SciELO - Scientific Electronic Library Online

 
vol.55 issue2ARBITRARINESS AND MOTIVATION IN COLLOCATIONSAPPEAL AND ATTENUATION: INTERGENERIC COMPARISON BETWEEN ORAL TRIALS AND PARLAMENTARY DEBATES IN SPAIN author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google

Share


RLA. Revista de lingüística teórica y aplicada

On-line version ISSN 0718-4883

RLA vol.55 no.2 Concepción Dec. 2017

http://dx.doi.org/10.4067/S0718-48832017000200143 

ARTICULOS

ROLES AND REGISTERS IN DIGITAL FORUM INTERACTION: DEVELOPING A COMMUNICATIVE IDENTITY-BASED APPROACH TO REGISTER VARIATION

ROLES Y REGISTROS EN LA INTERACCIÓN EN LOS FOROS DIGITALES: DESARROLLO DE UN ENFOQUE BASADO EN INDENTIDADES COMUNICATIVAS PARA EL ESTUDIO DE LA VARIACIÓN DE REGISTRO

ROSA GIMÉNEZ MORENO 1  

JUAN JOSÉ MARTÍNEZ SIERRA 2  

1University of Valencia, Spain. rosa.gimenez@uv.es

2University of Valencia, Spain. juan.j.martinez@uv.es

ABSTRACT

The paper studies how communicative roles and registers are expressed in an interde pendent way, influencing each other in current forums. In the first part, by combining previous insights in register variation analysis, social identity theory and cognitive linguis tics, three new concepts for linguistic analysis are proposed: 'Register-Modulating Role' (RMR), 'Register compass' (R-compass) and 'pivotal register'. The way they function in digital discourse is illustrated in the second part by analysing an open discussion forum sample corpus. The results show that each forum's profile displays a different array of RMR, together with their associated registers, which tend to spin around a prominent pivotal register. In our corpus three salient pivotal registers are identified, together with their associated registers and distinctive language features. The study substantiates the close connection between relational identities or roles and registers in digital discourse, and facilitates linguistic devices for a better use and understanding of register variation and forum discourse.

Keywords: Register-modulating role; communicative identities; computer-mediated com munication; discussion forum; register variation

RESUMEN

El presente artículo estudia cómo los roles y registros comunicativos se expresan de forma interdependiente, influenciándose mutuamente en los actuales foros digitales. En la pri mera parte, y a partir de la combinación de estudios previos sobre variación de registros comunicativos, la teoria de la identidad social y la lingüistica cognitiva, se proponen tres nuevos conceptos para el análisis discursivo: 'Rol Modulador de Registro' (RMR), 'brújula del registro' y 'registro-pivote'. La forma en la que funcionan en el discurso digital se ilus tra en la segunda parte, con el análisis de un corpus de foros de discusión. Los resultados demuestran que cada perfil de foro despliega una gama diferente de RMR, junto con sus registros asociados, y que todos ellos tienden a emerger y alternar alrededor de un registro-pivote prominente. En nuestro corpus se identifican tres claros registros-pivote con sus rasgos lingüisticos distintivos. El estudio sustenta la estrecha relación entre identidades relacionales o roles comunicativos y registros lingüisticos en el discurso digital y facilita herramientas discursivas para el análisis, la comprensión y el uso de la variación del registro en la interacción actual a través de internet.

Palabras clave: Rol modulador de registro; identidades comunicativas; comunicación por ordenador; foro de discusión; variación de registro

1. INTRODUCTION

The relationship between the linguistic choices of the speakers and their at titudes, roles and forms of expression has been studied in the last sixty years by many specialists in ethnography of communication, functional linguistics, dis course analysis, pragmatics, sociolinguistics, social psychology and other related disciplines. Thanks to pioneer studies on politeness theory, register variation (RV) and social identity theory, today we know that significant language choices de pend on the speakers' attitudes and manners (Brown and Gilman, 1960; Goff-man, 1967), on their purposes and relationships (Halliday, 1985), on their sense of group membership (North, 2007) and on selves variation (Tajfel and Turner, 1986). However, much work remains to be done to define and regulate the rela tionship between these parameters of variation in a way that allows its systematic analysis, the development of new computer tools for its detection and, subse quently, educational materials to facilitate its teaching and effective use.

The relationship between communicative role or identity and register has been persistent in previous research carried out on interactive genres outside the open Internet environment, such as email business correspondence (Giménez-Moreno, 2011a) and private conversations (Skorczynska and Giménez-Moreno, 2017). These studies have evidenced a significant relationship between relational identity variation and RV in current daily communication. Following this line of research, and the above mentioned foundational studies on variationist sociolinguistics (Bayley, 2013), the present research intends to put forward a new methodology for RV analysis by focusing on the interdependence between communicative roles and registers on online open access interaction platforms.

In order to fulfil this objective, after the introduction, we will summarise our approach to communicative identity and RV, focusing on the connection between identities, roles and registers. The concepts of 'register-modulating role' (RMR), 'pivotal register' and 'register compass' will also be introduced and illustrated in this section. The following part will be dedicated to highlight significant aspects of forum interaction. In the third part, we will concentrate on the methodology used to analyse communicative roles and registers in our corpus, and finally show the resulted RMRs, pivotal registers and main linguistic features as they are expressed in this type of communication.

Registers and roles in daily communication

In Applied Linguistics, RV, also called contextual variation, has been approached from very different perspectives, although only a few have had an extensive im pact (Halliday, 1980; Biber, 1995). As a result, and especially during the last half century, the concept of 'linguistic registers' have been conveyed through differ ent terms such as social dialects, diatypes, situational codes, language dimensions and contextual varieties, among others, covering and even often overlapping with other common concepts in the field of language variation analysis, such as genre and style (Giménez-Moreno, 1997; Biber and Conrad, 2009). This heterogene ity of terms and concepts, coupled with its inherent methodological complexity, has blurred the relevance and distinctiveness of RV, being gradually relegated to a lesser position in linguistics and discourse studies. This trend might reverse with new theoretical insights and practical applications of RV in educational and pro fessional settings. The transfer of results to other areas of communication is one of the main goals of the present study and a priority in our line of research within RV, as shown by the work done on registers in academic and business contexts (Giménez-Moreno, 2010, 2011a and 2011b).

A comprehensive and practical approach to language registers

Over the past ten years, we have been trying to make RV accessible to all learners and professionals approaching it from a unifying perspective which could en able effective applications and wider coverage studies (Giménez-Moreno, 2006; Giménez-Moreno and Skorczynska, 2013). Our framework is based on four main criteria which specify: (a) the main parameters of study, (b) its coverage, (c) its fo cus and (d) the methodological approach. These criteria intend to bring together the most significant insights of previous researchers in situational and RV (cf. Biber, 1995, 2012; Biber and Conrad, 2009; Halliday 1980, 1988).

The first principle focuses RV research around two distinctive and defining parameters: (1) the communicative settings, regarded as mental models and lan guage choices dependent on 'where we are', and (2) the participants' roles, viewed as mental models and language choices dependent on 'who we act as being'. Both concepts are addressed on the basis of their socio-cognitive nature as shared men tal constructs (Van Dijk, 2006, 2008). Secondly, a comprehensive approach to RV needs to assume the interdependence of registers in a language and therefore include coverage of all the situations that speakers are exposed to throughout their daily communication, from family to professional settings. Thirdly, for an extensive application, the initial theoretical framework needs to find the way to define a practical set of registers and their distinctive language features, prior to other more complex and in-depth linguistic insights. Finally, on a methodologi cal level, although quantitative data and the use of modern corpus analysis tools for studying RV is a must, native speakers' observation and collaboration should be considered as a priority in research design, data collection and corpus analysis.

Results obtained from research based on these four criteria indicate that regis ter varies as a dynamic continuum from intimate and private interaction to pro fessional and public communication, covering a wide range of everyday roles and situations. Depending on those roles and situations, during daily communication native speakers clearly identify four macro-registers (Giménez-Moreno, 2006): family, amicable, social and professional. Under this framework, native English speakers also identify that each of these registers has at least three communicative versions or tones, as illustrated in Table I: (a) a more relaxed, flexible and informal, (b) a neutral or conventional, and (c) a more ceremonial, rigid and formal.

Table I Professional register: examples of internal variation (Giménez-Moreno, 2006: 102). 

Depending on the specific situational and discursive context, the participants' profile and specially their intentions, they might use one or several of these tones, or shift from one register to another.

From this perspective, as mentioned above, several genres and communicative environments have been analysed: business correspondence (Giménez-Moreno, 2011a), business meetings and phone conversations (Giménez-Moreno, 2011b), email writing (Giménez-Moreno and Skorczynska, 2013) and private conversa tions in the family register (Giménez-Moreno and Skorczynska, 2015; Skorczyn-ska and Giménez-Moreno, 2017). For the understanding of register fluctuation and for the instruction of professional practitioners on the functioning of RV in daily communication, this approach has proven to be helpful (Giménez-Moreno, 2010). However, in order to delve deeper into this strategic use of registers, their fluctuation and the application of our findings to other genres and contexts, we need to define and classify more accurately the main parameters of variation, in particular the notion of role/identity as a modulating element of RV. In the initial studies carried out on RV, both the results of our analyses and the observations of our native collaborators, highlighted a recurrent interdependence between the roles adopted by the communicators and their chosen registers of expression (Giménez-Moreno, 2006). However, not all types of relational identity seem to have the same effect on RV, making it necessary to delve further into the register regulating function of these identities and into how their connexion and variation operates at a cognitive level. The following two sections propose some insights into this approach.

In search of register-modulating roles

Comprehensive books on human identity, such as The Sage Handbook of Identities (Wetherell and Mohanty, 2010) usually include studies on many types of identity such as biological, racial, multicultural, technological, religious, gender, social, sexual, parental and so on. Most of the specialists who contribute to these volumes show broad agreement in understanding identity not only as 'who we think we are' (individually or collectively), but also as 'who we act as being' in interperson al and intergroup interactions (Baumeister, 1986; Butler, 1990; Reicher, 2000), clearly differentiating among three main levels of the self: personal, relational and collective identities (Sedikides and Brewer, 2001).

This initial differentiation directs us towards our primary goal which is on a relational and interpersonal level. According to Chen, Boucher and Kraus's model (2011: 149) 'the relational self is self-knowledge linked to knowledge about sig nificant others; (...) capable of being contextually or chronically activated'. This dimension of human identity in action dependent on our relation with significant others, which focuses on the study of 'who we act as being', may help us to ap proach the study of 'how we express who we act as being', establishing correlations and assumptions between the relational self and its linguistic expression. From this perspective the deep relationship between relational identities ('who we think we are'), communicative roles ('who we act as being') and linguistic registers ('how we express who we act as being') becomes evident, their joint study being necessary to develop a comprehensive theory of RV, particularly the last two: roles and registers.

On this point, it is important to highlight the influence of socio-cultural varia tion on these two concepts which, as Biber and Conrad (2009: 23) points out, should always be studied with reference to the socio-cultural contexts to which they belong. For various reasons (e.g. accessibility, proximity, own specialisation, educational transfer, etc.) our approach has focused on British culture and our collaborators have been British English speakers. For example, in the case of the present research, although the identity of the participants is hidden under their nickname, in the corpus selection we were careful in choosing forums which were created and administered from Britain, in which the majority of participants used British English and/or provided cues in their messages about their British back ground. Therefore, in order to start profiling the way British speakers act as being and subsequently their main communicative roles, it is necessary to determine whether there is some consensus among them on the most frequent role models: the name and type of roles that relate and are linked to each register in everyday communication. To clarify terms, these salient roles or identities that influence RV in a direct and significant manner are called 'register-modulating roles' (RMRs).

RMRs are multifaceted conceptualisations which comprise the existence of at least two relational identities (e.g. the role 'mother' implies the existence of the role 'son'/'daughter'). Each RMR is concomitant with a particular register (e.g. family register), generally and usually recognized and used in a given socio-cultural context (e.g. current British society). This concept helps to objectify and reinforce the interdependence between roles and registers, as opposed to other roles that can coexist in a same communicative setting but which do not affect categorically the RV operating in that context. For example, a woman may per form the roles of 'mother', 'heroine' and 'protagonist' simultaneously in a family situation; however, at a communicative level, the first one may become the RMR. Conversely, the role 'mother' can coincide with 'teacher' in an educational setting and this latter become the RMR in most of that particular interaction.

Although the correlation between the four macro-registers above mentioned and their potential associated roles may seem predictable a priori, it is worth checking if it is so in the current English socio-cultural context. In order to start the research at source, a straight forward survey was carried out with 55 native collaborators between the ages of 19 and 54 years, all of them with language description skills from basic to advanced. Our informants' profiles were: 27 na tive English university students (undergraduate and postgraduate) and 28 native English professionals (most of them university staff and business workers). The survey contained a copy of Table II without information in the second column (only the heading 'roles/identities'), and instructions to carry out the following two tasks (see appendix):

  1. In the first task informants had to classify a list of given standard roles/iden tities (e.g. husband, boss, neighbour, girlfriend, etc.), including each role in its corresponding setting, adding any comments they might consider interesting about the roles, the settings or their own way of sorting them.

  2. In the second task informants were asked to complete the four groupings with all other roles they considered common or characteristic of each set ting, both from their own experience or by their observation of other peo ple living or working around or near them.

The RMR suggested by our collaborators followed different parameters (i.e. hierarchy, function, relationship, etc.) and were accompanied by diverse types of personal comments (e.g. 'my brother went to Canada when I was a kid so when we meet up at Christmas at my mum's it's like talking to a neighbour'), however there is a high degree of homogeneity in their outcomes: regarding first task 48 of our informants (87% of the total) coincided in the classification of the given standard roles, and 36 informants (72%) coincided in the additional roles they included in each setting in task two. As a synthesis of the results, Table II shows these RMR that generated greater agreement in both tasks among our informants.

Table II Basic register-modulating roles in daily communication (based on the register typology proposed in Giménez-Moreno, 2006). 

The asterisk indicates those RMR where there is a certain degree of cross-over between registers/tones, as our collaborators were aware of the fact that they could be dependent on the particular relationship that leads to the expression of that RMR. For example, twelve of them indicated that they had some neighbours whom they regarded as friends. In the results there are also cases of friends who are regarded as relatives, and even a doctor who sometimes behaves like a friend. Yet, even considering these shifts, there was a significant consensus in the RMR of Table II. Our collaborators, not only shared a number of role models, but also clearly assumed an interdependence between certain roles ('who I act as being'), certain contexts ('where I am') and certain registers ('how I express who I act as being in the setting where I am'). It seems that naturally each context prompts the speaker to adopt a particular role; either imposed by the circumstances (e.g. role of 'father') or selected depending on his/her intentions and circumstances (e.g. a father trying to be a 'close friend' of his son). However, there are still many unre solved questions here.

As our previous research showed, these identities often do not follow each other in a linear way (Giménez-Moreno, 2011a). In many contexts, beyond inten tions and circumstances, RMR alternate or fluctuate very quickly depending on the demands of the moment. As an example will illustrate later (in Table III), an executive may receive an urgent personal call on his mobile phone while he is in a meeting. In this context the family register will switch to the professional register briefly forcing the speaker to modulate both almost simultaneously. Something similar often happens to doctors and patients, when they are required to com municate with each other at public, private and intimate levels (Bonnin, 2013). Therefore the question that arises is how speakers deal with these changes and adjustments so naturally. It appears that the next logical step is to observe how these connexions happen.

The 'register compass'

From a socio-cognitive perspective, experts point out that human beings have an epistemic internal device, or special knowledge-management function, called by Van Dijk (2004, 2006, 2008) 'K-device' ('knowledge device'), that 'examines which knowledge may or should be expressed in discourse' (Van Dijk, 2006: 171) and regulates 'which knowledge of mental models may or must (not) be more or less explicitly expressed in discourse, or be left wholly or partially implicit' (Van Dijk, 2004: 12). This K-device has to be relatively simple as it has to be able to tell us, in fractions of seconds, how everything we say should be adapted to what we think our interlocutors' knowledge is; not only in terms of roles and registers, but also in reference to other types of identities that interact at the same time (e.g. national, cultural, professional, etc.) and to all the information shared and exchanged in that situation. At the interpersonal level, this K-device is constant ly adapting itself and finding strategies to approach each interlocutor and each context. Thus, logically, this device must include an internal 'situational code-switching' system (Auer, 2002; Gumperz and Hymes, 1986) or 'register compass' ('R-compass') which regulates RV pointing 'to the north' (i.e. to the adequate register at any time) as soon as interlocutors play a certain role in a certain context. We see this R-compass as a pointing and adjusting device, an internal detector or sensor that distinguishes the salient role or RMR in every situation and activates its mental model together with its related communicative features.

But how does this register compass works on a psychosocial level? How does this compass detect the RMR operating in a certain context and fluctuate across them? In order to look for evidence in current everyday communication within the SILVA Group (Research Group of Support for Investigation on Language Variation Analysis, registered within the Inter-university Institute of Applied Modern Languages, IULMA-University of Valencia, Spain) we are recording and gradually compiling a corpus of RMR samples. Table III includes a recorded frag ment of interaction in a business lunch in London last year which illustrates how the R-compass works at a time of rapid fluctuation between registers:

Table III Example of RMR fluctuation. 

In this fragment we see a sequence of RMR which occurs in seconds: our target communicator (speaker 1) is using the RMR 'partner to partner' to interact with his associate (a business woman) expressed in professional neutral register. As they seem to be the most frequent and dominant in the meeting, we might call them the axial or 'pivotal' RMR and register around which the others rotate during that encounter. As we see, the register changes into professional-casual when his R-compass detects that the lunch is coming to an end and his phone is ringing. Then, he takes the RMR 'costumer-to-waiter' manifested in social neutral register, which changes in seconds to 'father-to-daughter' expressed in neutral (towards intimate) family register; always according to the majority consensus of our native collaborators.

Examples such as this, in which there is a high level of fluctuation in our range of relational identities, communicative roles and linguistic registers, constantly happen in our daily communication, probably in most cases without us noticing. The following pages will show how it occurs in current online forum interaction.

Forum Communication

In the past fifteen years ODF (Online discussion forum) has proven a highly ef fective instrument in the teaching world, especially in the language teaching en vironment when using methodological approaches based on online learning (see Hammond, 1999; Ruiz-Madrid, 2006; Montero, Watts and García-Carbonell, 2007; Balaji and Chakrabarti, 2010; Anderson, 2014).

However, most of the studies that have focused on the nature and behaviour of forums in current online communication come from the field of computational linguistics. Within this field we find studies on text segmentation (Fragkou, 2014) and on latent user groups and filtering techniques (Qu and Liu, 2012). For ex ample, in the latter, a new system is proposed that can intelligently recommend threads from online community according to a user's interest. The system uses both content-based filtering and collaborative filtering techniques. In content-based filtering, we solve the problem of data scarcity in online content by smooth ing, using latent topic information. In collaborative filtering, we model users' par ticipation in threads with latent groups under an LDA framework. The two sys tems complement each other and their combination achieves better performance than individual ones. Their experiments across different forums demonstrate the robustness of our methods and the difference among forums.

Within the abovementioned field we also found the study of Chan, Hayes and Daly (2010) on forum user 'roles'. These authors presented an empirical statistical analysis of user communication 'roles' in a medium-sized bulletin board. They analysed the composition of 20 forums in terms of these roles, demonstrating similarities between forums based on underlying user behaviour rather than topic. In their analysis, the authors found that forums are typically composed of eight behaviour types such as: joining conversationalists, popular initiators, taciturns, supporters, elitists, popular participants, grunts and ignored (Chan et al., 2010: 2017). Their analysis also shows that depending on the topic of the forum (e.g. personal issues, Christianity, weather, development, humanities, accommodation, etc.) the clusters in which these roles appear varies, although there is certain degree of consistency in the presence of taciturns, grunts and supporters in a great ma jority of the corpus analysed. However, the meaning of 'role' in this study seems to be purely functional and technical or strategic as it is related to the strategic function that each user plays in the technical development of the forum, not on the identities that they express to develop the contents of the suggested topic. As their perspective is not linguistic, the study does not provide much information about the way these roles are expressed or the language they use to communicate their identities.

There are not many studies from discourse analysis, pragmatics or functional systemic linguistics. Ehrhardt (2012) analysed politeness and face work in German forum communication and observed that forum users seem to be more tolerant to face-threats than speakers in face-to-face conversations, always taking for granted a 'kind of default face' that all participants are expected to respect (although this is not always the case). From the previous studies we can also extract useful in formation such as the fact that a thread in an online forum typically contains several topics (Qu and Liu, 2012: 371) and that users participating in a thread seem used to infer the topic distribution of the thread and also infer the 'best' or more 'interesting' by amount of target user's participation (Qu and Liu, 2012: 374). It is worth highlighting the study of Delahunty (2012) on identity in on line discussion forums. According to her, users (students in her study) project an image of themselves negotiating their positioning within the group and therefore constructing their identities through the interpersonal process. Thus the forum becomes a process of multiple identity construction - as far as identity assump tion is concerned, it is of interest the work by Grant and Macleod (2016). This genre and its development of asynchronous discussion also becomes a platform or vehicle of collaborative knowledge building (Schire, 2006). In fact, specialists suggest that cultivating interpersonality online leads to increased participation and expands the depth of discussion, thus facilitating online collective knowledge building (Beuchot and Bullen, 2005).

Regarding types of forums, on websites we may find three main types of fo rums which most online applications, such as Moodle (Dougiamas, 2010:4), use:

  • Standard forums for general use. These are forums for large discussions where members can propose new topics and create new threads, while clicking on a list of previous topics to read and join other threads and re spond other participants' postings.

  • Single simple discussion forums, which involve short/time-limited discus sion on a single topic.

  • Q&A forums, which maintain a format similar to the popular 'question and answer' websites.

Although some applications, such as the above mentioned Moodle, add to these three types other forum versions which might be interesting in a certain context (for example, 'Each person posts one discussion forum'). Hence, there are other more punctual typologies - for example, Barak and Gluck-Ofri (2007) differentiate between support forums and discussion forums highlighting the oc currence and reciprocity of self-disclosure mechanisms in both of them.

That being said, and regarding identity construction in online interaction, our focus is not on how identity is built and developed throughout the interac tion that takes place, but on how participants in forums currently express their relational identities via their language choices and their RV.

2. CORPUS AND METHODOLOGY

Given the great amount of forums we can find on the Internet today, we needed to follow some selection parameters to collect our sample corpus. However, while carrying out this selection process emerged some methodological considerations.

The first parameter concerns the type of forum. For this piece of research we decided to concentrate on standard forums for general use with the format of mere opinion or discussion forum in which a participant proposes a topic for open discussion in order to gather different views on the subject (Montero et al., 2007). The first consideration that emerges is that, although they all seem to have this specific goal a priori (i.e. to collect opinions), the threads analysed often have other underlying objectives (i.e. need to make decisions, solve problems, correct mistakes, express feelings, receive comfort, improve skills, etc.). These underlying purposes largely influence: (a) the jointly set up virtual context in which a thread develops, (b) its overall similarities with other communicative genres (e.g. politi cal debate, legal advisory meeting, group therapy meeting, expert consultation, cooking show, etc.), and consequently (c) the roles/identities emerging through that thread. In this regard, we have observed that this type 'opinion or discussion forum' is actually a 'multi-purpose forum' which progressively shifts into a sup port forums, learning forum, entertainment forum, therapeutic forum, advisory forum, etc. So this type is very wide-ranging.

The second parameter concerns accessibility. It was not difficult to access threads where the salient registers were the family register or amicable register. However, as it is necessary to apply for membership to participate in professional forums, the access to threads with pivotal professional register was more complex. Eventually we managed to access forums focused on activities related to the aca demic profession.

Regarding the length of the threads, we noticed a significant difference in length between similar forums in terms of number of posts and users participat ing in the thread, and also in the length of their contributions. Finally we decided to include only forums with at least 10 different users and 15 posts with at least one full content unit. Proper consideration was given to the ethics issues raised.

Bearing these parameters and considerations in mind, the following three threads with different pivotal registers were eventually selected for the analysis in detail:

Table IV Selected threads for analysis. 

The analysis was carried out in two stages: the first aimed at RMR detection and labelling with the collaborators, and the second focused on our linguistic analysis based on their observations. Being consistent with our approach to RV, we searched for the help of six native English collaborators in the identification of pivotal registers, RMRs and roles: two undergraduate students (in their 20s), two postgraduate students (in their 30s) and two professionals (in their 40s). We purposely sought for different profiles, so that interferences caused by factors such as occupation and chronolectal variation were minimised. They had experience in language learning and basic linguistic description skills. In this first stage, our function was informative, orientative and explanatory, but not cooperative in their RMR detection process. They were asked to spot instances that they would associ ate with a certain type of interlocutor or communicator (e.g. a family member, friend, workmate, etc.) in a particular context (in the kitchen at home, at school, at work, etc.), trying to describe them as thoroughly as possible. We assumed the role of covert non-participant observers so that the observer effect was avoided. It should be noted that spending enough time interacting with our collaborators to get the needed data was necessary and that we had to deal with different restric tions, such as our collaborators availability or the fact that each working session could not last long since it involved a great mental effort from them. After this phase, we started our process of linguistic register analysis, classification and in terpretation. Altogether, the whole process was extended for about two months.

3. RESULTS AND DISCUSSION

The results of the analysis are presented below. First, the salient linguistic features in each pivotal register will be shown, paying special attention to the openings and closings of the messages since they are noteworthy moves in the analysis of conver sational genres, especially in the detection of the type of communicative roles cho sen by their authors (Paltridge, 2000: 86). Second, each one of the three forums selected for our study will be contemplated. Additionally, the register-modulating roles and pivotal registers in each forum will be considered and illustrated with some examples.

Salient linguistic features in each pivotal register

Regarding the register analysis, in order to identify the pivotal register we con centrated on the opening and closing of each post, and we also observed some recurrent features in the body of the messages. For example, the omission of the salutation is a characteristic feature of most posts in forums, especially in those pivoting around the family register. However, when we see the all openings used by the users in a same thread, it seems quite easy to identify what type of openings belongs to each register.

Table V Openings used by the users. 

As it can be seen in Table V, in Forum 1 (family neutral) the most recurrent element (15 %) to open a conversation is an exclamation. This is not the case in Forum 2 (amicable neutral), where the hello+name option appears as the most repeated (30 %) option. Finally, in Forum 3 (professional casual) the situation is, once again, different, with three possibilities as the most frequent (18 % each) ones: introductory statement with name, introductory praise with name and in troductory sentence.

Table VI Closings used by the users. 

As far as closing formulas are concerned, the Table VI shows that, in Forum 1 (family neutral) icons are used in 37 % of all cases. Forum 2 (amicable neutral) resorts most frequently (18 %) to the take care (+name/+kiss) option. Regarding Forum 3 (professional casual), a conclusion statement is the most repeated (23 %) selection.

As it has been seen, there is a clear relationship between the openings and clos ings in posts pivoting around each register. Let us now consider each one of the three forums selected for our study.

Register-modulating roles and pivotal registers

FORUM 1

Pivotal register: family neutral

Forum B: 'Family meals_simple recipes wanted ... I'm stuck in a rut (thread of 41 posts, 8 users)

From gingerbread.org.uk

General features: in this register the input is more similar to sms and mobile text messages (e.g. 'have you tried to make it?'), sharing their main features: short length, shortenings, icons, high presence of emotional/subjective language with exclamation marks and direct questions with question marks, no capital letters, lack of punctuation marks and so on (Tagg, 2009). There are endings which com bine many options such as im waiting huff lol icon' (MA) (ellipsis+indirect question+exclamation+icon). In these registers we find many posts with missing and unconnected information as their function does not seem informative but, following Jakobson (1963), rather emotive (providing information about the user's internal state), conative (trying to engage other users) or phatic (trying to maintain the communication channel/the forum working).

In the Table VII we see some examples from Forum 1. As indicated, the fam ily neutral register appears as pivotal. Indications about the concrete roles are also provided:

Table VII Registers, roles and examples (Forum 1 - sic). 

Forum 2

Pivotal register: amicable neutral

Forum A: 'Your relationship with your child's other parent' (thread of 28 posts, 15 users)

From gingerbread.org.uk

General features: the most frequent register seems to be amicable neutral (e.g. salutation with just 'Hello' or 'Hi), swinging on some occasions between a more formal register typically used with acquaintances or distant friends and a more intimate register, used with close friends, even relatives in some cases (family reg ister). See Table VIII for some examples:

Table VIII Registers, roles and examples (Forum 2 - sic). 

Forum 3

Pivotal register: professional casual

Forum C: 'University lecturing?' (17posts, 12 users)

From singletrackworld.com

General features: the posts are longer with higher information density (with a few exceptions of shorter messages, the information is structured and provided in lists, and there are less spelling and grammar mistakes, less abbreviation and eli sion. However, we find frequent exclamation and question marks, also the use of (...), still instances of bad punctuation, asides and icons, swear words like FFS or (***) and emphatic language. In Table IX we offer some relevant examples:

Table IX Registers, roles and examples (Forum 3 - sic). 

Table X summarises the different register-modulating roles in our corpus of analysis:

Table X Register-modulating roles. 

As it has been shown, some differences can be detected regarding the salient linguistic features in each pivotal register. Pivotal register family neutral shows a preference to open conversations by means of an exclamation. This is not the case of amicable neutral, where the hello+name option appears as the most repeated one, whereas professional casual seems to resort on a frequent basis to as many as three possibilities: introductory statement with name, introductory praise with name and introductory sentence. As far as closings are concerned, pivotal register family neutral typically makes use of icons, while amicable neutral most frequently employs the take care (+name/+kiss) formula and professional casual shows a con clusion statement as the most repeated choice. Hence, a clear correlation between the openings and closings in posts pivoting around each register can be detected.

Regarding the register-modulating roles and pivotal registers in each one of the three forums, they range from maternal/paternal figure, big sister and daughter/ little sister (family) in Forum 1 to acquaintance and (close) friend (amicable) in Forum 2 and student, junior/inexperience teacher, workmate, colleague and se nior/experienced worker (professional) in Forum 3.

As the results show, in these apparently uncomplicated forums in terms of roles, where the presumed dominant RMR a priori seems to be that of "an ordi nary citizen who expresses his/her opinion about a topic of interest", at a linguistic and cognitive level, many other relational identities emerge through which the fo rums' RV fluctuates. The recognition of these roles may help to better understand the intentions of the forum participants and communicate with them in a more natural and effective way. The study therefore reinforces the interdependence be tween roles and registers in digital discourse, and provides some new conceptual and methodological devices to approach RV in forum discourse.

Much work remains to be done to discover the multidisciplinary theoretical and practical foundations of RV, as it currently operates in human communica tion across languages and cultures, not only through forum interaction but also in other genres and communicative platforms. Another pending challenge is the development and application of this theory to the diverse educational and pro fessional environments in which RV unawareness and lack of competence are generating issues and confusion in communication. Many students of modern languages and professionals with an international profile would appreciate and use new materials, including software tools, that could give them access to RV cor respondences between their mother tongues and other target languages. In order to create these innovative facilities, it would be necessary to collect representative samples of communicative encounters taking place in family, amicable, social and professional settings and generate large corpora that would allow us a compre hensive analysis of RV. Due to the fact that this is such a complex and expensive enterprise we are making every effort to further develop this RV approach little by little, often against all odds.

4. CONCLUSION

Modern professional, social and personal dynamics sometimes lead us to situa tions in which the mixture, ambiguity and/or confusion between roles can hinder communication and even generate conflicts at different levels. As Shumate and Fulke (2004) indicate, some telecommuters find difficult to separate and differ entiate between professional and domestic roles and identities, needing help to regulate and express their roles' boundaries. This type of findings show that the world of communication and language teaching has a demand on linguistic analy ses which approach the interrelationship among identities, roles and registers from a wide but practical angle which incorporates and applies insights from other border knowledge areas related to human communication. The development of interdisciplinary joint approaches to RV across the diverse situations in daily com munication might help establish a clearer relationship between 'who we think we are' (identities), 'how we act as being' (roles) and 'how we express what we are' (registers).

The present paper proposes and illustrates a method that may be useful for the detection and analysis of these variables in diverse communicative genres within current online communication. The results display the variety of identities and roles that interact in forum interaction and highlight the main linguistic features used in those forums where the pivotal registers are family, amicable and profes sional registers. They also reinforce the fact that RV in digital discourse also de pends on the participants' register compass that helps them select which linguistic options are most appropriate and effective for expressing the diverse identities and roles they want to perform.

The study is only based on a sample of the model that we propose, but we believe that the analysis of a wider corpus could be useful to identify other corre spondences and to observe the different linguistic options used by users to express them. The results from the implementation of the model explained here could facilitate comprehension and communication by students, non-native English speakers and new platform users. Besides, the control of RV may improve their confidence, consistency, efficacy and interpersonal cohesion.

REFERENCES

Anderson, Chris J. 2014. Use discussion board forums to promote active student learning. UCAN. Available at http://www.ucan-cja.blogspot.com. [ Links ]

Auer, Peter. 2002. Code-Switching in Conversation: Language, Interaction and Iden tity. London: Taylor & Francis. [ Links ]

Balaji, M. S. and Diganta Chakrabarti. 2010. Student Interactions in Online Dis cussion Forum: Empirical Research from "Media Richness Theory" Perspec tive. Journal of Interactive Online Learning, 9(1), 1-22. [ Links ]

Barak, Azy and Orit Gluck-Ofri. 2007. Degree and Reciprocity of Self-Disclosure in Online Forums. CyberPsychology & Behavior, 10(3), 407-417. [ Links ]

Baumeister, Roy F. (ed.) 1986. Public Self and Private Self. New York: Springer. [ Links ]

Bayley, Robert. 2013. Variationist Sociolinguistics. Robert Bayley, Richard Cameron and Ceil Lucas (eds.) The Oxford Handbook of Sociolinguistics. Oxford: Oxford University Press, 11-30. [ Links ]

Beuchot, Albert and Mark Bullen. 2005. Interaction and interpersonality in on line discussion forums. Distance Education, 26(1), 67-87. [ Links ]

Biber, Douglas. 1995. Dimensions of Register Variation: A Cross-Linguistic Com parison. Cambridge: Cambridge University Press. [ Links ]

Biber, Douglas. 2012. Register as a predictor of linguistic variation. Corpus Lin guistics and Linguistic Theory, 6(1), 9-37. [ Links ]

Biber, Douglas and Edward Finegan. (eds.) 1994. Sociolinguistic Perspectives on Register. New York: Oxford University Press. [ Links ]

Biber, Douglas and Susan Conrad. 2009. Register, Genre and Style. New York: Cambridge University Press. [ Links ]

Bonnin, J. E. 2013. The public, the private and the intimate in doctor-patient communication: Admission interviews at an outpatient mental health care ser vice. Discourse Studies, 15(6), 687-711. [ Links ]

Brown, Roger and Albert Gilman. 1960. The pronouns of power and solidarity. In Thomas A. Sebeok (ed.) Style in Language. Cambridge, MA: MIT Press, 253-276. [ Links ]

Butler, Judith. 1990. Gender Trouble: Feminism and the Subversion of Identity. New York: Routledge. [ Links ]

Chan, Jeffrey, Conor Hayes and Elizabeth Daly. 2010. Decomposing discussion forums and boards using user roles. Proceedings of the Fourth International AAAI Conference on Weblogs and Social Media. Washington, D.C., 215-218. [ Links ]

Chen, Serena, Helen Boucher and Michael W. Kraus. 2011. The Relational Self. In Seth J. Schwartz, Koen Luyckx and Vivian L.Vignoles (eds.) Handbook of Identity Theory and Research. New York: Springer, 149-176. [ Links ]

Conrad, Susan and Douglas Biber. 2001. Variation in English. Multi-Dimension al Studies. CRAT: Corpus Research Analysis Tool. Available at http://perezparedes.blogspot.com/search/label/CRAT. [ Links ]

Delahunty, Janine. 2012. "Who am I?": Exploring identity in online discussion forums. International Journal of Educational Research, 53, 407-420. [ Links ]

Dougiamas, Martin. 2010. Using Moodle in ESOL writing classes. The Electronic Journal of English as a Second Language, 13(2), 1-11. [ Links ]

Ehrhardt, Claus. 2012. Politeness and face work in German forum communica tion. In Kristina Bedijs, Gudrun Bachleitner-Held and Christiane Maafi (eds.) Face Work and Social Media. Berlin: LIT, 83-107. [ Links ]

Fragkou, Pavlina. 2014. Text Segmentation for Language Identification in Greek Forums. Procedia - Social and Behavioral Sciences, 147, 160-166. [ Links ]

Giménez-Moreno, Rosa. 1997. The boundaries of the concepts of genre, register and style in Academic Discourse. Jordi Piqué and David J. Viera (eds.) Applied Languages: Theory and Practice in ESP. Valencia: Universitat de Valência, 37- 45. [ Links ]

Giménez-Moreno, Rosa. 2006. A new approach to register variation: the missing link. Ibérica, 12, 89-110. [ Links ]

Giménez-Moreno, Rosa. 2010. English for higher education studies. Rosa Gi ménez Moreno (ed.) Words for Working. Professional and Academic English for International Business and Economics. Valencia: PUV, 295-334. [ Links ]

Giménez-Moreno, Rosa. 2011a. Register variation in international business cor respondence. International Journal of English Studies, 11(1), 15-34. [ Links ]

Giménez-Moreno, Rosa. 2011b. Role fluctuation in professional English reg ister variation. Ma Luisa Carrió, Josefa Contreras, Françoise Olmo, Hanna Skorczynska, Inmaculada Tamarit & Debra Westall (eds.) La Investigación y la Enseñanza Aplicadas a las Lenguas de Especialidad y a la Tecnología. Valencia: Universitat Politêcnica de Valência, 71-79. [ Links ]

Giménez-Moreno, Rosa and Hanna Skorczynska. 2013. Corpus analysis and reg ister variation: A field in need of an update. Procedia - Social and Behavioral Sciences, 95, 402-408. [ Links ]

Giménez-Moreno, Rosa and Hanna Skorczynska. 2015. Family register in British English: the first approach to its systematic study. Procedia - Social and Behav ioral Sciences, 173, 222-227. [ Links ]

Goffman, Erving. 1967. Interaction Ritual: Essays in Face to Face Behavior. Garden City, NY: Anchor Books. [ Links ]

Grant, Tim and Nicci Macleod. 2016. Assuming Identities Online: Experimen tal Linguistics Applied to the Policing of Online Paedophile Activity. Applied Linguistics, 37(1), 50-70. [ Links ]

Gregory, Michael and Susan Carroll. 1978. Language and Situation. Language Va rieties and their Social Contexts. London: Routledge & Kegan Paul Ltd. [ Links ]

Gumperz, John J. and Dell H. Hymes. 1986. Directions in Sociolinguistics: The Ethnography of Communication. Oxford: Basil Blackwell. [ Links ]

Halliday, Michael A. K. 1980. Register variation and identity of a text. Sophia Linguistica, 6, 60-79. [ Links ]

Halliday, Michael A. K. 1985. Spoken and written language. Oxford: Oxford Uni versity Press. [ Links ]

Halliday, Michael A. K. 1988. On the Language of Physical Science. In Mohsen Ghadessy (ed.) Registers of Written English: Situational Factors and Linguistic Features. London: Pinter Publishers, 162-178. [ Links ]

Hammond, Michael. 1999. Issues associated with participation in on line forums -the case of the communicative learner. Education and Information Technolo gies, 4(4), 353-367. [ Links ]

Jakobson, Roman. 1963. Essais de linguistique genérale. Paris: Minuit. [ Links ]

Montero, Begoña, Frances Watts and Amparo García-Carbonell. 2007. Discus sion forum interactions: Text and context. System, 35(4), 566-582. [ Links ]

North, Sarah. 2007. "The Voices, the Voices": Creativity in Online Conversation. Applied Linguistics, 28(4), 538-555. [ Links ]

Paltridge, Brian. 2000. Making Sense of Discourse Analysis. Queensland: Gerd Sta bler, AEE Publishing. [ Links ]

Qu, Zhonghua and Yang, Liu. 2012. User participation prediction in online fo rums. Proceedings of the 13th Conference of the European Chapter of the Associa tion for Computational Linguistics . Avignon, 367-376. [ Links ]

Reicher, Steve. 2000. Social identity definition and enactment: A broad SIDE against irrationalism and relativism. In Tom Postmes, Russell Spears, Martin Lea and Stephen Reicher (eds.) SIDE Issues Centre Stage: Recent Developments in Studies of De-individuation in Groups. Amsterdam: Royal Netherlands Acad emy of Arts and Sciences, 175-190. [ Links ]

Ruiz-Madrid, M. Noelia. 2006. The role of forums in the construction and de velopment of the teaching-learning discourse in the ESP classroom. A new autonomising tool within the language teaching-learning context? Anglogermanica Online, 27-43. [ Links ]

Schrire, Sarah. 2006. Knowledge building in asynchronous discussion groups: Going beyond quantitative analysis. Computers & Education, 46(1), 49-70. [ Links ]

Sedikides, Constantine and Marilynn B. Brewer. 2001. Individual Self, Relational Self, Collective Self. Philadelphia: Psychology Press. [ Links ]

Shumate, Michelle and Janet Fulk. 2004. Boundaries and role conflict when work and family are colocated: A communication network and symbolic interaction approach. Human Relations, 57(1), 55-74. [ Links ]

Skorczynska, Hanna and Rosa Giménez-Moreno. 2017, in press. Analysing meta phor in the family register through sitcom scripted conversations. Journal of Metaphor and Social World. John Benjamins. [ Links ]

Tagg, Caroline. 2009. A Corpus Linguistics Study of SMS Text Messaging. Unpub lished PhD. Thesis presented in the Department of English of Birmingham University on March 2009. Accessed from the E-Theses Repository of the University of Birmingham. [ Links ]

Tajfel, Henri and John C. Turner. 1986. The social identity theory of inter-group behavior. In Stephen Worchel and William G. Austin (eds.) Psychology of Inter-group Relations. Chicago: Nelson-Hall, 33-48. [ Links ]

Van Dijk, Teun A. 2004. Discourse, knowledge and ideology. In Martin Pütz, JoAnne Neff and Teun A. van Dijk (eds.) Communicating Ideologies. Multi-disciplinary Perspectives on Language, Discourse and Social Practice. Frankfurt/ Main: Peter Lang, 5-38. [ Links ]

Van Dijk, Teun A. 2006. Discourse, context and cognition. Discourse Studies, 8(1), 159-177. [ Links ]

Van Dijk, Teun A. 2008. Discourse and Context. A Sociocognitive Approach. Cam bridge: Cambridge University Press. [ Links ]

Wetherell, Margaret and Chandra T. Mohanty (eds.) 2010. The SAGE Handbook of Identities. London: Sage. [ Links ]

Appendix. Survey and Instructions for Informants

PERSONAL INFORMATION: MOTHER TONGUE: …………………..OTHER LANGUAGES: PLACE OF BIRTH: …………………….CURRENT PLACE OF RESIDENCE: AGE: ……………………………….….GENDER: OCCUPATION: TASK 1: Please, classify in the table the following list of standard roles and re lational identities according to its associated communicative setting. Please add any interesting comment about your own experience when you adopt (or have adopted) these roles or when other people communicate with you adopting these roles / identities.

TASK 2: Please, complete each setting in the table with other roles/identities that you think are also associated, or tend to be associated, with them in current com munication.

Received: August 02, 2017; Accepted: October 30, 2017

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License