Algo-Ritmo

Data as Assemblage

The number is no longer a universal concept measuring elements according to their emplacement in a given dimension, but has itself become a multiplicity that varies according to the dimensions considered. (Deleuze & Guattari, 1987/1980, p. 9)

Since the Enlightenment, quantitatively measured data have been understood to be abstract and objective extractions from the world. They’re said to be value free and with the application of rigorous and systematic analyses can provide information about the “truths” of the world. This view of data has not only been deconstructed by the post-structuralist aphorism “there is no outside of the text” (Derrida, 1967/1998), the above quote by Deleuze and Guattari (1980/1987) posits that the “number” can no longer be understood as a universal or abstract concept but rather must be reconceived as a multiplicity, a substantive entity that emerges from the enfolding of other elements, particles, or multiplicities. By positing number as a multiplicity, they imbue number with ontology: not a simple signifier representing the measureable social world, but possessing an ontology that is not fixed or reducible. Building on the critical theorizing on data assemblages by Rob Kitchin (2014), I will argue here that data are assemblages that are more-than-human ontologies that consist of the forces of sociopolitical relations.

To further interrogate the assumed ontology of numerical data, I’d like to consider some of the work in the cultural studies of mathematics. Mathematician and cultural studies scholar Brian Rotman (2000) argues that the Platonic assumption that mathematical logic is the language of nature assumes a metaphysical deity. Rather than maintaining this assumption, he postulates that mathematics is a semiotic system and human cultural invention and practice no different from the symbolic systems of the alphabet. Thus, according to Rotman, mathematical laws do not ascertain the “truth” of natural phenomena but rather the cultural inventions from human perception.

Work in new materialisms illuminates the ways in which nature has always engaged in intellective, communicative, interpretive, and inventive processes (Barad, 2006; Coole & Frost, 2010; Kirby, 2011). Kirby (2011) argues that culture not only pre-dates the human organism but that the human inherited culture from nature as another entangled expression of Earth’s ontology. In pushing these boundaries, Kirby leans on Brian Rotman to consider the “unreasonable effectiveness” of mathematics. She argues that the Platonic imbuing of a theology, a deity, behind mathematics as the language of nature is undermined by the notion of math being a cultural invention. But, Kirby questions, what gives mathematics its “unreasonable effectiveness”? How do we make sense of the impressive functionality of advanced technologies that all have mathematical algorithm(s) that inform their theoretical engineering and enable the auto-correct or predictive analytics of the interactive keyboards of most smart phones? Leaning on Rotman, Kirby argues it is the power of perception as “empirically originated patterns, processes, and regularity.” However, Kirby wants to question whether this cultural invention begins with the human rather than being inherited from Nature. Thus, for Kirby, Plato’s notion of mathematics preceding human inquiry and creation would include the fabrication and non-fixed prescription in time and space. The further implication of Kirby’s new materialist deconstruction of mathematics is that while data are often measured or generated by human intervention, the human inherits data from the cultural inventions of the more-than-human ontologies of Nature. These are not ontologies of “truth” but rather an ongoing process of the world trying to become intelligible to itself.

The ontology of data was also sharply placed under the purview of critical examination in Rob Kitchin’s The Data Revolution. He postulates that data are like the bricks and mortar that make up the buildings and neighborhoods that are often the focus of urban space. Data are not just taken-for-granted but rarely have data been philosophically interrogated. Kitchin insightfully argues that data are not independent of ideology, techniques, political interest, economic forces, people, cultural practices, and context. Data have histories and are temporally and spatially produced. He advances that data are a complex sociotechnical assemblage that consists of a multiplicity of entwined and mutating apparatuses. Each apparatus constitutes what is possible and impossible for data, while data intra-actively informs what the apparatus constitutes as possible and impossible within the entangled fabric of social phenomena. Kitchin (2014) states, “Data and their assemblage are thus co-determinous and mutually constituted, bound together in a set of contingent, relational and contextual discursive and material practices and relations” (p. 25). Thus, Kitchin provides a beginning framework for the critical examination and understanding of data.

Table 1 lists each apparatus and their elements of the assemblage of data. In addition to each apparatus listed by Kitchin, I also add sociopolitical relations. Sociopolitical relations are the structural relations of power and “difference” that include race, gender, class, sexuality, dis/ability, and linguistic difference among others. As will be discussed further later in this article, sociopolitical relations consist of converging forces of differentiation and hierarchization that discipline the humanity of bodies (Weheliye, 2014). While these are materially and discursively produced structural relations of “difference,” they are also one of the many forces producing the multiplicities of data. More concerning, if data are produced from a multiplicity of relational forces then are the ontologies of data acting, enacting, and intra-acting with other assemblages?

OPEN IN VIEWER

Table 1.

The Apparatus and Elements of Data Assemblage.

Apparatus	Elements
Systems of thought	Modes of thinking, philosophies, theories, models, ideologies, rationalities, etc.
Forms of knowledge	Research texts, manuals, magazines, websites, experience, word of mouth, chat forums, etc.
Finance	Business models, investment, venture capital, grants, philanthropy, profit, etc.
Political economy	Policy, tax regimes, public and political opinion, ethical considerations, etc.
Governmentalities and legalities	Data standards, file formats, system requirements, protocols, regulations, laws, licensing, intellectual property regimes, etc.
Materialities and infrastructures	Paper/pens, computers, digital devices, sensors, scanners, databases, networks, servers, etc.
Practices	Techniques, ways of doing, learned behaviors, scientific conventions, etc.
Organizations and institutions	Archives, corporations, consultants, manufacturers, retailers, government agencies, universities, conferences, clubs and societies, committees and boards, communities of practice, etc.
Subjectivities and communities	Of data producers, curators, managers, analysts, scientists, politicians, users, citizens, etc.
Places	Labs, offices, field sites, data centers, server farms, business parks, etc., and their agglomerations
Marketplace	For data, its derivatives (e.g., text, tables, graphs, maps), analysts, analytic software, interpretations, etc.
Sociopolitical relations	Structural relations of power and “difference,” including race, gender, class, sexuality, dis/ability, language, etc.

Source. Adapted from Kitchin (2014).

Although it’s not clear who is informing Kitchin’s theory of assemblage, I’d like to further theoretically unpack assemblages so as to further advance a conception of data that accounts for their lively and vibrant ontologies. For Deleuze and Guattari (1980/1987), assemblage is a system made up of the organization, arrangement, relations, and connections of actualities, objects, or organisms that seemingly appear as a functioning whole. It’s not a collection of actualities, as would be understood in the English definition of assemblage, but rather a sticky constellation of a multiplicity of forces producing an event, situation, or composite grouping or body. It is a process that functions in relation and connection with other assemblages.

There is no longer a tripartite division between a field of reality . . . and a field of representation . . . and a field of subjectivity. . . . Rather, an assemblage establishes connections between certain mutliplicities drawn from each of these orders. (Deleuze & Guattari, 1980/1987, p. 23)

Thus, it is less about the content or meaning of the assemblages, in the sense of a signifier and signified, but rather the relations and connections between forces and multiplicities. Puar (2012) argues that assemblages have the theoretical virtue of not treating bodies as separate and unique phenomena but rather directly connected to all matter beyond it; deconstructing the human/non-human division by acknowledging the ontologies that live both within and beyond the human body; understanding that the meaning of a substance is not only from signification but rather a doing in matter, a “mattering”; and, social categories such as race, gender, and class are situated in events, acts, and situations rather than characteristics of human subjects. Bennett (2005) further explicates that assemblages are “a web with an uneven topography: some of the points at which the trajectories of actants cross each other are more heavily trafficked than others, and thus power is not equally distributed across the assemblage” (p. 445).

Assemblages of data, thus, function in relation to other assemblages. They are not inert productions but rather lively intra-acting ontologies that are in relation and connection with all bodies beyond them. Data are both materially and discursively produced from the multiplicity of forces that include human and more-than-human ontologies as inclusive of the list of apparatuses in Table 1. In an era of massively and rapidly produced data from our digital footprints, the assemblages of digital data are a new reconfigured relations of production of data; a reconfiguring of the structural relations of production wherein the produced commodity has little to no labor cost for production (or modes of production) but rather mainly has cost by way of the development and maintenance of infrastructural technology, software, and analytics (i.e., means of production). Although imbued with sociopolitical relations of asymmetric distributions of power and “difference,” the assemblages of data are produced in events, acts, and situations and not based on inherent human characteristics. This has direct implications for the interpretation of statistical estimates of “difference.” Data are better understood not by an interpretation of their supposed content but rather what they do in relation to other assemblages, how they act, enact, and intra-act with other assemblages.

Algo-Ritmo 1.0: More-Than-Human Performative Acts

Computation has become central to sociocultural systems, what critical theorist Luciana Parisi (2013) has characterized as the “age of the algorithm”. To consider the ways in which data enact, form, and shape bodies, we have to turn to an engagement of how the algorithmic operations that process and analyze data function. We need to consider if data are not abstractions from nature, but rather lively assemblages of materializations and discursive formations, then how do we make sense of the mysterious power of algorithmic prediction? I begin here by discussing what Parisi (2014) refers to as “soft(ware) thinking” to frame the immanent performative acts of algorithms that make up digital architectures.

Algorithms, in computer science, are understood to be a finite operation of step-by-step instructions of calculation. They are the computational processes that make up digital architectures and, via predictive analytics and machine learning, are what produce the recommendations and suggestions that are ubiquitous digital acts of the Internet of things. The dominant tendency is to treat algorithms as mechanical operations that are contingent on human intervention or design. Their systematic operations are said to be humanly designed and modeled and thus a prosthetic tool to human cognition. However, according to Parisi (2014), algorithms are much more independent than this more traditional understanding conveys. Algorithms are actual entities that consist of finite operations of calculation as well as incomputable data sequences. For Parisi, it is the actuality of incomputabilities that provide instantiation of the immanent cognition of algorithms. That is, operating between the space of finite algorithmic operations and the incomputability of the world’s infinite complexity (i.e., information) are manifestations of forms of speculative reason that are immanent to computation. To more fully explicate Parisi’s argument, I turn to a discussion of the incomputable quantity of Omega, the name given by Gregory Chaitin (2005) of the random real number to prove the limits of the formal axiomatic systems of mathematical reasoning, and first-order and second-order cybernetics.

Mathematician and information theorist Gregory Chaitin (2005) calls the infinities of incomputable probabilities Omega. Omega is a

never-ending sequence of 0s and 1s, without pattern or structure, generated through an infinitely long computer program. Omega has a maximally unknowable numerical value and is thus irreducible. Omega’s numerical value written in binary or base-two algorithm (Omega = .11011 . . . ) is an infinite string. (Parisi, 2014, p. 265)

Omega demonstrates the irreducibility of infinite complexity of mathematics and the finite complexity of formal axiomatic systems. According to Chaitin, Omega is his digital proof of Turing’s computational halting problem. Turing’s halting problem indicates that there is no formal axiomatic system that can determine in advance whether a program will stop or continue its step by step calculations within a finite period of time. The halting problem demonstrates the incompleteness of formal axiomatic systems and the irreducibility of complexity to the finite operations of algorithmic code. Rather than considering whether a program will stop or continue, Omega demonstrates the probability that any program chosen at random will halt. Thus, Omega illuminates that there is no elegant or universal algorithm that is able to solve the problem of irreducible complexity (Chaitin, 2005).

As a science by which computational processes communicate with and attempt to model the universe, first-order cybernetics assumed the universe to be made up of mathematical logic and probabilistic entities in which the computer would be able to produce all possible solutions (Parisi, 2014). In contrast, second-order cybernetics assumes the individual to be part of the world and was developed based on the notion of reflexivity (Parisi, 2014). The reflexivity of second-order cybernetics attempted to design a feedback relation between the individual and the world, thus, producing a self-making and evolving entity. In response to the limit of computation, the self-making and self-generating algorithm reprograms itself, yet it does not have to match its initial conditions. Thus, the limit of computation would become an evolutionary threshold, enabling the system to evolve, adapt, and change over time (Chaitin, 2005; Parisi, 2014). Parisi (2014) argues that second-order cybernetics is the dominant interactive model of contemporary Capitalism and that it is this self-making system via the re-programming of itself that produces immanent forms of speculative thought within algorithmic architectures.

Speculative thought, or “soft thought,” is not a form of reasoning modeled after any form of human reasoning. It is a particular form of algorithmic cognition that is independent of human thought or intervention. As Parisi states,

. . . it may also be misleading to assume that computation is yet another extension of living thought. Soft thought is instead the mental pole of an algorithmic actual object. It is the conceptual prehension of infinite data that defines computation actualities or spatiotemporalities as the point at which algorithms stop being determined by the efficient order of sequences and rather prehend their incomputable limit. Soft thought thus explains algorithmic computation as an actual mode of thinking that cannot be reproduced or instantiated by the neuroarchitecture of the brain (the neurosynaptic network), or to the neurophenomenology of the mind (the reflexive ability of the mind to become aware of its actions on the world). Soft thought, in consequence, is autonomous from cognition and perception. (Parisi, 2013, p. 169)

Soft thought is an actuality of the more-than-human cognitions and agencies that are immanent within digital architectures.

The immanent cognitive processes of algorithms provide us with one perspective of how to make sense of the mysterious power of algorithmic prediction. What about the effectiveness of the materializations of the digital acts of prediction? Here, I’d like to turn to the Spanish word for algorithm, algo-ritmo, for symbolic help. I hyphenate algo-ritmo to refer to the Spanish prefix and postfix of the word algoritmo. “Algo” translates in English to “something” and “ritmo” translates to “rhythm.” Thus, algo-ritmo could also be translated as a “regular repeated pattern of something.” Although not the same, this nuanced translation provides another understanding of algorithms; one I argue that points more insightfully toward what makes algorithmic prediction so mysteriously effective.

The translation of algo-ritmo, I postulate, symbolically points toward Derrida’s (1972/1982) concept of iterability. Derrida argues that what makes language, speech acts, or social practices effective, intelligible, and communicable is repetition. However, rather than a simple, linear repetition, Derrida posits the repetition to be an iterability. Building on the Sanskrit meaning of the prefix iter- (or more correctly itara-) as “other,” Derrida conceptualizes iterability as repetition with alterity, that is, repetition with differentiation and not uniform sameness. First, the functionality of language in the general sense “must be repeatable—iterable—in the absolute absence of the addressee or of the empirically determinable set of addressees” (Derrida, 1972/1982, p. 315). Second, for the structural context to exhaustively determine within the limits of produced conditions of a speech act or social practice, there must be full conscious intent and presence. Without this there is no way to enclose or harness the multiplicity of fleeting and drifting effects. As Derrida (1972/1982) argues,

For context to be exhaustibly determinable . . . it at least would be necessary for the conscious intention to be totally present and actually transparent for itself and others, since it is a determining focal point of the context. (p. 327)

Thus, Derrida would be suspicious of the over-determining of the structural conditions of speech acts. In fact, for Derrida, every sign or act can be cited, and removed from its “originary” context. As such, the citationality of data assemblages and the digital acts of algorithms is a break from the given context and thereby can be re-deployed in a new context in a non-saturable way. Thus, it is not that context doesn’t matter but that every data assemblage and digital act is not totally determined by context, enabling the emergence of new assemblages when cited in a new context.

Algo-ritmo suggests the same Derridean argument of iterability regarding the algorithmic speech acts of data. In fact, algo-ritmo can be more aptly understood as the repetition of data assemblages with alterity. However, given Parisi’s arguments of algorithmic soft thought, the computationally produced speech acts of digital architectures must be conceived of as more-than-human performatives. As material and discursive formations, algorithms are actualities with histories that are in relation and connection to other actualities. Barad (2006) states that “matter is a dynamic intra-active becoming that is implicated and enfolded in its iterative becoming” (p. 151). As Barad defines, intra-acting (in contrast to interacting) is the relational acts within entangled entities, not between them. Thus, Barad postulates that matter has always engaged in an intellective, communicative, and relational process and cannot be understood separate from the performative acts and social practices of culture.

Matter(ing) is a dynamic articulation/configuration of the world. In other words, materiality is discursive (i.e., material phenomena are inseparable from the apparatuses of bodily production; matter emerges out of, and includes as part of its being, the ongoing reconfiguring of boundaries), just as discursive practices are always already material (i.e., they are ongoing material [re]configurings of the world). (Barad, 2006, pp. 151-152)

By reading Foucault’s discursive theory and Butler’s theory of performativity via Bohr’s conception of matter and vice versa, Barad develops a more-than-human theory of the performative. The force of more-than-human performatives enact, form, shape, and produce bodies. Via Barad’s reconceptualizing of performative acts, algo-ritmos come to matter through their iterable intra-activities whereby, through their immanent forms of soft thought and materialized performative acts of “prediction,” they enact, form, shape, and produce both human and more-than-human bodies. Algo-ritmo, then, is an agency that is both enabled and bound by the material and discursive limits of digital architectures while also immanently reconfiguring social borders and boundaries and re-shaping and re-making the bodies and actualities of those orders.

Although others have developed performative theories of digital acts, they have reduced digital technology and data to being either a medium of or extension to human culture and sociality. For instance, Isin and Ruppert (2015) develop a performative theory of digital acts of citizenship that focuses on the speech acts of rights claims within digital architectures. While helpful, their theory is an anthropocentric focus on human interventions of digital acts of citizenship. Here, I am focused on the performative acts of the more-than-human ontologies of algorithms that enact, form, shape, and produce bodies. Algo-ritmo does not challenge Isin and Ruppert’s performative theory of digital acts but rather pushes the ontological scope of their theory beyond the human. In other words, are there more-than-human digital claims acts of citizenship (e.g., automated letters, messages, or petitions)? Or, in what ways do the more-than-human performative acts of digital architectures enact, form, shape, or enable digital claims acts? Although the political theoretical question of citizenship is not the focus of this article, it does bring us to the larger question of this article: In the “age of the algorithm,” what happens when the immanent forms of digital acts tell us something about ourselves?

Algo-Ritmo 2.0: Racializing Assemblages of Algorithmic Architectures

A Google search on “brilliant people in the world” in September 2015 produces results of predominantly White males. The same search on YouTube results in a list of videos highlighting predominantly White and East Asian males. And, when a Google search is done on “poverty” or “oppressed people,” the search results produce a list of images that are predominantly of Black and Brown bodies. The Internet of things is riddled with these instantiations of sociopolitical relations of “difference.” In fact, when Microsoft launched its AI chatbot Tay—designed to tweet like a teen—on twitter on March 23rd, 2016, it quickly began to tweet racist, mysoginist, and transphobic messages within 24-hours (for further details see http://bit.ly/22xx4lR). In what ways are bodies formed and shaped from the more-than-human performative acts of algorithmic architectures? As the Spanish word for algorithm, algo-ritmo also points toward more nuanced understandings of the formations of “difference” as immanent forms of algorithmic acts. In Western dominant English speaking countries and communities, Spanish has been reconfigured to discursively constitute Brown bodies, especially within the context of particular discourses such as immigration, mass incarceration, poverty, and educational inequity. Thus, as a concept, algo-ritmo also symbolically refers to the more-than-human performative acts that are forming, shaping, and reconfiguring “difference” via their exponentiated iterability in digital architectures. Here, I lean on Alexander Weheliye’s (2014) theory of racialized assemblages to conceptualize the racialized, classed, gendered, queered, and disabled shaping and forming of bodies within algorithmic architectures.

As a project that seeks to push back against the Western academic popular employment of Foucault’s biopolitics and Agamben’s barelife, in Habeas Viscus (“you shall have the flesh”), Weheliye seeks to insert the theoretical work from Black feminist literary studies to more adequately account for the processes of power and racializations of the body/flesh. For Weheliye, racialization is not to be reduced to race or racism but is the very process of differentiation and hierarchization that produces the entangled race, gender, class, sexuality, and dis/ability among other social categories of “difference.” He argues that posthumanist and anti-humanist theories assume that everyone equally occupies the space of humanity, without accounting for the ongoing historicity of sociopolitical relations and the ways in which political violence has been constitutive of the hierarchy of humanity. In particular, he puts to work Sylvia Wynter’s sociogenic principle and Hortense Spiller’s theory of the flesh so as to develop a theory of racializations that accounts for the ways in which sociopolitical relations and violence mark the flesh and discipline the ontologies of humanity into full humans, not-quite-humans, and nonhumans.

As a way of developing a theory of racializations that is situated in sociopolitical assemblages and accounts for the anchoring of “difference” in the ontogenic flesh, Weheliye leans on Wynter’s (2001) sociogenic principle. By incorporating theoretical work in neurobiology, Wynter rejects cultural and biological explanations of race while still accounting for the ways in which the fabrications of race, as sociogenic, become ontogenic via the flesh. Neurobiology, for Wynter, provides a theoretical route to explain how racializations become part of the ontologies of the body via neurochemical processes that reconfigure the experience of the self. Thus, the always-already ontologies of the anatomy of the body are positioned in intra-action with the sociogenics of race. As a product of sociopolitical forces, race is then not inherent to anatomical ontologies but rather those ontologies become racialized assemblages through their encounters of racialized events, situations, or acts. Weheliye (2014) further states,

Consequently, racialization figures as a master code within the genre of the human represented by Western Man, because its law-like operations are yoked to species-sustaining physiological mechanisms in the form of a global color line—instituted by cultural laws so as to register in human neural networks—that clearly distinguishes the good/life/fully-human from the bad/death/not-quite-human. (p. 27)

The “master code” that Weheliye posits of racializations, I argue, are not just enacted via human cultural laws and neural networks but also inherited in the data and (re-)programmed code of algorithms. The data produced from the millions upon millions of human intra-actions with the more-than-human actants of algorithms are both products and producers of the sociopolitical forces of racializing assemblages. Thus, regardless of the initial code of the algorithm, as it intra-acts with myriad persons and algorithms and analyzes and learns from the data, the ontology of the algorithm becomes a racialized assemblage.

As a way of drawing a distinction between the legal constitution of the body and the social designations of the flesh, Weheliye also calls upon Spillers (2003) theorizing of the flesh. As Spillers insightfully states “before the ‘body’ there is ‘flesh,’ that zero degree of social conceptualization that does not escape concealment under the brush of discourse or the reflexes of iconography . . . ” (as quoted by Weheliye, 2014, p. 39). Prior to the legal constitution of the body is the formation of the flesh, a formation that is bound by the markings or traces of political violence designating a hierarchy of humanity. The traces of political violence of the flesh are what Spillers refers to as “hieroglyphics of the flesh” that are produced from the instruments or acts of violence such as whips, police brutality, mass shootings, or more subtly from the silence in speech acts. Spillers argues that the “hieroglyphics of the flesh” are transmitted to future generations and is concealed in what is narrativized to be pathologized or biological explanations of hierarchies of “difference.” “Racializing assemblages translate the lacerations left on the captive body by apparatuses of political violence to a domain rooted in the visual truth-value accorded to quasi-biological distinctions between different human groupings” (Weheliye, 2014, p. 40). It is the political violence and disciplining of the flesh that designates bodies as full humans, not-quite-humans, and nonhumans, rendering certain bodies as exceptional and the outside of exceptional as disposable (Weheliye, 2014).

Despite Spillers sharp theorizing of the flesh, one might ask what does this have to do with the racializing assemblages of algo-ritmo? After all, algorithms don’t have a flesh nor are they able to “see” human flesh (though we know that to not necessarily be the case). My answer is quite simple: They have everything to do with the flesh. As I discuss above, the data and code of algorithmic acts inherit the sociopolitical forces of racializing assemblages from the iterability of millions upon millions of algorithmic intra-actions with the sociogenics of “difference.” Whether the software is designed to calculate the pigmentation of the flesh from a digital image or not, the data and the algorithm inherit the “hieroglyphics of the flesh” and become racializing assemblages. The historicity of the sociopolitical forces of racializations has materially and discursively produced racial formations beyond the visual sense. As Smith (2006) has delineated, while the visual sense of the flesh may have been the inaugurating sense, colonialists began to associate their other senses with that which they observed of the flesh (e.g., “I smell nigger.”). Moreover, in the Google or YouTube examples I provided earlier, in the visual images of the search results are the digital flesh of the persons captured. It is this digital phenotype of the flesh that is part of the intra-actions with other actants. Thus, algo-ritmo posits that human actors are not the only heirs of the “hieroglyphics of the flesh,” but the more-than-human ontologies of algorithmic architectures too. The algorithmic performative acts become part of the relational and connected forces of racializing assemblages.

To further elucidate this point, I want to refer to another digital instantiation of algo-ritmo. In the Apple iOS update 8.3, one of the improvements was a redesigned emoji keyboard. For those who downloaded the update, they were able to have access to a new array of skin color pigmentations in the human faces of the emojis. However, it also turns out that if you did not download the new iOS update, you not only did not have access to this newly designed human phenotype diversity of emojis but if someone sent you one from their updated phone, the phenotypically diverse emojis appeared as a combination of the original white-person emoji next to a picture of an alien (see Figure 1 for an example).

OPEN IN VIEWER

Figure 1.

Phenotypically diverse emojis.

While it can likely be inferred that the alien symbol is supposed to refer to a foreign object to the algorithm of the previous iOS, the algorithmic response to the incomputability in this context has a long political history. The iterability of this data assemblage within the context of this speech act will have a performative force that is also contingent on the recipients. Regardless of the degree of human subjectivity behind the code for this algorithm, it is the case that the performative act of this algorithm can be a powerful force in shaping and disciplining the flesh. In this example, algo-ritmo quite explicitly disciplines the flesh and designates humanity into full humans and nonhumans.

As an immanent act beyond human intervention, algo-ritmo is a performative force that may do more than simply reify “difference.” With the ubiquity of algorithms in society, algo-ritmo has the capacity to reconfigure the boundaries of “difference” as well as further magnify the sedimentation of “difference.” Even if humans socially got their acts together, the more-than-human performative acts of algo-ritmo would also need algorithmic intervention. It is to these ethical implications that I will now turn to for my concluding remarks.

Ethical Implications and Digital Social Inquiry Considerations

If digital architectures are not just reifying “difference” but potentially magnifying the sedimentation of racialized assemblages then there are at least two areas of implications or considerations. The first implication is regarding ethical and social policy considerations of more-than-human algorithmic acts that are further perpetuating the sociopolitical forces of racializing assemblages. These are implications that would suggest the need for greater regulatory practices and accountability of algorithmic acts. That is, there may need to be built into the design of the code of digital architectures an “ethical code” that could challenge the hegemony of algo-ritmo from within. Facebook and Twitter already do a form of this digital practice but with extremely inflammatory content such as a video of a beheading. These regulatory practices may need to be expanded to include algorithmic acts of racializing assemblages.

The second area of consideration is with respect to the ontological and epistemological implications to social inquiry within digital architectures. Social science research in digital architectures assume that digital technology is merely an instrument or prosthetic for human capacities and relationality, but if the algorithmic acts are more-than-human ontologies that are magnifying “difference,” then this would reconfigure human ontologies with digital technology. Thus, social science research would need to theoretically and empirically account for algo-ritmo of digital architectures. How much of the produced effects of algo-ritmo are a product of the sociopolitical relations of racializing assemblages of human ontologies in contrast to the more-than-human ontologies of digital architectures? The extent to which they are being produced beyond human intervention and with the digital acts of algorithms will have profound implications for the extent to which racializing assemblages are beyond the ontological sovereignty of the human.

As a concept, algo-ritmo provides a theoretical framework for the racializing assemblages of the more-than-human performative acts of algorithms. In other words, the sociopolitical relations of “difference” are also inherited by the more-than-human ontologies of algorithms and materialize in the digital acts that performatively discipline humanity. The performative acts of algorithms are more-than-human ontologies that are produced from the immanent forms of soft thought. Soft thought is a form of cognition that is unique to the speculative reasoning of algorithms, a result of the algorithmic process of self-adaptation and evolution in response to incomputabilities (i.e., infinity). Like the ontologies of the human anatomy, the algorithm is not inherently racialized but becomes racialized through the analysis of data assemblages. Data assemblages are materially and discursively produced within space and time. The lively intra-acting ontologies of data assemblages are always-already imbued with sociopolitical relations of “difference.” Thus, perhaps the algorithmic acts of prediction are doing more than recommending and suggesting but also telling us something about ourselves by hierarchizing and differentiating humanity and society.