Discourse analysis for social change: voice, agency and hope

1 Drinking mate without a bombilla

When we first started working here, back in ’94, the rulebook said “you may not drink mate with a bombilla”, right? So, in ’97, when we had become bolder, whenever the supervisor came along, we would remove the bombilla (laughs). When we saw him coming, we would hide it. He would say “you’re drinking mate”, and we’d reply “but without a bombilla!” (laughs).

Ana, who has worked for the Buenos Aires subway company for more than 25 years, recounts this anecdote to tell the story of how the subway workers founded their union. She could have described what their 12-hour working days were like, how they relieved themselves in the ticket office because they couldn’t go to the restroom, how their complaints were violently repressed. But instead, she chose to tell us that they used to take the bombilla out of the mate, making a mockery of supervisors and rulebooks alike.

Those of us who critically investigate the use of language need to learn from Ana and her peers – from their creativity, organization and sense of humor. In the current context of global crisis, we have the opportunity to design a research agenda for an uncertain future from a dark present. But there is no reason why that agenda should also be uncertain and dark.

An agenda thus established can start from three aspects that are present in this anecdote: the recognition and appreciation of multiple voices, organized and collective agency, and an unwavering and explicit bias for hope.

2 Voice

“Within the community of language (that is, within what is understandable: a limit right next to the infinite) each person’s duty is to find their own voice.” Jorge Luis Borges, The language of the Argentines (1928)

“Dentro de la comunidad del idioma (es decir, dentro de lo entendible: límite que está pared por medio de lo infinito) el deber de cada uno es dar con su voz”. Jorge Luis Borges, El idioma de los argentinos (1928)

There are multiple voices in Ana’s story: the voices of the rulebook, of the supervisor and of the workers who collectively reinterpret institutional regulations – literally – in order to resist them. These voices are a mise-en-scène of characters which, being discursive, do not lose historical substance. They belong to a foundational scene of the union’s history, which brings together many experiences in a single narrative. In this regard, as Ana is a spokesperson for the group, her colleagues’ voices resonate in hers. At the same time, however, her voice is unique. When I reread the transcript of the interview, I remember her laughter, intonation, gestures and other expressive dimensions that cannot be recovered in the written version. This is the aspect – individual and social, sociological and discursive – that I would like to analyze in greater depth.

The concept of voice, in the studies of language in use, has imprecise limits and broad conceptual dispersion, derived from the work of Bakhtin (1981, 1986). We can identify three different meanings.^[1]

The first meaning, present in discourse and narrative studies, designates the perspective and style of discursive characters, including the narrator himself (Wortham and Gadsden 2006: 318–319). Referred speech mechanisms introduce cited voices into a citing speech that arranges and organizes them (Reyes 1993, 1994; Clark and Van Der Wage 2003; Mey 2001). This perspective is paradoxically monological, since it is not concerned with the effective interaction between different subjects (oral, written or multimodal), but with their discursive representation in texts – usually written texts (cfr. Bolívar 2018).

Second, the concept of voice is frequently used to describe a “series of typeable semiotic regularities that indexicalize a context and a type of social subject, thereby enabling the reconstruction of socially available representations of classes of speakers and ways of speaking” (Bonnin 2019 a: 234). This definition is more common in the field of sociolinguistics, and is based on the distinction between “a diversity of social speech types (sometimes even diversity of languages) and a diversity of individual voices, artistically organized” (Bakhtin 1986: 262–263). Social voices enable us to understand the sociolinguistic dimension of discourse, since they describe indexical meanings linked to the sociocultural context of each event (cf. Agha 2005; Blommaert 2008; Strunck and Lassen 2011). Individual voices are frequently ignored, with voice and discourse being treated practically as synonyms (cf. Tovares 2016). The preferred methodological strategy is the analysis of deixis, which treats the voices of individuals as “textually individuated discursive figures that are typified through a system of person deixis” (Agha 2005: 45), in a sense that is practically identical to that of French theories of enunciation (Ducrot 1984; Maingueneau and Charaudeau 2002).

Third, however, it is possible to use the concept of “voice” without canceling the tension between the social and individual dimensions (Gumperz 1999; Hymes 1996). This type of analysis is especially productive in the field of ethnographic sociolinguistics (Codó et al. 2012), and defines the concept of voice from its understanding as a socially recognizable discourse – for example, Mishler’s (1984) voice of medicine and voice of the world – over which participants actively take agency, including emotions and evaluative orientations (Archakis 2014; Ribeiro 2006: 51).

This is the aspect I want to enquire into in greater depth. I understand that the individual voice is a set of semiotic regularities that indexicalize a biography: knowledge, resources and ways of speaking emerge from a singular experience of discursive practices (Bonnin 2019 b). From this perspective, individual voices are built on social voices to gain legitimacy, indexicalize authenticity, polemicize, etc. Voice, as an individual phenomenon, is thus rooted in social discourses.

In empirical terms, the individual voice is a unique interweaving of social discourses that signals the speaker’s biographical trajectory. Paraphrasing Bourdieu (1971), the voice is the representation that agents have of their distinctive position in social space and in relation to other agents. Thus, the individual voice is built by combining the capital that an individual accumulates in the social fields in which he or she participates: teacher, linguist, musician, etc. The most obvious way to index these distinctive capitals is through lexical choices, idioms, etc. that point to specific positions within those fields: leftwing within the political arena, human sciences within the academic field, jazz music within the artistic field, etc.

The focus on the participants’ biographies, on their trajectories through socially regulated spaces, enables the definition of a unique entry point to social relationships. From this perspective, social and biographical trajectory is not simply being socialized in practices, ideologies and discursive formations defined a priori, but rather contribute to (re)producing emergent discourses in everyday interactions. It is a social interpretation of the production of individuality, and an individual perspective on social relationships (Blommaert 2009; Bonnin 2019 a, 2019 b; cf. Otsuji and Pennycook 2010).

This does not mean that the unit of analysis is the individual’s speech. On the contrary, discourse is dialogical, and voice can only be analyzed in its effective interaction with other voices (Bolívar 2018; Voloshinov 2009 [1929]), with which it establishes relationships of closeness or distance, validation or rejection. Hence the need to analyze the phenomenon of co-construction of this voice, individual and social at the same time, in actual conversational contexts between participants with different and discursively recognizable biographical trajectories.

3 Agency

Ana’s story establishes two moments: one in 1994, “when we began”, and another in 1997, “when we had become bolder”. The prohibition to drink mate with a bombilla, which had been in place since 1994, is creatively reinterpreted in 1997 by the same workers (clearly indicated by the first-person plural). They propose not simply to disobey the regulation, but to act on it – on its literal text – collectively, by combining objects, words and actions to generate an alternative interpretation. When the group of individuals was organized into a collective, it became an agent of an action – drinking mate – and of a discourse of resistance.

The concept of “agency” – like that of voice – has a grammatical equivalent: it is the semantic role that initiates an action or event. Although often confused with grammatical subject, they are entities of different levels. Intuitively, the “agent” is associated with entities that have more or less “human” features: volition, perception or movement. Grammar, however, does not concede this humanistic argument: in “joy filled me”, “joy” is the agent of a process that affects me, even though it is an abstract entity.

In discourse studies and sociolinguistics, the concept is approached when discussing the interpretation of human behaviors – including communicative behaviors. On a theoretical level, nearly all contemporary currents of research on the use of language highlight the “dialectical” relationship between structure and agency in language policy and planning (Barakos 2016: 27; Cassels Johnson 2018; Unamuno and Bonnin 2018), in the relationships between discursive events and order of discourse (Fairclough 2006: 27), between discursive formation and enunciative scenes (Maingueneau 2004, 2014), etc. In contrast, on the level of empirical analysis and practical orientation of these investigations, the role of agency and change seems to be forgotten, in order to emphasize the place of the reproduction of an order of discourse. Although it is not an ontological distinction, but rather a distinction of focus (Tollefson 2012), the emphasis placed on the reproduction of structures rather than on the action that transforms them has a profoundly conservative effect. This is why I believe that we need to interrogate ourselves actively, empirically and programmatically about the role of discursive agency. Such interrogation can be based on three core ideas: agency emerges in interaction, it is not exercised in a homogeneous and intentional manner, and it cannot be limited to an agent defined in the terms of the humanist paradigm.

First, is agency a property of the individual? It is usually characterized as an individual’s ability to combine given elements in a novel way within the confines of an order of discourse (cfr. Blommaert 2005: 99; Fairclough 2006: 27). However, the individual speaker is an abstraction realized in a process that always necessarily involves others (Heller 2001: 260). Concepts such as “footing” (Goffman 1981) or “positioning” (Harré 2001) describe a bi-directional agency, in which participants define and condition each other, while being conditioned by dominant discourses that restrict certain options and enhance others. Thus, we are agents with and against others, mobilizing previous discourses – whether about gender, ethnicity or identity – even unintentionally.

In this interactional organization we observe “the emergence of structure out of agency” (Blommaert and Rampton 2011: 7), in its role as both reproducer and producer of identities, relationships and social processes. In the sociolinguistic field, the micro-macro distinction does not imply a “jump” from one level to another, but rather a continuum of layers of indexicality (Blommaert 2007; Mortimer and Wortham 2015; Silverstein 2000) which is more or less opaque (i.e., more or less transparent) to speakers. In this sense, Blommaert (2005: 160) proposes a “stratified” concept of ideology, with multiple dimensions and areas of operation, as well as different degrees of accessibility to consciousness and agency. That is to say, we can act on some levels and aspects of social discourse, but not on all of them simultaneously. This does not make the subject less an “agent”, because nobody “acts” on society as a whole. Fortunately, there is no puppeteer – no individual, institution, government or corporation – that can pull all the strings.

Finally, just as agency is not unidirectional and we are not homogeneously agents over an undifferentiated totality, agency is never concentrated, since it involves the distribution of skills, affordances and resources. Nor can it be seen as an exclusively human quality, as long as we abandon individual “intention” as an independent variable that explains behaviors (Pennycook 2019: 43). This conception of agency – which reminds us of the grammatical category that cannot be described in anthropomorphic terms – is extended and distributed through networks of entities and processes that go beyond the traditionally “humanist” limits of agency and cognition (Hayles 2012: 3). By erasing this radical distinction between the human and the non-human, agency becomes a distributed property that includes other objects and entities with which we need to be “attuned to”, “not just to the alterity of our interlocutors but also to the world of animals, objects and places” (Pennycook 2019: 106). Nature and the environment are agents, and they act on human societies: SARS-CoV-2 and climate change are some of the effects of this lack of attunement.

The posthuman perspective of agency is probably the most radical, and one worth developing in an upcoming agenda of research, action and activism. If we want to analyze and understand how people change the world, we must understand how they do it with others; how they interact with nature, technology, other people, languages ​​and resources available, or made available, for hope.

4 Hope

she undresses in the paradise

of her memory

she does not know the fierce fate

of her visions

she is afraid of not knowing how to name

what does not exist.

Alejandra Pizarnik, “6”, Diana’s Tree (1962)

ella se desnuda en el paraíso

de su memoria

ella desconoce el feroz destino

de sus visiones

ella tiene miedo de no saber nombrar

lo que no existe.

Alejandra Pizarnik, “6”, Árbol de Diana (1962)

The story of Ana and her colleagues at the beginning of this essay is exemplary: it shows transformative actions achieved in and through language, involving a world of people, objects and institutions in a process of change while maintaining both hope and a sense of humor. This horizon of hope, however, is frequently overlooked by discourse studies.

Indeed, the critical orientation in the studies of language in society is deep and powerful: the denunciation of linguistic inequality, of dominant ideologies, of racism, of sexist discourses, etc. is currently a part of its epistemological common sense.

But it has at least two limits. The first is in its own conditions of production: if we denounce from specialized journals, addressed to equally specialized colleagues, with the vague objective of “becoming conscious” or “raising awareness”, it is most likely that our complaint will have no practical effect on that reality that we describe critically. The second is in its object: if we only privilege the analysis of practices of social control, ideological domination, discursive hegemony, sociolinguistic orders or dominant ideologies, we block our perception, and even our own imagination, to those voices that act for change. In other words, we know, theoretically, that hegemony is not seamless, but in our analyses we cover up those seams, reproducing the same hegemonic effect that we are meant to question.

In this essay I have proposed two ways to show these seams. In the first place, by enriching our investigations with the voices of the people with whom we investigate, not as a return to romantic individualism, nor as uncritical empiricism, but as an entry point to language in society from the perspective of those who use it, who embody and transform it. Second, we need to show that these people act on language in an imperfect, contingent, shared and distributed manner; and that by their doing so, the world changes.

Thus, I propose to investigate with a “bias for hope” (Sikkink 2017: 14) based on evidence – not simply on optimism – in order to describe and understand this change, which is always underway.

In the short poem that heads this last section, Alejandra Pizarnik speaks of someone who fears not what exists, but not being able to name what does not exist. Today, studies of language in society are responsible for helping us to think about what does not exist, for discovering those who are already thinking about it and, in our own way, for helping to find a name for it.