The Phatic Function in Translation

Nord, Christiane 2007. The Phatic Function in Translation: Metacommunication as a Case in Point. Belgian Journal of Linguistics 21(1): 171-184.

Metacommunication is regarded as a specific variety of the phatic function, which, according to Jakobson (1960), is responsible for establishing, maintaining and closing the communicative channel between senders and receivers. It is an empirical fact that the phatic function relies more on culture-specific conventions than any other function in communication. This means that translation can only work properly if the receivers recognize phatic elements as such. Therefore, adaptation to target-culture conventions may very often be the best way to make sure that the phatic function "works" for them. (Nord 2007: 171)

What in the world? I've retyped Jakobson (1960) and know for certain that he says nothing about metacommunication. After the phrase "It is an empirical fact" one should put a Wikipedian "[citation needed]".

After a definition of metacommunication (as opposed to object communication) and metalanguage (as a subcategory of metacommunication), the paper seeks to explore some of the conventions of metacommunication, drawing on a corpus of English, German, Spanish and French university manuals and textbooks. (Nord 2007: 171)

Huh. This sounds like an interesting topic, especially because Ruesch is not in the references - so how is Nord going to define metacommunication?

Drawing on the models of language function presented by Karl Bühler (1934) and Roman Jakobson (1960), I have suggested as a schema of four basic communicative functions, with their respective sub-functions, for the purposes of translation-relevant text analysis both in translation practice and training (cf. Nord 1997: 44ff.). The four functions are the referental or representative, the expressive or emotive, the appellative or operative functions included by both Bühler and Jakobson, and the phatic function which, as I see it, is unique in Jakobson's model. (Nord 2007: 171)

Operative function? What? My interest has peaked. She is very correct in the last part. The first three functions come from Bühler; the fourth, poetic, feels more like Mukarovsky's contribution, code and message were added by Jakobson, but these are essentially Saussure's langue and parole. It is exactly the latest addition, the phatic function, that is most unique in Jakobson's scheme. Although, to be fair, the channel component was not ignored by Ruesch, who treated it in terms of "medium".

Most scholars take the view that the phatic function is limited to the channel as such, as in small-talk or hesitation phenomena. In my view, however, "making the communicative channel work" presupposes an appropriate communicative relationship between sender and receiver. If this is true, one of the most important purposes of the phatic function will be to specify this relationship (e.g. formal/informal or symmetrical/asymmetrical) and perhaps even to shape it in the course of the communication process. Forms of address and other relation markers will therefore play an important part in phatic utterances, as I have tried to show in another context (cf. Nord 2002). (Nord 2007: 172)

Wow. How? How did Nord reach an interpretation of the phatic function as the mu-function (communication about relationship) without the aid of Bateson? This is incredible!

As opposed to object communication (i.e. communication referring to any object that is out there in the world), metacommunication can be defined (drawing on Lauer 1986: 33, annd Tegmeier 1984: 133) as the sum total of verbal and nonverbal (including paraverbal) means used by a writer or speaker to comment on the conditions and factors of the communicative occurrence in progress, including the interacting partners (author and audience) and the relationship holding between them. Some authors speak of metadiscourse or text reflexivity (cf. Mauranen 1993: 145ff.). I prefer the term metacommunication because it seems more appropriate if one wants to indicate the whole range of aspects included. (Nord 2007: 172)

Huh. So she reached this term through some German writers. I have to admit, this is a pretty concise and exact statement on metacommunication. The references: (1) Lauer, I.-A. 1986. Fachtextlinguistische Untersuchungen zum Kommunikationsbereich der Pädagogischen Psychologie - dargestellt an ausgewählten Fachtextsorten im Englischen. Diss. A. Leipzig University. (2) Techtmeier, B. 1984. Das Gespräch. Funktionen, Normen und Strukturen (Sprache und Gesellschaft 19). Berlin: Akademie-Verlag. and (3) Mauranen, A. 1993. Cultural Differences in Academic Rhetoric: A Textlinguistic Study. Frankfurt a. M.: Peter Lang.

Metacommunication aims at monitoring one's own communicative activities and the reactions of the audience, which are observed in face-to-face and anticipated or assumed in written communication. In order to make communication successful, these activities and reactions have to be in line with the overall goal of the communicativ eact. Metacommunication thus seems to be an efficient way of controlling social interaction as it is developing between author and audience (cf. Nord 2002: 38). (Nord 2007: 172)

Monitoring one's own communicative activities I would not categorize as metacommunication because I have a separate cluster of terms for that very purpose (autocommunication, intrapersonal communication, self-communication, self-indication, etc.). The reactions of the audience, on the other hand, was explicitly part of Ruesch's metacommunication (cf. Communication, the social matrix of psychiatry, 1951: 152). Specifying how metacommunication is related to the regulative function (e.g. controlling social interaction) is on my agenda.

Metalanguage, which refers to linguistic objects only, is a specific form of metacommunication. If metalanguage refers to language as such or a particular natural language system or the use of language in an occurrence outside the communicative action in progress, it can be regarded as a subcategory of the referential function, the object of reference being language or language use (which, unlike Jakobson 1960, I do not regard as being in principle different from othe robjects of reference like horses or the values of democracy). If, however, it refers to the actual communicative event in progress, it is phatic according to the definition of metacommunication given above. (Nord 2007: 173)

I think Jakobson regarded metalingual reference as a separate category because he was an überlinguist (he gave language a privileged status in everything) and it may have a little something to do with Bakhtin's metalinguistics. I, on the other hand, wouldn't identify the phatic function and metacommunication so easily because that would be, in a sense, to dismiss the original meaning of phatic communion, which was not at all about commenting on the channel or relationship. Also, the liquidation of functions can be dangerous, because in the end all functions are referential - the emotive aspects of speech have a reference to the speaker's state of mind just like the phatic function - in Jakobson's strict sense - makes reference to the fact of communicating (e.g. "Here we are. Yeap...").

If the object of metacommunication is the actual communicative action in progress, metacommunication itself has to be regarded as located on a hyperlevel above the level of object communication. [...] [...] We probably learnt to use and identify metacommunicative markers in the course of acquiring communicative competence, but to my knowledge the subject has not really been studied very extensively, either in Linguistics or in Translation Studies. I have often noticed that young students find it difficult to distinguish between object communication and metacommunication. When asked to name the overall topic of at ext they start retelling the story or paraphrasing the words of the text, instead of moving to a hyper-level in order to look down at the communication presented by the text. Obviously, recognizing the meta-level requires a specific ability that has to be acquired. Students, therefore, should be made aware of the metacommunication markers used in their own culture, first, and then proceed to learn about those of the other cultures they work with. (Nord 2007: 173)

Well, it doesn't have to be. In Peeter Torop's semiotics of translation metacommunication is understood in a way that made me invent an arbitrary distinction between synchronic metaommunication (communication about the ongoing communication process) and diachronic communication (communication about previous communications). I find Torop's intertextual metacommunication problematic because it presupposes the assumption of identification between culture and communication.

In phatic communication in general, i.e. in salutations, in small-talk about the weather and the like, we observe that people tend to use stereotypical forms of expression, standardized formulae which are almost void of any other meaning than that of operating or closing the communicative channel or keeping it open. In a greeting like "How are you?", the sender does not expect a detailed account of the receiver's health and mood but only a "Oh, fine, how are you?" (with a slightly different intonation). We might even say that phatic communication works on the basis of conventionalized forms. (Nord 2007: 173)

Dipti Kulkarni pointed out that these conventionalized forms are like desemanticized utterances. But the whole channel-maintenance through small talk theme seems to begin with Jakobson's reference to "talking birds" in Mowrer. I'm not sure how well this aspect was represented in Malinowski's phatic communion (I'll have to read his essay again as soon as possible).

a) If metacommunication directly or indirectly refers to the relationship between the participants in the communicative interaction, it may be assumed to be of particular importance in those cases where this relationship is essential to a successful outcome of the communication. Or to put it the other way round: when the receiver has a natural or professional interest in the object of communication (like in a scholarly debate published in a specialized journal for experts of the discipline), metacommunication is probably much less important than in an asymmetrical relationship in which the sender tries to persuade the receiver to take an interest in an object the latter would not have chosen of his or her own accord. Therefore, metacommunication may be assumed to be particularly important in manuals and textbooks, which can be regarded as a kind of simulation of classroom teaching. (Nord 2007: 174)

Hmm. Kulkarni's study on the phatic function in instant messages reached a similar assumption but her point was exactly the reverse: when the relationship is asymmetrical (as between a student and a teacher) then phatic utterances are "skipped" for sake of immediate getting-to-the-point. Otherwise it may come across as a social engagement and this is... bad? This difference may amount to a cultural difference between Nord the German and Kulkarni the Hindu.

Communicative occurrences are interactions between two or more participants in a situation that is fixed with regard to time and space. The participants use a certain combination of signs, such as a text, to convey messages between them. We can thus schematically identify the three possible objects of metacommunicative utterances: the situation, the participants and the text itself (see Figure 1). (Nord 2007: 176)

So we got rid of some functions in Jakobson's scheme only to supplant them into a new model of metacommunication? I still think that perhaps the only object of metacommunication is communication. Otherwise you're bending the concept of metacommunication quite arbitrarily. It is beginning to seem that Nord confuses metacommunication with metatextual communication - or, in effect, something like metatextual instructions (or, to use her own favored term, markers).

It is interesting to note that comments on the sender's activities or intentions are frequently (although not always) adjusted to the deixis of the situation of production (I am not saying..., I shall not enter further into...), whereas references to the receiver's reactions take the receiver's perspective, anticipating the situation of reception (as in You may wonder whether...). In these cases, metacommunication seems to simulate a face-to-face communication, which is also indicated by the use of tenses. (Nord 2007: 178)

But this is the conative function! "You may wonder whether..." is a literal appeal to the reader. It is not a reference to communication but a reference to the receiver's anticipated reaction. It turns out that if you take writing and reading a text as your case of communication then "metacommunication" isn't a very productive concept - it seems to turn everything into a jumbled mess.

To be classified as metacommunicative, a text reference has to be directed at the receivers and must be related to the communication in progress. (Nord 2007: 179)

Under what conditions can a text be thought of as communication in progress?

Acknowledgements are therefore not regarded as metacommunicative; nor are evaluations of other works or studies. However, bibliographical references are metacommunicative if tehy are expressed in the (conventional) form of an imperative or using verbs referring to steps the audience is asked to take (like see, cf. = confer, vgl. = verleiche, voir, véase). (Nord 2007: 180)

But bibliographic references are intertextual and imperatives are conative! I see no metacommunication here.

Since Grundlegung, Introduction à, Análisis or Suggestions refer to the communicative activities the sender wants to bring about by means of the text, they are metacommunicative. (Nord 2007: 180)

Here's the problem: we have different understandings of what constitutes communicative activities. Following Nord's own previous definition ("Communicative occurrences are interactions between two or more participants in a situation that is fixed with regard to time and space") I can't really see how a student reading the description of the task in a textbook constitutes a "communicative occurrence".

Since my focus in this paper was on the methodological aspects, the results presented are no more than indicators of the culture-specificity of metacommunicative behaviour. They are mainly intended to raise the interest in a hithero neglected aspect of the phatic function. I believe that there is still plenty of gold in this mine. (Nord 2007: 183)

I have to admit that this is an interesting topic and, at least in my opinion, quite relevant for translation studies, but! I was hoping this to be about the culture-specificity of metacommunicative behaviour and it turned out to be about the culture-specificity of some meta- or intratextual markers. I consider this to be an abuse of the concept of metacommunication because the original concept of metacommunication was about behaviour: how verbal and nonverbal messages interact in the communicative situation. Ultimately, metacommunication implies a metachannel. The title of a chapter or the instructions of a task in a textbook lack this essential component. In the end this is an example of the communication paradigm being more of a hindrance than benefit for linguistics, semiotics and translation studies.