἔργα τοῦ πέμψαντος με ἕως ἡμέρα

ἐστίν, we must work [be doing] the works of him that sent me while

it is day.

 

99. The Future Optative does not occur in the New Tes.

tament.


48                                            THE TENSES.

 

The Future Infinitive denotes time relatively to the time of

the principal verb. It is thus an exception to the general prin-

ciple of the timelessness of the dependent moods.

Acts 23:30; μηνυθείσης δέ μοι ἐπιβουλῆς εἰς τὸν ἄνδρα ἔσεσθαι, and

when it was shown to me that there would be a plot against the man.

 

100. The Infinitive μέλλειν with the Infinitive of another

verb dependent on it has the force of a Future Infinitive of the

latter verb. The dependent Infinitive is usually a Present,

sometimes a Future. It is regularly a Future in the New

Testament in the case of the verb dILL.

Acts 28.6; οἱ δὲ προσεδόκων αὐτὸν μέλλειν πίμπρασθαι καταπίπτειν
           
ἄφνω νεκρόν, but they expected that he would swell or fall down sud-

denly. See also Acts 19: 27; 27: 10, etc.

 

101. The Perfect of the Dependent Moods is used of

completed action. As in the Indicative, the thought may

be directed both to the action and its result, or only to the

result. The time of the action is indicated, as in the

Present and Aorist, not by the tense but by the context or

by the function of the mood.

Acts 25:25; ἐγὼ δὲ κατελαβόμην μηδὲν ἄξιον αὐτὸν θανάτου πεπραχέναι,

but I found that he had committed nothing worthy of death.

Acts 26:32; ἀπολελύσθαι ἐδύνατο ἄνθρωπος οὗτος, this man might have

            been set at liberty.

Mark 4:39; Σιώπα, πεφίμωσο, peace, be still.

 

102. AN INTENSIVE PERFECT may occur in the dependent

moods as in the Indicative.

1 Tim. 6:17; τοῖς πλουσίοις ἐν τῷ νῦν αἰῶνι παράγγελλε μὴ ὑψηλοφρονεῖν
           
μηδὲ ἠλπικέναι ἐπὶ πλούτου ἀδηλότητι,

charge them that are rich in this present world, that they be not high

minded, nor have their hope set on the uncertainty of riches.


OF THE DEPENDENTT MOODS.                                  49

 

103. PERIPHRASTIC FORM OF THE PERFECT. In the New

Testament as in classical Greek, the Perfect Subjunctive Pas-

sive is formed by adding a Perfect Participle to the Present

Subjunctive of the verb εἰμί. These forms are in the New

Testament most commonly Perfects of Existing State. John

16:24; 11:19; 2 Cor. 1:9; etc. See also Luke 12:35,

which furnishes an instance of a periphrastic Perfect Impera-

tive, enjoining the maintenance of the state denoted by the

Perfect Participle. Cf. 20 and 431.

 

104. TENSES OF THE INFINITIVE AFTER PREPOSITIONS.

The general principle that the tenses of the dependent moods

characterize the action of the verb only as respects progress

and are properly timeless holds also respecting the Infinitive

after prepositions, The Infinitive itself is properly timeless,

though the time-relation is usually suggested by the meaning

of the preposition or by this combined with that which the

tense implies respecting the progress of the action.

 

105. By μετά with the Infinitive antecedence of the action denoted by

the Infinitive to that denoted by the principal verb is expressed, but this

meaning manifestly lies in the preposition, not in the tense of the verb.

That the Aorist Infinitive is almost constantly used (the Perfect occurs

once, Heb. 10:15) is natural, since in dating one event by another the

latter is usually conceived of simply as an event without reference to its

progress, See Matt. 26:32; Luke 12:5; Acts 1:3; 1 Cor. 11:25, etc.

 

106. By πρό with the Infinitive antecedence of the action of the prin-

 cipal verb to that of the Infinitive is expressed, and the action of the

Infinitive is accordingly relatively future. But here also the time relation

is expressed wholly by the preposition. The reason for the almost uniform

use of the Aorist (the Present EIp!1L occurs John 17: 5) is the same as in

the case of meta<. See Luke 2:21; 22:15; John 1:48.

 

107. After εἰς and πρός the Infinitive usually refers to an action which

is future with respect to the principal verb. This also results from the

meaning of the prepositions, which, expressing purpose or tendency,

necessarily point to an action subsequent to that of the verb which the


50                                            THE TENSES.

 

prepositional phrase limits. When πρός means with reference to, the time-

relation is indicated only by the necessary relation of the things spoken

of. See Luke 18: 1. All three tenses of the Infinitive occur after ei]j

and both Present and Aorist after πρός, the difference marked by the tense

being not of time but of progress. See Rom. 12:2; Phil. 1:23; Heb.

11:3; Matt. 6:1; Mark 13:22. Cf. 409-414.

108. After διά the three Infinitives distinguish the action as respects

the writer's conception of its progress, as continued, completed, or indefi-

nite. Time relations are secondary and suggested. The Aorist Infinitive

occurs only in Matt. 24:12, where τό πληθυνθῆναι τήν ἀνομίαν apparently

refers to the multiplication of iniquity as a fact of that time without

exclusive reference to its preceding the action of the principal verb. The

Present Infinitive refers to action in progress usually shown by the con-

text to be contemporaneous with the action of the principal verb. See

Matt. 13:5, 6; Acts 12:20; Heb. 10:2; Jas. 4:2. The Perfect Infini-

tive has its usual force, denoting an action standing complete. The time

of the state of completeness appears from the context; it is usually that

of the principal verb. See Acts 8:11; 18: 2; 27:9; but cf. Mark 5:4,

where δεδέσθαι denotes an action whose result was existing, not at the

time of speaking, but at an earlier time. Cf. 408.

109. After ἐν we naturally expect to find only the Present Infinitive,

the preposition by its meaning suggesting an action thought of as in

progress; and this is indeed the more common usage. Luke, however,

who uses ἐν with the Infinitive far more frequently than all the other New

Testament writers, has ἐν with the Aorist Infinitive nine times, and the

same construction occurs in Hebrews twice, and in 1 Corinthians once.

Since the Aorist Infinitive conceives of an action simply as an event with-

out thought of its continuance, it is natural to take ἐν with it in the same

sense which the preposition bears with nouns which denote an event rather

than a continued action or state (cf. 98), viz. as marking the time at which

the action expressed by the principal verb takes place. The preposition

in this sense does not seem necessarily to denote exact coincidence, but

in no case expresses antecedence. In 1 Cor. 11:21 and Heb. 3:12 the

action of the Infinitive cannot be antecedent to that of the principal verb;

see also Gen. 19:16. In Luke 9:34 such a relation is very difficult,

and in Luke 14:1 improbable in view of the Imperfect tense following.

In Luke 2:27; 11:37; 19:15 ; 24:30; Acts 11:15, the action denoted

by the Infinitive, strictly speaking, precedes the action of the principal

verb, yet may be thought of by the writer as marking more or less exactly

the time at which the action of the verb takes place. As respects the


OF THE DEPENDENT MOODS.                                     51

 

relation of the action to that of the principal verb, the Aorist Infinitive

after ἐν may be compared to the Aorist Indicative after ὁτε, which simply

marks in general the time of the event denoted by the principal verb,

leaving it to the context to indicate the precise nature of the chronological

relation. See Matt. 12:3; 21:34; 27:31; John 19: 6, 30. Similarly

indefinite is the use of the English preposition on with verbal nouns, as,

e.g., On the completion of his twenty-first year he becomes of legal age;

On the arrival of the train the procession will be formed. Luke 3:21

cannot in view of the Aorist tense be rendered, while all the people were

being baptized, nor in view of the preposition ἐν, after all the people had

been baptized, but must be understood as affirming that the baptism of

Jesus occurred at the time (in general) of the baptism of all the people.

Luke 9: 36 can only mean, when the voice came, a meaning entirely

appropriate to the context. Cf. 415.

 

110. THE TENSES OF THE DEPENDENT MOODS IN INDI-

RECT DISCOURSE. The Optative and Infinitive in indirect

discourse preserve the conception of the action as respects

progress which belonged to the direct discourse. The Present

Optative and Infinitive represent tense forms which in the

direct discourse denoted action in progress. Similarly the

Aorist of these moods represents forms which expressed action

indefinitely, and the Perfect stands for forms denoting com-

pleted action. The Future represents a Future Indicative of

the direct discourse. In the majority of cases each tense of

the Optative or Infinitive in indirect discourse stands for the

same tense of the Indicative or Subjunctive of the direct form.

Yet it is doubtful whether, strictly speaking, the dependent

moods in indirect discourse express time-relations. The cor-

respondence of tenses probably rather results from the neces-

sity of preserving the original conception of the action as

respects its progress, and the time-relation is conveyed by the

context rather than by the tense of the verb.

REM. Cf. Br. 161. "Der opt. und info aor. von vergangenen Hand-

lungen als Vertreter des indo aor. in der or. obl. entbehrten ebenso wie opt.

und info praes. (§ 158) des Ausdrucks der Zeitbeziehung, die nur aus der


52                                            THE TENSES.

 

Natur der in der Rede in Verbindung gebrachten Verbalbegriffe oder aus

der ganzen in Rede stehenden Situation erkannt wurde." Cf. G.MT.

85, contra.

 

111. The Present Optative in indirect discourse in the New

Testament usually represents the Present Indicative of the

direct form. Luke 1:29; 3:15; Acts 11:11; etc. In Acts

25:16, it stands for a Present Subjunctive of the direct form.

The Optative with ἄν is taken unchanged from the direct dis-

course. Luke 1:62; 6:11; etc. The Aorist Optative occurs

in indirect discourse only in Acts 25:16, where it represents

a Subjunctive of the direct form referring to the future.

Neither the Perfect Optative nor the Future" Optative occurs in

the New Testament.

 

112. The Present Infinitive in indirect discourse in the

New Testament stands for the Present Indicative of the direct

form. Matt. 22:23; Luke 11:18; 20:41; Acts 4:32; 1 Cor.

1:36; 1 John 2:9. Similarly the Perfect Infinitive rep-

resents the Perfect Indicative of the direct discourse. Luke

22:34; John 12:29; Acts 14:19; 2 Tim. 2:18. The Pres-

ent Infinitive as the representative of the Imperfect, and the

Perfect Infinitive as the representative of the Pluperfect

(G. MT. 119, 123) apparently do not occur in the New.Testa-

ment. The Future Infinitive is, as stated above (99), an

exception to the general rule of the timelessness of the de-

pendent moods. It represents a Future Indicative of the

direct form. John 21:25; Acts 23:30; Heb. 3:18.

 

113. The Aorist Infinitive occurs in the New Testament,

as in classical Greek, as a regular construction after verbs

signifying to hope, to promise, to swear, to command, etc. In

this case the action denoted by the Aorist Infinitive is, by the

nature of the case, future with reference to that of the princi-


OF THE PARTICIPLE                               53

 

pal verb, but this time-relation is not expressed by the tense.

The' Aorist Infinitive is here as elsewhere timeless. These

instances, though closely akin in force to those of indirect

discourse, are not usually included under that head. Cf.

G.MT.684.

 

114. The Aorist Infinitive referring to what is future with

t reference to the principal verb also occurs in a few instances

after verbs of assertion. These must be accounted cases in

which the Aorist Infinitive in indirect discourse is timeless.

Luke 24:46; ὅτι ὅυτως γέγραπται παθεῖν τὸν χριστὸν καὶ ἀναστῆναι ἐκ
           
νεκρῶν τῷ τρίτῃ ἡμέρᾳ, thus it is written, that the Christ should suffer,

and rise again from the dead the third day. See also Luke 2 : 26 ;

Acts 3:18. Cf. Hom. Od. 2. 171, φημὶ τελευτηθῆναι ἅπαντα, the

accomplishment being still future (Carter in Cl. Rev. Feb. 1891,

p. 5). Plat. Euthyd. 278, C. ἐφάτην ἐπιδείψασθαι τὴν προτρεπτικὴν
           
σοφίαν, they said that they would give a sample of the hortatory

wisdom.

Protag. 316 C. τοῦτο δὲ οἴεταί οἱ μ΄λιστα γενέσθαι, εἰ σοὶ ξυγγένοιτο, and he supposes that he would be most likely to attain this if he should associate with you; and other examples in Riddell, Digest of Platonic Idioms, § 81; also in G.MT. 127.

 

There is apparently no instance in the New Testament of

the Aorist Infinitive in indirect discourse representing the

Aorist Indicative of the direct form. Cf. 890.

 

TENSES OF THE PARTICIPLE.

 

115. The participle is a verbal adjective, sharing in part

the characteristics of both the 'verb and the adjective; it de-

scribes its subject as a doer of the action denoted by the verb.

For the proper understanding of a participle three things must

be observed:

(a) The grammatical agreement.

(b) The use of the tense.

(c) The modal significance, or logical force.


54                                THE TENSES.

 

116. In grammatical agreement, a participle follows the

rule for adjectives, agreeing with its noun or pronoun in gen-

der, number, and case.

 

117. The logical force of the participle, usually the most

important consideration from the point of view of interpreta-

tion, will be treated at a later point. See 419 ff. The matter

now under consideration is the significance of the tense of a

participle.

 

118. The tenses of the participle, like those of the otherc

dependent moods, do not, in general, in themselves denote time.

To this general rule the Future Participle is the leading ex-

ception, its functions being such as necessarily to express time-

relations. The fundamental distinguishing mark of each of

the other tenses is the same for the participle as for the

dependent moods in general. The Present denotes action in

progress; the Aorist, action conceived of indefinitely; the

Perfect, completed action. These distinctions, however, im-

pose certain limitations upon the classes of events which may

be expressed by the participle of each tense, and thus indirectly

and to a limited extent, the tense of the participle is an indica-

tion of the time-relation of the event denoted by it. Since for

purposes of interpretation it is often needful to define the

time-relation of an event expressed by the participle, it becomes

expedient to treat the tenses of the participle apart from

those of the dependent moods in general.

 

THE PRESENT PARTICIPLE.

 

119. The Present Participle of Simultaneous Action.

The Present Participle most frequently denotes an action

in progress, simultaneous with the action of the principal

verb. HA. 856; G. 1288.


THE PRESENT PARTICIPLE.                             55

 

Mark 16:20; ἐκεῖνοι δὲ ἐξελθόντες ἐκήρυξαν πανταχοῦ, τοῦ κυρίου
           
συνεργοῦντος, and they went forth and preached everywhere, the Lord

working with them.

Acts 10:44; ἔτι λαλοῦντος τοῦ Πέτρου τὰ ῥήματα ταῦτα ἐπέπεσε τὸ
           
πνεῦμα τὸ ἅγιον ἐπὶ πάντας τοὺς ἀκούοντας τὸν λόγον, while Peter

was yet speaking these words, the Holy Ghost fell on all them which

heard the word.

 

REM. The action of the verb and that of the participle may be of the

same extent (Mark 16:20), but are not necessarily so. Oftener the

action of the verb falls within the period covered by the participle (Acts

10: 44).

Even a subsequent action is occasionally expressed by a Present

Participle, which in this case stands after the verb. Cf. 145.

Acts 19:9; ἀφώρφισεν τγοὺς μαθητὰς, καθ' ἡμέραν διαλεγόμενος ἐν τῇ
           
σχολῇ Τυράννου, he separated the disciples, reasoning daily in the

school of Tyrannus. See also Acts 17:13; 18:23.

 

120. The Present Participle of Identical Action.

The Present Participle not infrequently denotes the same

action which is expressed by the verb of the clause in

which it stands.

John 6:6; τοῦτο δὲ έλεγεν πειράζων αὐτόν, and this he said trying him.

See also Matt. 27 : 41; John 21 : 19; Acts 9 : 22; Gal. 3 : 23.

 

121. The verb and the participle of identical action, though

denoting the same action, usually describe it from a different

point of view.. The relation between the different points of

view varies greatly. It may be the relation of fact to method,

as in Acts 9:22; 15:24, 29; of outward form to inner sig-

nificance or quality, as in Luke 22:65; or of act to purpose

or result, as in Matt. 16:1; John 6:6.

 

122. A Present Participle of Identical Action, since it de-

notes action in progress, most naturally accompanies a verb

denoting action in progress. Sometimes, however, a Pres-

ent Participle accompanies an Aorist verb denoting the same


56                                THE TENSES.

 

action; regularly so in the phrase ἀπεκρίνατο (ἀπεκρίθη) λέγων;

see Mark 15:9; Luke 3:16; John 1:26; etc.

 

Acts 15:24; ἐτάραξαν ὑμᾶς λόγοις ἀνασκευάζυντες τὰς ψυχὰς ὑμῶν,

they have troubled you with words, subverting your souls. See also

Acts 1:3; 22:4; Gen. 43:6.

 

Similarly a Present Participle representing the action as in

progress, may accompany an Aoristic Future, which conceives

of it simply as an event. Acts 15:29; 1 Macc. 12:22.

 

123. The General Present Participle. The Present

Participle is also used without reference to time or prog-

ress, simply defining its subject as belonging to a certain

class, i.e. the class of those who do ,the action denoted

by the verb. The participle in this case becomes a simple

adjective or noun and is, like any other adjective or noun,

timeless and indefinite. B. pp. 296 f.; WM. p. 444; WT.

p. 353.

Acts 10:22; Κορνήλιος ἑκατοντάρχης, ἀνὴρ δίκαιος καὶ φοβούμενος
           
τὸν θεόν, Cornelius a centurion, a righteous and God-fearing man.

Mark 5:16; πῶς ἐγένετο τῷ δαιμονιζομένῳ, what had happened to the

            demoniac.

Ga1. 6:6;  κοινωνείτω δὲ κατηχούμενος τὸν λόγον τῷ κατηχοῦντι ἐν
           
πᾶσιν ἀγαθοῖς, but let him that is taught in the word communicate to

him that teacheth in all good things.

 

124. A class may consist of those who habitually or con-

stantly do a given act, or of those who once do the act the

single doing of which is the mark of the class. The former

case is illustrated in Matt. 5:6; the latter in Rev. 14:13.

Matt. 5:6; μακαριοι οἱ πεινῶντες καὶ διψῶντες τὴν δικαιοσύνην, blessed

            are they that hunger and thirst after righteousness.

Rev. 14:13;  μακάριοι οἱ νεκροὶ οἱ ἐν κυρίῳ ἀποθνήσκοντες, blessed are

the dead which die in the Lord. See also Matt. 7:13.


THE PRESENT PARTICIPLE.                             57

 

In the first class of cases the Present Participle only can be

used; in the second class either an Aorist (as in Matt. 23:20;

26:52; John 16:2, et al.) or a Present may occur, and that,

either in the plural designating the class as such, or in the

singular designating an individual of the class.

 

Thus παντὶ ἀνθρώπῳ περιτεμνομένῳ (Gal. 5:3; cf. 6:13) does not mean, to every man that is wont to be circumcised, but, to every man that is

circumcised, i.e. that receives circumcision (R. V., correctly though not

literally). So also in Heb. 5:1 λαμβαλόμενος does not mean, one that is

wont to be taken, but, that is taken. Being once taken is the mark of the

class here referred to, as being once circumcised is the mark of the class

referred to in Gal. 5:3. The customariness applies not to the action of

the individual member of the class, but to that of the class as a whole; as

in Heb. 5:1, the Present Indicative καθίσταται may be rendered, is wont

to be appointed, not in the sense, each one is wont to be [repeatedly]

appointed, but, it is wont to happen to each that he is appointed. Cf. 125.

In Luke 16:18 πᾶς ἀπολύων means not, everyone that is wont to

divorce, still less, every one that has divorced, but, every one that divorces

 

125. Through the ambiguity of the English Passive form,

such Present Participles as those just referred to (124) are

easily taken by the English interpreter as equivalent to Per-

fect Participles, but always to the greater or less distortion of,

the meaning of the passage}

Thus in Gal. 5:3 (see 124) περιτεμνομένῳ not equivalent to a Perfect.

every circumcised man. The apostle is not speaking of circumcision as

an accomplished fact, but of becoming circumcised. Similarly Heb. 5:1

refers not to one that has been taken (German: ist genommen worden),

but that is taken (German: wird genommen). In Heb. 5:4 καλούμενος

is one that is (not, has been) called. In Luke 13:23, εἰ ὀλίγοι οἱ

 

1 This ambiguity of the English may be illustrated by the form is

written. In the sentence, It is written in your law, etc., is written is a

Perfect of Existing State, and is expressed by the Greek Perfect γέγραπται.

The German would be ist geschrieben. In the sentence, The name of each

I scholar is written in the register as he enters the school, the same form

is a Present of customary action, and would be expressed in Greek by

γράφεται, and in German by wird geschrieben.


58                                            THE TENSES.

 

σωζόμενοι, the participle is undoubtedly a General Present, the inquiry

being neither on the one hand as to the number of those that are already

saved (Perfect of Existing State) or that have been saved (Perfect of Com.

pleted Action) nor, on the other, with reference to those that are being

saved (Progressive Present of Simultaneous Action), but with reference

to those that are [i.e. become] saved. Cf. Luther's version, meinst du,

dass wenige selig werden? and Weizsacker's, sind es wenige, die gerettet

werden.?

The same participle in Acts 2:47; 1 Cor. 1:18; 2 Cor. 2:15, may

be understood in the same way, and be rendered, we that are (in the sense

we that become) saved, or may be taken as in R. V. as a Progressive

Present of Simultaneous Action. It cannot mean the saved in the sense

of those that have been saved. The statement of Dr. T. W. Chambers in

J.B.L. June 1886, p. 40, that "the passive participle of the present tense

in Greek is often, if not generally, used to express a completed action,"

is wholly incorrect, and derives all its verisimilitude from the ambiguity

of the English Passive forms.

 

126. A General Present Participle sometimes occurs in the singular

when the person to whom it refers constitutes the class designated. This

limitation of the phrase to an individual is accomplished, however, not by

the participle, but by its limitations. John 13:11, τὸν παραδιδόντα αὐτόν,

probably means simply his betrayer. The participle paradidou<j alone

designates anyone belonging to the class of betrayers. It is the addition

of the article and an object that restrict the participle to one person.

 

127. The Present Participle for the Imperfect. The

Present Participle is also sometimes used as an Imperfect

to denote a continued action antecedent to that of the

principal verb. H.A. 856, a; G.1289; G.MT.140.

Matt. 2:20; τεθνήκασιν γὰρ οἱ ζητοῦντες τὴν ψυχὴν τοῦ παιδίου, for

they are dead that were seeking the young child's life. See also

John 12:17; Acts 4:34 (cf. v. 37); 10:7; Gal. 1:23.

 

128. The following uses of the Present Participle are

closely analogous to the uses of the Present Indicative already

described under similar names. They are of somewhat infre-

quent occurrence in the New Testament.


THE AORIST P ARTIOIPLE.                               59

 

129. (a) THE CONATIVE PRESENT.

Matt. 23:13 (WH. et al., 14); οὐδὲ τοὺς εἰσερχομένους ἀφίετε εἰσελθεῖν,

            neither suffer ye them that are entering in to enter. See also Acts

28:23.

 

130. (b) THE PRESENT FOR THE FUTURE, the action de-

noted being thought of as future with reference to the time of

the principal verb.

Acts 21:3;  ἐκεῖσε γὰρ τὸ πλοῖον ἦν ἀποφορτιζόμενον τὸν γόμον, for

there the ship was to unlade her burden.

 

131. (c) THE PRESENT OF PAST ACTION STILL IN PROGRESS,

the action denoted beginning before the action of the principal

verb and continuing in progress at the time denoted by the

latter.

Act 9:33; εὗρεν δὲ ἐκεῖ ἄνθρωπόν τινα ὀνόματι Αἰνέαν ἐξ ἐτῶν ὀκτὼ
           
κατακείμενον ἐπὶ κραβάττου, and there he found a certain man named

AEneas, who had been lying on a bed eight years. See also Matt.

9:20; Mark 5:25; Luke 8:43; John 5:5; Acts 24:10.

 

THE AORIST PARTICIPLE.

 

132. The general statement made under 118, that the

tenses of the participle do not in general in themselves denote

time, applies also to the Aorist Participle. It is very impor-

tant for the right interpretation of the Aorist Participle that

it be borne in mind that the proper and leading function of the

tense is not to express time, but to mark the fact that the

action of the verb is conceived of indefinitely, as a simple

event. The assumption that the Aorist Participle properly

denotes past time, from the point of view either of the speaker

or of the principal verb, leads to constant misinterpretation of

the form. The action denoted by the Aorist Participle may

be past, present, or future with reference to the speaker, and


60                                THE TENSES.

 

antecedent to, coincident with, or subsequent to, the action of

the principal verb. The Aorist Participle, like the participles

of the other tenses, may be most simply thought of as a noun

or adjective, the designation of one who performs the action

denoted by the verb, and like any other noun or adjective

timeless. The distinction of the Aorist Participle is not that

it expresses a different time-relation from that expressed by

the Present or Perfect, but that it conceives of the action de-

noted by it, not as in progress (Present), nor as an existing

result (Perfect), but as a simple fact. Such an adjective or

noun will not ordinarily be used if contemporaneousness

with the action of the principal verb is distinctly in mind,

since contemporaneousness suggests action in progress, and

action in progress is expressed, not by the Aorist, but by

the Present tense. Nor will it be used when the mind

distinctly contemplates the existence of the result of the

action, it being the function, not of the Aorist, but of

the Perfect, to express existing result. Nor, again, will

the Aorist noun be used if the writer desires distinctly

to indicate that the doer of the action will perform it in

time subsequent to that of the principal verb, the Aorist be-

ing incapable in itself of suggesting subsequence or futurity.

But, when these cases have been excluded, there remains a

considerable variety of relations to which the Aorist is appli-

cable, the common mark of them all being that the action

denoted by the participle is thought of simply as an event.

Among these various relations the case of action antecedent

to that of the principal verb furnishes the largest number

of instances. It is thus, numerically considered, the leading

use of the Aorist Participle, and this fact has even to some

extent reacted on the meaning of the tense, so that there is

associated with the tense as a secondary, acquired, and wholly

subordinate characteristic a certain suggestion of antecedence.


THE AORIST PARTICIPLE.                                61

 

Yet this use is no more than the other uses a primary function

of the tense, nor did it ever displace the others, or force them

into a position of subordination or abnormality. The instances

in which the action denoted by the participle is not antecedent

to the action of the principal verb are as normal as that in

which it is so, and were evidently so recognized alike in clas-

sical and in New Testament Greek. The Aorist Participle of

Antecedent Action does not denote antecedence; it is used of

antecedent action, where antecedence is implied, not by the

Aorist tense as a tense of past time, but in some other way.

The same principle holds respecting all the uses of this tense.

The following section (133) is accordingly a definition of the

constant function of the Aorist Participle, while 134, 139, and

142 enumerate the elasses of events with reference to which it

may be used.

 

REM. Compare the following statements of modern grammarians:

1.1 Since the participle, like the other non-augmented forms of the

aorist, has nothing whatever to do with the denotation of past time, and

since time previous to a point in past time is not the less a kind of past

time, we do not here understand at once how the participle became used

in this sense. But the enigma is solved when we examine the nature of

the aorist and participle. The latter, an adjective in origin, fixes one

action in relation to another. The action which is denoted by the finite

verb is the principal one. When the secondary action continues side by

side with the principal action, it must stand [παρατατικῶς] in the participle

of the present; if, again, referred to the future, the proper sign of the

future is needed; and similarly, the perfect participle serves to express

an action regarded as complete in reference to the principal action. If,

however, it is intended to denote the secondary action without any

reference to continuousness and completion and futurity, but merely

as a point or moment, the aorist participle alone remains for this

purpose. We indeed, by a sort of necessity, regard a point which

is fixed in reference to another action as prior to it, but, strictly

speaking, this notion of priority in past time is not signified by the

aorist participle."--Curtius, Elucidations of the Student's Greek Gram-

mar, pp. 216 f.


62                                THE TENSES.

 

"An und fur sich bezeichnet das aoristische Particip ebenso wenig als

irgend eine andere aoristische Form ausser dem Indicativ, der in seinem

Augment ein deutliches Merkmal der Vergangenheit hat, etwas Vergan-

genes. Das Particip. des kurzesten und yon uns genauer betrachteten

Aorists, dessen Stamm eben nur die Verbalgrundiorm selbst ist, ist also

nur Particip an und fur sich, das heisst es bezeichnet eine Handlung, mit

der noch kein Satz als abgeschlossen gedacht werden soll; im Uebrigen

liegt sein Characteristisches fur uns nur darin, dass es als aoristisches

Particip nicht wie das prasentische Particip auch die Bedeutung der

Dauer in sich enthalt, sondern etwas bezeichnet, bei dell die Zeitdauer,

die es in Anspruch genommen, nicht weiter in Frage kommen, oder das

uberhaupt nur als ganz kurze Zeit dauernd bezeichnet werden soll."

--Leo Meyer, Griechische Aoriste, pp. 124,125.

"In satzen wie ἐπειδὴ εἶπεν, ἀπῄει; εἰπὼν ταῦτα ἀπῇει; ἐάν τι φάγωσιν, (Xen. An. IV. 5, 8) erschien die syntaktisch untergeordnete aoristische Handlung gegenuber dem anderen Vorgang darum als vergarigen, weil die beiden Handlungen sachlich verschieden waren. Das Bedeu.

tungsmoment der ungeteilten Vollstandigkeit und Abgeschlossenheit der

Handlung liess die Vorstellung, dass die Haupthandlung in den Verlauf

der Nebenhandlung hineinfalle und neben ihr hergehe (Gleichzeitigkeit),

nicht zu. Die Vorstellung der Vergangenheit in Bezug auf das Haupt-

verbum war also nicht durch die Aoristform an sich, sondern durch die

besondere Natur der beiden Verbalbegriffe, die zu einander in Beziehung

gesetzt wurden, gegeben. Man erkennt diesen Sachverhalt am besten

durch Vergleichung mit Satzen wie E 98, καὶ βάλ' ἐπΐσσοντα τυχὼν κατὰ
δεξεὸν ὦον, Herod. 5, 24, εὖ ἐποίησας ἀπικόμενος. An. I. 3, 17, βου-

λοίμην δ' ἂν ἄκοντος ἀπιὼν Κύρου λαθεῖν αὐτὸν ἀπελθών, Thuk. 6, 4,

ἐγγύτατα ὀκτὼ καὶ ἑκατὸν μετὰ τὴν σφετέραν οἴκισιν Γελῷοι Ἀκράγαντα
ᾤκισαν, τὴν μὲν πόλιν ἀπὸ τοῦ Ἀκράγοντος ποταμοῦ ὀνομάσαντες, οιἰκιστὰς δὲ
ποιήσαντες Ἀριστόνουν καὶ Πυστίλον, νόμιμα δὲ τὰ Γελῴων δόντες, wo die Vorstellung einer Zeitverschiedenheit darum nicht entstehen konnte, weil es sich um ein und denselben V6rgang handelte und das Partizip oder die Partizipien nur eine, bezjehungsweise mehrere besondere Seiten der Handlung des regierenden Verbums zum Ausdruck brachten." Br. 161.

 

133. The Aorist Participle is used of an action

ceived of as a simple event.

It may be used with reference to an action or event in

its entirety (indefinite), or with reference to the inception


THE AORIST PARTICIPLE.                                63

 

of a state (inceptive), or with reference to the accomplish-

ment of an attempt (resultative). WheR indefinite it may

be used of momentary or extended actions or of a series of

events. Cf. 35, and 39, and see examples below.

 

134. The Aorist Participle of Antecedent Action.

The Aorist Participle is most frequently used of an action

antecedent in time to the action of the principal verb.

Matt. 4:2; καὶ ωνστεύσας ἡμέρας τεσσεράκοντα καὶ νύκτας τεσσερά-

Κοντα ὕστερον ἐπείνασεν, and having fasted forty days and forty

nights, he afterward hungered.

Mark 1:31; ἤγειρεν αὐτὴν κρατήσας τῆς χειρός, and taking her by the

            hand he raised her up.

John 5:13; δὲ ἰαθεὶς οὐκ ᾔδει τίς ἐστιν, but he that had been healed

            wist not who it was.

Acts 14:19; καὶ πείσαντες τοὺς ὄχλους καὶ λιθάσαντες τὸν Παῦλον,
           
ἔσυρον ἔξω τῆς πόλεως, and having persuaded the multitudes they

stoned Paul, and dragged him out of the city.

Acts 27:13; δόξαντες τῆς προθέσεως κεκρατηκέναι ἄραντες ἆσσον
           
παρελέγοντο τὴν Κρήτην, supposing that they had obtained their pur-

pose, they weighed anchor, and sailed along Crete.

Rom. 5:1; δικαιωθέντες οὖν ἐκ πίστεως εἰρήνην ἔχωμεν πρὸς τὸν θεόν,

having therefore been Justified by faith, let us have peace with God.

1 Cor. 1:4; εὐχαριστῶ τῷ θεῷ ... ἐπὶ τῇ χάριτι τοῦ θεοῦ τῇ δοθείσῃ
           
ὑμῖν, I thank God. . . for the grace of God which was given you.

Col 1:3, 4;  εὐχαριστοῦμεν τῷ θεῷ ... ἀκούσαντες τὴν πίστιν ὑμῶν,

we give thanks to God. . . having heard of your faith.

2 Tim. 4:11; Μάρκον ἀναλαβὼν ἄγε μετὰ σεαυτοῦ, take Mark and

            bring him with thee.

 

135. The Aorist Participle of Antecedent Action is fre-

quently used attributively as the equivalent of a relative

clause; in this case it usually has the article, and its position

is determined by the same considerations which govern the

position of any other noun or adjective in similar construction.

See John 5:13; 1 Cor. 1:4, above.


64                                THE TENSES.

 

136. It is still more frequently used adv-erbially and is

equivalent to an adverbial clause or coordinate verb with and,.

in this case the article does not occur, and the participle

usually precedes the verb, but sometimes follows it. See

Rom. 5:1; and Col. 1:3, 4 (134).

 

137. In some instances of the Aorist Participle of Ante-

cedent Action, it is the inception of the action only which pre-

cedes the action of the principal verb. And this occurs not

only in verbs of state (cf. 35, and see Mark 5:33; Acts 23:1),

but also in verbs of action; which in the Indicative are not

inceptive. Acts 27:13 (134); 13:27; 2 Tim. 4:10.

 

138. The Aorist Participle of Antecedent Action is by no

means always best translated into English by the so-called

Perfect Participle. The English Present Participle is very

frequently placed before a verb to express an antecedent ac-

tion, and that, too, without implying that the action is thought

of as in progress. It is accordingly in many cases the best

translation of an Aorist Participle. See Mark 1:31 (134);

also Mark 5:36; Acts 13:16, R.Y. Frequently also the

Aorist Participle of the Greek is best reproduced in English

by a finite verb with and. See Acts 14:19; 27:13; 2 Tim.

4:11 (134); also Luke 21:1; Acts 21:1; Acts 10:23, R.Y.

 

139. The Aorist Participle of Identical Action. The

Aorist Participle agreeing with the subject of a verb not

inn'equently denotes the same action that is expressed

by the verb. HA. 856, b; G. 1290; G .MT. 150.

Matt. 27:4; ἥμαρτον παραδοὺς αἷμα δίκαιον, I sinned in that I betrayed

innocent blood.

Acts 10:33; σύ τε καλῶς ἐπίησας παραγενόμενος, and thou hast well

done that thou hast come. See also Matt. 19:27 (and the numerous

instances of the phrase ἀποκριθεὶς εἶπεν); Acts 27:3; 1 Cor. 15:18;

Eph. 1:9; Heb. 7:27; Gen. 43:5.


THE AORIST PARTICIPLE.                                65

 

140. The verb and the participle of identical action, though

denoting the same action, usually describe it from a different

point of view. 'Respecting this difference in point of view,

see 121.

 

141. An Aorist Participle of Identical Action mobt fre-

quently accompanies an Aorist verb, both verb and participle

thus describing the action indefinitely as a simple event. It

occurs also with the Fnture, with which as an aoristic tense

it is entirely appropriate (Luke 9: 25; 3 John 6), with the

Present and Imperfect (Mark 8 : 29; Acts 7: 26), and with the

Perfect (Acts 13: 33; 1 Sam. 12: 19).

 

142. The Aorist Participle used attributively as the equiva...

lent of a relative clause sometimes refers to an action subse-

quent to that of the principal verb, though antecedent to the

time of the speaker. Instances occur both in classical Greek

(see G.MT. 152; Carter and Humphreys in Cl. Rev. Feb. 1891)

and in the New Testament.

Acts 1:16; ἣν προεῖπε τὸ πνεπυμα τὸ ἅγιον διὰ στόματος Δαυεὶδ περὶ
           
Ἰούδα τοῦ γενομένου ὁδηγοῦ τοῖς συλλαβοῦσιν Ἰησοῦν, which the Holy Spirit spake before by the mouth of David concerning Judas who be-

came guide to them that took Jesus. See also Matt. 10:4; 11:21;

John 11:2; Col. 1:8.

 

143. It should be clearly observed that the participle in

these cases does not by its tense denote either antecedence to

the time of speaking or subsequence to that of the principal

verb. The participle is properly timeless, and the time-rela-

tions are learned from the context or outside sources.

 

144. Whether the Aorist Participle used adverbially, as the

equivalent of an adverbial or coordinate clause, ever refers to

an action subsequent to that of the principal verb is more

difficult to determine. No certain instance has been observed

in classical Greek, though several possible ones occur. See


66                                THE TENSES.

 

Dem. XIX. (F.L.) 255 (423) cited by Carter, and Thuc. II.

49. 2, cited by Humphreys, in Cl. Rev. Feb. 1891. See also

Rom. n. E. 369; N. 35, and Pindar, Pyth. IV. 189.

 

145. The New Testament furnishes one almost indubitable

instance of an Aorist Participle so used if we accept the best

attested text.

Acts 25:13,  Ἀγρίππας βασιλεὺς καὶ Βερνίκη κατήτησαν εἰς Καισα-

ρίαν ἀσπασάμενοι τὸν Φῆστον, Agrippa the King and Bernice arrived

at Cresarea and saluted Festus.

 

The doubt concerning the text rests not on the insufficiency

of the documentary evidence, but on the rarity of this use of

the participle. Cf. Hort in WR II. App. p. 100. "The

authority for –άμενοι is absolutely overwhelming, and as a

matter of transmission –όμενοι can only be a correction. Yet

it is difficult to remain satisfied that there is no prior corrup-

tion of some kind." With this case should also be compared

Acts 16:23; 22:24; 23:35; 24:23, where the participle,

which is without the article and follows the verb, is most

naturally interpreted as referring to an action subsequent in

thought and fact to that of the verb which it follows, and

equivalent to καί with a coordinate verb. These instances are

perhaps due to Aramaic influence. See Ka. § 76. d; and cf.

Dan. 2:26, 27; 3:13, 24, 26, 27, etc.

 

In Rom. 4:19, καὶ μὴ ἀσθενήσας τῇ πίστει κατενόησεν τὸ ἑαυτοῦ σῶμα
[
ἤδη] νενεκρωμένον, the participle ἀσθενήσας, though preceding the verb, is naturally interpreted as referring to a (conceived) result of the action

denoted by κατενόησεν. It is in that case an inceptive Aorist Participle

denoting a subsequent action. Its position is doubtless due to the

emphasis laid upon it. In Heb. 9:12 the symmetry of the figure is best

preserved if εὑράμενος is thought of as referring to an action subsequent to

that of εἰσῆλθεν. But it is possible that εἰσῆλθεν is used to describe the

whole highpriestly act, including both the entrance into the holy place and

the subsequent offering of the blood, and that εὑράμενος is thus a participle

of identical action. In either case it should be translated not having


PARTICIPLE.                                  67

 

obtained as in R. V ., but obtaining or and obtained. In Phil. 2:7 γενόμενος

is related to λαβών a participle of identical action; the relation of λαβών

to ἐκένωσεν is less certain. It may denote the same action as εκένωσεν

viewed from the opposite point of view (identical action), or may be

thought of as an additional fact (subsequent action) to ἐκένωσεν. In Rom.

4:21 the participles δούς and πληροφορηθείς may be understood as together

defining ἐνεδυναμώθη τῇ πίστει, though δούς is strictly subsequent to ἐνδευναμώθη.  Somewhat similar is 1 Pet. 8:18, where ζωοποιηθείς is clearly

subsequent to ἀπέθανεν [or ἔπαθεν], but is probably to be taken together with

θανατωτείς as defining the whole of the preceding clause Χριστὸς ἅπαξ περὶ
ἁμαρτιῶν ἀπέθανεν, δίκαιος ὑπὲρ ἀδίκων, ἵνα ὑμᾶς προσαγάγῃ τῷ θεῷ.

 

            146. The Aorist Participle used as an integral part of the

object of a verb of perception represents the action which it

denotes as a simple event without defining its time. The ac-

tion may be one which is directly perceived and hence coinci-

dent in time with that of the principal verb, or it may be one

which is ascertained or learned, and hence antecedent to the

action of the principal verb. In the latter case it takes the

place of a clause of indirect discourse having its verb in

the Aorist Indicative.

Acts 9:12; καὶ εἶδεν ἄνδρα . . . Ἁνανίαν ὀνόματι εἰσελθόντα καὶ ἐπι- 

            θέντα αὐτῷ χεῖρας, and he has seen a man named Ananias come in

and lay hands upon him. See also Luke 10:18; Acts 10:3; 11:3;

26:13; 2 Pet. 1:18.

Luke 4:23; ὅσα ἠκούσαμεν γενόμενα, whatever things we have heard

            to have been done.

 

147. The Aorist Participle with λανθάνω denotes the same

time as the principal verb. It occurs but once in the New

Testament (Heb. 13:2), the similar construction with φθάνω

and τυγχάνω, not at all. HA. 856, b; G. 1290.

 

148. The categories named above, Aorist Participle of An-

tecedent Action, of Identical Action, etc., which, it must be

remembered, represent, not diverse functions of the tense, but

only classes of cases for which the Aorist Participle may be

 


68                                THE TENSES.

 

used, do not include absolutely all the instances. There are,

for example, cases in which the time-relation of the action of

the participle to that of the verb is left undefined. John

16:2, ἀποκτείνας [ὑμᾶς] δόξῃ λατρείαν προσφέρειν τῷ θεῷ, means,

every slayer of you will think, etc. Whether he will have such

thought before he shall slay, when he slays, or after he shall

have slain, is not at all defined. Cf. Gen. 4:15.

 

149. Very rarely also the Aorist Participle used adverbially

refers to an action evidently in a general way coincident in

time with the action of the verb, yet not identical with it.

Heb. 2:10; ἔπρεπεν γὰρ αὐτῷ, δι' ὃν τὰ πάντα καὶ δι' οὗ τὰ πάντα,
           
πολλοὺς υἱοὺς εἰς δόξαν ἀγαγόντα τὸν ἀρχηγὸν τῆς σωτηρίας αὐτῶν
           
διὰ παθημάτων τελειῶσαι, for it became him, for whom are all

things, and through whom are all things, in bringing many sons unto glory,

to make the author of their salvation perfect through sufferings. The

participle ἀγαγόντα is neither antecedent nor subsequent to τελειῶ-

σαι, nor yet strictly identical with it. Nearly the same thought

might be expressed in English by when he brought or in bringing,

and in Greek by ὅτε ἤγαγεν or ἐν τῷ ἀγαγεῖν (cf. 109).

 

The choice of the Aorist Participle rather than the Present

in such cases is due to the fact that the action is thought of,

not as in progress, but as a simple event or fact. Concerning

a similar use of the Aorist Participle in Homer, see Leo Meyer,

Griechische Aoriste, p. 125; T. D. Seymour in T.A.P.A., 1881,

pp. 89, 94. The rarity of these instances is due not to any

abnormality in such a use of the tense, but to the fact that

an action, temporally coincident with another and subordinate

to it (and not simply the same action viewed from a different

point of view), is naturally thought of as in progress, and

hence is expressed by a Present Participle. Cf. exx. under 119.

 

150. As an aid to interpretation it may be observed that the Aorist

Participle with the article may sometimes be used instead of a relative


THE AORIST PARTICIPLE.                                69

 

clause with the Aorist Indicative, sometimes instead of such a clause with

the verb in the Aorist Subjunctive.1 But it should not be supposed that

from the point of view of the Greek language these were two distinct

functions of the Aorist Participle. The phrase ὃς ἔλαβε referred in Greek

to past time, ὃς ἂν λάβῃ to present or future time. It is not probable that

in the mind of a Greek λαβών was the precise equivalent of both of

these, standing alternately for the one or the other, so that when he wrote

λαβών he sometimes thought ὃς ἔλαβε, sometimes ὃς ὰν λάβῃ. The fact is

doubtless rather that the Aorist Participle was always, strictly speaking,

timeless, and that λαβών meant simply the receiver, the act of receiving

being thought of as a simple fact without reference to progress. Thus for

o[ labw<n in Matt. 25:16 ὃς ἔλαβε might have stood, and it may be trans-

lated, he that received; while for ὀμόσας in Matt. 23:20 ὃς ὰν ὀμόςῃ

might have stood, and it may be trauslated, whoever sweareth; and for

ὑπομείνας in Matt. 24:13 ὃς ἂν ὑπομείνῃ might have stood, and it may

be translated, whoever shall endure. Cf. Luke 12:8-10. But these

differences are due not to a difference in the force of the tense in the

three cases. In each case a translation by a timeless verbal noun-

receiver, swearer, endurer--would correctly (though from the point

of view of English rather awkwardly) represent the thought of the

Greek. As respects the time-relation of the action of the participle

to that of the principal verb λαβών and ὑπομείνας are participles

of antecedent action, ὀμόσας is a participle of identical action. But

these distinctions, again, as stated above, are made, not to mark different

functions of the Greek tense, but to aid in a fuller interpretation of the

facts of the case.

 

151. Some scholars have endeavored to explain all participles with

the article as equivalent to the relative pronoun with the corresponding

tense of the Indicative. It is true that such participial phrases may often

be resolved in this way and the sense essentially preserved. But that

this is not a general principle will be evident from a comparison of the

fuuction of the tense in the Indicative and in the participle.

(a) All the tenses of the Indicative express time-relations from the

Ii point of view, not of the principal verb, but of the speaker. This principle

holds in a relative clause as well. as in a principal sentence. An Aorist

verb standing in a relative clause may indeed refer to an action antece-

dent to the time of the principal verb, but this antecedence is not expressed

by the tense of the verb. All that the Aorist tense does in respect to

 

1 W. G. Ballantine, Attributive Aorist Participles in Protasis, in Bio.

Sac. Apr. 1889.


70                                THE TENSES.

time is to place the action in past time; its relation in that past time to

the action of the principal verb must be learned from some other source.

The corresponding thing is true of the Present tense, which in a relative

clause denotes time not contemporaneous with the action of the principal

verb, but present from the point of view of the speaker. See, e.g., Matt.

11:4; 13:17.

(b) The participle, on the other hand, is in itself timeless, and gains

whatever suggestion of time-relation it conveys from its relation to the

rest of the sentence. It is not affirmed that the Aorist Participle denotes

time relative to that of the principal verb, but that its time-relations are

not independent, like those of the fudicative, but dependent.

It is thus apparent that the whole attitude, so to speak, of the parti-

ciple toward time-relations is different from that of the Indicative, and no

formula of equivalence between them can be constructed. A timeless

noun or adjective cannot by any fixed rule be translated into a time-

expressing verb.

Somewhat less of error is introduced if the rule is made to read that

the participle may be trauslated into English by a relative clause using

that tense of the English fudicative which corresponds to the tense

of the Greek participle. Relative clauses in English frequently use the

tenses apparently to denote time relative to that of the principal verb.

Thus in the sentence, When I am in London I will come to see you, the

present tense, am, really denotes time future with reference to the speaker,

time present relative to that of the principal verb. SiIililarly in the

sentence, They that have done good shall come forth to the resurrection of

life --have done is past, not with reference to the time of speaking, but

to that of the principal verb. But such uses of tenses in English are

merely permissible, not uniform. Shall have done would be more exact

in the last sentence. Moreover, the rule as thus stated is false in principle,

and not uniformly applicable in fact. It would require, e.g., that a

Present Participle, standing in connection with an Aorist verb, should be

rendered by an English Present, instead of by an English Past as it .

should usually be. See John 2:16; Acts 10:35.

 

THE FUTURE PARTICIPLE.

 

152. The Future Participle represents an action as

future from the point of view of the principal verb. HA.

856; G. 1288.


THE PERFECT PARTICIPLE.                             71

 

Acts. 24:11; οὐ πλείους εἰσίν μοι ἡμέραι δώδεκα ἀφ' ἧς ἀνέβην προσκυ-

νήσων εἰς Ἰερουσαλήμ, it is not more than twelve days since I went

up to worship at Jerusalem.

1 Cor. 15:37; οὺ τὸ σῶμα τὸ σύμα τὸ γενησόμενον σπείρεις, thou sowest not the

body that shall be.

 

REM. The Future Participle is of later origin than the participles of

the other tenses, and is a clearly marked exception to the general time-

lessness of the participle. While its function was probably not primarily

temporal, the relations which it expressed necessarily suggested subse-

quence to the action of the principal verb, and hence gave to the tense a

temporal force. Del. IV. pp. 97 ff.; Br.163.

 

153. The Present Participle μέλλων followed by an Infini-

tive of another verb is used as a periphrasis for a Future

Participle of the latter verb, but with a somewhat different

range of use. To express that which is to take place, either

form may be used. But μέλλων is not used to express the

purpose of an action, and is used, as the Future Participle is

not, to express intention without designating the intended

action as the purpose of another act. See John 12:4 (cf. John

6:64); Acts 18: 14; 20:3,7.

 

THE PERFECT PARTIOIPLE.

 

154. The Perfect Participle is used of completed ac-

tion. Like the Perfect Indicative it may have reference

to the past action and the resulting state or only to the

resulting state. The time of the resulting state is usually

that of the principal verb. HA. 856; G.1288.

 

Acts 10:17;  οἱ ἄνδρες οἱ ἀπεσταλμένοι . . .  ἐπέστησαν ἐπὶ τὸν πυλῶνα,

the men who had been sent. . . stood before the gate.

Rom. 15:14; πεπληρωμένοι πάσης τῆς γνώσεως, filled with all knowledge.

Luke 8:46;  ἔγνων δύναμιν ἐξεληλυθυῖαν ἀπ' ἐμοῦ, I perceived that power

            had gone forth from me.


72                                THE TENSES.

 

155. The Perfect Participle stands in two passages of the New Testa-

ment as the predicate of the participle ὤν. The effect is of a Perfect

Participle clearly marked as one of existing state. See Eph. 4:18;

Col. 1:21.

 

156. The Perfect Participle is occasionally used as a Plu-

perfect to denote a state existing antecedent to the time of the

principal verb. The action of which it is the result is, of

course, still earlier.

 

John 11:44;  ἐξῆλθεν τεθνηκὼς δεδεμένος τοὺς πόδας καὶ τὰς χεῖρας
           
κειρίαις, he that was [or had been] dead came forth bound hand and

foot with grave-clothes. See also Mark 5:15, ἐσχηκότα, noting the

Present Participle in the same verse and the Aorist Participle

in v. 18; also 1 Cor. 2:7, ἀποκεκρυμμένην, comparing v. 10.


 

   THE MOODS.

 

MOODS IN PRINCIPAL CLAUSES.

 

       THE INDICATIVE MOOD.

 

157. The Indicative is primarily the mood of the un-

qualified assertion or simple question of fact. HA. 865;

G.1317.

 

John 1:1; ἐν ἀρχῇ ἦν λόγος, in the beginning was the Word.

Mark 4:7; καὶ καρπὸν οὐκ ἔδωκεν, and it yielded no fruit.

Matt. 2:2; ποῦ ἐστὶν τεχθεὶς βασιλεὺς τπων Ἰουδαίων, where is he

that is born King of the Jews?

John 1:38; τί ζητεῖτε, what are ye seeking?

 

158. The Indicative has substantially the same assertive

force in many principal clauses containing qualified assertions.

The action is conceived of as a fact, though the assertion of

the fact is qualified.

 

John 13:8;  ἐὰν μὴ νίψω σε, οὐκ ἔχεις μέρος μετ' ἐμοῦ, if I wash thee not,

thou hast no part with me.

 

159. (a) When qualified by particles such as ἄν, εἴθε, etc.,

the Indicative expresses various shades of desirability, improb-

ability, etc. Respecting these secondary uses of the Indicative

in principal clauses, see 26, 27, 248.

(b) Respecting the uses of the Future Indicative in other

than a purely assertive sense, see 67, 69, 70.

73


74                                THE MOODS.

 

(c) Respecting the uses of the Indicative in subordinate

clauses, see 185-360, passim.

 

REM. The uses of the Indicative described in 157 and 158 are substan-

tially the same in English and in Greek and occasion no special difficulty

to the English interpreter of Greek. The uses referred to in 159 exhibit

more difference between Greek and English, and each pal'ticular usage

requires separate consideration.

 

THE SUBJUNCTIVE MOOD.

 

The uses of the Subjunctive in principal clauses are as

follows:

 

160. The Hortatory Subjunctive. The Subjunctive

is used in the first person plural in exhortations, the

speaker thus exhorting others to join him in the doing of

an action. HA. 866, 1; G. 1344; B. p. 209; WM. p. 355;

G.MT. 255, 256.

 

Heb. 12:1;  δι' ὑπομονῆς τρέχωμεν τὸν προκείμενον ἡμῖν ἀγῶνα, let us

            run with patience the race that is set before us.

1 John 4:7; ἀγαπητοί, ἀπαπῶμεν ἀλλήλους, beloved, let us love one

another.

 

161. Occasionally the first person singular is used with

ἄφες or δεῦρο prefixed, the exhortation in that case becoming a

request of the speaker to the person addressed to permit him

to do something.

 

Matt. 7:4; ἄφες ἐκβάλω τὸ κάρφος ἐκ τοῦ ὀφθαλμοῦ σου, let me cast

            out the mote out of thine eye. See also Luke 6:42; Acts 7:34.

 

The sense of ἄφες in Matt. 27:49 and of ἄφετε in Mark 15:36 is doubt-

ful (see R.V. ad loc. and Th., ἀφίημι, 2, E.).

In Matt. 21: 38 (Mark 12:7) δεῦτε is prefixed to a hortatory first per-

son plural without affecting the meaning of the Subjunctive.


THE SUBJUNCTIVE.                                75

 

In none of these cases is a conjunction to be supplied before the Sub-

junctive.  Cf. the use of ἄγε, φέρε, etc., in classical Greek. G.MT. 257;

B. p. 210; WM. p. 356.

 

162. The Prohibitory Subjunctive. The Aorist Sub-

junctive is used in the second person with μή to express a

prohibition or a negative entreaty. H.A. 866, 2; G. 1346;

G.MT. 259.

 

Matt. 6:34; μὴ οὖν μεριμνήσητε εἰς τὴν αὔριον, be not therefore anxious

            for the morrow.

Heb. 3:8; μὴ σκληρύνητε τὰς καρδίας ὑμῶν, harden not your hearts.

Matt. 6:13; καὶ μὴ εἰσενέγκῃς ἡμᾶς εἰς πειρασμόν, and bring us not into

temptation.

 

163. Prohibitions are expressed either by the Aorist Sub-

junctive or by the Present Imperative, the only exceptions

being a few instances of the third person Aorist Imperative

with μή. The difference between an Aorist Subjunctive with

μή and a Present Imperative with μή is in the conception of

the action as respects its progress. H.A. 874. Thus

 

164. (a) The Aorist Subjunctive forbids the action as a

simple event with reference to the action as a whole or to its

inception, and is most frequently used when the action has

not been begun.

 

Acts 18:9; λάλει καὶ μὴ σιωπήσῃς, speak and hold not thy peace.

Rev. 7:3; μὴ ἀδικήσητε τὴν γῆν, hurt not the earth.

 

165. (b) The Present Imperative (180-184) forbids the

continuance of the action, most frequently when it is already

in progress; in this case, it is a demand to desist from the

action.

 

Mark 6:50; ἐγώ εἰμι, μὴ φοβεπισθε, it is I, be not afraid.

John 5:14; μηκέτι ἁμάρτανε, sin no more.


76                                THE MOODS.

 

When the action is not yet begun, it enjoins continued

abstinence from it.

 

Mark 13:21;  καὶ τότε ἐάν τις ὑμῖν εἴπῃ Ἴδε ὧδε χριστός Ἴδε
           
ἐκεῖ, μὴ πιστεύετε, and then if any man shall say unto you, Lo, here

is the Christ; or, Lo, there; believe it not. Cf. Matt. 24 : 23.

 

166. The Prohibitory Subjunctive occurs rarely in the third

person. 1 Cor. 16:11; 2 Thess. 2:3.

 

167. The strong negative, οὐ μή, occurs rarely in prohibi-

tions with the Aorist Subjunctive.

 

Matt. 13:14 and Acts 28:26, from Septuagint, Isa. 6:9, are probably

to be understood as prohibitory (as in the Hebrew of the passage in Isa.),

rather than emphatically predictive, as in R. V. Cf. Gen. 3:1, οὐ μὴ φάγητε

which is clearly prohibitory. G.MT. 297. Cf. 162.

In Matt. 21 : 19, on the other hand, the emphatic predictive sense, there

shall be no fruit from thee henceforward forever, is more probable, being

more consistent with general usage and entirely appropriate to the con-

text. The imperative rendering of the R.V. makes the passage doubly

exceptional, the Imperative Subjunctive being rare in the third person,

and οὐ μή being unusual in prohibitions.

 

168. The Deliberative Subjunctive. The Subjunctive

is used in deliberative questions and in rhetorical questions

having reference to the future. HA. 866, 3; G. 1358.

 

Luke 3:10; τί οὖν ποιήσωμεν, what then shall we do?

Luke 11:5; τίς ἐξ ὑμῶν ἕξει φίλον . . .  καὶ εἴπῃ αὐτῷ, which of you

            shall have a friend. . . and shall say to him?

 

169. Questions may be classified as questions of fact and

questions of deliberation. In the question of fact the speaker

asks what is (or was or will be). In the question of delibera-

tion, the speaker asks what he is to do, or what is to be done;

it concerns not fact but possibility, desirability, or necessity.

But questions may be classified also as interrogative or real

questions, and rhetorical questions. The former makes a real


THE SUBJUNCTIVE.                                77

 

inquiry (for information or advice) ; the latter is a rhetorical

substitute for an assertion, often equivalent to a negative

answer to itself, or, if the question is negative, to a positive

answer.

Since both questions of fact and questions of deliberation

may be either interrogative or rhetorical, it results that there

are four classes of questions that require to be distinguished

for purposes of interpretation.

 

(a) The interrogative question of fact.

Matt. 16:13;  τίνα λεγουσιν οἱ ἄνθρωποι εἶναι τὸν υἱὸν τοῦ ἀνθρώπου,

who do men say that the Son of man is? See also Mark 16:3;

John 7:45; Acts 17:18.

(b) The rhetorical question of fact.

1 Cor. 9:1; οὐκ εἰμὶ ἀπόστολος, am I not an apostle'

Luke 23:31; ὅτι εἰ ἐν ὑγρῷ ξύλῳ ταῦτα ποιοῦσιν, ἐν τῷ ξηρῷ τί γένηται,

for if they do these things in a green tree, what will be done in the dry?

See also Luke 11: 5; 16: 11.

 

(c) The interrogative deliberative question.

Mark 12 : 14; δῶμεν, μὴ δῶμεν, shall we give, or shall we not give' See

also Matt. 6 : 31; 18: 21; Luke 22 : 49.

 

(d) The rhetorical deliberative question.

Rom. 10:14; πῶς οὖν ἐπικαλέσωνται εἰς ὃν οὐκ ἐπίστευσαν; πῶς δὲ
           
πιστεύσωσιν οὗ οὐκ ἤκουσαν; . . . πῶς δὲ κηρύξωσιν ἐὰν μὴ ἀποστα-

λῶσιν, how then shall they call on him in whom they have not believed? how shall they believe in him whom they have not heard? . . . how shall they preach except they be sent? See also Matt. 26:54; Luke

14:34; John 6:68.

 

Interrogative questions of fact, and rhetorical questions of

fact having reference to the present or past, employ the tenses

and moods as they are used in simple declarative sentences.

Rhetorical questions of fact having reference to the future,

and all deliberative questions, use either the Subjunctive or

the Future Indicative.


78                                THE MOODS.

 

170. The verb of a deliberative question is most frequently

in the first person; but occasionally in the second or third.

Matt. 23:33; Rom. 10:14. The verb of a rhetorical question

may be of any person.

 

171. The Deliberative Subjunctive is sometimes preceded

by θέλεις, θέλετε, or βούλεσθε.  No conjunction is to be supplied

in these cases. The verb θέλειν is sometimes followed by a

clause introduced by ἵνα, but  ἵνα never occurs when the verb

θέλειν is in the second person, and the following verb in the

first person, i.e. when the relations of the verbs are such as to

make a Deliberative Subjunctive probable.

 

Luke 22:9; ποῦ θέλεις ἑτοιμάσωμεν, where wilt thou that we make ready?

See also Matt. 26:17; 27:17, 21; Mark 10:36, 51; 14:12; 15:9;

Luke 9:54; 18:41; 1 Cor. 4:21 (N.B.), and cf. (i!na) Matt. 7:12;

Mark 6:25; Luke 6:31; 1 Cor. 14:5.

 

172. The Subjunctive in Negative Assertions. The

AOlist Subjunctive is used with οὐ μή in the sense of an

emphatic Future Indicative. HA. 1032; G. 1360.

 

Heb 13:5;  οὐ μή σε ἀνῶ οὐδ' οὐ μή σε ἐγκαταλίπω,  I will in no wise

fail thee, neither will I in any wise forsake thee. See also Matt. 5:18;

Mark 13:30; Luke 9:27, et freq. Cf. Gild. in .A.J.P. III. 202 f.

 

REM. In Luke 18:7 and Rev. 15:4 the Subjunctive with ou] mh< is used

in a rhetorical question. The Subjunctive may be explained as occasioned

by the emphatic negative or by the rhetorical nature of the question.

 

173. This emphatically predictive Subjunctive is of frequent occurrence

in Hellenistic Greek. The Present Subjunctive is sometimes used with

οὐ μή in classical Greek, but no instance occurs in the New Testament.

Concerning the rare use of the Future with οὐ μή see 66; cf. Gild. u.s.


THE OPTATIVE.                                        79

 

      THE OPTATIVE MOOD.

 

174. The Optative Mood is much less frequent in the New

Testament, and in Hellenistic writers generally, than in clas-

sical Greek. Cf. Harmon, The Optative Mood in Hellenistic

Greek, in J.B.L. Dec. 1886. .

It is mainly confined to four uses, two of which are in prin-

cipal clauses.

 

175. The Optative of Wishing.  The Optative is used

without ἄν to express a wish. HA. 870; G. 1507.

1 Pet. 1:2; χάρις ὑμῖν καὶ εἰρήνη πληθυνθείη, grace to you and peace

            be multiplied.

2 Thess. 3:16; αὐτὸς δὲ κύριος τῆς εἰρήνης δῴη ὑμῖν τὴν εἰρήνην,

now the Lord of peace himself give you peace.

 

176. The Optative of Wishing occurs thirty-five times in the New

Testament: Mark 11:14; Luke 1:38; 20:16 ; Acts 8:20; Rom. 3:4;

3:6; 3:31; 6:2, 15; 7:7, 13; 9:14; 11:1, 11; 15:5, 13; 1 Cor.6:15;

Gal. 2:17; 3:21; 6:14; 1 Thess. 3:11, 12; 5:23; 2 Thess. 2:17; 3:5,

16; 2 Tim. 1:16, 18; Philem. 20; Heb. 13:21; 1 Pet. 1:2; 2 Pet. 1:2;

always, except Philem. 20, in the third person singular. It most frequently

expresseS a prayer. Mark 11 : 14 and Acts 8: 20 are peculiar in being im-

precations of evil.

 

177. The phrase μὴ γένοιτο is an Optative of Wishing which strongly

deprecates something suggested by a previous question or assertion.

Fourteen of the fifteen New Testament instances are in Paul's writings,

and in twelve of these it expresses the apostle's abhorrence of an inference

which he fears may be (falsely) drawn from his argument. Cf. Mey.

on Rom. 3:4, and Ltft. on Gal. 2: 17. On Gal. 6: 14 cf. 1 Macc. 9: 10.

 

178. The Potential Optative. The Optative with ἄν

is used to express what would happen on the fulfilment of

some supposed condition. It is thus an apodosis correla-


80                                THE MOODS.

 

tive to a protasis expressed or implied. It is usually to be

translated by the English Potential. HA. 872; G. 1327 ff.

 

Acts 8:31; πῶς γὰρ ἂν δυναίμην ἐὰν μή τις ὁδηγήσει με, how should I be

            able unless some one shall guide me?

Acts 17:18; τί ἂν θέλοι σπερμολόγος οὖτος λέγειν, what would this

            babbler wish to say?                                                     

 

179. The Optative with ἄν occurs in the New Testament only in Luke's

writings: Luke *1:62; *6:11; *9:46; [*15:26; 18:36] ; Acts *5:24;

†8: 31; *10:17; †17:18; [26:29]. Of these instan~es the six marked

with * are in indirect questions; the two marked with t are in direct

questions; those in brackets are of doubtful text; others still more

doubtful might be added. In only one instance (Acts 8:31) is the con-

dition expressed.

 

THE IMPERATIVE MOOD.

 

180. The Imperative Mood is used in commands and

exhortations. HA. 873; G. 1342.

Matt. 5:42; τῷ αἰτοῦντί σε δός, give to him that asketh thee.

1 Thess. 5:19; τὸ πνεῦμα μὴ σβέννυτε, quench not the spirit.

 

REM. Respecting other methods of expressing a command, see 67,

160-167,364.

 

181. THE IMPERATIVE MOOD is also used in entreaties and

petitions.

Mark 9:22; ἀλλ' εἴ τι δύνῃ, βοήθησον ἡμῖν σπλαγχνισθεὶς ἐφ' ἡμᾶς,

but if thou canst do anything, have compassion on us and help us.

Luke 17:5;  καὶ εἶπαν οἱ ἀπόστολοι τᾦ κυρίῳ Πρόσθες ἡμῖν πίστιν,

and the apostles said to the Lord, Increase our faith.

John 17:11;  πάτερ ἅγιε, τήρησον αὐτοῦς ἐν τῳ ὀνόματί σου, holy

Father, keep them in thy name.

 

182. THE IMPERATIVE MOOD is also used to express con-

sent, or merely to propose an hypothesis.


FINITE MOODS IN SUBORDINATE CLAUSES.                   81

 

Matt. 8:31, 32; οἱ δὲ δαίμονες παρεκάλουν αὐτόν λέγοντες  Εἰ ἐκβάλ-

λεις ἡμᾶς, ἀπόστειλον ἡμᾶς εἰς τὴν ἀγέλην τῶν χοίρων. Καὶ εἶπεν
 
αὐτοις Ὑπάγετε, and the demons besought him saying, If thou cast

us out, send us away into the herd of swine. A nd he said unto

them, Go.

John 2:19; ἀπεκρίθη Ἰησοῦς καὶ εἶπεν α'τοῖς  Λύσατε τὸν ναὸν τοῦτον
           
καὶ [ἐν ] τρισὶν ἡμέραις ἐγερῶ αὐτόν, Jesus answered and said unto

            them, Destroy this temple, and in three days I will raise it up.

1 Cor. 7:36; καὶ (εἰ) οὕτως ὀφείλει γίνεσθαι, θέλει ποιείτω· οὐχ
           
ἁμαρτάνει·  γαμείτωσαν, and if need so require, let him do what he

will; he sinneth not; let them marry.

 

183. An Imperative suggesting a hypothesis mayor may

not retain its imperative or hortatory force.

Luke 6:37; μὴ κρίνετε, καὶ οὐ μὴ κριθῆτε, judge not, and ye shall not

            be judged. Cf. John 2: 19, above.

 

184. Any ,tense of the Imperative may be used in positive

com mands, the distinction of force being that of the tenses of

the dependent moods in general. Cf. 95 ff. In prohibitions,

on the other hand, the use of the Imperative is confined almost

entirely to the Present tense. A few instances only of the

Aorist occur. Cf. 163.

 

FINITE MOODS IN SUBORDINATE CLAUSES.

 

185. Many subordinate clauses employ the moods and

tenses with the same force that they have in principal

clauses. Others, however, give to the mood or tense a force

different .from that which they usually have in principal

clauses. Hence arises the necessity for special treatment of

the moods and tenses in subordinate clauses. Principal clauses

also require discussion in so far as their mood or tense affects

or is affected by the subordinate clauses which limit them.


82                                THE MOODS.

 

186. Clauses considered as elements of the sentence may be classified il i

as follows:

 

    I. SUBSTANTIVE.

(1) As subject or predicate nominative (211-214, 357-360).

(2) As object in indirect discourse (334-356).

(3) As object after verbs of exhorting, etc. (200-204).

(4) As obJect after verbs of striving, etc; (205-210).

(5) As object after verbs of fear and danger (224-227).

 

   II. ADJECTIVE.

(1) Appositive (211, 213).

(2) Relative (289-333, in part).

(3) Definitive (215, 216, in part).

 

  III. ADVERBIAL, denoting

(1) Time (289-316, in part; 321-333).

(2) Place (289-316, in part).

(3) Condition (238-277, 296-315).

(4) Concession (278-288).

(5) Cause (228-233, 294).

(6) Purpose ([188-196], 197-199, 317).

(7) Indirect object, etc. (215, 217, in part; 318, 319).

(8) Result (218, 219, 234-237).

(9) Manner (217, 289-316, in part).

(10) Comparison, expressing equality or inequality (289-316,

in part).

 

REM. Conditional relative clauses introduced by relative pronouns,

and relative clauses denoting cause and purpose introduced in the same

way, partake at the same time of the nature of adjective and of adverbial

clauses.

 

187. The arrangement of the matter in the following sections (188-

347) is not based upon a logical classification of clauses, such as is indi-

cated in the preceding section, but in part on genetic relationships, and

In part on consIderations of practIcal converuence. The followmg 18 the

general order of treatment:

Moods in clauses introduced by final particles. . . . . . . 188-227.

Moods in clauses of cause  . . . . . . . . . . . . . . . . . . . . . .  228-233.

Moods in clauses of result. . . . . . . . . . . . . . . . . . . . . . .  234-237.


CLAUSES INTRODUCED BY FINAL PARTICLES.             83

 

Moods in conditional sentences. . . . . . . . . . . . .  238-277.

Moods in concessive sentences . . . . . . . . . . . . . 278-288.

Moods in relative clauses. . . . . . . . . . . . . . . . . .  289-333.

Definite relative clauses . . . . . . . . . . . . . 292-295.

Conditional relative clauses . . . . . . . . . . 296-316.

Relative clauses expressing purpose. . . .  317-320.

RelatIve clauses mtroduced by ἑως, etc.   321-333.

Indirect Discourse. . . . . . . . . . . . . . . . . . . . . . . . 334-356.

Construction after καὶ ἐγένετο, etc. . . . . . . . . .    357-360.

 

MOODS IN CLAUSES INTRODUCED BY PINAL PARTICLES.

 

188. CLASSIFICATION. Under the general head of clauses

introduced by final particles are included in New Testament

Greek:

(1) Pure final clauses.

(2) Object clauses after verbs of exhorting, etc.

(3) Object clauses after verbs of striving, etc.

(4) Object clauses after verbs of/earing.

(5) Subject, predicate, and appositive clauses.

(6) Complementary and epexegetic clauses.

(7) Clauses of conceived result.

 

189. General Usage. The relations expressed by the

clauses enumerated in 188 are in classical Greek expressed

in various ways, but, in the New Testament, these differ-

ences have, by a process of assimilation, to a considerable

extent disappeared. Clauses modeled after final clauses

take the place of Infinitives in various relations; the Opta-

tive disappears from this class of clauses; the distinction be-

tween the Subjunctive and the Future Indicative is par-

tially ignored. It results that the seven classes of clauses

named above conform in general to one rule, viz.:


84                                            THE MOODS.

 

Clauses introduced by a final particle usually employ the

Subjunctive after both primary and secondary tenses, less

frequently the Future Indicative.

 

REM. Concerning the Present Indicative after i!na, see 198, Rem.

 

190. Final Particles. The New Testament employs as

final particles ἵνα, ὅπως, and μή.

 

REM. The usage of the final particles in classical Greek is elaborately

discussed by Weber in Schanz, Beitrage zur historischen Syntax der

griechischen Sprache, Hefte IV., V., and by Gild. (on the basis of Weber's

work) in A.J.P. IV. 416 ff., VI. 53 if.

 

191. NEW TESTAMENT USE OF ἵνα.  Ἵνα occurs very fre-

quently in the New Testament, and with a greater variety of

usage than in classical Greek. Not only does it assume in

part the functions which in classical Greek belonged to the

other final particles, but clauses introduced by it encroach

largely upon the function of the Infinitive. This extension

of the use of lva is one of the notable characteristics of the

Greek of the New Testament and of all later Greek.  !Ina oc-

curs in the New Testament in

(1) Pure final clauses.

(2) Object clauses after verbs of exehorting, etc.

(3) Object clauses after verbs of striving, etc.

(4) Subject, predicate, and appositive clauses.

(5) Complementary and epexegetic clauses.

(6) Clauses of conceived result.

 

Of these clauses, the first class is the only one that regularly

employs i!na in classical Greek. Cf. G.MT. 311.

 

192. NEW TESTAMENT USE OF ὅπως.  Ὅπως occurs in the

New Testament, as in classical Greek, in


CLAUSES INTRODUCED BY FINAL PARTICLES.             85

 

(1) Pure final clauses.

(2) Object clauses after verbs of exhorting, etc.

(3) Object clauses after verbs of striving, etc. Cf. G.MT.

313.

 

193. NEW TESTAMENT USE OF μήΜή is used in the New

Testament, as in classical Greek, in

(1) Pure final clauses.

(2) Object clauses after verbs of striving, etc.

(3) Object clauses after verbs of fear'ing. Cf. G.MT.

807-310,339, 352.

 

194.  Ὡς, which occurs as a final particle in classical prose,

appears in a final clause in the New Testament in only one

passage and that of doubtful text, Acts 20:24.  Ὄφρα, which

was used as a final particle in epic and lyric poetry, does not

occur in the New Testament. Cf. G.MT. 312, 314.

 

195. In classical Greek, final clauses and object clauses after verbs

of striving, etc., frequently have ὅπως ἄν or ὡς ἄν. G.MT. 328; Meist.

p. 212. According to Gild. ἄν gives to the clause, except in the formal

language of inscriptions, a relative or conditional force, ?πως ἄν being

equivalent to ἤν πως. A.J.P. IV. pp. 422, 425; VI. pp. 53-73; L. and S.

o!pwj. In the New Testament ὅπως ἄν occurs four times (ὅπως alone forty-

nine times), always in a final clause proper. In Luke 2: 35; Acts 3:19

15:17 the contingent color may perhaps be detected; but in Rom. 3:4,

quoted from the Sfiptuagint, it is impossible to discover it.

 

196.   Ὅπως after verbs of fearing, which is found occasionally in

classical Greek, does not occur in the New Testament.

 

197. Pure Final Clauses. A pure final clause is one

whose office is to express the purpose of the action stated

in the predicate which it limits.

In classical Greek, final clauses take the Subjunctive


86                                THE MOODS.

 

after primary tenses; after secondary tenses either the

Optative or the Subjunctive. HA. 881; G.1365.

In the New Testament, the Optative does not occur.

The Subjunctive is regularly used after primary and sec-

ondary tenses alike.

Matt. 7:1; μἠ κρίνετε, ἵνα μὴ κριθῆτε, judge not, that ye be not judged.

Rom. 1:11; ἐπιποθῶ γὰρ ἐκεῖν ὑμᾶς, ἵνα τι μεταδῶ χάρισμα ὑμῖν πνευ-

ματικόν,  for I long to see you, that I may impart unto you some spiritual

gift.

Rom. 9:17;  εἰς αὐτὸ τοῦτο ἐξήγειρά σε ὅπως ἐνδείξωμαι ἐν σοὶ τὴν
           
δύναμίν μου, for this very purpose did I raise thee up that I might

show in thee my power.

Acts 28:27;  καὶ τοῦς ὀφθαλμοὺς αὐτῶν ἐκκάμμυσαν· μή ποτε ἴδωσιν
           
τοῖς ὀφθαλμοῖς, and their eyes they have closed; lest haply they should

perceive with their eyes.

 

198. Pure final clauses occasionally take the Future Indica-

tive in the New Testament as in classical Greek. HA. 881, c;

G.1366; B. pp. 234 f.; WM. pp. 360f.; WT. pp. 289f.

Luke 20:10; ἀπέστειλεν πρὸς γεωργοῦς δοῦλον, ἵνα . . .  δώσουσιν,

he sent to the husbandmen a servant, that they might give. See also 199.

 

REM. Some MSS. give a Present Indicative after ἵνα in John 5:20;

Gal. 6:12; Tit. 2:4; Rev. 12:6; 13:17. In 1 John 5:20 γινώσκομεν is

probably pregnant in force, "that we may know, and whereby we do

know." Ζηλοῦτε in Gal. 4: 17, and φυσιοῦσθε in 1 Cor. 4:6 are regarded

by Hort (WH. II. App. p. 167), Schmiedel (WS. p. 52), and Blass

(Grammatik, p. 207), as Subjunctives.. On John 17:3 see 213, Rem.

 

199. The Future Indicative occurs in pure final clauses in classical

Greek chiefly after ὅπως, rarely after μή, ὡς, and ὄφρα, never after ἵνα.

G.MT. 324; Weber, u.s.; Gild. u.s. The New Testament instances are

chiefly after ἵνα; a few instances occur after μή (μήποτε) and one after ὅπως.

The manuscripts show not a few variations between Subjunctive and Future

Indicative, and both forms are sometimes found together, after the same

conjunction. The following passages contain the Future, or both Future

and Subjunctive: Matt. 7:6; 13:15; Mark 14:2; Luke 14:10; 20:10;

John 7:3; 17:2 ; Acts 21:24; 28:27 ; Rom. 3:4; Gal. 2:4; 1 Pet. 3:1.


CLAUSES INTRODUCED BY FINAL PARTICLES.             87

 

200. Object Clauses after Verbs of Exhorting, etc.

In classical Greek, verbs of exhorting, commanding, entreat-

ing, and persuading are sometimes followed by an object

clause instead of the more usual Infinitive. Such a clause

usually employs ὅπως and the Future Indicative, sometimes

the Subjunctive. G. 1373; G.MT.355;

In the New Testament, object clauses after such verbs"

are frequent; they use both ἱνα and ὅπως; and employ

the Subjunctive to the exclusion of the Future Indicative.

Mark 5:18;  παρεκάλει α'τὸν δαιμονισθεὶς ἵνα μετ' αὐτοῦ , he who

had been possessed with a demon besought him that he might be with him.

Luke 10:2;  δεήθητε οὖν τοῦ κυρίου τοῦ θερισμοῦ ὅπως ἐργάτας ἐκβάλῃ
           
εἰς τὸν θρερισμὸν αὐτοῦ, pray ye therefore the Lord of the harvest that

he send forth laborers into his harvest. See also Matt. 4:3; 14:36;

16:20; Acts 23:15; 1 Cor. 1:10; 2 Cor. 8:6; Mark 13:18

(cf. Matt. 24 :20); Luke 22: 46 (cf. v. 40).

 

REM. In Eph. 1:17 δώῃ (Subjunctive) should be read rather than δῴη

(Optative).  Cf. 225, Rem. 2.

 

201. The use of  i!na, in an object clause after a verb of exhorting is

almost unknown in classical Greek. G.MT. 357. In the New Testament

fila, occurs much more frequently than ὅπως in such clauses.

 

202. The regular construction in classical Greek after verbs

of exhorting, etc., is the Infinitive. This is also in the New

Testament the most frequent construction, occurring nearly

twice as often as the  ἵνα and ὅπως clauses. Κελεύω and the

compounds of τάσσω take only the Infinitive.  Ἔντέλλομαι

employs both constructions.

 

203. Under the head of verbs of exhorting, etc., is to be in-

cluded the verb θέλω when used with reference to a command

or request addressed to another. It is frequently followed by

an object clause introduced by ἵνα. Here also belongs the verb


88                                THE MOODS.

 

εἶπον, used in the sense of command; also such phrases as

κάμπτω τὰ γόνατα (Eph. 3: 14), and μνείαν ποιοῦμαι ἐπὶ τῶν

προσευχῶν (Eph. 1:16; Philem. 4; cf. Col. 4:12), which are

paraphrases for προσεύχομαι.

 

204. In many cases a clause or Infinitive after a verb of commanding

or entreating may be regarded as a command indirectly quoted. It is

then a species of indirect discourse, though not usually included under

that head. Cf. 337, and G.MT. 684. Matt. 16:20; Mark 9:9; 13:34.

 

205. Object Clauses after Verbs of Striving, etc. In

classical Greek, verbs signifying to strive for, to take care,

to plan, to effect, are followed by ὅπως with the Future

Indicative, less frequently the Subjunctive, after both pri-

mary and secondary tenses. HA. 885; G.1372.

In the New Testament, the Subjunctive occurs more

frequently than the Future Indicative, and ŁIlia more fre-

quently than ὅπως.

John 12:10; ἐβουλεύσαντο δὲ οἱ ἀρχιερεῖς ἵνα καὶ τὸν Λάζαρον ἀπο-
           
κτείνωσιν, but the chief priests took counsel to put Lazarus also to death.

Rev. 3:9; ἰδοὺ ποιήσω αὐτοῦς ἵνα ἥξουσιν καὶ προσκυνήσουσιν ἐνώπιον
           
τῶν ποδῶν σου, καὶ γνῶσιν ὅτι ἐγὼ ἠγάπησά σε, behold, I will make them      
            to come and worship before thy feet, and to know that I have

loved thee. See also 1 Cor. 16:10; Col. 4:16, 17; Rev. 13:12, 16.

 

206. When the object clause after a verb meaning to care

for, to take heed, is negative, classical Greek sometimes uses

μή (instead of ὅπως μή) with the Subjunctive, "or less fre-

quently with the Future Indicative. G. 1375; G.MT. 354.

This is the common New Testament usage. See Matt. 24:4;

Acts 13:40; 1 Cor. 8:9; 10:12; Gal. 6:1; Col. 2:8; 1 Thess.

5:15; Heb. 3:12.

 Ὅπως μή with the Future in classical Greek, and ἵνα μή with

the Subjunctive in New Testament Greek, also occur. John

11:37; 2 John 8.


CLAUSES INTRODUCED BY FINAL PARTICLES                89

 

207.   Ὅπως occurs in the New Testament in such clauses (205) only

in Matt. 12:14; 22:15; Mark 3:6, and in all these cases after a phrase

meaning to plan. The clause thus closely approximates an indirect de-

liberative question. Cf. Mark 11:18. See Th. o!pwj, II. 2.

 

208. The Optative sometimes occurs in classical Greek after a

secondary tense of verbs of striving, etc., but is not found in the New

Testament.

 

209. It is sometimes difficult to say with certainty whether μή with

the Subjunctive after ὅρα or ὁρᾶτε is an objective clause or an independent

Prohibitory Subjunctive. In classical Greek the dependent constmction

was already fully developed (cf. G.MT. 354, 307); and though in the

New Testament lJplt is sometimes prefixed to the Imperative (Matt. 9:30;

24:6), showing that the paratactic constmction is still possible, μή with

the Subjunctive in such passages as Matt. 18:10; 1 Thess. 5:15 is best

regarded as constituting an object clause.

Μή with the Subjunctive after βλέπω is also probably to be regarded as

dependent. It is true that ble<pw does not take an objective clause in

classical Greek, that in the New Testament only the Imperative of this

verb is followed by a clause defining the action to be done or avoided, and

that in a few illstances the second verb is an Aorist Subjunctive in the

second person with μή, and might therefore be regarded as a Prohib-

itory Subjunctive (Luke 21:8; Gal. 5:15; Heb. 12:25). Yet in a

larger number of cases the verb is in the third person (Matt. 24:4; Mark

13:5; Acts 13:40; 1 Cor. 8:9, etc.), and in at least one instance is in-

troduced by i!na (1 Cor. 16:10). This indicates that we have not a coor-

dinate imperative expression, but a dependent clause. In Col. 4:17

βλέπε, and in 2 John 8 βλέπετε, is followed by i!na with the Subjunctive;

the clause in such case being probably objective, but possibly pure final.

In Heb. 3:12 the Future Indicative with μή is evidently an objective

clause.

 

REM. Concerning Luke 11:35, see B. p. 243; WM. p. 374, foot-note,

and p. 631; WT. p. 503; Th. fJ.'6, lli. 2; R. V. ad loc.

 

210. Verbs of striving, etc., may also take the Infinitive as

object. With Matt. 26:4, and John 11:53, cf. Acts 9:23;

with Rev. 13:12 cf. 13:13.

The verbs ζητέω and ἀφίημι, which are usually followed by


90                                THE MOODS.

 

an Infinitive, are each followed in one instance by ἵνα with tihe

Subjunctive. See Mark 11:16; 1 Cor. 14:12; cf. also 1 Cor.

4:2.

 

211. Subject, Predicate, and Appositive Clauses intro-

duced by ἵνα. Clauses introduced by ἵνα are frequently

used in the New Testament as subject, predicate, or appos-

itive, with a force closely akin to that of an Infinitive.

The verb is usually in the Subjunctive, less frequently in

the Future Indicative.

 

These clauses may be further classified as follows:

 

212. (a) SUBJECT of the passive of verbs of exhorting,

striving, etc., which in the active take such a clause as object,

and of other verbs of somewhat similar force. Cf. 200, 205.

 

1 Cor. 4:2.; ζητεῖται ἐν τοῖς οἰκονόμοις ἵνα πιστός τις εὑρεθῇ, it is

required in stewards that a man be found fazthful.

Rev. 9:4; καὶ ἐρρέθη αὐταῖς ἵνα μὴ ἀδικήσουσιν τὸν χόρτον τῆς γῆς,

and it was said unto them that they should not hurt the grass of the

earth. See also Mark 9:12 (γέγραπται implies command or will);

Rev. 9:5.

 

213. (b) SUBJECT, PREDICATE, OR APPOSITIVE with nouns

of various significance, especially such as are cognate with the

verbs which take such a clause as object, and with pronouns,

the clause constituting a definition of the content of the noun

or pronoun.

John 4:34; ἐμὸν βρῶμά ἐστιν ἵνα πιοήσω τὸ θέλημα τοῦ πέμψαντός
           
με καὶ τελειώσω τὸ ἔργον αὐτοῦ, my meat is to do the will of him

that sent me and to accomplish his work.

John 15:12; αὕτη ἐστιν ἐντολὴ ἐμὴ, ἵνα ἀγαπᾶτε ἀλλήλους, this

is my commandment, that ye love one another. See also Luke 1:43;

John 6:29, 39, 40; 15:8, 13; 18:39; 1 Cor. 9:18; 1 John 3:1;

2 John 6; 3 John 4.


CLAUSES INTRODUCED BY FINAL PARTICLES.               91

 

REM. The Present Indicative occurs in MSS. of John 17: 3 and is

adopted by Tisch. and Treg. (text).

214. (c) SUBJECT of phrases signifying it is profitable, it is

sufficient, etc.

 

Matt. 10:25; ἀρκετὸν τᾦ μαθητῇ ἵνα γένηται ὡς διδάσκαλος αὐτοῦ,

it is enough for the disciple that he be as his master. See also Matt.

5:29, 30; 18:6; Luke 17:2; John 11:50; 16:7; 1 Cor. 4:3.

 

215. Complementary and Epexegetic Clauses intro-

duced by  ἵνα. Clauses introduced by i!na are used in the

New Testament to express a complementary or epexegetic

limitation, with a force closely akin to that of an Infinitive.

The verb of. the clause is usually in the Subjunctive, some-

times in the Future Indicative.

 

These clauses may be classified as follows:

 

216. (a) Complementary limitation of nouns and adjec-

t tives signifying authority, power, fitness, need, set time, etc.

 

Mark 11:28;  τίς σοι ἔδωκεν τὴν ἐξουσίαν ταύτην ἵνα ταῦτα ποιῇς

or who gave thee this authority to do these things?

John 12:23; ἐλήλυθεν ὥρα ἵνα δοξασθᾖ υἱὸς τοῦ ἀνθρώπου, the

hour is come that the Son of man should be glorified. See also Matt.

8:8; Luke 7:6; John 1:27; 2:25; 16:2, 32; 1 John 2:27;

Rev. 21:23.

 

217. (b) Complementary or epexegetic limitation of verbs

of various significance; the clause defines the content, ground,

or method of the action denoted by the verb, or constitutes an

indirect object of the verb.

 

John 8:56;   Ἄβραὰμ παὴρ ὑμῶν ἠγαλλιάσατο ἵνα ἴδῃ τὴν ἡμέραν
           
τὴν ἐμήν, your father Abraham rejoiced to see my day.

 


92                                            THE MOODS.

 

Phil. 2:2; πληρώσατέ μου τὴν χαρὰν ἵνα τὸ αὐτὸ φρονῆτε, fulfil ye

my joy, that ye be of the same mind. (See an Infinitive similarly

used in Acts 15: 10.) See also John 9:22; Gal. 2:9; in both

these Jatter passages the ἵνα clause defines the content of the agree-

ment mentioned in the preceding portion of the sentence. See also

John 5:7. Cf. Martyr. Polyc. 10. 1.

 

218. Clauses of Conceived Result introduced by ἵνα.

Clauses introduced by tva are used in the New Testament

to express the conceived result of an action.

 

John 9:2;  τίς ἥμαρτεν, οὗτος οἱ γονεῖς αὐτοῦ, ἵνα τυφλὸς γεννηθῇ,

who did sin, this man or his parents, that he should be born blind?

1 Thess. 5:4;  ὑμεῖς δέ, ἀδελφοί, οὐκ ἐστὲ ἐν σκότει, ἵνα ἡμέρα
           
ὑμᾶς ὡς κλέπτας καταλάβῃ, but ye, brethren, are not in darkness, that

that day should overtake you as thieves. See also 1 John 1:9 (cf.

Heb. 6:10--Infinitive in similar construction); 2 Cor. 1:17; Rev.

9:20 (cf. Matt. 21:32); 14:13; 22:14.

 

219. The relation of thought between the fact expressed in

the principal clause and that expressed in the clause of con-

ceived result introduced by ἵνα is that of cause and effect, but

it is recognized by the speaker that this relation is one of

theory or inference rather than of observed fact. In some

cases the effect is actual and observed, the cause is inferred.

So, e.g., John 9:2. In other cases the cause is observed, the

effect is inferred. So, e.g., 1 Thess. 5:4. In all the cases the

action of the principal clause is regarded as the necessary con-

dition of that of the subordinate clause, the action of the sub-

ordinate clause as the result which is to be expected to follow

from that of the principal clause.

It is worthy of notice that in English the form of expres-

sion which ordinarily expresses pure purpose most distinctly

may also be used to express this relation of conceived result.

We say, He must have suffered very severe losses in order to be

80 reduced in circumstances. Such forms of expression are

 


CLAUSES INTRODUCED BY FINAL PARTICLES.             93

 

probably the product of false analogy, arising from imitation

of a construction which really expresses purpose. Thus in the

sentence, He labored diligently in order to accumulate property,

the subordinate clause expresses pure purpose. In the sen-

tence, He must have labored diligently in order to accumulate

such a property, the sentence may be so conceived that the sub-

ordinate clause would express purpose, but it would usually

mean rather that if he accumulated such a property he must

have labored diligently; that is, the property is conceived of

as a result the existence of which proves diligent labor. This

becomes still more evident if we say, He must have labored

diligently to have accumulated such a property. But when we

say, He must have suffered severe losses to have become so re-

duced in circumstances, it is evident that the idea of purpose

has entirely disappeared, and only that of inferred result

~ remains. Actual result observed to be the effect of observed

causes is not, however, thus expressed except by a rhetorical

figure. With these illustrations from the English, compare

the following from the Greek. Jas. 1:4; δὲ ὑπομονὴ ἔργον
τέλειον ἐχέτω, ἵνα ἦτε τέλειοι καὶ ὁλόκληροι, and let patience have

its perfect work, that ye may be perfect and entire. Heb. 10:36;

ὑπομονῆς γὰρ ἔχετε χρείαν ἵνα τὸ θέλημα τοῦ θεοῦ ποιήσαντες κομίσησθε
τὴν ἐπαγγελίαν, for ye have need of patience, that, having

done the will of God, ye may receive the promise. In the first sentence

the tva clause expresses the purpose of ἐχέτω. In the second,

though the purpose of ὑπομομή is contained in the clause ἵνα

. . . ἐπαγγελίαν, yet the function of this clause in the sentence

is not telic. Its office is not to express the purpo.se of the

principal clause, but to set forth a result (conceived, not act-

ual) of which the possession of ὑπομονή is the necessary condi-

tion. In John 9:2 the idiom is developed a step further, for

in this case the ἵνα clause in no sense expresses the purpose of

the action of the principal clause, but a fact conceived to be

 


94                                THE MOODS.

 

the result of a cause concerning which the principal clause

makes inquiry.

This use of  ἵνα with the Subjunctive is closely akin in force

to the normal force of w!ste with the Infinitive. Cf. 370, c, and

especially G.MT. 582~84.

 

220. Some of the instances under 215-217 might be considered as ex-

pressing conceived result, but the idiom has developed beyond the point

of conceived result, the clause becoming a mere complementary limita-

tion. The possible course of development may perhaps be suggested by

examining the following illustrations: John 17:2; Mark 11:28 ; Luke

7:6; 1 John 2:27. In the first case the clause probably expresses pure

purpose. In the last the idea of purpose has altogether disappeared.

 

221. In all these constructions, 211-218, which are distinct

departures from classical usage, being later invasions of the

lva clause upon the domain occupied in classical Greek by

the Infinitive, the Infinitive remains also in use in the New

Testament, being indeed in most of these constructions more

frequent than the i!na clause.

 

222. There is no certain, scarcely a probable, instance in

the New Testament of a clause introduced by ἵνα denoting

actual result conceived, of as such.

Luke 9:45 probably expresses pure purpose (cf. Matt. 11:25; WK.

p.574; WT. p. 459). Gal. 5:17 is also best explained as expressing the

purpose of the hostility of the flesh and the Spirit, viewed, so far as the

fila clause is concerned, as a hostility of the flesh to the Spirit. So, ap-

parently, R,V. Rev. 13:13 is the most probable instance of ἵνα denoting-

actual result; ἵνα . . . ποιῇ is probably equivalent to ὥστε ποιεῖν, and is

epexegetic of μεγάλα. It would be best translated, so as even to make.

Respecting ἵνα πληρωθῇ, Matt. 1:22 and frequently in the first

gospel, there is no room for doubt. The writer of the first gospel never uses

i!na to express result, either actual or conceived; and that he by this

phrase at least intends to express purpose is made especially clear by his

employment of o!pwj (which is never ecbatic) interchangeably with ἵνα.

With 1:22; 2:15; 4:14; 12:17; 21:4; 26:66, cf. 2:23 ; 8:17 ; 13:35.

 


CLAUSES INTRODUCED BY FINAL PARTICLES.             95

 

223. Concerning the post-classical usage of  ἵνα in general see Jebb in

Vincent and Dickson, Modern Greek, pp. 319-821. Concerning whether

rva in the New Testament is always in the strict sense telic, and whether

it is ever ecbatic (two distinct questions not always clearly distinguished),

see Meyer on Matt. 1:22:  “  Ἵνα ist niemals ἐκβατικόν, so dass, sondem

immer τελικόν, damit,"--the first half of which is true, the second half

far from true. Fritzsche on Matt. pp. 836 ft.; WT. pp. 457-462; WH.

pp. 573-578; B. pp. 235-240: "And although it [ἵνα] never stands in

the strict ecbatic sense (for w!ste with the finite verb), it has nevertheless

here reached the very boundary line where the difference between the

two relations (the relic and the ecbatic) disappears, and it is nearer to the

ecbatic sense than to its original final sense. Necessary as the demand is,

that in a systematic inquiry into the use pf the particle, even within a

comparatively restricted field, we should always make its original telic

force, which is the only force it has in earlier Greek writers, our point

of departure, and trace out thence the transitions to its diverse shades of

meaning; the interests of exegesis would gain very little, if in every in-

dividual passage of the N.T. even (the language of which has already

departed so far from original classic Greek usage) we should still take

pains, at the cost of the simple and natural sense, and by a recourse to

artificial means, always to introduce the telic force," p. 239. Hunzinger,

"Die in der klassischen Gracitat nicht gebrauchliche finale Bedeutung

der Partikel ἵνα im neutestamentlichen Sprachgebrauch," in Zeitschrift

für Kirchliche Wissenschaft, 1883, pp. 632-643--a valuable article which

elaborately disproves its own conclusion--"dass  ἵνα im N. T. in allen

Fallen final verstanden werden kann," UIiless a very broad and loose

sense be given to the term final.

 

224. Object Clauses after Verbs of Fear and Danger.

In classical Greek, clauses after verbs of fear and danger

employ μή with the Subjunctive after primary tenses; the

Optative, more rarely the Subjunctive, after secondary

tenses. HA. 887; G. 1378.

In the New Testament the Subjunctive only is used.

 

2 Cor. 12:20; φοβοῦμαι γὰρ μή πως ἐλθὼν οὐχ οἵους θέλω εὕρω ὑμᾶς, for

I fear, lest by any means, when I come, I shall find you not such as I

would. See also Acts 23:10; 27:29; 2 Cor. 11:3; Heb. 4:1.

 

REM. 1. Acts 5:26 may be understood as in R.V., τὸν λαόν denoting

the persons feared, and μὴ λιθασθῶσιν the thing feared (cf. the familiar


96                                            THE MOODS.

 

idiom with οἶδα illustrated in Mark 1:24; see also Gal. 4:11), so that the

meaning would be expressed in English by translating, for they were

afraid that they should be stoned by the people; or ἐφοβοῦντο... λαόν may

be taken as parenthetical, and mh> liqasqw?sin made to limit ἦγεν αὐτούς,

οὐ μετὰ βίας; so Tisch. and WH.

 

REM. 2. Some MSS. and editors read a Future Indicative in 2 Cor.

12:21.

 

225. The verb of fearing is sometimes unexpressed, the idea

of fear being suggested by the context; so, it may be, in Acts

5:39, and Matt. 25:9.

 

REM. 1. 2 Tim. 2:25, μή ποτε δώῃ αὐτοῖς θεὸς μετάνοιαν is

probably best explained in the same way. For the gentleness and meekness in

dealing with those that oppose themselves, which he has enjoined, the

apostle adds the argument, [fearing] lest God may perchance grant them

repentance, i.e. lest on the assumption that they are past repentance you

be found dealing in harshness with those to whom God will yet grant

repentance.

 

REM. 2. Δώῃ (Subjunctive) is to be preferred to δῴη (Optative) in

this passage as in Eph. 1:17. See the evidence in WS. p. 120 that this

form occurs as a Subjunctive not only in tbe Old Ionic language, but in

inscriptions of the second century B.C. Cf. WHo II. App. p. 168.

 

226. It is evident that object clauses after verbs of fear are closely

akin to negative object clauses after verbs signifying to care for. G.MT.

354. Some of the instances cited under 206 might not inappropriately

placed under 224. On the probable common origin of both, and their

development from the original parataxis, see G.MT. 307, 352.

 

227. When the object of apprehension is conceived of as

already present or past, i.e. as a thing already decided, al-

though the issue is at the time of speaking unknown, the In-

dicative is used both in classical and New Testament Greek.

HA. 888; G. 1380.

 

Gal. 4:11; φοβοῦμαι ὑμᾶς μή πως εἰκῆ κεκοπίακα εἰς ὑμᾶς, I am afraid

I have perhaps bestowed labor upon you in vain. See also Gal. 2 : 2 ;

1 Thess. 3:5; Gen. 43:11.


MOODS IN CLAUSES OF CAUSE.                   97

 

MOODS IN CLAUSES OF CAUSE.

 

228. A causal clause is one which gives either the cause or

the reason of the fact stated in the principal clause. Causal

causes are introduced by ὅτι, διότι, ἐπεί, ἐπειδή, ἐπειδήπερ, ἐφ' ,
etc. HA. 925; G. 1505.

 

229. Moods and Tenses in Causal Clauses. The

moods and tenses are used in causal clauses with the same

force as in principal clauses.

John 14:19; ὅτι ἐγὼ ζῶ καὶ ὑμεῖς ζήσετε, because I live, ye shall live also.

1 Cor. 14:12; ἐπεὶ ζηλωταί ἐστε πνευμάτων, πρὸς τῆν οἰοδομὴν τῆς
           
εκκλησίας ζητεῖτε ἵνα περισσεύητε, since ye are zealous of spiritual

gifts, seek that ye may abound unto the edifying of the church. See

also Luke 1:1; Acts 15:24; Rom. 5:12.

 

230. From the significance of a causal clause it naturally

results that its verb is usually an Indicative affirming a fact.

Any form, however, which expresses or implies either qualified

or unqualified assertion may stand after a causal conjunction.

Thus we find, e.g., a rhetorical question, or an apodosis of a

conditional sentence. In the latter case the protasis may be

omitted. In the folloWIng Instances all three of these phe-

nomena coincide; the causal clause is an apodosis, its protasis

is omitted, it is expressed in the form of a rhetorical

question.

 

1 Cor. 15:29; ἐπεὶ τί ποιήσουσιν οἱ βαπτιζόμενοι ὑπὲρ τῶν νεκρῶν, else

what shall they do which are baptized for the dead? i.e., since [if the

dead are notraisedJ they that are baptized for the dead are baptized

to no purpose.

Heb. 10:2;  ἐπεὶ ούκ ἄν ἐπαύσαντο προσφερόμεναι, else would they not

have ceased to be offered? i.e., since [if what was said above were not

true] they would have ceased to be offered. Cf. also Acts 5 : 38.


98                                THE MOODS.

 

231. From the nature of the causal clause as making an

assertion, it results that it is easily disjoined from the clause

which states the fact of which it gives the cause or reason,

and becomes an independent sentence.

 

Matt. 6:5;  καὶ ὅταν προσεύχησθε, οὐκ ἔσεσθε ὡς οἱ ὑοπκριτά· ὅτι
           
φιλοῦσιν ἐν ταῖς συναγωγαῖς καὶ ἐν ταῖς γωνίαις τῶν πλατειῶν
           
ἐστῶτες προσεύχεσθαι, and when ye pray, ye shall not be as the

hypocrites: because they love to .'?tand and pray in the synagogues and

in the corners of the streets (cf. 6: 16, where in a closely similar

sentence, γάρ is used instead of ὅτι). See also Luke 11:32;

1 Cor. 1:22, and cf. v. 21, where the same conjunction ἐπειδή intro-

duces a subordinate clause.

 

232. The distinction between a subordinate causal clause and an

independent sentence affirming a cause or reason is usually one of the

degree or emphasis on the causal relation between the two facts. When

the chief thing asserted is the e:xistence of the causal relation, as happens,

e.g., when one fact or the other is already present as a fact before the

mind, the causal clause is manifestly subordinate. When the emphasis

is upon the separate assertions as assertions, rather than on the relation

of the facts asserted, the causal clause easily becomes an independent

sentence. Thus in Rev. 3: 16, because thou art lukewarm, and neither

hot nor cold, I well spew thee out of my mouth, the causal clause is subor-

dinate. So also in John 16:3, and these things they will do, because they

have not known the Father nor me, where the words these things refer

to an assertion already made, and the intent of the sentence is to state

why they will do these things. See also John 20:29. On the other hand,

in Matt. 6:5; Luke 11:32; 1 Cor. 1:22 (see 231); and in 1 Cor. 15:29;

Heb. 10:2 (see 230), the casual clause is evidently independent, and the

particles ὅτι, ἐπεί, ἐπειδή have substantially the force of γάρ.

 

233. Causal relatiqns may also be expressed by a relative

clause (294), by an Infinitive with the article governed by διά

(408), and by a participle (439).

 


MOODS IN CLAUSES OF RESULT.                 99

 

MOODS IN CLAUSES or RESULT.

 

234. A consecutive clause is one which expresses the result,

actual or potential, of the action stated in the principal clause

or a preceding sentence.

 

In the New Testament consecutive clauses are introduced

by ὥστε. HA.927; G. 1449.

 

235. A consecutive clause commonly takes either the Indic-

ative or the Infinitive. The Indicative properly expresses the

acttIal result produced by the action previously mentioned, the

InfinitIve the result which the action of the principal verb

tends or is calculated to produce. Since, however, an actual

result may always be conceived of as that which the cause in

question is calculated or adapted to produce, the Infinitive

may be used when the result is obviously actual. Thus

if senselessness tends to credulity, one may say οὕτως ἀνόητοί
ἐστε ὥστε τὸ ἀδύνατον πιστεύτε or οὕτως ἀνόητοί ἐστε ὥστε τὸ ἀδύ-
νατον πιστεύειν, with little difference of meaning, though strictly

the latter represents believing the impossible simply as the

measure of the folly, while the former represents it as the act-

ual result of such folly. G.MT.582, 583; HA. 927; G.1450,

1451.

 

The use of the Infinitive is the older idiom. Attic writers show on

the whole a tendency to an increased use of the Indicative, Aristophanes

and Xenophon, e.g., using it more frequently than the Infinitive. See

Gild. A.J.P. VII. 161-175; XIV. 240-242. But in the New Testament the

Infinitive greatly predominates, occurring fifty times as against twenty-

one instances of the Indicative, but one of which is in a clause clearly

subordinate.

On w!ste introducing a principal clause see 237. On different concep-

tions of result, and the use of the Infinitive to express result, see 369-371.

 

236. The Indicative with w!ste. expresses actual result.

 

John 3:16; οὕτως γὰρ 'ἠγάπησεν θεὸς τ[ον κόσμον ὥστε τὸν υἱὸν τὸν
           
μονγενῇ ἔδωκεν, for God so loved the world that he gave his only

begotten Son.


100                                          THE MOODS.

 

REM. With John 3:16, which is the only clear instance in the New

Testament of ὥστε with the Indicative so closely joined to what precedes

as to constitute a subordinate clause, is usually reckoned also Gal. 2:13.

 

237. The clause introduced by w!ste is sometimes so dis-

joined from the antecedent sentence expressing the causal fact

that it becomes an independent sentence. In such cases w!ste

has the meaning therefore, or accordingly, and the verb intro-

duced by it may be in any form capable of standing in a prin-

cipal clause. HA. 927, a; G. 1454.

Mark 2:28;  ὥστε κύριός ἐστιν υἱὸς τοῦ ἀνθρώπου καῖ τοῦ σαββάτου, so

that the Son of man is lord even of the sabbath.

1 Cor. 5:8; ὥστε ἑορτάζωμεν, wherefore let us keep the feast.

1 Thess. 4:18; ὥστε παρακαλεῖτε ἀλλήλους ἐν τοῖς λόγοις τούτοις,

wherefore comfort one another with these words.

 

MOODS IN OONDITIONAL SENTENCES.

 

238. A. conditional sentence consists of a subordinate clause

which states a supposition, and a principal clause which states

a conclusion conditioned on the fulfilment of the supposition

stated in the subordinate clause. The conditional clause is

called the protasis. The principal clause is called the apodosis.

 

239. Suppositions are either particular or general. When

the protasis supposes a certain definite event and the apodosis

conditions its assertion on the occurrence of this event, the

supposition is particular. When the protasis supposes any

occurrence of an act of a certain class, and the apodosis states

what is or, was wont to take place in any instance of an act of

the class supposed in the protasis, the supposition is general.

Thus in the sentence, If he believes this act to be wrong, he will not

do it, the supposition is particular. But in the sentence, If [in any in-

stance] he believes an act to be wrong, he does not [is not wont to] do it,


MOODS IN CONDITIONAL SENTENCES.                 101

 

the supposition is general. In the sentence, If he has read this book, he

will be able to tell what it contains, the supposition is particular. But

in the sentence, If he read a book, he could always tell what it contained,

the supposition is general.

 

240. It should be noted that the occurrence of an indefinite pronoun

in the protasis does not necessarily make the supposition general. If the

writer, though using an indefinite term, refers to a particular instance,

and in the apodosis states what happened, is happening, or will happen

in this case, the supposition is particular. If, on the other hand, the

supposition refers to any instance of the class of cases described, and

the apodosis states what is or was wont to happen in any such instance,

the supposition is general. Thus, in the sentence, If anyone has eaten

any of the food, he is by this time dead, the supposition is particular.

In the sentence, If anyone [in any instance] ate any of the food, [it was

wont to happen that] he died, the supposition is general. In 2 Cor. 2:5,

but if anyone hath caused sorrow, he hath caused sorrow not to me, but

. . . to you all, the supposition refers to a specific case, and is particular.

Even the mental selection of one of many possible instances suffices to

make a supposition particular. So in 1 Cor. 3:12, it is probable that we

ought to read, if any man is building, and in 3:17, if any man is destroy-

ing, and take the clauses as referring to what was then, hypothetically,

going on rather than to what might at any time occur. On the other

hand, in John 11:9, if a man walk in the day, he stumbleth not, the sup-

position refers to any instance of walking in the day, and is general.

Concerning a protasis which refers to the truth of a general principle as

such, see 243.

 

241. Of the six classes of conditional sentences which are

found in classical Greek, five occur in the New Testament, not

however without occasional variations of form.

 

REM. 1. The classification of conditional sentences here followed is

substantially that of Professor Goodwin. The numbering of the Present

General Suppositions and Past General Suppositions as fifth and sixth

classes respectively, instead of including them as subdivisions under the

first class, is adopted to facilitate reference.

 

REM. 2. It should be observed that the titles of the several classes of

conditional sentences describe the suppositions not from the point of view

of fact, but from that of the representation of the case to the speaker's

own mind or to that of his hearer. Cf., e.g., Luke 7:39; John 18:30.


102                              THE MOODS.

 

242. A. Sinlple Present or Past Particular Supposi-

tion. The protasis simply states a supposition which refers

to a particular case in the present or past, implying

nothing as to its fulfilment.

The protasis is expressed by εἰ with a present or past

tense of the Indicative; any form of the finite verb may

stand in the apodosis. HA. 893; G.1390.

John 15:20; εἰ ἀμὲ ἐδίωξαν, καὶ ὑμᾶς διώξουσιν, if they have persecuted

me, they will also persecute you.

Gal. 5:18; εἰ δὲ πνεύματι ἄγεσθε, οὐκ ἐστὲ νόμον, but if ye are led

by the Spirit, ye are not under the law. See also Matt. 4:3; Luke

16:11; Acts 5:39; Rom.4:2; 8:10; Gal. 2:17; Rev. 20:15.

 

REM. Concerning the use of the negatives μή and οὐ in the protasis of

conditional sentences of this class, see 469, 470.

 

243. When a supposition refers to the truth of a general-

principle as such, and the apodosis conditions its assertion

on the truth of this principle, not on the occurrence of any

instance of a supposed class of events, the supposition is

particular. It is expressed in Greek by εἰ with the Indica-

tive, and the sentence belongs to the first class.

Matt. 19:10; εἰ οὕτως ἐστὶν αἰτία τοῦ ἀνθρώπου μετὰ τπης γυναικός, οὐ
           
συμφέφει γαμήσαι, if the case of the man is so with his wife, it is not

expedient to marry. See also Matt. 6 : 30; Gal. 2 : 21; cf. Plat. Prot.

340, C. In Rom. 4:14; 8:17; 11:6, the verb is omitted. The

use of ei] and the nature of the sentence, however, easily suggest

what form of the verb would be required if it were expressed.

 

244. Conditional clauses of the first class are frequently

used when the condition is fulfilled, and the use of the hypo-

thetical form suggests no doubt of the fact. This fact of ful-

filment lies, however, not in the conditional sentence, but in the

context. John 3:12; 7:23; Rom. 5:10.


MOODS SENTENCES.                              103

 

245. On the other hand, conditional clauses of the first class

may be used of what is regarded by the speaker as an unful-

filled condition. But this also is not expressed or implied by

the form of the sentence, which is in itself wholly colorless,

suggesting nothing as to the fulfilment of the condition.

Luke 23:35, 37; John 18:23; Rom. 4:2; Gal. 5:11.

 

246. Even a Future Indicative may stand in the protasis of

a conditional sentence of the first class when reference is had

to a present necessity or intention, or when the writer desires

to state not what will take place on the fulfilment of a futur6

possibility, but merely to affirm a necessary logical consequence

of a future event. 1 Cor. 9:11. Cf. G.MT.407.

 

247. In a few instances iav is used with the Present Indicative in the I

protasis of a conditional sentence, apparently to express a simple present

supposition. 1 Thess. 3:8; 1 John 5:15.

 

248. B. Supposition contrary to Fact. The protasis

states a supposition which refers to the present or past,

implying that it is not or was not fulfilled.

The protasis is expressed by el with a past tense of the

Indicative; the apodosis by a past tense of the Indicative

with ἄν. HA. 895; G. 1397.

The Imperfect denotes continued action; the Aorist

a simple fact; the Pluperfect completed action. The

time is implied in the context, not expressed by the

verb.

John 11:21; Κύριε, εἰ ἦς ὧδε οὐκ ἂν ἀπέθανεν ἀδελφός μου, Lord, if

            thou hadst been here, my brother would not have died.

Gal. 1:10; εἰ ἔτι ἀνθρώποις ἤρεσκον, Χριστοῦ δοῦλος οὐκ ἂν ἤμην, if I

            were still pleasing men, I should not be a servant of Christ. See also

John 14:28; Acts 18:14; Heb. 4:8; 11:15.


104                              THE MOODS.

 

249.  Ἄν is sometimes omitted from the apodosis. Cf. 30.

B. pp. 216 f., 225 f.; WM. pp. 382 f.; WT. pp. 305 f.; cf. G.MT.

pp. 415 ff., esp. 422, 423.

John 9:33; εἰ μὴ ἦν οὗτος παρὰ θεοῦ, οὐκ ἠδύνατο ποιεῖν οὐδέν, if this

            man were not from God, he could do nothing. See also Matt. 26:24;

John 15:22; 19:11; 1 Cor. 5:10; Gal. 4:15; Heb. 9:26.

 

250. C. Future Supposition with More Probability.

The protasis states a supposition which refers to the

future, suggesting some probability of its fulfilment.

The protasis is usually expressed by ἐάν (or ἄν) with

the Subjunctive; the apodosis by the Future Indicative or

by some other form referring to future time. RA. 898;

G. 1403.

Matt. 9:21; ἐὰν μόνον ἅψωμαι τοῦ ἱματίου αὐτοῦ σωθήσομαι, if I shall

but touch his garments, I shall be made whole.

John 12:26; ἐάν τις ἐμοὶ διακονῇ τιμήσει αὐτὸν πατήρ, if any man

serve me, him will the Father honor.

John 14:15; ἐὰν ἀγαπᾶτέ με, τὰς ἐντολὰς τὰς ἐμὰς τηρήσετε, if ye love

me, ye will keep my commandments. See also Matt. 5:20; 1 Cor.

4:19; Gal. 5:2; Jas. 2:15, 16.

 

251. In addition to ἐάν with the Subjunctive, which is the

usual form both in classical and New Testament Greek, the

following forms of protasis also occur occasionally in the New

Testament to express a future supposition with more proba-

bility:

 

252. (a) Εἰ with the Subjunctive.

Luke 9:13; οὐκ εἰσὶν ἡμῖν πλεῖον ἀρτοι πέντε καὶ ἰχθύες δύο, εἰ μήτι
           
πορευθέντες ἡμεῖς ἀγοράσωμεν εἰς πάντα τὸν λαὸν τοῦτον βρώματα,

            we have no more than five loaves and two fishes,' unless we are

to go and buy food for all this people. See also 1 Cor. 14:5; 1 Thess.

5:10; Judg. 11:9.

 


MOODS IN CONDITIONAL SENTENCES.                 105

 

253. This usage also occurs in Homer and the tragic poets, but is very

rare in Attic prose. It is found in the Septuagint and becomes very com-

mon in later Hellenistic and Byzantine writers. G. MT. 453, 454; Clapp in

T.A.P.A.1887, p. 49; 1891, pp. 88 f.; WT. pp. 294 f.; WM. pp. 368, 374, f.n.

For the few New Testament instances there is possibly in each case a

special reason. Thus in Luke 9: 13 there is probably a mixture of a con-

ditional clause and a deliberativequestion: unless indeed--are we to go?

i.e., unless indeed toe are to go. In 1 Cor. 14:5 and 1 Thess. 5:10 a

preference for the more common εἰ μή and εἴτε... εἴτε over the somewhat

unusual ἐάν μή and ἐάντε...ἐάντε may have led to the use of the former

in spite of the fact that the meaning called for a Subjunctive. 1 Thess.

5:10 can hardly be explained as attraction (B. and W.), since the nature

of the thought itself calls for a Subjunctive. On Phil. 3:11, 12, cf. 276.

It is doubtful, however, whether the discovery of any difference in force

between el with the Subjunctive and ἐάν with the Subjunctive in these

latter passages is not an over-refinement.

 

254. (b) Εἰ or ἐάν with the Future Indicative.

2 Tim. 2:12; εἰ ἀρνησόμεθα, κἀκεῖνος ἀνήσεται ἡμᾶς, if we shall deny

him, he also will deny us.

Acts 8:31; ἐὰν μή τις ὁδηγήσει με, unless some one shall guide me.

`See also Luke 19: 40.

 

255. Ei] with the Future Indicative occurs as a protasis of a condition

of the third form not infrequently in classical writers, especially in

tragedy. G.MT. 447. Of the New Testament instances of el followed

by a Future (about twenty in number), one, 2 Tim. 2:12, illustrates the

minatory or monitory force attributed to such clauses by Gild., T.A.P.A.

1876, pp. 9 ff.; A.J.P. XIII. pp. 123 ff. Concerning the other instances,

see 246, 254, 272, 276, 340.

 

256. (c) Εἰ with the Present Indicative. The protasis is

then apparently of the first class (242). The instances which "

belong here are distinguished by evident reference of the prot-

asis to the future.

Matt. 8:31; εἰ ἐκβάλλεις ἡμᾶς, ἀπόστειλον ἡμᾶς εἰς τὴν ἀγέλην τῶν
           
χοίρων, if thou cast us out send us away into the herd of swine. See

also 1 Cor. 10:27 (cf. v. 28); 2 John 10; Gen. 4:14; 20:7;

44:26; and as possible instances Matt. 5: 29, 30; 18: 8, 9;

Luke 14:26; 2 TIm. 2:12.

 


106                                          THE MOODS.

 

257. There is no distinction in form either in Greek or in English

between a particular and a general supposition referring to the future.

The distinction in thought is of course the same as in the case of present

or past suppositions (239). Thus in Matt. 9:21, if I shall but touch his

garment, I shall be made whole, the supposition evidently refers to a spe-

cific case, and is particular. But in John 16:23, if ye shall ask anything

of the Father, he will give it you in my name, the supposition is evid'ently

general. A large number of the future suppositions in the New Testa-

ment are apparently general. It is almost always possible, however, to

suppose that a particular imagined instance is mentally selected as the

illustration of the class. Cf. 240, 261.

 

258. When a conditional clause which as originally uttered

or thought was of the first or third class and expressed by

d with the Indicative or ἐάν with the Subjunctive is so

incorporated into a sentence as to be made dependent on a

verb of past time, it may be changed to εἰ with the Optative.

This principle applies even when the apodosis on which the

protasis depends is not itself strictly in indirect discourse.

Cf. 334-347, esp. 342, 347. See G.MT. 457, 694 ff.

Acts. 20:16;  ἕσπευδεν γὰρ εἰ δυνατὸν εἴη αὐτῷ τὴν ἡμέραν τῆς πεντη-
           
κοστῆς γενέσθαι εἰς  Ἰεροσόλυμα, for he was hastening, if it were pos-

sible for him, to be at Jerusalem the day of Pentecost. In this

sentence εἰ δυνατὸν εἴη represents the protasis of the sentence ἐάν

δθνατὸν γενησόμεθα which expressed the original thought of Paul,

to which the writer here refers. The same explanation applies to

Acts 24:19, and to 27:39 (unless εἰ δύναιντο is an indirect ques-

tion); also to Acts 17: 27 and 27: 12, but on these cases see

also 276.

 

259. D. Future Supposition with Less Probability.

The protasis states a supposition which refers to the future,

suggesting less probability of its fulfilment than is sug-

gested by e]a<n with the Subjunctive.

The protasis is expressed by εἰ with the Optative; the

apodosis by the Optative with ἄν.  H.A. 900; G. 1408.


MOODS IN CONDITIONAL SENTENCES.                 107

 

There is no perfect example of this form in the New Testa-

ment. Protases occur in 1 Cor. and 1 Pet., but never with a

regular and fully expressed apodosis. Apodoses occur in Luke

and Acts, but never with a regular protasis.

1 Pet. 3:17; κρεῖττον γὰρ ἀγαθοποιοῦντας, εἰ θέλοι τὸ θέλημα τοῦ θεοῦ,
           
πάσχειν κακοποιοῦντας, for it is better, if the will of God should so

will, that ye suffer for well doing than for evil doing. See also 1 Cor.

14:10; 15:37; 1 Pet. 3:14.

 

260. E. Present General Supposition. The supposi-

tion refers to any occurrence of an act of a certain class in

the (general) present, and the apodosis states what is wont

to take place in any instance of an act of the class referred

to in the protasis.

The protasis is expressed by ἐάν with the Subjunctive,

the apodosis by the Present Indicative. HA. 894, 1; G.

1393, 1.

John 11:9; ἐάν τις περιπατῇ ἐν τῇ ἡμέρᾳ, οὐ προσκόποτει, a man

walk in the day, he stumbleth not.

2 Tim 2:5; ἐὰν δὲ καὶ ἀθλῇ τις, οὐ στεφανοῦται ἐὰν μὴ νομίμως
           
ἀθλήσῃ, and if also a man contend in the games, he is not crowned,

unless he contend lawfully. See also Mark 3:24; John 7:51;

12:24; 1 Cor. 7:39, 40.

 

261. Εἰ with the Present Indicative not infrequently

occurs in clauses which apparently express a present general

supposition. G.MT.467. Yet in most New Testament pas-

sages of this kind, it is possible that a particular imagined

instance in the present or future is before the mind as an illus-

tration of the general class of cases. Cf. 242, 256. It is

scarcely possible to decide in each case whether the supposi-

tion was conceived of as general or particular.

 


108                              THE MOODS.

 

Luke 14:26; εἴ τις ἔρχεται πρός με καὶ ού μισεῖ . . .  τῆν ψυχὴν
            ἑαυτοῦ, οὐ δύναται εἶναί μου μαθητής,
if any man cometh unto me,

and hateth not. . . his own life, he cannot be my disciple. Cf. John 8:51;

12:26; where in protases of apparently similar force ἐάν with the

Subjunctive occurs, and the apodosis refers to the future.

Rom. 8:25; εἰ δὲ οὐ βλέπομεν ἐλπίζομεν, δι' ὑπομονῆς ἀπεκδεχόμεθα,

but if we hope for that which we see not, then do we with patience wait

for it. See also Jas. 1:26.

 

262. The third and fifth classes of conditional sentences are very

similar not only in form, but also in meaning. When the subject or other

leading term of the protasis is an indefinite or generic word, the third

class differs from the fifth only in that a sentence of the third class tells

what will happen in a particular instance or in any instance of the fulfil-

ment of the supposition, while a sentence of the fifth class tells what

is wont to happen in any such case. Cf., e.g., Mark 3:24 with 25; also

the two sentences of Rom. 7:3.

 

263. It should be observed that a Present Indicative in the principal

clause after a protasis consisting of tall with the Subjunctive does not

always indicate that the sentence is of the fifth class. If the fact stated

in the apodosis is already true at the time of speaking, or if the issue

involved has already been determined, though not necessarily known, the

Present Indicative is frequently used after a protasis referring to future

time. The thought would be expressed more fully but less forcibly by

supplying some such phrase as it will appear that or it will still be true

that. In other instances the true apodosis is omitted, that which stands

in its place being a reason for the unexpressed apodosis. In still other

cases the Present is merely the familiar Present for Future (15).

John 8:31; ἐὰν ὑμεῖς πείηντε ἐν τῷ λόγῳ τῷ ἐμῷ, ἀληθῶς μαθηταί μού
           
ἐστε, if ye shall abide in my word, [ye will show that] ye are truly

my disciples. Observe the Future in the next clause.

1John 1:9; ἐὰν ὁμολογῶμεν τὰς ἁμαρτίας ἡμῶν, πιστός ἐστιν καὶ
           
δίκαιος ἵνα ἀφῇ τὰς ἁμαρτίας, if we confess our sins, [he

will forgive us, for] he is faithful and righteous to forgive us our sins.

See also Mark 1 : 40; John 19: 12; Acts 26 : 5.

 

264. The difference in force between the fifth class of suppositions and

the class described under 243 should be clearly marked. There the issue

raised by the protasis is as to the truth or falsity of the principle as a. gen-


MOODS IN CONDITIONAL SENTENCES.                             109

 

eral principle, while the apodosis affirms some other general or particular

statement to be true if the general principle is true. Here the protasis

raises no question of the truth or falsity of the general principle, but

suggests as an hypothesis, that a general statement is in any single case

realized, and the apodosis states what is wont to take place when the

supposition of the protasis is thus realized. Thus in Matt. 19:10 (243)

the disciples say that if the principle stated by Jesus is true, it follows as

a general principle that it is not expedient to marry. On the other hand,

ἐὰν οὕτως ἔχῃ, οὐ συμφέρει γαμῆσαι would mean, If in any instance the

case supposed is realized, then it is wont to happen that it is not expedient to

marry. Cf. examples under 260.

 

265. F. Past General Supposition. The supposition

refers to any past occurrence of an act of a certain class,

and the apodosis states what was wont to take place in any

instance of an act of the class referred to in the protasis.

The protasis is expressed byel with the Optative, the,

apodosis by the Imperfect Indicative. HA. 894, 2; G.

1393, 2.

There is apparently no instance of this form in the New

Testament.

 

266. Peculiarities of Conditional Sentences. Nearly

all the peculiar variations of conditional sentences men-

tioned in the classical grammars are illustrated in the New

Testament. See HA. 901-907; G. 1413-1424.

 

267. (a) A protasis of one form is sometimes joined with

an apodosis of another form.

Acts 8:31; πῶς γὰρ ἄν δυναίμην ἐὰν μή τις ὁδηγήσει με, how can I,

unless some one shall guide me.'

 

268. (b) An apodosis may be accompanied by more than

one protasis; these protases may be of different form, each

retaining its own proper force.


110                              THE MOODS.

 

John 13:17; εἰ ταῦτα οἴδατε, μακάριοί ἐστε ἐὰν ποιῆτε αὐτά, if ye know

these things, blessed are ye if ye do them. See also 1 Cor. 9:11.

 

269. (c) The place of the protasis with d or Łav is some-

times supplied by a participle, an Imperative, or other form of

expression suggesting a supposition.

Matt. 26:15; Τί θέλετέ μοι δοῦναι κἀγὼ ὑμῖν παραδώσω αὐτόν, what are

ye willing to give me, and I will deliver him unto you.

Mark 11:24;  πάντα ὅσα προσεύχεσθε καὶ αἰτεῖσθε, πιστεύετε ὅτι ἐλ΄-
           
βετε, καὶ ἔσται ὑμῖν, all things whatsoever ye pray and ask for, believe

that ye have received them, and ye shall have them. See also Matt.

7:10; Mark 1:17; and exx. under 436.

 

REM. In Jas. 1:5, αἰτείτω is the apodosis of εἰ δέ τις ὑμῶν λείπεται
 
σοφίας, and at the same time fills the place of protasis to δοθήσεται.

See also Matt. 19: 21.

 

270. (d) The protasis is sometimes omitted. Luke 1:62;

Acts 17:18.

 

271. (e) The apodosis is sometimes omitted.

Luke 13:9; κἂν μὲν ποιήσῃ καρπὸν είς τὸ μέλλον εἰ δὲ μήγε, ἐκκόψεις
           
αὐτήν, and if it bear fruit thenceforth, --but if not, thou shalt cut it

down. See also Luke 19:42; Acts 23:9.

 

272. Εἰ with the Future Indicative is used by Hebraism

without an apodosis, with the force of an. emphatic negative

assertion or oath. Cf. Hr. 48, 9, a.

Mark 8:12; ἀμὴν λέγω, εἰ δοθήσεται τῇ γενεᾷ ταύτῃ σημεῖον, verily I

say unto you, there shall no sign be given unto this generation. See

also Heb. 3:11; 4:3, 5. On Heb. 6:14 see Th. ei], III. 11.

 

273. (f) The verb of the protasis or apodosis may be
omitted.

Rom. 4:14;  εἰ γὰρ οἱ ἐκ νόμου κληρονόμοι, κεκένωται πίστις καὶ
κατήργηται ἐπαγγελία, for if they which are of the law are heirs,


MOODS IN CONDITIONAL SENTENCES.                             111

 

faith is made void, and the promise is made of none effect. See also

Rom. 8:17; 11:16; 1 Cor.7:5, 8; 12:19; 1 Pet. 3:14. In 2 Cor.

11:16 κἄν stands for καὶ ἐὰν δέξησθε.

 

274. (g) Εἰ μή without a dependent verb occurs very fre-

quently in the sense of except. It may be followed by any

form of expression which could have stood as subject or as

limitation of the principal predicate. The origin of this usage

was of course in a conditional clause the verb of which was

omitted because it was identical with the verb of the apodosis.

Both in classical and New Testament Greek the ellipsis is un-

conscious, and the limitation is not strictly conditional, but ex-

ceptive. Like the English except it states not a condition on

fulfilment of which the apodosis is true or its action takes

place, but a limitation of the principal statement. It is, how-

ever, never in the New Testament purely adversative. Of.

Lift. on Gal. 1:7, 19.

 

275. (h) Ε'δ δὲ μή and εἰ δὲ μήγε are used elliptically in the

sense of otherwise, i.e. if so, or if not, to introduce an alterna-

tive statement or command. Having become fixed phrases,

they are used even when the preceding sentence is negative;

also when the nature of the condition would naturally call for

lav rather than d. Matt. 9:17; Luke 10:6; 13:9; Rev. 2:5.

G.MT. 478; B. p. 393.

 

276. (i) An omitted apodosis is sometimes virtually con-

tained in the protasis, and the latter expresses a possibility

which is an object of hope or desire, and hence has nearly the

force of a final clause. In some instances it approaches the

force of an indirect question. G.MT.486-493. In classical

Greek such protases are introduced by εἰ or ἐάν. In the New

Testament they occur with ei] only, and take the Subjunctive,

Optative, or Future Indicative.


112                              THE MOODS.

 

Phil. 3:12; διώκω δὲ εἰ καὶ καταλάβω, but I press on, if so be that I

            may apprehend.

Acts 27:12; οἱ πλείονες ἔθεντο βουλὴν ἀναχθῆναι ἐκεῖθεν, εἴ πως
           
δύναιντο καταντήσαντες εἰς Φοίνικα παραχειμάσαι, the more part

advised to put to sea from thence, if by any means they could reach

Phoenix, and winter there. See also Mark 11:13; Acts 8:22; 17:27;

Rom. 1:10; 11:14; Phil. 3:11.

 

277. (j) After expressions of wonder, etc., a clause intro-

duced by εἰ has nearly the force of a clause introduced by ὅτι.

Mark 15:44; Acts 26:8; cf. 1 John 3:13.

 

MOODS IN CONCESSIVE SENTENCES.

 

278. A concessive clause is a protasis that states a sup-

position the fulfilment of which is thought of or represented

as unfavorable to the fulfilment of the apodosis.

The force of a concessive sentence is thus very different

from that of a conditional sentence. The latter represents

the fulfilment of the apodosis as conditioned on the fulfilment

of the protasis; the former represents the apodosis as fulfilled

in spite of the fulfilment of the protasis. Yet there are cases

in which by the weakening of the characteristic force of each

construction, or by the complexity of the elements expressed

by the protasis, the two usages approach so near to each other

as to make distinction between them difficult.

In Gal. 1:8, e.g., the fulfilment of the element of the

protasis expressed in παρ' εὐηγγελισάμεθα is favorable to the

fulfilment of the apodosis ἀνάθεμα ἔστω, and the clause is so

far forth conditional. But the element expressed in ἡμεῖς
ἄγγελος ἐξ οὐρανοῦ, which is emphasized by the καί, is unfavor-

able to the fulfilment of the apodosis, and the clause is so far

forth concessive. It might be resolved into two clauses, thus,


MOODS IN CONCESSIVE SENTENCES.                    113

 

If anyone shall preach unto you any gospel other than that we

preached unto you [let him be anathema]; yea, though we or an

angel from heaven so preach, let him be anathema.

 

279. A concessive clause is commonly introduced by εἰ (ἐάν)

καί or καὶ εἰ (ἐάν). But a clause introduced by εἰ or ἐάν alone

may also be in thought concessive, though the concessive

element is not emphasiz.ed in the form. Matt. 26:33 (cf. Mark

14:29); Mark 14:31 (cf. Matt. 26:35).

 

280. Εἰ  (ἐάν) καί concessive in the New Testament generally

introduces a supposition conceived of as actually fulfilled or

likely to be fulfilled. See examples under 284, 285. Yet,

in concessive as well as in conditional clauses (cf. 282),

καί may belong not to the whole clause but to the word next

after it, having an intensive force, and suggesting that the

supposition is in some sense or respect an extreme one, e.g.,

especially improbable or especially unfavorable to the fulfil-

ment of the apodosis. So probably Mark 14:29.

281. Καὶ εἰ (ἐάν) concessive occurs somewhat rarely in the

New Testament. See Matt. 26:35; John 8:16; 1 Cor. 8:5;

Gal. 1:8; 1 Pet. 3:1 (but cf. WH). The force of the καί is

apparently intensive, representing the supposition as actually

or from a rhetorical point of view an extreme case, improbable

in itself, or specially unfavorable to the fulfilment of the

apodosis.

 

REM. Paley, Greek Particles, p. 31, thus distinguishes the force of εἰ

καί and καὶ εἰ, "generally with this difference, that εἰ καί implies an ad-

mitted fact' even though,'  καὶ εἰ a somewhat improbable supposition;

'even if.'" See other statements and references in Th. εἰ III.7; and

especially J. 861. It should be observed that a concessive supposition

may be probable or improbable; it is not this or that that makes it con-

cessive, but the fact that its fulfilment is unf.avorable to the fulfilment of

the apodosis.

 


114                                          THE MOODS.

 

282. Carefully to be disting1lished from the cases of καὶ εἰ (ἐάν) and εἰ

(ἐάν) καί concessive are those in which εἰ (ἐάν) is conditional and καί means

and (Matt. 11:14; Luke 6:32, 33, 34; John 8:55, etc.), or also (Luke

11:18; 2 Cor. 11:15), or is simply intensive, emphasizing the following

word and suggesting a supposition in some sense extreme (1 Cor. 4:7; 7:

11). Such a supposition is not necessarily unfavorable to the fulfilment of

the apodosis, and hence may be conditional however extreme. Cf. 280.

 

283. Moods and Tenses in Concessive Clauses. In

their use of moods and tenses concessive clauses follow in

general the rules for conditional clauses. The variety of

usage is in the New Testament, however, much less in the

case of concessive clauses than of conditional clauses.

 

284. Concessive clauses of the class corresponding to the

first class of conditional sentences are most frequent in the

New Testament. The event referred to in the concessive

clause is in general not contingent, but conceived of as actual.

2 Cor. 7:8;  ὅτι εἰ καὶ ἐλύπησα ὑμᾶς ἐν τῇ ἐπιστολῇ, οὐ μεταμέλομαι,

for, though [made you sorry with my epistle, I do not regret it. See

also Luke 18:4; 2 Cor. 4:16; 7:12; 11:6; 12:11; Phil. 2:17;

Col. 2:5; Heb. 6:9.

 

285. Concessive clauses referring to the future occur in

two forms.

(a) They take εἰ καί or εἰ, and a Future Indicative referring

to what is regarded as certain or likely to occur. In logical

force this construction is closely akin to that discussed

under 246.

Luke 11:8; εἰ καὶ ὀυ δώσει αὐτῷ ἀναστὰς διὰ τὸ εἶναι φίλον αὐτοῦ, διά
           
γε τὴν ἀναιδίαν αὐτου ἐγερθεὶς δώσει αὐτῷ ὅσων χρῄζει, though

he will not rise and give him because he is his friend, yet because of his

importunity he will arise and give him as many as he needeth. See

also Matt. 26:33; Mark 14:29.

 


MOODS IN CONCESSIVE SENTENCES.                    115

 

(b) They take ἐὰν καί, καὶ ἐάν, or ἐάν, with the Subjunctive

referring to a future possibility, or what is rhetorically con-

ceived to be possible. Kat eav introduces an extreme case,

usually one which is represented as highly improbable.

Gal. 6:1: ἐὰν καὶ προλημφθῇ ἄνθρωπος ἔν τινι παραπτώματι, ὑμεῖς οἱ
           
πνευματικοὶ καταρτίζετε τὸν τοιοῦτον ἐν πενύματι πραΰτητος,

even if a man be overtaken in any tre.'rpass, ye which are spiritual, restore

such a one in a spirit of meekness.

Gal. 1:8; ἀλλὰ καὶ ἐὰν ἡμεῖς ἄγγελος ἐξ οὺρανοῦ εὐαγγελίσηται
            [
ὑμῖν ] παρ' εὐγγελισάμεθα ἡμῖν, ἀνάθεμα ἔστω, but even if we,

or an angel from heaven, preach unto you any gospel other than that

which we preached unto you, let him be anathema. See also Luke

22:67, 68; John 8:16; 10:38; Rom. 9:27.

 

REM. The apodosis after a concessive protasis referring to the future,

sometimes has a Present Indicative, affirming what is true and will still

be true though the supposition of the protasis be fulfilled. See John

8:14; 1 Cor. 9:16. Cf. 263.

 

286. The New Testament furnishes no clear instance of a concessive

clause corresponding to the fourth class of conditional clauses. In 1 Pet.

3:14, εἰ καὶ πάσχοιτε διὰ δικαιοσύνην, μακάριοι, the use of καί before πάσχοιτε suggests that the writer has in mind that suffering is apparently opposed
to blessedness. Yet it is probable that he intends to affirm that blessed-

ness comes, not in spite of, but through, suffering for righteousness' sake.

(On the thought cf. Matt. 5:10 f.) Thus the protasis suggests, even

intentionally, a concession, but is, strictly speaking, a true causal con-

ditional clause. Cf. 282.

 

287. The New Testament instances of concessive clauses correspond-

ing to the fifth class of conditional clauses are few, and the concessive

force is not strongly marked. See 2 Tim. 2:5 (first clause) under 260;

2 Tim. 2:13.

 

288. C'oncessive clauses in English are introduced by

though, although, and even if, occasionally by if alone. Even

if introduces an improbable supposition or one especially

unfavorable to the fulfilment of the apodosis. Though and


116                              THE MOODS.

 

although with the Indicative usually imply an admitted fact.

With the Subjunctive and Potential, with the Present Indica-

tive in the sense of a Future, and with a Past tense of the

Indicative in conditions contrary to fact, though and although

have substantially the same force as even if. Even if thus

corresponds in force very nearly to καὶ εἰ; though and although

to εἰ καί.

 

MOODS IN RELATIVE CLAUSES.

 

289. Relative Clauses are introduced by relative pronouns

and by relative adverbs of time, place, and manner.

They may be divided into two classes:

 

I. Definite Relative Clauses, i.e. clauses which refer to a

definite and actual event or fact. The antecedent may be ex-

pressed or understood. If not in itself definite, it is made so

by the definiteness of the relative clause.

 

II. Indefinite or Conditional Relative Clauses, i.e. clauses

which refer not to a definite and actual event, but to a sup-

posed event or instance, and hence imply a condition. The

antecedent may be expressed or understood; if expressed, it is

usually some indefinite or generic word.

 

290. It should be observed that the distinction between the definite

and the indefinite relative clause cannot be drawn simply by reference

to the relative pronoun employed, or to the word which stands as the

antecedent of the relative. A definite relative clause may be introduced

by an indefinite relative pronoun or may have an indefinite pronoun as

its antecedent. On the other hand, an indefinite relative clause may

have as its antecedent a definite term, e.g., a demonstrative pronoun, and

may be introduced by the simple relative. A clause and its antecedent

are made .definite by the reference of the clause to a definite and actual

event; they are made indefinite by the reference of the clause to a sup-

posed event or instance. Thus if one say, He received whatever profit

was made, meaning, In a certain transaction, or in certain transactions,

profit was made, and he received it, the relative clause is definite, because


MOODS IN RELATIVE CLAUSES                    117

 

it refers to an actual event or series of events. But if one use the same

words meaning, If any profit was made, he received it, the relative clause

is indefinite, because it implies a condition, referring to an event--the

making of profit-which is only supposed. In John 1: 12, but as many

as received him, to them gave he the right to become children of God, we

are doubtless to understand the relative clause as definite, not because

of the expressed antecedent, them, but because the clause refers to a

certain class who actually received him. In Rom. 8: 24, on the other

hand, who hopeth for that which he seeth? the relative clause apparently

does not refer to a definite thing seen and an actual act of seeing, but

is equivalent to a conditional clause, if he seeth anything. In Mark 3:11,

whensoever they beheld him, they fell down before him, the form of the

Greek sentence shows that the meaning is, If at any time they sa'to him,

they were wont to fall down before him. That is, while the class of events

is actual, the relative clause presents the successive instances distribu-

tively as suppositions. These examples serve to show how slight may

be the difference at times between a definite and an indefinite relative

clause, and that it must often be a matter of choice for the writer whether

he will refer to an event as actual, or present it as a supposition.

 

291. Relative clauses denoting purpose, and relative clauses

introduced by e!wj and other words meaning until, show special

peculiarities of usage and require separate discussion. For

purposes of treatment therefore we must recognize four classes

of relative clauses.

 

I. Definite relative clauses, excluding those which express

purpose, and those introduced by words meaning until.

II. Indefinite or Conditional relative clauses, excluding

those which express purpose, and those introduced by words

meaning until.

III. Relative clauses expressing purpose.

IV. Relative clauses introduced by words meaning until.

 

I. DEFINITE RELATIVE CLAUSES.

 

292. Under the head of definite relative clauses are included

not only adjective clauses introduced by relative pronouns, ὅς,


118                                          THE MOODS.

 

ὅστις, οἷος, ὅσος, but all clauses of time, place, manner, and com-

parison, such clauses being introduced by relative words, either

pronouns, or adverbs, ὅτε ὡς (expressing either time or man-

ner), ὅπου, ὥσπερ, etc.

 

293. Moods in Definite Relative Clauses. Definite

relative clauses in general (excluding III. and IV. above)

show no special uses of mood and tense, but employ the

verb as it is used in principal cIa uses. EA. 909; G. 1427.

John 6:63; τὰ ῥήματα ἐγὼ λελάληκα ὑμῖν πνεῦμά ἐστιν καὶ ζωή
           
ἐστιν, the words that I have spoken unto you are spirit, and are life.

John 12:36; ὡς τὸ φπως ἔχετε, πιστεύετε εἰς τὸ φῶς, while ye have the

light, believe on the light.

Gal. 4.4. ὅτε δὲ ἦλθεν τὸ πλήρωμα τοῦ χρόνου, ἐξαπέστειλεν θεὸς
           
τὸν υἱὸν αὐτοῦ, but when the fulness of the time came, God sent forth

his son.

Jas. 2:26;  ὥσπερ τὸ σῶμα χωρὶς πνεύματος νεκρόν ἐστιν, οὕτως καὶ
           
πίστις χωρὶς ἔργων νεκρά ἐστιν, as the body apart from the spirit

is dead, even so faith apart from 'lvorks is dead.

Rev. 3:11; κράτει ἔχεις, hold fast that which thou hast.

Rev. 21:16; καὶ τὸ μῆκος αὐτῆς ὅσον τὸ πλάτος, and the length thereof

is as great as the breadth. Cf. Heb. 10:25. See also Matt. 26:19;

Col. 2:6.

 

294. A definite relative clause may imply a relation of

cause, result, or concession, without affecting the mood or tense

of the verb. HA. 910; G. 1445.

 

Rom. 6:2. Οἵτινες ἀπεθάνομεν τῇ ἁμαρτίᾳ, πῶς ἔτι ζήσομεν ἐν αὐτῇ,

we who died to sin, how shall we any longer live therein?

Jas. 4:13, 14; ἄγε νῦν οἱ λέγοντες Σήμερον αὐριον πορευσόμεθα
           
εἰς τὴνδε τὴν πόλιν καὶ ποιήσομεν ἐκεῖ ἐνιαυτὸν καὶ ἐμπορευσόμεθα
           
καὶ κερδήσομεν· οἵτινες οὐκ ἐπίστασθε τῆς αὔριον ποία ζωὴ ὑμῶν,

go to now, ye that say, To-day or to-morrow we will go into this city,

and spend a year there and trade and get gain " tvhereas [i.e.

although] ye know not of what sort your life will be on the morrow.


MOODS IN RELATIVE CLAUSES.                   119

 

295. All relative clauses whether adjective or adverbial may

be distinguished as either restrictive or explanatory. A re-

strictive clause defines its antecedent, indicating what person,

thing, place, or manner is signified. An explanatory clause

adds a description to what is already known or sufficiently

defined. The former identifies, the latter describes.

 

Restrictive clauses: John 15:20; μνημονεύετε τοῦ λόγου οὗ ἐγὼ εἶπον
           
ὑμῖν,, remember the word that I said unto you.

Matt. 28:6; δεύτε ἴδετε τὸν τόπον ὅπου ἔκειτο, come, see the place where

            he lay.

Mark 2:20;  ἐλέσονται δὲ ἡμέραι ὅταν ἀπαρθῇ ἀπ' αὐτῶν νυμφίος,

but days will come when the bridegroom shall be taken away from them.

Explanatory clauses: Luke 4:16; καὶ ἦλθεν εἰς Ναζαρά, οὗ ἦν τεθραμ-
           
μένος, and he came to Nazareth, where he had been brought up.

Eph. 6:17; τὴν μάχαιραν τοῦ πνεύματος, ἐστιν ῥῆμα θεοῦ, the sword

of the Spirit, which is the word of God.

 

 

II. CONDITIONAL RELATIVE SENTENCES.

 

296. An indefinite relative clause, since it refers to a sup-

posed event or instance, implies a condition, and is therefore

called a conditional relative clause. HA. 912; G. 1428.

 

Mark 10: 43;  ὃς ἂν θέλῃ μέγας γενέσθαι ἐν ὑμῖν, ἔσται ὑμῶν διάκονος,

            whosoever would become great among you, shall be your minister. Cf.

Mark 9:35; εἴ τις θέλει πρῶτος εἶναι ἔσται πάντων ἔσχατος καὶ
            πάντων διάκονος.  
It is evident that the relative clause in the former

passage is as really conditional as the conditional clause in the

latter.

 

297. Since a conditional relative clause implies a supposi-

tion, conditional relative sentences may be classified according

to the nature of the implied supposition, as other conditional

sentences are classified according to the expressed sup-

position.


120                                          THE MOODS.

 

298. The implied supposition may be particular or general.

When the relative clause refers to a particular supposed event

or instance, and the principal clause conditions its assertion on

the occurrence of this event, the implied supposition is partic-

ular. When the relative clause refers to any occurrence of an

act of a certain class, and the principal clause states what is

or was wont to take place in any instance of an act of the

class supposed, the implied supposition is general.

 

Thus in the sentence, The act which he believes to be wrong he will not

do, if reference is had to a particular occasion, or to one made particular

in thought, so that the sentence means, If on that occasion, or a certain

occasion, he believes an act to be wrong, he will not do it, the implied

supposition is particular. But in the sentence, Whatever act he [in any

instance] believes to be wrong, he does not [is not wont to] do, the implied

supposition is general. Cf. 239.

 

299. The distinction between the relative clause implying a particular

I supposition and the relative clause implying a general supposition is not

marked either in Greek or in English by any uniform difference in the

pronouns employed either in the relative clause or in the antecedent

clause. The terms particular and general apply not to the relative or its

antecedent, but to the implied supposition. Thus if one say, He received

whatever profit was made, meaning, If [in a certain transaction] any

1 profit was made, he received it, the relative clause implies a particular

! coudition. But if one use the same words, meaning, If [in any transac-

.tion] any profit toas made, [it was wont to happen that] he received it, the

implied condition is general. So also in John 1:33, upon whomsoever

thou shalt see the Spirit descending, and abiding upon him, the same is

he that baptizeth with the Holy Spirit, we have not a general principle

applying to anyone of many cases, but a supposition and an assertion

referring to a particular case. But in 1 John 3:22, whatsoever we ask, we

receive of him, the supposition refers to any instance of asking, and is

general.

Whether the implied supposition is particular or general can usually

be most clearly discerned from the nature of the principal clause. If

Ithis states what is true in a particular case, or expresses a command with

reference to a particular case, the implied supposition is particular. If

it states a general principle, or expresses a general injunction which


MOODS IN RELATIVE CLAUSES.                   121

 

applies in any instance of the event described in the relative clause, the

implied supposition is usually general. Cf. 240.

 

300. Of the six classes of conditional relative sentences

found in classical Greek, but four occur in the New Testament,

and these with considerable deviation from classical usage.

They are designated here according to the kind of condition

implied in the relative clause.

 

301. A. Simple Present or Past Particular suppo-

sition. The relative clause states a particular supposition

which refers to the present or past. It has a present or past

tense of the IndicatIve. The princIpal clause may have

any form of the verb. EA. 914, A; G. 1430.

 

Rom. 2:12; ὅσοι γὰρ ἀνόμως ἥμαρτον, ἀνόμως καὶ ἀπολοῦνται· καὶ
           
οσοι ἐν νόμῳ ἥμαρτον, διὰ νόμου κριθήσονται, for as many as have

sinned without law shall also perish without law: and as many as have

sinned under law shall be J"udged by law.

Phil 4:8;  τὸ λοιπόν, ἀδελφοί, ὅσα ἐστὶν ἀληθῆ, ὅσα σεμνά, ὅσα δίκαια,
           
ὅσα ἁγνά, ὅσα προσφιλῆ, ὅσα εὔφημα, εἴ τις ἀρετὴ καὶ εἴ τις ἔπαινος
           
ταῦτα λογίζεσθε. See also 2 Cor. 2:10.

 

REM. Respecting the use of the negatives μή and οὐ in relative clauses

of this class, see 469, 470.

 

302. B. Supposition contrary to Fact. The rela-

tive clause states a supposition which refers to the present

or past implying that it is not, or was not, fulfilled. It

has a past tense of the IndicatIve. The principal clause has

a past tense of the Indicative with avo HA. 915; G. 1433.

No instance occurs in the New Testament.

 

303. C. Future Supposition with More Probability.

The relative clause states a supposition which refers to the


122                                          THE MOODS.

 

future, suggesting some probability of its fulfilment. It

has the Subjunctive with llv. The principal clause may

have any form referring to future time. B.A. 916; G.1434.

Matt. 5:19; ὃς δ' ἂν ποιήσῃ καὶ διδάξῃ, οὗτος μέγας κληθήσεται ἐν τῇ
           
βασιλέίᾳ τῶν οὐρανῶν, but whosoever shall do and teach them, he shall

be called great in the kingdom of heaven.

Mark 13:11; καὶ ὅταν ἄγωσιν ὑμᾶς παραδιδόντες, μὴ προμεριμνᾶτε τί
           
λαλήσητε, ἀλλ' ἐὰν δοθῇ ὑμῖν ἐν ἐκείῃ τῇ ὥρα τοῦτο λαλεπιτε, οὐ
           
γάρ ἐστε ὑμεῖς οἱ λαλοῦντες ἀλλὰ τὸ πνεπυμα τὸ ἅγιον, and when

they lead you to judgment, and deliver you up, be not anxious before-

hand what ye shall speak: but whatsoever shall be given you in that

hour, that .peak ye: for it is not ye that speak, but the Holy Ghost.

See also I.iuke 13:25; Rev. 11:7; instances are very frequent in

the New Testament.

 

304. In the New Testament ἐάν not infrequently stands in a condi-

tional relative clause instead of the simple ἄν. Matt. 7 : 12; Mark 3 : 28 ;

Luke 9: 57; Acts 2: 21, et al. See WE: II. App. p. 173.

 

305. The Subjunctive with av in a relative clause is in the

New Testament usually retained in indirect discourse, or in a

sentence having the effect of indirect discourse, even after a

past tense. Matt. 14: 7; Rev. 12: 4. Of. 251. On Acts 25: 16

see 333, 344, Rem. 1.

 

306. In addition to the relative clause having the Subjunc-

tive with ctv (303), which is the regular form both in classical

and New Testament Greek, the following forms of the relative

clause also require mention as occurring in the New Testa-

ment to express a future supposition with more probability:

 

307. (a) The Subjunctive without ctv. This is very unusual

in classical Greek in relative clauses referring to the future.

In the New Testament also it is rare. J as. 2: 10 probably

belongs here; Matt. 10: 33 also, if (with Treg. and WH. text)

we read  ὅστις δὲ ἀρνήσηταί με . . . ἀρνήσομαι.


MOODS IN RELATIVE CLAUSES.                   123

 

308. (b) The Future Indicative with or without av.

Matt, 5:41;  ὅστις σε ἀγγαρεύσει μίλιον ἓν, ὓπαγε μετ' αὐτοῦ δύο, who

soever shall compel thee to go one mile, go with him two. See also

Matt. 10:32 (cf. v. 33); .18:4 (cf. v. 5); 23:12; Mark 8:35;

Luke 12:8,10; 17:31; Acts 7:7; Rev. 4:9. Cf. WH. II. App.

p. 172.

 

309. (c) The Present Indicative with or without ἄν
.

Mark 11:25; "ταν στήκετε προσευχόμενοι, ἀφίετε, whensoever ye stand

praying, forgive. See also Matt. 5: 39; Luke 12: 34; John 12: 26;

14:3.

 

310. There is no distinction in form either in Greek or in English

between a relative clause implying a particular supposition, and a relative

clause implying a general supposition, when the supposition refers to the

future. The difference in thought is the same as that which distinguishes

particular and general suppositions referring to the present or past. Cf.

298, 299. In Matt. 26:48, whomsoever I shall kiss, that is he, the sup-

position is particular, referring to a specific occasion and event. So also

in 1 Cor. 16:3. But in Luke 9:4, into whatsoever house ye enter, there

abide, and thence depart, the supposition is general, referring to anyone

of a class of acts. A large part of the conditional relative clauses referring

to the future found in the New Testament are apparently general. See,

e.g., Matt. 5:19; 10:14; 16:25; Mark 11:23; Luke 8:18, etc. Yet

in many cases it is possible to suppose that a particular imagined instance

was before the mind of the writer as an illustration of the general class

of cases.

 

311. D. Future Supposition with Less Probability.

The relative clause states a supposition which refers to the

future, suggesting less probability of its fulfilment than is

implied by the Subjunctive with c1:v. It has the Optative

witlwut ἄν. The principal clause has the Optative with

ἄν.  HA. 917; G. 1436.

No instance occurs in the New Testament.

 

312. E. Present General Supposition. The relative

clause refers to any occurrence of a class of acts in the


124                                          THE MOODS.

 

general present, and the principal clause states what is

wont to take place in any instance of the act referred to

in the relative clause. The relative clause has the Sub-

junctive with ἄν, the principal clause the Present In-

dicative. HA. 914, B. (1); G. 1431, 1.

 

1Cor. 11:26; ὁσάκις γὰρ ἐὰν ἐσθίητε τὸν ἄρτον τοῦτον καὶ τὸ ποτήριον
           
πίνητε, τὸν θάνατον τοῦ κυρίου καταγγέλλετε, ἄχρι οὗ ἔλθῃ, for as

often as ye eat this bread, and drink the cup, ye proclaim the Lord's

death, till he come. See also Matt. 15 : 2j Mark 10 : 11; Rev. 9 : 5.

 

REM. Concerning the use of ἐάν for ἄν, see 304.

 

313. The Present Indicative not infrequently occurs in con-

ditional relative clauses which apparently imply a present

general supposition. G.MT. 534. Yet in most such passages

in the New Testament, it is possible that a particular imagined

instance in the present or future is before the mind as an

illustration of the general class of cases. Of. 301, 309. It is

scarcely possible to decide in each case whether the supposi-

tion is particular or general. The difference of meaning is in

any case slight.

 

Luke 14:27;  ὅστις οὐ βαστάζει τὸν σταυρόν ἑαυτοῦ καὶ ἔρχεται ὀπίσω
           
μου, οὐ δύναται εἶναί μου μαθητήςwhosoever doth not bear his own

cross and come after me, cannot be my disciple. See also Matt. 10:38

13:12 (cf. Luke8:18); Luke7:47; John3:8; Rom.6:16; 9:18;

1 Cor. 15:36, 37; Heb. 12:6.

 

314. Concerning the similarity of the third and fifth classes of condi-

tional relative clauses, cf. 262. The statements there made respecting

ordinary conditional sentences are applicable also to conditional relative

sentences. See Mark 8:28, 29; Luke 9:24, 48; 1 John 8:22.

 

315. F. Past General Supposition. The relative clause

refers to any occurrence of a certam act or class of acts,

and the principal clause states what was wont to take


MOODS IN RELATIVE CLAUSES.                   125

 

place in any instance of the act referred to in the relative

clause. In classical Greek, the relative clause has the Op-

tative without ἄν, the principal clause the Imperfect Indic-

ative. HA. 914, B. (2); G. 1431, 2.

 

In the New Testament, the Optative does not occur

in such clauses, the Imperfect or Aorist Indicative with

ἄν being used instead. Cf. 26.

 

Mark 3:11;  καὶ τὰ πνεύματα τὰ ἀκάθαρτα, ὅταν αὐτὸν ἐθεώρουν, προσ-
           
έπιπτον αὐτῷ καὶ ἔκραζον, and the unclean spirits, whensoever they

beheld him, were wont to fall down before him and cry out. See also

Mark 6:56; 11:19; Acts 2:45; 4:35; 1 Cor. 12:2; ct. Gen.

2:19; 1 Sam. 2:13, 14.

 

316. In the New Testament, relative clauses conditional in

form are sometimes definite in force.

Mark 2:20; ἐλεύσονται δὲ ἡμέραι ὅταν ἀπαρθῇ ἀπ' αὐτῶν νυμφίος,
             but days will come when the bridegroom shall be taken away from them. 
            See also Luke 5:35; 13:28; Rev. 8:1.

 

III. RELATIVE CLAUSES EXPRESSING PURPOSE.

 

317. Relative Clauses of Pure Purpose. Relative

clauses expressing purpose take the Future Indicative

both in classical and New Testament Greek. HA. 911

G. 1442; B. p. 229; WM. p. 386, f. n.

 

Matt. 21:41; τὸν ἀμπελῶνα ἐκδώσεται ἄλλοις γεωργοῖς, οἵτινες ἀπο-
           
δώσουσιν αὐτῷ τοὺς καρπούς, he will let out the vineyard unto other

husbandmen, which shall render him the fruits. See also Acts 6:3.

 

318. Complementary relative clauses expressing that for

which a person or thing is fitted, or other similar relation, take

the Subjunctive or the Future Indicative both in classical and

New Testament Greek. G.MT. 572; Hale in T.A.P.A. 1893,

pp. 156 ff.

 


126                                          THE MOODS.

 

Heb. 8:3; ὅθεν ἀναγκαῖον ἔχειν τι καὶ τοῦτον προσενέγκῃ, wherefore

it is necessary that this high priest also have somewhat to offer. See also

Mark 14:14; Luke 11:6; 22:11. In Luke 7:4 a complementary

relative clause limiting the adjective ἄξιος has the Future Indicative.

 

319. The clauses referred to in 318 are to be distinguished from true

relative clauses of purpose in that they do not express the purpose with

which the action denoted by the principal clause is done, but constitute a

complementary limitation of the principal clause. Of. the clause with

i!na (215-217) and the Infinitive (368) expressing a similar relation.

The Subjunctive in such clauses is probably in origin a Deliberative

            Subjunctive. Thus in Mark 14:14, ποῦ ἐστὶν τὸ καάλυμά μου ὅπου τὸ
           
πάσχα μετὰ τῶν μαθητῶν μου φάγω, the relative clause ὅπυ . . . φάγω
            reproduces in dependent construction the thought of the deliberative

question ποῦ . . . φάγω. The same explanation doubtless applies, though

less obviously, to the Subjunctive in Acts 21:16, and to the Future in

Luke 7:4. In both instances the thought of a deliberative question is

reproduced in the relative clause. Of. the clauses similar in force, but

employing an interrogative pronoun, 346. See Tarbell in Cl. Rev. July

1891, p. 302 (contra, Earle in OZ. Re'/). March 1892, pp. 93-95); Hale

in T.A.P.A., 1893.

 

320. The Optative sometimes occurs after a past tense in these delib-

erative relative clauses in classical Greek. There are, however, no New

Testament instances of the Optative so used.

 

 IV. RELATIVE CLAUSES INTRODUCED BY WORDS MEANING

UNTIL, W RILE, AND BEFORE.

 

321.   Ἕως is properly a relative adverb which marks one

action as the temporal limit of another action. It does this

in two ways, either (a) so that the beginning or simple occur-

rence of the action of the verb introduced by twr; is the limit

of the action denoted by the principal verb, or (b) so that the

continuance of the former is the limit of the latter. In the

former case twr; means until, in the latter, while, as long as.

On the classical use of twr; and similar words, see HA. 920-

924; G. 1463-1474; G.MT. 611-661; Gild. in A..J.P. IV.


MOODS IN RELATIVE CLAUSES.                               127

 

416-418. On  ἕως in Hellenistic Greek see G. W; Gilmore in

J.B.L., 1890, pp. 153-160.

 

322. Clauses Introduced by ἕως and referring to the

Future. When the clause introduced by ἕως depends on

a verb of future time, and refers to a future contingency, it

takes the Subjunctive with liv both in classical and New

Testament Greek.

Mark 6:10; ἐκεῖ μένετε ἕως ἂν ἐξέλθητε ἐκεῖθεν, there abide till ye

depart thence. See also Matt. 5: 18; 12: 20; Luke 9: 27; 1 Cor. 4: 5.

 

323. In classical Greek, especially in tragic poetry, the

Subjunctive without av sometimes occurs with talf; after a

verb of present or future time. G.MT. 620. In the New

Testament this construction is frequent.

Luke 15:4;  καὶ πορεύεται ἐπὶ τὸ ἀπολωλὸς ἕως εὕρῃ αὐτό, after that which

            is lost, until he find it. See also Matt. 10 : 23; Luke

12:59; 22:34.

 

324. Clauses Introduced by e!wj and referring to

what was in Past Time a Future Contingency. When

the clause introduced by ἕως depends on a verb of past

time and refers to what was at the time of the principal

verb conceived of as a future contingency, it takes the

Optative without ἅν in classical Greek. In the New Tes-

tament it takes the Subjunctive without ἄν.

 

Matt. 18:30; ἔβαλεν αὐτὸν εἰς φυλακὴν ἕως ἀποδῷ τὸ ὀφειλόμενον,

he cast him into prison till he should pay that which was due.

 

325. The Subjunctive after ἕως in the New Testament is always an

Aorist, the action denoted being conceived of as a simple event, and ἕως

meaning properly until. Thus the accurate translation of Mark 14:32

(Matt. 26:36 is similar), καθίσατε ὧδε ἕως προσεύξωμαι, is, Sit ye here till


128                                          THE MOODS.

 

I pray, or have prayed (cf. 98). While I pray CR. Y.) is slightly para-

phrastic. Cf. Luke 17:8.

 

326. Clauses Introduced by ἕως (until) and referring

to a Past Fact. When ἕως means until and the clause

introduced by it refers to an actual past occurrence, the

verb of this clause is in a past tense of the Indicative, as in

an ordinary relative clause referring to past time.

Matt. 2:9; ἀστήρ . . . προῆγεν αὐτούς, ἕως ἐλθὼν ἐστάθη ἐπάνω οὗ
           
ἦν τὸ παιδίον, star. . . went before them, till it came and stood

over where the young child was.

 

327. Clauses Introduced by ἕως (while) and referring

to a Contemporaneous Event. When ἕως means while

and the clause introduced by it refers to an event contem-

poraneous with that of the principal verb, it has the con-

struction of an ordinary relative clause. Of. 293.

John 9:4;  ἡμᾶς δεῖ ἐργάζεσθαι τὰ ἔργα τοῦ πέμψαντος με ἕως ἡμέρα
           
ἐστίν, we must work the works of him that sent me, while it is day.

 

328. In John 21: 22, 23; 1 Tim. 4:13, the exact meaning of  ἑως

ἔρχομαι at is probably while I am coming, the coming being conceived of as

in progress from the time of speaking. Cf. Luke 9:13. In 1 Cor. 4:5

on the other hand it is thought of as a future event. In Mark 6:45  ἕως

ἀπολύει represents ἕως ἀπολύω of the direct form (cf. 347), the original

sentence meaning, go before me while I am sending away, etc.

 

329. When the ἕως clause refers to the future or to what was at the

time of the principal verb the future (322-326), it frequently has the

force of a conditional relative clause. See Matt. 18:30; Luke 15:4.

When it refers to an actual event (327, 328), it is an ordinary temporal

clause (293), requiring special mention here only to distinguish these

usages from those described above.

 

330. In the New Testament ἕως is sometimes followed by οὗ or ὅτου.

Ἕως  is then a preposition governing the genitive of the relative pronoun,


MOODS IN RELATIVE CLAUSES.                               129

 

but the phrase ἕως οὖ or ἕως ὅτου is in effect a compound conjunction hav-

ing the same force as the simple ἕως. The construction following it is

also the same, except that dv never occurs after ἕως οὖ or ἕως ὅτου. See

Matt. 5:25; 13:33; John 9:18; Acts 23:12.

 

331. Clauses introduced by ἄχρι, ἄχρι ο+θ, ἄχρι ἧς ἡμέρας, μέχρι
            and
μέχρις οὗ have in general the same construction and force

as clauses introduced by ἕως, ἕως οὗ, and ἕως ὅτου.

Mark 13:30 οὐ μὴ παρέλθῃ γενεὰ αὕτη μέχρις οὗ παῦτα πάντα
           
γένηται.

Acts 7:18;  ηὔξησεν λαὸς καὶ ἐπληθύνθη ἐν Αἰγύπτῳ, ἄχρι οὗ
           
ἀνέστη βασιλεὺς ἕτερος ἀπ' Αἴγυπτον.  See also Rev. 15:8; 20:3;

Luke 17:27; Acts 27:33.

Rev. 7:3;  μὴ ἀδικήσητε τὴν γῆν . . . ἄχρι σφραγίσωμεν τοὺς δούλους
           
τοῦ θεοῦ.

 

332. Gal. 3:19 [WH. text] furnishes one instance of ἄχρις ἄν with a

word meaning until after a verb of past time [WH. margin, Tisch., and

Treg. read ἄχρις οὗ]; cf.324. Rev. 2:25 contains the combination ἄχρι

οὗ ἄν with the Future Indicative; cf. 330. Rev, 17:17 contains a Future

Indicative with ἄχρι after a past tense.

 

333. Clauses introduced by πρίν and employing a finite

mood have in general the same construction as clauses intro-

duced by ἕως.

The New Testament, however, contains but two instances of a finite

verb after πρίν, Luke 2:26; Acts 25:16. In both cases the clause is in

indirect discourse, and expresses what was from the point of view of the

original statement a future contingency. In Luke 2:26 the Subjunctive

with ἄν is retained from the direct discourse. In Acts 25:16 the Optative

represents a Subjunctive with or without dv of the direct discourse.

Cf. 341-344.

 

REM. 1. The employment of a finite mood rather than an Infinitive

in these instances is in accordance with classical usage. Cf. 382, and

G. 1470.

 

REM. 2. In Acts 25:16 occurs after πρίν, and in Luke 2:26 it

appears as a strongly attested variant reading. Attic writers used the

simple πρίν with the finite moods. Cf. 381.


130                                          THE MOODS.

 

MOODS IN INDIRECT DISCOURSE.

 

334. When words once uttered or thought are afterward

quoted, the quotation may be either direct or indirect. In a

direct quotation the original statement is repeated without

incorporation into the structure of the sentence in the midst of

which it now stands. In an indirect quotation the original

sentence is incorporated into a new sentence as a subordinate

element dependent upon a verb of saying, thinking, or the like,

if and suffers such modification as this incorporation requires.

The following example will illustrate:

 

Original sentence (direct discourse), I will come.

Direct quotation, He said, "I will come."

Indirect quotation, He said that he would come.

 

REM. The distinction between direct discourse and indirect is not one

of the exactness of the quotation. Direct quotation may be inexact. In-

direct quotation may be exact. Suppose, for example, that the original

statement was, There are good reasons why I should act thus. If one say,

He said, "I have good reasons for acting thus," the quotation is direct

but inexact. If one say, He said that there 'Were good reasons 'Why he

should act thus, the quotation is exact though indirect.

 

335. Direct quotation manifestly requires no special discus-

sion, since the original statement is simply transferred to the

If new sentence without incorporation into its structure.

 

336. Indirect quotation, on the other hand, involving a re-

adjustment of the original sentence to a new point of view,

calls for a determination of the principles on which this re-

adjustment is made. Its problem is most simply stated in

the form of the question, What change does the original form

of a sentence undergo when incorporated into a new sentence

as an indirect quotation? All consideration of the principles


MOODS IN INDIRECT DISCOURSE.                           131

 

of indirect discourse must take as its starting point the origi-

nal form of the words quoted.

            For the student of Greek that expresses his own thought in

another language, it will also be necessary to compare the

idiom of the two languages. See 351 ff.

 

337. The term indirect discourse is commonly applied only

to indirect assertions and indirect questions. Commands,

promises, and hopes indirectly quoted might without impro-

priety be included under the term, but are, in general, ex-

cluded because of the difficulty of drawing the line between

them and certain similar usages, in which, however, no direct

form can be thought of. Thus the Infinitive after a verb of

commanding might be considered the representative in indi-

rect discourse of an Imperative in the direct discourse; some-

what less probably the Infinitive after a verb of wishing might

be supposed to represent an Optative of the direct; while for

the Infinitive after verbs of striving, which in itself can

scarcely be regarded as of different force from those after

verbs of commanding and wishing, no direct form can be

thought of.

            338. Concerning commands indirectly quoted, see 204. Con-

cerning the Infinitive after verbs of promising, see 391.

            339. Indirect assertions in Greek take three forms:

            (a) A clause introduced by ὅτι or ὡς. In the New Testa-

ment, however, ὡς not so used.

            (b) An Infinitive with its subject expressed or understood.

See 390.

            (c) A Participle agreeing with the object of a verb of per-

ceiving, and the like. See 4.60.


132                                          THE MOODS.

 

            340. Indirect Questions are introduced by εἰ or other inter-

rogative word; the verb is in a finite mood. HA. 930; G.

1605.

            341. Classical Usage in Indirect Discourse. In indi-

rect assertions after ὅτι and in indirect questions, classical

usage is as follows:

            (a) When the leading verb on which the quotation de-

pends denotes present or future time, the mood and tense

of the direct discourse are retained in the indirect.

            (b) When the leading verb on which the quotation de-

pends denotes past time, the mood and tense of the direct

discourse may be retained in the indirect, or the tense may

be retained and an Indicative or Subjunctive of the direct

discourse may be changed to an Optative. EA. 932; G.

1497.

            342. The above rule applies to all indirect quotations in

which the quotation is expressed by a finite verb, and includes

indirect quotations of simple sentences and both principal and

subordinate clauses of complex sentences indirectly quoted.

            The classical grammars enumerate certain constrnctions in which an

Indicative of the original sentence is uniformly retained in the indirect

discourse. These cases do not, however, require treatment here, the gen-

eral rule being sufficient as a basis for the consideration of New Testament

usage.

            343. New Testament Usage in Indirect Discourse. In

indirect assertions after ὅτι and in indirect questions, New

Testament usage is in general the same as classical usage.

Such peculiarities as exist pertain chiefly to the relative

frequency of different usages. See 344-349.

 


                        MOODS IN INDIRECT DISCOURSE.                           133

 

John 11:27;  ἐγὼ πεπίστευκα ὅτι σὺ εἶ χριστος υἱὸς τοῦ θεοῦ,

            I have believed that thou art the Christ, the Son of God.

Gal. 2:14; εἶδον ὅτι οὐκ ὀρθοποδοῦσιν, I saw that thell were not walking

            uprightly.

Matt. 20:10; ἐλθόντες οἱ πρῶτοι ἐνόμισαν ὅτι πλεῖον λήμψονται,

            when the first came, thell supposed that they would receive more.

Mark 9:6; οὐ γὰρ ᾔδει τί ἀποκριθῇ, for he wist not wh,at to answer.

Luke 8:9;  ἐπηρώτων δὲ αὐτὸν οἱ μαθηταὶ αὐτοῦ τίς αὕτη εἴη παρα-
           
βολή,  and his disciples asked him what this parable was.

Luke 24:23; ἦλθαν λέγουσαι καὶ ὀπτασίαν ἀγγέλων ἑωρακέναι, οἳ
           
λέγουσιν αὐτὸν ζῇν, they came salling that they had also seen a vision

            of angels, which said that he was alive. In this example the principal

            clause of the direct discourse is expressed in the indirect discourse

            after a verb of past time by an Infinitive, while the subordinate

            clause retains the tense and mood of the original.

Acts 5:24;  διηπόρουν περὶ αὐτῶν τί ἂν γένοιτο τοῦτο, they were per-

            plexed concerning them whereunto this would grow. Bnt for ἄν in

            this sentence, it might be thought that the direct form was a

            deliberative question having the Subjunctive or Fnture Indicative.

            But in the absence of evidence that av was ever added to an

            Optative arising under the law of indirect discourse, it must be

            supposed that the indirect discourse has preserved the form of

            the direct unchanged, and that this was therefore a Potential

            Optative with protasis omitted. See also Luke 6: 11; 15: 26 ;

            Acts 10:17.

 

            344. The Optative occurs in indirect discourse much less I

frequently in the New Testament than in classical Greek. II

It is found only in Luke's writings, and there almost exclusive- I

ly in indirect questions.

 

            REM:. 1. Acts 25:16 contains the only New Testament instance of an

Optative in the indirect quotation of a declarative sentence. (But cf.

347 and 258. ) It here stands in a subordinate clause which in the direct

discourse would have had a Subjunctive with or without ἄν. If the ἄν, be

supposed to have been in the original sentence (cf. Luke 2: 26), it has

been dropped in accordance with regular usage in such cases. HA.934;

G. 1497, 2.11


134                                         THE MOODS.

 

            REM. 2. The clause μήποτε δῴη [or δώῃ] αὐτοῖς θεὸς μετάνοιαν 2 Tim. 2:25 is regarded by B. p. 256, Moulton, WM. pp. 374, 631, foot

            notes, as an indirect question. But concerning the text and the

            interpretation, see 225.

 

            345. In quoting declarative sentences the indirect form. is

comparatively infrequent in the New Testament, the direct

form either with or without ὅτι being much more frequent.

The presence of ὅτι before a quotation is in the New Testament

therefore not even presumptive evidence that the quotation is

indirect. The ὅτι is of course redundant.

 

Luke 7:48;  εἶπεν δὲ αὐτῇ Ἀφέωνταί σου αἱ ἁμαρτίαι, and he said

            unto her, Thy sins are forgiven.

John 9:9; ἐκεῖνος ἔλεγεν ὅτι Ἐγώ εἰμι, he said I am he.

 

            REM. The redundant o!ti sometimes occurs even before a direct ques-

tion. Mark 4:21, et al.

 

            346. Indirect deliberative questions are sometimes found

after ἔχω and other similar verbs which do not properly take a

question as object.  The interrogative clause in this case serves

the purpose of a relatIve clause and Its antecedent, while retain-

ing the form which shows its origin in a deliberative question.

 

Mark 6:36;  ἵνα . . .  ἀγοράσωσιν ἑαυτοῖς τί φάγωσιν, that . . . they

            may buy themselves somewhat to eat.

Luke 9:58;  δὲ υἱὸς τοῦ ἀνθρώπου ούκ ἔχει ποῦ τὴν κεφαλὴν κλίνῃ,

            but the Son of man hath not where to lay his head. See also Matt.

            8:20; Mark 8:1, 2; Luke 12:17.

 

            347. The principles of indirect discourse apply to all sub-

ordinate clauses which express indirectly the thoughts of

another or of the speaker himself, even when the construc-

tion is not strictly that of indirect discourse. HA. 937; G.

1502. See New Testament examples under 258.

 

            348. Both in classical and New Testament Greek, the Im-

perfect occasionally stands in indirect discourse after a verb of


                        MOODS IN INDIRECT DISCOURSE.                           135

 

past time as the representative of a Present of the direct dis-

course, and a Pluperfect as the representative of the Perfect.

Thus exceptional Greek usage coincides with regular English

usage. HA. 936; G. 1489.

 

John 2:25; αὐὸς γὰρ ἐγίνωσκεν τί ἦν ἐν τῷ ἀνθρώπῳ, knew what was in

            man. See also Acts 19:32.

 

            349. In classical Greek, ὅστις is used in introducing indirect

questions. HA. 1011; G. 1600. In the New Testament it is

not so employed, but there are a few passages in which it is

apparently used as an interrogative pronoun in a direct ques-

tion.

            It is so taken by Mey., B., WH., et al, in Mark 9:11, 28, and by WH.

in Mark 2:16. See B. pp. 252 f.; Th., ὅστις, 4; also (contra) WM. p.

208, f.n.; WT. p. 167.

           

            350. The simple relative pronouns and adverbs are some-

times used in indirect questions in the New Testament as in

classical Greek. HA. 1011, a; G. 1600; J. 877, Obs. 3; B.

pp. 250 f.

 

Luke 8: 47; δι' ἣν αἰτίαν ἥψατο αὐτοῦ ἀπήγγεθιλεν, she declared for

            what cause she had touched him. See also Mark 5:19, 20; Acts

            14:27; 15:14.

 

            351. INDIRECT DISCOURSE IN ENGLISH AND IN GREEK.

From what has been said above, it appears that the tense of a

verb standing in a clause of indirect discourse in Greek does

not express the same relation between the action denoted and

the time of speaking as is expressed by a verb of the same

tense standing in a principal clause; or, to speak more exactly,

does not describe it from the same point of view.  A verb in

a principal clause views its action from the point of view of

the speaker.  A verb in an indirect quotation, on the other

hand, views its action from the point of view of another person,

viz. the original author of the words quoted. It has also


136                                          THE MOODS.

 

appeared that in certain cases the mood of the Greek verb is

changed when it is indirectly quoted. Now it is evident that

in order to translate the Greek sentence containing a clause of

indirect discourse into English correctly and intelligently, we

must ascertain what English usage is in respect to the tenses

and moods of the verbs of indirect discourse; otherwise we

have no principle by which to determine what English tense

and mood properly represent a given Greek tense and mood

in indirect discourse. Furthermore, since Greek usage has

been expressed in terms of the relation between the original

utterance and the quotation, it will be expedient to state Eng-

lish usage in the same way. An example will illustrate at the

same time the necessity of formulating the law and of formu-

lating it in terms of relation to the direct form.

 

      (1) He has seen a vision.                         (2) ὀπτασίαν ἑῶρακεν

      (3) They said that he had seen a vision. (4) εἶπον ὅτι ὀπτασίαν ἑώρακεν.

 

            The sentences marked (1) and (2) express the same idea

and employ corresponding tenses. The sentences marked (3)

and (4) represent respectively the indirect quotation of (1)

and (2) after a verb of past time, and express therefore the

same meaning. They do not, however, employ corresponding

tenses, the Greek using a Perfect, the English a Pluperfect.

It is evident therefore that the principle of indirect discourse

is not the same in English as in Greek, and that we cannot

translate (4) into (3) by the same principle of equivalence of

tenses which we employ in direct assertions. To translate

(4) we must first restore (2) by the Greek law of indirect dis-

course, then translate (2) into (1), and finally by the English

law of indirect discourse construct (3) from (1) and the trans-

lation of the Greek εἶπον. This process requires the formula-

tion of the law of indirect discourse for English as well as for

Greek.

 


                        MOODS IN INDIRECT DISCOURSE.                           137

 

            352. English usage in indirect discourse is illustrated in the

following examples:

 

Direct form. . . . . . . . . . . . . . . .      I see the city.

Indirect, after present tense. . . .     He says that he sees the city.

     "          "     future      "   . . . .     He will say that he sees the city.

     "          "     past         "   . . . .      He said that he saw the city.

Direct form . . . . . . . . . . . . . . . .     I saw the city.

Indirect, after present tense. . . .     He says that he saw the city.

     "          "      future     "   . . . .     He will say that he saw the city.

     "          "      past        "    . . . .     He said that he had seen the city.

Direct form. . . . . . . . . . . . . . . . .    I shall see the city.

Indirect, after present tense . . . .    He says that he shall see the city.

       "        "     future     " . . . . . .    He will say that he shall see the city.

       "        "     past        " . . . . . .     He said that he should see the city.

Direct form. . . . . . . . . . . . . . . . .    I may see the city.

Indirect, after present tense  . . . .   He says that he may see the city.

       "        "      future    ". . . . . . .   He will say that he may see the city.

       "        "      past       ". . . . . . .    He said that he might see the city.

 

            From these examples we may deduce the following rule for

indirect discourse in English:

            (a) After verbs of present or future time, the'mood and tense

of the direct discourse are retained in the indirect discourse.

            (b) After verbs of past time, the mood of the direct dis-

course is retained, but the tense is changed to that tense which

is past relatively to the time of the direct discourse.

            Thus, see becomes saw; saw becomes had seen; shall see becomes

should see (the change of mood here is only apparent); may see

becomes might see, etc.

           

            REM. In questions and in conditional clauses a Present Indicative of

the direct form may become a Past Subjunctive in indirect quotation after

a verb of past time. See Luke 3:15; Acts 10:18; 20:17, E.V.

 

            353. Comparing this with the Greek rule, we may deduce

the following principles for the translation into English of

clauses of indirect discourse in Greek:


138                                          THE MOODS.

 

            (a) When the quotation is introduced by a verb of present

it or future time, translate the verbs of the indirect discourse

by the same forms which would be used in ordinary direct

discourse,

            (b) When the quotation is introduced by a verb of past

time, if there are Optatives which represent Indicatives or

Subjunctives of the direct discourse, first restore in thought

these Indicatives or Subjunctives, then translate each Greek

verb by that English verb which is relatively past to that

which would correctly translate the same verb standing in

direct discourse,

 

            354. The statement of English usage in indirect discourse

is presented in the form adopted above for the sake of brevity

and convenience of application, It is, however, rather a for-

mula than a statement which represents the process of thought.

In order to apprehend clearly the difference between English

and Greek usage it must be recognized that certain English

tenses have, not like the Greek tenses a two-fold function, but

a three-fold. They mark (1) the temporal relation of the point

of view from which the action is described to the time of

speaking; (2) the temporal relation of the action described to

this point of view; (3) the conception of the action as re-

spects its progress. Thus in the sentence, I had been read-

ing, (1) the point of view from which the act of reading is

viewed is past, (2) the action itself is previous to that point of

view, and (3) it is viewed as in progress. He will not go is a

Future from a present point of view presenting the action as a

simple event. In the sentence, When he came, I was reading,

I was reading would be more accurately described as a Present

progressive from a past point of view, than as a Past progres-

sive from a present point of view, In other instances the same

form might be a Past from a present point of view. These


                        MOODS IN INDffiECT DISCOURSE.                           139

 

triple-function tenses have perhaps their chief use in English

in indirect discourse, but are used also in direct discourse.

Many of them are derived by the process of composition, out

of which so many English tenses have arisen, from verb-forms

which originally had only the two-fold function, but their

existence in modern English is none the less clearly estab-

lished. Professor W. G. Hale1 in A.J.P., vol. VIII. pp. 66 ff.,

has set forth the similar three-fold function of the Latin tenses

in the Indicative Mood. But it should be noticed that the

English has developed this three-fold function more clearly

even than the Latin. For example, the antecedence of an

action to a past point of view is in Latin only implied in the

Iassertion of its completeness at that past point of time. But

in English this antecedence may be affirmed without affirming

the completeness of the act.

 

            Bearing in mind this three-fold function of certain English

tenses, the difference between Greek and English usage in in-

direct discourse may be stated comprehensively as follows:

 

            The Greek, while adopting in indirect discourse the point of

view of the person quoting as respects the person of verbs and

pronouns, and while sometimes after a verb of past time mark-

ing the dependent character of the statement by the use of the

Optative in place of an Indicative or SubjunctivE) of the origi-

nal statement, yet as respects tense, regtuarly carries over into

the indirect discourse the point of view of the original state-

ment, treating it as if it were still present. What was present

to the original speaker is still treated from his point of view,

as present; what was past, as past; what was future, as future.. .

In English, on the other hand, in quoting a past utterance,

 

            1 Professor Hale's article furnished the suggestion for the view of the

English tenses presented here.


140                                         THE MOODS.

 

the fact that it is past is not only indicated by the past tense

of the verb which introduces the quotation, but still further

by the employment of a tense in the quotation which marks the

point of view from which the act is looked at as past.  Thus

in Greek a prediction expressed originally by a Future tense,

when afterward quoted after a verb of past time, is still ex-

pressed by a Future, the act being viewed as future from the

assumed point of view, and this point of view being treated as

present or its character as past being ignored.  But in English

such a prediction is expressed by a Past-future, i.e. by the

English tense which describes an action as future from a past

point of view.  Thus in quoting ὄψομαι, I shall see, in indirect

discourse, one says in Greek, εἶπεν ὅτι ὄψεται; but in English, he

said that he should see.  Similarly, a statement made originally

by the Perfect tense, when quoted after a verb of past time,

is still expressed by a Perfect tense in Greek, but in English

by a Pluperfect.  Thus ἡμάρτηκα, I have sinnedεἶπεν ὅτι
ἡμάρτηκενhe said that he had sinned.

 

            When we pass to quotations after verbs of present time, the

usages of the two languages naturally coincide, since the differ-

ence between the point of view of the original utterance and

the quotation, which in English gave rise to a change of tense

not however made in Greek, disappears.  The point of view of

the original statement is in both languages retained and

treated as present, because it is present.  Thus e]leu<somai,  I shall

come, requires only a change of person in quotation after a verb

of present time, λέγει ὅτι ἐλεύσεται, he says that he shall come.

 

            It might naturally be anticipated that in quotations after

verbs of future time, where again the time of the original

statement differs from that of the quotation, there would arise

a difference of usage between English and Greek. Such how-

ever is not the case.  What the Greek does after a verb of


                        MOODS IN INDIRECT DISCOURSE.                           141

 

past time, the English as well as the Greek does after a verb

of future time, viz. treats the point of view of the original

utterance as present. Thus let us suppose the case of one

predicting what a person just now departing will say when he

returns. He has not yet seen anything, but it is imagined that

when he returns he will say, I have seen all things. The asser-

tion of this by he will say, takes the form he will say that he

has seen all things; just as in Greek one quoting ἑώρακα πάντα

after ἐπρῖ says ἐρεῖ ὅτι ἑώρακεν πάντα. Thus the person quoting

does not describe the event from his own point of view--this

would require he will see, nor does he mark the fact that the

point of view of the utterance is different from his own--this

would require he will have seen; but treats the point of view

of the person whose expected language he quotes in advance,

as if it were present. Thus while the Greek is consistent in

simply adopting the conceived point of view of the future

statement, the English departs from the principle which it fol-

lows after past tenses, and follows here the same method as

the Greek.

            355. These facts enable us to see that it would be incorrect to say

that the tense of the direct discourse is in Greek determined from the

point of view of the original speaker, in English from the point of view of

the person who makes the quotation. The correct statement is that in

both languages the act is looked at from the point of view of the original

speaker, but that the two languages differ somewhat in their method of

indicating the relation of this point of view to the time of the quotation.

This difference, however, pertains only to quotations whose point of view

is past. Its precise nature has already been stated (354). When the

point of view is present or future the usage of the two languages is

identical.

 

            356. The comparison of English and Greek usage may

be reduced to articulated statement as follows: English usage

is like Greek usage in three respects, and different in two

respects.

 


142                                          THE MOODS.

 

I. It is like Greek in that,

            (a) It adapts the person of the pronouns and verbs of the

original utterance to the point of view of the quoter.

            (b) It looks at the act described in the quotation from the

point of view of the original statement.

            (c) After a verb of present or future time this point of view

of the original utterance is treated in the quotation as present,

as after verbs of present time it is in fact.

 

II. It differs from Greek in that,

            (a) While it looks at the act from the point of view of the

original statement, if that point of view is past it designates

it as past, using a tense which describes the action from a past

point of view. A Past of the original utterance becomes in

the quotation a Past-past; a Future becomes a Past-future,

etc. This the Greek does not do, having in general no tense

which has this double temporal power.

            (b) It does not as a rule change the original mood of the

verb in quotation. Most apparent changes of mood, such as

will to would, are changes of tense. But cf. 352, Rem.

 

                        CONSTRUCTION AFTER Καὶ ἐγένετο.

 

            357. Clause or Infinitive as the Subject of ἐγένετο.

By a Hebraism  καὶ ἐγένετο and ἐγένετο δέ, Septuagint ren-

derings of וַיְהִי, are used in the New Testament (Matt.,

Mark, Luke, Acts) to introduce a clause or an Infinitive

which is logically the subject of the ἐγένετο. The ἐγένετο

is usually followed by a phrase or clause of time; the event

to be narrated is then expressed by καί with an Indicative,

or by an Indicative without καί, or by an Infinitive. It

thus results that the construction takes three forms:


                                    THE INFINITIVE.                                       143

 

            358. (a) Καὶ ἐγένετο, or ἐγένετο δέ, and the phrase of time are

followed by καί with an Indicative.

Luke 5:1;  ἐγένετο δὲ ἐν τῷ τὸν ὄχλον ἐπικεῖσθαι αὐτῷ καὶ ἀκούειν
           
τὸν λόγον τοῦ θεοῦ καὶ αὐτὸς ἦν ἑστὼς παρὰ τὴν λίμνην Γεννησα-  
           
ρέτ, now it came to pass, while the multitude pressed upon him

            and heard the word of God, that he was standing by the lake of Gennesaret.

 

            359. (b) Καὶ ἐγἐνετο, or ἐγένετο δέ, and the phrase of time are

followed by an Indicative without καί.

 

Mark 1:9 Καὶ ἐγένετο ἐν ἐκείναις ταῖς ἡμέραις ἦλθεν Ἰησοῦς ἀπὸ
           
Ναζαρὲτ τῆς Γαλιλαίας, and it came to pass in those days, that Jesus

            came from Nazareth of Galilee.

 

            360. (c) καὶ ἐγένετο, or ἐγένετο δέ, and the phrase of time are

followed by an Infinitive, the narrative being continued either

by an Infinitive or an Indicative.

 

Acts 9:32;  ἐγένετο δὲ Πέτρον διερχόμενον διὰ πάντων κατελθεῖν, and

            it came to pass, as Peter went throughout all parts, he came down.

            See also Mark 2:23; Luke 6:12. B. pp. 276-278.

 

 

                                    THE INFINITIVE.

 

            361. That the Infinitive in Greek had its origin as respects

both form and function in a verbal noun, and chiefly at least in

the dative case of such a noun, is now regarded as an assured

result of comparative grammar. At the time of the earliest

Greek literature, however, the other cases of this verbal noun

had passed out of use, and the dative function of the form that

remained had become so far obscured that, while it still re-

tained the functions appropriate to the dative, it was also used

as an accusative and as a nominative. Beginning with Pindar

it appears with the article, at first as a subject-nominative.

Later it developed also the other cases, accusative, genitive,

and dative. By this process its distinctively dative force was

obscured while the scope of its use was enlarged. In Post-

 


144                                          THE MOODS.

 

Aristotelian Greek, notably in the Septuagint and the New

Testament, another step was taken. The Infinitive with the

article in the genitive began to assume some such prominence

as at a much earlier time the dative had acquired, and as

before, the sense of its case being in some degree lost, this

genitive Infinitive came to be used as a nominative or accusa-

tive. We mark therefore four stages of development. First,

that for whIch we must go back of the histonc penod of the

Greek language itself, when the Infinitive was distinctly a

dative case. Second, that which is found in Homer: the Infin-

itive begins to be used as subject or object, though the strictly

dative functions still have a certain prominence, and the arti-

cle is not yet used. Third, that of which the beginnings are

seen in Pindar and which is more fully developed in classical

authors of a later time: the Infinitive without the article,

sometimes with dative functions, sometimes with the force of

other cases, is used side by side with the articular Infinitive

in the nominative, genitive, dative, and accusative singular.

Fourth, that which appears in the Septuagint and the New

Testament: all the usages found in the third stage still con-

tinuing, the Infinitive with the article in the genitive begins

to lose the sense of its genitive function and to be employed

as a nominative or accusative.

            From the earliest historic period of the Greek language the

Infinitive partakes of the characteristics both of the verb and

the noun. As a verb it has a subject more or less definite, and

expressed or implied, and takes the adverbial and objective

limitations appropriate to a verb. As a noun it fills the office

in the sentence appropriate to its case. Many of these case-

functions are identical with those which belong to other sub-

stantives; some are peculiar to the lIifinitive.

 

            REM. Concerning the history of the Infinitive, see G.MT. 742, 788 ;

Gild. in T.A.P.A. 1878, and in A.J.P. III. pp. 193 ff.; IV. pp. 241 ff.,


                                    THE INFINITIVE.                                       145

 

pp. 418 fl.; VIII. p. 329; Birklein, Entwickelungsgeschichte des sub-

stantivierten Infinitivs, in Schanz, Beitrage zur. historischen Syntax der

griechischen Sprache, Heft 7.

 

            362. In the Greek of the classical and later periods, the functions of

the Infinitive as an element of the sentence are very various. They may

be classified logically as follows:

 

            I. As A PRINCIPAL VERB (364, 365).

            II. As A SUBSTANTIVE ELEMENT.

                        (1) As subject (384, 385, 390, 393, 404).

                        (2) As object in indirect discourse (390).

                        (3) As object after verbs of exhorting, striving, promising,

                                    hoping, etc. (387-389, 391, 394, 404).

                        (4) As object after verbs that take a genitive (401-403).

            III. As AN ADJECTIVE ELEMENT.

                        (1) As appositive (386, 395).

                        (2) Expressing other adnominallimitations (378, 379, 400).

            IV. As AN ADVERBIAL ELEMENT, denoting,

                        (1) Purpose (366, 367, 370 (d), 371 (d), 372, 397).

                        (2) Indirect object (368).

                        (3) Result (369-371, 398).

                        (4) Measure or degree (after adjectives and adverbs) (376,399).

                        (5) Manner, means, cause, or respect (375, 377,396).

                        (6) A modal modification of an assertion (383).

 

            The articular Infinitive governed by a preposition (406-417) expresses

various adverbial relations, the precise nature of which is determined by

the meaning of the preposition employed. Similarly πρίν or πρίν with

the Infinitive (380-382) constitutes an adverbial phrase of time, the

temporal idea lying in πρίν rather than in the Infinitive.

 

            363. To arrange the treatment of the Infinitive on the basis of such a

logical classification as that given above (362) would, however, disregard

the historical order of development and to some extent obscure the point

of view from which the Greek language looked at the Infinitive. It seems

better, therefore, to begin with those uses of the Infinitive which are most

evidently connected with the original dative function, and proceed to

those in which the dative force is vanishing or lost. This is the general

plan pursued in the following sections, though it is by no means affirmed

that in details the precise order of historical development has been

followed.


146                                          THE MOODS.

 

            THE INFINITIVE WITHOUT THE ARTICLE.

 

            364. The Imperative Infinitive. The Infinitive with-

out the article is occasionally used to express a command

or exhortation. This is the only use of the Infinitive as a

principal verb. It is of ancient origin, being especially

frequent in Homer. EA. 957; G. 1536.

            The New Testament furnishes but one certain instance

of this usage.

Phil. 3:16; πλὴν εἰς ἐφθάσαμεν, τῷ αὐτῷ στποχεῖν, only whereunto

            we have attained, by the same rule walk.

 

            365. Rom. 12:15 affords another probable instance of the imperative

use of the Infinitive. Buttmann supposes an ellipsis of λέγω, and Winer

a change of construction by which the writer returns from the independ-

ent Imperatives used in v.14 to the construction of an Infinitive dependent

on λέγω employed in v. 3. This explanation of change of construction

probably applies in Mark 6:9 (cf. the even more abrupt change in

Mark 5:23); but in Rom. ch. 12 the remoteness of the verb λέγω (in v.

3) from the Infinitive (in v. 15) makes the dependence of the latter upon

the former improbable. B. pp. 271 f.; WH. pp. 397 f.; WT. 316.

 

            366. The Infinitive of Purpose. The Infinitive is used

to express the purpose of the action or state denoted by

the principal verb. EA. 951; G. 1532.

 

Matt. 5:17;  μὴ νομίσητε ὅτι ἦλθον καταλῦσαι τὸν νόμον τοὺς προ-
           
φήτας· οὐκ ἦλθον καταλῦσαι ἀλλὰ πληρῶσαι, think not that I came

            to destroy the law or the prophets: I came not to destroy, but to fulfil.

Luke 18:10; ἄνθρωποι δύο ἀνέβησαν εἰς τὸ ἱερὸν προσεύξασθαι, two

            men went up into the temple to pray.

Acts 10:33;  νῦν οὖν πάντες ἡμεῖς ἐνώπιον τοῦ θεοῦ πάρεσμεν ἀκοῦσαι
           
πάντα τὰ προστεταγμένα σοι ὑπὸ τοῦ κυρίου, now therefore we are

            all here present in the sight of God, to hear all things that have been

            commanded thee of the Lord.

 

            367. The Infinitive expressing purpose is sometimes intro-

duced by ὥστε or ὡς. See 370 (d), 371 (d), 372.


                        THE INFINITIVE WITHOUT THE ARTICLE.                        147

 

            368. THE INFINITIVE AS AN INDIRECT OBJECT. Closely

akin to the Infinitiv:e of Purpose is the Infinitive of the indi-

rect object. The former is a supplementary addition to a

statement in itself complete, and expresses the purpose .had in

view in the doing of the action or the maintenance of the state.

The Infinitive of the indirect object on the other hand is a

complementary limitation of a verb, expressing the direct ten-

dency of the action denoted by the principal verb, or other

similar dative relation. Some of the instances of this usage

are scarcely to be distinguished from the Infinitive of Purpose,

while in others the distinction is clearly marked.

Luke 10:40; Κύριε, οὐ μέλει σοι ὅτι ἀδελφή μου μόνην με κατέλειπεν
           
διακονεῖν, Lord, dost thou not care that my sister has left me to serve

            alone?

Acts 17:21;  Ἀθηναῖοι δὲ πάντες καὶ οἱ ἐπιδημοῦντες ξένοι εἰς οὐδὲν
           
ἕτερον ηὐκαίρουν λέγειν τι ἀκούειν τι καινότερον, now all the

            Athenians and the strangers sojourning there spent their time in nothing

            else than either to tell or to hear some new thing. See also Mark 4:23;

            6:31; 10:40; Luke 7:40; 12:4; Acts 4:14; 7:42; 23:17, 18, 19;

            25:26; Tit. 2:8.

 

            369. The Infinitive of Result. The Infinitive may be

used to denote the result of the action expressed by the

principal verb. When so used it is usually introduced by

ὥστε. HA. 953; G. 1449.

Mark 4:37; καὶ τὰ κύματα ἐπέβαλλεν εἰς τὸ πλοῖον, ὥστε ἤδη γεμί-
           
ζεσθαι τὸ πλοῖον, and the waves beat into the boat, insomuch that the

            boat was now filling.

1 Thess. 1:8; ἐν παντὶ τόπῳ πίστις ὑμῶν πρὸς τὸν θεὸν ἐξελή-
           
λυθεν, ὥστε μὴ χρείαν ἔχειν ἡμᾶς λαλεῖν τι, in every place your

            faith to God-ward is gone forth, so that we need not to speak anything.

 

            370. Under the general head of expressions of result it is

necessary to distinguish three different conceptions:

            (a) Actual result, conceived of and affirmed as actual; in this

case classical Greek uses w!ste with the Indicative. See 236.


148                           THE MOODS.

 

            (b) Tendency or conceived result which it is implied is an

actual result. In this case the result is thought of as that

which the action of the principal verb is adapted or sufficient

to produce, and it is the context or the nature of the case only

which shows that this result is actually produced. In this

case classical Greek uses w!ste with the Infinitive.

            (c) Tendency or conceived result thought of and affirmed

simply as such. In this case the result is one which the

action of the principal verb is adapted or sufficient to

produce, though the actual production is either left in doubt,

or is indicated by the context not to have taken place. Clas-

sical Greek employs ὥστε with the Infinitive (in Homer the

Infinitive without ὥστε).

            To these three may be added as a closely related conception

which the Greek also expressed by ὥστε with the Infinitive:

 

            (d) Purpose, i.e. intended result.

 

            The constructions by which these several shades of meaning

are expressed are substantially the same in the New Testament

as in classical Greek, except that the construction appropriate

to the second meaning has apparently encroached upon the

realm of the first meaning, and the line of distinction between

them has become correspondingly indistinct.  Ὥςστε with the

Indicative occurs very rarely except with the meaning there-

fore, introducing a principal clause; and this fact, together

with the large number of instances in which ὥστε with the

Infinitive is used of a result evidently actual, makes it probable

that the use of ὥστε with the Infinitive is no longer restricted,

as in classical Greek, to instances in which the result is thought

of as theoretical, but is used also of result in fact and in

thought actual. Cf. G.MT. 582-584. There remain, however,

instances entirely similar to those found in classical Greek, in

which a result shown by the context to be actual is appa.rently


            THE INFINITIVE WITHOUT THE ARTICLE.                        149

 

presented simply as one which the event previously expressed

tends to produce. Between these two classes it is evidently

impossible to draw a sharp line of distinction. Oases of the

third class are expressed in the New Testament by the Infini-

tive with or without w!ste. Cf. also 218 and 398.

 

            371. The following examples illustrate New Testament

usage:

 

            (a). Actual resul: conceived and affirmed as such.

Indicative after w!ste.

 

John 3:16; οὕτως γὰρ ἠγάπησεν θεὸς τὸν κόσμον ὥστε τὸν υἱὸν
           
τὸν μονογενῆ ἔδωκεν, for God so loved the world that he gave his only

            begotten Son.

            Infinitive after w!ste.

 

Mark 9:26;  ἐγένετο ὡσεὶ νεκρὸς ὥστε τοῦς πολλοὺς λέγειν ὅτι ἀπέ-
           
θανεν, he became as one dead; insomuch that the more part said that

            he was dead.

 

            (b) Tendency, by implication realized in actual result. In-

finitive, usually after ὥστε.

 

Luke 12:1;  ἐν οἶς ἐπισυναχθεισπων τῶν μυριάδων τοῦ ὄχλου, ὥστε
           
καταπατεῖν ἀλλήλους, in the meantime, when the many thousands of the

            multitude were gathered together, so as to tread one upon another.

Rev. 5:5;  ἰδοὺ ἐνίκησεν λέων ἐκ τῆς φυλῆς Ἰούδα, ῥίζα Δαυείδ,
           
ἀνοῖξαι τὸ βιβλίον,  behold the lion that is of the tribe of Judah, the

            Root of David, hath overcome, to open the book. See also Acts 1:25;

            2 Cor. 1:8; 2 Thess. 2:4.

 

            (c) Tendency or conceived result thought of as such. In-

finitive, usually after ὥστε.

 

1 Cor. 13:2;  κἂν ἔχω πᾶσαν τὴν πίστιν ὥστε ὄρη μεθιστάνειν, and if

            I have all faith, so as to remove mountains.

Matt. 10:1; ἔδωκεν αὐτοῖς ἐξουσίαν πνευμάτω ἀκαθάρτων ὥστε ἐκ-
           
βάλλειν αὐτά, he gave them authority over unclean spirits to cast them

            out. Here probably belongs also Rom. 1:10. See also 2 Cor. 2:7;

            Rev. 16:9.


150                                          THE MOODS.

 

            REM. The Infinitive in Heb. 6:10, οὐ γὰρ ἄδικος θεὸς ἐπιλαθέσθαι,

must also be accounted an Infinitive of conceived result. The origin of

this idiom may be an impersonal construction (cf. G.MT. 762), but it

has departed in meaning as well as in form from its original. The mean-

ing of this sentence is not, It would not be unjust for God to forget, but,

God is not unjust so as to forget.

            (d) Purpose, i.e. intended result.

Luke 4:29; καὶ ἤγαγον αὐτὸν ἕως ὀφρύος τοῦ ὄρους ὥστε κατα-
           
κρημνίσαι αὐτόν, and they led him unto the brow of the hill that

            they might throw him down headlong. See also Luke 20:20.

 

            REM. In Matt. 27:1, ὥστε with the Infinitive stands in definitive appo-

sition with συμβούλιον, defining the content of the plan, rather than ex-

pressing the purpose of making it.

 

            372. The Infinitive is used with cJs in Luke 9 : 52 according to the

reading adopted by WHo (most editors read ὥστε) and in Acts 20: 24

according to,the generally adopted reading (WH. read a Subjunctive).

In both cases the phrase dEJnotes purpose. No instance of ὡς with the

Infinitive denoting result occurs in the New Testament. See Th. ὡς, III.,

and references cited there, and cf. G. 1456. In 2 Cor. 10:9 ὡς ἄν is used

with the Infinitive. This usage also occurs rarely in classical and later

Greek. See Alf. ad loco and Gr. p. 230. The phrase is elliptical, the In-

finitive most probably expressing purpose and ὥς ἄν modifying it in the

sense of quasi. WH. p. 390; WT. p. 310.

 

            373. In the New Testament the Infinitive is not used either with ὥστε

or ἐφ' or ἐφ' ᾦτε in the sense on condition that. HA. 953, b; G.

1453, 1460.

 

            374. The classical usage of an Infinitive (of conceived result) with ,

or ὥστε, or ὠς, after a comparative, does not occur in the New Testa-

ment. The Infinitive after in the New Testament is used as the correIa-

tive of some preceding word or phrase, and usually as a nominative. See

Luke 18:25; Acts 20:35. On Acts 17:21 cf. 368.

 

            375. Somewhat akin in force tο the Infinitive of (conceived)

result, but probably of Hebraistic origin, is the Infinitive used

to define more closely the content of the action denoted by a

previous verb or noun. Of. Hr. 29, 3, e.

 


            THE INFINITIVE WITHOUT THE ARTICLE.                        151

 

Acts 15:10; νῦν οὖν τί πειράζετε τὸν θεόν, ἐπιθεῖναι ζυγὸν ἐπὶ τὸν
           
τράχηλον τῶν μαθητῶν, now therefore why tempt ye God, that ye

should put (i.e. by putting, or in that ye put) a yoke upon the neck

of the disciples? Cf. Ps. 78: 18 (Hebrew).

Heb. 5:5;  οὕτως καὶ χριστὸς οὐχ ἑαυτὸν ἐδόξασεν γενηθῆναι ἀρχιερέα,
            so Christ also glorified not himself to be made a high priest.

            See also Luke 1:54, 72; cf. 1 Sam. 12:23, ἀνιέναι; 22:13; Ps. Sol. 2:28,

            39, 40. See Ryle and James, Ps. Sol. p. lx:xxiii.

 

376. The Infinitive limiting Adjectives and Adverbs.

The Infinitive is used with adjectives and adverbs of abil-

ity, fitness, readiness, etc., to denote that which one is or is

not able, fit, or ready to do. HA. 952; G. 1526.

 

Mark 1:7;  οὖ ούκ εἰμὶ ἱκανὸς κύψας λῦσαι τὸν ἱμάντα τῶν ὑποδημάτων
           
αὐτῦ, the latchet of whose shoes 1 am not worthy to stoop down and

unloose.

2 Tim. 2:2; οἵτινς ἱκανοὶ ἔσονται καὶ ἑτέρους διδάξαι, who shall be able

            to teach others also.

Rev. 4:11;  ἄξιος εἶ, κύριος καὶ θεὸς ἡμῶν, λαβεῖν τὴν δόξαν καὶ
           
τὴν τιμὴν καὶ τὴν δύναμιν, worthy art thou, our Lord and our God, to

receive the glory and the honor and the power. See also Luke 14: 31

2 Cor. 12:14.

 

377. The Infinitive may be used after any adjective to limit

its application to a particular action. HA. 952; G. 1528.

 

Heb. 5:11;  περὶ οὗ πολὺς ἡμῖν λόγος καὶ δυσερμήνευτος λέγειν,

of whom we have many things to say, and hard of interpretation--a

felicitous free translation. More literally it would read, concerning

whom our discourse is much, and hard of interpretation to state, i.e.

hard to state intelligibly.

 

378. The Infinitive limiting Nouns. The Infinitive is

used with abstract nouns of ability, authority, need, hope,

etc., to denote that which one has, or has not, ability,

authority, need, etc., to do. Here may also be included


152                                          THE MOODS.

 

the Infinitive after ὥρα, which implies a necessity. HA.

952; G. 1521. .

 

Matt. 3:14; ἐγὼ χρείαν ἔχω ὑπὸ σοῦ βατισθῆναι, I have need to be

            baptized of thee.

John 1:12; ἔδωκεν αὐτοῖς ἐξουσίαν τέκνα θεοῦ γενέσθαι, to them gave

            he the right to become children of God.

Rom. 13:11; καὶ τοῦτο εἰδότες τὸν καιρόν, ὅτι ὥρα ἤδη ὑμᾶς ἐξ ὕπνου
           
ἐγεθῆναι, and this, knowing the season, that now it is high time for you

to awake out of sleep. See also 2 Cor. 10:15; Rev. 9:10.

 

379. The Infinitive is also occasionally used after con-

crete nouns cognate with verbs which take an object In-

finitive.

 

Gal. 5:3; ὀφειλέτης ἐστὶν ὅλον τὸν νόμον ποιῆσαι, he is a debtor to

do the whole law.

 

380. The Infinitive is used after πρίν or πρίν .  HA.

955; G. 1469-1474.

 

Mark 14:30; πρὶν δὶς ἀλέκτορα φωνῆσαι τρίς με ἀπαρνήσῃ, before

the cock crow twice, thou shalt deny me thrice.

John 4:49; κύριε, κατάβηθι πρὶν ἀποθανεῖν τὸ παιδίον μου, Sir, come

            down ere my child die.                                                                

 

381. The use of after πρίν, which occurs twice in the Iliad, fre-

quently in Herodotus, and rarely in Attic writers, is well attested in three

of the thirteen instances in the New Testament in which πρίν is used

with the Infinitive, and occurs as a variant in other passages. G.1474.

 

382. As respects the mood which follows πρίν or πρίν , New Testa-

ment usage is the same as that of Post-Homeric Greek in general, in that

the Infinitive is generally (in the New Testament invariably) used when

the leading clause is-affirmative; the Subjunctive and Optative occur only

after a negative leading clause. The Indicative after πρίν which some-

times occurs in classical Greek, chiefly after a negative leading clause, is

not found in the New Testament. HA. 924, a; G. 1470.

 


THE INFINITIVE WITHOUT THE ARTICLE.                        153

 

383. The Infinitjve used absolutely in a parenthetic clause

occurs but once in the New Testament. HA. 956; G. 1534.

Heb. 7:9 ὡς ἔπος εἰπεῖν, so to speak.

 

384. The Infinitive as Subject. The Infinitive may

be used as the subject of a finite verb. HA. 949, 959;

G.1517.

Matt. 3:15;  οὕτω γὰρ πρέπον ἐστὶν ἡμῖν πληρῶσαι πᾶσαν δικαιοσύνην,

for thus it becometh us to fulfil all righteousness.

Luke 18:25; Εὐκοπώτερον γάρ ἐστιν κάμηλον διὰ τρήματος βελόνης
           
εἰσελθεῖν, for it is easier for a camel to enter in through a needle's eye.

See a.lso Mark 3:4; Luke 16:17; 20:22; Gal. 4:18.

 

385. The Infinitive with subject accusative sometimes

stands as the subject of an impersonal verb (δολεῖ, etc.). Fre-

quently, however, the personal construction is employed, that

which is properly the subject of the Infinitive being put in the

nominative as the subject of the principal verb. But the logi-

cal relation is the same in either case. HA. 944.

In the New Testament the personal construction is regularly

employed with δοκεῖ.

Acts 17:18; ξένων δαιμονίων δοκεὶ καταγγελεὺς εἶναι, he seemeth to be

            a setter forth of strange gods. See also Gal. 2:9; Jas. 1:26, etc.

 

REM. Concerning the Infinitive as subject of ἐγένετο, see 357, 360.

 

386. The Infinitive as Appositive. The Infinitive may

stand in apposition with a noun or pronoun. EA. 950;

G.1517.

 

Jas. 1:27;  θρησκεία καθαρὰ καὶ ἀμίαντος . . . αὕτη ἐστίν, ἐπι-
           
σκέπτεσθαι ὀρφανοὺς καὶ χήρας ἐν τῇ θλίψει αὐτῶν, pure religion

and undefiled. . . is this, to visit orphans and widows in their afflic-

tion. See also Acts 15:28; 1 Thess. 4:3.

 

387. The Infinitive as Object. The Infinitive may be

used as the object of a verb. The verbs which are thus


154                                          THE MOODS.

 

limited by an Infinitive are in part such as take a noun or

pronoun in the accusative as object, in part such as take a

noun or pronoun in the genitive as object, in part verbs

which cannot take a noun or pronoun as object but require

an Infinitive to complete their meaning. HA. 948; G.

1518, 1519.

 

Matt. 19:14; ἄφετε τὰ ποιδία καὶ κωλύετε αὐτὰ ἐλθεῖν πρός με,

suffer the little children, and forbid them not, to come unto me.

Mark 12:12;  καὶ ἐζήτουν αὐτὸν κρατῆσαι, and they sought to lay hold

            on him.

Luke 16:3;  σκάπτειν οὐκ ἰσχύω, ἐπαιτεῖν αἰσχύνομαι, I have not strength

to dig; to beg I am ashamed.

Heb.7:25; ὅθεν καὶ σώζειν εἰς τὸ παντελὲς δύναται, wherefore also he

            is able to save to the uttermost. See also Matt. 1:19; John 5:18;

Rom. 14:2; Gal. 3:2, et freq.

 

388. The Infinitive χαίρειν in salutations is to be regarded

as the object of an unexpressed verb of bidding.

 

Acts 23:26; Κλαύδιος Λυσίας τῷ κρατίστῳ ἡγεμόνι Φήλικι χαίρειν,

Claudius Lysias unto the most excellent governor Felix, greeting.

Jas. 1:1;   Ἰάκωβος. . . .ταῖς δώδεκα φυλαῖς ταῖς ἐν τῇ διασπορᾷ
           
χαίρειν, James. . . to the twelve tribes which are of the Dispersion,

greeting.

 

389. The verbal idea governing the Infu1itive is sometimes implied

rather than expressed. The Infinitive τεκεῖν in Rev. 12:2 is doubtless an

object Infinitive governed by the idea of desire implied in the preceding

participles. The Infinitive ψεύσασθαι in Acts 5:3 may be regarded as an

object Infinitive governed by the idea of persuading implied in ἐπλήρωσεβ
τὴν καρδίαν, or as an Infinitive of conceived result. Cf. 370 (c).

 

390. The Infinitive in Indirect Discourse. The Infini-

tive is frequently used in the indirect quotation of asser-

tions. It is usually the object of a verb of saying or of

thinking, or the subject of such a verb in the passive

voice. HA. 946; G. 1522.


THE INFINITIVE WITH THE ARTICLE.                     155

 

Mark 12:18; οἵτινες λέγουσιν ἀνάστασιν μὴ εἶναι, which say that there is

            no resurrection.

John 21:25; οὐδ'αὐτὸν οἶμαι τὸν κόσμον χωρήσειν τὰ γραφόνμενα
           
βιβλία, I suppose that even the world itself will not contain the books

that will be written.

Heb. 11:5; πρὸ γὰρ τῆς μεταθέσεως μεμαρτύρηται εὐαρεστηκέναι τῷ
           
θεῷ, for before his translation he had witness borne to him that he

had been well-pleasing unto God. See also Luke 2:26; 22:34;

24:46 (?); John 12:29; Acts 16:27; Rom. 15:8; 2 Tim. 2:18;

1 John 2:9.

 

REM. 1. Respecting the force of the tenses of the Infinitive in indirect

discourse, see 110-114.

 

REM. 2. Respecting the use of negatives with the Infinitive in indirect

discourse, see 480-482.

 

391. The Infinitive occurs frequently as object after verbs

of hoping, promising, swearing, and commanding, with a force

closely akin to that of the Infinitive in indirect discourse.

Such instances are not, however, usually included under that

head. Cf. 337, and G.MT. 684.

 

THE INFINITIVE WITH THE ARTICLE.

 

392. The prefixing of the article to the Infinitive tends to

the obscuring of its original dative force, while it emphasizes

its new substantive character as a noun which can be used in

any case. Some of the uses of the Infinitive with the article

differ from those without the article only by the greater em-

phasis on the substantive character of the form. This is the

case with its use as subject and object. Others express nearly

the same relations which were expressed by the Infinitive

without the article, but with a different thought of the case-

relation involved. Thus the use of the Infinitive without the

article after adjectives of fitness, worthiness, etc., doubtless


156                                          THE MOODS.

 

sprang originally from the thought of the Infinitive as a dative.

The Infinitive with the article after such adjectives is thought

of as a genitive, as is evident from the use of the article tou?.

The difference in meaning is, however, very slight. Compare

the English worthy to receive and worthy of receiving. Still

other uses of the Infinitive with the article are wholly new,

being developed only after the Infinitive had begun to be used

with the article. To this class belongs the use of the Infini.

tive after prepositions.

 

REM. The Infinitive with the article being by means of that article

practically a declinable noun, the various uses are grouped in the follow-

ing sections according to cases.

 

393. The Infinitive with τό as Subject. The Infinitive

with the article τό is used as the subject of a finite verb.

HA. 959; G.1542.

 

Matt.  15:20;  τὸ δὲ ἀνίπτοις χερσὶν φαγεῖν οὐ κοινοῖ τὸν ἄνθρωπον,

but to eat with unwashen hands defileth not the man. See also Matt.

20:23; Mark 9:10; 12:33; Rom. 14:21.

 

394. The Infinitive with τό as Object. The Infinitive

with the article τό is used as the object of a transitive

verb. This usage is far less common than the object

Infinitive without the article. HA. 959; G.1543.

 

Acts 25:11; οὐ παραιτοῦμαι τὸ ἀποθανεῖν, I refuse not to die. See

also 2 Cor. 8:11; Phil. 2:6.

 

395. The Infinitive with the Article, in Apposition.

The Infinitive with the article may stand in apposition with

a preceding noun or pronoun.

 

Rom. 4:13; οὐ γὰρ διὰ νόμου ἐπαγγελία τῷ  Ἀβραὰμ τῷ σπέρ-
           
ματι αὐτοῦ, τὸ κληρονόμον αὐτὸν εἶναι κόσμου, for not through the

law was the promise to Abraham or to his seed, that he should be heir

of the world.


THE INFINITIVE WITH THE ARTICLE.                     157

 

2 Cor. 2:1; ἔκρινα γὰρ ἐμαυτῷ τοῦτο, τὸ μὴ πάλιν ἐν λύπῃ πρὸς ὑμᾶς
           
ἐλθεῖν, for I determined this for myselj; that I would not come again to

you with sorrow. See also Rom. 14:13.

 

396. The Infinitive with τῷ. The Infinitive with the

article τῷ is used in classical Greek to express cause, man-

ner, means. In the New Testament it is used to express

cause. Its only other use is after the preposition ἐν. HA.

959; G. 1547.

 

2 Cor. 2:13; τῷ μὴ εὑρεῖν με Τίτον τὸν ἀδελφόν μου, because I found

            not Titus my brother.

 

397. The Infinitive of Purpose with τοῦ. The Infini-

tive with the article τοῦ is used to express the purpose of

the action or state denoted by the principal verb. HA.

960; G. 1548.

 

Matt. 2:13; μέλλει γὰρ Ἡρῴδης ζητεῖν τὸ παιδίον τοῦ ἀπολέσαι αὐτό,

for Herod will seek the young child to destroy him. See also Matt.

24: 45; Luke 2 : 24, 27; Acts 26 : 18; Phil. 3 : 10.

 

REM. That the Infinitive with τού expresses purpose with substan-

tially the same force as the simple Infinitive appears from the joining of

the two together by καί.

 

Luke 2:22, 24; ἀνήγαγον αὐτὸν εἰς  Ἰεροσόλυμα παραστῆσαι τῷ κυρίῳ,

. . . . καὶ τοῦ δοῦναι θυσίαν, they brought him up to Jerusalem, to

present him to the Lord, and to offer a sacrifice. Ct. also Luke 1:76,

77; 1:79.

 

398. The Infinitive of Result with τοῦ. The Infinitive

with the article τοῦ is occasionally used in the New Tes-

tamen't to express conceived result. Of. 218 and 369-371.

 

Matt. 21:32;  ὑμεῖς δὲ ἰδόντες οὐδὲ μετεμελήθητε ὕστερον τοῦ πιστεῦ-
           
σαι αὐτῷ, and ye, when ye saw it, did not even repent afterward,

so as to believe him. See also Acts 7:19; Rom. 7:3; probably also

Acts 18:10; cf. Gen.3:22; 19:21; 34:17, 22; Isa.5:14.


158                                          THE MOODS.

 

REM. Meyer takes the Infinitive phrase τοῦ μὴ εἶναι in Rom. 7:3 as

expressing a divine purpose, and adds that τοῦ with the Infinitive never

expresses result, not even in Acts 7:19. But this is grammatical purism

not justified by the evidence. The uniformly telic force of tou? with the

Infinitive can be maintained only by evasive definition or forced inter-

pretation.

 

399. The Infinitive with τοῦ after Adjectives. The

Infinitive with the article τοῦ is used with such adjectives

as may be limited by a simple Infinitive. HA. 959; G.

1547. Cf. 376.

 

Acts 23:15 ἕτοιμοί ἐσμεν τοῦ ἀνελεῖν αὐτόν, we are ready to slay him.

See also Luke 24:25.

 

400. The Infinitive with τοῦ after Nouns. The Infini-

tive with the article tou? is used to limit nouns. The rela-

tions thus expressed are very various and are not always

easy to define exactly. Instances occur not only, as in

classical Greek, of the objective genitive, but also of the

genitive of characteristic, the genitive of connection, and

the .appositional genitive. HA. 959; G. 1547.

 

Heb. 5:12 πάλιν χρείαν ἔχετε τοῦ διδάσκειν ὑμᾶς, ye have need again

            that some one teach you.

Luke 2:21; καὶ ὅτε ἐπλήσθησαν ἡμέραι ὀκτὼ τοῦ περιτεμεῖν αὐτόν,
            and when eight days were fulfilled for circumcising him.

Rom. 11:8; ἔδωκεν αὐτοῖς θεὸς πνεῦμα κατανύξεως, ὀφθαλμοὺς τοῦ
           
μὴ βλέπειν καὶ ὦτα τοῦ μὴ ἀκούειν, God gave them a spirit of stupor,

eyes that see not, and ears that hear not. See also Luke 1:57, 74;

2:6; 10:19; 21:22; 22:6; Acts 14:9; 20:3; Rom. 1:24;

l Cor. 9:10; 2Cor. 8:11;  I Pet. 4:17; cf.Gen.16:3; I Sam. 2:24.

 

401. The Infinitive with τοῦ after Verbs that take the

Genitive. The Infinitive with τοῦ is used as the object of

verbs which take a noun in the genitive as object, especially

of verbs of hindering, etc. HA. 959, 963; G. 1547, 1549.


THE INFINITIVE WITH THE ARTICLE.                     159

 

Luke 1:9; ἔλαχε τοῦ θυημιᾶσαι, it was his lot (prop. he obtained by lot)

to burn incense.

2 Cor. 1:8; ὥςτε ἐξαπορηθῆναι ἡμᾶς καὶ τοῦ ζην, insomuch that we de-

spaired even of life.

Rom. 15:22; διὸ καὶ ἐνεκοπτόμην τὰ πολλὰ τοῦ ἐλθεῖν πρὸς ὑμᾶς,

wherefore also I was hindered these many times from coming to you.

Cf. Gen. 34:19; Ps. Sol. 2:28, 29.

 

402. In classical Greek, verbs of hindering are followed by

three constructions, (a) Infinitive without the article, (b) In-

finitive with τοῦ, (c) Infinitive with τό.  Μή may be used or

omitted with the Infinitive withput difference' of meaning.

HA. 963; G. 1549, 1551; G.MT. 791 (exx.). In the New

Testament, all these constructions occur except that with τὸ μή.

See Matt. 19:14; Rom. 15:22; 1 Cor. 14:39; Gal. 5:7;

Acts 10:47.

 

403. The Infinitive with τοῦ μή after verbs of hindering is closely akin

to the Infinitive of Result. Cf. Luke 24:16; Acts 14:18.

 

REM. Meyer's interpretation of τοῦ μὴ ἐπιγνῶναι αὐτόν in Luke 24:16

as expressing a divine purpose (the English translation does not correctly

represent the meaning of the German original), is not required by New

Testament usage. The Greek most naturally means, Their eyes were

held from knowing him. Cf. 398, Rem.

 

404. The Infinitive with τοῦ as Subject or Object.

The Infinitive with τοῦ is used even as the subject of a

finite verb or as the object of transitive verbs which regu-

larly take a direct object. This is a wide departure from

classical usage, and indicates that the sense of the genitive

character of the article τοῦ before the Infinitive was partly

lost in later Greek. B. p. 270; WM. pp. 411 f.; WT. pp. 327 f.

 

Acts 27:1; ἐκρίθη τοῦ ἀποπλεῖν ἡμᾶς εἰς τὴν Ἰταλίαν, it was determined

that we should sail for Italy. See also Luke 4:10; 5:7; Acts 3:12;

10:25; 15:20; 21:12; 23:20; 1 Sam. 12:23; Eccl. 4:13, 17;

1 Macc. 3:15.

 


160                                          THE MOODS.

 

405. The origin of this use of the Infinitive with τοῦ is perhaps in such

usages 88 appear in Luke 17:1; 1 Cor. 16:4; and still more in such as

that in Luke 4:10. In Luke 17:1 the genitive is apparently suggested

by the idea of hindering or avoiding in the adjective ἀνένδεκτον; in

1 Cor. 16:4 it is the adjective ἄξιον which gives occasion to the genitive;

but in both cases the Infinitive seems to be logically the subject of the

copulative verb, the adjective being the predicate. Whether this con-

struction represents the thought in the mind of the writer, or whether

the expression is rather to be regarded as an impersonal one, the Infini-

tive being dependent on the predicate adjective, cannot with confidence

be decided. Such usages as Luke 4:10 and 5:7 doubtless owe their

origin to the same mental process by which a clause introduced by ἵνα

came to stand as the object of a verb of exhorting. Ps. Sol. 2:28 com-

pared with Luke 12:45 is also suggestive. It is doubtless the idea of

hindering in χρονίζω that gives rise to the genitive in the former passage;

in the latter the Infinitive is a direct object.

 

406. The Infinitive with the Article governed by

Prepositions. The Infinitive with the article τό, τοῦ, τῷ

is governed by prepositions. HA. 959; G.1546.

The prepositions so used in the New Testament are:

with the accusative, διά, εἰς, μετά, προός; with the genitive,

ἀντί, διά, ἐκ, ἕνεκεν, ἕως, πρό; with the dative, ἐν.

 

Mark 4:6; καὶ διὰ τὸ μὴ ἔχειν ῥίζαν ἐξηράνθη, and because it had no

root, it withered away.

1 Thess. 3:5; ἔπεμψα εἰς τὸ γνῶναι τὴν πίστιν ὑμῶν, I sent that I might

            know your faith.

Mark 14:28; ἀλλὰ μετὰ τὸ ἐγερθῆναί με προάξω ὑμᾶς εἰς τὴν Γαλι-
           
λαίαν, howbeit, after I am raised up, 1 will go before you into Galilee.

Matt. 6:1; προσέχετε [δὲ] τὴν δικαιοσύνην ὑμῶν μὴ ποιεῖν ἔμροσθεν
           
τῶν ἀνθρώπων πρὸς τὸ θεαθῆναι αὐτοῖς, take heed that ye do not

your righteousness before men, to be seen of them.

Gal. 3:23; πρὸ τοῦ δὲ ἐλθεῖν τὴν πίστιν ὑπὸ νόμον ἐφρουρούμεθα, but

before faith came, we were kept in ward under the law.

Luke 24:51; καἰ ἐγένετο ἐν τῷ εὐλογεῖν αὐτὸν αὐτοῦς διέστη ἀπ'
           
αύτῶν, and it came to pass, while he blessed them, he parted from them.

 

407. These prepositions vary greatly in frequency in the

New Testament. Εἰς occurs with the Infinitive 63 times


THE INFINITIVE WITH THE ARTICLE.                     161

 

(Infinitives 72); ε'ν 52 times (Infinitives 56); διἀ with the

Accusative 27 times (Infinitives 31); μετά 15 times; πρός 12

times; πρό  9 times; each of the others once (WH. text). See

Votaw, Infinitive in Biblical Greek, p. 20; cf. G.MT. 800-802.

 

408. Διά governing the Infinitive with τό denotes cause, and is nearly

equivalent to ὅτι or διότι with the Indicative, differing in that the Infini-

tive gives in itself no indication of the time of the action.

 

Jas. 4:2, 3;  οὐκ ἔχετε διὰ τὸ μὴ αἰτεπῖσθαι ὑμᾶς· αἰτεῖτε καὶ οὐ λαμ-
           
βάνετε, διότι κακῶς αἰτεῖσθε, ye have not, because ye ask not. Ye

ask, and receive not, because ye ask amiss.

 

In Mark 5:4 διά with the Infinitive expresses the evidence rather than

the cause strictly so called.

 

409. Εἰς governing the Infinitive with τό most commonly expresses

purpose. It is employed with special frequency by Paul, but occurs also

in Heb., 1 Pet., and Jas.

 

Rom. 8:29; ὅτι οὓς προέγνω, καὶ προώρισεν συμμόρφους τῆς εἰκόνος
           
τοῦ υἱοῦ αὐτοῦ, εἰς τὸ εἶναι αἠτὸν πρωτότοκον ἐν πολλοῖς ἀδελφοῖς,

for whom he foreknew, he also foreordained to be conformed to

the image of his Son, that he might be the first-born among many brethren.

See also Rom. 1:11; 3:26; 7:4; Eph. 1:12; Phil. 1:10; Heb. 2:17; Jas.1:18;

1 Pet. 3:7.

 

410Εἰς with the Infinitive is also used, like the simple Infinitive, to

represent an indirect object. Cf. 368.

 

1 Cor. 11:22;  μὴ γὰρ οἰκίας οὐκ ἔχετε εἰς τὸ ἐσθίειν καὶ πίνειν,  what?

have ye not houses to eat and to drink in?  See also Matt. 20:19;

26:2.

 

411. Eἰς with the Infinitive also expresses tendency, measure of effect,

or result, conceived or actual.

 

Heb. 11:3;  πίστει νοοῦμεν κατηρτίσθαι τοὺς αἰῶνας ῥήματι θεοῦ, εἰς
           
τὸ μὴ ἐκ φαινομένων τὸ βλεπόμενον γεγονέναι, by faith we under-

stand that the worlds have been framed by the word of God, so that

what is seen hath not been made out of things which do appear. See

also Rom. 12:3; 2 Cor. 8:6; Gal. 3:17; 1 Thess. 2:16.


162                                                      THE MOODS.

 

Εἰς τὸ ἐσθίειν in 1 Cor.8:10 either expresses measure of effect or is the

indirect object of οἰκοδομηθήσεται. Εἰς τὀ εἶναι αὐτοὺς ἀναπολογήτους in

Rom. 1:20 might appropriately be interpreted as expressing purpose but

for the causal clause which follows. This clause could be joined to an

expression of purpose only by supposing an ellipsis of some such expres-

sion as καὶ οὕτως εἰσίν, and seems therefore to require that εἰς τὸ εἶναι be

interpreted as expressing result.

 

REM. Meyer's dictum (see on Rom. 1:20) that εἰς with the articular

Infinitive is always telic, is, like his similar dictum respecting τοῦ with

the Infinitive, a case of grammatical purism, not justified by the evidence.

 

412. Εἰς with the Infinitive is also used, like ἵνα with the Subjunctive,

or the simple Infinitive, as the direct object of verbs of exhorting, etc.

1 Thess. 2: 12; 3: 10; 2 Thess. 2 : 2.

 

413. Εἰς  with the Infinitive is still further used, like the simple Infini-

tive, to limit an adjective, as in Jas. 1:19, or a noun, as in Phil. 1:23.

 

414. Πρός governing the Infinitive with τό usually expresses purpose;

it is occasionally used with the sense, with reference to.

 

Matt. 6:1; προσέχετε [δὲ] τὴν δικαιοσύνην ὑμῶν μὴ ποιεῖν ἔμπροσθεν
           
τῶν ἀνθρώπων πρὸς τὸ θεαθῆναι αὐτοῖς, but take heed that ye do not

your righteousness before men, to be seen of them.

Matt. 26:12;  βαλουσα γὰρ αὕτη τὸ μύρον τοῦτο ἐπὶ τοῦ σώματός μου
           
πρὸς τὸ ἐνταφιάσαι με ἐποίησεν, for in that she poured this ointment

upon my body, she did it to prepare me for burial. See also Matt.

5:28; 13:30; 2 Cor. 3:13; Eph. 6:11, etc.(purpose); Luke 18:1

(reference).

 

415.  Ἐν governing the Infinitive with τῷ is most commonly temporal,

but occasionally expresses other relations, such as manner, means, or

content. This construction is especially frequent in Luke and Acts.

Luke 8:5; καὶ ἐν τῷ σπερειν αὐτὸν μὲν ἔπεσεν παρὰ τῆν ὁδόν, and

as he sowed, some fell by the way side.

Acts 3:26;  ὑμῖν πρῶτον ἀναστήσας θεὸς τὸν παῖδα αὐτοῦ ἀπέστειλεν
           
αὐτὸν εὐλογοῦντα ὑμᾶς ἐν τῷ ἀποστρέφειν ἕκαστον ἀπὸ τῶν πονη-
           
ριῶν [ὑμῶν], unto you first God, having raised up his Servant, sent

him to bless you, in turning away everyone of you from your iniquities.

See also Luke 1:8; Acts 9:3; 11:15; Ps. Sol. 1:1 (temporal);

Luke 12:15; Acts 4:30; Heb. 2:8; 3:12, 15; Ps. Sol. 1:3; Gen.

19:16; 34:15.


 

THE PARTICIPLE.                                     163

 

416. The force of the other prepositions used with the Infinitive

scarcely needs special definition, the meaning of each being in general

the same as that of the same preposition governing nouns. Respecting

the force of the tenses after prepositions, see 104-109.

 

417. Concerning the Infinitive without the article governed by prepo-

sitions, see G.MT. 803, and cf. Gen. 10:19. The Infinitive γίνεσθαι in

Acts 4:30, which is by R. V. taken as the object of δός is more probably

governed by the preposition ἐν. It is however not strictly without the

article, the τῷ which precedes ἐκτείνειν belonging in effect also to γίνεσθαι.

 

 

THE PARTICIPLE.

 

418. The Participle is a verbal adjective, sharing in part

the characteristics of both the verb and the adjective. As a

verb it has both tense functions and functions which may be

designated as modal functions, being analogous to those which

in the case of verbs in the Indicative, Subjunctive, or Optative

belong to the mood. For the proper understanding of a par-

ticiple, therefore, it is necessary to consider (a) The grammat-

ical agreement, (b) The use of the tense, and (c) The logical

force or modal function. The first and second of these have

already been treated, grammatical agreement in 116, the uses

of the tenses in 118-156. It remains to consider the logical

force or modal function of the participle. From the point of

view of the interpreter this is usually the matter of most

importance.

 

419. In respect to logical force, participles may be classified

as Adjective, Adverbial, and Substantive.

 

REM. 1. The terminology here employed for the classification of parti-

ciples differs somewhat from that commonly employed. It is adopted

substantially from the article of Professor Wm. Arnold Stevens, "On

the Substantive Use of the Greek Participle" in T.A.P.A. 1872. The

Adjective Participle corresponds nearly to the Attributive Participle


164                                          THE MOODS.

 

as treated in G. and EA., the Adverbial Participle to the Circum-

stantial Participle, and the Substantive Participle to the Supplementary

Participle.

 

REM. 2. Respecting the use of the negatives μή and οὐ with partici-

ples, see 485.

 

 

THE ADJECTIVE PARTICIPLE.

 

420. The Adjective Participle limits its subject directly

and exclusively. It attributes the action which it denotes

to the subject as a quality or characteristic, or assigns the

subject to the class marked by that action. EA. 965; G.

1559.

Acts 10:1, 2; ἀνὴρ δέ τις ἐν Καισαρίᾳ ὀνόματι Κορνήλιος, ἑκατοντάρ-
           
χης ἐκ σπείρης τῆς καλουμένης Ἰταλικῆς, εὐσεβὴς καὶ φοβούμενος
           
τὸν θεὸν σὺν παντὶ τῷ οἴκω αὐτοῦ, ποιῶν ἐλεημοσύνας πολλὰς τῷ
           
λαῷ καὶ δεόμενος τοῦ θεοῦ διὰ παντός, now there was a certain man
            in  Cresarea, Cornelius by name, a centurion

of a band called the Italian band, a devout man and one that feared God with all his house, who gave much alms to the people and prayed to God alway. The four participles in this sentence are all Adjective Participles, de-

scribing their subject. This is especially clear in the case of φοβού-

μενος, which is joined by καί to the adjective εὐσεβής. For other

similar examples see Col. 1:21; Jas. 2:15; see also examples

under the following sections.

 

421. The Adjective Participle may be used attributively or

predicatively. When used attributively it may be either re-

strictive or explanatory.

 

422. The Restrictive Attributive Participle. An at-

tributive Adjective Participle may be used to define or

identify its subject, pointing out what person or thing is

meant. It is then equivalent to a restrictive relative clause.

Cf. 295.

 


THE ADJECTIVE PARTICIPLE.                                    165

 

John 6:50 οὗτός ἐστιν ἄρτος ἐκ τοῦ οὐρανοῦ καταβαίνων, this is
            the bread which cometh down out of heaven.

Jude 17;  μνήσθητε τῶν ῥημάτων τῶν προειρημένων ὑπὸ τῶν ἀπο-
           
στόλων τοῦ κυρίου ἡμῶν Ἰησοῦ Χριστοῦ, remember the words which

have been spoken before by the apostles of our Lord Jesus Christ.

 

423. The subject of the Restrictive Attributive Participle is

often omitted. The participle is then an Adjective Participle

used substantively. Such a participle usually has the article,

but not invariably. HA. 966; G. 1560.

Matt. 10:37;  φιλῶν πατέρα μητέρα ὑπὲρ ἐμὲ οὐκ ἔστιν μου ἄξιος,

he that loveth father or mother more than me is not worthy of me.

Acts 5:14; προσετίθεντο πιστεύοντες τῷ κυρίῳ πλήθη ἀνδρῶν τε καὶ
           
γυναικῶν, believers were added to the Lord, multitudes both of men and

            women.

Acts 10:35;  ἀλλἐν παντὶ ἔθνει φοβούμενος αὐτὸν καὶ ἐργαζόμενος
           
δικαιοσύνην δεκτὸς αὐτῷ ἐστίν, but in every nation he that feareth

him, and worketh righteousness, is acceptable to him.

Rev. 1:3; μακάριος ἀναγινώσκων καὶ οἱ ἀκούοντες τοὺς τῆς
           
προφητείας καὶ τηροῦντες τὰ ἐν αύτῇ γεγραμμένα, blessed is he

            that readeth, and they that hear the words of the prophecy, and keep the

            things which are written therein.

 

424. A noun without the article, or an indefinite pronoun, is

sometimes limited by a participle with the article. The article

in this case does not make the noun strictly definite. The

person or thing referred to is placed within the class charac-

terized bythe action denoted by the participle, and the atten-

tion is directed to some one or to certain ones of that class,

who are not, however, more specifically identified. Nearly the

same meaning is expressed by a participle without the article,

or on the other hand by a relative clause limiting an indefinite

substantive. For classical examples of this usage see WM. p.

136; WT. pp. 109 f.

Col. 2:8; βλέπετε μή τις ὑμᾶς ἔσται συλαγαγῶν διὰ τῆς φιλοσο-
           
φίας, take heed lest there shall be anyone that maketh spoil of you

through his philosophy. See also Acts 5:11; 10: 41; Gal. 1:7.

 


166                                         THE MOODS.

 

425. A neuter participle with the article is sometimes

equivalent to an abstract noun. It is then limited by a geni-

tive like any other abstract noun. HA. 966, b.; G.1562.

 

Phil. 3:8; διὰ τὸ ὑπερέχον τῆς γνώσες Χριστοῦ Ἰησοῦ τοῦ κυρίου
           
μου, because of the excellency [superiority] of the knowledge of Christ

Jesus my Lord. See also the similar use of neuter adjectives in

Rom.2:4; 1 Cor. 1:25; Phil. 4:5; Heb.6:17. WM.pp.294f.;

WT. pp. 234 f.

 

426. The Explanatory Attributive Participle. An

attributive Adjective Participle may be used to describe a

person or thing already known or identified. It is then

equivalent to an explanatory relative clause. Cf. 295.

2 Tim. 1:8, 9; κατὰ δύναμιν θεοῦ, τοῦ σώσαντος ἡμᾶς καὶ καλέσαντος
           
κλήσει ἁγίᾳ, according to the power of God; who saved us, and called

us with a holy calling.

1 Thess. 1:10;   Ἰησοῦν τὸν ῥυόμενον ἡμᾶς ἐκ τῆς ὀργῆς τῆ ἐρχομένης,

Jesus, which delivereth us from the wrath to come. In this example

ῥυόμενον is explanatory, ἐρχομένης is restrictive. See also Acts

20:32; Heb. 7:9.

 

427. An Attributive Participle when used to limit a noun

which has the article, stands in the so-called attributive posi-

tion, i.e. between the article and the noun, or after an article

following the noun; but when the participle is limited by an

adverbial phrase, this phrase may stand between the article

and the noun, and the participle without the article follow the

noun. It thus results that all the following orders are possi-

ble:

(1) article, partieiple, modifier of the participle, noun;

(2) art., mod., part., noun;

(3) art., mod., noun, part. ;

(4) art., part., noun, mod.;

(5) art., noun, art., mod., part.;


THE ADJECTIVE PARTICIPLE.                                    167

 

(6) art., noun, art., part., mod. See Professor Charles Short's

essay on The Order of Words in Attic Greek Prose, in Yonge's

English-Greek Lexicon, pp. xlix. f.; K. 464, 8; HA. 667, a.

 

Acts 13:39; καὶ ἡμεῖς ὑμᾶς εὐαγγελιζόμεθα τὴν πρὸς τοὺς πατέρας
           
ἐπαγγελίαν γενομένην, and we bring you good tidings of the promise

made unto the fathers. See also Acts 12: 10; 26: 4,6; Heb. 2: 2;

and especially Rom. 2:27, where ἐκ φύσεως ἀκροβυστία τὸν νόμον
 
τελοῦσα should doubtless be rendered, the uncircumcision which

by nature fulfils the law (cf. v. 14).

 

428. An Attributive Participle equivalent to a relative

clause, may like a relative clause convey a subsidiary idea of

cause, purpose, condition, or concession (cf. 294, 296 ff., 317

ff.). It then partakes of the nature of both the Adjective

Participle and the Adverbial Participle. Of. 434.

Rom. 3:5; μὴ ἄδικος θεὸς ἐπιφέρων τὴν ὀργήν,  is God unright-

eous, who [because he] visiteth with wrath?

Matt.  10:39; εὑρὼν τὴν ψυχὴν αὐτοῦ ἀπολέσει αὐτήν, καὶ ἀπολέ-
           
σας τὴν ψυχὴν αὐτοῦ ἕνεκεν ἐμοῦ εὑρήσει αὐτήν, he that findeth his

life shall lose it, and he that loseth his life for my sake shall find it.

See also vss. 37,40,41; cf. vss. 38, 42, and Luke 14: 26.

 

429. The Predicative Adjective Participle. A parti-

ciple may be used as the predicate of the verb el/.l.t or other

copulative verb.

Matt.  3:15;  οὕτω γὰρ πρέπον ἐστιν ἡμῖν πληρῶσαι πᾶσαν δικαιοσύνην,

for thus it is becoming for us to fulfil all righteousness.

Gal. 1:22;  ἤμην δὲ ἀγνοούμενος τῷ προσώπῳ ταῖς ἐκκλησίαις τῆς
           
Ἰουδαίας, and I was unknown by face unto the churches of Judea.

Rev. 1:18; καὶ ἐγενόμην νεκρὸς καὶ ἰδοὺ ζῶν εἰμὶ εἰς τοῦς αἰῶνας τῶν
           
αἰώνων,  and I became dead, and behold, I am alive for evermore.

 

430. The Predicative Participle always stands in the so-

called predicative position, i.e. not in attributive position,

which is between the article and its noun or after an article

following the noun. Cf. 427.


168                                          THE MOODS.

 

431. Under the head of the Predicative Participle belong

those Present and Perfect Participles which, with the Present,

Imperfect, and Future of the verb, form periphrastic Presents,

Imperfects, Perfects, Pluperfects, Futures, and Future Perfects.

Cf. 20, 34, 71,84, 91, 97; G.MT. 45, 830, 831; B. pp. 308-313;

S. pp. 131 ff. See Rev. 1 : 18; Matt. 27 : 33; Mark 2 : 18; Luke

1 : 21; 13: 10; J as. 5 : 15; 2 Cor. 9 : 3; Luke 2 : 26; John 13: 5 ;

Matt. 18: 18.

 

432. To the Greek mind there was doubtless a distinction

of thought between the participle which retained its adjective

force and its distinctness from the copula, and that which

was so joined with the copula as to be felt as an element

of a compound tense-form. This distinction can usually be

perceived by us; yet in the nature of the case there will

occur instances which it will be difficult to assign with cer-

tainty to one class or the other. Since, moreover, an Adjec-

tive Participle used substantively without the article may

stand in the predicate, this gives rise to a third possibility.

A. participle without the article standing in the predicate is

therefore capable of three explanations:

(a) It may be an Attributive Participle used substantively.

So probably

Mark 10:22; ἦν γὰρ ἔχων κτήματα πολλά, for he was one that had

            great possessions. See also John 18:30.

 

(b) It may be a Predicative Participle retaining its adjec-

tive force. So probably the examples under 429, especially

Gal. 1:22.

 

(c) It may form with the copula a periphrastic verb-form.

 

Luke 5:17;  καὶ ἐγένετο ἐν μιᾷ τῶν ἡμερῶν καὶ αὐτὸς ἦν διδάσκων,

and it came to pass on one of those days that he was teaching.


THE ADVERBIAL PARTICIPLE.                       169

 

433. An Adjective Participle used substantively with the

article may of course occur as a predicate with a copula. This,

however, is not properly a Predicative Participle. The presence

of the article makes its use as a noun easily evident. The

participle without the article may be as really substantive

(432, a), but is not so easily distinguished as such.

 

Luke 7:19;  σὺ εἶ ἐρχόμενος; art thou he that cometh?  See also

Luke 16 : 15 j 22: 28.

 

 

THE ADVERBIAL PARTICIPLE.

 

434. The Adverbial Participle logically modifies some

other verb of the sentence in which it stands, being equiva-

lent to an adverbial phrase or clause denoting time, condi-

tion, concession, cause, purpose, means, manner, or attend-

ant circumstance. HA. 969; G. 1563. Thus we find:

 

435. The Adverbial Participle of Time, equivalent to

a temporal clause.

 

Luke 24:36;  ταῦτα δὲ αὐτῶν λαλούντων αὐτὸς ἔστη ἐν μέσῳ αὐτῶν,

and as they spake these things, he himself stood in the midst of them.

John 16:8; καὶ ἐλθὼν ἐκεῖνος ἐλέγξει τὸν κόσμον, and he, when he is

come, will convict the world.

 

436. The Adverbial Participle of Condition, equiva-

lent to a conditional clause.

Heb. 2:3; πῶς ἡμεῖς ἐκφευξόμεθα τηλικαύτης ἀμελήσαντες σωτηρίας,

how shall we escape, if we neglect so great salvation?

1Tim. 4:4; ὅτι πᾶν κτίσμα θεοῦ καλὸν, καὶ οὐδὲν ἀπόβλητον μετὰ
           
εὐχαριστίας λαμβανόμενον, for every creature of God is good, and

nothing is to be rejected, if it be received with tltanksgiving. See also

1 Cor. 11:29; Gal. 6:9.

 


170                                          THE MOODS.

 

437. The Adverbial Participle of Concession, equiva-

lent to a concessive clause. The concessive force is some-

times emphasized by prefixing καὶπερ or καί γε to the

participle.

Acts 13:28;  καὶ μηδεμίαν αἰτίαν θανάτου εὑρόντες ᾐτήσαντο Πειλάτον
           
ἀναιρεθῆναι αὐτόν, and though they found no cause of death in him,

yet asked they of Pilate that he should be slain.

Heb 5:8; καίπερ ὢν υἱός, ἔμαθεν ἀφ' ὧν ἔπαθεν τῆν ὑπακοήν,

he was a Son, yet he learned obedience by the things which he suffered.

See also Matt. 14:9; Mark 4:31; Acts 17:27.

 

438. A concessive participle refers to α fact which is

unfavorable to the occurrence of the event denoted by the

principal verb. Cf. 278. It should be distinguished from the

participle which is merely antithetical. A participle denoting

accompanying circumstance, or even condition or cause, may

be antithetical. See 1 Cor. 4:12, διωκόμεμνοι ἀνεχόμεθα; 2 Cor.

8:9; Gal. 2:3.

 

439. The Adverbial Participle of Cause, equivalent

to a causal clause.

 

Col. 1:3, 4; εὐχαριστοῦμεν τῷ θεῷ . . . ἀκούσαντες τὴν πίστιν ὑμῶν ἐν
           
Χριστῷ Ἰησοῦ, we give thanks to God. . . having heard (because

we have heard) of your faith in Christ Jesus.

1 Tim. 4:8;  δὲ εὐσέβεια πρὸς πάντα ὠφέλιμός ἐστιν, ἐπαγγελίαν
           
ἔχουσα ζωῆς τῆς νῦν καὶ τῆς μελλούσης, but godliness is profitable

for all things, having promise of the life which now is, and of that which

is to come. See also Matt. 2:3, 10; Acts 9:26.

 

440.   Ὡς prefixed to a Participle of Cause implies that the

action denoted by the participle is supposed; asserted, or pro-

fessed by some one, usually the subject of the principal verb,

to be the cause of the action of the principal verb. The

speaker does not say whether the supposed or alleged cause

actually exists. HA. 978; G. 1574.


THE ADVERBIAL PARTICIPLE.                       171

 

1 Cor. 4:18;  ὡς μὴ ἐρχομένου δέ μου πρὸς ὑμᾶς ἐφυσιώθησάν τινες,

but some are puffed up, as though I were not coming to you, i.e. be-

cause (as they suppose) I am not coming. See also Acts 23: 15,

20; 27:30; 28:19; 1 Pet. 4:12.

 

441. The origin of this idiom is probably in a clause of manner con-

sisting of ὡς and a finite verb, the latter modified by a Participle of Cause.

Thus κολάζεις με ὥς κακοποιήσαντα, you punish me as having done evil, i.e.

you punish me because, as you allege, I have done evil, may have its

origin in such a sentence as κολάζεις με ὡς κολάζεις τινὰ κακοποιήσαντα, you punish me, as you punish one who has (or because he has) done evil. Yet

it is not to be supposed that the Greek any more than the English required

the supplying of a finite verb after ὡς. Such phrases in classical Greek

or in the New Testament are, as they stand and without the addition of

other words, expressions of cause, the use of ὡς indicating that the phrase

describes the opinion or assertion of the snbject of the sentence rather

than of the speaker.

 

442. The Adverbial Participle of Purpose, equivalent

to a final clause. This is usually, but not invariably, in

the Future tense.

 

Acts 8:27;  [|ος] ἐληλύθει προσκυνήσων εἰς Ἰερουσαλήμ, who had come

            to Jerusalem to worship.

Acts 3:26;  ἀπέστειλεν αὐτὸν εὐλογοῦντα ὑμᾶς, he sent him to bless you.

 

443. The Adverbial Participle of Means. This can-

not usually be resolved into a clause.

Matt. 6:27;  τίς δὲ ὑμῶν μεριμνῶν δύναται προσθεῖναι ἐπὶ τὴν
           
ἡλικίαν αὐτοῦ πῆχυν ἕνα, and which of you by being anxious can

add one cubit unto his stature.f See also Acts 16 : 16 ; Heb. 2 : 18.

 

444. The Adverbial Participle of Manner, describing

the manner in which the action denoted by the verb is

done.

Acts 2:13; ἕτεροι δὲ διαχλευάζοντες ἕλεγον, but others mocking said.

See also Luke 19:48.


172                                          THE MOODS.

 

445. The manner of an action is frequently expressed by

ὡς with the participle.

 

Mark 1:22; ἦν γὰρ διδάσκων αὐτοὺς ὡς ἐξουσίαν ἔχων καὶ οὐχ ὡς οἱ
           
γραμματεῖς, for he taught them as one having authority, and not as the

scribes.

1 Cor. 9:26; οὕτως πυκτεύω ὡς οὐκ ἀέρα δέρων, so fight I as not beat-

ing the air.

2 Cor. 5:20; ὑπὲρ Χρθστιῦ οὖν πρεσβεῦομεν ὡς τοῦ θεοῦ παρακαλοῦν-
           
τος δι' ἡμῶν, we are ambassadors therefore on behalf of Christ, as

though God were intreating by us.

 

446. When ὡς with the participle is used to express manner, the parti-

ciple itself may be either an Adjective Participle used substantively or an

Adverbial Participle of Manner. The origin of such expressions is doubt-

less, in either case, in a clause of manner consisting of ὡς and a finite verb

similar to the principal verb, the participle being either the subject of

such a verb or an adverbial (or other) limitation of it. Thus διδάσκει ὡς

ἔχων ἐξουσίαν is equivalent to διδάσκει ὡς ἔχων ἐξοουσίαν διδάσκει, he

teaches as one having authority teaches, or διδάσκει ὣς τις διδάσκει ἔχων ἐξουσίαν, he teaches as one teaches hamng authority. Yet m neither case is to be

supposed that the Greek, any more than the English, required the sup-

plying of a finite verb after ὡς. The phrase as it stood was an expres-

sion of manner. That the participle, however, was in some cases still

felt as a substantive (Adjective Participle used substantively) seems

probable from its being used correlatively with an adjective or noun and

from the occasional use of the participle with the article. See Mark 1:22

above; also 1 Cor. 7:25; 2 Cor. 6:9, 10; 1 Pet. 2:16; and cf. Mark

6:34; Luke 22:26, 27. That this is not always the case, but that the

participle itself is sometimes adverbial is evident from such cases as 2 Cor.

5:20 (see above, 445).

 

447. The participle expressing manner or means often

denotes the same action as that of the principal verb, describ-

ing it from a different point of view. In this case the participle

is as respects its tense a (Present or Aorist) Participle of

Identical Action (cf. 120, 139), while as respects its modal

function it is a participle of manner or means.


THE ADVERBIAL PARTICIPLE.                                   173

 

Acts 5:30;  θεὸς τῶν πατέρων ἡμῶν ἤγειρεν Ἰησοῦν, ὃν ὑμεῖς
           
διεχειρίσασθε κρεμάσαντες ἐπὶ ξύλου, the God of our fathers raised

up Jesus, whom ye slew by hanging him on a tree. See also Acts

9:22; 10:33; 1 Tim. 5:21.

 

448. In quotations from the Old Testament a participle is

sometimes placed before a personal form of the same verb.

The idiom arises from an imitation of the Hebrew construction

with the Infinitive Absolute. The force of the participle is

in general intensive. Hr. 28, 3, a; B. pp. 313 f.; WM. pp.

445 f.; WT. pp. 354 f.

Heb. 6:14; εὐλογῶν εὐλογήσω σε καὶ πληθύνων πληθυνῶ σε, blessing

I will bless thee, and multiplying I will multiply thee.

 

449. The Adverbial Participle of Attendant Circum-

stance.

Mark 16:20; ἐκεῖνοι δὲ ε'ξελθποντες ἐκήρυξαν πανταχοῦ, τοῦ κυρίου
           
συνεργοῦντος καὶ τὸν βεβαιοῦντος, and they went forth and

preached everywhere, the Lord working with them and confirming the

word.

Luke 4:15; καὶ αὐτὸς ἐδίδασκεν ἐν ταῖς συναγωγαῖς αὐτῶν, δοξαζό-
           
μενος ὑπὸ πάντων, and he taught in their synagogues, being glorified

of all.

Acts 15:22;  τότε ἐδοξε τοῖς ἀποστόλοις καὶ πρεσβυτέροις . . .
           
ἐκλεξαμένους ἄνδρας ἐξ αὐτῶν πέμψαι εἰς Ἀντιόχειαν, then it

seemed good to the apostles and the elders. ..to choose men out of their

company and send them to Antioch.

Acts 18:18;  δὲ Παῦλος . . . ἐξέπλει εὶς τὴν Συρίαν, καὶ σύν α'τῷ
           
Πρίσκιλλα καὶ Ἀκύλας, δειράμενος ἐν Κενχρεαῖς τὴν κεφαλήν

and Paul. . . sailed thence for Syria, and with him Priscilla and Aquila;

having shorn his head in Cenchrere.

2 Tim. 4:11; Μάρκον ἀναλαβῶν ἄγε μετὰ σεαυτοῦ, take Mark and

            bring him with thee. See also Luke 5:7; 11:7.

 

450. The term "attendant" as used above does not define the tem-

poral relation of the participle to the verb, but the logical relation. The

action of a Participle of Attendant Circumstance may precede the action


174                                          THE MOODS.

 

of the principal verb, accompany it, or even follow it. But as respects

logical relation, it is presented merely a.s an accompa,niment of the action

of the verb. It does not, e.g., define the time or the cause, or the means

of the action of the prinicipal verb, but simply prefixes or adds an asso-

ciated fact or conception. It is thus often equivalent to a coordinate verb

with καί. Though grammatically not an independent element of the

sentence, the participle in such cases becomes in thought. assertive,

hortatory, optative, imperative, etc., according to the function of the

principal verb.

The position of the Participle of Attendant Circumstance with refer-

ence to the verb is not detennined by any fixed rules, but by the order

of the writer's thought, this being in turn governed of course to a certain

extent by the order of the events. If the action of the participle is ante-

cedent to that of the verb, the participle most commonly precedes the verb,

but not invariably. Such a participle is usually in the Aorist tense (134),

but occasionally in the Present (127). If the action of the participle is

simultaneous with that of the verb, it may either p1'ecede or follow the

verb., more frequently the latter. It is of course in the Present tense

(119). If the action of the participle is subsequent to that of the priJici-

pal verb, it almost invariably follows the verb, the tense of the participle

being detennined by the conception of the action as respects jts progress.

The instances of this last-named class are not frequent in the New Testa-

ment and are perhaps due to Aramaic influence. Cf. 119, Rem.; 146.

 

451. The various relations of time, cause,manner, etc., being

not expressed, but implied by the participle, cases arise in

which it is impossible to assign the participle unquestionably

to anyone of the above heads. Indeed, more than one of these

relations may be implied by the same participle.

 

452. THE GENITIVE ABSOLUTE. An Adverbial Participle

may stand in agreement with a noun or pronoun in the geni-

tive without grammatical dependence upon any other part of

the sentence, the two constituting a genitive absolute phrase

and expressing any of the adverbial relations enumerated in

435-449. HA. 970, 971; G. 1568.

 

Rom. 9:1; ἀλήθειαν λέγω ἐν Χριστῷ, οὐ ψεύδομαι, συνμαρτυρούσης
           
μοι τῆς συνειδήσεώς μου ἐν πνεύματι ἁγίῳ, I say the truth in Christ,

I lie not, my conscience bearing witness with me in the Holy Ghost.

See also John 8: 30; Acts 12: 18; 18: 20.


THE SUBSTANTIVE PARTICIPLE.                              175

 

453. The noun or pronoun of the genitive absolute phrase regularly

refers to a person or thing not otherwise mentioned in the sentence.

Occasionally, however, this principle is violated, and the genitive phrase

may even refer to the subject of the sentence. This irregularity is some-

what more frequent in the New Testament than in classical Greek. HA.

972, d.; G .MT. 850. See Matt. 1:18; Acts 22:17, and other examples in

B. pp. 315 f.

454. A participle in the genitive absolute occasionally stands alone

without an accompanying noun or pronoun, when the person or thing

referred to is easily perceived from the context. HA.. 972, a.; G. 1568;

G.MT.848. See Luke 12: 36; Rom. 9: 11.

455. The Adverbial Participle always stands in the so-called

predicative position, i.e. not in attributive position, which is

between the article and its noun or after an article following

the noun. Of. 427.

 

 

THE SUBSTANTIVE PARTICIPLE.

 

456. The Substantive Participle is employed as itself

the name of an action. It thus performs a function which

is more commonly discharged by the Infinitive. HA. 980-

984; G.1578-1593.

457. The Substantive Participle as Subject. The

Substantive Participle may be used as an integral part of

the subject of a verb, the action which it denotes being

itself an essential part of that of which the predicate is

affirmed.

Matt. 6:16; ὅπως φανῶσιν τοῖς ἀνθρώποις νηστεύοντες, that they may

be seen of men to fast. (Not only they, but their fasting, is to be seen.)

Acts 5:2; οὐκ ἐπαύοντο διδάσκοντες καὶ εὐαγγελιζόμενοι τὸν χριστὸν
           
Ἰησοῦν, they ceased not teaching and preaching Jesus as the Christ.

See also Matt. 1:18, ἔχουσα; Mark 6:2, γινόμεναι; Luke 5:4,

λαλών.


176                                          THE MOODS.

 

458. The Substantive Participle as Object. The Sub-

stantive Participle may be used as an integral part of the

object of a transitive verb. This occurs especially after

verbs of perception, the action denoted by the participle

being itself that which one perceives.

Luke 8:46; ἐγὼ γὰρ ἔγνων δύναμιν ἐξεληλυθυῖαν ἀπ' ἐμοῦ, for I per-

            ceived power to have gone out of me.

John 7:32; ἥκουσαν οἰ Φαρισαῖοι τοῦ ὄχλου γογγύζοντος, the Pharisees

             heard the multitude murmuring.

 

            459. With verbs of finishing, ceasing, etc., the Substantive

Participle agrees grammatically with the subject of the verb.

Since, however, certain of these verbs are transitive, the

action denoted by the participle must in these cases be re-

garded as logically the object of the verb.

 

Matt. 11:1; ὅτε ἐτέλεσεν Ἰησοῦς διατάσσων τοῖς δώδεκα μαθηταῖς
           
αὐτοῦ, when Jesus had finished commanding his twelve disciples. Cf.

            Matt. 13:53; see also Luke 7:45.

 

            460. THE SUBSTANTIVE PARTICIPLE IN INDIRECT DISCOURSE.

A Substantive Participle forming a part of the object of a verb

is sometimes equivalent to a clause of inqirect discourse.

 

1 John 4:2; πᾶν πνεῦμα ὁμολογεῖ  Ἰησοπυν Χριστὸν ἐν σαρκὶ ἐλη-
           
λυθότα ἐκ τοῦ θεοῦ ἐστίν, every spirit which confesseth that Jesus

            Christ has come in the flesh is of God. See also Luke 4:23;

            Acts 7:12; 8:23; 3 John 4.

 

            461. The Substantive Participle as a Limiting Gen-

itive.  The Substantive Participle may be used as an

integral part of a genitive limiting phrase.

 

John 4:39;  πολλοὶ ἐπίστευσαν εἰς αὐτὸν τῶν Σαμαρειτῶν διὰ τὸν λόγον
           
τῆς γθναικὸς μαρτυρούσης, many of the Samaritans believed on him

            because of the word of the woman testifying, i.e. of the woman's testi-

            mony. See also Heb. 8: 9; and cf. Jos. Ant. 10. 4. 2, where a Sub-

            stantive Participle occurs after a preposition.


                        THE SUBSTANTIVE PARTICIPLE.                  177

 

            462. The Substantive Participle, like the Adverbial Parti-

ciple, always stands in the so-called predicative position. Of.

455, and 427.

            463. The Substantive Participle must be carefully distin-

guished from the Adjective Participle used substantively.

The latter designates the doer of an action, the former the

action itself. "In the one it is the adjective force of the word

which is substantivized, in the other, the verbal force." See

Stevens, u.s., 419, Rem. 1.


 

 

 

                        THE USE OF NEGATIVES WITH VERBS.

 

 

            464. In the use of the simple negatives οὐ and μή and

their compounds, οὐδί, οὐδείς, οὔτε, οὐτε, οὐκέτι, μηδέ, μηδείς, μήτε,
μηκέτι, etc., as also of the double negatives οὐ μή and μὴ οὐ,

New Testament Greek conforms in the main to classical

usage, yet exhibits several important variations. The fol-

lowing sections exhibit the essential features of New Testa-

ment usage in comparison with that of classical writers;

rarer .and. more delicate classical usages which have no

analogies In New Testament usage are not mentIoned; state-

ments which are not restricted to classical or New Testament

usage are to be understood as referrIng to both. What IS

said respecting the simple negatives οὐ and μή applies in

general also to their respective compounds when standing

alone.

 

                        NEGATIVES WITH THE INDICATIVE.

 

            465. The Indicative in an independent declaratory sentence

regularly takes ou as its negative. HA. 1020; G. 1608.

John 1:11; εἰς τὰ ἴδια ἦλθεν, καὶ οἱ ἴδιοι αὐτὸν οὐ παρέλαβον, he

            came unto his own, and they that were his own received him not.

 

            REM. On the use of negatives in later Greek, see Gild., Encroach-

ments of μή on οὐ in later Greek, in A.J: P. I. pp. 45ff.

 

                                                            178


                        NEGATIVES WITH THE INDICATIVE.                       179

 

            466. In classical Greek, the Future Indicative used to ex-

press a prohibition sometimes has οὐ, sometimes μή. HA.

844; G.MT. 69, 70.

            In the New Testament a Prohibitory Future takes οὐ.

 

Matt. 6:5;  καὶ ὅταν προσεύχησθε, οὐκ ἔσεσθε ὡς οἱ ὑποκριταί, and

            when ye pray, ye shall not be as the hypocrites.

 

            467. In questions that can be answered affirmatively or

negatively, οὐ is used with the Indicative to imply that an

affirmative answer is expected; μή to imply that a negative

answer is expected. HA. 1015; G. 1603.

 

Matt. 13:55; εὐχ οὗτός ἐστιν τοῦ τέκτονος υἱός, is not this the car-

            penter's son?

John 7:51; μὴ νόμος ἡμῶν κρίνει τὸν ἄνθρωπον ἐὰν μὴ ἀκούσῃ
           
πρῶτον παρ' αύτοῦ, doth our law judge a man, except it first hear from

            himself?

 

            468. In Rom. 10:18, 19; 1 Cor. 9:4, 5; 11:22, μὴ οὐ is

used in rhetorical questions equivalent to affirmative state-

ments. Each negative has, however, its own proper force,

οὐ making the verb negative, and μή implying that a negative

answer is expected to the question thus made negative.

 

            469. In classical Greek, the Indicative in conditional and

conditional relative clauses is regularly negatived by μή. But

οὐ sometimes occurs in conditions of the first class. In this

case ou] negatives the verb of the clause or other single element

rather than the supposition as such. HA.1021; G.1610, 1383.

            In the New Testament, conditional clauses of the second

class (248) are regularly negatived by μή. In other conditional

clauses and in conditional relative clauses, the Indicative

usually takes ov as its negative, occasionally μή. In con-

cessive clauses the Indicative takes οὐ.


180                             THE USE OF NEGATIVES WITH VERBS

 

John 9:33;  εἰ μἠ ἦν οὗτος παρὰ θεοῦ, οὐκ ἠδύνατο ποιεῖν οὐδέν, if this

            man were not from God, he could do nothing. See also Matt. 24:22.

Rom. 8:9;  εἰ δέ τις πνεῦμα Χριστοπυ οὐκ ἔχει, οὗτος οὐκ ἔστιν αὐτοῦ,

            but if any man hath not the Spirit of Christ, he is none of his. See

            also Luke 14: 26.

Matt. 10:38; καὶ ὃς οὐ λαμβάνει τὸν σταυρὸν αὐτοπυ καὶ ἀκολουθεῖ
           
ὀπίσω μου, οὐκ ἔστιν μου ἄξιος, and he that does not take his cross

            and follow after me, is not worthy of me. See also Luke 9:50;

            14:33; cf. 2 Pet. 1:9; 1 John 4:3.

Luke 18:4, 5; εἰ καὶ τὸν θεὸν οὐ φοβοῦμαι οὐδὲ ἄνθρωπον ἐντρέπομαι,
           
διά γε τὸ παρέχειν μοι κόπον τὴν χήραν ταύτην ἐκδικήσω αὐτήν,

            though I fear not God nor regard man, yet because this widow

            troubleth me, I will avenge her.

2 Cor. 13:5; οὐκ ἐπιγινώσκετε ἑαυτοὺς ὅτι Ἰησοῦς Χριστὸς ἐν
           
ὑμῖν; εἰ μήτι ἀδόκιμοί ἐστε, or now ye not as to your own selves that

            Jesus Christ is in you?  unless indeed ye are reprobate. See also

            1 Tim. 6:3; Tit. 1:6.

 

            REM. In Matt. 26:24; Mark 14:21, οὐ occurs in the protasis of a

conditional sentence of the second class.

 

            470. It is possible that ou in conditional and conditional relative

sentences in the New Testament is usually to be explained as negativing

the predicate directly (cf. G. 1383. 2; Th. ei], III. 11.), μή on the other

hand as negativing the supposition as such. Yet the evidence does not

clearly establish this distinction; to press it in every case is certainly

an over-refinement. Cf., e,g., 1 John 4:3, πᾶν πνεῦμα μὴ ὁμολογεῖ τὸν
Ἰησοῦν ἐκ τοῦ θεοῦ οὐκ ἔστιν, and 1 John 4:6, ὃς οὐκ ἔστιν ἐκ τοῦ θεοῦ οὐκ
 
ἀκούει ¨ημῶν. See also 1 Tim. 6:3 and Tit. 1:6, where μή is used after εἰ, yet

quite evidently belongs to the verb rather than to the supposition as such.

 

            471. Εἰ μή in the sense of except is used as a fixed phrase,

without reference to the mood which would follow it if the

ellipsis were supplied. Of. 274.

 

Matt. 17:8;  οὐδένα εἶδον εἰ μἠ αὐτὸν Ιησοῦν μόνον, they saw no one

            save Jesus only.

Mark 9:9; διεστείλατο αὐτοῖς ἵνα μηδενὶ εἶδον διηγήσωνται, εἰ μὴ
           
ὅταν υἱὸς τοῦ ἀνθρώπου ἐκ νεκρῶν ἀναστῇ, he charged them that

            they should tell no man what things they had seen, save when the Son

            of man should have arisen from the dead.


            SUBJUNCTIVE, OPTATIVE, AND IMPERATIVE.                181

 

            472. In clauses introduced by μή as a conjunction, the

Indicative takes oil as its negative. After other final particles

its negative is μή. HA. 1021, 1033; G. 1610.

 

Rev. 9:4; καὶ ἐρρέθη αὐταῖς ἵνα μὴ ἀδικήσουσιν τὸν χόρτον τῆς γῆς,

            and it was said unto them that they should not hurt the grass of the

            earth. The continuation of this sentence by ou8l ...ou8l is a

            syntactical irregularity. Col. 2:8 illustrates the rule.

 

            473. In indirect discourse the negative of the direct form

is retained. HA. 1022; G.1608.

 

Matt. 16:11;  πῶς οὐ νοεῖτε ὅτι οὐ περὶ ἄρτων εἶπον ὑμῖν,

            that ye do not perceive that I spake not to you concerning bread,

 

            REM. In 1 John 2:22 a clause of indirect discourse depending on a

verb meaning to deny contains a redundant οὐ. Cf. 482, and B. p. 355.

 

            474. In causal clauses, and in simple relative clauses not

expressing purpose or condition, the Indicative is regularly

negatived by οὐ. HA. 1021; G. 1608.

 

John 8:20;  καὶ οὐδεὶς ἐπίασεν αὐτόν, ὅτι οὔπω ἐληλύθει ὥρα αὐτοῦ,

            and no man took him; because his hour was not yet conte.

Mark 2:24;  ἴδε τί ποιοῦσιν τοῖς σάββασιν οὐκ ἔξεστιν, behold, why

            do they on the sabbath day that which is not lawful?

 

            REM. 1. In John 3:18 a causal clause has an Indicative with μή. This

is quite exceptional in the New Testament, but similar instances occur in

later Greek. B. p. 349, Gild. u.s. p. 53.

 

            REM. 2. Tit. 1:11, διδάσκοντες μή δεῖ  is an exception to the general

rule for relative clauses, unless indeed the relative clause is to be taken

as conditional. Cf. 469.

 

            NEGATIVES WITH THE SUBJUNCTIVE, OPTATIVE, AND

                                                IMPERATIVE.

 

            475. The negative of the Subjunctive both in principal and

in subordinate clauses is μή, except in clauses introduced by


182                  THE USE OF NEGATIVES WITH VERBS.

 

the conjunction μή, lest. In these the negative is οὐ. Con-

cerning οὐ μή with the Subjunctive see 487, 488. HA. 1019,

1033 ; G. 1610.

 

1 John 3:18; μὴ ἀγαπῶμεν λόγῳ, let us not love in word.

Heb. 4:7; μὴ σκληρύνητε τὰς καρδίας ὑμῶν, harden not your hearts.

2 Cor. 12:20; φοβοῦμαι γὰρ μή πως ἐλθὼν οὐχ οἵοθς θέλω εὕρω
           
ὑμᾶς, for I fear, lest by any means, when I come, I should find you

            not such as 1 would. See also Acts 20: 16; Rom. 10: 15; 1 Cor.

            2:5.

 

            REM. 1. In Matt. 25:9 a Subjunctive after the conjunction μή is

negatived by ou] (WH. margin), or, according to other MSS., followed by

WH. (text) by the strong negative οὐ μή.

 

            REM. 2. In Rom. 5:11 οὐ μόνον limits a verb understood which is

probably to be taken as a Subjunctive. Of. 479, 481.

 

            476. In classical Greek, ou] is used with the Potential

Optative; μή with the Optative of Wishing. In the New

Testament, no instance of a negatived Potential Optative

occurs. With the Optative of Wishing μή is used as in

classical Greek. HA.l020; G. 1608.

 

Gal. 6:14;  ἐμοὶ δὲ μὴ γένοιτο καυχᾶσθαι, but far be it from me to glory.

            See also Mark 11:14; Rom. 3:4, 6, 31, etc.

 

            477. In classical Greek, the Optative in subordinate clauses

takes μή as its negative except in indirect discourse and after

μή, lest. HA. 1021, 1022; G. 1610.

            In the New Testament, no instance of a negatived Optative

in a subordinate clause occurs.

 

            478. The negative of the Imperative is μή. HA. 1019;

G. 1610.

            This rule holds in the New Testament with very rare

exceptions.

 

Luke 12:11; μὴ μεριμνήσητε πῶς [ τι] ἀπολογήσησθε, be not anxious

            how or what ye shall answer. See also under 165.


                        INFINITIVE AND PARTICIPLE.                                    183

 

            479. Of the apparent exceptions to the rule stated above (478), some

are to be explained as parenthetic non-imperative phrases in the midst of

imperative sentences. So, clearly, in 1 Cor. 5: 10, [I meant] not [that

you should have no company] at all, with the fornicators of this world, etc.

So also 2 Tim. 2: 14, that they strive not about words, [a thing which is]

profitable for nothing. The use of ou]x rather than μή in 1 Pet. 3:3 seems

to indicate that the following words, . . . κόσμος, are excluded from the

injunction rather than included in a prohibition. In 1 Pet. 2:18 οὐ μόνον

occurs, perhaps as a fixed phrase, after a participle with Imperative of the

verb εἶναι understood. On the other hand, it is noticeable that elsewhere

limitations of the Imperative when negatived regularly take μή. Thus μή

μόνον occurs in John 13:9; Phil. 2:12; Jas. 1:22. Of. 481.

 

 

NEGATIVES WITH THE INFINITIVE AND PARTICIPLE.

 

            480. In classical Greek, the Infinitive usually takes ou] as

its negative in indirect discourse; elsewhere μή. HA. 1023,

1024; G. 1611; but see also Gild. u.s. (465, Rem.) pp. 48 ff.

on the use of μή with the Infinitive in indirect discourse.

            In the New Testament, the Infinitive regularly takes μή

as its negative in all constructions.

Matt. 22:23;  λέγοντες μὴ εἶναι ἀνάστασιν, saying that there is no

            resurrection.

Luke 11:42; ταῦτα δὲ ἔδει ποιῆσαι κἀκεῖνα μὴ παρεῖναι, but these ought

            ye to have done, and not to leave the other undone.

 

            481. When a limitation of an Infinitive or of its subject

is to be negatived rather than the Infinitive itself, the negative

ou] is sometimes used instead of μή. See Rom. 7:6; 1 Cor.

1:17; Heb. 7:11; 13:9. This principle applies especially

in the case of the adverb μόνον. In the New'Testament at

least, οὐ μόνον rather than μή μόνον occurs regularly with the

Infinitive, and this both when the phrase as a whole belongs

to the Infinitive itself, and when it applies rather to some

limitation of the Infinitive. See John 11:52; Acts 21:13;

26:29; 27:10; Rom. 4:12, 16; 13:5; 2 Cor. 8:10; Phil.


184                  THE USE OF NEGATIVES WITH VERBS.

 

1:29; 1 Thess. 2:8.  Μή μόνον is found with the Infinitive

only in Gal. 4:18. It is perhaps as a fixed phrase, unaffected

by the Infinitive, that εἰς οὐθέν limits λογισθῆναι in Acts 19:27.

 

            482. A compound of οὐ may occur with an Infinitive depend-

ent on a principal verb limited by οὐ, in accordance with the

principle of 488.

 

John 5:30;  οὐ δύναμαι ἐγὼ ποιεῖν ἀπ' ἐμαυτοῦ οὐδέν, I can of myself

            do nothing. See also Mark 7:12; Luke 20:40; John 3:27, etc.

            Probably Acts 26 : 26 should be translated, I am not persuaded (i.e.

            I cannot believe) that any of these things was hidden from him. B. p.

            350.

 

            483. The Infinitive after verbs of hindering, denying, etc.,

may take μή without change of meaning. Such a negative

cannot be translated into English. HA. 1029; G. 1615.

 

Acts 14:18; καὶ ταῦτα λέγοντες μόλις κατέπαυσαν τοὺς ὄχλους τοῦ
           
μὴ θύειν αὐτοῖς, and with these sayings scarce restrained they the

            multitudes from doing sacrifice unto them. See also under 402.

 

            484. In classical Greek, an Infinitive which would regularly

take μή, usually takes μὴ οὐ when it depends on a verb which

is itself negatived by οὐ. HA. 1034; G.1616.

            In the New Testament, the simple negative μή is retained

in such a case.

 

Acts 4:20;  οὐ δυνάμεθα γὰρ ἡμεῖς ειδαμεν καὶ ἠκούσαμεν μὴ λαλεῖν,

            for we cannot but speak the things which we saw and heard.

 

            485. In classical Greek, the participle takes μή if it is

equivalent to a conditional, or conditional relative clause;

otherwise it takes οὐ. HA. 1025; G. 1612; Gild. u.s. (465,

Rem.) pp. 55 ff.

            In the New Testament, participles in all. relations usually

take μή as the negative. But participles not conditional in


            SUCCESSIVE AND DOUBLE NEGATIVES.              185

 

force occasionally take οὐ, there being in all some seventeen

instances in the New Testament.

 

Acts 13:28;  καὶ μηδεμίαν αἰτίαν θανάτου εὑρόντες ᾐτήσαντο Πειλᾶτον
           
ἀναιρεθ[ῆναι αὐτόν, and though they found no cause of death in him,

            yet asked they of Pilate that he should be slain.

Luke 12:33; ποιήσατε ἑαυτοῖς βαλλάντία μὴ παλαιούμενα, make for

            yourself purses which wax not old.

John 5:23; μὴ τιμῶν τὸν υἱὸν οὐ τιμᾷ τὸν πατέρα, he that honoreth

            not the Son honoreth not the Father.

Matt. 22:11; εἶδεν ἐκ εῖ ἄνθρωπον οὐκ ἐνδεδυμένον ἔνδυμα γάμου, he

            saw there a man which had not on a wedding-garment.

Acts 17:6;  μὴ εὑρόντες δὲ αὐτοὺς ἔσυρον Ἰάσονα καί τινας ἀδελφοὺς ἐπὶ
           
τοὺς πολιτάρχας, and not finding them they dragged Jason and certain

            brethren before the rulers of the city. See also Matt. 22:29; Luke

            6:42; 9:33; John 10:12; Acts 7:5; 13:28; 26:22; Gal. 4:8.

 

 

                        SUCCESSIVE AND DOUBLE NEGATIVES.

 

            486. When two simple negatives not constituting a double

negative, or a compound negative followed by a simple nega-

tive, occur in the same clause, each has its own independent

force. The same is also true of course when the negatives

occur in successive clauses. HA. 1031; G. 1618.

1Cor 12:15;  οὐ παρὰ τοῦτο οὐκ ἔστιν ἐκ τοῦ σώματος, it is not

            therefore not of. the body. See also 1 John 3: 10; 5: 12.

Matt. 10:26; οὐδὲν γάρ ἐστιν κεκαλυμμένον οὐκ ἀποκαλυφθήσεται,

            for there is nothing covered, that shall not be revealed. See also

            1 John 2:21.

 

            REM. Concerning μή οὐ in questions, see 468.

 

            487. The double negative οὐ μή is used with the Subjunc-

tive, and more rarely with the Future Indicative, in emphatic

negative assertions referring to the future. Cf. 172, 66. HA.

1032; G. 1360, 1361.

 

 


186                  THE USE OF NEQATlVES WITH VERBS.

 

Mark 13:2; οὐ μὴ ἀφεθῇ ὧδε λίθος ἐπὶ λίθον ὃς οὐ μὴ καταλυθῇ,

            there shall not be left here one stone upon another, which shall not be

            thrown down.

Rev. 2:11;  νικῶν οὐ μὴ ἀδικηθῇ ἐκ τοῦ θανάτου τοῦ δευτέρου,

            that overcometh shall not be hurt of the second death.

Rev. 7:16;  οὐ πεινάσουσιν ἔτι οὐδὲ διψήσοθσιν ἔτι, οὐδὲ μὴ πέσῃ ἐπ'
           
αὐτοὺς ἥλιος, they shall hunger no more, neither thirst any more,

            neither shall the sun strike upon them at all. On Matt. 25: 9 see

            475, Rem. 1.

 

            488. Οὐ μή is occasionally used with the Subjunctive or

Future Indicative expressing a prohibition. Cf. 167, 67,

Rem. 2.

 

            489. When a negative is followed by one or more similar

compound negatives or by the double negative οὐ μή the effect

is a strengthened negation. HA. 1030; G. 1619.

 

Luke 23:53; οὗ οὐκ ἦν οὐδεὶς οὔπω κείμενος, where never man had yet

            lain.

Heb. 13:5; οὐ μή σε ἀνῶ οὐδ'οὐ μή σε ἐγκαταλίπω, I will in no wise

            fail thee, neither will I in any wise forsake thee.

Rom. 13:8; μηδενὶ μηδὲν ὀφείλετε, owe no man anything.


 

                                                INDEX OF SUBJECTS.

 

                                    [The Numbers refer to Sections.]

 

Adjective Participle, 420-433, see                166, 167; Optative in indirect dis-

   under Participle.                                           course, 110, 111; Imperative in

Adjectives of ability, fitness, readi-              commands and prohibitions, 163,

   ness, etc., followed by clause with            184; Infinitive after prepositions,

   ἵνα, 216; followed by Infinitive,                104-109; Infinitive after verbs

   376; followed by elf with the ar-               signifying to hope, etc., 113; In-

   ticular In:finitive, 413.                                 finitive in indirect discourse, 110,

Adverbial Participle, 434-455, see                114.

   under Participle.                                               Participle: properly expresses

Adverbs, limited by Infinitive, 3.76.           not time but action conceived of

Aoristic Present, 13; Aoristic Fut-                as a simple event, 132, 133; used

   ure, 59, 62; Aoristic Perfect, 80.                of antecedent action, 134-138; of

Aorist: constant characteristic, 35;               identical action, 139-141; of sub-

   Indefinite, Inceptive, Resultative,              sequent action, 142-145; as inte-

   35, 37.                                                            gral part of the object of a verb

   Indicative: Historical, 38; mo-                   of perception, 146; with lanqa<nw,

   mentary, comprehensive, collec-               147; leaving time-relation unde-

   tive, 39, 40, 54; Inceptive, 41;                    fined, 148; denoting action in

   Resultative, 42, 87; Gnomic, 43;               general simultaneous with that

   Epistolary, 44; Dramatic, 45; for               of principal verb, 149; with the

   English Perfect, 46, 52, 54; with                article, equivalent to relative

   force of Greek Perfect, 47; for                   clause with verb in Indicative or

   English Pluperfect, 48, 52-54; in               Subjunctive, 135, 144, 150, 151.

   indirect discourse, 49; used pro-             Apodosis: defined, 238; force and

   leptically, 50; in condition con-                 form of, after simple present and

   trary to fact, 248; in apodosis of                past particular suppositions, 242,

   such condition, 248, 249; with ἄν             243; after supposition contrary to

   in past general supposition, 26,                  fact, 248, 249; after future sup-

   315; expressing an unattained                    position more probable, 250, 263 ;

   wish, 27; English equivalents, 52,             after (implied) future supposition

   63; distinction between Aorist                   with less probability, 259; after

   and Imperfect, 56, 57; between                  present general supposition, 260,

   Aorist and Perfect, 86, 87.                          263; after past general supposi-

      Dependent Moods, 98; Sub-                    tion, 265 ; may have two protases,

   junctive in prohibitions, 162-164,             268; may be omitted, 271; its

 

                                                            187


188                                          INDEX OF SUBJECTS.

 

   verb may be omitted, 273; im-                Deliberative Future, 70.

   plied in the protasis, 276; after                Deliberative Subjunctive, 168; in

   concessive protasis, 278.                           relative clauses, 319.

Article with the participle, position           Dependent moods, enumerated, 3;

   of, 427.                                                          term not strictly applicable, 3,

Attendant circumstance expressed                Rem.

   by participle, 449, 119, 145.                    Direct Quotations, 334, 335; intro-

Attributive Participle, 421-428, see              duced by ὅτι, 345; frequent in

   under Participle; cf. 419, Rem. 1.              New Testament, 345.

Causal clauses, 228-232; other                   Dramatic Aorist, 45.

   methods of expressing causal re-            English Equivalents of Greek Aorist

   lations, 233; negatives in, 474.                   Indicative, 52, 53.

Circumstantial Participle, see 419,            English tenses, three-fold function,

   Rem. 1.                                                          354.

Commands: expressed by Future               Epistolary Aorist, 44.

   Indicative, 67; by Subjunctive,               Exhortations, expressed by the Sub-

   162-164, 166, 167; by Imperative,            junctive, 160-162; by the Impera-

   180; by Infinitive, 364; negatived              tive, 180.

   by IJ.1l, 478; apparently negatived        Exegetical grammar, scope and re-

   by 00, 479; negatived by 00 IJ.1l,              lation to historical grammar, 2.

   488 ; indirectly quoted, 204.                    Explanatory relative clauses, 295.

Complete and completed, sense of             Final clauses (pure), 197-199. For

   as grammatical terms, 85.                           relative clauses of purpose, see

Conative tenses: Present Indicative,             Relative clauses.

   11; Imperfect Indicative, 23;                   Final Particles, 190; clauses intro-

   Present Participle, 129.                                duced by, 188; general usage of

Concessive clauses, 278-288: general          such clauses, 189; usage of several

   definition, 278; particles intro-                  particles in detail, 191-195.

   ducing, 279-282; use of moods               Form and function distinguished, 1.

   and tenses, 283; various classes,             Future:

   284-287 ; English translation, 288 ;             Indicative: Predictive, 58-66;

   participle equivalent to, 437.                      Aoristic, 59;. Progressive, 60;

Conditional sentences, 238-277 ;                  value and significance of these

   definition, 238; express particular             distinctions, 61-64; assertive and

   or general supposition, 239, 240 ;              promissory force, 65; with οὐ μή,

   classification, 241; six classes,                   66; Imperative Future, 67; in

   242-265; peculiarities of, 266-277 ;          third person, 68; Gnomic, 69;

   negatives in conditional clauses,               Deliberative, 70 ; periphrastic

   469 475. See also Conditional                   forms, 71, 72 ; in final clauses and

   Relative sentences, under Relative            clauses introduced by final par-

   clauses.                                                          ticles, 198, 199, 205, 211, 215,

Consecutive clauses, 234-236; see               218, 224, Rem. 2; in cor:ditional

   also Result.                                                    clauses, 246, 254, 255; with εἰ by

Definite Relative clauses, see under             Hebraism with the force of an

   Relative clauses.                                           oath, 272; with εἰ expressing an

'"

                            INDEX OF SUBJECTS. 189


   object of desire, 276; in con-                   Imperative: in commands and ex-

   cessive clauses referring to the                   hortations, 180; in entreaties and

   future, 285 (a); in conditional                    petitions, 181; expressing con-

   relative clauses, 308; in rela-                      sent or hypothesis, 182, 183 ; force

   tive clauses of purpose, 317; in                  of tenses, 184, 163, 165; intro-

   complementary relative clauses,    duced by ὡστε, 237; negative of,

   318; in relative clauses intro-                     478, 479.

   duced by ἄχρι, 332; negatives                 Imperative Future, 67.

   with Future Indicative, 465, 466.            Imperfect Indicative: Progressive,

      Dependent moods, 99; peri-                    21, 22; Conative, 23; of repeated

   phrastic form made from μέλλειν,             action, 24; expressing an unat.

   100.                                                                tained wish, 27; translated by

      Participle: represents action                    English Perfect, 28 ; translated by

   relatively future, 152; of later                   English Pluperfect, 29; Imper-

   origin than other participles, 152,              fect of verbs of obligation, etc.,

   Rem.; periphrastic form made                    30-32; of verbs of wishing, 33;

   from μέλλων and Infinitive, 153.               in condition contrary to fact, 248;

Future Perfect Indicative, in New                 in apodosis of such condition, 248,

   Testament only in periphrastic                   249; with ἄν in conditional rela-

   form, 93, 94.                                                 tive clauses, past general suppo-

General and particular suppositions:            sition, 315; in indirect discourse

   expressed, 239, 240; implied in                 for Present Indicative, 348; peri-

   relative clause, 298, 299.                             phrastic form, 34; distinction

General Present: Indicative, 12;                    between Imperfect and Aorist, 56,

   Participle, 123-l26.                                      57.

Genitive absolute, 452-454.                        Inceptive Aorist, 35, 37; Indicative,

Gnomic tenses: Present, 12; Aorist,              41; dependent moods, 98; parti-

   43; Future, 69; Perfect, 79.                         ciple, 137.

Grammar, relation of, to interpreta-          Indefinite Aorist, 35, 98.

   tion, 2. ;                                                      Indicative:

Hebraisms in the New Testament:       In principal clauses: in un-

   Εἰ with Future Indicative with                    qualified assertions, etc., 157; in

   force of an emphatic assertion,                  qualified assertions, 158, 159.

   272; clause or Infinitive as sub-                    In subordinate clauses: in final

   ject of ἐγέντο, 357; Infinitive                     clauses and clauses introduced by

   defining content of action of a                   final particles, 198, 199, 205, 211,

   previous verb or noun, 375; in-                  215,224, Rem. 2,227; in clauses

   tensive participle, 448.                                of cause, 229, 230; in clauses of

Historical grammar, relation to                     result, 235, 236; in conditional

   exegetical grammar, 2.                                clauses, 242, 248, 254-256, 261; in

Historical Present, 14; Historical                  concessive clauses, 284, 285 (a) ;

   Aorist, 38; Historical Perfect,                     in definite relative clauses, 293,

   78.                                                                  294 ; in conditional relative

Hortatory Subjunctive, 160, 161;                  clauses, 301, 308, 309, 313; with

   introduced by ὡστε, 237.                            ἄν, 315; in relative clauses of pur-


190                      INDEX OF SUBJECTS.

   pose, 317-319; in relative clauses                 Constructions with the article:

   introduced by ἕως, etc., 826-332 ;             general effect of prefixing the

   as logical Sllbject of ἐγένετο, 857-            article, 392; as subject, 398; as

   360.                                                                object, 394; in apposition, 395;

      Negatives with Indicative, 465-              with τῷ expressing cause, 396;

   474.                                                                with τοῦ expressing purpose, 397

Indirect Discourse, 384-856; defini-            with τοῦ expressing result, 398

   tion, 834, 837; various methods                 with τοῦ after adjectives, 399;

   of expressing, 389, 840; classical              with τοῦ after nouns, 400; with

   usage, 341, 342, 847-350; New                  Toil after verbs that take the geni-

   Testament usage, 343-850; Eng-               tive, 401-408; with τοῦ as subject

   !ish usage compared with Greek,              or object, 404, 405; governed by

   851-356; Infinitive in, 390; par-                 prepositions, 406; various prepo-

   ticiple in, 460; conditional clauses            sitions used with it and their

   in,258; conditional relative clauses           force, 407-416; force of tense,

   in, 305; negatives in, 478.                           104-109.

Indirect object, expressed by clause          Intensive Perfect: Indicative, 77;

   introduced by ἵνα, 217; by In-                    dependent moods, 102.

   finitive, 368; by articular Infini-             Interpretation, relation of, to gram-

   tive after εἰς, 410.                                         mar, 2.

Indirect Questions, how introduced           Latin tenses, two-fold function, 854.

   in Greek, 840; after ἔχω, etc.,                  Moods: enumeration of, 3; in prin-

   346. introduced by ὅστις, 849;       cipal clauses, 157-184; finite

   by simple relatives, 350.                             moods in subordinate clauses, see

Infinitive: origin and stages of de-                Indicative, Subjunctive, etc., or

   velopment, 361; classification of               Contents, §§ 185-360.

   uses, 862, 363; negatives used                Negatives, 464-489: classical and

   with it, 480-484.                                           New Testament use in general,

      Constructions without the arti-               464; with the Indicative, 465-

   cle: with imperative force, 364,                 474; with the Subjunctive, 475;

   865; expressing purpose, 366,                    with the Optative; 476,477; with

   367; as indirect object, 368; ex-                 the Imperative, 478, 479; with

   pressing result, 369-372; defining             the Infinitive, 480-484; with

   content of action of a previous                   participles, 485; successive and

   verb or noun, 375; limiting ad-                  double negatives, 486-489, 468

   jectives and adverbs, 876, 377;                  compound negatives, 464, 486,

   limiting nouns, 378, 379; after                   489.

  πρίν or πρὶν , 380-382; used ab-            Object clauses: classification, 186;

   solutely, 383; as subject, 384,                    after verbs of exhorting, 200-204;

   885, 390, 857, 360; as appositive,             after verbs of striving, etc., 205-

   386; as object, 387-391, 202, 210;             210; after verbs of fear and

   in indirect discourse, 890 j force of           danger, 224-227; in indirect dis-

   the tensee in indirect discourse,                 course, 834, 339-850.

   110,11.2-:114; without article after        Optative, infrequent in New Testa-

   prepositions, 174.                                         ment, 174.


                                  INDEX OF SUBJECTS.                                      191


      In principal clauses: of wishing,             use of the participle with intensive

   175-177; Potential, 178, 179.                     force, 448; participle of attendant

      In subordinate clauses: in οb-                 circumstance, 449,450; more than

   ject clauses after verb of exhort-                one relation expressed by one

   ing, 200, Rem.; in conditional                    participle, 451; genitive absolute,

   clauses, future supposition more               452-454; position of adverbial

   probable, in indirect discourse,                  participle, 455.

   258; in conditional clauses, future              Substantive Participle: defined,

   supposition less probable, 259;                  456; as subject, 457; as object,

   with εἰ expressing an object of                   458-460; in indirect discourse,

   desire, 276.                                                    460; as a limiting genitive, 461;

      Negatives with Optative, 476,                 its position, 462; distinction from

   477.                                                                adjective participle used substan-

Participle: general nature, 115,418;              tively, 463.

   grammatical agreement, 116.                        Negatives with participle, 485.

      Tenses: general significance,               Particular and general conditions:

   118; use of each tense in detail,                 expressed, 239, 240; implied in

   see Present, Aorist, Future, Per-                 relative clause, 298, 299.

   fect, or Contents, §§ 119-156.                 Perfect:

   Classification respecting logical                   Indicative: of completed action,

   force, 419.                                                     74, 76, 85; of existing state, 75,

      Adjective Participle: defined,                 76 ; Intensive, 77; Historical, 78;

   420; restrictive attributive parti-                Gnomic, 79; Aoristic, 80, 88; in

   ciple, 422; with subject omitted,                indirect discourse, 81; of a past

   423, 433; with the article after                   event thought of as separated

   noun without the article, 424;                     from the moment of speaking

   neuter participle for abstract                       (incapable of adequate English

   noun, 425; explanatory attribu-                 translation), 82; used prolepti-

   tive participle, 426; attributive                   cally, 50; periphrastic form, 84;

   participle conveying subsidiary                 distinction between Perfect and

   idea of cause, etc., 428; predicate              Aorist, 86, 87.

   adjective participle, 429; its posi-                 Dependent moods: denoting

   tion, 430; forming periphrastic                  completed action, or existing re-

   verbs, 431; possible explanations              sult, 101; Intensive, 102; peri-

   of participle in the predicate, 432.             phrastic form, 103 ; Infinitive

      Adverbial Participle: defined,                 after prepositions, 104, 105, 107,

   434; temporal, 435; conditional,                108; Optative not found in New

   436; concessive, 437,438; causal,              Testament, 111; Infinitive in in.

   439; causal with cJs, 440, 441; of              direct discourse, 110, 112.

   purpose, 442; of means, 443;                   Participle: of completed action

   of manner, 444; manner ex-                        or existing state, 154 ; periphrastic

   pressed by ὡς with participle, 445,            form, 155; for a Pluperfect, 156.

   446; participle of manner or                    Periphrastic forms: in general, 20,

   means denoting same action as                  431; Present Indicative, 20; Im-

   that of the verb, 447; Hebraistic                 perfect, 34; Future Indicative, 71,


192                       INDEX OF SUBJECTS.

 

   72; Perfect. Indicative, 84; Plu-                 120-122; General, 123-126; for

   perfect, 91; Future Perfect, 94;                   the Imperfect, 127; Conative, 129 ;

   Present Infinitive and Imperative,              for the Future, 130; of action

   97; Perfect Subjunctive and Im-                still in progress, 131.

   perative, 103; Future Participle,              Primary and Secondary tenses, 4.

   153; Perfect Participle, 155.                    Progressive tenses:

Pluperfect Indicative: of completed                 Indicative: Present, 8-11; Im-

   action, 89; of existing state, 90,                 perfect, 21-23; Futtlre, 60.

   92; periphrastic form, 91; in in-                   Dependent moods: Present, 96.

   direct discourse for Perfect, 348.                  Participle: Present, 119-122,

Potential Optative: force of, 178,                  127-131.

   179; negatives with, 476.             Prohibitions, expressed by Aorist

Predicative Participle, 429-432, see             Subjunctive or Present (rarely

   under Participle.                                           Aorist) Imperative, 163; by Aorist

Predictive Future, 58.                                      Subjunctive, 162-164, 166, 167;

Present:                                                             by Present Imperative, 165,184.

   Indicative: most constant char-               Prohibitory Future Indicative, 67,

   acteristic of, 9; Progressive, 8,                   Rem. 2.

   10, 11; Conative, 11; General or             Prohibitory Subjunctive, 162-164,

   Gnomic, 12; Aoristic, 13; His-                   166, 167; used only in Aorist,

   torical, 14; used for Future, 15;                  163; force of tense, 164.

   ἥκω, πάρειμι, etc., 16; of past                  Promissory Future, 65.

   action still in progress, 17; in in-            Protasis: defined, 238; force and

   it. direct dIscourse, 19; periphrastic          form of, in simple present and

   forms, 20; in pure final clauses,                 past particular supposition, 242-

   198; in an appositive clause, 213,              247; in supposition contrary to

   Rem.; in conditional clauses, pres-            fact, 248; in future supposition

   ent particular supposition, 242,                  more probable, 250-258, 263; in fu-

   future supposition, 256, present                 ture supposition less probable, 259;

   general supposition, 261 ; in con-              in present general supposition,

   cessive clauses, 284; in condi-                   260-262, 264; in past general sup-

   tional relative clauses, present                   position, 265; joined to an

   particular supposition, 301, future             apodosis of a different form, 267 ;

   supposition, 309, present general              two protases with one apodosis,

   supposition, 313; after ἕως and                  268; substitutes for, 269; omitted,

   referring to the future, 326; after               270; equivalent to an oath, 272;

   Ews and referrIng to the present,               its verb omitted, 273; containing

   328.                                                                an apodosis, 276; after expressions

      Dependent moods, 96; peri-                    of wonder has the force of ὅτι,

   phrastic.form, 97; Infinitive after               clause, 277.; concessive protases,

   prepositions, 104, 106-109; Opta-             278; partIcIple equivalent to pro-

   tive and Infinitive in indirect dis-              tasis, 436,437.

   course, 110-112.                                        Purpose, expressed by clause intro-

      Participle: of simultaneous                      duced by ἵνα, 197-199; by rela-

   action, 119; of identical action,                  tive clause, 317; by Infinitive


                                     INDEX OF SUBJECTS.                                       193

   without the article, 366; by In-                Result: several conceptions of, 370 ;

   finitive with ὥςτε, 370(d), 371(d);            methods of expressing, 371 ; actual

   by Infinitive with ὥς, 372; by In-              result expressed by ὥστε with In-

   finitive with τοῦ, 397; by articular            finitive or Indicative, 235, 236,

   Infinitive after εἰς, 409; after πρός,            369, 370 (a) (b), 371 (a) (b); by

   414.                                                                articular Infinitive with εἰς, 411;

Questions: various classes of, 169;               conceived result expressed by

   negatives with, 467, 468. See clause         introduced by ἵνα, 218,219,

   also Indirect Questions. cf. 222; by           Infinitive usually with

Relative clauses: classification, 289-            ὥστε, 369, 370 ( c), 371 ( c); by

   291; distinction between definite              Infinitive with τοῦ, 398; by artic-

   and indefinite relative clauses,                   ular Infinitive with εἰς, 411; in-

   289,290.                                                         tended result (purpose), 370 (d),

      Definite Relative clauses: in-                  371 (d).

   troduced by relative pronouns and         Resultative Aorist, 35, 37, 42, 87.

   adverbs, 292; use of moods and              Shall and will in translating the

   tenses, 293; may imply relation                 Greek Future, 65.

   of cause, result, or concession,               Subjunctive:

   294; classified as restrictive and                   In principal clauses: Hortatory,

   explanatory, 295; conditional in                160, 161; with ἄφες or δεῦρο pre-

   form, 316.                                                      fixed, 161; Prohibitory, 162-164,

      Conditional Relative sentences:             166, 167; Deliberative, 168-171;

   defined, 289, 290, 296; imply par-            with θέλεις etc. prefixed, 171; in

   ticular or general supposition, 298,           negative assertions referring to

   299; six: classes, 300-315; clauses            the future, 172, 173.

   conditional in form but definite                    In subordinate clauses: in pure

   in thought, 316; introduced by                   final clauses, 197 ; in object

   ἕως, 329.                                                        clauses after verbs of exhorting,

      Relative clauses expressing pur-            etc., 200; after θέλω 203; in

   pose, 317-319.                                              clauses after verbs of striving,

      Relative clauses introduced by               etc., 205-207, 209; in subject,

   words meaning until, etc., 321-                  predicate, and appositive clauses

   333; introduced by ἕως, and re-                 introduced by ἵνα, 211-214; in

   ferring to the future, 322, 323,                   complementary and epexegetic

   325, 326; referring to what was in             clauses introduced by ἵνα, 215-

   past time a future contingency,                  217; in clauses of conceived re-

   324, 326; referring to a past fact,               suIt introduced by ἵνα, 218-220;

   327; referring to a contemporane-             in object clauses after verbs of

   ous event, 328; introduced by ἕως            fear and danger (expressed), 224;

   οὗ or ἕως ὅτου, 330; introduced by            (implied), 225; after ἐάν in con-

   ἄχρι, ἄχρι οὗ, etc., 331, 332; intro-            ditional clauses, future supposi-

   duced by πρίν, 333.                                      tion, 250; after εἰ in conditional

      Negatives in relative clauses,                  clauses, future supposition, 252,

   469,470, 474.                                                253; with εἰ, expressing an object

Restrictive Relative clauses, 295.                 of desire, 276; changed to Opta-


194                                  INDEX OF SUBJECTS.

 

   tive in indirect discourse, 258;                Tenses: enumeration of, 4; primary

   after lap ill condItional clauses,                 and secondary, 4; two-fold func-

   present general supposition, 260;              tion of, 5 ; chief function, 5, Rem. ;

   in concessive clauses referring to              interchange of, apparent, not real,

   the future, 285 (b); in conditional              15, Rem.

   relative clauses, future supposi-                    In Indicative Mood: general

   tion (with ἄν or ἐάν), 303, 304;                  significance, 6 ; two-fold function,

    (without ἄν), 307; retained in                    5 ; denote time relative to that of

   indirect discourse, 305; in condi-              speaking,7; apparent exceptions,

   tional relative clauses, present gen-           7 ; use of each tense in detail,

   eral supposition (with ἄν), 312;                 8-94 (see Present, Imperfect,

   in complementary relative clauses,           Aorist, etc.).

   318, 319 ; after ἕως [ἄν], 322-325 ;              In Dependent Moods: general

   after ἕως οὗ or ἕως ὅτου, 330 ; after         significance, 95; use of each tense

   ἄχρι, μέχρι, etc., 331; after πρίν,                ill detail, 96-114; tenses of the

   333.                                                                Infinitive after preposltlons, 104-

   Negatives with Subjunctive, 475.              109; of the Infinitive in indirect

Subordinate clauses classified, 186,             discourse, 110-114.

   187.                                                                   Of the Participle: general sig-

Suppositions: distinction between                nificance, 118; use of each tense

   particular and general, 239, 240,                in detail (see Present, Aorist,

   257; implied in relative clause                  Future, Perfect).

   289, 290, 296, 297; particular and          Will and shall in translating the

   general, 298, 299; expressed by                 Greek Future, 65.

   an Imperative, 182, 183; ex-                    Wishes: expressed by Optative, 175-

   pressed by a participle, 428, 436.               177; by the Future Indicative, 27,

Substantive Participle, 456-463, see             Rem. 2 ; unattainable, expressed

   under Participle.                                           by the Imperfect or Aorist In-

Supplementary Participle, see 419,               dicative, 27.


                                  INDEX OF GREEK WORDS.

 

                                    [The Numbers refer to Sections.]

Ἀκούω, Present with force of Per-             Ἄφες and ἄφετε prefixed to Ηorta
   fect, 16.                                                          tory Subjunctive, 161.

Ἄν :                                                                Ἄχρι, 331, 332.

      With Indicative: with Imper-                Βούλεςθε prefixed to Deliberative

   fect and Aorist to denote a cns-                  Subjunctive, 171.

   tomary past action, 26 ; in past                Γέγονα, Aoristic Perfect in Matthew

   general supposition, 315 ; in apodo-          (Mark?) only, 88.

   sis of condition contrary to fact,             Δεῦρο or δεῦτε prefixed to Hortatory

   248; omitted in such apodosis, 31,             Subjunctive, 161.

   249; cases in which it is not to be           Διά with τό and the Infinitive, 108,

   regarded as having been omitted,              406-408.

   30 (cf. 32, 33); with Future In-               Διότι as a causal particle, 228.

   dicative in future supposition,                 Διά with Infinitive as subject, 385.

   308; with Present Indicative in                Ἐάν :

   future supposition, 309.                             Conditional: with Present Indic-

      With Subjunctive: in condi-                    ative in present particular sup-

   tional relative clauses, implying                position, 247; with Future In-

   future supposition, 303; implying              dicative in future supposition,

   present general supposition, 312 ;             254; with Subjunctive in future

   relative clauses introduced by                    supposition, 260; in present gen-

   ἕως, 322; after ἄχρι, 332; after                   eral supposition, 260 ; in condi-

   ὅπως in final clauses, 196; for ἐάν             tional relative clauses for ἄν, 304,

   in conditional clauses, 250.                        312, Rem.

      With Potential Optative, 178,                     Concessive, 279-281,286 (b).

   179.                                                             Ἐὰν καί, concessive, 279,280,286 (b),

      With Infinitive, 372.                                 287; conditional, 282.

   In definite relative clauses con-              Ἐβουλόμην without ἄν, 33.

   ditional in form, 316; retained                Ἔγένετο δέ, construction after, 357-

   in indirect discourse with Sub-                   360.

   junctive retained unchanged, 306;          Ἔγνων with force of Perfect, 47.

   omitted when Subjunctive is                   Ἔδει with Infinitive denoting pres-

   changed to Optative, 344, Rem. 1.             ent obligation, 32.

Ἄντί with τοῦ and the Infinitive,               Εἰ :

   406, 407.                                                        Conditional: with present or

Ἄπέθανον with force of Perfect, 47.             past tense of the Indicative in

                                                              195

 


196                     INDEX OF GREEK WORDS.

 

   simple present or past particular             Εἴς with τό and the Infinitive, 107,

   supposition, 242, 243; with Pres-               406, 407, 409-413.

   ent Indicative in future supposi-             Ἔμελλον with the Infinitive, 73.

   tiοn, 266; with Present Indicative           Ἐν with τῷ and the Infinitive, 109,

   ill presεnt general supposιtιon,                 406, 407, 415.

   261; with a past tense of the                    Ἔνεκεν with τοῦ and the Infinitive,

   Indicative in condition contrary                 406, 407.

   to fact, 248; with Future Indica-             Ἐξέστην with force of Perfect, 47.

   tive in supposition referring to                Ἐπεί, ἐπειδή, ἐπειδήπερ as causal

   present intention, etc., 246; with                particles, 228.

   Future Indicative in future sup-               Ἔσχηκα as Aoristic Perfect, 88.

   position, 264, 256; with Future               Ἔφ' as a causal particle, 228.

   Indicative with the force of an                Ἔχω, followed by indirect delibera-

   oath, 272; with the Subjunctive                 tive questions, 346; by relativε

   in future supposition, 252, 253;                clause of similar force, 318.

   with the Optative in indirect dis-            Ἑώρακα as Aoristic Perfect, 88.

   course for ἐάν with Subjunctive             Ἕως: force as a relative adverb, 321,

   or fl with the Indicative of the                    introducing a clause referring to

   direct discourse, 258; with the                   the future, 322, 323, 326, 329; in-

   Optative in future supposition                   troducing a clause referring to

   less probable, 259; with Sub-                     what was in past time a future con-

   junctive, Optative, or Future In-                tingency, 324, 326, 329 ; how trans-

   dicative expressing an object of                lated when followed by the Sub-

   desire, 276; after expressions                     junctive, 325; introducing a clause

   of wonder with nearly the force                 referring to a past fact, 327; in-

   of ὅτι, 277.                                                    troducing a clause referring to a

      Concessive, 279-281; with a                   contemporaneous event, 328; fol-

   present or past tense of the In-                   lowed by οὖ or ὅτου, 330; with τοῦ

   dicative, 284; with Future Indie-                and the Infinitive, 406, 407.

   ative, 285.                                                   , Infinitive after, 374; after 1rpl",

      Interrogative, in indirect ques-                333, Rem. 2, 381.

   tions, 340.                                                   Ἤθελον without ἄν, 33.

Εἰ δέ and εἰ δὲ μήγε used elliptically,        Ἥκω, force of Present tense, 16.

   275.                                                             Ηὐχόμην without ἄν, 33.

Εἰ καὶ concessive, 279, 280, 281, Rem.,   θέλεις and θέλετε prefixed to De-

   288; conditional, 282, 286.                         liberative Subjunctive, 171.

Εἰ μή without dependent verb, mean-       Ἵν: New Testament usage, 191, 221,

   ing except, 274, 471.                                   223; in pure final clauses, 197-

Εἴθε in unattainable wishes, 27,                    199; in object clauses after verbs

   Rem. 1.                                                          of exhorting, etc., 200-203; in ob-

Εἴληφα as Aoristic Perfect, 88.                     ject clauses after verbs of striving,
Εἰμί used in the formation of peri-                etc., 205, 206; in subject, predi-

   phrastic verb-forms, 20, 34, 71,                 cate, and appositive clauses, 211-

   84, 91, 97, 103, 155, 431.                           214 ; in complementary and

Εἴρηκα as Aoristic Perfect, 88.                     epexegetic clauses, 215-217; in


                       INDEX OF GREEK WORDS.                           197

   clauses of conceived result, 218,            Μὴ γένοιτο, 177.

   219; not used to express actual               Μὴ οὐ, in questions, 468 ; after verbs

   result, 222; post-classical usage                 of hindering, etc., 484.

   in general, 223.                                          Ὅστις as an interrogative, 349.

Καί γε with concessive participle,             Ὅπως, New Testament usage,. 192,

   437.                                                                196; in pure final clauses, 197,

Καὶ ἐάν, concessive, 279, 281, 285              199; in object clauses after verbs

   (b); conditional, 282.                                   of exhorting, 200-202; in object

Καὶ ἐγένετο, construction after, 357-           clauses after verbs of striving, etc.,

   360.                                                                205, 207.

Καὶ εἰ concessive, 279, 281, 288;              Ὅτι as a causal particle, 228.

   conditional, 282.                                       Ὅτι introducing indirect quotation,

Καίπερ with concessive participle,               339 (a), 343 ; redundant before a

   437.                                                                direct quotation, 345.

Κέκραγεν, functionally a Present, 78.       Οὐ and its compounds: classical and

Λαμθάνω participle with, 147.                      New Testament usage in general,

Μέλλω, μέλλειν, etc., with Infinitive,          464.

   72, 73, 100, 153.                                              With the Indicative: in inde-

Μετά with τό and the Infinitive, 105,           pendent declaratory sentences,

   406, 407.                                                        465; with Imperative Future, 466 ;

Μέχρι, 331.                                                      in questions expecting an affirma-

Μή as a negative, and its compounds :         tive answer, 467; in conditional,

   classical and New Testament use               conditional relative, and con-

   in general, 464.                                             cessive clauses, 469, 470; after μή

      With the. Indicative: in ques-                  as a conjunction, 472; in indirect

   tions expectmg a negative answer,            discourse, 473; in causal and

   467, 468; in conditional and con-              simple relative clauses, 474.

   ditional relative clauses, 469; in                    With the Subjunctive after μή

   causal and relative clauses, 474,                as a conjunction, 475.

   Rem. 1, 2.                                                         With limitations of the Im-

      With the Subjunctive, 475, 162,             perative, 479.

   163.                                                                   With limitations of the Infini-

      With the Optative of Wishing,                tive, 481; compounds of ov with

   476.                                                                an Infinitive depending on a verb

      With the Imperative, 478, 163.               limited by οὐ, 482.

      With the Infinitive, 480; re-                        With participles, 485.

   dundant after verbs of hindering,            Οὐ μή, emphatic negative: with Pre-

   denying, etc., 402, 483, 484.                      dictive Future, 66, 487; with the

      With participles, 485.                               Imperative Future, 67, Rem. 2,

Μή as a final particle: New Testa-                488; with the Prohibitory Sub-

   ment uses, 193 j in pure final                     junctive, 167, 488; with the Sub.

   clauses, 199; in object clauses                   junctive in negative assertions,

   after verbs of striving, 206, 209 ;              172,173,487; after another nega-

   in object clauses after verbs of                   tive, 489.

   fear and danger, 224-227.


198                          INDEX OF GREEK WORDS.


Ὅφελον in expression of wishes, 27,        Ὑπάρχω used in the formation of

   Rem. 1, 2.                                                      periphrastic verb-forms, 20.

Πάρειμι, force of Present tense, 16.           Ὥς: New Testament usage as a final

Πεπίστευκα as Intensive Perfect, 77.            particle, 194; with the Infinitive

Πεποίηκα as Aoristic Perfect, 88.                denoting purpose, 372; with a

Πέποιθα as Intensive Perfect, 77.                causal participle, 440, 441; with

Πέπρακεν as Aoristic Perfect, 88.              the participle expressing manner,

Πρίν with a finite mood, 333; with               445, 446.

   the Infinitive, 380-382.                            Ὥστε : denoting result with Indica-

Πρό with τοῦ and the Infinitive, 106,           tive and Infinitive, 234, 235; with

   406, 407.                                                        Indicative, 236, 370 (a), 371 (a);

Πρός with τό and the Infinitive, 107,           introducing principal verb, any

   406, 407, 414.                                               mood, 237; with Infinitive denot-   

Σωζόμενοι, force of tense, 125.                     ing result, 369-371; with the In-

Τό, τοῦ, τῷ, Infinitive with, 392-417.          finitive denoting purpose, 367.

 


INDEX OF PASSAGES REFERRED TO.

 

[The numbers refer to sections. Passages referred to in Remarks are cited

by the number of the section to which the Remark is appended.]

 

                              I. NEW TESTAMENT PASSAGES.

 

SECTION                                                                  SECTION

Matt. 1: 18                453, 457                                 Matt. 5: 41                 308

Matt.l:19                    387                                          Matt.5:42                   180

Matt. 1: 22                 222, 223                                 Matt. 6: 1                   107, 406, 414

Matt. 1: 23                 20                                            Matt. 6: 5                   231, 232, 466

Matt.2:2                     157                                          Matt.6:11                   98

Matt.2:3                     439                                          Matt. 6:13                  162

Matt. 2: 9                   326                                          Matt. 6: 16                 231, 457

Matt. 2:10                  439                                          Matt. 6:25.                 171

Matt. 2: 13                 72, 397                                   Matt. 6: 27                 443

Matt. 2:15.                 222                                          Matt. 6:30.                 243

Matt. 2:20                  127                                          Matt. 6:31                  169

Matt. 3: 14                 23, 378                                   Matt. 6: 34                 162

Matt. 3: 15                 384, 429                                 Matt. 7: 1                   197

Matt. 3:17                  55                                            Matt.7:4                     161

Matt.4:2                     134                                          Matt.7:6                     199

Matt. .4 : 3                 200, 242                                 Matt. 7: 10                 269

Matt. 4: 4                   68                                            Matt. 7: 12                 171, 304

Matt. 4:14.                 222                                          Matt. 7:13                  124

Matt.5:6                     124                                          Matt.7:17                   12

Matt. 5:10                  286                                          Matt. 8:3                    39

Matt.5:17                   98, 366                                   Matt.8:8                     216

Matt. 5: 18.                172, 322                                 Matt. 8: 17                 222

Matt. 5: 19                 303, 310                                 Matt. 8: 20                 346

Matt. 5: 20.                250                                          Matt. 8: 31                 182, 256

Matt. 5: 21                 46, 52, 54                               Matt. 8: 32                 182

Matt. 5: 23                 96                                            Matt. 9: 17                 275

Matt. 5: 25                 97, 330                                   Matt. 9: 20                 131

Matt. 5: 28                 414                                          Matt. 9: 21                 250, 257

Matt. 5: 29, 30          214, 256                                 Matt. 9: 25                 53

Matt. 5: 39                 309                                          Matt. 9: 30                 209

                                                              199

 


200              INDEX OF PASSAGES REFERRED TO.

 

SECTION                                                                  SECTION

Matt. 10: 1                 371                                          Matt. 15: 20               393

Matt.l0:4                    142                                          Matt.16:1                   121

Matt.l0:14                  310                                          Matt.16:11                 473

Matt. 10: 23               323                                          Matt. 16: 13               169

Matt. l0:25                 214                                          Matt.16:19                 94

Matt. 10: 26               486                                          Matt. 16: 20               200, 204

Matt.l0:32                  308                                          Matt.16:22                 66

Matt. 10. 33               307, 308                                 Matt. 16. 25               310

Matt.10:37                 423, 428                                 Matt.16:26                 66

Matt.10:38                 313, 428, 469                        Matt.16:27                 72

Matt.l0.39                  428                                         Matt.16.35                 66

Matt.10:40,41.          428                                         Matt.17:5                   55

Matt.10:42                 65,428                                    Matt.17:8                   471

Matt.l1:14                  59                                            Matt.18:4,5                308

Matt.11:41                 51                                            Matt.18:6                   214

Matt.11:14                 282                                         Matt.18:8, 9.              256

Matt.11:21                 142                                         Matt.18:10                 209

Matt.11:25                 222                                         Matt.18:18                 94, 431

Matt.11:28                 65                                            Matt.18:21                 169

Matt.11:29                 65                                            Matt.18:30                 324, 329

Matt.12:3                   109                                         Matt.18:33                 30

Matt.12:14                 207                                         Matt.19:4                   52

Matt.12:17                 222                                         Matt.19:10                 243, 264

Matt.12:18                 55                                            Matt.19:14                 387, 402

Matt.12:20                 322                                         Matt.19:21                 269

Matt.12:31                 65                                            Matt.19:27                 139

Matt.13:5,6                108                                         Matt.20:10                 343

Matt.13:12                 313                                         Matt.20:19                 410

Matt.13:14                 167                                         Matt.20:22                 72, 393

Matt.13:15                 199                                         Matt.20:26,    27.       68

Matt.13:17                 151                                         Matt.21:42                 22

Matt.13:30                 414                                         Matt.21:19                 167

Matt.13:33                 330                                         Matt.21:32                 218, 398

Matt.13.35                 222                                         Matt.21.34                 109

Matt.13:46                 78, 88                                     Matt.21:38                 161

Matt.13:53                 459                                         Matt.21:41                 317

Matt.13:55                 467                                         Matt.22:11                 485

Matt.14:3                   48                                            Matt.22:14                 218

Matt.14:4                   29                                            Matt.22:15                 207

Matt. 14: 7                 305                                          Matt. 22: 23               112, 480

Matt.14:9                   437                                          Matt. 22:28                39

Matt. 14: 13               218                                          Matt. 22: 29               485

Matt. 14: 36               200                                          Matt. 23: 2                 55

Matt. 15: 2                 312                                          Matt. 23: 12               308

Matt. 15: 6                 67, 68                                     Matt. 23: 13               129


                     INDEX OF PASSAGES REFERRED TO              201

 

SECTION                                                                  SECTION

Matt. 23: 20               124, 150                                 Matt. 28: 15               18, 52

Matt. 23: 23               30                                            Matt. 28: 18               15

Matt. 23: 33               170

Matt. 24: 4                 206, 209                                 Mark 1: 7                   376

Matt.24:6                   209                                          Mark 1:9                    359

Matt. 24: 12               108                                          Mark 1: 11                 55

Matt.24:13                 150                                          Mark1:17                   269

Matt. 24: 20               200                                          Mark 1: 22                 445, 446

Matt. 24: 22               469                                          Mark 1: 24                 224

Matt. 24: 23               165                                          Mark 1: 31                 134, 138

Matt. 24: 45               397                                          Mark 1: 40                 263

Matt.25:6                   80                                            Mark 2:5                    13

Matt. 25: 8                 8                                              Mark 2: 16                 349

Matt. 25: 9                 225, 475, 487                        Mark 2: 18                 34, 431

Matt. 25: 16               150                                          Mark 2: 20                 295, 316

Matt. 25: 27               30                                            Mark 2: 23                 360

Matt.26:2                   410                                          Mark2:24                   474

Matt. 26: 4                 210                                          Mark 2: 28                 237

Matt. 26: 15               269                                          Mark 3: 4                   384

Matt. 26: 17               171                                          Mark 3: 6                   207

Matt. 26: 18               15                                            Mark 3:7                    67

Matt. 26: 19               293                                          Mark 3: 11                 290, 315

Matt. 26: 24               249,469                                  Mark 3: 21                 47

Matt. 26: 32               105                                          Mark 3: 24                 260, 262

Matt. 26: 33               279, 285                                 Mark 3: 25                 262

Matt. 26: 35               66, 279, 281                          Mark 3: 28                 304, 314

Matt. 26: 36               325                                          Mark 3: 29                 314

Matt. 26: 43               91                                            Mark 4: 6                   406

Matt. 26: 48               310                                          Mark 4: 7                   157

Matt.26:52                 124                                          Mark 4:21                  345

Matt. 26: 54               169                                          Mark 4: 23                 368

Matt.26:56                 222                                          Mark   4:31                437

Matt.27:1                   371                                          Mark   4:37                369

Matt.27:4                   139                                          Mark 4:39                  101

Matt. 27: 8                 18, 52                                     Mark 5: 4                   108, 408

Matt. 27 :17               171                                          Mark 5: 15                 166

Matt. 27: 20               42                                            Mark 5: 16                 123

Matt. 27: 21               171                                          Mark 5: 18                 156, 200

Matt. 27: 31               48, 109                                   Mark 5: 19, 20.         350-

Matt. 27 : 33              20, 431                                   Mark 5 : 24                57

Matt. 27: 41               120                                          Mark 5: 25                 131

Matt. 27: 43               75                                            Mark 5: 33                 137

Matt. 27: 49               161                                          Mark 5: 35                 47

Matt.27:63                 15                                            Mark 5:36                  138

Matt. 28: 6                 295                                          Mark 5: 41                 20

202        INDEX OF PASSAGES REFERRED TO.


SECTION                                                                  SECTION

Mark 6: 2                   457                                          Mark 12: 12               387

Mark 6: 9                   365                                          Mark 12: 18               390

Mark 6: 10                 322                                          Mark 12: 33               96, 393

Mark 6: 25                 171                                          Mark 12: 41               21, 56

Mark 6: 31                 368                                          Mark 12: 44 ,            56

Mark 6: 34                 446                                          Mark 13: 2                 487

Mark 6: 36                 346                                          Mark 13: 5                 209

Mark 6: 45                 328                                          Mark 13: 11               303

Mark 6:50                  165                                          Mark 13:18                200

Mark 6: 56                 315                                          Mark 13: 21               165

Mark 7: 12                 482                                          Mark 13: 22               107

Mark 8: 1, 2.              346                                          Mark 13: 30               172, 331

Mark 8:12                  272                                          Mark 13:34                204

Mark 8:14                  48, 53                                     Mark l4:2                   199

Mark 8:29                  141                                          Mark 14:12                171

Mark 8: 35                 308                                          Mark 14: 14               318, 319

Mark 9: 6                   343                                          Mark 14: 21               469

Mark 9: 9                   204, 471                                 Mark 14: 28               406

Mark 9: 10                 393                                          Mark 14: 29               279, 280, 285

Mark 9 : 11                349                                          Mark 14: 30               380

Mark 9: 12                 212                                          Mark 14 : 31              66, 279

Mark 9 : 22                181                                          Mark 14: 32               325

Mark 9: 26                 371                                          Mark 14: 41               52

Mark 9: 28                 349                                          Mark 15: 9                 122, 171

Mark 9: 31                 15                                            Mark 15: 36               161

Mark 9: 35                 296                                          Mark 15: 44               277

Mark 10: 11               312                                          Mark 16: 3                 169

Mark 10: 22               432                                          Mark 16: 20               119, 449

Mark 10: 32               34

Mark 10: 36               171                                          Luke 1: 1                   229

Mark 10: 40               368                                          Luke 1: 8                   415

Mark 10: 43               296                                          Luke 1: 9                   401

Mark 10: 51               171                                          Luke 1: 10                 34

Mark11:13                 276                                          Luke1:21                   431

Mark 11 : 14              176, 476                                 Luke 1: 22                 34

Mark 11 : 16              210                                          Luke 1: 29                 1l1

Mark l1:18                 207                                          Luke 1:33                  59

Mark 11: 19               315                                          Luke 1: 38                 176

Mark 11: 23               310                                          Luke 1: 43                 213

Mark 11: 24               269                                          Luke 1: 54                 375

Mark 11: 25               309                                          Luke 1: 57                 400

Mark 11 : 27              14                                            Luke 1: 59                 23

Mark 11 : 28              216, 220                                 Luke 1: 62                 111, 179

Mark 11: 31               65                                            Luke 1: 66                 21

Mark 12: 7                 161                                          Luke 1: 72                 375

                    INDEX OF PASSAGES REFERRED TO.                      203

                                    SECTION                                                                  SECTION

Luke 1: 74                 400                                          Luke 7: 39                 73, 241

Luke 1 : 76,               77, 79 397                              Luke 7: 40                 368

Luke2:6                      400                                          Luke7:45                   459

Luke 2:18.                 57                                            Luke 7:47                  313

Luke 2 : 21.               106, 400                                 Luke 7: 48                 345

Luke 2: 22, 24           397                                          Luke 8: 2                   89

Luke 2: 26                 91, 114, 333,                         Luke 8: 5                   415

344, 390, 431                        Luke 8: 9                   343

Luke 2: 27.                109, 397                                 Luke 8: 10                 366

Luke 2 : 35                195                                          Luke 8: 18                 310, 313

Luke 2 : 44                37                                            Luke 8: 27 29,           48

Luke 2:49.                 28                                            Luke 8:43                  131

Luke 2: 51                 22                                            Luke 8: 46                 154, 458

Luke 3: 9                   15                                            Luke 8: 47                 350

Luke 3: 10.                168                                          Luke 8: 49                 14, 47

Luke3:15                   111                                          Luke 9:4                     310

Luke 3: 16.                122                                          Luke 9: 13                 252, 253

Luke3:21                   109                                          Luke 9:18                  97

Luke 3: 22.                55                                            Luke 9: 24                 314

Luke 4: 10                 404, 405                                 Luke 9: 25                 141

Luke 4 : 15.               449                                          Luke 9: 27                 172, 322

Luke 4: 16.                295                                          Luke 9: 33                 485

Luke 4:22.                 67                                            Luke 9:34                  109

Luke 4: 23.                146, 460                                 Luke 9: 36            41, 78, 88, 109

Luke 4:29.                 371                                          Luke 9:44                  72

Luke 4 : 41.               90                                            Luke 9: 45                 222

Luke 5: 1                   358                                          Luke 9: 46                 179

Luke 5: 4                   457                                          Luke 9: 48                 314

Luke 5: 7                   404, 405, 449                        Luke 9: 50                 469

Luke 5:10.                 71                                            Luke 9:52                  372

Luke5:16                   34                                            Luke 9:54                  98, 171

Luke 5:17.                 432                                          Luke 9:57                  304

Luke 5:26.                 62                                            Luke 9:58                  346

Luke 5:35                  316                                          Luke 10:2                  200

Luke 6:11.                 111, 179, 343                        Luke 10:6                  275

Luke 6: 12.                360                                          Luke 10: 18               146

Luke 6: 31.                171                                          Luke 10 : 19              400

Luke 6: 32, 33, 34    282                                          Luke 10 : 40              368

Luke 6: 37.                183                                          Luke 11 : 1                97

Luke 6: 42.                161, 485                                 Luke 11: 3                 98

Luke 7: 2                   13                                            Luke 11 : 5                168, 169

Luke 7 : 4                  318, 319                                 Luke 11 : 6                318

Luke7:6 216,             220                                          Luke 11: 7                 449

Luke7:19                   433                                          Luke 11:8                  285

Luke 7 : 35.               43                                            Luke 11 : 18              112, 282

204                      INDEX OF PASSAGES REFERRED TO.

SECTION                                                                  SECTION

Luke 11: 32               231, 232                                 Luke 16: 17               384

Luke 11 : 35              209                                          Luke 16: 18               124

Luke 11: 37               109                                          Luke 16: 31               59

Luke 11: 42               480                                          Luke 17: 1                 405

Luke 12: 1                 371                                          Luke 17: 2                 214

Luke 12: 4                 368                                         Luke 17: 4                 98

Luke 12: 5                 105                                          Luke 17: 5                 181

Luke 12: 8                 308                                          Luke 17: 8                 325

Luke 12: 8-10           150                                          Luke 17: 27               331

Luke 12: 10               308                                          Luke 17: 31               308

Luke 12 : 11              478                                          Luke 18: 1                 107, 414

Luke 12: 15               415                                          Luke 18: 4                 284, 469

Luke 12: 17               346                                          Luke 18: 5                 469

Luke 12: 32               55                                            Luke 18: 7                 172

Luke 12: 33               485                                          Luke 18: 10               366

Luke 12: 34               309                                          Luke 18: 25               374, 384

Luke 12: 35               103                                          Luke 18: 36               179

Luke 12: 36               454                                          Luke 18: 41               171

Luke 12: 45               405                                          Luke 19: 9                 46

Luke 12: 52               94                                            Luke 19: 15               109

Luke 12: 59               323                                          Luke 19: 17               97

Luke 13: 7                 17                                            Luke 19: 40               93, 254

Luke 13: 9                 271, 275                                 Luke 19: 42               271

Luke 13: 10               431                                          Luke 19: 47               34

Luke 13: 16               30                                            Luke 19: 48               444

Luke 13:23                125                                          Luke 20:6                  84

Luke 13: 25               303                                          Luke 20: 10               198, 199

Luke 13: 28               316                                          Luke 20: 16               176

Luke 14: 1                 109                                          Luke 20: 20               371

Luke 14: 10               199                                          Luke 20: 22               384

Luke 14: 26               256, 261, 428, 469                Luke 20: 40               482

Luke 14: 27               313                                          Luke 20: 41               112

Luke 14: 31               376                                          Luke 21 : 1                138

Luke 14: 33               469                                          Luke 21: 8                 209

Luke 14: 34               169                                          Luke 21 : 22              400

Luke 15: 4                 323, 329                                 Luke 21 : 24              71

Luke 15: 16               23                                            Luke 22: 6                 400

Luke 15: 24               92                                            Luke 22: 9                 171

Luke 15: 26               179, 343                                 Luke 22: 11               318

LukeI5:29                  17                                            Luke 22:15                106

Luke 15: 32               41, 54                                     Luke 22: 26, 27.        446

Luke 16: 3                 387                                          Luke 22 : 28              433

Luke 16: 4                 45                                            Luke 22 : 34              112, 323, 390

Luke 16: 11               169, 242, 472                        Luke 22 : 40              200

Luke 16: 15               433                                          Luke 22 : 46              200

                   INDEX OF PASSAGES REFERRED TO.                       205


SECTION                                                                  SECTION

Luke 22: 49               70, 169                                   John 4: 52                  41

Luke 22 : 61              66                                            John 5: 5                    131

Luke 22:65                121                                          John 5:6                     17

Luke 22: 67, 68         285                                          John 5: 7                    217

Luke 23: 16               84                                            John 5: 13                  134, 136

Luke 23: 19               20                                            John 6: 14                  165

Luke 23 : 31              169                                          John 6: 18                  387

Luke 23: 35               246                                          John 5:23                   485

Luke 23 : 37              245                                          John 5: 30                  482

Luke 23: 53               489                                          John 5:36                   86

Luke 24: 16               403                                          John 5:37                   88

Luke 24: 23               343                                          John 6:38                   86

Luke 24 : 25              399                                          John 6: 6                    120, 121

Luke 24 : 26              30                                            John 6: 25                  82

Luke 24 : 30              109                                          John 6:29                   213

Luke  24 : 32             22                                            John 6: 39, 40           213

Luke 24:36                435                                          John 6:50                   422

Luke 24 : 46              75, 114, 390                          John 6: 63                  293

Luke 24: 51               406                                          John 6: 64                  153

John 6: 68 70,           169

John 1:1                     157                                          John 6:69                   77

John 1:11                   38, 465                                   John 6:71                   73

John 1: 12                  290, 378                                 John 7: 3                    199

John 1:15                   78                                            John 7:23                   244

John 1:18                   88                                            John 7:26                   47

John 1: 26                  122, 250                                 John 7: 32                  458

John 1:27                   216                                          John 7:39                   73

John 1:33                   299                                          John 1:45                   169

John 1: 38                  157                                          John 7: 51                  260, 467

John 1:48                   106                                          John 8: 14                  285

John 2:4                     16                                            John 8:16                   281, 285

John 2: 16                  151                                          John 8: 20                  474

John 2:17                   84                                            John 8:30                   452

John 2: 19                  182, 183                                 John 8: 31                  263

John 2: 25                  216, 348                                 John 8: 33                  88

John 3:8                     313                                          John 8:51                   261

John 3:12                   244                                          John 8:52                   47

John 3: 16                  236, 371                                 John 8: 55                  282

John 3:18                   474                                          John 8:56                   217

John 3: 27                  482                                          John 9: 2                    218, 219

John 4:4                     30                                            John 9:4                     98, 327

John 4:34                   213                                          John 9:9                     345

John 4:39                   461                                          John 9:18                   330

John 4:47                   16                                            John 9:22                   89, 217

John 4: 49                  380                                          John 9: 33                  31, 249, 469


206                   INDEX OF PASSAGES REFERRED TO.

 

SECTION                                                                  SECTION

John 10: 12                485                                          John 16: 8                  435

John 10: 32                11                                            John 16: 13                65

John 10: 38                285                                          John 16: 23                257

John 11 : 2                 142                                          John 16 : 24               52, 103

John 11 : 9                 240, 260                                 John 16: 32                216

John 11 : 21               248                                        John 17: 2                  199, 220

John 11:27                 343                                          John 17:3                   198

John 11 : 36               21                                            John 17 : 5                 55, 106

John 11 : 37               206                                          John 17 : 11               181

John 11: 44                156                                          John 17: 19                103

John 11: 50                214                                          John 17: 24                55

John 11 : 52               481                                          John 18: 16                90

John 11: 53                210                                          John 18: 18                90

John 12:3                   142                                          John18:23                  245

John 12:4                   153                                          John 18: 24                48

John12:10                  205                                          John 18:28                 14

John 12: 17                48, 127                                   John 18: 30                241, 432

John 12: 18                53                                            John 18: 39                213

John 12 : 23               216                                          John 19: 6                  109

John 12 : 24               260                                          John 19: 11                249

John 12 : 26               260, 261, 309                        John 19: 12                263

John 12 : 29               112, 390                                 John 19: 30                48, 109

John 12 : 32               62                                            John 20 : 29               232

John 12 : 36               293                                          John 21 : 19               120

John 13 : 5                 431                                          John 21 : 22, 23        328

John 13 : 8                 158                                          John 21 : 25               112, 390

John 13:9                   479

John 13: 11                126                                          Acts 1: 3                    105, 122

John 13:12                 48                                            Acts 1:10                   90

John 13: 17                268                                          Acts 1: 16                  30, 142

John 13:23                 34                                            Acts 1:25                   371

John 14 : 3                 309                                          Acts 2: 13                  84, 444

John 14: 15                250                                          Acts 2:21                   303

John 14: 19                229                                          Acts 2: 45                  315

John14:26                  59                                            Acts 2:47                   125

John 14:28                 248                                          Acts 3:2                     24

John 15:6                   43                                            Acts 3:12                   404

John 15:8,                  50, 213                                   Acts 3:18                   114

John 15:9                   98                                            Acts 3:19                   195

John 15 : 12, 13        213                                          Acts 3: 26                  415, 442

John 15: 20                242, 295                                 Acts 4: 14                  368

John 15: 22                249                                          Acts 4: 20                  484

John 16 : 2                 124, 148, 216                        Acts 4: 27                  92

John 16 : 3                 232                                          Acts 4: 30                  415, 417

John 16 : 7                 65, 214                                   Acts 4: 31                  92

                    INDEX OF PASSAGES REFERRED TO.                  207

SECTION                                                                  SECTION

Acts 4: 32.                 112                                          Acts 9: 34                  13

Acts 4: 34.                 127                                          Acts 9: 39                  29

Acts 4: 35.                 315                                          Acts 10: 1, 2              420

Acts 4: 37                  127                                          Acts10:3                    146

Acts 5: 3                    389                                          Acts l0:7                    127

Acts 5: 5                    39                                            Acts 10: 17                154, 179, 343

Acts 5: 14.                 423                                          Acts 10: 22                123

Acts 5: 17.                 424                                          Acts 10: 23                138

I Acts 5: 24.               179, 343                                 Acts 10: 25                404

Acts 5: 26.                 224                                          Acts 10: 33                139, 366, 447

Acts 5: 28.                 74                                            Acts 10: 35                151, 423

Acts 5:30.                  447                                          Acts 10:41                 424

Acts 5:35.                  72                                            Acts 10:44                 119

Acts 5:38                   230                                          Acts10:47                  402

Acts 5:39                   225, 242                                 Acts11:3                    146

Acts 5 : 42.                457                                          Acts 11 : 15               109, 415

Acts 6:3                     317                                          Acts 11:17                 98

Acts 6:11                   86                                            Acts 12:10                 427

Acts 7:5                     485                                          Acts 12:18                 452

I Acts 7:7                   308                                          Acts 12:20                 108

Acts 7:12                   460                                          Acts13:16                  138

Acts 7: 18.                 331                                          Acts 13:27                 137

4 Acts 7: 19.              398                                          Acts 13: 28                437, 485

Acts 7 : 26.                23, 141                                   Acts 13: 32                427

Acts 7:34                   161                                          Acts13:33                  141

Acts 7 : 35.                82                                            Acts 13: 40                206, 209

Acts 7:36                   37, 42                                     Acts14:9                    400

Acts 7: 42.                 368                                          Acts 14: 18                403, 483

Acts 7 : 44.                89                                            Acts 14: 19                112, 134, 138

Acts 7:52                   62                                            Acts14:27                  350

Acts 7: 60.                 41                                            Acts 15: 10                217, 375

Acts 8: 11.                 108                                          Acts 15: 12                41

Acts 8: 20.                 176, 452                                 Acts 15: 13                37, 98

.Acts 8: 22.                276                                          Acts 15: 14                350

Acts 8: 23.                 460                                          Acts 15: 17                195

Acts8:27                    442                                          Acts 15:20                 404

Acts 8: 31                  178, 179, 254, 267                Acts 15: 21                17

Acts 9:3                     415                                          Acts15:22                  449

Acts 9: 12.                 146                                          Acts 15: 24        86, 121, 122, 229

Acts 9:21                   89                                            Acts15:26                  343

Acts 9: 22.                 89, 120, 121, 447                  Acts 15: 28                386

Acts 9 : 23.                210                                          Acts 15: 29                121, 122

Acts 9 : 26.                439                                          Acts 16: 16                443

Acts 9: 32.                 360                                          Acts 16: 18                13

Acts 9: 33.                 131                                          Acts 16: 23                145

208                    INDEX OF PASSAGES REFERRED TO.

 

SECTION                                                                  SECTION

Acts 16: 27 37,          390                                          Acts 23 : 30               44, 99, 112

Acts 17: 3.                 30                                            Acts 23: 35                145

Acts 17 : 6.                16, 485                                   Acts 24: 10                131

Acts 17: 11                111                                          Acts 24: 11                152

Acts 17: 13                119                                          Acts 24 : 19               30, 258

Act8 17: 18                169, 178, 179,                       Acts 24: 23                145

270, 385                                 Acts 25: 10                84

Acts 17: 21                368, 374                                 Acts 25: 11                394

Acts 17: 27                268, 276, 437                        Acts 25: 13                142, 145

Acts 18: 2.                 108                                          Act8 26 : 16               111, 333, 344

Acts 18: 9.                 98, 164                                   Acts 25: 22                33

Acts 18: 10                398                                          Acts 25 : 26               101

Acts 18: 14                153, 248                                 Acts 25 : 26               368

ActS 18:18                449                                          Acts 26:11                 13

Acts 18: 20                452                                          Acts 26 : 4                 427

Acts 18: 23                119                                          Acts 26 : 5                 263

Acts 19:9                   119                                          Acts 26:6                   427

Acts 19: 27                100, 481                                 Acts 26 : 8                 277

Acts 19: 32                89, 348                                   Acts 26: 11                23

Acts 20: 3.                 163, 400                                 Acts 26 : 13               146

Acts 20:7                   153                                          Acts 26:18                 397

Acts 20: 16                268, 476                                 Acts 26 : 22               485

Acts 20: 24                194                                          Acts 26 : 26               84, 482

Acts 20: 32                426                                          Acts 26: 29                179, 481

Acts 20 : 36               374                                          Acts 26: 32                30, 101

Acts 20:38                 72                                            Acts 27:1                   404

Acts 21:1                   138                                          Acts 27:3                   139

Acts 21 : 3.                130                                          Acts 27 : 9                 108

Acts 21 : 12               404                                          Acts 27 : 10               100, 481

Acts 21 : 13               481                                          Acts 27 : 12               258, 276

Acts 21 : 16               319                                          Acts 27 : 13               134, 137, 138

Acts 21 : 24               199                                          Acts 27 : 21               30

Acts22:4                    122                                          Acts 27:29                 224

Acts 22 : 17               453                                          Acts 27 : 30               440

Acts 22:22                 32                                            Acts 27:33                 331

Acts 22 : 24               145                                          Acts 27 : 39               268

Acts 23 : 1.                137                                          Acts 27 : 43               42

Acts 23 : 9.                271                                          Acts 28 : 6                 100

Acts 23 : 10               224                                          Acts 28: 19                440

Acts 23 : 12               330                                          Acts 28 : 23               129

Acts 23 : 13               146                                          Acts 28 : 26               167

Acts 23 : 16               200, 399, 440                        Acts 28 : 27               197, 199

Acts 23 : 17, 18, 19  368                                          Acts 28 : 30               39

Acts 23 : 20               404, 440

Acts 23: 26                388                                          Rom. 1: 10                 276, 371

INDEX OF PASSAGES REFERRED TO.                209

SECTION                                                                  SECTION

Rom. 1: 11                 197, 409                                 Rom. 9: 17                 197

Rom.l:20                    411                                          Rom. 9:18                  313

Rom.l:24                    400                                          Rom. 9:27                  285

Rom. 2: 4.                  11, 425                                   Rom. 10 : 14             169, 170

Rom.2:12                   54, 301                                   Rom. l0:15                 475

Rom. 2 : 27                427                                          Rom. 10: 18, 19        468

Rom.3:4.                    16, 177, 195,                         Rom. ll:l                     176

199, 476                                 Rom. 11 : 6.               243

Rom. 3:5.                   428                                          Rom.11:8.                  400

Rom. 3:6.                   176, 476                                 Rom.11: 11                176

Rom. 3:23                  54                                            Rom.11:14                 276

Rom. 3:26                  409                                          Rom. ll:16                  273

Rom. 3:31                  176, 476                                 Rom.12:2                   107

Rom. 4:2.                   242, 245                                 Rom.12:3                   411

Rom. 4:12                  481                                          Rom.12:15                 365

Rom. 4:13                  395                                          Rom.13:5.                  481

Rom. 4:14                  243, 273                                 Rom.13:8                   489

Rom. 4:16                  481                                          Rom.13:11                 378

Rom. 4:19                  145                                          Rom.14:2                   387

Rom. 4:21                  145                                          Rom.14:9                   41

Rom. 5:1                    134, 136                                 Rom.14:13                 395

Rom. 5:5                    74                                            Rom.14:21                 393

Rom. 5:7                    69                                            Rom.15:6                   176

Rom. 5:11.                 475                                          Rom.15:8.                  390

Rom. 5:10                  244                                          Rom.15:13                 176

Rom. 5: 12                 64, 229                                   Rom. 16 : 14             154

Rom. 6: 2.                  60, 176, 294                          Rom. 15: 22               28, 401, 402

Rom. 6:15                  176                                          Rom.16:7                   82

Rom. 6:16                  313

Rom. 7: 3.                  69, 262, 398                          1 Cor. 1: 4                  134, 135

Rom. 7:4                    409                                          l Cor. l:l0                   200

Rom. 7:6                    481                                          l Cor. 1:17                 481

Rom. 7:7                    176                                          1 Cor. 1:18                126

Rom. 7:13                  176                                          1 Cor. 1:21                231

Rom. 8:9.                   469                                          1 Cor. 1:22                231, 232

Rom. 8:10                  242                                          1 Cor. 1:25                425

Rom. 8:13                  72                                            l Cor. 2:7                    156

Rom. 8: 17                 243, 273                                 1 Cor. 2: 10               156

Rom. 8:24                  290                                         l Cor.2:5                     476

Rom. 8:25                  261                                          1 Cor. 3:12                240

Rom. 8:29                  409                                          1 Cor. 3:17                240

Rom. 9:1.                   462                                          l Cor. 4:2                    210, 212

Rom. 9:3                    33                                            l Cor. 4:3                    214

Rom. 9:11                  454                                          l Cor. 4:6                    322

Rom. 9:14                  176                                          l Cor. 4:6                    198

210                        INDEX OF PASSAGES REFERRED TO.


SECTION                                                                  SECTION

1 Cor. 4:7                   282                                          1 Cor.15:4                  86

1 Cor. 4:8                   27, 54                                     1 Cor.15:18               139

1 Cor. 4: 12               438                                          1 Cor. 15: 29             230, 232

1 Cor. 4:18                440                                          1 Cor.15:36               313

1 Cor. 4: 19               250                                          1 Cor. 15: 37             152, 259, 313

1 Cor.4:21                  171                                          1 Cor. 15:51, 52        59

1 Cor. 5:8                   237                                          1 Cor. 16:3                310

1 Cor. 5: 10               31, 249, 479                          1 Cor. 16 : 4              405

1 Cor. 5: 11               44                                            1 Cor. 16 : 10            205, 209

1 Cor. 6: 15               176                                          1 Cor. 16: 11             166

1 Cor. 7:5                   273

1 Cor. 7:8                   273                                          2 Cor. 1:8                   371, 401

1 Cor.7:11                  282                                          2 Cor. 1:9                   80, 103

1 Cor.7:25                  446                                          2 Cor.l:10                  77

l Cor.7:28                  50                                            2 Cor.1:17                  218

l Cor.7:36                  112, 182                                 2 Cor. 2:1                   395

1 Cor. 7:39                79, 260                                   2 Cor. 2:3                   30

1 Cor. 7:40                260                                          2 Cor. 2:5                   240

1 Cor. 8:5                   281                                         2 Cor. 2:7                   371

1 Cor. 8:9                   206, 209                                 2 Cor. 2:10                301

1Cor. 8:10                  411                                          2 Cor. 2:13                80, 396

1 Cor. 9:1                   169                                          2 Cor. 2:15.               125

1 Cor. 9:4, 5              468                                          2 Cor. 2:17                20

l Cor. 9:10                 400                                          2 Cor. 3:13                414

1 Cor. 9:11                246, 268                                 2 Cor. 4:3                   84

1 Cor. 9:16                285                                          2 Cor. 4:16                284

1 Cor. 9:18                213                                          2 Cor. 5:13                47

1 Cor. 9: 26               445                                          2 Cor. 5: 20               445, 446

1 Cor. 10: 12             206                                          2 Cor. 6: 9, 10            446

1 Cor. l0:27,28          256                                          2 Cor. 7:5                   80

1 Cor. 11:2                75                                            2 Cor. 7:8                   85, 284

1 Cor. l1:18               16                                            2 Cor. 7:12                284

1 Cor. 11 : 21            109                                          2 Cor. 8: 6                  200, 411

1 Cor. 11 : 22            410, 468                                 2 Cor. 8: 9                  37, 41, 438

1Cor.11:25                105                                          2 Cor.8:10                  481

1 Cor. 11 : 26            312                                          2 Cor. 8: 11               394, 400

1Cor. 11:29               436                                          2 Cor. 9:3                   431

1 Cor.12:2                  315                                          2 Cor. 9:7                   12

1 Cor.12:15               486                                          2 Cor.9:12                  20

l Cor.12:19                273                                          2 Cor.10:9                  372

1Cor.13:2                   371                                          2 Cor. l0:15.              378

1 Cor. 14: 5               171, 252, 253                        2 Cor. 11 : 1              27

1 Cor.14:10.              259                                          2 Cor. l1:3                 224

1 Cor. 14: 12             210, 229                                 2 Cor. 11 : 6              284

1 Cor.14:39               402                                          2 Cor. 11:15              282

                      INDEX OF PASSAGES REFERRED TO.                       211


SECTION                                                                  SECTION

2 Cor. 11 : 16            273                                          Gal. 5: 17                   222

2 Cor. 11 : 25            39, 80, 88                               Gal. 5: 18                   242

2 Cor. 12:9                88                                            Gal. 6:1                      206, 285

2Cor. 12:11               284                                          Gal. 6:6                      123

2Cor. 12:14               376                                          Gal. 6:9                      436

2 Cor. 12: 17             78, 88                                     Gal. 6: 13                   124

2 Cor. 12:20             224, 475                                 Gal. 6:14                    176, 177, 476

2 Cor. 12: 21             224

2Cor.13:5                   469                                          Eph.1:9                      139

Eph.1:12                    409

Gal. l:6. 8,10,            424                                          Eph.1:16                    203

Gal 1.7                       274                                         Eph 1.17                    200

Gal. 1: 8                     278, 281, 285                        Eph. 2: 4.                   39

Gal.1:10.                    248                                          Eph. 3:14                   203

Gal.1:11.                    13                                            Eph. 4 : 18                 155

Gal. 1.19                    274                                          Eph. 5.4                     32

Gal. 1 : 22                  429, 432                                 Eph. 6 : 11                 414

Gal. 1 .23                   34, 127                                   Eph. 6.17                   295

Gal. 2:2                      227                                          Eph. 6:22                   44

Gal. 2:3                      438

Gal. 2:4                      199                                          Phil. 1:6                     60

Gal. 2: 9                     217, 385                                 Phil. 1 : 10                 409

Gal. 2:13                    236                                          Phil. 1:18                   60

Gal. 2: 14                   343                                          Phil. 1 : 23                 107, 413

Gal. 2 : 17                  176, 177, 242                        Phil. 1: 29                  481

Gal. 2:21                    243                                          Phil. 2:2.                    217

Gal. 3:2                      387                                          Phil. 2:6                     394

Gal. 3:17                    411                                          Phil. 2:7                     145

Gal. 3: 19                   332                                          Phil. 2:12                   479

Gal. 3:21                    176                                         Phil.2:17                    284

Gal. 3 : 23                  120, 406                                 Phil. 2 : 28                 44

Gal. 4:4                      293                                          Phil.3:1                      96

Gal. 4:8.                     85,485                                    Phil.3:8.                     425

Gal. 4: 11                   224, 227                                 Phil. 3: 10                  397

Gal. 4 : 15                  249                                          Phil. 3 : 11, 12           253, 276

Gal. 4: 17                   198                                          Phil. 3 :16                  364

Gal 4 : 18                   384, 481                                 Phil. 4 : 5.                  425

Gal. 4:20                    33                                            Phil. 4:8                     301

Gal. 5:2.                     250                                          Phil. 4: 11                  46, 55

Gal. 5 : 3                    124, 125, 379

Gal. 5 : 4.                   11                                            Col. 1: 3, 4                 134, 136, 439

Gal. 5:7                      402                                          Col. 1:8                      142

Gal. 5:11                    245                                          Col. 1:21                    155, 420

Gal. 5:12                    27                                            Col. 2:5                      284

Gal. 5:15                    209                                          Col. 2:6                      293

212                   INDEX OF PASSAGES REFERRED TO.


                                    SECTION                                                                  SECTION

Col. 2 : 8                    206, 424, 472                        2 Tim. 2 : 13              287

Col. 3:18                    32                                            2 Tim. 2 : 14              479

Col. 4:8                      44                                            2 Tim. 2:18                112, 390

Col. 4:12                    203                                          2 Tim. 2:25                225, 344

Col. 4:16                    205                                          2 Tim. 3:15                17

Col. 4 : 17                  205, 209                                 2 Tim. 4 : 7                74, 85

2 Tim. 4:10                137

1 Thess. 1 : 8             369                                          2 Tim. 4 : 11              134, 138, 449

1 Thess. 1 : 10.          426

I Thess. 2:8                481                                          Tit. 1:6                       469, 470

1 Thess. 2:12             412                                          Tit.1:11                      474

1 Thess. 2:16             411                                          Tit.2:8                        368

1 Thess. 3 : 5             227, 406

1 Thess. 3 : 8             247                                          Philem. 4.                  203

1 Thess. 3 : 10.          412                                          Philem. 11.                44

1 Thess. 3: 11, 12     176.                                        Philem. 13, 14           33

1 Thess. 4 : 3             386                                          Philem. 20.                176

1 Thess. 4 : 18.          237

I Thess. 5:4                218, 219                                 Heb. 2:2                     427

I Thess. 5:10              252, 253                                 Heb. 2:3                     436

1 Thess. 5:15.            206, 209                                 Heb. 2:8                     415

I Thess. 5:19              180                                          Heb. 2:10                   149

I Thess. 5:23             176                                          Heb. 2:13                   94

                                                                                    Heb. 2:17                   409    

2 Thess. 2:2               412                                          Heb. 2:18                   443

2 Thess. 2:3               166                                          Heb. 3:8                     162

2 Thess. 2:4               371                                          Heb. 3:11                   272

2 Thess. 2: 17            176                                          Heb. 3: 12         109, 206, 209, 415

2 Thess.3:5                176                                          Heb. 3:15                   415

2 Thess. 3:11             16                                            Heb. 3:18                   112

2 Thess. 3: 16.           175, 176                                 Heb. 4: 1                    224

Heb. 4:2                     84

1 Tim. 4:4                  436                                          Heb. 4:3                     272

1 Tim. 4:8                  439                                          Heb. 4:5                     272

1 Tim. 4:13.               328                                          Heb. 4:7                     475

1 Tim. 5:21                447                                          Heb.4:8                      248

1 Tim. 6 : 3                469, 470                                 Heb. 5 : 1                   124, 125

1 Tim. 6 : 17              102                                          Heb. 5:4                     125

Heb. 5:5                     375

2 Tim. 1:8, 9.            426                                          Heb. 5:8                     437

2 Tim. 1:16                176                                          Heb. 5:11                   377

2 Tim. 1:18                176                                          Heb. 5:12                   400

2 Tim. 2:2                  376                                          Heb. 6:9                     284

2 Tim. 2 : 5                260, 287                                 Heb. 6 : 10                 218, 371

2 Tim. 2 : 12              254, 255, 266                        Heb. 6 : 14.                272, 448

                    INDEX OF PASSAGES REFERRED TO.                           213


                                    SECTION                                                                  SECTION

Heb. 6: 17                  425                                          Jas. 4: 2                      108, 408

Heb.7:6                      82                                            Jas. 4:3                       408

Heb. 7: 9                    82, 383, 426                          Jas. 4 : 13, 14.           294

Heb. 7:11                   481                                          Jas. 5:15                     431

Heb.7:13                    88

Heb.7:25                    387                                          1 Pet. 1:2                    175,176

Heb. 7:27                   139                                          1 Pet. 1:24                 43

Heb. 8:3                     318                                          1 Pet. 2:16                 446

Heb. 8:5                     82                                            I Pet. 2:18                  479

Heb. 8:9                     461                                          1 Pet. 3:1                    199, 281

Heb. 9:12                   145                                          I Pet. 3:3                    479

Heb. 9:18                   88                                            1 Pet. 3:7                    409

Heb. 9: 26                  31, 249                                   1 Pet. 3: 14                259, 273, 286

Heb. 10: 2                  108, 230, 232                        1 Pet. 3 : 17               259

Heb. l0:9.                   88                                            1 Pet. 3:18                 145

Heb. 10:15                 105                                          I Pet. 4:12                  440

Heb. l0:25                  293                                          1 Pet. 4:17                 400

Heb. l0:36                  219

Heb. ll:3                     107, 411                                 2 Pet.l:2                      176

Heb. ll:5.                    390                                          2 Pet. 1:9                    469

Heb.11:15                  248                                         2 Pet. 1:17                 55

Heb. 11 : 28               80, 88                                     2 Pet. 1: 18                146

Heb.12:1                    160

Heb.12:6                    313                                          1 John 1:1                  86

Heb. 12: 25                209                                          1 John 1: 9                 218, 263

Heb.13:2                    147                                          I John1:10                 88

Heb. 13: 5.                 172, 489                                 1 John 2: 5                 79

Heb.13:9                    481                                          1 John2:7                   28

Heb. 13: 21                176                                          1 John 2: 9                 112, 390

I John 2:21                 486

Jas. 1: 1                      388                                          1 John 2: 22               473

Jas. l:4                        219                                          I John 2:27                 216, 220

Jas. l:5                        269                                          1 John 3:1                  213

Jas. 1:11                     43                                            I John 3:10                 486

Jas. l:18                      409                                          I John 3:13                 277

Jas. l:19                      413                                          1 John 3:18                475

Jas. 1: 22                    479                                          1 John 3: 22               299, 314

Jas. l:24                      43, 78, 79                               1 John 4 : 2                460

Jas. 1: 26                    261, 385                                 1 John 4: 3                 469, 470

Jas. 1:27                     386                                          1 John 4:6                  470

Jas. 2:8                       67                                            1 John 4:7                  160

Jas. 2: 10                    50, 307                                   1 John 4:9, 10           86

Jas.2:15                      250, 420                                 I John 5:12                 486

Jas. 2:16                     250                                          1 John 5:15                247

Jas. 2: 26                    293                                          1 John 5: 20               198

214               INDEX OF PASSAGES REFERRED TO.

 

                                    SECTION                                                                  SECTION

2 John 6                     213                                          Rev. 8:1                     316

2 John 8                     206, 209                                 Rev. 8:5                     80

2 John 10                   256                                          Rev. 9:4.                    212, 472

Rev. 9: 5.                   212, 312

3 John 4                     213, 460                                 Rev.9:6                      60

3 John 6                     141                                          Rev. 9:10                   378

Rev. 9:20                   218

Jude 17                       422                                          Rev. ll:7                     303

Rev. 11 : 17               54

Rev. 1:3                     423                                          Rev.12:2.                   389

Rev. 1:18                   429, 431                                 Rev.12:4                    305

Rev. 2:5                     275                                          Rev.13:12                  205, 210

Rev. 2:11                   487                                          Rev.13:13                  210, 222

Rev. 2:25                   332                                          Rev.13:16                  205

Rev. 3:2                     28                                            Rev.14:13                  124, 218

Rev. 3:3                     88                                            Rev.15:4                    172

Rev. 3:9                     205                                          Rev.15:8                    331

Rev. 3:11                   293                                          Rev.16:9                    371

Rev. 3:15.                  27                                            Rev.17:12                  52

Rev. 3:16.                  232                                          Rev.17:17                  332

Rev. 4:9                     308                                          Rev.19:3                    80

Rev. 4:11                   376                                          Rev.19:13                  75

Rev. 5:5                     371                                          Rev.20:3                    331

Rev. 5:7                     88                                            Rev.20:15                  242

Rev. 7:3                     164, 331                                 Rev.21:16                  293

Rev. 7:14.                  80                                            Rev.21:23                  216

Rev. 7:16                   487                                          Rev.22:14                  218


                                 II. OLD TESTAMENT.

Gen. 2:19                   315                                          Gen. 43:5                   139

Gen. 3:1                     167                                          Gen. 43:6                   122

Gen. 3:22                   398                                          Gen. 43:11                 227

Gen. 4:14                   256                                          Gen. 44:15                 69

Gen. 4:15                   148                                          Gen. 44:26                 256

Gen. l0:19                  417

Gen. 16:3                   400                                          Deut.8:3                     68

Gen. 19: 16                109, 415                                 Judg. 11 : 9                252

Gen. 19: 21                398

Gen. 20:7                   256                                          1 Sam. 2:13, 14         315

Gen. 34:15                 415                                          1 Sam. 2:24.              400

Gen. 34:17                 398                                          1Sam.12:19               141

Gen. 34: 19                401                                          1 Sam. 12 : 23.          375, 404

Gen. 34: 22                398                                          1 Sam. 22 : 13.          375


                          INDEX OF PASSAGES REFERRED TO.                215


SECTION                                                                  SECTION

Ps. 78: 18, Heb. (LXX. 77 : 18) 375                      Jer. 2 : 19                   56

Ps. 101: 16, LXX. (Heb.102: 16) 65

                                                                                    Dan. 2: 26, 27           145

Eccl.4:13,17;             404                                          Dan. 8:13                   145

Dan. 3:24                   146

Isa. 5:14                     898                                          Dan. 8:26, 27            146

Isa. 6:9                       167

Isa. 42:1                     56                                            Mal. 2:17                   66

 

III. APOCRYPHA, PSEUDEPIGRAPHA, ETC.

1 Esdr. 4: 26, 27       52                                            Ps. Sol. 2 : 29.           401

Ps. Sol. 2 : 39, 40.    376

1 Macc. 3: 16.           404

1 Mace. 6: 13.           47                                            Ev. Pet. 23. 20,         88

I Macc. 9:10              177                                          Ev.Pet.81                   88

1 Macc. 12 : 22         122

Mart. Polyc. 8: 2, 8 37 f.n.

Ps. Sol. 1: 1               416                                          Mart. Polyc. 10: 1     217

Ps.Sol. 1:8                 416

Ps. Sol. 2: 28             375, 401, 406                        Jos. Ant. 10. 4.2. 461

 

 

 

 

                                    PRINTED IN THE U.S.A.