Four Data Discourses and Assemblage Forms: A Methodological Framework

“four discourses have emerged: data as an asset, data as latent knowledge, data as a technology of governing, and data as resistance. These four discourses align with Deleuze and Guattari’s capitalist, territorial, statist, and nomadic social formations, respectively.” ⤴️
“that struggle for discursive dominance. This paper argues that four discourses have emerged: data as an asset, data as latent knowledge, data as a technology of governing, and data as resistance. These four discourses align with Deleuze and Guattari’s capitalist, territorial, statist, and nomadic social formations, respectively. T” ⤴️
“The social forms exist in reciprocal relationship to and are the aggregates of concrete assemblages” ⤴️
“opportunity to interrogate dominant conceptualisations of Big Data before they become embedded in our everyday lives ‘foreclosing alternative possibilities’. Dalton and Thatcher (2014) exhorted that it is a characteristic of all successful technologies that our attention on them dissipates as they become normalised.” ⤴️
“data as an economic asset, data as means to create knowledge, data as a technology of governing, and data governance subjects resisting subjugation and claiming the value and power of their data for themselves. It will demonstrate how categorising data discourses in this four-part framework can be used to link up the CDS corpus both conceptually and methodologically via Deleuze and Guattari’s assemblage theory.” ⤴️
“a procedural approach that follows how the discourses manifest through the creation of data, insights, and their use within concrete assemblages. Finally, the paper outlines how combining these two analytic approaches could be employed as the first two steps of Deleuze and Guattari’s assemblage analysis.” ⤴️
“CDS ‘applies critical social theory to data’ (Kitchin and Lauriault 2014). It challenges the naive assumption that data are ‘neutral, objective, independent, raw representations of the world’ (Kitchin and Lauriault, 2014; also see Iliadis and Russo, 2016: 1; boyd and Crawford, 2012: 667; Dalton and Thatcher, 2014). Rather data are regarded as subjective, ‘situational, contingent, relational, contextual’ and doing ‘active work in the world’ (Kitchin and Lauriault, 2014). Data is an expression of power (D’lgnazio and Klein, 2020; Iliadis and Russo, 2016; Kitchin and Lauriault, 2014).” ⤴️
“Big Data does not drive social change. Instead, it supports already present forces (Dalton and Thatcher, 2014) such as rising authoritarianism, political polarisation, and capital accumulation.” ⤴️
“discourse is a ‘historically contingent social system that produces knowledge and meaning’ (Adams, 2017)” ⤴️
“Discourses structure social relations by fixing—or attempting to fix—meaning and logic (Adams, 2017).” ⤴️
“ey are products of social power, and ‘this power prescribes particular rules and categories which define the criteria for legitimating knowledge and truth within the discursive order’ (Adams, 2017; also 2” ⤴️
“e Jørgensen and Phillips, 2002). However, discourses generally hide their political intentions, masquerading as a-historical, universal, objective, and stable (Adams, 2017; Hook, 2001: 524).” ⤴️
“Crawford (2012) argue that the ‘the rules about how Big Data will be used, and who gets to participate’ are set by a privileged few with the means to collect and analyse the data (675), potentially reinforcing existing power asymmetries.” ⤴️
“t materialises within the microcosm of discrete data assemblages with discourses becoming fixed within their structures, such as in roles, data architecture and governance arrangements (Beer, 2016: 5; also see Hook, 2001: 529-30).” ⤴️
“grey literature (e.g. media, industry, and government” ⤴️
“n some instances, data is traded as a commodity in generally unregulated markets formed around the buying and selling by data brokers of ‘raw’ data or enhanced data products (Fourcade and Healy, 2017: 11, 16; Kitchin, 2014b: 42-5).” ⤴️
“who has a right to the data. Data is often seen as property—something to be owned—with disputes, and increasingly, regulation to designate who has ownership rights (Birch, Cochrane and Ward, 2021: 3-4; Kitchin and Lauriault, 2014). However, there are also disputes about whether ownership is actually possible or the appropriate legal mechanism (Birch, Cochrane and Ward, 2021: 3-4).” ⤴️
“quantification of the material world of, for example, people, spaces, events, materials; labelled ‘datafication’ (Van Dijck, 2014: 198 referring to Mayer-Schönberger and Cukier, 2013). The second pass is the monetisation of the data (if sold or licensed); the data itself is in some instances quantified (Van Dijck, 2014: 200-1). However, Birch, Cochrane and Ward (2021) argue that, regarding personal data captured by Big Tech firms, users are monetised rather than the data per se (4).” ⤴️
“quantification of the data and its outcomes is used to justify both public and private investment in data infrastructure.” ⤴️
“overnments and businesses are assured that increases in productivity (i.e. in the company or economy) can be achieved with investment in enhanced collection, sharing, or use of data, i.e. exploiting data is tied to economic growth (Kitchin, 2014b: 16; König, 2017: 5; McAfee and Brynjolfsson, 2012).” ⤴️
“data is not necessarily collected purposefully, that is, related to a clearly articulated analytic question and value proposition. Instead, it is increasingly captured speculatively either as incidental ‘exhaust’ from existing processes or proactively datafying all aspects of experience for both known and unknown future uses (Fourcade and Healy, 2017: 13; Kitchin, 2014b: 6-7; Sadowski, 2019: 4-” ⤴️
“Sadowski (2019) argues that a ‘rhetoric of universality’ that data is everywhere and free for the taking ‘reframes everything as within the domain of surveillance/platform/digital capitalism’ (2-3). H” ⤴️
“quating data with naturally found resources such as ‘data is the new oil’ (Sadowski, 2019: 4), a resource to be ‘mined’ (Sadowski, 2019: 2-3, 6), or value extracted through a ‘production line’ or ‘value chain’ of data collection, storage, analysis, and use (Chun, Adam and Novek, 2018; Löfgren and Webster, 2020). F” ⤴️
“entities that become data custodians on behalf of others, for example holding data in ‘trusts’, somewhat like financial trusts (Micheli et al., 2020: 8-9).” ⤴️
“Regulation and controls are resisted by industry using arguments regarding risks to the economy (Sadowski, 2019: 5, 9). This discourse ‘provides a convenient narrative’ for software and hardware vendors to justify the purchase of their products and services (Kitchin, 2014a: 5; also see Löfgren and Webster, 2020: 2).” ⤴️
”, data is perceived as latent knowledge, a pre-factual material to be transformed by data scientists into valued insights (Kitchin, 2014b: 3-4, 19)” ⤴️
“he outcome prized within this discourse are insights not possible within the limits of the computational powers of the human brain.” ⤴️
“ncreasing the speed and types of analysis provide opportunities such as real-time or fast feedback, profiling, and prediction.” ⤴️
“readily highlighted by CDS scholars and noted earlier, an epistemological tension within this discourse is between a vision projecting data as providing objective truths and acknowledging data’s biases, limitations, culturally and historically contingent subjectivity. It includes a debate about the scope of data analytics in helping us to understand the world; whether it can enable us to be all-knowing or is limited to a subset of questions ignoring those related to, for example, meaning, politics, and ethics (Kitchin, 2014a: 8). K” ⤴️
“itchin (2014a) highlights that within the empiricist epistemology, ‘meaning transcends context or domain-specific knowledge, thus can be interpreted by anyone who can decode a statistic or data visualisation’ (4).” ⤴️
“the promise of cleansing the process of interfering intermediaries which ‘corrupt data’ with their vested interests (Kitchin, 2014b: 19).” ⤴️
“data is often discussed in terms that suggest a living oracle-like being. It is common to hear phrasings such as ‘What does the data say?’, ‘Let’s consult the data’, and ‘The data does not lie’. T” ⤴️
“proponents of data-driven science argue that the size and relationality of the data enable insights to be born inductively from the data (6). Promoters of this way of thinking claim that analysts can find answers to questions they did not know they had (Kitchin, 2014a: 4).” ⤴️
“he idea that data science transcends human analytic capabilities extends to the notion that data holders can know you better than you know yourself (Zuboff, 2019: 163). In the commercial sphere, the ideal is to anticipate and seamlessly satisfy consumers needs and wants ‘before they are fully formed’ (Fourcade and Healy, 2017). To do this, ‘the market sees you from within, measuring your body and emotional states, and watching you as you move around your house, the office, or the mall’ (Fourcade and Healy, 2017: 23).” ⤴️
“Power and influence are concentrated in those with such expertise and data access (boyd and Crawford, 2012), while outsiders, e.g. academic sociologists and domain rather than data experts, are marginalised (Burrows and Savage, 2014; Kitchin, 2014a: 5; Savage and Burrows, 2007).” ⤴️
“o proclamations of data science as the ‘sexiest job of the twenty first century’ (Davenport and Patil, 2012). T” ⤴️
“use of data for shaping the behaviour of self, others, materials, and environments. Data and its analytic outputs provide the capability to act or intervene in a more informed manner. This framing aligns with the Foucaul” ⤴️
“dian-inspired concept of governmentality, or the ‘conduct of conduct’ (Dean, 2009: 17-8; Ruppert, Isin and Bigo, 2017: 3; Rose, O’Malley and Valverde, 2006).” ⤴️
“governing with data’ is distinct from ‘governing the data’ or data governance as it is commonly termed.” ⤴️
“ataveillance (Van Dijck, 2014), ‘If you’re not the customer, you’re the product’ (Sadowski, 2019: 4), and ‘You can’t manage what you don’t measure’ (McAfee and Brynjolfsson, 2012).” ⤴️
“the diabetic monitoring their insulin levels to self-manage their condition, customers receiving pre-emptive service provision, drivers avoiding traffic jams, infrastructure owners monitoring wear and tear on equipment, businesses obtaining improved revenues through more effective advertising and fraud detection, governments able to better target services to needy citizens or identify threats such as disease outbreaks or criminal activity. I” ⤴️
“ata as resistance There are two ways of framing this discourse. Either reactively as a discourse of resistance in opposition to the logic or singular pursuit of the other three discourses. Alternatively, it can be framed actively as ‘data as agency’ (see Kennedy, Poell and Van Dijck, 2015; Smith, 2016: 267, 280). That is, the governed claiming the enhanced agency and value provided by the data, including its potential monetary, knowledge producing or governance value (Kennedy, Poell and Van Dijck, 2015; Micheli et al., 2020; Pybus, Coté and Blanke, 2015).” ⤴️
“ctors attempt to shape the behaviour of a government agency or elected representative using data about citizens, objects, or environments within their jurisdiction, and they therefore may resist.” ⤴️
”. Given that knowing you are being governed is often a precursor to resistance, a logical first step is identifying the power, governance structures, and the flow of benefits underlying data arrangemen” ⤴️
“Corporate resistance in the interests of data subjects includes industry calls for greater regulation and technological giants (e.g. Amazon, IBM) resisting government access to data created on their platforms, for example, police access to their facial recognition technology (Magid, 2020; O’Loughlin and Bukowitz, 2021: 1). Another industry response supporting data subject resistance has been the emergence of competitor services that prioritise privacy (Hodge, 2021).” ⤴️
“Sadowski (2019) argues that data is more accurately described as capital rather than the freely available and abundant raw commodity” ⤴️
“The ‘data imperative’ is to create commodities that generate data in an unceasing flow (Sadowski, 2019: 4). To resist any restriction on this flow, corporate datafication has leveraged similar tactics developed for financial capital, such as avoiding oversight through opaque technologies and avoiding regulation through political influence (Sadowski, 2019: 5, 9).” ⤴️
“Kitchin (2014a) juxtaposes empiricist programs with data-driven science (including positivist and post-positivist forms), and knowledge-driven or purely deductive science (1-5). Rather than rejecting Big Data approaches, Kitchin (2014a) makes a bid for an epistemological framing that acknowledges research’s inherent politics and is situated, reflexive, and contextualised (9).” ⤴️
“Beer (2016) draws primarily on Foucault and Hacking to illustrate Big Data’s social and conceptual origins in statistics (also see Ruppert, Isin and Bigo, 2017: 3). Statistics emerged in the early nineteenth century as an important component of population governance by states through intense categorisation and the concepts of normalcy and deviancy (Beer, 2016: 3). ‘These emerging numbers quickly came to define how people saw themselves’ and ‘how limits and boundaries were placed around actions and opportunity’ (Beer, 2016: 3). Beer (2016) emphasises how Big Data’s impact extends beyond direct use of the data and analysis to govern.” ⤴️
“collecting data as evidence for injustice (Van der Velden, 2015), b” ⤴️
”, becoming active and conscious” ⤴️
“rather than passive generators of data (Pybus et al., 2015), facilitating participatory democracy through open d” ⤴️
“ata (Baack, 2015), and democratising data governance (Kennedy and Moss, 2015; Micheli et al., 2020).” ⤴️
“Assemblages for Deleuze and Guattari answer the question ’what is a thing?’ such as an object, subject, idea, mood, or institution, within a metaphysics of the continuous rather than the discrete (Adkins, 2015: 10-11).” ⤴️
“he sensible and intelligible coexist on one continuous ontological plane rather than two discrete planes (Adkins, 2015: 1-2, 11-3).” ⤴️
“ollowing the logic of a metaphysics of continuity, things must be animated internally or immanently (in Deleuze and Guattari’s terms by ‘desire’) rather than externally by a transcendent universal law, truth, or direction (Adkins, 2015: 2-3)” ⤴️
“flux, change, and becoming are as pertinent to understanding ‘things’ as stasis and being (Adkins, 2015: 10-” ⤴️
“From their ethical perspective, universals are illegitimate illusions of transcendence and capture that separate us from our power to act (Smith, 2011: 125-6; also see Buchanan, 2008: 73); points of fixation that masquerade as immutable to channel power in particular directions.” ⤴️
“t is common to describe assemblages along the lines of a temporary collection of heterogeneous connections with tendencies towards both stability and change (Adkins, 2015: 14)” ⤴️
“assemblages are desire invested, socialised, or captured (Buchanan, 2008: 54-5, 92-4)” ⤴️
“hey are the means through which desire—the immanent conflicting drives within all objects and subjects (Smith, 2011)—invests or fixes itself in connections in the hope of increasing its potential to act, its ‘power to’ (Adkins, 2015: 96-7, 138, 153).” ⤴️
“he two are independent but exist in reciprocal presupposition; a dialectical, oppositional relationship (see Buchanan, 2015: 390-1 for an example). The form of content and form of expression do not map perfectly to each other, ’something always escapes’ (Buchanan, 2021: 33).” ⤴️
“Social formations—the focus of this section—are no ‘different in kind’ than concrete assemblages, except that they are ‘determined by historical conditions’ and are ‘desire on a social scale’ (Buchanan, 2008: 95-6” ⤴️
”). In Anti- Oedipus, Deleuze and Guattari refer to the capitalist, territorial, and statist forms as being ’molar’; however, not concerning scale, but use the term molar to refer to a tendency towards stasis and the discrete (Adkins, 2015: 48; also see 40-1 and Buchanan, 2008: 24-5). ‘Molecular’, in contrast, is a tendency towards fluidity and the continuous (Adkins, 2015: 48). Deleuze and Guattari (1983) describe capitalist, statist, and territorial assemblages or social forms as the statistical molar aggregates of concrete assemblages (183; Buchanan, 2008: 90). Concrete assemblages (or desiring-machines) both constitute these social forms and are shaped and permeated by them (Buchanan, 2008: 90-1; Deleuze and Guattari, 1983: 183). Social formations and concrete assemblages never exist independent of each other but in a reciprocal relationship (Buchanan, 2008: 90-1; Deleuze and Guattari, 1983: 183).” ⤴️
“e in a constant state of flux and mutation, sometimes in support and sometimes at odds with each other (Adkins, 2015: 219, 222; Nail, 2017:” ⤴️
”, social forms also do not exist in ’degrees of perfection measured against a pre-existing template or model, but states of equilibrium in a more or less constant movement of variation whose end result cannot be known’ (Buchanan, 2008: 91; also see Adkins, 2015: 219). The task is not to classify and judge, for example dividing things into ‘bad’ statist things and ‘good’ nomadic things (Adkins, 2015: 191). Rather, the ‘project of becoming, of creating the new, begins with seeing the nomadic in everything’ (Adkins, 2015: 13).” ⤴️
“her individualist, enclave, hierarchical, and isolate institutional forms (6 and Richards, 2017: 11) resemble Deleuze and Guattari’s assemblage types” ⤴️
“ue to the greater clarity I find in Deleuze and Guattari’s metaphysical and ethical foundations and the utility of their methodological approach.” ⤴️
“ouglas sought to identify universal, essential forms to explain culture (6 and Richards, 2017: 48-9, 78). Deleuze and Guattari, in contrast, vigorously worked to critique concepts of a universal and ordered underlying structure or ontological realm distinct from the sensible realm. T” ⤴️
”. In A Thousand Plateaus, Deleuze and Guattari add a fifth social form, the networked and polarising urban centre (Adkins, 2015: 220). Deleuze and Guattari would not advocate that you accept their analysis of the four or however many social formations as the last word, quite the contrary. What is most pertinent is the methodological approach (Buchanan, 2008: 65). They encourage us to consider a nomadic science alongside a statist science; the former in which ‘axioms are not settled once and for all’ but ‘can also be the starting points for experimentation’ (Adkins, 2015: 228).” ⤴️
“Data as latent knowledge aligns to a territorial social form that qualitatively defines and maintains tribes, territories, and truths.” ⤴️
“data as resistance aligns to a non-conformist form focused on revolution, creative transformation, and resisting repression (” ⤴️
“he discourse of ‘data as an asset’ maps to Deleuze and Guattari’s capitalist assemblage or social form (Buchanan, 2008: 107-16; Nail, 2017: 31-2), which works towards stripping all things of their qualitative relations (Nail, 2017: 31). It replaces them with homologous quantitative relations a” ⤴️
”). The territorial assemblage ‘attempts to ward off the chaos of a meaningless world’ by classifying and qualitatively ‘coding’ or giving meaning to elements (e.g. persons, objects, words) and defining their ‘nature’, or proper purpose (Buchanan, 2008: 94; Nail, 2017: 29). The remaining uncoded elements—the residual code that falls outside the classification system—are blocked from attaching to the assemblage through, for example, prohibition and taboo, and boundaries are formed, e.g. borders of towns, boundaries to gender identities (Nail, 2017: 29; also see Buchanan, 2008: 96-7). These ‘abstract relations change, but only one concrete element at a time…like a game of leapfrog. They set up some limits and by doing so create a new limit to cross…in a perpetual disequilibrium making its very dysfunction an essential element of its ability to function…Change happens progressively, one concrete point at a time’ (Nail: 2017: 30).” ⤴️
“Coding ’is not enough by itself to produce an enduring social machine’; relationships including alliances and affiliations are also required (Buchanan, 2008: 99). Alliances attempt to prevent the incorporation or ‘fusion’ of the uncoded elements by redistributing this residual or ‘surplus code’ (Nail, 2017: 29-30). These processes prevent stockpiling of wealth and create debt relation” ⤴️
“s” ⤴️
“This surplus redistribution has the effect of warding off capitalist (Buchanan, 2008: 93) or statist forms” ⤴️
“Deleuze and Guattari do not specifically relate these territorial processes to knowledge creation, but for Deleuze and Guattari, territorial assemblages are about meaning making. As noted in the earlier discussion, discourse is a process and struggle to fix meaning and define truth in the interests of certain power blocs or alliances. Knowledge creation is an ongoing process of classification, boundary pushing, and maintenance.” ⤴️
“his specialised body brings the elements ’into resonance around a [vertical and hierarchical] centralized point of transcendence’ (Nail, 2017: 30; additions not in original). It exerts controls or an apparatus of capture, also known as ‘overcoding’ (Adkins, 2015: 221-2). For instance, territory is overcoded as land or property such that ‘different properties are now comparable to one another’, that is ‘in terms of the rent they’ll support’ (Adkins, 2015: 222). This overcoding has the effect of cutting elements off from their relations to other elements (Nail, 2017: 30)—i.e. people’s historical connection to place, community and land based subsistence, and controlling their new connections, e.g. to taxed work—‘and in this way extracts its required surplus value’ (Buchanan, 2008: 105)” ⤴️
“Adkins (2015) draws on the imagery of clay squeezing from a hand; the clay conforms to the hand, but the more one squeezes, the more that escapes (223). The ‘imperial state remains only to the degree that it is able to recapture these flows’ (Adkins, 2015: 224).” ⤴️
“a site of transformative and creative revolution devoid of hierarchical and—in its pure form—social limits (Nail, 2017: 32-4). Nomadic assemblages ‘are able to change and enter into new combinations without arbitrary limit or so-called “natural” or “hierarchical” uses and meanings’ (Nail, 2017: 32). Where the capitalist assemblage provides limitless quantification, the nomadic assemblage provides unlimited qualitative transformation (Nail, 2017: 33).” ⤴️
“a nomadic assemblage: a direct participation without representation or mediation [Deleuze and Guattari, 1987: 471” ⤴️
“This kind of participation and self-management thus offers a political alternative absolutely incompatible with territorial hierarchies based on essentialist meanings, state hierarchies based on centralized command, and capitalist hierarchies based on globally exchanged generic quantitie” ⤴️
“(Nail, 2017: 33; reformatted reference in italics)” ⤴️
“This form aligns with the data discourse of resistance and agency.” ⤴️
”, in Deleuze and Guattari’s assemblage theory, nomadic ‘lines of flight’ are not a response to power per se; they ’are not phenomena of resistance of counterattack in an assemblage’ (Deleuze and Guattari, 1987: 531, footnote 39). Rather, they are primary; ‘the first given of a society is that everything takes flight’ (Deleuze, 1997).” ⤴️
“desire rather than power ’assembles the social field’ (Deleuze, 1997; also see Smith (2016) for a comparison of Foucault’s concept of resistance with Deleuze and Guattari’s). Power is secondary; it is the fixed, static, molar component of an assemblage (Deleuze and Guattari, 1987: 531, footnote 39).” ⤴️
“My aim here is to show how to realise assemblage theory’s potential more fully through the work of Deleuze and Guattari and methodologically bring together the analysis of discourse, social forms, and concrete assemblages” ⤴️
“In aggregate, concrete assemblages maintain and mutate these relatively stable, historically derived social formations.” ⤴️ ^c4560a
“influence the creation and transformation of the smaller, concrete assemblages in a reciprocal relationship.” ⤴️
“leverage points of potential change, ‘lines of flight’ or escape.” ⤴️
”. a discourse analysis. Buchanan (2008) describes this as a destructive task of isolating and unravelling transcendent illusions (Buchanan, 2008: 117). Destruction must be approached carefully to avoid re-establishing illusions of transcendent capture in the same or a different guise (Buchanan, 2008: 120).” ⤴️
“o unearth how and why those meanings ‘were constructed in the first place’ (Buchanan, 2008: 121). For Deleuze and Guattari, this step requires going beyond interpretation or questions of ‘what does it mean?’ or ‘where does it come from?’ ‘into what Deleuze and Guattari specify as the functional or machinic realm’ (Buchanan, 2008: 121; 2017: 461)” ⤴️
“elucidate the ratio and points of stasis and change within an assemblage, look for lines of flight, new and unique connections, or points at which the assemblage might become something else, something less repressive (Adkins, 2015: 92)” ⤴️
“How does the assemblage work? What are its flows and blockages? What logics are the form of content and expression indicative of, how do the two forms hold together, where do they conflict, or what points of escape do they revea” ⤴️
“What is the mix of the assemblage types? What would it mean to break free from capitalist, territorial, and statist assemblage forms and enhance the nomadic (Adkins, 2015; Buchanan, 2008, 2015, 2017)?” ⤴️
“s liberating desire from its capture or repression, reaching ’the investments of unconscious desire in the social field’ and enabling them to coexist ’in opposite modes’ (Buchanan, 2008: 123 drawing from Deleuze and Guattari, 1983: 350).” ⤴️
“seeing fixation points as a collective choice that are open to change; they are not immutable, transcendent law. L” ⤴️
“iberating desire involves dissolving these fixation points as required if it increases our collective ’power to’.” ⤴️
“Too much stasis or too much change is hazardous; ‘both are a kind of death. Life occurs in the middle’ (Buchanan, 2021: 52).” ⤴️
“s studying how different actors compete to entrench the discursive logics within practice and structure, embedding the discourses in data infrastructures and processes. A” ⤴️
“how these fixations create flows and blockages that move data, knowledge, and power in particular directions and away from others.” ⤴️
“Deleuze and Guattari’s methodology indicates that opportunities for new disruptive connections should also get special attention as potential sites of emancipatory change.” ⤴️
“Relationships between the stages and elements within them—for instance, actors, objects, places, rituals, discourses, institutions—may be convoluted with feedback loops between the four domains in terms of economies, truth- claims, governing, and resistance.” ⤴️
“This analytic approach is not to be confused with the metaphor of the production line or the concept of the value chain in which value is seen to be progressively added at each stage of the proce” ⤴️
“value might not increase for all parties.” ⤴️
“As the data becomes increasingly distal to the context in which it was collected the risk of spurious interpretation increases, or as it progresses to the stage of governing its use may threaten the welfare of governance subjects.” ⤴️
”. Löfgren and Webster (2020) illustrate how a procedural approach of breaking the data process into stages can be used to analyse blockages created when the logics of the four discourses and assemblage types collide in practice.” ⤴️
“Kennedy, Poell, and Van Dijck (2015) highlight case studies of ‘practices of data work’ that resonate with the nomadic tendencies of direct participation in transformation, and self-governance.” ⤴️
“y supporting and amplifying emancipatory findings from these sites of action, CDS scholars can move from critique to tangible and positive change, which in aggregate can be powerful forces.” ⤴️
“lassification as an imperfect means to mediate a chaotic and disordered world (6 and Richards 2017, p. 45) and the importance of anomalies or the unclassifiable (6 and Richards 2017, p. 25).” ⤴️
“Data as resistance  A means to threaten or enhance one’s agency  Provides agency (to self or others)  Data’s value is contingent on who controls and benefits from it, rather than being inherently valuable” ⤴️
“Data activism  Data sovereignty  Tech lash  This discourse can be framed reactively (as resistance) or proactively (as agency)” ⤴️

Four Data Discourses and Assemblage Forms: A Methodological Framework

“four discourses have emerged: data as an asset, data as latent knowledge, data as a technology of governing, and data as resistance. These four discourses align with Deleuze and Guattari’s capitalist, territorial, statist, and nomadic social formations, respectively.” ⤴️
“that struggle for discursive dominance. This paper argues that four discourses have emerged: data as an asset, data as latent knowledge, data as a technology of governing, and data as resistance. These four discourses align with Deleuze and Guattari’s capitalist, territorial, statist, and nomadic social formations, respectively. T” ⤴️
“The social forms exist in reciprocal relationship to and are the aggregates of concrete assemblages” ⤴️
“opportunity to interrogate dominant conceptualisations of Big Data before they become embedded in our everyday lives ‘foreclosing alternative possibilities’. Dalton and Thatcher (2014) exhorted that it is a characteristic of all successful technologies that our attention on them dissipates as they become normalised.” ⤴️
“data as an economic asset, data as means to create knowledge, data as a technology of governing, and data governance subjects resisting subjugation and claiming the value and power of their data for themselves. It will demonstrate how categorising data discourses in this four-part framework can be used to link up the CDS corpus both conceptually and methodologically via Deleuze and Guattari’s assemblage theory.” ⤴️
“a procedural approach that follows how the discourses manifest through the creation of data, insights, and their use within concrete assemblages. Finally, the paper outlines how combining these two analytic approaches could be employed as the first two steps of Deleuze and Guattari’s assemblage analysis.” ⤴️
“CDS ‘applies critical social theory to data’ (Kitchin and Lauriault 2014). It challenges the naive assumption that data are ‘neutral, objective, independent, raw representations of the world’ (Kitchin and Lauriault, 2014; also see Iliadis and Russo, 2016: 1; boyd and Crawford, 2012: 667; Dalton and Thatcher, 2014). Rather data are regarded as subjective, ‘situational, contingent, relational, contextual’ and doing ‘active work in the world’ (Kitchin and Lauriault, 2014). Data is an expression of power (D’lgnazio and Klein, 2020; Iliadis and Russo, 2016; Kitchin and Lauriault, 2014).” ⤴️
“Big Data does not drive social change. Instead, it supports already present forces (Dalton and Thatcher, 2014) such as rising authoritarianism, political polarisation, and capital accumulation.” ⤴️
“discourse is a ‘historically contingent social system that produces knowledge and meaning’ (Adams, 2017)” ⤴️
“Discourses structure social relations by fixing—or attempting to fix—meaning and logic (Adams, 2017).” ⤴️
“ey are products of social power, and ‘this power prescribes particular rules and categories which define the criteria for legitimating knowledge and truth within the discursive order’ (Adams, 2017; also 2” ⤴️
“e Jørgensen and Phillips, 2002). However, discourses generally hide their political intentions, masquerading as a-historical, universal, objective, and stable (Adams, 2017; Hook, 2001: 524).” ⤴️
“Crawford (2012) argue that the ‘the rules about how Big Data will be used, and who gets to participate’ are set by a privileged few with the means to collect and analyse the data (675), potentially reinforcing existing power asymmetries.” ⤴️
“t materialises within the microcosm of discrete data assemblages with discourses becoming fixed within their structures, such as in roles, data architecture and governance arrangements (Beer, 2016: 5; also see Hook, 2001: 529-30).” ⤴️
“grey literature (e.g. media, industry, and government” ⤴️
“n some instances, data is traded as a commodity in generally unregulated markets formed around the buying and selling by data brokers of ‘raw’ data or enhanced data products (Fourcade and Healy, 2017: 11, 16; Kitchin, 2014b: 42-5).” ⤴️
“who has a right to the data. Data is often seen as property—something to be owned—with disputes, and increasingly, regulation to designate who has ownership rights (Birch, Cochrane and Ward, 2021: 3-4; Kitchin and Lauriault, 2014). However, there are also disputes about whether ownership is actually possible or the appropriate legal mechanism (Birch, Cochrane and Ward, 2021: 3-4).” ⤴️
“quantification of the material world of, for example, people, spaces, events, materials; labelled ‘datafication’ (Van Dijck, 2014: 198 referring to Mayer-Schönberger and Cukier, 2013). The second pass is the monetisation of the data (if sold or licensed); the data itself is in some instances quantified (Van Dijck, 2014: 200-1). However, Birch, Cochrane and Ward (2021) argue that, regarding personal data captured by Big Tech firms, users are monetised rather than the data per se (4).” ⤴️
“quantification of the data and its outcomes is used to justify both public and private investment in data infrastructure.” ⤴️
“overnments and businesses are assured that increases in productivity (i.e. in the company or economy) can be achieved with investment in enhanced collection, sharing, or use of data, i.e. exploiting data is tied to economic growth (Kitchin, 2014b: 16; König, 2017: 5; McAfee and Brynjolfsson, 2012).” ⤴️
“data is not necessarily collected purposefully, that is, related to a clearly articulated analytic question and value proposition. Instead, it is increasingly captured speculatively either as incidental ‘exhaust’ from existing processes or proactively datafying all aspects of experience for both known and unknown future uses (Fourcade and Healy, 2017: 13; Kitchin, 2014b: 6-7; Sadowski, 2019: 4-” ⤴️
“Sadowski (2019) argues that a ‘rhetoric of universality’ that data is everywhere and free for the taking ‘reframes everything as within the domain of surveillance/platform/digital capitalism’ (2-3). H” ⤴️
“quating data with naturally found resources such as ‘data is the new oil’ (Sadowski, 2019: 4), a resource to be ‘mined’ (Sadowski, 2019: 2-3, 6), or value extracted through a ‘production line’ or ‘value chain’ of data collection, storage, analysis, and use (Chun, Adam and Novek, 2018; Löfgren and Webster, 2020). F” ⤴️
“entities that become data custodians on behalf of others, for example holding data in ‘trusts’, somewhat like financial trusts (Micheli et al., 2020: 8-9).” ⤴️
“Regulation and controls are resisted by industry using arguments regarding risks to the economy (Sadowski, 2019: 5, 9). This discourse ‘provides a convenient narrative’ for software and hardware vendors to justify the purchase of their products and services (Kitchin, 2014a: 5; also see Löfgren and Webster, 2020: 2).” ⤴️
”, data is perceived as latent knowledge, a pre-factual material to be transformed by data scientists into valued insights (Kitchin, 2014b: 3-4, 19)” ⤴️
“he outcome prized within this discourse are insights not possible within the limits of the computational powers of the human brain.” ⤴️
“ncreasing the speed and types of analysis provide opportunities such as real-time or fast feedback, profiling, and prediction.” ⤴️
“readily highlighted by CDS scholars and noted earlier, an epistemological tension within this discourse is between a vision projecting data as providing objective truths and acknowledging data’s biases, limitations, culturally and historically contingent subjectivity. It includes a debate about the scope of data analytics in helping us to understand the world; whether it can enable us to be all-knowing or is limited to a subset of questions ignoring those related to, for example, meaning, politics, and ethics (Kitchin, 2014a: 8). K” ⤴️
“itchin (2014a) highlights that within the empiricist epistemology, ‘meaning transcends context or domain-specific knowledge, thus can be interpreted by anyone who can decode a statistic or data visualisation’ (4).” ⤴️
“the promise of cleansing the process of interfering intermediaries which ‘corrupt data’ with their vested interests (Kitchin, 2014b: 19).” ⤴️
“data is often discussed in terms that suggest a living oracle-like being. It is common to hear phrasings such as ‘What does the data say?’, ‘Let’s consult the data’, and ‘The data does not lie’. T” ⤴️
“proponents of data-driven science argue that the size and relationality of the data enable insights to be born inductively from the data (6). Promoters of this way of thinking claim that analysts can find answers to questions they did not know they had (Kitchin, 2014a: 4).” ⤴️
“he idea that data science transcends human analytic capabilities extends to the notion that data holders can know you better than you know yourself (Zuboff, 2019: 163). In the commercial sphere, the ideal is to anticipate and seamlessly satisfy consumers needs and wants ‘before they are fully formed’ (Fourcade and Healy, 2017). To do this, ‘the market sees you from within, measuring your body and emotional states, and watching you as you move around your house, the office, or the mall’ (Fourcade and Healy, 2017: 23).” ⤴️
“Power and influence are concentrated in those with such expertise and data access (boyd and Crawford, 2012), while outsiders, e.g. academic sociologists and domain rather than data experts, are marginalised (Burrows and Savage, 2014; Kitchin, 2014a: 5; Savage and Burrows, 2007).” ⤴️
“o proclamations of data science as the ‘sexiest job of the twenty first century’ (Davenport and Patil, 2012). T” ⤴️
“use of data for shaping the behaviour of self, others, materials, and environments. Data and its analytic outputs provide the capability to act or intervene in a more informed manner. This framing aligns with the Foucaul” ⤴️
“dian-inspired concept of governmentality, or the ‘conduct of conduct’ (Dean, 2009: 17-8; Ruppert, Isin and Bigo, 2017: 3; Rose, O’Malley and Valverde, 2006).” ⤴️
“governing with data’ is distinct from ‘governing the data’ or data governance as it is commonly termed.” ⤴️
“ataveillance (Van Dijck, 2014), ‘If you’re not the customer, you’re the product’ (Sadowski, 2019: 4), and ‘You can’t manage what you don’t measure’ (McAfee and Brynjolfsson, 2012).” ⤴️
“the diabetic monitoring their insulin levels to self-manage their condition, customers receiving pre-emptive service provision, drivers avoiding traffic jams, infrastructure owners monitoring wear and tear on equipment, businesses obtaining improved revenues through more effective advertising and fraud detection, governments able to better target services to needy citizens or identify threats such as disease outbreaks or criminal activity. I” ⤴️
“ata as resistance There are two ways of framing this discourse. Either reactively as a discourse of resistance in opposition to the logic or singular pursuit of the other three discourses. Alternatively, it can be framed actively as ‘data as agency’ (see Kennedy, Poell and Van Dijck, 2015; Smith, 2016: 267, 280). That is, the governed claiming the enhanced agency and value provided by the data, including its potential monetary, knowledge producing or governance value (Kennedy, Poell and Van Dijck, 2015; Micheli et al., 2020; Pybus, Coté and Blanke, 2015).” ⤴️
“ctors attempt to shape the behaviour of a government agency or elected representative using data about citizens, objects, or environments within their jurisdiction, and they therefore may resist.” ⤴️
”. Given that knowing you are being governed is often a precursor to resistance, a logical first step is identifying the power, governance structures, and the flow of benefits underlying data arrangemen” ⤴️
“Corporate resistance in the interests of data subjects includes industry calls for greater regulation and technological giants (e.g. Amazon, IBM) resisting government access to data created on their platforms, for example, police access to their facial recognition technology (Magid, 2020; O’Loughlin and Bukowitz, 2021: 1). Another industry response supporting data subject resistance has been the emergence of competitor services that prioritise privacy (Hodge, 2021).” ⤴️
“Sadowski (2019) argues that data is more accurately described as capital rather than the freely available and abundant raw commodity” ⤴️
“The ‘data imperative’ is to create commodities that generate data in an unceasing flow (Sadowski, 2019: 4). To resist any restriction on this flow, corporate datafication has leveraged similar tactics developed for financial capital, such as avoiding oversight through opaque technologies and avoiding regulation through political influence (Sadowski, 2019: 5, 9).” ⤴️
“Kitchin (2014a) juxtaposes empiricist programs with data-driven science (including positivist and post-positivist forms), and knowledge-driven or purely deductive science (1-5). Rather than rejecting Big Data approaches, Kitchin (2014a) makes a bid for an epistemological framing that acknowledges research’s inherent politics and is situated, reflexive, and contextualised (9).” ⤴️
“Beer (2016) draws primarily on Foucault and Hacking to illustrate Big Data’s social and conceptual origins in statistics (also see Ruppert, Isin and Bigo, 2017: 3). Statistics emerged in the early nineteenth century as an important component of population governance by states through intense categorisation and the concepts of normalcy and deviancy (Beer, 2016: 3). ‘These emerging numbers quickly came to define how people saw themselves’ and ‘how limits and boundaries were placed around actions and opportunity’ (Beer, 2016: 3). Beer (2016) emphasises how Big Data’s impact extends beyond direct use of the data and analysis to govern.” ⤴️
“collecting data as evidence for injustice (Van der Velden, 2015), b” ⤴️
”, becoming active and conscious” ⤴️
“rather than passive generators of data (Pybus et al., 2015), facilitating participatory democracy through open d” ⤴️
“ata (Baack, 2015), and democratising data governance (Kennedy and Moss, 2015; Micheli et al., 2020).” ⤴️
“Assemblages for Deleuze and Guattari answer the question ’what is a thing?’ such as an object, subject, idea, mood, or institution, within a metaphysics of the continuous rather than the discrete (Adkins, 2015: 10-11).” ⤴️
“he sensible and intelligible coexist on one continuous ontological plane rather than two discrete planes (Adkins, 2015: 1-2, 11-3).” ⤴️
“ollowing the logic of a metaphysics of continuity, things must be animated internally or immanently (in Deleuze and Guattari’s terms by ‘desire’) rather than externally by a transcendent universal law, truth, or direction (Adkins, 2015: 2-3)” ⤴️
“flux, change, and becoming are as pertinent to understanding ‘things’ as stasis and being (Adkins, 2015: 10-” ⤴️
“From their ethical perspective, universals are illegitimate illusions of transcendence and capture that separate us from our power to act (Smith, 2011: 125-6; also see Buchanan, 2008: 73); points of fixation that masquerade as immutable to channel power in particular directions.” ⤴️
“t is common to describe assemblages along the lines of a temporary collection of heterogeneous connections with tendencies towards both stability and change (Adkins, 2015: 14)” ⤴️
“assemblages are desire invested, socialised, or captured (Buchanan, 2008: 54-5, 92-4)” ⤴️
“hey are the means through which desire—the immanent conflicting drives within all objects and subjects (Smith, 2011)—invests or fixes itself in connections in the hope of increasing its potential to act, its ‘power to’ (Adkins, 2015: 96-7, 138, 153).” ⤴️
“he two are independent but exist in reciprocal presupposition; a dialectical, oppositional relationship (see Buchanan, 2015: 390-1 for an example). The form of content and form of expression do not map perfectly to each other, ’something always escapes’ (Buchanan, 2021: 33).” ⤴️
“Social formations—the focus of this section—are no ‘different in kind’ than concrete assemblages, except that they are ‘determined by historical conditions’ and are ‘desire on a social scale’ (Buchanan, 2008: 95-6” ⤴️
”). In Anti- Oedipus, Deleuze and Guattari refer to the capitalist, territorial, and statist forms as being ’molar’; however, not concerning scale, but use the term molar to refer to a tendency towards stasis and the discrete (Adkins, 2015: 48; also see 40-1 and Buchanan, 2008: 24-5). ‘Molecular’, in contrast, is a tendency towards fluidity and the continuous (Adkins, 2015: 48). Deleuze and Guattari (1983) describe capitalist, statist, and territorial assemblages or social forms as the statistical molar aggregates of concrete assemblages (183; Buchanan, 2008: 90). Concrete assemblages (or desiring-machines) both constitute these social forms and are shaped and permeated by them (Buchanan, 2008: 90-1; Deleuze and Guattari, 1983: 183). Social formations and concrete assemblages never exist independent of each other but in a reciprocal relationship (Buchanan, 2008: 90-1; Deleuze and Guattari, 1983: 183).” ⤴️
“e in a constant state of flux and mutation, sometimes in support and sometimes at odds with each other (Adkins, 2015: 219, 222; Nail, 2017:” ⤴️
”, social forms also do not exist in ’degrees of perfection measured against a pre-existing template or model, but states of equilibrium in a more or less constant movement of variation whose end result cannot be known’ (Buchanan, 2008: 91; also see Adkins, 2015: 219). The task is not to classify and judge, for example dividing things into ‘bad’ statist things and ‘good’ nomadic things (Adkins, 2015: 191). Rather, the ‘project of becoming, of creating the new, begins with seeing the nomadic in everything’ (Adkins, 2015: 13).” ⤴️
“her individualist, enclave, hierarchical, and isolate institutional forms (6 and Richards, 2017: 11) resemble Deleuze and Guattari’s assemblage types” ⤴️
“ue to the greater clarity I find in Deleuze and Guattari’s metaphysical and ethical foundations and the utility of their methodological approach.” ⤴️
“ouglas sought to identify universal, essential forms to explain culture (6 and Richards, 2017: 48-9, 78). Deleuze and Guattari, in contrast, vigorously worked to critique concepts of a universal and ordered underlying structure or ontological realm distinct from the sensible realm. T” ⤴️
”. In A Thousand Plateaus, Deleuze and Guattari add a fifth social form, the networked and polarising urban centre (Adkins, 2015: 220). Deleuze and Guattari would not advocate that you accept their analysis of the four or however many social formations as the last word, quite the contrary. What is most pertinent is the methodological approach (Buchanan, 2008: 65). They encourage us to consider a nomadic science alongside a statist science; the former in which ‘axioms are not settled once and for all’ but ‘can also be the starting points for experimentation’ (Adkins, 2015: 228).” ⤴️
“Data as latent knowledge aligns to a territorial social form that qualitatively defines and maintains tribes, territories, and truths.” ⤴️
“data as resistance aligns to a non-conformist form focused on revolution, creative transformation, and resisting repression (” ⤴️
“he discourse of ‘data as an asset’ maps to Deleuze and Guattari’s capitalist assemblage or social form (Buchanan, 2008: 107-16; Nail, 2017: 31-2), which works towards stripping all things of their qualitative relations (Nail, 2017: 31). It replaces them with homologous quantitative relations a” ⤴️
”). The territorial assemblage ‘attempts to ward off the chaos of a meaningless world’ by classifying and qualitatively ‘coding’ or giving meaning to elements (e.g. persons, objects, words) and defining their ‘nature’, or proper purpose (Buchanan, 2008: 94; Nail, 2017: 29). The remaining uncoded elements—the residual code that falls outside the classification system—are blocked from attaching to the assemblage through, for example, prohibition and taboo, and boundaries are formed, e.g. borders of towns, boundaries to gender identities (Nail, 2017: 29; also see Buchanan, 2008: 96-7). These ‘abstract relations change, but only one concrete element at a time…like a game of leapfrog. They set up some limits and by doing so create a new limit to cross…in a perpetual disequilibrium making its very dysfunction an essential element of its ability to function…Change happens progressively, one concrete point at a time’ (Nail: 2017: 30).” ⤴️
“Coding ’is not enough by itself to produce an enduring social machine’; relationships including alliances and affiliations are also required (Buchanan, 2008: 99). Alliances attempt to prevent the incorporation or ‘fusion’ of the uncoded elements by redistributing this residual or ‘surplus code’ (Nail, 2017: 29-30). These processes prevent stockpiling of wealth and create debt relation” ⤴️
“s” ⤴️
“This surplus redistribution has the effect of warding off capitalist (Buchanan, 2008: 93) or statist forms” ⤴️
“Deleuze and Guattari do not specifically relate these territorial processes to knowledge creation, but for Deleuze and Guattari, territorial assemblages are about meaning making. As noted in the earlier discussion, discourse is a process and struggle to fix meaning and define truth in the interests of certain power blocs or alliances. Knowledge creation is an ongoing process of classification, boundary pushing, and maintenance.” ⤴️
“his specialised body brings the elements ’into resonance around a [vertical and hierarchical] centralized point of transcendence’ (Nail, 2017: 30; additions not in original). It exerts controls or an apparatus of capture, also known as ‘overcoding’ (Adkins, 2015: 221-2). For instance, territory is overcoded as land or property such that ‘different properties are now comparable to one another’, that is ‘in terms of the rent they’ll support’ (Adkins, 2015: 222). This overcoding has the effect of cutting elements off from their relations to other elements (Nail, 2017: 30)—i.e. people’s historical connection to place, community and land based subsistence, and controlling their new connections, e.g. to taxed work—‘and in this way extracts its required surplus value’ (Buchanan, 2008: 105)” ⤴️
“Adkins (2015) draws on the imagery of clay squeezing from a hand; the clay conforms to the hand, but the more one squeezes, the more that escapes (223). The ‘imperial state remains only to the degree that it is able to recapture these flows’ (Adkins, 2015: 224).” ⤴️
“a site of transformative and creative revolution devoid of hierarchical and—in its pure form—social limits (Nail, 2017: 32-4). Nomadic assemblages ‘are able to change and enter into new combinations without arbitrary limit or so-called “natural” or “hierarchical” uses and meanings’ (Nail, 2017: 32). Where the capitalist assemblage provides limitless quantification, the nomadic assemblage provides unlimited qualitative transformation (Nail, 2017: 33).” ⤴️
“a nomadic assemblage: a direct participation without representation or mediation [Deleuze and Guattari, 1987: 471” ⤴️
“This kind of participation and self-management thus offers a political alternative absolutely incompatible with territorial hierarchies based on essentialist meanings, state hierarchies based on centralized command, and capitalist hierarchies based on globally exchanged generic quantitie” ⤴️
“(Nail, 2017: 33; reformatted reference in italics)” ⤴️
“This form aligns with the data discourse of resistance and agency.” ⤴️
”, in Deleuze and Guattari’s assemblage theory, nomadic ‘lines of flight’ are not a response to power per se; they ’are not phenomena of resistance of counterattack in an assemblage’ (Deleuze and Guattari, 1987: 531, footnote 39). Rather, they are primary; ‘the first given of a society is that everything takes flight’ (Deleuze, 1997).” ⤴️
“desire rather than power ’assembles the social field’ (Deleuze, 1997; also see Smith (2016) for a comparison of Foucault’s concept of resistance with Deleuze and Guattari’s). Power is secondary; it is the fixed, static, molar component of an assemblage (Deleuze and Guattari, 1987: 531, footnote 39).” ⤴️
“My aim here is to show how to realise assemblage theory’s potential more fully through the work of Deleuze and Guattari and methodologically bring together the analysis of discourse, social forms, and concrete assemblages” ⤴️
“In aggregate, concrete assemblages maintain and mutate these relatively stable, historically derived social formations.” ⤴️
“influence the creation and transformation of the smaller, concrete assemblages in a reciprocal relationship.” ⤴️
“leverage points of potential change, ‘lines of flight’ or escape.” ⤴️
”. a discourse analysis. Buchanan (2008) describes this as a destructive task of isolating and unravelling transcendent illusions (Buchanan, 2008: 117). Destruction must be approached carefully to avoid re-establishing illusions of transcendent capture in the same or a different guise (Buchanan, 2008: 120).” ⤴️
“o unearth how and why those meanings ‘were constructed in the first place’ (Buchanan, 2008: 121). For Deleuze and Guattari, this step requires going beyond interpretation or questions of ‘what does it mean?’ or ‘where does it come from?’ ‘into what Deleuze and Guattari specify as the functional or machinic realm’ (Buchanan, 2008: 121; 2017: 461)” ⤴️
“elucidate the ratio and points of stasis and change within an assemblage, look for lines of flight, new and unique connections, or points at which the assemblage might become something else, something less repressive (Adkins, 2015: 92)” ⤴️
“How does the assemblage work? What are its flows and blockages? What logics are the form of content and expression indicative of, how do the two forms hold together, where do they conflict, or what points of escape do they revea” ⤴️
“What is the mix of the assemblage types? What would it mean to break free from capitalist, territorial, and statist assemblage forms and enhance the nomadic (Adkins, 2015; Buchanan, 2008, 2015, 2017)?” ⤴️
“s liberating desire from its capture or repression, reaching ’the investments of unconscious desire in the social field’ and enabling them to coexist ’in opposite modes’ (Buchanan, 2008: 123 drawing from Deleuze and Guattari, 1983: 350).” ⤴️
“seeing fixation points as a collective choice that are open to change; they are not immutable, transcendent law. L” ⤴️
“iberating desire involves dissolving these fixation points as required if it increases our collective ’power to’.” ⤴️
“Too much stasis or too much change is hazardous; ‘both are a kind of death. Life occurs in the middle’ (Buchanan, 2021: 52).” ⤴️
“s studying how different actors compete to entrench the discursive logics within practice and structure, embedding the discourses in data infrastructures and processes. A” ⤴️
“how these fixations create flows and blockages that move data, knowledge, and power in particular directions and away from others.” ⤴️
“Deleuze and Guattari’s methodology indicates that opportunities for new disruptive connections should also get special attention as potential sites of emancipatory change.” ⤴️
“Relationships between the stages and elements within them—for instance, actors, objects, places, rituals, discourses, institutions—may be convoluted with feedback loops between the four domains in terms of economies, truth- claims, governing, and resistance.” ⤴️
“This analytic approach is not to be confused with the metaphor of the production line or the concept of the value chain in which value is seen to be progressively added at each stage of the proce” ⤴️
“value might not increase for all parties.” ⤴️
“As the data becomes increasingly distal to the context in which it was collected the risk of spurious interpretation increases, or as it progresses to the stage of governing its use may threaten the welfare of governance subjects.” ⤴️
”. Löfgren and Webster (2020) illustrate how a procedural approach of breaking the data process into stages can be used to analyse blockages created when the logics of the four discourses and assemblage types collide in practice.” ⤴️
“Kennedy, Poell, and Van Dijck (2015) highlight case studies of ‘practices of data work’ that resonate with the nomadic tendencies of direct participation in transformation, and self-governance.” ⤴️
“y supporting and amplifying emancipatory findings from these sites of action, CDS scholars can move from critique to tangible and positive change, which in aggregate can be powerful forces.” ⤴️
“lassification as an imperfect means to mediate a chaotic and disordered world (6 and Richards 2017, p. 45) and the importance of anomalies or the unclassifiable (6 and Richards 2017, p. 25).” ⤴️
“Data as resistance  A means to threaten or enhance one’s agency  Provides agency (to self or others)  Data’s value is contingent on who controls and benefits from it, rather than being inherently valuable” ⤴️
“Data activism  Data sovereignty  Tech lash  This discourse can be framed reactively (as resistance) or proactively (as agency)” ⤴️

Mentions

Four Data Discourses and Assemblage Forms: A Methodological Framework