1142

Addenda 3rd ed.

CHAPTER XXI

PARTICLES ( AI PARAQHKAI)

I. Scope. The word particle is a Latin diminutive, particula (cf. French particule) from pars. It is a small part of something. Longinus terms this part of speech paraqh,kh with the notion that it was a word placed beside another. No portion of syntax is treated with so little satisfaction in the grammars. The grammarians are not agreed as to what parts of speech should be called "particles." Riemann and Goelzer¹ treat under this term (Les Particules) negative particles, particles of comparison and prepositions. Jannaris² includes prepositions, conjunctions and negative particles. Kuhner-Gerth³ here discuss conjunctions, prepositions and the modal adverbs, though they use the phrase "die sogenannten Partikeln." Blass⁴ almost confines the discussion of particles to conjunctions. He makes the two terms equivalent: "Particles (Conjunctions)." Winery uses the word broadly to cover all adverbs, prepositions and conjunctions. Monro⁵ limits the designation to certain conjunctions and adverbs "that are mainly used to show the relation between other words and between clauses." But he does not treat all conjunctions (paratactic and hypotactic) nor all modal adverbs. He passes by prepositions. Brugmann⁶ sees clearly that, as there is no real distinction between adverbs and prepositions, so there is no fast line ("keine feste Grenze") between "particles" and other adverbs. All languages have a large group of words that pass over into the category of particles, but Brugmann cuts the Gordian knot by declaring that it is not a function of scientific grammar to delimit these words. That is a matter of subjective standpoint. He takes little interest in the various subdivisions of the particles, but he extends the term to its widest sense to

PARTICLES ( AIPARAQHKAI) 1143

cover all modal adverbs, prepositions and conjunctions. Brugmann notes that many of these particles go back to the IndoGermanic time and hence their etymology is unknown. He treats the particles from the standpoint of their origin so far as known. Hartung⁷ takes a much narrower view of particles. He discusses the paratactic conjunctions and the intensive particles. He⁸ conceives that the greater portion of the particles have no meaning in themselves, but are merely modifications on other words or on whole sentences. This is not strictly correct. We are not always able to discover the original import of these words, but it is probable that they originally had a definite meaning. It is true that the particles are all subordinated to other words in various ways. In a broad way it may be stated that there are four classes of words (verbs, nouns, pronouns, particles) in the sentence. From this point of view the word particle covers all the adverbs, prepositions, conjunctions and interjections. But it is impossible, as Brugmann holds, to make a perfectly scientific treatment of the particles without much overlapping. The interjections in one sense do not belong to grammar. The negative and the interrogative particles cannot be properly treated under adverbs, though they are adverbs. So also conjunctions are adverbs, but a good deal more. Intensive particles again are adverbs, but more. It is not worth while to recount the story of the adverbs and the prepositions at this stage. They are particles, but they have received sufficient discussion in special chapters. In the same way the construction of hypotactic conjunctions came in for somewhat careful treatment in connection with subordinate sentences under Mode. Hence, hypotactic conjunctions do not here demand as much discussion as the paratactic conjunctions. One has to be, to a certain extent, arbitrary in this field, since the ground is so extensive and so much remains to be clone. There is still need of a modern and exhaustive treatise on the Greek Particles. It was in 1769 that the Dutch scholar Hoogeveen⁹ wrote his book. He was followed by Hartung.¹⁰ Klotz¹¹ reworked the writings of Devarius. In

1144 A GRAMMAR OF THE GREEK NEW TESTAMENT

1861 Baumlein produced his Untersuchungen uber griech. Partiheln. Paley¹² has carried the work on, as has Navarre.¹³ There are, to be sure, a great number of monographs on special groups or on single particles.¹⁴ "If any particular section of Greek grammar were taken as a specimen to illustrate the historical evolution of the Greek language, no better representative could be selected than the section of the particles."¹⁵ Jannaris speaks thus, not because the grammars have treated the particles with such skill, but because the particles best show the growth and decay of parallel words before other new synonyms that are constantly coming into existence. The particles come to a sharp point and gradually lose the edge and whittle down into platitudes. Then they give way to others with more freshness. In general, the particles mark the history of the effort to relate words with each other, clause with clause, sentence with sentence, paragraph with paragraph. They are the hinges of speech, the joints of language, or the delicate turns of expression, the nuances of thought that are often untranslatable. We must here confine our attention to Intensive Particles, Negative Particles, Interrogative Particles, Conjunctions and Interjections. This order is chosen for logical reasons simply, not because this was the order of development. That we do not know. The particles that are linked to single words logically come before conjunctions which have to do with clauses and sentences. Interjections stand apart and so are put last in the list. Some of the particles are employed with words, clauses and sentences (like a;raà de,à ou=n), so that a strict division on this basis is not possible.¹⁶

II. Intensive or Emphatic Particles ( paraqh/kai evmfatikai, or paraplhrwmatikoi. su,ndesmoi according to Dionysius Thrax).

1. LIMITATIONS. Here again there is no absolute agreement as to what particles are considered "emphatic" or "intensive." Winer, indeed, has no separate discussion of the intensive particles like geà per. He admits¹⁷ that, while the Greek of the N. T. uses adverbs well in an extensive sense, it is defective in the intensive use. Adverbs of place, time, manner, all come in abundance in the N. T. Thompson¹⁸ follows Winer in the absence of discussion of the intensive particles. The intensive particles, in

PARTICLES ( AI PARAQHKAI) 1145

Addenda 2nd ed.

fact, as a rule receive poor handling in the grammars.¹⁹ But Paley²⁰ properly sees that they are "an elaborately finished part of a most complex and beautiful machinery." Poetry, especially tragic poetry, uses these emphatic particles more than other kinds of writing. In Homer "they sustain and articulate the pulses of emotion. By them alone we can perceive that Greek was the language of a witty, refined, intellectual, sensitive and passionate people. It would be impossible in any book to tabulate the delicate shades of meaning, the subtle, intricate touches of irony or pathos, the indescribable grace and power which the particles lend to many of the grandest passages in ancient literature."²¹ It is only by a close study of the entire context that these can be felt. They can never be fully translated from one language to another. Thus it is impossible to reproduce in English the various shades of meaning of me,n and de, when in contrast. "The attempt to translate a particle leads to curious results. Dr. Cyril Jackson used always to render Trw/e,j r`a by 'the Trojans, God help them,' and a former head-master of Eton always distinguished between soi, 'Sir, to you', and toi, 'at your service' (Coleridge, Greek Classic Poets, p. 221).<sup>22</sup> Indeed, it is not possible to put into mere written language all that the look, the gesture, the tone of voice, the emphasis of the accent carried when heard and seen. Cf. a Frenchman in conversation. The spoken vernacular thus has all the advantage of the written style. All the vernacular cannot be reproduced on the page. Cf. the charm of the actual speech of Jesus and Paul. The N. T. is in the vernacular koinh,, but even so it does not reproduce to any great extent the witchery of the old Greek particles. Time has worn them down very much. Still, we do find them here and there. There is a good example in Ph. 3:8, avlla. me,n ou=n ge kai. h`gou/mai. So also ei; pwj h;dh pote, (Ro. 1:10) and ti, e;ti kavgw. w`jgrk grk(3:7). Cf. P. B. M. 42 (B.C. 168) ouv mh.n avll v evpei. kai, and 0. P. 1164, 5 (vi/vii A.D.) ouv mh.n de. avlla. kai,. This shows that Paul at least knew how to indicate the finer shades of thought by means of the Greek particles. Blass²³ notes that, in comparison with the Semitic languages, the N. T. seems to make excessive use of the particles, poor as the showing is in comparison with the classic period. "Modern Greek has lost the classical Greek wealth of connective and other particles which lend nicety and

1146 A GRAMMAR OF THE GREEK NEW TESTAMENT

precision of thought. Only kai, ( ou;teà ouvde,%, h; and the less commonly used conjunctions avlla,à plh,nà o[mwj have been retained. The loss of ga,rà a;ra has been compensated by new formations; but the ancient Greek te,à de,à me,n- de,à me,ntoià mh,nà ou=n ( gou/n), e;tià dh,à ge,à pe,r have left no successors" (Thumb, Handb., p. 185). The papyri seem barren of intensive particles in comparison with the older Greek. Jannaris²⁴ observes how these postpositive particles ( ge,à dh,à me,n pe,rà toi, and their compounds) tend in the later Greek either to disappear or to become prepositive. The N. T. is in harmony with this result. The same thing occurs with a;ra, which sometimes becomes prepositive, but that is not true of ga,rà de, ou=n. Dionysius Thrax²⁵ has a very extensive list of "expletive particles" or paraplhrwmatikoi. su,ndesmoi ( eivsi. de. oi[de\ dh,à r`a,à nu,à tou/à toi,à qh,n a;rà dh/taà pe,rà pw,à mh,nà a;nà au=à nu/nà ou=nà ke,nà ge,à avlla,à mh,nà toi,nunà toigarou/n). Some of these (like a;raà ou=nà avlla,, and one might add ga,rà de,% are so prevailingly conjunctival that they are best treated under conjunctions. Others (like ke,nà r`a,) belong to earlier stages of the language. The discussion of a;n could have come here very well, since it is undoubtedly intensive whatever its actual meaning, whether it is blended with eiv into eva,n or used with o[jà o[stijà i[naà o[pwjà w`j, etc., or used with the verb itself in the apodosis of a condition. It is a modal adverb of emphasis (now definite as in Rev. 8:1, now indefinite as in Mt. 23: 18). It is like a chameleon and gets its colour from its environment or from its varying moods. This fickleness of meaning is true of all the intensive particles. Indeed, Dionysius Thrax is rather slighting in his description of these words, o[soi paro,ntej ouvde.n wvfelei/n du,nantai ou;te mh.n cwrisqe,ntej lumai,nontai. He contradicts his disparagement by the use of mh,n in this very sentence.

The adverbial nature of the intensive particles is well shown by the variety of usage of the modal adverb ou;twj. See Thayer's Lexicon for the N. T. illustrations, which are very numerous (some 200). In Jo. 4:6, evkaqe,zeto ou[twj evpi. th|/ phgh|/, we have a good example of the possibilities of ou[twj. The local adverb pou, dwindles from 'somewhere' (Heb. 2:6) to 'somewhat' in Ro. 4:19. Cf. also dh, pou ('surely') in Heb. 2:16. Some of the temporal adverbs also at times approach the emphatic particles. Cf. to. loipo,n in Ph. 3:1; 4:8 (see Kennedy in loco) almost²⁶ = ou=n. But in the N. T. a;rti and h;dh are always strictly temporal. How-

PARTICLES ( AIPARAQHKAI) 1147

ever, pote, sometimes loses its notion of 'once upon a time' (Gal. 1:23) and fades into that of 'ever' as in 1 Cor. 9:7; Eph. 5: 29. In h;dh pote, (Ro. 1:10; Ph. 4:10) it is more the notion of culmination ('now at last') than of time. But in mh, pote the notion of time may be wholly gone before that of contingency ('lest perchance'), as in Lu. 12:58. In the N. T. we find undoubted instances of the non-temporal use of nu/n and nuni, where the sense differs little from dh, or ou=n. Some of the passages are in doubt. But the logical and emotional use, as distinct from the temporal, is clear in Jo. 15:22, 24 where nu/n de, gives the contrast to the preceding conditions, 'but as it is.' Cf. also 1 Jo. 2:28, kai. nu/n tekni,a, where John's emotional appeal is sharpened by the use of nu/n. Cf. likewise kai. nu/n deu/ro in Ac. 7:34 (LXX). Cf. kai. nu/n, B. U. 530 (i/A.D.). In general, the N. T. language, like the English, leaves most of the emotion and finer shades of thought to be brought out by the reader himself. "The historical books of the N. T., and especially their dialogues and discourses, are only fully and truly intelligible to us in reading them in high voice in the original Greek text, and in supplying the intonation, the gestures, the movement, that is to say, in reconstituting by the imagination the scene itself."²⁷

2. THE N. T. ILLUSTRATIONS.

(a) Ge,) We may begin with ge,. The origin of ge, is by no means certain. In the Boeotian, Doric and Eleatic dialects it is ga,. It seems to correspond²⁸ to the k in the Gothic mi-k (German mi-ch). Cf. Greek evme,Äge. Brugmann sees also a kinship to the g in the Latin ne-g-otium, ne-g-legere, ne-g-are. Hartung²⁹ connects it with the adverb za,) It may also be the same word as the Vedic Sanskrit gha, which is used in the same way.³⁰ Cf. further qui in the Latin qui-dem. It is not so common in the koinh, as in the classic Attic (Radermacher, N. T. Gr., p. 29). Its function is to bring into prominence the particular word with which it occurs. It is enclitic and so postpositive. The feelings are sharply involved when ge, is present. It suits the Greek,³¹ which "delights in pointed questions, irony and equivocal assent." But there is no English equivalent and it frequently cannot be translated at all. Hartung³² sees in ge, a comparative element, while

1148 A GRAMMAR OF THE GREEK NEW TESTAMENT

kai, is cumulative and arithmetical. As a matter of fact, ge, brings to the fore the idea of the word with which it is used, but adds no distinctive notion of its own.³³ Hubner³⁴ calls it a concessive particle on a par with o[mwj. But that is not always true of ge,. The distinction made by ge, may be either the least important or the most important (Thayer). The resultant idea may be 'at least,' this much if no more, a concessive notion. We find this to be the significance of ge, in Lu. 11:8, dia, ge th.n avnaidi,an auvtou/. Here, however, the ge, more properly belongs to avnaidi,an, since that is the point, not the preposition dia,) The same slight variation from the classic idiom appears in 18: 5, dia, ge to. pare,cein moi ko,pon th.n ch,ran tau,thn. The concessive minimizing idea comes out clearly in Jo. 4:2, kai,toige vIhsou/j auvto,j. See further a;ra ge and kai, ge in Ac. 17:27, and, in particular, avlla, ge u`mi/n eivmi, (1 Cor. 9:2) where again the ancient idiom would prefer u`mi/n ge, 'to you at least' (if not to others). Once more note ei; ge in Eph. 3:2; 4:21; Col. 1:23, and eiv de. mh, ge in Mt. 6:1; 9:17, etc. There is a keen touch of irony in Ro. 9:20, w= a;nqrwpeà menou/nge su. ti,j ei=* Cf. a;rage in Mt. 17:26. On the other hand ge, means 'this much,' 'as much as this,' in other contexts. So in Lu. 24:21, avlla, ge kai. su.n pa/si tou,toij, where the ascensive force is accented by kai,à su,n and avlla, (affirmative here, not adversative), and the climax of the crescendo is reached in ge,. The same climacteric force of the particles occurs in Ph. 3:8, avlla. me.n ou=n ge kai. h`gou/mai pa,nta zhmi,an ei=nai. 'I go,' says Paul, 'as far as to consider all things to be loss.' Cf. a;rage in Mt. 7:20 and kai, ge in Ac. 2:18 (Joel 3:2). So we have a=ra, ge in Ac. 8:30. A fine example is o[j ge tou/ ivdi,ou ui`ou/ ouvk evfei,sato (Ro. 8:32). So 10:18. There is irony again in kai. o;felo,n ge evbasileu,Ä sate (1 Cor. 4:8), and note the position of ge, apart from kai,. In Homer ge, is very common with the pronouns,³⁵ but in the N. T. we have only o[j ge (Ro. 8:32). We no more find e;gw ge, but evgw. me,n (Mt. 3:11), evgw,ÄÄsu,grk grk(3:14), evgw. de,grk grk(5:22), auvto.j evgw, (Ro. 9:3). Indeed all of the thirty examples of ge, in the N. T. occur with conjunctions (paratactic or hypotactic) or other particles except those in Lu. 11:8; 18:5; Ro. 8:32. Cf. a`marti,a ge, evstin ('indeed it is sin') in Hermas, Vis., i, 1.8. The particles with which ge is found in the N. T. are avlla, ge (Lu. 24:21); a;ra ge (Mt. 7:20); a=ra, ge (Ac. 8:30); ei; ge (Eph. 3:2); eiv de. mh, ge (Mt.

PARTICLES ( AIPARAQHKAI) 1149

6:1); kai, ge (Ac. 17:27); kai,toige (Jo. 4:2); mh,tige (1 Cor. 6: 3); o;felon ge (1 Cor. 4:8); menou/nge (Ro. 9:20). Cf. dia, ge in Lu. 11:8; 18:5. Ga,r is compounded of ge, and a;ra, but it will be treated under conjunctions, though it is sometimes not much more than an intensive particle. Cf. ti, ga.r kako.n evpoi,hsen (Mt. 27:23).

(b) Dh,. It has likewise an uncertain etymology.³⁶ It appears in the Attic poets as dai, (cf. nh,à nai,) and is seen in composition with dh/Ätaà dh,Äpouà evpeiÄdh,à h;Ädh)³⁷ In h;Ädh we probably have³⁸ h= and dh,. It was originally temporal in idea and goes back to the Indo-Germanic period. Jannaris³⁹ thinks de, that and dh, are one and the same word (cf. me,n and mh,n) and holds that the difference is due to the transliteration from the old to the new alphabet when alone a distinction was made between e and e ( h). Thus the spelling dh, was confined to the intensive particle, while de, was the form for the conjunction. It is certain that in Homer there is confusion between and before vowels.⁴⁰ In Homer also dh, may begin a sentence, but in the N. T. as elsewhere all the examples are postpositive (but not enclitic). Blass⁴¹ does not treat it as an intensive particle, but as a consecutive particle. It is hard to follow Blass' theory of the particles. Like the other intensive particles it has no English or German equivalent and is a hard word to translate. It is climacteric and indicates that the point is now at last clear and may be assumed as true.⁴² Cf. Latin jam nunc, nu/n- h;dh (1 Jo. 4:3); h;dh pote, (Ro. 1:10). The similarity in sense between dh, and one usage of de, may be seen in Ac. 6:3, evpiske,Ä yasqe de, $dh,%, where W. H. put dh, in the margin. Cf. kai. su. de, in Lu. 1:76. Dh, is not genuine in 2 Cor. 12:1. There are left only six N. T. illustrations, counting dh, pou in Heb. 2:16, ouv ga.r dh, pou avgge,lwn evpilamba,netai. In Mt. 13:23, o[j dh. karpoforei/, it occurs in a relative sentence, 'who is just the man who.'⁴³ The other examples are all with the hortatory subjunctive (Lu. 2: 15; Ac. 15:36) or the imperative (Ac. 13:2; 1 Cor. 6:20) in accord with the classical idiom. There is a note of urgency in avfori,sate dh, (Ac. 13:2) and doxa,sate dh, (1 Cor. 6:20). The passage with dh, pote in Jo. 5:4 has disappeared from the critical text.

1150 A GRAMMAR OF THE GREEK NEW TESTAMENT

(c) Ei= mh,nà nh, and nai,. Somewhat akin to the positive note in dh, is the use of h= mh,n which is read by many MSS. in Heb. 6:14. The etymology of this adverb is again quite uncertain, though it is possible that it may have the same root as h; ( h=F eà h`F e,%.<sup>44</sup> Cf. h= dh, ( h;dh). In h;per (Jo. 12:43) and h;toi (Ro. 6:16) we have the comparative or disjunctive h;. In Homer it was often used in connection with other particles.<sup>45</sup> We may pass mh,n for the present. If h= were genuine in Hebrews the usage would be in strict accord with classic construction for a strong asseveration. But certainly ei= mh,n is the true text. This queer idiom appears a few times in the LXX (Ezek. 33:27; 34:8; 38:19, etc.). It occurs also in the papyri and the inscriptions<sup>46</sup> after iii/B.C. Cf. ei= mh,n P. Oxy. 255 (A.D. 48). So that it is mere itacism between h= and ei=. The Doric has ei- for h= where Moulton<sup>47</sup> holds against Hort<sup>48</sup> that the distinction is strictly orthographical. See further chapter VI, Orthography and Phonetics, ii, (c). So then ei= mh,n has to be admitted in the koinh, as an asseverative particle. It is thus another form of h= mh,n. Jannaris<sup>49</sup> gives a special section to the "asseverative particles" nh, and ma,. We do not have ma, in the N. T. and nh, only once in 1 Cor. 15:31, kaq v h`me,ran avpoqnh,skw nh. th.n u`mete,ran kau,chsin) Nh, is a peculiarity of the Attic dialect and is used in solemn asseverations (oaths, etc.) and means 'truly,' 'yes.' It is probably the same word as nai,, the affirmative adverb which occurs over thirty times in the N. T. Nai, may be simply 'yes,' as in Mt. 13:51. It may introduce a clause as 'yea' or 'verily,' as in Mt. 11:9. It is used in respectful address, Nai,à Ku,rie (Jo. 11: 27). It may be used as a substantive (like any adverb) with the article (2 Cor. 1:17) or without the article (Mt. 5:37), where it is repeated. It occurs with avmh,n in Rev. 1:7. It stands in contrast with ouv in Mt. 5:37 and 2 Cor. 1:17. There was an old form nai,Äci (cf. ouvÄci,). But we do not know the etymology, though Brugmann⁵⁰ compares it with the Latin ne and nae and possibly also with the old Indo-Germanic naÄna ('so - so').

(d) Me,n. We know a little more about which is postpositive, but not enclitic. It is only another form of mh,n which occurs in the N. T. only in Heb. 6:14. The Doric and Lesbian use ma,n and the Thessalian ma,ÄÄde,) So then it seems probable⁵¹ that ma,n

3 Moulton, Prol., p. 46.

PARTICLES ( AI PARAQHKAI) 1151

( ma, used with words of swearing after a negative), mh,n and me,n are one and the same word. Indeed, in Homer⁵² all three forms occur in the same sense. That original sense is affirmative, meaning 'surely,' 'indeed,' 'in truth.' It is overrefinement to find in me,n ( mh,n) the subjective confirmation and in dh, the objective attestation.⁵³ It is probable that in the change from the old alphabet to the new the transcribers adopted the two ways of spelling, common in Attic and Ionic ( me,n and mh,n) with a notion that mh,n was merely emphatic with single words, while me,n was correlative (forwards or backwards) or antithetical.⁵⁴ Questions of metre may also have entered into the matter. But there is no doubt at all that in itself me,n does not mean or imply antithesis. The original use was simply emphatic confirmation of single words, usually the weightiest word in the sentence. This use was gradually left more and more to mh,n and other particles, but it is not anacoluthic, as Winer⁵⁵ holds, for me,n to occur without the presence of de, or avlla,. The older language is naturally richer⁵⁶ in this original idiom with mh,n, but it survives in the N. T. and is not to be regarded as unclassical or uncouth. For an example in the papyri see B. U. 423 (ii/A.D.), pro. me,n pa,ntwn. The old idiom survived best in the vernacular and in poetry, while the literary prose was more careful to use the antithetical or resumptive me,n. This me,n, solitarium, as the books call it, may have a concessive or restrictive force.⁵⁷ Cf. eiv me.n ga.r o` evrco,menoj (2 Cor. 11:4), where there is no thought of de, or avlla,. It is seen also rather often in the Acts. Cf. 1:18 ou-toj me.n ou-n evkth,sato cwri,on,grk grk(3:13) oa}n u`mei/j me.n paredw,kate (cf. u`mei/j de, in next verse which is copulative, not adversative),adversative adversative(3:21) oa}n dei/ ouvrano.n me.n de,xasqai,grk grk(3:22) Mwush/j a;nqrwpoj me,ngrk grk(17:12) polloi. me.n ou=n evx auvtw/n evpi,steusan,grk grk(21:39) evgw. a;nqrwpoj me,n eivmi,grk grk(23:18) o` me.n ou=n paralabw,n (cf. also 23:31), (27: 21) e;dei me,n,grk grk(28:22) peri. me.n ga.r th/j ai`re,sewj tau,thj, and the instances of oi` me.n ou=n Acts 1:6; 2:41; 5:41; 8:25, where no contrast is intended. See eiv me,n ou=n in Heb. 7:11; h` me.n euvdoki,a in Ro. 10:1; evf v o[son me.n ou=n eivmi. evgw, in 11:13. Cf. 2 Cor. 12:12; 1 Th. 2:18, evgw. me,n) Cf. also the single instance of menou/n as one word (Lu. 11:28) which is obviously without contrast. The same thing is true of menou/nge (Ro. 9:20; 10:18; Ph. 3:8) however it is printed. The main word is sharpened to a fine point and there is a hint of contrast in Ph. 3:8. Indeed, most

1152 A GRAMMAR OF THE GREEK NEW TESTAMENT

of the instances of me,n ou=n in the N. T. are resumptive, not correlative or antithetical.⁵⁸ There remain the instances where me,n implies contrast. It is just a step in advance of the original idiom. Cf. Mt. 8:21, evpi,treyo,n moi prw/ton avpelqei/n, where there is nothing to correspond to prw/ton. The e;peita is involved in what precedes. So with prw/ton and te- kai, in Ro. 1:16 and prw/ton - kai, in 2 Cor. 8:5. The kai, does not answer to the prw/ton.⁵⁹ Just so we have to.n me.n prw/ton lo,gon in Ac. 1:1 without a deu,teron de, though the clear implication is that the Acts is the second book. In 1 Cor. 11:18, prw/ton me.n ga,r, the contrast is implied⁶⁰ in verses 20 ff., but in Ro. 1:8, prw/ton me.n euvcaristw/, there is no hint of other grounds of thanksgiving. This instance may be a change of thought on Paul's part (anacoluthon), or it may be the original use of me,n, meaning 'first of all in truth.' Cf. prw/ton me,n in Ro. 3:2. In Ro. 7:12, o` me.n no,moj, there is no contrast stated, but in verse 14 it is given by de,Ã yet without me,n. In Col. 2:23, a[tina, evstin lo,gon me.n e;conta sofi,aj, the antithesis is really stated in ouvk evn timh|/Ã ktl) without an adversative particle. In 1 Cor. 5:3 the me,n stands alone, while avpw,n and parw,n are contrasted by de,) In Heb. 12:9 there is contrast between the me,n clause and the next, which has no particle (only polu. ma/llon). In Ac. 26:4, 6, me,n is followed by kai. nu/n by way of contrast and by ta. nu/n in 17:30. Cf. me,n - kai. in 1 Th. 2:18, me,n - te, in Ac. 27:21, where there is practically no contrast. But see oa} me,n- kai. e[teron in Lu. 8: 5 ff., oa} me,n- kai. a;llo in Mk. 4:4 ff. We have me,n- e;peita in Jo. 11:6; Jas. 3:17; 1 Cor. 12:28. These are all efforts to express antithesis. We see this also in me,n- plh.n in Lu. 22:22 and in me,n- avlla, in Ac. 4:16; Ro. 14:20; 1 Cor. 14:17. In Mk. 9: 12 f. avlla,, is independent of the me,n. But it is the me,n- de, construction that is the most frequent in the N. T. as in the Attic Greek. There are two and a half pages of examples of me,n in its various uses in the N. T. given in Moulton and Geden's Concordance, but even so the particle has made a distinct retreat since the Attic period.⁶¹ It is wholly absent from 2 Peter, 1, 2 and 3 John, 2 Thessalonians, 1 Timothy, Titus (critical text) and Revelation. It occurs thrice in Jude, only once in Eph. Eph.(4:11), Col. Col.(2:23), 1 Th. 1 Th.(2:18), Jas. Jas.(3:17). It is most frequent in Matthew, Acts,

PARTICLES ( AI PARAQHKAI) 1153

Romans, 1 Corinthians and Hebrews. Paley⁶² thinks that me,n and de, may contain the roots of one ( mi,a) and two ( du,o). But certainly the correlative antithesis is not necessary to either of them, though with de, there is the notion of addition. Cf. in this connection me,n - kai, (Mk. 4:4; Lu. 8:5) and to,te me,n (Jo. 11:6). There are varying degrees of contrast where me,n and de, occur together. There may be no emphasis on the me,n and very little on thewhich is not essentially adversative. The me,n may preserve almost its original idiom while de, has slight contrast. So Lu. 11:48, a;ra ma,rture,j evste kai. suneudokei/te toi/j e;rgoij tw/n pate,rwnà o[ti auvtoi. me.n avpe,kteinan auvtou/j u`mei/j de. oivlkodomei/te) The whole sentence is quoted to show that it is agreement (correspondence), not opposition, that is here accented. In verse 47 we have 8, but not me,n, which is hardly felt in 48. See also Ac. 13:36 f.; Ph. 3:1; Heb. 7:8. In particular we note this slight contrast when a whole is distributed into its parts as in Mt. 25:14 ff.; 1 Cor. 9:25. Cf. also Ac. 18:14 f. But the distribution may amount to sharp division, as in 1 Cor. 1:12, vEgw. me,n eivmi Pau,louà vEgw. de. vApollw,à vEgw. de. Khra/à vEgw. de. Cristou/. It is thus the context that decides how pointed is the contrast. It is not the words me,n and de, that inherently mean opposition. Indeed, the contrast may be indicated by de, alone as in Mt. 5:22, 28, 32, 34, 39, 44; 25:46; Ac. 12:9; Heb. 2:8; 4:13; 6:12.<sup>63</sup> We see a good illustration of clear antithesis in John's words about his baptism and that of Christ in Mt. 3:11, evgw. me,n- o` de,) See further 20:23; 22:8; 23:28; 25: 33, kai. sth,sei ta. me.n pro,bata evk dexiw/n auvtou/ ta. de. evri,fia evx euvwnu,mwn. The examples are numerous. See oi` me,n - oi` de, (Ph. 1:16 ff.); oua}j me,n- oua}j de, (Jude 1:22); tine.j me,n- tine.j de, (Ph. 1:15); eivj me,n- eivj de, (Heb. 9:6 f.); oi` me,n- a;lloi de, (Mt. 16:14); a;llh me,nÄÄa;llh de, (1 Cor. 15:39); tou/to me,n- tou/to de, (Heb. 10:33); prw/ton me,n ÄÄÄe;peita de, (Heb. 7:2); eiv me.n ou=n- eiv de, (Ac. 19:38 f.); eiv me,n- nu/n de, $Heb. 11:15 f.), etc. These examples fairly exhibit the N. T. usage of me,n. It is often a matter of one's mood how much emphasis to put on me,n and de,à as in Mt. 9:37 and Mk. 14:38. In me,ntoi there is always strong contrast. As examples of me,n - avlla, in sharp contrast see Ro. 14:20; 1 Cor. 14:17. So also me,n- plh,n (Lu. 22:22).

(e) Pe,r. It is probably a shortened form of peri, (cf. perfect) or pe,ri more exactly.⁶⁴ It is both postpositive and enclitic and is usually in the N. T. printed as a part of the word with which it

1154 A GRAMMAR OF THE GREEK NEW TESTAMENT

Addenda 2nd ed.

occurs. But in Homer this is not true, while pe,r follows kai, only once.⁶⁵ There is no doubt about the etymology of this particle.⁶⁶ Some⁶⁷ even connect it directly with pe,ran or pe,ra. Cf. peraite,rw (critical text in Ac. 19:39). But this idea does not conflict with the other, for pe,ri, is the locative of pe,ra. It is an IndoGermanic root and the original notion of pe,ri occurs in periÄ pi,mplhmià periÄplhqh,j, nu-per, per-manere, per-tinax, sem-per, etc. It means then to do a thing to the limit (beyond), thoroughly. There is a note of urgency in pe,r. It is intensive as ge, but probably tends to be more extensive also.⁶⁸ Sometimes the emphasis in pe,r is in spite of opposition⁶⁹ as in kai,per which occurs six times in the N. T. (Ph. 3:4; Heb. 5:8; 7:5; 12:17; 2 Pet. 1:12), and always with participles, as kai,per w;n ui`o,j (Heb. 5:8). The Textus Receptus has o[nper in Mk. 15:6, but W. H. read only o[n, but dio,per appears twice as an inferential conjunction (1 Cor. 8:13; 10:14). See a;sper, 0. P. 1125, 6 (iii/A.D.). The other examples are all with conjunctions, as eva,nper (Heb. 3:14; 6:3); ei;per (a half-dozen times, all in Paul, as Ro. 8:9; 1 Cor. 15:15); evpei,per (some MSS. in Ro. 3:30, but the best MSS., as W. H. give, have ei;per); evpeidh,per (only Lu. 1:1); h;per (only the critical text in Jo. 12:43); kaqa,per (some 17 times, all in Paul save Heb. 4:2), kaqw,sper (Heb. 5:4 and a varia lectio in 2 Cor. 3:18), w[sper (some 36 times, chiefly in Matthew, Luke and Paul, as Mt. 6:2), w`sperei, (once only, 1 Cor. 15:8).

(f) Toi, does not occur alone in the N. T., but only in composition. It is enclitic as in h;toià kai,toià me,ntoi, but it comes first in toigarou/n and toi,nun. The etymology is not certain. Brugmann⁷⁰ takes it to be a fixed form of the ethical dative soi, ( toi,). Others⁷¹ take it as the locative of the demonstrative to,. Kuhner-Gerth⁷² consider it the locative of the indefinite ti.. There seems no way of telling for certain. But it seems to have the notion of restriction and in Homer⁷³ is often combined with adversative particles. In the N. T. we find h;toi, once (Ro. 6:16), kai,toi twice (Ac. 14: 17; Heb. 4:3), kai,toige once (Jo. 4:2), me,ntoi eight times, five in John's Gospel as Jo. 4:27 and once in Paul (2 Tim. 2:19), toigarou/n twice (1 Th. 4:8; Heb. 12:1), toi,nun three times (Lu. 20:25; 1 Cor. 9:26; Heb. 13:13). [Omwj is an adversative par-

PARTICLES ( AIPARAQHKAI) 1155

tide that occurs three times in the N. T. (Jo. 12:42, here with me,ntoi; 1 Cor. 14:7; Gal. 3:15), twice with a participle.

III. Negative Particles ( sterhtikai. paraqh/kai). The use of the negative particles has been discussed already in various parts of the grammar in an incidental way in connection with the modes, verbal nouns and dependent clauses. But it is necessary at this point to treat the subject as a whole. It is not logical negative that one, has here to deal with. Many words are negative in idea which are positive in form. Thus "empty" is negative, "cold" is negative, "death" is negative. Aristotle uses sterhtiko,j for this negative conception. It is in reality an ablative idea as stere,w implies. But the grammarian is concerned simply with those words that are used to make positive words (or clauses) negative. This is the grammatical negative. There are, indeed, in Greek, as in English, negative post-fixes.⁷⁴ But there is a common negative Greek prefix av( n) called alpha privative, Sanskrit a(n), Latin in, Gothic un, English un. In Sanskrit this prefix does not occur with verbs and is rare with substantives. It is there found chiefly with adjectives and participles.⁷⁵ In Greek it occurs with verbs, but chiefly denominative verbs like avtima,zw.⁷⁶ The use of av- ( avn- before vowels) is in the Greek still more common with adjectives and verbals. See the chapter on Formation of Words for details. Cf. avdo,kimojà avdiki,aà avpeiqh,jà avsu,netojà avsu,nqetojà a;storgojà avneleh,mwn (Ro. 1:28-30).

1. THE OBJECTIVE Ouv AND ITS COMPOUNDS.

(a) Origin. This is unknown. Hubschmann⁷⁷ sees a connection with the Latin haud as do other scholars.⁷⁸ Fowler⁷⁹ takes it as an original intensive particle like pas in the French ne pas and - ci, (Indo-Ger. -ghi) in ouvÄci,) The Zend ava is also noted and the Latin au (au-fero).⁸⁰ But there is no doubt that a in the Greek took the place of the Sanskrit na, Latin ne- (ne-que, ne-scio; the relation of ne ne-quidem, ne-quam to this ne is not known), Gothic ni. The use of the Greek ou corresponds to the Sanskrit na.

1156 A GRAMMAR OF THE GREEK NEW TESTAMENT

(b) History. As far back as Greek goes we find ouv, but a did not hold its own with mh, in the progress of the language. Within the past century ouv has become obsolete in modern Greek outside of a few proverbs save in the Laconian and the Pontic dialects.⁸¹ The Pontic dialect uses ki, from Old Ionic ouvki,. But modern Greek has ouvde, and ou;te (Thumb, Handb., p. 200). In the Boeotian dialect, it may be noted, ouv never did gain a place. We have seen ouvde,n used as an adverb, an idiom that goes back to Homer.⁸² Jannaris⁸³ explains that the vernacular came to use ouvde,n and mhÄ de,n, for emphasis and then on a par with ouv and mh,) Then ouvde,n dropped ouv and mhde,n lost de,n, leaving de,n and mh, for the modern Greek. At any rate this is the outcome. De,n is the negative of the ind. in modern Greek except after na,, and final clauses when we find na. mh, (Thumb, Handb., p. 200). And de,n is the regular negative in the protasis of conditional sentences both with ind. and subj.⁸⁴ The distinction between ouv and mh, did become more or less blurred in the course of time, but in the N. T., as in the koinh, generally, the old Greek idiom is very well preserved in the main. Buttmann⁸⁵ even thinks that the N. T. idiom here conforms more exactly to the old literary style than in any other point. De,n may represent mhde,n (Rendel Harris, Exp., Feb., 1914, p. 163).

(c) Meaning. Ouv denies the reality of an alleged fact. It is the clear-cut, point-blank negative, objective, final.⁸⁶ Jannaris⁸⁷ compares ouv to o[ti and mh, to i[na, while Blass⁸⁸ compares ouv to the indicative mode and mh, to the other modes. But these analogies are not wholly true. Sometimes, indeed, ouv coalesces with the word as in ou; fhmi = not merely 'I do not say,' but 'I deny.' So ouvk eva,w (Ac. 16:7) = 'I forbid.' Cf. ouv qe,lw (Mk. 9:30); ouvk e;cw (Mt. 13:12); ouvk avgnoe,w (2 Cor. 2:11). See also to.n ouv lao,n in Ro. 9:25 (LXX) where ouv has the effect of an adjective or a prefix. Delbruck⁸⁹ thinks that this use of ouv with verbs like the Latin ne-scio was the original one in Greek. In the LXX ouv translates alo.

(d) Uses. Here it will be sufficient to make a brief summary, since the separate uses (pp. 917 f., 929 f., etc.) are discussed in detail in

PARTICLES ( AI PARAQHKAI) 1157

the proper places. The point here is to show how all the varied uses of ouv are in harmony with the true meaning of the particle.

(i) The Indicative. We meet ouv with the indicative in both independent and dependent clauses.

(a) Independent Sentences. Here the negative ouv is universal with the indicative in declarative sentences. The force of ouv ( ouvk before vowels, ouvc before aspirate) is sometimes very power- ful, like the heavy thud of a blow. Cf. ouvk evdw,kateà ouvk evpoti,sateà ouv sunhga,geteà ouv perieba,leteà ouvk evpeske,yasqe (Mt. 25:42 f.). The force of all these negatives is gathered up in the one ouv in verse 44. In verse 45 ouv and ouvde, are balanced over against each other. See ouvk e;pesen, in Mt. 7:25. Cf. ouv pare,labon in Jo. 1:11. In Mt. 21:29 see the contrast between evgw,à ku,rie and ouvk avph/lqen. Note the progressive bluntness of the Baptist's denials till ou; comes out flat at the last (Jo. 1:21 f.). In the N. T. ouv alone occurs with the future indicative used as a prohibition, though the classic idiom sometimes had mh,. Cf. ouv foneu,seij (Mt. 5:21); ouvk e;sesqe w`j oi` u`pokritai,grk grk(6:5), etc. Still, Blass⁹⁰ quotes mhde,na mish,sete in Clem., Hom., III, 69. The volitive subjective nature of this construction well suits mh,, but ouv is more emphatic and suits the indicative. In Mt. 16:22, ouv mh. e;stai soi tou/to, we have ouv mh.) in the prohibitive sense. When ouv occurs alone = 'no,' as at the end of a clause, it is written ou; as in ou;à mh, pote (Mt. 13:29); to. Ou; ou; (2 Cor. 1:17).

But in interrogative (independent) sentences ouv always expects the answer 'yes.' The Greek here draws a distinction between ouv and mh, that is rather difficult to reproduce in English. The use of a negative in the question seems naturally to expect the answer 'yes,' since the negative is challenged by the question. This applies to ouv. We may leave mh, till we come to it. Ouv in questions corresponds to the Latin nonne. Cf. Mt. 7:22, ouv tw|/ sw| / ovno,mati evprofhteu,samen ktl., where ouv is the negative of the whole long question, and is not repeated with the other verbs. See further Mt. 13:55; Lu. 17:17; 1 Cor. 14:23. In 1 Cor. 9:1 we have ouv four times (once ouvci,). The form ouvci, is a bit sharper in tone. Cf. Mt. 13:27; Lu. 12:6. In Lu. 6:39 we have mh, with one question, mh,ti du,natai tuflo.j tuflo.n o`dhgei/n; and ouvci, with the other (side by side) ouvci. avmfo,teroi eivj bo,qunon evmpersou/ntai; There is a tone of impatient indignation in the use of ouv in Ac. 13:10, ouv pau,sh| diastre,fwn ta.j o`dou.j tou/ kuri,ou ta.j euvqei,aj; In Ac. 21:38, ouvk a;ra su. ei= o` Aivgu,ptioj; the addition of a;ra means 'as I supposed,

1158 A GRAMMAR OF THE GREEK NEW TESTAMENT

but as I now see denied.⁹¹ In Mk. 14:60 note the measured use of ouv and ouvde,n in both question, ouvk avpokri,nh| ouvde,n; and the description of Christ's silence, kai. ouvk avpekri,nato ouvde,n. In Lu. 18:7, ouv mh. poih,sh|ÄÄkai. makroqumei/ evp v auvtoi/j; we come near having ouv mh, in a question with the present indicative as well as with the aorist subjunctive. In a question like mh. ouvk e;comen; (1 Cor. 9:4) ouv is the negative of the verb, while mh, is the negative of the sentence. Cf. Ro. 10:18, 19. In 1 Cor. 9:8 we have mh, in one part of the question and ouv in the other, mh. kata. a;nqrwpon tau/ta lalw/Ã h' kai. o` no,moj tau/ta ouv le,gei; In Mt. 22:17 (Lu. 20:22; Mk. 12:14) we have h' ou;* as the alternative question, and Mark adds h' mh,. Babbitt⁹² holds that " ouv is used in questions of fact, while in other questions (e.g. questions of possibility) mh, is used." I doubt the correctness of this interpretation.

In declarative sentences the position of ouv is to be noted when for emphasis or contrast it comes first. Cf. ouv and avlla, in Ro. 9:8. So ouv ga,r- avll v o[ in 7:15. In 7:18 f. note ou;\ ouv side by side. Cf. also position of ouv in Ac. 1:5; 2:15; Ro. 11:18 ( ouv su,ÄÄavlla,%. So avll v ouvk evgw, in 1 Cor. 6:12.

( b) Subordinate Clauses. In principle the use of ouv is the same as in independent sentences. But there are some special adaptations which have already been discussed and need only brief mention here.

In relative clauses with the indicative ouv is almost the only negative used in the N. T., the examples of mh, being very few as will be seen directly. This is true both with definite relative clauses where it is obviously natural, as in 2 Cor. 8:10, oi[tinej ouv mo,non- proenh,rxasqe (cf. Ro. 10:14; Jas. 4:14), and in indefinite relative clauses where mh, is possible, but by no means necessary, as in Mt. 10:38, oa}j ouv lamba,nei (cf. Lu. 9:50; 14:33, etc.). The use of ouv in the relative clause which is preceded by a negative is not an encroachment⁹³ on mh,. Cf. ouv mh. avfeqh|/ w=de li,qoj evpi. li,qon oa}j ouv kataluqh,setai (Mt. 24:2). It is a common enough idiom in the old Greek, as we see it in 10:26 (Lu. 12:2), ouvde,n evstin kekaÄ lumme,non oa} ouvk avpokalufqh,setai. Cf. Lu. 8:17, where the second relative has ouv mh. gnwsqh|/ and Ro. 15:18 for the negative ouv in principal and relative clause. In Mk. 4:25 note oa}j e;cei and oa}j ouvk e;cei. Cf. oa} qe,lw and oa} ouv qe,lw (Ro. 7:15, 19). Practically the same⁹⁴ construction is oh with the relative in a question, as ti,j

PARTICLES ( AI PARAQHKAI) 1159

evstin oa}j ouv in Ac. 19:35; cf. Heb. 12:7. For further illustration of ouv with relative clauses see Mt. 12:2= Mk. 2:24; Jo. 6:64; Lu. 14:27; Jo. 4:22; Ro. 15:21; Gal. 3:10; Rev. 9:4.

In temporal clauses with the indicative ouv comes as a matter of course.⁹⁵ This is true of a definite note of time as in Ac. 22: 11, w`j ouvk evne,blepon, and of an indefinite period as in Jo. 4:21, w[ra o[te ou;te (cf. also 9:4, nu.x o[te ouvdei,j%)

In comparative clauses with the indicative the negative comes outside in the principal sentence, since comparison is usually made with a positive note. So ouv kaqa,per (2 Cor. 3:13); ouv kaqw.j hvlpi,samengrk grk(8:5); ouvk eivmi. w[sper (Lu. 18:11); ouvc w`j (Ro. 5:15 f.). We do have w`j ouvk ave,ra de,rwn, in 1 Cor. 9:26 (participle) as in 2 Cor. 10:14 we have ouv ga,rà w`j mh. evfiknou,menoià where the two negatives are in good contrast.

In local clauses likewise the use of ouv is obvious, as in o[pou ouvk ei=cen gh/n pollh,n (Mt. 13:5); o[pou ouv qe,leij (Jo. 21:18. Here the ouv is very pointed); ou= de. ouvk e;stin no,moj (Ro. 4:15).

In causal sentences ouv is not quite universal, though the usual negative. Cf. Mt. 25:45 evf v o[son ouvk evpoih,sate e`ni. tou,twn tw/n evlaÄ ci,stwn,grk grk(2:18) o[ti ouvk eivsi,n (Heb. 6:13) evpei. kat v ouvdeno.j ei=cen, (1 Cor. 14:16) evpeidh. ouvk oi=den. See further Lu. 1:34; Jo. 8:20, 37; Ro. 11:6. In Heb. 9:17 evpei. mh. to,te [ mh, pote marg. of W. H.] ivscu,ei may be a question as Theophylact takes it, but W. H. do not print it so in the text. But it is not a departure from ancient Greek idiom to have mh, with the ind. in causal sentences as will be shown. Cf. Jo. 3:18 with 1 Jo. 5:10.

In final clauses with the ind. ouv does not occur. The reason for mh, in clauses of purpose is obvious even though the ind. mode be used (cf. Rev. 9:4, 20). It is only with clauses of apprehension that ouv is found with the verb when mh, occurs as the conjunction. Cf. 2 Cor. 12:20, fobou /mai mh, pwj ouvc eu[rw. But this is the subj., not the ind. Cf. here ouvc oi[ouj qe,lw and oi-on ouv qe,lete. Cf. also Mt. 25:9. In Col. 2:8 we have ble,pete mh, tij e;staiÄÄkai. ouv kata. Cristo,n. The kai. ouv is in contrast with kata. ta. stoicei/a tou/ ko,smou, though as a second negative it would properly be ouv anyhow. But in Rev. 9:4 we have i [na mh. avdikh,sousin- ouvde, ÄÄouvde,. This⁹⁶ does seem unusual and is almost an example of i[na ouv. No example of a clause of result with a negative occurs in the indicative, but it would, of course, have ouv.

The use of ouv in conditional sentences has already received

1160 A GRAMMAR OF THE GREEK NEW TESTAMENT

adequate treatment. See Conditional Sentences, ch. on Mode. The details need not be gone over again here. There is no doubt of the fact that eiv ouv made encroachments on eiv mh, in the later Greek.⁹⁷ Blass⁹⁸ puts it "in direct contradistinction to the classical language." Thouvemin⁹⁹ likewise treats this use of eiv ouv as "contrairement a l'usage classique - ou on le trouve exceptionnellement." It is only the frequency, the normality of eiv ouv in the N. T. that is remarkable. This is in full accord with the koinh, development, since¹⁰⁰ in the modern Greek de,n "is regularly used in the protasis of a conditional sentence, alike with the indicative and with the subjunctive mood." So a' de.n ph,gaina, 'if I had not gone' (Thumb, Handb., p. 195). See Mt. 26:42; Lu. 12:26; Jo. 1:25; 3:12; 5:47; 10:37; 2 Pet. 2:4; Ro. 8:9; 11:21; 1 Cor. 16:22; 2 Cor. 12:11; Heb. 12:25, etc. They are all conditions of the first class (determined as fulfilled) save one of the second class (determined as unfulfilled) in Mt. 26:24. In 26:42 eiv ouv and eva.n mh, stand out sharply. It is so nearly the rule with conditions of the first class in the N. T. that it is hardly necessary to follow out the analysis of Winer¹⁰¹ to bring the examples into accord with ancient usage. It is gratuitous to take eiv ouvde, as causal in Lu. 12:26, or to make eiv ouvk eivmi, in 1 Cor. 9:2 a denial of a positive idea. There are cases of emphatic denial, as ei; tij ouv filei/ (1 Cor. 16:22). Cf. also 2 Jo. 1:10, ei; tij e;rcetai kai. ouv fe,rei. Cf. also eiv ouv poiw/ and eiv poiw/ in Jo. 10:37 f., where the antithesis is quite marked. See also the decisive negation in Jo. 1:25. But, when all is said, eiv ouv has made distinct inroads on eiv mh, in the later Greek.

As to the negative in indirect discourse with the indicative, it only remains to say that the use of ouv is universal. Cf. Mt. 16: 12, sunh/kan o[ti ouvk ei=pen prose,cein. In 16:11 note pw/j ouv noei/te o[ti ouv peri. a;rtwn ei=pon u`mi/n; where each negative has its own force. Cf. also 1 Cor. 6:9.

(ii) The Subjunctive. In Homer ouv was the negative with the futuristic subjunctive¹⁰² as in ouv de. i;dwmai, Iliad, I, 262. This futuristic use of the subj., as we have seen (Modes), largely passed over to the future indicative,¹⁰³ so that a disappears from the subjunctilre almost entirely both in principal and subordinate clauses.

PARTICLES ( AI PARAQHKAI) 1161

One may compare the final disappearance of ouv before mh, with participles. In Jer. 6:8 B reads h[tij ouv katoikisqh|/ where aAQ* have katoikisqh,setai. It is to be remembered also, as already noted, that in the modern Greek de,n occurs in the protasis with subjunctive as well as with the indicative, as a a' de.n pisteu,h|j (Thumb, Handbook, p. 195). This is partly due, no doubt, to the obscuration of the ouv in de,n, but at bottom it is the futuristic use of the subj. We have already noted the use of mh. ouvc in 2 Cor. 12:20 with eu[rw after fobou/mai, where the ouv is kept with the subj. (classic idiom) to distinguish it from the conjunctional mh,. It is also a case of the futuristic subj., not volitive as in final clauses with i[na or o[pwj. In Mt. 25:9 the margin of W. H. has mh, pote ouvk avrke,sh| without a verb of fearing, though the notion is there. The text has mh, pote ouv mh,) Jannaris¹⁰⁴ boldly cuts the Gordian knot by denying that mh, in ouv mh, is a true negative. He makes it merely a shortening of mh,n. If so, 'all the uses of ouv mh, with the subj. would be examples of ouv with the subj. Some of these, however, are volitive or deliberative. This view of Jannaris is not yet accepted among scholars. It is too simple a solution, though Jannaris argues that ouv mh,n does occur as in Soph. El. 817, Eur. Hec. 401, and he notes that the negation is continued by ouv de,Ã not by mh. de,) Per contra it is to be observed that the modern Greek writes mh,n as well as mh,Ã as na. mh.n ei=ce para,dej, 'because he had no money' (Thumb, Handb., p. 200). But, whatever the explanation, we do have ouv mh, with the aorist subj. in the N. T. We have had to discuss this point already (Tense and Mode), and shall meet it again under Double Negatives. But in Jo. 18:11, ouv mh. pi,w; the answer is in accord with ouv.

(iii) The Optative. In the N. T. there are no instances of the use of ouv with the optative. It is only in wishes (volitive) that the optative has a negative in the N. T. and that is naturally mh,.¹⁰⁵ But this is just an accident due to the rapid disappearance of the optative. There is no reason why a should not be found with the potential optative (futuristic) or the deliberative which was always rare.

(iv) The Imperative. The most striking instance is 1 Pet. 3:3, w-n e;stw ouvc o` ÄÄko,smojà avll v o` krupto,jà ktl) It is the sharp contrast with avll v that explains the use of ouvc. Cf. also ouv mo,non in 1 Pet. 2:18, where the participle stands in an imperative atmosphere.

1162 A GRAMMAR OF THE GREEK NEW TESTAMENT

Cf. also ouv with the inf. in the imperatival sense in 1 Cor. 5:10. Elsewhere with the imperative we have mh. mo,non (Jo. 13:9; Ph. 2:12; Jas. 1:22). Ouv is used in an imperatival connection with the fut. ind. (Mt. 5:21) and in questions of like nature (Ac. 13:10).

(v) The Infinitive. It is common to say that in the N. T.¹⁰⁶ ouv does not occur with the infinitive, not even in indirect assertion. In Homer and in the classic Attic we do find ouv with the inf. in indirect assertion. This is usually explained on the ground that the ouv belonged to the original indicative in the direct and is simply preserved in the indirect. Monro (Hom. Gr., p. 262) observes that in the old Sanskrit only finite verbs have the negative particles. This question received full discussion under Mode and Verbal Nouns. Only a brief word is allowed here. The oldest use of the negative in indirect discourse was in the form ou; fhsin dw,swn where ouv formally goes with fhsin, but logically with dw,sen. From this use Monro conceives there came a with the inf. itself. But the situation in the N. T. is not quite so simple as Blass¹⁰⁷ makes it. In Jo. 21:25, ouvd v auvto.n oi=mai cwrh,sein, the negative does go with oi=mai. But this is hardly true in Mk. 7:24, nor in Ac. 26:26. Besides ouv occurs in a number of clauses dependent on the inf., as in Heb. 7:11; Ho. 8:12; Ac. 10:41; Ro. 7:6; 15:20; Heb. 13:9; 1 Cor. 1:17; Ac. 19:27. For the discussion of these passages see Infinitive, ch. XX, 5, (l). It is proper to say that in the N. T. we still have remnants of the old use of ouv with the inf., though in general mh, is the negative. In Ro. 15:20 ouvc o[pou after euvaggeli,zesqai stands in sharp contrast with avlla. kaqw,j. In 2 Cor. 13:7 we have mh. poih/sai u`ma/j kako.n mhde,n ouvc i[na- avll v i[na where the ouvc is clearly an addendum. Burton¹⁰⁸ explains eivj ouvqe.n logisqh/nai, in Ac. 19:27, "as a fixed phrase," but even so it is in use. Besides, there is mh. logomacei/n evp v ouvde.n crh,simon in 2 Tim. 2:14. See also kai. ouv a after w[ste douleu,ein in Ro. 7:6. The use of ouvde,n, with the inf. after ouv with the principal verb is common enough. Cf. Mk. 7:12; Lu. 20:40; Jo. 3:27; 5:30; Ac. 26:26, etc. Burton¹⁰⁹ notes that in the N. T. ouv mo,non occurs always (cf. Jo. 11:52; Ac. 21:13; 26:29; 27:10; Ho. 4:12, 16; 13:5; 2 Cor. 8:10; Ph. 1:29; 1 Th. 2:8) except once mh. mo,non in Gal. 4:18. The use of ouv mo,non occurs both in limiting clauses and in the sentence viewed as a whole.

(vi) The Participle. There is little to add to what was given on

PARTICLES ( AIPARAQHKAI) 1163

Addenda 3rd ed.

the subject of ouv and mh, with the participle under the Verbal Aspects of the Participle (see Verbal Nouns). Galloway¹¹⁰ thinks that it was with the participle that ouv was first used (as opposed to the Sanskrit negative prefix) before the infinitive had ouv. At any rate ouv is well established in Homer. We may simply accent the fact that the encroachment of mh, on ouv with the participle gives all the greater emphasis to the examples of ouv which remain. Cf. o` ouvk w'n poimh,n (Jo. 10:12); w`j ouvk de,rwn (1 Cor. 9:26). There is no trouble in seeing the force of ouv wherever we find it with the participle in the N. T.

(vii) With Nouns. Here we see a further advance of the negative particles over the Sanskrit idiom which confined them to the finite verb. The Greek usually employs the negative prefix with nouns, but in a few instances in the N. T. we have ouv) So to.n ouv lao,n in Ro. 9:25 (LXX), ouv lao,j in 1 Pet. 2:10 (LXX), ouvk e;qnei in Ro. 10:19 ( ~[' alo; Deut. 32:21). But this is by no means a Hebraism, since it is common in the best Greek writers. Cf. h` ouv dia,lusij in Thuc. 1, 137. 4 and h` ouvk evxousi,a in 5, 50. 3. Cf. ouvk avrciere,wj in 2 Macc. 4:13. As Thayer well says, ouv in this construction "annuls the idea of the noun." The use of ouv to deny a single word is common, as in ouv qusi,an (Mt. 9:13). Cf. ouvk evme, in Mk. 9:37. In general for ouv with exceptions see ouvk evn sofi,a| (1 Cor. 1:17), ouv me,lani (2 Cor. 3:3). In 2 Tim. 2:14, evp v ouvde.n crh,simon, it is possible that chr,simon is in the substantival sense. There is, of course, nothing unusual in the use of ouv with adjectives like ouv polloi. sofoi, (1 Cor. 1:26). What is noteworthy is the litotes so common in the N. T. as in the older Greek. Cf. met v ouv polu, (Ac. 27:14); met v ouv polla.j h`me,raj (Lu. 15:13); ouvk ovli,ga (Ac. 17:4); ouvk avsh,mougrk grk(21:39). Cf. ouvk evk me,Ä trou (Jo. 3:34); ouv metri,wj (Ac. 20:12). Ouv pa/j and pa/j ouv have received discussion under Adjectives, and so just a word will suffice. Ouv pa/sa sa,rx (1 Cor. 15:39) is 'not every kind of flesh.' Cf. ouv panti. tw|/ law|/ (Ac. 10:41); ouv pa,ntej (Mt. 19:11); ouv pa,ntwj (1 Cor. 5:10). But ouvk a'n evsw,qh pa/sa sa,rx (Mt. 24:22) means 'no flesh,' like the Hebrew alo-lB. The construction in both senses is more common in John than in the Synoptic Gospels. It is perhaps worth while to note the use of ouvde,n or ouvqe,n (1 Cor. 13:2) as an abstract neuter in the predicate. In general, attention should be called to the- distinction made by the Greeks between negativing a word and a sentence. This is one reason why with the imper., subj. and inf. we find ouv with

1164 A GRAMMAR OF THE GREEK NEW TESTAMENT

single words or phrases, where mh, is the normal negative of the clause.

(e) Kai. Ouv. In general when a positive clause is followed by a negative we have kai. ouv as in classic Greek. Cf. Ro. 7:6 (with inf. as in Heb. 7:11). See also Col. 2:8, 19. So Lu. 8:14, sunÄ pni,gontai kai. ouv telesforou/sin.¹¹¹ Cf. Mt. 9:13. Once, indeed, in a peculiar case, we find kai, connecting two negative clauses, Lu. 6:37, kai. mh. kri,nete kai. ouv mh. kriqh/te)

(f) Redundant or Pleonastic Ouv. There is one instance of ouv in indirect discourse where it is pleonastic according to the classic idiom (see also the French ne). It is in 1 Jo. 2:22, o` avrnou,Ä menoj o[ti vIhsou/j ouvk e;stin. Some MSS. have the pleonastic ouv in Mk. 9:39.

(g) Repetition of Ouv. When the second is a single negative, the full force of each is retained. It is seldom that we find two examples of ouv in the same clause, as in 1 Cor. 12: 15 f., ouv para. tou/to ouvk e;stin evk tou/ sw,matojà 'It is not therefore not of the body.' There are instances of ouv followed by mh, where both preserve the full force, Ac. 4:20, ouv duna,meqa- mh. lalei/n. Cf. also ouv- mh, in 1 Cor. 9:6. So also o` mh. poiw/n diÄ kaiosu,nhn ouvk e;stin evk tou/ qeou/ (1 Jo. 3:10). Cf. 5:12. The examples are numerous enough when the second a is in a dependent clause. So ouvde.n ga,r evstin kekalumme,non oa} ouvk avpokalufqh,setai (Mt. 10:26); pw/j ouv noei/te o[ti ouvà ktl)grk grk(16:11) ouv tolmh,sew ti lalei/n w-n ouv kateirga,sato Cristo,j (Ro. 15:18); ouvk oi;date o[ti- ouv klhronomh,Ä sousin (1 Cor. 6:9). In Mt. 24:2 ouv follows ouv mh,. See also Lu. 8:17. The uses of mh. ouv and ouv mh, are treated later. But note ou;à mh, pote- evkrizw,shte (Mt. 13:29) where ouv stands alone. The solemn repetition of ouv- ouv in 1 Cor. 6:10 is rhetorical.

(h) The Intensifying Compound Negative. We have seen how ouv can be made stronger by ci, ( ouvci,, as in Lu. 1:60). Brugmann¹¹² considers this an intensive particle and different from the Homeric¹¹³ ki, ( ouvÄki,% which is like ti $kijà kià tijà ti%) So also ouvde, was originally just ouv de, ('and not,' 'but not') and is often so printed in Homer.¹¹⁴ In the sense of 'not even' see Mt. 6: 29. The form ouvdei,j is intensive also, originally 'not one indeed'¹¹⁵ and was sometimes printed ouvde. ei-j (Ro. 3:10) for even stronger emphasis. But ouv ÄÄtij also occurs (Jo. 10:28). Cf. also ouvde, tij (Mt. 11:27); ouv du,nh| e;ti (Lu. 16:2); ou;teÄÄtij (Ac. 28:21);

PARTICLES ( AI PARAQHKAI) 1165

ouv- pote. (2 Pet. 1:21). The adverbial form ouvde,n, occasionally occurs in Homer. The form ouvqei,j (cf. Ac. 26:26), which flourished for a limited period in the koinh, has already had sufficient discussion. Various other compound negatives were built up on ouv, as ouvdamw/j (Mt. 2:6); ouvde,pw (Jo. 20:9); ouvde,pote (Mt. 7:23); ouvke,ti (Mt. 19:6). Ouvkou/n was used so much in questions that it lost its negative force (Jo. 18:37), unless one writes it ou;koun) Ou;te is, of course, only ouv and te,. These compound negatives merely strengthen the previous negative. This emphatic repetition of the compound negative was once good vernacular in both English and German, but it gave way in literary circles before the influence of the Latin.¹¹⁶ It was always good Greek. This discussion does not apply to subordinate clauses (as in Jo. 8:20) where each negative has its own force. The use of ouvde, and ou;te belongs to the discussion of conjunctions (cf. ou;te ou;teÄÄouvde, in Ac. 24:12 f.), but the examples in the N. T. of the other compound negatives with ouv are numerous. Farrar¹¹⁷ gives some good illustrations of old English. "No sonne were he never so old of years might not marry," Ascham, Scholemaster. Modern English vernacular refuses to give up the piling-up of negatives. "Not nohow, said the landlord, thinking that where negatives are good, the more you heard of them the better" (Felix Holt, ii, 198). Again: "Whatever may be said of the genius of the English language, yet no one could have misunderstood the query of the London citizen, Has nobody seen nothing of never a hat not their own?" So likewise the Hebrew uses two negatives to strengthen each other (cf. 1 Ki. 10:21; Is. 5:9). A good example is Mk. 5:3, ouvde. ouvke,ti ouvdei,j. So ouvdei.j ou;pwgrk grk(11:2). The commonest kind of example is like ouv du,nasqe poiei/n ouvde,n (Jo. 15:5). Cf. 2 Cor. 11:8. Another instance of triple negative is Lu. 23:53, ouvk h=n ouvdei.j ou;pw. The ouv is sometimes amplified¹¹⁸ by ou;teare as in Mt. 12:32, as well as by ouvde,ÄÄouvde, as in Jo. 1:25. Plato shows four negatives, ouvdeni. ouvdamh/ ouvdamw/j ouvdemi,an koinwni,an (Phaedo 78 (I). The combinations with ouv mh, may also be noticed, as ouvde.n ouv mh, (Lu. 10:19); ouv mh, se avnw/ ouvd v ouv mh, se evgkatali,pw (Heb. 13:5); ouvke,ti ouv mh, (Rev. 18:14). There is no denying the power of this accumulation of negatives. Cf. the English hymn

"I'll never, no never, no never forsake."

(i) The Disjunctive Negative. We frequently have ouv "where one thing is denied that another may be established."¹¹⁹ Here

1166 A GRAMMAR OF THE GREEK NEW TESTAMENT

Addenda 3rd ed.

there is sharp antithesis. The simplest form is ouv - de, as in Jas. 2:11, or ouv - avlla, as in Mt. 15:11; Mk. 15:11; Lu. 8:52; Ac. 5:4; 1 Cor. 15:10; 2 Cor. 3:3, etc. In Jo. 7:22 we have ouvc o[ti- avlla,, as also in Ph. 4:17. In Ph. 4:11 ouvc o[ti occurs alone without avlla,. In 2 Cor. 7:9 we have ouvc o[ti- avll v o[ti. In 1 Jo. 2:21 we have ouvk e;graya u`mi/n o[ti- avll v o[ti where more naturally we might expect e;graya ouvc o[ti- avll v o[ti. Winer¹²⁰ makes rather overmuch of the possible rhetorical distinctions between the varying shades of emphasis in the different contexts where ouv - avlla,, occur. Cf. further ouvc i[na- avlla, (Jo. 6:38); ouvc i[na- avll v i[na (Jo. 3:17). We usually have a mo,non- avlla. kai, (Jo. 5:18; Ro. 1:32, etc.), but sometimes merely ouv mo,non- avlla, (Ac. 19:26; 1 Jo. 5:6). Sometimes the negative is not expressed, but is to be supplied in thought as in Mt. 11:7-9. Then again we may have only the negative as in ouv brw,masin (Heb. 13:9), leaving the contrast to be supplied in the thought. The contrast may even be expressed by kai. ouv as in Mt. 9:13, e;leoj qe,lw kai. ouv qusi,an (A, LXX). But we have already entered the sphere of the conjunctions as in the parallel ou;te kai, in Jo. 4:11. So 3 Jo. 1:10.

2. THE SUBJECTIVE NEGATIVE Mh, AND ITS COMPOUNDS.

(a) The History of Mh,. The Ionic, Attic and Doric dialects have mh,, the Eleatic has ma,, like the Sanskrit ma. In the old Sanskrit ma was used only in independent sentences, while ned occurred in dependent clauses.¹²¹ In the later Sanskrit ma crept into the dependent clauses also. It was originally a prohibitive particle with the old injunctive which was in the oldest Sanskrit always negative with ma.¹²² In the later Sanskrit ma was extended to the other modes. In the Greek we see mh, extended to wish and then denial.¹²³ Wharton¹²⁴ undertakes to show that mh, is primarily an interrogative, not a prohibitive or negative particle, but that is more than doubtful. Already in Homer " mh, had established itself in a large and complex variety of uses, to which we have to appeal when we seek to know the true nature of the modal constructions as we come to them."¹²⁵ The distinction between ouv and mh, goes back to Indo-Germanic stock and has

PARTICLES ( AIPARAQHKAI) 1167

survived into modern Greek. But from the very start mh, made inroads on ouv, so that finally mh, occupies much of the field. In the modern Greek mh, is used exclusively with participle, in prohibitions and with the subj. except in conditions, and occurs with na, ( na. mh,) and the incl. Gildersleeve¹²⁶ has shown in a masterly way how mh, made continual encroachments on ouv. In the N. T., outside of eiv ouvà advance of mh, is quite distinct, as Gildersleeve shows is true even of Lucian. So as to the papyri and the inscriptions. The exact Attic refinements between ouv and mh, are not reproduced, though on the whole the root-distinction remains.¹²⁷

(b) Significance of Mh,. Max Miller¹²⁸ gives an old Sanskrit phrase, ma kaphalaya, 'not for unsteadiness,' which pretty well gives the root-idea of mh,. It is an "unsteady" particle, a hesitating negative, an indirect or subjective denial, an effort to prevent (prohibit) what has not yet happened. It is the negative of will, wish, doubt. If ouv denies the fact, mh, denies the idea. made one advance on ouv. It came to be used as a conjunction. We see this use of ma in the late Sanskrit.¹²⁹ But the origin of this conjunctional of mh, is undoubtedly paratactic in clauses of both fear and piirpose.¹³⁰ It is obviously so in indirect questions¹³¹ where mh, suggests 'perhaps.' Campbell¹³² argues that "the whole question of the Greek negatives is indeterminate." This is an extreme position, but there is no doubt a border-line between ouv and mh, which is very narrow at times. One's mood and tone have much to do with the choice of ouv or mh,. Cf. Jo. 4:29, mh,ti ou-to,j evstin o` Cristo,j; where a would have challenged the opposition of the neighbours by taking sides on the question whether Jesus was the Messiah. The woman does not mean to imply flatly that Jesus is not the Messiah by using mh, ti, but she raises the question and throws a cloud of uncertainty and curiosity over it with a woman's keen instinct. In a word, mh, is just the negative to use when one does not wish to be too positive. Mh, leaves the question open for further remark or entreaty. Ouv closes the door abruptly.¹³³ The LXX uses mh, for la'.

1168 A GRAMMAR OF THE GREEK NEW TESTAMENT

(c) Uses of Mh,. In general we may follow the outline of ouv.

(i) The Indicative. Blass¹³⁴ expounds the two negatives by saying that " ouv negatives the indicative, mh, the other moods, including the infinitive and participle." But, unfortunately, the case is not so simple as that. "In reviewing Blass, Thumb makes the important addition that in modern Greek de,n belongs to the indicative and mh,( n) to the subjunctive."¹³⁵ But de,n occurs in the protasis with the subj. in modern Greek, as we have seen. Besides, as Moulton¹³⁶ adds, " mh, has not been driven away from the indicative" in the N. T. It may be said at once that with the indicative is as old as historic Greek.¹³⁷ The Sanskrit suggests that originally mh, was not used with the indicative. But already in Homer mh, occurs with the indicative in prohibition, wish, oath, fear, question.¹³⁸ "The essence of these idioms is the combination of the imperative tone - which shows itself in the particle - with the mood proper to simple assertion."¹³⁹ But in the N. T. we no longer have mh, with the fut. ind. in prohibition, except in case of ouv mh,)

In independent sentences we have with the indicative only in questions. "It's use in questions is very distinct from that of ouv and is maintained in the N. T. Greek without real weakening."¹⁴⁰ In Jo. 21:5, paidi,aà mh, ti prosfa,gion e;cete* we have a typical example with the answer Blass¹⁴¹ expresses needless objection to this "hesitant question," as Moulton rightly expounds it. Cf. Jo. 4:33; 7:26; and Ro. 11:1, mh. avpw,sato* with the answer in verse 2, ouvk avpw,sato. See Jo. 7:51, where Nicodemus adroitly uses mh, in a question and the sharp retort of the other members of the Sanhedrin mh. kai. su,; The difference between ouv and mh, in questions is well shown in Jo. 4:33, 35. In the use of mh, the answer in mind is the one expected, not always the one actually received as is illustrated in the question of the apostles at the last passover. They all asked mh, ti evgw, eivmià r`abÄ bei,; The very thought was abhorrent to them, 'It surely is not I.'¹⁴² But Judas, who did not dare use ouv, received the affirmative answer, su. ei=paj (Mt. 26:25). Mh,ti comes to be used intensively much like ouvci, (both chiefly in questions). In the case of mh. ouv

PARTICLES ( AIPARAQHKAI) 1169

Addenda 3rd ed.

in question (Ro. 10:13 f.; 1 Cor. 9:4 f.; 11:22) mh, is the interrogative article while ouv is the negative of the verb.

In dependent clauses mh, occurs with the indicative with the second class conditions ( eiv mh,) always except in Mt. 26:24 (Mk. 14:21). Cf. eiv mh, in Jo. 15:22, etc. There are also five instances of eiv mh, with the ind. in conditions of the first class.¹⁴³ So Mk. 6:5; 1 Cor. 15:22 Cor. 13:5; Gal. 1:7; 1 Tim. 6:3. Cf. mh, in a few relative clauses, as aa} mh. dei/ (Tit. 1:11); w|- mh. pa,restin tau/ta (2 Pet. 1:9); oa} mh. o`mologei/, (1 Jo. 4:3, W. H. text). Cf. Ac. 15:29 D. There is a certain loofness about mh, here that one can feel as in Plato who, "with is sensitiveness to subtle shades of meaning, had in mh, an instrument singularly adapted for purposes of reserve, irony, politeness of suggestion."¹⁴⁴ This use of with the relative and indicative is clearly a remnant of the literary construction.¹⁴⁵ This literary use of mh, with the relative was often employed to characterize or describe in a subjective way the relative. There is a solitary instance of mh, in a causal sentence, o[ti mh. pepi,steuken (Jo. 3 18), which may be contrasted with o[ti ouv pepi,steuken (1 Jo. 5:10). For o[ti mh. e;ceij see Epictetus, IV, 10. 34, and o[ti soi ouv, IV, 10. 35. Radermacher (N. T. Gr., p. 171) quotes fasi.n o[ti mh. dei/, Diog. of Oinoanda, Fragm. IV, 1. 9. There is, besides, evpei/ mh. to,te ivscu,ei in Heb. 9:17, according to the text of W. H., though they give in the margin evpei. mh, pote- diaqe,menoj; In that case (the marginal reading) mh, pote would introduce a question. See further Causal Clauses. In clauses of design we have i[na mh, with the ind., as in Rev. 9:4, i[na mh. avdikh,sousin. The margin of W. H. in 13:17 has i[na mh, tij du,natai. Moulton¹⁴⁶ explains mh, with the ind. after verbs of apprehension as not originally a conjunction, but mh, in the sense of 'perhaps' (paratactic, not hypotactic). So Lu. 11:35, sko,pei mh. to. fw/j- sko,toj evsti,n) Cf. also Col. 2:8; Heb. 3:12; Gal. 4:11; 1 Th. 3:5. The papyri give abundant parallels. Moulton (Prol., p. 193) cites avgwniw/ mh, pote avrrwstei/, P. Par. 49 (ii/B.C.). The use of mh, as a conjunction in clauses of design and fear with the indicative is parallel to the use of the negative particle mh,, but does not fall here for discussion.

(ii) The Subjunctive. After all that has been said it is obvious that mh, was destined to be the negative of the subj., first of the volitive and deliberative uses and finally of the futuristic also. The few remnants of ouv with the subj. have already been discussed. For the rest the normal and universal negative of the

1170 A GRAMMAR OF THE GREEK NEW TESTAMENT

subj. is mh,) Cf. mh. evnkakw/men (Gal. 6:9). In Mk. 12:14, dw/men h' mh. dw/men; (cf. ou; just before), we see how well mh, suits this deliberative question. The use of mh, with the aor. subj. in prohibitions need not be further stressed. Wherever the subj. in a dependent clause has a negative (save after the conjunction mh, after verbs of fearing) the negative is mh,. Cf. o[j aa}n mh. e;ch| (Lu. 8:18); i[na mh. e;lqhte (Mk. 14:38), etc. It is needless to give more examples.

(iii) The Optative. It is only the optative of wish that uses mh,. It was rare to have the negative precative optative in the old Sanskrit.¹⁴⁷ But already in Homer mh, is used with the optative for a future wish. In the N. T. there is no example of mh, with the optative except in wish. It is seen chiefly in mh. ge,noito, as in Ro. 3:4, 6, 31; Gal. 6:14, etc. But note also the curse of Jesus on the fig-tree in Mk. 11:14, mhdei.j karpo.n fa,goi.

$iv) The Imperative. It seems that the imperative was originally used only affirmatively and the injunctive originally only negatively with ma. The oldest Sanskrit does not use ma with the imperative.¹⁴⁸ In Homer we find once mh. e;nqeo (Il., IV, 410) and once mh. katadu,seo (Il., XVIII, 134) and once mh. avkousa,tw (Od., XVI, 301). The second person aorist imper. in prohibitions did not take root and the third person only sparingly (cf. p. 856). See Mt. 6:3, mh. gnw,tw.¹⁴⁹ The original negative injunctive ap- pears in the form mh. poih,sh|j (Latin ne feceris). The imperative in Greek follows the analogy of this construction and uses mh, uniformly. Cf. Lu. 11:7, mh, moi ko,pouj pa,rece. For the difference between mh, with the present imperative and mh, with the aorist subjunctive see Tenses and Modes. Cf. Mk. 13:21, mh. pisteu,ete, wit Lu. 12:11, mh. merimnh,shte, and mh. fobei/sqe with mh. fobhqh/te (Mt. 10:28, 31). It is obviously natural for mh, to be used with the imperative. For a delicate turn from ouv to mh, see Jo. 10:37. But Radermacher (N. T. Gr., p. 171) cites ouvdeni. evxe,stw from an inscr. (Benndorf-Niemann, Reisen in Lykien und Karien, 129 N. 102).

(v) The Infinitive. As we have already seen, the oldest Sanskrit inf. did not use the negative particles, and in Homer¹⁵⁰ ouv appears to be the original negative. But there are a few instances of pit with the inf. in Homer. They occur when the inf. is used as an imperative (cf. in the N. T. 1 Cor. 5:9; 2 Th. 3:14), for an oath, a will or an indirect command. It is thus from the imperative and other finite modes that mh, crept into constant use with the inf.

PARTICLES ( AIPARAQHKAI) 1171

It came to be the normal idiom with the inf. outside of indirect assertion and in antithetical or emphatic phrases (see under ouv). Thompson¹⁵¹ challenges the statement of Gildersleeve: "Not till the infinitive, came to represent the indicative (in indirect statement) could ouv have been tolerated with the infinitive." Thompson adds: "But this toleration is established in Homer." Just as we saw mh, make inroads on ouv in other constructions (cf. participles), so it was with the inf. Even in indirect statement.¹⁵² The came to be the rule (cf. the Atticist Lucian). Even in the Attic ouv did not always occur with the inf. in indirect statement.¹⁵³ The facts as to the use of mh, with the inf. in the N. T. have been already given. (see Infinitive and Indirect Discourse). Cf., for instance, le,gousin avna,stasin mh. ei=nai (Mk. 12:18); avpekri,qhsan mh. eivde,nai (Lu. 20:7). In short, Blass¹⁵⁴ says that in the N. T. " mh, is used throughout." That is not quite true, as we have seen, but the limitations have already been given under ouv. Cf. Lu. 11:42, tau/ta de. e;dei poih/sai kavkei/na mh. parei/nai. Cf. 21:14. The use of mh. lalei/n after ouv duna,meqa (Ac. 4:20) has already been noticed. Here mh, retains its full value. We need not pursue the matter. Cf. tou/ mh, (Ac. 21:12); pro.j to. mh, (2 Cor. 3:13); eivj to. mh,grk grk(4:4); dia. to. mh, (Mt. 13:5) tw|/ mh, (2 Cor. 2:13); w[ste mh, (Mt. 8:28), etc. The redundant or pleonastic use of mh, with the inf. has likewise come up for consideration under the Infinitive. In Lu. 20:27 some MSS. read avntiÄle,gontej and thus mh, is redundant after avnti,--, but aBCDL do not have avnti-. Then in 22:34 aBLT reject with mh, with eivde,nai after avparnh,sh|. In Heb. 12:19 W. H. put mh, in the margin after parh|th,santo. But there is no doubt of the use of the redundant mh, in the N. T. Cf. Lu. 17:1 avne,nÄ dekto,n evstin tou/ ta. ska,ndala mh. evlqei/n,grk grk(24:16) evkratou/nto tou/ mh. evpignw/nai auvto,n. See also Lu. 4:42; 1 Pet. 3:10; Gal. 5:7. But this pleonastic mh, is by no means necessary (cf. Ac. 8:36; Ro. 15:22). It does not usually occur with kwlu,w in the N. T., but note Ac. 10:47, mh,ti to. u[dwr du,natai kwlu/sai, tij tou/ mh. baptisqh/Ä nai; Here mh,ti is the interrogative particle expecting the answer 'no,' while mh, is redundant after kwlu,ein. But in Ac. 24:23 mhde,na is not pleonastic. We do not have mh. ouv, with the inf. in the N. T. Here (after ouv) mh, stands alone and is not redundant (cf. Ac. 4:20) or is redundant redundant(20:20, 27), as the case may be. The use of mh, and mh. ouv was not compulsory in the ancient Greek.¹⁵⁵

1172 A GRAMMAR OF THE GREEK NEW TESTAMENT

(vi) The Participle. We have seen already how the oldest Sanskrit did not use the negative particles with the participle. In Homer we have only one instance of with the participle (Od., IV, 684).¹⁵⁶ But mh, gradually made its way with participles even in Attic Greek. In the modern. Greek mh, has driven ouv entirely from the participial use. In the N. T. ouv still hangs on, as we have seen, but that is all. The drift of the koinh, is for mh,, and a writer like Plutarch shows it.¹⁵⁷ Mh, is the usual negative of the participle. The details were given in connection with Participles. In the N. T. we need pay no attention to the Attic refinements on this point, which were not always observed even there. We have mh, with the participle in the N. T. as a matter of course. Cf. Mt. 12:30 o` mh. w;n and o` mh. suna,gwn, (1 Tim. 5:13) ta. mh. de,onta (Lu. 4:35) mhde.n bla,yan, (Ac. 20:22) mh. eivdw,j. In Mt. 22:11 f. and 1 Pet. 1:8, a distinction, as was shown, seems to be drawn between ouv and mh, with the participle. Cf. Mt. 18:25; Lu. 12: 33; Jo. 7:15; Ac. 9:9; 17:6; 1 Th. 4:5 (cf. Gal. 4:8), etc. The downright denial of a lingered on awhile in the koinh, (cf. papyri), but mh, is putting ouv to rout.¹⁵⁸

(vii) Nouns. The ancient Greek¹⁵⁹ used mh, with substantives as o` mh. ivatro,j (Plato, Gorg. 459 b), adjectives as oi` mh. kaqaroi, (Ant. v. 82), or adverbs as to. mh. evmpodw,n, (Thuc. ii, 45. 1). In the N. T., so far as I have noticed, mh, with substantives and adjectives occurs only in contexts where it is natural. Thus in Lu. 10:4, mh. ph,ranà mh. u`podh,mata, we have just before mh. basta,zete balla,nÄ tion. In Jo. 13:9, mh. tou.j po,daj mou mo,non, we have no verb, but ni,pte is to be supplied from the preceding sentence. Cf. also Eph. 5:15; Jo. 18:40. So in Ro. 12:11 mh. ovkhroi, is in the midst of participles used in an imperatival sense. In 1 Tim. 3:3, mh. pa,roinonà mh. plh,kthn, the construction is dei/ ei=nai. This infinitival construction is carried on in verse 6 (in spite of the parenthesis in verse 5) by mh. neo,futon. So as to verse 8 and Tit. 1:7. There is no difficulty as to the use of mh, in Col. 3:2 and 2 Th. 3:6.

(d) The Intensifying Compounds with Mh,. The same story in the main that we found with ouv is repeated with mh,. There is no mhci,, but we have mh,ti in this sense. The examples in the N. T. are all in questions (cf. Mt. 7:16; Jo. 18:35) except one, eiv mh,ti (Lu. 9:13). The position of mh, may give it emphasis as in Jas. 3:1 (cf. ouv in Mt. 15:11). The use of the compound

PARTICLES ( AIPARAQHKAI) 1173

negative as a second (or third) negative is simply to strengthen the negative s is true of ouv. Cf. Mk. 11:14 mhke,ti mhdei.j fa,goi, (Ac. 25:24) evpibow/ntej mh. dei/n auvto.n zh/n mhe,ti, (Ro. 13:8) mhdeni. mhde.n ovfei,lete, (2 Cor. 13:7) mh.ÄÄmhde,n, etc. Besides mhdei,j there is mhqe,n (Ac. 27:33), mhde, in the sense of 'not even' (Eph. 5: 3), mh,ge (Mt. 6:1), mhde,pote (2 Tim. 3:7), mhde,pw (Heb. 11:7), mhke,ti (Mk. 9:25), mh,pote (margin of W. H. in Heb. 9:17. Elsewhere in the N. T. a conjunction), mhdamw/j (Ac. 10:14), mh,pou (Ac. 27:29), mh,pw (Ro. 9:11), mh,tige (1 Cor. 6:3), mh,tij (2 Th. 2:3). Mh,pwj is only a conjunction in the N. T. If mh, is followed by ouv as in 1 Jo. 3:10, o` mh. poiw/n dikaiosu,nhn ouvk e;stin evk tou/ qeou/, the last negative retains its force. So vice versa in Ac. 4:20. In Gal. 6:3 there is a sharp contrast between ti and mhde,n (both neater abstracts referring to a person.).

(e) Kai. mh,) We saw that after a positive statement the negative was carried on by kai. ouv. So also we have kai. mh, as in Eph. 4:26, ovrgi,zesqe kai. mh. a`marta,nete, and in Lu. 1:20; 2 Cor. 12:21 In Ac. 18:9 note mh. fobou/ avlla. la,lei kai. mh. siwph,sh|j, where a positive command comes in between the two examples of p,77. In Jas. 3:14, per contra, mh. katakauca/sqe kai. yeu,desqe kata. th/j avlhqei,aj, the negative mh, seems to cover both verbs connected by kai, rather than mhde,) Cf. also Lu. 3:14. We have instances. also of kai, connecting a clause with the conjunction mh, pote (Mt. 13:15 = Mk. 4:12).¹⁶⁰ In Lu. 14:29, i[na me, pote qe,ntoj auvtou/ qeme,Ä kion kai. mh. ivscu,ontojÄÄa;rxwntai we have mh, pote with a;rxwntai and mh, with ivscu,ontoj.

(f) Disjunctive Use of Mh,. The simplest form of this contrast is mh,ÄÄde, as in Lu. 10:20, mh. cairete- cai,rete de,) Then we have mh,ÄÄ avlla, as in mh. tou/ton avlla. to.n Barabba/n, Jo. 18:40; mh. fobou/ avlla. la,lei, Ac. 18:9. We have mh,ÄÄplh,n, in Lu. 23:28. In Lu. 10:20 we really have mh, o[ti - de. o[ti. Moulton (Prol., 240) does not find mh. o[ti in the N. T., but considers mh,tige in p. 1 Cor. 6:3 as tantamount to it. See Jo. 13:9 for mh. mo,non- avlla. kai,. So Ph. 2:12. We need not trench further upon the conjunctions.

3. COMBINATION OF THE TWO NEGATIVES.

(a) Mh. ouv. This is very simple. It is in the N. T. confined to questions where is the interrogative particle and ouv is the negative of the verb. Each negative thus has its own force, though it is a bit difficult to translate the combination into good English. But it is good Greek. Moulton (Prol., p. 192) quotes

1174 A GRAMMAR OF THE GREEK NEW TESTAMENT

Addenda 3rd ed.

Addenda 3rd ed.

Addenda 3rd ed.

Plato's Protag. 312 A, avll v a;ra mh. ouvk u`polamba,neij. Cf. also mh. ouvci,, in Jer. 23:24. So Ro. 10:18, mh. ouvk h;kousan; We may render it 'Did they fail to hear?' expecting the answer 'No.' Paul repeats the same idiom in 10:19. See further 1 Cor. 9: 4 f.; 11:22. 1 Cor. 9:8 is not an instance, since mh, comes in one part of the question and ouv in the other. We do have mh, pwj ouvc eu[rw after fobou/mai, in 2 Cor. 12:20, but here mh, is a conjunction and ouvc is the negative of eu[rw, both retaining their full force. The construction in 1 Jo. 3:10 is not pertinent.

(b) Ouv mh,. The use of ouv- mh, in Ac. 4:20 is not under discussion, nor the redundant mh, after ouv (Ac. 20:20, 27), but only the idiomatic ouv mh, with the aorist subj. (rarely present) or occasionally the fut. ind. Cf. ouv mh. fa,gwà ouv mh. pei,nw in the boy's letter, P. Oxy. 119 (ii/iii A.D.). See Is. 11:9, ouv mh. katkopoih,sousin ouvde. mh. du,nwntai. Whatever the origin of this vexed problem, the negative is strengthened, not destroyed, by the two negatives. We need not here recount the various theories already mentioned.¹⁶¹ See Tense and Mode. Let it go at Gildersleeve's suggestion that was originally ou; mh,. Moulton (Prol., p. 249) quotes Giles to the effect that this explanation was offered in the Middle Ages the ancients have all our best ideas) and notes "in one if not both of the best MSS. of Aristophanes it is regularly punctuated ou; mh,)" In Mt. 13:29 we have ou; mh, pote- evkrizw,shte where mh, is a conjunction. Gildersleeve notes that ouv mh, is more common in the LXX and the N. T. than in the classic Greek.¹⁶² But Moulton (Prol., pp. 187-192) will not let it go at that. "In the LXX alo is translated ouv or ouv mh, indifferently within a single verse, as in Is. 5:27." It seems probable that the force of ouv mh, has worn down in the LXX and the N. T. In the non-literary payri " ouv mh, is rare, and very emphatic," Moulton notes. He urges also that in spite of the 100 examples in the text of W. H. the idiom in the N. T. is as rare as in the papyri when the 13 LXX quotations and the 53 from the words of Christ are removed, "a feeling that inspired language was fitly rendered by words of a peculiarly decisive tone." But in these examples the force of ouv mh, is still strong. Of the other 34 some are probably weakened a bit as in Mt. 25:9; Mk. 13:2; Jo. 18:11. It is only in the Gospels and the Apocalypse (66 and 18 respectively) that ouv mh, occurs with frequency. It is interesting to observe that on this point Moulton gets the Gospels and Revelation in har-

PARTICLES ( AI PARAQHKAI) 1175

mony with the papyri by eliminating the 70 passages due to Semitic influence. Cf. Gildersleeve (A. J. P., iii, 202 ff.) and Ballentine (ib., xvni, 453 ff.). But Radermacher (N. T. Gr., p. 172) explains Mt.24:21, oi[a- ouvd v ouv mh. ge,nhtai, not as a Hebraism, but as a "barbarism" like the Wesseley Papyrus xxvi, ouvd v ouv mh. ge,nhtai, moi gunh,. He quotes also Pap. Lugd. II, p. 107, 9, eva.n qe,lh|j gunai/kaj ouv mh. sceqh/nai (Rev. 2:11); ouv mh. e;stai (Mt. 16:22). There is a climax in Rev. 7:16, ouv - ouvde, se evgkatali,pw. Even ouv mh, was not strong enough sometimes, so that we have ouvde, and ouv mh, in Heb. 13:5, ouv mh, se avnw/ ouvd v ouv mh, se evgkatali,pw. So also ouvde.n ouv mh. avdikh,sei (Lu. 10:19). In Mk. 13:2 we have ouv mh, in both the principal and the subordinate (relative) clause.

IV. Interrogative Particles ( evperwthkai. paraqh/kai). It is not the mode thatIlwe have under discussion here, but simply the particles used in the various forms of questions.¹⁶³

1. SINGLE QUESTIONS.

(a) Direct Questions.

(i) No Particle at all. So sunh,kate tau/ta pa,nta; (Mt. 13:51). So 13:28 and fiery often. Here the inquiry is colourless except as the tone of voice or context may indicate one's attitude. In fact, most interrogative sentences have no interrogative word at all. Cf. Lu. 13:2; Jo. 7:23; 13:6; Ac. 21:37, etc. Hence it is sometimes a matter of doubt whether a sentence is interrogative or declarative. Cf. Jo. 16:31; Ro. 8:33; 14:22; 1 Cor. 1:13; 2 Cor. 3:1; Heb. 10:2; Jas. 2:4, etc. It may be doubtful also at what point the question ends. Cf. Jo. 7:19; Ro. 4:1. Winer¹⁶⁴ rightly says on this point grammar cannot speak.

(ii) The Use of Negative Particles. They are used to indicate the kind of answ, expected. This subject has already had sufficient discussion. See under ouv and mh,. Ouv expects the answer 'yes' (cf. Mt. 7:22) and mh, the answer 'no' (cf. Jo. 7:31). In Jo. 18:37 we ouvkou/n, according to W. H., which has lost its negative force, but ou;koun would preserve it. Probably Pilate was hardly ready to go that far unless in jest. The use of mh, varies greatly in tone. The precise emotion in each case (protest, indignation, scorn, excitement, sympathy, etc.) depends on the context. Cf. Jo. 4:29; 6:67; 7:47; Lu. 6:39; Ro. 10:18; 11:1. In Jo. 3:10 the first part of the question has no negative and the second part has ouv.

1176 A GRAMMAR OF THE GREEK NEW TESTAMENT

(iii) Other Particles. There are not many. There is a;ra (akin to root of avrÄarÄi,skw, 'to join'), an illative particle which occurs with ouvk as in Ac. 21:38, mh,ti as in 2 Cor. 1:17, or with ti,j as in Mt. 18:1. This classic use is not strictly interrogative, but illative in the interrogative sentence. But a=ra, from the same root¹⁶⁵ with more vocal stress, is interrogative. Indeed, it is sometimes doubtful which accent is correct, as in Gal. 2:17, where a=ra is probably correct. In Ro. 14:19, however, W. H. give a;ra ou=n We have a=ra in Lu. 18:8 and a=ra, ge in Ac. 8:30. ;Ara looks backward, a=ra forward. But the accent is a question of edit- ing. The use of eiv in direct questions is either a Hebraism¹⁶⁶ or involves ellipsis. Cf. Mt. 12:10, eiv e;xesti toi/j sa,bbasin qerapeu,ein; So also 19:3. It is common in the LXX (cf. Gen. 17:17) but is foreign to the old Greek. The classic Greek, however, did use eiv in indirect questions, and this fact may have made it easier for the direct use of eiv to arise. Radermacher (N. T. Gr., p. 136) takes this eiv= h=. The N. T. does not use h=, but the papyri have it: h= mei,nwi evn bakcia,di* h= me,l$l%w evntunca,nin; P. Fay. 137 (i/A.D.). So the question to the oracle.

(iv ) Interrogative Pronouns. The most common in the N. T. is ti,j cf. Mt. 3:7). Other words are frequently added, as a;ra (24: 45); ga,rgrk grk(9:5); ou=n (Lu. 3:10). The various uses of ti, as adverb (Mk. 10:18, Lu. 16:2); with prepositions, as dia. ti, (Mt. 9:11) and eivj ti, (Mk. 14:4) or ca,rin ti,noj (1 Jo. 3:12); or elliptically, as ti, o[ti (Lu. 2:49) and i[na ti, (Mt. 9:4), need not detain us. The double interrogative ti,j ti, appears in Mk. 15:24. Both ti,j and poi/oj occur in 1 Pet. 1:11. For potapo,j see Mt. 8:27, and po,Ä soj see 15:34. We need not tarry longer on these elementary details.

(v) Interrogative Conjunctions. These are common besides ti, (as in Mk. 10:18). The possible exclamatory use of ti, in Lu. 12: 49= 'how' is sustained by the modern Greek ti, kala,= 'how fine.' Cf. posa,kij (Mt. 18:21); po,tegrk grk(25:38); e[wj po,tegrk grk(17:17); pou/ (Lu. 8:25); pw/jgrk grk(10:26); po,qen (Mt. 13 : 27), etc.

(b) Indirect Questions. Here there must be either a pronoun or a conjunction.

(i) Pronouns. The use of ti,j ( ti,) is common. Cf. Mt. 6:25; Lu. 9:46; Jo. 2:25; Ac. 19:32. We find o[ti so used in Ac. 9:6 and a[ apparently so in 1 Tim. 1:7. Certainly o`poi/oj occurs in this construction (1 Cor. 3:13). The same thing is true of

PARTICLES ( AI PARAQHKAI) 1177

Addenda 3rd ed.

o[soj (Mk. 5:19) and o`poi/oj (Jas. 1:24). Cf. also poi/oj (Mk. 11: 29); po,soj (M 27:13); potapo,j (Lu. 7:39); phli,koj (Heb. 7:4), and h`li,koj in Gal. 6:11 (margin of W. H.) if this reading be accepted. Cf. ti, in Ac. 12:18.

(ii) Conjunctions. These are also common, as eiv (Mk. 15:44); po,qen and tou/ (Jo. 3:8); po,te (Mk. 13:33); pw/j (1 Th. 1:9); o [pw,j (Lu. 24 :20); o[pou (Mk. 14:14); mh, pote (Lu. 3:15), etc.

2. DOUBLE QUESTIONS. These are rare.

(i) Direct. There is no instance of po,teron- h;. We do have ti,j- h; (Mt. 9:5; 23:17; 27:17), the later Greek caring little for the dual idea in po,teron. We more commonly have simply with the second part of the question and nothing in the first, as in Lu. 20:2, 4; Ro. 2:3 f. We may have h' ou' (Mt. 22:17) and h' mh, (Mk. 12:14). Sometimes we have simply at the beginning of the question with a reference to an implied alternative (1 Cor. 9:6; 2 Cor. 1:17). This h; may come in the middle of the sentence as in 1 Cor. 9:8. The may even precede ti,j as in Mt. 7:9.

(ii) Indirect. There is one instance of po,teron- h; in an indirect question Jo. 7:17).

V. Conjunctions ( su,ndesmoi). In the nature of the case much had to be sail about the conjunctions¹⁶⁷ in the treatment of the Sentence and also Subordinate Clauses. The syntactical principles controlling both paratactic and hypotactic sentences have received adequate discussion. But conjunctions play such an important part in the language that it is best to group them all together. They connect words, clauses, sentences and paragraphs, and thus form the joints of speech. They have a very good name, since they bind together (con-jungo) the various parts of speech not otherwise connected, if they need connection, for asyndeton is always possible to the speaker or writer. The point here is to interpret each conjunction as far as possible so that its precise functin may be clear.

1. PARATACTIC¹⁶⁸ CONJUNCTIONS ( su,ndesmoi parataktikoi,).

(a) Copulative. Conjunctions which connect words and clauses are evidently later in development than the words and clauses. The use of conjunctions came to be very common in the Greek so that the absence was noticeable and was called asynde-

1178 A GRAMMAR OF THE GREEK NEW TESTAMENT

ton.¹⁶⁹ But it is a mistake to suppose that these connectives are necessary. One may fail to use them as a result of rapidity of thought as the words rush forth, or they may be consciously avoided for rhetorical effect. Cf. ble,peteà ble,peteà ble,pete in Ph. 3:2, with Tennyson's "Break, break, break." All this is entirely within the province of the speaker. Cf. 1 Cor. 3:12, cruso,n, a;rguronà li,qouj timi,oujà xu,laà co,rtonà kala,mhn. Cf. also 1 Cor. 13: 4-7 where the verbs follow one another in solemn emphasis with no connective save one de,) In the same way contrast may be expressed without conjunctions as in 1 Cor. 15:43 f.¹⁷⁰ In Luke and John there is a pleasing alternation of asyndeton and conjunctions. Cf. Gal. 5:22. The first conjunctions were the paratactic or co-ordinating, since language was originally in principal sentences.¹⁷¹ The copulative (connecting) conjunctions are the simplest and earliest type of the paratactic structure. They simply present the words or clauses as on a par with each other.¹⁷² The primitive conjunctions were monosyllabic like kai,à te,à de,.¹⁷³

(i) Te,. This word appears to be related to the Sanskrit ca, the Latin que (with labio-velar qu), and the Gothic -h.¹⁷⁴ These words are all enclitic and postpositive. The Sanskrit is almost devoid of conjunctions which were so highly developed by the Greek and Latin, but ca is one of the few possessed by this ancient tongue.¹⁷⁵ There is a striking connection between quis, que, quis-que and ti,jà te, ti.j. The Thessalian dialect has ki.j for ti.j and ki,sÄke. We have ti,j te in the old Greek. Te, shows this double pronomial origin in its use for and and ever (just like que, quis-que).¹⁷⁶ The indefinite use is distinctly Homeric.¹⁷⁷ The use of evpei, teà o[j te was old Ionic and continued in Attic tragedy, as oi-o,j te did in Attic prose. Cf. Rademacher (N. T. Gr., p. 5). Indeed, some scholars¹⁷⁸ hold that the correlative use ( te,ÄÄte,) was the original one, but this is doubtful. It seems certain that te, indicates a somewhat closer unity than does kai,. This close correlative use is certainly very old. Cf. su, t v evgw, te in Homer.¹⁷⁹ In the N. T. it is rare except in the Acts, where it occurs some 175 times. It is common in all parts of the book and is thus a subtle argument

PARTICLES ( AIPARAQHKAI) 1179

Addenda 2nd ed.

for the unity of the work (we-sections and all). It is something additional, but in intimate relation with the preceding. We find te alone as in Ac. 2:33, where u`ywqei,j and labw,n are united by te. Cf. also 10:22, where again two participles are connected. In 23:24, kth,nh te parasth/sai, the change from the direct to the indirect discourse is marked by te,, whereas kai, is used twice before to join minor phrases. Te, puts parasth/sai on a par with e`toima,Ä sate)¹⁸⁰ In the same way in 20:11 the first two participles are joined by kai, and then both are related to the next by te,. The same idiom occurs in Jo. 6:18, where te gives an additional item somewhat apart from the kai, - kai, just before. In Jo. 4:41 kai, - te, are not co-ordinate. Kai, introduces the whole sentence and te, connects the two parts. Cf. thus de,ÄÄte, in Ac. 2:37. But te,ÄÄte, is strictly correlative. Cf. the Latin que - que, English as - so. See Ac. 2:46 where the two participles are co-ordinated. In Ro. 14:8 -- have te four times in succession with eva,n. There are here two pairs of conditions. The parts of each pair are balanced carefully. The disjunctive ei;teÄÄei;te (cf. 1 Cor. 12:26) is at bottom this same correlative use of te,. So as to ou;te - ou;te (Mt. 12:32) and mh,te- mh,te (Ac. 27:20). The use of te,ÄÄkai, is also common where there is an inner bond, though no hint is given as to the relative value of the matters united. Cf. avrcieÄ rei/j te kai. grammatei/j (Lu. 22:66); poiei/n te kai. dida,skein (Ac. 1:1); a;ndrej te kai. gunai/kejgrk grk(8:12); evkinh,qh te- kai. evge,netogrk grk(21:30); diÄ kai,wn te kai. avdi,kwngrk grk(24:15); mikrw| / te kai. mega,lw|grk grk(26:22); [Ellhsi,n te kai. barba,roij (Ro. 1:14); vIoudai,ou te prw/ton kai. [Ellhnojgrk grk(2:9), etc. For te. kai,ÄÄte, see Ac. 9:15, and for te. kai,ÄÄte,ÄÄkai, 26 20. In Jo. 4:11, ou;teÄÄkai,, we really have the te,ÄÄkai, ('both and') construction. Cf. Latin non que- et. We even have ou;te ÄÄou;teÄÄkai, in Jo. 5:37 f. In Ac. 27:20 mh,te- mh,te stand together and both are parallel to te following. Per contra we find te, ÄÄ de, in Ac. 19:2 and also 3. The manuscripts often vary between te, and de, $cf. Ac. 3:10; 4:14, etc.). We have te. ga,r (common from Aristotle on¹⁸¹) in Ro. 1:26 followed by o`moi,wj te kai,. In Heb. 2:11 note te. ga,r- kai,. As a rule te stands after the word or words that are paralleled, but this is not always so.

(ii) Kai,. The etymology of this conjunction is disputed. Curthis¹⁸² makes it he locative case of the pronominal stem kaÄÄÃ koÄÄÃ so

1180 A GRAMMAR OE THE GREEK NEW TESTAMENT

that it would ultimately come from the same root as te, (que). It would thus mean 'in this respect,' 'this besides.' Brugmann¹⁸³ finds its original sense in koino,j, Latin co-, cum, Gothic ga. The idea would then be 'together with,' 'in addition to.' The Arkadian, South Achaean and Cypriote dialects use ka,j and ka,Êkai,) Whatever the origin, it all comes to the same thing in the end. It is by far the most frequent of all the conjunctions or other particles in the N. T. It is so common in fact that Moulton and Geden do not list it in their concordance. This in itself is in accord with the later Greek idiom, as Thumb¹⁸⁴ notes in Aristotle and in the modern Greek and Moulton¹⁸⁵ in the papyri. Moulton cites Par. P. 18, e;ti du,o h`me,raj e;comen kai. fqa,somen eivj Phlou,si, as parallel to Mk. 15:25; Jo. 4:35. But there can be little doubt that the extreme fondness for parataxis in John's Gospel, for instance, is partially due to the use of kai, in the LXX for the Hebrew which "means a hook and resembles a hook in shape."¹⁸⁶ It was certainly used to "hook" together all sorts of sentences. There is not the same unity in the older Greek in the matters united as is true of te,. Kiaà "connects in a free and easy manner"¹⁸⁷ and the Hebrew w. still more loosely. There are three main uses of kai, which appear in the N. T. as in all Greek.

The Adjunctive Use ('Also'). This is possibly the original use, though one cannot tell. It is thus like the Latin et-iam, English too (to) = addition to something already mentioned, and is common enough in all stages of the language.¹⁸⁸ A good example of this use of kai, is seen in Mt. 8:9, kai. ga.r evgw. a;nqrwpo,j eivmi u`po. evxousi,an. The kai, here points to Christ's relation to the boy. The centurion, like a true soldier, does not say that he is a man who gives orders, but rather one who obeys them. He has the true military spirit and knows therefore how Jesus can cure the boy without going to see him. The kai, is here very significant. Cf. ou[twj kai. u`mei/j in Mt. 7:12, where the Golden Rule is applied to Christ's hearers by kai,. Cf. Jo. 7:3 i[na kai. oi` maqhtai, souÃgrk grk(12:10) i[na kai. to.n Lazaro,n. This use of kai, is more frequent in Luke than elsewhere in the N. T.¹⁸⁹ Cf. kavgw, (Lu. 20:3); h' kai, (Lu. 12:41); de. kai,grk grk(12:54, 57); ti, kai, (1 Cor. 15:29); kai. ga,r (Mt. 8:9); eva.n kai, (Gal. 6:1); eiv kai, (2 Cor. 11:15); kai. de, (Mt. 10:

PARTICLES ( AIPARAQHKAI) 1181

18); w`j kai, (Ac. 11:17); kaqw.j kai, (Ro. 15:7); ou[tw kai, (Ro. 6:11); o[j kai, (Ac. 2:6, 8); o`moi,wj kai, (Jo. 6:11); w`sau,twj kai, (1 Cor. 11:25); kaqa,per kai, (1 Th. 3:12); dio. kai, (Lu. 1:35); dia. tou/to kai, (Lu. 11:49); avlla. kai,grk grk(24:22), etc. So then kai, in the sense of 'also' occurs with nouns, pronouns, verbs, adverbs, conjunctions. It may refer to a word or a clause. Cf. a;llwj te kai,, B. G. U. 530 (i/A.D.). For the use of o` kai, see the Article, and for su.n kai, see Prepositions.¹⁹⁰ It is common for kai, to sum up a sentence that precedes. For the relative and articular participle see the kai, in the sentences in Mt. 5:39-43. Here kai, balances the principal and the subordinate clauses. So in the apodosis of a conditional sentence we find kai,, as in Jo. 14:7. Cf. Heb. 7:26, where kai, almost means 'precisely,' and Mt. 6:10, where, it means 'just so.' Cf. Ro. 11:1. So with a[ we find it in the apodosis (Jo. 5:19). Cf. also after w[sper in 5:26. Sometimes the kai, seems to be redundant as in Lu. 11:1, kaqw.j kai,, or w`j kai, in 1 Cor. 7:7. We may indeed have kai, ('also') in both parts of the comparison, a studied balancing of the two members of the sentence as in Mt. 18:33, kai. se,ÄÄw`j kavgw,) So Ro. 1:13, kai. evn u`mi/n kaqw.j kai. evn toi/j loipoi/j e;qnesin. See oi=da kai,ÄÄoi=da kai, (Ph. 4:12).

The Ascensive Use ('Even'). The notion of 'even' is an advance on that of mere addition which is due to the context, not to kai,. The thing that is added is out of the ordinary and rises to a climax like the crescendo in music. Cf. Latin adeo. Cf. ouv mo,nonà avlla. kai, (Ac. 21:13; Ro. 13:5). This use of kai, depends wholly on the context. Cf. Mk. 1:27, kai. toi/j pneu,masi toi /j avkaqa,rtoij evpita,ssei. (So Lu. 10:1 ). Cf. also kai. oi` telw/nai and kai. oi` evqnikoi,, Mt. 5: 46 f. See further Ac. 10:45; 11:1, 20; Gal. 2:13. The use of kai. eiv belongs here. (Cf. 1 Cor. 8:5.)

The Mere Connective ('And'). The difference between kai, as 'and' and kai, as 'also' is very slight, whichever was the original idea. The epexegetic or explicative use of kai, occupies a middle ground between 'also' and 'and.' Blass¹⁹¹ treats it under 'also.' Cf. Lu. 3:1:, polla. kai. e[tera parakalw/n, where the "connective" force of kai, is certainly very slight. So also Jo. 20:30, polla. kai. a;lla shmei/a. See further Jo. 1:16, kai. ca,rin avnti. ca,ritoj, where the clause is an explanatory addition. Cf. (Ac. 22:25) kai. avkata,kriton, (1 Cor. 2:2) kai. tou/ton evstaurwme,non, (Ro. 13:11) kai. tou/to (Latin idque) which is our 'and that too' where we combine 'and' and 'also' ('too') in the kai,, (Heb. 11:12) kai. tau/ta (frequent in ancient

1182 A GRAMMAR OF THE GREEK NEW TESTAMENT

Greek). See in particular Eph. 2:8, kai. tou/to ouvk evx u`mw/n, where tou/to refers to the whole conception, not to ca,riti. The simple copulative idea is, however, the most common use of kai, where words are piled together by means of this conjunction. Sometimes the connection is as close as with te,. Thus o` qeo.j kai. path.r (2 Cor. 1:3); kalh|/ kai. avgaqh|/ (Lu. 8:15). But the words may be very loosely joined in idea, as oi` Farisai/oi kai. Saddoukai/oi (Mt. 16:1). Kai, may be used to connect all sorts of words, clauses and sentences. Thus le,gw ;Ercouà kai. e;rcetai (Mt. 8:9). The use of kai, after the imperative is seen in Mt. 11:29. The chain with kai, as the connective may go on indefinitely. Cf. the four examples in Ph. 4:9; five in Ro. 9:4; the six in Rev. 7:12 (so 5:12). So we have kai.. o[ti three times in 1 Cor. 15:4 ( kai, to connect o[ti clauses). In Rev. 12-16 every paragraph and most of the sentenes begin with kai,. In fact it is true of much of the Apocalypse. If one turns to First Maccabees, it is true even to a much grater extent than in the Apocalypse. In First Maccabees kai, translates the Hebrew w.. But Thumb¹⁹² has found this repetition of kai, in Aristotle so that the Hebrew influence simply intensified a Greek idiom. We have noted the use of kai, with te, ( te,ÄÄkai,. Cf. Ro. 1:20). The use of kai,ÄÄkai, is far more common in the seise of 'both - and' as in Ac. 2:29, kai. evteleu,thse kai. evta,fh. Cf Mk. 4:41; Ph. 2:13; Ac. 26:29. Sometimes the connection almost amounts to 'not only, but also.' In Col. 2:16 note kai,ÄÄh;. Cf. ka;nÄÄka;n (Lu. 12 : 38). A. Brinkmann contends that in the papyri and late Greek ka;n is sometimes 'at any rate' and is never a mere link (Scriptio continua und Anderes, Rhein. Mis. LXVII, 4, 1912). In Lu. 5:36 we have kai,ÄÄkai,ÄÄkai. ouv (so Jo. 6:36), and in Jo. 17:25 kai. ouv- de.ÄÄkai.. It is usual to have kai. ouv after an affirmative clause as in Jo. 10:35. Cf. kai. mh, in 2 Cor. 9:5. See Negative Particles. In Lu. 12:6 kai. ouv follows a question with ouvci,. Kai, connects two negative sentences in Lu. 6:37. For ou;teÄÄkai, see Jo. 4:11. Sometimes kai, begins a sentence when the connection is with an unexpressed idea. Children use "and" thus often in telling stories and asking questions. Cf. kai. su. h=sqa in Mt. 26:69 (and 73) like Et tu, Brute. See also Mk. 10:26, kai. ti,j du,natai swqh/nai. So also Lu. 10:29; Jo. 9:36; 2 Cor. 2:2. Cf. also the use of kai, in parenthesis as in Ro. 1:13, kai. evkwlu,qhn a;cri tou/ deu/ro. The context gives other turus to kai, that are sometimes rather startling. It is common to find kai, where it has to bear the content 'and yet.' So Jo.

PARTICLES ( AIPARAQHKAI) 1183

Addenda 2nd ed.

Addenda 3rd ed.

3:19; 4:20; 6:49; 7:30; 1 Jo. 2:9. The examples are common in John's Gospel (Abbott, Joh.. Gr., pp. 135 ff.). See Jer. 23:21. In Mk. 4:4 note me,n- kai,. In 1.Cor. 10:21 we have ouv- kai, in contrast. Cf.Mt. 3:14, kai. su. e;rch| pro,j me; So also Ph. 1:22, kai. ti, ai`rh,somai. This idiom occurs in Plato, and Abbott notes a number of them in the Gospel of John. Cf. 1:5; 2:20; 3:13; 5:39 7:27 f.; 8:57, etc. In Lu. 12:24 kai, is almost equal to avlla,, that is, the context makes contrast. Cf. also Mt. 6:26 ( ouv- kai,); Mk. 12:12; Lu. 20:19; Jo. 18:28. Tholuck¹⁹³ so takes kai, in Ro. 1:13 (the parenthetical kai,%. Sometimes kai, seems imitative of the Hebrew w. by almost having the sense of o[ti or i[na ('that') as in Mt. 26:15; Mk. 14:40; Lu. 9:51; 12:15. In particular note kai. evge,neto kai, (as in Lu. 5:1, 12, 17, etc.). In Mt. 16:6 observe o[ra/te kai,. So Lu. 12:15 and Mt. 26:15. In modern Greek kai, has so far usurped the field that it is used not only in all sorts of paratactic senses like 'and,' 'but,' 'for,' 'or,' 'and so,' but even in hypotactic senses for na, or pou/, declarative and even consecutive (Thumb, Handb., p. 184). In Mk. 3:7 kai, comes near takng the place of o[, for in the next verse there are five instances of kai. co-ordinate with each other, but subordinate to kai, in verse 7. Sometimes after kai, we may supply 'so' as in kai. la,mpei, Mt. 5:15; kai. ble,pomen, Heb. 3:19. See also Ph. 4:7. This is a kind of consecutive¹⁹⁴ use of kai,. Cf. Lu. 24:18. The fondness for co-ordination in the Gospels causes the use of kai, where a temporal conjunction ( o[te% would be more usual. Cf. Mk. 15:25, h=n w[ra tri,th kai. evstau,rwsan (Lu. 23:44). But Blass¹⁹⁵ admits that this is a classic idiom. Cf. Mt. 26:45; Lu. 19:43, where kai, drift further away from the ancient idiom. Cf. also kai. ivdou, in the apodosis, 'and behold,' as in Lu. 7:12. In 2 Tim. 2:20 note kai, allowed by aa} me,n- aa} de,. In Ph. 4:16 note kai, thrice (one= 'even,' two = 'both - and').

(iii) De,) This conjunction is generally ranked wholly as an adversative particle.¹⁹⁶ Monro¹⁹⁷ says: "The adversative de, properly indicates that he new clause stands in some contrast to what has preceded. Ordinarily, however, it is used in the continuation of a narrative." As a matter of fact, in my opinion, Monro has the matter here turned round. The ordinary narrative use (continuative) conceive to be the original use, the adversative the developed and later construction. The etymology confirms

1184 A GRAMMAR OF THE GREEK NEW TESTAMENT

this explanation, though it is largely conjectural. Brugmann¹⁹⁸ associates it with the aksl. ze and possibly also¹⁹⁹ with dh, and the enclitic ending - de, $oi;kaÄdeà o[Ädeà toso,jÄde%, while Hartung²⁰⁰ connects it with du,oà di,j, and Baumlein²⁰¹ with deu,Äteroj. The enclitic - de thus means 'again,' 'back,' while the conjunction de, would mean 'in the second place' or 'a second comment' or an important addition' ( dh,). But, however we take it, there is in the word no essential notion of antithesis or contrast. What is true is that the addition is something new²⁰² and not so closely associated in thought as is true of te, and nkai,. I prefer therefore to begin with the narrative and transitional (copulative) use of de,) Kuhner-Gerth²⁰³ call this use of de, for 'something new' (etwas Neues) copulative and give it separate discussion. Abbott²⁰⁴ has the matter correctly: "In classical Greek, calling attention to the second of two things, may mean (1) in the next place, (2) on the other hand." The first of these uses is the original one and is copulative. The second is adversative. Abbott notes also that de, in both senses occurs in Matthew and Luke nearly three times as often as in Mark and John. Its use is mainly in the historical books of the N. T. It is so common there that, as with kai,, Moulton and Geden do not give any references. A good place to note the mere copulative force of de, is in the genealogy in Mt. 1:2-16 where there is no notion of opposition at all. The line is simply counted from Abraham to Christ. In verses 6 and 12 there are breaks, but the contrast is made by repetition of the names, not by deà which appears with every name alike. In Mt. 23:4 we have both uses of de,) The first is properly translated 'yea' and the second 'but' (adversative). See further 1 Cor. 4:7 ( de, and de. kai.% where there is a succession of steps in the same direction. So 15:35; 2 Cor. 6:15 f.; Heb. 12:6; and in particular the list of virtues in 2 Pet. 1:5-7. Sometimes a word is repeated with de, for special emphasis, as dikaiosu,nh de, in Ro. 3:22 (cf. 9:30). A new topic may be introduced by de, in entire harmony with the preceding discussion, as the Birth of Jesus in Mt. 1:18 ('Now the birth of Jesus Christ,' etc.). The use of de, in explanatory parenthesis is seen in Jo. 3:19 ('And this is,' etc.); 19:23 ('Now the coat,' etc.). For w`j de, ('and when,' 'so when') in John see 2:9, 23. In John

PARTICLES ( AI PARAQHKAI) 1185

as elsewhere it is sometimes not clear whether de, is copulative or adversative. Cf. 3:1, h=n de,) Is Nicodemus an illustration or an exception?²⁰⁵ The resumptive use of de,Ã after a parenthesis, to go on with the main story, is also copulative. Cf. Mt. 3:1; Lu. 4:1. There is continuation, not opposition, in the use of kai. de,Ã as in Lu. 1:76 kai. su, de, where means 'and' and kai, 'also.' Cf. further Mt. 10:18; 16:18; Jo. 15:27. In Jo. 6:51 we have kai. de, in the apodosis of the condition in this sense. De, is always postpositive and may even occupy the third place in the sentence (Mt. 10:11) or even the fourth (Jo. 6:51) or fifth (1 Jo. 2:2) or sixth (Test. xiii, Patr. Jud. 9:1) as shown in chapter on Sentence.

In accord with the copulative use of de, we frequently have ouvde, and mh,de, in the continuative sense, carrying on the negative with no idea of contrast. Cf. Mt. 6:26, ouv spei,rousin ouvde. qeri,zousin ouvde. suna,gousin. So also 6:28; Mk. 4:22, etc. In Jo. 7:5, ouvde. ga,r, we have ouvde, in the sense of 'not even' as often (Mt. 6:29, etc.). In Mt. 16:15 ouvde, means 'not also' (cf. also 21:27, etc.). All three uses of kai, are thus paralleled in ouvde, (merely ouv de,%. For mhde, in the continuative sense see Mt. 7:6. It means 'not even' in 1 Cor. 5:11. For the repetition of continuative mhde, see 1 Cor. 10:7-10. In Mk. 14:68, ou;te oi=da ou;te evpi,stamai (some MSS. ouvk- ouvde,), we come pretty close to having ou;teÄÄou;te in the merely continuative sense as we have in ou;teÄÄkai, (Jo. 4:11; 3 Jo. 1:10).

(iv) vAlla,. Here there is no doubt at all as to the etymology. vAlla, is a virtual proclitic (cf. e;pi and evpi,), and the neuter plural was avlla, ( a;lla, 'other things'). Baumlein²⁰⁶ does take avlla, as originally an adverb. But in reality it is 'this other matter'²⁰⁷ (cf. tau/ta and tou/to). In actual usage the adversative came to be the most frequent construction, but the original copulative held on to the N. T. period. It is a mistake to infer that a;lloj means 'something different.' In itself it is merely 'another.' Like de, the thing introduced by avlla, is something new, but not essentially in contrast.²⁰⁸ So the classic Greek used avlla. mh,n in the emphatic continuative sense.²⁰⁹ Blass²¹⁰ observes that "the simple avlla, also has this force of introducing an accessory idea." Cf. 2 Cor. 7:11, po,shn kateirga,sato u`mi/n spoudh,nà avlla. avpologi,anà avlla. avgana,kthsinà avlla. fo,bonà avlla. evpipo,qhsinà avlla. zh/lonà avlla. evkdi,khsin. All these six examples are confirmatory and continuative. See further Lu. 24:21, avlla, ge kai. su.n pa/sin tou,toij, where it is cli-

1186 A GRAMMAR OF THE GREEK NEW TESTAMENT

Addenda 2nd ed.

acteric, not contradictory. The story is carried on by avlla. kai, in verse 22. Cf. also 2 Cor. 1:9; Lu. 12:7; 16:21. In Ph. 1:18, cai,rwà avlla. kai. carh,somai, the connection is very close. The most striking example of all is Ph. 3:8, avlla. menou/nge kai. h`gou/mai. In 2 Cor. 11:1, avlla. kai. avne,cesqe, the tone of irony akes it doubtful whether to take avlla, as copulative or adversative. These and similar passages are not a dropping of the adversative idea, but merely the retention of the original copulative meaning. Abbott<sup>211</sup> sees that "it is hard to find a satisfactory explanation of Jo. 8:26" along the usual line. If one no longer feels impelled to translate by 'but,' the trouble vanishes. Just make it 'now' or 'yea' and it is clear. Abbott<sup>212</sup> likewise considers axm "inexplicable" in 4:23, because it has to mean 'but.' Cf. Jo. 16:2, avll v e;rcetai w[ra, 'yea, the hour comes.' The same use of avlla, occurs also in negative sentences. In 1 Cor. 3:3, avll v ouvde. nu/n du,nasqe after ou;pw evdu,nasqe. In 4:3, avll v ouvde. after an affirmative clause. In Ac. 19:2, avll v ouvd v, the thought answers the preceding question and is probably adversative, as is possible in 1 Cor. 3:3. The avlla, at any rate is negative like the ouvde,) So as to avll v ouvde. `Hrw|,dhj (Lu. 23:15).

(b) Adversative. It should be stated again that not all of these conjunctions mean contrast (antithesis) or opposition, but the context makes the matter clear. The modern Greek keeps avlla,à o[mwjà plh,n, but not de, and me,ntoi (Thumb, Handb., p. 185).

(i) De,) In Jas. 1:13 f. note the two uses of de, (continuative and adversative). Sometimes the positive and the negative are sharply contrasted and then de, is clearly adversative as in Mt. 23:4, auvtoi. de. ouv qe,lousin. More obvious still is 6:14 f., eva.n avfh/te ÄÄeva.n de. mh. avfh/te. Cf. also 6:23. So mh. qhsauri,zete- qhsauri,zete de, grk(6:19 f.). Cf. 1 Cor. 1:10, etc. The contrast may lie in the nature of the case, particularly where persons stand in contrast as in evgw. de, (Mt. 5:22, 28, 32, etc.), su. de, (Mt. 6:6; 1 Tim. 6: 11; h`mei/j de, (1 Cor. 1:23); u`mei/j de, (Mk. 8:29); the common o` de, (Mk. 1:45), oi` de, (Mt. 2:5); auvto.j de, (Lu. 8:37), auvto.j de. vIhsou/j (Jo. 2:24), etc. The contrast is made more manifest by the use of me,n, (see Intensive Particles) as in Mt. 3:11. In 1 Cor. 2:6, sofi,an de. ouv tou/ aivw/noj tou,tou, an exception is filed to the preceding. This adversative use of de, is very common indeed. Cf. further Mk. 2:18; Lu. 5:5; 9:9, 13; 24:21; Ac. 12:15; Ro. 8:9 ff.

(ii) vAlla,. Just as a;lloj (cf. 2 Cor. 11:4) can be used in the sense of e[teroj (when it means 'different,' not merely 'second'), so

PARTICLES ( AIPARAQHKAI) 1187

Addenda 3rd ed.

Addenda 3rd ed.

avlla, can mean 'another' in contrast to the preceding. With a negative the antithesis is sharp as in Lu. 1:60, ouvci,à avlla. klhqh,Ä setai vIwa,nhj. So Jo. 6:32, ouv Mwush/j- avll v o` path,r (cf. 6:38). Cf. Mk. 9:37; 1 Cor. 15:37. In verse 39 of 1 Cor. 15 note avlla. a;llh me,n- a;llh de, where both avlla,, and a;llh have the notion of difference due to the context. In 1 Cor. 9:12 note avlla,, twice. In Mt. 15:11 ouv begins one clause and avvlla, the other. Cf. 2 Cor. 4:5, ouv ga.r e`autou.j khru,ssomenà avlla. Cristo.n vIhsou/n ku,rion. So Mt. 5:17. In Lu. 12:51 note ouvci, avll v h; and in 2 Cor. 1:13, a;llaÄÄavll v h; a sort of pleonastic use of avlla,. This is a classical idiom.1 Cf. also ouv mo,non- avlla,, (Ac. 19:26) or avlla. kai, (Ro. 5:3). See Negative Particles. For ouvc o[ti- avlla, see Jo. 7:22, for ouvc i[na- avlla, see 6:38. For avlla, ge in apodosis see 1 Cor. 9:2, for avlla,. Col. 2:5, for avll v ouv, 1 Cor. 4: 15. Sometimes avll v i[na may be elliptical as in Mk. 14:49; Jo. 1:8. vAlla, alone may refer to an interruption in thought not expressed, as in Jo. 12:27. One of the most striking instances of avlla, occurs in Ac. 16:37, ouv ga,rà avlla,, where ouv ga,r means 'not much' with fine shorn (cf. kai. nu/n; just before). Both Winer and W. F. Moulton (W.-M., p. 566) felt certain that avlla, never equalled eiv mh,, not even in Mt. 20:23 and Mk. 4:22. But J. H. Moulton (Prol., p. 241) quotes Tb. P. 104 (i/B.C.), kai. mh. evxe,stw Fili,skwi gunai/ka a;llhn evpagage,sqai avlla. vApollwni,an, where avlla, means practically 'except.' See also Gen. 21:26. Moulton suggests that, since eiv mh, (brachylogy) in Lu. 4:26 f.; Rev. 21:27, means 'but only,' the same may be true of avlla,.

(iii) Plh,n. Curtius gets it from ple,on ('more'), but Brugmann²¹³ finds its original --meaning to be 'near by.' At any rate it was a preposition (Mk. 12:32). Cf. Ac. 15:28, ple,on plh.n tou,twn where the two werds exist together. Probably its original use as a conjunction is seen in the combination plh.n o[ti (Ph. 1:18). It is chiefly confined to Luke's writings in the N. T. As a conjunction it is always adversative (cf. Lu. 6:24; 12:31, etc.). In Mt. 26:39 note plh.n ouvc w`j - avll v w`j. The classical language used it as a preposition and with o[ti, but Aristotle²¹⁴ shows the existence of plh,n as a conjunction which developed in the vernacular. Blass²¹⁵ notes that Paul uses it at the end of an argument to single out the main point. Cf. 1 Cor. 11: 11; Eph. 5:33; Ph. 3: 16; 4:14.

1188 A GRAMMAR OF THE GREEK NEW TESTAMENT

(iv) Me,ntoi. This word is a combination of two intensive parisles ( me,nà toi,), and is used to mean 'however.' Cf. Jo. 4:27; 12:42. It occurs in the N. T. only eight times.

(v) [Omwj. This word is even more rare than me,ntoi. It occurs with two participles (1 Cor. 14:7; Gal. 3:15) and once with me,nÄ toi (Jo. 12:42).

(vi) Eiv mh,. This phrase marks an exception, as in Mt. 12:4; Jo. 17:12. We even have evkto.j eiv mh, (1 Cor. 14:5; 15:2; 1 Tim. 5:19).

(c) Disjunctives. Dionysius Thrax calls this construction su,nÄ qesij diazeutikh,. It was always possible to express alternative ideas without any conjunction (cf. the Latin nolens volens) or by copulative conjunctions ( de,Ã kai,), a construction common in the vernacular²¹⁶ (cf. Hebrew w.). Dissimilar things may be united by kai, as in Col. 3:11, but we do not have to take kai, as being h; or vice versa.²¹⁷

(i) ;H. Its origin from hve, (enclitic) is held by Brugmann.²¹⁸ They are equivalent in Homer. We may have just h; as in Mt. 5:17. For h' kai, see Mt. 7:10; Lu. 18:11. In the sense of 'or' h; may be repeated indefinitely (Ro. 8:35). In Ro. 1:21 we have ouvc- h; as in 4:13. See mh,pw- mhde,ÄÄh; (Ro. 9:11). This use in negative clauses appears in Thuc. 1, 122, and later writers. In 1 Th. 2:19 note h' ouvci. kai,. In Mt. 21:23 we have kai. ti,j, while in Lu. 20:2 (parallel passage) the reading is h' ti,j. This does not prove kai, and h; to be synonymous. The logion was translated differently. The modern Greek retains ou;teà mh,te and h; (Thumb, Handb., p. 185). In 1 Cor. 11:27, oa}j a'n evsqi,h| to.n a;rton h' tpi,nh| to. poth,rion tou/ kuri,ou, some MSS. have Kai, but ij is the true text. This, however, does not mean that some partook of one element and some of the other, but that, whatever element was taken in this way, there was guilt. The correlative use of h; - h; ('either --or') is also frequent.²¹⁹ Cf. Mt. 6:24; 1 Cor. 14:6. In Ro. 6: 16 note h;toiÄÄh;. As a disjunctive we have po,teron- h; in Jo. 7:17 and h; - h; - h;- h; in Mk. 13:35. For pri.n h; see Mt. 1:18; for h; after qe,lw see 1 Cor. 14:19; after kalo,n, Mt. 18:8; after cara,, Lu. 15:7; for avll v h;, Lu. 12:51. Radermacher (N. T. Gr., p. 27) finds h; toi- h;, B. G. U. 956; h; toi- h; toi,

PARTICLES ( AIPARAQHKAI) 1189

Vett. Val., p. 138, h;teÄÄh;, I. G. XII, 2, 562, 5 (Roman time); h;teÄÄh;te, Quaest. Barth., pp. 24, 30.

(ii) Ei;teÄÄei;te $eva,nte- eva,nte). These conditional particles are like the Latin sive - sive. Cf. 1 Cor. 10:31, ei;teÄÄei;teÄÄei;te. So 12:13; 14:7. We have ei;te eight times in 3:22. In 14:7 it follows h; - h; in verse 6. For eva,nte- eva,nte see Ro. 14:8.

(iii) Ou;teÄÄou;te ( mh,te- mh,te). We have seen that there is nothing inherent in ou;te to make it disjunctive. Cf. Jo. 4:11; 3 Jo. 1:10. It is simply ouv and te, (cf. ouv de,, a negative copulative conjunction. In Rev. 5:3 f. we have ouvde,ÄÄou;te (cf. Gal. 1:12) and the next verse ouvdei,j- ou;te. In Ac. 24:12 f. we have ou;te ÄÄou;teÄÄou;teÄÄouvde,) Cf. Lu. 20:35 f. In Jo. 5:37 f. note ou;te - ou;te - kai. ouv. In 1 Cor. 6:10 note ou;teÄÄou;teÄÄouv - ouv. In Jas. 3:12 cf. are after question. A good-example of the correlative ou;teÄÄou=te is 1 Cor. 3:7. In Ro. 8:38 f. ou;te occurs ten times. In Ac. 23:8 we find mh,ÄÄmh,te- mh,te. This is also just a copulative negative conjunction ( mh, te%. In Mt. 5:34-36 we have mh,ÄÄmh,te- mh,te - mh,te- mh,te. In 2 Th. 2:2 we have mhde,ÄÄmh,te- mh,te. In Lu. 7:33 mh,ÄÄmh,te, while in 9:3 mhde,n is followed by mh,te five times. There is often some confusion in the MSS between mhde, and mh,teà ouvde,, and ou;te. Blass²²⁰ rejects ou;te oi=da ou;te evpi,stamai in Mk. 14:68 ( aBDL), but on whimsical grounds.

(d) Inferential Conjunctions. It is not easy to draw a distinction between "inferential" and "causal." There is no doubt about a;ra and ou=n. These are inferential paratactic particles. What about ga,r? Monro²²¹ calls it causal. Kuhner-Gerth²²² treat all three as causal. Perhaps it is just as well to reserve the term "causal" for the hypotactic particles o[tià evpei,, etc. One has to be arbitrary sometimes. And even so these particles ( a;ra, ou=n, ga,r) were originally just transitional or explanatory in sense. Blass²²³ calls them "consecutive" co-ordinate conjunctions.

(i) ;Ara. The etymology seems to be clear, though not accepted by all scholars. The root avrÄÄ ( avr- ar- i,skw, 'to fit') suits exactly.²²⁴ It means then 'fittingly, accordingly.' Cf. our "articulate" (ar-ticulus). The word expresses some sort of correspondence between the sentences or clauses. It was postpositive in the ancient Greek, but in the N. T. it is not always so. Cf.

1190 A GRAMMAR OF THE GREEK NEW TESTAMENT

Mt. 12:28; Ac. 17:27. It occurs some 50 times in the N. T., in Matthew, Mark, Luke, Acts, Paul's Epistles, and Hebrews. The original notion of mere correspondence is apparently preserved in Lu. 11:48, a;ra ma,rture,j evste, 'so ye are witnesses.' Cf. alsd Ac. 11:18. In Mk. 11:13; Ac. 17:27, eiv a;ra has the idea of 'if haply.' Klotz takes a;ra to describe the unexpected and strange, something extrinsic, while Baumlein considers it a particle giving point to what is immediately and necessarily conclusive. Most of the N. T. instances seem to be clearly illative. Cf. Mt. 17:26 f.; Ro. 7:21. It has ge added three times (cf. Mt. 7:20; 17:26 f.; Ac. 17:27). Paul is specially fond of a;ra ou=n (Ro. 5:18; 7:3, 25, etc.). Once he has a;ra nu/n (Ro. 8:1). ;Ara occurs also in the apodosis (Mt. 12:28; Gal. 2:21). We have mh,ti a;ra in a question in 2 Cor. 1:17.

(ii) Ga,r. There is no doubt as to the origin of this word. It is a compound of ge, and a;ra and is always postpositive. It is called su,ndesmoj aivtiologiko,j, but it does not always give a reason. It may be merely explanatory. We have seen that a;ra itself was originally just correspondence and then later inference. So then ge, can accent as an intensive particle either of these ideas. It is a mistake, therefore, to approach the study of ga,r with the theory that it is always or properly an illative, not to say causal, particle. It is best, in fact, to note the explanatory use first. Thayer wrongly calls the illative use the primary one. The word is common in all the larger books of the N. T. It is least common in the Gospel of John and in Revelation. In Matthew and Luke it is much more frequent in the discourses and is rare in the strict narrative. In Mark and John it is about half and half.²²⁵ In general the N. T. use of ga,r is in accord with that of the classic period. The explanatory use is common in Homer.²²⁶ The N. T. examples are numerous. Cf. Mt. 19:12; Mk. 5:42; 16:4; Lu. 11:30; 18:32. Here the explanation follows immediately. Sometimes the explanation comes in by way of appendix to the train of thought. So Mt. 4:18, h=san ga.r a`liei/j. Cf. also Mk. 2:15; Ro. 7:2. In questions we have good examples, particularly ti, ga,r. So Mt. 27:23, ti, ga.r kako.n evpoi,hsen; Cf. Ro. 3:3. In Ac. 16: 37, ouv ga,r avlla,, we have to resolve ga,r into its parts and make the phrase= 'not much, but.' In Jo. 9:30, evn tou,tw| ga,r, the man uses ga,r with fine scorn, 'why, just in this,' etc. In Jo. 19:6 it is hardly creditable to Pilate's common sense to take ga,r as illative. Cf. also Jo. 7:41; Ac. 19:35; Mt. 9:5. Ga,r sometimes

PARTICLES ( AIPARAQHKAI) 1191

gives the major premise (Mt. 26:52), more often the minor premise (2 Pet. 1:15 f.), sometimes both (Jo. 3:19 f.). The purely illative use of ga,r is simple enough, though the force of the ground or reason naturally varies greatly. See Mt. 1:21, auvto.j ga.r sw,sei;grk grk(6:24) h' ga,r; (Ro. 8:18) logi,zomai ga,r. Paul begins every sentence with ga,r in Ro. 8:18-24. For kai. ga,r see Ro. 11:1; 15:3. The precise relation between clauses or sentences is not set for by ga,r. That must be gathered from the context if possible. Cf. Jo. 4:44. Note ga,r - o[ti in 1 Tim. 6:7.

(iii) Ou=n. The etymology of ou=n is unknown. Brugmann²²⁷ thinks it probable that it is derived from * o` evn or o` o`n (cf. o;ntwjà tw|/ o;nti). The Ionic also has w=n (so Lesbian, Doric, Boeotian). But, however that may be, it is important to note that the particle is not illative nor even consequential in Homer.<sup>228</sup> It is merely a transitional particle relating clauses or sentences loosely together by way of confirmation. It was common in this sense in Homer, though rare in fthe Attic writers save in me.n ou=n. But it is very frequent in the Gospel of John as a mere transitional particle. In this Gospel it occurs about 200 times, nearly as frequent as all the rest of the N.T., though it is rare in the other Johannine writings. In John's Gospel, outside of 8 examples in the words of Jesus, the rest occur in the narrative portion.<sup>229</sup> Abbott<sup>230</sup> seems puzzled over the many non-illative instances of ou=n in John and suggests that "the writer perhaps had in view the objections of controversialists" But this is wholly gratuitous and needless in the light of a history of the particle. Probably a majority of the instances in John's Gospel are non-illative as in Homer, the original use of the word.<sup>231</sup> Luke preserves the literary Attic idiom by the common use of me.n ou=n as in Ac. 15:3, 30, etc. But John boldly uses ou=n alone and needs no apology for doing so. It just carries along the narrative with no necessary thought of cause or result. It is, because of John's free use, one of the commonest particles in the N. T. and is oftener in the narrative books than in the epistles.<sup>232</sup> It is interesting in John to take a chapter and note when ou=n is merely continuative and when illative. Cf. ch. 11, for instance, verses 3, 6, 12, 14, 16, 17, 20, 21, 31, 32, 33, 36, 38, 45, 47, 54, 56. So we start off again in 12:1 with o` ou=n vIhsou/j

1192 A GRAMMAR OF THE GREEK NEW TESTAMENT

(continuative). It is the commonest connective between sentences in this Gospel. We moderns do not feel the same need for connecting-particles between independent sentences. The ancient Greeks loved to point out these delicate nuances. The interrogative ouvkou/n occurs only in Jo. 18:37. A good instance of the purely illative use is in Mt. 3:8, poih,sate ou=n karpo,n. It is common in Paul's Epistles (Ro. 5:1; 6:12, etc.). Paul is fond also of a;ra ou=n (Ro. 8:12) and of ti, ou=ngrk grk(6:1, 15; 7:7; 8:31, etc.). Ou=n is always postpositive.

2. HYPOTACTIC CONJUNCTIONS ( su,ndesmoi u`potaktikoi,%. The conjunctions used in the N. T. with subordinate clauses have been discussed and the constructions given in detail already. See Modes (Subordinate Clauses). The relative, temporal, comparative, local, causal, final and consecutive, apprehensive, conditional and declarative conjunctions make a goodly list. But it is not necessary to go over the same ground again. Most of these conjunctions, as previously shown, are of relative origin.<sup>233</sup> All are adverbs. It was necessary to treat at length the paratactic conjunctions which antedate the hypotactic in origin and were always exceedingly abundant in the vernacular. The hypotactic belong to the more highly developed speech, but one must not think that the hypotactic conjunctions regulate the construction of the sentence. They get their meaning from the sentence, not the sentence from the conjunction. The other view is a mechanical theory of language out of harmony with the historical growth of both mode and particle.<sup>234</sup> Hypotaxis grew out of parataxis. This paratactic origin survives in many ways. Cf., for instance, the relative at the beginning of sentences, as evn oi=j (Lu. 12:1). So also o[ti in 1 Jo. 3:11 f. The Greek is particularly rich in its subordinating conjunctions as compared with the Sanskrit and the Hebrew. Each subordinate clause possesses a case-relation toward the principal sentence as substantive, adjective or adverb, so that the sentence expansion is on the lines of the word-relations. In general the disappearance of the ancient Greek conjunctions from the modern Greek is noticeable. `Opo,te $o`po,tan%à a;crijà me,crijà eivà evf v w|- "have entirely disappeared" (Thumb, Handb., p. 186). Thumb goes on with the story. We have w`j in sa,n and w[ste na,= 'until.' [Oti is gone before pou/ and na,, though o[pwj has revived.

PARTICLES ( AIPARAQHKAI) 1193

Na, has greatly extended its functions. Some survive greatly modified, like avfou/à eva,nà ei;teÄÄei;teà evnw|/à evpeidh,à pri,nà w`j pou/ $e[wj), pou/ $o[pou%à protou/, etc. The paratactic conjunctions are "pressed into service to form dependent clauses" as at the beginning. Parataxis turfs into hypotaxis.

VI. Interjecitions. Winer²³⁵ considers interjections to be mere sounds, and so entirely outside of the sphere of syntax and indeed of graminar. But one²³⁶ of the imperatival forms ( a;ge) is exclamatory in origin. Or is the interjection an imperative in origin? We this form still used as an interjection in Jas. 4:13. So also i;de in Jo. 1:29, i;de o` a;mnoj tou/ qeou/. Cf. deu/ro (Mk. 10:21), deu/te (Mt. 11:28). Deu/ro is very vivid in Jo. 11:43, La,zare deu/ro e;xw. vIdou, is either used absolutely (Mt. 11:10) or with the nominative (Rev. 4:1) and is of frequent occurrence. Kai. ivdou, is gold Greek, but its frequency reminds one of the Hebrew idiom. We have e;a in Lu. 4:34. Once ouva, occurs (Mk. 15:29) with the vocative. So ouvai, is found with the vocative in Lu. 6:25. It is found absolutely in Rev. 18:10, 16, 19, ouvai,, ouvai,. Twice it is used with the accusative (Rev. 8:13; 12:12), as the object of thought. Usually the dative is found with ouvai, as in Mt. 11:21; Lu. 6:24 f.; 11:42. The word occurs mainly in Matthew and Like. Sometimes we w= have with the vocative as in Mt. 15:28, w= gu,nai. So Ac. 13:10; Ro. 2:1; Gal. 3:1. There is usually some vehemence or urgency when w= is used. But not always. See Ac. 1:1; 18:14. In Ro. 10:15 w`j is an exclamatory particle, as ti, is in Lu. 12:49. It is not quite true, therefore, to say that interjections lie quite outside of grammar. Indeed, language may come from just these ejaculatory sounds, like "mama" with the babe. Tragedians²³⁷ naturally use interjections more frequently. People differ greatly in the use of "Oh" and "Ah." The English audiences are fond of "Hear, hear," while the American crowds love to clap their hands or stamp their feet. Farrar²³⁸ follows Scaliger and Destutt de Tracy in regarding them as words par excellence and as having high linguistic importance. Grammar can deal with emotion as well as with thought.

¹ Synt., pp. 802-820.

² Hist. Gk. Gr., pp. 365-433.

³ II, pp. 113-347.

⁴ Gr. of N. T. Gk., pp. 259-275.

⁵ W.-Th., pp. 356-512.

⁶ Hom. Gr., pp. 240-269.

⁷ Griech. Gr., pp. 525-550.

⁸ Lehre von den Partikeln der griech. Spr., Tl. I, 1832; Tl. II, 1S33.

⁹ Ib., Tl. I, p. 37. Schroeder (Uber die formelle Untersch. der Reclet., 1874, p. 35 f.) writes well on the obscurity of the origin of particles and the use of the term.

¹⁰ Doctrina Particularum Linguae Graecae. Ed. Secunda, 1806.

¹¹ See above.

¹² De Graecae Linguae Particulis, vol. I, 1840; II, 1842.

¹³ The Gk. Particles, 1881.

¹⁴ Etudes sur les particules grecques, R. E. A., VII, pp. 116-130.

¹⁵ Cf. Hubner, Grundr. zu Vorlesungen uber die griech. Synt., pp. 70-87.

¹⁶ Jann., Hist. Gk. Gr., p. 365.

¹⁷ Monro, Hom. Gr., p. 240.

¹⁸ W.-Th., p. 462.

¹⁹ Synt. of Attic Gk.

²⁰ Paley, The Gk. Particles, p. vi.

²¹ Ib., p. ix.

²² Farrar, Gk. Synt., p. 195.

²³ Ib.

²⁴ Gr. of N. T. Gk., p. 259.

²⁵ Hist. Gk. Gr., p. 400.

²⁶ Cf. Uhlig's ed., p. 96, and Schol. Dion. Thrax in Bekk. An., 970. 10.

²⁷ So mod. Gk., Thumb, Handb., p. 184.

²⁸ Viteau, Etude stir le grec, 1896, p. ii.

²⁹ Cf. Brun., Griech. Gr., p. 541.

³⁰ Partikellehre, I, p. 344 f. Cf. K.-G., II, pp. 171-178.

³¹ K.-G., II, p. 171.

³² Paley, The Gk. Particles, p. 14.

³³ Partikellehre, I, p. 326.

³⁴ Baumlein, Griech. Partikeln, 1861, p. 54.

³⁵ Grundr., p. 85. Cf. also Nagelsbach, Comm. de particulae ge, usu Hom. 1830, p. 4.

³⁶ Monro, Hom. Gr., p. 258.

³⁷ Brug., Griech. Gr., p. 547.

³⁸ Prellwitz, Et. Worterbuch, p. 73.

³⁹ Monro, Hom. Gr., p. 256.

⁴⁰ Hist. Gk. Gr., p. 410.

⁴¹ Monro, Hom. Gr., p. 256.

⁴² Gr. of N. T. Gk., p. 273 f.

⁴³ Klotz ad Devar., II, p. 392.

⁴⁴ Blass, Gr. of N. T. Gk., p. 274.

⁴⁵ Cf. Brug., Griech. Gr., p. 541; K.-G., II, p. 144.

⁴⁶ Monro, Hom. Gr., p. 248.

⁴⁷ Ib., p. 46.

⁴⁸ App., p. 151.

⁴⁹ Hist. Gk. Gr., p. 410.

⁵⁰ Griech. Gr., p. 544.

⁵¹ Ib.

⁵² Monro, Hom. Gr., p. 251.

⁵³ K.-G., II, p. 135.

⁵⁴ Jann., Hist. Gk. Gr., p. 409.

⁵⁵ W.-Th., p. 575.

⁵⁶ Jann., Hist. Gk. Gr., p. 409.

⁵⁷ Hartung, Partikellehre, II, p. 404.

⁵⁸ Blass, Gr. of N. T. Gk., p. 267. Jann. (Hist. Gk. Gr., p. 410) gives a very large list of illustrations of the original use of men, from anc. Gk.

⁵⁹ Cf. W.-Th., p. 576.

⁶⁰ But Blass (Gr. of N. T. Gk., p. 267) takes it to be 'from the very outset' and so the original use of 1.47).

⁶¹ Ib., p. 266,

⁶² The Gk. Particles, p. 34.

⁶³ Blass, Gr. of N. T. Gk., p. 266.

⁶⁴ Brug., Griech. Gr., p. 545.

⁶⁵ Monro, Hom. Gr., p. 257.

⁶⁶ Hartung, Partikellehre, I, p. 327.

⁶⁷ Baumlein, Partikeln, p. 198.

⁶⁸ K.-G., II, p. 168.

⁶⁹ Monro, Hom. Gr., p. 257.

⁷⁰ Griech. Gr., pp. 402, 525.

⁷¹ Cf. Monro, Hom. Gr., p. 252

⁷² II, p. 149.

⁷³ Hom. Gr., p. 252.

⁷⁴ Anon., Notes on Negative Postfixes in Gk. end Lat., 1884, p. 6.

⁷⁵ Thompson, Synt. of Attic Gk., p. 447.

⁷⁶ Brug., Griech. Gr., p. 529.

⁷⁷> Cf. Das indoger. Vol:al-System, p. 191.

⁷⁸> Cf. Gildersl., Am. Jour. of Philol., XVIII, pp. 4, 123 f.; Horton-Smith, ib., pp. 43 ff.; Brug., Griech. Or., p. 52S.

⁷⁹ The Negatives of the Indo-Europ. Lang., 1896. Cf. Delbruck, Grundr., IV, p. 519.

⁸⁰ But Draegcr (Hist. Synt., p. 133) says that this connection with the Lat. haud cannot be shown.

⁸¹ Farrar, Gk. Synt., p. 182; Jann., Hist. Gk. Gr., p. 425.

⁸² Monro, Hom. Gr., p. 259.

⁸³ Hist. Gk. Gr., p. 426.

⁸⁴ Thumb, Handb., p. 194 f.; Jebb, in V. and D., p. 339.

⁸⁵ Gr. of the N. T. Gk., Thayer's Transl., p. 344.

⁸⁶ Cf. Thouvemin, Les Negations dans le N. T., Revue de Philol., 1894, p. 229.

⁸⁷ Hist. Gk. Gr., p. 427.

⁸⁸ Gr. of N. T. Gk., p. 253.

⁸⁹ Synt. Forsch., IV, p. 147.

⁹⁰ Gr. N. T. Gr., p. 254.

⁹¹ W.-Th., p. 511.

⁹² Harv. Stu. in Class. Philol., 1901, The Use of Mid in Questions, p. 307.

⁹³ W.-Th., p. 451.

⁹⁴ Thouvemin, Les Negations, etc., p. 233 f.

⁹⁵ Blass, Gr. of N. T. Gk., p. 255.

⁹⁶ Burton, N. T. M. and T., p. 181.

⁹⁷ Jann., Hist. Gk. Gr., p. 429.

⁹⁸> Gr. of N. T. Gk., p. 254.

⁹⁹> Les Negations, etc., p. 233.

¹⁰⁰ Jebb, V. and D.'s Handb., p. 339.

¹⁰¹ W.-Th., pp. 477 ff.

¹⁰² Monro, Hom. Gr., p. 198.

¹⁰³ Thompson, Synt. of Attic Gk., p. 498. Cf. W. G. Hale, The Anticipatory Subjunctive in Gk. and Lat., Cornell Stu., 1895.

¹⁰⁴ Hist. Gk. Gr., p. 433.

¹⁰⁵ Robertson, Short Gr. of the Gk. N. T., p. 200.

¹⁰⁶ Cf. Jann., Hist. Gk. Gr., p. 430.

¹⁰⁷ Gr. of N. T. Gk., p. 255.

¹⁰⁸ N. T. M. and T., p. 184.

¹⁰⁹ Ib., p. 183 f.

¹¹⁰ On the Use of Mh, with the Participle in Class. Gk., 1897, p. 6.

¹¹¹ Cf. W. H. S. Jones., Cl. Rev., Mar., 1910.

¹¹² Griech. Gr., p. 528.

¹¹³ Monro, Hom. Gr., p. 259,

¹¹⁴ Ib.

¹¹⁵ Brug., Griech. Gr., p. 528.

¹¹⁶ W.-Th., p. 499.

¹¹⁷ Gk. Synt., p. 1S9.

¹¹⁸ Cf. W.-Th., p. 499.

¹¹⁹ Thayer's Lex., p. 461.

¹²⁰ W.-Th., pp. 495 ff.

¹²¹ Thompson, Synt., p. 448; Brug., Griech. Gr., p. 528.

¹²² Thompson, ib., p. 499.

¹²³ Monro, Hom. Gr., p. 260.

¹²⁴ The Gk. Indirect Negative, 1892, p. 1. Cf. also Babbitt, The Use of Mh, in Questions, Harv. Stu. (Goodwin Vol.).

¹²⁵ Moulton, Prol., p. 170.

¹²⁶ Encroachments of Mh, on Ouv in Later Gk., Am. Jour. of Philol., I, pp. 45

¹²⁷ Moulton, Prol., p. 170. Cf. also Birke, De Particularum mh, et ouv Usu Polybiano Dionysmco Diodoreo Straboniano, 1897, p. 14 f.

¹²⁸ Oxford inaugural Lecture, Note C.

¹²⁹ Thompson, Synt., p. 448.

¹³⁰ Moulton, Prol. p. 192 f.

¹³¹ Ib.

¹³² On Soph. Truch., 90.

¹³³ Cf. Postgate, Contrasts of Ouv and Mh,, Cambridge Philol. Jour., 1886.

¹³⁴ Gr. of N. T. Gk., p. 253.

¹³⁵ Moulton, Prol., p. 170.

¹³⁶ Ib.

¹³⁷ Vierke, De mil Particulae cum Indicativo Conjunetae Usu Antiquiore, 1876.

¹³⁸ Monro, Hom. Gr., pp. 260 ff.

¹³⁹ Ib., p. 261.

¹⁴⁰ Moulton, Prol., p. 170 f. Moulton gives an interesting note on the use of paidi,a as "lads" in the mod. Gk.

¹⁴¹ Gr. of N. T. Gk., p. 254.

¹⁴² Ib., p. 254.

¹⁴³ Moulton, Prol., p. 171.

¹⁴⁴ Thomson, Synt., p. 441.

¹⁴⁵ Moulton, Prol., p. 171.

¹⁴⁶ Ib., p. 192.

¹⁴⁷ Thompson, Synt., p. 499.

¹⁴⁸ Ib., p. 495 f.

¹⁴⁹ Ib.

¹⁵⁰> Monro, Hom. Gr., p. 263.

¹⁵¹ Synt., p. 414.

¹⁵² Jann., Hist. Gk. Gr., p. 430.

¹⁵³ Gr. of N. T. Gk., p. 255.

¹⁵⁴ Cf. Goodwin, M. and T., pp. 324 ff.; Thompson, Synt., pp. 425 ff.

¹⁵⁵ Monro, Hom. Gr., p. 263.

¹⁵⁶ Thompson, Synt., p. 255.

¹⁵⁷ Cf. Moulton, Prol., p. 231 f.

¹⁵⁸ Thompson, Synt., p. 410 f.

¹⁵⁹ Cf. W.-Th., p. 494.

¹⁶⁰ Cf. Goodwin, M. and T., pp. 389 ff.; Thompson, Synt., pp. 431-438.

¹⁶¹ Justin Martyr, p. 169.

¹⁶² Cf. W.-Th., pp. 508 ff.; Robertson, Short Gr., pp. 177 ff.

¹⁶³ W.-Th., p. 508.

¹⁶⁴ Jann. (Hist. Gk. Gr., p. 411) notes the pre-Attic h= r`a)

¹⁶⁵ Blass, Gr. of N. T. Gk., p. 260.

¹⁶⁶ The distinction between adv. and conj. is, of course, arbitrary. Conjs. are advs. just as the other particles are. Cf. Paul, Principles of the Hist. of Lang., p. 406.

¹⁶⁷ "Co-ordinating'' is from co-ordino, to range together.

¹⁶⁸ Brug., Griech. Gr., p. 551.

¹⁶⁹ Cf. W.-Th., p. 538.

¹⁷⁰ Brug., ib., p. 552.

¹⁷¹ Cf. C. Pitman, Conjunctions., p. 5 f.; Blass, Gr. of N, T. Gk., p. 261.

¹⁷² W.-Th., p. 434.

¹⁷³ Brug., Griech. Gr., pp. 529, 541 f.

¹⁷⁴ Whitney, Sans. Gr., p. 417.

¹⁷⁵ Brug., Griech. Gr., p. 530. Cf. K.-G., II, pp. 536 ff.

¹⁷⁶ Monro, Hom. Gr., p. 242.

¹⁷⁷ K.-G., II, p. 246.

¹⁷⁸ Brug., Griech. Gr., p. 530.

¹⁷⁹ This classic iJiom is a mark of Luke's literary style. But in the koinh, te is on the retreat before kai,) Jann., Hist. Gk. Gr., p. 401.

¹⁸⁰ Cf. Hammer De te Particulae Usu Herodoteo Thucydideo Xenophonteo, 1904, p. 92.

¹⁸¹ Gk. Etymology.

¹⁸² Griech. Gr., p. 542.

¹⁸³ Hellen., p. 129.

¹⁸⁴ Prol., p. 12.

¹⁸⁵ Farrar, Gk. Synt., p. 196.

¹⁸⁶ Jann., Gk. G-., p. 401.

¹⁸⁷ Cf. M. W. Humphreys, The Cl. Rev., 1897, vol. XI, pp. 140

¹⁸⁸ Abbott, Joh. Gr., p. 140.

¹⁸⁹ Cf. Deiss., B. S.; Hatch, Jour. of Bib. Lit., 1908, p. 142.

¹⁹⁰ Gr. of N. T. Gk., p. 263.

¹⁹¹ Hellen., § 129.

¹⁹² Beitr. zur Spracherklarung d. N. T., p. 35.

¹⁹³ Blass, Gr. of N. T. Gk., p. 262.

¹⁹⁴ Ib.

¹⁹⁵ So Jann., Hist. Gk. Gr., p. 407.

¹⁹⁶ Hom. Gr., p. 245.

¹⁹⁷ Griech. Gr., p. 547.

¹⁹⁸ Ib. Cf. also Hist. Gk. Gr., p. 410. Cf. Klotz ad Dev., II, p. 355.

¹⁹⁹> I, p. 156 f.

²⁰⁰ Part., p. S9.

²⁰¹ II, p. 274.

²⁰² W.-Th., p. 443.

²⁰³ Joh. Gr., p. 104.

²⁰⁴ Cf. Abbott, Joh. Gr., p. 105.

²⁰⁵ Unters. uber griech. Partikeln, p. 7.

²⁰⁶ Paley, Gk. Particles, p. 1.

²⁰⁷ K.-G., II, p. 286.

²⁰⁸ Ib.

²⁰⁹ Gr. of N. T. Gk., p. 269.

²¹⁰ Joh. Gr., p. 100.

²¹¹ Ib., p. 99.

²¹² Blass, Gr. of N. T. Gk., p. 269.

²¹³ Griech. Gr., p. 550.

²¹⁴ Blass, Gr of N. T. Gk., p. 268.

²¹⁵ Ib.

²¹⁶ Jann., Hist. Gk. Cr., p. 406.

²¹⁷ W.-Th., p. 440.

²¹⁸ Griech. Gr., p. 541.

²¹⁹ Cf. Margolis, The Particle h; in O. T. Gk. (Am. Jour. of Sem. Lang. and Lit., July, 1909).

²²⁰ Gr. of N.T. Gk., p. 265.

²²¹ Hom. Gk., p. 253.

²²² II, p. 317.

²²³ Gr. of N. T. Gk., p. 272.

²²⁴ Cf. K.-G., II, p. 317 f., for the discussion of the theories. So Brug., Griech. Gr., p. 519.

²²⁵ Abbott, Joh. Gr., p. 102.

²²⁶ Monro, Hom. Gr., p. 253.

²²⁷ Griech. Gr.,p.549.

²²⁸ Monro, Hom, p. 255.

²²⁹ Abbott, Joh. Gr., p. 165.

²³⁰ Ib., p. 168.

²³¹ Cf. K.-G., II. p. 326. See also Weymouth, App. A, Rendering into Eng. of the Gk. Aorist and Perfect, 1894.

²³² Blass, Gr. of N. T. Gk., p. 272.

²³³> On the relative origin of conjs. like o[tià o[teà o[pwjà w`jà e[wj see Baron, Le Pronom Relatif et la Conjonction, 1891, pp. 95 ff.

²³⁴ Cf. Nilsson, Die Kausalsiitze im Griech. his Arist. See also Gildersl., Am. Jour. of Philol., 1907, p. 354 f.

²³⁵ W.-Th., p. 35

²³⁶ Cf. Moulton, Prol., p. 171 f.

²³⁷ Muller, De interjeetionum apud Sophoclern, Euripidem que Usu, 1885, p. 3.

²³⁸ Gk. Synt., p. 201.