The development of Germanic analytical future tenses

Abstract

The stages that encompass the future tense development are singled out as discrete phenomena within the process of the Germanic language development. The Gothic verb system can serve as the background for the investigation of the tense transformations in question. The difficulties of tense examination in the Old Germanic languages were connected with some conceptions about the Indo-Iranian and Greek languages that used to dominate in the scientific circles for a long time. Those conceptions were based on Latin and Greek patterns and postulated the use of present, past and future tenses in all Indo-European languages. The above conceptions were ruined when the study of Tokharian and Hittite demonstrated the use of the present tense for the description of future actions. The idea of losing "the protolanguage inheritance" was proved wrong, and it was incorrect to transfer the complex tense system of Sanskrit, Greek, and Latin to other Proto-Indo-European languages. The examination of the tense differentiation in Gothic (as the main source of the Old Germanic language) demonstrates that the Gothic infinitive functioned as a no-particular-time unit, while personal verb forms were involved in performing tense functions. The Gothic present tense verbs represented present and future tenses and no-particular-time phenomena. Some periphrastic forms containing preterite-present verbs with the infinitive occurred sporadically. The periphrastic forms correlated with Greek and Latin patterns of the same future tense meaning. The periphrastic future forms in Gothic often contained some modal shades of meaning. The Gothic present tense functioned as a colony-forming archi-unit and a pluripotential (temporal) precursor. The periphrastic Gothic future forms are recognised as a monopotential (temporal) precursor with some modal meaning. The key research method used in the present article is the comparative historical method. The authors viewed it as the most reliable and appropriate for the study of tense forms.

Key words: colony-forming archi-unit, future tense, monopotential precursors, periphrastic forms.

Анотація

Publication analysis

A common conception dominated in classical linguistics for a long period of time. The main idea of this conception was that the Indo-European ancient language had a highly well-developed system of verb forms by the time of its separation into groups and subgroups. These forms are best preserved in the Indo-Iranian and Greek languages. In fact, the ProtoIndo-European verb system was reconstructed by transferring the forms, which were recognised mostly in Greek and Sanskrit, to the Proto-Indo-European. In particular, the tense paradigm was reconstructed by encompassing present, imperfect, aorist, perfect, pluperfect and future. The absence of certain reconstructed archetypes in the Slavic, Germanic and Baltic languages was attributed to the loss the "protolanguage inheritance" [18, p.125126]. Naturally enough, under these conditions when all the factors of the Old Indian (Sanskrit) and Ancient Greek languages were grossly exaggerated, all the deviations from Greek and Sanskrit norms in the structure and function of the verb units were treated as new formations. This idea was represented clearly and consistently in Brugmann and Delbruk's compendium [7]. It was also represented in the subsequent investigations of the Indo-European languages and was reflected as a steady tendency in the investigations of the following decades, for example Kronasser's book about Hittite [15]. The same scheme was involved for the elucidation and interpretation of the Germanic verb system genesis in the description of Comparative Grammar of the Germanic languages. The old system of the Germanic verb was reconstructed applying the abovementioned model. It was possible to find old aorist, imperfect, perfect participle in the field of Germanic verb forms.

The first comparative grammar of the Germanic languages written by W. Streitberg proclaimed that the Germanic languages "had lost" imperfect, pluperfect, sigmatic aorist, sigmatic future tense and preserved only two forms - present and perfect. The book reflected the position of Neogrammarian Schools. It emphasised that perfect had got preterite meaning [22, p. 280-281]. Later these theses with some variations were mentioned in Deiter and Bethge's compendium of the Germanic languages [8, p. 345-346], in Kluge's [13, p. 54] and Krahe's [14, p. 92-93] works. A. Meillet also supported the above theses, but partially [17, p. 91-93]. This tradition was stable in the history of some Indo-European languages. H. Brinkmann wrote about the "will to simplification". This "will" caused the transposition from the developed Indo-European verb system to the very simple system of the Germanic verb. In particular, the absence of imperfect, pluperfect and future in early Old German texts was explained by the assumption that Germanic tribes "did not have the sense of time (Zeitgefuhl)" [6, p. 47]. H. Brinkmann repeats Gabelenz Junior's idea about the absence of the sense of time in Old Germanic tribes as can be seen from the tense system of the Gothic verb. In those days the idea correlated with the traditional ideas about the Germanic verb. It was viewed as the result of gradual movement away and separation from the Proto-Indo-European patterns that were preserved in Greek and Sanskrit.

The process may be reconstructed within three stages. The first stage is associated with the Proto-Indo-European language which had a distinctively differentiated tense system. The second stage includes two substages. The Proto-Germanic language lost the inherited variety of Indo-European forms under the simultaneous influence of the "striving for economy (Streben zur Sparsamkeit)". The third stage was connected with the compensation of lost forms by means of periphrastic forms. If we take into consideration the presence or absence of the distinctive tense differentiation, the Old Germanic tribes demonstrated the process of degradation as compared with their Proto-Indo-European ancestors. This degradation was connected not only with "the loss of some forms". It was associated with the simplification and primitivisation of qualitative thinking. The qualitative thinking had been the characteristic feature of the higher degree of mankind's development. Only later and due to the collision of Greek-Roman culture with Germanic culture there appeared the necessity of forming the tense categories, which needed the periphrastic forms. The artificialness of this theory is evident, because the Germanic regress and degradation seem pointless and ungrounded. The revision of this conception had taken place before Tokharian and Hittite discovery. The involvement of these languages helped to reject the traditional linguistic schemes. H. Hirt wrote in his works that the Proto-Indo-European verb system had not been so complex and sophisticated as it had been in the Greek and Indo-Iranian languages. He was sure that the tense paradigm had only three constituents: present, aorist and perfect [11, p. 146]. These ideas influenced K. Karstein [12], H. Ammann [1, p. 336-340], H. Arntz [2, p. 441] and later they developed a new conception that proclaimed the fact that the Proto-Indo-European language did not have a well-developed verb system by the moment of its division into separate language groups. The formation of the Indo-Iranian verb system and the verb systems of other Indo-European languages were viewed as results of individual development [1, p. 341].

A. Meillet was sure that it was wrong to transfer the complex tense system of Sanskrit, Greek and Latin to other Indo-European languages. Many elements of that system were the secondary, marginal phenomena. A. Meillet stimulated considerable progress in investigating the Indo-European verbs. He introduced new methods and principles of linguistic research: fundamentals of prototypical linguistics, partly the ways of inner reconstruction, the rules of systematisation and relative chronology. All these methods were very effective for further linguistic analysis when applied to the selected facts. Completely new points of view on the archaic type of the Indo-European verb were formed between 1908 and 1930. The first date was connected with the publication of A. Meillet's work. It described the dialects of the common Indo-European language [17]. The second date was remarkable because of the introduction of Hittite and Tokharian records and clay tablets into the linguistic comparison. It caused the revision of old schemes of the Indo-European system. It is evident that the use of the Hittite and Tokharian samples for the explanation of the Indo-European model genesis was the marker that indicated the instability and destruction of the old scheme of the Indo-European verb. This scheme had been based on the classical tradition of the comparative linguistics. The present state of the Indo-European verb investigation was described by Chr. Stang in 1942: "Concerning the verb system, the IndoEuropean linguistics is now at a stage when scholars have to confess they know about Common Indo-European situation far less than they did a few decades before. The comparative simple picture of the Indo-European verb system, introduced by K. Brugmann, happened to have been ruined; but the new system was not introduced instead" [20, p. 2]. Evidently this difference may be explained by the specification of the initial point when the separation of discrete language groups and subgroups took place. This initial point specification embraced two existing facts. The first fact was a fully formed and well-developed noun system, while the second fact was an undeveloped diffusive verb system. The verb system got its further development later, when the intensive formation of the Indo-European language community took place. It was a product of the separate language groups. germanic future tense gothic verbal

The 21^st century's research of the Germanic future tense development looks rather incomplete and multitargeted without steady and gradual approach to the reconstruction of the analytical structures for indicating the future action in the system of the Germanic verbs. The investigation of the future tense development in English is focused on the functions of the individual verb will in different situations based on various semantic backgrounds [23], but the diachronic aspect looks very subtle in the above-mentioned research and does not reveal the specification and polyaspectness of the modal verb will. The situation is partly better in the historical research [19], which is connected with the problem of gradual grammaticalization; but the sphere of research is restricted only by the boundaries of English itself. The comparison of the closely connected and related west Germany languages (English and Dutch) demonstrates [4] the formation and function of the present tense with the future meaning enumerating the cases of that grammatical form usage. The research would be far better if all other West Germanic languages (German, Frisian, Africans, in particular) were involved into the process of comparison. The process of semi-auxiliary verb werden grammaticalization broadens the future tense aspect investigation grasping the German language, too [9], but the research is narrowed by involving only werden and avoiding wollen, sollen. The lack of information about the future forms in the North Germanic (Scandinavian) languages is made up for in the synchronic research [10] connected with the Swedish language. The investigation reveals the future meaning only in the synchronic aspect without taking into consideration the diachronic one. The drawbacks of above-mentioned research may be explained by the lack of steady diachronic investigation. The Gothic language as the main source of the East Germanic languages is not involved into the above-mentioned research. It makes all the attempts of outlining the future tense development in the Germanic languages incomplete and uncertain, without definite understanding and steady penetration into the formation of verb forms involving the processes of grammaticalisation. Our research tries to reconstruct the earliest initial processes of grammaticalisation, which may be traced in the Gothic language. The present day investigation of the German verb system is connected only with the separate, individual phenomena, the modal verbs and their equivalents [5], but there is a lack of generalisation and balance between the separate Germanic languages and the Common Germanic language comparison.

Main information presentation

When the grammatical categories of the Germanic verbs are analysed, it is very important to distinguish between personal and non-personal verb forms because their grammatical categories differ. Such grammatical categories as the person and the mood are irrelevant for non-personal forms; for some of them the category of number is also irrelevant (e.g. the infinitive). The structure of the tense and voice oppositions is different for personal and non-personal forms. The Gothic infinitive did not denote any particular time at all. Its only form represented all the varieties of the indefinite mood, which occurred in the original biblical text. There are three different correlations between the Gothic infinitive and the Greek infinitive forms. The first correlation can be represented with a formula Goth. inf. = Greek inf. aorist: bi biuhtja gudjinassaus hlauts imma urrann du saljan (Lk. I, 9) (according to the custom of the priest's office, his lot was to burn incense), saljan translates Greek inf. aorist Ouuiaoai. Ip Aileisabaip usfullnoda mel du bairan (Lk. I, 57) (it was time for Elizabeth to bear (a child)); bairan corresponds to Greek inf. aorist II. Gamelid ist auk patei aggilum seinaim anabiudip bi puk du gafastan puk (Lk. IV, 10) (For it is written, He shall give his angels charge over thee, to keep thee); gafastan corresponds to Greek diayu/.acai inf. aorist I. The second formula is the following: Goth. inf. = Greek inf. perf. act.: hausidedup ina siukan (Php. II, 26) (ye had heard that he had been sick); where siukan = Greek inf. perf. Man auk ni waihtai mik minnizo gataujan (Cor.II.XI, 5) (For I suppose I was not a whit behind the very chiefest apostles); where gataujan = inf. perf. The third formula is Goth. inf. = Greek inf. pres.: insandida ins merjan piudangardja gudis (Lk.IX, 2) (and (he) sent them to preach the kingdom of God); where Goth. merjan = Greek inf. pres. Consequently, the Gothic infinitive equally translates Greek inf. pres., aor. and perf. This confirms the conclusion about the absence of any tense differentiation in the Gothic infinitive. It means that the whole system of Gothic verbum infinitum is beyond any tense differentiation.

As far as verbum finitum is concerned, it is common knowledge that the oldest nucleus of tense opposition is based on contrasting the two tenses: present and past. The essence of the opposition lied in contrasting two different processes. The first process took place or happened in the past, while the second process was not characterised by this feature (present future). There was a strong component in the above opposition. It was the past tense, the semantic differentiational feature of that was rather clear, whereas the process itself could mean the whole chain of different processes. The Gothic present represents three different processes. It represents the process correlating with the present tense: qipa auk izwis (Mt. V, 20) (I say unto you). It may represent the process belonging to no particular time: witup, broprjus kunnandam auk witop rodja, patei witop fraujinop mann, swa lagga hveila swe libaip (Rom.VII, 1) (the law hath dominion over a man as long as he liveth). The most interesting case is the representation of the future tense involving the present tense forms: managai fram urrunsa jah saggqa qimand (Mt.VIII, 11) (many shall come from the east and west). It is possible to represent the future tense meaning using the oblique mood.

It is quite difficult to understand the influence of Vulgar Latin, because in the postclassical period Vulgar Latin mixed the functions of the oblique mood and the future tense. This was due to the modal character of the Vulgar Latin future tense. But still the influence of Vulgar Latin was rather possible [3, p. 231]. In any case, the Gothic oblique mood represents the Greek future tense equally in the cases when the tense has the imperative mood shade in the orders and recommendations of the New Testament and when it represents the future tense protocategory.

1) The future tense with the imperative mood shade: hausidedup patei qipan ist: frijos Nehvundjan peinana, jah fiais fiandpeinana (Mt. V, 43) (Ye have heard that it hath been said, Thou shalt love thy neighbour, and hate thine enemy); where frijos, fiais are optative forms representing Greek future ayanqmiq каї uiaqmic. Ni ufarswarais, ip usgibais fraujin aipans peinans (Mt. V, 33) (Thou shalt not forswear thyself, but shalt perform unto the Lord thine oaths); where ufarswarais, usgibais = Greek, anoddjmi Q. The Greek future represents only the category of the future tense without any modal shade: jah sijaina po twa du leika samin (Mk. X, 8) (and they twine shall be one flesh); where sijaina = Greek єаоутаї. Hva nuh taujai frauja pis weinagardis? (Mk. XII, 9) (What shall therefore the lord of the vineyard do?) where taujai = Greek noiqoei. In parallel with all these forms of the Greek future action representation the Gothic language used the present tense of simple and prefixed verbs.

Independently of Streitberg's ideas about the Gothic present of prefixed verbs and its correlation to the future perfect [21, p. 121], there is a tendency to use the present tense of durative and perfective verbs (according to Streitberg's theory) for representing equally the present and the future tenses. This demonstrates the absence of definite differentiation of these two categories. There is a case when a simple verb equally correlates with the present tense and the future tense of the Greek text: saei Galaubeid mis: po waurstwa poei ik tauja, jah is taujip jah maizona paim taujip (Jn. XIV, 12) (he that believeth on me, the works that I do shall he do also; and greater works than these shall he do); where the first tauja = поко, the second = noiqosi. The opposite case is in Cor. II.I, 10: izei us swaleikaim daupum uns Galausida jah Galauseip (єррЬоато), du pammei wenidedum ei Galauseip (рЬоєтаг) (Who delivered us from so great a death, and doth deliver: in whom we trust that he will yet deliver us).

The introduction of periphrastic forms was caused by the necessity of describing the newly formed conceptions. It was an innovation in the Gothic language. It was possible to trace the dependence of periphrastic forms on Greek and Latin patterns. In Gothic these forms are extremely rare. This demonstrates that they are alien elements for Common Germanic. The modal character can be seen in the periphrastic forms and in the oblique mood forms. The modal shade underlines the necessity and the desire to act. The future in the Gothic language, when it has some specific arrangement, demonstrates the tendency to represent not only tense but modal relations as well. In particular, modal relations prevail over tense relations. This is due to periphrastics involving skulan, haban. These preterite-present (later modal) verbs represent Greek pekkeiv and express the shade of obligation. The future, as a purely temporal category, uses abstract werden and gets its final arrangement in the Germanic languages at the very end of the Middle Ages. It belongs to the bookish style. Even now this form is absent from some dialects of the German language [6, p. 294-295].

There is a correspondence in the Gothic language with duginnan (to start, to begin). It occurs only twice in Gothic texts. The Gothic periphrastic future is formed using duginnan, haban, skulan. In Gothic it is possible to see the influence of the Greek-Latin periphrastic future. In Latin there are incipere, habere, in Greek there is ud/.eiv These forms compete with old forms in Vulgar Latin and Coine. In Gothic these forms are of sporadic nature, and there was no need to form the category of future in the Proto-Germanic language [6, p. 17]. The most abstract form with duginnan occurs only twice. This form correlates with Latin incipere [18]. The first example: jah in pamma fagino, akei jah faginon duginna (Php. I, 18). (and I therein do rejoice, yea, and will rejoice). It correlates with Greek каї ev аірю, аХХа. каї харпоиаі. where the contrast of Greek Хаірю - харцоораг correlates with Gothic fagino - faginon duginna. Here the use of a periphrastic form, as in many other cases, serves to represent the Greek contrast of present and future. Here is another case with duginnan: wai izwis, jus hlahjandans nu, unte gaunon jah gretan duginnid (Lk.VI, 25) (Woe unto you that laugh now! for ye shall mourn and weep), where duginnan emphasises inchoative action. Haban and skulan, as a rule, represent of the original text. Latin influenced the development of the periphrastic future in Gothic. Latin habere had the shade of obligation with the contrast to the future simple, and there was a complete correspondence with Greek ue.eiv later, in the period of Vulgar Latin, when the traditions of Classical Latin were ruined under the influence of Colloquial Latin. Habere became widely used, substituting all the old future forms. The modal future gained a victory over the purely temporal category that represented the future tense in Classical Latin.

Gothic haban + inf. represents Greek игл/.eiv + inf.: jah andnimands aftra pans twalif dugann im qipan poet habaidedun ina gadaban (Mk. X, 32) (and he took again the twelve, and began to tell them what things should happen unto them); in Greek та иел/.ovTa аитю rnupaiveiv. Ip silba wissa patei habaida taujan (Jn. VI, 6) (he himself knew what he would do); in Greek - epeAAev noieiv. Qapuh pan pana Judan Seimonis, Iskariotu, sa auk habaida ina Galewjan (Jn. VI, 71). (He spake of Judas Iscariot the son of Simon: for he it was that should betray him); in Greek - гиел/.ev napabibovai.

Skulan + inf. in the same way corresponds to Greek ue./.Јiv + inf.: Sa ist Helias, saei skulda qiman (Mt. XI, 14). (This is Elias, which was for to come); skulda qiman = Greek o pekkmv e/y/eaOai, Pai gasaihvanans in wulpau qepun urruns is, poei Skulda usfulljan in lairusalem (Lk. IX, 31) (those who saw him in glory said about his going out that should happen in Jerusalem); skulda usfuljan = eue /.ev пqpohv. The same constructions occurred in Lk. IX,44; Lk. XIX, 11; Jn.VII, 35; Jn.VII,39; Jn.XII, 33; Jn.XVIII, 32; T. II.IV, 1. The specification of the verb skulan and its semantics that represents obligation did not permit its use in other cases. In three cases haban represents Greek future without psAAeiv. In two cases out of these three it specially emphasises the future in contrast to the present, formed from one and the same verb. Compare: tauja jah taujan haba (Cor. II.XI, 12). (But what I do, that I will do); taujip jah taujan habaip (Th. II.III, 4). (Ye both do and I will do).

Periphrastic future forms in Gothic were taken from Greek and Latin, and they are predominantly connected with a modal shade. They were created due to the necessity of representing this very moment in the category of the future tense. Later, as an exception, when the traditional representation of the future with the help of the forms of the present tense of the oblique mood became ineffective and stopped functioning sufficiently, the periphrastic forms were introduced. There are only a few periphrastic forms for future representation in the Gothic Bible texts. It is unknown whether bishop Ulfilas emphasised the special shade of future which was not represented in the Greek original.

The analysis of the future tense development in all Old Germanic languages shows that there are common tendencies to broaden two- component tense paradigm (present~past) involving different combinations of Auxiliary Verb + Notional Verb. These tendencies are evident in other Indo-European languages and, in particular, in such archaic and very distant from the Germanic group as Hittite [15, p. 213-214]. It is likely that in the same system of the active voice of the Germanic languages there were the oldest word combinations which were aimed at the limitation of the present tense meaning. These word combinations were connected with the presentation of the future tense projection. In the Gothic language there are word combinations with the initially weak future meaning. These word combinations correlated with further analytical forms of perfect and past perfect of other Germanic languages. Their introduction may have been connected with the present tense forms in future meaning, but the semantic representation was inadequate. The function of the future tense specification can be performed by a phrase or supraphrase context which has two types of indicators. The first type of indicators indirectly implies temporal features of the process. The second type of indicators has lexical units or syntactic forms: in pizai usstassai, pan usstandand, hvarjamma ize w airpip qens? (Mk. XII, 23) (In the resurrection therefore whose wife shall she be?).

It is possible that the presence of these specificators did not always represent future meaning. Only in the subordinate adverbial clauses of condition and time was this function of specificators sufficient. That is why this tendency is present in the Modern Germanic languages. The analysis and comparison of Gothic and Greek tense forms demonstrates that the Greek future tense correlates with the Gothic present tense as the most traditional way of tense representation. In the temporal aspect the Gothic present tense may be recognised as a colony-forming archi-unit. Functioning in the language this unit acts as a pluripotential (temporal) precursor. The pluripotentiality means the use of that archi-unit for representing present, future and no-particular time actions. Periphrastic Gothic future, in particular skulan + inf. may be recognised as a monopotential (temporal) precursor since it predominantly represents future tense with some modal meaning.

Conclusion and further investigation

Gothic analytical future tense forms were created against the background of the dominating Gothic present colony-forming archi-unit. This unit is recognised as a pluripotential precursor. The introduction of future tense analytical forms was caused by the insufficient effectiveness of the future tense specificators or the necessity of giving some additional peculiarities of the action described, which was not reflected in the Greek original texts. It is necessary to take into consideration the fact that the Gothic preterite-present verbs functioned as a firmly-drawn group that gradually carried out functional-semantic expansion taking upon itself the function of the future tense specificator. The Gothic analytical future realised its target into future, using preterite-present or inchoative verbs. In fact, these two variants determine the formation of the analytical future tense in all Germanic languages. In Old Scandinavian languages there were periphrastic forms with munu, skuly + inf. (modal future) and a widely used model with kommer [16, p. 231-232]. In English the future tense forms involved a modal verb; in German there were two ways of its formation in different periods [3, p. 208-263]. Further investigation should focus on tracing the future tense in the related Germanic languages which belong to the West Germanic group. The contact interaction with the North Germanic Group should also be taken into consideration.

References

1. Ammann H. (1936). Germanischen und indogermanischen. Germanen und Indogermanen. Volkstum, Sprache, Heimat, Kultur, Festschrift fur Herman Hirt. Heidelberg: Fachbuchverlag, 328-342 [in German].

2. Arntz H. (1936). Gemeingermanisch. Germanen und Indogermanen. Volkstum, Sprache, Heimat, Kultur, Festschrift fur Herman Hirt. Heidelberg: Fachbuchverlag, 429-451. [in German].

3. Behaghel O. (1923). Deutche Syntax. Heidelberg: Carl Winter Universitat. Bd.II. B. Adverbum, C. Verbum, 372 s. [in German].

4. Beheydt G. (2005). The absolute and the relative present tense with future time references in English and Dutch. Crosslinguistic Views on Tense, Aspect and Modality. Amsterdam: Rodopi, 15-32 [in English].

5. Botsman, A.V. & Morenets, I.M. (2015). Modals and periphrastics in English and Dutch. Science ans Education. A New Dimension. Philology, 68 (III), 7-10. [in English].

6. Brinkmann, H. (1931). Sprachwandel und Sprachbewengungen in althochdeutscher Zeit. Jena: Friedrich - Schiller - Universitat, 428 s. [in German].

7. Brugmann, K. & DelbruckB. (1897). Grundrif der vergleichenden Grammatik den indogermanischen Sprachen. Strassburg: Fachbuchverlag, 386 s. [in German].

8. Deiter, F. & Bethge R. (1898). Laut und Formenlehre der altgermanischen Dialekte. Leipzig: S. Hirzel, 542 s. [in German].

9. Diewald, G. & Mechthild H. (2005). Die Entwicklung von werden and Infinitiv als Futurgrammem: Ein Beispiel fur das Zusammenwirken von Grammatikalisierung, Sprachkontakt und soziokulturellen Faktoren. Grammatikalisierung im Deutschen. Berlin: Mouton de Gruyter, 229-250 [in German].

10. Hilpert, M. (2006). A synchronic perspective on the grammaticalization of Swedish future constructions. Nordic Journal of Linguistics, 29(2), 151-173 [in English].

11. Hirt, H. (1928). Indogermanische Grammatik. Leipzig, Veb. F.A. Brokhaus Verlag. Bd.IV, 482 s. [in German].

12. Karstein, K. (1936). Indogermanisch und Germanisch. Germanen und Indogermanen: Volkstum, Sprache, Heimat, Kultur, Festschrift fur Herman Hirt. Heidelberg: Fachbuchverlag, 297-327 [in German].

13. Kluge, F. (1913). Urgermanisch. Strassburg: Fachbuchverlag, 385 s. [in German].

14. Krahe, H. (1957). Germanische Sprachwissenschaft. Berlin: Akademie Verlag. Bd.II: Formenlehre, 438 s. [in German].

15. Kronasser, H. (1956). Vergleichende Laut und Formen lehre des Hethitischen. Heidelberg: Carl Winter Universitatsverlag, 526 s. [in German].

16. Meillet, A. (1917). Caracteres generaux des langues germaniques. Paris: Academia, 273 p. [in French].

17. Meillet, A. (1908). Les dialects indoeuropeens. Paris: Academia, 428 p. [in French].

18. Merold, F. (1875). Futurum und futurische Ausdrucke in Gotischen. Wissenschaftliche Monats blatter. Wien: Lang, 283-298 [in German].

19. Nevalainen, T. (2004). Three perspectives on grammaticalization: Lexico-grammar, corpora and historical sociolinguistics. Corpus Approaches to Grammaticalization in English. Amsterdam: John Benjamins, 1-31 [in English].

20. Stang Chr., S. (1942). Das slavische und baltische Verbum. Oslo: Akademiska Bokhandeln, 342 s. [in German].

21. Streitberg, W. (1910). Gotisches Elementarbuch. Heidelberg: Carl Winter Universitatsverlag, 275 s. [in German].

22. Streitberg, W. (1896). Urgermanische Grammatik. Einfuhrung in das verleichende Studium del altgermanishcen Dialekte. Heidelberg: Carl Winter Universitatsverlag, 488 s. [in German].

23. Ziegeler, D. (2006). Omnitemporal will. Language Sciences, 28(1), 76-119 [in English].

Размещено на allbest.ru