It kind of is a proof if we assume that single words can be translated at all. Translate a single word from Language X (more words) to language Y (fewer words) and back. I can't uniquely recover all the words in Language X that way.
I don't know about that. For many practical purposes probably not?
I'm just on the thread following this idea: "The article seems to think that a word is untranslateable if there is no single word in the target language"
So we're talking about "translatability" of single words. Mapping multiple words of language X to one word of language Y is going to have some effect on translation.
That is the crux of the article premise: each synonym conveys similar denotations (principle component is I think what the article called it), but usually with some difference in connotations (the off axis contributions). You can nudge the languages vectors towards each other by adding enough synonyms and modifiers together, but they are always a little bit off even still
So, really, this can be simplified to the question "can written text fully convey all human concepts", some of which having labels in only some languages, which is an obvious "no".
I thought there was a difference between those two in how they start burning, like one needs an external flame to start while the other can burst into flame without an obvious ignition source.
But what joyful means to you likely differs from what it means to me, simply because we haven’t read the exact same literature and had the same conversations.
True, but many languages now have words that were absent from their earlier vocabularies. Shakespeare did not have the option to use 'telephone', 'semiconductor' or 'entropy'.
