Affective infrastructures and political organisation

Bosworth's (2023) ‘What is “Affective Infrastructure”?’ is doubly useful to those of us working at the intersection of affect and politics. First, because it offers a succinct critical overview of the growing literature in which the concepts of ‘affect’ (and neighbouring notions like ‘emotion’, ‘feeling’, and ‘mood’) and ‘infrastructure’ (‘materiality’, ‘economy’, and ‘determination’) combine. Second, because it challenges us to think more rigorously about the ways in which we connect these two semantic fields and how we apply them to phenomena.

If interest in affective infrastructures is partly due to ‘a concern with … political-affective organization’ (Bosworth, 2023), it is from this perspective that I would like to respond to that challenge. I will try to clarify what I have taken the concept to mean since first coming across it (in Dean, 2016), while fleshing it out in reference to a concrete case, that of far-right politics in contemporary Brazil. Through that, however, my goal will be to address some of the questions that Bosworth raises, shedding light on a few of the theoretical decisions we must make in order to establish what is unique to ‘affective infrastructure’ as a concept.

Pinpointing the notion's double function of denoting constraints and affordances, Bosworth suggests that we should view affective infrastructures as encompassing at once ‘a certain form of passive technical alienation … in which a collective's powers are separated from it by the social subjection of work and everyday life’ and efforts to channel ‘the indignation arising from such a situation back into an affirmative politics that would grow the capacity to organize the infrastructures of everyday life’. Yet this formulation immediately highlights the need for a choice regarding the level of generality at which the concept is expected to operate. For it would be possible to stretch it so far as to cover an entire social formation: we could then speak of a single affective infrastructure for a society or period as the ‘metastable solute out of which different sorts of affective politics are individualized or formed’ (Bosworth, 2023). The concept would thus be essentially coextensive with something like ‘infrastructure’ or ‘material basis’ in the Marxist lexicon, with the added idea that these necessarily involve the production and reproduction of certain affects that in turn contribute to their reproduction.

The problem of extending the scope so is that making ‘affective infrastructure’ refer to the most general affective conditions in a given formation would prevent us from using it where it can be most illuminating: in finer investigations of the specific ways in which social, technical and affective conditions interlock for different groups. Another option would be to have the concept vary in scale – a move that is theoretically justified insofar as each particular affective infrastructure can always be delimited as a part of the larger affective infrastructure of a society. (The idea of multiple interacting affective infrastructures is considered by Bosworth when the writes of a ‘field’ of such entities as ‘a skein, a cacophony, a battleground’.) But if we are talking about political organisation, the more restricted use is evidently the most useful. In order to show what it can do and what it implies, I turn to a concrete example: the supporters of far-right leader Jair Bolsonaro in Brazil – or, more precisely, those we can describe as ‘Bolsonaristas’.

As I have argued elsewhere (Nunes, 2021), that term is not necessarily synonymous with ‘Bolsonaro voter’. People voted for Bolsonaro in 2018 for myriad reasons, and even now there are those who may choose him despite himself. One of the ways in which we can distinguish this circumstantial voter from the hardcore Bolsonarista is precisely the latter's participation in a collective infrastructure of affect.

In what does this consist? First of all, a communicational infrastructure. This encompasses WhatsApp and Telegram groups, social media influencers, allied voices and outlets in the corporate media (radios, newspapers, and TV channels). A lot of this structure was built from the bottom up, but then incorporated and expanded into something more organised and pyramidal from the 2018 campaign onwards (Nemer, 2021). This resulted in a system of top-down diffusion and feedback in which the general lines of discourse to be adopted filter down from an extra-official group directly connected to the president (the so-called ‘Hate Cabinet’), while responses, themes, and material produced at the base filter up to the top –– where they might be amplified, recycled, or induce adaptations in the original messages. Yet the structure also includes non-digital spaces, in both the physical and the social sense, like churches and church groups, clubs, friends, and family networks that are connected in one way or another to the main stem of the communicational network system. Both digital and non-digital spaces enable regular practices and encounters that create and reproduce certain affective habits.

All of this is evidently built on material infrastructure, both public and private: mobile phones, computers, cables, servers, antennas, electricity, printers, buildings, and whatever facilitates and organises their use. This, I would suggest, provides a criterion through which to distinguish affective infrastructure from ‘a relational concept of affect’ or ‘the generic concept of an “affective condition”’ (Bosworth, 2023): its reference to a specific material support. Obviously, it is not the case that affective conditions and structures of feeling can exist without relying on a material support, not least bodies endowed with the capacity to affect and be affected. What makes ‘affective infrastructure’ unique, however, is the way in which the concept foregrounds that reliance and opens it up to closer examination, putting the spotlight on a concrete material infrastructure, with its own determinate architecture (and what that affords and constrains) as well as the specific practices that activate it. ¹

Now, what is it that enables us to characterise a certain material support as an affective infrastructure? It is, first of all, the fact that what circulates in it is not just information or goods but affects. By means of repetition (of frames, messages, practices, encounters) and recursion (adjusting said frames, messages, practices, and encounters to changes in the group or its environment), an affective infrastructure locks participants in certain affective patterns, sustains the intensity with which these are felt, and thus reproduces a certain relation of the group to itself and to its various out groups (which may be compressed into a single, compact figure of the ‘other’ or ‘enemy’). It is through that, rather than merely rhetorical means, that an affective infrastructure can produce the chains of equivalence (Laclau and Mouffe, 2001) that split the social field into two camps, ‘us’ and ‘them’: affective and discursive patterns interact so as to define objects of love and hate, admiration and repulsion, fear and hope, emulation and aversion, as well as the relations between them and certain dispositions and possible courses of action. Finally, affective infrastructures also function as media for positive feedback effects that ramp up the intensity with which certain affects are experienced, resulting in phenomena of affective synchronisation (Nunes, 2014: 22ff) (in which large numbers of people find themselves simultaneously experiencing the same strong feeling) that may or may not lead to individual and collective discharges (in the form of stochastic violence and protests, for instance).

This gives us a hint as to how affective infrastructures might be individuated –– that is, on what grounds we can fence off a certain socio-technical assemblage as constituting something that can be so described. (This returns us to the issue of generality raised above.) Obviously, both the material support and the affects that an affective infrastructure latches on to, feeds off, and sustains belong to society at large; it is because they already exist and are structurally or conjuncturally reproduced that they can be seized upon. What then makes them predicable of that particular infrastructure and not just of a general ‘mood’ or ‘affective condition’? The answer lies in the fact that affective infrastructures possess a certain degree of closure and self-reference. That is, even if what they do is essentially fold affects and materials that are already there, they involve the effort to establish and maintain an inside through iteration (of patterns and investments), identification (with those patterns, with other participants, with leaders), and orientation (regarding themselves, their others, and the courses of action that this polarised relation suggests). This is also what allows us to distinguish affective infrastructures from infrastructures designed for the circulation of information and affects, such as digital platforms: while the latter can function as media for the formation of affective infrastructures, they do not possess the requisite level of consistency that would make them thus describable.

Something that follows from conceiving affective infrastructures in this fashion, on the other hand, is that they should be understood as media in their own right: not only in the sense that they are at once the means for the propagation of information and affects and the milieu through which these propagate, but also in the sense that one can speak in optics of a translucent medium, that is, one which transmits the waves that go into it, but not without some loss or distortion. This is precisely one of the things that characterises them as an ‘inside’: they mediate between those who participate in them and information and affects coming from the outside by means of affective biases that select and/or modify what passes within. This is to say that, in Spinozist and Simondondian terms, they amount to a collective ingenium or a transindividual zone of individuation for the singular ingenia of participants. ² This entails the need for a certain perspectivism when dealing with affective infrastructures: messages and actions can signal and trigger different things depending on whether one is inside or outside.

One final issue to be addressed concerns the relation between potentia and potestas, or of the collective capacity to act and its alienation. Choosing a far-right affective infrastructure as my example stresses the need to see that relation as taking place on more levels than Bosworth appears to envisage. If political organisation can always be seen as ‘a practice of collectively apprehending and augmenting … mediating settings to produce social analysis and grow social power’ (Bosworth, 2023) among its participants, this does not automatically tell us anything about its broader societal goals.

The first level of analysis concerns the passivity and alienation produced by the dominant socio-technical arrangements in the society in which an affective infrastructure forms, and which it seeks somehow to address; political organisation on the far right as much as on the left responds to broadly shared feelings of subjection and powerlessness. The second level concerns the ratio of potentia and potestas within the infrastructure itself. Regardless of its political orientation, an affective infrastructure always provides its participants with the means to experience their own collective capacity to act, to feel it wax and wane as it affects and is affected by the world. At the same time, this inevitably depends on some degree of passivity, from submission to leaders or written and unwritten rules to the technical alienation of dealing, for example, with digital platforms one neither fully understands nor controls. Finally, there is the level at which an outside observer can judge a given affective infrastructure according to whether its political orientation points in the direction of greater collective activity and emancipation or, on the contrary, passivity, alienation, and subjection. It is at this level, then, that one can pose ‘the fundamental problem of political philosophy’, ‘the one that Spinoza saw so clearly, and that Wilhelm Reich rediscovered: “Why do men fight for their servitude as stubbornly as though it were their salvation?”’ (Deleuze and Guattari, 1983: 29).