“New media studies,” Brian Lennon announced a couple of years ago, “has discovered temporality.”

The digital field’s preoccupation with presentism and futurism – and the periodizing that balancing between them requires (marking off “first” from “second” waves, “golden” from “silver” ages, and so on) – lends itself to fantasies of change that are familiar to anyone who has thought much about how we have got to where we are. But the temporalities of media, and in their own way – different from but dependent on the qualities of their objects – the temporalities of media disciplines, are more irregular than talk of ages and waves can encompass. As Lennon observes, the always already new figures “a more absorbent fold, perhaps” than remediation. Bolter and Grusin’s influential formulation of that concept entails a modern, progressive medial evolution, repeatable and accessible to measurement – and on that basis, accessible also to a progressive media theorization [Bolter 1999]. In contrast, the always-already aims at experiences of change that are tangled in their way of marking time: looking ahead, but also visited by atavisms and negativities that make imagining what lies ahead a messier undertaking, subject to uncanny feedback systems.

A case in point is represented by the two episodes that frame Lennon’s reflection on new media’s turn to historical reflexivity. The first of these is the publication during the 1990s of Michael Joyce’s critical writings on digital media and culture, “untimely” in Lennon’s estimation and overshadowed by the attention given to Joyce’s hypertext fictions of the same period. The second is Joyce’s subsequent withdrawal from literary-critical circles whose other members imagined themselves to be following his lead – the significance of which, Lennon notes, may be discerned in the shocked silence with which that withdrawal has been met.[2] Here, Lennon ingeniously traverses this disappearing act which continues to leave its mark. Joyce’s secession in or around 2004, he proposes, should be read not as a retreat from literary-critical dialogue, but an act of provocation directed at a culture of scholarship that had been too hastily constituted from Joyce’s precedent, and was unselfconsciously (and is still) spinning away from the complexities of that precedent. Lennon contrasts the essayism of Joyce’s criticism of this period – the errant, unfinished, and dissatisfied voice of his writing, more radical in its way of acting out than was his hyperfiction – with a more controlled and compact critical attitude that seems increasingly characteristic of American new media studies.

The reduction of Joyce to hyperfiction author, in the new media studies scholarship that more or less brackets Joyce’s critical project, is in fact nothing new as a disciplinary gesture; rather, it is a repetition of the founding ruse of literary-critical modernity, in what Clifford Siskin has called “novelism”: the generation, from a heterogenous and yet unified (or combined and uneven) field of writing, of the separate positions of the self-identified critic and her critical object, produced by the writer. In the subordination of writing, a discourse and an institution, to the (fabricated) research object “the novel,” modern disciplinarity in the literary sphere naturalizes writing as mimesis – that is, gives it a job to do, in determined opposition to the radical self-reflexivity of writing as mass professionalization itself, in its capacity simultaneously to expand and to contract the division of intellectual labor. [Lennon 2009]

Joyce’s stopping out, Lennon surmises, was at least in part in response to his reduction to the role of its most important critical test case [Harpold 2008, 6.01]. It was also (and I think this among the most important implications of Lennon’s reading of his framing episodes) an expression of dissatisfaction derived from Joyce’s earlier way of writing out his relation as critic to the objects of his criticism. Sacrificed in the field’s coalescing around the novelist mode of research is some of the raw productivity of writing and reading in the essayist mode. There, media are met in a more relational way, and without much assurance concerning what they make happen – in, as Joyce had observed, the “momentary awkwardness” of an encounter with something that may be unprecedented [Joyce 1995, 219–26]. If, as Lennon proposes, new media’s discovery of an always-already-new temporality is a sign of its arriving at a position of more unified disciplinarity, this has involved setting aside the excesses of first wave theorists such as Joyce. That is a way of effectively focusing on the job at hand, but only at the cost of repressing some of its complexity, or the complexity of our imagining that there is a job or jobs, and not instead some aggregates of unfinished and uncompletable – unendliche [Freud 1937] – encounters. Repressing, that is, styles of critical engagement that are less confident of their security and more anxious (or euphoric) about media’s inmixing of subjects and objects, which anxiety (and euphoria) Joyce’s essayist precedents brilliantly exemplified.

I want to give shape to this line of reasoning along four turns of language, which I draw from the graphs French psychoanalyst Jacques Lacan named the “four discourses,” introduced in his XVIIth seminar, L’Envers de la psychanalyse (“The Other Side of Psychoanalysis,” 1969–70; [Lacan 1991a]; [Lacan 2007]).[3] The graphs figure (in the order in which I will discuss them) the discourses of the University, the Master, the Hysteric, and the Analyst (Figures 5–8). Each describes relations of the linguistic foundations, subjects, and objects of knowledge in a different way, according to different structural relations of those elements. By extension, each of the graphs figures distinctive, but related, disciplinary practices, the efficiencies of which are characterized by these different structural relations. At the risk of disclosing the punch line before most of the work of setting it up, my claim is the four discourses describe the dominant effects of modes of critical engagement and teaching in the contemporary digital field. The graphs’ descriptions of these modes demonstrate how it is that we got to where we are, and point toward other, I think more productive, paths of disciplinary practice.

The graphs of the four discourses are highly formalized, and the typographic and terminological units they circulate have specific meanings in Lacan’s thought. Once these units are defined, the graphs operate on them with a notable directness and concreteness; they perform the relations they describe. (Dēscrībō: they write out these relations.) In this regard, the graphs are uniquely productive, in that they generate from determinate conditions of knowledge, new forms of knowledge. What I am after in this essay is a way, with the graphs of the four discourses also in hand, of taking hold of recent critical and teaching practices of the digital field, with the dual aim of illustrating logics of those practices and of making something new out of them. Before putting the graphs to work in this way, it is necessary to move programmatically through their terms and shapes, so as to engage their effects.

Prior to 1969, Lacan’s uses of the term “discourse” [discours] are largely consistent with Émile Benveniste’s influential definition: “discourse is language put into action, and necessarily between partners”  [Benveniste 1971, 223]; “[it is] every utterance assuming a speaker and a hearer, and in the speaker, the intention of influencing the other in some way”  [Benveniste 1971, 209].[4] That definition fits with the specific social-linguistic relation of the psychoanalytic clinic (see for example, Benveniste 1971, 223–30); for Lacan, it applies more generally to the encounters of the speaking subject with the entire field of language (the symbolic order, the “big-O” Other), which disturbances of speech in the clinic bring to the foreground. Thus, “every bungled action is a successful, even ‘well-phrased’ discourse”  [Lacan 2006, 222]; “the first resistance analysis faces is that of discourse itself”  [Lacan 2006, 348]; and the many variations in Lacan’s teaching on “the unconscious is the Other’s discourse” (for example, [Lacan 2006, 316]).[5]

Beginning with the introduction of the four discourses in Seminar XVII, Lacan amplifies discourse’s knotting of its interlocutors: discourse is now characterized as “a social bond [lien social], founded on language [langage]” (1972–73, [Lacan 1998b, 16–17]). This formula echoes Saussure’s definition of language [langue] as a “social product of the faculty of speech”  [Saussure 1966, 13] but, crucially, shifts the locus of sociality from the abstract code of langue (the system of signs deployed in speech [parole]) to the material circulation of signifiers between speaker and addressee.[6] In this reading, the social bond is realized in a concrete, dynamic exchange in which subjects recognize each other and themselves according to (conscious and unconscious, present and past) relations of signifiers. Those relations are determinate; the sociality of the bond obtains, so far as Lacan is concerned, wholly in signifying operations, not in the conventions of langue or as a consequence of the positive content of speech.[7]

Lacan builds the graphs of the four discourses within this material-relational schema of speech by mapping it onto circulations of four terms which are of particular significance in his teaching: S1, the master signifier; S2, the field of knowledge; $, the divided subject; and a, the objet petit a.

The master signifier (S1, Figure 2) orients the expression and reception of discourse [Lacan 2007, 13]. In this role it is comparable to the famous “quilting point” [point de capiton], the site of convergence in a signifying chain in relation to which the rest of the chain is situated, retrospectively and prospectively [Lacan 1993, 268], [Lacan 2006, 681]. We are familiar with the master signifiers of the digital field; as scholars and teachers we traffic in them and appropriate them to new ends: terms like avatar, home, interface, screen, trail, and proper names like Bush, Engelbart, Nelson, and so on. Our fluency in them is a sign of disciplinary competence, and the invention of new master signifiers a key to our professional success. Our speech is bound to contours introduced by S1; we can’t imagine what we achieve in discourse without it [Lacan 2007, 129–30, 188]. Consequently, Lacan accords this signifier a special importance: there is no anchor of a discourse’s reference, or the subject’s position in the field of speech, except by way of S1s cleaving off from other signifiers as their support and alibi [Lacan 2007, 189]. As Lacan envisages it, this is a strictly formal operation; whatever positive content S1 may be imagined to convey in other contexts is irrelevant to its function in this regard.

S2 (Figure 2) figures the battery of signifiers whose network constitutes the knowledge (savoir) transmitted by a discourse, in which S1 intervenes as a point of reference, as the term that anchors the expression of knowledge. The temporality of the intervention is ambiguous. Strictly speaking, S2 may precede S1 as a mere aggregate of signifiers, gathered incidentally to the situation in which speech (an ordered chain of signifiers) comes to function as discourse (a social bond founded on language). S1s intervention in S2 makes it appear – that is, effectively makes it come to pass – that S2 specific to this discourse articulates an “already structured field of knowledge”  [Lacan 2007, 13], for which S1 stands, retrospectively, as its support. In this way, S1 impresses order on the expression of S2, in relation to which the speaking subject recognizes itself and others by way of their adherence to S2, as a style or idiom of a shared vocabulary. When we identify ourselves as “new media scholars and teachers” or “digital humanists,” and more so, as scholars and teachers operating within more narrowly-defined subfields (“game studies,” “platform studies,” “software studies”), we situate our speech within fields of knowledge (S2 ) that are contoured by specific master signifiers; we acknowledge the subjection of our speech to the relations of S1 and S2.

$ (Figure 3) figures the split or divided subject. In Lacan’s thought, the speaking subject is said to be split by effects of language in two respects, in that what is possible for the subject to say is always at a remove from that to which speech can refer, and in that the aim of the speech act is always ahead of or behind what it signifies.

The first dimension of this split follows from Saussure’s division of the sign into signifier and signified [Saussure 1966]. From the breakthrough of the 1953 Rome Discourse forward,[8] Lacan insisted that this division be rigorously respected in psychoanalysis: anything that the subject communicates of its desires must be recognized as always-already destabilized by the non-coincidence of the signified with the signifier [Lacan 2006, 227, 595–6, 712], [Lacan 2007, 33]. And more so: Lacan’s revision of Saussure’s doublet – as indicated in the “algorithm” S/s [Lacan 2006, 428] – upends semiology’s privileging of signified (s) over signifier (s), and accords the signifier and relations between signifiers priority in the scene of speech. His famous opening parry in the “Seminar on ‘The Purloined Letter,’ ” that the symbolic order is “constitutive for the subject,” which is determined by the “itinerary of a signifier”  [Lacan 2006, 7] might be rewritten as: what is concretely said in the clinic (especially unaware wordplay and slips of the tongue) matters more – it is more directly productive – than what is meant (to be said).[9] In the graphs of the four discourses, the expression of $ – that is, the expression of the subject qua speaking subject – is oriented by S1s interventions in S2 specifically along the fissure of this primordial non-coincidence [Bracher 1994, 113], [Lacan 2007, 170].

The second dimension of the subject’s split in language is signaled in grammatical and temporal ambiguities of speech, as described by two related concepts which Lacan draws from the work of Roman Jakobson – the “shifter” – and Émile Benveniste – the distinction between the subject of the statement and the subject of the enunciation. A shifter [embrayeur] is a linguistic element whose meaning cannot be determined without taking into account that a sender and a receiver are engaged in an act of communication [Jakobson 1990, 398].[10] There are many species of shifters [Ducrot 1979, 252, 324ff], but first and second person pronouns represent the plainest examples: the terms “I” and “you” designate a particular person – the one who speaks, or the one to or about whom is spoken – only in the context of a speech act. There, the shifter is associated with that person by convention (“you” = the woman to whom I speak; in Peircian terms, the shifter “symbolizes” her), while at the same time it points to her by virtue of an immediate, existential relation to the speech act; it concerns specific conditions of the speech act as well as she. (“You” = the woman to whom I speak here and now, in the scene of speech; the shifter “indexes” her. Jakobson calls shifters “indexical symbols”  [Jakobson 1990, 388]).[11] For Lacan, the duplex aspect of the shifter is evidence of the relationality of discourse, particularly with regard to the speaker. When I speak of myself (“I speak”), the agent for and of whom I speak, and the signifier that figures that agent in speech are not ontologically coincident; the first person pronoun designates only a point of contact between this supposed agent and a signifier. Furthermore, the signifier and this supposed agent can only seem to coincide in that my speech is directed to an other (my addressee – “you” – but more concretely, another signifier, ju:), in relation to which the signifier I (aɪ) has meaning. (According to Lacan’s famous dictum, “The definition of a signifier is that it represents the subject not for another subject but for another signifier” [Lacan 1972b, 194] [Lacan 2007, 47–48].) Moreover, because Lacan treats the shifter as an indexical signifier [Evans 1996, 182], its notional reference, contingent on the structure of the speech act, is subordinated to patterns of the signifying chain. Whereas linguists treat the shifter as a special case in problems of linguistic reference, for Lacan the shifter (and especially the I of the first person) exemplifies the speaking subject’s fundamental destabilization ($) in the field of speech [Lacan 1958, 19 Nov. 1958], [Lacan 2006, 556].

Correlatively, Lacan radicalizes the implications of Benveniste’s distinction between the subjects of enunciation and statement [Lacan 1971, 223–30]. The statement [énoncé] is the chain of signifiers that appears to stand apart from the speech act in place of things and events to which the speech act refers. The subject of the statement is the agent designated (or implied to have been designated) thereby. The enunciation [énonciation] is the speech act as it takes place in the here and now of an utterance. Its agent exists only in the moment of the utterance, and is immediately thereafter eclipsed by the subject of the statement. It is the illusion of a continuous present in speech, Benveniste says, which sutures the temporal split between these subjects [Benveniste 1971, 227]. For Lacan, this split (refente) is another consequence of the determinate effect of the signifying chain as evidenced in the shifter [Lacan 2006, 650], which designates the subject of the enunciation (viz., associates that vanishing entity retrospectively with the subject of the statement) but does not signify that subject, in the way that Lacan understands “signifierness” [significance] to mean a dynamic interplay of elements of the chain in the here-and-now [Lacan 2006, 677, 764–65]. The subject qua speaking subject is therefore suspended between ($) the registers of the enunciation and statement [Lacan 1998a, 139], [Lacan 2006, 758].

The fourth term of the graphs, the a, is the glyph of Lacan’s objet petit a – “the little a object,” considered by devotees of Lacan to be the most original and nuanced theoretical element of his teaching [Evans 1996, 124–26], [Miller 1990]. For my purposes, objet petit a can be characterized as a thing excluded from discourse while also determining its structure. Partly real, partly imaginary, the a persists outside of the symbolic, a remnant of language’s parceling of the undifferentiated real into series of manipulable elements (signifiers), which the speaking subject combines and recombines in an (always unsuccessful) effort to bring the a into representation.[12] Thus Lacan describes the a as the object-cause of desire, urging on desire's expressions, which find their motive force in being unable to resolve a’s status within the field of language [Lacan 2001, 207].[13] This is the impasse of desire disclosed by the psychoanalytic clinic: desire’s primordially extrinsic cause (a) can’t be brought into speech (S2), as an unmediated experience of the object-cause is forbidden to the speaking subject as such ($); it can only be named in absentia by retroactively pinning speech to a particular signifier (S1); language can convey no positive object that satisfies desire.[14]

The radical eccentricity of objet petit a compels the neurotic to grasp her relation to the object-cause by way of fantasy, “an image set to work in the signifying structure”  [Lacan 2006, 532]; there, the imaginary aspect of the a, its correlatively real impossibility, and the chain of signifiers contoured by S1 are brought into contact.[15] This, Lacan represents with the matheme of the neurotic’s “fundamental fantasy,” $ ◊ a, in which the lozenge between the $ and the objet petit a designates a circuit of “envelopment-development-conjunction-disjunction”  [Lacan 2006, 542], such that imagined satisfaction and its actual impossibility are inscribed within the same scenario. Satisfaction would have arrived in a real and lasting way, the neurotic imagines, if things had been “just-so,” if everything had lined up “as it should have,” “if only” this or that had happened, etc. – if, that is, a could have been made to come forward by way of S1 and S2, which is categorically prohibited.[16]

In the graphs of the four discourses, these terms – S1, S2, $ and a – are slotted into four positions (Figure 4).[17]

The positions are:

• in the lower left: the truth of the discourse: the term that sets the discourse in motion;
• in the upper left: the agent of the discourse: the term that puts the discourse’s effects into action through speech;
• in the upper right: the other: the addressee toward whom the agent’s speech is directed;
• in the lower right: the production of the discourse: the term generated by or left over from the discourse.[18]

Between the upper terms, the arrow figures the agent’s interrogation of the other. This may be a demand for recognition or satisfaction; in the most general sense, it is a pure event of discourse: a spoken provocation aimed at making something happen. As in other Lacanian mathemes that include this typographic element [Lacan 2006, 428], the horizontal bars on each side of the graph figure a breach between the upper and lower registers, in this case a refusal or ignorance on the part of the term above bar with regard to the term below the bar. Thus, the agent takes itself to be the instigator of the interrogation, but this is only apparently the case; the actual instigator of the interrogation (as the expression of a social relation) is the truth on which discourse is based [Lacan 2007, 62]. Similarly, the other takes itself to be the definite target of the interrogation, but discourse actually aims at bringing the production forward into a social relation.

With these definitions out of the way, let’s allow the graphs to operate.[19]

The graph of the Discourse of the University (DU, Figure 5) is characteristic of research and pedagogy in their institutional settings, such as the university or the laboratory, in which practices of investigation, description, and indoctrination are dominant [Lacan 2007, 31]. Because the disciplinary formation of knowledge in the digital field is among my primary concerns in this essay, I will begin with this discourse.

In the DU, a field of knowledge (S2) is the discourse’s agent, propped upon a master signifier (S1), the choice of which guarantees S2s apparent consistency. Knowledge interrogates something outside of itself (S2 → a), the incitements of which it wishes to appropriate to itself and to bring into closer association with S1. Produced by this is the subject specific to the DU ($), divided along the fissure of its affiliation with institutional thought and its repression of other modes of relation to the object-cause (a). In this way, the DU expresses, by its single-minded devotion to investigation and truth-finding [Lacan 2007, 105], also a desire-to-not-know some aspects of the objects that trouble it, namely those which designate a relational framework of knowledge ($).[20] Interrogation under the DU may amount to little more than obstinate fiddling with terms of a defined lexicon (S2 /S1 ), in willful ignorance of the irreducibility of desire ($). In its more supple forms, the DU acknowledges the master’s caprice and the subject’s split, but goes on about its business anyway, in hope of eventually arriving at another (a more authentic) form of mastery [Lacan 2007, 63].[21]

The relevance of the DU to the present state of teaching and research in the digital field is marked in our labors to rigorously establish the field’s disciplinarity: to, in Lennon’s formulation, naturalize the writing out of critical desire as mimesis ( S2 ), in place of something more unsettled ($ – which, ironically, is precisely that which the DU produces). Taxonomic and descriptive schemes of the finest granularity are the epitome of the DU’s aim to bury its troublesome object under a heap of positives. Consensus building and canon formation have this quality of beginning from definite a priori (S1) by which to interrogate the eccentric term (a) so as to arrive at a scenario that seems to accrete out of the interrogation (S2 → a). Anthologies and archaeologies depend on such frameworks (Boluk 2009); crowdsourcing and folksonomies foreground the process of accretion and shuffle truth (S1) and product ($) a little behind. We haven’t yet our own brand of the DSM (Diagnostic and Statistical Manual of Mental Disorders – each revision of which represents the purest expression of the DU yet produced), but we have proliferating encyclopedias, anthologies, and textbooks. No accrediting professional organizations to determine best practices, but emerging cohorts of more voluntary, yet still disorderly, nature, which are sustained (S2 / S1) by master signifiers of exemplary resonance – Digital Hum… Electronic… Literature… O… – and on which are founded collegial but also doctrinally distinct debates within the field (e.g., [Gold 2012]), which from another vantage we might read as the productive circulation of master signifiers within the DU’s quadrature.

It’s not that consensus and canons, anthologies, archaeologies and folksonomies, and the like don’t have to be built, or that debates among practitioners about what is it, exactly, they do, don’t have to take place. These activities are essential to the development of a field, the consolidation of its research, and the training of its adherents, in short, to its formation as a discipline. In this regard, the DU operates with notable, even laudable, effect. It’s just that the DU isn’t efficacious only in the way that we might wish as we go about revising our research reports, forming working groups, lobbying our administrators for funding and hires, preparing our syllabi, and training our students. We recognize this on some level, and so we condition the products of our labors with forecasts of refinements and convictions that old problems will be resolved or set aside as no longer applicable under new conditions. Or, conversely, with convictions that the old problems may be better addressed under more expansive rubrics (hypertext studies and game studies are thus drawn into the penumbras of comparative media studies or digital humanities), and if we house their practitioners in already-established divisions of the university. (Matthew Kirschenbaum’s widely-cited 2010 essay “What is Digital Humanities and What’s It Doing in English Departments?” appears to have invited nearly as much normalization as the destabilization of its key terms [Kirschenbaum 2012].[22] In the present intellectual and educational funding climates in the United States, a strategic alignment of the digital field with the humanities is increasingly promoted as a method of reinvigorating the humanities, when it may be also a method of sequestering disruptive potentials of both the humanistic and the digital. One consequence of a Lacanian-discursive reading of the strategic alignment, via the DU, is that sequestration and disruption are unavoidable.)

Missing from most scenarios of refinement or containment is an acknowledgement of persistent, even irreducible, impasses of language which may be expressed in a particular way in the digital field, may require their own, idiomatic, master signifiers and their own, idiomatic, strategies of interrogation (S2/S1 → a). Similarly, periodizing – i.e. temporalizing the field in a progressive mode that relegates first waves and golden ages to a historical past, or contrasts modest beginnings with new-found largesses of administrators looking for the Next Thing – may function to contain the presentness of these impasses. Losing sight of the perdurability of our dissatisfaction means giving up also the continuing advantage of our awkwardness.

In the graph of the Discourse of the Master (DM, Figure 6) pure prestige and efficiency are paramount. Here, the master signifier (S1) directs the interrogation of knowledge (S2). Confident of the unassailability of his agency (S1 → S2), the master loses sight of the truth of his division ($) and the eccentricity of the object-cause that his efforts reveal (a).[23] That his mastery is in the final analysis an imposture is irrelevant to him; he doesn’t care to know otherwise because his aim is to get something done in an authoritative mode [Lacan 2007, 24]. In the master’s mind, the DM is unrelentingly pragmatic, but its pragmatism is touched with a millenarian or apocalyptic style, barely holding off a crisis of faith which could be catastrophic to it. That’s why there’s something of the obsessional in the DM’s way of thinking that is comparable to the DU’s more classic obsessional everything-and-the-kitchen-sinking of reason, except that the DM’s compulsion is more aggressive because there is more at stake in this undertaking for the master than for the bureaucrat.[24] Whereas the DU’s passion for detail is directed toward refinement and exhaustion, the DM is associated with predictions of radical transformation, bold ends and beginnings. When it is retrospective, the DM summons deep genealogies and traditions so as to prop up a brave or catastrophic future.

Expressions of the DM in the digital field are traced in signifiers of the theoretical-critical propositions which have had the greatest consequence, the field’s “greatest hits,” as it were. These include signifiers of the field’s primal scenes: Vannevar Bush’s “intimate supplement”  [Bush 1991], Douglas Engelbart’s “augmentation”  [Engelbart 1963], and Ted Nelson’s “twingle”  [Nelson 1990] – the resonances of which, I have argued elsewhere, are undiminished [Harpold 2008].[25] And, further along the chain: Jay David Bolter’s declaration of the new “writing space” of electronic textuality (e.g., [Bolter 1991], [Bolter 2001]); Robert Coover’s proclamation of the imminent “end of books” ([Coover 1992] – a millenarian thought experiment that many of us were pleased to entertain – hadn’t Victor Hugo given us our script 160 years earlier?); George Landow’s prediction of “convergence” between literary theory and digital practice [Landow 1997], [Landow 1992], [Landow 2006]; Espen Aarseth’s shifting of the discussion away from differences of electronic and printed texts to the more varied registers of cybertexts [Aarseth 1997]; Bolter and Grusin’s remediation [Bolter 1999], which came to stand for general feedback systems of mediality; Friedrich Kittler’s analyses of the “symbolic world” of code (e.g., [Kittler 1997], [Kittler 2008]); Gregory Ulmer’s celebrations of the new logic of electracy [Ulmer 2003]; N. Katherine Hayles’s descriptions of the intermediation of print, code, and bodies [Hayles 2005] [Hayles 2008]; Lev Manovich’s assertion that software has taken command of cultural production [Manovich 2001], [Manovich 2010]; and Ian Bogost’s advocacy of unit analysis as a technic of medial interpretation [Bogost 2006].

Measuring the effects of these master signifiers is difficult.[26] Some have lost their ability to hold our attention, which may be the most reliable indication of the master’s general authority. Others persist in attenuated or negated forms, and this also may be a sign of their effectiveness, by a kind of inversion. Among Lennon’s targets in his 2009 essay are versions of bright new beginnings that have taken hold by way of rejections of the first wave’s recursive negativities in favor of more empirical and measured (e.g., less “theoretical,” less “bellettristic,” which is to say less essayist) practices of writing out the digital field.[27] The turn away from the master’s delight in transformation toward the rewards of the DU’s finer granularities doesn’t surprise – that is the classic trajectory of disciplinary formation – but it’s a little too easy to mistake the turn for progress or a refinement of one’s assumptions and methods. The DM’s penchant for bold ruptures and the DU’s devotion to evidence are not incompatible; both have the alibi of S1 to guarantee that we will figure out what we’re about once we get down to real work. What undercuts that guarantee is that a fundamental maladaptation of the subject to the cause of this work: ($ ◊ a); clever obscurantism and pellucid empiricism function as two faces of a single coin.

But more generally, the mutuality of the DM and the DU is evidence of a principle that is clear in Lacan’s treatment of the discourses and which has to be kept in mind in any squaring off of the four: none operates in complete isolation from the others. If one can be said to dominate in a given situation, it is because the others travel under its sponsorship, and deliver their effects according to its structure [Lacan 2007, 43].[28] With respect to the DM’s determination of the DU (DM → DU), it is clear that its efficacy may have more than one dimension: the more inflexibly S1 prevails, the more it limits what is knowable in relation to S2, and the more difficult it will be to grasp the object-cause (a), as it will be more operationally estranged from the subject ($). This estrangement is more evident in the DM than in the other discourses because the fundamental fantasy ($ ◊ a) is expressed in its underside, beneath the interpellation of knowledge by the master signifier [Lacan 2007, 108].[29]

The graph of the Discourse of the Hysteric (DH, Figure 7) expresses a relation of dissatisfied and uncompleted inquiry. Here, the divided subject ($) is in the position of agent, supported by the truth of her object-cause (a).[30] Her interpellation of the master produces knowledge in the form of new signifiers (S1/S2), which take on the significance of symptoms of her relations to the master and the object-cause; fantasy ($ ◊ a) operates by way of by S1s determination of S2.

It is easy to see why Lacan associates this structure of discourse with the hysteric’s characteristic styles of speech. Freud’s pathbreaking insight at the birth of the psychoanalytic clinic was that hysterical suffering, whatever its dysfunctions, communicates a truth ($ / a) of the hysteric’s situation [Lacan 2007, 73]. It would be incorrect to say that the hysteric cannot take hold of the conditions of her suffering, as she is perfectly capable of demonstrating these to one who cares to listen [Lacan 2006, 34, 83].[31] Her knowledge is, however, distinctly productive in a way in that it cannot on its own be brought to closure or stable reflection; hers is, says Lacan, a knowledge that doesn’t know that it knows itself (un savoir qui ne se sait pas, [Lacan 2007, 33], trans. modified.), but which finds its expression in the presence of an other, the analyst who knows how to listen.

A substantial body of work in the digital field is characterized by one of two styles of the DH. The first style resembles the subject Verhaeghe (ironically) terms the “good” hysteric [Verhaeghe 1995]. She is devoted to the master, and acts as if expressions of her faith in him (S1/S2) are something in which he is deeply invested. She obscures her frustrations and measures her provisional satisfactions in the conviction that these labors will one day bring about the master’s enjoyment, and vicariously her own. The paradox lies in that the hysteric wants the master’s dissatisfaction – she seeks it, she lacks for it – in order that she may (also) desire [Lacan 2006, 518]; [Lacan 2007, 129].[32] In the digital field, such devotion to the master’s enjoyment is expressed in visions of seamless immersion and the erasure of divisions between language and the real – which divisions, if we take Lacan seriously, must be understood to be uneraseable because they determine the subject qua speaking subject ($ / a; [Lacan 2006, 712]). Early 1990s fantasies in this vein, of totalizing virtualities and “post-symbolic” communication now have been set aside, even by ardent proponents (e.g., [Lanier 2011]), but related fantasies of transcendence are still vigorous in some strands of posthumanist thought. In actuality, they are only more exacting forms of subjection, the hysteric’s version of the master’s millennial breakout. In pop-cultural accounts of the technological singularity in particular (e.g., [Kurzweil 2005]), the master’s imprint on desire is evident in the conviction that human self-interest, or the interest of an Other that recognizes humanity as its friend or enemy, must span the divide of the transformative event.

In contrast, the second style of the DH resembles Verhaeghe’s “bad” hysteric [Verhaeghe 1995]. She is uncooperative and openly frustrated with the master’s responses to her elaborate demonstrations. In this uncooperative mode, her style more nearly grasps her discourse’s dependence on S1, in that she presses the master to show his hand ($ → S1) by way of innovating signifiers (S2) expressing her dissatisfaction. This style of the DH more openly acknowledges the efficacy of its structure, as it puts the master’s caprice to productive work, not as the DU would (conjecturing that eventually the effects of structure may be transcended), but by insisting on the irresolution of the hysteric’s desire.

In the digital field, examples of this style of the DH include in particular those interrogations that openly confront the subject’s inmixing with her objects of analysis. Joyce’s criticism (1995, 2000, 2001), written in a voice I would characterize as unambiguously hysterical, is the type of this form of writing out that acts out the uncompletability of the critic’s desire – in Lennon’s terminology, writing out desire in the essayist mode, in contrast to the DU’s effort to situate critic and object on distinct registers, in the novelist mode. Other landmark examples include: Shelley Jackson’s dexterous stitchings of her fictional and critical personae (e.g., 1995, 1998); John Cayley’s (e.g., 2004, 2010), Loss Pequeño Glazier’s (e.g., 2002, 2006) and Talan Memmot’s (e.g., 2001) contributions to a digital poetics in which coding and coded voices are robustly inmixed; Stuart Moulthrop’s excursions in the garden and the library (e.g., 1991, 1999), which were more concerned with the relationality of hypertext writing and reading than their efficiencies; Mark Hansen’s intimate reflections on new media affectivity (2004, 2006); Hayles’s method of “autocritography” (2002); and Ulmer’s invention methodologies of the “mystory” and “MEmorial” (1989, 1994, 2005). This is the style of interrogation I aimed to activate in Ex-foliations (2008) by first, insisting that critics should hew closely to Roland Barthes’s original definition of the lexia in order to seize the term’s generative potential in hypertext criticism, as a name (S1) for field effects (S2) of reading [Harpold 2008, 1.44]; and second, by associating these effects with what I termed the reader’s historiation, her conscious and unconscious recollection of her reading history. Or, to write this out in the style of the bad hysteric: a retrospective gathering signifiers in a dissatisfied present ($ / a → S1) – leading to further, unresolved dissatisfactions and further gatherings – is a way forward to new forms of knowledge, when this activity is directed at the one who cares to listen [Harpold 2008, 1.55–1.56]. Such a generative writing out in the style of the DH need not dominate all aspects of the critical encounter. It may be expressed as a punctual interruption of a writing out in another discourse or style. The pattern of discourse is invariable (Figure 4), but the positions of the terms in play will be determined by specific provocations of the social bond. The key is to remain open to these provocations and their unsettling effects, even as disciplinary formations tend to dampen those effects.[33]

The graph of the Discourse of the Analyst (DA, Figure 8) figures the social bond in the scene of the psychoanalytic clinic. Here, the analyst takes up the agency of the object-cause (a) [Lacan 2007, 42], supported by the truth of the signifiers (S2) that the analysand presents to him. The analyst interrogates the analysand ($), prompting her to produce the master signifiers (S1) that orient her speech and her suffering [Lacan 2007, 172–73].

In the strictest sense, the analyst cannot occupy the place of the objet petit a, as it is eccentric to the field of speech by which the analyst and analysand are united. But he may be supposed to occupy its place for the analysand, and this supposition is determinate of her coming to recognize, beyond the analyst, the master toward whom her speech is directed.[34] (Here, the analyst plays for the analysand the role of the “subject-supposed-to-know” or “subject-supposed-to-knowledge” [sujet supposé savoir], which ruse is essential to establishing the transference [Lacan 1998a, 233]; [Lacan 2006, 254].) But the analyst must not confuse supposed knowledge with actual knowledge, as that would mean falling prey to effects of the counter-transference. The knowledge that founds their social relation is not in him but pertains only to the signifiers they exchange [Lacan 2001, 249], and he must remain skeptical of the authority invested in him, and even willing to surrender it when this is called for.[35] In contrast to the hysteric’s classic partner, the obsessional, who matches her misdirections by stretching out the misrecognition of his own desire, the analyst presents the hysteric with an encounter wherein she can grasp the degree to which her knowledge (S2) is oriented by S1. Of priority here is the social relation of this encounter, founded on language, which is determinate of its effects. What the analysand says in this scene is, in an important sense, immaterial – her speech can be so much blah blah blah and the analyst will encourage her to keep at it [Lacan 2006, 275]; [Lacan 2007, 52].[36] What matters is the pinning of S2 to S1 in relation to the object-cause [Lacan 2007, 33].[37]

If we accept that the DH, and more especially, discourse in the style of the “bad” hysteric, points forward to new forms of knowledge (S2) in our encounter with media, then it would appear that the most generative variants of this encounter will take place under the hysteric’s mandate to keep desire productive, specifically within a scene of interrogation in which the object-cause is understood to urge on her dissatisfaction (a → $), and her relations to master signifiers ($ → S1) are recognized to found her speech. In other words, in the social relation of the analyst and the hysteric (DA → DH).

In a nutshell, that is the diagnosis and the prescription that I propose here. But such a formula has important consequences for the future of our field, if that is to be as productive as it has been during the last two decades, and if a shift to a more unified disciplinarity and a more secure institutional footing is not to come at the cost of a former uncompletability that made very interesting things come to pass. This is nowhere more true than in the social bond of our teaching. The graphs of the four discourses show us that the more productive pedagogy is not in the style of the master, which occludes the subject’s division and spits out as its remainder the object-cause of desire (S1 → S2 / a). Nor is it in the style of the University, which reifies the subject’s division by promising eventual mastery over desire (S2 /S1 → a). Much like the neurotic’s just-so stories, that is just a form of temporizing. The more productive pedagogy is one that sustains desire by way of the hysterization of the subject in response to something enigmatic, the operation of which is signaled in the field of speech qua social bond [Lacan 2007, 33]. There, the place of the agent is not occupied by the digital object, its technical frameworks, or a conjectural entity supposed to “interact” with these things (a user) – but, more generally, by those aspects of the interrogation which take on the role of an enigmatic, generative element (a), to which we and our students respond, a little unnerved and uncertain. In the model of the DH we can grasp that the eccentricity of the objet petit a is foundational to the subject’s interrogation of the conditions of her knowledge. In the model of the DA, by way of its engagement with the DH, we can grasp a positive effect of this always-outsideness of the a: it can serve as the agent of the production of new and destabilized forms of knowledge – the DA as a disappearing act that leaves (behind) its mark [Lacan 2007, 23].[38]

Such an engagement of the DH and DA would be consistent with Joyce’s observation that our awkwardness may (still) be to our advantage. Dissatisfaction, befuddlement, and persistent misprision (the hysteric’s modes of relation) and a hesitation to be, or appear to be, confident of one’s reading of things (the analyst’s reflective response) need not be taken as evidence of lapsed knowledge; they may be acknowledgements of how the scene of knowledge is structured and carried through to its aims. If we profess from the position of supposed knowledge that our charges imagine for us, this will be most effective if we don’t take the supposition seriously, or not for long. We may foster supposition in situations in which this tends toward the building of foundations and the applications of lexicons (DM → DU): for example, instructing our students in the combinatorial play of the field’s master signifiers, which is a prerequisite for disciplinary competence. Such a tactic is in the short run disingenuous. But in the long run it may be productive of new signifiers if we, after a time, also renounce some of the authority invested in us, so as to desuppose the knowledge imagined to to be attached to that authority (DA → DH). We might call into question – even peremptorily, as is our due – master signifiers and the scaffolding built from them; perhaps even the notion of disciplinary competence itself. Shifting the scene of discourse from supposition to desupposition does not preclude speaking with conviction or making profitable use of the tools at our disposal (“consensus and canons, anthologies, archaeologies and folksonomies, and the like”). But desupposing, not supposing, is the impulse of a more innovative knowledge production [Lacan 1998b, 67]. It entails attending to, even signaling to our charges, our and their irreducibly relational engagements with the elements of the social bond. A lesson of the four discourses is that such an attending and a signaling must be initiated by more radical breaks with the master than readjusting our (or their) a prioris. And that these breaks have to be perpetually renewed so that they don’t lapse into merely new variants of supposed knowledge.[39]

We are committed, by institutional and professional conditions, to practices that tend toward the imperiousness of the DM and the recursive fiddling of the DU. But another of the lessons of the four discourses is that any one, two, or three discourses must operate in some relation to the others. Within a general economy of knowledge transmission, the primary stated goals of which are to “advance” the discipline and maximize the understanding of the scholar and the student, we could take such a lesson to heart and press the misdirections of imperiousness and fiddling to other, more relational ends. This means, along with the analyst, siding with the hysteric instead of the master, with the deliberate aim of getting her to grasp that what she wants (lacks) is the expression of the master’s desire (that which he lacks, but cannot have). And of making something come to pass thereby: new turns of the circuit of discourses and new master signifiers, in relation to which the hysteric and her master will be newly subjectified. (Doesn’t my description of the exemplary hysteria of Joyce’s early criticism, and Lennon’s accounting of Joyce’s withdrawal into spectrality as an act of provocation, imply that Joyce has, by his own admission [Joyce 1995]; [Joyce 2001a], played exactly this game of double-mindedness?)[40]

Playing at the tangle of those relations is neither a heroic nor a cynical response to the laws of desire; the hysteric and the analyst don’t aim to nullify the laws, only to recalibrate their effects, shaking loose the master’s confidence that he knows what he is talking about, and the bureaucrat’s determination that he will do even better – to, in other words, make something new come out of the laws of desire that confidence and determination might hide away ($ → S1 / S2 : a → $ / S1). In a mode of rigorous relationality in which the irreducibility of the laws of desire is acknowledged, one can look ahead, along a path from the impostures of the DM and DU toward (and beyond) the impasses of the DH and DA. Such a change in our point of view would be endlessly unsettled and unsettling, and for that reason, of authentic consequence for our discipline.

The graphs of the four discourses strip the social bond founded on language to its barest formalization. This enables us to grasp discourse as merely, but consequentially, an operation in which something new is made to happen within signification. That for Lacan this innovation takes place determinately, if not exclusively, in the concrete register of speech becomes clear if we associate the graphs of the four discourses with another famous Lacanian graph.

The elements on the left (“theme”) and right (“phoros”) of the (•) symbol figure two composite terms that are related in the substitution. The upper case letters (S, S´1, S´2) represent signifiers involved in this operation, with S´1 and S´2 figuring the signifiers that fall out of the signifying chain, and S the (general instance of the) signifier that is carried over into the product. The lower case x is the “unknown signification”  [Lacan 2006, 494] that completes the metaphor. I/s´´ figures the new signified that is induced by this operation (or something like a new signified; I will return to this in a moment).[42]

The full import of this diagram is evident only if we understand that Lacan conceives of the operation of metaphor in strictly formal terms; the relations between the signifiers involved (S – S´1 – S´2) are differential and unmotivated with regard to their reference. Metaphor is merely, sufficiently “one word for another”  [Lacan 2006, 422]; an implied analogy between theme and phoros, or the new meaning intended by their association in this way – what, it seems safe to say, is usually meant when we speak of “metaphor” – is not of especial relevance to the efficacy of metaphor’s structure. Indeed, one might characterize as obdurate Lacan’s insistence that the substitution is in other respects without significance for the subject.[43]

The reason for his unwavering formalism is this coda for the subject – that is, the subject qua speaking subject, for whom the machinations of the signifying chain are of importance in and of themselves. (This subject is, we may remember, represented in the chain by a signifier for another signifier.) As shown in the expanded formula, metaphor is the substitution of a signifier (S) for one or more other signifiers (S´1, S´2), such that their presence in the chain is elided and they are functionally driven below the bar of the signifier/signified doublet (S / s; [Lacan 2006, 594]). This produces a “new species of signification”  [Lacan 2006, 757] that preserves a phantom imprint of S´1 and S´2 (perhaps this is figured in the double priming of s, which carries some residue of the elided signifiers?), and which is something new in the field of speech that did not exist before: a surplus (→ SI/s´´) derived from the mere associations of signifiers on the left side of the formula. Lacan refers to the element below the bar on the right side as a “signified,” but we may interpret it as something more evocative and uncompleted that accounts for metaphor’s role in Lacan’s thought during this period. First, I/s´´ is not, as Ed Pluth has observed, a “signified” in the strictly Saussurean sense – a concept associated with the acoustic image of the signifier – but something more like a “signified effect” [effet de signifié] or “resonance” of signification generated by interactions of signifiers, which functions as if it were a signified [Pluth 2007]. [44] This is how Lacan understands the production of meaning in the signifying chain, and primordially in the “paternal metaphor” that founds Oedipus.[45] Second, because Lacan correlates this driving of a signifier below the bar with secondary repression (Verdrängung, refoulement: [Freud 1915]), metaphor’s production of a “new species of signification” is homologous with repression’s production of new signified effects from the exclusion of signifiers from consciousness [Lacan 1998a, 218]. At issue in both metaphor and repression as Lacan understands these operations is the brute productivity of signifying substitution, which is realized without regard to the positive content of the elements involved.[46]

Grasping metaphor thus as a purely formal operation of substitution and production, concretely realized in a specific scene of speech, we can see metaphor’s relevance to Lacanian discourse theory in two respects.

First, metaphor constitutes a primary axis of the productivity of signification within the social bond of language. There, relations between signifiers are determinate; the sociality of the bond obtains wholly in signifying operations. In the expanded formula of metaphor, the production of new signification emerges from the impertinence of the substitution, insofar as it leaves something behind.[47] In the graphs of the four discourses, this impertinent product is returned to the circuit of the graph as a term of their subsequent turns: the truth, the agent, the other. This is the trace within the social bond of discourse’s perpetually-renewing innovation.

Second, the graphs of the four discourses are not a metaphor of the social bond in the usual sense of the term “metaphor.” For Lacan, the graphs express structures of the bond directly and concretely; they perform social relations of language. (They write out the social bond.) Putting the graphs to work – putting them through the turns I have described in this essay – demonstrates the instability and uncompletability of the bond, expressly in that putting them to work will generate new signifying effects. (Which will lead to further turns of the graphs, and so on.)

Correlatively, mapping specific scenes of discourse to the graphs, if performed scrupulously, is not to make a metaphor of the graphs (i.e., a metaphor for the scenes of discourse), but to take them in hand (the scenes and the graphs) and to make something new come out of them that was not in evidence before.[48] To write this more precisely: the signifying substitution of each of the graphs for specific turnings of practice in and of itself (S / S´1 • S´2 / x) should make something new (→ S (I/s´´)). The character of the product of each substitution ($, a, S1 or S2) alone determines the efficacy of such a tactic.

A further lesson of the discourses: to aim, merely but consequentially, at producing a response from the other is both the narrowest and most rigorous (DA: a → $ / S1), and the most extravagant and indecent (DH: $ → S1 / S2 ), of methodological ambitions.

A first version of this essay was presented at “Futures of Digital Studies,” the 2010 Conference of the Digital Assembly, University of Florida, February 25–27, 2010. I am grateful to the conference organizers, Mauro Carassai and Elisabet Takehana, and the editors of this journal, for the opportunity to revise and expand on those remarks. Jeanne Ewert, Julia Flanders, Loss Pequeño Glazier, Brian Lennon, and Gregory Ulmer offered invaluable advice as the text threaded to its present form.