Page 45 →2 Processual Magnitude, the Sublime, and Computational Poiesis

As a rhetorical tactic, magnitude is often discussed with regard to the piling of diction or the use of grand metaphors, or even the practice of creating and visualizing immense archives of information—verbal and visual discourses, created and spoken by humans to cultivate a sense of conviction.³ For example, Jenny Rice builds from the work of Aristotle and explains that the aesthetic experience of magnitude (megethos) can be facilitated by networked, digital media technologies to become archival magnitude: “an aesthetic inflection of a quantitative mass that gives a sense of weightiness, a sense that sustains the epistemic without relying on epistemology to structure it.”⁴ Magnitude, performed in the construction and citation of large databases of “evidence” by science denialists, for instance, offers a feeling of resolution to their claims, by seemingly locating it in the transcendence of individual perspective while emulating networked rituals of science communication.⁵ “Look, we have a database. We’ll visualize Page 47 →the data. HIV doesn’t cause AIDS.” Important to this, for Rice, is that the performance of archival magnitude, say in the aesthetics of big data visualizations, is “a sense of something coherent, a sense that possibly transcends the individual pieces of datum that are contained within that aesthetic whole.”⁶ The performance of archival magnitude is a means of amassing human rhetorical energies in support of one’s claims (regardless of factual basis), seemingly to foreclose reproach—to negate the drive to ask more questions. Magnitude can also be used in new media contexts to add coherence in the face of oppressive social structures that might support unmerited questions, as Stephanie Larson demonstrates in her compelling analysis of the #MeToo hashtag, in which she analyzes the hashtag as a vast list of stories, representing the magnitude of the problem of sexual violence, by accumulating sensations of pain to garner attention and motivate audiences to action in the face of patriarchy.⁷ In either archival magnitude or the magnitude of vast lists, magnitude can be characterized as coherentizing, or “beautiful” (in the technical sense of the word), which Immanuel Kant, in his Critique of the Power of Judgement, defines as an aesthetic construction, which represents harmony and symmetry, fixed in a bounded object (such as a data visualization or a hashtag).⁸ And as Rice and Larson have compellingly demonstrated, phenomena like archival magnitude or the magnitude of vast lists are beautiful ones, which leverage immense amounts of human energy to afford feelings of coherence to a public, either to foreclose reproach or to maintain resilience, despite unwarranted reproach.

By contrast, there exists another aesthetic, manifest as magnitude, but which can emerge from beyond the human: the sublime. Sublime experiences are moored in difficult-to-comprehend quantities and forces, which are characterized by a complex mixture of enjoyment and horror found in the pleasurable experience of an object that transcends human control and comprehension as a boundless object worthy of respect.⁹ (Thunderstorms are often used as an example of the sublime.) In contrast to beautiful magnitude, sublime magnitude, rather than encouraging a sense of coherence, cultivates a sense of disarray, located in the overwhelming of the imagination’s ability to comprehend. With concern for magnitude as an overwhelming experience in the context of machine communication, sublime magnitude is distinct from something like Jonathan Bradshaw’s rhetorical exhaustion. In cases of rhetorical exhaustion, technologies, like bots, can be used to repeat and amplify one’s claims in ways that overwhelm publics, offering not a sense of coherence, but rather a sense of exhaustion, sapping energy through amplification, reducing motivation to engage.¹⁰

Page 48 →Computational performances characterized by sublime magnitudes of computing, on the other hand, even if they are attended by angsty attunements (more on this later), overwhelm with their unpredictable, uncontrollable energies, nonetheless motivating further questioning: “What is my story? What is theirs? What matters in a story?” The immense storytelling of @censusAmericans makes clear that the processual magnitude of some computational performances represent sublime energies, which can be leveraged strategically to influence publics by invoking a sense of irresolution, located not just in the widgets produced by their operations (“Look at this procedurally generated story!”), but also in their (un)natural scale of movement (“Whoa! This bot will continue to tweet poems for the next millennium!”).

The sublime blurring of the “natural” and the “artificial” in technological discourses is not necessarily new territory for rhetorical scholarship. But within this territory, the sublime is often invoked with concern for utopian myths perpetuated in discourse about technology, rather than artistic performances and experiences entangled with lively movements of technology. For example, In their description of the rhetoric of the electrical sublime, James W. Carey and John J. Quirk point to an underlying myth of “electrical utopia,” a technologically deterministic discourse, guiding policy and relations. Electrical utopia “invests electricity with the capacity to produce automatically, on the one hand, power, productivity and prosperity and, on the other, peace, a new and satisfying form of human community and a harmonious accord with nature.”¹¹ Such is an approach to a problematic discourse, facilitated by an aesthetic of technology as a natural force. In Carey and Quirk’s words: “The first task is to demythologize the rhetoric of the electronic sublime. Electronics is neither the arrival of apocalypse nor the dispensation of grace. Technology is technology; it is a means for communication and transportation over space, and nothing more.”¹²

Similarly, David Nye has noted that the sublime in technology can be accompanied by appeals moored in the rational domination of nature (e.g., the technological sublime) or the movement of forward progress (e.g., the electrical sublime).¹³ Similarly, Vincent Mosco tracks the utopian promise of “cyberspace” to transcend physical geography and to flatten political hierarchy, as it is entangled with corporate discourse.¹⁴ Nye’s and Mosco’s ideas have been considered in the context of algorithms by Morgan Ames. In Ames’s view, algorithms are accompanied by the myth of algorithms as impenetrable “black boxes.”¹⁵ Consequently, the goal in studying algorithms and their entanglements with culture is to “dispel the algorithmic sublime that characterizes contemporary Page 49 →discourses on algorithms” by cracking open the black box—to learn about how particular algorithms operate—to counter the myth by showing how algorithmic processes are shaped by, and shape, cultural practices of surveillance, for instance.¹⁶ These approaches to the sublime and technological discourse are enlightening with regard to how we conceive of, and talk about, technology, enhancing critical sensibilities by offering tools for parsing (and illustrating the blurring of) the mythic, the natural, the technical, and the cultural. And, as such, the dispelling of the myth of the sublime in technological contexts is useful for critical thinking. By the same token, however, such approaches largely interrogate the sublime in the pejorative, leaving by the wayside the potential artistic merits of sublime experiences, such as that offered by the performance of a Twitterbot, which will ceaselessly tweet stories—exert energies—without rest or repose, for the next millennium.

To demonstrate the leveraging of the sublime energies of computational performance, I now move to an analysis of both the front and the back ends of @censusAmericans, a Twitter bot that is splendidly simple in its programming but represents vastness in the energies of its computational performance.

@censusAmericans is a Twitterbot that earned some buzz within the botmaker community, and in the broader artistic community, where, for instance, its impacts can be found as resonating into the work of other artists who wrote music to accompany the bot.¹⁷ The bot runs on Twitter, sourcing US census data to generate and tweet small biographical statements about individual, anonymous persons. For example: “I work in machine shops; turned products; screws, nuts and bolts. I don’t have health insurance. I have been married twice. I am divorced.”¹⁸ The creator, Jia Zhang, crafted the bot as a graduate student who was affiliated with the Massachusetts Institute of Technology Social Computing Group. As she writes in a FiveThirtyEight article, which details the project: “CensusAmericans will insert strangers into your life at regular intervals and will continue its automated task until it gets to the end of the 15,450,265 rows in the data set. That’ll only take about 1,760 years.”¹⁹ Such is an instance of computational performance that confronts persons with processes beyond control and prediction; it is a sublime experience.

Artists / scholars Jon McCormack and Alan Dorin have developed an account of the aesthetic experience enmeshed with generative computational art (i.e., art made via random or mutative processes in a program). Specifically, Page 50 →they develop an account of the computational sublime: “the instilling of simultaneous feelings of pleasure and fear in the viewer of a process realized in a computing machine, a duality in that even though we cannot comprehend the process directly, we can experience it through the machine–hence we are forced to relinquish control. It is possible to realize processes of this kind in the computer due to the speed and scale of its internal mechanism, and because its operations occur at a rate and in a space vastly different to the realm of our direct perceptual experience.”²⁰

Integral to McCormack and Dorin’s computational sublime is the concept of emergence, that which manifests beyond human control and prediction—poiesis, born of the human hand, but “at work” beyond it, and thus, in turn, taking on a character of natural processes: “It offers both pleasure and fear: pleasure in the sense that here inside a finite space is the representation (and partial experience) of something infinite to be explored at will; fear in that the work is in fact infinite, and also in that we have lost control.”²¹ The energies of some computational performances, despite being artificial, move in ways that feel natural, as in emerging beyond human control. Within this framing, the performance of a Twitter bot can be aesthetically powerful, not just because of what it says, but also how it moves; its lively energies strike the body in ways (un)natural.

Every hour a new micro-story about a United States citizen is shared by @censusAmericans, representing an ongoing, unresolved performance: “Last time I got married was in 1996. I have been married twice. I am divorced. I work in medical equipment and supplies. I drive to work.”²² The richness of the bot’s stories, including its blemishes, grows over time, amassing the outputs of its work on the front end: “I work from home. I don’t have health insurance. I work in miscellaneous retail stores. I have multiple ancestries.”²³ One could copy one of these tweets from @censusAmericans and share it elsewhere; one might even print out all of the tweets to date and publish them in a book. However, those copies would be missing the experience of visiting the site of the bot directly. Intimately woven into this is the exhilaration of seeing the newest tweet to pop up in the feed; the result of an automated system working on the back end, randomly parsing data and tweeting within the parameters of a Python script—the “other half” of the movements of the machine as it performs its art-making. @censusAmericans, as it exists at its Twitter account, is characterized by a lively computational performance, something which is only decayed when copied away from its site online, where the bot performs the work of art.

Page 51 →From looking at Zhang’s GitHub profile, where she has made her source code public, it can be seen that the bot is composed of three different scripts, written in the programming language Python.²⁴ The first script, “draft.py,” sources and structures raw data from the United States Census Bureau’s “2013 Public Use Microdata.”²⁵ The second script, “refine.py,” takes that structured data and randomly generates three or four sentences, sourcing individual categories from the census data form, while checking that they fit within a 140-character limit.²⁶ The third script, “censusAmericansBot.py,” posts the final set of sentences to Twitter.

According to the “Read Me” file accompanying the code, Zhang wrote the third script while working from the basis of “everywordbot,” a Twitter bot written by Allison Parrish, an instructor in the Interactive Telecommunications Program at New York University. Also important to note is that the code located in the Python script posted to Zhang’s public GitHub repository indicates a “time.sleep(14400),” which means that each tweet should happen every four hours, after a 14400-second “sleep.” As it exists on Twitter, however, it is evident that @censusAmericans is posting every hour. There is also some evidence that the version of the code on the GitHub account is set to make “tweets,” 3–4 sentences long, whereas the bot “in the wild” shows evidence of at least 5. As such, the analysis will attempt to triangulate between both the source code accessible on GitHub and the output on the Twitter profile itself in accounting for the energies of its computational performance.²⁷

Surely Zhang’s artistic hand is present in the coding. In fact, some would be quick to recognize the “craftiness” of her ability to hack census data and work within the constraints of Python as a programming language to create art.²⁸ In addition to appreciating Zhang’s artistry, however, the viewer also comes to the site to see what the bot will do. Each tweet is the result of computational processes. For example, the script “draft.py” makes it possible for “column JWTR with value 06” to be “translated into ‘I take a ferryboat to work’.”²⁹ Then, after that script runs, it is passed off to “refine.py,” which tries multiple permutations of the conversational sentences generated from “draft.py” until a tweet is crafted containing at least three sentences, but not more than four, falling within a 140-character restriction.³⁰ Finally, “censusAmericansBot.py” sources each set of sentences and, using the Twitter application programming interface, it then posts that set as a unique message to the Twitter account, @censusAmericans, once every hour.³¹ For example: “I work in electric lighting and electrical equipment manufacturing. I got married in 1969. I have a high Page 52 →school diploma. I am married.”³² The scripts undergirding @censusAmericans, “draft.py” and “censusAmericansBot.py” are the least exciting scripts; they are simply populating a data object with the values of the census data or automatically posting results to Twitter. However, there is vital randomness involved in the third script, “refine.py,” which demonstrates the sublimity of the bot’s lively movements.

With “refine.py,” Zhang is essentially handing off the census data to the system to compose a tweet, sourcing the original sentences culled from the census data file, ultimately, to decide what will make it into the final tweet and in what order. Note the random.sample() method employed in the following excerpt of the script, highlighted in bold. Note also the bits of code that follow “#”s, as these are essentially working scripts, likely leftover from originally testing the operation of the bot, which would be helpful to the coder for tracking specific portions of the processes of the bot (i.e., to print to screen the values of a working variable).

while len(newRowText) > 140 and rowLength > 3:

sampling = random.sample(range(1, len(newRow)), rowLength- 1)

rowLength = rowLength-1

#print sampling

shorttweet = ““

for i in sampling:

shorttweet += str(newRow[i])

#print len(shorttweet)

#print newLine

#print shorttweet

if shorttweet in existingtweets:

print “repeat”

else:

spamwriter.writerow([shorttweet])³³

What random.sample() is doing in this script is referencing a list of sentences previously generated by “draft.py” and choosing 3–4 of them. For a simple example, if one has a list of four items that is defined as cutesyExampleList = [“kitten,” “puppy,” “bear,” “squirrel”] and they apply random.sample() while defining that they only want two items from the list, it could be expressed as random.sample(cutesyExampleList, 2), which would produce something like [“squirrel,” “kitten”]. Important here is that random.sample() does not Page 53 →just grab random items, it also randomizes the ordering of the output. In this sense, Zhang has designed the system to engage in the four basic rhetorical operations: omission, addition, transposition, and transmutation, defined in Rhetorica Ad Herennium.³⁴ The machine is omitting, adding, transposing, and transmuting content. The initial script, “draft.py” transmutes census data into sentences, and “refine.py” adds, omits, and transposes those sentences into individual tweets. While Zhang is the writer of the bot, the bot is the lively entity performing the work of writing the tweets, enacting rhetorical action, characterized by a vastness, beyond words. As such, @censusAmericans invites the witness to simultaneously feel gratification and anxiety amid the open-ended, unanswerable character of infinity “at work,” implied by a computational performance wherein control has been surrendered to the generative activities of the machine, beyond prediction. Existing in @censusAmericans are Zhang’s energies in writing Python scripts. But alongside Zhang’s energies are those of the machine, carrying out its scripts to create on its own, invoking awe, born of the smashing physis (the chaos of nature) into poiesis (the organization of nature for human purpose).

Just as we might appreciate a painter’s ability to work from primary pigments to create signature tones and work with brushes to create an image of a human face—to work within “constraints”—so too can one appreciate Zhang’s brilliance of programming. Notable here is that artists and constraints are different. Artists work within constraints to turn things into art; conversely, the automated system that undergirds @censusAmericans is not merely a constraint on creation, but rather an entity also working under constraints set by Zhang. Integral to the energies one can experience at the site of @censensusAmericans is the difficulty in distinguishing which choices belong to Zhang, and which ones belong to the automated system. “Is the program a constraint on Zhang, or the other way around?” The notion that the creation of the bot exists between the nonhuman and the human, where, apparently, art “just happens” offers an experience, characterized by energies that enigmatize distinctions between the artificial and the natural, but manifest as machinic movement.

Crucially, the performance of infinity and the loss of control that characterize the performance of @censusAmericans is a unique melding of two types of sublime experience. Drawing on a Kantian framework of the sublime, McCormack and Dorin explain that some computational art can demonstrate both the mathematical sublime (the idea of infinity) as well as the dynamical sublime (the Page 54 →distanced experience of threat). The mathematical sublime designates experiences in which the limits of one’s ability to comprehend an idea are surpassed, exposing to a person the limits of their own capacity, leaving them to feel inadequate, but in a manner that springs forth an enjoyable moment for having the thought in the first place.³⁵ The immensity of St. Peter’s Basilica, for example, can leave the viewer with “a feeling of the inadequacy of . . . [their] imagination for presenting the ideas of a whole, in which the imagination reaches its maximum and, in the effort to extend it, sinks back into itself, but is thereby transported into an emotionally moving satisfaction.”³⁶ The dynamical sublime, on the other hand, exists in moments where one is confronted with an uncontrollable threat, but experienced at a distance, leaving the viewer to ponder limits of their own power, but while also offering an enjoyable consideration of one’s own strength beyond the usual limits.³⁷ A violent thunderstorm off in the distance or the eruption of a volcano instantiate experiences that “allow us to discover within ourselves a capacity for resistance . . . which gives us the courage to measure ourselves against the apparent all-powerfulness of nature.”³⁸ In McCormack and Dorin’s frame, computational artworks can include both categories in that they can imply infinity beyond human comprehension (the mathematical sublime) as well as loss of control to the processes of the machine (the dynamical sublime). In both cases, however (and important to the current discussion of rhetorical energy, which is concerned with the moods/ attunements that attend computational performances), there is a need to adapt McCormack and Dorin’s Kantian framework of the sublime slightly.

That is, Kant’s sublime is an experience that involves emotion, but which is actualized in cerebral exercise, finding resolution in rational reflection—thoughts, rather than moods. The sublime, in Kant’s view, “is the disposition of the mind resulting from a certain representation occupying the reflective judgment,” for “that is sublime which even to be able to think of demonstrates a faculty of the mind that surpasses every measure of the senses.”³⁹ Consequently, Kant’s sublime, even if it originates from an emotional experience, finds resolution in rational thought more than feeling. As it concerns an investigation of the rhetorical energies of vast computational performance, which is expressly concerned with the attunements of the body, Kant’s sublime is left wanting as it seems to locate rational thought, rather than embodied attunement, as the linchpin of sublime experience. This is a limitation in that computational performance can capitalize on vastness not to offer a cerebral resolution, but rather a sustained, ambient feeling of irresolution as can be seen in @censusAmericans’ lively movements.

Page 55 →To better account for such irresolution born of magnitude, I expound upon the Kantian notion of the sublime by drawing on Carolyn Kane’s X-ray sublime and connecting it to computational performance. Kane, building on the work of Giles Deleuze, explains her X-ray sublime as an inversion of the longstanding Kantian view of the sublime as an “unhinged state of mind in the subject, triggered by worldly representations that occupy the energies of both the imagination and reflective judgment, but are ultimately appeased by the latter.”⁴⁰ The X-ray sublime is an experience of the sublime, characterized not by rational resolution, but rather the feeling of unresolvedness—the lack of closure, continual open-endedness. (The removal of the “taking” of imagination by reason also is a clever way to approach the sublime in a manner that carefully moves forward from an aesthetic tradition marred by problematic gender binaries such as reason as masculine or beauty as feminine, for instance.)⁴¹

As Kane elaborates, “the failure to provide resolution is also at the heart of my concept of the X-ray sublime, which turns on the inversion of classical and modern aesthetic pursuits for unification and cohesive symbolization. The X-ray sublime does not overcome itself but instead lands in constant and perpetual chaos.”⁴² As an example of the X-ray sublime, Kane gives an aesthetic reading of Edward Burtynsky’s e-waste photographs, which depict the horrible working and living conditions wrought by e-waste disposal around the globe, foregrounding intricate messes of color and figure in the form of wires or discarded electronics parts. According to Kane, the images create a tension between pleasure and fear, posing a question, rather than an answer, productively invoking a state of openness to reconsider one’s place in, and effect on, the world. The “impossibility” of Burtynsky’s photographs “implicates us, not just ‘them’,” withholding closure, asking the viewer: “Can you look at this peacefully, undisturbed? And even if the answer is yes, then the image has at least done the work of raising the question.”⁴³

For all of the promise Kane finds in the sublime for helping persons move toward better relations, she is doubtful of the mathematical sublime, for the experience, which is moored in the grandness of ideas—like infinity—beyond the human capacity to grasp them, “is further removed and abstracted, not only from landscape, but also from concrete experience.”⁴⁴ It is easy to agree with Kane, for it seems that invoking unresolved tensions tracks better with the dynamical sublime—powerful forces, beyond human control, threatening imposition into one’s sphere of living—because it affords embodied considerations of one’s own, and others’, lifeworlds, to feel differently, rather than just think differently. But within the frame of rhetoric as energy, and within the Page 56 →context of the computational performance of @censusAmericans, it is evident that, by way of the lively movements of the performance, even the mathematical sublime—the idea of infinity—is rendered through the movements of the machine; it is transduced into the body, and in a way that cultivates an attunement to those borders between human and nonhuman, artificial and natural, instantiating an embodied experience with the uncanny.

Distinct from still images, or elevated language, computational performances are multimodal in their liveliness, and by consequence, as Minsoo Kang explains in his Sublime Dreams of Living Machines, their movements (especially movements that mimic animal and human behavior) pushes one toward an uncanny sublimity, which “pos[es] an ever greater danger to our reality schema based on the categories of the animate/inanimate, natural/artificial, and living/dead,” while at the same time, “it is indeed only a machine, no matter how good it is at pretending to be a living being.”⁴⁵ Consequently, dynamical sublimity of computational performances is spawned from a loss of control (as McCormack and Dorin maintain), but also in flashes of existential threat to humanity’s understanding of itself, wherein the imagination reaches its own limits of understanding (as Kang posits). “What is this? What am I?” As such, @censusAmericans animates an X-ray sublime experience, realized in the existential threat of an entity, which can move beyond the scope of human control and comprehension, imposing energies that strike the viewer as threatening precisely because they represent an unbounded magnitude, moored in creation beyond prediction—the idea of infinity “at work.”

Consequently, the sublimity of @censusAmericans is uncanny, performing more-than-human energies, entangled with human ones. For example, @censusAmericans will continue to generate micronarratives for 1760 years (an unfathomable length of time in contrast to the human lifetime). Knowing that this is something that could continue well after one’s own lifetime brings the realization that the bot will create things that, as of yet, are waiting to emerge, but which are also incomprehensible. This can be just as exciting as it can be disconcerting, for the bot is confronting human viewers with their own mortality—their own insignificance—contrasted to the magnitude of the cosmos. Bewildering and astonishing, @censusAmericans invites the viewer to experience vastness, open-endedness, unpredictability—things that characterize the awe of natural processes—but distinctly as the consequence of (non)human poiesis. Put differently, each new biographical tweet generated by the steadfast computational processes that undergird @censusAmericans Page 57 →encourages questions rather than answers: “Who is this person?” “To what life projects do they belong?” “How do my projects relate?” And enlivening these questions are the sublime energies of the machine, operating with the vastness of nature, imposing simultaneously a dynamical sublimity, located in the existential threat posed by a nonliving, but lively, entity and a mathematical sublimity, located in the movements of the machine as it performs infinity “at work.”

In summary, @censusAmericans offers an illustration of the uniquely influential energies of vast computational performance, which can be leveraged rhetorically to offer not a sense of cohesion—a reification of belief—but a sustained sense of unresolvedness, an X-ray sublime experience, not just in dynamical terms (i.e., immense force), but also in mathematical terms (i.e., infinity “at work”). What @censusAmericans teaches us is that the loss of control and existential threat of machinic movement in some computational performances can vitalize not a sense of coherence, but rather a sense of irresolution.

Internet artist and fellow bot-maker Darius Kazemi praises Zhang’s @census Americans as delivering “devastating summaries of the lives of real Americans culled from census data. It is poetry. Jia Zhang is brilliant.”⁴⁶ With concern to the transformation of data into art, it is more than easy to agree with Kazemi. Zhang’s bot takes “found” things that exist in something as humdrum and monotonous as census data and transforms them into pithy opportunities for viewers to ask the question that Kane articulates in her explanation of the X-ray sublime: “Can I look at this peacefully undisturbed?”⁴⁷ But beyond being a brilliant piece of art, @censusAmericans also represents a unique tactic that can be leveraged influentially and that is moored specifically in the movements of a lively, but not alive, performance of computing.

Debra Hawhee helps us understand the uniqueness of the aesthetics of @censusAmericans through her reading of On the Sublime, and specifically, in (Pseudo-)Longinus’s commentary on the divine horses of Homer’s Illiad, wherein, the “cosmic strides [of the divine horses] are awe-inspiring precisely because they bear reference to—and radically outpace—the comprehensible, earth-bound strides of human-guided horses.”⁴⁸ Homer, a human, wrote the story, but the power of the magnitude is facilitated by the energies of the nonhuman. Important here is that the nonhuman does not make the sublime per Page 58 →se, but rather it is the energy of the nonhuman operating beyond one’s sphere of influence and understanding, which Longinus would elaborate does “not . . . persuade the audience but rather . . . transport[s] them out of themselves.”⁴⁹ Similarly, Zhang wrote @censusAmericans, but the bot nonetheless moves on its own, representing energies in its stories, located precisely in the fact that they are written beyond human control. Rather than tapping into the energies of the machine metaphorically through language, it performs infinity “at work,” energies at magnitudes that dwarf human comprehension and ability, to transport one “out of themselves.”

Because it confronts the witness with a displacement of self-understanding, @censusAmericans can be described as cultivating an attunement to angst, a ground-mood, ever-present, concealed under the stability of one’s everyday lifeworld.⁵⁰ As Matthew Ratcliffe explains, the phenomenological experience of angst implies a momentary but “radical transformation of the ordinarily taken-for-granted sense of belonging to a world, where the usual sense of things as practically significant is gone from experience.”⁵¹ With the experience of angst, even if only for a moment, comes “the sense that anything is significant, ever was significant or ever could be significant is absent.”⁵² Preconceptions of value and commitment are suspended. Plans for the future are forgotten. Dwelling on the past is traded for open attunement to new possibilities of possibility. With this shift in attunement comes increased openness, a receptivity to ideas and feelings perhaps closed off by the previously concealed mores of one’s own lifeworld. Where some might argue that angst is something to be considered as an undesirable state, wherein rhetoric can be employed to make noisy archives more “tolerable” by offering structure, @censusAmericans demonstrates that perhaps angsty attunements can also be desirable, exposing that the sublime energies of vast computing can cultivate openness, rather than closedness, in association with the biographical data of real people and their own situations in the world.⁵³

In the psychological literatures, “perspective-taking,” or “actively imagining how the other is affected by [their] plight,” is considered an antecedent to “empathic concern,” which is a state associated with behaviors like helping others in need.⁵⁴ In these literatures, the ability to appreciate and understand others is often conceived as a “work in progress,” requiring practice, including hypothetical, imaginative practice. Psychologists Michael W. Myers and Sara D. Hodges explain further that, “as interactions unfold over time, mental constructions such as schemas and simulations may become increasingly more Page 59 →important sources of information for the perceiver than the actual behavior of the target person during the interaction.”⁵⁵ That is, the more scenarios one can imagine themselves into, the better equipped they will be to attune to the other when confronted with an actual interaction with another person. Important to note is that the effect of @censusAmericans is similar to the ambient music that Rickert analyzes in his Ambient Rhetoric: it “organizes an experience, not so much to persuade in any direct sense, but to attune and inflect our sense of bodily inhabitance and the cradle of intelligibility within which we comport ourselves.”⁵⁶ In the case of @censusAmericans, perspective-taking is enlivened not simply by the energies of other humans, but the more-than-human energies imbued by the “natural” movements of a lively (but not alive) entity, which implies infinity “at work,” encouraging an attunement to angst, a permeating of lifeworld, opening newly embodied considerations of the lifeworlds of others: “I walk to work. I get to work around 1:15pm. I am looking for work. I work for the state government. I don’t have health insurance.”

While one could argue that the ambiguity of the individual stories told by @censusAmericans makes it possible for a person to easily put themselves in the other’s “shoes,” it is more profound to notice that the machinic energies of the bot’s performance invoke a uniquely sublime magnitude, affecting the viewer with a sustained sense of unresolvedness and further implicating an attunement of angst. Consequently, processual magnitude is not just something to behold as a novelty of machine-generated art; it is something that can be strategically employed in contexts that demand expansions of lifeworld, broaches toward empathy.

The positive implications of the sublime energies of vast performance are not positioned here as a retort to scholars who remain skeptical of communication technology or those who might find automation itself to be entangled with dangerous ideologies, hidden under appeals to progress.⁵⁷ Nor is the analysis a naive perpetuation of technological utopianism. Rather, the analysis should be read for what it is: an effort to understand the rhetorical implications of vast computational performance. As a case, @censusAmericans demonstrates that the energies of computational performance can be leveraged through processual magnitude affording an X-ray sublime experience, located in the simultaneous activation of the “artificial” and the “natural” by an entity that moves as Page 60 →infinity “at work,” enlivening the imagination, not with a sense of coherence, but a sense of unresolvedness. Jia Zhang’s computational art, moreover, is positive, deep, and moving, not simply because of the stories it tells, but also the lively way it moves.

Kevin Brock has made an extended argument for the importance of analyzing computationally driven texts with specific regard for the back-end processes that undergird them, for those processes, in a manner unique to computational media, shape experience in ways distinct from strictly verbal texts, moored in time-based performance and often the processes themselves.⁵⁸ The current chapter extends Brock’s insights to show that when connections are made between the front and back ends of computational performances, one can expose the sorts of confusing, complex, and incoherent habits of being that can be activated by the energies of computing. The case study of this chapter illustrates those habits in which the magnitude of vast computing can be approached as dynamical, mathematical, and uncanny, distrupting the distinctions one might make between the “natural” and the “artificial.”

There exist avenues for continuing to examine the deep end of performances, characterized by the sublime. For example, @censusAmericans, while seemingly open-ended in the outputs that it makes, remains a fairly closed system, working solely within the parameters of its three Python scripts and 2013 US census data. Other, more open-ended kinds of generative art systems also exist, such as music bots, which improvise based on the input of a human musician, creating a feedback loop wherein both the bot and human “follow” each other while playing—machine–human negotiations of improvisation, making for more complex interplays of human and nonhuman energies.⁵⁹

While the example of @censusAmericans underscores the productivity of vast computing for cultivating angsty attunements with concern to relations between persons, processual magnitude is also foreseeably a viable means by which to approach the cultivation of environmental relations between humans and nonhumans. For example, a computational system meant to randomly generate soundscapes of bird songs, representing the dramatic loss in bird populations over the last fifty years, would offer an embodied experience in which the “voice” of nature is placed in the forefront. But further than this, it would offer an experience accompanied by the computational sublimity of unpredictable and ephemeral of unique soundscapes generated in real time by the machine, but which also represent species loss across the larger bird population. Such a performance would offer an embodied understanding of what three billion Page 61 →fewer birds in the ecology sounds like—to feel the conclusion and to feel it in a way that invites one to listen beyond themselves.⁶⁰

Having pursued the aesthetic end of rhetoric in a computational performance that is highly constrained by the parameters of its programming, in the next chapter I turn to the political end of rhetoric, locating it in a computational performance that is much looser in its preprogrammed constraints, which, in turn, energetically signals from its processes a “pure” mathematics.