Reading : John Cayley

Andrea Geyer / Evidence (Criminal Case 40/61)
Andrea Geyer / Criminal Case (40/61)

Reading : John Cayley


The President does not read. This statement, my own, is derived from two articles in the Washington Post and the New York Times that were published before and after the 2016 election, respectively.1 I might have quoted something like it, out of context, from the title of the Washington Post article, but Trump was not yet president and I would have been pointedly discarding the adverb “much.” From the New York Times piece, written immediately after Trump’s inauguration, I will, however, quote its writer as telling us—perhaps intending some kind of reassurance—that the President, “who does not read books, is able to end his evenings with plenty of television.”

If you read, or write to be read, you tend to take care with words. I don’t think it’s a coincidence that I have not been able to find writing by experienced journalists from which I could quote the opening statement that I have made in the form that I have made it. Reading is a political concept. It is clear from what I’ve read, however, that the Trump of this presidency has a problem with reading. He can read, we must assume, but habitually he does not, and, when he does, he freely admits to being regularly and happily interrupted by far more pressing concerns. He’s “a very efficient guy,” he says. He “loves” to read but often only gets to look at his books. Still, he can reach smart decisions, he says, “with very little knowledge other than knowledge I [already] had, plus ‘common sense,’ because I have a lot of common sense and I have a lot of business ability.”2

The concept of literacy is not at issue here. We are not discussing the sense of reading that can be done by anyone who has been schooled to read. Trump is, at least minimally and conventionally, literate. We are discussing the reading that presidents are expected to do and may even be admired for doing, the type of reading that is done by the lettered rather than the merely literate, the type of reading that we do in the university. But if “read” in my original statement is also a veiled passive, if it is also the case that the President does not read, if he is not, and cannot be, read by us, then we may have to concede that something more troubling, conceptually, is being brought to our attention, something that may well have a bearing on literacy as well as on the culture of letters and the reconfiguration of political economy.

“Attention” is the mark, in this context, of a certain contemporary analysis and critique of reading. The “attention economy” is an acknowledged aspect of the wider computational economy.3 As advertising, it monetizes and drives the most powerful and significant enterprises in this realm. The attention economy is enabled by networked and programmable media, and its transactional universe, the internet, provides us with what is arguably the predominant infrastructure for all practices of reading in the developed world, including the reading that we do in the university, and in the world of letters. The overabundance of “information” and minimal costs for the production and dissemination of minimally readable texts, however, ensure that the attention economy is characterized by overload, segmentation, and distraction. At least in terms of temporality, human attention is finite and thus over-producers must compete for a type of consumption that is intended to yield, ultimately, commercial transaction of some sort.4 If we read within these infrastructural software architectures, then, inevitably, we read in some new and different way, and this must be the case, a fortiori, for those of us who have been educated, more or less exclusively, in the brave new world of our encompassing infrastructure. In a world now driven by vectoralized attention, fueling the computation economy, reading is changing or has changed.

Higher journalism and popular non-fiction have gone so far as to assemble evidence for a cognitive degeneration that is produced by this change in reading.5 It becomes stupid reading, shallow reading. This resonates in contemporary critical thought with the political economic degeneration of “systematic stupidity.”6 For Bernard Stiegler, who warns us of this condition, there is, however, a crucial choice that all of us may make with regard to developments in the technics of culture as instigated by corporate capital, even when neoliberalism guarantees that these developments are overdetermined by short-termism and benefits defined more or less exclusively in terms of self-centered individuality, instrumentality, and profit. If we adapt to these technics we consume the pharmakon of networked, computational information and infrastructure as a poison; if we adopt them with care we may be able to experience new ways of reading, for example, as therapeutic.7

Along those branches of the humanities that are sensitive to and schooled in contemporary technics, there is some evidence of a more optimistic, therapeutic reading of contemporary reading. This is best represented in the work of N. Katherine Hayles who established a distinction between reading that is either “deep” or “hyper,” qualifiers which are also applied to attention.8 In subsequent work it is clear that deep and hyper are to be seen as non-exclusive modes of reading practice, and Hayles goes on to add a third mode, that of “machine” reading. She claims generative synergies between all these varieties of reading and, while keeping faith with the close and deep reading of traditional humanities—and higher education in general—she proposes that hyper reading may be a sensitive, nuanced, effective human and humanities response to the information economy and computational culture, a matter, perhaps, of adoption rather than adaptation.9 Hayles is, moreover, concerned to emphasize the role of what she now designates as “nonconscious” cognitive processes in human experience generally, and, in particular, for reading as hermeneutic, aesthetic, and scientific project.10 Those underlying aspects of synergistic reading that are, in Hayles’ accounts, nonconscious allow her to accommodate analogous (or, speculatively: identical) aspects of machine reading—for example, inferences from the quantitative analyses of distant reading—which exploit computation’s ability to process the big data of linguistically motivated tokens at velocities that greatly exceed the temporalities of human attention, to the extent that most machine reading takes place more or less beyond the horizon of human perception, let alone consciousness.

Whatever the merits of the case for such new and alternate modalities of reading—implied by synergies such as those that Hayles proposes—these are all analyses of one particular understanding of reading, most summarily designated as the reading of (typo)graphically instantiated text. The prehistory, technics, and archive of print culture are presupposed by the reading of text, and it is the contemporary ubiquity of newer technics, exceeding those of print, that have caused us to challenge the much-analyzed traditional concept of reading. If print were not under threat then it would be far less likely that we should imagine that reading faced any sort of existential crisis.

But—without referring to linguistic literacy—what if we were to consider the possibility that a more fundamental conception of reading is being subjected, historically, to existential challenge? That, rather than trying to determine whether, for example, Trump’s failure to read books might be redeemed by multi-modal, synergetic but relatively shallow practices of deep, hyper, and machine reading, we were to accuse him of an even more troubling reading failure, one that might be expressed in a statement such as, “The President does not read in our own or anyone else’s language”? What if we were to take into consideration a concept of reading that has comprehended certain implications of Jacques Derrida’s Of Grammatology and goes on to reconfigure reading as, straightforwardly, that which brings human language into being?

Etymologically, “reading” and to “read” are well positioned to bear this ontological burden, deriving from cognates used to suggest something like the ability to make well-advised, convincing guesses, particularly concerning utterances and other forms of signifying trace. In contemporary English, we have no one word for to hear-and-understand(-as-language), and I suggest that “reading” and to “read” can and should serve as such. The move in my argument that disregards linguistic and philosophical differences with regard to the two operative support media for language—aurality and visuality, and speech and writing—is deliberate, and I take it to be authorized by the insights of Derridean grammatology. One of Derrida’s crucial arguments reconfigures the logic of the supplement, especially with respect to structural linguistics and its paradoxical misapprehension of writing as, at once, secondary to speech while pragmatically supplying primary research materials for its syntactic, morphological, lexicographic, and often even phonological analyses. I take seriously Derrida’s thesis, meticulously traced through his close reading of Rousseau, that language comes into being as, in a sense, the supplement, foreclosing all but our delusions of full, authentic speech, or of language as brought forth by presence and invested with its manifest authoritative meanings.

Any speculative first speech and, simultaneously, all speech requires readings from its speakers and from any others who know a language. The participation of others in community with ourselves is an equally essential aspect of the constitutive events. There is no way that any of the readings that we do—inherently the supplement of natural (evoking Rousseau) human experience—can be identified with anything that supports them materially or to which they refer by conventions of différance or, for that matter, difference. It is the readings themselves that constitute language and that allow us to grasp its elements, and to know, in some sense, that we have language as a faculty. To flesh out this concept of reading, I characterize it as grammaleptic, and designate the process by which it brings language into being as grammalepsis. Grammalepsis comprises all of those numberless human encounters with indeterminate but perceptible, artificial forms and gestures that resolve themselves into the readable grammé of language as it is grasped, collectively understood, and as it becomes, suddenly, the readable phenomenon in and of our world, where it had not been before.

Framed by constitutive reading, our perspective on what language is, ontologically, shifts to coincide with what we can grasp of language as we read. This all-but-tautological statement has significant consequences. It implies that language is something that human beings perform together in productive and receptive acts of reading. And because such reading is, to the extent of our current knowledge, unique to our species, this is a radically humanist view of language.11 Regardless of its humanism, grammaleptic reading—as what brings language to be in the world—is applied, by definition, to our evolved faculty of language and as such it is applied equally to what Walter Ong calls primary orality, whereas historically, our common sense of “reading” has been overwritten by history itself: by the spectacular cultural and political economic success of literacy. Myself, I would go so far as to claim that history is the linguistic imperialism of what Ong, following earlier scholars, calls grapholects; that all writing (and common-sense reading, the grammaleptic reading of graphic traces) is practice in what is properly regarded as another language or at least a distinct dialect with respect to any practice of language as speech, however correspondent.12 But you do not need to agree with this position to see that learning the language of literacy has convinced us (correctly, despite the misdirections) that graphically represented differences of writing are all that we need to read language into being. And because these differences present themselves as the discrete graphically abstracted idealities of orthography, we have, since structuralism and particularly since the advent of post-war computation, tended to assume that the operations of language are homologous with those of formal “language,” that the infinite expressiveness of language is merely a function of combinatorial operations on finite sets of structured linguistic elements.13 It was Derrida who showed us, conclusively, the inadequacy of this position. Différances that may be “invisible,” “inaudible,” or—properly stated in the terms of this argument—unreadable when expressed in one support medium, or from within any phenomenological horizon of perception—enter, at every moment, into the serious play of reading and of language. This defers (even in terms of countability, by several orders of magnitude), if it does not rule out, the possibility that différances are finite.14 The question is: who will decide, us or the robots? And will this decision be made beyond our horizons of perception?

Whatever the robots may perceive as readable and we will not, one of the most important consequences of framing linguistic ontology in terms of grammaleptic reading is now perceptible to us. We can appreciate that the entire discourse of new, changed, degenerate, synergistic, multimodal reading is constrained by its concern for the reading of written, graphically recorded, text. When this text is subjected to deep or hyper reading, this may well be reading in terms of one or more practices of human language, but even so, these readings will orbit structured differences that are graphically or typographically instantiated. When it comes to machine reading, these differences are encoded as transcriptions of graphic differences and nothing more. Currently, including at the bleeding edge of machine learning, these are the only differences that count as so-called data. Whole realms of différances in language practice remain literally unreadable to computation.15 These circumstances foreclose neither certain types of computational, data-driven inference (concerning whatever remains literally unreadable to computation in text produced by humans), nor varieties of relation and engagement across human nonconscious reading and whatever it is that computation is being “trained” by machine learning to do with the tokens of literacy. What we can say, however, is that to argue for the adaptation of reading as a whole to machine reading would be as stupid and shallow as to argue for the adaptation of a diverse, complex linguistic culture to a single dialect. We can also be fairly clear, given available sources of investment, that machines are being trained to read, however non-consciously, for reasons that must bias their readings in terms of a commercially motivated relation with transaction, rather than—by way of contrastive example—with beauty or with contemplative thought, or science for that matter. Our “smartest” machines “read” in order to promote transaction, and this implies, with respect to the culture and even the history of reading, that they read, at best, in a narrowly focused grapholect, stupidly and shallowly.

Trump does not deep read, more or less by his own admission, and he shows no evidence of being able to do so. We can say this with assurance because we are the community of language practice for this reading. Neither does he, himself, machine read. Actual machine reading is left to his campaign network, which reads and writes data points, records extracted and archived by private corporations, such as indeterminately motivated “likes” on Facebook. These data points can nonetheless be tallied with specific individual citizens in order to translate, suggestively, ostensibly “anonymous” data into tractable, transactable votes.16 “An efficient guy,” meanwhile, Trump hyper reads, “bigly” (judging by his inability to hold in mind the names and details of bereaved family members as far as official phone calls). He reads television and book covers (we know), briefings (we surmise), and, of course, tweets. We all have access, in some manner, to his tweets and we see that they are, to a certain extent, responsive to other Twitter texts. Twitter is the scene of reading for the Trump presidency.

Twitter reconfigures Trump’s reading with respect to machine reading. He hyper reads within an infrastructure that is programmatically attuned to machine reading. Twitter will only grasp a highly constrained (although recently doubled) number of symbolic, sub-lexical virtual linguistic tokens, and human readers adapt to these constraints. This does not entirely foreclose the possibility of grammaleptic reading. Human tweeters may well grasp différances that are beyond the horizons of the machine when—reading tweets—they encounter, for example, tone, punning, ellipsis, and style in general. For most readers including Trump, however, constraint is constraint in this context. And, in the last analysis, so as to survive in a world of super-corporate winners and losers, the Twitter corpus is, of course, commercially dedicated to big data analyses that build vectors for advertisement and transaction—for, that is, distraction and exploitation.

Nonetheless, this Trump edition of the Presidency chooses to read and to be read in such circumstances—publicly, and with respect to policy and executive action. In terms of my own critical thought, this implies that Trump reads, as I suggested before, in a language that is no longer our own or anyone else’s. Nonetheless I contend that the true delusions and degeneracies of these circumstances lie elsewhere, in the construction of a pseudo-public agora, a non-mutual forum for virtual language, where assumptions of presence—at the troubled heart of western metaphysics—become a catastrophic tyranny that erases language and more especially the language of shared polity. That the infrastructure of these circumstances is software-architected and silently provisioned with terms of service that are explicitly unequal and non-mutual is not my chief concern here, but requires brief mention.17 The tweeting entities of @POTUS and @realDonaldTrump are, for different reasons, granted powers of Twitter presence far, far beyond that of ordinary users. There is no software-structural necessity for this. The empowering code is constructed by private corporations that set their own terms of use, establishing artificial institutions of authority and authenticity more or less arbitrarily, particularly with respect to state regulation which seems, typically, to intervene in a manner that is responsive and compliant rather than caring or proactive.18

“[H]e keeps reading if he’s mentioned.”19

In the midst of Derrida’s close reading of Rousseau, in the second part of Of Grammatology, there is a section with the title, “Writing, Political Evil, and Linguistic Evil.”20 Derrida interrogates Rousseau’s rendition of a deep-seated prejudice against writing, one that resonates with Plato’s rhetorical, mnemonic, pharmacological warning, but which directly associates writing and political subjection, writing as the deathly, malevolent enslavement of free speech. For Rousseau, the moment of political community is “naturally” established by societies when all their subjects are present within one another’s hearing-understanding and are able to affirm in full speech the promises of collectivity. These speech-enabled promises may be destroyed, according to Rousseau (and many others), by writing, by the letter that controls without presence, and even after death. If we substitute reading for writing in this scenario—reading as equally pertinent to performances of both speech and writing—while acknowledging that reading is, as I have argued, constitutive of language, then Derrida’s underlying deconstruction of Rousseau’s pathological, paradoxical relationship with writing, political evil, and linguistic evil becomes more construable. Paradox, as a crux for deconstructive investment, is signaled by Rousseau’s reputation as writer par excellence, as self-confessed practitioner of the dangerous supplement. It’s not that these so-called supplementary practices are themselves irredeemable or that they necessarily entail political evil. Our pragmatic, ethical, political impetus must shift so as to care for a less delusional, potentially therapeutic, pharmacological understanding of individual and collective practices of language regardless of form or dialect. What we need to do is to read better, despite and against delusional infrastructures of cultural formation and political economy.

Twitter, social media, and networked, programmable media generally, pretend to make human subjects like the President and the real Donald Trump directly and immediately present—“connected” in the jargon of the vectoralist superpowers—to vast numbers of those persons who make up our institutional collectivities. Addressed, regardless of and effectively divorced from institutional affiliation, as authenticated individual accounts at terminal points of a universal internetwork, those persons, we ourselves, are rendered structurally and psycho-politically complicit with this delusion of presence, willfully ignoring the vectoral corporate infrastructures that deliver celebrity tweets “directly” to the human faculties attempting to read them. Anyone with privileged access to this inherently delusional infrastructure—delusional in terms of its representation and mediation of presence and absence—is empowered to speak and to read in any language at all, including a singular language like that of @POTUS or @realDonaldTrump. If the President does not read as we do, particularly in these circumstances, this signals a terrifying threat to language and to the political concept of reading. Our reading of the Presidency is reduced to the tweets that are delivered to our terminals, and the President himself reads only in an auto-affective narcissolect, the language of someone who will not read anything that does not bear the presence of his name, in praise of his delusional authority.21


John Cayley is a poet and Professor of Literary Arts at Brown University.


Published on July 28, 2018

1. Marc Fisher, “Donald Trump Doesn’t Read Much. Being President Probably Wouldn’t Change That,” Washington Post, July 17, 2016; Maggie Haberman, “A Home Body Finds the Ultimate Home Office,” New York Times, Jan 25, 2017. Notable also is the Fox TV interview with Tucker Carlson reported on as ‘inexplicable’ in the United Kingdom’s rightwing tabloid, the Daily Mail: Geoff Earle, “Trump’s Inexplicable Answer to the Interviewer’s Question: ‘What Do You Read?’,” Daily Mail Online, March 17, 2017. When googling, verbatim, for “The President does not read,” the author was referred, with brutal irony, to the Congressional Record, vol. 150, part 11, June 25–July 14, 2004, 15032. Jim McDermott, Democrat, was given five minutes during a debate on the Patriot Act under the rubric “Knowledge is power in American politics” and quotes the then President, George W. Bush, “I don’t read newspapers. I don’t read books. Except for children’s books when there is a photo-op possibility. I only take information that is pre-chewed by my staff and brought in to me and given to me.” This essay helps to explain how it can be that the circumstance of “the non-reading president”—clearly, not without infamous precedent—has, nonetheless, worsened.

2. Earle, “Trump’s Inexplicable Answer to the Interviewer’s Question: ‘What Do You Read?.”

3. Yves Citton, The Ecology of Attention [Pour une écologie de l’attention] (Cambridge: Polity Press, 2017).

4. We are, provisionally, assuming that reading is a type of consumption here but we must bear in mind that this is in fact a crucial question for any interrogation of our political concept. The economically implicated address to human attention is often couched in terms that cast the human consumer as a “product” (cf. Citton, The Ecology of Attention, p. 9). “If something is free, you are the product.” (This is, apparently, an internet meme, of indeterminate origin.) I believe that this reading is a failure of political economic analysis with respect to any putative “attention economy.” The power and authority of corporate entities in the computational economy is substantive and material. It establishes, following McKenzie Wark, a new exploitative class of vectoralists, who are also now heavily invested in more traditional capital. McKenzie Wark, A Hacker Manifesto (Cambridge: Harvard University Press, 2004). Compliant, adaptive consumers of vectoral exploits may be an artifact or, perhaps, a by-product of these exploits but it is, I believe, a misdirection to see any sense of “them” or “us” as produced. Our agencies are compromised by complicity with processes of extraction and exploitation, but they have not been produced or owned as such. They can and should be redirected.

5. Nicholas G. Carr, “Is Google Making Us Stupid? What the Internet Is Doing to Our Brains,” The Atlantic, July/August, 2008; The Shallows: What the Internet Is Doing to Our Brains, 1st ed. (New York: W.W. Norton, 2010). The Atlantic article by Carr has its own Wikipedia page.

6. Bernard Stiegler, States of Shock: Stupidity and Knowledge in the Twenty-First Century [États de choc: Betise et savoir au XXIè siecle], trans. Daniel Ross (Cambridge: Polity, 2015).

7. Bernard Stiegler, For a New Critique of Political Economy [Pour une nouvelle critique de l’économie politique], trans. Daniel Ross (Cambridge: Polity, 2010), p. 82 and passim; What Makes Life Worth Living: On Pharmacology [Ce qui fait que la vie vaut la peine d’être vécu], trans. Daniel Ross (Cambridge: Polity, 2013), pp. 105-106, 127-130, and passim.

8. N. Katherine Hayles, “Hyper and Deep Attention: The Generational Divide in Cognitive Modes,” Profession 2007, no. 1 (2007).

9. N. Katherine Hayles, “How We Read,” in How We Think: Digital Media and Contemporary Technogenesis (Chicago: University of Chicago Press, 2012).

10. N. Katherine Hayles, Unthought: The Power of the Cognitive Unconscious (Chicago: University of Chicago Press, 2017).

11. James R. Hurford, The Origins of Language: A Slim Guide (Oxford: Oxford University Press, 2014); Robert C. Berwick and Noam Chomsky, Why Only Us: Language and Evolution (Cambridge: MIT Press, 2016).

12. Walter J. Ong, Orality and Literacy: The Technologizing of the Word, ed. Terence Hawkes, New Accents (London: Routledge, 1982; repr., 1995). Ong’s paraphrase: “A grapholect is a transdialect language formed by deep commitment to writing” (p. 7).

13. It is salutary to reflect, in this context, on potential distinctions to be made between “spelling” and “orthography” with regard to practices of language that are constituted by the reading of written, graphically recorded, text. “Orthography” in the sense of “correct” writing is a relatively recent phenomenon, coincident with the advent of authoritative dictionaries and the (contestable—see Ivan Illich and Barry Sanders, The Alphabetization of the Popular Mind [London: Marion Boyars, 1988], pp. 52–65) correspondence of (countable-noun) languages with national jurisdictions. “Spelling” can be taken as something else, something that, for example, Shakespeare was doing as he (and/or those who transcribed his work) playwrote and before orthography could pretend to collapse what may have been différances into what we moderns think of as the Shakespearian “vocabulary” in “corrected” spelling. “Différance” itself, clearly, represents a transgression of orthography until—how is this determined?—it becomes a new “word” and is thus re-subjected to the underlying regime of combinatorial linguistics. Consider what might be the effects on the projects of structural linguistics (even today), Chomskian (combinatorial) linguistics, or machine translation if human writing was practiced as spelling in a différantial sense rather than as orthography. If reading is a political concept, spelling could rediscover a place at the forefront of activism.

14. My readers may already have noted that I am taking the liberty of using “différance” in a manner that is as applied as it is theoretical. Specifically, some readers may have been troubled by the pluralization of différances. I read Of Grammatology as an essay in the philosophy of language. The proposal of grammatology as a positive science of writing is shown, by the essay itself, to be a more or less doomed project, particularly so long as this proposal continues to be contained (read) within the episteme that it critiques (Catherine Malabou, “Grammatology and Plasticity,” in Changing Difference: The Feminine and the Question of Philosophy [Cambridge: Polity Press, 2011]), but also for the reason that, as Catherine Malabou points out, Derrida himself conceded that différance might be superseded. Derrida makes this concession in the very essay on différance that elaborates its crucial role for his critique of ontotheology, in the course of arguments that can be read as reserving the term for a singular—and, perhaps, singular only—role in critical thought (Jacques Derrida, ‘Différance,’ in Margins of Philosophy [Chicago: University of Chicago Press, 1982], p. 7). Personally, I am compelled and engaged by the ontologically implicated philosophical claims that Derrida explores by way of différance, those that trouble even “God” or the “ineffable,” but, in practice, I prefer to play with the traces of this term in what I take to be the political actualities of language. If “différance” is a word that is not a word, it is most certainly an unabashed witticism and a playful, pretentious instance of itself, the singular instantiation of a concept that is not a concept. (Geoffrey Bennington and Jacques Derrida, “Différance,” in Jacques Derrida [Chicago: University of Chicago Press, 1993], p. 70.) Granted, it introduces deferral and temporalization into the economy of what might otherwise be comprehended by “difference” but its mute—necessarily imperceptible—difference is also meant to distinguish whatever it is(n’t) from the concept of difference because, clearly, difference doesn’t quite cut it, philosophically, regardless of temporalization. Difference can be perceived by us, and its instances can be transcribed, formally, regularly. “Différance” cannot. However, in practice, in an applied grammatology, it seems to me to be entirely legitimate to use the same orthography (technically speaking), for a (non)concept in the discourse of language or philosophy or both, and also for certain countable instances of linguistic forms that are related to that concept. Despite any nuances of structuralist analyses, it is untroubling for us to consider differences that are indicative of difference. If you want to insist that conceptual “difference” is a different word with respect to countable “difference(s),” this is also legitimate. Our comprehension of the relationship between the terms has not been foreclosed by usage or lexicographic distinction. Thus, in my applied grammatology, “différance” is a (non)concept but it is also, manifestly and historically, a linguistic form. As Geoffrey Bennington puts it, “[différance] can be neither a word nor a concept . . . but at the same time it is only a word/concept that is not sheltered from its own effects” (pp. 73-74). In my argument here, the concept of différance allows me to distinguish both combinatorial, formal systems working with finite sets of differences (structuralism) and also our graphically recorded literature of transcribed differences (history-as-language) from the human practice of language as such (thus demonstrably poststructural), within which we also enjoy the constant, continuous play of generative différances that are in principle, and may remain, imperceptible (mute, invisible) while they are nonetheless simultaneously grasped and read in the processes of what I here call grammalepsis. I think it’s a great joke (or: “it’s jokes” as young Brits would say) that, in spoken French, without the assistance of an article, there is no difference between “différance” and “différances.”

15. This is not to say, however, that such différances are perpetually or in-principle inaccessible to computation. If a human person grammaleptically reads a virtual linguistic form and, in doing so grasps one or more new différances, they may choose to encode these features according to one of the schemes of computation, feeding new differences (now differences as such) back into the system. There is also the possibility—acknowledged in the text—that computational process is able to infer différances from data within which they were not explicitly encoded.

16. The United Kingdom based company, Cambridge Analytica, has become the most notorious example of a “marketing” or public persuasive “service” that is politically biddable (thus tending to bias in favor of commercial power) and driven by big data analytics. Basically, these allow, initially, the correlation of a relatively small number of data points with a pragmatic set of “personality types.” Claims are made in the press for the effectiveness of a single data point, a single Facebook “like,” in this context. See Ian Sample, “One Facebook ‘like’.” The data is purchased or otherwise obtained from the vectoralist superpowers, where it is aggregated after having been freely harvested, i.e., property-grab-enclosed, from vast numbers of terminal accounts, i.e., ourselves. Subsequently, all the “anonymized” terminal accounts can nonetheless be individually targeted with psychologically personalized persuasive material. These methods are likely to have had a significant if not conclusive effect on the Brexit referendum in the United Kingdom and the 2016 U.S. presidential election. See, amongst many such reports, Jamie Doward and Alice Gibbs, “Did Cambridge Analytica Influence the Brexit Vote and the Us Election?,” The Guardian, March 4, 2017 and Stephanie Kirchgaessner, “Cambridge Analytica Used Data from Facebook and Politico to Help Trump,” The Guardian, October 26, 2017.

17. For more on this by the author, see John Cayley, “Terms of Reference & Vectoralist Transgressions: Situating Certain Literary Transactions over Networked Services,” Amodern 2 (2013); “Of Capta, Vectoralists, Reading and the Googlization of Universities,” in Digital Humanities and Digital Media: Conversations on Politics, Culture, Aesthetics, and Literacy, ed. Roberto Simanowski (London: Open Humanities Press, 2016).

18. A prime example of arbitrary and secretly-certified corporate-sponsored authority/authenticity on Twitter is “verification,” a tick mark indicating, ostensibly, that the identity of the entity operating the account has been verified. This has become a much sought-after badge of honor, mysteriously bestowed and conveying significant prestige to the account and account holder. Cf. Alison Hearn, “Verified: Self-Presentation, Identity Management, and Selfhood in the Age of Big Data,” Popular Communication 15, no. 2 (2017). For a recent study that provides a nuanced and scholarly approach—in the context of social science metrics—to related issues of inequitable social media representation and access see Carolin Gerlitz and Bernhard Rieder, “Tweets Are Not Created Equal: Investigating Twitter’s Client Ecosystem,” International Journal of Communication 12 (2018).

19. Steve Holland and Jeff Mason, “Embroiled in Controversies, Trump Seeks Boost on Foreign Trip,” Reuters [Online], May 17, 2017.

20. Jacques Derrida, Of Grammatology, trans. Gayatri Chakravorty Spivak, Corrected ed. (Baltimore and London: Johns Hopkins University Press, 1997), pp. 167-171.

21. This concluding statement was composed as a consequence of arguments set out in the text itself and in response, particularly, to Derrida’s readings of Rousseau. Its significance is, nonetheless, startlingly supported by reports in the press that staffers, preparing briefings for Trump, select material that contains his name, to keep him reading. Holland and Mason; Andrew Griffin, “Officials Put Trump’s Name in ‘as Many Memo Paragraphs as We Can Because He Keeps Reading If He’s Mentioned’,” The Independent, May 17, 2017. I am grateful to my student, Griffin Smith, who referred me to the Reuters article.