To further clarify, until after the war Japan used non-phonetic "spelling" (much like modern English does now). 伴う and 向う used to be written (and centuries ago, pronounced) like ともなふ and むかふ.
Actually, I was / am somewhat worried that this column makes too many demands on the non-Japanese reader. However, I think it is very important that people see the problems that translators of O Sensei, like John Stevens, for example, have had to grapple with.
I think that's a valid concern. But what you've done here is an amazing effort.
Some tangential thoughts --
In view of the enormous number of homonyms that exist in Japanese, this would make reading very difficult and this, I think, is why Chinese characters have never been banished from the written language.
The necessary rejoinder to this is that children and illiterate people have no problem understanding spoken Japanese, with all of its homonyms. Context is very often quite clear. In fact, moving to a system of historical kana usage would likely clear up most of the misunderstandings that would remain. Rather than any necessity for kanji in understanding written Japanese, I believe it's mostly a matter of inertia and tradition. Removing kanji would be a bigger change than English or German spelling reform, and those have not exactly proceeded smoothly. Further, a Japanese-speaking population that learns kanji then has access to the entire history of Japanese writing in its original form. If Japan went to an all kana or all romaji system, it would require the transliteration of the entire corpus of Japanese literature, an undertaking of a scale that I believe would be unprecedented in the history of the world.
Those accustomed to the alphabet might well also ponder at the immense rote-learning process expected of every young Japanese entering the Japanese school system, in order to master reading and writing the 5,000-odd characters required for day-to-day written communication in Japanese.
I work in two Japanese elementary schools, so I have some experience with this. Actually, this is neither so rote nor so immense as it might seem at first glance. A key thing to understand is that while kanji certainly contain a core meaning, to everyday Japanese they more often represent sounds
rather than meaning. Often, when native Japanese make mistakes of kanji, they substitute kanji that sound the same or contain the same phonetic element. Aside from a few simple, core pictographs (e.g., 日, sun, 山, mountain, 水, water, etc.) Japanese children are not taught the meaning of kanji in isolation, but rather as parts of compounds, and as words in context. Kanji themselves are made up of a finite number of commonly appearing components, and children learn to match these components together to make a work, much like English speaking children learn how to match together non-phonetic or pseudo-phonetic combinations of letters to spell. Elementary school children generally only rote copy the kanji out ten times or so, and this is less to impart muscular memory as it is to practice balance and appearance. By the time they go to junior high school, they don't need to spend much time practicing writing new characters -- they can easily break them down into their recognizable components, and remember how to match them back to together again.
Finally, the majority of Japanese people do not and need not learn upwards of 5,000 characters. By the time they leave high school, they know the 1,945 daily use kanji, plus the 284 kanji commonly used in names, for 2,229 kanji. In the vast majority of media, kanji and terms outside of these will be accompanied by yomigana -- superscript hiragana used to indicate how a kanji is read. (I'm sure you are aware of this, Professor Goldsbury; I'm adding it just as general supplementary information for others.)
To bring this more on topic, I look at the translations and can't help be frustrated at how much is not
translated, or translated poorly. But that does not necessarily reflect on Messrs Stevens and Bieri. Translating even modern Japanese is a difficult, unsatisfying task. More often than not for discourses such as Ueshiba's what is needed is not mere translation, but impractical explanation and commentary, like Professor Goldsbury has provided above.
As a further example, I prefer Steven's Essence of Aikido
translation of the Shochikubai poem, but both neglect a very real possibility -- that 錬り清めゆく(neri kiyome yuku) is a modifier
of 気の仕組 (ki no shikumi). In other words, it could read "The Pine, Bamboo and Plum -- the mechanism of ki
that purifies and refines..." Unlike English, Japanese transitive verbs require no specified object, and it's not a given that either "ki no shikumi" or "Shochikubai" must be the object of the purifying and refining.
A further note on "shikumi" -- while Stevens renders it "make-up" and "basis", it's sense is more of "mechanism, working, the nuts-and-bolts". Explicit in the meaning is the sense of variable components working in concert. Of course, I can explain all this, but if demanded to create a somewhat lyrical translation that captures all of this, I doubt I could come up with one. Particularly since Japanese poetry is designed to be vague and evocative, so that no two people will necessarily understand the original in the same way.