This was written partly as a response to an article I recently read in which alt-right trolls were compared with Kierkegaardian ironists. To the degree that trolling deserves any theoretisation at all, the argument seems initially seductive, but I have very serious reservations about it.

With the prospect of generalized irony as a feature of all language, a la Cleanth Brooks (who, in a definition that constituted a famous landmark in terms of its sheer scope, claimed that irony was simply the way in which a text registers the pressures of context), we have to face the extinction of a certain opposition of irony to serious language. That is to say, we must stop insisting that there is a clearly non-ironic use of language in which the speaker coincides entirely with the subject of the statement, standing completely within or behind it without residue, and entirely committed to the social meaning of the utterance. Rather, the gap between subject and subject of the statement must be seen to be operative in all language. Although, to be sure, this gap is not always ‘operative’ to the same extent, it nonetheless remains a fundamental non-relation between the subject and the subject. The paradigm of this gap’s operation would today be the internet chatroom or below-the-line readers’ comments sections, in which no-one can be sure that a participant is committed in any way to their comments, that they are not simply ‘trolling’ (or indeed trolling in a non-simple, multilayered or ambiguous, complex way). In such an environment, one sees the most ridiculously anachronistic and wilfully transgressive of political views expressed as if earnestly, and one sees the attempt to defuse or counter-argue as either ‘taking the bait’, ‘feeding the troll’, or ‘white-knighting’. The ‘better’ player emerges in these comment streams as the one who can up the ante, amplify the trolling of others, and so on. The internet is couched as a space of freedom: principally the freedom of the anonymous to transgress, to voice or enact puerile fantasies detached from the meaty encumbrance of socialised identities with their materially-consequential responsibilities, substantial duties, bonds and relations; freedom from personal consequences, freedom from having to endure the returning effects of what one circulates. It might appear as if posting on the internet was a realisation of precisely that substancelessness, that ‘infinite negativity’, which Kierkegaard (after Hegel) criticised in the Schlegelian theory of romantic irony. Hegel and Kierkegaard both criticised romantic irony as being potentially limitless. After a certain point the ironic conscious is entirely free to dispose of all substantial (that is, normative, social, intersubjective, moral: ‘political’ in its modern sense) concerns, even those which might appear to further its own subsistence. This is where the line between ‘constructive’ (i.e. Socratic) and ‘destructive’ (Schlegelian) irony (as understood by Kierkegaard and Hegel) is often drawn.

While the possibility of anacoluthon (the abrupt interposition of a changed direction, e.g. of argumentation, position, etc.) and parabasis (the sudden shift to a metaregister) mark ironic speech, in Schlegelian destruction they leave nothing but the empty position of irony itself, as infinite negativity — according to Hegel. One drops bombs and dances onwards in the same floating dissociation from whatever is said, because whatever is said is merely noise or pixels, and can’t really hurt anyone, or so the dangerous, evil ironist thinks. But while this criticism certainly functions to describe one of the ways in which today’s trolls excuse their excesses, Hegel (and subsequently Kierkegaard) may have misunderstand Schlegel. We’ll come to that later; first let’s examine why Hegel and Kierkegaard are right about infinite negativity, but are nonetheless quibbling since it represents a position that is impossible to assume.

Such a position of infinite negativity should not be possible according to Wittgenstein: one can not enunciate a statement without context, use a totally ‘private language’ uncommitted to social meaning, etc. Language ‘is’, in a certain sense, nothing but the rules that constitute it: dependent for its status as meaningful on a system of social rules. Or, one can say with the conviction of the Saussurean school of linguistics, language is predicated on a differential system of meanings: meaning arises from differences, comparatives, relatives, and specific arrangements, not from positive terms.

From this perspective the idea of producing entirely uncommitted speech behind which stands no substantial interest is bogus: there is a process of selection — enchainment, arrangement, both syntagmatic and paradigmatic — to achieve a desired effect in terms of reception: meaning. Thus to speak is to incur responsibility for the articulation , the choosing and stringing together, that produces meaning, even if it is not ‘meant’, is ambiguous, open to interpretation, or was merely ‘banter’. The speaker, tweeter, memer or troll are not simply the producer of locutions with illocutionary sense overlaid by a readership; on the contrary these groups encode perlocutionary force in what they produce, indeed depend upon it as the primary motivating factor. Just as WWE wrestling baits the audience, producing ‘heat’ by various mechanisms such as repeated ‘heel’ beatdowns on ‘faces’, in a predictable and controllable formula that does not fail to elicit the mass emotional response sought after, so is trolling a purposeful elicitation of contextual response. The troll is akin to a wrestling script-writer, a provocateur, except in one point: here the kayfabe is not an open secret to be ‘suspended’ (i.e. as ‘disbelief’) for the purpose of entertainment, but functions as a screen or excuse to be insisted upon, in order to pretend that the action isn’t socially real, and doesn’t generate real effects. Besides ritual beatdowns to characterise various wrestling actors as heels and faces (or to perform ‘turns’ from one to other), WWE sports entertainment utilises a panoply of stock characters, redolent of pantomime, as managers, coaches and trainers. In 2013, for example, the ‘Real American’ wrestler Jack Swagger was mentored by a character calling himself Zeb Colter (portrayed by Dutch Mantel), a racist and anti-immigration, Mexican-hating, welfare-bashing MAGA demagogue with a long luxurious Yosemite-Sam moustache and a Southern secessionist veneer (Swagger’s logo clearly echoing a confederate flag). At the time it was obvious the Tea Party was being parodied. The audience simultaneously recognised Colter as a simulacrum and also consistently refused his performative efficiency, through booing, heckling and jeering, all despite a not-too-dissimilar version of this persona being voted into the office of POTUS just three years later. To an uninformed viewer, it might appear as if social attitudes had somehow swung from a liberal majority to an alt-right majority in those three years, or that the WWE audience represented an unduly liberal crowd compared to outsiders. But on the contrary, the WWE network is tight with the republican party; the chairman is married to a republican senator, and the show had actually featured Donald Trump in its ‘Hall of Fame’ segment, immortalising a face off between the latter and Vince McMahon. In other words, kayfabe itself can function in a more complex way than a pre-critical understanding might allow: as the pretence of a pretence. Trump would, more than any romantic ironist, better characterise the modern troll, who hides behind the screen of an assumed political kayfabe – a redoubled pretence of pretence – since he just so happens to really mean the things he ‘jokes’ about. This strategy of also meaning it was identified, with great wit, by the comedian Stewart Lee in a performance criticising the show Top Gear (more precisely its presenters). Not coincidentally, when Jeremy Clarkson first began presenting a new version of Top Gear hosted not by the BBC but online, he boasted ‘no-one can fire me now… we’re on the internet, which means I could pleasure a horse’.

In other words, the alt-right (and the wider manosphere) is worlds apart from romantic irony. While it claims for itself the social disconnection, the infinite negativity, attributed to romantic irony by Kierkegaard and Hegel, alt-right sites such as Breitbart profit socially through the clicks and shares of their readerships. If the alt-right’s self-appointed ‘intellectuals’ would like to believe they merely play with signifiers as punks played with swastikas, they are clearly mistaken; perhaps the illusions of omnivalence and omnivocity granted by the adoption of internet anonymity are all too captivating. However, they do not escape the obvious: they are white men, politically economic-libertarian, and their substantial interests are clearly visible. Whatever theoretisation of irony the alt-right may try to deploy, it is extremely deficient.

A careful reading of romantic irony can act as a corrective to Kierkegaard’s and Hegel’s warning-off. Far from floating free of social investments, romantic irony always implicates everyone: narrator, author, reader. The Early German Romantic theory of irony allows for no Archimedean point, no privileged or anonymous position from which all may appear as a joke, and from which laughter may erupt in a ‘cult of kek’. Such an attitude indeed leaves one feeling that the alt-right must consider the interactive space of the internet to exist over and above social existence, rather than as an integrated part of the social world. Claire Colebrook clarifies:

Until Romanticism, the literary or rhetorical function of irony was seen as a special case within an otherwise simple and literal language of representation. Irony was deemed to be an ornament or trope within representational language. For the Romantics, however, it was only possible to have a seemingly simple and representational world through the forgetting and repression of the creativity and poetry of language. Irony – or the gap between words and world – was, for the Romantics, original. Speech and language originate or come into being only when ideas or concepts give form and imagination to the actual world; all language is essentially and originally figural, or different from the world it supposedly names. Literal language is the denial or forgetting of this gap. If we think of our language as a simple one-to-one label or picture of the world, then we forget the creative and disruptive birth of language. To see all language as ironic, the Romantics argued, would be to restore life to its once open, fluid and productive past. Life would no longer be frozen into the fixed forms of grammar and syntax, or reduced to what is sayable. Irony recognises a sense that is always other than what is said.

Once Romanticism established that the truth of life did not lie in adequate representation but in a questioning and imaginative play of representations – such as poetry – then it became possible to see literature as the privileged mode of human understanding. Literature would be the truth of life because literature was essentially ironic: adopting a permanently distanced and questioning attitude to all language and fixed positions.


Further, for the Romantics, this fall is one of ‘buffoonery.’ German Romantic irony was defined through a constellation of concepts, including, in addition to buffoonery, humour, wit and satire. The joke or Witz undoes the mastery of the subject, as laughter and nonsense disrupt logic and sense. Irony is related to buffoonery not just because subjective mastery is undermined; buffoonery falls, enjoys the humour of the fall, laughs from on high at the falling buffoon, and remains implicated in the fall. One can never master the ironic process, never recognise or stand above one’s finitude: ‘Irony is the clear consciousness of eternal agility, of an infinitely teeming chaos’ (Schlegel). The minute we see ourselves as other than what has fallen, as beings who can overlook and describe the fall, we fall further into smug self-recognition: ‘One can only become a philosopher, not be one. As soon as one thinks one is a philosopher, one stops becoming one’ (Schlegel). Irony must recognise that we can never overcome singular viewpoints and achieve a God-like point of view; we are always subject to a cosmic joke. For any idea we have of our selves or our world will be part of a process of creation and destruction that we can neither delimit nor control. If humour often relies on a feeling of superiority or elevation above life’s misfortunes, irony recognises – but never fully realises – the implication of all life in this chaos. The ironic attitude must not just take a delight in seeing the clown slip on a banana skin; it must not just laugh at this fall from human coordination into an animal or thing-like buffoonery. It must recognise that we are all part of this falling; we are always dupes and effects of a life with a power well beyond our control and recognition.


This brings us to the heart of irony and dialogue. To acknowledge poeisis is to acknowledge that the creativity of life can never be encompassed or reflected in an overarching point of view. Conscious activity is never at one with the forms it creates.

At the same time, Colebrook advises us that if language use is as primordially rhetorical (and thus catachrestic), as Paul de Mann claims, then it is also and by the same token unavoidably committed to some meaning. For Lacan this is also true; the demands of the gurgling infant are always already semanticised, always addressed to an Other, always already mediate and never immediate. Demand cannot be isolated from desire in practice. The encoding of perlocutionary force, the rhetorical gamble of language, is always there at the very beginning of language acquisition, not something super-added to it and from which one can then opt-out from as anonymous trolls seem to believe. All of this seems obvious to philosophers like Colebrook who, as part of their profession, practice thinking and address themselves to their thoughts, in what is a deeply social process. Reasoning is often said to rely on the weighing of arguments, searching for the most persuasive. Thought is, in this sense, never disinterested. Nor is the troll.

In Hegelian parlance, ‘substantial’ interests and the term ‘substance’ itself refers to the intersubjective (or ‘ethical’) domain of socially-shared morality, normativity, etc. It is the forerunner of the later idea of the ‘socially objective’ domain, as reified and misused by Stalinism. Today we tend to understand the socially objective domain simply as the domain of all our various forms and relations of social co-operation — it is the largely agreed-upon, although always culturally mediated, world of standards of taste and decency. It is precisely this realm, the realm of the rule (be it formal or informal) that the troll time and time transgresses or attempts to get others to transgress. In order to transgress it, however, one must first acknowledge its existence. We can discuss various forms of relation to the Big Other in psychoanalytic terms. We can approach it in terms of discourse analysis. We can discuss socialisation as a process. We can approach the hypocrisy (conscious or not) of the alt-right troll in many ways. We do not lack the tools to deal with the advent of the internet.

However, I believe that a direct comparison between the concept of romantisch ironie in Frühromantik and the absolutely modern phenomena of anonymous trolling on alt-right websites and sections of reddit is misled. Many theorists in the past have contended that romantic irony and post-modernism, particularly Derridean deconstruction, are related if not genetically connected. This argument always revolves on a secondary interpretation of deconstruction which inflates the ‘play of the signifier’ to the degree that the theoretical underpinnings of deconstruction, and its insistence on textual commitment are lost: Derrida expressly refuted the possibility that an uncommitted speech act could take place, and this is systematically forgotten whenever the ‘free-play of signifiers’ argument is invoked as a defence of offensive speech or language. A similar elision occurs when romantic irony is compared with trolling. Schlegelian irony is defined as an act of ‘self-creation and self-destruction’ not in a phenomenological or psychological register, but in a social one: the theory of irony is advanced in a particular historical juncture in which the productive capacity of art, and particularly poetry, freed from patronage systems and available to reason (contemporaneously leading the French Revolution) and its self-reflective possibilities, are becoming far more visible.

‘The French Revolution, Fichte’s philosophy of science, and Goethe’s Meister are the greatest tendencies [Tendenzen] of the age. No revolution that is not loud and material will appear important to the person who objects to this compilation. That kind of person had not yet climbed to the high, broad point of view of human history’

– Friedrich Schlegel, Athenaeumfragment 216

The situation of the Early German Romantics such as Schlegel is is a unique historical conjuncture, in which enlightenment reason, in the wake of Kant, and the arts (particularly the literary arts) encounter and affect each other. The comparison with the age of digital media and the sudden ability to self-author through providing ‘content’ for social networks is tempting, but ultimately misled. Today’s digital aestheticism of the self is a world away from the intent of the Früromantik movement to make art announce itself as art. The misunderstanding can, in fact, be traced to how Fichte is read.

From reading de Man’s summarisation of Fichte’s dialectic of the self in the Wissenshaftslehre (‘On the Concept of Irony’, Aesthetic Ideology p.172) it becomes apparent that the frequent mischaracterisation of Fichte’s system as a form of Subjective Idealism a la Berkeley, or as solipsism, stems from attempts to read Fichte’s ‘self’ from a phenomenological, or even a vulgar psychological vantage. For de Man, the Fichtean self is purely linguistic, and its act of self-positing occurs not in experience or in the inner mental life of any individual but purely within language. The enunciation or marking of the I, the act by which language marks and posits itself as self-generating, is a linguistic act and not a subjective one. Fichte’s non-I which is generated out of the linguistic I is the linguistic I’s reference to the outside of the language it commands – a reference which can be seen, true to Fichte’s system, to occur only within language itself; thus the non-I is generated from the I. It is only on the basis of this primordial linguistic act by which language refers to its own taking place (see the significance of grammatical shifters and deixis in Benveniste and in Agamben) that something like subjectivity in the usual, phenomenological sense, can follow. Here then we have something like a Lacanianism avant la lettre: the signifier precedes; the subject is constituted, or receives its being, from the Other (that is, from language). We must consider then the Fichtean ‘self’ as a presubjective or nonsubjective I, a purely ‘formal’ self-inscription or self-registration of language, a virtual point which through its singular self-indicative status, enables reference (and thus both subjectivisation and worldliness). The (Hegelian) criticism that the Fichtean ‘I’ is desubstantialised would thereby miss the point; it would be substantive – its materiality would be that of a language pliable and sensitive enough to register its own capacities. Nor would the (de Mannian) Fichtean ‘I’ pose for us some figure of inmost inwardness, or point of absolute subjectivism; indeed it would be radically desubjective and impersonal, but thoroughly social. Its detachment from any individual phenomenological field would be that of the narrative voice of an author writing a novel in the first person: the ‘I’ of the narrative need bear no relation to the psychological self of the author, it merely enables the unary trait of a linguistic voice through which some kind of minimal narrative consistency might be established, a feature that is a wholly separate matter to that of how ‘unreliable’ the narrator and the narrative may prove to be. The relation of Fichte’s self to the ‘self’ envisaged in folk psychology would be strictly arbitrary. What we need to question, like Derrida, is how and when signification – the deferral of a signifier to another in some kind of enchainment whereby metaphysical presence is never finally achievable – can be said to occur. In other words, it is far from being the positivity of an extradiscursive realm which is in question, and much more akin to being the metaphysical basis of the positivisation process itself which is under fire here. What does positivity even mean? The posited, the positioned, and so on. Positivity, as Foucault surely saw in using the term, can only indicate the operation of language, just as the status of ‘facts’ can only indicate the operation of some factum, some fundamental act of making. Nothing permits us to summarise these Fichtean realisations (as Schopenhauer’s vulgarisation did) as a form of subjective idealism in which a spider-like ego weaves the world purely out of its own mentation. At stake is not the physicality of the world nor the materiality of matter: indeed, far from this, at stake is a bewildering understanding that language itself exists materially as part of the world and not as a neutral medium that simply overlays and describes a world that pre-exists it.

The point of this discursus was simply this: a recognition of the social materiality of language. To speak, write, declaim in any way, is to enter a social space. One of Derrida’s influential mentors, the literary theorist Maurice Blanchot, mused that he was never less alone than when writing. One is directly exposed to the the material sociality of language, one is indeed always already ‘buried’ within the social. Part of the structural affinity of the Früromantik movement’s loose literary theory with these more recent observations lies in the ability of language to reflect and evince its own material status, which is to say, its historicality.

At this point it is necessary to state my own commitments and interests. I am currently studying Art History at postgraduate level and intend to focus my dissertation on forms of irony in visual culture. Given that irony has been defined right from the beginning as a phenomenon rooted in orality and literacy rather than in plastic and visual cultures, my problem will be to contextualise irony in artwork from an art historical rather than a literary theoretical perspective. Thankfully some of this work has already been done, although there is much to do.

Until Schlegel’s famous expansion of the term, irony had been regarded in only two modes: the rhetorical-oratorial and the philosophical. Schlegel’s redefinition and expansion of irony, leading to literary scholarship’s embrace of the concept of ‘romantic irony’, made it possible for Strohshneider-Kohr’s assertion that ‘ironie ist Mittel der Selbstrepräsentation von Kunst’ (“[Romantic] Irony is the means whereby art represents itself”, Strohschneider-Kohrs, I., Die Romantisch Ironie in Theorie und Gestalltung, Tübingen: Niemeyer 1977 [1960], p.70) It is precisely this capacity of art to subjectify and identify itself qua art through the social mechanisms of romantic irony which, hopefully, will form the subject of my dissertation.

With this definition of romantisch ironie in mind, it seems to be wholly incompatible to compare the internet-based culture of the political alt-right anonymous trolling with anything happening in Jena in the 1790s-1800. On the one hand we find a developing theory of art’s social reflexivity and an enlarged capacity for the expressive and experimental capacity of art (more specifically life as a productive, open poesis centred not on a transcendental self but on a larger, historical view of language); on the other hand we find a network of white supremacists posing as deconstruction-literate ‘free-players’ who don’t ‘get’ Derrida at all, understand irony even less, and who all-too-conveniently happen to be politically aligned with some of the ridiculous and extreme transgressions they post online. All this while desperately fearing their anonymity will be broken and they will be outed to their employers and family. Banter, in other words, is deeply committed, deeply politically egregious, and a long way from being in any way self-reflective.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.