Some of my books and articles are translated into a variety of languages: Danish, Norwegain, Finnish, German, French, Spanish, Portuguese, Lithuanian, Slovenian, Chinese, Japanese, Korean and Farsi. I may have forgotten some.
I have translated myself from English to Swedish and back. It is not really translating because it is my own text, and I know what I want to say and can say it slightly differently in the other language.
With Danish and Norwegian, I can check that the translation is correct. With other languages I can at least ask how they have managed some things that can be problematic in translation. Most translators are grateful if the author is willing to help.
I have also translated myself into Russian. The problem with Russian is that it is not my working language, unlike Swedish and English. I used to do guest lectures in Russian, but not for a long while now. Still I typically prefer to translate myself for the reasons given above. As an author, I can allow myself liberties that a translator woudn't be comfortable with. Yet some time ago I was approached by a professional journal in Russia that wanted to translate one of my articles, and I didn't have time to do it myself so I said go ahead, but I want to see the translation. Well, yesterday they sent it. They had three minor quiries. I thought I'd read through it quickly, just to see that all my metadiegeses and crossfocalisations were in place. £&!^$%£"&%*;!!! It took me all day yesterday and all morning today to get it somewhat right. Surely it would have been quicker if I had translated it myself.
There are many things that can go wrong in a translation. Occasionally, a small error makes the meaning the opposite. All translators have done this. I have done this. Usually you yourself or the editor picks it up because something feels wrong and does not make sense. But if I say in my article that something is invaluable, and the translator says it's impossible, a very important argument goes profoundly wrong.
If a scholarly article uses, all the way through, a particular term, I would expect the translator to find out what it is called in the target language rather than use long and clumsy circumscriptions. It took me five seconds to find it on Russian Wikipedia.
However, I would be prepared to forgive this. It is actually easy to correct.
I am very proud of writing simple, comprehensible academic prose. My motto comes from Kurt Vonnegut: "A scholar who cannot explain to a five-year-old what he is doing is a charlatan". I frequently receive compliments for the clarity of my writing, and therefore I am not scared of writing about complicated subjects. If I can explain them clearly, I have understood them myself. Therefore I was thoroughly upset when I started reading the translation. If I for some reason picked up this article as it was published, I'd say: This author writes gibberish. I don't want to read gibberish. I cannot get through to what the author is trying to say.
Russian is a rich language with a huge vocabulary. A translator doesn't have to invent new words the meaning of which readers would not know. Russian is also a remarkably flexible language. It uses gerunds and participles where English uses embedded subordinate clauses, which allows for a variety of sentence structures. I used to assess translations into Russian for various publishing houses and grant-awarding bodies. My first check was always for gerunds. Russian also has free word order that can be used for rhythm and emphasis. My second assessment would therefore be for word order. Russian has instrumental case that comes handy. Translators who ignore these options produce bland, dull prose. Academic prose does not have to be bland. It has to be particularly enjoyable if it is to convey complex argument.
Shall I go on? Or do you believe me that I spent at least twelve hours editing twenty pages of horrendously bad academic writing? And mind, it's not like reading a student essay with deficient argument and poor structure. If you could only get through to them, the structure and the argument were perfect (the article had gone through a rigorous peer-review procedure and a fierce editor before it was published).
You may ask why I did it. It isn't a prestigious publication. I could have asked them to go to Hampshire and Hereford where hurricans hardly happen. But it is a matter of honour. My name is on it, and I am proud of being able to write simply about difficult subjects... see above. If there is a handful of colleagues in Russia who might be interested in my work, I don't want to scare them off with slovenly language.