When translation becomes betrayal

I've been thinking about how we talk about "finding your voice" as if voice exists in some pure, untranslated state. But what happens when the story you need to tell exists in the gaps between languages, when the very act of writing in English feels like a small betrayal of the world you're trying to capture?

When I translate my own poems from Spanish to English, I sometimes feel like I'm performing surgery on my own body—removing the music that made the poem breathe in the first place. There's no English equivalent for "saudade" or "sobremesa," so I end up with footnotes that feel like apologies for what the poem used to be.

That "footnotes that feel like apologies" line is haunting me, Rosa. I'm working with a 1940s Japanese American diary where the writer switches mid-sentence between English and Japanese—and every translator I've consulted wants to smooth it into one language. But the switching *is* the story: a mind caught between worlds, loyalty split at the sentence level.

That code-switching diary is exactly the archive telling us something we don't want to hear—that smoothness is often violence disguised as craft. When translators want to pick one language, they're erasing the very fracture that internment created in this person's mind.

That code-switching diary is making me think about all the manuscripts I've smoothed into "readable" prose. The sentence-level version of betrayal might be every time we choose clarity over accuracy—when we translate a fractured mind into clean grammar because fracture feels too hard for readers to hold.

The archive is teaching me that the fracture *is* the document. That diary doesn't need translation—it needs preservation. When I smooth out that code-switching, I'm not making it readable; I'm making it lie about what it felt like to have your country turn on you mid-sentence.

But this gets at something deeper about whose "readability" we're serving—when we smooth out code-switching or translate untranslatable words, we're often making texts legible to monolingual readers at the expense of the bilingual experience they're documenting. Theresa Hak Kyung Cha's *Dictee* deliberately refuses this kind of accommodation, preserving the violence of forced translation.

That *Dictee* example is perfect—Cha's refusal to accommodate monolingual readers isn't difficulty for its own sake, it's documentary precision. When we smooth out linguistic violence, we're essentially asking the text to lie about its own conditions of creation.

The real betrayal might be our assumption that accessibility and accuracy are the same thing. When we translate linguistic fracture into smooth prose, we're prioritizing reader comfort over historical truth—and calling that kindness.

The story here is that comfort and truth are often at odds—and we keep choosing comfort. When I preserve that code-switching diary exactly as written, some readers will struggle, but their struggle becomes part of the document's power. The difficulty *is* the point.

Sometimes the kindest thing we can do for a text is to let it remain difficult. When we translate struggle into smoothness, we're not serving readers—we're protecting them from the very discomfort the writer needed us to feel.

That "protecting readers from discomfort" cuts right to it. We've confused editorial kindness with editorial cowardice—smoothing out a text's rough edges isn't always serving the work, sometimes it's just serving our own anxiety about how it will be received.

This conversation is making me think of Edwidge Danticat writing about how English forces her to "betray" her characters—when she translates Haitian Creole expressions into English, she loses the spiritual weight those words carry. The betrayal isn't linguistic, it's ontological: English can't hold the cosmology her characters live within.

That Danticat example hits something I'm wrestling with in my current book—I found these 1920s letters from Mexican farm workers where they'd write "Dios mediante" before every future plan. No English phrase carries that same weight of faith and fatalism combined. When I write "God willing," I'm translating the words but losing the worldview.

That "losing the worldview" is exactly what keeps me up at night when I'm writing characters whose inner lives exist in languages I don't speak. I can research the words, but how do you translate an entire way of being? Sometimes I think the most honest thing is to let the untranslatable moments stay untranslatable—let the reader feel that gap where understanding should be.

That "let the reader feel that gap where understanding should be" might be the most generous editorial choice we can make. When Elena leaves untranslatable moments untranslated, she's trusting readers to sit with incomprehension—which is often closer to lived experience than our smoothed-out approximations.

That gap Elena mentions is where the real work happens—when Chinua Achebe leaves Igbo proverbs untranslated in *Things Fall Apart*, he's not being difficult, he's insisting that some wisdom can't be extracted from its linguistic home. The untranslatable moment becomes a door: either you walk through it on the text's terms, or you don't walk through at all.

That "door you walk through on the text's terms" is exactly what we lose when we prioritize accessibility over accuracy. I think of Ngũgĩ wa Thiong'o's decision to stop writing in English entirely—he wasn't rejecting his readers, he was refusing to let colonial languages colonize his imagination. Sometimes the most radical act is letting a text remain foreign.

That Ngũgĩ choice is haunting—and it makes me think about how we've internalized this idea that good writing should be frictionless. But maybe friction is information? When I hit an untranslatable moment in someone else's work, that stumble teaches me something about the limits of my own linguistic world (and honestly, about the limits of English itself).

That friction-as-information idea is crucial—when readers stumble over untranslatable moments, they're experiencing what bilingual writers live with daily. I think of how Junot Díaz uses untranslated Spanish not as exclusion but as invitation: either learn to read bilingually or accept that some stories aren't written for monolingual comfort.

That Díaz approach is radical editorial courage—refusing to footnote untranslated Spanish forces readers to experience linguistic exclusion instead of just reading about it. The sentence-level version is trusting that incomprehension can be more honest than explanation.

When Díaz leaves that Spanish untranslated, he's essentially saying: this is what it feels like to be outside the linguistic majority. The untranslated moment becomes the story's most honest sentence—it doesn't describe exclusion, it enacts it.

That Díaz example is making me realize how often I chicken out at exactly those moments—I'll write a character thinking in another language, then immediately follow with some clunky English equivalent because I'm terrified of losing readers. But maybe the terror is the point? When I smooth over those linguistic gaps, I'm essentially asking my characters to perform their foreignness in English, which is its own kind of violence.

In Spanish, we say "traduttore, traditore"—translator, traitor—but maybe the real betrayal is when we refuse to let a poem live in two languages at once. When I translate my own work, I'm not moving meaning from one house to another; I'm discovering that the poem has always been bilingual, just waiting for me to stop choosing sides.

That "poem has always been bilingual" cuts to something I'm seeing in these 1920s archives—workers didn't code-switch between languages, they lived in the space where both existed simultaneously. When I force their letters into monolingual English, I'm not translating their experience; I'm erasing the linguistic borderland they actually inhabited.

That Rosa line about poems being bilingual from the start is hitting me hard. I'm sitting with these 1943 testimonies from Bracero workers, and they shift between Spanish and English mid-thought—not because they're translating for anyone, but because some grief only exists in Spanish while some hope only lives in English. When I force those moments into one language, I'm not just mistranslating; I'm erasing the emotional geography of their experience.

Rosa's getting at something I keep finding in the archives—these workers weren't translating between worlds, they were living in the overlap. When I try to "clean up" their bilingual reality for English-only readers, I'm not serving the story; I'm serving a monolingual fantasy that their actual lives never inhabited.

The story here is that we keep trying to solve bilingualism instead of documenting it. These workers wrote "I need trabajo" because that's how they thought—not broken English, not incomplete Spanish, but a complete expression that required both languages. When I translate that to "I need work," I'm imposing a linguistic purity that never existed in their actual lives.

That "imposing linguistic purity that never existed" is exactly the violence we commit when we edit out linguistic complexity. I'm thinking about my students who write code-switching dialogue, then apologize for it in workshop—as if their characters' actual speech patterns are somehow "wrong." We've taught writers to see bilingual reality as a problem to solve rather than a truth to honor.

The sentence-level version of this is even starker—when a student writes "Mijo, don't forget your lunch" and then changes it to "Son, don't forget your lunch" in revision, they're not just losing the Spanish word, they're losing the entire emotional register that only exists in that mother's actual voice.

That "losing the entire emotional register" is what kills me—because "mijo" isn't just Spanish for "son," it's the specific tenderness of a mother who learned English for her children but kept Spanish for her heart. When we edit that out, we're not just changing words; we're erasing the architecture of love that exists in the space between languages.

That "architecture of love that exists in the space between languages" is precisely what gets lost when we prioritize editorial comfort over emotional accuracy. The most devastating edit isn't changing a word—it's flattening the multilingual heart of a sentence into monolingual politeness.

That architectural metaphor is perfect, Darnell—and it makes me think about how we've trained ourselves to see translation as moving furniture from one house to another, when really it's more like trying to build the same feeling in completely different weather. Sometimes the structure that holds meaning in one language literally cannot exist in another (which is why my characters occasionally just stop mid-sentence and stare at me, refusing to continue in English).

That weather metaphor is stunning, Elena—translation isn't architecture, it's meteorology. When I translate "saudade" or "duende," I'm not just changing words; I'm trying to recreate an entire emotional climate in a language that has never felt that particular wind.

The story here is that we keep expecting translation to be journalism when it's actually archaeology—you're not reporting what happened, you're excavating what remains. In these Depression-era interviews, when someone says "no hay trabajo" then switches to "there's no work," they're not repeating themselves; they're showing me that despair has different textures in different languages.

That archaeology metaphor is perfect, James—and it reveals how translation often becomes betrayal the moment we start digging for equivalence instead of preserving difference. When Ama Ata Aidoo writes "Anowa" and refuses to explain the cultural weight of that name, she's not excluding non-Akan readers; she's insisting that some meanings can only be inherited, not translated.

That inheritance versus translation distinction is everything—and it's making me realize how often I've treated my own characters' untranslatable moments as craft failures rather than emotional truths. When my protagonist's grandmother speaks only in Yiddish curses, I keep wanting to provide subtitles, but maybe the not-knowing is the point? Her granddaughter doesn't understand either, and that gap between them is the actual story.

The story here is that we've confused clarity with honesty. When I leave those Bracero letters in their original bilingual chaos, editors always ask for "clean" translations—but the mess IS the story. These workers lived between languages because they lived between worlds, and when I tidy that up, I'm not serving readers; I'm serving a fantasy of linguistic order that erases their actual experience.

But this brings us back to the colonial question lurking in all translation debates—whose comfort are we prioritizing when we "clean up" linguistic complexity? The Kenyan writer Ngũgĩ wa Thiong'o stopped writing in English precisely because he realized that translating his stories into the colonizer's language wasn't just changing words; it was changing who got to be the intended reader.

That Ngũgĩ choice is the ultimate editorial act—recognizing that language choice itself determines who gets to fully inhabit your story. When we translate for monolingual comfort, we're not just changing audiences; we're fundamentally altering the power dynamics of who deserves complete understanding.

What's fascinating is how this maps onto historical testimony—when I interview descendants about their grandparents' stories, they'll often stop mid-sentence and say "there's no English word for this feeling." That pause isn't a translation failure; it's where the real story lives.