Because we don't have much other evidence of this sort, and because earlier uses of the term "shanty" for a work song have been so elusive....AND because the presence of lumberjacks makes it sensible to suspect he was referring to their hut.... I concede that this is a difficult case to argue, and may just have to be filed in the "Who Knows?" drawer.

My problem -- and why I thought this example was worth considering before filing it -- is that although I don't see cinching evidence that it refers to chanties, the other explanation (lumberjack hut) is *also* problematic.

Sure, it may be most plausible that "shanty", isolated, refers to the hut. But that use in context just doesn't make sense to me. Do we have any other evidence to suggest lumberjacks called their off-duty songs "shanty songs," i.e. that they named them after their dwellings? And even if they did, if those were more of the nature of "focsle songs", why call this one a "shanty song." It is clearly (to me) a rowing/work song.

The phrase "shanty song" (in various spellings) has been used by others to refer to chanties somewhat in the manner of "chai tea" and "ATM machine." Those references are much later, however.

My guess is that O'Grady has coined a new "poetic" verb, to "shant," to correspond to the noun "shanty." It would thus mean "to stay in a shanty."

That interpretation has the virtue at least of coming from within the poem itself.

I agree that O'Grady may have coined a new verb. My opinion differs, however, in that I think it is more plausible that the hypothetical "to shant" would mean "to chant/to chaunt/ to sing songs at work". I don't think "to stay in a shanty" comes from within the poem any more than does "to sing as one rows." It's just that the shanty-hut is familiar to us whereas as we don't know whether "shant" referred to a work song then. [ shant, "song" > shanting "singing"] seems morphologically more plausible than [shanty, "hut" > shanting (not shantying) "staying in a hut"].

I think too that if O'Grady had meant "shantying," he would have used that word,

Not if it didn't exist yet in that form. Remember the whole issue with how Nordhoff's "chant" became "chanty", or why the leady singer was called "chanty-man" rather than "chant-man." It seems like the morphology of this new concept was not fixed in a definite was a noun, verb, or adjective. It may be that "shanty" (in this Quebec experience) was the noun and "shant" was the verb. At least, I think that is just as plausible as O'Grady making the form "shanting" to mean "staying in a hut" rather than making that one be "shantying." The "y" is problematic in either case. It comes down to opinion as to whether O'Grady was coining (poetic license) or reflecting something that was non-standardised.