Exploring Jakobson's 'phatic function'

Kulkarni, Dipti 2014. Exploring Jakobson's 'phatic function' in instant messaging interactions. Discourse & Communication 8(2): 117-136.

This research investigates the nature of phatic communion in instant messaging interactions. It adopts and expands Jakobson's much-quoted definition according to which 'phatic' is the language in an interaction whose primary purpose is to maintain contact between the speakers. Adapting conversation analysis for the study of textual interactions, the research observes the linguistic means used by interlocutors to signal attention, interest, and agreement - these being identified as important constituents of contact. The corpus comprises 60 chats, collected from 20 participants who chat in a mixture of English and Indian languages such as Marathi and Hindi. Openings, middles, and closings of these interactions are analyzed to study the ways in which participants establish, maintain, and terminate contact. The use of various linguistic means in these interactions such as back-channels, evaluations, expressives, and questions draws attention to a significant amount of interactional work done by interlocutors towards maintaining contact. (Kulkarni 2014: 117)

This is one extremely interesting abstract (abstracts are not often this interesting). I see that Kulkarni has made the connection between phatic function and phatic communion, but probably is not critical towards the simplicity of Jakobson's interpretation of phatic communion, as this seems focused on his interpretation. On the other hand, one could probably replace attention, interest and agreement with Ruesch's understanding, acknowledging and agreeing and reach new frontiers. Dipti Kulkarni defended a 150-page doctoral thesis on this subject in 2012 and uploaded it to academia.edu - which means that I may read the complete work, too.

The idea of phatic communion is traced to Malinowski (1923), who coined the phrase to refer to the language used to build ties of union with other members of the community. The linguistic function so conceived as some overlaps with the 'interpersonal' in Halliday's functional model (1973) and the 'social' in Lyons's classification (1996). The concept is important because it counteracts the emphasis on the descriptive function of language and draws attention to the social, interpersonal use of language. (Kulkarni 2014: 117-118)

Firstly, it was the demon of terminological invention that coined the phrase through Malinowski and I'm not so sure it originally referred to "phatic language". Secondly, wow. I'm only dimly aware of Halliday and Lyons so I didn't know about these parallels, but reached a similar conclusion through Ruesch - that the phatic function should rather be the social function.

The predominance of internet-based media such as instant messaging and social networking sites in facilitating our social interactions makes it pertinent to study phatic communion in this environment. Amongst the various communication platforms available on the internet, phatic communion will be observed here in instant messaging (IM). IM is a computer-based programme offered by websites such as Gmail, Yahoo and MSN that allows two people (in each other's contact list) to interact in real time through the continuous exchange of text messages. (Kulkarni 2014: 118)

I agree - the growing role of social media and new means of communication in contemporary society call for investigation. The definition of instant messaging I find weird, though, mainly because of the outdated and/or out of place word "programme". Instant messaging is not a specific programme but rather a type of communication characterized by it's temporal aspect. Technically, face-to-face interaction is also instant messaging, but instant messaging as such is the face-to-face analogy of internet communication: while face-to-face communication implies being in the same space, instant messaging implies being in the same time. These are very crude characterizations, but capture the spatiotemporal aspects quite well.

The foremost difficulty in studying phatic communion empirically is the problem of identifying a justifiable definition which at the same time can assist in recognizing phatic phenomena in natural conversation. I discuss these problems in the next section and review the approaches adopted in extant literature. Following this, I discuss Jakobson's (1960) interpretation of phatic and the ways in which it is adopted and expanded here. Jakobson (1960) defined phatic as the language in an interaction whose predominant function is to maintain contact between the speakers. Through this interpretation has been much quoted, it has not been pursued empirically. I take up the task and following this definition of phatic, the objectives of my analysis are to identify the communicative means used by interlocutors to maintain contact in their interactions and to see what these suggest about the nature of contact in IM. (Kulkarni 2014: 118)

Just like I had guessed - this expands on Jakobson's interpretation on phatic. This is probably why Kulkarni wishes to identify "phatic phenomena" - something like phatic language. Much could be written about expressive language and the emotive function, for example, in terms of the former being a heuristic device - a fiction useful for scientific analysis - and the latter being a relation between language and how it is used. I guess "phatic phenomena" is better than "phatic language" because maintaing or eliminating contact has verbal as well as nonverbal aspects. In IM, I imagine, there are similar chronemic rules as there are in oral communication - e.g. saying something, anything, after a period of quiet just for the sake of continuing the conversation. This is very narrow and not at all what I see in phatic communion, but I'm sure Kulkarni can lead the way towards a better understanding of phatics.

In order to investigate phatic communion in natural interactions we first need to know what is phatic. Though there is a broad agreement which suggests that phatic is language in an interaction directed at being social, there is considerable variance in the specific interpretations. Malinowski, who coined the phrase, primarily uses the method of instantiation to suggest the scope of the phenomenon: when a number of people sit together at a village fire, after all the daily tasks are over, when they chat resting from work, a mere phrase of politeness, inquiries about health, comments on weather, affirmations of some superbly obvious state of things (1923: 314-315). Functionally, phatic communion according to Malinowski serves to avoid silence, diffuse hostility, acknowledge the presence of another person, and enjoy each other's company (1923: 314-315). This description, along with the examples offered, does not point to a homogeneous set of interactional phenomena that can be labeled as phatic. It is unclear whether phatic is limited to short, routine exchanges or it includes longer stretches of interaction, whether it refers to communion building or contact maintenance, whether it is limited to desemanticized utterances or it also includes meaningful utterances; further, whether it includes only intances of bonding and coming together of people or also occasions of disagreement and estrangement. (Kulkarni 2014: 118-119)

I wish to distance my approach from the understanding of phatic communion as merely "niceties", "politenesses" or "small talk" as well as from management of involvement (channel management). Rather, I would make my way from phatic to emphatic, that is, empathy (from em + pathos - "in, at" + "passion"; like German Einfühlung - "co-feeling"). In effect, phatic communion in this sense would mean something like fellow-feeling, a sense of social unity. The phrase "desemanticized utterances" is relevant here, because it is close to what Charles Morris and Jurgen Ruesch mean by "communization" (sharing by means other than signs) and, following the empathy-line, phatic language should indeed - like expressive language - be introversive (desemanticized in the sense of not having external reference). The difference between phatic and emotive is in this sense very simple: emotives refer to the emotions or attitudes of the speaker while emphatics refers to the emotions or attitudes shared by the sender and receiver (e.g. what Jakobson generalizes as the "psychological connection").

Amidst these interpretations we find another definition of phatic communion in Jakobson's essay 'Linguistics and Poetics' (1960). Here he defined 'phatic' as the language in an interaction whose primary function is the maintenance of contact between the spakers; where 'contact' is the physical channel and psychological connection between the addresser and the addressee, enabling them to enter and stay in communication. (Kulkarni 2014: 119)

My aim is to show that there are other, nonverbal or simply broader, factors that enable people to enter and stay in communication. This is what Ruesch calls "common experience". For example, I am able to get what Kulkarni is writing about phatics because I already have a lot of background knowledge about phatics. Much like the Trobriand villagers who can exchange a few words with their neighbours and feel like they have something in common, I can read this text and feel that I have something in common with Kulkarni - interest in phatics. For my purposes it is relevant that interest, experience, emotions, etc are nonverbal and thus untouched by Jakobson's definition of the phatic function.

If we compare Jakobson's definition of phatic to that of Malinowski, we see that while Malinowski emphasizes the creation of ties or the relational goal of phatic messages, for Jakobson, phatic utterances were those that help to maintain contact between the speakers. This restricted interpretation has been questioned and criticized (Schneider, 1988: 24; Senft, 1996). (Kulkarni 2014: 119)

Oh my god, YES! Jakobson is restrictive in his interpretation of phatic communion just as he is of basically all other components and functions of the communication system. Also, now that I think about it, reinventing the phatic function as something like the social function would make a lot of sense in terms of groups (collective "I"-s) or figurations as well - it is often (sometimes even solely) phatics that establishes a collection of individuals as a distinct whole, a unity, a community. References to these critics: Schneider KP (1988) Small Talk: Analysis of Phatic Discourse. Malburg: Hitzeroth. and Senft G (1996) Phatic communion. In: Verschueren J. Östman J and Blommaert J (eds) Handbook of Pragmatics. Amsterdam: John Benjamins (loose leaf installment, 10 pp.). - I was not aware that so much has been written about phatics - Kulkarni's references are a gold mine!

However, I would like to argue that ther are overlaps between the two interpretations and Jakobson's definition is not as off-the-mark as it may look at first sight. First, some of Malinowski's examples point to the contact function of phatic utterances - uttering a polite phrase on seeing someone or saying something to avoid silence are examples of establishing contact (1923: 314-315). Second, Jakobson's notion of contact as not only the physical channel but also the psychological connection between the speakers takes care of the relational aspect of phatic utterances. The only significant difference is that Jakobson firmly situates phatic communion in a communicative context or speech situation as opposed to seeing it as a general social phenomenon. (Kulkarni 2014: 119)

This is exactly why I deem it necessary to connect phatic communion to communization: communication is not the only means of establishing contact or relations between people. That is, besides the explicit exchange of signs there are other forms of sharedness, some quite physical or materialistic, but others simply either too effervescent or too complicated for strictly linguistic or even semiotic analysis.

Jakobson does not specify what he means by these terms and neither is there any elaboration in existing literature. However, it appears that the physical channel is a notion associated with some necessary part of contact that must be present for the conversation to go on, whereas the psychological connection must be related to the attitudes that speakers must have and display so that the interlocutors find the environment conductive enough for conversation. Following these preliminary cues, I interpret 'physical channel' as the attention that speakers must pay towards each other. That the interlocutors should be mutually attentive seems to be the most minimal requirement for human communication (when all other material conditions are in place). It is only when the other interlocutor is attending to the conversation that a speaker can continue. Amongst the attitudes that form a part of the 'psychological connection', the current corpus suggests that showing interest and expressing agreement are the most crucial. That attention, interest and agreement are important in conversation and work towards contact maintenance has also been suggested by others (Jones, 2009; Schegloff and Sacks, 1974: 261; Schneider, 1988: 160). (Kulkarni 2014: 119)

Exactly! Jakobson elaborates the other functions and uses them for different purposes in speech and poetic analysis, but the newcomer - the phatic function - is almost abandoned by him. (Or at least that is the impression I have reached from reading a large portion of Jakobson's corpus.) Thus far I have associated the physical channel with - in Jakobson's terms - "some kind of contiguity between the participants of any speech event" and the psychological connection with "the perceiver's attention and fatigue". It is actually quite interesting how Jakobson - in his "Linguistics and Poetics" - very elliptically draws together the social, physical and psychological universes and imposes a hierarchy upon them - the social (phatic function) depends on the physical (channel) and the psychological (connection). Mutual attention is indeed the most minimal requirement for human communication and Ruesch has captured this in his definition of the communication system, which is contingent on mutual perception and awareness between the participants. My interest in Kulkarni's work has peaked and so far I agree with almost everything.

Following Jakobson's definition of phatic, the objectives of the analysis are twofold: first, to identify the various linguistic means used by participants to establish, maintain, and terminate contact (thus identifying instances of phatic communion), and second to see what these suggest about the nature of contact in IM. (Kulkarni 2014: 119-120)

My enter similarly consists of trying to identify the various nonverbal or bodily means used by participants to establish, maintain and terminate communication and to see what these suggest about nonverbal communication more generally.

From the point of view of contact maintenance, the absence of non-verbal cues in IM is significant, as these aret he primary carriers of contact-related information (Lyons, 1996: 53). (Kulkarni 2014: 120)

Huh. From Lyons's Semantics but Lyons himself refers to the reader on face-to-face communication by Laver & Hutcheson and their "interaction-management information" on this matter.

It is interesting to contrast this situation with Malinowski's observation of phatic communion as: '...personal accounts of the speaker's views and life history, to which the hearer listens under some restraint and with slightly veiled impatience, waiting till his own turn arrives to speak' (1923: 314). In IM, participants do not have to wait. Also, like all computer-mediated communication (CMC), IM exchanges are made possible because of a host of hardware and software technologies, and the affordances of this technology has an effect on the nature of interaction. With regard to the circumstances of interaction, IM participants are spatially separated and therefore, unlike face-to-face interactions, information regarding the interlocutor's availability and commitment to conversation is not readily available; rather this has to be inferred. Nardi et al. (2000) have observed that this allows the users to 'negotiate availability' and respond to messages at a time suitable to them. (Kulkarni 2014: 120)

I vaguely remember this part in Malinowski but now understand it differently. This makes me question whether it really is related to pathos. I must read some papers on Aristotle's rhetoric (e.g. ethos, pathos and logos). The bit about "the speaker's views and life history" does indicate that it has something in common with Ruesch's communization (although not perhaps with Morris's communization, which seemingly isn't about common experience). The interlocutor's availablitiy and commitment is related to Goffman's "involvement idiom", but I'm trying to distance my own thing from mere contact-maintenance and get at the underlying "base", so to say (I don't yet have specific terms chosen to handle what I'm trying to bring attention to).

Using Jakobson's definition of phatic as the heuristic tool the chats are scanned to identify utterances wich were, it can be argued, predominantly used to signal contact. Semantically similar instances are grouped together to form categories and then, drawing on contextual information like the relationship between the participants, properties of IM, and established understanding of conversational sequencing, I describe the nature of contact maintenance in IM. (Kulkarni 2014: 121)

But what does it mean "to signal contact"? I would argue that the phatic function should be approached as the social function because it's not so much about the transient physical channel as it is about the more lasting psychological connection - the social relation. I fully understand that Jakobson means psychological connection as something transient as well, but it can be reviewed in terms of Bateson's mu-function (communication about relationships), for example.

A third set of cues is the non-verbal behavior of the participants such as the alignment of the body, facial gestures such as smiling while reading messages, etc. The closest attempt at capturing the interaction in its entirety would be to have three cameras at each end which would record screen activity, keyboard activity, and the participant. The analyst would then draw on all these video recordings to fill out the chat script. Past research (e.g. Beisswenger, 2008; Marcoccia et al., 2008) has demonstrated the significance of this information in studying chats. However, it needs to be noted that while there could be some non-verbal behavior that may have been interactionally significant, largely as the hands are tied to the keyboard and the body is aligned towards using the computer, the body is not free for movement and gesture as it is in the case of face-to-face interactions. Evene the gaze is directed at the screen and not at the other participant. The textual cues that are mutually available thus are the only cues that are being communicatetively produced to a large extent. Recording the non-verbal behavior also raises the question of whether the analyst is trying to study the interaction as it was mutually available or as it was available to the researcher as a third party. However, as this research is based on the analysis of chat scripts alone, other interactionally significant information such as the system generated messages was not available for analysis. (Kulkarni 2014: 121)

This is an interesting topic which I know very little about but which is becoming increasingly important as more and more people spend more and more of their waking life interacting with technology (personal computers, laptops, tablets, smart phones, smart watches, and other smart objects). I can only quibble with the terminology that concerns nonverbal behaviour. I understand now that she means alignment of the body with the computer, but posture itself is not so much the alignment of the body as it is the alignment of body parts (I may be wrong because I don't specialize in postures). "Facial gesture" has the connotation of being interactionally significant (communicative) and intentional. If someone smiles to him- or herself when receiving an instant message then it is a "facial expression".

Laver (1975) observes the openings in face-to-face conversations and suggests that the indexical tokens used in the openings help speakers to identify the roles they will adopt in the oncoming interaction. (Kulkarni 2014: 121)

These indexes belong to a larger category that Ruesch calls metacommunicative instructions - e.g. the janitor's uniform is an index that instructs passers-by about his role as janitor.

Unlike face-to-face interactions, in CMC, potential interlocutors do not occupy the same physical space; rather, they only happen to be online at the same time. And being online at the same time does not place an obligation on them to acknowledge someone's presence in the way in which actually encountering someone does. In such circumstances it is through the first message, or the 'ping' as it is known, that one interlocutor invites the other to a conversation. The opening ping is identified as phatic irrespective of its content because its location in the conversation is such that it plays an important role towards establishing contact. (Kulkarni 2014: 122)

I would argue that it does place such an obligation, but in a significantly different way. I don't much enjoy instant messaging exactly because I feel obliged to answer and I can't concentrate on anything else when a conversation is ongoing. It feels weird letting someone's message wait for a reply indefinitely, especially when you know the content like in facebook chat where you can see the message in an open tab but don't send the "X has seen the message" if you don't click on the tab (I can do this because I use several different browsers, so that I can migrate to another browser to avoid acknowledging the ping). Ultimately I prefer e-mail because even if you're a distant and aloof person like me an instant message feels like it requires an answer in 24 hours max, while an e-mail can be answered 1 week later. In short, these different mediums/channels have different "involvement idioms". (I'm actually pretty sure that I'm misusing Goffman's term because he wrote at a time when there were no personal computers and the involvement idiom didn't apply to all forms of communication but only to face-to-face interaction.)

In IM, as the programme prefixes the message with the user's name, interlocutors do not have to identify themselves. Further, as IM interactions are often between friends, information about the availability and interest of the other person in having a conversation is also often known. Therefore much of the work that an opening in intended to do is already in place. In this context it is not surprising that speakers choose to omit the opening sequence completely. Interestingly, such non-routine pings were found between two extremes of the solidarity continuum. They were found either between people who share a very close relationship or between those who shared a very formal relationship. (Kulkarni 2014: 123)

This is what I'm trying to get at: friends and colleagues are communized (I'm not sure if this is the correct back-formation from communization) - they already know each other, have communicated before, have common experience and thus can engage in more atypical (non-routine) patterns of interaction. What I'm trying to get at, in the end, comes to an explanation of why Juri Lotman identifies the channel with memory in his model of autocommunication. It was actually very insightful and can explain some fundamental aspects of socialization as such. One of the points that I wish to make is that phatic communion and/or communization has a very important role to play in any kind of group-formation and community-building.

In a situation like IM, where non-verbal cues are not available, phatic utterances play a very crucial role. By 'phatic utterances' I mean those utterances that are directed primarily at signaling that the person is attending to the conversation. Unlike informative utterances that are considered valuable precisely because they give some information about the world, phatic utterances merely convey that the recipient of this information has attended to what has been said - besides this they may not convey any further information. (Kulkarni 2014: 125)

Yeap, this is exactly Gregory Bateson's mu-function or communication about relationship. The canonical nonverbal parallel to a phatic utterance is a headnod that signals that the nodding person is attending to the speaker's content. There are of course cultural variations in such "phatic gestures" as to a Western person a headnod may signal agreement while for the Japanese it signals understanding or acknowledgement.

In contrast, Shreyasi's 'wow!' provides Manavi with no new information. Yet what I want to argue is that this utterance does a significant amount of work towards maintaining contact. (Kulkarni 2014: 126)

It could be argued that this wow is more emotive than phatic. Irregardless of whether Sheryasi is really amazed or surprised at Manavi's information, Sheryasi is expressing amazement or surprise for Manavi's sake. But that's where the beauty of Jakobson's scheme actually lies: any given utterance can have a dominant function as well as subordinate functions. Kulkarni in this sense points out, in effect, that the phatic function is dominant, while emotive function is subordinant in this specific "wow!".

b) Evaluations/expressives. While a back-channel can be considered as a minimal response to incoming information, an evaluation does something more - as the name suggests, it evaluates the information. Evaluations and expressives are a little different from each other and yet they are so close that it is difficult to differentiate one from the other in all instances. Evaluations evaluate the incoming information with remarks such as good, nice, cool, too bad, etc. Expressives, on the other hand, are like interjections which exclaim using utterances like oh!, wow! or emoticons like :( and :). (Kulkarni 2014: 127)

I think they are difficult to differentiate because they are in fact the same thing. Jakobson subsumes both under the term "emotive" because he drew from the German phenomenological linguist Anton Marty, a student of Franz Brentano, according to whom emotive utterances inherently contain a judgment (evaluation). Interjections are the purely emotive stratum as Jakobson puts it, while most any kind of language can express an emotion or attitude towards what is spoken about.

In the absence of non-verbal cues like head nods, eye contact, and body posture which are significant carriers of contact-related information, all the work of signaling attention and interest during the conversation has to be done by the textual messages. (Kulkarni 2014: 134)

It is my task to investigate exactly how significant nonverbal communication is to maintaining contact. But since I'm an undergrad and my work is not empirical, I can go into an almost philosophical discussion of what it really means to have communicative contact.

Also, this research interpreted phatic communion in a specific sense (contact) which took attention away from certain other (relational, communion building) aspects of phatic communion. There is a need to observe phatic communion and community building as it happens online, from a macro-sociological perspective. It is through many such studies that we will gain a fuller understanding of phatic communion in the online environment. (Kulkarni 2014: 134)

Yup. Phatic communion is a curious phenomenon and despite the word being coined by a concrete text (Malinowski's essay) it has taken on very different connotations. It's as if Malinowski's demon of terminological invention was met by a whole host of demons of terminological appropriation. That is why I prefer not to talk about phatic communion, phatic function or even phaticity (as one of the next papers does) but about Phatics, although I'm not yet sure whether it signifies a field of investigation (spanning interdisciplinary efforts) or a host of phenomena related to phatic communion in some measure. There is much that needs to be covered before I can have a say in this veritably interesting and growing subject.