In the spring of 2019, I chanced upon Lee Sumyeong’s book 마치 in a small campus bookstore. What initially drew my attention was the book’s clipped adverbial title, which stood out both for its brevity and for its defiant lack of “poetic” ballast. Picking up the book and opening it to the first poem, I proceeded to read something like the following:
things like cement vegetable paper
A man runs a field and the field of the man running the field caves in. A man lacking a caved-in field to the man runs. Things like cement vegetable paper sweltering he haphazardly plucks cabbage. Picks up the cabbage and walks the field a man bored through.
The mice escaping the field to escape the field is uncomfortable.
A man runs a field and because the field has no lid the era of the field closes the man. Closes the field. The man running the field drags over the field. The field is an expanding use. Into the man the man is ousted. Brings up things like cement vegetable paper.
Even though I was already in my mid-thirties at the time, my encounter with Lee’s poem strikes me on reflection as having been somehow formative. The strangeness of the poem—its uncanny grammar, its entanglements of man and field, its subtle yet vast thematic implicature, its permutational richness—hit me almost immediately. By that point, I had been living in Korea for just under a decade, and I had already been intermittently translating Korean poetry (albeit in a quite dilettantish manner and without any plans for publication) for several years. But reading Lee’s poem, I felt something that I had not felt previously—a compulsion to translate, perhaps even a need to translate.
Looking back, I recognize the intensity of this response as the first stirring of a kind of psychic imprimatur—a reflexive license to translate something seriously for the first time in my life. In my previous dabblings with translation, I had convinced myself that the practice was not something that, for me, could ever amount to more than a private diversion. While doubts about my Korean ability were surely a factor in my trepidation, I now realize that the mental blockage stemmed not purely from concerns about competency or aptness; it was also the result of a certain literary pessimism—a pessimism of which Lee’s book, more so than any other, has since succeeded in disabusing me. This change was triggered in part by the cognitive polarities that Lee’s poetry elicits. On the one hand, there is an undeniable immediacy to the poems: not only are they comparatively pared down in terms of diction, but they are also largely free of any discursive or allusive freight. But from clarity comes confusion: crystalline sentences give way to complex run-on syntagms; grammatical moods shift without warning; conjunctive clauses prove tenuous or illogical; poems end abruptly or with apparent non-sequiturs. For whatever reason, this dichotomy between order and confusion made me feel that genuine ingress into the world of translation had suddenly become more possible.
1 From Homeophrase to Allopoem
Before the first poem proper in 마치, there is a highly suggestive sliver of prefatory matter:
3월, 행진, 망치,
그리고
Als Ob
This preface seems designed both to repel a facile translation and to highlight translation itself as being somehow key to the poetics of the book. In a mere handful of words, there are tinctures of multiple scripts or languages: Arabic, Korean, Chinese, English, German, and Latin. The “literal” semantic translation of the words is: “March, march, hammer / and [also] / as if.” Transliterated, the Korean words read: “samwol, haengjin, mangchi / geurigo.” If examined purely from the point of view of their semantic and phonological Korean properties, the words in the first string bear seemingly no relation to each other. However, in the process of translating the first two words into English, one can quickly discern what some theorists would call the tertium comparationis (hereafter “tertium”), the “factor which links or is the common ground between two elements in comparison” (OED). The tertium in this instance is the English word “March/march” qua its analysis into the phonetic unit “mɑːtʃ.” Appropriately, the link is weak, identifiable only via a kind of recognized difference whereby the monosyllabic “mɑːtʃ” bends according to the rules of Korean phonology, becoming the bisyllabic “machi” (마치) of the book’s title. Thus, the language that is translated out of is almost instantaneously translated back into. All of this is carried out furtively: “March” and “march” exist merely as implied translations, while “마치” is observable only via its deformation into the word “망치” (hammer), whose jamo “ㅇ” functions both as the welt of the violence and the instrument of it, while also reinforcing the spoliation by turning the syllable into a homophone of the Sinitic “亡” (destroy, collapse, etc.). There is a similar play in the remaining lines: “Als Ob” bears a graphical similarity to “also” (a latent translation of “그리고”), such that we might say that it is as if one locution were potentially just like the other—or, given the refractory capitals, the typographical space, and the appended “b,” as if it were almost just like the other. Within this brief textual field, then, there is a kind of semantico-phonological shell game going on, by which the kernel of likeness engendered by the word relationships keeps shifting into mere similarity. On the one hand, the strategic arrangement of Lee’s words suggests that a hyper-translation of them might read “마치, 마치, 마치, / 마치 / 마치” (or “as if, as if, as if, / as if / as if,” etc.). But the lines are also set up to resolutely thwart such a translation, to deny the machi or the als ob or the as if within their midst.
This preface heralds one of the major concerns of Lee’s book—how within the concept of likeness there is an inherent vacillation between identity and difference. While translating the collection, I thought constantly about distinctions between the homo- (the same) and the homeo- (the similar). It is perhaps worth noting that these two combining forms bear a strong resemblance to each other, and they can in certain guises appear almost indistinguishable.[1] Likewise, words and phrases in Just Like constantly hover between these contiguous states of identity and similarity. A few excerpts will hopefully prove illustrative:
between new gradation and gradation
pointing at similar gradation (“raise arm”)
just like
dead leaves will surely cover the ground
dead leaves will surely cover the ground completely [ . . . ]
just like
as if overflowing (“just like”)
into a complex into a complex
crying children enter and because the complex spills out
there are so many complexes so
they are living in the same complex (“residents”)
Note first the lines “just like / as if overflowing.” On the one hand, the simile is placed at a further remove from the referent via the conjunctive doubling, while on the other hand it seems to overflow into the referent via the suggestiveness of the line break. An examination of the other excerpts can help us clarify this condition still further. Collectively, this suite of quotations would seem to suggest that Lee is strongly drawn to the rhetorical figure of epizeuxis—the proximate repetition of words or phrases. It should be noted that Korean admits of repetition more readily than English does—whereas the latter tends to eschew nominal repetition and opt instead for endophoric pronouns or the like, the former is more forgiving of lexical reduplication. Korean is also far more homeostatic in terms of its sound culminations—sentence endings are conventionally fixed or limited, tending in the written language to culminate with a “-다” (da) particle. In this way, Korean comes pre-installed with something akin to what, since Aristotle’s Rhetoric, has been known as homoeoteleuton, repetition of the same or similar word endings. But if Lee is already operating within a linguistic system that is comparatively built for repetition, her repetitions nonetheless move strikingly beyond those that are standard in the language. Instead of mere nominal reduplication or inbuilt line-ending repetition, there is something like a confluence of the two, a significant or wholesale repetition of word or phrase—something that we might call homeophrasis. A homeophrastic line works differently to an epizeuctic one; whereas epizeuxis is rhetorically concerned with emphasis or reiteration, homeophrasis is predicated on ontological mutability or uncertainty. Thus, “gradation” can be similar to or identical with “gradation”; to “cover the ground” and to “cover the ground completely” can be “just like” each other, even if neither is propositionally sufficient; and “so many [different] complexes” can simultaneously be “the same complex.” For Lee, the homeophrase is one of the fundamental units of a poetics in which image, referent, and category constantly flicker between uniqueness and equivalence, between form and nebulosity, between immanence and its evacuation.
the night drives everything into the night.
The night returns to the night and
along with the night the night is completely full. (“night’s formation fliers”)
This poetics of similarity expands beyond the homeophrase up to the level of the poem itself. Like elusive dreams, Lee’s poems are often no more than semi-apprehensible. When we fall asleep or wake up, the opposed logics of dream and wakefulness coalesce into a hypnagogic pool in which things become “without reason”:
Woke slowly. Woke crying. Tears fell without reason from within sleep. The shoulders are scattered. Who was crying who was briefly hidden there are times even when I’m a girl times even when morning becomes perforated. When morning appears give me morning. Well-known costumes were invariably established I keep wanting to be your costume. Since the future initially passed we continued to use the future. Throwing the beach ball the girls are reiterated and who was briefly hidden who was crying I want to be your empty beach ball. Today misplaced sleep. I fit wherever. (“someone briefly”)
As elsewhere in Lee’s poetry, the syntactic unfolding is highly mutable, with short, clear sentences nestling against mercurial syntagms that forgo punctuation. While much modern poetry eschews the full stop altogether, Lee incorporates it into almost all of her poems—never as an atavistic feature, but rather as a highly estranging device whose demarcating energies are as much fluid, oblique, and arbitrary as they are propositionally terminal. Such syntactic vacillation provides a good foil for the above poem, wherein the attempt to relay an oneiric scene is full of porosity, dispersal, and confusion. The awakened crier cannot uncomplicatedly be the agent of the crying, as the tears fall “from within sleep.” But neither is it the case that there is a monolithic dream-crier. Just as the lines juxtapose multiple modal conditions (interrogative, indicative, optative) and undulate between various temporal horizons, so too does the “I” shift (or potentially shift) to various loci (“girl,” “costume,” “empty beach ball,” “today,” etc.). In short, the rendering into language bilaterally compromises the waking and the dream states, obscuring the essence of what is being described—there is similarity, perhaps, but too much uncertainty for there to be identity or identification.
This uncertainty, which runs throughout Just Like, has the effect of suggesting that Lee’s poems are not quite resolved. Both within and outside of the poems there are, I believe, other poems, variants of the given, secret iterations occulted in the undertows of disintegrative logic, isomorphic grammar, surreptitious homophony, and various other phenomena. Such allopoems, as I will call them, should not be understood as being akin to Platonic forms, ideals towards which the given poems stand merely as defective simulacra (such an idea would imply a conceptual purity that I do not think obtains in Lee’s poetry). Nor are they rejectamenta akin to the negations of apophatic theology—dross removable from the ore so that we might get closer to the refulgent metal of the given poem. Instead, allopoems are potentially constitutive of the given poem, even if they are not perceivable. A somewhat crude analogy might be the biological phenotype—the observable set of traits exhibited by an organism. Just as beneath the phenotype there can be hidden traits in the genotype, so too do Lee’s poems harbor unexpressed possibilities. As the problematizing deictic markers in the poem “carries me” hint, we should be wary about suffusing a given space with too tangible an immanence: “Even if I enter here here does not come to me [ . . . ] here being at there there being at over there.” In other words, allopoetic features are part of a given poem even if they are elsewhere. Because of this condition, it is possible to describe the aggregate of allopoems within a given poem as the tertium by which the given poem becomes comparable to itself.
Logically speaking, the above claim is of course absurd, evoking Wittgenstein’s illustratively nonsensical question of “whether the good is more or less identical than the beautiful.”[2] But we are not dealing with logic here, and the apparent absurdity of an identicalness that is gradable into “more” or “less” does make sense in relation to Lee’s poems. It will be useful here to turn our attention to Walter Benjamin’s groundbreaking essay “The Task of the Translator,” a piece that has directly influenced Lee’s thought and writing. In an interview that I conducted with Lee for the Australian magazine Rabbit, she stated: “I am indebted to Benjamin for convincing me that my poetry contains not simply that which is already apparent in it but also a latent generativity that might appear only through translation.”[3] For Benjamin, the process of translation facilitates partial access to the realm of “pure language” that subtends the original text. This “pure language” is a kind of linguistic plenitude, wherein expression is not limited by the strictures of a particular language. Anyone who has spent a considerable amount of time translating and re-translating a poem from one language to another will know what Benjamin means when he says that it is “the task of the translator to release in his own language that pure language which is under the spell of another, to liberate the language imprisoned in a work in his re-creation of that work.”[4] Unquestionably, there are vectors and dimensions within texts that are not obviously present in the original, non-translated versions.
But while Benjamin is undoubtedly right in pointing out that translation can liberate caches of constitutive material from an original text, it could be argued both that his diagnosis doesn’t go far enough and that his nomenclature is in a certain regard misleading. My own sense, which I felt especially keenly while translating Lee’s poems, is that there is an essential impurity to literary language, one that would remain even if all possible languages were able somehow to resound simultaneously and to compensate for each other’s deficiencies. Paul Ricoeur can be a useful reagent to Benjamin’s thinking on this point. Ricoeur writes that “Every language’s struggle with the secret, the hidden, the mystery, the inexpressible is above all else the most entrenched incommunicable, initial untranslatable.”[5] As for Benjamin, there is for Ricoeur something “hidden” in every language; however, what is hidden is not something that might be theoretically explicable into an ultimate purity—because of the incredibly complex, contradictory, and context-dependent ways in which any given living language functions, “no universal language can succeed in reconstructing its infinite diversity.”[6] The upshot is that for understanding to take place, there must already be an “internal translation” that occurs within a specific linguistic community.[7] As the philosopher Richard Kearney elegantly puts it, we “are dealing with both an alterity residing outside the home language and an alterity residing within it.”[8] Yet while Ricoeur so perspicaciously lays all of this bare, he still describes the “initial untranslatable” as a product of the “multiplicity of languages.”[9] Can we not go further still and say, however oxymoronically, that the “initial translatable” is a product of language as such, something that would be immanent even in a sole, pre-Babelized, perfect language?
Here, we need only look at poetry itself. Good poetry is not ambiguous or equivocal by accident: ambiguity and equivocation are part of its very essence. This fact would surely still obtain if there were only a single, perfect language—unless of course poetry itself were somehow obviated by the perfection, rendered a mere vessel for the conveyance of information. While Benjamin describes the crude impulse to relay information as the “hallmark of bad translations,”[10] for Ricoeur, value judgments are more complicated:
The faithfulness/betrayal dilemma claims to be a practical dilemma because there is no absolute criterion of what would count as a good translation. This absolute criterion would be the same meaning, written somewhere, on top of and between the original text and the target text. This third text would be the bearer of the identical meaning, supposed to move from the first to the second. Hence, the paradox, concealed behind the practical dilemma between faithfulness and betrayal: a good translation can aim only at a supposed equivalence that is not founded on a demonstrable identity of meaning. An equivalence without identity.[11]
Throughout his writings on translation, Ricoeur regularly invokes the absence of a tertium (or, as above, “third text”) as evidence for an insuperable theoretical impasse between two languages. But if we believe that the “initial untranslatable” is immanent in language as such, without reference to a second language even being necessary, then what happens to this tertium? Where vital poetry is concerned, I would argue that the tertium is already subsumed in the poem itself as a product of the work’s inherent ambiguity and equivocation, and even if we cannot see this tertium, we can sense it. We can thus invoke Ricoeur’s “equivalence without identity” to ask to what degree a poetic text is equivalent to itself.
Lee’s collection is an exemplary text about which to ask the above question. Its title, with the absence of an object to which to tether the averred likeness, suggests that what characterizes the poetry therein is likeness tout court—a likeness that can thus also be reflexive. But of course, if a poem can be described as being like or just like itself, then it must also, in a certain regard, be unlike itself. If there is a latent generativity within a poem—if, as I have suggested, there are allopoems within it—then we must conclude that it is not simply through translation that comparison becomes possible; translation is simply the activity that, in making the latent apparent, best brings into relief the conditions for the comparison.
2 Requisite Apologia
However much a translator may feel compelled to explain or defend their decisions, it is seldom feasible to be exhaustive—it will not be possible here to talk about every interpolation, excision, or divergence in Just Like. I hope, therefore, that some broader principles might be extrapolated from the few examples that I invoke below. One thing that certainly warrants mention is my decision to translate the title 마치 as Just Like rather than As If. Although, context depending, both of these English phrases are equally sound translations of the Korean “마치,” the “Als Ob” of Lee’s gnomic preface may perhaps suggest (per standard glosses of the German phrase and the philosophical discourse that attaches to it) that As If would have been the more appropriate translation for this book. But of course, language tends to do its own thing, and the phrase “as if” has come to take on strong frequencies of sarcastic incredulity in colloquial English. And while I certainly believe that translators must be amenable to letting in new resonances during the process of translation, they should also be judicious as regards what gets admitted, and the resonance in question seems too shrilly incongruous for the context. What is defection in one particular is thus often fidelity in another.
It is also worth touching briefly upon the general tone of the translations. I have always read Lee’s poems as being somewhat clinical, even though they never assume the kind of rigid precision that this condition might imply. Certainly, unlike most lyric poetry, Lee’s poems are not subservient to an overriding anthropocentric impulse. All the same, they retain an organic suppleness that should not be jettisoned. The tone I favored for these translations is therefore slightly cold and detached while still, I hope, malleable enough to capture the proliferative energies of Lee’s poetry. In terms of how the punctuation mediates this tone, it was largely a case simply of following Lee’s suit: I mirrored Lee’s full stops exactly, and I more or less did the same with her commas, adding my own only when the absence of a comma would introduce a strongly equivocal note not present in the source poem. Certain lexical or syntactic features of my translations (e.g. a preponderance of non-contracted words or the deployment of slightly formal-sounding absolute constructions) were also put in the service of this tone. It is perhaps worth noting that I had the privilege of alighting on this tone in isolation. A natural impulse when beginning to translate a poet is to contrast one’s translations with existing ones, but I could find no English translations of Lee’s poems when I embarked on this project.[12] Though an absence of antecedent translations can in a certain respect be daunting, it also removes the risk of the poet’s voice being already cauterized by an “authoritative” precedent.
Especially important to mention are the twinned categories of error and defection, both of which should be borne in mind when approaching this (or any) poetic translation. It is almost certain that some mistakes will have crept into Just Like. To the captious critical eye, mistakes are often one of the main criteria by which a translation might be judged to be “deficient” or “bad.” The truth, though, is that mistakes can be found in the work of even the best and most vital translators—a critic who is inclined to find errors will certainly be able to do so. Just as the translator should be willing to forgo the hubris of imagining that they will at all times have a full purchase on accuracy, so the critic should be cognizant of the fact that being wrong is not always “wrong.” Don Mee Choi, one of the most important translator-thinkers of our time, has written the following about wrongness:
I think I was wrong, to begin with, because I was Korean, but when I first came to the States, people constantly tried to correct my English spelling and pronunciation. My British English was wrong because it was uttered from a mouth attached to an unexpected face, a wrong face. So naturally I have become intrigued with displaced things—things that are wrong. And translation is in a perpetual state of being wrong because it isn’t the original.[13]
Choi reminds us not only that wrongness is inevitable, but that sometimes we must get things “wrong” on purpose.[14] Whether unwitting or deliberate, wrongness is frequently a corollary of the distance between the source language and the target language—or rather, of the translator’s attempts to traverse this distance. For centuries, a key point of contention (usefully codified by Friedrich Schleiermacher, even though it certainly preceded him) has been whether translations should be alienating or naturalizing—that is, whether they should incline more towards the source language or to the target language:
Either the translator leaves the writer alone as much as possible and moves the reader toward the writer, or he leaves the reader alone as much as possible and moves the writer toward the reader.[15]
It isn’t simply a case of one of these approaches being more faithful than the other; as per some later nomenclature proffered by Eugene Nida, equivalence can be either formal or dynamic.[16] A formal equivalence may well instantiate a broad overlap (formally and semantically) with the source text but at the cost of sounding jarring in the target language. A dynamic equivalence may well (through looser, more localized language) come closer to eliciting a reader response that is similar to the reception-context of the source text, though it may also in the process sacrifice some of the formal and semantic intricacies at play. Neither of these choices will necessarily be ethically neutral: a translation that is excessively naturalizing risks a descent into parochialism, whereas a translation that is excessively alienating risks an unseemly exoticization of the source language.
The problems consequent to the above binarisms have inevitably meant that there have emerged translation or translation-adjacent praxes that value deviation as much as or more than equivalence. There are vernacular “transcreations” of Sanskrit classics; “feminist interventionist translations” that upend baked-in patriarchal assumptions; “translucinations” that intergraft the foreign and the domestic; “faithless translations” that take linguistic incommensurability as one of their constitutive conditions; and even “fake translations” that are not linked to any extant text by the writer to whom they are ascribed.[17] If the given contextual framing is right, there is no reason why any of these strategies of defection shouldn’t be considered legitimate. At the same time, the attempt to facilitate a faithful encounter with a poet should not be considered passé; the aim of conveying a poet into a new language (with sufficient fidelity that a reader who does not know the source language might still be capable in some measure of reading this poet) is a worthwhile one. Ultimately, the task that I set myself with this translation was to create a text that would be uncompromisingly like Lee’s book without ever purporting to be the same as it.
To the above end, one thing that I found useful to think about was the different ways in which Korean and English function. Korean is a far more morphologically rich language than English, and it is capable of economically encoding a wide array of nuances relating to register or mood. Such differences can sometimes be so extreme as to thwart notions of adequation a priori. An obvious example is the speech level system of Korean, whose honorifics and addressee-lowering forms do not travel at all felicitously into English. To my mind, the same can also be true of some of the language’s realis moods, its subtle interrogatives, and its miratives (verbal inflections that express surprise). Regarding these particulars, some considerable tonal loss is inevitable during translation. On the other hand, there are some aspects of Korean (such as the elided subject) that I think can carry over well from challenging Korean poetry into challenging English poetry, even if the elisions are not necessarily unnatural in Korean. From the opposite end, there are things that English has at its disposal that Korean doesn’t have: in particular, a comparative richness of articles and number declensions. For me, it was not simply a question of downplaying the Korean features while accentuating the English ones (or vice versa); I paid no particular fealty to an alienating or naturalizing impulse. Instead, I simply leaned into whichever particularities of either Korean or English I felt worked well for the poems, sometimes allowing features of the Korean to pass like phrasal or grammatical calques into the English, at other times allowing English-language particularities to suggest new reverberations or inflections in the original.
My approach has at times created a certain degree of noise that is not there (or is differently there) in the original poems. Consider the excerpt below from “four-lane road”:
The four-lane road spreads like a contagion. Spreads in front of the eyes. Glistening all the while in the sun. On the four-lane road are people wearing shorts between the legs of people putting up billboards and people frozen with billboards tar slides down. The four-lane road extends out and seems to have go firsted and seems to have don’t goed [ . . .]
In the fourth sentence, I strove to approximate the freneticism and breathlessness of the Korean original. However, doing so required that the image be torqued somewhat. In the original, the main ganglion of ambiguity occurs around the detail of “people putting up billboards and people frozen with billboards,” wherein Lee brilliantly foreshadows the denoted freezing-together via the suggestion of a compound noun (“people billboards”). This feat is achieved via a highly disruptive comma elision between “people” and “billboards,” and if we imaginatively reinsert a comma at this juncture we can cleave the compound noun and make the syntagm more readable. Because it is not possible to render such a compound into English while still leaving room for the line to be resolvable into sense, I instead opted to omit a conjunction (“people wearing shorts [and]”). The result, I hope, is a sentence that can parse grammatically while still allowing for a pictorial equivocation comparable to that of the original (even if the particular chains of ambiguity are different). Another instance of noise from the same excerpt can be seen in the “go firsted” and “don’t goed” formations, which are perhaps more protrusive than their Korean counterparts. In Lee’s poem, the sonic patterning allows the implication of reported speech to be resolved quite easily, despite the atypical grammar. Even so, there is technically no ascription of saying in Lee’s words; rather, there is a subsumption of uttered language into the form of a verb. Some might argue that such notes are too subtle in Korean to warrant amplification in English and that the better approach would be to mute or downplay them for a smoother translation. But because I wanted to retain the implication of an utterance denuded of actual speech—a nuance that felt appropriate, given the implied non-human agent of the utterance—the aural protrusion seemed to me an allopoetic resonance worth indulging.
Though such choices may be tendentious, this very tendentiousness raises an important question concerning the allopoetic: namely, if the specific configurations of a given language seem not to invite particular translation choices, does that necessarily mean that such choices are precluded? Is it legitimate to convey a subject elision into English when, in the Korean, the subject is either implied or is ambiguous in a way that is more endemic to the poetic conventions of the source language than to those of the target language? Is it okay for an English translator to take a highly protean approach to articles and grammatical numbers, even though the Korean poem is not—and indeed, cannot be—comparatively protean as regards these particulars? Questions like these return us to Benjamin’s thinking about the imprisoning dimensions of individual languages. Given that certain English solecisms such as “seems to have don’t goed” cannot be reliably encoded in Korean as probable translations, we might conclude that the Korean source poems are necessarily closed off to such solecisms. But such is far from the case if the mode is a consciously allopoetic one. Sometimes, the allopoetic choice might be justified via a mere cognitive inversion. For instance, we might ask how the English “seems to have don’t goed” would be translated into Korean. At other times, the license may come not from any local equivalence in the source language but rather from the broader ambiance or grain of the poetry (in “person doing gymnastics,” for instance, the article and number shifts were governed by the general nominative slipperiness of the poem). If some of my choices may seem in isolation to foreclose certain potentials, I hope that the opposite might be true when these choices are considered in their aggregation.
Taken on their own terms, Lee’s original poems and my translations of them will certainly exhibit local discrepancies. Some parts of my translations will congeal where there is not a corresponding congelation in the Korean, just as other parts of them will flow where the Korean is more static. But what I think Lee’s book shows us is that her poems—or at least her given poems—should not simply be taken on their own terms. As the critic Bo-Won Kang has insightfully written:
[D]espite all its difficulties, translation will not ruin Lee’s poetry, because the words and sentences in her poems are optimized in the first place not for preservation or conveyance but destruction. They attempt to lose direction, and translation will help them do so . . . One day we will read Lee Sumyeong’s poems translated back into Korean from their translated versions, without reference to the originals.[18]
I read Kang’s lines with a pleasurable recognition. I am content for there to be some loss or even some “destruction” in my translation of Lee’s poems since loss is an ineluctable condition of the poems themselves. If such is to a degree true of all poems, it is especially true of Lee’s poems. This very “optimization” was one of the things that emboldened me to undertake a serious translation of Lee's poetry in the first place. While the spurious notion of a definitive translation had already long been anathema to me, such distaste was not enough in itself. Before I could immerse myself diligently in the act of translation, my resistance to the definitive needed to be complemented by a feel for the allopoetic. I needed first to be shown the allopoem. I needed to be shown poetry that was just like itself—and thus, by extension, unlike itself. Such is what Lee’s book showed me. My hope for this translation is that it might be capable, in its way, of showing something similar.
