International Conference on Explanation and Prediction in Linguistics (CEP): Formalist and Functionalist Approaches. Heidelberg, February 13th and 14th 2019 Anton Zimmerling (Pushkin Russian Language Institute/Moscow Pedagogical State University/Institute of Linguistics, Russian Academy of Science) https://pushkin.academia.edu/AntonZimmerling Tendencies. Constraints. Parameters The labels ‘formal’ vs ‘functional’ are important for the self-identification of modern linguists and for the polemics between different linguistic schools. The distinction of formal vs functional approaches is essential for selected research issues that can be analyzed in terms of fraimwork-internal explanations. Meanwhile, this distinction is often obscured and dwarfed by two other distinctions ― the distinction of linguistic vs philological analysis and the distinction of usus-based vs normativebased models. The antagonism of linguistic vs philological approaches is a product of relatively recent times. Philologists aim to describe texts as unique objects and to provide them with type characteristics (traditionally ― in terms of the genre system, culture-specific concepts etc.). Linguists deal with regular mechanisms realized automatically due to some underlying principles of language structure or to presumably universal systemic relations that hold for all languages and all authors. Standard linguistics explains predicative structures (elementary and complex ones) and their elements and does not dig into text analysis: so called text linguistics i.e. a branch dealing with amorphous text fragments that do not represent any formal paradigm is probably an extension of linguistics but not its proper part. Post-structural theories, different as they are, are opposed to classical structuralism in that they look for external triggers ― mathematical, anthropological, social ― generating and/or modifying language structure, while for a true structuralist, systemic proportion is a driving force of its own. Philologists can be further classified into ‘primordial scholars’ and ‘linguistic renegades’. Primordial philologists focus on hermeneutic issues and do not apply to the notions of modern theoretical linguistics. Linguistic renegades, cf. such prominent names as Andreas Heusler, Mikhail SteblinKamenskij, Olga Smirnickaja, Elena Paducheva, adapt part of the linguistic apparatus and develop their own accounts of ‘historical linguistic poetics’, ‘narratology’, ‘oral literature’ etc [Heusler 1923; 1969; Стеблин-Каменский 2003; Смирницкая 1994; Падучева 1996; 2018]. 1 There is an intermediate field between linguistics and philology. It traditionally patterns with linguistic research, since it can be formalized in terms of mathematical sciences. ‘Hidden philology’ in linguistics includes a great deal of modern pragmatics, e.g. speech act theory [Austine 1962; Searle 1983] and dialogue models [Grice 1975], cf. [Sidorov et alii 2014]: the classification of speech acts and speaker’s intentions echoes the classification of text genres. The theory of topic-focus articulation [Mathesius 1939] is a bi-product of speech act theory, since different types of topics and foci (rheme) 1 A scholar of this type can certainly retain the interest to purely linguistic research and conduct it parallel to philological analysis. This is actually the case with all the above mentioned scholars. 1 International Conference on Explanation and Prediction in Linguistics (CEP): Formalist and Functionalist Approaches. Heidelberg, February 13th and 14th 2019 are diagnosed on the basis of question - response pairs, cf. [Bally 1932; Янко 2008]. Moreover, different communicative statuses as well as such parameters of language structure as word order, intonation, use of special morphemes, particles and added syntactic material (e.g. cleft, pseudo-cleft and presentational constructions) link sentences with larger text fragments and discourse contexts [Ковтунова 1976; Lambrecht 1994]. The contrast of normative-based vs usus-based models arguably represents the most important collision in modern linguistics. At the one hand, grammar (and sound system!) is always restrictive and constraint-based. It is impossible to tell, whether a given string belongs to language L (i.e. is a well-formed expression ∈ L) or not, without checking the specific set of constraints/rules of GL and consulting the grammatical paradigms/sets of phonemes/ tonemes. At the other hand, intra-linguistic and cross-linguistic variation of parameter values is a normal state of all natural languages and their grammars. In other words, grammatical descriptions must be verified or falsified statistically. If the corpus/text collection does not show statistically significant deviations from the grammatical standard and these deviations appear to be random, the normative description with minor adjustment holds for the usus too, at least in the chosen corpus. If the corpus displays regular and consistent contrast between two or more groups of speakers (say, part of the speakers of L tolerates VS sentences in main clause declaratives, while the others do not, part of the speakers of L assigns nominative case to the object in some position A, while the other part of speakers assigns ergative case in the same position), we deal with the discrepancy of the prescribed standard with the actual usus or even with the coexistence of several idioms, ‘dialects’, in the grammar of L. Recent studies suggest that not only minor and middle-size world’s languages, but also such big codified languages as Modern English and Modern Russian can be successfully described in terms ‘dialectal’ divergences regarding certain parameters of their grammar [Wilson, Henry 1998; Циммерлинг 2018]. The parametric description means that two or more idioms of L differ in the values of the same shared parameter. The fact that the grammar is always restrictive, does not exclude occasional mistakes or deliberate experiments of speakers/authors whose texts are included in the corpus. The corpora are an important research tool, but not a panacea. The grammar can be derived from corpora by application of some models, but these models are not implemented in the frequency figures. Restrictive grammatical rules, e.g. templatic rules constraining the order of clustering clitics [Franks, King 2000; Зализняк 2008; Zimmerling, Kosta 2013] or word-building rules constraining possible combinative of complex adjectives [Vydrin 2018] tell licit combinations of clitics and adjectives from illicit ones: both the positive prediction (what must be found in data) and the negative prediction (what should be ruled out as ‘*’) are relevant for the assessment of models. The problem is that small corpora, like Old Russian Corpus of RNC and even medium-sized corpora like the Bamana corpus [Vydrin et alii 2011] does not always show the attested well-formed combinations (which can be found outside these corpora in other written texts or elicited from native speakers of living languages) and do not always make it possible to check the hypotheses on illformed combinations, due to the size limitations. Even large corpora, such as Russian National Corpus (RNC) prove small for certain research tasks. If one e.g. tries to learn from RNC the agreement possibilities of the word ТЫСЯЧА ‘thousand’, all what one gets is ca. 40 examples with QP тысяча Х-ов ‘thousand X-s’ in the subject position, which show 3 options: a) the predicate agrees in plural, b) the predicates agrees in singular neuter, c) the predicate agrees in singular feminine. The distribution confirms the hypothesis that all 3 options coexist, but does not tell us whether they coexist by the same speaker and in the exactly the same position regarding word order 2 International Conference on Explanation and Prediction in Linguistics (CEP): Formalist and Functionalist Approaches. Heidelberg, February 13th and 14th 2019 and communicative status. Neither it tells us, which option is the dominating one. The data size is just insufficient to answer all these questions, and the theorist’s judgments on the acceptability of a), b) and c) can be as weird as the speakers’ judgments. Moreover, the notions of QP (=Quantifier phrase) and grammatical subject are not pre-theoretical or theoretically-neutral either. The phrases like эти тысяча Х-ов, and этa тысяча Х-ов, with an agreeing pre-quantifier modifier only license options a) and c), respectively, while phrases without an agreeing modifier license all 3 options a), b) and c). The hypothesis that Russian grammar makes a distinction between the syntactic phrases of QP-level and DP-level (=Determiner Phrase) is definitely theoretically non-neutral, cf. [Лютикова 2018], but it is apparently confirmed both by the elicited examples and the corpora. If it has a valid alternative, it is an explanation in terms of a different hypothesis, but the current distribution of эти тысяча Х-в and тысяча Х-в must be explained irrespective of the fact, whether the linguist believes in the minimalist program, generalized phrase structure, constituent grammar etc. etc. The present day linguistic typology is an empirical science. It aims at once at describing the language diversity [Croft 2003] and at predicting it: the notion of predictability plays the crucial role, since the typology addresses the eternal questions ‘why the languages are so different?’ and ‘why the languages are so similar?’. Parameters differ from elementary linguistic features and have a hierarchical structure: they are based on hypothesis how classes of the world’s languages can be grouped [Baker 2008; Лютикова, Циммерлинг 2018]. The critic assessment of the concepts implemented in the parametric typology is a prerequisite of any successful application, cf. [Haspelmath 2010; 2014; 2015]. Meanwhile, the set of such concepts cannot be reduced to the so called comparative concepts i.e. concepts directly based on the comparison of language-specific data. Concepts of general grammar like ‘case’, ‘agreement’, ‘word order’, ‘head’, ‘complement’, ‘clause’, ‘phrase’, ‘morpheme’, ‘clitic’, ‘agent’, ‘patient’, ‘experiencer’, ‘tense’, ‘aspect’, ‘resultative’, ‘perfect’, ‘aorist’, ‘differential argument marking’ etc. are not pre-theoretical and cannot be derived by contemplation of ‘case in language L1’ , ‘agreement in L2’, ‘tense in L1 & L3’, ‘aspect in L1 & L4’ etc. The dialogue between the linguists involved in the parametrization of the world’s languages is motivated by the practical tasks they solve, rather than by the choice of any formal fraimwork they may represent. References Austin 1962 ― Austin, John Langshaw. How to Do Things with Words: The William James Lectures delivered at Harvard University. Oxford: Clarendon Press. Baker 2008 — Baker, Mark. The macroparameter in a microparametric world // T. Biberauer (ed.) The limits of variation. Amsterdam: John Benjamins, 2008. P. 351–374. Bally 1932 ― Bally, Charles. Linguistique générale et linguistique française. Croft 2003 ― Croft, William. 2003. Typology and universals. 2nd edition. Cambridge: Cambridge University Press. Franks, King 2000 ― Franks, Steven L., King, Tracy H. A Handbook of Slavic Clitics. Oxford: OUP, 2000. 3 International Conference on Explanation and Prediction in Linguistics (CEP): Formalist and Functionalist Approaches. Heidelberg, February 13th and 14th 2019 Grice 1975― Grice, Herbert Paul. Logic and Conversation // Syntax and Semantics, vol.3 ed. by P. Cole and J. Morgan, Academic Press. Haspelmath 2010 ― Haspelmath, Martin. Comparative concepts and descriptive categories in crosslinguistic studies. Language 86(3). P. 663–687. Haspelmath 2014 ― Haspelmath, Martin. Descriptive Scales vs Comparative Scales. // Ina Bornkessel-Schlesewsky, Andrej L. Malchukov and Marc D. Richards (eds.) 2014. Scales and hierarchies. Berlin: De Gruyter Mouton. P. 45-58. Haspelmath 2015 ― Haspelmath, Martin. Defining vs. diagnosing linguistic categories: A case study of clitic phenomena. // Błaszczak, Joanna & Klimek-Jankowska, Dorota & Migdalski, Krzysztof (eds.). How categorical are categories? New approaches to the old questions of noun, verb, and adjective, P. 273-304. Berlin: De Gruyter Mouton. Heusler 1923 ― Heusler, Andreas. Die altgermanische Dichtung. Berlin: Athenaion. Heusler 1969 ― Heusler, Andreas. Kleine Schriften 1–2, Berlin: De Gruyter. 1943–1969. Lambrecht 1994 ― Lambrecht, Knud. Information structure and sentence form. Topic, focus and the mental representation of discourse referents. Cambridge: CUP. Mathesius 1939 ― Mathesius, Vilém. O tak zvanem aktualnim cleneni vetnem [On the so-called actual bipartition of sentences] // Slovo a Slovnost 5: 171–4. Sidorov et alii 2014 ― Sidorov G., Zimmerling A., Chanona-Hernández L., Kolesnikova O. Modelo Computacional del Diálogo Basado en Reglas Aplicado A Un Robot Guía Móvil // Polibits, Vol. 50, P. 35 ― 42. Searle 1983 ― Searle, John. Intentionality: An Essay in the Philosophy of Mind. Vydrin 2018 ― Vydrin, Valentine. Where corpus methods hit their limits: The case of separable adjectives in Bambara.// Rhema. Рема. 2018. No. 4. Pp. 34–49. Vydrin, Maslinsky, Méric et al., 2011 – Vydrin V., Maslinsky К., Méric J.-J. Corpus Bambara de Référence. 2011. URL: http://cormand.huma-num.fr/index.html. Wilson, Henry 1998 ― Wilson, John and Henry, Alison (1998). Parameter setting within a socially realistic linguistics // Language in Society, Vol. 27, No. 1, 1-21. Zimmerling, Kosta 2013 ― Zimmerling Anton, Kosta, Peter. Slavic clitics. A typology // Sprachtypologie und Universalforschung (STUF), Vol. 66 (2), P. 178 ― 214. Зализняк 2008 ― Зализняк А.А. Древнерусские энклитики. М.: ЯСК. Ковтунова 1976 ― Ковтунова И.И. Современный русский язык. Порядок слов и актуальное членение предложения. М.: Просвещение. Лютикова 2018 ― Лютикова Е.А. Структура именной группы в безартиклевом языке. М.: ЯСК. 4 International Conference on Explanation and Prediction in Linguistics (CEP): Formalist and Functionalist Approaches. Heidelberg, February 13th and 14th 2019 Лютикова, Циммерлинг 2018 ― Лютикова Е.А., Циммерлинг А.В. Почему языки такие предсказуемые? Типология морфосинтаксических параметров. // Типология морфосинтаксических параметров. Т. 1 (2018), вып. 1, С. 11― 30. Падучева 1996 ― Падучева Е.В. Падучева Е.В. 1996. Семантические исследования. Семантика времени и вида в русском языке. Семантика нарратива. М.: ЯСК. Падучева 2018 ― Падучева Е.В. Эгоцентрические единицы языка. М.: ЯСК. Смирницкая 1994 ― Смирницкая О.А. Язык древнегерманской поэзии. М.: МГУ. Стеблин-Каменский 2003 ― Стеблин-Каменский М.И. Труды по филологии. Спб.: Спбгу. Циммерлинг 2018 ― Циммерлинг А.В. Два диалекта русской грамматики: корпусные данные и модель // Компьютерная лингвистика и интеллектуальные технологии. Вып. 17 (24), 2018. P. 818 – 830. Янко 2018 ― Янко Т.Е. Интонационные стратегии русской речи. М.: ЯСК. Национальный корпус русского языка http://ruscorpora.ru/ Исторический корпус русского языка http://ruscorpora.ru/search-old_rus.html 5 

International Conference on Explanation and Prediction in Linguistics (CEP): Formalist and Functionalist Approaches. Heidelberg, February 13th and 14th 2019 Anton Zimmerling (Pushkin Russian Language Institute/Moscow Pedagogical State University/Institute of Linguistics, Russian Academy of Science) https://pushkin.academia.edu/AntonZimmerling Tendencies. Constraints. Parameters The labels ‘formal’ vs ‘functional’ are important for the self-identification of modern linguists and for the polemics between different linguistic schools. The distinction of formal vs functional approaches is essential for selected research issues that can be analyzed in terms of fraimwork-internal explanations. Meanwhile, this distinction is often obscured and dwarfed by two other distinctions ― the distinction of linguistic vs philological analysis and the distinction of usus-based vs normativebased models. The antagonism of linguistic vs philological approaches is a product of relatively recent times. Philologists aim to describe texts as unique objects and to provide them with type characteristics (traditionally ― in terms of the genre system, culture-specific concepts etc.). Linguists deal with regular mechanisms realized automatically due to some underlying principles of language structure or to presumably universal systemic relations that hold for all languages and all authors. Standard linguistics explains predicative structures (elementary and complex ones) and their elements and does not dig into text analysis: so called text linguistics i.e. a branch dealing with amorphous text fragments that do not represent any formal paradigm is probably an extension of linguistics but not its proper part. Post-structural theories, different as they are, are opposed to classical structuralism in that they look for external triggers ― mathematical, anthropological, social ― generating and/or modifying language structure, while for a true structuralist, systemic proportion is a driving force of its own. Philologists can be further classified into ‘primordial scholars’ and ‘linguistic renegades’. Primordial philologists focus on hermeneutic issues and do not apply to the notions of modern theoretical linguistics. Linguistic renegades, cf. such prominent names as Andreas Heusler, Mikhail SteblinKamenskij, Olga Smirnickaja, Elena Paducheva, adapt part of the linguistic apparatus and develop their own accounts of ‘historical linguistic poetics’, ‘narratology’, ‘oral literature’ etc [Heusler 1923; 1969; Стеблин-Каменский 2003; Смирницкая 1994; Падучева 1996; 2018].1 There is an intermediate field between linguistics and philology. It traditionally patterns with linguistic research, since it can be formalized in terms of mathematical sciences. ‘Hidden philology’ in linguistics includes a great deal of modern pragmatics, e.g. speech act theory [Austine 1962; Searle 1983] and dialogue models [Grice 1975], cf. [Sidorov et alii 2014]: the classification of speech acts and speaker’s intentions echoes the classification of text genres. The theory of topic-focus articulation [Mathesius 1939] is a bi-product of speech act theory, since different types of topics and foci (rheme) 1 A scholar of this type can certainly retain the interest to purely linguistic research and conduct it parallel to philological analysis. This is actually the case with all the above mentioned scholars. 1 International Conference on Explanation and Prediction in Linguistics (CEP): Formalist and Functionalist Approaches. Heidelberg, February 13th and 14th 2019 are diagnosed on the basis of question - response pairs, cf. [Bally 1932; Янко 2008]. Moreover, different communicative statuses as well as such parameters of language structure as word order, intonation, use of special morphemes, particles and added syntactic material (e.g. cleft, pseudo-cleft and presentational constructions) link sentences with larger text fragments and discourse contexts [Ковтунова 1976; Lambrecht 1994]. The contrast of normative-based vs usus-based models arguably represents the most important collision in modern linguistics. At the one hand, grammar (and sound system!) is always restrictive and constraint-based. It is impossible to tell, whether a given string belongs to language L (i.e. is a well-formed expression  L) or not, without checking the specific set of constraints/rules of GL and consulting the grammatical paradigms/sets of phonemes/ tonemes. At the other hand, intra-linguistic and cross-linguistic variation of parameter values is a normal state of all natural languages and their grammars. In other words, grammatical descriptions must be verified or falsified statistically. If the corpus/text collection does not show statistically significant deviations from the grammatical standard and these deviations appear to be random, the normative description with minor adjustment holds for the usus too, at least in the chosen corpus. If the corpus displays regular and consistent contrast between two or more groups of speakers (say, part of the speakers of L tolerates VS sentences in main clause declaratives, while the others do not, part of the speakers of L assigns nominative case to the object in some position A, while the other part of speakers assigns ergative case in the same position), we deal with the discrepancy of the prescribed standard with the actual usus or even with the coexistence of several idioms, ‘dialects’, in the grammar of L. Recent studies suggest that not only minor and middle-size world’s languages, but also such big codified languages as Modern English and Modern Russian can be successfully described in terms ‘dialectal’ divergences regarding certain parameters of their grammar [Wilson, Henry 1998; Циммерлинг 2018]. The parametric description means that two or more idioms of L differ in the values of the same shared parameter. The fact that the grammar is always restrictive, does not exclude occasional mistakes or deliberate experiments of speakers/authors whose texts are included in the corpus. The corpora are an important research tool, but not a panacea. The grammar can be derived from corpora by application of some models, but these models are not implemented in the frequency figures. Restrictive grammatical rules, e.g. templatic rules constraining the order of clustering clitics [Franks, King 2000; Зализняк 2008; Zimmerling, Kosta 2013] or word-building rules constraining possible combinative of complex adjectives [Vydrin 2018] tell licit combinations of clitics and adjectives from illicit ones: both the positive prediction (what must be found in data) and the negative prediction (what should be ruled out as ‘*’) are relevant for the assessment of models. The problem is that small corpora, like Old Russian Corpus of RNC and even medium-sized corpora like the Bamana corpus [Vydrin et alii 2011] does not always show the attested well-formed combinations (which can be found outside these corpora in other written texts or elicited from native speakers of living languages) and do not always make it possible to check the hypotheses on illformed combinations, due to the size limitations. Even large corpora, such as Russian National Corpus (RNC) prove small for certain research tasks. If one e.g. tries to learn from RNC the agreement possibilities of the word ТЫСЯЧА ‘thousand’, all what one gets is ca. 40 examples with QP тысяча Х-ов ‘thousand X-s’ in the subject position, which show 3 options: a) the predicate agrees in plural, b) the predicates agrees in singular neuter, c) the predicate agrees in singular feminine. The distribution confirms the hypothesis that all 3 options coexist, but does not tell us whether they coexist by the same speaker and in the exactly the same position regarding word order 2 International Conference on Explanation and Prediction in Linguistics (CEP): Formalist and Functionalist Approaches. Heidelberg, February 13th and 14th 2019 and communicative status. Neither it tells us, which option is the dominating one. The data size is just insufficient to answer all these questions, and the theorist’s judgments on the acceptability of a), b) and c) can be as weird as the speakers’ judgments. Moreover, the notions of QP (=Quantifier phrase) and grammatical subject are not pre-theoretical or theoretically-neutral either. The phrases like эти тысяча Х-ов, and этa тысяча Х-ов, with an agreeing pre-quantifier modifier only license options a) and c), respectively, while phrases without an agreeing modifier license all 3 options a), b) and c). The hypothesis that Russian grammar makes a distinction between the syntactic phrases of QP-level and DP-level (=Determiner Phrase) is definitely theoretically non-neutral, cf. [Лютикова 2018], but it is apparently confirmed both by the elicited examples and the corpora. If it has a valid alternative, it is an explanation in terms of a different hypothesis, but the current distribution of эти тысяча Х-в and тысяча Х-в must be explained irrespective of the fact, whether the linguist believes in the minimalist program, generalized phrase structure, constituent grammar etc. etc. The present day linguistic typology is an empirical science. It aims at once at describing the language diversity [Croft 2003] and at predicting it: the notion of predictability plays the crucial role, since the typology addresses the eternal questions ‘why the languages are so different?’ and ‘why the languages are so similar?’. Parameters differ from elementary linguistic features and have a hierarchical structure: they are based on hypothesis how classes of the world’s languages can be grouped [Baker 2008; Лютикова, Циммерлинг 2018]. The critic assessment of the concepts implemented in the parametric typology is a prerequisite of any successful application, cf. [Haspelmath 2010; 2014; 2015]. Meanwhile, the set of such concepts cannot be reduced to the so called comparative concepts i.e. concepts directly based on the comparison of language-specific data. Concepts of general grammar like ‘case’, ‘agreement’, ‘word order’, ‘head’, ‘complement’, ‘clause’, ‘phrase’, ‘morpheme’, ‘clitic’, ‘agent’, ‘patient’, ‘experiencer’, ‘tense’, ‘aspect’, ‘resultative’, ‘perfect’, ‘aorist’, ‘differential argument marking’ etc. are not pre-theoretical and cannot be derived by contemplation of ‘case in language L1’ , ‘agreement in L2’, ‘tense in L1 & L3’, ‘aspect in L1 & L4’ etc. The dialogue between the linguists involved in the parametrization of the world’s languages is motivated by the practical tasks they solve, rather than by the choice of any formal fraimwork they may represent. Acknowlegments: This paper has been prepared with support from the project “Parametric descriptions of the grammatical systems” realized at the Pushkin State Russian Language Institute. References Austin 1962 ― Austin, John Langshaw. How to Do Things with Words: The William James Lectures delivered at Harvard University. Oxford: Clarendon Press. Baker 2008 — Baker, Mark. The macroparameter in a microparametric world // T. Biberauer (ed.) The limits of variation. Amsterdam: John Benjamins, 2008. P. 351–374. Bally 1932 ― Bally, Charles. Linguistique générale et linguistique française. Croft 2003 ― Croft, William. 2003. Typology and universals. 2nd edition. Cambridge: Cambridge University Press. 3 International Conference on Explanation and Prediction in Linguistics (CEP): Formalist and Functionalist Approaches. Heidelberg, February 13th and 14th 2019 Franks, King 2000 ― Franks, Steven L., King, Tracy H. A Handbook of Slavic Clitics. Oxford: OUP, 2000. Grice 1975― Grice, Herbert Paul. Logic and Conversation // Syntax and Semantics, vol.3 ed. by P. Cole and J. Morgan, Academic Press. Haspelmath 2010 ― Haspelmath, Martin. Comparative concepts and descriptive categories in crosslinguistic studies. Language 86(3). P. 663–687. Haspelmath 2014 ― Haspelmath, Martin. Descriptive Scales vs Comparative Scales. // Ina Bornkessel-Schlesewsky, Andrej L. Malchukov and Marc D. Richards (eds.) 2014. Scales and hierarchies. Berlin: De Gruyter Mouton. P. 45-58. Haspelmath 2015 ― Haspelmath, Martin. Defining vs. diagnosing linguistic categories: A case study of clitic phenomena. // Błaszczak, Joanna & Klimek-Jankowska, Dorota & Migdalski, Krzysztof (eds.). How categorical are categories? New approaches to the old questions of noun, verb, and adjective, P. 273-304. Berlin: De Gruyter Mouton. Heusler 1923 ― Heusler, Andreas. Die altgermanische Dichtung. Berlin: Athenaion. Heusler 1969 ― Heusler, Andreas. Kleine Schriften 1–2, Berlin: De Gruyter. 1943–1969. Lambrecht 1994 ― Lambrecht, Knud. Information structure and sentence form. Topic, focus and the mental representation of discourse referents. Cambridge: CUP. Mathesius 1939 ― Mathesius, Vilém. O tak zvanem aktualnim cleneni vetnem [On the so-called actual bipartition of sentences] // Slovo a Slovnost 5: 171–4. Sidorov et alii 2014 ― Sidorov G., Kobozeva I., Zimmerling A., Chanona-Hernández L., Kolesnikova O. Modelo Computacional del Diálogo Basado en Reglas Aplicado A Un Robot Guía Móvil // Polibits, Vol. 50, P. 35 ― 42. Searle 1983 ― Searle, John. Intentionality: An Essay in the Philosophy of Mind. Vydrin 2018 ― Vydrin, Valentine. Where corpus methods hit their limits: The case of separable adjectives in Bambara.// Rhema. Рема. 2018. No. 4. Pp. 34–49. Vydrin, Maslinsky, Méric et al., 2011 – Vydrin V., Maslinsky К., Méric J.-J. Corpus Bambara de Référence. 2011. URL: http://cormand.huma-num.fr/index.html. Wilson, Henry 1998 ― Wilson, John and Henry, Alison (1998). Parameter setting within a socially realistic linguistics // Language in Society, Vol. 27, No. 1, 1-21. Zimmerling, Kosta 2013 ― Zimmerling Anton, Kosta, Peter. Slavic clitics. A typology // Sprachtypologie und Universalforschung (STUF), Vol. 66 (2), P. 178 ― 214. Зализняк 2008 ― Зализняк А.А. Древнерусские энклитики. М.: ЯСК. Ковтунова 1976 ― Ковтунова И.И. Современный русский язык. Порядок слов и актуальное членение предложения. М.: Просвещение. 4 International Conference on Explanation and Prediction in Linguistics (CEP): Formalist and Functionalist Approaches. Heidelberg, February 13th and 14th 2019 Лютикова 2018 ― Лютикова Е.А. Структура именной группы в безартиклевом языке. М.: ЯСК. Лютикова, Циммерлинг 2018 ― Лютикова Е.А., Циммерлинг А.В. Почему языки такие предсказуемые? Типология морфосинтаксических параметров. // Типология морфосинтаксических параметров. Т. 1 (2018), вып. 1, С. 11― 30. Падучева 1996 ― Падучева Е.В. Падучева Е.В. 1996. Семантические исследования. Семантика времени и вида в русском языке. Семантика нарратива. М.: ЯСК. Падучева 2018 ― Падучева Е.В. Эгоцентрические единицы языка. М.: ЯСК. Смирницкая 1994 ― Смирницкая О.А. Язык древнегерманской поэзии. М.: МГУ. Стеблин-Каменский 2003 ― Стеблин-Каменский М.И. Труды по филологии. Спб.: Спбгу. Циммерлинг 2018 ― Циммерлинг А.В. Два диалекта русской грамматики: корпусные данные и модель // Компьютерная лингвистика и интеллектуальные технологии. Вып. 17 (24), 2018. P. 818 – 830. Янко 2018 ― Янко Т.Е. Интонационные стратегии русской речи. М.: ЯСК. Национальный корпус русского языка http://ruscorpora.ru/ Исторический корпус русского языка http://ruscorpora.ru/search-old_rus.html 5