Jan Blommaert 

What follows is part of the “Durkheim and the Internet” project – a self-conscious attempt at drawing theories of wider relevance from recent sociolinguistic studies and evidence. The theories are designed to provide a generalizable heuristic for new research, a set of potentially productive “grounded” hypotheses to be deployed in a wide variety of domains of investigation.

One of the theories emerging from this project is a theory of power – not a general one (power per se) but a specific one, about one kind of institutional power. Two points of departure underlie the effort here.

1. In The Utopia of Rules, David Graeber describes the fundamental stupidity of contemporary bureaucratization, observing the spread of what he calls “power without knowledge”: “where coercion and paperwork largely substituted for the need for understanding (…) subjects” (2015: 65). The contemporary power of bureaucrats often involves an assumption of total knowledge (articulated, e.g. in Foucault’s work). Graeber, however, disagrees: “situations of structural violence invariably produce extremely lopsided structures of imaginative identitification” (69): rulers have no clue about who and what their subjects are, what it is they do, what they attach importance to, how they live. The schematization and simplification of bureaucracy serve as a substitute for intimate and experience-based knowledge, but evidently fail to match up to that.
2. A decent amount of applied-linguistic work, notably on bureaucratic procedures such as asylum applications, shows how transnational subjects, often carrying the traces of a checkered diasporic biography, are nonetheless caught in administrative templates in which their “origins” are determined on the basis of imaginations of nation-state regimes of bureaucratic identity and on “modernist” theories of language (cf. Maryns 2006; Blommaert 2001, 2009; Jacquemet 2015). Concretely: if applicants’ claims as to origin (being from country X) are being disputed, knowledge of the official, national languages of Country X is used as a definitive test. If one fails this criterium, asylum is being denied. The same happens whenever an applicant provides discourse which is sensed to violate the rules of denotational purity: whenever s/he produces contradictions, silences, a muddled chronology or a lack (or overload) of detail, the applicant is judged to be untrustworthy and the success of his/her application is jeopardized.

The “lopsided structures of imaginative identitification” described by Graeber, we can see, in actual fact assume the shape of anachronisms: schemes of social imagination, and thus of patterns of meaning-making,  perhaps valid in an earlier stage of development, but not adjusted to recent changes and thus inadequate to do justice to the phenomenology of present cases. At the same time, these obsolete schemata are strongly believed to have an unshakeable, persistent relevance as a rationality of administrative information-organization, and are enforced from within that rationality. Thus, an important part of contemporary institutional power is based on anachronisms.

Anachronisms are, of course, an inevitable feature of social change, and we know that governmentality – the logic of institutional bureaucracy and governance – is widely characterized by inertia. It represents a segment of society which develops more slowly than the segments it is supposed to deal with. The gap between the phenomena to be addressed, and the schemata by means they are addressed, is a grey zone of uncertain understanding and often arbitrary judgment – and thus, increasingly, of miscarriage of justice and of litigation.

In terms of research, such anachronistic gaps offer a very rich site for investigating social change itself. It is based on the general image of social change described elsewhere: an image of different layers developing at different speeds. The different speeds manifest themselves in actual, situated cases of misunderstanding (or rather: the incapacity for understanding) and/or of experienced injustice.

The awareness of anachronisms is nothing new, needless to say. Durkheim’s own efforts, we have seen, were grounded in his convicition that “society” had not been adjusted to an important range of innovations caused by the industrialization and urbanization of France. Similar views, of an old social order being crushed under the weight of a new one, are widespread in the sociological literature. What this theory of anachronisms as power now offers, is accuracy. When earlier generations saw “society” being ill addapted to innovation, they couldn’t possibly mean all of society, for the parts that had been innovated were also very much part of that society. What we can contribute, therefore, is a highly precise focus when we look at such phenomena. The anachronisms are particular modes of organizing social interaction through specific patterns of meaning-making: categorization, the connection of different phenomena, objects or persons in specific sets of relationships to each others (as when an asylum seeker is brought in a certain relationship with national languages in determining his/her origins), patterns of argumentation and the ways in which we attribute judgments of persuasiveness to certain such patterns. Our theory enables us to look for very precise objects of analysis that can document change and the anchronistic effects that accompany it.

Evidently, the internet as a technology that has brought substantial innovation to the modes of social interaction now common around the world, is prone to such anachronisms. It is a segment of contemporary social life that develops at very high speed, while our modes of meaning-making are slow to be synchronized. Thus, we talk about, and in, new modes of internet communication very much in ways reflecting an pre-internet complex of social relationships.

A very clear example of this is the fact that Facebook, the largest social media platform in the world founded in 2004, uses one of the oldest and most primitive terms in the vocabulary for human relationships as its core tool: “friends”. Evidently, Facebook “friends” are not necessarily coterminous with offline friends. Facebook also uses a similarly ancient and primitive term to describe the most common interaction function on its platform: “like”. And evidently, this “like” function covers a very broad and extraordinarily heterogeneous range of actual meanings. No-one needs to actually like an update in order to “like” it, and no-one needs to be an actual friend in order to become a Facebook “friend” (which is why s/he can be easily and swiftly “defriended” whenever differences of opinion arise).

Those are of course innocent phenomena, merely indexing the anachronistic gaps caused by developments in social media. Less innocent, but very difficult to pinpoint, are the effects of some of the organizing principles behind social media: the algorithmic engines used by e.g. Google and Facebook to bring people, messages and zones of social activity together on the basis of aggregations of huge amounts of data and metadata generated by users. These algorithms belong to the world’s best-kept industrial secrets, and the assumptions on which they are based can, consequently, not be directly researched. But some of their effects are known.

All of us, I am sure, have at times error-clicked some advertisement on a social media page – let’s say, an advertisement for the newest model of urban SUV by Peugeot. All of us must have noticed how in the days following that erroneous click, multiple advertisements for cars appear on almost any page we open, usually cars in the same price range as the Peugeot we error-clicked. Less visible, perhaps, is the fact that in our social media newsfeeds, we are likely to encounter more people who recently clicked such advertisements in the days following our error-click, most likely people from our contacts network and people in the same geographical area as us. And also less visible, perhaps, is the fact that our perceived interest in cars of a certain brand and price range will be correlated with other data we produce through our social media usage – other products we express an interest in, other aspects of lifestyle, other persons, perhaps political views or preferences for certain sports or sports teams – all of this resulting in a permanently updated “algorithmic identity”, of certain interest for marketing and security professionals, over which we ourselves do not have any control, let alone agency.

Altough we can, as I said, gauge these procedures from a distance only, we can infer from what we know that these algorithms are anachronisms too. They are overwhelmingly linear and reductionist: linear, for clicking an item is interpreted as necessarily rational and deliberate – the mind-reading procedures of the algorithm exclude the possibility that we clicked the button by acccident. And reductionist in the sense that clicks are seen as inspired by very specific forms of interest in the thing we clicked – an interest, for instance in buying that object rather than to just admire it or confirm our opinion that such things are absurdly expensive.

The algorithmic identities thus ascribed to us may be light years removed from the actual motives driving our social conduct and from the ways in which we see ourselves. Well known, for instance, is that at a certain time when terrorism alert worldwide was red-hot, googling for information on pressure cookers was algorithmically flagged as suspicious because these mundane receptacles happened to be widely used in manufacturing home-made explosive devices. Which is an activity performed, fortunately, by very few individuals; but in order to locate these individuals, a great many more must have come under close scrutiny by security and intelligence officials – for no reason other than, perhaps, they contemplated buying a very nice pressure cooker so as to boost the quality of their bowl of evening soup.

Patterns of human interaction and meaning-making are the most sensitive indicators of social change – every neologism in our everyday language usage demonstrates this. If we wish to understand the fine grain of social change, close attention to these patterns is therefore sure to offer far more analytical purchase than almost any other aspect of social life. Power, too, can be investigated by looking at the anachronisms characterizing patterns of interaction and meaning-making deployed in governance; it can be looked at in very great detail.

References

Blommaert, Jan (2001) Investigating narrative inequality: African asylum seekers’ stories in Belgium. Discourse & Society 12/4: 413-449

Blommaert, Jan (2009) Language, Asylum and the National Order. Current Anthropology 50/4: 415-445.

Graeber, David (2015) The Utopia of Rules: On technology, Stupidity, and the Secret Joys of Bureaucracy. Brooklyn: Melville House.

Jacquemet, Marco (2015) Asylum and superdiversity: The search for denotational accuracy during asylum hearings. Language & Communication 44: 72-81.

Maryns, Katrijn (2006) The Asylum Speaker: Language in the Belgian Asylum Procedure. London: Routledge.

Jan Blommaert 

Sociolinguistics has over the past six decades provided a wealth of empirical evidence of great theoretical value, while it has quite consistently failed to formulate crucial insights as more generally valid social and cultural theories (cf. Williams 1992; Coupland 2016). In the “Durkheim and the internet” project, I extract and formulate theories from contemporary sociolinguistic research inflected, notably, by research on the connection between online and offline phenomena. The goal is to formulate social and cultural theories often fundamentally revising dominant ones, or, alternatively, lending support to views articulated earlier but left largely underexplored.

Several building blocks for the “Durkheim and the Internet” project are available in embryonic shape here. See, for instance, the essays on mobility, context and the chronotope, chronotopic identities, on conviviality and new social formations on social media, on the notion of culture, and several others. In what follows, I will offer some brief reflections on “structure” – as in “social structure” or “structuration theory”.

What I wish to avoid

Let me first sketch the field of arguments in which I shall situate my reflections. I wish to steer clear from two quite widespread frames of reference for discussing structure.

• First, “structure”, certainly in a Lévi-Straussian variety of structuralism, has acquired strong suggestions of absoluteness, abstractness, predictability, anonymity, a-temporality and staticity. Structure, as the guiding value system of a society, is that which provides enduring stability to a social system and makes it resilient – as Parsons suggested – to the onslaught of cultural revolutions from within youth culture (Parsons 1964). And even if structure is the outcome of active structuration at a variety of scale levels in social life (Giddens 1984; Thompson 1984), most scholars would still use the term to describe dominant (if not determining) rules, values or principles driving the development of societies across spacetime. It is also quite often presented as a social force operating below the level of consciousness and agency of people, a set of tacit and not always “emicly” well-understood aspects of social life.
• Two, “structure” is often seen as something antagonistic to “postmodernist” and “mobility/complexity” approaches to social life. While traditional (“modernist”) social science would be on the side of anonymous structure, “postmodernist” science would favor individual agency and thus become at once “poststructuralist” – in an unrealistic either/or frame in which methodological preferences appear to lead directly to ontological strictures. (Nik Coupland walks into the trap of such false antagonism: 2016: 440-442). It is rarely observed that scholars such as Bourdieu and Foucault do not just reject any concept of structure but reject a specific one: the Lévi-Straussian one referred to above. They reject a certain kind of structuralism (“poststructuralism” would be more accurately defined as “post-Lévi-Straussism”) but not “the structural” as a dimension of social systems. In general, this false antagonism often renders more nuanced understandings of structure impossible.

Since I am very much on the side of mobility/complexity in this debate, I am likely to be misread, given the lines of debate just sketched. Many fail to recognize that complexity is not the absence of order, but a different kind of order. I shall therefore use another term to make my point. Rather than using “structure”, I shall use “constructure” in what follows. New terms enable us to examine the validity of the older ones, and they also afford some measure of detachment from unwarranted intertextual readings. “Constructure” is not technically speaking a neologism – it is an archaic term that offers a nice collocation of “structure” and “construction” (and, if you wish, “conjuncture”). The term “construction”, as can be seen, can easily be changed into “agency”, and so we have a concept in which both dimensions, often seen as antagonistic, are heuristically and analytically joined.

Constructures

The baseline assumption – one that, I hope, is unproblematic – is that any social event is structured: there is always “order” in any observed social event. But from a complexity perspective on sociolinguistic phenomena and processes, this order is always:

• dynamic and unstable: order is always a temporally contingent quality because systems are perpetually unfolding and changing; (E.g. describing language at one point in time will necessarily result in a description which is different from what was current a generation ago, as well as from what will be current in the next generation).
• unfinished: given the perpetual change, any momentary observation of “order” will contain open-ended, quickly evolving features anticipating new forms of “order”; it will also contain features that are contested and conflictual, and features in the process of being eliminated or established; (E.g. archaisms and neologisms, short-lived as well as more lasting ones, are always part of any synchronic observation of language).
• non-unified: any “order” consists of a mixture of different forces, developing at different speeds and with different scope and range; (E.g. the different registers and genres in anyone’s repertoire have different speeds of development, with “standard” registers usually slower in development than e.g. youth registers – hence our sense of “trendiness”).

In these different forces, we are used to reserve the term “structure” for the slower, more persistent forces, the durée. I suggest we avoid this distinction and consider the entire mix when we use the term “constructure”, because given the complexity perspective, there is no telling a priori which of the features in the mix will determine future developments – change often happens in the margins and begins a statistical minority or exception, often negatively qualified. Think of the spectacular rise of emoticons as part of several mainstream genres of writing nowadays. Emoticons have not replaced the conventional forms of alphabetic writing – we still write from left to right, and we still use the conventional “orthographic” symbols we associate with the written form of the language we are using. Emoticons have been added to the mix of contemporary writing, so to speak, they represent what we could call a “light” feature, blended with the “big” features of conventional orthography.

Constructures are, thus, a permanently unfolding mix of various separate “structures”, the momentary deployment of which in social practice grants the latter a degree of orderliness, recognizable and ratifiable for others.

Constructures as change

In constructures, we can unify traditional notions of “structure” and “agency”; slightly rephrased, we have a tool for recognizing two essential characteristics of social life – iterativity and performativity. Most of the behavior we deploy socially is overwhelmingly iterative, but slightly inflected by unique, creative and situated performativity – something Piia Varis and I earlier called “culture as acccent” (Blommaert & Varis 2015).

Observe that I do not equate iterativity with “stability” and performativity with “change”. The entire mix is continuously changing, including the “iterative” aspects of it. Detaching the performative “accent” from the iterative “structure” obscures the fact that, for people in everyday practice, the “accent” is often the essence of what they perceive as meaningful in social action. And it is by means of the performative “accent” that the iterative features of behavior are also transformed into unique and creative characteristics of specific social actions performed by specific people. To make a superficial comparison: every novel is a token of a genre-type “novel”; but we have our favorite novels as well as novels we can’t possibly enjoy because of the actual unique “accent” brought to each novel by its author. And the entire genre changes due to such unique, individual and innovative acts of creation, all of which develop within the genre of the novel.

Rather than as a concept that points towards the stability of social systems – the simplistic interpretation of “structure”, noted above – constructure thus points to the permanently changing nature of social systems. When we read Erving Goffman’s observations on social life in the US of the 1950s and 1960s, we can still recognize a great deal of it today, even if much of our social life these days is performed in a social space that didn’t exist in Goffman’s world: the virtual space of social media. Interaction in this virtual world is organized along different sets of norms many of which differ strongly from the ones Goffman detected in face-to-face engagements. Online sociality, however, has not replaced the Goffmanian world of social interaction – the mix has changed. Which is why we can still recognize ourselves in Goffman’s work, even if we realize that large chunks of our lives are led in very different ways. The constructures have changed.

References

Blommaert, Jan & Piia Varis (2015) Enoughness, Accent, and Light Communities: Essays on Contemporary Identities. Tilburg Papers in Culture Studies, paper 139. https://www.tilburguniversity.edu/research/institutes-and-research-groups/babylon/tpcs/item-paper-139-tpcs.htm

Coupland, Nikolas (2016) Five Ms for sociolinguistic change. In Nikolas Coupland (ed.)Sociolinguistics: Theoretical Debates: 433-454. Cambridge: Cambridge University Press.

Giddens, Anthony (1984) The Constitution of Society: Outline of the Theory of Structuration. Cambridge: Polity Press.

Parsons, Talcott (1964) Social Structure and Personality. New York: Free Press.

Thompson, John B. (1984) Studies in the Theory of Ideology. Cambridge: Polity Press.

Williams, Glyn (1992) Sociolinguistics: A Sociological Critique. London: Longman.