... THE USUAL CONCEPTION of a comparative grammar of the Slavic languages or a comparative grammar of the Indo-European languages is based on the assumption that languages are pure, that the history of languages is an uninterrupted and undisturbed development of an original system in various directions without the interference of foreign languages. Fundamental linguistic studies, such as the Grundriss der vergleichenden Grammatik by Brugmann and Delbrück, give such an impression in fact by working with several dozens of roots or groups of etymologically related words. Yet it is sufficient to consult any dictionary of such a supposedly pure language as Indo-European or Slavic to convince oneself that it contains a far larger number of assimilated words or words of obscure or unclear origin than words suitable for comparative grammar in the current sense of the word. The latter, indeed, constitute a negligible number, if we consider that ordinary comparative grammars do not cover entire languages, do not give an insight into the entire structure of the languages in question. Many people, however, seem to be unaware of this, treating the thesis of the absolute purity of language as a linguistic dogma. They admit, it is true, the possibility of foreign elements being borrowed, but at the same time add that the basic character of a given language remains unchanged. Thus, for example, they do not deny that many Romance elements have entered English, but they insist that all these foreign elements have been completely assimilated without in the least affecting the Germanic base of the language. And once such an allegation is made by an authoritative scholar, it is bound to be repeated by a whole crowd of imitators.
It is well known that teachings drawn from foreign grammars which have been obsolete even within their own areas have hampered the development of original but true-to-fact views on the nature of particular languages.
How a countrys public opinion may be offended by any attempt to emancipate a given science from prevailing views, from prejudice and preconceived ideas, can be shown by an example from the history of Russian grammar. Thirty-five years ago, N. P. Nekrasov attempted to treat the Russian verb in an independent fashion (in his O znacenii form russkogo glagola, St. Petersburg, 1865). He was shouted down and bitterly attacked. How dare he, a Russian, look with his own eyes at the facts of the Russian language and see in it what actually exists instead of what is supposed to exist according to the stereotypes of medieval Latin grammars! This peculiar kind of Westernism was inspired by the fear that the acceptance of Nekrasovs teaching would force one to think and, as is known, denken ist schwer und gefährlich! It is all right to go on repeating the ideas of others, but, for goodness sake, let us not stir things up!
Do not think, however, gentlemen, that my remark is intended to champion the ideas of the Slavophile imitators of those German philosophers who claimed special originality for the Russian or Slavic mind. On the contrary, I do not recognize any national logic, but only identical laws of thought for all mankind. There is no European, American, French, English, German, Russian, or Polish science—there is only one science, common to all men. For the same reason I believe that Western European scholars have no monopoly on scientific discoveries and generalizations, that a Russian or Pole may also have original ideas, and that there is therefore no need to slavishly follow so-called European science and to hold to its tenets blindly and uncritically.
Each of us has a duty to look with his own eyes. And if we examine the question of the mixing or non-mixing of languages with our own eyes, we will have to admit that no language is pure or unmixed.
Mixing, be it of a physical or psychological nature, is the beginning of all life. Mixing takes place in the development of any individual language beginning in early childhood. The child is influenced by his parents and other persons of his immediate milieu. The language of any of these persons is surely, even if negligibly, different from the language of other persons, and the various individual languages shape the language of the child who will almost invariably select those linguistic peculiarities which he can master more readily and with less effort.
For example, if in a Russian or Polish family, at least one person pronounces ŭ instead of ł, or a velar r instead of an apical r, the children of this family will, in the majority of cases, tend to make the same sound substitution, pronouncing ŭa instead of ła, and ra (with a velar r) instead of ra (with an apical r).
Marriage produces mixed family languages, and the contact of tribes and nations leads to interaction and mixing of dialects and, on a wider scale, to mixing of linguistic groups and of tribal and national languages.
Nations and tribes live in close proximity or are interspersed. Along the boundaries of tribal and national languages, multilingualism inevitably arises, and with it, mixed languages.
Nomadic life, military campaigns and war service, abduction of foreign women and slaves, trade, scientific exchange, etc.,—all these factors contribute to the mixing of languages.
We could easily hypothesize that at a certain time the Slavs of Eastern Europe were scattered among the Finns (to a certain extent this is still the case even now); in other parts of Europe the Finns mixed with Germanic and Lithuanian tribes; in Central Europe Slavs lived among Germans; Celts, who were scattered over almost all Europe and in Asia Minor, underwent the influence of peoples of different origin who, in turn, influenced their languages; the Aryans of Eastern India were scattered among peoples of Dravidian origin; in Western Europe, tribes related to the modern Basques were constantly in contact with Celtic, Romance, and other tribes; the language of the nomadic Gypsies, who are still nomads, absorbed all kinds of foreign elements; the Jews, who have lived among other peoples, have felt the impact of the languages of their fellow citizens and have influenced their languages to a greater or lesser extent.
But this interaction and mixing of languages is determined not only geographically and territorially, but also chronologically. An old language used as a language of the church, of ritual or in traditional formulas and expressions, may influence the living language of a given period which in turn undergoes the influence of the latter. It is enough to think of the interaction between modern Latin and the Romance languages, or between the varieties of Church Slavonic and the modern South and East Slavic languages.
The formation of a mixed language, that is, of a language whose elements prevail over another language, though not without being seriously modified by the language spoken by the linguistically assimilated people, can in some cases be documented by historical records. Thus, for example, Pastor Bielenstein1 was able to trace the formation of Latvian, a language of the Baltic or Aistian branch of the Indo-European languages, and to show the indelible traces which Finnish has left in its phonetics, morphology, etc.
The Armenian language is, for a variety of reasons, justly treated as one of the Indo-European languages, but there are good reasons (considerations of its basic structure, as well as its peculiarities) why it should be grouped also with the Turco-Tatar or Uralo- Altaic or some such closely related group of languages. For example, external, physical, and spatial relations are in the Armenian declension expressed in the Tatar manner (in the locative, ablative, and instrumental cases), while social relations continue to be expressed through Indo-European forms (in the genitive, dative, and accusative). The Armenian suffix of the plural is obviously not of Indo-European origin, but a borrowing, if not from Tatar, at least from some other language. The loss of gender distinctions and the lack of sexualization of reality must be due to foreign, non-Indo-European influence. The question of the ethnic origin of the Armenians is even more complicated by the fact that the Armenians have historically mingled with the Jews and other Semites who had shared their habitat with Caucausian, and especially Georgian peoples whose languages must have affected to a noticeable degree the character of the Armenian language.
The influence of foreign peoples upon the development of the Romance languages from vulgar Latin dialects can likewise be traced historically.
In recent times, mixed dialects have been formed in border areas where Chinese have come into contact with Europeans. Such are, among others, the Kjaxta or Majmak dialect of Russian- Chinese and the international dialects of English-Chinese and Portuguese-Chinese.2 Recent decades have witnessed the creation of artificial mixed languages which are to serve as vehicles of international communication: Volapük, Esperanto, Bolak.
The effects of language mixing are twofold: on the one hand, they enrich a given language with elements of another, foreign language (with a stock of new words, new syntactic phrases, forms, pronunciation), and, on the other hand, they weaken the distinctions peculiar to the given language. They bring about a more rapid simplification and fusion of forms, the disappearance of irrational distinctions, the assimiliation of some forms to others (through the process of analogy), the loss of declension and its replacement by single forms with prepositions, the loss of conjugation and its replacement by single forms with prefixes of a pronominal origin and auxiliary particles, the loss of a morphologically mobile stress, etc.
In the contact and interaction of two languages that mix in a natural way, the victory goes in each case to that language which is marked by greater simplicity and clarity. Simpler and clearer forms prevail over difficult and irrational ones. Thus, if one language has gender distinctions and the other language lacks them, the language resulting from the mixture will either lack these distinctions completely or exhibit them in a weakened form. If only one of the two interacting languages has an article, or personal possessive suffixes (that is, suffixes denoting the belonging of an object or person to another person: my, your, his, her, our, your, their), it is much more likely that the resulting mixed language will inherit this analytic or decentralizing feature than not.
The same can be said regarding the predominance of one of two mixing languages in their struggle for survival: victory belongs to that language which is more easily acquired and which requires less physiological and psychological energy. Thus, for example, in localities where Roumanians live side by side with Germans or Slavs, the predominating language, the language of intertribal communication, is Roumanian; this is understandable, since Roumanian is more easily mastered by Germans and Slavs than the other way around. Similarly, because of the relative simplicity of Tartar as compared with the far more difficult Russian, the language of communication between Russian and Tartar peasants within Russia is usually Tartar. Of course, this takes place only in the natural course of things, in the absence of conscious interference on the part of administrative authorities and of other political and social forces which employ preventative or coercive measures.
Looking, then, at the whole Slavic linguistic world, we can find in it a considerable number of cases of language mixing. In some instances this mixing is occurring now and can be observed in actu; in other instances, we are faced with the results of mixing that took place a long time ago.
Right now mutual influence can be seen to take place, for example, between literary Russian and the various dialects of Great Russian and Ukrainian or varieties of Polish, with Great Russian gaining the upper hand as a result of existing peculiar conditions. Similar mutual influence takes place between Polish and Ukrainian, Polish and Czech, Polish and Slovak, Polish and Kashubian, Serbian and Bulgarian, etc. Mutual influence is also observed between the various Slavic literary languages and the territorial dialects where the respective literary languages are being used.
Several peripheral Southwest Slavic areas (mainly the Slavic areas of Northern and Southern Italy) are subjected to Romance influence, as are Slavic areas bordering on Roumania. Some Southwestern and Northwestern Slavs (e.g., the Lusatians and the Poles of Prussia, including the Kashubians) are subject to strong German influence. The Slavs of the Hungarian kingdom are exposed to the influence of Hungarian. Within Russia, the Slavic linguistic element is in a state of interaction with various foreign, non-Slavic elements (Lithuanian, Latvian, Estonian, Tartar, čuvaš,3 etc.). In America, the speech of Poles, Czechs, Slovenes, and other Slavs is subjected to the influence of English and to gradual Anglicization.
Conversely, we can note the Slavicization of foreigners. Thus, the Russians are gradually assimilating the ethnic minorities; the Czech and Slovenes are partially assimilating the Germans; a partial, though negligible, Polonization and Russification takes place in Lithuania.
The final results of language mixing can be seen in the different parts of the German linguistic territory where Slavs have been absorbed, not however without leaving more or less clear linguistic traces in the German dialects which they have made their own. The same can be said of those parts of the Romance linguistic territory where we find descendants of Romanized Slavs.
Within the boundaries of the Russian state there are quite a few regions whose populations constitute a mixture of a foreign and a preponderant Slavic element. In the Northwest such Slavicized areas are the former territories of Lithuania and Estonia. The Polish cities were at one time occupied, in addition to the Jews, by settlers from various parts of Germany; now these cities are mostly Polonized, but the formerly predominant German element is of necessity reflected in the language of their present ethnically mixed populations.
With our own eyes, so to speak, we could observe the final Slovenization of several settlements of German exiles or colonists in Krajna and Gorica : in Nemški Rovt (Deutschruth), in Koritnica, in Stršišče, etc. As early as the first half of the 19th century, the inhabitants of these villages spoke a characteristic South German dialect (which is closely related to the dialects of Tyrol); in the 1870% only the old people could express themselves in this dialect; their children, who were then middle-aged, could still understand this dialect, but they no longer could speak it; but the youngest generation could not even understand this language of its forefathers. The commonly used dialect at that time was a Slovenian dialect, borrowed from nearby Slovenian neighbors. The local dialect, however, preserved clear traces of the German origin of its speakers in its phonetics, morphology, and syntax. Phonetically it was so thoroughly German that at a distance it gave the impression of a German dialect. Later, of course, under the influence of the school, sermons, and communication with the adjacent purely Slovenian villages, this German-Slovenian dialect has more and more relinquished its German imprint, though it is unlikely to lose it completely.
In view of such historically proved cases of the influence of foreign linguistic traits on the basic structure of Slavic languages, we have the right to admit a similar influence of foreign traits in wide areas of the Slavic linguistic world. We can regard Bulgarian, Macedonian, Upper and Lower Lusatian, Great Russian, Ukrainian, Slovenian, Serbo-Croatian and other areas as being linguistically mixed. The tiny group of the Rezija dialects shows certain basic features that cannot be explained without admitting external, non-Slavic influence. Czech, Slovak, and Polish exhibit peculiarities which are reminiscent of the Finno-Ugric branch of languages. Among these is the fixation of stress on a certain syllable of the phonetic word (on the penultimate syllable in Polish, and on the initial syllable in Czech and Slovak), and the loss of distinction in most Polish territorial dialects between the hushing consonants, s, z, c, and the hissing consonants, s, z, c, in favor of the latter. The original Slavic stress was morphologically mobile, and characteristic of certain morphemes, i.e., morphologically connected with significant parts of the word, and not bound to a definite syllable of the phonetic word. The mobile stress could not have disappeared spontaneously while the original structure of the Slavic word remained unchanged. The basis of this structure was a centralized and cohesive unity of the various significant parts of the word; such a structure is even now characteristic of all varieties of Czech, Slovak, and Polish. Only the impact of ethnic mixing with tribes incapable of perceiving the morphological mobility of stress could have caused the loss of this stress among the linguistic ancestors of the present-day Slovaks, Czechs, Poles, as well as of the Upper and Lower Lusatians.
The inability to pronounce hushing consonants (which characterizes most Polish dialects) is peculiar to the western Finns (Estonians, Suomi, etc.), but there is no doubt that in former times the Finns reached farther west, and it is most probable that some of them lived scattered amidst the Northwestern Slavs. That the Finns once lived together with the Lithuanian (Baltic, Aistian) tribes is attested by the development of the Latvians who formed a nation only in historical times, as is shown by the investigations of Bielenstein, which I mentioned above. And, in fact, the number of hushing consonants is much smaller in Latvian than it is in Lithuanian, whereas the hissing consonants prevail in the Latvian dialects and are less common in Lithuanian. The patterns of old Lithuanian stress were likewise destroyed in Latvian, evidently under the influence of the Kurs and Livonians, who are closely related to the present-day Estonians, and who were totally absorbed by the Lithuanian-Latvian tribes. The relationships of stress and quantity of vowels and syllables in the Latvian dialects are very similar to those found in the Estonian linguistic area.
In view of all the above-mentioned instances of interaction between Slavic and foreign elements, of all the completed and continuing influences of foreign dialects upon Slavic and of Slavic upon foreign dialects, we believe that the ideal Slavic linguist should master not only all the Slavic languages in their multifarious variety, but also all the languages that influenced or might have influenced the Slavic languages, and vice versa. Such a goal is, of course, only an ideal. I myself am quite remote from it; in the first place, I do not have perfect mastery of all Slavic languages, and in the second place, I am completely or almost completely unfamiliar with some foreign dialects that influenced or were influenced by the Slavs. But I can, at least, console myself that there is hardly a scholar who fits this ideal, that is, who knows the Slavic languages in all their variety together with all the abovementioned foreign languages.
Realizing full well the demands of such an ideal, we shall be content with a more modest goal: we shall limit our investigation to certain phenomena of the Slavic linguistic world and examine them in their mutual relation from a scientific, primarily psychological, point of view; on the basis of this investigation we shall then attempt to arrive at conclusions of a general linguistic nature.
Thus, we shall compare languages in their mutual relation. But the comparison and examination of languages in their mutual relation opens up two possible approaches.
In one approach, we can compare languages independently of their genetic (or historical) relationship. We can examine the identical features, changes, historical processes, and transformations of languages which are unrelated historically and geographically. From this viewpoint we can compare the development of the Romance languages with that of the modern Indie languages, the development of the Slavic languages with that of the Semitic languages, the development of Russian with that of Coptic, the development of English with that of Chinese, etc. Everywhere we are confronted with the question of the cause of the similarities and differences in the structure and evolution of language. This kind of linguistic comparison provides the basis for the broadest linguistic generalizations in the fields of phonetics, morphology, and semasiology (the science concerning the meaning of words and expressions).
The other approach enables us to compare languages on the basis of their historical kinship and geographic, social, and literary proximity. The comparison of historically related languages constitutes the foundation of comparative grammar in the current sense of the word. On it are based the comparative grammars of the Indo-European, Semitic, Uralo-Altaic (or Turco-Tatar), Germanic, Romance, Slavic, and other languages.
Less common is the comparison of languages on the basis of their geographic, social, and literary proximity, that is, the kind of comparison which takes into consideration their mutual influence in the widest sense of the word. Geographic contiguity, compact or diffuse proximity, commercial and other relations, wars, various types of cultural influence, even at a geographical and historical distance—all these provide the basis for comparative studies of two or more languages against a historical background. A most rewarding topic of study along these lines would be a comparative grammar of the Slavic and Baltic (Lithuanian-Latvian) languages as representatives of both the Indo-European and Finno-Ugric languages. Of equal interest would be comparative studies of the Slavic and Uralo-Altaic (Turco-Tartar) languages, of the West Slavic dialects and the neighboring German dialects (including the German literary language), of some Slavic languages and Hungarian, of the South Slavic languages and Roumanian, Albanian, modern Greek, and others.
Having entitled my course Comparative grammar of the Slavic languages with relation to other Indo-European languages, I intend to present, of course, a comparison based on the posited historical, or genetic, kinship of these languages. We shall, consequently, consider not the borrowed, but the internal material as it has evolved in time in various directions along the line of historical succesion from proto-Indo-European to Common Indo-European, and from proto-Slavic to Common Slavic. Thus, we shall devote our main attention, first, to the comparison of those aspects of the Slavic languages which are considered purely Slavic with those aspects of other Indo-European languages which are considered purely Indo-European, and second, to the formulation of general linguistic conclusions on the basis of these comparisons.
1. Die Grenzen des lettischen Volksstammes und der lettischen Sprache in der Gegenruart und im 13. Jahrhundert. Ein Beitrag zur ethnologischen Geographie und Geschichte Russlands von Dr. A. Bielenstein. (Mit einem Atlas von 7 Blättern.), St. Petersburg, 1892.
2. Through the study of mixed languages, Professor Hugo Schuchardt of Gratz has rendered a great service to science.
3. I call attention to the very interesting investigation by V. A. Bogorodickij, Dialektologiceskie zametki, II. Nepravilnosti russkoj reci u Cuvas, Kazan, 1900.