No, in Esperanto the tense of subordinate clauses is independent of the tense of main clauses. The sickness is in the present tense (even in the English sentence "I didn't know that you were sick"), so the verb "esti" is used in present tense ("estas").
In English, you take the tense from the main clause as the starting point, and you need a perfect tense ("had been") to look back even further to the past, but Esperanto is different in that regard.
I guess what I should have added to my message is that in Esperanto, the subordinate clause takes the point of time from the main clause as a starting point. So at the time of "knowing" (sciis), it was present tense that the person was sick. Therefore, it is written in the present tense in Esperanto. In English, since these tenses are dependent of one another, that verb takes over the past tense of the main verb, even though it means the same as in Esperanto, that the person is sick at the time of the knowing.
That's a good question that actually crossed my mind while writing that last post. I'm not entirely sure, my guess is it might actually be ambiguous in Esperanto as is, and if you really wanted to specify, you would have to add a word like nun or ankoraŭ or some such. I'm sure there are better Esperanto speakers than myself who can help out better.
That said, it is a bit of a weird thing to want to say. Basically you are saying that you did not know then that the person would be sick now. Usually when you say such a thing, you probably just learned about it before saying it, and so the then is only slightly prior to the now, and in that case there is hardly any meaning difference between "I did not know that you were sick" (the more typical phrase in English) and "I did not know that you are sick". In fact, the latter kind of sounds off somehow; if you really did want to talk about the present tense, as compared to a (significantly) earlier time of knowing, you would probably be more likely say "I did not know, that you would be sick". And in that case, estos would work just fine in Esperanto ("Mi ne sciis, ke vi estos malsana").
I can't reply to Vincent for some reason, so I'm adding my reply here:
"I didn't know that you are sick" sounds just fine if the speaker is referring to an ongoing situation. Similar: I didn't know (in the past) that you are a a vegetarian (an ongoing situation).
So, if Vincent is right, then there would be ambiguity, and the speaker could perhaps add a few words to clarify, as I did parenthetically.
I think it's because you can only go so deep with comment replies.
I'm still a bit doubtful about the "didn't know that you are" situation; I think in most instances, you would say "were", even if it is an ongoing situation. ("I didn't know that you were a vegetarian.") But I'm not 100% sure, and it might be fine in many cases.
Thank you for your input, David.
You were right in your first reply: in Esperanto the tense of subordinate clauses is independent of the tense of main clauses. "Mi ne sciis, ke vi estas malsana." means that the knowing is in the past and the being sick in the present, both of them relative to when you are speaking.
Maybe you're confusing this kind of sentence with indirect speech, in which you use the same tense you'd use in direct speech, and it's therefore relative to the time of the main clause.
I'm still not so sure it's right that both are determined relative to when you are speaking. As far as I understand, "Mi ne sciis, ke vi estas malsana" just means that you didn't know that the person was sick at the time of your not-knowing, not at the time of your speaking that sentence. Which is why a sentence like "Mi ne sciis, ke vi estos malsana" is possible. I know it is like that for indirect speech, but I believe it does extend at least also to verbs such as scii, pensi, etc. (Try Tekstaro for
[^ ]+is, ke (la )?[^ ]+ [^ ]+os\b—there are many results involving those other verbs, not just indirect speech; some of which also involve things that should have ended by the time of the sentence being produced, so are no longer future tense relative to that.)
You're right, it does say that, and it could be done like that. But it also doesn't mention the thing about indirect speech at all, and I found the many examples on Tekstaro that I mentioned.
Still, you're probably right. But if it is indeed an error, it is at least a widespread one, apparently. And it would bother me a little bit that the language made an arbitrary exception for indirect speech, even though the example discussed here is structurally identical to such a case. But, well, the mere fact of it bothering me is not a good cause for me to challenge it—I still should follow the rules as they are, in these situations.
Indirect speech is not an exception. It is addressed in the same page you linked to. The sentence I quoted was where Bertilo explains that relative clauses work the same way as indirect speech. The tense stays the same in indirect speech as it would in direct speech.
“Li venos.” Tion mi diris. → Mi diris, ke li venos.
The reason indirect speech tense is relative to the main verb tense is because it already is relative in direct speech, not because of an arbitrary exception.
Alright, I see what you're saying. I guess I just don't like the rule in general, then—but again, that's of little consequence. As is the fact that apparently I always misunderstood it a little bit.
It does seem awkward for some cases, though. To go back to the original example, "Mi ne sciis, ke vi estas malsana." According to the rule you have laid out, if the sickness is in the past, but was at the same time as the 'knowing' (present relative to the main verb), you would say "Mi ne sciis, ke vi estis malsana", correct? That feels unnatural to me (and I don't believe I ever hear people use it like that, although of course people could just be doing it wrong, same as myself). Would there be a way to distinguish between that case and the case where it was in the past even relative to the main verb? Perhaps with the addition of a word like antaŭe ("Mi ne sciis, ke vi estis malsana antaŭe"), but even then, if that subordinate phrase were put on its own, that word antaŭe could just as easily refer to the present relative to the main verb sciis, as further in the past beyond the time of the main clause.
I still think there's something arbitrary to the rule that the tense should stay the same as it were in direct speech when it's not direct speech, but that's neither here nor there.
My 2 cents: In English, if you show up to someones house and they answer the door, tissues in hand blowing their nose, you could say "I didn't know you are sick" and even though that makes the most technical sense, most people would actually say "I didn't know you were sick", and they don't mean it in the past tense. That's English for you.