Ruby? Agate?

I was curious about the term ルビ and the only place I could find anything about its origins was in Wikipedia:

“In British typography, ruby was originally the name for type with a height of 5.5 points, which printers used for interlinear annotations in printed documents. In Japanese, rather than referring to a font size, the word became the name for typeset furigana. When transliterated back into English, some texts rendered the word as rubi, (a typical romanization of the Japanese word ルビ). However, the spelling “ruby” has become more common since the W3Cpublished a recommendation for ruby markup. In the US, the font size had been called “agate“, a term in use since 1831 according to the Oxford English Dictionary.”

I don’t tend to type my ruby in size agate as that would make them pretty small for Japanese NNS.

Pittsburgh Post-Gazette: Saturday Diary / Dealing with a name that people can’t pronounce. http://google.com/newsstand/s/CBIwnYWQxwI

It’s interesting to consider the difficulties teachers have with written names – those in Roman letters as well as in Japanese. This is something we grew up with in the US and is now fairly common in Japan.

The Languages of Japan

In 1990 Masayoshi “Matt” Shibatani at Rice University published what is perhaps the seminal modern overview of the languages of Japan in a book under the same name. I had planned on teaching a course using this book supplemented with a few other sources in the spring semester of 2012, but that class was cancelled. I have decided to record my thoughts and findings on these languages here instead.

Shibatani’s book begins with a section on the Ainu language, which is now relegated to some very small enclaves in the northernmost island of Hokkaido. There is very little hope of resurrecting this mostly extinct language and without direct governmental intervention it is unlikely to survive the passing of the few remaining octogenarian native speakers.

The next section of Shibatani’s book discusses the Japanese language. Shibatani offers some information on existing views of the origins of the Japanese language, but does not, himself, accept the concept that Japanese is one branch of a Japonic family. This is a viewpoint that falls in line with the dominant view of Japanese scholars and the Japanese government since the latter 19th century. Something important to consider, however, is that all languages and dialects (local variants) fall on a dialectical continuum in which distance may be measured most clearly by mutual intelligibility. Modern English, Dutch and German are all Germanic languages, historically produced from the same mother language, however, the lack of mutual intelligibility tells us that they are no longer the same language and, in fact, only loosely resemble their mother language. The fact that monolingual native speakers of the Ryukyuan “languages” (regarded as “dialects” by most Japanese linguists) cannot communicate effectively with monolingual speakers of the various dialects of Modern Japanese reveals a distance that is similar to that of Dutch and English, for example. I would prefer to regard the Ryukyuan “languages” as a singular language with a number of dialects which have each had extensive contact with other languages of the area. The most commonly spoken of these Ryukyuan languages is Uchinaaguchi (“Okinawan”).

My current research looks at the effects of diaspora on the Uchinaaguchi language and concepts of identity in the Uchinaan (Okinawan) community here in Georgia.

I believe that there are several categories of lexical evidence of vowel harmony in Japanese.

1. Original Japanese words (very abbreviated selection)


2. Onomatopoeia [also originally Japanese]


3. Loanwords ending in consonants

sarada (salad)
chinki (tinktuur)
daburu (double)
desuku (desk)
juusu (fruit juice)

igirisu (England)

I propose that Japanese has an underlying matrix which encodes foreign phonemes (or those that do not already exist in Japanese) into understandable phonetic units. This matrix relies on a system of vowel harmony to determine phoneme compatibility.

What is ‘vowel harmony’? http://en.wikipedia.org/wiki/Vowel_harmony

Rules of vowel harmony have likely historically come into play in three areas:

1. Creation of “Yamato kotoba” (originally Japanese/Korean words)
– Example: kutu
2. Encoding of Chinese loanwords
– Example: Creation of kanbun (漢文)
3. Encoding of Western loanwords


Differentiation in Manyougana (万葉仮名) usage (especially in case particles) in the Nara period indicates the likely presence of 8 vowel sounds in early Japanese. These have been presented as follows:

a i1 i2 u e1 e2 o1 o2

This may point towards an existing system of vowel harmony, however, modern Japanese only distinguishes between 5 vowels.

a i u e o

These early vowels may have been allophones of three of the remaining vowel sounds and may still exist in certain dialects. That is worth researching.


Another indication of an underlying system of vowel harmony in Japanese may be found in foreign loanwords (外来語) and their Japanization. The fact that loanwords are incorporated into the language at different times reveals a great deal about the natural proclivity to assign one sound or another when inserting vowels into consonant clusters or in final position (to accommodate the Japanese phonological requirement of V, CV, or CV/n/) .

(Note: Shibatani reports that 80% of loanwords are from English as of 1964. That percentile has surely risen in the computer age. During the Taisho period he says that 48% of loanwords were from languages other than English, with Dutch and Portuguese having the strongest showing.)


1. final /i/

penki < Dutch ‘pek’
< English ‘(labor) strike’
suteiki < English ‘steak’
< English ‘three’

2. final /u/

< German ‘Trachom’
< English ‘size’

Note: sutoraiku < English ‘(baseball) strike’ [used to differentiate from older ‘sutoraiki‘]

3. final /o/

narusisuto < English ‘narcisist’
< English ‘test’

4. pluralization with English /s/ or allophone

This pluralization allows for disambiguation with existing native Japanese words.

takosu < Spanish ‘taco’ (Similar to tako – ‘octopus’)


1. medial /i/

supiido < English ‘speed’

2. medial /u/

sutereo karute < German ‘Karte’
wairudo < English ‘wild’
3. medial /e/

gerende < German ‘Glände’

4. medial /0/

sutoresu < English ‘stress’

5. medial vocalic gemination

sutaato < English ‘start’
< French ‘casserole’


Some loanwords adopted later may purposely not follow the principles of vowel harmony so as to semantically distinguish them from earlier loanwords that follow identical encoding.

gurasu vs. garasu
< Dutch ‘glas’

Both of the above terms mean ‘glass’, both likely Dutch loanwords, however, one refers to a drinking glass and the other to the material glass. Which is which?


1. medial consonant gemination

kizzu < English ‘kids/kid’s/kids”

… in progress.

I strongly disagree with the currently held position that the sound represented as (fu/hu) in Japanese is pronounced /ɸɯ/.

The initial consonant in , now represented as /ɸ/, is certainly an allophone of /h/ when preceding the vowel /ɯ/, but that consonant is in actuality a voiceless labiodental approximant instead of the widely believed voiceless bilabial fricative. The concept of bilabial fricativity implies that the lips are actually in contact in some way, when this is clearly not the case. The upper teeth also play a crucial role in producing this sound, which is not addressed in the bilabial aspect of the current representation. Other approximants, like /w/, more closely mimic the articulatory gestures and spacing required to produce this sound.

As this letter currently has no logographic symbol, I propose that it may possibly be written with a /ʋ /, the symbol for a voiced labiodental approximant, with a /˳/ voiceless diacritic below it.


Turkish, Japanese and Korean words for existence (and non-existence) also cover availability, as in “available for purchase” at a shop. There is no singular/plural element to any of these words.

is/exists isn’t/doesn’t exist
Japanese アル (aru) ナイ (nai)
Korean 있다 (idda) 없다 (opda)
Turkish var yok

   (There is/are) no problem(s).

T    Problem yok.
K    문제 없다. Munje opda.
J     問題ナイ。 Mondai nai.

   Is there any tea?

T    Çay var mı?
K    차가있는가? Cha ga innun ka?
J    オ茶ガアル(ノ)カ? Ocha ga aru (no) ka?

Japanese alone further subcategorizes existence into “being” and “living” (being alive).

is a living being

イル iru

is not a living being

イナイ inai

Turkish has a suffix which similarly serves as a “living” existence term. Consider the following present progressive sentences in both languages:

I am drinking.

İç -iyor -um.
drink -ing -am

non -de -iru
drink -ing -am

All three languages also have a differentiated copula (and the corresponding negative) separate from these existence terms:

Japanese デス (desu)
Korean (yo)
Turkish dir
Japanese デハナイ (dewanai)
Korean 아니다 (anida)
Turkish değil

Turkish seems to drop this copula in all informal cases, whereas Japanese shortens the copula in informal situations.

This is my passport.

Bu pasaportum.
this passport-GEN(me)

(kore ha) boku no pasupouto da.
(this NOM) me GEN passport COP.