Facebook Posts as Performatives in Bulgarian Political Life: A Sociolinguistic Investigation

This work was supported by the Bulgarian Ministry of Education and Science. Competing interests: no competing interests have been declared. Publisher: Institute of Slavic Studies, Polish Academy of Sciences. This is an Open Access article distributed under the terms of the Creative Commons Attribution 3.0 PL License (creativecommons.org/licenses/by/3.0/pl/), which permits redistribution, commercial and non-commercial, provided that the article is properly cited. © The Author(s) 2019. Nelly Tincheva Sofia University “St. Kliment Ohridski” http://orcid.org/0000-0001-7383-5939

on non-verbal, bodily aspects of communication such as gestures, facial expressions, quality of voice, length of pauses, etc. (see e.g. Brown & Yule, 1983;. Moreover, linguistically inclined investigations of performatives, similarly to performing arts-associated studies, have also demonstrated extensive interest in communicative functions such as evading a direct answer and lying . In other words, they also focus on 'doing' and on the distinction between reality and fiction. Whatever the viewpoint on performativity, however, there are two generalizations to be made regarding the very notion as well as the way(s) it has been approached analytically. First, in its language -and discourse-directed interpretation, the notion of performing is predominantly academic, and -to the best of my knowledge -has not been studied as a general-audience concept. Second, performativity, especially in its language -and discourse-related interpretation, has not been studied quantitatively.
The study presented here addresses the research niche delineated by those two combined generalizations. It aims to quantitatively investigate generalaudience interpretations of the main concepts pertaining to the notion of 'performing': the concepts of 'doing', 'reality', 'fictional reality' and 'expressing oneself publicly'. More specifically, the study will focus on the genre of the Facebook post as a central modern instrument for understanding 'doing', 'reality', 'fictional reality' and 'expressing oneself publicly'.
The choice of this specific genre is based on the results of a previously conducted sociolinguistic study that will be discussed in the next Section. Section 3 will offer data obtained from a questionnaire involving 120 respondents and their answers to 8 questions. It will also discuss the data obtained through the questionnaire.

Previous Research and Theoretical Framework
The analytical point of departure and prototypology The point of departure for the present investigation is a sociolinguistic study I conducted which focuses on Bulgarian perceptions of genres in political discourse . Its main objective is to outline trends in Bulgarian society's assumptions and expectations of the communicative efficiency of genres in political discourse. That (initial) study includes 120 participants who were asked to fill in a questionnaire and provide their own evaluations of a list of genres. After answering the questions, the respondents were also invited to provide any comments or background information they saw as relevant.
The results obtained in that study reveal two peculiarities which prompted the present investigation. The first one is the un-hypothesized yet emphatic presence of the Facebook status as a genre of significant communicative importance in Bulgarian political life. The second peculiarity is associated with the additional comments and background information the respondents provided after answering each question. A significant number of those comments contain references to the performing -and performativity-related concepts singled out above (i.e. 'doing', 'reality', 'fictional reality', 'expressing oneself publicly'). Hence the decision to conduct a second one, i.e. a follow-up study, and thus place the results of the original study in a broader perspective. The present paper reports on that follow-up study.
The original study from which the present investigation departs involved 120 respondents differentiated along four main parameters as follows: (a) Educational background. The respondents' population included 20 linguists, 20 engineers, 20 economists, and 60 school graduates without a university degree; (b) Age. The respondents' population included 36 respondents from 21 to 34 years of age; 32 respondents aged between 35 and 45; 20 respondents aged between 46 and 59; and 32 respondents aged between 60 to 75; (c) Gender. The respondents' population included 64 female respondents and 56 male respondents, (d) Residence. The respondents' population included 65 city residents and 55 who live in a town or a village.
The questions in that study's questionnaire were formulated so as to follow a communicative-function-based type of progression. The progression rests on the premise that communicative text functions operate prototypically (term as in  rather than in terms of clear-cut, departmentalized categories (see, e.g., . In other words, text functions in the original study  are represented on a scale which ranges from most passive reception to the most active communicative involvement. Along this scale, text functions frequently overlap. Importantly, the functions selected for inclusion in that study associate with the text addressee's viewpoint in the process of communication. In other words, while answering the questionnaire questions, the respondents in the Study were asked to identify with the text receiver in a hypothetical political communicative act. The study presented here (i.e. the follow-up study) also adheres to the prototypology-related principles of overlapping and scalarity in communicative textual functions. However, in contrast to the original study, which adopted the viewpoint of the text receiver , the present study adopts the viewpoint of the text producer. Even more simply put, while answering the questionnaire questions the respondents in the Study reported here are asked (a) to imagine themselves as participants in a hypothetical political communicative situation, and (b) to identify with the text-producer role in that political communicative situation.
As far as the formulation of the questions in the present Study's questionnaire is concerned, they derive from the additional comments and background information provided by the participants in the original Study. In those comments the respondents attested to the role of the Facebook status genre in political communication as one that allows them the freedom of introducing themselves to various and new audiences. This genre, some of the comments maintained, also gives them a 'real sense of community' and a confidence in their own 'political empowerment'. Above all, the comments singled out this genre as the only one within political discourse which has the potential to help them 'change the system'.
Such comments appeared in roughly 8% of the completed questionnaires. On the one hand, the number of such comments suggests what the respondents shared can be of general validity; on the other hand, the (possibly) general validity of the comments needs to be supported by opinions across the whole group of respondents. Therefore, the present investigation: (a) hypothesizes the general validity of the existing comments, and (b) sets out to test the validity of those comments against all of the 120 members of the follow-up study's respondents' group.

The Facebook post as a genre
A theoretical point which might need clarification concerns the very status of personal Facebook contributions as a separate genre. In choosing to treat Facebook statuses, comments, and, generally, posts as a separate genre, I will uphold the theoretical view expounded in, for instance, Page (2010), ,  and . "In the 21st century, Bakhtin may have well added the Facebook status update, the blog post, the comment on an online newspaper story, TV quiz shows, and more, as forms of speech genre", writes Whitworth (2014, p. 125). In a similar vein, Palmquist agrees it is the internet that has led to the redefinition of many genres; he argues that in present days "[w]riters are living, in the fullest sense of the ancient Chinese proverb, in interesting times. Not since the fifteenth century, when Gutenberg perfected a workable system of movable type, has there been such a change in how information and ideas are exchanged" (Palmquist, 2005, p. 219).
Generally, the prevailing modern view on the topic can be summarized in what Abercrombie characterizes as the "steady dismantling of genre" (1996, p. 45). Hence my present interest in the upcoming and socially transformative 'new genres' such as Facebook posts.

'Politics', 'political discourse', 'performing in politics'
A last point which might need clarification concerns the way this paper's key notions of 'politics' and 'political discourse' are employed for the present purposes.
To start with, it seems rather important to note that the sociolinguistic nature of the present investigation (which by definition involves actual language users of various social backgrounds) requires that each notion employed in the investigation is commonly known and not of strictly technical or strictly academic use. In other words, the terms and notions employed in the present investigation need to be clarified but from a statistically verifiable culture-based viewpoint. Thus, employing any corpus-driven source (e.g. the Longman & Macmillan Dictionaries) would seem a reasonable starting point, as such sources register real-life users' authentic conceptualizations and actual discourse preferences.
Importantly, employing such sources (and these two sources specifically) is not a new idea when it comes to analyzing actual language use. For instance, the Macmillan Dictionary lies at the center of all the analytical techniques employed in the so-called MIP approach (i.e. the Metaphor Identification Procedure) that was proposed by the Pragglejaz . What is more, the MIP's later version of MIPVU (e.g.  uses both the Macmillan Dictionary and secondary references to the Longman Dictionary of Contemporary English and to the Oxford English Dictionary. The present investigation follows such corpus-driven cross-check procedures.
Going back to the central notion of 'politics', any probe across corpusderiving sources will establish that the notion is invariably conceptualized along the following lines: (1) 'Politics' may refer to an individual's (personal) ideas, beliefs, attitudes, and activities involved in the process of acquiring and exercising power in a country, community, or a particular area of the world; (2) 'Politics' may refer to ideas, beliefs, attitudes, and activities involved in the process of a particular group of people acquiring and exercising power in a country, community, or a particular area of the world; (3) 'Politics' may refer to an individual's (personal) ideas, beliefs, and attitudes about how a country, community, or a particular area of the world should be governed; (4) 'Politics' may refer to the study of (1), (2), and/or (3).
As the definitions reveal, the notion of 'politics' rests upon two major conceptual factors. First, the notion may vary in dependence with scope (i.e. it may encompass a single individual, a group of people, a community, a whole nation, etc.). Second, the notion may vary in terms of vantage point: it may adopt the perspective of individuals who govern, or, more simply put, individuals who are in power; it may adopt the perspective of individuals who are being governed; it may view a political process, or political processes, from the perspective of an impartial academic. All these conceptual options are approached here as functioning prototypically (as argued above): they may overlap and share a considerable number of features. In actual language users' minds, they may represent fuzzy-boundary, non-prototypical cases.
What is more, in actual language users' minds, the conceptual options listed above, i.e. conceptual options which associate with the domain of politics, tend to overlap rather frequently with the domain of political discourse. As Wodak maintains, "[p]olitics necessarily includes persuasion, rhetoric, deceptive devices, and so forth" (2009, p. 578). In the literature, there is considerable agreement (Miller, 1991) that politics is about reconciling differences in society and that as a consequence politics is highly dependent on communicative practices such as discussing, persuading and bargaining. Consequently, differentiating between communicative functions and political functions can be a rather complicated activity. Due to the high degree of conflation between political practices and communicative practices, the two can prove extremely hard to delimit from each other.
However, if the perspective of prototypology is adopted, along with its major principles of overlapping and fuzzy boundaries, the issue of differentiating between the domains "may simply be a matter of emphasis" (Wilson, 2005, p. 399). Hence the heightened role of performativity in the joint operation of the domains of politics and political discourse: through conceptual overlapping, discourse about politics becomes more than simply describing a social phenomenon. As Diez maintains, participating in political communication means "tak[ing] part in the construction of the polity itself " (1999, p. 599).

Data and Discussion
As the study reported here is a follow-up study that draws and builds upon a previous investigation, it aimed to involve the same set of respondents as the original study did. However, for various scheduling reasons unrelated to the essence of the two Studies, 7 of the original respondents were unable to participate in the follow-up. Participants of similar social backgrounds were selected for inclusion. The resulting profiles of the 120 respondents involved in the present Study are represented along four main parameters as follows: a) Educational background: 20 linguists, 20 engineers, 20 economists, 60 school graduates without a university degree; b) Age: 36 respondents aged 21 to 34; 32 respondents aged 35 to 45; 20 respondents aged 46 to 63; 32 respondents aged 60 to 78; c) Gender: 61 female respondents; 59 male respondents; d) Residence: 65 live in a city; 55 live in a town or a village.
The participants were asked to complete a questionnaire and provide their own evaluations of a list of genres. While answering the questions they were expected to: 1. Identify with the text producer, 2. Provide information only about their immediate responses, 3. Provide evaluations drawing on their personal subjective preferences only. They were expected not to generalize about any common social/political views and/or behavior.
After answering each question, the respondents were also allowed to provide information they perceived as relevant to the question they were answering.
Following the original study, the follow-up questionnaire offers several genre options as possible responses to each question. However, as stated above the original Study required respondents to identify with the text receivers in the communicative situation, while the present study asks them to identify with the text producer. In both cases, the social context against which these identification alternatives exist is political life, and text producers in political discourse are generally politicians or media employees . Thus, identifying with political text producers was expected to be an ambiguous activity for the respondents as none of them were a politician and only 4 were professional journalists. In other words, there was a strong possibility that the genres in the original study would not fit the purposes of the present study. The genre of 'political speech', for instance, would have to be interpreted by the participants as having a far narrower audience scope than the speeches typically delivered by politicians. Nevertheless, to preserve the patterns and principles of the original Study and to make the results of the follow-up comparable to the original ones, the list of genres had to be kept the same as in the original Study.
In the participants' questionnaires, the genres were listed alphabetically to avoid any implications of priming techniques. In the tables included below, the genres appear in the same alphabetical order. The data on Question 1 demonstrate that when the 'express oneself' aspect of political performance is in focus and the text addressees are people with whom the text producer is familiar: a. Debate is the most frequently preferred genre. This positioning, combined with the speech genre, which came in 3 rd place, strongly suggests Bulgarians prefer face-to-face communication when expressing themselves politically in familiar environments. b. Second in frequency of preference comes the Facebook post. c. The 2 answers selecting the article genre as their communicative vehicle of choice were both given by professional journalists.  Table 2 The data on Question 2 demonstrate that when the 'express oneself' aspect of political performance is in focus and the text addressees are people with whom the text producer is not familiar: a. Facebook post is the most frequently preferred genre and comes first with more than 7 times the number of preferences as the genre in 2 nd place (debate). b. In contrast to the data obtained on Question 1, Facebook post demonstrates an increase of 40 preferences. This number represents a swing of as much as 1/3 of the total votes. Bearing in mind that the only parameter in Question 2 that differs from the parameters involved in Question 1 is the one that relates to text addressees being strangers, it can be argued that when expressing themselves politically to strangers Bulgarians prefer non-face-to-face communication. More specifically, in such situations they opt for the Facebook post genre and not for other mediated genres such as Twitter messages or TV/ radio interviews. c. In comparison to Question 1, the speech genre still occupies 3 rd position; however, it also registers a slump from 15 (Question 1) to 7 (Question 2) in the number of actual preferences. d. All 4 answers that indicate the article genre as their vehicle of choice in expounding political views to strangers are given by professional journalists. e. In contrast to Question 1, there are 10 'None of the above' responses.

Question 2. Which genre would you prefer if you wanted to express yourself politically to strangers?
Arguably, this choice indicates a reluctance to expound one's political views before strangers.  Table 3 The data on Question 3 demonstrate that when the 'perform within political reality' aspect of performing politically is in focus and the text producer initiates the community-creating process:

Question 3. Which genre would you prefer if you wanted to attract politically like-minded people?
a. Facebook post is the most frequently preferred genre and comes first with a lead of more than 6 times the number of preferences for the genre in 2 nd place (debate). b. The 1 answer that points to the article genre as the vehicle of choice is given by a professional journalist. c. There are 5 'None of the above' responses. Arguably, those choices point to a reluctance to initiate a political movement or a community-creating process.  Table 4 The data on Question 4 demonstrate that when the 'perform within political reality' aspect of performing politically is in focus and the text producer does not initiate the community-creating process:

GENRE NUMBER OF RESPONDENTS
a. Facebook post is the most frequently preferred genre and comes first with a lead of more than 28 times the number of preferences for the genres in 2 nd (speech) and 3 rd place (debate) combined. b. All genres other than Facebook post register very low numbers of choices and, thus, could not be interpreted as registering actual tendencies. c. The respondents who preferred speech and debate later commented that they prefer to "argue their way" into being accepted in a political community.  The data on Question 5 demonstrate that when the 'perform verbally within political reality' aspect of performing politically is in focus and the text producer wants to create a truthful rendition of that political reality: a. Facebook post is the most frequently preferred genre and comes first with a lead of more than 3 times the number of preferences for the genre in 2 nd place (debate). b. Debate is the 2 nd most frequently preferred genre. This positioning, combined with the 3 rd place of the speech genre, suggests a significant number of Bulgarians (roughly 30%) prefer face-to-face communication when they want to convince others on political issues. c. The numbers of choices registered by all other genres are very low and could not be interpreted as registering actual tendencies. d. There are no 'None of the above' responses. Arguably, to Bulgarian non--politicians (as mentioned above, the group of respondents includes no politicians) the communicative function of 'lying' does not overlap with the communicative function of 'convincing'.  Table 6 The data on Question 6 demonstrate that when the 'perform verbally within political reality' aspect of performing politically is in focus and the text producer does not want to create a truthful rendition of that political reality: a. Facebook post is the most frequently preferred genre here and comes first with a lead of almost 15 times the number of preferences for the genre in 2 nd place (debate).

Question 6. Which genre would you prefer if you wanted to deceive people about what the political reality is?
b. In contrast to the answers to Question 5 above (where a truthful rendition of reality was the task), in Question 6 (where a non-truthful rendition of reality is the task), there is a very significant rise in preferences for the Facebook post genre. Very simply put, Bulgarian respondents appear to find it considerably easier to lie about a political issue via Facebook than tell the truth via Facebook. In other words, performing untruthfully via Facebook proves considerably easier than performing truthfully via Facebook. This conclusion, however, should not be interpreted as a counter-argument against Facebook posts as they are still by far the most frequently preferred genre when rendering a political reality in general is the task. c. Debate is the 2 nd most frequently preferred genre here; however, the number of choices registered for it is very low and comparable to all other choices except the top-most one. d. All other genres but Facebook post demonstrate very low numbers of preferences. The numbers are so low that they could not be claimed to register tendencies.  Table 7 The data on Question 7 demonstrate that when the 'actually doing' aspect of performing politically is in focus: a. Facebook post is the most frequently preferred genre and comes first with a lead of almost 13 times the number of preferences for the genre in 2 nd place (debate).
b. All other genres but Facebook post demonstrate very low numbers of preferences. The numbers are so low that they could not be claimed to register tendencies. c. There are 17 'None of the above' responses. None of the other 6 questions so far has displayed a comparable number of 'None of the above' choices.
A possible interpretation would be that Bulgarian respondents would not like to see their political system change; however, their comments after this question strongly suggest that the reason for these 17 non-responses is disbelief that the Bulgarian political system can be changed.  Table 8 The data on Question 8 demonstrate that when the 'performative' aspect of performing politically is in focus: a. Facebook post is the most frequently preferred genre and comes first with a lead of more than 8 times the number of preferences for the genre in 2 nd place (debate). b. Debate is the 2 nd most frequently preferred genre. In comparison to Question 7 and in contrast to the Facebook post, which appears with exactly the same number of preferences, the number of preferences for debate increases. Arguably, for roughly 8% of respondents, face-to-face arguments count as political action. c. There are 13 'None of the above' responses. None of the other questions so far except Question 7 has displayed a comparable number of 'None of the above' choices. Arguably, for more than 10% of Bulgarian respondents, verbal communication does not classify as actual political action. This is in contrast to the roughly 74% of Bulgarian respondents who view the Facebook post as a truly performative political action genre.

Conclusion
The Study reported here provides a sociolinguistic, quantitative perspective on actual language users' perceptions of genres in Bulgarian political discourse. The presented results strongly suggest the Facebook post genre is seen in present-day Bulgarian society as a community-building instrument through which Bulgarians do not simply communicate but believe they actually perform politically.
The data obtained reveal the Facebook post genre performs most evenly across all the questions in the questionnaire. The debate genre appears consistently in second place across the questionnaire. These are the only two genres which are seen as so communicatively efficient as to represent actual political action(s).
Through the study reported here, the Facebook post is indubitably confirmed to serve the purpose of expressing oneself politically (to text addressees that the text producer is familiar with and to addresses that the text producer is not familiar with). It is also confirmed to serve the purposes of joining and initiating a new political community, movement, or group. The Facebook post is also confirmed to serve purposes of verbal communication when the creation of a non-truthful or truthful rendition of political reality is required. Above all, employing this genre proves that one is seen as 'actually doing' politics. All those conclusions combined characterize the Facebook post as the top genre that enables political action in Bulgarian political life.
Admittedly, the above conclusions derive from information provided by only 120 respondents. Nevertheless, the results are so unambiguous that it seems fully founded to argue that even if the numerical data obtained may not represent exact statistics and percentages, they still represent pronounced social tendencies in Bulgarian politics. Abercrombie, N. (1996).