Our work is made possible through the support of the following organisations:

SRB logoSRB logoSRB logoSRB logo

On Equivalence

Sean Pryor on poetry, code, and Mez Breeze

Email to a friendEmail

Ten years ago, Mez Breeze published ‘\oriGaM[e,]i[,+ U]/’ online. Sean Pryor revisits this work in the context of Breeze’s experiments mixing code and poetry. Can literary-critical tools help close the gap between natural and artificial languages?

The first thing you encounter is a title: ‘\oriGaM[e,]i[,+ U]/’. 

The title takes the word origami, from the Japanese oru, ‘to fold’, and kami, ‘paper’, and finds in it – hears in it – the wholly unrelated English word me. So, the title adds an e in square brackets: me, as well as mi. This addition rather sets the i apart, producing an I alongside the me, the subjective alongside the objective case of the English first-person singular pronoun. The title then adds a capital U, suggesting the second-person pronoun, whether subject or object, and precedes it with the mathematical symbol for addition. Further square brackets separate that plus sign and the U from the origami and the me, and commas sequence the three pronouns, almost as if the self as object and the self as subject were separate people. Me, I, and you – but only a lowercase i, as a mark of modesty, perhaps, or as though something about the subject in question, or something about subjectivity itself, has been curtailed or devalued. Usually, it’s I which gets capitalised, but this title gives the capital to me, as though objecthood were somehow newly significant. Of course, the capital M also serves to indicate, in lieu of a space, that another word begins here, mid-word: it indicates the me which has been folded into – or which has itself unfolded from – the origami. In much the same fashion, the title capitalises g as well. No need to worry overmuch about the subordination of the subject: it’s only a game

These may or may not be the first thoughts you have when, browsing online late at night or sitting down with next week’s readings for a university course, you find yourself confronted by the title ‘\oriGaM[e,]i[,+ U]/’. But such thoughts won’t be surprising to anyone versed in the work of Mez Breeze, who published ‘\oriGaM[e,]i[,+ U]/’ ten years ago, in 2015. (Who is Mez Breeze? It’s a complicated question, to which we’ll return.) Breeze’s title is an instance of what she calls mezangelle, ‘a type of merging of programming languages/code with poetic elements’. The word mezangelle, as Sally Evans observes, is both noun and verb, and the word itself mezangelles: it merges or mangles Mez – a name under which Breeze, born Mary-Anne, often writes – with mangle, angle, angel, and elle, the French feminine third-person singular pronoun. It’s nothing if not an evocative compound. Recalling mezangelle’s origins, Breeze once explained that it began as a ‘style of writing//textual construction’ in the late 1990s, when, on mailing lists and in other network exchanges, her ‘particular “angle” was to take various information text tracts and “mangle” them through free/multi-word associative techniques’. Breeze and her readers frequently refer to mezangelle in terms of mangling, and mangling happens also to have a specific meaning in computer programming: ‘name mangling’ is a process by which functions and variables are given unique names to avoid conflicts in compiling. 

As a title, ‘\oriGaM[e,]i[,+ U]/ is certainly unique, and it also suggests computer programming. Those square brackets recall the use of square brackets in programming languages to specify the index number of an element in an array – that is, to specify the number which identifies a particular value or variable in a collection of values or variables. But of what arrays might ‘e,’ and ‘,+ U’ be the index numbers, and to what elements might they point? If nothing else, the brackets imply some kind of equivalent: an equivalent in the realm of subjects and objects for arrays in the realm of computer science. At the same time, Breeze’s square brackets indicate a logic of alternatives. If parentheses conventionally clarify or expand upon what has been said, these square brackets specify optional inclusions: either mi, on the one hand, or mi and me, on the other; either i, on the one hand, or i + U, on the other. Coming after that plus sign, the capital U recalls ∪, the mathematical symbol for the union of a collection of sets, as though adding you to I were to add to the self a union of selves. Alternatively, and insofar as that U means you, Breeze’s title evokes the irreverent orthography of online language or netspeak, as well as of popular culture more broadly. (‘I Would Die 4 U’, sings Prince.) In any case, the title’s combination of mathematical notation, backslash, forward slash, and square brackets connotes code. Moreover, while origami is the art of folding paper, many a source-code editor allows for code folding, whereby programmers can ‘fold’ or hide parts of a text, usually according to the syntactical structure of the programming language in use. Fold the ‘e,’ and the ‘,+ U’ and you’re left with ‘\oriGaMi/’. But if, in all these ways, ‘\oriGaM[e,]i[,+ U]/’ connotes code, it’s not itself code you could ask a computer to execute. Instead, the title’s typographical in[ter]vention suggests programming language and, by implication, an information technology far removed from – even, some might say, in opposition to – the handicraft of origami. 

Mangling makes good sense as a figure for all of this – for ‘\oriGaM[e,]i[,+ U]/’ in particular and for mezangelle in general – though some readers reach instead for other figures, from splicing to fragmentation. Likening mezangelle to Dada cut-ups, Evans speaks of Breeze’s ‘technique of lexical splicing’, a ‘mechanism by which words are fused into new polysemantic splicewords’. For C. T. Funkhouser, Breeze ‘divides and imaginatively re-joins fragments of words’. That invocation of imagination is in keeping with a common emphasis on play, and Breeze herself has identified ‘curiosity and play’ as essential to her work. For John Reep, this playfulness both emerges from programming language and undoes its smooth functioning: ‘mezangelle disrupts and destroys the efficacy of computer code by revealing its potential for linguistic play’. 

The proliferation of figures for Breeze’s ‘technique’ registers, even as it conceals, a further problem. It is easy to move from text to interpretation, from lowercase i to subordinated subject, or from mezangelle to Mez’s angelic angle as the feminine mangling of phallogocentrism. It is easy, too, to proliferate interpretations since, as Breeze urges, ‘there is never only one level of interpretation’ in her works, never ‘an ultimately correct [or incorrect] option’ (Breeze’s square brackets). Mezangelle ‘invites its readers into radically free interpretative relationships’, Dani Spinosa proposes. Different readers will, moreover, generate different interpretations, according to background and experience. Though ‘many people parse only the poetic underpinnings,’ Breeze reports, some readers, and especially those ‘in the code-loop’, ‘absorb the programming elements’ instead, or as well. But beyond the ease of interpretation, it is hard to say exactly how Breeze’s neologism and her title have been constructed, harder to specify the logic of their formation. What governs mezangelle as making? 

While Evans likens mezangelle to Dada cut-ups, Florian Cramer compares it to the portmanteaus of James Joyce and Lewis Carroll. The comparisons are plausible enough and the impulse to position Breeze in relation to precursors, be they avant-garde, modernist, or Victorian, is telling, but the comparisons don’t get us much further than do figures of mangling and splicing. Funkhouser, Rita Raley, and N. Katherine Hayles in her 2008 study Electronic Literature point to Breeze’s puns and homophones, concepts which do at least specify something about what Breeze calls ‘textual construction’. But to specify a logic for the formation of Breeze’s texts more precisely would, it seems, be to cancel the play. Breeze emphasises the freedom of those ‘associative techniques’, and Brian Kim Stefans neatly characterises her ‘rich’ verbal ‘surfaces’ as ‘part freed-signifier and part charismatic tsunami’. ‘Breeze’s mezangelled net.wurks are not designed according to strict, logical rules’, Reep explains, while Beatrice Beaubien muses that Breeze’s ‘word fragments’ operate ‘according to an alchemical formula’. But why ‘alchemical’, exactly? Is the ‘formula’ unknowable or only unknown? Might there after all be some hidden set of ‘rules’ governing mezangelle, some occult key to ‘\oriGaM[e,]i[,+ U]/’? 


Poetry, Language, Code 

As a title, ‘\oriGaM[e,]i[,+ U]/’ does not set generic expectations, in the way that a title like ‘Ode on the Spring’ (1748) does. But when ‘\oriGaM[e,]i[,+ U]/’ was published in 2015, it appeared online in Cordite Poetry Review, and that does seem unambiguously to frame the work as poetry, as a poem. Breeze and her readers sometimes label her works this way, too, whether as digital poetry, code poetry, or codework poetry. Still, something about these works complicates categorisation. Confronted by Breeze’s ‘_cross.ova.ing ][4rm.blog.2.log][_’ (2003), Spinosa notes that the work’s language is ‘part code’ and ‘part poetry’, decides that this language is ‘ultimately neither’ code nor poetry, and then refers to the work as a ‘series’ of ‘poems’. Hayles is more circumspect in a 2002 essay, calling Breeze’s works ‘poem-like objects’. The difficulty of categorisation leads other readers to abandon labels like poetry and literature for codework, and Breeze, too, has on occasion explicitly resisted the category of poetry, opting instead for codewurk and net.wurk. Poetry is ‘a label I’m uncomfortable with’, she explained in an early interview

The difficulty extends from Breeze’s works to Breeze herself. She has been called a poet, a digital poet, a code poet, and a codework poet, but also an ‘electronic author’ (by Kent Aardse), a ‘digital writer’ (by Spinosa), an ‘Internet artist’ (by Wendy Hui Kyong Chun), and a ‘net.artist and game designer’ (by Asun López-Varela Azcárate). Her work is nothing if not diverse, moving across media and between literature, visual art, gaming, programming, and networked communication. Breeze’s Bluesky account describes her as an ‘OG digital artist/writer’, a ‘Permaculturalist’, and a ‘Narrative Designer + Games Developer’. As the website for Mez Breeze Design puts it, her ‘genre-defying work often dodges neat categorization, continually challenging traditional aspects of the games/literature/arts and spatial computing industries’. Even in the case of a printed book, seemingly so old-fashioned, Breeze is keen to advertise freedom from established or inherited genres. In her short preface to Attn: Solitude, published in 2017 by Cordite Books, Breeze explains that the work is neither ‘a straight poetry book’ nor ‘a strict collation of cyborgian-emulated [chap+lady]book texts’. 

In formulating this generic indeterminacy, Breeze invokes an original or essential distinction between poetry and code, though the distinction is far from straightforward or static. Recently, the production of poetry by Large Language Models has seemed to some to bridge the gulf between poetry and code, human and machine, soul and algorithm. The title of a recent book of poems generated by one such model, I Am Code, might seem to be at once a declaration by that model and, by implication, a declaration on behalf of poetry itself, identifying poetry as code. But the editors of I Am Code never suggest that the model’s own code is poetry; indeed, though they use the model, they never read its code. Moreover, in his manifesto for a critical approach to the reading of code, Mark C. Marino declares that ‘Code Is Not Poetry’, except in the unusual case of codework, because code is primarily functional, poetry primarily aesthetic. Nevertheless, Geoff Cox, Alex McLean, and Adrian Ward have argued that code, too, has an ‘aesthetic value’ and that, though code may not be poetry, the comparison of code with poetry is illuminating. As long ago as 1968, Donald E. Knuth proposed, in The Art of Computer Programming, that programming could be ‘an aesthetic experience much like composing poetry’. Indeed, Marino also acknowledges that there can be an aesthetic dimension to code, and even to code ‘written to solve a utilitarian problem’. Code can be ‘artful’, even when it is not art. 

Still, that sense of an original or essential distinction persists. Most descriptions of mezangelle distinguish either between code and poetry or between code and natural language. On the one hand, as we have seen, Breeze defines mezangelle as a merging of poetic elements with programming languages or code. On the other hand, she also describes mezangelle in a 2004 piece as ‘the arrangement, dissection, and splicing of […] code scripting practices in2 english’ (my ellipsis). In digital poetry more broadly, writes Norbert Bachleitner, ‘natural language’ is ‘contaminated by machine language’. Hayles termed the resulting hybrid of code with natural language a ‘creole’ at a conference in 2001 and then again in her 2002 study Writing Machines, and other readers follow her lead, though John Cayley objects to the label because, unlike a true creole, codework is not the first language of a community. 

Regardless, it’s tempting to accept the distinction and to read ‘\oriGaM[e,]i[,+ U]/’ as the grafting of code onto natural language or onto ‘straight’ poetry. Structured in five steps, ‘\oriGaM[e,]i[,+ U]/’ offers something like the instructions for an origami crane or lantern. Here is the first step: 

Step 1: Start with a 6_itch x 6_itch identity with desire side down. Fold yearnings in half on the fan[boi+gal_ecs]tasy axis. Sleep_crease well and unfold. 

A natural or conventional syntax structures each of these three imperatives. The nouns and verbs are then troped according to a central conceit, a metaphor: the making of a relationship is like the making of an origami design. So, we begin not with a sheet of paper but an ‘identity’, placed ‘desire side down’, much as, in starting a relationship, one might keep one’s desire to oneself, hidden from sight. The metaphor is then extended when we learn or remember that, in mathematics, an identity matrix is a square matrix whose principal diagonal consists of ones, with zeroes everywhere else – that is to say, an identity matrix rather resembles the illustration of an origami sheet with a diagonal line showing where it should be folded. In ‘\oriGaM[e,]i[,+ U]/’, however, we fold in half not a sheet of paper but ‘yearnings’, and when we ‘crease’ those yearnings we do so, after ‘ecs]tasy’, by ‘Sleep[ing]’. This turn to a language of emotion, together with the use of metaphor, suggests, beyond natural language, the art form of poetry – or at least a certain idea of poetry. Having expressed her discomfort with the category of poetry, Breeze notes that, nevertheless, some of her work does approach ‘orthodox poetry boundaries in that I use succinct emotive phrases/words’. Words such as desire and ecstasy, for instance. But here, it seems, ‘orthodox’ or ‘straight’ poetry has been mangled or contaminated by code: the underscores, the brackets, the plus sign. 


The Principle of Equivalence 

Mangled how, exactly? For a start, ‘6_itch x 6_itch’ substitutes itch for inch and x for by. Again, it is easy to move from text to interpretation. One standard size for origami paper is a square measuring six inches on each side; the making of a relationship begins, ‘\oriGaM[e,]i[,+ U]/’ explains, with an identity measured in itches to be scratched, cravings to be satisfied. In that context, the x for multiplication may mark a kiss, while ‘6_itch’ punningly sounds a ‘sick sitch’, a situation of sickness. ‘6_itch x 6_itch’ equals sick sitch kiss sick sitch. In many programming languages, underscores serve to string multiple words together as identifiers for variables, and ‘6_itch x 6_itch’ might be called an ‘identity’ insofar as it holds for all values of the variables in question. Whoever you are, whatever your itches, this is where you start. Have one sick sitch kiss another sick sitch and you’ve begun. 

Let’s move instead from interpretation to construction, to mangling as making. In ‘fan[boi+gal_ecs]tasy’, Breeze unfolds fanboi and fangal from the first syllable of fantasy, and ecstasy from its second and third syllables. These are, we might say, kinds of rhyme. In this way, the first step in ‘\oriGaM[e,]i[,+ U]/’ performs what Roman Jakobson famously terms ‘the poetic function’: ‘The poetic function projects the principle of equivalence from the axis of selection into the axis of combination’ (italics removed). What Jakobson means by this is that, in some forms of language use, ecstasy is more or less equivalent to its synonyms: bliss, rapture, euphoria, and so forth. These, for instance, could all be words for the state or experience achieved when two people come together. In wanting to name that state or experience, in many discursive situations, we select one of these equivalents and it does not matter much which equivalent we choose. In a situation governed by the poetic function, however, the ‘principle of equivalence’ determines not the set from which we choose, but the combination of words chosen. We choose ecstasy because, unlike bliss, rapture, and euphoria, it’s equivalent to fantasy, in the sense that it has the same suffix. We choose fanboi and fangal because they share a first syllable with fantasy. This is true, despite the fact that those first syllables are semantically and etymologically distinct: fantasy derives ultimately from Ancient Greek phainein, ‘to show’ or ‘to appear’, while fanboi and fangal derive, through fan and fanatic, from Latin fānum, ‘temple’. In ‘\oriGaM[e,]i[,+ U]/’, the effect is, through equivalence in sound and spelling, to bridge that semantic divide, to propose that romance is a fantasy consisting in a kind of fandom, and even that the fantasy is false because the fandom, though reciprocal, is structured unevenly, each the idol to the other’s fan. But it’s the specific logic of construction, the poetic function, which makes that interpretation and other interpretations possible. 

The bridging of a semantic divide is sometimes valued as rhyme’s proper purpose: it’s the kind of rhyme celebrated in the work of Alexander Pope by W. K. Wimsatt, for whom, as for many critics, rhyme is at its best when similarity in sound works productively with the words’ meanings, whether those meanings be similar or different, and especially when the words serve different grammatical or syntactic functions. Better to rhyme noun with verb than noun with noun, for instance. The rhyme of fantasy with ecstasy would seem poor in comparison: two nouns, similar only in their common suffix, rather than in their roots. But in fact, rhyme of this kind can be equally winning. In ‘Goblin Market’ (1862), Christina Rossetti rhymes raspberries with mulberries, cranberries, dewberries, and blackberries to as good effect as one could wish for. Ultimately, it’s not rhyme as a poetic device or technique, with its particular histories of practice and theory, which is at issue in ‘\oriGaM[e,]i[,+ U]/’. Breeze can be breezily indifferent to inherited canons of rhyme, because it’s the underlying principle of equivalence which matters. 

For Jakobson, the poetic function extends well beyond rhyme, too. Alliteration and assonance are also good examples, as are metaphor and metre. Take the fifth and final step in ‘\oriGaM[e,]i[,+ U]/’: 

Step 5: This is an interesting step that we'll repeat several times. RIP the [re]active side back to the left. Fold and boil your_[d]raw[n]_self_dry. Heave through your rou[Sal]tines while [D]ang[er]St_breathing. Murder_shift the table and p[b]o[lt]st the door. Then flIP your sIT_st[K]ill switch and howl. 

Here, equivalence is a feature of both sound and spelling. Alongside the assonance of your, raw, and drawn, the initialism R.I.P. and the verb rip, though pronounced very differently, share the same letters. The /eɪn/ in danger differs from the /æŋ/ in anger and angst, but all three words are spelt with ang. (In this, Breeze’s technique differs from the rhyming of Rossetti or Pope, because that rhyming is largely indifferent to spelling.) Beyond phoneme and letter, the poetic function extends as well to stress and so to rhythm. Then flip your sit-still switch and howl is as perfect an ‘iambic tetrameter’ as Andrew Marvell ever penned, because, as Jakobson explains, the poetic function treats each stressed syllable as equal to every other stressed syllable, each unstressed syllable as equal to every other unstressed syllable. The equivalence can function, moreover, not only within a work but between works, so that in reading ‘Then flIP your sIT_st[K]ill switch and howl’ one may hear a rhythm heard before. 

Jakobson is careful to argue that the poetic function is by no means unique to poetry. It can appear in any use of language, from odes to election slogans, casual conversation to advertising jingles. At the same time, poetry can perform other linguistic functions: it can refer or express or command. ‘The poetic function is not the sole function of verbal art’, Jakobson writes, ‘but only its dominant, determining function’. So, I don’t at all mean to insist that ‘\oriGaM[e,]i[,+ U]/’, because it demonstrates Jakobson’s poetic function at work, must be categorised as poetry. Nor do I mean to claim that Jakobson’s poetic function explains everything that is distinctive or significant about Breeze’s work. I do want to say that ‘\oriGaM[e,]i[,+ U]/’ shares a logic of construction with many other and much older works which have been categorised as poetry – works that rhyme, works that are metrical, works that are metaphorical. And I want to propose that this logic of construction is most apparent precisely when one might least expect it, when Breeze’s text seems furthest from ‘straight’ or ‘orthodox’ poetry, as well as from natural language. The poetic function appears most powerfully when Breeze’s text also involves code. It’s not that in ‘\oriGaM[e,]i[,+ U]/’ code belatedly mangles a pre-existing poetry. Instead, the code and the poetry are intimately interanimated: each becomes the other’s condition of possibility. 


Mezangelle in History 

I say ‘the poetry’, but really that is to make universal what is actually particular. Jakobson’s notion of the poetic function describes some of what’s important to some works which have been called poetry. No more than that. Across history, poets, philosophers, critics, and others have proposed for poetry other ‘dominant, determining’ features or qualities, from succinct emotive phrases and words to the act of making, to poiesis. No single concept of poetry plucks out the heart of its mystery, because poetry isn’t a mystery; it’s a history of diverse practices and theories. Still, recognising Jakobson’s poetic function at work in ‘\oriGaM[e,]i[,+ U]/’ helps in several ways. 

First, it offers a longer historical perspective. Comparing Breeze’s work to avant-garde and modernist writing and art – Joyce, Dada, the found object, concrete poetry, language poetry – is popular, but attention to phonemic equivalence opens up other, and older, comparisons. There’s reason to think that Breeze might be open to this, too. When offering a list of ‘influences>tools>inspirations’ in a 2007 interview, she cited Samuel Taylor Coleridge’s ‘The Rime of the Ancient Mariner’ (1798) alongside, amongst others, George Orwell, Kathy Acker, Duke Nukem, and post-structuralism. Ten years later, Breeze quoted Coleridge’s characterisation of poetry as the ‘willing suspension of disbelief’, calling the formulation ‘beautifully phrased’. 

Second, recognising the poetic function at work in ‘\oriGaM[e,]i[,+ U]/’ may check the impulse to read Breeze’s work and codework more broadly as a radical rupture, as an effort, in the words of Andy Carruthers, ‘to change the ontological ground of poetry itself’. The impulse is undoubtedly strong. For Anna Gibbs, codework allows Breeze to ‘highlight the complexities both of agency and address that print-based technologies allow us to overlook’. For Bachleitner, digital poetry by Breeze and others is newly self-reflexive and newly determined by ‘indeterminacy and différance’. Breeze, too, has said that mezangelle is ‘designed to disrupt traditional poetic construction’. But to identify a rupture in the history of poetry requires a specific understanding of that history. Is print-based poetry so very simple in its agency and address? Are older poems so unself-reflexive, so free from indeterminacy and différance? Clearly both the practice and the concept of poetry change, but attending to continuities across its long history might help us better to see the differences. After all, as Breeze remarked in that 2007 interview, she doesn’t ‘demand a type of polaris[blank]e[t]d coupling 2 the new|future or against the old|past’. 

Jakobson writes of the principle of equivalence being projected into the ‘axis of combination’, by which he means the sequencing of rhyme words, of stressed and unstressed syllables, and so forth. That is the kind of combination which governs the line I unfolded from ‘\oriGaM[e,]i[,+ U]/’: Then flip your sit-still switch and howl. This is a combination which works with the flow of natural language. But ‘st[K]ill’ works differently; the kind of combination it represents works against or interrupts that flow. I can hear the /i/ chime three times in sit-still switch, but I can’t hear it chime in still and kill together; I seem instead to have to choose. Combination as splicing cuts across combination as sequencing. Readers have often identified this quality of interruption, this disruption of linearity, as important to mezangelle, with Stephanie Strickland suggesting that Breeze’s work rewards ‘a scanning multidirectional view that is not restricted to movement in lines’. Rather than choosing whether to voice still or kill, I’m invited to move back and forth between them, in a movement which suspends the movement of natural language. I’m invited to consider the relation between stillness and death, between sitting still and a kill switch, between – in the aftermath of a relationship gone wrong – the paralysis of grief and the desire to die. In similar fashion, ‘flIP’ splices the flipping of that switch with the Internet Protocols which govern the transfer of data across networks, so that to flip the kill switch might be, after a breakup, to cut communication. Is it this disruption of sequence, then, which constitutes, if not a change in poetry’s ontological ground, at least Breeze’s signal innovation? 

Insofar as mezangelle thwarts the easy functioning of conventional syntax, it asks us to take each word separately. There is no conventional syntax to determine the relation between still and kill, as there is to ensure that ‘flIP’ takes ‘switch’ for its object and that ‘and’ coordinates ‘flIP’ with ‘howl’, whether those actions be simultaneous or successive. You can invent a configuration for anger, angst, and danger, but other configurations will be possible. True, code suggests an alternative syntax, an alternative logic for relating elements. When we are instructed, in that fifth step, to ‘Murder_shift the table and p[b]o[lt]st the door’, the underscore combines murder and shift to coin a new verb, while the square brackets offer bolt as an alternative to or gloss upon post. At this point in the relationship, perhaps, it is as if moving the table in the living room is to murder it, as if murdering the table by hacking it to pieces is the only way to shift it, as if the inanimate table is alive while I want only to sit still and die. At the same time, shift tables are used in the analysis of clinical trials to measure change in subjects over time, so perhaps it is as if the trial of a relationship in ‘\oriGaM[e,]i[,+ U]/’ leads not to cure but to death. What about post and bolt? Is it that one posts a door by posting its picture on social media? Would that then also be, in effect, to bolt the door, by advertising its closure? Or is one to bolt the door so firmly it’s as if the bolt’s as long and strong as a post? As ever, multiple interpretations are possible. When Breeze explained her use of square brackets in 2004, she said that they ‘indicate a meaning-mode-hyperjump’, which rather suggests interpretative freedom, not a fixed function specified by a programming language. Without the conventional syntax of natural language to configure post with bolt and murder with shift, each word’s every shade of sense, and even the most remote of connotations, becomes newly relevant, newly significant. 

It is the word in this new light, in suspension, which may then be placed in relation to the words around it or to the words folded into, or out of, its very self. But this is not inarguably a radical rupture either, for it is precisely what Roland Barthes, writing seven years before Jakobson, identifies as the mark of modern poetry. In ‘classical’ or pre-modern poetry, Barthes proposes, each individual word is subordinate to its relations; it has its meaning in relation to other words, subject to the ‘classical flow’ of language; ‘connections lead the word on’. In modern poetry, beginning in the second half of the nineteenth century with Arthur Rimbaud, and intimated before that by Victor Hugo, those ‘[f]ixed connections’ are ‘abolished’ and the word stands like a ‘monolith’; it ‘plunges into a totality of meanings, reflexes and recollections’. The reader of modern poetry ‘encounters the Word frontally, and receives it as an absolute quantity, accompanied by all its possible associations’. ‘\oriGaM[e,]i[,+ U]/’ may not look much like Une saison en enfer (1873), but they share a poetic ground. 

Or they do if we accept Barthes’ concepts of classical and modern poetry. We could instead wonder whether, say, the tumbling rhyme, alliteration, and assonance in John Skelton’s ‘Collyn Clout’ (ca. 1521-22), even as they depend upon the flow of language, also interrupt or suspend that flow. ‘What can it avayle’, Skelton asks, 

To ryme or to rayle, 

To wryte or to indyte, 

Other for delyte 

Or elles for despyte? 

Or bokes to compyle 

Of dyvers maner style, 

Vyce to revyle 

And synne to exyle? 

To teche or to preche 

As reason wyll reche? 

If Skelton’s repeated or separates rhyming from railing, teaching from preaching, the phonemic equivalences bid us continually question that separation. If the rhyming of ‘revyle’ and ‘exyle’, together with Skelton’s ‘And’, promises that the reviling of vice and the exiling of sin are more or less the same, the rhyming of ‘revyle’ and ‘exyle’ with ‘style’, alongside the assonance of all three words with ‘dyvers’, also suggests that the reviling of vice and the exiling of sin need, as examples of those divers styles, to be distinguished. Is ‘to indyte’ merely to ‘wryte’? Or, instead of imposing semantic equivalence, does the rhyme point up semantic difference, reminding us that ‘to indyte’ is also, more generally, to put into words or to compose in a poetic or otherwise literary form, and then also to level an accusation, even to make a criminal charge in a legal process? What exactly is the relation between law and religion in this satire on the clergy, the nobility, and social corruption in general? Skelton’s poetry amply demonstrates how Jakobson’s poetic function may also produce a movement back and forth, a suspension of syntactic flow in the search for meaningful relations between the equivalents it combines or even, it may be, a suspension of syntactic flow in the suspension of that search. From this perspective, Breeze shares a poetic ground with both Skelton and Rimbaud, and countless others beside. Or, if differences remain, they are differences of degree, rather than of kind. 

But the risk with any concept of poetry – be it Barthes’ or Breeze’s, Coleridge’s or Jakobson’s – is that it reduces particulars to a universal. The risk is that, insofar as ‘\oriGaM[e,]i[,+ U]/’ abolishes fixed connections and offers words as absolute quantities, it becomes merely an instance of modern poetry, or that, in its performance of Jakobson’s poetic function, it becomes an indifferent equivalent of ‘Collyn Clout’. On the contrary, ‘\oriGaM[e,]i[,+ U]/’ puts that common logic of construction to work in a particular way, and that particularity makes all the difference. To appreciate how ‘\oriGaM[e,]i[,+ U]/’ is more than a clever exercise in linguistic play and generic indeterminacy, to see how it also critiques the logics which condition love, sex, and life more broadly in an age of code, we need to understand precisely what ‘\oriGaM[e,]i[,+ U]/’ does with equivalence. 


For ‘\oriGaM[e,]i[,+ U]/’ is about equivalence. The whole work, after all, is a series of steps to be repeated, and the fifth and final step compounds that repetition with its own: ‘This is an interesting step that we'll repeat several times.’ This quality of repetition belies the finality of Breeze’s last sentence, ending with a ‘howl’ beyond which, it might otherwise appear, there is nowhere to go. This quality of repetition also extends the central conceit, from the making of a relationship to its maintenance, from relationship as product to relationship as process. Follow the instructions over and over, and you can make a thousand origami lanterns; follow ‘\oriGaM[e,]i[,+ U]/’ over and over, and you can make a thousand broken relationships or break your relationship a thousand times. 

This, too, is an aspect of Jakobson’s poetic function, for that function ‘makes reiterable not only the constituent sequences […] but the whole message as well’. A poem repeats a rhythm or a rhyme or whatever other feature of the language is determined by the principle of equivalence, and a poem itself elicits repetition. You can read a poem or voice a poem over and over again, and however differently you perform or interpret it, it remains itself, it retains its identity. ‘This capacity for reiteration whether immediate or delayed,’ Jakobson continues, ‘this conversion of a message into an enduring thing […] represents an inherent and effective property of poetry.’ But whatever it may or may not mean for defining poetry, in ‘\oriGaM[e,]i[,+ U]/’ reiteration hurts. However acute the pain, each time you ‘flIP your sIT_st[K]ill switch and howl’ is equivalent to the last. However singular it seems, each new fantasy of your idol or idol of your fantasy is equivalent to the last. You, in fact, are equivalent to every other fanboi or fangal. 

The breaking of a relationship is routine, says ‘\oriGaM[e,]i[,+ U]/’, and in some ways ‘\oriGaM[e,]i[,+ U]/’ operates like a programming routine. You can’t ask a computer to execute it, but you can execute ‘\oriGaM[e,]i[,+ U]/’ yourself, over and over, to your heart’s [dis]content. Each time you ‘Heave through your rou[Sal]tines’, the time is indifferent: it doesn’t matter how many times this particular relationship has broken down or how many other, earlier relationships have already broken down. It’s all one and the same, like your routine rue or ruth, like the predictably salty sobs you heave, like the salt crackers you binge and binge again. The routine is indifferent to people, too: 

Step 3: Fold XY to meet XX. Seduce_crea[m]se well and u[E]nfold. 

Step 4: Now fold XX to meet XY. C[|G]rease well. 

That vertical bar or ‘pipe’ indicates, in some programming languages, the logical operation or, as if to notate crease well or grease well, or both, and as a typographical character, the pipe neatly figures the crease or fold itself. But in these two steps, code extends from computing to genetics. The DNA that determines sex is troped as the points on a sheet of origami paper labelled in instructions as X and Y to indicate lines for folding. This reduction of the person to biology excludes other chromosomal configurations (as, for instance, XXY, XYY, XXX, X0) and ignores phenotype and gender. It also makes one individual with two X chromosomes equivalent to any another, one individual with an X chromosome and a Y chromosome equivalent to any another. 

That is to say, the third and fourth steps introduce a gap between, on the one hand, the particularity of two chromosomal configurations and the gender identities with which they are associated and, on the other, the seeming universality of the imperatives, as if addressed to anyone and everyone: ‘Fold’, ‘u[E]nfold’. In Breeze’s title, too, the personal pronouns appear as shifters without a situation to fix their reference, as if both you and I and anyone else can be me or I or you. But it becomes clear, in the way these steps extend code to genetics, that this routine scripts relationships in specifically heteronormative terms. In the first step, too, Breeze unfolds fanboi and fangal from fantasy as if in accord with the fantasy of boy meets girl, but the choice of gal, rather than girl, casts that fantasy as old-fashioned caricature and the choice of boi, rather than boy, recalls that word’s use in the 1990s as a term for men who refuse traditional forms of masculinity and for women who appear or identify as masculine. (Soon enough, of course, boi was borrowed by others for other uses, and by 2002 Avril Lavigne could sing about her ‘Sk8er Boi’.) Breeze has herself actively resisted gendered scripts by varying the avatars she uses online and in publication: some of these avatars are gender neutral, some are masculine, some are feminine, and some, like ‘tech.no.whore’, advertise Breeze’s feminism. In this light, ‘\oriGaM[e,]i[,+ U]/’ critiques the codes, both social and technological, which condition modern relationships. Until 2016, the year after ‘\oriGaM[e,]i[,+ U]/’ was published, Tinder allowed its users only two gender options. So, too, in ‘\oriGaM[e,]i[,+ U]/’ the routine conscripts the reader, whomever the reader may be, and it does so over and over again. 

It’s not only that the hurt is reiterated, then. It’s that the reiteration itself hurts – it’s the seemingly inescapable indifference, the reductive equivalence. Partly it hurts because ‘\oriGaM[e,]i[,+ U]/’ extends that central conceit from origami to programming, so that relationships are likened not only to origami designs, but also to programming routines. The conceit pushes past the consolation of handicraft, which grounds repetition in the rhythms and affordances of the human body. It hurts to think the human has been reduced to the machine. But relationships are also not like infinitely reiterable programming routines, for they are not indifferent to particulars. The person, the place, the time, the desire, the fantasy, the anger, the howl – each itself matters. As an instance of equivalence, therefore, Breeze’s metaphor both compares and contrasts; even as it proposes a similarity, the metaphor displays a difference. 

One sure sign that particulars matter to ‘\oriGaM[e,]i[,+ U]/’ is its performance of Jakobson’s poetic function. When we are instructed to ‘Seduce_crea[m]se well and u[E]nfold’, the principle of equivalence both depends upon and reveals a rich set of particulars. There’s the history by which Latin chrisma, ‘chrism’, comes via Old French cresme into Middle English as creme, creem, or creyme, comes then in the sixteenth century to be spelled with an a, and so converges in its digraph with crease, whose etymology is obscure. There’s the fact that, though crease and fold are in some ways synonyms, one can fold, unfold, and enfold but one cannot uncrease, and though one could as late as the nineteenth century encrease, that word meant to increase, not to enclose with a crease. (Unlike crease, increase comes from Latin crēscĕre, ‘to grow’.) So, Breeze can fold cream into crease, but not into fold or its cognates, and she can combine the making and unmaking of a fold, but not those of a crease. Although seduce and crease both end in /s/, the difference in spelling means that they must be combined in sequence, with an underscore, rather than by splicing, with square brackets. Both cream and crease, moreover, can be verb or noun. (‘Cream’, sings Prince, but is that a flirtatious command or the naming of an object of desire?) This particular nexus of etymology, morphology, and grammar permits plenty of interpretative play. It may be that we’re instructed to seduce by creasing, which would also be to seduce by creaming, or it may be that we’re instructed to seduce a crease, a crease in the flesh of the idol of our fantasy, which would in turn be to seduce a cream, to educe the cream of that idol. None of the available synonyms would have the same effect: not lotion or ejaculate, not fold or groove or furrow. ‘Seduce_crea[m]se well’ is an imperative, but who exactly is to crease and who is to cream? Whose crease and whose cream are in question? Then, when you ‘u[E]nfold’, do you unfold the other to enfold them in an embrace, or do you unfold from them, after the embrace, and enfold yourself, alone? In this fantasy of ecstasy, the individual as subject or object is, again, merely indifferent, but the formulation of the fantasy is acutely particular, unrepeatable. 

Is there a lesson in this for thinking about poetry, that always insufficient, seemingly indispensable, astonishingly generative concept? A lesson for histories of poems and poem-like objects? For anxieties about the place of poetry in an age of code? For thinking, in turn, about code? Maybe, though if so, never as bodiless generalisation and never without commitment to differences, care for the details. 

‘\oriGaM[e,]i[,+ U]/’ is made from the particulars of language, it is itself a particular linguistic artefact, and it puts that particularity to a particular purpose: ‘\oriGaM[e,]i[,+ U]/’ in itself holds out an alternative to the reiteration that hurts; it opposes the indifference, the equivalence. Though the principle of equivalence is fundamental to mezangelle in general, each work of mezangelle works with its own specific and historically contingent materials. In this respect, ‘\oriGaM[e,]i[,+ U]/’ is by no means merely an example of mezangelle, and mezangelle is not a routine or technique which can simply, automatically be applied to any and all materials to the same, inevitable effect. In its particulars and in its way with the particular, ‘\oriGaM[e,]i[,+ U]/’ opposes the logic too often imposed by social media, by romance conventions, by the misappropriation of science, by the production of commodities for exchange. It affirms the value of the particular in relationships and in the world at large. ‘\oriGaM[e,]i[,+ U]/’ is itself neither a salt cracker, one of a myriad identical factory products, nor a profile pic, one of a myriad indifferent images presented for swiping left or right; it is a singular examination of equivalence.