I assume the little kids reading those books can read hiragana a lot better than romaji… in any case, encrypted JN-25 messages were not their problem!
Backing up to this and making sure the real issue is clear to people:
Japanese does not have an alphabet. They have a syllabary. The difference is is having different characters for, say, the letters “T” and “E” versus “TE”. This, per se, has some advantages and disadvantages as a writing system but regardless does match the language quite well. However, the awkward thing about Japanese was that not only does it have the extraordinary number of homophones, but also adopted writing multiple times.
This resulted in Japanese using two separate syllabaries (hiragana and katakana), one of which is somewhat restricted to loanwords and often signage (there’s some story behind that but I can’t explain it) as well as kanji, which are classical Chinese logographs as interpreted and developed through the separated Japanese culture/s over the ages (and, in reality, a relatively narrow section of that culture/s). Let me counter poster Isamu: hiragana and katakana most certainly do carry the “full meaning” because that is actual Japanese language being spoken.
In writing, kanji are often convenient to explain the meaning of ordinary messages, because context is more difficult in impersonal communications. However, the Japanese people evidently felt that this all wasn’t complicated enough and decided that adding the Latin alphabet looked like a fun and exciting way to liven up a dull holiday, so they did that as well. Which, fortunately, is relatively easy for Westerners to pick up because many Japanese syllables transliterate easily. However, a few do not because there irregular formations in Japanese as with all human languages.
Now, I can’t speak to specific WW2 codes. That being said, this offers some opportunities and weaknesses for Japanese cryptography. First, if you have some secure reference material, the code-maker has some interesting opportunities. For example, imagine having a cipher that changes characters from a reference work (any arbitrary book or whatever) into numbers, like page+line+character. Then this number is itself coded and sent out. To decode it, you not only need to break the first layer of encryption, which might be quite difficult, then find the reference work. But you could further complicate it by using traditional works with vertical writing instead of horizontal writing, which would really mess with codebreakers. Or you can turn the syllables into a code and use that (and trust to context to determine any ambiguities), or mix all kinds of approaches. In practice, however, speed may be a necessity in retrieving the secreted message.
There’s some information listed here: Japanese naval codes - Wikipedia and also here: U.S. in World War II: How the Navy broke Japanese codes before Midway.
Also, sometimes you don’t have to technically break the code to understand it. For example, during the war the IJN referred to Midway as location “AF”. I have no idea whether that was randomly decided or somehow encoded the name of the island, or what. However, the codebreakers didn’t need to decipher the meaning; all they needed was to discover that the IJN referred to Midway as “AF”.
This is correct. Kana carries meaning as much as any spoken Japanese words. What kana lacks is extra context to differentiate homophones, in the same way we might get confused about “see” vs. “sea” without written context.
It’s only “childish” in the same sense that all-caps in English may be considered “childish” because it’s sometimes done for (or by) children. But there’s nothing wrong with it, and it doesn’t really affect meaning.
This is not correct. If the sound of the word didn’t carry the full meaning, then it would be impossible to speak Japanese without supplementing it by writing kanji.
The advantage of kanji is that it makes literature more compact by being physically smaller and by carrying context that helps differentiate homophones. But it’s not a requirement… there’s a lot of Japanese literature entirely in kana for the benefit of children, foreigners, and other less-literate readers.
These sound to me like the sort of thing that would be of no obstacle at all to a codebreaker who was fluent in Japanese, and I assume that a major nation would have plenty of fluent Japanese speakers available. If they didn’t, then the Japanese wouldn’t have needed any code at all.
And it’s possible that I missed a technical nuance in the definition of “alphabet”, but the point that I was making with that term is that Japanese, like English but unlike Chinese, has a relatively small set of characters which can, in combination, phonetically represent all of the spoken language. The Japanese set is a bit larger than English’s, which might make keyboards a little more awkward; and it has a hard time representing certain sound combinations that occur in other languages, but not Japanese (such as syllables that end in a consonant); but neither of those is a particular obstacle to encryption or decryption.
Let’s not overlook that @DesertDog answered this question as far as the specific WW2 cipher in question is concerned. What was being encrypted was not katakana and was not kanji at all, they were 5-digit code words.
Well, the naval codes were, in fact, basically all broken so in the end it didn’t change the outcome. However, if you can code for a specific syllable or logograph, without even giving an indication which is which, you can potentially make a very strong code indeed. The biggest problem is training, issuing all the codebooks, and then securely destroying them all. Codes done in this way are inherently time-limited, however, since you must assume the code will be broken eventually.
I also wonder - did / does Japanese have strong regional accents the way England does? how would this affect the interpretation of Kanji?
For example, depending on the accent, the sound “eye” can be “A” or “i”. In New Zealand, 'harbour" sounds like “hah-bah” and “Ben Smith” is “been smeth”. This would do wonders for phonetic spelling, as can be seen in English writing (especially names) from the 1500’s before spelling was standardized.
Does Japanese phonetic spelling suffer from this?
Japanese regional accents most definitely exist, but I am not sure how they affected spelling before standardization.
Yes. More dialects. A lot more variation.
Basically not at all. It’s like how you and I might say the word “pen” differently. You might hear me say “pin” and get confused, but I can resolve the confusion by writing the word, which doesn’t change no matter how we say it.
Yes. The “standard” dialect is the one from the Tokyo reigon, and some of the others are seen as “contrified”/“hick” dialects. I don’t know exactly how they represent the different dialects in Japanese (probably use kana) but here are a few examples of English translations of manga that depict regional dialects:
(When they are doing English dubs of anime, they tend to use heavy “southern” accents to represent the differences.)
Japanese lacks the space character and has only 5 vowel sounds (versus English’s 14) and 15 consonant sounds (versus 24 in English).
When you’re speaking the language, breaks and emphasis help to isolate the discreet words so that you can figure out which words they are. When you’re looking at a full sentence, written entirely in katakana or hiragana, it’s nigh impossible to isolate the words because they have so many plausible ways to break up the sentence into valid sentences.
Chinese characters always come at the start of a word, so that helps to break up the sentence in lieu of spaces.
It’s actually called Romaji.
Regarding Japanese Morse code, there’s a system called Wabun for the kana.
True. As I mentioned in the Christmas gift thread, I gave a cuckoo clock imported from Germany as a gift. The brochure with the unpacking instructions was in ten languages and the Japanese section took up the least space with Chinese* just a bit more.
Of the European languages Swedish was the most compact and Spanish took up the most room, although just ahead of English and Russian.
*The symbol for Chinese was the PRC flag but the symbols look like traditional Chinese to me, still used in Taiwan but not in China proper.
I find it so weird that there are English words in the message. Of a Japanese code. So how would a Japanese naval captain get such a message? Morse (or similar) code?
Go ask a Japanese friend if they like reading books in hiragana only. Katakana only texts will make them stabby, hiragana only makes them toss the book.
And you go ask your Japanese friend if hiragana books have any different meaning than kanji books. I’ll wait right here.
As I said before, all-kana is analogous to allcaps in English. It’s annoying to read, but as far as carried meaning, it’s no different to talking.
This is the hill you want to die on? The information contained in hiragana (without kanji) is not the same as the information contained in kanji (with hiragana).
That’s just not true.
Of course it is. Look, if you’re correct, then it would be impossible to understand a Japanese text if someone has read it aloud to you, because spoken language doesn’t have kanji. Obviously that’s an absurd proposition, therefore you’re incorrect.
Insistences aren’t persuasive at all. Show us a normal kanji-containing sentence, then explain to us how and why you think it loses meaning when transliterated into hiragana.
You say it would be impossible to understand? No way, I never said that. I implied it may be confusing.
Have you not noticed that Japanese to Japanese conversations need more confirmation as to what was said than English conversations do?