Reconstructing texts, reconstructing texts 1

Summary : A. Schleicher was so enthusiastic about the idea of the Indo-European languages reconstruction that in 1868 he even used it to write a fable. Inspired by his work, from the Schleicherian reconstruction of text in the usual sense of this word, some Moscow semioticians turned to the reconstruction of texts in one of the senses of this particular concept accepted in the Moscow-Tartu semiotic school. circle,

"In principle, in historical and comparative linguistics, it is possible to reconstruct both systems and texts. […] if it is possible to reconstruct a system, then a possibility to reconstruct a text must also exist […]" 2 .
In the contemporary sense of the word, the idea of reconstruction has spread relatively late in language sciences. In many respects, it is explained by the fact that the notion of language reconstruction presupposes the comparison of facts from different languages which are analyzed in a diachronic perspective; meanwhile, the idea of a historical method (if one speaks of the predominant "paradigms", of course) is comparatively recent in linguistics, like the idea of language comparison in general (which had arisen earlier). For instance, in the center of ancient "linguistic traditions", as a rule, was only one particular language (Chinese, Greek, etc.), whereas other languages did not attract the attention of scholars; their comparison with "central" languages did not predominate. When an interest in the comparison of languages appeared, in the beginning it was deprived of any historical dimension: for instance, in 1660, in the Port-Royal Grammar [Grammaire de Port-Royal], its authors Antoine Arnauld and Claude Lancelot compared French and Latin in a typological aspect, that is, as just two different languages and not as a language-ancestor and a languagedescendant 3 . Historical and comparative linguistics was born in the late 18th -early 19th century, and it changed the relations between linguistics and the adjoining fields (like language philosophy for instance) radically: it was then that linguistics acquired its own method.
Two factors favoured the development of historical and comparative linguistics at that time: Romanticism with its interest in history in general (including the history of particular languages) and, for the Europeans, the "official discovery" of Sanskrit in 1786 by William Jones. The following ideas spread out: first, the resemblance of languages could be explained by their common origin; second, by comparing the related languages one could reconstruct not only their evolution, but also their common languageancestor. Thus, the idea of language reconstruction arose.
Historians of linguistics more or less agree as to the year in which the idea of language reconstruction appeared: it was in 1816, when Franz Bopp published the work On the Conjugation System of Sanskrit in Comparison with that of Greek, Latin, Persian and Germanic [Über das Conjugationssystem der Sanskritsprache in Vergleichung mit jenem der griechischen, lateinischen, persischen und germanischen Sprache]. Establishing 2 Ivanov, Toporov 1963, рp. 88, 90. 3 Cf., for instance, Alpatov 1998, p. 51. correspondences between verbal endings in different languages, Bopp insisted particularly on their regular character.
Another important step forward in the development of the notion of historical and comparative reconstruction (more precisely, the transition from the reconstruction of languages to the reconstruciton of texts) was made by August Schleicher. Schleicher was so enthusiastic about the idea of the Indo-European languages reconstruction that in 1868 he even used it to write a fable called "The sheep and the horses" [Avis akvasas ka]. Below is its translation into English: "[On a hill,] a sheep that had no wool saw horses, one of them pulling a heavy wagon, one carrying a big load, and one carrying a man quickly. The sheep said to the horses: 'My heart pains me, seeing a man driving horses.' The horses said: 'Listen, sheep, our hearts pain us when we see this: a man, the master, makes the wool of the sheep into a warm garment for himself. And the sheep has no wool.' Having heard this, the sheep fled into the plain" 4 .
Since then, Schleicher's idea to compose a literary text in the reconstructed Indo-European proto-language has been criticized for the following reasons: first of all, it is impossible to speak about the reconstruction of any language as a whole, but only of its particular fragments. Secondly, even the reconstructed elements can go back to different epochs -especially if it is not possible to determine their relative chronology. Thirdly, possible dialectal differences in the reconstructed proto-language are usually not taken into account in the reconstruction. Besides, there exists a point of view that in the process of historical and comparative reconstruction, no proto-language as such is reconstructed, but only an artificial system of correspondences between the compared languages. Finally, in the case of Schleicher's reconstructed fable it was not clear why its content was precisely such. According to Tamaz Gamkrelidze and Vjačeslav Ivanov, "the procedure of reconstruction of a proto-language text [prajazykovoj tekst] by the application of rules determining the compatibility of particular words with each other (cf. the famous reconstruction of the Indo-European text -a 'fable' by Schleicher […]) does not provide any concrete real text in a protolanguage, but only presents a kind of illustration of rules of the words' compatibility, reflecting a particular state of knowledge about the Indo-European proto-language" 5 .
Nevertheless, in spite of the criticism, in 1939 Herman Hirt and Helmut Arntz "rewrote" Schleicher's fable anew, taking into account the achievements of the Indo-European linguistics after Schleicher's death (in 4 Beekes 1995(in 4 Beekes [2011]. 5 Gamkrelidze, Ivanov 1984, p. 832. the same 1868) 6 . Also, once again despite criticism, Schleicher's reconstructed fable became a source of inspiration for some Moscow semioticians (a number of whom were -and still are -professional linguists working on historical and comparative reconstruction) one hundred years after it had been composed.
This direction of work of the Moscow semiotic circle 7 was based on a number of principles, mainly that of semantic reconstruction; it was in many respects inherited by Moscow semioticians directly from Schleicher. In particular, in the work published in 1969 and written "for the centenary of Schleicher's death", Vladimir Toporov expresses his conviction that during over a hundred years of linguistic reconstructions, those of the forms of the proto-language (its phonetics, morphology and even syntax) have worked themselves out: "After a long period of disappointment as to the possibility to reconstruct ancient Indo-European ('proto-language', in Schleicher's terms) texts, some prerequisites spring up for a more sober and, at the same time, for a more optimistic view of the possibility of such a reconstruction. It is clear that the contemporary view of this possibility and of the appropriateness of such a reconstruction (including the question about the degree of reliability of the reconstruction of different text levels) differs considerably from the one which was accepted a hundred years ago. […] Now there are first of all phonetic and morphological levels which present the biggest difficulties for the reconstruction. At the same time, within every particular Indo-European tradition are brought to light, to a greater or lesser extent of reliability, specific texts and formulae which can be considered as late examples of the verbal art of the Indo-European epoch. To Schleicher […], the phonetic (and morphological) part of the matter seemed the most reliable, he was ready to be responsible for them before the verdict of the historical and comparative linguistics of his time. The problem of the content of text in its proto-language state did not interest him in principle and he did not even ask any questions about the foundations as to the choice of the reconstructed text with this particular content" 8 .
According to Toporov, it is necessary to shift the focus from the reconstruction of language form to the reconstruction of semantics: "The progress of particular philologies (Indo-Iranian, Hittite-Luwian, ancient Greek, Celtic, Germanic, Balto-Slavic, etc.) and the creation of reliable bases for the reconstruction of a number of Indo-European metric systems together with the ever deepening approach to the problem of reconstrucion of cultural phenomena reflected in the language, make appropriate the launch of particular attempts to reconstruct the ancient Indo-European texts. In this connection, semantics acquires a particular importance. Explicitly or implicitly, one should start precisely with semantics, and sometimes restrict oneself to 6 Hirt, Arntz 1939; cf. Toporov 1969, р. 9. Besides, there were several other attempts to "rewrite" Schleicher's fable anew. Further in his work, Toporov gives "several particular exemples of reconstruction which can be traced to a more ancient period than the existence of independent Indo-European cultural and linguistic traditions" 10 : these reconstructions are based on the texts of incantations. For instance, comparing the Slavic, Baltic, ancient Indian and Germanic incantation formulae of the type "bone to bone…" (which had been known to researchers for a long time already), Toporov emphasizes that such formulae also exist "far beyond the Indo-European tradition" and refers, in particular, to the "exact Estonian parallel" 11 . However, according to him, "the extension of a certain cultural phenomenon beyond a particular tradition cannot be considered as a foundation for not trying to reconstruct it as applied to the ancient state of this tradition, if only inside the latter there exist conditions for such a reconstruction" 12 . Here, there is an evident difference in the approaches to the reconstruction of linguistic and "cultural" phenomena.
Insisting on the priority of semantic reconstructions, Toporov nonetheless undertakes the first reconstruction (the "bone to bone…" formula) also on the formal level. Namely, according to him, semantics of the corresponding Indo-European formula concerns "putting in touch, joining of one part of body with another one, having the same name, with the purpose to heal the damaged part of body" 13 . As to the formal reconstruction, at the beginning the researcher confines himself to the most general conclusions: he says, for instance, that the "substantives designating various parts of the body are joined together immediately, that is, by particular case forms, or by prepositions expressing rapprochement or joining" 14 and he establishes the "use, in connection with the names of these parts of the body, of the verbs like to gather, to join, to couple, to sew together and so on" 15 . However, later on, Toporov specifies that on condition that there are no doubts "in the ancient and non arbitrary [motivirovann(yj)] character […] of phraseology within one particular tradition", it is also possible to reconstruct the corresponding text on a "language level" 16 . In doing so, Toporov proposes 17 a notation of the reconstructed Indo-9 Ibid., p. 10. 10 Ibid.
European formula on the "phono-morphological level" 18 -with reserve that in his reconstruction "some forms are traced not to the 'Indo-European' level […], first, and not to the stems […], secondly" 19 (besides, according to him, an "important complication" can be constituted by the fact that, for instance, in case of the reconstruction of the Indo-European formula "bone to bone…" "for many parts of body it is impossible to reconstruct the only one variant of name which would have an obvious advantage over the other ones" 20 ).
However, already in his posterior reconstructions based on the texts of incantations and cited in the same article, Toporov pays much less attention to the reconstruction of language forms as such. For instance, speaking about the formulae sending away diseases in the Indo-European incantations, Toporov only states that "as to the texts of incantations sending diseases away, the resemblance between Slavic and other Indo-European traditions concerns the whole sequence of semantic and syntactic units" 21 . A similar reconstruction of a "prototype-text" only on the "semantico-syntactic level" is also discussed in the analysis of the Indo-European incantations against worms ("the process of their consecutive reduction -destruction" 22 ): according to the researcher, in this case in general "there is still no necessity [sic. -Е.V.] to bring the proposed reconstruction to the phono-morphological level, even if the latter can probably be achieved with a certain probability, at least in certain key elements of the reconstructed archetype" 23 .
Finally, in the closing part of his work, Toporov gives up even the reconstruction of "semantico-syntactic units", only proposing some "material for the reconstruction of the ancient Indo-European love incantations" and reconstructing the "motives of the prototype of the love incantation" 24 . Therefore this time, in accordance with his scientific credo stated in the beginning of his article and quoted above, Toporov begins the reconstruction with semantics, to which he limits his analysis.
To define the level of reconstruction in Toporov's work of 1969 more precisely, let us refer to the preceeding research by Toporov and Ivanov (which is mentioned in Toporov's article in 1969) about the "theoretical questions of reconstruction" 25 . In this article, using the general model of communication system elaborated by Claude Shannon, Ivanov and Toporov determine the reconstruction as "that of the message М, transfered from А' and received in A'' after an interval of time t" 26 . According to the authors, all kinds of reconstruction in natural languages can be reduced to the following multilevel scheme of textual transcoding: General idea of text → Level of large semantic "units" → Syntactico-semantic structure of sentence → Level of words → Level of morphemes → Level of phonemic groups (syllables) → Level of phonemes 27 . If Schleicher, reconstructing the fable in the Indo-European proto-language, was "responsible" for the last five levels, in the final examples of his work of 1969, Toporov limited the reconstruction to the level of "large semantic 'units'".
The same peculiarity -the limitation of reconstruction mainly (though with some exceptions, as we shall see later) on the level of "large semantic 'units'" -is typical for the works of Moscow semioticians in the field of semantic reconstruction of the so-called "basic myth" of the Indo-European mythology 28 .
Beside Schleicher's (evidently reinterpreted) heritage 29 , among other theoretical premises allowing the Moscow semioticians to reconstruct ancient mythological texts, was the idea of structuredness of the space of text in general, and the idea of the bilateral nature of sign of any length, together with the syntagmatic direction of research in the Moscow semiotic circle. If the orientation towards paradigmatics presupposes, in principle, a complete description of the material of study, a syntagmatic approach allows to describe only a part of it, keeping some lacunas -in particular, in the reconstructed texts. These gaps can be filled with time, in analyzing new texts 30 .
According to Nikolaeva, another important theoretical premise for the Moscow semioticians was the thesis that language signs in texts can be connected with other textual units on the levels both of semantics and form: for instance, the pagan god Perun can be semantically connected with Elijah the Biblical prophet riding around the sky on a thundering 26 Ibid., p. 46. 27 Ibid., p. 48. 28 According to Tat'jana Nikolaeva, these works constitute the strongest aspect of researches in the Moscow semiotic circle and have "no equivalent in the world semiotics" (Nikolaeva 1997, р. XXV). In addition to this "global" reconstruction which concerns immediately the "basic" myth, many other reconstructions have been done in the Moscow semiotic circle (cf. Ivanov 1987;Toporov 1986;1987;Nevskaja 1987;1988;1993а;1993b, etc.: in the present article, it is impossible to enumerate all the works of Moscow semioticians [Vjačeslav Ivanov, Vladimir Toporov, Tat'jana Civ'jan, Tamara Sudnik, Tat'jana Svešnikova, Lidija Nevskaja, etc.], dealing with this problematics: there are dozens of them. That is why, here and further, we shall limit the enumeration to a few examples only). They are however considered by their authors as particular cases of reconstruction of the same "basic" myth. 29 Of course, it is hardly possible to reduce all historical theoretical premises of the Moscow semioticians' works only to Schleicher's heritage, and we insist particularly on their reception of Schleicher's ideas in this article, supposing the comparison of several approaches to the historical and comparative reconstruction in linguistics. 30 Nikolaeva 1997, рp. XXV-XXVI. chariot, while "formally" Perun is associated with textual units having similar forms: for instance, in Russian, petruška 'parsley', perec 'pepper', etc. 31 Semantics of textual units (let us remember that in their reconstructions, Moscow semioticians gave priority precisely to semantics), in accordance with such an interpretation, can depend directly on their forms, because as it happens, "formal" connections finally also lead to semantic ones (parsley and pepper are considered as "Perun's plants", etc.). It means that, in contrast to one of the basic theses of the Saussurean Course in General Linguistics, language sign is considered to be non arbitrary. In addition (this principle is connected with the previous one), a border disappears between proper and common names as constituting two (usually) opposite ways to designate reality: individualization (proper names) and generalization (common names).
The following is, in its general outline, the "basic" myth of the Indo-European mythology reconstructed by the Moscow semioticians (they wrote much about it beginning the late 1960s) 32 -though the corresponding work of reconstruction has not been finished yet 33 and neither have the different parts of the plot been reconstructed so well 34 : 31 Ibid., р. XXVI. 32 Toporov 1986, р. 48. By Indo-European mythology, in this case is obviously meant a hypothetic totality of myths (wide)spread among those who spoke the hypothetical Indo-European proto-language, which presupposes an implicit parallel of study of language and mythology. In the above analyzed article by Toporov about the Indo-European reconstructions (Toporov 1969) some attempts to reconstruct the texts within the "ancient Slavic tradition" (Ivanov, Toporov 1963;1965;cf. also Ivanov, Toporov 1983b) are also mentioned. In these works, for the Slavic "level" some theses had been specified which later were transposed to the Indo-European "level" (cf. the article Toporov 1969). According to the authors, "primary attention to the reconstruction of higher levels is explained by the fact that, owing to the peculiarities of the evolution from the proto-Slavic language to the particular Slavic ones (cf. the insignificant time interval separating the proto-Slavic language from the first manuscripts written in particular Slavic languages, and the still remaining proximity of Slavic languages, facilitating the reconstruction of more ancient stages in the language evolution), the reconstruction of text on the inferior levels (for instance, phonological and morpholgical) does not present any particular difficulties. At the same time, a sharp break in the cultural history of all Slavic peoples in the period from the existence of late proto-Slavic language till the appearance of early Slavic cultures (fixed in the written sources after the christianization) makes particularly difficult the problem of reconstruction of the content of text, that is, of the semantic message, the rules of encoding of which on inferiour levels have already been described before. Hence the necessity to study and to reconstruct the texts semantics in the large sense, supposing the reconstruction of the ancient model of the world" (Ivanov, Toporov 1965, pp. 5-6). Besides, for a series of Moscow semioticians' researches, it is a common tendency to give particular examples of reconstruction done on the semantic level only. 33 Is it, in general, possible to imagine this work finished? It seems not: orientation towards the syntagmatic approach, filling the lacunas (and discovery of new ones) can in principle continue endlessly (in connection with this, cf. also below about the infinite number of sources of reconstruction, for the Moscow semioticians). 34 Nikolaeva 1997, р. XXVII. "А. The God of Thunder is located on high, in particular, on a mountain (together with the Sun and the Moon), near the top of the tripartite world tree pointing to the four corners of the earth. В. The Serpent is located underneath, near the roots of the tripartite world tree, on the black wool. С. The Serpent steals the oxen (hidding it in a cave behind a rock); the God of Thunder breaks the rock and liberates the oxen (or people). D. The Serpent hides […] by taking the appearance of different living beings, one after another (a man, a horse, a cow, etc.); the Serpent hides under a tree or under a stone. E. The God of Thunder, on a horse or in a chariot, strikes [the] tree with his arm (a hammer, a lightning) and burns it, or he strikes the stone and splits it. F. After the victory of the God of Thunder, water appears over the Serpent (it rains); the Serpent hides himself in the earth's waters" 35 .
For the majority of their reconstructions, Moscow semioticians use mainly folklore texts, but from the point of view of their importance for semantic reconstruction, all folklore genres are not equivalent for them: some genres are more valuable than others, for instance, the texts of incantations and charms, which can be explained by several reasons. In particular, "the reconstruction of proto-Slavic texts of incantations presents considerable advantages because of two circumstances -their simple structures and the limited character of the fields they concern" 36 . Besides, the texts of incantations are usually considered to be very ancient ones: already their sacral character seems to prevent them from any changes (to change only one word in incantations would be like changing the whole texts, depriving them of their magic potential). That is why, in the research of 1969, Toporov carries out semantic reconstruction using precisely the texts of incantations, and it is not a coincidence that some Moscow semioticians working on semantic reconstruction have studied and still study incantations 37 . However in the reconstruction of the "basic" myth, Moscow semioticians do not use the texts of incantations only: the number of potential sources (including the verbal ones) seems to them, in principle, infinite and, in addition, endlessly diverse. According to Toporov, the "basic" myth 35 Ibid., р. XXVIII. As Toporov emphasizes (cf., for instance, Toporov 1981, р. 274), in the most detailed way the totality of motives constituting the Indo-European versions of the "basic" myth's plot has been expounded in the work Ivanov, Toporov 1974. In the process of reconstruction, much attention is paid to the analysis of "semantic relations between the elements of the plan of content" (Ivanov, Toporov 1965, р. 5; italics ours. -Е.V.), especially of binary oppositions -undoubtedly, in this tendency, the stucturalist orientation of the corresponding researches becomes evident. 36 Ivanov, Toporov 1963, р. 148. 37 Though even as concerns the most "archaic" folklore genres, according to Toporov, the problem of time here is much more complicated, and it would be erroneous to see in them "only a kind of preserve where archaisms are conserved" (Toporov 1986, р. 37; in this casein the article of 1986 -it was the question of Lettish popular songs which conserve many relics of the "Indo-European" epoch, according to the researcher).
"is fixed not only in the proper mythological texts, but also in a considerably larger group of texts (by the way, also in those which, unfortunately, are not always taken into account because they supposedly satisfy none of the two indispensable conditions -'to be text' and 'to be mythological')" 38 .
Hence, some elements important for the reconstruction can be kept even in "secondary" and "third-order" sources 39 .
It is precisely in this respect that one can speak about the semiotic orientation of the researches of Moscow scholars -if by semiotics we mean not only a science of signs, but also a "global science", a dialogue, if not a synthesis, of various disciplines. The semiotic approach to the reconstruction of ancient mythological texts presupposes, above all, the analysis of most various codes -not only of verbal texts, but also, for instance, of the elements of material culture, of archeological finds, etc.
Some theoretical premises of the semiotic approach to the reconstrucion of the "basic" myth can be found in the above-mentioned work by Ivanov and Toporov about the theoretical problems of reconstruction, published in 1966 40 . As we have noted before, the work justifies the theoretical possibility of a reconstruction on the semantic level only. Additionally, according to its authors, "the necessity of particular operations for the reconstruction is explained, first of all, by the demands of encoding and decoding, because of which the sequence of physical signals as such still does not allow to perceive the communicated message; secondly, these signals take place in time with some 'noise' imposed: the latter can be withdrawn, in particular, owing to the existence of 'noise'-immune codes, of codes allowing to find out and to correct mistakes […], for instance one same message can be transfered by several channels of communication, or several times, successively, by one same channel, etc. In such cases, the reconstruction of the initial message is facilitated by the comparison of a series of its representations" 41 . 38 Toporov 1981, р. 275. Cf. in this connection about the transformation of the notion of text (Ivanov, Toporov 1965, р. 7). 39 Toporov 1986, р. 48. 40 However, the problem of reconstruction had also interested Ivanov and Toporov before: according to the researchers, the "general theoretical theses" of their article "correspond to the purposes" of their earlier work, pubished in 1963(Ivanov, Toporov 1963. As to their work of 1966, as particular examples of reconstructions, they analyse decoding of unknown written languages, reconstruction in historical and comparative linguistics, translation, reconstruction of lost texts and of "texts which are initial for a whole group of presentations of most various texts" (Ivanov, Toporov 1966). In other cases, the content of the notion of reconstruction in the works of Moscow researchers is expressly specified in a more detailed manner, which is confirmed, for instance, by a whole series of specifications made in the article about the "reconstruction of ancien Prussian" (Palmajtis, Toporov 1984). 41 Ivanov , Toporov 1966, Toporov [1997. In the final part of their work, outlining the "further applications" of the idea of reconstruction, Ivanov and Toporov propose -in particular, in the reconstruction of higher levels of text structures -to rely on the "pattern", "standard" modes of the texts' organisation (ibid., р. 70): as we have stressed before, Toporov also did it, analyzing the formulae repeated in the incantations of various traditions. Speaking more concretely concerning the "basic" myth, "as the results of the reconstruction show, the basic myth is variable, in principle -both in terms of content and in terms of expression. Of course, this variability has its origin, first, in the imperfection of the ways of transmitting information in time -in particular defects of the conserving mechanism and second, in the 'perturbing' role of the unequal interpretation of sense by those using the texts (and, consequently, interpreting them)" 42 .
Hence the "possibility of reciprocal [vzaimnyj] translations between various modelling semiotic systems and the existence of many systems of this kind in any human collective make 'noise'-immune and reliable the transfer of the plan of content of messages concerning the culture of this collective" 43 .
For example, as concerns not only the reconstruction of the "basic" myth, but also the reconstruction of semantics of ancient (designated as "lost" and "initial") texts in general, according to Ivanov and Toporov, "the reconstruction of a text can be carried out on condition of either the existence of at least two of its representations (i.e. the texts transfering one same message), or the existence of its representation which, being heterogeneous, can be divided into two texts […]" 44 .
But of course, in order for the reconstruction to be complete, it is necessary to study not only two, but as many as possible versions of the "basic" myth which are available in the corresponding epoch: variability is considered as an indispensable consequence of evolution in time 45 ; it is considered that the "initial text (= archetype) is reconstructed by the bringing together of various code versions" 46 . In other words, the "basic elements of the studied myth reached us in the texts using different code systems". That is why, "it seems necessary to analyze various 'code' versions of this myth which would allow to determine the message itself, understood as an invariant scheme" 47 .
In general, the difference between the "basic" (for a particular tradition) myth and all the others consists in the fact that "it is precisely the basic myth (and this myth alone) which is transfered by the maximum number, in the mythopoetic tradition, of code versions (ideally, by all versions available in this tradition)" 48 -though in a number of cases the Moscow semioticians discussed the plots of the "basic" myth in connection 42 Toporov 1981, р. 274. 43 Ivanov, Toporov 1966 Ibid., р. 277. This maximum number of the code versions of the "basic" myth is directly connected with its completeness and with the finish of its plot (Ivanov, Toporov 1983b, р. 155). with other mythological plots, for instance, those about the creation of the world 49 .
Moreover, these versions can even be considered as important as the hypothetical initial text itself: "At the same time, the problem (so painful for researchers) of choice of the 'initial' version among the multitude of all the existing ones becomes much simpler in the light of the more and more firmly established point of view on the reality of the versions, and not of a certain 'initial' text, with respect to them (cf. the ideas of C. Lévi-Strauss and of certain other scholars who insisted, in particular, on the primary role of relations between the versions, rather than on that of particular texts 50 ). The 'initial' text is most often an imaginary construct, but from this it does not follow that it loses its heuristic value for the reconstructions, on the one hand, and that all the versions are of equal worth and that they don't need any stratification presupposing their reasonable valuation" 51 .
Among the different variants and "code" versions of the "basic" myth, there are, for instance, zoomorphic 52 , entomological 53 , vegetable 54 ones. Besides, the Moscow semioticians considered the "recoding" of certain elements of the myth as possible, their conversion from one code to others 55 .
However, the study of the basic plot and of the main characters of the "basic" myth represents only one level of semantic reconstruction; in addition to that, (in various Indo-European "traditions" -Slavic, Baltic, etc.) some minor characters, their different personifications, incarnations, behaviour, etc. 56 are reconstructed. The theoretical basis of such a reconstruction, once again, goes back to the early works of Ivanov and Toporov: "The comparison of two (or more) versions of one text is realized by their step-by-step division into the smaller and smaller fragments and by the establishment of correspondences between these fragments" 57 . Therefore, having reconstructed the general semantic structure of the "basic" myth, Moscow semioticians pass on to its smaller semantic fragments, to the specification of its more concrete details. 49 Sudnik, Civ'jan 1982, р. 146. 50 Cf. the final part of the footnote 35. -Е.V. 51 Toporov 1986, р. 45. 52 Cf., for instance, Sudnik, Civ'jan 1982. 53 Toporov 1981. 54 Sudnik, Civ'jan 1981Toporov 1977. Like in the previous cases, it is impossible to enumerate here all the works of the Moscow semioticians in which these versions are examined; that is why we limit ourselves to some examples only. 55 Sudnik, Civ'jan 1981;Toporov 1981, etc. 56 Once again, we shall limit ourselves to few references only: Ivanov, Toporov 1983а;Sudnik, Civ'jan 1980;1982;Toporov 1977;. 57 Ivanov, Toporov 1966].
At the same time, these researchers did not ignore the reconstruction of formal elements of the "basic" myth completely. In 1986, pointing out to the main achievements of the reconstruction of the myth, which began in the late 1960s, Toporov stressed that progress had been made in: 1) the reconstruction of the scheme of the plot, 2) the establishment of the language form of the key elements of this plot and of their concrete realisations [otraženija], 3) the detailed definition of the nature of the Slavic version of the "basic" myth, 4) the outlining of the basic traits of the myth's Baltic version, 5) the indicating of its connections with the Slavic version 58 . As to the second point, concerning the formal reconstruction, Toporov gives the example of reconstructed forms of names of the two main antagonists in the myth: the name of the God the Thunderer and the name of its principal enemy 59 .
Of course, it is impossible to "prove", in the positivist sense of the word, that the "basic" myth (if such a myth has ever existed) had precisely this particular semantic structure. As to the discussions about the existence of the only one "basic" myth, they are partly reminiscent of the linguists' debates about the mono-and polygenetic theories of language origin, which once again allows to return to the parallelism, implicitly established by the Moscow semioticians, of (the study of) language and folklore: had there once existed a unique language -ancestor of all ancient and modern languages? Had there once existed (at least, in the "Indo-European tradition") a unique "basic" myth, from which all other myths descended? Besides, even while accepting the monogenetic point of view, one can continue to discuss whether the real proto-myth is reconstructed (like a proto-language which has supposedly really existed) or whether it is a question of the creation of an artificial system of correspondences between various myths belonging to particular, more concrete traditions (like the artificial system of correspondences between cognate languages) 60 .
In reconstructing the "basic" myth, the Moscow semioticians obviously adhere to the monogenetic point of view -if not in linguistics, at 58 Toporov 1986, р. 49. The fundamental difference between the Slavic and the Baltic versions of the "basic" myth consists, according to Toporov, in the fact that the Slavic version presents almost entirely the result of a reconstruction, while the Baltic data constitutes a much more "alive" system, presented in its original mythopoetic locus (ibid.). Besides, of course, the corresponding "sub-versions", for instance, inside the "Baltic" version -the Lettish and the Lithuanian ones (ibid., р. 48), etc. can be compared with one another. 59 Ibid. At the same time, in the process of reconstruction, the proper noun as such can be considered as a "minimum […] text" (Ivanov, Toporov 1963, р. 120). 60 Cf. however the opinion of Ivanov and Toporov about the distinction, in this connection, between proto-Slavic and Indo-European reconstruction: "The provenance of Slavic languages from the dialects of proto-Slavic is beyond any doubt, which justifies the task of reconstruction of the real proto-Slavic text (in contrast to the conventional notation of the morphonological correspondences with reference to Indo-European)" (ibid., р. 91; italics ours. least in their study of mythology. However, anyhow, it is a question of finding a model allowing to look at many phenomena in a new way, in particular, in folklore (which has, for a long time, been considered by many researchers as a "receptacle of antiquities").
If, now, we go back to the question about the reception, by the Moscow semioticians, of the Schleicherian "paradigm" of historical and comparative reconstruction, would it be possible to maintain that they continue his line of research or, on the contrary, break with it? At first sight, the answer seems evident: indeed, the Moscow semioticians follow the "paradigm" of Schleicher who was the first to compose a whole text in the reconstructed Indo-European language. However the difference between the two approaches is also striking: first of all, unlike Schleicherian reconstructions, the majority of texts reconstructed by the Moscow semioticians are not only reconstructed on the semantic level par excellence, but they don't have any fixed formal basis in general. Therefore in the majority of cases the Moscow semioticians reconstructed texts which had no forms: only semantics was reconstructed. Let us remember that, according to Toporov, sometimes the reconstruction must be limited to semantics only: it is not surprising that many Indo-European texts, paradoxically though it may seem, can be reconstructed only to the 'sub-language' [pod''jazykovoj] level; in other words, in some cases the pretensions of those trying to reconstruct the text must be limited by the reconstruction of not that text as such [text understood in a usual sense of this word. -Е.V.], but of its sublanguage substrate [pod''jazykovoj substrat] -the real situation which could (or not) be reflected in the corresponding texts in this particular language" 61 .
Thus, the object of reconstructions of the Moscow semioticians turns out to be closer not to the text in the usual understanding of the word, but to the text in one of the terminological interpretations of the Moscow-Tartu semiotic school 62 . In this sense, the text of a particular object or phenomenon (text of Petersburg, text of space, text of shepherd, etc.) does not present any "objectivized" [ob''ektivirovannyj] description of a real object or phenomenon, but refers to their particular mental representation(s) 63 . As a semiotic concept, the text of an object or of a phenomenon is composed, first of all, after the reading of ordinary, primary texts where the corresponding object or phenomenon are discussed, and, secondly, after a 61 Toporov 1969, р. 10. 62 Cf. for instance Toporov 1984Toporov [1995; Civ'jan 1995, etc. A detailed comparison of the two corresponding directions of Moscow semioticains' work would require a particular study. Besides, the analysis of the notion of semantic reconstruction (which is, in general, different from the notion of semantic reconstruction in the works of researchers from the Moscow semiotic circle) in the book Gamkrelidze, Ivanov 1984 could also be of interest. 63 Cf. Nikolaeva 1997, р. XXXI. According to Nikolaeva, the Moscow semioticians (first of all, Toporov) were the first to introduce into the philological and semiotic language the general notion of text of X in its above-mentioned sense (ibid., р. XXXIX). generalization which follows the reading. In this way, speaking about the reconstruction of semantics, the Moscow semioticians obviously had in view semantics (sense) reflected in the most diverse forms and (that is why) often having no fixed forms at all 64 .
Therefore from the Schleicherian reconstruction of text in the usual sense of this word, the Moscow semioticians turned to the reconstruction of texts in one of the senses of this particular concept accepted in the Moscow-Tartu semiotic school. That is why, in this case one should speak rather about a break with the Scheicherian "paradigm", than about its direct continuation.
© Ekaterina Velmezova 64 Comparison of methods and results of semantic reconstruction outside any fixed language forms in the works of Moscow semioticians, on the one hand, and, on the other hand, in the works of pupils and successors of Nikolaj Marr could also constitute a subject of a particular research: in this connection, deserves attention the fact that the "semiotic" component of the "New theory of language" was important also for Marr and for his many followers (Velmezova 2007, p. 341, etc.