Časopis Slovo a slovesnost
en cz

The role of extra-linguistic means in written texts

Jan Čadil



Role mimojazykových prostředků v psaných textech

Language is, of course, the principle means of human communication and plays a key role in human thinking. However, human thinking does not just involve ”linguistic” thinking: similarly, human communication does not just consist of linguistic means. There is a large range of means between those that are very distant to the language (e. g., gestures; illustrations) and those closely connected with the language (e. g., intonation; punctuation). This applies to both spoken and written communication.

It is not necessary to recall the different views of the mutual relationship between spoken and written languages nor the disputes over whether spoken and written languages are two languages or two forms (i. e. two modes of existence) of one and the same language. Any solution of this question always depends upon the particular viewpoint one holds on the notion of language. There was always something to be explained, no matter what the solution was – either why the two languages are so close to each other, or why the two modes of existence of one language are so different. On the assumption that written and spoken texts are not the texts of two independent languages but the texts of one and the same language in its two forms (i. e. the texts of one language mediated by two channels – the aural and the visual) we would have to try to explain the differences between these texts. Is it possible to account for these differences by the two channels only, the two substances? We can find many differences that concern all levels of the language system (i. e. not only the phonemic and the graphemic levels). Let us take into account only the languages using an alphabet. Phonemes and graphemes present mutually corresponding units in these systems; nevertheless, their basic correspondence differs between individual languages and to a varying extent is modified and supplemented by other types of correspondence (cf. Sgall, 1986, p. 7).

If we consider higher levels of a language system, we will find many other apparent and obvious differences between spoken and written texts. Whereas in spoken texts there are often more incomplete syntactic structures, ellipses, juxtapositions, repetitions, corrections, contractions, etc., in written texts there are usually more complete and more intricate syntactic structures, condensations, as well as a higher level of arrangement and regularity. Though these characteristics occur more frequently within the respective texts, they are not the characteristics connected directly with the phenomenon of speaking or writing, but rather, they result from the situation and context and thus the characteristics are connected primarily with certain stylistic forms.

The intrinsic difference between written and spoken texts is, of course, the difference in the realization, i. e. the different substance (graphic signs or sounds). Many other differences are rather related to the situation that presents a framework for using certain types of texts. In many respects, the number of differences between a delivery of a spoken text prepared in advance and a written text is fewer than the number of differences between two spoken texts or two written texts that are stylistically distant from each other. As far as vocabulary and grammar are concerned, a spoken text prepared in advance can be closer to a written text rather than to a spontaneous spoken text, which, on the contrary, can be closer to a written dialogue in prosaic text or to a spontaneous, unedited personal letter. Certain characteristics are more frequent in one or another substance, but the grammatical characteristics (mentioned earlier) can be transferred quite easily to the opposite substance. These characteristics are not necessarily connected with written or spoken texts. Nevertheless, there is still some distinction between written and spoken texts just as there is something shared by the whole range of written texts.

[170]The text production is based on the same grammatical rules both in spoken and in written texts. However, we cannot manage only with these rules. Linguistic means are not the only means used in texts. We also use extra–linguistic means (ELM) both in spoken and in written communication. These ELM are connected directly and immediately either with speaking or with writing; these means cannot be transferred from one substance to another exactly and in an unambiguous way. Nevertheless, we cannot miss them. Natural language takes them into account and – to take it to the extreme – it is possible to say that an attempt to clear language texts of the extra–linguistic means would be similar to an attempt to create an artificial language.

In the comparison of speaking and writing it can be useful to imagine any text as an entirety created by the means of two kinds, or, in other words, to discern two components (or two layers) in any text. The first component can be called the neutral component of the text (NCT). It is governed by the rules that are shared by spoken and writter texts. It consists of the linguistic means in the narrow sense (i. e. everything that can be transferred from a spoken text to a written text or vice versa; in other words, the means corresponding in a stable, conventionalized, generally intelligible way). This can be noted in cases when graphic signs strictly, though not necessarily directly, are in accord with sounds (or the other way round). Actual language texts usually (and spoken texts almost always) contain something more than what is presented by the NCT. This ”something” is what cannot be transferred (transposed), but only ”translated”, i. e. replaced by the specific means used in the opposite substance, eventually substituted by periphrases. These means which do not have any unambiguous counterpart in the opposite substance, constitute the second component of a text; we can call it the marked component of a text (MCT). The differentiation of two components of a text is just a theoretical construction that can facilitate the recognition of the intrinsic qualities of the text in its respective substances). While the NCT is the common denominator of spoken and written texts (something that is linguistic in a narrow sense), the intrinsic qualities of speaking and writing are immediately manifested in the MCT. The more typical the text–type is for speaking or writing, the more substantial the MCT is. Compare, for example, the proportion of the linguistic and the ELM in the spoken argument (which has a variety of gestures, facial expressions, special voice qualities) with the proportion found in the news (which can be both spoken and written and where the ELM are relatively rare) and in dictionary entries (which present an exclusively written text – with a variety of punctuation marks; various types as italics, extra–bold print, capitals; abbreviations; special marks). There are many types of written texts, which are considerably dependent upon ELM, that is, their communicative efficiency is dependent upon the use of ELM. This applies to various charts, forms, scientific texts, etc.

As far as spoken texts are concerned, practically all sound qualities that have no grammatical function (that are not grammatically necessary) pertain to the MCT. This is, admittedly, a somewhat vague formulation. But it is difficult to say generally which sound qualities have a grammatical function and which do not. This differs across individual languages, e. g. the forms and the functions of individual patterns of intonation are different. Additionally, in spoken texts, both components of the text are more closely interconnected, in that they influence each other more than they do in written texts. Certain grammatical meanings (e. g. the modality of a sentence) are signified by a certain pattern of intonation (among other things). But the pattern of intonation is just a pattern. In actual spoken texts, the intonation is often more complicated and may also indicate other meanings. In other words, to separate the spoken text and its sound shape is a fictious undertaking (even the inner speech probably involves a certain ”quasi–intonation”).

Where exactly the border–line between the NCT and the MCT in written text lies, is – again – difficult to say. It depends not only on the differentiation of NCT and MCT in spoken text within a particular language, but also on the character of the writing system [171]that is employed. In written texts, the MCT consists of a large range of heterogeneous ELM, for example, the use of various characters or types, the arrangement of the text on the sheet (the lay–out), underlining, the use of titles, subtitles, supertitles, headings, the use of footnotes, glosses, annotations, marginal fingers, the use of other special marks, the use of colour in the text, etc., and also certain cases of the use of punctuation marks. In languages employing an alphabetical writing system, the NCT is constituted primarily by graphemes of letters, and in certain cases, by the use of some punctuation marks. Stability, general intelligibility and the obligatory nature of correspondence between the means in question of written and of spoken texts can serve as a criterion for inclusion in a NCT. The inclusion of certain graphic means in a NCT or to a MCT can differ not only between writing systems, but it can also be difficult to define the exact borderline between the neutral and the marked use of certain graphic means within the framework of a single writing system. This, for instance, is the case of the use of initial capitals or the case of the use of some punctuation marks. Certain punctuation marks are in some positions obligatory (they then often correspond to a certain intonation). Let us use the question mark as an example. In Czech, the question mark is usually used at the conclusion of a sentence as a signal of its modality. This function of the question mark is entirely grammatical. The question mark in this function is included in a NCT. However, the question mark is sometimes used in other functions – as a means of MCT (then we can consider it an ELM). The question mark can be used almost anywhere in a text, e. g. as a signal of misgiving or discord. This misgiving or discord can refer to any part of the text, to anything between whole sentence and a single word. A question mark in this function is often placed within the parentheses and/or combined with other punctuation marks, especially with an exclamation mark and with ellipses. The whole utterance can be substituted with a question mark, an exclamation mark, or any combination of the two.

The punctuation marks usually bear upon sound qualitites only loosely (though this varies between languages). It is said that the principal function of punctuation marks is the division, or the analysis, of a text. Punctuation marks can divide the text not only into sentences, clauses, or phrases, but they also can separate sections of the text that are longer than a sentence, or separate sections that are shorter than a word. Quotation marks and parentheses can separate a whole paragraph, and parentheses can also separate a part of a word (and thus create, in fact, two words in the same position of a sentence), as in the following example: česk(oslovensk)ého – ”Czech(oslovak) adj. gen. sg.”. The latter use of punctuation marks (or other ELM, such as, italics, capitalization, or colour) suggests two parallel possibilities of reading; that is, the text is, in a way duplicated (or the information provided by the text is duplicated, i. e. there are two parallel – not only potential – interpretations of (the part of) the text). This is a manifestation of the ability of the employed graphic means to cross (in a certain sense) the border–lines provided by the linearity of a linguistic text.

It is sometimes assumed that whereas a spoken text is linear (has only one dimension, the dimension of time), a written text has two or even three dimensions. Presumably, the linear nature of a text applies only to the neutral component of both the spoken and the written text. The additional dimensions of a written text apply only to the MCT. As the facial expressions, gestures, and certain sound qualities bring additional dimensions into spoken texts, the ELM used in written texts have the same role. This multi–dimensional nature of a written text is not rooted in the fact that the writing sheet has more than one dimension; it is rooted in the ability of ELM to convey some additional information. The ELM supplement and/or modify the information that is contained in the NCT. In the sign system limited to one dimension (as with a stream of speech, flowing in time, and with the graphic record imitating this stream) we are not able to communicate more information concurrently. ELM allow this (in part) because of their capacity to suggest more than one interpretation and to shift the meaning of the [172]NCT. This suggestion and shift represent, to put it metaphorically, the shift or diversion in the linear stream mentioned earlier. The role of ELM in written texts is to enrich the linguistic (basically linear) text with other dimensions. Producers of texts are aware of the ELM, employ them and sometimes even rely upon them. ELM, in general, present an undispensable extension of language in communication.




Augst, G. (ed.): New Trends in Graphemics and Orthography. Walter de Gruyter, Berlin 1986.

Gelb, I. J.: Principles of writing systems within the frame of visual communication. In: P. A. Kolers – M. E. Wrolstad – H. Bouma (ed.), Processing of Visible Language 2. Plenum Press, New York 1980, p. 7–24.

Hausenblas, K.: Výstavba jazykových projevů a styl. UK, Praha 1971.

Sgall, P.: Towards a theory of phonemic orthography. In: Eksplicitnoje opisanije jazyka i avtomatičeskaja obrabotka tekstov XII. MFF, Praha 1986, p. 1–46.

Vachek, J.: Written Language: General Problems and Problems of English. Mouton, The Hague 1973.

Vachek, J.: Written Language Revisited. John Benjamins, Amsterdam 1989.



Role mimojazykových prostředků v psaných textech

Při srovnávání psaných textů s mluvenými lze odlišit jazykové prostředky v užším smyslu (které jsou společné oběma substancím, i když mnohé z těchto prostředků jsou běžnější v psaných/mluvených textech) od prostředků mimojazykových, které nelze převést z jedné substance do druhé, ale lze je pouze nahradit jinými prostředky, vlastními zvukové/grafické substanci. Jazykové prostředky tvoří neutrální složku textu. Mimojazykové prostředky, tvořící příznakovou složku textu, umožňují (i) v psaném textu překonat jistá omezení daná lineárností neutrální složky textu a představují tak rozšíření komunikačních možností psaného textu.

Filozofická fakulta Univerzity Karlovy

Slovo a slovesnost, ročník 54 (1993), číslo 3, s. 169-172

Předchozí Helena Confortiová: Relations between written and oral forms of language in teaching Czech foreigners

Následující František Čermák: Lexis in spoken and written language