LANGUAGE LEVELS: EUGENIO COSERIU’S LINGUISTIC THEORY

ABSTRACT

From his very early writings on general linguistics onwards, the Romanian linguist Eugenio Coseriu, like his own contemporary Noam Chomsky and Ferdinand de Saussure before them, delineated the concept of linguistic knowledge . His pursuit of this concept led him to elaborate his famous distinctions of the levels of language , which came to be one of his most important contributions to linguistic theory. In addition to detailing such distinctions, this article discusses the author’s view of the nature, content and structure of linguistic knowledge. It also presents some external and internal circumstances to his theory that could explain the widespread ignorance of Coseriu’s ideas in the realm of contemporary general linguistics, especially in the English-speaking world, before recent claims that Coseriu’s original contributions could have given linguistics a different direction if half a century ago his work had received proper attention. Finally, this paper addresses the question whether Coseriu is a neglected genius of twentieth century linguistics as it is frequently argued by his former students and devotees.

Eugenio Coseriu; linguistic competence; linguistic knowledge; language levels

RESUMO

Desde seus primeiros escritos de linguística geral, o linguista romeno Eugenio Coseriu, como o seu contemporâneo Noam Chomsky, ou mesmo Ferdinand de Saussure antes deles, delineou o conceito de conhecimento linguístico. Sua busca pelo refinamento desse conceito o levou a elaborar suas famosas distinções dos níveis da linguagem, que figuram entre suas contribuições mais importantes para a teoria linguística. Além de detalhar tais distinções, este artigo discute a visão do autor sobre a natureza, o conteúdo e a estrutura do conhecimento linguístico. Apresenta ainda algumas circunstâncias externas e internas a sua teoria que explicariam o desconhecimento generalizado de suas ideias no âmbito da linguística geral contemporânea, especialmente no mundo anglófono, diante de reivindicações recentes de que as contribuições originais de Coseriu poderiam ter dado um rumo diferente à linguística se meio século atrás o seu trabalho tivesse recebido a devida atenção. Finalmente, discute se Coseriu é um gênio negligenciado da linguística do século XX como frequentemente argumentam ex-alunos e devotos de sua teoria.

Eugênio Coseriu; saber linguístico; competência linguística; níveis da linguagem

Introduction

Thirty years ago, Eugenio Coseriu’s book Competencia linguística: elementos de la teoría del hablar [Linguistic competence: elements of the theory of speaking, 1992] was published in Spanish. The book, first published in German as Sprachkompetenz. Grundzüge der theorie des sprechens (1988), was edited by Heinrich Weber from the lectures delivered by Coseriu at the University of Tübingen between 1983 and 1985 and presents Coseriu’s formulation of linguistic competence¹ as the knowledge the speaker has about his own language. The methodology used by Weber in his edition of Coseriu’s lectures echoes Charles Bally and Albert Sechehaye’s procedures with Ferdinand de Saussure’s Course in General Linguistics [ CGL ], which was edited from the latter’s lectures at the University of Geneva between 1906 and 1911 (the publication of which, it has been argued, Saussure would not have supported).

Unlike the methodology in the composition of CGL which was edited mainly through handwritten notes of Saussure’s students, Coseriu’s text was edited from transcriptions of audio recordings of his lectures. The project also had Coseriu’s consent and support, and he carefully reviewed the content, providing clarification to ensure that his ideas were faithfully rendered before it was sent to publication. Differently from other books by Coseriu, which are usually compilations of articles, this one systematises in a unitary way the main themes established by him during his academic trajectory and, as such, it can be regarded as a faithful introduction to the Romanian linguist’s ideas, or at least to his theory of linguistic competence.

Coseriu’s work has been read and contested over time. As the author of dozens of books and hundreds of articles on the nature of language, for the student of general linguistics he should not require any introduction. His theory, nevertheless, may be a novelty for someone reading this article, mainly if they are outside the world of Romance languages, languages in which Coseriu published most of his writings. Despite his intense intellectual activity over more than half a century which resulted in important reflections about what language is and how it works, his theories remain understudied. The reasons for this remain unclear, but this article draws together some considerations about the reasons for his obscurity in the realm of linguistics outside the domain of German and Romance languages. The fact is that Coseriu was one of the most brilliant thinkers of twentieth century linguistics. This is evident from his time at the University of the Republic in Uruguay (1950-1963), when he first formulated the themes that he would continue to develop for the rest of his life at the University of Tübingen in Germany, where he died in 2002.²

Someone unfamiliar with Coseriu’s theory might ask what his main contribution to linguistic theory was. Coseriu once addressed this question, and, without hesitation, he summarised the work of an entire and profitable academic life down to a “relatively simple distinction” (COSERIU, 1985). Indeed, it is true that the distinction he refers to is the cornerstone of his linguistic theory. It was early formulated in 1955 in his now famous text Derterminácion y entorno: dos problemas de una linguística del hablar [Determination and environment: two problems of a linguistics of speaking], since Coseriu was uncomfortable at that time with Saussure’s langue / parole distinction. The distinction that he claims to be his most relevant contribution to linguistics concerns “the levels of language”, something that “applies first of all to what in the last decades has been called ‘linguistic competence’” ( COSERIU, 1985 , p. 1). Although Coseriu refers to his distinctions as relatively simple, they are not simple at all: they involve a complex set of terminologies and concpets used to provide a picture of language at different levels and wiewpoints, something that he considers capital to the understanding of what linguistic competence or linguistic knowledge is.

This foundational concept of modern linguistics, regarded as the knowledge the speaker possesses of his language, is, nevertheless, not a creation of the twentieth century, when linguistics underwent a radical change of perspective. Following the great developments and transformations of this science, the historical linguistics of the nineteenth-century gave way to structuralism, either under Saussure’s or Leonard Bloomfield’s orientation. The new approach was stronger in the first half of this century, with a decline in the following decades, as Chomsky’s linguistics of competence emerged as the primary concern in linguistic theorising.

Chomsky’s first readings of Saussure’s theory sought to find analogies between the concepts of langue and generative grammar: “The generative grammar internalised by someone who has acquired a language defines what in Saussurian terms we may call langue ” ( CHOMSKY, 1964 , p. 8). Nevertheless, unsatisfied with Saussure’s definition of langue as merely a “store of signs’’, Chomsky breaks with Saussure’s concept and borrows from Humboldt (1999 [1836]) the concept of underlying competence: “it is necessary to reject [Sassure’s] concept of langue as merely a systematic inventory of items and to return rather to the Humboldtian conception of underlying competence as a system of generative processes” ( CHOMSKY, 1965 , p. 4). Similarly, Coseriu takes up Humboldt to develop his analysis of linguistic competence, but, unlike Chomsky, he separates the biological and the cultural levels of language and undertakes an analysis only of the latter.

This article has the goal of inquiring into Coseriu’s ideas on linguistic competence as a product of culture and, as such, to highlight his views on its nature, content and structure. Coseriu approaches this question by exploring what he calls the three levels of the structuring of linguistic knowledge: the universal (elocutional knowledge), the historical (idiomatic knowledge) and the individual (expressive knowledge) levels. Thus, in order to best contextualise his theory, the following section presents the background of the main conceptual distinctions in linguistic theory and its developments, discussing Saussure’s langue / parole and Chomsky’s competence/performance distinctions. Section 2 deals with Coseriu’s distinctions of the levels of language as a way to analyse linguistic competence. Section 2.1 presents Coseriu’s view of the nature of linguistic competence and his claims that linguistic knowledge is intuitive knowledge, opposed to the linguist’s reflexive knowledge. Section 2.2 discusses Coseriu’s statement that linguistic competence contains both signs and operations, a vision that conciliates Saussure and Chomsky's viewpoints. Section 2.3 shows Coseriu’s analysis of how linguistic knowledge is structured. Finally, the concluding section discusses the obscurity of Coseriu’s theorising in the realm of general linguistics and draws some possible reasons for it.

1. The background of Coseriu’s distinctions

The distinction between language and speaking is not an invention of modern linguistics; it was the subject of discussion in the disciplines of classical antiquity, such as grammar, rhetorics and dialectics, each one dealing with a specific aspect of linguistic knowledge. Nevertheless, it was only in the nineteenth century that such concepts were widely debated by great philosophers of language such as Friedrich Hegel (who distinguishes between language and speaking [ Sprache / Rede ] and presents language as the system of speaking), Wilhelm von Humboldt (who discusses the concepts of energeia and ergon and concludes that language is energeia rather than ergon , since it is not a product, but is in production), and Georg von der Gabelentz³ (who distinguishes language in general – Sprache , language – Einzelsprache , speaking – Rede, and language faculty – Sprachvermövem ).

One of the lessons that can be learned from those differentiations is that to make distinctions is an essential operation for the process of understanding an object of study. This conceptual operation aims to study objects that are not considered separate outside the linguist’s viewpoint. Of course, the act of abstraction may deform objects when, through a conceptual operation, the researcher separates their different characteristics in order to study them independently. Thus, when a linguist makes claims about the different aspects of language, this does not necessarily mean that he is making claims about the reality of things, or that the object is to be apprehended in the empirical world as a split, fragmented entity. It means, first of all, that a priori methodological decisions are necessary for researchers to proceed to any analysis of the nature of an object.

Apparently, it is by following this assumption that linguists tend to abstract language from speaking or to study these two objects independently. Coseriu (1985) starts his paper Linguistic competence: what is it really ? by quoting the Italian philosopher Benedetto Croce for whom “ conoscere è distinguere ” [“to know is to distinguish”], and Coseriu adds that “to attain scientific knowledge is, first of all, to make distinctions” (COSERIU, 1985, p. xxv).

Hence, in order to attain linguistic knowledge, the historicist school of thought is concerned mainly with the genesis and evolution of words. Saussurian structuralism, instead, emphasises the intelligibility of word arrangements and systematic organisation, since, for Saussure, the history of a word was not enough for the establishment of its current meaning. He introduces the distinctions langue / parole and, although he talks about a linguistics of parole , a project that he does not concentrate on, he establishes langue as the unique object of linguistics. This object was to be regarded as a system of signs constituted by relations of reciprocal determination between the signified ( signifié ) and the signifier ( signifiant ).

It is a commonplace in the history of linguistics, since then, to attribute to Ferdinand de Saussure the paternity of linguistics due to the epistemological break⁴ represented by the creation of langue , an abstract object within the concrete facts of language. Indeed, linguists of the first half of the twentieth century marvelled at the discussions surrounding the structure of langue and placed Saussure among the most prestigious linguists of all time.

However, from the second half of the twentieth century onwards, linguists became interested in what enables the individual to speak and understand his interlocutor. Indeed, the best explorations were by Noam Chomsky, and his work represented a revolution in this field. Within Chomsky’s generative enterprise, linguistic competence is the speaker’s capacity, based on innate principles, that enables him to use particular languages through experience. Therefore, the task of linguistics is to describe this component of the human mind, by establishing its innate principles and investigating its use.

Coseriu, by contrast, placed himself mainly in von Humboldt’s linguistic tradition, since early writings such as Derterminácion y entorno: dos problemas de una linguística del hablar [Determination and environment: two problems of a linguistics of speaking, 1955], which antedates Chomsky writings concerning his theory of linguistic competence, to elaborate a theory that he then called a theory of linguistic knowledge. The text is considered (by some of his devotees and former students) a milestone in the change of perspective in the study of language (although this text might be unknown by many), since, inverting Saussure’s viewpoint, Coseriu makes speaking ( parole ) the norm for all manifestations of language, the starting point of linguistic inquiry.

Notwithstanding the almost always fruitful rereading of Saussure’s work, with honourable mention to his text originally written in Spanish about the concepts of system, norm and speech ( COSERIU, 1952 ), which, for beginners in linguistics, is mandatory to dilute the apparent solidity of the Saussurian dichotomy langue and parole , the inversion of Saussure’s point of view proposed by Coseriu seems, on the contrary, to have been developed from a somewhat limited vision of the Genevan linguist’s thought. Saussure’s Course in general linguistics is quoted by Coseriu when the former says that the study of language should be rooted in langue , “from the very outset” one should “put both feet on the ground of language [ langue ] and use language [ langue ] as the norm of all other manifestations of speech [ langage ]” ( SAUSSURE, 1959 , p. 9), whereas Coseriu claimed that he would start everything from parole . Yet it is clear throughout Saussure’s Course that knowledge of langue is possible only through analysis of parole , so the difference between Saussure and Coseriu’s positions are not so apparent.

Nevertheless, Coseriu’s distinctions of language levels could help him find some singularity in expressing his approach to linguistic inquiry. Thus, he delineates the three dimensions which can be differentiated in the cultural nature of speaking: the universal level (speaking in general); the historical level (speaking one particular language); and the individual level (the discourse). Speaking, in these three dimensions, can be studied from different viewpoints: as activity, as the knowledge that enables this activity, and as the product of this activity.

This Aristotelian distinction (i.e., a free activity that can be taken as energeia , dynamis , or ergon ) is widely known through Humboldt’s theories on language. For Coseriu, each of the three levels can be studied from the three viewpoints, which extends the complexity of the study of speaking to nine fields (see Table 1 in section 2.3). He asserts that the study of linguistic competence must begin with the linguistic knowledge that the speaker possesses, and the linguist must subsequently address the problem of its nature, its content and its structure. In the following sections, each of the three aspects of linguistic knowledge is detailed.

Table 1
– Coseriu’s Matrix

	Points of view
Levels	Activity	Knowledge	Product
Universal level	Speaking in general	Elocutional knowledge	Totality of utterances
Historical level	Concrete particular language	Idiomatic knowledge	(Abstract particular language)
Individual level	Discourse	Expressive knowledge	Text

Coseriu (1985)

2. Linguistic Knowledge

2.1 The nature of linguistic competence: from intuitive to reflexive knowledge

One of the goals of linguistic research, as Coseriu points out, is to highlight the knowledge that the speaker accesses when using a language. It follows that the object of linguistics is the linguistic knowledge of the speaker. In search of substance for his point of view, Coseriu undertakes an analysis of how Saussure and Chomsky approach the problem of the nature of linguistic knowledge, pointing out that Chomsky conceives it as an intuitive knowledge of the grammatical rules of a language, assuming the linguist’s own knowledge of his language as the starting point for the description of this knowledge. For Saussure it is an unconscious knowledge, for “people use their language without conscious reflection, being largely unaware of the laws which govern it” ( SAUSSURE, 2013 , p. 84).

For Coseriu, the idea that langue is something whose power is beyond individual consciousness is found in various points in Saussure’s theory of langue . Thus, Coseriu argues that the problem involving Saussure’s conception is that he confuses unconscious knowledge with intuitive knowledge: “Linguistic knowledge is a certain knowledge, and as such, it is fully conscious, but it is a knowledge that can not be sufficiently justified. And this (...) is precisely what characterises intuitive knowledge.” (COSERIU, 1992, p. 218).

Intuitive linguistic knowledge is the knowledge the speaker has about his language which makes him/her competent to speak and to understand. In this respect, Coseriu (1979 , p. 52) states that “a non-pathological activity of the awakened consciousness is not and cannot be unconscious.” Thus, against Saussure’s statements that the linguistic mechanism “cannot be grasped without reflexion”⁵ and that “even speakers who use it daily may be quite ignorant in this regard” ( SAUSSURE, 2013 , p. 85), Coseriu protests that speakers are fully conscious of the linguistic system. They know what they say, and also how to say something, and they also know what they cannot say, what is grammatically “correct” or acceptable, “otherwise they could not speak” (COSERIU, 1979, p. 54). However, for Coseriu, the speaker rarely reflects about the way the linguistic instrument works (this is a task for linguists and their reflexive knowledge).

Chomsky, unlike Saussure, states that linguistic knowledge is an intuitive⁶ knowledge of the grammatical rules of a language. Thus, he makes intuitive linguistic knowledge the basis for the description of linguistic competence. For him, the starting point for the description of this intuitive knowledge is the linguist’s own knowledge of his language. Besides, the linguist must ask other naive speakers whether certain sentences are considered “correct” or not. This idea converges with Coseriu’s position on linguistic knowledge, but he opposes the methodology of generative grammar for not differentiating intuitive from reflexive knowledge. Indeed, this is, for Coseriu, a key distinction when addressing the nature of linguistic knowledge:

[…] generative grammar often alludes to the express opinions of speakers. It supposes that they correspond to their intuitive knowledge, but it is not like that. The expressed opinions of speakers are no longer merely intuitive knowledge, but attempts or beginnings in the direction of reflexive knowledge. ( COSERIU, 1992 , p. 221).

Thus, aligned with Humboldt and Friedrich Hegel’s conception of knowledge as creative and productive, Coseriu maintains that in the use of language, in the process of verbal creation, the speaker always goes beyond their experience and creates something new. For him, generative grammar, while accepting creativity as a universal property of languages, does not represent it as the speaker’s capacity to create culture, but tries to link it with “a linguistic knowledge which is innate” (COSERIU, 1992, p. 225).

Coseriu’s attempt to distinguish intuitive from reflexive knowledge finds support in Friedrich Hegel’s (1770-1831) distinction between what he calls “known” knowledge and “acknowledged” knowledge, or between non-justified knowledge and justified knowledge. In short, justified knowledge is the knowledge that emerges from the process of reflection about unjustified or intuitive knowledge. Analogously, Coseriu asserts that reflexive knowledge, the linguist’s knowledge, takes the speakers’ intuitive knowledge as a starting point: “The task of linguists is precisely to turn speakers’ non-justified, non-scientific knowledge into scientific, reflexive knowledge” (COSERIU, 1992, p. 229).

However, more appropriate than Hegel’s distinction for the determination of the nature of linguistic knowledge, Coseriu claims to be the distinctions that G. W. Leibniz (1646-1716) made in his treatise Meditationes de cognitione, veritate et ideis (1978 [1684]). In short, Leibniz firstly distinguishes between obscure ( obscura ) and clear ( clara ) knowledge ( cognitio ). Clear knowledge is, then, distinguished into confused ( confusa ) and distinct ( distincta ), and distinct knowledge is, finally, distinguished into inadequate ( inadaequata ) and adequate ( adaequata ). The table below summarises Leibniz’s degrees of knowledge:

Image 1
– Leibniz’s degrees of knowledge

These distinctions reveal Leibniz’s intention, which is to get to the clear distinct adequate knowledge, or to the nature of scientific and philosophical knowledge. From the distinctions established by Leibniz, Coseriu states that linguistic knowledge is ‘knowing-how’. In other words, knowing how to speak a language is a knowledge that is justified only immediately. It means that although the speaker knows exactly which set of words and structures to choose for the purpose of a communicative situation, he usually does not know the reasons why languages work the way they work, as he does not regard language as a structured system of signs with its internal relations. Therefore, Coseriu argues that rather than being reflexive knowledge, linguistic competence is intuitive knowledge. Conversely, linguistics is clear distinct adequate knowledge , which has the speaker’s linguistic knowledge as the object of reflection. Thus, for Coseriu, this fundamental difference between the speaker’s ‘intuitive’ knowledge and the linguist’s ‘reflexive knowledge’ licences linguistics to make the speaker’s linguistic knowledge explicit, and to formulate it in a deeper degree of reflection.

2.2 The content of linguistic competence: signs and operations

After outlining the nature of linguistic competence and conceptualising it as an intuitive knowledge that must be made reflexive by the linguist, Coseriu seeks to highlight the content of this knowledge by posing the question of what constitutes a single language. Again, to address the problem regarding the content of linguistic competence, Coseriu establishes a penetrating analysis of Saussure and Chomsky’s theories, showing that Saussure’s point of view highlights linguistic knowledge as being, essentially, an ability to relate a certain linguistic form to its meaning, since he conceives langue as a static system of forms and contents with their corresponding structures and paradigms. Chomsky, conversely, conceives of competence as a system of rules that operates dynamically to construct forms and linguistic structures correctly.

Regarding these different points of view, Coseriu takes a conciliatory position that expands the content of linguistic competence, attesting that it involves both signs and operations. Thus, one can approach language focusing on forms and contents or on a system of rules. By undertaking this theoretical position, Coseriu seeks to place himself between Saussure and Chomsky positions: “we have to maintain that the content of linguistic competence cannot be reduced simply to signs or operations, but that there are both” (COSERIU, 1992, p. 279). Hence, linguistic inquiry can approach it from a viewpoint that privileges the signs as forms and contents or can consider it as a system of rules that creates sentences. Nevertheless, the dichotomy (signs vs. operations) that Coseriu sets up between Saussure and Chomsky is rather simplistic and does not seem to accurately reflect either linguists’ true position.

As one can see below (Section 2.3.3), Coseriu establishes the three viewpoints according to which language can be regarded: as activity, as knowledge and as product. Nevertheless, it needs to be clear that when he is discussing the nature, content or structure of linguistic competence, he is addressing the viewpoint of language exclusively as knowledge . It is, for Coseriu, a threefold knowledge which he distinguishes into elocutional (universal), idiomatic (historical) and expressive (individual).

In short, the content of elocutional knowledge “comprises everything that applies in principle to all languages independently of their respective linguistic structuring, that is, a number of principles of thought and the general knowledge of the world” ( COSERIU, 1985 , p. xxix). Therefore, using expressions like The five continents are four: Europe, Asia, and Africa ; or The apostles were twelve ; Peter was an apostle, therefore Peter was twelve , Coseriu argues that they can not be considered as linguistically incorrect from the viewpoint of a particular language, as they are not deviant of “any syntactic or semantic rules of a given language but because they violate certain principles of thought” (COSERIU, 1985, p. xxix). Similarly, Coseriu states that sentences like This tree sings and I boiled the piano should not be regarded from the viewpoint of a particular language as deviant, but as “conditioned exclusively by our knowledge of things: in our empirical world, trees do not sing nor do we boil pianos’’ ( COSERIU, 1985 , p. xxx). Thus, when the speaker hears such utterances, he will seek to interpret them as coherent before judging them as incoherent, and such judgement he makes before those sentences reveals the elocutional knowledge he possesses.

Regarding the content of idiomatic knowledge, Coseriu asserts that in all sectors of a particular language (phonology, syntax, lexicon, semantics) the content of linguistic knowledge is represented by procedures made explicit as rules and elements that combine with each other. To illustrate it in the phonetic level, the German speaker knows the procedures for the aspirated pronunciation of sounds /p/, /t/, /k/; the French speaker knows the procedures for the aspirated realisation of /r/, and the English speaker knows the procedures for the realisation of the voiced and voiceless interdental (th) realisation. These examples reveal the content of idiomatic knowledge at the phonetic level, which is represented by the forms of expression and the articulation procedures involved in sound production. This way, one may see that “in the phonetic level there are both things, that is, forms of expression that the speakers want to perform and procedures for their realisation” (COSERIU, 1992, p. 277).

Similarly, one can see the procedures involved in the morphological level. Regarding the procedures in word-formation, it might be useful to have a glance at the procedure involving the sequence impact - impacting – ?impactful? – ??impactfully?? which will be discussed in the next section. Since the speaker knows both the linguistic signs (linguistic units) as well as the procedures of word-formation, although the construction ??impactfully?? may not exist in the English language, the suffix -fully will be recognised as a procedure to form adverbs in this language. This procedure reveals a kind of knowledge the speaker has regarding the linguistic mechanism involved in word formation.

Finally, Coseriu considers syntax the sector of combinations par excellence. Thus, “if one departs from syntax, i.e. from the combinations in the sentence, procedures will be the main occurrences. If, on the other hand, the lexicon is taken as a starting point, more combined units will occur” (COSERIU, 1992, p. 279). Nevertheless, it is important to highlight that the knowledge the speaker has of the procedures involved in sound articulation, word formation, or sentence combination is intuitive knowledge, which means that the speaker, although using language consciously, rarely reflects on how such procedures work when he/she speaks.

Regarding the content of expressive knowledge, third level of the threefold knowledge, which enables the speaker to construct oral or written texts according to the context of language use, it is related to the “knowledge of procedures [of text formation] with their inherent norms” ( COSERIU, 1992 , p. 282) and operates with linguistic signs of a particular historical language tradition. One can see how it manifests through the capacity the speaker has to distinguish different kinds of texts within the different text formation procedures. For Coseriu, the expressive knowledge enables the speaker to choose the proper procedures to start or to continue the discourse: “If a particular text begins with a procedure for the continuation of the speech, the speaker will recognise it as a deviation and possibly reconstruct for himself the ideal beginning of the text. Thus, the deviation reveals the norm of the beginning of the text” (COSERIU, 1992, p. 282).

Thus, conjunctions and linking words are, according to Coseriu, signs in particular languages that operate at the level of discourse and reveal the speaker’s expressive knowledge. Similarly, fixed forms to start a discourse as once upon a time in English language or its equivalent forms in other languages enable the speaker to access the knowledge he/she possesses of particular types of texts and interpret it as a fable or other type. Likewise, certain norms of text production as greetings may be considered deviant if the speaker does not know the historically adopted form in a speech community. Thus, if a non-native utters the French greeting Bon matin , taking it as an equivalent of English Good morning , he ignores the “greeting formula in the French linguistic tradition” ( COSERIU, 1985 , p. xxx), although Bon matin is not deviant or ungrammatical in French, but exclusively in this context of text production.

These examples highlight the procedures and norms inherent in text production in the individual level of discourse, making evident the speaker’s expressive knowledge. As Coseriu (1992 , p. 284) argues, the knowledge the speaker possesses about these procedures is more important since “the formulas and linguistic units are only indicative of the reference procedures.” Furthermore, it emphasises Coseriu’s position that the content of linguistic competence is not only about signs or operations, but rather it is necessarily about signs and operations.

2.3 The configuration of linguistic competence: the levels of language

Coseriu’s distinction between universal, historical and individual language levels is central to his linguistic theory and to the understanding of how linguistic competence is structured. Relying on Humboldt’s dictum that language is not “product ( Ergon ), but an activity ( Energeia )” ( HUMBOLDT, 1999 [1836], p. 49), Coseriu reformulates it and states that language is not essentially a product (although it can be abstracted and studied as such), but a creative activity (a free activity that creates signs and procedures), that goes beyond its power or its historicity (since the creative aspect of language enables the creation of new signs which keep language in a constant process of creation). Coseriu considers these distinction “indispensable” for the understanding of what language is and how it works: “we cannot understand language if we consider it only as activity, only as knowledge, or only as product, or if we consider these viewpoints as equivalents” (COSERIU, 1985, p. xxix). Therefore, “language is a universal activity that is performed by particular individuals, as members of historical communities” ( COSERIU, 1979 , p. 43). For Coseriu, a full description of a language can only be possible if these three aspects of the same reality which is language are taken into consideration.

To sum up, at the universal level, language is seen as a creative activity, speaking in general, which involves the technique of knowing how to speak (elocutional knowledge), and has as its product the totality of utterances. At the individual level, language as an activity is discourse, that is, the linguistic act of a given individual in a given situation, involving an expressive knowledge (discourse oriented knowledge), which has the text (spoken or written) as a product. At the historical level, language as an activity is the historically concrete language (Italian, French, English etc.), which involves a community’s traditional knowledge ( idiomatic knowledge). At this level, language, as a product, is never concrete, since what the creative activity produces is adopted by the community and becomes part of its traditional knowledge. Language considered as a product at the historical level “can only be the abstract language, that is, the language extracted from speaking and objectified in a grammar or a dictionary” ( COSERIU, 1980 , p. 93). The following table summarises Coseriu’s positions on the general structuring of language, which he calls “the levels of language”.

Coseriu’s matrix of language levels is well-known in linguistics and is largely rendered as a product of Coseriu’s geniality of theorising about language and its study. However, one can find it striking for instance that although Coseriu insists in his rejection of Saussure’s langue and parole distinctions, and yet when one gets to Table 1 , it certainly looks as though universal level/speaking in general equates with langage , historical⁷ level/concrete particular language matches with langue , and individual level/discourse with parole . As a matter of fact, the core of these distinctions antedates Saussure and one can find them in authors of the nineteenth century as Gabelentz, Hegel and Humboldt, or yet, in some respect, they can be traced back to Aristotle. Nevertheless, no one can find them distinguished in a complex way in which language can be effectively regarded in three levels and in nine viewpoints, as found in Coseriu, all of them offering a real possibility of approaching language, but whose autonomy is recognized to be only theoretical as they are inseparable in the activity of speaking.

Frequently in his writings, Coseriu criticises traditional and modern linguistics for having focused essentially on the historical level of language, which makes it a linguistics of languages. By criticising this, he advocates in favour of a linguistics of speaking, a theory that would concentrate its efforts on the study of speaking, or on “the universal technique of speaking”, a study that, for him, goes beyond linguistic facts since it involves non-linguistic knowledge, such as knowledge of contexts and situations – the general knowledge of the world⁸ . Consequently, Coseriu (1966) introduces the distinction between historical language and functional language . For him, a historical language contains not only linguistic knowledge but also extra-linguistic (knowledge of the outside world, about things). Therefore, historical language represents a diasystem that contains diatopic ,⁹ diastratic and diaphasic varieties. Coseriu does not indicate from whom he borrows the term “diasystem”. However, it is well known that it was coined in 1954 by Uriel Weinreich in his now famous paper “Is a structural dialectology possible?” to refer to a “system of systems”. As Weinreich states (1954, p. 390), a “diasystem¹⁰ can be constructed by the linguistic analyst out of any two systems which have partial similarities”

Regarding Coseriu’s uses of Weinreich’s diasystem concept and Flydal’s diatopic and diastratic terminologies, Völker (2009) emphasises that “Coseriu has taken over, unified, modified and, above all, promoted the terminological instruments proposed by Flydal and Weinreich” ( VÖLKER, 2009 , p. 32). Nevertheless, although Coseriu must be relying on Weinreich’s concept of diasystem, he does not quote Weinreich when discussing this concept. At least, he does not do so either in Coseriu (1966), the paper that Völker refers to, or in other papers where he continues to talk about this concept, such as Coseriu (1980) . On the other hand, Flydal is quoted in both papers and in others.

All in all, Coseriu defines historical language as a diasystem, only to abandon it shortly afterwards as he concentrates his theorising on functional language. Thus, for him, in contrast to historical language that does not have a concrete realisation, functional language does, and since it is related to the communicative event, it should be approached as having a syntopic , synstratic , and symphasic nature. However, that does not mean that in a single discourse the speaker will use only one functional language. On the contrary, different functional languages may intervene in a discursive act, but the analyst will realise that every discourse will be mainly guided by a specific given functional language: “it is always a functional language that is present at every point of the discourse” ( COSERIU, 1966 , p. 202).

Therefore, functional language is the proper object of structural description (see COSERIU, 1966 , 1980 ) since its relative homogeneity facilitates inquiry into the structure of linguistic knowledge. This theoretical position echoes Chomsky’s (1965 , p. 3) claim of the scientific need for linguistic homogeneity in the object of linguistic description, which led him to introduce the concept of ideal speaker-listener . Coseriu disagrees with Chomsky in a crucial aspect, since, for him, Chomsky sought to identify this homogeneous language simply as historical language. Since a historical language is composed of several functional languages, it could not be described as a “linguistic system with a unitary and homogeneous structure” (COSERIU, 1980, p. 113).

Coseriu does not ignore the fact that every speaker knows more than one functional language and can communicate in different dialects, levels and styles. Thus, he acknowledges that “the functional language has the disadvantage of never corresponding to the totality of the speech of a certain speaking subject” (COSERIU, 1966, p. 202). Nevertheless, he refuses to take the historical language as the object of linguistic description because it exceeds the linguistic knowledge the speaker possesses, since no speaker possesses knowledge of all a historical language: “a historical language – e.g. “French” – cannot be realized as such in the discourse: it is always in the form of one or the other of the many functional languages that it contains” ( COSERIU, 1966 , p. 202). This obstacle on the path of studying linguistic knowledge forces Coseriu to propose a conciliatory approach. The challenge of a “functional-integral” linguistic description should “conciliate the homogeneity requirement of the structural description with the requirement to correspond to the real knowledge of a language” (COSERIU, 1980, p. 117).

The question of how linguistic knowledge is structured is the third problem that Coseriu seeks to demonstrate in his description of linguistic competence. Again, he argues that such comprehension can be found in his distinction between the three levels of language structuring. Therefore, one should search for structure in strict sense, i.e. the internal relations that organize linguistic knowledge, on the level of idiomatic ¹¹ knowledge , through the analysis of a functional language: “a language that idealiter is homogeneous and corresponds to a single dialect, a single level and a single style” ( COSERIU, 1992 , p. 291). On the other hand, one should look outside the functional language structure itself to see how elocutional and expressive knowledge are externally conditioned. The first depends on the laws of thought and knowledge of things; the second is related to the purpose and circumstances of the communicative act in a given moment.

The idea of investigating a functional language by searching for its structure does not mean reducing linguistic knowledge to a unitary and homogeneous body of knowledge. As Coseriu states, “ structural description does not mean any reduction of the historical language to a single system”, but it means “that all oppositions must be established and described in the functional language to which it belongs” ( COSERIU, 1966 , p. 202). Hence, the assumption of homogeneity¹² can be regarded as an artifice that guides linguists in their investigation of the structure of language, that is, of the stable forms of its internal relations.

Still considering language as knowledge ( Table 1 above), a standpoint that regards language as intuitive knowledge, composed of signs and operations and is structured internally and externally, Coseriu then proceeds to distinguish three “functional strata”, or “three levels of grammaticalness”, according to which one could find how the functional language is structured: norm ,¹³ system and linguistic type . Norm is a hierarchically inferior dimension of the structuring of idiomatic knowledge, a body of linguistic realisations shared by the linguistic community. System is hierarchically superior to the norm since it includes much more than what is linguistically realised and it also establishes the conditions for the various possibilities of linguistic realisation. Finally, the linguistic type is conceptualised as the highest dimension of language structuring, seen as a possibility for the system.

To illustrate the difference between norm and system, one can find in the English word impact ¹⁴ and its possible derived forms an account of how these two concepts relate to each other. For instance, one can question if the sequence impact - impacting – ?impactful?? – ??impactfully?? – are English language forms. The first two forms are indeed well established in the English language norm. The form ? impactful? , although it is attested in English use since the mid-1960s, has sometimes been dismissed by the English norm and rejected as “bad usage” or a “meaningless word”. Lastly, the form ??impactfully?? certainly does not belong to the English norm, it has not been historically or socially established yet, or, at least, one can not find it registered in dictionaries. However, both ?impactful? and ??impactfully?? exist in the English language system since they are forms which are functionally possible: -ful and -fully are recognized suffixes. They can be regarded as belonging to English and not to another language since they are constructed according to English functional structures.

Indeed, according to Coseriu’s conceptual framework (COSERIU, 1979, p. 50), the system means “system of possibilities” since it indicates what is possible and what is not possible in the creative aspect of a language. The norm, by contrast, is a “system of obligatory realisations” since it does not have to do with “what can be said” in a speech community, but, instead, with “what has already been said” and traditionally “is said” in the same speech community. Therefore, the system represents the dynamicity of a given language and its possibilities of creating new words; the norm, conversely, corresponds to the fixation of the newly created items in the tradition of a speech community.

The system presents itself to us as a system of possibilities … it admits infinite realisations and only demands that the functional conditions of the linguistic instrument are not affected; rather than imperative, its nature is advisory. What, in reality, imposes itself on the individual, limiting his expressive freedom and compressing the possibilities offered by the system within the framework set by traditional realisations, is the norm: the norm is, in effect, a system of obligatory realisations, of social and cultural impositions, and varies according to the community. ( COSERIU, 1992 , p. 294).

In a later review of his system, norm and speech distinctiveness ( COSERIU, 1968 ), and from the concept introduced by Humboldt (1999 [1836]) of “characteristic form”, a structuring principle of a language, and Gabelentz’s analogous idea of “type of language” which he developed from his readings of Humboldt, Coseriu would position the type of language as the highest dimension of language structuring, which enables not only functions and oppositions already existing in the system, but many others that may or may not be formed. As he had already arguied ( COSERIU, 1966 ), one will find at the level of linguistic type the classes of oppositions and lexical distinctions proper to a language. For instance, he states that some types of language are more noun-structuring of reality, with relatively few verbs, which is the case of Persian. Conversely, some languages are more verb-structuring, with many verb-based derivatives which are the case of Ancient Greek and German.

Moreover, Coseriu (1968) postulates that different parts of the system of Romance languages are configured according to the same principles and can be analysed as belonging to the same linguistic type. Finally, he argues that “in principle, what is possible in the norm is already given in the system; what is possible in the system is already given in the linguistic type” ( COSERIU, 1968 , p. 280). Norm, system and language type are, then, three functional language dimensions that jointly operate in the configuration of a language. These dimensions form part of the speaker’s linguistic knowledge which enables him to speak and to understand his language. Jordan Zlatev (2011) , a contemporary cognitive linguist who recently came into contact with Coseriu’s theory, makes the following statement regarding the Coserian matrix of language structuring:

[…] this three-level conception of (linguistic) meaning is broad enough to include the multiple perspectives and levels of the phenomenon, while most theorists have typically focused on one at the expense of others… Such remarks show that Coseriu’s framework is not only a taxonomy of perspectives/levels of language, but can be used to explicate how these interrelate ( ZLATEV, 2011 , p. 131).

To sum up, it may be said that the knowledge the speaker has of his/her language, defined as linguistic competence, is a highly complex kind of knowledge. It is at the same time a creative activity that is constantly operating and a product of human beings’ ability to create culture. The Coserian distinctions have the merit of separating what in knowledge may be considered as universal, historical or individual: that is, what it means to know how to speak a language, what a language is and what a linguistic circumstance or act is. Indeed, in the use of language, the speaker intuitively disposes of the complex nature of this knowledge, which involves signs and operations that are structured internally and externally. This knowledge enables the speaker not only to speak and understand, but to exercise his creativity, so that he can go beyond the uses already foreseen and formulate others. This creative process keeps the linguistic mechanism in constant development. As Coseriu (1992 , p. 305) asserts, “every language is an open or dynamic technique”.

In lieu of a conclusion: the obscurity of Coseriu’s theories in the realm of general linguistics

This article briefly outlined a particular chapter of twentieth century linguistic theory in which the Romanian linguist Eugenio Coseriu played an important role in conceptualising the object of linguistics as the speaker’s linguistic knowledge. In lieu of a conclusion, it discusses the question of whether the obscurity of Coseriu’s theories in the realm of general linguistics is due to a kind of academic ideological boycott, language barriers, strategic errors, or some combination of these.

As this article shows, from his early writings ( COSERIU, 1955 ) as a young Romanian scholar with a strong linguistic and philosophical background, Coseriu had a firm position about what he considered to be the object of linguistics and how one should proceed in inquiry into it. Instead of supporting Saussure’s postulate that one should take langue “as the norm of all other manifestations of language” ( SAUSSURE, 1959 , p. 9), Coseriu proposes a radical change of perspective which was to invert Saussure’s viewpoint so that one should place speaking “as the norm of all other manifestations of language” (COSERIU, 1955, p. 32).

Indeed, it is evident from the beginning that Coseriu had a clear intention to make a name for himself as an author of theoretical linguistic models. Kabatek (2017) refers to Coseriu’s “glorious time in Montevideo” as a period when he brought to light remarkable texts such as Sistema, norma y habla (1952), Forma y sustancia en los sonidos del lenguaje (1954), Determinación y entorno (1955-1956), La geografía lingüística (1955) and Sincronía, diacronía y historia (1958). Of these, mainly Sistema, norma y habla (1952), Determinación y Entorno (1955-1956), alongside Sincronía, diacronía y historia (1958) were responsible for making Coseriu’s name known in linguistic circles around the world and for delineating what one can call a Coserian conception of language. All these texts were written in Spanish: because of this, Coseriu’s theory may have received unequal reception in general linguistics despite the relevance of his reflections about language.

Some of Coseriu’s collaborators (LOUREDA-LAMAS, 2007; KABATEK, 2017 ) have suggested that the unequal impact of his theorising on contemporary linguistics was due the fact that his favourite languages for disseminating his theories were Spanish, German, French, Italian and Romanian. In his productive period in Uruguay (1950-1963), during which he formulated the basis of his theoretical enterprise —themes that he would continue to develop during his long scholar trajectory — his main texts were published in Spanish.

Coseriu himself, presenting a paper in English (UCLA Conference on Causality and Linguistic Change, Los Angeles, May 1982), complains about the fact that his theory on linguistic change outlined in his paper Sincronía, diacronía e historia (1958) “has not always been understood, because of the ‘ Hispanicum est, non legitur (It is Spanish, won’t be read) ’ ” ( COSERIU, 1988 , p. 147). Coseriu adds that the ignorance that surrounds his theory may also have to do with “the oddity” of his “background in the spirit of that time , especially in the English speaking world”. Perhaps, what Coseriu means with the last complaint concerns with his humboldtian background in a context of Saussure’s ideas effervescence, since he argues in the beginning of his talk that “today, thanks not least to some notions of generative grammar and to a better knowledge of Humboldt, the times are much better, so that I hope not to surprise you with a completely heterodox conception” (COSERIU, 1988, p. 147).

Moreover, Kabatek (2017) points out some other circumstances that could explain the neglect that surrounds Coseriu’s theory. One is that Coseriu deliberately refused to be part of some circles as, for example, the emerging field of pragmatics, since he believed that “what pragmatics describes had already been outlined in a more coherent and clear way in Determinación y entorno ” ( KABATEK, 2017 , p. 29). For this reason, Coseriu would complain that the researchers involved with pragmatics ignored his works on this subject. Besides, Coseriu’s “critiques of generative grammar and, later, cognitive linguistics” reveal his strategy, which consisted in insisting “on his own ideas for the creation of a powerful school of thought before the others” ( KABATEK, 2017 , p. 29). Thus, outside the strictly Coserian Romanist circle, Coseriu’s ideas would circulate with certain restrictions.

Another circumstance involves a kind of ideological boycott of Coseriu’s work, at least within the scope of pragmatics, a field whose emergence is conventionally dated to the late 1960s/early 1970s, a period marked by the 1968 protests in many different parts of the world. As Kabatek (2017) asserts, since Coseriu had a conservative or right-wing reputation, he openly opposed the students’ protests which “led to a rejection of his work by broad sectors of the Germanists who defended the ideas of 1968” ( KABATEK, 2017 , p. 30). This way, Kabatek argues that one of Coseriu’s most independent students, Brigitte Schlieben-Lange [1943-2000], “a progressivist daughter of 1968, had the merit of opening the doors to Coserian ideas also in circles where he was rejected” ( KABATEK, 2017 , p. 30).

Outside external circumstances such as language barriers and the boycott that affected Coseriu’s work, there is the problem that some of his theories have been misunderstood. For instance, Coseriu, himself, recognises that the complexity involved in one his most important works ( Determinación y entorno , 1955) may be responsible for its low receptivity by the linguistics community: “It is a complete treatise condensed in a study. Everything is so compressed that it can hardly be comprehended” ( KABATEK, 2017 , p 23). Another important testimony is given by Esa Itkonen, who attended Coseriu’s course on the history of the Romance languages in 1965, and for whom Coseriu had great esteem. Itkonen describes an episode that took place during the 14th International Congress of Linguists, in East Berlin, in August 1987. During the congress, Coseriu introduced Itkonen to his entourage saying that he was “one of the very few people” who really understood him (ITKONEN, 2011, p. 25). This statement is very significant since it reveals Coseriu’s complaints about what he considered to be a misinterpretation of his linguistic ideas.

Apart from that, but still concerning the internal coherence of Coseriu’s theory, it is evident that mainly where the main focus of this article, the theory of linguistic competence, is concerned, one cannot find it systematically developed through the random papers Coseriu published for over half a century. It was not until 1988 that his ideas were consistently exposed as a whole with the publication of Coseriu (1988 [1992]). While Chomsky¹⁵ , who had a systematic theory, and from the beginning was surrounded by a group who had a strategy for the development of generative grammar (which included funding for its research and dissemination through international linguistic congresses), Coseriu was confined in the Universidad de la Republica , with a library that lacked books about linguistics, and where he was the first to create a linguistic circle seeking to connect South-American linguists in the area.

In light of the above, it may be argued that although Coseriu’s work has been read and contested over time, he is usually not regarded as a theorist of a new linguistic model. On the contrary, his work has been frequently read as a criticism of classical structuralism (being Coseriu himself considered a structuralist due to his studies in structural semantics) and as an introduction to theorists such as Humboldt, Saussure, Hjelmslev and Chomsky, rather than as a specific theorist of language ( ITKONEN, 2011 ; ALTMAN, 2017 ). Nevertheless, in recent years, after Coseriu’s death in 2002, there has been an increasing interest in the diffusion of his work. Every year, there have been conferences in many countries around the world such as Peru, Germany, Italy, Spain, Romania, France and Uruguay to debate his theories and eventually to give him a correct position in the history and philosophy of linguistics.

As a final point, it is appropriate to quote Coseriu’s own solemn statement about Hjelmslev’s glossematics on the occasion of the Danish linguist’s passing in 1965. For Coseriu, the merits of glossematics are far superior to its mistakes ( ITKONEN, 2011 ). Likewise, a reader of Coseriu will undoubtedly assert that despite the setbacks his theory may have faced, or the shortcomings it may have had, it is vastly outweighed by its merits.

Acknowledgements

I am very grateful to Dr Nick Riemer for his considerable contribution to this paper while I was a Visiting Scholar at The University of Sydney under his supervision. His detailed reviews were fundamental to give this text the form and content it now presents. Any remaining errors are my own. I am also indebted to Dr Marie Quinn (The University of Technology Sydney) for early reading and discussion of this work.

REFERÊNCIAS

• ALTMAN, C. Eugenio Coseriu entre a Filologia e a Linguística Brasileira (1950-1963). Revista de la Academia Nacional de Letras , Montevideo, n. 13, p. 97-117, 2017.
• BENEDETTO, C. Estetica come scienza dell’espressione e linguistica generale . Milano: Sandron, 1902.
• CHAMBERS, J. K.; TRUDGILL, P. Dialectology . Cambridge: Cambridge University Press, 1998.
• CHOMSKY, N. Aspects of the Theory of Syntax . Cambridge, Massachusetts: The MIT Press, 1965.
• CHOMSKY, N. Current Issues in Linguistic Theory . The Hague: Mouton, 1964.
• COSERIU, E. Competencia lingüística: elementos de la teoría del hablar. Tradução espanhola por Francisco Meno Blanco. Madrid: Editorial Gredos, 1992.
• COSERIU, E. Linguistic change does not exist. In: ALBRECHT, J.; LUDKE, J.; THUN, H. (ed.). Energeia und Ergon: sprachliche variation, sprachgeschichte, sprachtypologie. Studia in honorem Eugenio Coseriu. Tubingen: Gunter Narr, 1988. p. 147-157.
• COSERIU, E. Linguistic competence: what is it really? Modern Language Review , Cambridge, n.80, v.4, p.xxv-xxxv, 1985.
• COSERIU, E. Lições de linguística geral . Tradução de Evanildo Bechara. Rio de Janeiro: Livro Técnico, 1980.
• COSERIU, E. Sincronia, diacronia e história: o problema da mudança linguística. Tradução de Carlos Alberto da Fonseca e Mário Ferreira. Rio de Janeiro: Presença, 1979.
• COSERIU, E. Sincronía, diacronía y tipología. In: CONGRESO INTERNACIONAL DE LINGÜÍSTICA Y FILOLOGÍA ROMÁNICAS, 11., 1965, Madrid. Actas […], Madrid: Facultad de Filosofía y Letras, 1968. p. 269-281.
• COSERIU, E. Georg von der Gabelentz et la linguistique synchronique, Word , Philadelphia, v.23, n.1-3, p.74-100, 1967.
• COSERIU, E. Structure lexicale et enseignement du vocabulaire. In: PREMIER COLLOQUE INTERNATIONAL DE LINGUISTIQUE APPLIQUÉE, 1966, Nancy. Actes […], Nancy: Faculté des lettres et des sciences humaines de l’Université de Nancy, 1966. p. 175-252.
• COSERIU, E. Diacronía, sincronía e historia: el problema del cambio lingüistico. Revista de la Facultad de Humanidades y Ciencias/Universidad de la República , Montevideo, n.15, p. 201-361, 1958.
• COSERIU, E. Determinación y entorno: dos problemas de una lingüística del hablar. Revista de la Facultad de Humanidades y Ciencias/Universidad de la República , Montevideo, p. 29-54, 1955-1956.
• COSERIU, E. La geografía lingüística. Revista de la Facultad de Humanidades y Ciencias/Universidad de la República. Montevideo, n.14, p. 29-69, 1955.
• COSERIU, E. Forma y sustancia en los sonidos del lenguaje. Revista de la Facultad de Humanidades y Ciencias/Universidad de la República . Montevideo, n.12, p. 143-217, 1954.
• COSERIU, E. Sistema, norma y habla. Revista de la Facultad de Humanidades y Ciencias/Universidad de la República . Montevideo, n.9, p. 113-181, 1952.
• FLYDAL, L. Remarques sur certains rapports entre le style et l’état de langue . Oslo: Norsk tidsskrift for sprogvidenskap, 1951.
• HJELMSLEV, L. Langue et parole. Cahiers Ferdinand de Saussure , Berne, n.2, p.29-44, 1942.
• HUMBOLDT, W. V. Humboldt on language: on the diversity of human language construction and its influence on the mental development of the human species. Cambridge: Cambridge University Press, 1999.
• ITKONEN, E. On Coseriu’s legacy. Energeia, Zürich, n. 3, p.1-29, 2011.
• KABATEK, J. Determinación y entorno: 60 años después. In: HASSLER, G.; STEHL, T. Kompetenz, Funktion, Variation / Competencia, Función, Variación . Frankfurt am Mein: Peter Lang, 2017. p.19-37.
• LÓPEZ-SERENA, A. Intuition, acceptability and grammaticality: a reply to Riemer. Language Sciences , Oxford, v.5, n.31, p.634-648, 2009.
• LOUREDA LAMAS, O. Entrevista con Óscar Loureda Lamas, por Eugenia Bojoga. Revista electrónica de estudios filológicos , Murcia, n.14, dez. 2007. Disponível em: http://www.um.es/tonosdigital/znum14/secciones/entrevistas-%20loureda.htm Acesso em: 18 jun. 2019.
  » http://www.um.es/tonosdigital/znum14/secciones/entrevistas-%20loureda.htm
• MORENO-FERNÁNDEZ, F. A Framework for Cognitive Sociolinguistics . New York: Routledge, 2016.
• MOULTON, W. G. The short vowel systems of Northern Switzerland: a study in structural dialectology. Word , Philadelphia, v. 16, n.2, p. 155-182, 1960.
• NORMAND, C. Propositions et notes en vue d’une lecture de F. de Saussure. La Pensée, Paris, n. 154, p. 34-51, 1970.
• PULGRAM, E. Structural comparison, diasystems, and dialectology. Linguistics , Berlin, v. 2, n.4, p. 66-82, 1964.
• RIEMER, N. On not having read Itkonen: empiricism and intuitions in the generative data debate. Language Sciences , Oxford, v. 5, n. 31, p. 649-662, 2009.
• SAPIR, E. Language: an introduction to the study of speech. Harcourt: Brace, 1921.
• SAUSSURE, F. Course in general linguistics . Translated by Roy Harris. London: New York: Bloomsbury Academic, 2013.
• SAUSSURE, F. Course in general linguistics . Translated by W. Baskin. New York: Philosophical Library, 1959.
• TRUDGILL, P. The social differentiation of English in Norwich . Cambridge: Cambridge University Press, 1974.
• VÖLKER, H. La linguistique variationnelle et laa perspective intralinguistique. Revue de linguistique romane , Nancy, n.289, p.27-76, 2009.
• WEINREICH, U. Is a structural dialectology possible? Word , Philadelphia, v. 2, n.10, p.388-400, 1954.
• WEINREICH, U.; LABOV, W.; HERZOG, M. Empirical foundations for a theory of language change. In: LEHMANN, W. P.; MALKIEL, Y (ed.). Directions for Historical Linguistics . Austin: University of Texas Press, 1968. p. 95-195.
• WHITNEY, W. D. Whitney on language: selected writings of William Dwight Whitney. Massachusetts: The MIT Press, 1971.
• ZLATEV, J. From Gognitive to Integral Linguistics and Back Again. Intellectica , Compiègne, v.2, n.56, p.125-147, 2011.

• 1
  The term ‘competence’ has a strong Chomskyan association since it emerged as the proper object of the generative enterprise. However, as this article shows, this concept is not a creation of twentieth-century linguistics, but has its roots in Wilhelm von Humboldt’s (1999 [1836]) concept of ‘underlying competence’. Coseriu, likewise, establishes an analysis of Humboldt’s theory but returns to Aristotle to distinguish language as activity, knowledge and product.
• 2
  With this publication, I honour Coseriu and his academic trajectory on the occasion of the centenary of his birth (1921-2021).
• 3
  Coseriu discusses, in detail, Gabelentz’ distinctions and Saussure’s possible uses of them in “ Georg von der Gabelentz et la linguistique synchronique ” ( COSERIU, 1967 ).
• 4
  The idea of an epistemological break [ rupture épistémologique ] was introduced in the philosophy of science by Gaston Bachelard (1884-1962) in his celebrated book The formation of the scientific mind: a contribution to a psychoanalysis of objective knowledge (2002[1938]). By opposing Auguste Comte’s positivism, and his view of science as continuous development, Bachelard states that the history of science is a history of discontinuity. In the field of linguistics, the French linguist Claudine Normand (1970) , one of the most important Saussurean scholars, borrowed the term from Bachelard to answer the following question: “are we with Saussure facing an ‘epistemological break’, constitutive of a science, comparable to the example usually mentioned of Galileo’s work?” ( NORMAND, 1970 , 35). According to Normand, this break emerges essentially between the deductive and the inductive method, since Saussure starts with the empirical analysis to delimit the linguistic units. This methodological choice marks a rupture with the linguistic tradition of his time.
• 5
  Saussure here as in many other aspects of his theory of langue seems to follow William Dwight Whitney’s claims about the linguistic mechanism. Although, unlike Saussure, Whitney considered each individual as an agent who acts “in the great work of perpetuating and of sharing the general speech”, he argues that the work of each individual is done unpremeditatedly, or as it were unconsciously” (WHITNEY, 1971, p. 18).
• 6
  According to Coseriu, the idea that linguistic knowledge is an intuitive knowledge dates back to Benedetto Croce, in his Estetica come scienza dell’espressione e linguistica generale [Aesthetic as science of expression and general linguistics, 1902], in which he conceives language in general as a form of intuitive knowledge. This idea would have been passed to Sapir (1921) , who had much esteem for Croce, and from Sapir to Chomsky.
• 7
  As Jordan Zlatev (2011 , p. 129) rightly argues, Coseriu’s notion of historical language is not identical to Saussure’s conception of langue , since “it is not a monolithic unitary system” as it was for Saussure. As this article shows below, Coseriu opposes historical language to functional language. The first is never a single linguistic system, it is a “diasystem”. The concept of functional language, perhaps matches with Saussure’s conception of langue .
• 8
  Coseriu’s approach to these questions is referred to as integrational linguistics. Jordan Zlatev (2011) criticises integrational linguistics for, paradoxically, delaying in attempting to integrate with other current theories, what may be one of the explanations for the widespread ignorance of Coseriu’s theory in contemporary linguistics.
• 9
  The Greek prefix ‘diá’ (through, between) is used in linguistic terminology to denote the idea of variety or heterogeneity. The terms diatopic (relates to geographical space varieties) and diastratic (refers to social stratum varieties) were introduced by the Norwegian linguist, L. Flydal (1951) . Coseriu coined the term diaphasic to refer to expressive modalities and speech styles varieties ( COSERIU, 1966 , 1980 ). Opposing these three concepts, Coseriu introduces de terminologies syntopic , synstratic , and symphasic to refer to the ideal object of linguistic description, i.e., a functional language with relative homogeneity.
• 10
  Weinreich, with this work, intended to deliver a dialectology that was able to deal with the structural description of dialects. Differently from linguists’ practice of his time, whose position was that a linguistic system should be studied independently without comparison with other systems, Weinreich claimed that the comparison between dialects seeking its similarities and differences “could be not only meaningful but also revealing” ( CHAMBERS; TRUDGILL, 1998 , p. 35). Despite some linguists’ attempting in applying Weinreich’s claims to the study of different dialects, they would soon argue that it was not satisfactorily adequate (For further discussion see CHAMBERS; TRUDGILL, 1998 ; TRUDGILL, 1974 ; PULGRAM, 1964 ; MOULTON, 1960 ) and, therefore, it needed refinements. As some linguists have argued, Labov’s notion of variable rule provides a suitable framework to account for dialectal differences (In this respect see MORENO FERNÁNDEZ, 2016; CHAMBERS; TRUDGILL, 1998 ) and would have replaced Weinreich’s notion of diasystem in sociolinguistics studies. Since Weinreich supervised Labov’s master (1963) and PhD (1966) researches, it is not at all speculative to find Weinreich’s influence or even a joint creation of the concept of variable rule, which subsequently appears in Weinreich, Labov and Herzog (1968).
• 11
  As Zlatev (2011 , p. 130) explains, what Coseriu calls idiomatic knowledge “does not concern knowledge of ‘idioms’, but rather the ability to ‘speak idiomatically’, according to the structural rules of a language.
• 12
  Saussure’s concept of langue invokes a similar idea to Coseriu’s concept of functional language, something that abstracts itself from the elocutional and expressive knowledge. Besides, by assuming homogeneity, langue dismisses the varieties which constitute a historical language. A similar statement can be made about Chomsky’s notion of competence abstracted from the acts of performance.
• 13
  Certainly, the concept of norm is one of the most celebrated and original contributions by Coseriu, and it surely figures very centrally in his reconfiguration of Saussure’s langue and parole dichotomy. In his essay “ Sistema, norma y habla ” (System, norm and speech, 1952), Coseriu establishes norma as a link between Saussure’s langue and parole distinctiveness. By doing this, he showed his disagreement with Saussure’s dichotomy, but perhaps it was motivated by Hjelmslev (1942) distinctiveness of schéma ( langue forme pure ), norme ( langue forme matérielle ) and usage ( l’ensemble des habitudes ).
• 14
  This exemple came up through an informal conversation I had with my friend Dr Marie Quinn, University of Technology Sydney, as I was telling her about my research. For this and for her early reading of this paper, I am grateful.
• 15
  Jordan Zlatev (2011 , p.132), a scholar with great reputation in cognitive linguistics and semiotics, and who has recently “discovered” Coseriu’s integral linguistics, which departs from Coseriu’s Matrix exposed in Table 1 above, argues: “Even without having the privilege of reading Coseriu’s more extensive treatment of these topics in his books (due to the language barrier), … it is possible to surmise that linguistics would not have been in its present fragmented state if, sometime half a century ago, it had followed the lead of thinkers such as Coseriu rather than Chomsky”. A divergent viewpoint comes from Nick Riemer (2009) in an important debate with López-Serena (2009) , a Coseriu’s scholar. For Riemer (2009 , p. 657), “the distinctions between ‘(1) language in general; (2) particular languages; and (3) language as individual discourse’… or that between ‘the universal level, the historical level, that is, the level of languages of historically constituted communities, and the individual level: this or that fragment of language’ … are also hardly unavailable in generativism, which has simply chosen to make a different, though overlapping, set of distinctions”.

Publication Dates

• Publication in this collection
  26 June 2023
• Date of issue
  2023

History

• Received
  11 Apr 2022
• Accepted
  7 Nov 2022