Mixed language. Confusion of languages
The problem of language confusion has been of particular interest to linguists in late XIX century. Thanks to the numerous works of Schuchardt devoted to this issue, this topic has constantly remained in the field of scientific interests of linguists.
Then a pause followed - linguistics asserted its method and won new positions. Finally, relatively recently, the Soviet linguist
N. Ya. Marr, without specifically dealing with this issue, attached particular importance to the mixing of languages in connection with the problem of the emergence of new qualitative formations in the language or the emergence of new language systems.
It would be interesting to revisit this issue, given the latest developments in linguistics. This is what we intend to do in the proposed work.
STATEMENT OF A QUESTION
To clarify our terminology, we must first distinguish between two categories of facts:
1. Mixed language (langue mixte). This term is applicable in cases where there is an interpenetration of two morphological systems, for example, in Norwegian or in the Creole languages of America (V e n d y e s, 21, p. 348).
The reason for the process is bilingualism. A bilingual speaker confuses two language systems, and the closer
A. R o s e t t i, Langue mixte et langues melangees, Acta Linguistica, V, Copenhague, 1945-1949, pp. 73-79; see also A. R o s e t t i, "Linguistica", s'Grauenhague, 1965, pp. 65-70.
the same these systems, the easier it is to mix; so, as a result of mixing the Norwegian and Danish languages, riksmol arose.
2. Language with elements of confusion (langue melangee). A language with elements of confusion suggests borrowings from other languages that do not affect morphology, which only rarely perceives individual elements a foreign language
Such are, for example, in Romanian the -o vocative in feminine gender names, borrowed from Slavic languages, or in Welsh (Welsh) - the ending -s of the plural of names, borrowed from English.
As for suffixes - and the number of suffixes of Slavic origin in Romanian is very significant - they penetrated through the dictionary (being separated from the composition of Slavic words containing certain suffixes, and subsequently became productive in Romanian).
This is also the reason for the process in bilingualism, but, generally speaking, vocabulary borrowing does not imply mandatory knowledge of the language from which individual elements are borrowed.
The phenomenon of bilingualism can also explain language calques, i.e., the reproduction of the internal form of a foreign word. For example: German. Eindruck, Ausdruck are formed according to the type of im-pression, ex-pression; st.-sl. chrymnlo "ink" - tracing paper from Latin atramentum and goth. swartizl (M e i 1 let, 6, p. 68); rum. unt-de-lemn "vegetable oil" (literally, "wood oil") is formed according to the Bolg type. Darven oil, etc.
A German speaking French, a Slav speaking a Romance language, a Romanian speaking one of the Slavic languages, created tracing papers from foreign languages; we find the same phenomenon in the Latin auiare "bird-catchers" (cf. old-French oiseler - tracing-paper from the other-in-German fogalon, created by a certain native speaker of the Germanic language who settled in Italy, Duvau, 3).
The difference we have established between the two groups of phenomena is very significant. It is able to reconcile those who accept the theory of confusion of languages and those who reject it.
Max Müller (Schuchardt, 16, p. 5) and F. Géo Mol argued that mixed languages do not exist. G. Schuchardt (16, pp. 5-17, p. 131) believed that there are no languages without confusion. The same opinion was held by N. Ya-Marr (Marr, 4, I, pp. 55-56, III, pp. 5-6; I, pp. 23, 27, note 1; III, p. 5 ; V, p. 405; Meshchaninov, 9), who believed that all languages at the time of their occurrence were mixed.
These theories consider language from the outside (see Shcherba, 16, p. 7); the linguist expresses himself on this issue, based on the analysis of this language.
There is another approach to this issue: the language is considered as if from the inside, based on the linguistic sense of a native speaker. Using this method, Meillet argued that the speaker feels that he uses one language: "in all cases known so far, we are dealing with one continuous linguistic tradition" 6 .
Meillet, however, allowed for use in special cases, such as languages Far East, morphological means of another language .
So, we should decide: which of these two statements is correct: the one that states that mixed languages exist (after all, the existence of languages with elements of mixing is generally recognized), or the one that rejects the reality of their existence?
It has rightly been pointed out that it is dangerous to take the linguistic consciousness of the speaker as the basis of linguistic research (Magt i n e t, 5, pp. 36 et seq.). This danger is that you can draw conclusions based on the subjective assessment of the speaker. If for the knowledge of the subject it is necessary to reveal it, then it follows that objective criteria must be used to answer the question posed.
But if the objective means of analysis that we have at our disposal allow us to recognize elements in the morphology of a given language various origins, which belong to two different systems, then we will inevitably come to the conclusion about the correctness of the first proposition - that is,
E. to the fact that mixed languages exist.
1. Mixed language (Langue mixte)
A mixed language is a product of bilingualism. A mixed language is more likely to occur where there are two close language systems. The influence of a foreign language extends to the phonetics, morphology and vocabulary of a given language.
The less developed the borrowing language is, the more changes the language from which borrowings are made undergoes (cf. Creole languages) 13 . Native language people who borrow a foreign language are gradually degrading. That's what happened to German in America, which became mixed under the influence of English (Wundt, 23, pp. 404 et seq.) 14 . Similarly, in French, local dialects also gradually disappeared under the onslaught of a national language. Morphology offers the greatest resistance, but in the end it also gives way, and the old language system immediately disappears.
A mixed language is, for example, Norwegian Rixmol; its phonetic system is Danish-Norwegian, the distribution of phonemes is explained by the phonemic system of Danish, the morphology is mixed, Danish-Norwegian, the dictionary also contains elements of both languages (Sommerfelt, 19).
The Creole languages (Negro-Portuguese, -English, -French) have also long been regarded as mixed (Schuchardt, 17, p. 135 et seq.; Delafosse, 2, p. 559); the grammar of these languages is Negro-African with elements of respectively Portuguese, English, French (Delafosse, 2, p. 559; for example: stone is formed according to the type of dual ma-dale "stones", Schuchardt, 17, p. 137). So called. "lingua franca" is a Romance language with Turkish or Arabic grammar.
This is how a new language system is born, new language(Meshchaninov, 9).
Thus, bilingualism is to blame for the appearance of mixed languages, but there are cases when two languages exist side by side, and interpenetration does not occur. Both languages exist in isolation, and the speaker sharply distinguishes between their use. In this case, the transition from one language system to another is difficult or even completely impossible. There is a case when a woman, a peasant woman from a Transylvanian village, spoke Romanian and Hungarian from childhood and spoke both fluently, but was completely unable to translate at least one phrase from one language to another: in her mind, both languages were separated by a blank wall.
Undoubtedly, this is an exceptional case. Most often, the speaker simply experiences certain difficulties when switching from one language to another (Shcherba, 15, p. 7 et seq.).
A third case should also be distinguished: when two language systems coexist, forming in our minds one system of associations; every element in one language has a counterpart in another language. Here the transition from one system to another is easy. Such is the situation with the Sorbian language: the speaker equally uses the Sorbian and the corresponding German word (Suspensian is a mixed language with two terms, Shcherba, 15, p. 7).
2. Language with elements of mixing (Langue melangee)
There are no languages absolutely devoid of any elements of mixing, which means that all languages are mixed to some extent (see above).
Mixing affects primarily the vocabulary. The phonetics, syntax and morphology of a given language can also be mixed under the influence of a foreign language, but to a much lesser extent, while the vocabulary can change completely - for example, in Armenia, the morphological system of the Gypsy language is Armenian, and the vocabulary is Gypsy (V e nd y - e s, 21, p. 344), and among the gypsies of Spain it is Spanish with gypsy vocabulary (Schuchardt, 16, p. 10), while morphology only in exceptional cases perceives some elements of a foreign language *.
Let's take the Romanian language as an example. Its phonetics bears some features of Slavic influence (iotized pronunciation e: el "he" is pronounced TseP, etc.); Romanian syntax is not devoid of Slavic influence. Morphology, which acts as closed system language, not subject, according to some linguists (cf. T e s p i e g e, 20, p. 87), to foreign influences, also contains some elements of Slavic origin (vocative forms, suffixes, prefixes, numerals). But only in the dictionary did the Slavic influence fully manifest itself: according to the statistics of 1879, out of 5765 words of the Romanian language, 2/6 are of Slavic origin.
The morphological system of the Romanian language as a whole remained outside of foreign influences.
It is easy to note that in order to construct a Romanian phrase, one should refer to Latin elements.
Let's take a few words of Slavic origin and make a Romanian phrase out of them: Iubesc re prietenii mei dragi "I love my dear friends." This phrase contains three Slavic words: iubi (verb), prieten and drag. But iubi has latin ending(-esc), re (lat. per) - accusative construction with nouns with the meaning of the person, mei - pl. number from meu (
Phraseologism "Mixing languages" meaning
This expression is familiar to us from the biblical event, the so-called "". In ancient Babylon, people decided to build a tower, as high as the sky. However, God was angry with the people, and in order to prevent their proud plans, he mixed up all languages. People who previously spoke the same language suddenly began to speak many and ceased to understand each other.
The explanation for this legend is quite simple. Ancient Babylon stood at the crossroads of many trade routes and roads, therefore, there has always been a multilingual population. In those days, people did not understand why everyone does not speak the same way, but each in his own dialect. Many versions were invented, sometimes quite witty. The tale of the "Babylonian pandemonium" fit perfectly.
Curiously, even the name of the city of Babylon, according to some Hebrew books, means "mixing." However, this misconception, since the word "Babylon" ("Babilon" among the inhabitants of the city) came from the word "Bab Ilu" from the ancient Akkadian, which means "Gate of God". For comparison: on Arabic: "Bab-el-Mandeb", which means "gate of tears", in Hebrew: Gabriel - "God's man", Michael - "like God", Raphael - "God's help". Legends, in order to look true, can very cleverly twist everything in their own way!
Today the expression confusion of languages" is used when it comes to confusion, confusion, a motley crowd in which nothing can be made out. “Since yesterday, there has been a complete mixture of languages in the house - daughter graduation class finished!”
Language contacts are not limited to the processes of integration and differentiation. Representing a complex non-simultaneous phenomenon, they can take on different forms.
Intensive and long-term contacts of peoples often lead to bilingualism (or bilingualism Y ‘double, double’, lingua'language'). It has been found that about half of the total population the globe is either bilingual or multilingual, and in many countries of the world bilingualism is the norm (compare, for example, the situation in Russia, on whose territory, along with Russian, there are such languages as Tatar, Bashkir, Yakut, Buryat, Ossetian and many others , in connection with which the population of the respective republics speaks several languages; or in India, West Africa and New Guinea, where the inhabitants usually speak the local, regional and colonial language).
Bilingualism, therefore, is the functioning of two languages within the same society, whose members constantly use both languages in everyday life: at home, for example, they can speak one language, and easily switch to another at work or in a store. Many educated Africans living in cities speak the local language at home and public service use French or English.
The coexistence of languages within the same society (state) often leads to the fact that languages begin to differentiate functionally, resulting in a functional inequality of languages when one of them is used only in one area of communication, where the second language, as a rule, is not allowed. . This is how the phenomenon of functional diglossia arises (di ‘two’, glossa‘language’, i.e. literally "bilingual"). Diglossia is characterized by a whole set of features: 1) the functional distribution of languages leads to the fact that one of them is used in "high" areas and situations of communication (for example, in church, science, education), while the other is used in everyday communication or in some, strictly defined genres of writing (for example, in contracts, office work, advertising, etc.); 2) in the linguistic consciousness of the society, the language used in high spheres has a special prestige; 3) this language is a supra-ethnic language, i.e. it is not the native (mother) language for any ethnic group of the population; 4) mastery of this language is possible only in the process of special education, because naturally(i.e. in family and everyday communication) it is not transmitted. An example of such a functional diglossia is the situation in Muscovite Russia before the Petrine reforms, when two related languages - Old Russian and Church Slavonic were in a relationship of functional distribution: Church Slavonic was the “correct”, normalized language of the Russian Middle Ages (they spoke with God in this language, from Greek translated liturgical books), while in everyday life and in office work (for example, when inventorying property or making court decisions), Old Russian was used.
Language contacts often lead to the formation of so-called contact languages, which are auxiliary mixed languages with an extremely poor vocabulary and minimal, unsettled grammar. A contact language is the result of a failed attempt to learn the language of a neighbor, a communication partner, i.e. it is a language of interethnic communication, hybrid in its origin (since phonetics and most of vocabulary goes back to one of the contacting languages), limited in function (used most often as a language of trade in ports or markets). Among such intermediary languages, lingua franca and pidgins are distinguished.
Lingua franca (lingua franca ‘Frankish language’) is a trading language that developed in the Middle Ages in the Eastern Mediterranean based on French and Italian vocabulary and was used as a means of communication between Arab and Turkish merchants and Europeans. In modern sociolinguistics, this term has expanded its meaning and has come to mean any contact language in interethnic communication (for example, a simplified version of Swahili in East and Central Africa).
Pidgin (business ‘business’) is an oral language of trade and business contacts, which is based on a mixture of elements of one of the European languages (English, Dutch, Spanish, French, etc.) with elements of the native language. This language, as a rule, has a European vocabulary, and phonetics, word formation and grammar are native. The functional use of this language is limited only to business interethnic communication (an example of such a language is sea pidgin beachlamar on the English based: it was used on the islands of Oceania in the places of whaling camps and on the ships themselves, since the crews were recruited from oceanic sailors; another example - trade pidgin- language russenorsk established in the 19th century. and used by Norwegian fishermen and Russian merchants in the border regions: it has only 300 words and a rather simple grammar).
Sometimes these pidgin languages can expand their communicative functions and be used not only as a means of communication between natives and Europeans, but also as a means of communication between local ethnic groups in interethnic contacts.
This is how the Creole language arises, which gradually becomes the native language of a certain ethnic community. In this language, the vocabulary is expanding, the phonetic and grammatical structure is becoming more complicated, i.e. a pidgin language tends to become a natural language. An example of such a language is the French-based Creole languages of Fr. Haiti and about. Martinique, which became native to the vast majority of the population, as well as the Creole language that arose on the basis of English talk-pisin, one of national languages Papua New Guinea, which is a means of social communication between people who speak different languages, especially in cities; This is the main working language in Parliament and in public institutions, the language of the press, radio, television, and in Lately and schools traditionally taught in English.
Creole languages are an example of a true "mixed" language with its own substratum and superstratum elements. The study of them by scientists makes it possible to trace the formation and development of the grammatical system of the language, for all of them reveal an amazing structural similarity.
The question of the mixing of languages (in foreign linguistics this term usually does not have semantic differences from another - crossing) has come to the fore since the beginning of this century, although linguists of the last century, starting from W. Humboldt and J. Grimm, occasionally turned to it. I. A. Baudouin de Courtenay attached great importance to it. In the concept of G. Schuchardt and the linguists adjoining him, in the theoretical constructions of neolinguists, the mixing of languages takes the form of a methodological principle, since it turns out to be the driving force of all linguistic changes, the stimulus that forms languages. From these premises comes the conclusion about the mixed character of all languages.
Dedicated to this issue a large number of G. Schuchardt wrote: “Among all the problems that linguistics is currently dealing with, there is, perhaps, none as important as the problem of linguistic confusion.” And from the point of view of G. Schuchardt, such an assessment of this problem is understandable, since he believed that “the possibility of linguistic mixing knows no restrictions; it can lead to both maximum and minimum differences between languages.
Mixing can also take place during a permanent stay in the same territory, flowing in this case intensively and being carried out complicated ways". Emphasizing the special importance of mixing in the life of a language, the neolinguist J. Bonfante proclaims: “Thus, it can be argued (simplifying, of course, the actual state of things) that French is Latin + Germanic (Frankish); Spanish is Latin + Arabic; Italian is Latin + Greek and Osco-Umbrian; Romanian is Latin + Slavic; Czech is Slavic + German; Bulgarian is Slavic + Greek; Russian is Slavic + Finno-Ugric, etc.”
Crossing of languages occupied a special place in the theories of Acad. N. Ya. Marra. “Even in one of his works in 1914,” notes S. B. Bernstein in his article specially devoted to this issue, “N. Ya. this moment next and main theoretical problem.
Later, repeatedly returning to this question, he always spoke in the sense that all the languages of the globe are crossed languages and that the very process of crossing determines the real content of the development of any language. Here are a few quotes of this kind. “The fact is that according to the Japhetic theory there is not a single language, not a single people, not a single tribe (and there were none at their origin) simple, not mixed, or, in our terminology, not crossed.” "In the very emergence and, naturally, in the further creative development of languages, the main role is played by crossing." “Crossing is not an anomaly, but a normal way to explain the origin of species and even the so-called genetic relationship.”
In the theory of N. Ya. Marr, staged transformations played an important role, which suddenly, in the form of an explosion, changed the “quality” of the language. Mixing (or in this case, in the terminology of N. Ya. Marr, already crossing) created an impetus for such an explosive transformation of the language, moreover, according to N. Ya. Marr, as a result of crossing two linguistic "qualities" (i.e., simply speaking, two structurally different languages) a new “quality” arises (a structurally new language). Such theories, of course, could not find wide application in the practice of linguistic research, they required critical consideration; an attempt at such a consideration was made by Stalin during a discussion in 1950 in the work "Marxism and questions of linguistics",
“They say,” he wrote, “that numerous facts of the crossing of languages that have taken place in history give reason to assume that during crossing a new language is formed by an explosion, by a sudden transition from an old quality to a new quality. This is completely false.
The crossing of tongues cannot be regarded as a single act of a decisive blow, yielding its results within a few years. Crossing languages is a long process that continues for hundreds of years. Therefore, there can be no talk of any explosions here.
Further. It would be completely wrong to think that as a result of crossing, say, two languages, a new, third language is obtained, not similar to any of the crossed languages and qualitatively different from each of them. In fact, when crossing, one of the languages usually comes out victorious, retains its grammatical structure, retains its basic vocabulary and continues to develop according to the internal laws of its development, while the other language gradually loses its quality and gradually dies off.
Consequently, crossing does not give some new, third language, but preserves one of the languages, preserves its grammatical structure and basic vocabulary and gives it the opportunity to develop according to the internal laws of its development.
This speech, directed against the theory of N. Ya. Marr on the significance of crossing languages for the sudden transformation of their "qualities", contributed to a certain simplification of the very complex and multifaceted problem of mixing languages.
The processes of mixing play, of course, an enormous role in the life of languages, and in studying them it is equally important not to overestimate them as not to underestimate them. These processes take many forms, so reducing them to a single type does not give a correct idea of their actual essence and significance.
The processes of mixing languages can be considered in the frontal plan. In this case, we will deal with various types mixing (mutual influence) of languages. But these same processes can be studied in terms of individual aspects of languages. In this case, we will face the problem of the permeability of certain aspects or areas of the language (ie, its phonetic, grammatical and lexical systems). Let us turn to a consistent, consideration of the processes of mixing languages in the order indicated.
V.A. Zvegintsev. Essays on General Linguistics - Moscow, 1962
1. The concept of mixing languages is one of the most obscure in modern linguistics, so perhaps it should not be included among the linguistic concepts, as A. Meillet did (Bull. S. L., XIX, p. 106).1
Indeed, looking at some articles that deal with the question of mixing languages, we are inclined to think that the terms "Sprachmischung", "gemischte Sprache" were introduced only as a result of reaction to the well-known ideas of the last century, when the language was considered as a kind of organism and when people willingly spoke about the organic development of the language as the only legitimate one, as opposed to inorganic innovations, considered as diseases of the language. For the younger generation of linguists, this stage is already completely passed; however, we still remember what great importance was attached at one time to both the purity of the race and the purity of the language. True, the general public is still at the mercy of these high-sounding words.
Under such circumstances, it is not surprising that Schuchardt, in his large factual material, testifying to the influence of the Slavic language on German, on the one hand, and on the influence of the Slavic language on Italian, on the other,2 could argue that there is no language that would not be mixed, at least to a minimal extent, and it is quite clear that Baudouin de Courtenay was able to publish an article in 1901 (JMNP) entitled "On the mixed character of all languages."
Finally, we see that Wackernagel, in his interesting article "Sprachtausch und Sprachmischung" (Gotting. Nachr., Geschaftl. Mitt., 1904, S. 112), clearly says that in his exposition he only wanted to emphasize the changes in views that in his time in linguistics.
2. If you look closely at the facts given by various authors on the confusion of languages, you will notice that all or almost all of them can be divided into three categories (it goes without saying that, if considered from other points of view, one could come to other classifications):
1) Borrowings in the proper sense of the word, made by a given language from foreign languages.
2) Changes in this or that language, which he owes to the influence of a foreign language. Examples of such changes are numerous; it suffices to cite as an example the French haut, derived from the Latin altus, which derives its aspirated h from the Germanic synonym corresponding to the German hoch. The form of the French place name Eveque-mont is also the result of Germanic influence, cf. German Bischofsberg: in French we would expect Mont-Eveque (example taken from Wackernagel's article already mentioned). Wed also calques of Latin, German and Slavic, ultimately all made according to Greek models, such as conscientia, Gewissen, conscience and many others. etc. Wed. also the development of the use of the attributive genitive case in Russian under the influence of foreign languages, etc.
3) Facts resulting from insufficient mastery of any language. Everyday life replete with individual facts of this kind; but much more rare are facts of the same order that have acquired social significance, i.e., those errors of language that have become a generally recognized norm in a certain environment. Most often, due to the presence of a real norm of the acquired language, only more or less common mistakes remain. I could not give a completely convincing example of such a language, an example that I would be able to control myself. However, peculiar facts of this kind are numerous; it suffices to refer to the above-mentioned work of Schuchardt.
As for the numerous Creole and other dialects similar to them, they, however, also belong to this category, but with the proviso that the native speakers of the language that others strove to master, adapting it well to needs and opportunities, also participated in their formation. these latter (see Schuchardt's extremely important explanations on this point in his Die Sprache der Saramakkaneger in Surinam. Verh. d. K. Akad. v. Wet. te Amsterdam, Afd. Letterkunde. Nieuwe Reeks, Deel XIV, No. 6, 1914, pp. III et seq., which I know only from Hugo Schuchardt-Brevier).
3. It follows from this enumeration of facts that we have every right, in view of the fact that they all appear only where two languages are in direct contact, to unite them all under a common heading, giving it some name, for example, a mixture of languages \u003d Sprachmischung.
But there is hardly any benefit in this, since if the facts of the second category are in principle identical to the facts of the third, because they are often based on processes similar to those that take place within the same language, then borrowings in the proper sense of the word must its origin to a completely different process.
In any case, from the totality of these facts, apparently, nothing can be deduced that could shake the existing views on the connections possible between languages. Apparently, in all these cases there can be no doubt about what kind of language it is, within which certain changes have occurred, in one way or another caused by other languages. Windisch, in his article "Zur Theorie der Mischsprachen und Lehnworter" (B. d. K.-SGW Phil.-hist. Cl., B. 49, 1897, S. 113) points out that, no matter how much language is mixed , there is always one language that forms its basis.
So, perhaps it would be better to replace the term "mixing of languages" with the term "mutual influence of languages", which contains nothing in itself in relation to the facts described, while the word "mixture" suggests to some extent that both languages, being in direct contact, can equally participate in the formation of a new language.
4. However, this last conclusion can easily be arrived at by considering the facts of the "mutual influence of languages" from a different point of view than was done above. Especially when we are dealing with languages whose history is unknown to us. Analyzing such a language, one can sometimes state that its elements go back to different languages. As long as the number of its essential elements going back to one of these languages far exceeds the number of elements borrowed from any other language (but it may be less than the total number of all elements borrowed from these other languages), we state only borrowings and influences of foreign languages and we say that the language being studied is a continuation of the one that gave largest number elements. But if, by chance, it turns out that two languages have conveyed to one language or another an equal number of elements equally important in the ordinary use of the language, then we would be at a loss to say which of these languages is the continuation of the language being studied.
Perhaps it is this consideration that underlies the note on mixed languages Setala (bottom page 16 of his article "Zur Frage nach der Vermandschaft der finnisch-ugrischen und samojedischen Sprachen". Helsingfors, 1915).
Schuchhardt in his article "Zur methodischen Erforschung der Sprachverwandschaft" ("Revue Internationale des Etudes Basques", VI, 1912) writes: elements, we still would not know whether the former merged into the latter or vice versa, or whether both developed from one common basic language. In his article "Sprachverwandschaft" ("Sitzungsberichte der Akademie der Wiss.", Bd. XXXVII, Berlin, 1917, 8.526), Schuchhardt says in general: "Further, one should not start with the question: does language a belong to the language family A or not? We can never be limited to two possibilities in advance," and he compares languages to pictures, which give different images depending on where we look at them from. The very question of whether this or that element of the language is native or borrowed is not considered important by Schuchardt: "this distinction is both insignificant and cannot be made" (the first of the cited articles, p. 2 of a separate reprint).
All this shows us the concept of language confusion in a new light, assuming that a language can have multiple sources.
5. Meillet, in an article that appeared in 1914 in the journal "Scientia" (see now "Le probleme de la parente des langues" in his book "Linguistique hitorique et linguistique generale", 1921), rebelled with all his might against this point vision. He showed, with all his characteristic clearness, that we always have reason to ask ourselves what kind of language is the continuation of which is given language, in other words, look for a base language. The reason for this is that the phenomenon of the continuity of language, inaccurately called the kinship of languages, is a purely historical fact; it is based solely on the presence of the will of the speaker to use a certain language, either keeping it as unchanged as possible, or modifying it, or supplementing it with borrowed elements. enjoy. This is why Meillet disagrees with the expression "mixture of languages" as being suggestive of a language having two sources.
6. First of all, it seems to me that we have the right, without the risk of being suspected by Schuchardt of the materialization of language (see the already cited article "Sprachverwandschaft", the beginning of the note at the bottom of p. 522), to assert that languages in general form more or less separate systems ( at least in the normal case) and well felt as such by the speakers, which, of course, only comes to light on occasion. These systems may be subject to various changes under the influence of various kinds factors, but in no case are destroyed as a result. It follows from this that Meillet is quite right in allowing the continuity of the languages themselves, and not just their elements.
7. In addition, Meillet rightly asserts that anyone who wants to study the history of any language has to reckon with related languages, that is, that the very course of the history of a language is based on the sense of the continuity of the language in speakers. And all of this is in line with social entity language, since each language is the language of some more or less strictly limited social group.4 The sense of continuity of language increases or decreases in direct proportion to the self-awareness of the social group of which it is an organ. The weakening of ties within the group is one of the conditions for the complete disappearance of the sense of the continuity of the language, which ultimately I do not consider impossible, at least in principle (see below, 9, 15).
All the great historical descriptions of various languages, always regarded as national works, are based essentially on this sense of the continuity of language, but almost never take it into account, at least explicitly. However, it is more than likely that the acceleration of change that occurs in the course of the history of a language is always connected in some way with the weakening of social ties.
8. On the other hand, it seems to me that there are two circumstances on which Meillet did not dwell or on which he did not insist enough.
1) It may be of some interest to leave native speakers aside and consider only the history of all the elements of a language. A historical description thus composed would have several points of departure instead of one.5 This does not great advantage in the case when the language is clearly something unified; but if it had been profoundly influenced by other languages, then the overall picture would have benefited greatly from revealing the role of all these elements.
And this is all the more true because the speaker's sense of the continuity of language is mainly guided by the material side of the language. In my dialectological journeys, I have always observed that speakers are very inclined to establish the sound similarities of words and much less those similarities that belong to the field of semantics. It follows from this that the linguists themselves, under the hypnosis of the external side of linguistic signs, take less into account what Schuchardt calls the inner form (innere Form). Meanwhile, there are many languages in which "external form" and "internal form" go back to different languages, while in ordinary descriptions the external form always takes precedence over the internal, and thus that part of the language that goes back to the language that gave inner shape often remains in shadow.
2) Since the relationship of languages, based on the sense of the continuity of the language among the speakers, is recognized as a historical fact, it becomes obvious that it can be proved only by historical methods. Comparative linguistics may not have anything to do with it. In those cases where the language is clearly a single entity, this question is not difficult. But where we are dealing with a language that has heterogeneous elements in its composition, linguistic methods are insufficient. True, we have a number of cases in which we are able to use not only the linguistic method, but also the historical one, and it is quite possible to extract, by observing these cases, some rules of thumb; according to these rules, we are entitled to admit in certain cases the unattested historical fact of a sense of the continuity of a language developing in one direction or another; but these rules are too general and are only valid for languages with more or less the same structure.
9. Finally, can't we imagine such social conditions, under which the loss of a sense of the continuity of the language would be possible? Let's say we have two tribes the same value, but speaking different languages, who have lost all contact with kindred tribes and are forced to live together, forming one social group. Obviously, in this case, only the language, customs, etc., will remain from the social ties within each tribe. But since each member of the new group will be interested in being understood not only by his own, but also by representatives of another tribe, he will learn something -as the language of these latter. And since neither of these two "pure" languages will have advantages over the other and there will be no practical use in it, due to the complete weakening of social ties within each tribe, then only these poorly learned languages \u200b\u200bwill survive, which will be a mixture of both original languages, taken in different proportions. By eliminating everything too individual, and therefore difficult6 (for example, too complicated grammar), from this mixture they form a single language adapted to the needs of a new social group, a language that does not continue for speakers either of the two original languages.
The process would be the same as in the formation of Creole dialects, with the only difference that there really was specific language which one wanted to imitate, while in the example imagined above there would be little concern for imitating this or that language, in view of their equal social importance, and the decisive factor here would be only ease of understanding. All this does not aim at and should not in any way diminish the value of existing comparative grammars, but only admits that we may always be faced with a problem that we could not solve by means of our comparative methods; but not because there would be no correspondences that could be established, but because from these correspondences we could not infer a historical fact - the feeling of speakers that they continue this or that language.