Hello, one of my friends is a little bit crazy but I like sitting with her at café and chatting. I could not able to explain and oppose her about what she said, that the Japanese language is of extraterrestrial origin and their language is computer-generated. Unfortunately if I take a look at its system of syllables I have the feeling that it is too symmetrical to be just the result of a random evolution. I know little about Japanese, but I am not completely ignorant about it. The real Japanese words /not those of Chinese origin/ are really interesting because they are combination of symmetrical syllables. see below. So, do you know about any other (less known) language where you can find similar symmetry, so I can say it is nothing extraterrestrial? Thanks. Ecolpius.
How to explain regularity in Japanese language?
- Thread starter Encolpius
- Start date
What do you mean by "symmetrical syllables"?
Japanese only allows the following syllable structure:
(C)(ː)(j)V(N)
A syllable can start with a consonant C. That consonant can be geminated. Then there can be a glide, then a vowel V, then a nasal element N. The syllable itself can have lower or higher pitch.
There are languages with even simpler syllable structures, like Hawaiian, which has syllables like this:
(C)V(ː)(V)
And Hawaiian only has eight consonants and five vowels.
(C)(ː)(j)V(N)
A syllable can start with a consonant C. That consonant can be geminated. Then there can be a glide, then a vowel V, then a nasal element N. The syllable itself can have lower or higher pitch.
There are languages with even simpler syllable structures, like Hawaiian, which has syllables like this:
(C)V(ː)(V)
And Hawaiian only has eight consonants and five vowels.
Note that the table in post #1 is a table of katakana symbols.
The table is not a list of the syllables in Japanese (see table below).
The table is not words in Japanese.
The Japanese language has 100 different syllables, plus a 2-level "pitch accent" that changes meaning.
In general Japanese words have more syllables than English words ("car" is "karuma").
Here is a table showing the 100 different syllables in Japanese (except for the syllable N):
It's true that Japanese has far fewer different syllables (100) then Chinese (1,300 with tones), Korean (11,000) or English (more than 15,000). The small number of different syllables forces Japanese writing to use borrowed Chinese characters ("Kanji") for the roots of most nouns, verbs, and adjectives. Unlike Chinese, Japanese has word endings, so it adds word endings to those roots in phonetic "hiragana".
Japanese has second phonetic alphabet called "katakana" that is used to represent sounds (without meaning), so it is used in foreign loanwords. "Hamburger" is ハンバーガー (ha-n-ba-ga).
Last edited:
My mistake. I don't speak Japanese...except "I don't speak Japanese." (日本語を話せません).
I remember that from the 1980s. I have forgotten everything else. 本当!
Those are not all the possible syllables, but (perhaps) all the possible morae, which is a different concept. If you consider long and short vowels, the nasal coda -n (which is obviously not pronounced as an isolated syllable) and low-high vowel combinations such as in "hai", the total number of syllables must be much larger than 100.
EDIT: The paper "Bridging phonological system and lexicon: Insights from a corpus study of functional load" (Yoon Mi Oh et al.) gives the number of syllables in Japanese as 484. That number is calculated from a corpus, which means that the theoretical possible number is larger. For English, the equivalent figure is 6,469, though the number of theoretical syllable combinations (based on phonotactic constraints) is around 100,000.
Last edited:
But look a bit further: Proto-Slavic syllables can start with rather complex consonant clusters such as stv-.
The paper does not say that about English.
How languages compare with the number of different syllables from all words?
This comment says the Yoon Mi Oh thesis paper only counts the syllables in the 20,000 most common words in each language, and for Engish that number is 6,949. However, English has 520,000 words, so only looking at 20,000 words is not accurate. If you look at all the words, the number is 15,000+ syllables.
This linguistic thesis paper compares 8 languages. For that comparison, the paper invents a special definition of "syllable" in Japanese, which is different from "mora". That is not standard grammar. In standard grammar Japanese only has "mora", instead of "syllables". Most authors either use "Japanese syllable" to mean "Japanese mora", or never use the term "Japanese syllable".
Japanese does not have "long and short vowels". It only has short vowels. Two of those vowels in a row might be incorrectly called a "long vowel", but it is pronounced as two short vowels (in separate syllables). For example, there is no phonetic difference between two O's in midword, a word ending in O followed by the accusative particle O (を), and a word ending in O followed by a word starting with O.
"Syllabic N" is not part of the previous mora/syllable, like it is in Chinese. In Japanese, N is a separate syllable, with the same duration as KYA or TA or any other syllable. It is only a "coda" in the sense that it follows a vowel.
High/low variations only occur in multi-mora (multi-syllable) situations. The English fake-Japanese single syllable word "Hai" does not exist in Japanese. The diphthong "ai" does not exist in Japanese. The Japanese word is ha-i (yes), two syllables, where the bold shows the higher pitch. Other Japanese words include a-i (love) and a-o-i (blue).
All these concepts (high/low vowel combinations, long vowels, the coda -n) seem to be part of an attempt to create a "Japanese syllable" that matches "syllables" in English, French, etc. That is reasonable in a paper comparing 8 different languages linguistically, but it isn't part of normal Japanese grammar.
Japanese definitely has vowel length. The pronunciation as two short vowels is only really heard when singing*, otherwise, they're pretty much pronounced like long vowels, and there's even the vowel lengthening mark (which looks like a dash). Furthermore, for the most part, even the clusters ei and ou have reduced to long vowels, ee and oo, respectively, though some dialects may retain the cluster.
* In singing, I notice that oo, regardless of where that occurs, is usually sung as owo, which is likely a result of the o-wo merger, which is also why the wo particle is now pronounced o. I believe wi also merged into i, while we merged first into ye, then ye in turn merged into e. I need to see if these leave any traces when singing. Also, some Japonic languages go further and even merge wa into a, which is why, for example, Okinawa is Uchinaa in Okinawan.
Now, I do agree that long vowels are two morae long, but that does not negate them being long vowels, and the same is true about geminate consonants as well. And there just happens to be an Indo-European language that shares this property - Latin, where long vowels and geminate consonants were in fact origially wrote as double vowels and double consonants, then the apex was introduced for both, but it only stuck with long vowels, while it fell into disuse with geminate consonant which ended up being written as double consonants again. Japanese, on the other hand, writes long vowels in a variery of ways, including with the vowel length mark following the mora and writes geminate consonants with the geminantion mark (the small tsu in both higarana and katakana) before the mora.
Furthermore, Japanese vowel length is acknowledge by the Hepburn transliteration system, which uses the macron for the purpose, which just so happens to also be used in modern Latin transcription to replace the apex.
Now, the moraic n is indeed a peculiarity of Japanese, though in speech, it is basically pronounced like a normal n (or m or ŋ if preceding p/b or k/g, respectively), the peculiarity only arises in singing. That, lenghtening the correct vowel, and the correct shortening/near-elision of short u and i, are the three pitfalls I've observed non-Japanese speaker stumble on when singing in Japanese.
Incidentally, that short u and i that get shortened to the point of near-elision, seem to me to parallel the development in Slavic where the two yers (which were also ŭ and ĭ) developed the same way and in fact, in most Slavic languages, got completely elided in the end, which I think would have happened in Japanese as well, had it not been for the writing system not allowing such a loss, though it does seem that way in speech at least, eg. desu is often pronounced essentially as des, and yoshi as yosh.
* In singing, I notice that oo, regardless of where that occurs, is usually sung as owo, which is likely a result of the o-wo merger, which is also why the wo particle is now pronounced o. I believe wi also merged into i, while we merged first into ye, then ye in turn merged into e. I need to see if these leave any traces when singing. Also, some Japonic languages go further and even merge wa into a, which is why, for example, Okinawa is Uchinaa in Okinawan.
Now, I do agree that long vowels are two morae long, but that does not negate them being long vowels, and the same is true about geminate consonants as well. And there just happens to be an Indo-European language that shares this property - Latin, where long vowels and geminate consonants were in fact origially wrote as double vowels and double consonants, then the apex was introduced for both, but it only stuck with long vowels, while it fell into disuse with geminate consonant which ended up being written as double consonants again. Japanese, on the other hand, writes long vowels in a variery of ways, including with the vowel length mark following the mora and writes geminate consonants with the geminantion mark (the small tsu in both higarana and katakana) before the mora.
Furthermore, Japanese vowel length is acknowledge by the Hepburn transliteration system, which uses the macron for the purpose, which just so happens to also be used in modern Latin transcription to replace the apex.
Now, the moraic n is indeed a peculiarity of Japanese, though in speech, it is basically pronounced like a normal n (or m or ŋ if preceding p/b or k/g, respectively), the peculiarity only arises in singing. That, lenghtening the correct vowel, and the correct shortening/near-elision of short u and i, are the three pitfalls I've observed non-Japanese speaker stumble on when singing in Japanese.
Incidentally, that short u and i that get shortened to the point of near-elision, seem to me to parallel the development in Slavic where the two yers (which were also ŭ and ĭ) developed the same way and in fact, in most Slavic languages, got completely elided in the end, which I think would have happened in Japanese as well, had it not been for the writing system not allowing such a loss, though it does seem that way in speech at least, eg. desu is often pronounced essentially as des, and yoshi as yosh.
Last edited:
Hello. Next to maps of fragmented Japan, and the context of a Yayoi migration (one expanding layer), you could have had slight variation under isolation, to work towards that AAABBB around the middle (Wolfram; cf. a visualization of order from chaos, like a Galton board), if the 'clay' was not that far. Just based on Jomon dog burials,1 2 their own layer may not have been as incompatible. Any comparisons with older Ainu would be extremely indirect, but it is unlikely the Yayoi met something 'extraterrestrial.'
That ...a ni ho-no-n-i in MKorean (p. 33) feels like a different language. Possible reasons seem controversial. Cf. nawu=nyi=gya in Old Okinawan (2.4,11). Without unification, and a [ ] substrate, you could have had a map like other islands, with a hundred languages.
Before the kamikaze of 1274 and 1281, you'd isolate iterations of a common source, and this 'ordering' process. The influence of Chinese after the 5th C. can also be understood under this, i.e. not as a shock from conquest, with a hybrid or 'creoloid.'Those dogs "may have resembled Shiba Inus" (S; cf. yellow); the older 'breakaway' of the Jomon may explain some of the features, as an ancestor of a language isolate. But for more answers someone has to devote a lifetime to the question of 'isolate.'
Last edited:
The Japanese language has long vowels.
東外大言語モジュール｜日本語｜発音｜実践編| 2 円滑なコミュニケーションのために 2.2.1 拍感覚（長音）
[Tokyo University of Foreign Studies Language Modules]
Last edited:
Yes, from the phonetic point of view, long monophthongs [V:] and diphthongs [VV̯] definitely exist in Japanese. But The main question here is whether they are also phonologically distinctive, I quess.
The meanings of words, such as /'oo/ 'king' or /'ai/ 'love' remain the same, irrespective of whether they are pronounced as a sequence of two short monophthongs (i.e, ['o̞.o̞] and ['ä.i̞]), respectively or as a long monophthong (i.e., ['o̞ː]) or a diphthong (i.e., [äi̞̯]). Although Japanese has diphthongs from the phonetic perspective, they are not short diphthongs like those in German because its word like "Ei" 'egg' can only be a short diphthong ['aɪ̯], not a sequence of two short monophthongs ['a.ɪ]. In Japanese, the syllabicity of the vowels that occur after the left most nucleus can change. The same also holds true for the so-called "mora nasal", whose common phonological notation is /N/. A change in the syllabicity of /N/, is not phonologically distinctive because ['ho̞.ɴ̩] (the uvular nasal [ɴ] in the syllable nucleus) and ['ho̞ɴ] (the uvular nasal in the syllable coda) have the same meaning 'book'. The segments /V/ and /N/ are all "moraic" ones within the syllable rhyme, irrespective of their syllabicity.
And the so-called "mora obstruent", which is commonly represented by /Q/, is also a moraic segment because it occurs only in the syllable coda, followed by an obstruent /C/ occurring in the onset of the next syllable (i.e., .../Q/$/C/...). Although /C/ and /Q/ are all unsyllabic segments, we need to distinguish phonologically between /C/ and /Q/ because the consonants /C/ occur only in the syllable onset and are never moraic and the consonants /Q/ occur only in the syllable coda and always moraic. When the mora nasal /N/ and the mora obstruent /Q/ co-occur, /N/ always comes before /Q/.
The meanings of words, such as /'oo/ 'king' or /'ai/ 'love' remain the same, irrespective of whether they are pronounced as a sequence of two short monophthongs (i.e, ['o̞.o̞] and ['ä.i̞]), respectively or as a long monophthong (i.e., ['o̞ː]) or a diphthong (i.e., [äi̞̯]). Although Japanese has diphthongs from the phonetic perspective, they are not short diphthongs like those in German because its word like "Ei" 'egg' can only be a short diphthong ['aɪ̯], not a sequence of two short monophthongs ['a.ɪ]. In Japanese, the syllabicity of the vowels that occur after the left most nucleus can change. The same also holds true for the so-called "mora nasal", whose common phonological notation is /N/. A change in the syllabicity of /N/, is not phonologically distinctive because ['ho̞.ɴ̩] (the uvular nasal [ɴ] in the syllable nucleus) and ['ho̞ɴ] (the uvular nasal in the syllable coda) have the same meaning 'book'. The segments /V/ and /N/ are all "moraic" ones within the syllable rhyme, irrespective of their syllabicity.
And the so-called "mora obstruent", which is commonly represented by /Q/, is also a moraic segment because it occurs only in the syllable coda, followed by an obstruent /C/ occurring in the onset of the next syllable (i.e., .../Q/$/C/...). Although /C/ and /Q/ are all unsyllabic segments, we need to distinguish phonologically between /C/ and /Q/ because the consonants /C/ occur only in the syllable onset and are never moraic and the consonants /Q/ occur only in the syllable coda and always moraic. When the mora nasal /N/ and the mora obstruent /Q/ co-occur, /N/ always comes before /Q/.
Last edited:
Long vowels are simply vowels that are two morae long instead of one. And the only time you hear those moraic breaks in Japanese is in singing, which is also the only time the coda /N/ takes a moraic form - in normal speech, there's usually no moraic breaks in the long vowels, and the coda /N/ outright assimilates into what comes next, so it becomes variously n, m, or ŋ.
Also, I would contend that coda syllales tend to always be moraic in general. Take, again, Latin, for example, where syllables with a short vowel followed by a consonant in the coda are considered just as heavy as syllables with a long vowel and no coda, and from what I know, where they disappeared, the preceding vowel underwent compensatory lengthening as a result.
Something similar is observed in Ancient Greek as well, with compensatory vowel lengthenings caused by the loss of w: *odwos -> ōdos. If we do a moraic breakdown, it makes sense: o-d-wo-s -> o-o-do-s. As the coda consonant of the first syllable became the onsent consonant of the following syllable due to the loss of w, the vowel of the first sylable lenghtened in order to compensate for the loss.
So the moraic /N/ of Japanese would then not be anything special, but just the result of the n being in the coda. The unusual thing is that it can be drawled out in singing, but I think that's just a result of Japanese coda n being a simplification of earlier mu, which is evident from the fact its hiragana evolved from the mu hiragana. I think the coda n arose as an allphone for word-final mu (see i-ɸa-mu / iwamu -> i-ɸa-n / iwan), and when all the Middle Chinese loanwords were being borrowed, the allphone became phonetic because it now also found itself in words that can only be pronounced with n and not mu. But its mu origins live on in the hiragana character used for writing it, and in its drawling out in singing, where it does almost sound like mu.
Also, I would contend that coda syllales tend to always be moraic in general. Take, again, Latin, for example, where syllables with a short vowel followed by a consonant in the coda are considered just as heavy as syllables with a long vowel and no coda, and from what I know, where they disappeared, the preceding vowel underwent compensatory lengthening as a result.
Something similar is observed in Ancient Greek as well, with compensatory vowel lengthenings caused by the loss of w: *odwos -> ōdos. If we do a moraic breakdown, it makes sense: o-d-wo-s -> o-o-do-s. As the coda consonant of the first syllable became the onsent consonant of the following syllable due to the loss of w, the vowel of the first sylable lenghtened in order to compensate for the loss.
So the moraic /N/ of Japanese would then not be anything special, but just the result of the n being in the coda. The unusual thing is that it can be drawled out in singing, but I think that's just a result of Japanese coda n being a simplification of earlier mu, which is evident from the fact its hiragana evolved from the mu hiragana. I think the coda n arose as an allphone for word-final mu (see i-ɸa-mu / iwamu -> i-ɸa-n / iwan), and when all the Middle Chinese loanwords were being borrowed, the allphone became phonetic because it now also found itself in words that can only be pronounced with n and not mu. But its mu origins live on in the hiragana character used for writing it, and in its drawling out in singing, where it does almost sound like mu.
From the phonetic point of view, it is certainly unusual but not wrong. In Japanese, we can pronounce sequences of vowels /VV/ or that of a vowel and a mora nasal /VN/ with syllabic break between them without being phonologically ungrammatical. This is certainly not the case in German, for example. In German, we can't "resyllabify" words like ['aɪ̯] 'egg' or ['liː.bə] 'love' into ['a.i] or ['li.i.bə] without being phonologically ungrammatical, so I think that the phonological status of long monophthongs or diphthongs in Japanese and that in German are different. The same is also true for the syllabicity of the mora nasal /N/. Although the nasals in German can also be pronounced syllabically under certain conditions, we can't resyllabify sequences of /VN/ in the "Japanese" way described above.OBrasilo said:
The only nasal consonant that can occur in the position represented by /N/ in the nucleus or in the coda without being dependent on the following segment occurring in the next syllable is an uvular nasal ([ɴ̩]/[ɴ]) and the uvular nasal never occurs in the syllable onset. The mora nasal /N/ can not only be one of the five nasal consonants [m, n, ɲ, ŋ, ɴ] but also a nasalised vowel if /N/ is followed by a vowel in the next syllable as in ['hoɔ̝̃. ɔ̝] /'hoN o/ 'book + case marking postposition/particle'.
I think that the mora nasal /N/ and the mora obstruent /Q/ primarily stand for the underspecified slots corresponding to <ん/ン> and <っ/ッ>, respectively.
Last edited:
In German, the ressylabification can't work because of the mandatory glottal stop between two vowels that does not exist in Japanese. So ['a.i], really [Ɂa.i] would inevitably become ['Ɂa.Ɂi]. So [-a.i-] pretty much does not exist in German, and that impedes such resyllabification. The same is in ['li.i.bǝ] which is inevitably in fact ['li.Ɂi.bǝ] - again, no [-i.i-] in German.
Though, I wonder if the glottal stop insertion is exactly why you see a strong difference between, say, [hai] and [ha.i], because I personally don't, at least in native pronunciation other than it sounding drawled out, but the glottal stop could appear when you pronounce the latter due to ingrained habits from German.
I wonder if this was influenced by Portuguese, where /VNV/ became /ṼV/, and finally /V/. See eg. general /ʒene'raɫ/ first becoming gẽeral /ʒẽe'raɫ/, and now it's geral /ʒe'raɫ/.
Though, I wonder if the glottal stop insertion is exactly why you see a strong difference between, say, [hai] and [ha.i], because I personally don't, at least in native pronunciation other than it sounding drawled out, but the glottal stop could appear when you pronounce the latter due to ingrained habits from German.
I wonder if this was influenced by Portuguese, where /VNV/ became /ṼV/, and finally /V/. See eg. general /ʒene'raɫ/ first becoming gẽeral /ʒẽe'raɫ/, and now it's geral /ʒe'raɫ/.
In Japanese, the vowels can be accompanied by a glottal stop but the pre-vocalic glottal stop is not phonemic (/V.V/ = [ʔV.ʔV] and/or [V.V]). This doesn't have anything to do with another language such as German or an idiosyncrasy of mine.
Although unlike Japanese, the pre-vocalic glottal stop in German is not optional, it is still not phonemic. I think the question whether or not the pre-vocalic glottal stop is phonetically obligatory does not matter here because even if the pre-vocalic glottal stop in the environment /V.V/ were phonetically obligatory for these vowels to be pronounced in Japanese, ['ʔä.ʔi̞] and ['ʔäi̞̯] would be phonologically the same word /'a(.)i/ 'love'. ['(ʔ)ä.(ʔ)i̞] and ['(ʔ)äi̞̯] are simply two phonetic variants of the same word in Japanese. If the "phonological" status of long monophthongs or diphthongs in Japanese were the same as that in German, ['(ʔ)ä.(ʔ)i̞] and ['(ʔ)äi̞̯] could not refer to the same word in Japanese either, irrespective of whether the pre-vocalic glottal stop is phonetically obligatory or not.
N/ > /'hoõ/), which you might be assuming.
Although unlike Japanese, the pre-vocalic glottal stop in German is not optional, it is still not phonemic. I think the question whether or not the pre-vocalic glottal stop is phonetically obligatory does not matter here because even if the pre-vocalic glottal stop in the environment /V.V/ were phonetically obligatory for these vowels to be pronounced in Japanese, ['ʔä.ʔi̞] and ['ʔäi̞̯] would be phonologically the same word /'a(.)i/ 'love'. ['(ʔ)ä.(ʔ)i̞] and ['(ʔ)äi̞̯] are simply two phonetic variants of the same word in Japanese. If the "phonological" status of long monophthongs or diphthongs in Japanese were the same as that in German, ['(ʔ)ä.(ʔ)i̞] and ['(ʔ)äi̞̯] could not refer to the same word in Japanese either, irrespective of whether the pre-vocalic glottal stop is phonetically obligatory or not.
['hoɔ̝̃] in ['hoɔ̝̃. ɔ̝] is phonologically still /'hoN o/, not /'hoo o/ or something. The oral configuration of [ɔ̝̃] in /ho[ɔ̝̃] o/ is determined by [ɔ̝] of the case marking postposition "o" (regressive assimilation), not by the vowel /o/ of the noun /'hoN/. We don't have a compensatory lengthening for a loss of /N/ (/'hoN/ > /'hoOBrasilo said:
I don't understand how pyramids could have been built, so their origins must be extraterrestrial.
Stonehenge's stones are to heavy for human handling, so Stonehenge must be extraterrestrial.
Human eyes are too complex to have evolved according to Darwin's theory, so their origins must be extraterrestrial.
I don't understand the regularity of Japanese syllables, so their origins must be extraterrestrial.
And on and on and on.
[you may replace 'extraterrestrial' with 'Godly']
Stonehenge's stones are to heavy for human handling, so Stonehenge must be extraterrestrial.
Human eyes are too complex to have evolved according to Darwin's theory, so their origins must be extraterrestrial.
I don't understand the regularity of Japanese syllables, so their origins must be extraterrestrial.
And on and on and on.
[you may replace 'extraterrestrial' with 'Godly']
- Joschl: If you read the Portuguese example I gave, it's a case of /VNV/ not /VN/ with compensatory lengthening. There was no compensatory lengthening in portuguese, just /VNV/ turning into a nasal /VV/, and both the nasality and the length later being lost. /VNV/ turning into a nasal /VV/ is a general phenomenon in the evolution of Portuguese and happened even with two different vowels, ie. /V1NV2/ turned into a nasal /V1V2/ (optiones > *opçones > opçõẽs), thought it backed into ãũ (written ãu) when not followed by anything, most likely by analogy with mãũ (mão) from mano (hand).
The closes example to your hon o > hõ.o would be bono > *bõõ > bõ (bom). I do admit that in Portuguese, the nasalization engulfed both vowels, becoming a long nasal vowel, that then shortened. But, it may be the case in Japanese as well, ie. it might be hõ.õ, in which case Portuguese influence can't be excluded. Certainly, I've heard Nihon (Japan) pronounced essentially as Nihõ, which mirrors Renaissance-era Portuguese Iapam pronounced Iapã. So it could be that the Japanese at some point heard Portuguese sailors pronouncing Japanese with a Portuguese accent with all the nasalization processes mentioned here, and it ended up spreading across the Japanese population from there.
The closes example to your hon o > hõ.o would be bono > *bõõ > bõ (bom). I do admit that in Portuguese, the nasalization engulfed both vowels, becoming a long nasal vowel, that then shortened. But, it may be the case in Japanese as well, ie. it might be hõ.õ, in which case Portuguese influence can't be excluded. Certainly, I've heard Nihon (Japan) pronounced essentially as Nihõ, which mirrors Renaissance-era Portuguese Iapam pronounced Iapã. So it could be that the Japanese at some point heard Portuguese sailors pronouncing Japanese with a Portuguese accent with all the nasalization processes mentioned here, and it ended up spreading across the Japanese population from there.
Boy, those extraterristrials sure were busy!
In a recent TED talk, a scientist whose research team studies the human brain said "Every answer we find opens up two more questions. That is science: endless unknowns."
Hello. The few Port. "words were mostly confined to Kyushu,"1 just as their trade and slavery, in a country of "15–17 million by 1600" (cf. 12.3M). Apparently Hideyoshi "endeavored to obtain two galleons from the Portuguese, but without success,"2 before the Imjin War. This would be consistent with "one or two Portuguese naus [...] every year."3
For a connection, you may be seeing a similar source (Latin -n-), and a similar outcome ([hoɰ̃o], etc.; "the Tōkyō variety has a nasal vowel, Ōsaka has a syllabic nasal and Kagoshima has a nasal coda"). This is, that -n- is worn down across the seas.
We were not disagreeing. We were using the term "long vowel" with different meanings.
Japanese short and long vowels have the same vowel sound. Here "long" and "short" mean duration, because vowel sound duration is phonemic in Japanese (it changes meaning; it makes it a different word).
English short and long vowels have different vowel sounds. Here "long" and "short" do not mean duration, because vowel sound duration is not phonemic in English (it does not change meaning; it is the same word).
Others were using the phonetic meaning of "long", while I was using the English meaning of "long vowel". The WR dictionary lists both meanings:
- (of a speech sound) lasting a relatively long time: [Phonetics]
- having the sound of the English vowels in mate, meet, mite, mote, moot, and mute:
That second definition is simply never used when talking about any language but English.
I am not able to recognise anything similar to /'hoN o/ = ['hɔ̝ɔ̝̃ ɔ̝].OBrasilo said:
I actually meant that there is not a process like */'ho1N o2/ > /'ho1OBrasilo said:
We should keep in mind that the phonetic notation [ɰ̃] is sometimes used as a placeholder and can also stand for other individual and/or regional allophonic variants of the mora nasal /N/. /N/ seems to have a wide range of vocalic allophones.
If you are interested in the history of writing in Japan, I warmly recommend the following book to you.Encolpius said:
Seely, Christopher (2000) A history of writing in Japan. Honolulu: University of Hawaii Press.
Last edited:
Sorry, my fault, I mean something like "regular arrangement of syllables".
Any simplistic phonotactics would most likely result in regular arrangement of syllables, wouldn't it? I don't see how Japanese is different from, say, Tahitian in that regard.
Nothing *unexpected*. Highly simplistic phonotactics of that kind *is* rather uncommon, but for pretty natural reasons: it conflicts with compactness and/or unambiguity of speech. In fact, vowel reduction in Japanese (which affects /u/ and /i/ in certain positions) already results in more complex syllables on the actual phonetic level.