You as a translator have to alter the form of the original in order to convey a reasonable meaning across the languages. it's not a matter of being hard to quantify. It has to occur because Greek,for instance is very different than Japanese. You can't carry the form over intact.
I've been waiting for someone to ask me in regards to the OP, what if the target language doesn't have a passive form? A scholar friend wrote me and said that the Coptic language has no passive form. It's even more complicated than that, because in Japanese the passive form is not completely equivalent to the Greek passive. For example, there is an honorific use to the Japanese passive.
Let me say here that the universal grammar theory of linguistics is relevant. Here is a definition: "
universal grammar Traditionally of any system of grammatical categories, structures, rules, etc. seen as common to all languages. Thus one in which, e.g., all languages distinguish nouns and verbs, or all languages have rules for anaphora" (
Oxford Concise Dictionary of Linguistics, by P. H. Matthews, 421).
Here is an older (1967) definition:
“universal grammar: The study of languages in general, and the general principles basic to the grammatical phenomena of all languages, without confining itself to any particular language” (Dictionary of Linguistics, by Mario Pei and Frank Gaynor, 224).
Therefore, translation of grammatical form is possible. The target language may not have a certain form that the original has, but it will have a way to express the meaning of the grammatical form of the original.
And you said merely that you disagreed with the above with no explanation. But accuracy does demand the meaning of a given text and not the form of the original. Equivalent meaning is the aim,not the form. Otherwise,in English at least,we would not have true translations but nonsensical interlinears. The versions that are hyped as being so literal are in-fact adhering to functionally-equivalent methodologies because the form has to be changed so often so that meaning is expressed.
When I say that form has meaning, I am talking about individual grammatical forms, not sentence order. The sentence order of a language will have significance in the form of nuance or style in that particular language, but such significance is usually non-transferable. On the other hand, generally speaking grammatical forms (noun, pronoun, active, passive, etc.) will have equivalents in the target language.
When this is not so, as in the Greek infinitive which has not parallel in Japanese, I look for the form in the target language which fulfills the same function and conveys the same meaning. In Japanese this may be a present verb followed by the word "koto" (
行くこと = "to go") or what is called the "te form" of the verb (
行って= "to go"). Simply because there is no parallel sentence structure from language to language does not mean that there are no grammatical equivalents.