Lifetime Linguistic Inspirations

Foreword

We are happy to offer this collection of papers to Igor Mel’čuk, our mentor, colleague and friend, on the occasion of his 90^th birthday.

Since Igor Mel’čuk is a household name in linguistics, there is no need to expand on his numerous, trailblazing contributions to the discipline. Nor is it necessary to mention the influence (positive, of course) that the encounter with Igor’s work has had on many an aspiring linguist and seasoned researcher alike—for some of us, this has been a career-determining or even career-changing event. Therefore, it seems more fitting to simply express here our admiration and gratitude for the great scientist and great teacher that Igor is.

The Festschrift volume consists of 42 contributions by 55 authors from 13 countries (Australia, Canada, Czech Republic, Germany, Israel, Italy, Finland, France, Russia, South Korea, Spain, Switzerland, USA). The papers, written in Catalan, English, French, Russian and Spanish, treat data from these and some other languages (Arabic, BCS [Bosnian-Croatian-Serbian], Czech, Finnish, Hebrew, Korean, Upper Kuskokwim Athabaskan [Alaska], Shughni [Pamir]). They pertain to virtually all branches of “core” synchronic linguistics (with occasional excursions into diachrony): semantics, lexicology/lexicography, syntax and morphology. Some are more theoretically minded while others are focused on pedagogical and/or computational applications of language models, particularly those of Igor’s brainchild, the Meaning-Text model. The topics covered, many of which had been previously taken up by Igor, are as diverse: collocations and other types of phrasemes, lexical functions, syntactic dependencies, argument structure and grammatical voice, to mention just a few. A few papers explore topics linking linguistics with other disciplines: literary theory, philology, religious studies.

We should like to offer our heartfelt thanks to the many people who made the present volume possible: the colleagues who contributed the articles, the members of Editorial Board for selfless and diligent reviewing of the articles, Polina Mikhel for her expert assistance with the typesetting of the manuscript, Tilmann Reuther, Editorin-Chief of Wiener Slawistischer Almanach, and the publisher, Peter Lang Verlag.

Happy birthday, Igor! We hope you will enjoy the read.

Moscow, Halifax, Nancy
August 2022
Editors:
Leonid Iomdin
Jasmina Milićević
Alain Polguère

Devuška! Russian clausatives of café and restaurant interaction

Elena Akborisov
PLIDAM, INALCO
elena.akborisova@inalco.fr

Nikolay Chepurnykh
Université de Lorraine, CNRS, ATILF
chepurnikh@gmail.com

Tomara Gotkova
Université de Lorraine, CNRS, ATILF
tgotkova20@gmail.com

Lidia Kolzun
Université de Lorraine, CNRS, ATILF
lidia.kolzoun@gmail.com

Svetlana Krylosova
CREE, INALCO
svetlana.krylosova@inalco.fr

Polina Mikhel
Université de Lorraine, CNRS, ATILF
p.mikhel@gmail.com

Alain Polguère
Université de Lorraine, CNRS, ATILF
alain.polguere@univ-lorraine.fr

Abstract. We study a family of syntactically autonomous phraseological expressions that we term socializing clausatives. They play an important role in daily linguistic interaction and their mastering is crucial when acquiring a foreign language in order to smoothly participate in exchanges with native speakers. This paper focuses on Russian café and restaurant socializing clausatives examining their lexicographic modeling in the perspective of the teaching of Russian as a foreign language.

Keywords: socializing clausative, linguistic cliché, pragmateme, Russian Lexical Network (ru-LN), Russian as a foreign language

1. Objectives

Sentimentally, this paper is a token of affection and gratitude for Igor A. Mel’čuk from the LEC-ru team.¹ Scientifically, it deals with a subtype of non-free expressions – i.e. phrasemes – that have the four following characteristics:

C1 they are phraseological clausatives, which means that they syntactically behave as autonomous utterances;²

C2 they are essentially used in direct oral interaction;

C3 they are associated with prototypical daily-life contexts;

C4 they are so commonly used that their mastering is a key to a smooth linguistic social life and they should therefore be dealt with in foreign language teaching.

To terminologically highlight the importance we attribute to Characteristic 4, we call such expressions socializing clausatives. The Annex (p. 27) presents the list of socializing clausatives we analyzed in the context of the present study. In (1) and (2) below are two sets of illustrative examples, in English (a) with French (b) and Russian (c) counterparts. These examples are typical of café or restaurant interaction between the customer C and the waiter/waitress W (cf. Char. 2 and 3 above).3

Though socializing clausatives may be quite varied from a formal point of view – Cheers! (single lexeme) _≈ Here’s to you (full sentence) –, they all have to be handled together in language teaching as they fulfill the same type of communicational/social goal from the Speaker’s perspective. We are thus interested in examining how these expressions, in spite of their potential structural differences, can be uniformly modelled in lexicographic descriptions aimed at language learners and their teachers.

Our targeted language is contemporary Russian, in the perspective of the teaching of Russian as a foreign language. To narrow down our investigation, we focus on socializing clausatives that are specifically, or particularly relevant to the context of café and restaurant interaction: hereafter, café-restaurant clausatives. We believe that this lexical domain of the Russian language should be a focus of attention as it has experienced significant changes since the end of the Soviet Union. The patronizing of cafés and restaurants has become something quite ordinary in urban life over the past thirty years and, consequently, linguistic interaction in such places contains both old practices, that may have evolved (3), and new ones, that can be influenced by Western culture (4).

Comment 1. This clausative has become the norm in café-restaurant interaction – competing with Ščët, požalujsta! lit. ‘Check, please’ (N^o 37 in the Annex) – in spite of complaints from purists who argue that the pronominal verb rassčitat′sja (‘to pay one’s debt’) does not have a lexicalized corresponding transitive verb in dictionaries.

To sum up, eating, drinking, talking, reading, working, flirting in cafés and restaurants have become common facts in Russian daily life. It is important that language learners acquire a good command of current socializing clausatives used in these settings as they belong to the immediate needs for a successful communication in everyday situations.4 In the context of the present tribute to Igor Mel’čuk, the fact that he is not particularly fond of going out to cafés and restaurants is an additional reason for the naughty and contrary LEC-ru team to deal with this topic.

Our theoretical and descriptive framework is Explanatory Combinatorial Lexicology (Mel’čuk 2013: Chap. 11) from which we borrow most of our terminology. Important notions are signalled by italicised sans serif font, accompanied by either brief definitions or pointers to relevant publications. Due to lack of space, the discussion of theoretical notions is limited to the bare minimum, with a focus on descriptive considerations.

The remainder of the paper is structured as follows: formal diversity of socializing clausatives (Section 2); presentation of the set of clausatives collected for this study (Section 3); illustration of our lexicographic modeling with the two clausatives Vam povtorit′? and Devuška! (Section 4).

2. Lexical spectrum of socializing clausatives

Socializing clausatives form a semantically and structurally heterogeneous class of expressions; from this sole point of view, it is safe to say that the only, though very significant, property they share is that of being phraseological clausatives – cf. characteristic C1, Section 1, p. 11. All remaining common characteristics – C2 to C4 – are pragmatic in nature: they relate to the context of use, the social anchoring and the pedagogical relevance of these expressions. This section is dedicated to the identification of the parameters that have to be taken into consideration for the proper lexicographic treatment of socializing clausatives in the context of second language learning. To achieve this, we proceed in three steps: formal types of clausatives (2.1), criteria for identifying clausatives that are “true” pragmatemes (2.2) and axes of characterization that influence their lexicographic modeling (2.3).

2.1. Formal types

We identify in Table 1 three main formal types of clausatives based on their status as lexical entities. Each formal type is illustrated in the table with a given clausative together with the indication of the context of enunciation in which it is normally used by the Speaker.

As shown in Table 1 with Hello 2 – a specific sense of the vocable Hello that serves as prompt over the phone –, clausatives are not necessarily polylexical and a question arises: are they all phrasemes? We consider that monolexical clausatives are indeed phraseological as they are constituted of the non-free combination of several linguistic signs: a lexical sign and a prosody – i.e. a combination of suprasegmental signs expressing the communicative organization of the message. Hello 2 is interesting in that the lexical component Hello can combine with either an interrogative or an exclamative prosody: Hello? ∼ Hello! But even this choice is not free. It is odd to answer the phone with a flat, declarative Hello. It could be interpreted as a sign of depression, sadness or preoccupation. Note that the same observation applies to all types of orally realized clausatives: they are prosodically constrained.6

To sum up: (i) clausatives are either lexical units – lexemes or idioms – or semantically compositional expressions – linguistic clichés; (ii) they all are phrasemes (= non-free expressions), even clausative lexemes, because they “come” with a restricted prosody. However, the picture is more intricate as idioms and linguistic clichés can be constructed out of collocations in Hausmann (1979)’s sense, another formal type of phraseme. Here are two illustrations where lexical functions (Mel’čuk 1996; Mel’čuk and Polguère 2021) are used to elucidate collocational structures:

• the clausative idiom ⌜Sweet dreams⌝ ‘I wish you to sleep well’ is formally built out of the collocation sweat dream where dream is the base and sweet is the collocate expressing the lexical function application Bon(dream_(N));

• the cliché Have fun! is made up of a collocation with fun as base and have as support verb collocate expressing the lexical function application Oper₁(fun_(N)).

It should be stressed that these two phrasemes relate differently to the notion of collocation: because it is an idiom, ⌜Sweet dreams⌝ is “no longer” a collocation whereas Have fun! is, simultaneously, a full-fledged collocation and a linguistic cliché.

The leave-taking clausative (5a) – a linguistic cliché – is a pragmateme because it is meant to be used specifically at the end of a letter or email (ending with a coma and followed by a signature). It would be odd to part from someone on the street by saying Best regards to her/him. On the other hand, (5b) is not a pragmateme: it is what we call a general-purpose clausative as it can be used to say goodbye to someone in any context where it makes sense to say goodbye – even in contexts that are typical of the use of the pragmateme (5a).7

We adhere to the above notion of pragmateme, originally introduced by I. Mel’čuk (e.g. Mel’čuk 1996). However, we take this notion literally, which leads us to refuse the status of pragmateme to clausatives that are often classified as such. For instance, the linguistic cliché (6) is a general-purpose clausative in spite of appearances:

(6) Hold your fire!

All situational constraints on the use of (6) originate from its meaning, in particular: (i) it is an order expected to be used by a Speaker who is in a position of authority relative to the Addressee; (ii) it makes sense to use it only if the Addressee is currently firing a weapon. The Speaker can be a hunter talking to other hunters, an instructor at the firing range, etc. To know how to use (6), one only needs to know that it is the prototypical, unmarked clausative to use in a situation where someone intends to order others to stop firing their weapon. In this respect, (7) is a different type of clausative though it is synonymous to (6):

(7) Cease fire!

This linguistic cliché possesses the same meaning as (6), but the Speaker has to be a soldier talking to other soldiers. If used in the hunting situation, for instance, it is stylistically marked and the Speaker is “acting” as a soldier talking to subordinates.

In other words, for a constraint that applies to the use of a clausative to qualify as being pragmatic in nature – and for the corresponding expression to be a true pragmateme –, this constraint should not be inferable from the very meaning of the clausative. In our opinion, the semantic inferrability of the context is often overlooked ←15 | 16→with the consequence of diluting the very notion of pragmateme. In our study, we try to be more scrupulous because the classification of socializing clausatives as being pragmatemes or not has implications on their proper lexicographic modeling.

LEC-ru’s research is based on extensive lexicographic work (Krylosova 2017), in line with Explanatory Combinatorial Lexicology’s principles. Unlike for classical Russian lexicography conducted within this framework – that is based on the writing of dictionaries (Apresjan 2014 & 2017, Mel’čuk and Žolkovskij 20168) –, our model ←17 | 18→is non-textual and takes the form of a special type of lexical network called Lexical System (Polguère 2014). The most advanced Lexical System to date is the French Lexical Network (fr-LN) relative to which our Russian Lexical Network (ru-LN) can be considered as being in its infancy.9

From a formal point of view, the ru-LN – as all Lexical Systems – is a graph whose nodes are essentially lexical units of the language (lexemes and idioms) and arcs are essentially lexical function relations connecting these lexical units. Each node is itself a complex entity that incorporates lexicographic information associated with the corresponding lexical unit (grammatical characteristics, semantic content, etc.). The ru-LN deviates from these basic structuring principles in two ways.

1) Nodes of the graph can correspond to other lexical entities than just lexical units. In particular – and this is very important in the present context! – linguistic clichés (and similar lexical entities) are nodes of the ru-LN: they possess their own individual lexicographic description (more on this in 3.3 below).

2) Arcs of the graph are not lexical function relations only. In particular, the ru-LN encodes copolysemy relations that connect senses of polysemous vocables (Polguère 2018). Each vocable is modeled by a subgraph made up of lexical nodes connected by copolysemy relations (sense extension, metaphor, metonymy, etc.).

A parallel can be drawn between the structure of Explanatory Combinatorial Dictionaries or ECDs (Mel’čuk 2013: Chap. 11) and that of Lexical Systems such as the ru-LN: ECD entries for lexical units parallel lexical unit nodes of Lexical Systems together with the cluster of lexical function arcs they control; ECD super-entries for vocables parallel subgraphs of lexical unit nodes connected by copolysemy relations.

The main descriptive issue that has to be discussed is the formal status of caférestaurant clausatives (and other socializing clausatives) in the ru-LN and how their modeling is hooked up to the bulk of the lexical network. We focus first (3.3) on clausatives that are linguistic clichés, i.e. that are not lexical units. Later (3.4), we examine lexical relations controled by socializing clausatives in general.

3.3. Lexicographic articles for linguistic clichés

We said in the previous section that linguistic clichés possess their own entries in the ru-LN, as in any Lexical Systems. This can be justified as follows.

1) Each linguistic cliché is a linguistic entity that possesses a significant amount of individual properties (cf. 2.3 above). These properties have to be learned/known in order to use the cliché adequately. Additionally, for such complex lexical entity to be fully mastered, its description must contain sufficient exemplification (lexicographic examples), as it is the case for lexical units. Where can this information be provided in a lexical model, if not in an article dedicated to the cliché?

2) The fact that linguistic clichés, being semantically compositional expressions, do not require lexical definitions (⇒ are not lexical units) weighs little in making the choice to attribute them individual entries or not in light of the preceding observation on the density of information each cliché carries.

3) Because of points 1 and 2 above, embedding the description of clichés in the lexicographic article of lexical units – e.g. lexical units that are formally included in the clichés and/or contribute significantly to their meaning – is as faulty as embedding the description of idioms in lexemic articles like most traditional dictionaries do.

However, we shall not throw the baby out with the bathwater. The notion of lexical anchor is indeed relevant for information access in the Mental Lexicon. When thinking of a friend’s birthday, one may use the lexical unit birthday as lexical support for this conceptualization and try to access to “things that one says when it’s someone’s birthday.” Lexical anchors can help in some cases of lexical access, but there is no way dictionaries or other lexical models can encode all possible cases of lexical access originating from a conceptualization in the Speaker’s mind. Only two lexical anchors are envisaged here for Happy birthday! – birthday and wish_(N) –, but many others would be equally valid: anniversary, birthdate, birth_(N), born, ..., wish_(V), ... Lexical models have to focus on lexical relations, and lexical relations do exist, but they hold between linguistic clichés and lexical units that express specific elements of contents of these clichés, as illustrated in (9) above. The encoding of such links within the lexical network approach allows at the same time for an access to clichés from “anchoring” lexical units (or from fellow linguistic clichés that are paraphrases): this corresponds to upstream navigation in the lexical network, where one examines incoming rather than outgoing lexical relations. Such incoming relations can be displayed together with outgoing relations in “article-views” (textual visualizations) of lexical network nodes (Ollinger and Polguère 2020: 5).10

Semantic inclusion relations. Lexical meanings included in the meaning of a clausative are connecting points that integrate the clausative to the rest of the lexical network. If the clausative is a lexical unit, semantic inclusions of other lexical units are modeled by its lexicographic definition. For instance, the definition of the idiom ⌜Break a leg!⌝ includes the semanteme ‘[artist] performance’ and therefore points to the lexeme performance.11 The description of communicative intents includes lexical pointers as well. Such is the case for Gde možno ruki pomyt′? ‘Where can I wash my hands?’ – Ex. (8) in Section 2.3 and N^o 24 in Annex – whose communicative intents point to the lexeme tualet ‘toilet’. A concrete illustration is also provided below in the lexicographic article for Vam povtorit′? ‘Another round?’ (Section 4.1).

Lexical function relations. Both standard and non-standard lexical functions apply to socializing clausatives and connect them to other lexical entities in the ru-LN; this is illustrated in Section 4. The standard lexical functions we use (Syn, S₀, S_loc ...) do not need to be explained as their properties are well-known and can easily be found in the literature (e.g Mel’čuk 1996, Mel’čuk and Polguère 2021). Two non-standard lexical functions, on the other hand, play an important role in the description of socializing clausatives and have to be explained: Speaker and Addressee.

Clausatives denote acts of utterance production (cf. point 2, Section 2.3), which implies that they semantically encompass two participants: the Speaker and the Addressee. In the specific case of café-restaurant clausatives, these two participants are normally W (waiters) and C (customers), each of them in either the role of Speaker or Addressee. In a similar way lexical functions S₁, S₂ ... denote names for first, second ... actants, we use two non-standard lexical functions to point to names of (prototypical) Speakers and Addressees of clausatives when it is relevant to do so: Speaker – named Говорящий (Govorjaščij) in the ru-LN; Addressee – named Адресат (Adresat) in the ru-LN.

The Russian language offers lexicalizations for W: mainly, oficiant ‘waiter’ and ←20 | 21→oficiantka ‘waitress’ (see Table 3, Section 4.2 below).12 As in English however, there is no Russian term denoting C specifically for cafés and restaurants. Russian has nouns such as zavsegdataj or gost′ ‘patron’ that are used to designate the clientele of cafés and restaurants but are not restricted to it. The use of Speaker and Addressee in the ru-LN is therefore limited to pointing to the lexicalization of W.

Theoretical remark. Most café-restaurant clausatives are linguistic clichés. From a theoretical point of view, the application of (non-)standard lexical functions to other entities than lexemes or idioms (i.e. lexical units) is borderline. Nevertheless, we consider it acceptable because linguistic clichés are still lexical entities and it is not as if we were using lexical functions to connect “concepts” to the lexicon – which would be a true corruption of the notion of lexical function.

4. Illustration of lexicographic descriptions

Two clausatives are used below to illustrate our approach to lexicographic modeling: Vam povtorit′? (4.1) and Devuška! (4.2). They have been selected because they are diametrically opposed as regards to their formal nature (linguistic cliché vs. lexeme), their Speaker (W vs. C) and the tightness of their association with the café-restaurant context (pragmateme vs. general-purpose clausative).

Due to space limitation, we do not offer fully developed lexicographic articles; we focus on (i) grammatical characteristics, (ii) communicative intent, (iii) lexical function relations and (iv) lexicographic examples (a maximum of three). Articles of the ru-LN are formulated in Cyrillic Russian. To make the following illustrations easier to read by non-speakers of Russian, linguistic data they contain is transliterated (often accompanied with English gloses) and English is used as metalanguage.

Each illustration is in two parts: an introduction to characteristics of the clausative and the simplified lexicographic article proper. Taken together, these two components of the illustration can be used as “teaching card” in a pedagogical context.

4.1. Waiter’s phrasal pragmateme clausative (N^o 3 in Annex): Vam povtorit′?

The phrasal clausative Vam povtorit′? can be glossed as follows:

Vam povtorit′? is informally used by W to ask C if she/he wants an additional drink (or another serving of food, but this is less common). Note that W implies another complete serving, not just a refill. The implicit direct object of the verb povtorit′ – corresponding to the drink to be served again – can be explicitly expressed, as in (11) below. The clausative could therefore be named as Vam (N) povtorit′?

From the point of view of standard Russian, this use of the Vam N povtorit′? construction is incorrect because the direct object of the lexeme povtorit′ ‘to ←21 | 22→repeat’ normally denotes actions (‘to repeat doing something’) and therefore cannot denote names of drinks. We decided to take into consideration cases such as (11) because they are very common in café-restaurant interaction and are mentioned in phrasebooks/manuals for waiters.

Devuška! ‘Young lady!’ is a general-purpose clausative. Though it is not a pragmateme and can be used anywhere,17 it is emblematic of café-restaurant interaction, as a linguistic means for a customer C to attract the attention of a (rather young) waitress W. Singularly, it is the first clausative that came to mind of several members of the LEC-ru team when the topic of café-restaurant clausatives was originally discussed. The dialogue (12) below illustrates the use of Devuška! in a restaurant interaction; in this case, both W and C are women.

(12) Srazu že podošla oficiantka i, zabiraja tarelku, sprosila:

— Vam ponravilos′?

— Devuška, milaja, ponravilos′ – ne to slovo, – čestno priznalas′ ja. — Vaš povar volšebnik.

‘A waitress came immediately and, collecting the plate, asked:

— Did you like it?

— Young lady, my dear, to like is not the right word – I answered with sincerity. — Your chef is a magician.’¹⁸

The standardized expression Devuška! used to hail a young woman is built from the lexeme Devuška 2: a term of address (hence always in the vocative) derived from devuška 1 ‘young woman’, the basic lexical unit of the vocable devuška. Both senses have the same denotational content; they are in a relation of part of speech derivation corresponding to two reciprocal lexical function relations:

1. Claus(devuška 1) = Devuška 2

2. S₀(Devuška 2) = devuška 1

The orientation of the polysemy derivation devuška 1_→Devuška 2 is “standard:” a term of address sense derived from an non-clausative sense. However, the reverse derivation can also exist. For instance, the French term of address Madame 1 ‘Madam’ has derived a non-clausative sense madam 2 ‘woman’ that belongs to child language: Maman, la madame elle est méchante ! lit. ‘Mum, the madam [= woman], she is nasty!’ (madam 2 can be used by an adult as well, when talking to a child).

The relevance of Devuška! to café-restaurant interaction, in spite of the semantic vagueness of Devuška 2, may be explained by the fact that this latter fills a lexical gap among terms of address that C can use clausatively to attract the attention of W – cf. the gray cell in Table 3 below.

Table 3 – Derived Russian terms of address that can be used for hailing W

Table 3 shows that, for reasons unknown to us, the Russian language has not derived from oficiantka ‘waitress’ a corresponding term of address. Incidentally, ←23 | 24→this lexical gap is an illustration of the fact that such term of address is not systematically derivable. One can of course use clausatively oficiantka to attract the attention of a waitress, but it is an unconventional usage that will be perceived as rude and disrespectful, as in (13) with the juxtaposition of devuška and oficiantka as hailing clausatives, which conveys the impression that C is probably drunk:

‘— Young lady, waitress, – hailed one of them, – bring us another bottle of cognac!’19

Because Devuška! is “filling in” for the non-existent *Oficiantka! ‘Waitress!’, it may be considered as more relevant to café-restaurant interaction than its masculine counterpart Molodoj čelovek! ‘Young man!’ (N^o 28 in the Annex), this latter being in competition with the semantically specific Oficiant! ‘Waiter!’ (N^o 31 in the Annex).

There is debate among the LEC-ru team (see also, for instance, Krylov 2019) on whether the use of Devuška! is controversial and whether a usage note is required in its description. Our conclusion: it is the presence of the semanteme ‘[Addressee is] young’ in the meaning of the clausative that could contextually generate an undesirable effect. For example, for C to hail a “no-longer-so-young” W with Devuška! could be interpreted as bad behavior. Therefore, the social acceptability of using the clausative is purely contextual. The last example in the lexicographic description below is a good illustration of the fact that some waitresses may prefer to be addressed by their first name (which reflects an Americanization of practices in restaurants) rather than as devuška.

Ollinger Sandrine and Polguère Alain (2020): Guide de navigation Spiderlex (version du 19 mai 2020) [Guide to Spiderlex Browsing (version 19 May 2020)], tech. rep., Nancy, ATILF CNRS, downloadable from https://lexical-systems.atilf.fr/en/spiderlex-en/.

Ovejas Martín Vanesa (2022): “Pragmatemas en español: estudio teórico y experimental”, PhD thesis, Universidad Complutense de Madrid.

Pausé Marie-Sophie (2017): “Modelling French idioms in a lexical network,” Studi e Saggi Linguistici, 55:2, pp. 137–155.

Polguère Alain (2014): “From Writing Dictionaries to Weaving Lexical Networks,” International Journal of Lexicography, 27:4, pp. 396–418, doi: 10.1093/ijl/ecu017.

Polguère Alain (2016): “Il y a un traître par minou: le statut lexical des clichés linguistiques [The lexical status of linguistic clichés],” Corela [On line], 19.

Polguère Alain (2018): “A Lexicographic Approach to the Study of Copolysemy Relations,” Russian Journal of Linguistics, 22:4, pp. 788–820, doi: http://dx.doi.org/10.22363/23129182-2018-22-4, url: http://journals.rudn.ru/linguistics/issue/view/1142.

Academic lexical combinations for a writing tool: about their lexical nature

Margarita Alonso-Ramos
Universidade da Coruña
margarita.alonso@udc.es

Abstract. This paper focuses on recurrent strings of words that appear in Spanish academic discourse. We call them academic lexical combinations (ALC). With the purpose of building a tool that helps novice writers, named HARTA, we have compiled lists of ALCs that are of different nature: some can be classified as a type of phraseme, therefore, a lexical entity that has to be registered in the lexicon; however, others have an uncertain status as lexical entities, even though any writer recognizes them as not built on the fly. In this regard we discuss to what extent we can say that a given string of words is a lexical entity. Without being able to provide a definitive answer here, we lean towards a broad approach and we include in HARTA any recurrent string fulfilling a specific discourse function.

Keywords: academic discourse, phraseology, lexical entity, discourse function, writing tool.

Resumen. Este artículo se centra en cadenas recurrentes de palabras tal y como aparecen en el discurso académico a las que llamamos combinaciones léxicas académicas (CLA). Con el fin de crear una herramienta que ayude a los escritores noveles (llamada HARTA), hemos compilado listas de CLA de diferente naturaleza: algunas pueden ser clasificadas como un tipo de frasema, por lo tanto, una entidad léxica que tiene que ser registrada en el léxico; sin embargo, otras tienen un estatus vago como entidades léxicas, a pesar de que todo investigador las reconoce como no construidas sobre la marcha. A este respecto exploraremos hasta qué punto podemos decir que una cadena dada de palabras es una entidad léxica. Sin poder proporcionar una respuesta definitiva aquí, nos inclinamos por un enfoque amplio que nos lleva a incluir en HARTA toda cadena recurrente que cumpla una específica función discursiva.

Palabras clave: discurso académico, fraseología, entidad léxica, función discursiva, herramienta de ayuda a la redacción.

Introduction

The academic discourse swarms with recurrent strings of words that any novice writer has to learn if (s)he wants to be accepted in the academic community. The English literature on academic lexicon is huge: Biber et al. (2004), Ackerman and Chen (2013), Paquot (2010), Simpson-Vlach and Ellis (2010), Durrant (2016), to mention just a few titles. The academic lexicon is the one included in scientific papers coming from different disciplines. Therefore, it must be interdisciplinary and exclude domain-specific terminology. Although there is some debate in the literature ←29 | 30→as to whether this interdisciplinarity exists (Hyland and Tse 2007), we claim that there is a common core in the academic lexicon that any writer must know (see also Tutin 2014, Frankenberg-Garcia et al. 2019). Thus, for instance, in academic English, all the following expressions could appear used in any scientific paper: to confirm a hypothesis, the problem lies in, on the other hand, as previously stated, the aim of this paper is, etc. As we can see, these examples comprise different types of lexical entities. We are using this term in the sense introduced by Polguère (2016: 6): “an entity that belongs to the lexicon of the language, has individual properties and must therefore be learned and described as such” (the translation is mine). Some of them are usually treated as phrasemes, but others have a fuzzier status as lexical entities. We call them academic lexical combinations (ALC). By ALC we mean recurrent strings of words that may or may not be semantically compositional. Most of them fulfil discourse functions such as giving examples, concluding, expressing possibility or certainty, etc.

We have designed a combined dictionary-corpus tool than intends to help novice writers in the usage of ALC: HARTA.1 The object of HARTA are Spanish ALC, among which there are collocations such as análisis profundo (‘deep analysis’), el análisis revela (‘the analysis reveals’), llevar a cabo un análisis (‘to carry out an analysis’), and also other lexical combinations that include idioms (por otra parte ‘on the other hand’), as well as items with a less clear lexical status, such as en lo que se refiere a (lit. ‘in what refers to’, ‘with regard to’). Here we will examine the different types of ALCs, according to the typology of phrasemes established in the Meaning-Text Theory literature (MTT, Mel’čuk 2012, 2015, 2020, 2021). As we will explain later, collocations and idioms, the best-known phrasemes, are easily distinguishable and well described. In this framework, collocations are not lexical units and do not have their own lexicographic entry, unlike idioms. However, formulemes, a type of clichés (Mel’čuk 2015), are a doorway to strings of words with a fuzzy lexical status. This paper focuses on the frontiers between idioms, formulemes or free phrases. More particularly, we want to discuss to what extent an ALC such as en lo que se refiere a (lit. ‘in what refers to’, ‘with regard to’) is or not a lexical entity and if so, how we should describe it in our lexical tool.

In what follows, we will provide, in Section 1, a brief overview of how some of these ALC appear listed in Spanish and English dictionaries and other lexical resources. Section 2 will explore the notion of lexical entity by examining which ones have a “right” to a lexicographic entry in the framework of MTT. Besides presenting some differences between different types of phrasemes, we reflect about what Sinclair called “extended units of meaning.” Section 3 is devoted to explaining the strategy that we have chosen in HARTA. We claim that for our goal, the most important criterion is the discourse function. Irrespective whether en lo que se refiere a is or not a formuleme or an idiom, we will treat it as a lexical unit in its own right, since it should be part of the lexical competence of a speaker who has this string at his disposal for introducing a subject in academic discourse. Finally, we draw some conclusions on the presented work and give future lines of research.

Many ALC are variable and less fixed than clear idioms. For this reason, it cannot be predicted if these “loose” expressions are collected in dictionaries or not. We have looked up some of them in Spanish dictionaries (the DLE and another dictionary focused on discourse particles, the DPDE) and English dictionaries (MacMillan Dictionary and the Oxford Learner’s Dictionary, OLD), as well as in a bilingual dictionary (WordReference) and an online machine translator (https://www.deepl.com/es/translator).