A translation PhD student's runaway thoughts on language, translation and translated books.
Don't wanna be here? Send us removal request.
Text
Diary entry 16: ChatGPT and I have a talk
Firstly, it's been a while and although I have always wanted this blog to serve as a record of my PhD journey, as well as my runaway thoughts, I just don't feel like writing an update of where I'm at right now.
I want to talk about ChatGPT instead. My friend introduced me about a week ago and I have been obsessively asking ChatGPT anything and everything I can think of, seeing where the limitations are (besides the obvious: it's only up-to-date until September 2021 and doesn't learn).
Today, after seeing how best to communicate with it, finding the commonalities in its answers, I felt ready to tackle translation together.
I have been sporadically translating the Ukrainian novel Your M. I gave ChatGPT the first 200 words of the novel to translate, then compared that to my own translation and Google Translate. Of the two machine translations, ChatGPT was way better (GT got the tone weird and mistranslated some words). Then, I had a little interview with ChatGPT about its capabilities. Below is the interview, edited for the sake of not reading ChatGPT say 'as a language AI model I don't XXX' over and over again (I get it, you don't learn or remember, relax).
AN INTERVIEW WITH CHATGPT ON THE TOPIC OF TRANSLATION AND ITS PROGRAMMING
IB: How does your translation programming work?
CGPT: As a language model, I have been trained on a large dataset of text in multiple languages. I use statistical patterns and patterns in the data to generate translations. I do not have a specific "translation programming" per se, but rather I generate translations based on the patterns I have learned from the training data.
When translating text, I analyze the input text and generate the corresponding output text in the target language based on the patterns and structures I have learned during training. However, it's important to note that as a language model, I do not have a deep understanding of the meaning of the text or the context in which it is used. My translations are generated based on patterns and correlations learned from the training data, and may not always capture the nuances, cultural references, or idiomatic expressions accurately. Therefore, it's always recommended to have translations reviewed by a professional translator for accuracy and quality assurance.
IB: Do you translate all entirely originally or do you congregate translations of a text that are already out there on the internet?
CGPT: I do not directly access or gather translations from the internet. I generate translations based on the patterns and structures learned from a large dataset of text during my training. I do not search or retrieve translations from external sources or the internet in real-time.
However, it's important to note that the training data I am based on may include publicly available text from the internet, which can encompass a wide range of translations, among other types of text. During training, I learn from the statistical patterns and correlations present in this data, but I do not have access to or actively search for translations from specific sources during the translation process.
It's also important to mention that as an AI language model, I do not have the ability to verify the accuracy or reliability of the translations I generate. It's always recommended to consult professional human translators for accurate and reliable translations, especially for important or sensitive content.
IB: What would you say are the advantages and disadvantages of using your programming to translate literature versus a human translator?
CGPT: Advantages of using an AI language model for translating literature:
Speed: AI language models can process text quickly and generate translations in a relatively short amount of time, which can be useful for fast turnaround times or large volumes of content.
Cost-effectiveness: AI language models can be a more cost-effective option for certain translation tasks compared to human translators, especially for shorter texts or less complex content.
Consistency: AI language models can provide consistent translations, as they follow predefined patterns and rules, which can be helpful for maintaining consistency in terminology or style across a large body of text.
Disadvantages of using an AI language model for translating literature:
Contextual understanding: AI language models may struggle with understanding the nuances of context, cultural references, and literary elements, which can result in inaccuracies or loss of meaning in the translated text.
Language limitations: AI language models may have limitations in handling certain language pairs or translating complex or rare language structures, idioms, or dialects.
Human touch: Literature often requires creativity, interpretation, and subjective judgment, which may be challenging for AI language models to replicate as they lack human emotions, experiences, and cultural insights that contribute to literary translation.
Quality assurance: While AI language models can generate translations quickly, they may require additional quality assurance and editing by human translators to ensure accuracy, readability, and style.
In summary, AI language models can offer advantages in terms of speed and cost-effectiveness for certain translation tasks, but they may have limitations in terms of contextual understanding, language complexity, human touch, and quality assurance, especially for literature translation where artistic and cultural nuances are crucial. Human translators, with their cultural understanding, linguistic expertise, and creative interpretation skills, remain invaluable for accurate and nuanced translations of literature.
IB: Will AI ever be able to be programmed to have this 'human touch' or understand cultural context?
CGPT: While AI language models have made significant advancements in recent years, achieving a truly human touch or fully understanding cultural context remains a challenging task. Human touch in literature, including creative interpretation, emotion, and cultural insights, often involves subjective judgment, intuition, and contextual understanding that are inherently human traits.
AI language models, being based on statistical patterns and trained on vast amounts of data, may struggle with capturing the intricacies of human touch and cultural context, which are highly nuanced, ever-evolving, and often subjective. While advancements in natural language processing, machine learning, and AI technologies may continue to improve the performance of AI language models, achieving a truly human touch or understanding cultural context at the same level as human translators remains uncertain.
However, it's worth noting that AI language models can still be valuable tools in the translation process, assisting human translators by providing suggestions, aiding in terminology management, or helping with initial drafts. Human translators can leverage the strengths of AI language models while also applying their expertise, creativity, and cultural understanding to ensure high-quality translations that capture the nuances of literature and cultural context effectively. Collaboration between AI and human translators can result in better translation outcomes, combining the efficiency of AI with the human touch of professional translators.
IB: What do you mean by 'Collaboration between AI and human translators can result in better translation outcomes, combining the efficiency of AI with the human touch of professional translators'? What are you referring to as 'better'?
CGPT: By "better" in the context of collaboration between AI and human translators, I refer to translations that are accurate, fluent, and capture the intended meaning and nuances of the source text while also incorporating the creative and cultural insights that human translators bring to the table. This means that the translation should not only be linguistically correct but also convey the intended emotions, style, tone, and cultural references of the original text, in a way that resonates with the target audience.
AI language models can offer efficiency in tasks such as terminology management, sentence structure suggestions, and initial drafts. They can process and generate translations at a faster rate than human translators, helping to streamline the translation process and save time.
However, human translators, with their deep cultural understanding, contextual knowledge, and creativity, can provide the human touch that is often required in literature and other creative works. They can make subjective decisions, interpret nuances, and adapt the translation to resonate with the target audience, taking into consideration cultural context, local idioms, and other cultural references that may not be readily apparent to an AI language model.
The collaboration between AI and human translators can thus result in translations that are both efficient and culturally accurate, combining the strengths of AI in terms of processing power and speed with the human touch of professional translators, resulting in high-quality translations that effectively convey the intended meaning, style, tone, and cultural nuances of the original text.
IB: Can you define fluency in this context we're discussing?
CGPT: Fluency in the context of translation refers to the smoothness and naturalness of the translated text in the target language. A fluent translation reads as if it were originally written in the target language, without any awkwardness or choppiness that may arise from literal or word-for-word translations.
In literary translation, fluency goes beyond mere linguistic accuracy and includes capturing the intended style, tone, and rhythm of the original text. It involves conveying the emotions, nuances, and cultural references in a way that resonates with the target audience, evoking the same impact and effect as the original work.
Fluency also encompasses the ability to adapt the translation to the cultural and linguistic norms of the target audience, ensuring that the translation is not only accurate in terms of meaning, but also culturally relevant and appropriate. It requires understanding the cultural nuances, idioms, and references that may not have a direct equivalent in the target language, and finding creative ways to convey them effectively in the translation.
In summary, fluency in translation means producing a translated text that reads smoothly, naturally, and effectively in the target language, capturing not only the meaning but also the style, tone, emotion, and cultural nuances of the original work.
IB: How does domestication and foreignisation play into this definition of fluency? Does your translation programming sway to either method?
CGPT: Domestication and foreignisation are two different approaches in translation that can impact the fluency of the translated text.
Domestication, also known as localization or target-oriented translation, is an approach where the translation is adapted to the culture and language of the target audience. It prioritizes making the translation sound natural and familiar to the target audience, even if it deviates from the source text in terms of cultural references, idioms, or style. Domestication aims to produce a fluent and culturally relevant translation that resonates with the target audience.
Foreignisation, also known as source-oriented translation, is an approach where the translation retains more of the characteristics of the source text, even if it may sound less natural or familiar to the target audience. Foreignisation aims to preserve the uniqueness and cultural specificity of the source text, even if it requires introducing foreign elements into the translation. This approach may be used when the translator wants to maintain the original flavor, style, or tone of the source text, or when the cultural differences between the source and target languages are considered important.
My translation programming is based on patterns learned from vast amounts of text data, and I do not have a specific bias towards domestication or foreignisation. The approach I use in translation depends on the data and patterns I've learned during training. However, I can adapt to different translation approaches based on the input and guidance provided by the user or human collaborators during the translation process.
It's worth noting that finding the right balance between domestication and foreignisation depends on various factors, such as the purpose of the translation, the preferences of the target audience, the genre of the text, and the specific requirements of the translation project. It's up to the human translator or the collaborative efforts of human translators and AI to decide on the appropriate approach for a given translation task to achieve the desired level of fluency in the translated text.
IB: If I ask you to translate a text leaning on foreignisation, would you be able to do it? Or can you only translate in one way?
CGPT: I can adapt my translation output based on the input and guidance provided by the user or human collaborators. If you specifically request a translation that leans towards foreignisation, I can attempt to generate a translation that retains more of the characteristics of the source text and preserves the uniqueness and cultural specificity of the original, even if it may sound less natural or familiar to the target audience.
However, it's important to note that my translations are generated based on patterns learned from vast amounts of text data during training, and I do not have a specific bias towards domestication or foreignisation. My translations are not perfect and may not fully replicate the nuanced and creative decision-making abilities of a human translator. Human translators have subjective judgment, cultural understanding, and contextual awareness that can impact their translation choices. Therefore, while I can attempt to generate translations leaning towards foreignisation, the output may still be limited to the patterns and data I've learned during training, and it's always recommended to have translations reviewed and edited by professional human translators for accuracy, fluency, and cultural appropriateness.
IB: Okay, can you translate the following with more foreignisation than you did previously
I gave the same section of text as before the interview when I was comparing translations. Chat GPT translated almost exactly the same as before, save for a few seemingly random changes:
Seatmate → neighbour
Window → glass
Everything was completely different → everything was so different
I hope it comes true → Let it come true
Perhaps this text didn’t have enough truly colloquial language (moreso TOV) to focus on foreignisation.
IB: Do you have the same capabilities in all languages, or are some stronger than others?
CGPT: I am trained on a large corpus of text data in multiple languages, but my capabilities may vary depending on the language. My primary language is English, and I am generally most proficient in generating text in English. However, I also have capabilities to generate text in other languages, including but not limited to Spanish, French, German, Chinese, and many more.
While I can generate text in various languages, my proficiency may be higher in some languages compared to others, depending on the amount and quality of training data available for that language. Additionally, my responses in some languages may be more fluent and accurate than others, as they are influenced by the training data and patterns observed during training. It's always best to verify the accuracy of my responses in languages other than English with a native speaker or a reliable language source.
IB: Does that mean that you would create a more accurate translation out of French than out of Ukrainian? These are just example languages
CGPT: My translation accuracy depends on various factors, including the amount and quality of training data available for each language. Generally, if I have been trained on a larger and more diverse corpus of text data in a particular language, my translation accuracy for that language may be higher compared to languages with less available training data.
However, it's important to note that my translations are generated based on statistical patterns learned from data, and they may not always be perfect or completely accurate, regardless of the language. Translation accuracy can also be affected by the complexity of the text, domain-specific terminology, idiomatic expressions, and other linguistic nuances that can vary across languages.
It's always recommended to verify the accuracy of machine translations, including those generated by AI language models like me, with human translators or other reliable language sources, especially for important or sensitive content.
0 notes
Text
Diary entry 15: the beauty of things not translated
Often I feel like all my diary entries need to start or finish with Hugh Grant's voice saying 'If you look for it, I’ve got a sneaky feeling you’ll find that translation, actually, is all around.'
What I mean by that is so often I will be doing something unrelated to translation or my studies in a direct sense, and translation will come up. It's probably because much of my choice of entertainment somehow related to foreign language and we can't escape translation when it comes to foreign language, which I guess is the point in all of this.
The other night I went to see Dorr-e Dari at the Arts Center, which was a performance piece about love and Persian poetry. It was performed half in English and half in Persian. Sometimes a performer would be reciting a poem in Persian and the English translation would be being projected behind them. Other times a performer would be telling a story in Persian and another would be interpreting it into English in real time. Sometimes they would be speaking entirely in English and would be sprinkling Persian words here and there, leaving them untranslated. Sometimes the performers would start to chat to each other in Persian, seemingly chatting through what was to come next before the launch into it. They had also invited a poet to video call in and recite some poetry they had written in Persian - this poetry was not translated.
Doing so much work and thinking about translation, language, meaning and understanding at the moment has made me have a deep love for things that aren't translated. I think the biggest fear that people have around translated literature is it forces us to confront the fact that we can't know everything. Theoretically you could learn every language out there (though, actually, doing the maths in my head now I think perhaps you couldn't).
As people we don't like not knowing things, and we don't like other people knowing things we don't even less.
As a researcher (different to 'people'??) I am finding great comfort in not knowing things. I loved the moments that weren't translated in Dorr-e Dari, they probably weren't for me. They were for the people in the audience who spoke Persian, who laughed at the un-translated parts because they understood, who recited the poetry along with the performers.
When I think about translation I imagine a utopia, where every language and culture is represented and cherished, where everyone is okay with not understanding sometimes, where we feel grateful when we are allowed to understand and invited in on the joke through translation.
0 notes
Text
Diary entry 14: hehe freaking out a bit
A while ago I wrote about learning about my own practice and workflow, and having to learn to work with my brain and not against my brain. In that context, it was about how I can't translate for longer than an hour at a time. Now, I am learning more about myself: I need time to absolutely feel despair before rolling my sleeves up.
I will need to undergo my Confirmation of Candidature sometime before February next year, preferably this year, perhaps even October.
Do I feel ready for October? You have three guesses.
Today on the tram in to my little study sess (with my study buddy who is studying for his actuarial exams) I was wondering this is the time to pack it all in, you know, while I'm 'ahead'.
It's not, I have to try to pass the CoC but I am freaking out a little.
I haven't been doing much PhD reflection on here because so far it's just been me in my own little world creatively meandering through the world of literary translation. But now it's getting real and I hate reality.
On another, more straight-up depressing, note: I keep ruining my mood by looking at the news. My research is focusing more and more on translation as a bridge between the translator and the culture, and through my auto-ethnographic work it's more precisely about translation being a bridge for the immigrant back to their culture. But my nationality is Ukrainian ... so, like ... ??
I'll be having a nice day studying in the beautiful State Library of Victoria, starting to feel good about things and like maybe I have my life on track, and then I'll take a little social media break and see an article about footage being leaked of a Russian man castrating a Ukrainian soldier?
I just ... I don't really want to do anything after that. It's totally ruined my mood and I would rather escape into some other world than go back to thinking about my Ukrainian identity.
Also, I just found out you can use different fonts on here.
Anyway, the CoC preparation has begun. I am freaking out a bit, I am emotional drained. My plan is to giggle my way through this.
Also, P.S. I just noticed that I some entries ago I started calling these 'Dairy ##' instead of 'Diary entry ##' and the editor in me is having a meltdown and wants to go through and update, but the researcher is terrified that if I update it'll change the day of the diary entry and when I need to go back and it it'll be different. On that note, should I make a backup of all of these, with dates, on GDrive or something?
0 notes
Text
Diary 13: Back on the Bird
So, I'm back on Duolingo: I have my beginner Korean going with the intention of actually learning, my beginner German back up with the intention of messing around, my advanced French with the intention of making myself feel better when I fail at the Korean. I know Duolingo has it's shortcomings, many people don't like that it gamifies learning a language. But it's my favourite game, so there you go (and I am not under the assumption that I can learn a whole ass language just by playing on Duo).
I love the process of learning grammar of a language. I enjoy learning rules, it feels like an insight into the mindset of the culture (see: the work of Lera Boroditsky). This has come to the point where I now like guessing rules, which is now a past-time of mine when I watch Kdramas and listen to the patterns that come up when people speak.
The thing that I find so intriguing (and fun) about learning through Duolingo is app gives no grammar lessons. They are just plonk sentences in front of you with the English translation. You're not learning bit by bit, putting the puzzle pieces together, you're learning in translation.
My dad is a sailor, he used to go all around the world in a big boat and deliver various foods to different countries. In maritime college they had to learn English, as that is the main language that is used globally for trade. To speed learn English, I remember him telling me, they would just learn whole phrases and that they meant. When you have a job to do, it doesn't matter how or why the sentence is structured that way - that's just what you say when you need X.
Duolingo is great for learning vocab, but the grammar aspect is almost like ... like you're being taught how you would translate a sentence into English, not what the sentence is truly saying or what it's made up of.
I'm very tired right now and still hit with covid fatigue, so I'm not totally sure where I'm going with this. I guess I can't decide whether Duolingo teaches language through an already established translation, or whether it is devoid of translation as it doesn't share the essence of a phrase.
But then, learning a language without any aspect of translation at all is impossible ...
0 notes
Text
Diary 12: still thinking about who gets to translate ... and titles
I am at the part of the process that I have been dreading this whole time: coming up with a title. Technically, I have roughly another four months before I have to present a title but I need to start brainstorming now so that I don't panic cry in four months because I have put it off so long.
I had a big brainstorming session at the library* and noted down the key areas I have been looking at. One of which was translator identities, as in: who is the translator figure in our literary culture now? What do they represent. Etc. etc.
This may just be on my mind because I have been waiting (and have finally had the opportunity to stop waiting) for the release of the book I want to die but I was to tteokbokki, a Korean book translated into English by Anton Hur.
I have been really excited to see how I connect with this book, which was described online as part-memoir part-self-help. It is a recording of author Beak Se-Hee's therapy sessions. I find this interesting because, like Beak, I, too, am a young woman, who work(ed) in book publishing and sees a therapist. I feel like this book is written for me (in some ways ... I'm not Korean so the cultural aspects that would have brought her to see a therapist are different to mine).
What I was very excited about was the fact that this book would be marketing to young people like me (cue the pastel book cover), with mention that RM from BTS is a fan at any opportunity, but translated by a man in his 40s.
I like the translator, I think he is super clever and I like tuning in whenever he is on a translation panel because I find he is very candid about his translation practice. But the fact that it gave me pause that he would be translating this book was interesting to me.
Hur has said before that he mostly reads women so it makes sense for him to want to translate what he reads. He has also (in a panel where I asked the question weehee) mentioned that he translates as best and faithfully he can, and trusts that the original author's words will do the magic.
Do I need this book to be translated by a similarly-aged woman? I don't think so, but then why was this my brain's first instinct?
Anyway, I'm about halfway through now and the world has not caught flames. A book is a book, you know?
*Side note: I studied at St Kilda Library and their desks are so retro. They're wooden, with little compartments, and red chairs, and stood in little rows like some old-timey classroom. I could feel the desk telling me to sit up straight.
0 notes
Text
Rogue thought 8: intrusive interpretations
The most definite representation of the inability to extract one's own history with literature came to me in a flash when I was walking down Elisabeth street, a little tipsy, the other night. It came to me in the form of a Ravi Zupa print that was sitting in the window of an art shop:
Borges once wrote that literature doesn't happen in a straight line. Books may be written one after the other but we don't spend our lives reading them in that order, and yet each new book read will give new perspective to the books already read, and colour the interpretation of the books to be read. Some of Dostoyevsky's work could be deemed to be Kafkaesque, even though Kafka wrote after D.
Because of this Zupa print, two great works of literature will forever be changed and linked in my head. First, the obvious one, Dylan Thomas's 'Do not go gentle into that good night', and my all-time favourite novel Mikhail Bulgakov's The Master and Margarita.
Maybe Zupa is also a Bulgakov fan, maybe it's unintentional, but when I saw that black cat the first thing I thought of was all the renditions of Bulgakov's character Behemoth that I have seen. They all look, more or less, like this. So, my eyes saw the cat and my brain went *bang* it's Behemoth. Then I read the words around, half expecting it to be a Bulgakov quote, but rather found Thomas.
This feels tangential to the translation work and yet, that was the third thing that came to my mind. Interpretation is like an intrusive thought, once you interpret something you can't un-interpret it, even if you end up deciding you don't like that specific interpretation. The thought is still out there and it will still have some influence.
I guess, the point, after all that, is that it's important as translators to know where we're coming from and to listen attentively to our intrusive interpretations.
I want to buy this print for my walls.
0 notes
Text
Diary 11: what happens in the brain when we switch languages and why is mine so invasive?
I spent a long long time proclaiming that I would be terrible translator because I find it really difficult to switch between languages. Words don't connect in my head, not to their meanings, not to their counterparts in other languages.
When my parents would ask me what a word in English meant in Russian and I couldn't answer, that felt like a failure on behalf of my parents. When my parents would ask me how to say something French and I couldn't think of it on the spot ... that's my pride being messed with.
I stopped thinking about this phenomena because, well, I don't live with my parents anymore. But the other day it came back full-swing, in a way I haven't experienced it before.
Recently, something clicked in my brain and French is coming back to me. I'm talking to myself in French, I'm talking to my kitten in French, I'm doing French translation practice, I'm listening to French music, I watched a French movie the other night. They say the most surefire way to learn a language is through immersion, and they're probably right.
After a week or so of this, and after watching a movie where I was still reading the Eng subtitles (which I do because I'm terrified of finding out whether I'm better or worse at French than I think), I picked up my English-language book. At the moment, I'm reading Strangers I Know, a wonderful memoir by Italian translator Claudia Durastanti. And my brain went absolutely haywire with trying to translate the English words I was reading into French in my head.
Have you ever used the Google Translate app's camera feature? You choose the language you are translating from and the language you're translating into, click the camera and hold your phone over the text you don't understand. After a second or two of reading the image in front of it the phone app starts to switch the words on the screen into the language you chose to translate into. This is still not a smooth process, it's like *BANG* all the words are now in English. They sort of flicker into English one by one, and not in order.
This flickers is what was happening in my brain while I tried to read. I see 'children', my brain says 'enfants'; I see 'emotion', my brain says 'emossyonn' in a terrible French accent. And all this was happening fast. My eyes were reading and my Golden Retriever brain was sprinting to keep up, but wouldn't drop the heavy stick it had found that was slowing it down.
I find switching between languages difficult. Thinking or speaking in the one language is just easier, turns out even when I know that languages significantly less than the other.
0 notes
Text
Diary entry 10: le francais sans dictionnaire
I wrote an essay a while ago about Google Translate being my best frenemy in my translation practice. I lean on it too much and while I don't think that's a bad thing (because I'm still writing the mood, the tone, editing the bits that GT gets wrong, etc. etc.) I do wish I was better without it.
In what is now becoming a life habit: I was reading a Lydia Davis essay and she was talking about how when she was new to Dutch she would take simple stories and try to translate them without a dictionary, basically trying to infer the meaning of the words she didn't know by their context ... so I copied her.
I went to Heart Attack and Vine, my favourite place to study when I don't need wifi, and wrote out the lyrics to Orelsan's song 'Jour meilleur'. Why Orelsan's lyrics? Eh, just because his songs have been stuck in my head for months and I thought why not them?
The rule of the game was simple: translate the lyrics (just the meaning of the words, not the poetry) without checking the dictionary. If I got stuck, I would circle the word and move on. If, after coming back to it at the end I still couldn't get the meaning, I would check the dictionary. And then after all that I compared my meaning translation to Google's.
To my genuine surprise and delight, I translated 'Jour meilleur' very quickly and almost perfectly, I just got a couple of words NQR. It almost felt unsatisfactory, how quickly it all ended. So I have started on Orelsan's 'La Quête' and this one ... is harder haha. I have gone through my first go and there a lot more blanks and circled words in this one. I am too tired right now to keep going but tomorrow, with fresher eyes, I hope to try to find the meanings as best I can without a dictionary. See how far I get.
My high school French teacher once said that you're a good language learner if you're a good guesser ... I think she was right.
I do wonder why I feel the need to screen translation practice through French and not go straight to my Ukrainian. I started with a French novel before I felt confident enough to try a Ukrainian novel. I am doing this so I can use the same technique in my Ukrainian novel translation.
I wonder if it's saying something that I learnt translation theory (the rudimentary academic knowledge that I have of it) in French class at uni. Maybe it doesn't feel as scary to try these translation techniques because in French they're not new and I am very aware of my level of French, as opposed to Ukrainian, in which I find gaps in knowledge all the time. It's funny comparing my knowledge of a language that was taught to me, and a language that I know in a ... feeling, somewhere deep inside my brain, kind of way.
0 notes
Text
Rogue thought 7: translating 'девчонки'
I just had an interesting choice to make in regards to the word девчонки. In Russian девчонки is a popular slang word for women, or girls, usually meaning young ones. Like, you see a group of 20-somethings walking down the street or having drinks at a bar, those are девчонки.
I'm translating 'Чёрный Бумер' for a sort of poetry thing I want to create and there's the word девчонки. My initial instinct is to translate it to 'girlies', because that's what I would use. I love the girlies, I love to see them having fun. I don't necessarily call my friends girlies, though. I guess I could also translate девчонки to 'ladies' in another context, just not in the context of this rap. I don't think the man rapping about his BMW is calling the girls 'ladies'.
And I realised he wouldn't be calling them girlies either. No dude who looks like this would refer to women as girlies, that's weird. So I went with 'chicks', which I hate but feels right.
Funny how a slang word can mean something different depending on who is using it.
0 notes
Text
Rogue thought 6: getting to know one's own workflow
I am a year into this PhD research and this weekend cemented something for me: I cannot, for the life of me, work properly if I don't have time constraints.
It was a four-day weekend. Four. Whole. Days. Yes, I got food poisoning and that sucks, but really I should have been able to get more done than what I did.
Another thing that I learnt is that after roughly 1 hour of Ukrainian translation, which currently equals a page of the book I am translating, I can no longer function in any language. I see the Ukrainian sentence, not a word makes sense. I put it into Google Translate ... not a word makes sense. It's like a total jumble.
We have our limits, I guess I'm learning mine.
Now, to research productivity tools I can use to make this easier once I have a full-time job.
0 notes
Text
Diary entry 9: swapping one colloquialism for another
I am back to translating some Твоя М and found myself doing a funny thing today: I failed at translating a Ukrainian colloquialism in one sentence, and in the next translated into an Aussie one.
The word геть in Ukrainian basically means to leave a place, but depending on the context it could range from simply leaving to straight up fucking off. However, it's not a verb, though it can be used as a command. Confusing? This is why I couldn't translate it properly.
Google Translate translates it to 'away', which is right, but we don't use 'away' or 'off' in the same multifaceted way that Ukrainians use геть.
Not only this, but to my ears геть is such a ... Ukrainian word. Firstly, the way Ukrainians pronounce that г letter, which would translate closest to our English g, but in Ukrainian it's pronounced by sort of hollowing out the sound. It's a sound that Ukrainians get bullied for, but I like the way it sounds.
I couldn't find the right way of translating the word and keep leaving it out, or using slang-like qualifiers like 'totally' where appropriate.
Lamenting the fact that I had to take out something so wholly Ukrainian in my translation, in the next sentence, without a morsel of hesitation, I translated a sentence to read '[she] went off at them'.
The book is written very colloquially (I'm even trying my best to stick faithfully to the punctuation, which includes a lot of ellipses), so I'm trying to keep the same tone. But as soon as I typed the words 'went off at them' I stopped and wondered if that's a specifically Aussie way of putting things.
Do people go off at other people in England or America? Maybe!
It was novel to feel like I was trading a slang term from one beloved country for another, like I'm bringing the two together in a way that's it's brought together in my head.
0 notes
Text
Diary entry 8: I almost saw a movie in the cinema in a language I can't understand
Something funny happened to me today to once again reiterate the need for translation in order to be able to experience and participate in art.
I went to see Belle at the Nova. I went to see the Japanese version with English subtitles. I sat in the cinema, with one other person in the seats, and watched roughly ten minutes of a movie in Japanese (a language I know absolutely nothing about) with no subtitles. There was no English translation and I was just watching a very beautiful movie that I couldn't understand.
And I wondered, really truly, whether I should go tell someone or if I should just sit there and watch the thing. In some ways, it would have been interesting to watch a film that everyone has been describing as a visual spectacle (it is!) and not understand a word, try to see how much I can infer just with visual cues and tone of voice speaking the lines.
Then I got up and let the cinema managers know. Turns out the subtitles haven't been working for a week and a half, and they had rebooted the system in the hopes that its would work today but no dice. So, they put on the movie dubbed in English and I happily watched the whole story.
In Belle there is a virtual reality called U, where everyone has an avatar and lives their lives online. A detail of the animation and creation of the U universe is how the comments left by people in the universe self-translate in real time. Text in little text boxes comes up in Russian, Korean, Arabic, French, Indonesian (those are just the ones I could recognise quickly enough) and then flip into Japanese.
I liked that the creators of Belle understood that without translation this world couldn't exist. And without translation, I would have been sitting in an empty cinema totally dumbstruck.
0 notes
Text
Diary entry 7: Yet ANOTHER serendipitous moment
What is it with things fitting into place sometimes?
The other week I was working on an 'unimportant translation', which basically meant that I was translating something that wasn't culturally significant in any usual way and seeing how I felt, what would come about, etc. And then, I opened up my book of Lydia Davis essays, ESSAYS TWO, and read some essays exactly about some experiments she did that were ... well, unimportant in the grand scheme of things, but she learnt a lot. And I thought 'well, if Lydia Davis is doing this stuff, then so can I.'
Now, I have started translating some speeches by Zhirinovsky, the psychopath, which I want to turn into a poem (chilling) and opened up my ESSAYS TWO to an essay by Lydia of her forming a memoir into a poem. She writes in full detail how she went about changing the format, what she had to think about, what it did to the text.
Incredible! It's so helpful.
Lydia Davis and I are so in sync, next I'll start translating Proust!
0 notes
Text
Rogue thought 5: translation in audio
I had to have a real think about whether to title this 'translation in audio' or 'audio in translation', but have gone with the former because my current thinking is around books that have been translated and then performed for audiobook format in the translation.
There's layers of interpretation happening there, which I having been mulling over today.
I have listened to translated like in audiobook format before but today I started listening to Before the Coffee Gets Cold, written in Japanese by Toshikazu Kawaguchi and translated into English by ... Jeez Louise, the translator's name isn't on Google and it's not on the PanMac book page. Who translated this book? This should not be an investigation ... standby
Geoffrey Trousselot - an Australian man, turns out. Lovely, got there in the end.
Anyway, today I started listening to it and became acutely aware of the English words that I was hearing. Listening to a book is already a different way of engaging with the language than reading, some moments just sounds crisper when spoken aloud. Studying translation has already made me pay attention to word choices in a different way, but hearing them said aloud, given a new context, makes me pay ... even more attention.
'Scatterbrain' was used in the book and when I heard it something just pinged in my brain and I thought 'Scatterbrain ... what an interesting word, I have never paid attention to it before. What an English word! I wonder what the word is in Japanese, I wonder if there is a scatterbrain in Japanese or what the closest would be, I wonder ...'
And off and off I went, and then realised I had stopped listening.
0 notes
Text
Diary entry 6: Another serendipitous moment
A war has broken out in Ukraine and while my mind is now filled with so many different thoughts, this blog is specifically about translation and PhD research, so I will reflect on that.
I have been thinking about viewing translation as a bridge between translator and source culture. I think translation is a bridge between many things to many things (like a public transport network, maybe) but this is the particular link that's been on my mind lately. I am trying to narrow my focus in my studies and am wondering if examining how learning to translate, and all that entails, including reading and scouting, not just translating words into other words, could connect me with my culture. Perhaps I don't have to be an authority on Ukrainian literature now, but I can grow into it.
I went to see a film at the French Film Festival: A tale of love and desire, which I loved. It's about a young French man of Algerian background, who does not speak Arabic, trying to be palatable to everyone: the French and the French-Algerian community he is part of. At Sorbonne, he signs up to a literature class about erotic Arabic poetry, where he meets the newly-immigrated from Tunisia, Farah. And so they learn about culture, about being in love and lust, about Arabic poetry.
At the heart of cultural connection is literature, relationships, and translation.
Ahmed writes Farah a love poem in French. Farah writes him a letter in Arabic, which he treasures and finally asks his father to translate, i this way connecting with his dad as well.
Connection through literature is at the heart of what I am doing. I feel that now. This movie made me feel so happy and understood. I know there are immigrants who have tried (and, in some ways, failed) to fully assimilate the what they're made of inside, and feel inspiration when experiencing their language, their literature.
I'll write a bigger, better reflection on this for my research but wanted to put this hear, so my thoughts are dated.
0 notes
Text
Diary entry 5: I am not, and can never be, Maya Angelou
I had a go translating Maya Angelou’s ‘Still I Rise’, which is one of my favourite poems. Some things came up: I decided to translate ‘Still I Rise’ into Russian because it is a poem I love and love to spend time with, and because I wanted to show my parents some contemporary English-language poetry. They say you can’t translate something you don’t love (‘One can only really translate successfully those books which one would have liked to write oneself.’ Simon Leys) but they also say that in order to translate something well you need to understand it totally, whatever that means (‘the translator must know more about the work than the author knows himself’ Simon Leys). But I can never know what Maya Angelou was writing about. I can try to understand, I can use all my critical thinking skills and historical knowledge, but I will never know what it was to be a Black woman in America.
Even my love for this poem is not the same love other women would have for this poem.
I persevered in my translation because I know it will never see the light of day, and I was interested in these pauses and questions of identity I had to keep asking myself. But what really surprised me was how the Russian language was not built for these politics. The language itself was struggling with Maya.
‘Sassiness’, as in ‘Does my sassiness upset you’, does not exist in Russian. To dictionary-translate sassiness, you get дерзость, which, with the help of back-translation, means audacity, insolence, impudence. Дерзость is all the bad stuff, but it’s not sassiness. That word, a heady mix of politics, reclaiming language, sexiness, fun and strength, just never had to exist in Russia. Anyway, with the help of my mum, I went with ‘пофигизм’, which roughly means ‘I-don’t-give-a-fuck-ism’.
Another word that really stumped me was ‘soulful’, as in ‘ Weakened by my soulful cries’. All the possible translations I was getting were to do with literally the soul, but soulful, the emotions and sounds that word evokes, didn’t exist. I’m still mulling over that one.
But how do I do it? How do I translate Maya Angelou into a language that doesn’t have the words seeped in this particular history? Maybe I can’t. Maybe I shouldn’t.
Happy new year, by the way.
0 notes
Text
Rogue thought 4: an alternative literature
Today I started reading The Work of Literary Translation by Clive Scott and this caught my eye: ‘an alternative literature’. Scott doesn’t go on to extrapolate what he means by this, not yet at least, but I hope he does.
In the meantime, I haven’t been able to concentrate on anything else. These days I think often about how translated lit fits into traditional book publishing industry. By that I specifically mean that I’m coming to this not as a translator but as someone who works in sales & marketing; it’s my job to know how to talk about a book and know who would want to buy it. In the industry I often hear the term ‘the translation reader’ touted around, which I have often thought was a lazy category but maybe not.
My Twitter feed is basically all translators and translation-related stuff. The hype that I felt for Tilted Axis Press’s upcoming short story collection translated from the Indonesia - i.e. a language we don’t see many translations from - when it was first announced made me realise that I’m not engaging with translated literature like a general reader. I am a specific kind of reader and I do look out for specific things.
We can translate anything, right, any genre, any book. But maybe this idea of ‘an alternative literature’ is important in recognising that no matter what we do, the translated text is different to the organic text*.
It is literature, yes, but it is also somehow outside literature: a confusion for the publishing industry, a necessary evil.
I don’t know where I’m going with this but there was something about the notion of ‘an alternative literature’ that struck me.
*P.S. Organic text? Organically-grown text? What is this thought-train I’m on? Is it saying something?
0 notes