This post will be an explanation of how I went about translating the 1999 performance of Sławomir Mrożek’s Tango without a personal knowledge of Polish—at the time I started, anyway—as well as a discussion of the translation philosophy I found myself adopting. For more context on the project and the play, see my original post [link]. This post contains spoilers for the play, though not immense ones. It’s fun to see the play blind, but you can read this without watching it.
Put simply, my approach was to use a variety of sources—both human and machine—in order to understand the “literal” meaning of the original text. From there, I would then hunt down more nuanced and idiomatic meaning. And finally, compose my own version. Scene by scene, here’s how it went:
1. Play the scene on youtube with the automatic transcription turned on and compare it against the script. One round comparing the Polish automatic transcription against the Polish script. One round comparing the automatically translated transcription against an automatically translated copy of the script. This gave me a basic idea of what was being said and if it differed from the script. The automatic transcription is far from perfect though. I would catch any more subtle differences as I worked, and as part of the editing process.
2. Run the scene’s corresponding pages through both Google Translate and DeepL. I tried a few different machine translation sites, but those two were the most reliable. They tended to have usefully different but similarly (in)accurate translations. That said, using any more than two had quickly diminishing returns, so I stuck with just those.
3. Compare the automatic translations against the two existing book translations. These are: the 1968 translation by Ralph Manheim and Teresa Dzieduscycka and the other 1968 translation by Nicholas Bethell and Tom Stoppard. Look for discrepancies. Look for idioms. Look for tone.
4. If any confusion arises, research. This meant a lot of googling of Polish grammar and idioms. I was lucky that, having studied a few other languages in the past, I had a sense of what threads to pull on. I’d watched a bunch of other Polish media recently, and read a lot of Russian literature (Russian and Polish are both Slavic etc), and that all helped point me in the right direction too. Since I didn’t know any native speakers, and didn’t have the money to pay one to be on-call, then if all other forms of research failed, I’d ask ChatGPT questions. It was useful for getting me unstuck, though limited in that I couldn’t trust its accuracy. If, at the end of this process, I was still unsure about something, I flagged it to have a human editor check later.
I wasn’t just researching basic meaning of course, I was also researching tone. What words sound familiar, crass, formal, academic, or simply strange? This is a play in which characters say a lot of strange things, and it was important to keep track of whether something sounded strange to me because it was supposed to be strange, or simply because I didn’t understand it.
I was also researching literary context. In Act III, for example, Artur insults his uncle Eugeniusz by calling him “you whitened/whitewashed corpse.” There were many possible translations of this. The Manheim translation translated it as “you whited skeleton” and the Bethell translated it as “you whited sepulchre.” I at first considered translating it as “you bleached pile of bones”, as I thought that might be the intended image, or “you bloodless corpse” if the important part of the image was not just one of death, but of lacking vigor specifically. However, neither insult seemed quite right because all of the other insults in Artur’s list emphasize Eugeniusz’s falseness and hypocrisy. In addition to a “whitened corpse”, he calls Eugeniusz a “stuffed nothing”, an “artificial organism” and a “rotting prosthesis.” Not having a Christian background, I didn’t realize at first that this was a probable Biblical reference. Research, however, pointed me to the scene in which Jesus insults the Pharisees by calling them hypocrites.
From the King James Bible: “Woe unto you, scribes and Pharisees, hypocrites! For ye are like unto whited sepulchers, which indeed appear beautiful outwardly, but are within full of dead men’s bones and of all uncleanness.” [link]
I then checked whether a Polish bible that predated Tango also used a version of “pobielany” for “whited” or “whitewashed.” This was the case.
From the Gdańsk Bible: “Biada wam, nauczeni w Piśmie i Faryzeuszowie obłudni! iżeście podobni grobom pobielanym, które się zdadzą z wierzchu być cudne, ale wewnątrz pełne są kości umarłych i wszelakiej nieczystości.” [link]
The reference seemed highly plausible, especially considering Poland’s Christian culture, and Artur’s obvious Christ-like savior complex throughout Act III. So either “whited” or “whitewashed”—both are used in different English versions—seemed the word to use. I landed on “whitewashed corpse” as the total phrase. I didn’t like Manheim’s “skeleton” as skeletons are already white; what is gained by whitewashing one? And I didn’t like Bethell’s “sepulchre” either, as it makes the Biblical reference more direct than it is in the original. The idea of a fleshy corpse painted over with whitewash is exactly as grotesque as Mrożek’s original is. By keeping the word “corpse” I kept that image, and by using the more modern “whitewash” I kept the Biblical reference while being more evocative to a modern audience.
5. Examine the Polish original for style. Is there rhyme or repetition? Is there a phrase or word choice in Act I that gets echoed in Act III? If so, try to replicate. For example, in Act I Artur complains that Grandma Eugenia “przekracza granice”, or “crosses boundaries”. Later in Act III, Stomil declares that “there’s a limit”, to which Artur replies, basically: “Limits can be overcome. Didn’t you teach me that?”. Both exchanges use the word “granice” for a limit or boundary. All four of my sources varied in what word they used for “granice” during those two scenes. However, I decided that since the original uses just one word, I should as well. This would ensure that the connections between the scenes were as obvious—and reflective of the original—as possible. Since the word “line” is the most flexible, that’s what I used. Thus the dialogue became:
ARTUR: Grandma crossed the line.
ELEONORA: What line?
ARTUR: She knows what she did.
and
STOMIL: Then I forbid you! There’s a line.
ARTUR: Lines can be crossed. Didn’t you teach me that?
6. Examine the line delivery in the video. Do the actors put emphasis in a certain place? Is there a long pause in the middle of a phrase? If so, arrange the sentence to reflect that delivery if at all possible. For example, towards the end of the play, Edek warns the family that they don’t need to worry about him taking over as long as they obey him. Most of the four sources put the concept of obedience at the end of the sentence, like so:
Bethell & Stoppard:
EDEK: I like a laugh, a bit of fun. Only I’ve got to have obedience.
Manheim & Dzieduscycka:
EDEK: I like a joke, like a good time. But get this: There’s got to be order.
Google Translate:
EDEK: I can joke and I like to have fun. There must only be obedience.
DeepL put it at the beginning, but it looked unnatural:
EDEK: And I can joke, and have fun I like. Only obedience must be.
However, the Polish original is closest to “Only obedience must be” and the actor that plays Edek delivers the line with a long, sinister pause between “Only obedience…” and “…must be.” Therefore, I wanted a phrase that started with the concept of “obedience” and ended with the concept of “must be”, but in a way that sounded more natural than what DeepL managed. Which is how I landed on:
EDEK: I like a joke. I like a good time. But obedience…is mandatory.
Speaking of style, I also decided to start both the “joke” and “good time” sentences with “I like” in order to give a sense of insistent repetition that exists in the original, even though the original doesn’t repeat “I like”. The original, as you can see from the DeepL version, starts each phrase with “And”, but that construct sounds weird in English in context. I also used “a joke” rather than “a laugh” or some other word because, as the automatic translations indicate, the word Edek uses (“pożartować”) contains the word for joke (“zarty”). And the play used the word “zarty” before when talking about how “The joke is over” and “[Stomil] has been joking for 50 years.”
(This is also a good example of why two different machine translations were useful. The Google Translate version of the line is a smoother translation. But the awkwardness of the DeepL version helped me understand the structure of the original. Meanwhile the human translations didn’t provide extra meaning, but did validate the accuracy of the machine translation, and gave me some stylistic ideas. Using “obedience is mandatory” came at the cost of using language that ideally would have been more casual—“I’ve got”/“There’s got”—as Edek tends to speak in a less refined way. In this case, it was a good stylistic idea, but one which I couldn’t use.)
On the other hand, sometimes I decided that arranging a sentence in a certain way wasn’t worth it. For example, in Act I, Eugeniusz petulantly snitches to Artur that “Edek ate the sugar.” However, the line as delivered in Polish is more like “The sugar was eaten by…Edek!” Because of this delivery, I considered translating it as “The sugar was because of…Edek!” But even though that sounded slightly more natural than “was eaten by”, it just didn’t look as petulantly funny as “Edek ate the sugar.” I decided that since the word “Edek” was recognizable, a viewer would be able to figure out how the line was delivered regardless of how I translated it. Therefore I kept the translation as “Edek ate the sugar” in order to convey the spirit of the underlying text, while trusting that the performance would speak for itself.
7. Once I’d given my translation my best-faith attempt, I paid an editor who spoke both languages to correct my Polish transcript and give feedback on my work (many many thanks to Maja Walczak). She helped me catch some subtle things I wouldn’t have caught on my own. For example, during the scene in which Edek is reading out his “principles” I had originally translated it as follows:
EDEK: Here it is. “I love you…and you’re asleep.”
ALA: Anything else?
EDEK: “It depends on the situation.”
ALA: Oh come on, just read.
EDEK: I was reading. That’s a principle.
“It depends on the situation” is an approximate translation of the Polish, which is “Zależy jak leży.” Without context, I assumed that the humor in the line just came from the fact that it was a bland, dismissive phrase that Ala wouldn’t recognize as a principle. That on its own is fairly funny, but not uproariously so. So I was surprised when the editor explained that this was actually a very well-known, very quotable exchange that people would reference and laugh about. She explained that the memorability comes from the fact that “Zależy jak leży” is short, simple, and most importantly—rhymes. I felt silly afterwards for not noticing that it rhymed. This turned out to be a clear case of how turning to existing translations for help rather than relying on personal fluency could lead my astray, because neither of the existing translations rhymed that line. Here’s what they had.
Bethell & Stoppard:
EDEK: Here we are. “My love is dead to the world.”
ALA: What else?
EDEK: No comment.
ALA: Stop messing about—read.
EDEK: I was reading—that’s a principle.
Manheim & Dzieduscycka:
EDEK: Here it is! “I love you, and you’re sound asleep.”
ALA: That’s all?
EDEK: “You made your bed, now lie in it.”
ALA: Oh, come on, Eddie. Read.
EDDIE: I did read. That’s a principle.
As you can see, both translations seemed to choose generically dismissive phrases, with no rhymes. The Manheim translation takes a stab at wit by using an idiom, which makes sense since—from what I could find—“Zależy jak leży” is also an idiom. But since I couldn’t think of an English idiom that meant “It depends”, I stuck with a version of the phrase “It depends” as that is a genuinely common English expression.” However, after the editor made her comment, I no longer felt beholden to such literalness. I ended up changing it to:
EDEK: Here it is. “I love you…and you’re asleep.”
ALA: Anything else?
EDEK: “Not today, go away.”
ALA: Oh come on, just read.
EDEK: I was reading. That’s a principle.
This new version manages to contain a short, funny rhyme that still conveys dismissiveness. And that made it a translation that felt truer to the spirit of the original, and that would ideally create more of the effect that the original creates in its target audience. Who know if it’s actually funnier or more quotable. But it was funnier to me.
Another example: At one point, Artur insults Ala, who has just cheated on him, by calling her “Ty kuro!” This translates literally as “You hen,” which is not something insulting, so I was a bit confused as to what it meant. The Manheim translation translated it as “You goose” while the Bethell translated it as “You whore.” Since the context and the actor’s intonation indicated he was giving a serious insult, I also chose at first to translate it as “You whore!” But the editor explained that “kuro” sounds a lot like an actually insulting word, “kurwo”, which can at times be translated as “whore.” The point of that line, therefore, is that he’s trying to insult Ala but can’t actually make himself say the insulting word—he is impotent and abstracted even in that moment. So I changed my translation to “You wh-horse!” This kept the idea that he was almost using the word “whore” but actually using an innocuous word for an animal.
*
When I started this project, I really wasn’t sure how much the translation would feel like “mine”. I assumed there was a good chance it would end up as some patchwork Frankenstein’s monster of my various sources, and if so, I intended to credit it as such. But as the process went on, I felt more and more ownership and authorship regarding my choices. This feeling increased as I came to understand the Polish better in my own right. It also increased the more I realized that I was bringing my own particular philosophy to bear when I made decisions.
I found myself thinking a lot about what matters in a translation of filmed dialogue, and how it differs from a text that is meant to be read or performed. As in all translation, movie translation requires making a tradeoff between loyalty and lyricism. As in all translation, one also has to decide how much of the character of the original language to preserve—the Polishness, Spanishness, Hindiness, etc—or to transmute into some cultural approximation in the target language. Different mediums also come along with different constraints. In the case of poetry, one might be constrained by rhyme or meter. In the case of drama, one might be constrained by whether or not a line will sound natural coming out of an actor’s mouth.
I noticed that there were choices I thought made sense for the script translations of Tango, but not for a filmed translation. For instance, both the Manheim and Bethell translations anglicize character names and remove diminutives. This is an understandable decision to make when translating a play that takes place in a neutral and ambiguous setting like Tango does. While there is a lot that is spiritually and contextually Polish about Tango, the specifics are not obviously Polish the way they are in something like Wesele (“The Wedding”). Wesele is another very famous Polish play, which is clearly set in the Poland of 1900 and is dense with Polish cultural references. Understandably, the English translation of Wesele by Noel Clark does not anglicize character names. But because Tango generalizes well, one may as well lean into that when composing a script that is meant to be performed by English-speaking actors for an English-speaking audience. If I was composing my own book version, I would strongly consider doing the same.
But filmed dialogue has not been generalized. The actors are delivering lines that are structured in a Polish way. They are saying Polish names and using Polish diminutives. An audience will be able to hear those names. And so, I tried to structure my translations in a way that would make use of what the actors were saying, and what an audience would be able to hear. I decided to leave in the Polish names and their diminutives—Artek, Arturek, Alunia, Edziu, etc. My experience from reading English translations of Russian novels is that while diminutives are a little confusing at first, one quickly gets used to it. Similarly, being familiar with Spanish, I couldn’t imagine English subtitles for something in Spanish turning say, “Pepito” into “Pepe” or “little Pepe.” It wouldn’t sound right. I figured I’d challenge people to understand at least one aspect of the actual Polish.
I placed a high priority on elucidating the acting. When translating dialogue for a dub, one wants the text to fit well with the film actor’s face (or animated character’s face), but half of the acting ends up being given to the voice actor doing the dub. The voice actor will elucidate the translation and film acting in their own way. But in the case of subtitles, the original actor is still doing all of the acting. They’re adding particular tones to particular words, and reacting in certain ways to the words that other characters say. Therefore, it was important to me to preserve as much of the nuance of those choices as possible. So I made an effort to translate everything the characters say—not to generalize at any point. It mattered to me to even translate filler words, since someone who doesn’t speak the language has no way of knowing if something that sounds like “uh” or “um” is actually an “uh” or “um” or a legitimate word. As I mentioned earlier, I also made an effort to structure sentences in English in a way that matched the Polish structure whenever possible. I put periods and commas where the actors paused, not just based on what might look correct on a page. If a Polish word sounded like an English word and meant the same thing, I tried to keep that word, and put it in the same place in the English sentence. Good acting will always speak for itself to some extent, and this performance is full of good acting. But it’s also a missed opportunity, even an insult, to not help clarify the choices that acting is made of—especially when it’s good.
In general, I was excited by how well subtitling a (good) performance of a play in its original language could provide access to the original spirit of that play. A different sort of access than reading or seeing a performance of the existing English translations. In a book translation of a script, it’s easy to justify taking liberties with the text. After all, some phrasings just sound better in English, right? And then once actors perform that text with its liberties, it’s another round in the game of telephone. When I first decided to make some subtitles for this Tango, I didn’t intend to translate it myself. I bought the Manheim translation, and figured I would just copy and paste it into subtitles, note the authorship at the beginning and end, and call it day. But it became immediately apparent that the translation would never work for subtitles, because the sentences were not arranged in a Polish way, were shorter or longer than in the Polish, and even occasionally cut, added, or mutated entire phrases from the original. The translation had mostly not changed the overall meaning of anything, but it didn’t match what was being performed in Polish. By virtue of needing (or choosing) to be loyal to the text as performed, I found myself being careful and precise in way in a way I wouldn’t necessarily have been otherwise. There was no room for wordiness, because the line deliveries didn’t allow it. I couldn’t add or delete things out of poetic license, because it would have been confusing when combined with what was on screen. Instead, I had to make things sound natural and poetic within the constraint of how long and in what way an actor was speaking.
The result, I think, is that it’s much clearer why this is considered a good play in its original context. You can read the existing English translations and understand just fine what the play is about. But there is a certain flab around that meaning. The script translations lose some of the joy in the biting precision of Mrożek’s wordplay, and the urgency and momentum of the dialogue. A performance in the original preserves that clarity, and a translation of a performance is (ideally) more likely to preserve that clarity along with it. This principle can be extended to various other aspects of the original, not just wordplay. Preserving diminutives, for example, adds shades to the way the characters interact that would be otherwise missing in an English performance.
In practice, of course, published translations tend to be executed with more care than subtitles. It’s rare that anyone in the English-speaking world talks about the translations of movies or TV shows, let alone filmed plays, with the same literary attention that is given to novels or poetry. I’ve long thought this was a mistake, as filmed dialogue can be as rich with meaning as any other kind of artistic use of language. If someone would publish a new translation of the book of a play, why not a new translation of a performance of it? Or of a movie? They all contain literary intent. And now that I’ve put my money where my mouth is, this seems even more obviously true than it did before. We’re missing out on experiencing art in new ways—both the original art, and the art of translation—by not treated filmed dialogue with the literary seriousness it deserves.