How much reading do you have to do in a language until you read smoothly, without having to stop often and ask yourself about morphology or syntax, or to consult the dictionary? A simple question with a more complicated answer. It depends on the reader, on the language, on the text and genre, and even on the particular sentences within those texts (not all sentences within the same author or genre are of the same difficulty for learners), &c. And there are, of course, different kinds of reading, and many texts, too, for one reason or another merit, not only reading, but even multiple re-readings. Even with these variables, for most of us, fluid reading (or hearing) means the past mastering of several thousand lines wherein the dictionary did have to be frequently cracked, wherein the grammar did have to be checked, wherein the concordance did have to be probed, and wherein the original beside the version did have to be compared, and so on.
At the beginning of a unique Greek grammar for beginning students, Paula Saffire refers to a time in graduate school when reading Greek became less encumbered and more automatic for her.
The reason this happened was that I was reading Greek, happily, about eight hours a day, because of Harvard’s most powerful teaching tool, the Reading List. (Read all of Aeschylus, all of Sophocles, all of Homer, seven by Euripides, and so on.)[1]
The great Swedish scholar of Chinese, Bernhard Karlgren, wrote in 1908 of his reading assignments in some Germanic languages — to which family, it should be noted, belonged Karlgren’s mother tongue — while a student of Adolf Noreen:
300 pages Icelandic prose, 80 pages Icelandic poetry, 100 pages Gothic grammar, 40 pages Gothic text, 275 (difficult!) pages Old Swedish. I have very good reasons to rest a little while.[2]
At the time, Karlgren was working on two majors: one the subject just mentioned, and the other being Slavonic Languages. (Karlgren needed to master Russian because of the font of materials on Japanese and Chinese in that language. Students of Georgian are in a similar situation today.)
I don’t have a specific number, whether in hours or in lines, to answer the question asked above. But I know that it is a lot, and in many cases we may recognize the specific number only after the fact. One day, after hour upon hour and line upon line, we just realize that we’re moving along in a text with far fewer bumps in the road than before. And that’s when a new kind of enjoyment begins in the language.
If you have any studio-biographical references for scholars’ and learners’ time and efforts spent among the pages of foreign languages, please share them in the comments.
[1] P. Saffire and C. Freis, Ancient Greek Alive, 3d ed., p. xv.
[2] Letter of April 11, 1908 to his girlfriend Inna, quoted in N.G.D. Malmqvist, Bernhard Karlgren: Portrait of a Scholar, p. 38.
hmmlorientalia 
http://paleojudaica.blogspot.co.uk/2014_02_23_archive.html#1641156955604875898
Thanks a lot, Jim. I agree wholeheartedly about the consistent reading and about the utility of prudently using translations as aids. In the context of deeper exegesis, you mentioned concordances at the end of your post. I’ve found concordances to be very helpful in language learning, too: e.g. in Old Georgian, when I come across certain forms I want to know more about, I can check the corpus available through TITUS and see them in context, and for translated texts, perhaps also then find the underlying Greek, Armenian, etc.