984x 
 
\--:M. 
 
THE ENGLISH FUTURE; 
 
 ITS ORIGIN AND DEVELOPMENT. 
 
 A DISSERTATION 
 
 WRITTEN FOR THE DEGREE OF 
 DOCTOR OF PHILOSOPHY 
 
 FROM THE 
 
 UNIVERSITY OF LEIPZIG 
 
 BY 
 
 FRANCIS ADELBERT BLACKBURN 
 
 OF SAN FRANCISCO, U. S. A. 
 
 ^-*^^ 
 
 LEIPZIG -REUDNITZ 
 
 PRINTED BY OSWALD SCHMIDT. 
 
 MDrcCLXXXXTT. 
 
«••*•' 
 
 o^^^ 
 
 J^& 
 
THE ENGLISH FUTURE; 
 
 IT8 ORIGIN AND DEVELOPMENT. 
 
 The student of English, at the outset of his work in the 
 ■grammar of the language, finds in the inflection of the verb 
 a paradigm called the "future tense", made up of the two 
 verbs shali and will combined with the verb that serves as 
 t he model of conjugation ; e. g. 
 
 I shall lovBy Wc shall lave, 
 
 Thou wilt love, You tvill love. 
 
 He will lovCy They tvill love. 
 
 He is told, moi-eover, in the text of his book or by his teacher 
 that in English the future tense is composed of the "auxiliary 
 verbs" shall and will prefixed to the infinitive of the verb 
 whose future he may wish to form, shall being used in the 
 first person and irill in the second and third. 
 
 This statement of the elementary instruction books, 
 though not complete, is in a certain sense true; to express 
 a future action the language often makes use of the formula 
 given, though not to the exclusion of other modes of ex- 
 pression, and thougli this combination, in spite of the name 
 given to it, does not always have the sense of a future. 
 Whether it should be called a "tense" at all is a question of 
 terminology, the answer depending on the definition of the 
 word "tense", but when it expresses the same meaning as the 
 future tense of other languages there is no impropriety in 
 calling it a "future". In the following discussion this name 
 is used to denote the periphrasis composed of shall or will 
 and the infinitive, Avhen it has only the future meaning, i. e., 
 
when shall or will lias sunk to a mere tense-sign and no 
 longer suggests to the speaker or hearer any idea of obligation 
 or wish, but only that of time ; the same that was suggested 
 to a Roman by the h of the Latin future. When it is 
 desirable to make a distinction between this purely temporal 
 meaning and that which originally belonged to the auxiliary, 
 the name "pure future" or "simple future" is used to denote 
 the former and the latter is characterized as a "mixed future". 
 
 The variation between sluill and will according to the 
 person of the subject of the verb takes place in modern usage 
 only in two cases; 1st., when the formula has become a simple 
 future and is used to state that an action is to take place 
 later; 2nd., when it contains beside the future meaning that 
 of a promise or threat, or the expression of a resolve or 
 determination of the speaker. In the former case modern 
 usage requires the use of shall in the iirst person and will in 
 the second and third; in the latter, on the contrary, will is 
 generally, though not invariably, used in the first person 
 and shall in the others. When the formula retains anything 
 of the original meaning of the auxiliary verb, the choice of 
 that verb depends not on the person of the subject but on 
 the desired meaning. 
 
 The use of a periphrasis to express the modification in 
 time that is expressed in other languages by a flexional form 
 called the "future tense" is common to all the Teutonic lan- 
 guages, but the same auxiliary verb is not used in all, and 
 the variation of the auxiliary according to the person of the 
 subject is peculiar to English. The purpose of the following 
 essay is to trace the growth of the English usage from its 
 beginning to the time when it became the normal mode of 
 expressing the future. In doing so the following topics are 
 considered; — 
 
 1. The methods of expressing the future used in the 
 primitive Teutonic speech. 
 
 2. The formulas for such expression of the future that 
 tlie English tongue, at the beginning of its existence as a 
 separate language, inherited from the mother-speech. 
 
o — 
 
 3. The various modew of expressing the future that have 
 been iu use at different periods of the language, and the 
 development of the present usage. ^ 
 
 I. 
 
 Among the various formations that are called futures in 
 the grammars of the various Indo-European languages is one 
 which is found in several widely separated groups and which 
 serves as a future in all. It may therefcft-e be assumed that 
 this form existed and performed the function of a future in 
 the primitive Indo-European speech. Its special characteris- 
 tic is a stem formed from the verbal root by means of the 
 suffix sje, and its original force is assumed to have been 
 modal, suggesting the idea of willingness or inclination. It 
 is impossible to decide, of course, whether the use of this 
 form in the sense of a future was at a remote period universal 
 and was afterwards dropped by some of the groups of deri- 
 vative languages and kept by others, or whether its use 
 was a dialectical peculiarity shared b}^ certain dialects and 
 surviving in the corresponding groups of languages that have 
 been developed from them. That a flexional future may be 
 lost and replaced by a new formation is shown in the case 
 of French, for example, where a periphrasis with habeoj 
 •^have", replaced the older future of the Latin, and that the 
 same loss of the older future has taken place in certain 
 cases is proved by the Balto-slavic group, for the Baltic 
 languages have kept the old form though the Slavic branch 
 has lost it, only a single form surviving to prove its former 
 existence. Whether the Teutonic languages ever possessed it 
 cannot be proved; no trace of it is found in any language of the 
 
 ' The statistics given in the following pages are the result ot 
 personal examination; none are taken from other sources. It cannot 
 be claimed, of course, that they are absolutely without error, but they 
 have been made with care and it is believed that the errors are too 
 few to aifect the correctness of the conclusions drawn from them. 
 
— 6 — 
 
 group and there is no evidence that it existed in the primitive 
 mother-tongue from which they have sprung. If we assume 
 its existence in some remote period we must also assume 
 that it was already lost before the development of the separate 
 dialects. 
 
 How then was the future expressed in the primitive 
 Teutonic speech ? No positive answer to this ({uestion is 
 possible; we can only infer Teutonic usage from the usage 
 of the later tongues. If we iind in these a general agreement 
 in any point of grammatical usage, it is a fair inference that 
 the usage is an inheritance from the parent speech, and in 
 reconstructing the primitive forms and syntax of that speech 
 we may properly introduce into it whatever is common to 
 all the languages or to several of them. Moreover, we may 
 assume, as a general proposition, that the older a language 
 is, the nearer it stands in forms and usage to the primitive 
 tongue. With these principles in mind we therefore take 
 up the Teutonic languages separately with reference to their 
 method of expressing the future. 
 
 1. Gothic. 
 
 In the Greek text corresponding to those portions of the 
 Gothic Bible that have been preserved about 650 cases of 
 the future indicative occur. In some of these the tense has 
 a modal force ; they are therefore not counted in the statistics 
 given here. They include the use of the future in commands 
 and in rhetorical questions, in modifying clauses of purpose, 
 of condition and of result. About 75 such cases occur and 
 the present optative is the regular construction in Gothic; 
 the present indicative is found once or twice. 
 
 Aside from these modal uses of the Greek future and a 
 few cases where the Greek form is the same as the aorist 
 subjunctive, the syntax allowing either, or where the Greek 
 construction is not clear or is changed in translation, we 
 have the following cases with the corresponding construction 
 in Gothic. 
 
— 7 — 
 
 Direct statements, translated by 
 
 The present indicative ..... . 456 ^ 
 
 The i)resent optative 14- 
 
 Periphrasis, hahan and the infinitive . . 2 
 „ dnginnan „ » „ . . 2 
 
 Dependent statements with ort, translated by 
 
 The present indicative 46 
 
 The present optative 5 '^ 
 
 Periphrasis, hahan and the infinitive . . I 
 Simple relative clauses, translated by 
 
 The present indicative 28 
 
 The present optative 1 ^ 
 
 Direct questions, translated by 
 
 The present indicative 32 
 
 The present optative 6 ^ 
 
 Periphrasis, opt. of shilan and the infin. . 1 
 The Greek future indicative in the constructions named 
 is therefore translated into Gothic 
 
 By the present indicative, 562 times, 
 By the present optative, 26 times, 
 By a periphrasis, 6 times. 
 
 To these may be added the cases of ov ^it] with the 
 subjunctive used in the sense of a future indicative, as 
 follows ; 
 
 Direct statements, translated by 
 
 The present indicative .... 20 
 The present optative 1 ^ 
 
 ' Including 7 cases in which the indicative and optative are alike 
 in form. They are classified with the indicatives on account of the 
 great preponderance of that construction. 
 
 - Viz. Jno. viii, 55; ix, 21, Luke i, 20; vi, 40; x, 6; xvii, 8; Mark 
 ix, 35 ; II Cor. ix, 10, [twice] ; xi, 30, xii, 6 ; Gal. v, 10 ; vi, 4 ; I Tim. 
 vi, 8 ; Neh. vii, 3. 
 
 a Viz. Mat. vi, 7 ; Jno. v, 45 ; xvi, 26 ; Kom. viii, 49 ; Phil, i, 22. 
 
 4 Viz. II Thess. iii, 3. 
 
 5 Viz. Jno. viii, 22; Luke iii, 14; xviii, 8, 18; xx, 15; Mark x, 17. 
 ^ Viz. Jno. viii, 52. 
 
— 8 — 
 
 Dependent statements with (ko, translated by 
 The present indicative .... 1 
 Of the six cases of periphrasis three are cases of a fu- 
 ture coupled with the present of the same verb, viz. ; 
 II Cor. XI. 12. de nom, xai tcoo^gm, 
 
 il> I>atei tauja, jah taujan liaba ; 
 
 Philip. I, 18. iv Tovro) x^iQO) dlld xal ^aQf^GO^ao^ 
 
 , in pamma fagino akei jah faginon duginna; 
 II Tliess. Ill, 4. noiehe ycai noLtjaeze 
 
 taujip jah taujan habaip. 
 The repetition of the present in these cases would fail to 
 show the difference of the original^ and the use of a peri- 
 phrasis is justified as being the best means that the language 
 possessed to express the idea of the Greek text. 
 In a fourth case 
 
 Luke I, 66, tI aga TC navdiov tovto eGxai-^ 
 
 Hwa skuli pata barn wairtan? 
 we have not only a periphrasis but also the optative of the 
 auxiliary verb. The use of the optative is paralleled by its 
 occurrence in other questions of like character, expressing 
 wonder and surprise and not asked for the purpose of getting 
 an answer, and in the use of the optative the question belongs 
 with the other rhetorical questions, but is included in the 
 above statistics because of the periphrasis. Skulan with the 
 infinitive is common in Gothic, but is not used to translate 
 the future, though it regularly renders ^lellia with the infin- 
 itive, which does not differ much in meaning from a future. 
 As the connection gives to the question here the sense that 
 belongs to ^wfUw, it is possible that the translator purposely 
 avoids a strict verbal translation in order to give the implied 
 force of the question more clearly than the Greek word, 
 aside from the context, expresses it. 
 
 In the two remaining cases of periphrasis it is not easy 
 to see any good reason for departing from the usual method 
 of translation. They are 
 
 John XII, 26, bVroi' ii^il iyw, 8K8i ycai 6 dLaxovog b ifio^ earai^ 
 
 parei im ik paruh sa ambahts meins wisau habaip, 
 
— 9 — 
 ancl Luke VI, 25, ormi v^uv ol yeXolvrsg vvv, on nsv&riGere xui 
 
 yj.avaeie, wai izwis jus hlalijaiidaiis iiu, imte gaunon 
 
 jail gretan duginnid. 
 
 Streitbei'g finds a reason for the periphrasis in the con- 
 trast of time with that of the preceding verb. ^ But if such 
 a contrast prevents the use of the present, why is ^aycccQtoc 
 01 yclaiovreg vvv^ oto yelaGfre, only four verses before, rendered j^ 
 by "audagai jus gretandans nu, unte ufhlohjanda ?" Or is this 
 explained by the fact that ufhlohjanda is a compound and 
 therefore "perfective", while (jaunon and gretan form no such 
 compounds ? Such an explanation might be satisfactory if we 
 could see any reason why they should not be compounded 
 as well as Jdohjan, but it would not help at all with the other 
 case for there is a perfective verb corresponding to wisan, viz. 
 tvair\H(n, and its use is very frequent. If it be objected that 
 ivisan must be used here, because a ''durative" future is re- 
 quired, the question arises why tvair\>an is used in other passages 
 where the continuance of the state is implied as much as here, 
 e. g. Luke I, 15, "wair|)i|) auk mikils in andAvairl>ja fraujins", 
 Tie shall be great in the sight of the Lord, or Luke VI, 35, 
 "wairl)il) sunjus hauhistans", ge shall be the children of tlw 
 Highest. 
 
 In the 27 cases of the optative, also, no good reason is 
 apparent for the variation from the normal method of trans- 
 lation. In some of them it is possible that the translator 
 conceived that the original contained a command, but this 
 explanation will apply to but few. Especially striking are 
 two or three cases where the Greek has two futures coupled 
 by a conjunction, the first of which is rendered by the in- 
 dicative and the second by the optative; e. g. 
 
 Luke XVII, 8, xa/ ^lara xavra (fay^aai xai nleaat av, 
 
 jan bil)e gamatjis jah gadrigkais l>u. 
 Bernhardt finds an explanation of these cases in the fact 
 that the action of the second verb is in some sense a result 
 
 ' In his article in PBB, XV, 70 tf. The use of "perfective" verbs 
 is considered farther on. 
 
— 10 — 
 
 of that of the first, or otherwise conditioned by it. The optative, 
 according to his explanation, ''marks the more remote action, 
 dependent on the fulfilment of the first condition". He trans- 
 lates the passage given above, "dann issest du und spater 
 [d. h. vielleicht] trinkst du". 13ut in view of the fact that 
 this variation is exceptional, both verbs being usually rendered 
 in the same way, the explanation seems forced, and if the 
 translator meant by the change of mood to imply any such 
 idea as Bernhardt's translation suggests, he must have found 
 it outside of the Greek text. Even if the explanation be 
 accepted, it does not account for the cases in which a single 
 future is rendered by the optative. For example, eaim is 
 occasionally ^ translated by sijai or ivair[Hd^ though ist or 
 wair\iil> is far more frequent, and no difference is apparent in 
 the thought of the passages diff'erently rendered. 
 
 The statistics given above show that the normal way of 
 translating the Greek future into Gothic is by the present 
 indicative, this construction being used in nearly 95 per cent 
 of the cases that occur. Besides this normal usage we find 
 a few cases of the optative and a very few of periphrasis, 
 which latter in most cases, if not in all, have a special reason. 
 
 2. Old High German. 
 In Old High German the case stands much the same as 
 in Gothic. In the first hundred sections of Tatian the Latin 
 future is translated 
 
 By the present indicative . . 288 times. 
 By the present optative ... 7 times, '^ 
 By a periphrasis with seal, scalon 2 times. 
 There is also one case of a periphrasis with mmj, which 
 may perhaps be due to an intentional variation of the thought 
 to express more clearly the surprise of the speaker. It 
 occurs in the question of Mary at the annunciation; -'Quo- 
 modo fiet istud?" Cases of the future in commands are not 
 included ; they are regularly rendered by the imperative. 
 
 » E. g. Luke vi, 40; Mark ix, 35; II Cor. iii, 8. 
 
 2 Viz, 4,13; 13,3 [three times]; 39,2; 44,17 [twice]. 
 
— 11 — 
 
 The two cases of periphrasis offer no good reason for 
 a variation from the usual mode of translation. The first 
 occurs in 4, 1 ; vocabitiir, "sin namo seal sin." But the same 
 word is rendered by "uuirdit ginemnit" in 3, 17. The second 
 case is in 13, 16; quid facicmus'^ "uuaz sculon uuir tuon", 
 and in the next verse the same question is repeated and 
 rendered by "uuaz tuomes". In both cases the thought is 
 exactly parallel to that of the passages translated by a 
 periphrasis. 
 
 In the first twenty-five psalms of Notkei''s paraphrase 
 we find the Latin future rendered 
 
 By the present indicative . . 172 times, 
 By a periphrasis with sol, . . 3 times. 
 The usage of Old High German is therefore the same 
 as in Gothic, normally the present indicative; occasionally 
 the optative, less often a periphrasis. 
 
 3. Old Norse. 
 
 In old Norse the usage cannot be mathematically stated, 
 for translations are wanting for such an examination as we 
 have made in Gothic and Old High German. An examination 
 of the Eddie poems shows that the same devices for ex- 
 pressing the future were also in use, but the optative has 
 almost entirely disappeared and the present indicative does 
 not, as in the languages already examined, hold the leading 
 place. In round numbers we may estimate the occurrences 
 of the present at about one third of the Avhole ; the other 
 two thirds are expressed by a periphrasis with skulu or 
 nuinu, the latter being about twice as frequent as the former. 
 Especially striking among the instances of the present is the 
 large number of cases of vcr'^a. Of the use of this verb in 
 a future sense we shall have occasion to speak farther on. 
 
 4. Old Saxon. 
 
 In Old Saxon the use of periphrasis has gone much 
 farther. It is not easy to state the usage of tlie Heliand, 
 for though passages from the Gospels occur often, theuthor(a>) 
 
— 12 — 
 
 usually elaborates them into very loose parai)lirases, and the 
 auxiliary verbs are often introduced where the original con- 
 tains a straightforward statement. For example, Mat. V, 14 
 is extended to eleven verses, and the Vulgate esfis becomes 
 "gi . . nu for^ skulun . . wesan". The addition of the ad- 
 verbs nu for^ gives a future sense here, but it is not certain 
 that the writer uses sJculun wesan as a future, especially as 
 the original has the present. Not only shall but also other 
 auxiliaries are often used in the Heliand to form periphrases 
 that seem to express no more than the simple verb. For 
 example, in 1475 we have "Ok skal ik iu seggean noh", and 
 three verses farther on, without apparent difference of 
 meaning, "Than seggio ik iu te waron ok". So too in 1508, 
 "Than williu ik iu seggean that san ni swerea neoman" cor- 
 responds to the Vulgate Mat. V, 34, Ego autem dico vobis 
 non juraro omnino, and just after this passage "skal ik . . 
 gibeodan" is used in the same way. But though these phrases 
 are not necessarily future in sense there are many passages 
 where the context shows that a future sense is needed and 
 an examination of these shows that the same methods of 
 expressing it prevail here as in the other languages cited. 
 For example, the Vulgate "Pater tuus . . reddet tibi" is ren- 
 dered in 1559 "Than most thu eft geld niman", and in 1565 
 "Thanan thu skalt Ion neman"; the Vulgate "dimittet et 
 vobis pater vester celestis delicta vestra" becomes "than 
 alatid iu waldand god . . firin-werk". These three quotations 
 fairly represent the usage of the writer. The present, 
 though common, is less so than a periphrasis with skulan or 
 motan, between Avhich usage is about equally divided. There 
 are also frequent cases of a periphrasis with tvillian, but it 
 is not clear in any of them that the phrase formed is a 
 simple future. For example in the verse next after the last 
 quotation, the Vulgate "nee pater dimittet" is paraphrased 
 by "ni wil iu waldand-god grim-werk fargeban", and in both 
 cases the dependent condition, "si dimiseritis" is expressed 
 by willian. But the connection is such that an expression 
 of willingness is quite appropriate, and the variation of 
 
— 18 — 
 
 thought from "if ye forgive not" to "if ye will not forgive" 
 is much less than we find in many of the paraphrases of the 
 Heliand. 
 
 The case is clearer in the glosses and in the interlinear 
 version of the psalms, which Heyne has included in his edition 
 of the lesser Old Saxon remains, though in his preface he 
 calls them "niederfrankisch". In the Lipsius glosses, not count- 
 ing the cases where the verb-form is glossed by the infinitive, 
 or, if passive, by the passive participle, thus giving a simple 
 definition, the Latin future is glossed 
 
 By the present indicative ... 8 times, 
 By a periphrasis with sculan . . 35 times. 
 In the version of the Psalter the periphrasis is the normal 
 rendering. We have 
 
 The present indicative .... 9 times, 
 
 The present optative 3 times. 
 
 Periphrasis with sculan .... 180 times. 
 The cases of the optative are a little doubtful in view of 
 the inexactness of the spelling, and it is noticeable that eight 
 of the nine cases of the present indicative are passive and 
 contain the verb uuerthan. It might be supposed that the use 
 of the present here is due to the desire to avoid a long and 
 cumbrous circumlocution, since the periphrasis in the passive 
 requires three words, but we find four other passages in which 
 the periphrasis is used in spite of its length. Another ex- 
 planation will be offered later. We may state the usage of 
 these low German remains, therefore, to be normally a peri- 
 phrasis with shall, rarely the present indicative; the use of 
 the optative being very rare and uncertain. 
 
 5. Old Friesic. 
 In Friesic the character of the remains of the older 
 dialects is such that we cannot expect to find the future 
 except in a quasi command, or in a conditional clause refer- 
 ring to the future, where in Latin the completed future would 
 be the usual construction. In the latter case the present 
 indicative is the usual construction in Friesic; in the former 
 
— 14 — 
 
 we find tlie optative and periphrasis with shila [shall], and 
 aya [have to]. These correspond to the constructions in the 
 Latin laws, where we find not only the future indicative, but 
 also the imperative, the subjunctive and phrases with debet 
 tenetur^ etc., and the idea is expressed in Friesic in the waj^ 
 that we should naturally expect to find it, by words or con- 
 structions that convey the idea of obligation. It must be 
 noted that we are not dealing with a simple future here, and 
 that tliis fact, together with the late date of the Friesic laws, 
 renders the language of much less value for the question we 
 are here considering, the mode of expressing the future in 
 primitive Teutonic. 
 
 6. Old English (Anglo-Saxon). 
 
 The oldest translations are the glosses and tlte Vespasian 
 Psalter and Hymns. In the latter about 1100 cases of the 
 Latin future are rendered by the present indicative. This 
 includes all the cases that occur except 14, which seem to be 
 corrupt, the future being translated by the past in eight ^ of 
 them and by the past optative in one. ^ The other five -^ are 
 rendered by the present optative, which may be a genuine 
 usage, but in four of the five the change of a single letter 
 in the Latin word would make it a subjunctive, and in view 
 of the 1100 cases of the present indicative it seems at least 
 possible that the translator misread his text or that the 
 passages are corrupt. 
 
 The same results are obtained from an examination of 
 Wiilker's "Anglo-Saxon Glosses". There are 1H3 cases of Latin 
 futures glossed by the present indicative and two cases of 
 the optative, if they are not corrupt, as six other cases 
 plainly are. 
 
 The conclusions drawn from the above statistics in regard 
 to the method of expressing the future in primitive Teutonic 
 are the following; — 
 
 » Viz. XV, 10, xlv, 10; cix, 4; cxlviii, 6. Hymns 4 and 7 [three cases.] 
 
 '■J Viz. xlix, 21. 
 
 3 Viz. xxi, 8; xxiv, 3; xxxiii, 16; Iviii, 7, 15. 
 
— 15 — 
 
 1. There is no evidence that the Teutonic mother-speech 
 ever possessed a special form for the future tense. 
 
 2. The present indicative, less often the present optative, 
 was used as a future. 
 
 3. Periphrastic forms of expression were in use, as in 
 other languages, in a sense approaching that of a future, but 
 seem to have been used very rarely to express a simple future, 
 and apparently only for special reasons that rendered the 
 use of the other constructions unsuitable. The verbs most 
 often employed in these were the praeteritives. 
 
 These conclusions are not vitiated by the objection that 
 naturally suggests itself, that the old translations are too 
 slavishly literal to give us a picture of the language as it 
 really sounded in the mouths of the people, for the pains 
 taken by the writer to make his translation like the original 
 in every point, though it resulted in unidiomatic sentences 
 and stiff and awkward constructions, would at the same time 
 lead him to use special care to reproduce the forms of the 
 original as nearly as his own tongue was capable of doing it. 
 In the translation of a Latin or Greek future, therefore, he 
 would be likely to use the form that expressed the future 
 idea in his own language, even if the original occurred in a 
 construction that could be better expressed in his own speech 
 in some other way. The translations of a class of beginners 
 in Latin or Greek will furnish a very good representation 
 of the character of the old translations of the Scriptures, and 
 the painstaking care of the tyro to reproduce tenses and 
 moods in his translation is probably no greater than that 
 of the old versions from which we have taken a large part 
 of the statistics we have given. 
 
 Two of the three methods of expressing a future have 
 analogies in other languages. The use of a modal form like 
 the optative is almost universal; even those languages that 
 have a fully developed future have modal forms in use also. 
 Such forms are found, for example, in Indie, ^ in Iranic, 
 
 See illustrations in Whitney's Skt. Gram., 576, 577. 
 
— 16 — 
 
 and in Greek, though there is a special form for the tuture 
 in each. Moreover, so-called futures are often modal formations, 
 as in Latin and Keltic, and the original Indo-European 
 s-future was at first, no doubt, a mode rather than a tense. 
 We need not be surprised, therefore, to find the optative used 
 as a future in the Teutonic tongues ; it would not be strange, 
 in view of the lack of a special future form, if its use were 
 far more frequent. Periphrastic forms also are not unusual; 
 we need only refer to fiMo) with the infinitive 'in Greek, to 
 the Latin fadurus sum, to the periphrastic future of Sanscrit, 
 composed of the verb asmi, "am", and a nometi agentiSj and 
 in particular to the periphrasis in Church Slavonic, which is 
 entirely analogous to that of the Gothic. In general, however, 
 periphrasis is comparatively rare in the older languages and 
 increases in frequency in later works. The reason for its in- 
 crease in the Teuton! clanguages until it became the normal idiom 
 of the various dialects will be considered in the proper place- 
 Both these methods, however, as we have seen, were rare, 
 in spite of analogies in other languages and in spite of their 
 apparent adaptation to the end in view, and the normal way 
 of rendering the future in translation was by means of the 
 present indicative, which at first sight seems the least suitable 
 of the three. Its use would involve such ambiguity, one might 
 suppose, as to render it unfit for translation from Latin or 
 Greek, especially in the case of the canonical Scriptures, the 
 letter of which was so closely followed by the old translators. 
 The ambiguity seems to have been avoided, at least to some 
 extent, by the distinction made between "perfective" and 
 ^'imperfective" verbs, as in the Old Slavic languages. According 
 to this distinction, all verbs that imply the completion of the 
 action are perfective; such as do not contain this notion are 
 imperfective. By composition with prepositions the imperfec- 
 tive verbs are made perfective. 
 
 Now in Old Slavic the present of the perfective verbs 
 has a future sense, while the future of an imperfective verb 
 must be expressed by periphrasis, for which purpose the lan- 
 guage makes use of chosta, „I will," imami, „I have" or na- 
 
 J 
 
— 17 >- 
 
 cm^, „I begin," with the infinitive. These forms are less 
 common however, than one would expect, for there are but 
 few verbs that cannot be compounded with prepositions and 
 thus made perfective, so that the present form has the future 
 function. 
 
 The reason for this peculiarity lies in the idea of the 
 completion of the action that belong:s to such verbs. We are 
 accustomed to classify time as past, present and future, for- 
 getting that the present, unlike the past and the future, has 
 no duration but is merely the dividing line between the past 
 and the future. Such a statement as „He is finishing his 
 work" is logically contradictory, for „finishing" being an in- 
 stantaneous act, would, if present in time, become past before 
 the statement could be uttered. The sentence means „He is 
 at his work and near the end of it," „He will soon finish his 
 work." The present of a verb that implies the completion 
 of the action necessarily implies that completion in the future, 
 and may therefore properly be employed as a future in a 
 language that has no other way of expressing it. 
 
 This distinction, being one of meaning, exists in all lan- 
 guages and the effect of composition in changing the action 
 from imperfective to perfective is found in othei* tongues as 
 well as in the Slavic. That the primitive Teutonic stood on 
 the same plane as the Slavic in this respect has been main- 
 tained by many philologists and the question has been fully 
 considered by Streitberg in his article on ,,Perfective und 
 imperfective Actionsart im germanischen," in Paul & Braune's 
 Beitrage, Vol. XV, pp. 70—177. He cites from the Gothic 
 numerous examples in proof of the distinction and in particular 
 shows the effect of composition with a preposition-adverb in 
 changing an imperfective verb to the perfective meaning. His 
 views have been strongly opposed by others and we need 
 not examine them here in detail, as we are only concerned 
 with that part of the article that treats of the use of the 
 present of the perfective verbs to translate the Greek future, 
 as in the Slavic Scriptures. His citations furnish proof enough 
 to warrant the belief that the distinction existed at some 
 
 2 
 
_ 18 -^ 
 
 remote period, even if they do not convince us that such 
 was the case at the time of the Gothic remains, and it 
 seems quite probable that the use of the present in a 
 future sense, as in Slavic, was based on this difference of 
 meaning. 
 
 Streitberg points out, however, that the lack of iterative 
 verbs in Gothic necessarily makes the present of all the per- 
 fective verbs ambiguous in time, since the same form serves 
 for both present and future, and also that certain imperfective 
 verbs, which are not rendered perfective in Gothic by composi- 
 tion, their meaning not being capable of such modification, 
 use the present as a future, like the perfective verbs, instead 
 of resorting to periphrasis. The natural result of such ambi- 
 guity is the loss of the sense of difference between the two 
 classes of verbs and it seems possible, at first sight, that 
 this effect had already made itself felt at the time of the 
 Gothic Bible-translation. Streitberg is compelled, in order 
 to bring all cases into accord with his theory, to assume a 
 difference in the character of the action where the Gothic 
 shows a difference in the choice of the verb, though none 
 exists in the Greek and though the connection of thought 
 suggests none. This is especially noticeable in his discussion 
 of the future of eifii The Gothic has two words to render 
 this verb, wisan, [imperfective], and wair\mn, [perfective]. If 
 now the distinction were clearly felt, we should expect the 
 future to be translated by the latter wherever a completed 
 action in the future is implied, i. e. the attainment of a state 
 or condition, and by the former only in case of a durative 
 action, i. e. the continuance in a certain state or condition. 
 In the latter case, moreover, a periphrasis should be used, 
 but the Gothic, as was said above, fails to keep the distinction 
 in such cases. Streitberg maintains that the two verbs are, 
 used with this distinction, and translates Mark VI, 11, 
 avtycTOTeQov earai, ^odofioig ri Fo^bQQOig iv fnieQo. jtQiaewg, „SUtizo 
 ist Saudaumjam ail>ljau Gaumaurjam in daga stauos" by „ihre 
 lage wird eine angenehmere sein" and the similar passage 
 in Mat. XI, 24, yy ^odu^wv dvemOTe^ov taxai iv 7j^i8^a K^laewg, 
 
— 19 — ' 
 
 „airl>ai Saudaumje sutizo wairi>ii) in daga stauos" by „sie 
 wird ein milderes geschick treifen." There is no doubt that 
 either translation is a correct rendering of the original ; in the 
 latter case the translator chose to express only the coming of 
 punishment, leaving its continuance to be inferred from the 
 context, while the choice of isf in the former distinctly 
 expresses the durative idea. 
 
 An examination of the other cases gives like results. 
 The future of elfil occurs 77 times in the Greek text and 
 in the translation tvairpan is used 55 times and wisan 22. 
 A comparison of the passages shows that the f(>rmer is some- 
 times used though the action is durative, i. e. when the 
 continuance in a certain state or condition is foretold or 
 promised. For example. Mat. VI, 22, ^allata leik l^ein 
 liuhadein wairl>il)," thy tvhole body shall be full of light j certainly 
 implies the continuance of the state as much as y,ivhere thy 
 treasure is, there will thy heart be also/' the verse next preceding, 
 where arra* is translated by „ist". Nor is it apparent how 
 Luke XIV, 10, „l>anuh ist ])us hauhi^a," then shalt thou have 
 glory ^ implies continuance in a condition more than „audags 
 wair])is," thou shalt be blessed, four verses farther on, both 
 being the promised result of a specified course of conduct. 
 
 Nevertheless in a large majority of the cases the distinc- 
 tion that Streitberg tries to establish is entirely suitable to 
 the thought. The fact that the perfective verb wair^an is 
 used to translate the Greek future in five sevenths of the 
 cases which occur is good evidence that the distinction once 
 existed, and it is easier to explain the few cases that seem 
 to disregard the distinction by assuming that the translator 
 left the durative idea unexpressed, except by the connection, 
 than to account for the retention of the distinction between 
 perfective and non-perfective verbs, clearly observed in so 
 many cases, if the sense of this distinction had already van- 
 ished. In any case, even if the cases cited be regarded 
 as a proof that the distinction had already begun to dis- 
 appear, we are forced to admit that it must have existed 
 with its full force at an earlier period, and that the use 
 
 2* 
 
— 20 — 
 
 of the present in a future sense in the various Teutonic 
 languages is based upon it. 
 
 In the othei Teutonic languages, this distinction, as a 
 means of expressing the future, had to a great extent dis- 
 appeared. No doubt the difference was still felt, as it now 
 is, but with the exception of one or two cases, which seem 
 to be survivals of the older usage, and which were used, 
 no doubt, without any consciousness of their origin, there 
 is no sign in any of them that the writers, in their render- 
 ing of the Latin future, had any clear sense of the differ- 
 ence between perfective and imperfective action. In Tatian 
 we find vocabis rendered by "nemnis" in 2, 5 and by "gi- 
 nemnis" in 3, 4, showing that no distinction is made between 
 simple verbs and compounds; vocahitur is rendered by "uuir- 
 dit ginemnit" in 3, 7, and by "ist geheizan" in 25, 6, showing 
 that ist and uuirdit are used without distinction; exietj a 
 perfective verb is rendered by "quimit", an imperfective verb 
 in 8, 3; erit is translated fourteen times in the first fifty 
 sections by "ist^' and only once by "uuirdit", etc. The same 
 results follow an examination of the oldest English trans- 
 lations, and in the other languages, as we have seen, the 
 use of the present in a future sense was already on the 
 wane, having been replaced in a great degree by the use 
 of periphrasis. 
 
 Whether we should set the date of the beginning of the 
 loss of distinction between perfective and imperfective verbs 
 as far back as the primitive Teutonic or assign it to each 
 language separately after the division, is not, for our pur- 
 pose, a matter of importance. It is plain that in the mother- 
 speech the present was the normal way of expressing the 
 future, that along with this the optative was also in use, 
 though only occasionally, and that a periphrastic form of 
 expression could be resorted to in case of some special ambi- 
 guity or of a wish for special exactness in time. The di- 
 stinction between perfective and imperfective verbs served 
 to prevent the ambiguity in time involved in the double use 
 of the present, as long as this distinction lasted; how it 
 
— 21 — 
 
 may have been avoided in ordinary cases, after the distinc- 
 tion was lost, will be considered farther on. 
 
 11. 
 
 ENGLISH BEFORE THE NORMAN CONQUEST. 
 
 The remnants of the oldest English, the Anglo-saxon, as 
 we have seen above, render the Latin future by the present 
 indicative; the few cases of other constructions are doubtful 
 and may be corrupt or due to blunders. Moreover, there is 
 no hint of any difference of usage according to the character 
 of the action ; if such a distinction ever existed it was lost 
 before the date of the earliest writings. 
 
 This method continued to be the normal one up to the 
 time of the Norman Conquest, In Alfred's translation of 
 (jregory's Pastoral Care there are 61 cases of the Latin 
 future rendered by the pi'esent indicative, and 25 which 
 express a command and are rendered by the imperative. 
 Other constructions are found in a few cases, but seem to 
 be intentional variations. The translator, as one may easily 
 see and as his preface tells us, turned the book into Eng- 
 lish "sometimes w^ord for word and sometimes sense for 
 sense", making no effort in the latter case to keep the exact 
 construction of his original. 
 
 These other constructions comprise six cases of shall ^ ^ 
 nine of ivilV- and five of mayj^ all of which, however, have 
 their proper sense and cannot be regarded as futures. The 
 assumption that they are intentional variations is supported 
 not only by the sense of the passages in which they stand 
 but also by the frequent substitution in the translation of 
 different forms of expression, where the Latin allows such 
 variation without violence to the thought ; e. g. the rendering 
 
 1 Sweet's ed. pp. 91, 187, 255, 281, 311, 339. 
 
 2 „ , , 111, 243, 249, 285, 304, 419. 
 •■^ « „ „ 345, 393, 399, 443. 
 
— 22 — 
 
 of "vapulabit multis" by "he biS manigra wita wyr^e" (he 
 is worthy of many punishments), and various other similar 
 cases. 
 
 It should be noted however that the cases of the use 
 of /r^7/, with two exceptions, are in the first person and ex- 
 press a promise or resolve. This seems to have be the ear- 
 liest use of will that approaches the future meaning. In 
 Alfred's translation of Orosius, the future indicative is al- 
 ways rendered by ivill, but all the cases that occur are of 
 this kind. There are but six^ such in the work, all intro- 
 ductory to a new topic; "I will now relate" or some similar 
 phrase. We shall have occasion farther on to notice this 
 use of ivill in later writers. 
 
 Similar results are derived from an examination of 
 the translation of the Gospels. In the text of Matthew as 
 given by Bosworth and Waring, London, 1865, the present 
 indicative occurs 318 times ^ where the Vulgate shows the 
 future, and in only 15 cases is a diiferent construction found. 
 Of these fifteen cases four ^ are in the optative and one * ui 
 the imperative (the translator apparently regarding the 
 sentence as a command), and there are three ^ cases of 
 periphrasis, two with may and one with mote. The seven*' 
 other cases are due to a change of construction, or possibly 
 in one or two cases to a different reading of the Latin text. 
 It is possible, also, that the four cases of the optative and 
 the three of periphrasis may be intentional variations. They 
 occur in sentences where the construction used gives the 
 
 1 Sweet's ed. pp. 11, 15, 28, 27, 61, 251. 
 
 - Not including 18 cases where the future has an imperative sense. 
 In 15 of these the imperative is used ; the other three are rendered by 
 the indicative. For the latter there is no good reason apparent; all of 
 them occur in such connection that it is impossible to suppose that the 
 translator mistook them for statements. They are found in IV, 10 
 [twice], and V, 33. 
 
 3 VII, 19 ; X 26 [twice] ; XX, 27. 
 
 4 VIT, 15. 
 
 ^ XII, 26; XXVI, 54; XVIII, 21. 
 
 6 IV, 6 ; XII; 29 ; XVI, 22 ; XVII, 20 [twice] ; XXI, 28 ; XXVI, 53. 
 
— 23 — 
 
 general meauiiig quite as well as the future. As regards 
 the three cases of periphrasis, occasional cases of the use of 
 may or mote with an infinitive are found in the later language 
 with a sense closely approaching that of the future, but they 
 are not frequent enough to warrant us in assuming that they 
 are meant to express the future meaning here. It is more 
 probable that the variation is intentional. 
 
 That the present was regarded as the nearest equivalent 
 for the Latin future that the language could furnish is shown 
 very clearly by ^Ifric's Latin Grammar. In his translation 
 of the forms that he cites he uses the present invariably to 
 render the Latin future, but with the addition of an adverb 
 of time to make the meaning clear. Thus in his paradigms 
 we lind 
 
 staho, "ic stande nu rihte ob^e on suman timan", (p. 123) ^ 
 
 amabo, ''ic lufige gyt to daeg oSbe to mergeu", (p. 131) 
 
 amahorj "ic beo gelufod gyt", (p. 140) 
 etc. etc. etc. 
 
 So also in other forms which he calls futures, e. g. 
 '*optativo modo, tempore futuro", 
 
 Htinam amem, ''forgife god l>8et ic lufige gyt", (p. 132) etc. 
 "subjunctivo modo, tempore futuro", 
 
 cum amacero, ''^onne ic lufige gyt", (p. 133) etc. 
 But when he translates the Latin participles in -turns or 
 'Hclus he uses a periphrasis; e. g. 
 
 Iccturus Slim eras, "ic sceal raedan to merigen", (p. 136) ; 
 
 amanduSy "se be sceal beon gelufod", (p. 144); 
 
 amandus est tile, "he is to lufigenne", (p. 144); 
 
 video tcdod arum esse J 'Ic geseo paet bu wylt tsecan", (p. 150); 
 
 docturus sum eras pueros, "ic wylle tsecan to merigen ^am 
 cildum", (p. 152); 
 
 fisurus, "se be wylle oS^e sceal truwian", (p. 247); 
 
 osciilaturusj ''se be wylle obbe sceal cyssan", (p. 248) ; 
 
 oseulandus, '*se be sceal beon gecyssed", (p. 248) ; etc. etc. 
 From these examples we infer that ^Ifric did not regard 
 
 Citations are given by the pages of Zupitza's edition, Berlin, 1880. 
 
— 24 — 
 
 the periphrasis with simll or will as a satisfactory rendering 
 of the Latin future, though he uses it to translate construc- 
 tions that approach the future in meaning.^ It is noticeable 
 that the Latin participle in -ndus, which expresses the notion 
 of obligation, is rendered by shall, and that either shall or 
 will may be used to express the meaning of the active 
 participle in -turns, where no modal idea is suggested. The 
 comparative frequency of the two words cannot be inferred 
 from the examples given, but there is no choice according 
 to the person of the subject, as in modern English. 
 
 Other writings of ^Ifric show the same usage. In his 
 translation of Genesis there are about 225 cases where the 
 present corresponds to the Latin future of the Vulgate ; along 
 with these we find also 8 cases of shall and 11 of tvill. 
 These seem to be best explained, as in Alfred's translations, 
 on the supposition that the translator purposely varied from 
 the usual way of expressing the future idea because the 
 connection implied willingness or obligation, though the Latin 
 future, while not excluding such a notion, did not clearly 
 express it. Similar changes not involving the use of shall 
 or tvill are so common that the assumption of an intentional 
 change is easier than the belief that shall and ivill could 
 be used at this date to make simple futures. For example, 
 in XLIV, 16 we find "Hwset magon we cwe^an?" for the 
 Vulgate quid rcspoiidcbimus ? ; in XX, 4, Abimelech's question 
 "wilt thou slay?" becomes an imperative "slay not!"; and 
 similar changes ai'e found on almost every page. When the 
 translator thinks verbal exactness necessary or desirable, he 
 seems not to allow himself the license of periphrasis. A 
 striking instance is found in XXII, 14; "He liet l)a stowe 
 Dominus videt, \>2Bt is "god gesyh^" and gyt ys gesaed swa, 
 In monte dominus videbit, l>8et is "god gesyhcS on dune". If 
 the periphrasis with shall or ivill were regarded by the writer 
 
 * A similar case is found in the Gospel of Matthew, XXIV, 44. 
 Here, instead of the usual rendering of renturus est by "is towerd" or 
 ys to cumenne", we find "wyle cuman." 
 
— 25 — 
 
 as a simple fiUiire, we should expect to find it used in this 
 passage. 
 
 The same usage is shown in ^Ifric's Homilies. In the 
 nine that Assmann has published in Vol. Ill of Wtilker's 
 "Bibliothek der angelsachsischen Prosa" there are about 70 
 cases of the present with future meaning. Shall and ivill 
 also occur with meanings that closely approach that of the 
 future, the former 11 times, the latter 9 times. Of these 9 
 instances, one is apparently a simple future; "hit wile 
 hearmian ^inum cynerice", VIII, 152. The others occur in 
 the same use that has been observed in JElfi^ed's Orosius 
 and elsewhere in the introduction of a new topic; e. g. IIT, 
 49, "Now wylle we eow secgan", or a similar phrase. Whether 
 any of the cases of shall must be regarded as pure futures 
 without any suggestion of other meaning than that of time 
 I is not quite clear, but a part of them seem to have at least 
 no meaning beyond that of destiny, the original notion of 
 duty or obligation having entirely faded out; e. g. in VII, 
 204, "sceal eacnian" for the Latin concipiet. 
 
 The oldest English, then, seems to have no way of 
 expressing the future, but uses the present almost invariably. 
 To oui modern feeling the future seems as necessary as the 
 past and it is hard for us to conceive of a language without 
 some method of expressing it. Yet Teutonic speech lacked 
 not only the future but also the three complete tenses, which 
 the study of Latin grammar has taught us to consider almost 
 necessary to human speech, and it is plain that the primitive 
 Indo-European language had no such thing as a tense, i. e. 
 a variation of form for the sole purpose of distinguishing the 
 time of the action. That this lack was felt, however, is 
 proved by the later development of meaning that changed 
 modal forms into tense-forms, and by the evolution of new 
 forms to fulfil the functions of tenses. In like manner, we 
 cannot doubt, our Teutonic forefathers were sensible of the 
 ambiguity that followed the use of the present for both 
 present and future, though perhaps not to the same extent 
 as we, who have been trained to distinguish the tw^o. 
 
— 26 — 
 
 How then was the necessary distinction of time shown? 
 No doubt, as the quotations from ^Ifric's Grammar show, 
 by the addition of an adverb of time, whenever it seemed 
 necessary, just as the past was originally designated by the 
 adverbial particle that later became a flectional device under 
 the name of "augment". Reverence for the letter of the 
 Scriptures prevented the older translators from adding such 
 adverbs, lest they should incur the curse pronounced in the 
 Apocalypse on him that should add anything to the words 
 of the book. But in ordinary speech, we may assume, the 
 speaker would have no difficulty in making his meaning clear 
 by adding such words as were needed to define the time, 
 Even in modern English it is not unusual to use the present 
 in a future sense when an adverb of time makes the meaning 
 clear. Such expressions as T begin work to-morrow", "He 
 is coming next week", and the like, are true futures in spite 
 of their form, and suggest the future meaning to those who 
 use or hear them, but they cannot be used to convey the 
 future sense without the addition of an adverb of time or 
 such connection of thought as defines the time clearly. Even 
 the formulas which the grammar-books have named the 
 "future tense", e. g. "I shall begin work to-morrow", "He 
 will come next week", etc. contain only forms that were 
 once present in meaning; and the future sense which 
 they now convey is conventional and the outgrowth of 
 usage, as in fact it is in all futures, whether flexional or 
 otherwise. 
 
 But ^Elfric's paradigms show also the use of periphrasis 
 in the case of quasi futures like Icdurtis sum, etc., and both 
 his translations and these of Alfred show that these mixed 
 futures were close enough to pure futures in force to be used 
 for such when the translator did not feel compelled to seek 
 for verbal correctness in his rendering. The germ of the 
 later English usage was therefore already in existence, though 
 only the germ; but this seems to have been inherited from 
 the parent- speech and was available in cases where it seemed 
 necessary or desirable. Of the distinction between perfective 
 
— 27 — 
 
 and uon-peifective verbs in their way of ex])ressing the future 
 not the least trace remains. 
 
 III. 
 
 FROM THE NORMAN CONQUEST TO THE INTRODUCTION 
 OF PRINTING. 1066— 1474. 
 
 For nearly a century after the Conque&t we find but few 
 works that are of value for our purpose. Some of the Saxon 
 Chronicles were continued, one ot them for almost a century, 
 but most of the works of this period are either copies or 
 revisions of older works. 
 
 Among" the former may be mentioned the Hatton Gospels, 
 contained in a MS. later than 1150. Here we find the same 
 method of expressing the future as in the older version written 
 before the Conquest. A comparison of Hardwick's edition of 
 Matthew with the older version shows that the scribe made 
 very few changes except in orthography and in the occasional 
 substitution of a new word. In the expression of the future 
 there is no difference whatever; the passages refen-ed to 
 above in which the translator of the older version varied 
 from his usual way of rendering the future show exactly 
 the same variations in the Hatton MS. The future is regularly 
 rendered by the present, and the MS. is plainly a copy, not 
 a new version. Its value for our purpose lies in the proof it 
 furnishes that the older way of rendering the future by the 
 present was still intelligible at the middle of the twelfth century. 
 It is a fair assumption that the changes made by the copyist 
 were intended to make the language easier of comprehension 
 for his readers, and the absence of change in the expression 
 of the future idea is a negative proof that the old way of 
 expressing it was still intelligible. We are not compelled 
 to believe however that it was the normal usage in ordinary 
 speech. Versions of the Scriptures often preserve an older 
 style than that of the time when they are made, and other 
 
~ 28 — 
 
 works of an earlier date than the Hatton MS. show that a 
 change had already begun. 
 
 The Winteney version of the Benedictine Rules is contained 
 in a MS. which is believed to have been written in the iirst 
 quarter of tlie twelfth century and is therefore several decades 
 older than the Hatton MS. It contains the rules of the Order 
 in Latin and English as arranged for the use of nuns. This 
 arrangement compels frequent changes in the gender of ad- 
 jectives and pronouns, substitution of abbodesse for abbod, and 
 the like, but does not affect the use of the tenses. For our 
 purpose, therefore, a comparison with the older version is 
 instructive, since it shows the beginning of the change of 
 idiom which took place in the period we are now considering. 
 
 In the Latin text * there are 36 cases of the future. 
 Of these 22 are expressed by the present, one (a command) 
 by the imperative, in three the construction of the sentence 
 is changed and in three ^ the future idea is expressed by a 
 phrase. The remaining seven are expressed by a periphrasis, 
 in which shall occurs three times, will twice, mote and may 
 once each. Now a comparison with the older version shows 
 that all these cases are there rendered in the same way 
 except the five cases of shall and ivill ^ For these we find, 
 as in most other cases, the present. The two cases in which 
 will is substituted in the later version are in the first person 
 and express a promise; the three cases of shall are simple 
 futures. 
 
 It is not easy, of course, in the case of a revision of an 
 older work, to decide to what extent the reviser's style is 
 affected by his original. But that the use of the present in 
 a future sense was still current at the time of the Winteney 
 version is plain not only from its use in the Hatton MS^ as 
 mentioned above, some years later, but also by the usage in 
 a homily written before 1150. In this we find the present 
 
 ' As given in Schroer's edition, Halle, 1888. 
 
 2 e. g. p. 90 ; wen is ]'a3t (= erit), etc. 
 
 =< These are found; shall, pp. 19, 31 (twice); will, pp. 29, 39. 
 
 4 In Morris and Skeat's "Specimens of Early English". 
 
— 29 — 
 
 with the sense of the future 17 times, will once (a promise) 
 and shall seven times. We infer from these works that the 
 old use of the present still prevailed during the period 
 immediately following the Conquest, but we see also the 
 beginning of the use of shall, which soon becomes the normal 
 usage. 
 
 At the beginning of the thirteenth century we find a 
 revival of literary activity. To this date belong a number 
 of important works, some of which, e. g. Layamon's Brut 
 and the Orrmulum are large works and of special value for 
 grammatical study. An examination shows us that in the 
 way of expressing the future, as in many other respects, 
 these writings mark a new era in the development of the 
 language. The present has ceased to be the normal way of 
 expressing the future idea, though it has not entirely disappeared 
 and periphrasis, which is almost unknown before the Conquest 
 but begins to appear in the interval between that date and 
 the beginning of the thirteenth century, has now become the 
 regular method. 
 
 We have seen above that the use of a modal form in 
 the sense of a future, as we find it in the oldest remains of 
 the Teutonic languages has its analogy in nearly all the 
 groups of the Indo-European family ; also that the use of the 
 present indicative, based originally on a distinction in the 
 kind of action expressed by the verb, finds an analogy in 
 Slavic and was not so inapt in the beginning to express the 
 future meaning as it became later when the distinction between 
 perfective and non - perfective verbs had vanished from 
 consciousness. It seems to be appropriate here to ask in what 
 respect the new method of periphrasis is adapted to such 
 use, and all the more as this method has crowded out the 
 older ones and established itself in all the languages of the 
 Teutonic group. 
 
 The Teutonic languages possess a series of verbs, which 
 in later use form periphrases whose meaning corresponds in 
 general to that expressed by flexional forms called "moods" 
 in other languages. They may best be designated by the 
 
— 80 — 
 
 name ''mood - verbs". In modern grammar they are most 
 frequently called "auxiliary verbs". They are peculiar in 
 form ; one of them is an optative without a corresponding 
 indicative, the others are perfects without corresponding 
 presents ; all have the force of present indicatives, however. 
 That these verbs had the same peculiarities of form and use 
 in the parent-speech as in the later period of the separate 
 dialects is proved by their likeness in form and function in 
 the various languages. Several of them are suited, with such 
 modification of meaning as usage naturally brings, to express 
 the future, and we have seen that they are so used in the 
 oldest Teutonic remains, when some special circumstance made 
 it necessary to vary from the normal method. Tlie following 
 may be cited as suited to the purpose named. ^ 
 ik skal, "I am obliged", "I ought to . ." 
 ik wiljaUj "I should desire", "I wish to . ." 
 ik mag, "I have power", "I am able to . ." 
 ik marij ^'l intend", "I am about to . ." 
 ik mot, "I have opportunity", "I am in condition to . ." 
 ik kann, "I have knowledge", "I know how to . ." 
 ik \iarf, "I have need", "I need to . ." 
 ik aih, "I possess", "I have to . ." 
 It should be added that these praeteritive verbs are not 
 the only ones that are fitted by their meaning to become 
 auxiliaries and express the future meaning. A number of 
 other verbs, which are complete in their flexion, are equally 
 well suited to such a use, in particular tver\ian, "to become", 
 hahan, "to have", also any verb meaning "begin", "intend", 
 "plan", etc. The peculiarity of form in will and the prae- 
 teritives has nothing to do with their fitness for the use we 
 are treating here. Modern English has a variety of ex- 
 I^ressions that express a future meaning as truly as any 
 regularly formed tense can do ; e. g. "it is going to rain", 
 "he is to sail soon", etc. The fact that the verbs mentioned 
 
 ' The Gothic form is given as probably the nearest to the primitive 
 Teutonic. 
 
/ 
 
 — 31 — 
 
 above have come into use rather than others is probably 
 due to the fact that the periphrases formed with them are 
 shorter than the others, since they are all monosyllabic and 
 are followed by an infinitive without the preposition to. 
 
 The adaptation of these verbs, as in fact of other verbs, 
 is the result of their meaning. All are verbs of incomplete 
 action, i. e. to make complete sense they require the addition 
 as object of a noun naming the action, and in use they re- 
 gularly take an infinitive. It is plain that in such expressions as 
 "I wish to do . .", "I ought to do . .", "I have the power to do . ." 
 "I have the opportunity to do . .", and the like, the doing lies 
 in the future, not in the present or past, and so phrases of 
 this kind, used as a substitute for the lacking future, would 
 naturally sink to a mere future in meaning, the verb gra- 
 dually losing its significance and becoming a mere tense-sign. 
 At first, however, these combinations must have been what 
 we may for convenience call "mixed futures", i. e. formulas 
 that expressed the future idea but' not this idea only; there 
 was also suggested the notion of power, will, obligation, or 
 whatever other meaning originally inhered in the verb em- 
 ployed. In this mixed meaning these formulas are entirely 
 analogous to the flexional futures of other languages, which 
 had a modal force in the beginning but gradually lost it; a 
 development exactly like that of the Teutonic future. 
 
 Out of the large number of verbs in a language tliat 
 are suited by their meaning to use in a mixed future we 
 may naturally expect those to be chosen that are least often 
 used in other senses or in other constructions, and that fur- 
 nish the shortest and most pithy expressions, and this seems 
 to be the reason that these "mood-verbs" have been chosen 
 for this function. Gothic, as was noted above, uses hahan 
 three times and du-gmnan twice, and Modern German uses 
 werden, but with these exceptions the use of verbs to form 
 a phrasal future is limited, in general, to slwll, ivUlj motj and 
 num. A few cases of mag T)ccur, in which the future meaning 
 is suitable to the connection , and a careful investigation 
 would probably succeed in detecting occasional cases of the 
 
— 32 — 
 
 use of the others in a like sense, but it is not easy to fix 
 the exact force of such phrases in the older works. In the 
 modern languages shall and will have nearly driven out all 
 the others, though it is not easy to see what special fitness 
 for this use they have more than the rest. 
 
 In English the four verbs given above are in use to 
 form a future, but mot and mrni are so rare in comparison 
 with shall and will that it will not be necessary to treat 
 them very fully. Their use seems to have been dialectic, 
 and even in those works in which they are found they bold 
 only a subordinate place, the other two verbs being far more 
 frequent. We shall therefore concern ourselves in tracing 
 the growth of the periphrastic English future chiefly with 
 shall and tvill, considering the others only incidentally as 
 they occur. 
 
 At the very outset of our study of the development of 
 meaning of these forms of expression we meet with a dif- 
 ficulty that we might naturally expect, the decision of the 
 question whether in any particular occurence of the formula 
 we have a true future or not. In the course of the change 
 from the original meaning of slmll or ivill to a mere tense- 
 sign we must expect to find cases that stand on the line 
 between the old and the new, sentences that to one hearer 
 or reader will suggest something of the old meaning of the 
 auxiliary while to another only the future meaning appears. 
 In fact such is still the case. The exact and forcible use 
 of these two words is one of the niceties of English style, 
 a nicety that not only furnishes a peculiar difficulty to 
 foreigners but is also a stumbling-block to the great major- 
 ity of those whose mother-tongue is English. It must be 
 understood therefore that the statistics given are not abso- 
 lutely accurate, but are the result of judgment in many cases, 
 based on opinion and not on unquestioned fact. Even in 
 cases of translation it cannot be claimed that the numbers 
 given are correct beyond the postSibility of improvement, for 
 cases of intentional variation by the translator are not im- 
 possible. It is assumed, however, that the judgments of 
 
— 33 — 
 
 different persons would not vary enough to affect the truth 
 of the conclusions drawn from the figures given. 
 
 A special phase of this difficulty lies in the fact that 
 these auxiliaries not only vary in their force but seem at 
 times to be without special force, the phrase formed with 
 them having the same meaning that the simple verb would 
 express. Attention has been called above to this peculiarity 
 in Old Saxon and the same occurs in English also. In the 
 introduction to his edition of the Cura Pastoralis, Sweet 
 points out instances of it in Alfred's English and cites one 
 or two instances from Old Norse and Middle High German. 
 This peculiarity of use is still found in will^ which is often 
 used where there is no notion of either wish or futuritj^, 
 and where the phrase formed , if it differs at all from the 
 simple present, only does so in suggesting a customary or 
 frequentative action. Such cases are therefore not classified 
 as futures but as denoting repeated or customary action. It 
 must be admitted that even this idea is often absent, or nearly 
 so, and that it would be more nearly correct to say that 
 the auxiliary does not express any meaning at all. 
 
 We now take up the use of the auxiliaries to form a 
 future as we find it in the period we are considering. 
 
 In the "Owl and Nightingale" the future is regularly 
 formed by means of shall, which occurs 61 times, in 46 of 
 which the future meaning best suits the connection. In the 
 15 other cases there is an implication more or less strong 
 of the idea of duty or necessity, the original meaning of 
 the verb. This idea is softened down to that of destiny in 
 some cases, the intermediate step between the older meaning 
 and that of simple futurity. 
 
 Will is found 39 times but in none of the cases is it 
 certain that a merely future meaning is intended. In all 
 the idea of wish is more or less prominent;^ four or five 
 
 1 e. g. "hwat wulle ye . . . pais tobreke" (v. 1730); "for Pat hi 
 nelleb to feor go" (v. 653), etc. 
 
 3 
 
— 34 — 
 
 contain a promise or a threat ^ and in one or two the notion 
 of a customary action is seen. 
 
 The older use of the present as a future is found four 
 times and the phrase "is to come" occurs once. There are 
 also 11 cases of mote, a part of which seem to be simple 
 futures, though none of them are necessarily so and most 
 of them keep the older meaning. 
 
 In the "Life of St. Katherine" we find shall 20 times in 
 a future and 10 times with more or less of its original meaning. 
 Will is found 23 times, but only once in a future, viz. in 
 verse 96, "what me wole bifalle." All the others are either 
 promises or expressions of a wish. 
 
 In "Genesis and Exodus" the usage is the same. The 
 first 408 verses, as printed in Wiilker's Altenglisches Lese- 
 buch contain 25 cases of shall^ all futures, and 4 cases of 
 will, three 2 of which express a resolve, while the fourth^ 
 has no reference to time, but gives the iterative idea spoken 
 of above. It is coupled with a simple present and both 
 verbs contain general statements. 
 
 In Layamon's Brut, the narrative form gives but little 
 occasion for the use of a future, except in speeches, and as these 
 consist to a great extent of threats and promises, we need not 
 be surprised to find a large number of cases of will. In the first 
 volume of Madden's edition (over 10000 verses) will is found 
 192 times, but almost always to express a threat, a promise or a 
 resolve, or with its original force of wish or consent. There are 
 three ^ cases, however, where the meaning is that of a simple 
 future, and four'^ others in which it seems to be so, though 
 it is possible to trace also the notion of wish in them. 
 
 Sliall is found 114 times, and regularly expresses a 
 
 ^ e. g. "ich wille beoii of Te awreke" (v. 262); "nulle ich wij* I>e 
 plaidi" (v. 1639), etc. 
 
 2 Viz. 277, 279, 336. 
 
 3 Viz. 191. 
 
 4 Viz. hit wule |>e suggen minne gult. (8354.); to whan bis tocne 
 wule ten. (9134.); beos boc him wulle suggen. (10110). 
 
 '^ 969, 982, 3056, 3496. 
 
-- 35 — 
 
 future , or in some cases a command. Only a single ^ case 
 of the original idea of duty or obligation has been noted. 
 
 In verse 3841, "^yef ich mot libben" seems to mean 
 "if I live/' and may therefore be called a future. There 
 are also a few cases ^ of the use of the present with future 
 meaning, but the simple future is generally formed by the 
 use of sliall. 
 
 The '-Orrmulum" shows the same preponderance of shall 
 that has been noticed in the other works. Occasional instances 
 of its older meaning occur, e. g. in the Dedication, 73, "Witt 
 shulenn tredenn unnderrfot", the meaning is "we ought to 
 tread under foot", a meaning that would now be expressed 
 by the preterit, "should". But in the large majority of cases 
 shall has sunk to a mere tense -sign and the phrase formed 
 with it is a simple future, or if any other idea than that of 
 futurity is implied, it is one that might be expressed by the 
 future in Latin, e. g. a command. Such cases are quite 
 natural in an ecclesiastic like the author, trained in the Latin 
 tongue and regarding it as his model. 
 
 The case is different with tvUl Of 131 cases noted 82 
 clearly express the original idea of wish or willingness. The 
 difference in the use of the two words is well illustrated in 
 the Dedication, 95—98; 
 
 & whase wilenn shall l)iss boc 
 
 Efft o^err sil)e writenn, 
 Himm bidde ice |)att het wi-ite rihht 
 
 Swa summ |)iss boc himm t8echel)l>. 
 
 Here ivilenn has the meaning of the Latin "velle", but shall 
 is a mere sign of the future ; "whase wilenn shall", translated 
 into Latin, would be "^w? volef" or \ui voluerif. The case is 
 the same in at least two thirds of the instances where will 
 occurs with an infinitive, making the formula now called a 
 "future tense", and in the other cases it is by no means 
 certain that we can properly regard the meaning as future. 
 
 1 V. 5312. 
 
 • e. g. 716, 744, 978, 4556, etc. 
 
— 36 — 
 
 Nearly all of them occur in a formula which with slight 
 variation is often prefixed to the interpretation of a passage 
 or to the explanation of the mystical significance of events 
 or circumstances, e. g. Homilies 251 — 254; 
 
 & her ice wile shsewenn yuw 
 
 Off tise twa Goddspelless 
 Hu mikell god teyy Igerenn yuw 
 
 Off yure sawless nede. 
 
 This phrase occurs at least thirty times with will and 
 only once with shall In some cases it is so connected with 
 the thought that the meaning "I wish to show you" is quite 
 appropriate and it seems possible that this is the meaning 
 intended by the writer in the other cases also. It is to be 
 noticed, moreover, that the use of tvill in the first person, 
 when a promise is expressed, as here, and not a mere de- 
 claration of what is to be, is entirely in accordance with the 
 present usage of the language. It has been mentioned above, 
 and is found in all periods, as far back as JElfred's time. 
 
 Especially instructive is the use of shall and will with 
 subjects that denote inanimate things. It is plain that the 
 use off tvill in such cases would not be suitable as long as 
 it keeps its original force, for volition can only be predicated 
 of beings that possess life and mind. We find ten such cases 
 in the "Owl and Nightingale", all with shall. In the Orrmulum 
 shall occurs 42 times with an inanimate subject, will 4 times, 
 and always in the phrase "whatt itt wile uss tacnenn". But 
 this phrase is not a future and means, apparently, no more 
 than the simple present, "what it betokens", "what significance 
 it has for us". We also find shall used in the same way; 
 e. g. Homilies 17585 and 17609, 
 
 & forrt>i shall l^e werelld her 
 Bitacnenn mannkinn ane, 
 i, e. "the word world here means mankind". This use of shall 
 and will has been spoken of above. It is apparently a correct 
 statement of Orrmin's usage that tvill is never used to make 
 a simple future. We find the present a few times with future 
 
— 37 — 
 
 meaning and there are one or two cases of munc that may 
 be so regarded, but the regular usage throughout the whole 
 work is that of shall. 
 
 The works cited may be taken as representing the usage at 
 the beginning of the thirteenth century. It is also the usage of 
 Middle High German and of the Low German version of the 
 Psalms, as stated above. So too in the modern Teutonic 
 languages, taken as a whole, sJiall is the predominant future- 
 sign. Modern German uses iverden and Icelandic mun, but 
 shall is found in all the others, being used along with ivill 
 in English and Norwegian and exclusively in Dutch, Danish 
 and Swedish. 
 
 The reason of the predominance of shall lies, no doubt 
 in its meaning, which fits it for use as a future-sign better 
 than any of the other verbs mentioned above as suited to 
 such use. The idea of obligation or necessity includes that 
 of the necessity of fate or destiny, an idea that plays so 
 prominent a part in Teutonic Mythology, and the notion of 
 a destined event is scarcely to be distinguished from that of 
 a future event. It is not surprising, then, that the Teutonic 
 writers and speakers, when trying to avoid the ambiguity 
 caused by the use of the present, by means of phrases that 
 referred to the future but with the additional iSea of 
 knowledge, power, will, etc., chose for the simple future the 
 word that least implied any other idea. And the natural 
 result of this use was the loss of the inherent meaning of 
 the verb until it suggested at length only the notion of time. 
 
 This use of sliall continued for more than a century. In 
 King Horn (shortly before 1300) will occurs but twice, in 
 both cases to express a promise, while shall is found 72 times. 
 So too in Robert of Gloucester's Chronicle. The 490 verses 
 in Wiilker's Altenglisches Lesebuch show five cases of shall, 
 and five of will, but the latter are all promises. Moreover 
 in the introduction of sliall, there seems to be no dialectic 
 difference ; contemporaneous works, though arising in different 
 neighborhoods and representing different dialects, show no 
 appreciable difference in the expression of the future. 
 
— 38 — 
 
 The English language, however, did not remain content 
 with the use of shall and the next step is to trace the use 
 of will from its beginning until the establishment of the 
 present usage. We have seen that will was already in use 
 at the beginning of the Middle English period to form what 
 might perhaps be called a "mixed future", i. e. one in which 
 something of the meaning of the verb was still apparent. 
 Such must have been the character of the future formed with 
 shall at an earlier period, as well as with all the other 
 auxiliaries. But though shall had already lost its force at 
 this date to such a degree as to serve at need as a mere 
 sign of the future, ivill did not do so until much later. 
 
 In the York Plays, the composition of which falls early 
 in the fourteenth century, we find that slmll still predominates, 
 being three or four times as frequent as will. In both words 
 the original meaning is often plain, ^ especially in will^ as is 
 shown not only by the thought of the passages in which it 
 stands but in a special way by the choice of the auxiliary 
 with inanimate subjects. In the third person shall is found 
 nearly twelve times as often as will, though when the sub- 
 ject denotes an animate being it occurs only about two and 
 a half times as often. We find here therefore the beginning 
 of the use of ivUl to make a pure future, since with a inanimate 
 subject the idea of willingness or wishing is no longer present 
 and the auxiliary is a mere tense-sign. 
 
 The same is true of the Townley Mysteries, in which the 
 use of these verbs does not differ much from that of the 
 York Plays. In both moreover we find mon used occasionally. 
 It is clear that this is due to Norse influence, since we find 
 
 ^ A good instance of this is found in No. XXVIII, in the prayer of 
 Christ in the garden before his betrayal, 
 
 And Fadir, I shall dede taste, 
 I will it nogt deiFende. 
 Both these sentences are in the future but with an additional impli- 
 cation of duty or destiny in the first and of willingness or consent in 
 the second. It may not be out of place to point out that the flexional 
 future of Latin is incapable of so appropriate a variation of the thought. 
 
~ 39 — 
 
 it only in works that come from that part of the country 
 that was most subject to Danish influence. We may treat it 
 therefore as a foreign idiom that had only a temporary use 
 and did not influence the development of the language. 
 
 The case is much the same in the English poems of 
 MS. Harl. 2253. In this miscellany of writings, lyi'ic, didactic, 
 narrative and dramatic, many of which are clearly the pro- 
 duction of a plebeian muse, the cases of shall outnumber 
 those of uill in the ratio of three to one, but with an inani- 
 mate subject the ratio is eight to one. 
 
 Especially important for this period is the translation 
 of the Bible made by John Wyclif and his co-laborers, since 
 a comparison of the translation with the original leaves us 
 less in doubt what meaning the writer intends to convey. 
 In the Wyclif-Purvey translation of the Gospel of Matthew 
 the Latin future of the Vulgate is rendered by shall 322 
 times and by will only twice. ^ The two cases of ivill are 
 Mat. Vil, 9, 10; '^What man of you is, that if his sone axe 
 hym breed, whether he w^ole take hym a stoon ? Or if he axe 
 fishe, wiiether he wole take hym an edder ?" Here ''whether 
 he wole take" translates the Vulgate numquid porriget. 
 Another version, a few years older, has no case of will, but 
 uses shall in these passages, as in the others, and it is plain 
 that shall was considered at this time the proper English 
 equivalent of the Latin future. 
 
 The Wyclif translation, then, stands on the same footing 
 as the language of nearly two centuries before, while the 
 Harleian MS., though about three fourths of a century older, 
 shows undoubted instances of a pure future formed with 
 will. The explanation is probably to be looked for in the 
 
 ' These figures do not include cases of the future used imperatively, 
 all of which, 17 in number are rendered by shall. Shall is also used 
 nine times to translate the future active participle with sum and once 
 to translate tradendus est. The future is rendered five times by the 
 present and once by the past. The latter is due no doubt to the con- 
 fusion of -abit and -acit in the mediaeval MSS, ; perhaps the others 
 depend on variations in the Latin text. 
 
- 40 — 
 
 diiferent character of the works, unless it reflects local usage, 
 and no marked difference in the way of expressing the future 
 is noticeable in the different dialects before Wyclifs time. 
 It need not be thought unlikely, that at this period also the 
 common speech was farther along on the way to the later usage 
 than the style of literary men, and it may be that this stage of 
 development is shown to some extent in the popular satires 
 aad love songs of the collection named. This finds confir- 
 mation in the fact that in the political and secular poems 
 the cases of the use of ivill approach in number those of 
 shall, being 19 to 23 ; but in the religious lyrics and in the 
 more formal didactic and narrative writings ivill is found 
 less than one fourth as often as shall. These cases are not 
 all simple futures, of course, but the increase of frequency in 
 the use of ivill in what we have called above "mixed" futures 
 was no doubt the real change in the idiom, to which the 
 use of will in making a pure future was only a corollary, 
 or perhaps we may better say, of which it was the natural 
 consequence. 
 
 In the latter part of this period we find the usage much 
 the same, only modified by the increasing frequency of will. 
 We have spoken of Wyclifs translation of Matthew already, 
 because it shows no change from the older usage of a century 
 and a half earlier. That this was not due to the conventional 
 retention of an older style, as we often find it in Scripture 
 versions, is shown by the fact that his other writings have 
 the same peculiarity. In his ^'Sunday Gospels", the ten ex- 
 positions for the first ten Sundays after Trinity show 59 
 futures formed with shall and only two with will^ 
 
 In the expression of the future we find the same usage 
 in Barbour's Bruce as in Wyclif. The first ten books contain 
 119 instances of the use of shall and only three of will., ex- 
 cept in promises etc. where will is found 13 times. This 
 preponderance of shall is perhaps dialectic; other works of 
 
 * Viz. in No. 2. For there slial nothing faile ^at seintis wolen desire, 
 and No. 9. Crist wole axe of bis streite rekenynge. 
 
__ 41 — 
 
 the same period show a greater increase in the use of will 
 and it is possible that this increase begins in the South. 
 Such a hypothesis would explain the futures with will in the 
 Harleian Ms. and in other works, and does not necessarily 
 exclude the supposition that the use of will first occurs in 
 the folkspeech, since this theory is based on a comparison 
 of the more popular songs of the MS. with the more formal 
 works in it. 
 
 In the "Vision of William concerning Piers the Plowman" 
 the occurrences are^ 
 
 1st. pers, shall, 23 times; wUl, 30 times 
 2d. „ „ 22 „ ; „ 6 „ 
 
 oQ. „ „ oD „ ; „ 11 „ 
 
 The classification according to the person of the subject 
 shows that the modern distinction had not yet begun, and the 
 ratio of usage is about the same as in the Harleian MS., 
 but it is noticeable that the use of will in the first person 
 is on the increase. The same increase is still more marked 
 in Chaucer's Prologue and Knight's Tale. Here the occur- 
 rences are 
 
 1st. person, shall, . . 19 will, . . 29 
 2nd. , , . . 18 „ . . 3 
 
 3rd. , , . . 37 „ . . 6. 
 
 Many of these cases are not pure futures; sluill has its proper 
 force in many cases and will very much oftener. In fact, if 
 it were thought necessary, in order to establish a theory, to 
 find the idea of volition in as many cases as possible, there 
 are but few instances of the use of wiU where this notion 
 could not be introduced without doing violence to the thought, 
 and it is quite conceivable that a contemporary reader or 
 hearer may have felt the force of the verb in many cases 
 where it is not apparent to us, trained as we are to a 
 difterent conception of the force of the formula. 
 
 1 We begin the classification according to the person of the subject 
 at this point. Until about 1400 the use of uill is the exception, except 
 in its proper sense and in the cases mentioned, and there is no need, 
 therefore, of such a classification before this date. 
 
— 42 — 
 
 We also find in Cliaucer a few cases of mote, and in the 
 Vision there are five cases of the use of the present with a 
 future meaning, four of which are passives formed with wmih, 
 e. g. "worth chained", [shall he ehained]. These expressions 
 are noteworthy since they seem to be formulas that have 
 survived from the old distinction between perfective and non- 
 perfective verbs. It was mentioned above that the Gothic 
 makes use of wairlmn, the perfective verb, to render the 
 Greek future in the majority of cases, also that a large 
 number of the instances of the present in a future sense in 
 Old Norse contain the same verb and that eight of the nine 
 cases of the present in the Old Low German version of the 
 Psalms are passives formed with iverthan, like those in the 
 Vision of Piers Plowman. It does not seem bold to regard 
 these cases as surviving formulas dating from an older time 
 when the use of the present as a future was based on the 
 perfective meaning of the verb. 
 
 During this period, then, the future was formed by the 
 use of shall with an infinitive, but will begins to come into 
 use at the very beginning and steadily increases in frequency. 
 The cases in which will forms a pure future are limited 
 however to the latter part of the period and are few absolutely 
 and very few relatively to those of shall. The increase is 
 most marked in works of a popular stamp and it is a fair 
 inference that the more frequent use of will prevailed in 
 ordinary speech at an earlier date than that at which we 
 find it in books, and there are indications that it began in 
 the South earlier than in the North. 
 
 These conclusions, drawn from the examination of a few 
 representative works, are confirmed by the reading of the 
 extracts in Morris and Skeat's "Specimens of Early English", 
 and Skeat's "Specimens of English Literature". These selections 
 are arranged chronologically and extend from A. D. 1150 to 
 A. D. 1579. Most of them are too short to furnish a basis 
 for a statement of the usage of any particular writer, but 
 they may be used to illustrate the usage of the period. Taken 
 together as representative of the literature of these four 
 
— 43 - 
 
 centuries, they funiisli nothing contradictory to the conclusions 
 drawn from the works cited above. Only in one or two, 
 moreover, is anything additional discoverable; e. g. a few 
 cases of maij that seem to have a future sense and instances 
 of iviir^ making a passive future. Especially instructive 
 are variations of a common formula of contrast in time, as 
 follows ; 
 
 l)us hit haS ibi and is and wur^ oS domesdei. 
 
 Homily, before 1150. 
 
 al l)at is ant wes ant eauer schal iwurden, 
 
 St. Juliana, about 1210. 
 
 as |ni were ant art & schalt beon in eche, 
 
 SaAvles Warde, about 1210. 
 
 neuer nes ne neuermore ue w^irS iboren [>in iliche. 
 
 On god Ureison, about 1210. 
 
 In these extracts, all of which fall at the beginning of 
 the period we are considering, there is a contrast of future 
 time with present or past and the various ways of expressing 
 it are instructive. There seems to have been something of 
 uncertainty still as to the best way of bringing out the future 
 idea and a kind of groping around to find some satisfactory 
 expression. In another case we find "may beon and schal'* 
 meaning no more than ''shall be", and other cases of may 
 are found ; in one of them the later text has nioiven and the 
 earlier sulen. But shall is the predominant word throughout, 
 ivill scarcely appearing at all except in its proper sense until 
 toward the end of the period. 
 
 The first use of tcillj moreover, that can properly be said 
 to make a future, is in promises and threats, and naturally 
 confined to the first person, and it was here that it first 
 began to take the place of sluiU. This usage increases in 
 frequency toward the end of the period but does not become 
 the rule of the language until later. 
 
— 44 — 
 
 IV. 
 
 SINCE THE INTRODUCTION OF PRINTING. 
 
 There follows nearly a century of comparative barrenness 
 in literature, succeeded by the grand outburst of the Eliza- 
 bethan period. The invention of printing from movable types, 
 the discovery of the Western continent and the Protestant 
 Reformation may be named among the causes that led to this 
 intellectual awakening and to such a development of literature 
 that a single generation produced in England more great 
 writers than usually fall to the lot of a century. These events, 
 however, are not of the kind that influence language peculiarly, 
 and we need not look for any change like that which marks 
 the transition from the first to the second period. All that 
 we find that bears on the question we are investigating is 
 an increase in the use of will, which had now in many cases 
 lost enough of its original force to serve, like shall, as a mere 
 tense-sign. A statement of the usage of a few leading writers 
 will illustrate the change. 
 
 It has been pointed out above that in tlie Wyclif version 
 of Matthew the Latin future is regularly rendered by sliall. 
 This version is so scrupulous a rendering of the original, even 
 transferring Latin idioms and peculiarities,^ that one can 
 hardly regard it as a specimen of idiomatic English, as it 
 was spoken at the time. If Wyclif's use of sMU be not 
 dialectic, as was suggested above, it is conceivable that it 
 is due to the effort to set like expressions in English over 
 against like expressions in Latin, and that the use of will is 
 avoided because the word still kept its original meaning too 
 fully to make it an equivalent of the Latin future. 
 
 Nearly a century and a half later Tyndale printed his 
 translation of the New Testament. Unlike the former trans- 
 lators he rendered the original into clear and idiomatic 
 
 * Illustrations may bo found on almost every page ; e. g. "a thousand 
 pacis" (= mille passuum); "be seen" (= videri) instead of "appear" or 
 "seem"; the plural "derknessis" (= tenebrae), etc. 
 
— 45 ~ 
 
 English, and the closer resemblance of English to Greek in 
 the structure of the sentence gave him an advantage over 
 his predecessors who had followed Latin versions. His language 
 may fairl}^ be regarded as a good specimen of the usage ot 
 his time. 
 
 Now an examination of Tyndale's work shows that but 
 little change had taken place in the language in its way of 
 expressing the future. Of the cases of shall in Wyclif, about 
 330 in number, • 266 are also expressed by shall in Tyndale. 
 Of the remainder, a few differ from a variation ot the con- 
 struction and a few are due, apparently, to different readings 
 in the original works, and there are 47 cases ^ of will ^ of 
 which 34 are in the first person and express a promise, a 
 threat or a resolve, and 13 are in the third person and are 
 simple futures. It should be noted, however, that in all of 
 the latter the subject denotes an animate being and the act 
 is a voluntary one. It cannot be asserted therefore that will 
 had become simply a tense-sign in Tyndale's time. He writes 
 for example, in the Parable of the vineyard, (XXI, 40, 41), 
 "When the lorde of the vyneyarde commeth, what wyll he 
 do with those husbandmen? They sayde vnto hym, He will 
 evyll destroye those evyll persons and wyll lett out hys 
 vyneyarde vnto other husbandmen". But a few verses farther 
 on we find. "And whosoever shall fall on thys stone, shalbe 
 alto broken ; and whomsoever thys stone shall fall oppon, he 
 shall grynde him to powder". This and similar cases seem to 
 show that tvUl could be used to make a future tense, when 
 the voluntary act of an animate being was expressed, or, 
 in other words, when the original meaning of will was not 
 unsuitable to the case, but that it was not considered ap- 
 propriate to the expression of the future idea, when the action 
 
 ^ Cases are here included in which the auxiliary is omitted in 
 one or both translations, having been expressed just before ; e. g. XII, 19; 
 •'He shal nat stryve, ne crye" (W.); "He shall not stryve, he shall not 
 crye" (T.). So XIX, 5; XXIH, 34; XVII, 11; VII, 19; etc. 
 
 - Including four cases where tvill is omitted, having been previously 
 expressed. 
 
— 46 — 
 
 was involuntary and the subject inanimate. Even in the 
 former case we have only the beginning of the usage ; in the 
 great majority of cases the older use of shall prevails. 
 In Spenser's Faerie Queen, Book I, we find 
 
 1st. pers. shall^ 7 times; will, 5 times, 
 '^d. „ „ D ^ ; „ o „ 
 
 Here is a decided gain in the proportionate employment of 
 will, and in many cases the meaning is that of a pure future. 
 It is to be noted, however, that the modern distinction accord- 
 ing to person does not yet show itself but that two thirds 
 of the cases in the second person and three fourths of those 
 in the third still make use of shall. A part of this is due, no 
 doubt, to Spenser's intentional cultivation of an older style 
 of expression. The same is true of the Bible of 1611. This 
 version was only a revision of an older one and the older 
 language was carefully retained wherever it was idiomatic 
 and intelligible. In the Gospels of Matthew and Mark the 
 Greek future is translated by 
 
 Shall, 1st. person, . 10 times ; will, 1st. pers. 36 times. 
 „ 2nd. „ .63 „ ; „ 2nd. „ 6 „ 
 , 3rd. „ . 267 „ ; , 3rd. „ 35 , 
 
 Here the predominance of shall over ivill in the second 
 and third persons is still more marked than in Spenser, being 
 about 88 per cent of the whole, while will has increased in the 
 first person. If now we compare these figures with those 
 of a contemporary work not under the influence of antiquarian 
 taste or following an older model, we shall see that will had 
 gained much more than is apparent from the works cited. 
 For this purpose we take Shakspere's Tempest. The occur- 
 rences are 
 
 [a]. In simple futures, 
 
 1st. pers., sliall, . . 16 times, ivill, . . 3 times. 
 2nd. „ „ . . 5 „ , . . 2 , 
 3rd. , , . . 15 16 „ 
 
— 47 - 
 
 [b]. In promises and threats, 
 
 1st. pers., shall . . times, ivill . . 72 times. 
 2nd. , „ . . 12 „ „ . . „ 
 3rd. „ ... 12 „ ; . . „ 
 
 Further, shall occurs once in a conditional clause and 
 seven times in expressions of determination ; will occurs four- 
 teen times in the expression of a wish and three times to 
 express a customary action. These uses, as well as that in 
 promises and threats are in entire accordance with modern 
 idiom , and the usage in simple futures shows a tendency 
 toward it, though shall still holds its place in the second and 
 third persons, h^m^ used about as often as tvill. 
 
 The same tendency toward the present usage continued 
 for about half a century more by which time it may be considered 
 as established. In Milton's Areopagitica, pure futures very 
 seldom, if ever violate the modern rule of shall in the first 
 person and will in the others, though his use of these verbs 
 is so nice and exact that at first sight one would be incli- 
 ned to say that he still retains shall in the third person in 
 a large number of cases. These cases, about thirty in number, 
 when carefully studied, will be seen to retain something of 
 the older meaning of shall, and a good test of this fact is 
 the attempt to substitute will in them. Such a test shows 
 at once how the force of the expression is lost and proves 
 that there is more than a mere future meaning in them. We 
 may fjiirly assert that the modern usage is practically esta- 
 blished at the middle of the seventeenth century. 
 
 It cannot be asserted that the rule of modern usage as 
 given above permits no exceptions. In one class of con- 
 structions sh<^ll keeps its place still, viz., in modifying clauses 
 of condition and temporal clauses implying a condition. 
 Here shall is used even though the clause expresses only a 
 temporal meaning. The present indicative or subjunctive 
 is more common in such, but if a periphrasis is used it must 
 be formed with shall. It is quite correct to say "if he comes", 
 or "if he come", or "if he shall come", though the latter is 
 
— 48 — 
 
 less common, but "if he will come" can only be used when 
 the idea of willingness or consent is implied. In other words 
 "if he will come" means si vult venire or si volet venire; the 
 other forms mean si vcniet. 
 
 There is also another use of shall in the second and 
 third persons that cannot be said to be strictly idiomatic 
 according to modern every day usage but which may be met 
 with still in writings of a certain class, and which was much 
 more frequent formerly. The great predominance of shall in 
 the Authorized Version of the Scriptures has led grammarians 
 unfamiliar with the fact that the use of wUl is comparatively 
 late to speak of the ^^ shall of prophecy" with the assumption 
 that there is something in the meaning of the word suited 
 to prophetic utterances, a prophecy being in a certain sense 
 a declaration of destiny or of something established and 
 fore-ordained by a higher power than that of man. It is 
 apparent, however, that the use of slmll in such utterances 
 in the Scriptures is only the retention of the older idiom of 
 the earlier versions, and in other writings in many cases only 
 an imitation, more or less unconscious, of familiar scripture 
 phraseology. This is plain from the more frequent use of 
 shall in works that find their inspiration in the study of the 
 Scriptures and in writers whose style has been affected by 
 the study of them. Familiarity with the use of shall in the 
 prophecies has naturally led to the feeling that the word is 
 suited to the declaration in a formal manner of that which 
 in the view of the writer is sure to happen, especially on 
 such subjects as are dealt with in the prophecies. A good 
 illustration of this feeling is shown in Bunyan's Pilgrim's 
 Progress, Part I. The date of this book is a little later 
 than Milton, and the usage of the two auxiliaries, except in 
 quotations from the Bible, is about that of to-day, aave for 
 a fairly large number of cases of shall in the third person, 
 an echo of the prophetic style. A good instance occurs near 
 the beginning; "There shall be no more crying, nor sorrow; 
 For he that is owner of the place, will wipe all tears from 
 our eyes". Here the author involuntarily uses sluill in the 
 
— 49 — 
 
 first clause in imitation of the familiar passage in Rev. XXI, 4, 
 but in the second clause, having substituted for the word 
 God a different expression, he is not conscious of the influence 
 of scriptural style and instead of shall which stands in the 
 Bible text, he naturally employs tvill which had now become 
 the usual word in simple futures. 
 
 The use of will in pure futures, as we have seen, is 
 comparatively modern and is in the nature of an encroachment 
 on the use of sJmll which was at first universal, as in Middle 
 High German and several of the modern Teutonic tongues. 
 In those uses in which either verb keeps something of its 
 older meaning will is far more frequent than sliall, the earlier 
 use of the latter as a mere tense-sign having caused it to 
 lose its independent force to a far greater degree than will 
 has done. 
 
 The question naturally arises whether any adequate 
 reason for the substitution of wUl for sJuill can be offered. 
 It is often asserted that the English usage is based on a 
 feeling of courtesy; that it is quite in accordance with the 
 law of politeness to say "I shall", (i. e. "I am obliged"), but 
 "you will", (i. e. "you are willing"). The convei-se of this 
 statement would come nearer the truth; the use of sMll in 
 the second or third person often expresses a threat or a 
 feeling of determination on the part of the speaker, and may 
 therefore be discourteous. But it is quite improbable that 
 any such instinctive feeling caused the usage in the first 
 place. When wia fip^t began to take the place of sMU it 
 was the first pei-son that showed the change, as can be seen 
 in the statistics given above. This is just what one would 
 expect a priori, one can assert Ms own willingness more 
 confidently than that of another; but the fact seems a con- 
 clusive argument against the theory that a feeling of courtesy 
 was the original cause of the idiom. 
 
 The fii-st use of ivUl, moreover, without its full original 
 meaning of volition was in promises and threats and therefore 
 in the first person. Its fitness for such a use is evident, for 
 
— 50 — 
 
 a threat or promiRe is the expression of the will of the 
 speaker, and speaker and subject are the same in the first 
 person. But in the second and third persons, where speaker 
 and subject are different, the old use of shall continued, since 
 will^ if used, would have expressed the will of the subject 
 and would therefore have conveyed no idea of a promise or 
 threat on the part of the speaker. There was thus developed 
 in the first person a difference of form corresponding to a 
 difference of meaning, shall expressing as before a simple 
 future and ivill a promise or a threat, while in the other 
 persons only shall could be used. The next step in the 
 evolution naturally soon followed. The difference between 
 a simple future and a promise or threat may be very little 
 or very much, and whether more or less is denoted by tone 
 and manner more than by words. The introduction of tvill 
 into the first person and its frequent use there with a 
 weakened meaning was followed by a weakening of its 
 meaning in the other two persons, where it was already in 
 common use, though with its older force. Both shall and 
 will were now in use to form the future and the next step 
 was the differentiation of form to express different meanings. 
 This had already been done in the first person, where the 
 signification of the two verbs fitted them for different uses; 
 in the second and third persons shall was suited by its 
 meaning to express a threat or a promise and will was not; 
 will was therefore unconsciously chosen for the other meaning, 
 that of simple futurity. This differentiation of form in the 
 second and third persons was therefore in a certain sense 
 mechanical, the distinction not being made on account of the 
 superior fitness of will to express futurity, but as a conse- 
 quence of the difference of form in the first person, where 
 it was based on difference of meaning. 
 
 This hypothesis is supported by the chronology of the 
 usage of the two words, which we give here roughly in 
 tabular form. The dates are, of course, only approximate, 
 and there is no hard and fast rule of usage; the different 
 writers, too, as one would expect in cases of unsettled usage. 
 
51 — 
 
 vary widely from one another and the question is still further 
 complicated by the frequent cases of the older meaning of 
 obligation or wish as well as by the habit of the language 
 of using periphrasis without special force beyond that of the 
 simple verb, but a general survey of the language for the 
 four or five centuries during which the modern usage was 
 developed gives the general results shown in the following 
 table. 
 
 Period 
 
 Simple Future 
 
 Threat or Promise 
 
 1150-1400 
 
 shall 
 
 shall (or tviliy 
 
 1400-1550 
 
 shall 
 
 1. will (or shall) 
 2. 3. shall 
 
 1550-1650 
 
 1. shall 
 2. 3. shall or will 
 
 1. will 
 2. 3. shall 
 
 1650 
 
 1. shall 
 2. 3. unU 
 
 1. will 
 2. 3. sJmll 
 
 Alongside of these uses we find also the constant use 
 of each verb in its original sense ; shall expressing obligation, 
 duty, necessity, destiny, etc., will, wish, consent, etc.; further 
 the idiomatic use of both, but especially will, as frequenta- 
 tives, expressing little more than the simple verb. 
 
 The conclusions reached in this investigation of English 
 usage are the following; 
 
 I. 
 
 The oldest English, i. e. before the Norman Conquest, 
 expressed the future tense regularly by the present indicative, 
 adding an adverbial expression, no doubt, when it was ne- 
 cessary to guard against ambiguity in time. 
 
 Instead of the present indicative the present optative is 
 occasionally found. This is a survival of an older usage, once 
 much oftener employed. 
 
 1 The bracketed words are rare. 
 
— 52 — 
 
 Periphrastic forms of expression are also found, but 
 rarely, and never to make a simple future ; something of the 
 meaning of the verb used for this purpose is always apparent 
 in them. It is quite possible, however, that in the ordinary 
 usage of the people a periphrasis with shall was already in 
 use though not found in the writings of the period; such a 
 hypothesis would explain its general use at the beginning of 
 the thirteenth century, when the folkspeech came to the surface 
 again after a century or more of repression. 
 
 n. 
 
 In the next following period, i. e. from the Conquest to 
 the introduction of printing, the future is regularly expressed 
 by a periphrasis with shall. 
 
 The present indicative also occurs at the beginning of 
 the period but soon gives way to shall. After the first century 
 and a half scarcely more cases occur than are allowed by the 
 modern rules of the language. 
 
 Periphrases with murij mote and ivill are also found, the 
 first two being dialectic and rare, the third being rare at 
 first but increasing in frequency toward the end of the period, 
 especially in the first person in the expression of a promise 
 or a threat. There are indications that this use of will began 
 in the South, and also that it was in use in works of a 
 popular character earlier than in more formal and polished 
 style. 
 
 III. 
 
 Between the second and third periods there is no break, 
 but the use of will increases rapidly , caused apparently by 
 its use in the first person in promises and threats, where it 
 had lost something of its original force. The substitution of 
 will for shall in pure futures in the second and third persons 
 begins in the latter part of the second period, but is of rare 
 occurrence; it increases in frequency from the beginning of 
 the third period and becomes the rule of the language as 
 early as the middle of the seventeenth century. 
 
— 53 — 
 
 The distinction between shall and will in Modern English 
 according to the person of the subject does not rest on any 
 special fitness of either verb to express futurity with one 
 person more than with another, but is the result of an 
 unconscious effort to distinguish between a simple declaration 
 of that which is to take place and a threat or promise, i. e. 
 a declaration of the speakers determination that something 
 shall take place. In these latter mixed futures the use of 
 shall and will is determined by the meaning of the verb, will 
 being suited to the first person only and shall being kept in 
 the other two from the older usage. 
 
VITA. 
 
 I was born March 5. 1845, in the Township of Gaines. 
 Genesee Co. Michigan, U. S. A., and my first instruction was 
 obtained in the district school of the neighborhood. The re- 
 moval of my parents to Ann Arbor in 1860 gave me the 
 opportunity of continuing my studies in the High School of 
 that place, wliere I was prepared for the Clavssical Course 
 of the University of Michigan, which I entered in 1863. Five 
 years later I graduated with the degree of A. B., my College 
 course having been interrupted by service in the Union Army 
 during the last year of the Civil War. 
 
 Since leaving college I have been continuously engaged 
 in teaching or in special study. The latter was the case in 
 1869 — 70, which I spent at the University of Michigan in post- 
 graduate work in Classics and Natural History ; in 1875 — 76, 
 when I was pursuing classical and philological studies at the 
 University of Leipzig, Germany; and since 1889, during which 
 time I have been at the same University engaged in the 
 study of Comparative Grammar, especially of the Teutonic 
 languages. 
 
 During my first year in Leipzig I heard lectures by 
 Curtius, Ritschl and Hiibschmann; during the last two years 
 and a half I have had the advantage of instruction hy Profs. 
 Wiilker, Brugmann, Leskien, Windisch, von Bahder, Lindner 
 and Scholvin, and by Drs. Mogk, Schirmer and Holz. 1 have 
 also been a member of Prof. Brugmann's "Sprachwissenschaft- 
 liche Gesellschaft" and of the "Konigliches deutsches Seminar" 
 conducted by Profs. Fr. Zarncke and von Bahder. To all 
 the instructors named I am under obligation for repeated 
 courtesies and kindness. 
 
 Francis Adclhert Blackburn. 
 Leipzig, Germany; Nov. 4. 1891. 
 
RETURN CIRCULATION DEPARTMENT 
 
 TO^-^ 202 Main Library 
 
 LOAN PERIOD 1 
 HOME USE 
 
 2 3 
 
 4 
 
 5 6 
 
 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 
 
 Renewals and Recharges may be made 4 days prior to the due date. 
 
 Books may be Renewed by calling 642-3405 
 
 DUE AS STAMPED BELOW 
 
 .^'nfc;|;3xoQ7 
 
 
 
 JUL '-' ' '" 
 
 I 
 
 
 r^ioniii^"^' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 UNIVERSITY OF CALIFORNIA, BERKELEY 
 FORMNO. DD6 BERKELEY, CA 94720 ^^ 
 
Gay lord Bros. 
 
 Makers 
 
 U.C. BERKELEY LIBRARIE 
 
 CDmn35it 
 
 S-boSZ^Y 
 
 3i. 
 
 UNIVERSITY OF CALIFORNIA LIBRARY