I dare to add my 2 cents, even though I am definitely not as experienced as some of the other commenters here.
jammon39 wrote:Learning a dead language in pursuit of reading fluency seems like a significantly different task than learning a spoken language.
Not necessarily. Don't forget that many people learn living languages primarily for reading as well, or it is at least their favourite activity in those languages. But in case of a dead language, there is much less pressure on speaking, and you don't get judged as a bad learner, just because of your priorities.
There was a wonderful thread on Latin (and other languages), that started with a video guide on learning Latin through tons of reading. At first a rather modern coursebook (Lingua Latina per se illustrata), then lots of intensive and extensive reading, if I remember correctly. With nice ideas on how to construct a not too steep learning curve. I'm a bit lazy to search now, but it was a very good video.
How would methods differ?
That would depend a lot on the Dead Language. In case of Latin (and probably Ancient Greek, at least some historical variants of it, and two or three others), you could learn it like a living language too. There are coursebooks like Assimil, there are grammar workbooks, there are very living-like coursebooks for schools (I was very impressed, when a friend of mine showed me her Latin coursebook used in a German school. That was many years ago).
You could definitely go the traditional way, focusing a lot on grammar and analyses. And I think you'd have a solid chance to succeed, if you really followed those textbooks with lots and lots of reading. Whether or not the language is dead, I think the old learning method is too demonized and blamed for something it wasn't guilty of.
But in case of many other languages, you'd have no choice than to learn more like a sort of cryptography, or a rather theoretical thing, given the scarcity of available resources or incomplete knowledge of the language.
Would the reading/listening approach work well for this task? Or should listening be a lower priority?
Yes, I think it could work well, but it would depend only on the quality and amount of the available resources. In some cases, it would be rather impossible, you'd be inventing a lot of stuff (and probably wrongly. What a luck the natives cannot defend their language ). In others, it wouldn't be so hard.
In case of Latin, you'd struggle not only with the two official variants of the pronunciation (church vs. classical), but also all the variants or "accents" of the "normal" pronuncation. It doesn't get discussed much, but the way a French, a German, a Czech, an English, and an Italian natives will pronounce Latin "right" will sound very different. And all of them will probably swear they are using the official reconstructed classical Latin pronunciation. You can even get mocked for the "mistakes" you were officially taught. So, it's not that simple.
I'd say the main value of listening, no matter how artificial the recording may seem, and speaking, no matter how useless it is in the real life, is mainly a memory aid. You get to save information to your cortex with more tools. That's what I find very valuable. And something I honestly believe I should have been doing a decade ago.
How would you replace the time you would normally spend in conversation or with a language partner?
I don't usually do that much, so there is nothing to replace But more seriously: I'd say more reading is always a good idea.
But now that I think of it, I am luckier than most, when it comes to using a dead language in the normal professional life
Would you expect learning to be faster and easier or slower and harder?
Neither. It would depend on your goals, and also what do you compare the Dead language to. In my case, I'd expect Latin to be much easier than learning to read Japanese. But probably harder than Italian. And a different Dead language would be much harder. No clue whether I'd expect contemporary Mandarin to be harder than Classical Chinese.
I don't think it makes any sense to generalize in this case, as another variable will be the literature available to you.
Btw I love Einzelne's idea about the graded readers.