“I’ll grant you misleading, but I don’t think it could ever be as jarring. I never did get used to Isaacson’s unitalicized “ya”. ”
No, certainly not as jarring. I say misleading because the Lo! type language is usually an attempt to replicate the tone of written poetry based on oral literature, and I don’t think that’s very appropriate for highly stylized language of something like haiku.

Nope. I agree with LH and Xee on this one though — lots of people interested in haiku have beginner-to-medium level Japanese, so better there than not. (Of course, the ideal edition has Japanese orthography, romaji, several translations by different authors, detailed and thoughtful notes, a wet bar, and a solid gold robot pony. Everything else is just an imperfect shadow of this idea.)

TThis is an interesting approach, and I’m sure one that has been used many times in translating such poetry. In fact, I know it has. But the problem is that these English interjections have their own nuances and poetic uses, and their use as a substitute for Japanese particles could be just as jarring (or at least misleading) to a reader highly familiar with archaic English poetry.

I’ll grant you misleading, but I don’t think it could ever be as jarring. I never did get used to Isaacson’s unitalicized “ya”.

By the way, I’ve only ever seen the “Lo!” approach used in very old approaches to tanka/haiku, like William Porter’s Hyakunin Isshu. You seem to have something specific in mind too — do you know of any more recent attempts in this vein?

perhaps the rōmaji is also there so you can try and work out what it sounds like even if you don’t speak japanese? poetry being sound as well as lexical meaning.
That said, my editions of translated greek and russian poetry are either all-english or bilingual: they do assume you are either familiar with the script or don’t care about the original sound.

my ideal for translations of japanese poetry, i decided a while back, would be something rather like the iwanami nihon koten bungaku taikei, with the page in three parts (e.g., orig/translation/notes), so there’s a continuous section for notes, which for me is somehow easier to deal with than individual footnotes.

But Japanese readers don’t need the translations at all and probably won’t be reading the book, whereas I suspect there are a lot of English-speaking readers, like me, who have enough of a smattering of Japanese to make some use of the romaji but to whom kanji would be nothing but a decorative element.

“Archaic English interjections — “Lo!” “Alas!” “O!” — would arguably have gotten Isaacson much closer to the effect he claims to be seeking.”

This is an interesting approach, and I’m sure one that has been used many times in translating such poetry. In fact, I know it has. But the problem is that these English interjections have their own nuances and poetic uses, and their use as a substitute for Japanese particles could be just as jarring (or at least misleading) to a reader highly familiar with archaic English poetry.