Actually, you can get a readout on lexical semantics (the meaning of words) if you have a large enough corpus (much larger than we have for any extinct language

Actually, you can get a readout on lexical semantics (the meaning of words) if you have a large enough corpus (much larger than we have for any extinct language). The method uses a vector space of word cooccurrence (look up word2vec). It will tell you what words are semantically similar. What it won't tell you is what those words mean in the real world (unless possibly you have pictures with captions). It's like being able to see mountains in one direction and a lake in another and a valley in another direction; you can say you're going towards one or the other of these features, but (ignoring the sun and stars) you don't know whether that's north, south, east or west.