1 Transcription and Phonology
1.1 General
-
Square brackets [] indicate a phonetic transcription.
-
Slashes // indicate a phonemic transcription.
-
Angle brackets (or rather: single angle quotation marks) ‹› indicate a graphemic transcription.
1.2 Japanese
Non-linguistic examples are transcribed using the so-called Hepburn romanization system. On the other hand, the linguistic examples given in the present study are consistently provided using a phonemic transcription. The diachronic development of the phoneme inventory of the Japanese language is reflected in the transcription accordingly. Major phonemic changes are described below. The linguistic periodization very roughly corresponds to the historical periodization given in parentheses.
Old Japanese (OJ): 7th–8th centuries (~Asuka and Nara periods)
-
Transcription of OJ follows Unger’s (1977 [21993]) system as modified in Unger (2000).
-
Distinctions of /mwo/ :: /mo/ are only made for the Kojiki.
Classical Japanese (CJ): 9th–12th centuries (~Heian period)
-
OJ /p/ > CJ /f/.
-
OJ /Cwi, Cye, Cwo/ > CJ /Ci, Ce, Co/.
-
Sino-Japanese /kw-, gw-/ :: /k-, g-/.
-
A distinction is made between syllable-final /-n/ and /-m/. Moreover, Middle Chinese /-ŋ/ is reflected as Sino-Japanese nasal vowels /-ĩ/ and /-ũ/—depending on the preceding vowel in MC.
-
Middle Chinese /-p, -t, -k/ are reflected as /-fV (< -pV), -tV, -kV/ in Sino-Japanese, even though they were perhaps pronounced without a succeeding vowel, particularly in the case of /-t/.
Middle Japanese (MJ): 13th–17th centuries (~Kamakura and Muromachi periods)
-
CJ /o-/ > MJ /wo-/.
-
CJ /-f-/ > MJ /-w-/ before /a, o/; > MJ /-∅-/ elsewhere.
-
CJ /wi/ > MJ /i/.
-
CJ /e-, ye, we/ > MJ /ye/.
-
CJ /kwe, gwe, kwi, gwi/ > MJ /ke, ge, ki, gi/.
-
CJ /-n, -m/ > MJ /-ñ/.1
-
CJ /-ĩ, -ũ/ > MJ /-i, -u/.
-
No phonemic distinction is made between long open /ǒǒ/ (< /au/) and long rounded /ôô/ (< /ou/) as there is not sufficient evidence to believe that such a distinction existed (see Toyoshima 1984). Instead, /au/ and /ou/ are used.
New Japanese (NJ): 18th–19th centuries (~Edo period)
-
MJ /wo/ > NJ /o/.
-
MJ /ye/ > NJ /e/.
-
MJ /au, ou/ > NJ /oo/.
-
MJ /f-/ > NJ /h-/.
-
MJ /du, di, dy-/ > NJ /zu, zi, zy-/.
Modern Japanese (ModJ): 20th century onward
-
NJ /kwa, gwa/ > ModJ /ka, ga/.
1.3 Chinese
Linguistic and non-linguistic, modern and pre-modern examples are generally transcribed anachronistically using the non-phonemic Hànyǔ pīnyīn
‘Literary Chinese’ is used as a term referring to the pre-modern, predominantly written lingua franca of East Asia, which is based on the grammar of Old Chinese, particularly the varieties from the later Zhou dynasty (6th–3rd centuries BCE). Since it was in use up to the Qing dynasty (1616–1912), it naturally underwent various diachronic changes, which however are not relevant to the present study. The term ‘Literary Sinitic’ is deliberately not used, as it might potentially obscure a clear division between the linguistic dimensions of language and writing and suggests that more than one language is synchronically being referred to.
2 Morphology and Syntax
The morphological analysis follows the non-syllabic description of Japanese as presented in Rickmeyer (1995 [²2014]) for ModJ and Rickmeyer (2004 [⁵2017]) for CJ and is adapted accordingly for the other stages of the language evolution. The morphological symbols used in the analysis are the following.
-
A middle dot ‹ꞏ› is used to separate pre- and suffixes from lexemes, other pre- and suffixes, and derivational enclitics.
-
A double oblique hyphen ‹⸗› is used to separate enclitics from lexemes and other enclitics.
-
A hyphen ‹-› is used to separate two lexemes from each other.
-
When morphemes are cited in isolation, phonemes present only in some but not all allomorphs are put in parentheses, e.g., the morpheme -(a)zu ‘negative’ with the allomorphs -azu and -zu or the morpheme -((u)r)u ‘attributive’ with the allomorphs -uru, -ru and -u. This is the only difference from Rickmeyer’s notation system (cf. his -Azu, -URu).
-
A hashtag ‹#› is used to indicate the end of a sentence.
-
A dash ‹–› is used to indicate a syntactic inversion.
-
Quotation marks ‹“”› are used to indicate syntactically embedded quotes of (in)direct speech.
-
A slash ‹/› is used between lines within a poem. No morphosyntactic statement is implied by this purely metrical parsing aid.
3 Grapholinguistics
There are two ways in which example sentences are analyzed in regard to the level of writing. A simple one-level analysis, mixing morphosyntax and grapholinguistics, and a more detailed two-level analysis, used only when the need arises. The one-level analysis shows the following features:
-
Phonographic notation is recorded in lowercase letters, morphographic notation in SMALL CAPS.
-
Underspelled elements (phonemes or morphemes) are u̲n̲d̲e̲r̲l̲i̲n̲e̲d̲.
-
Non-sequentially written elements are rearranged according to their sequence in syntax.
The two-level analysis is specifically used when sequentiality as a phenomenon is discussed. An additional, abstract layer in the analysis is added, preceding the first level described above.
-
The first level of the analysis gives the values (phonographic or morphographic) of each graph in the same order as they appear in the writing sequence, separated by a hyphen, and marks them by consecutive indices.
-
Allomorphy is marked on this level by parentheses. Underspelled elements are not added on this level. Unread graphs are indicated by ‹∅›.
-
The second level then resolves non-sequential graphs while the indices remain, making it possible to backtrack the movement.
4 Dates
-
Dates according to the Gregorian calendar are written as ‘Day.Month.Year’ with the month given in Roman numerals.
-
Historical dates of the Japanese lunisolar calendar are written as ‘Year.Month.Day’ preceded by the era name.
5 Primary Sources and Abbreviations
5.1 Japanese
| AK |
Azuma kagami |
| GK |
Gonki |
| HJK |
Hōjōki |
| HM |
Heike monogatari |
| HMF |
Harima-no kuni fudoki |
| HMZ |
Honchō monzui |
| IJS |
Iroha jiruishō |
| KJK |
Kojiki |
| KKO |
Kōzanji-bon koōrai |
| KKW |
Kokin wakashū |
| KMJ |
Kokinshū manajikai |
| MGK |
Meigetsuki |
| MHI |
Mana Hyakunin isshu |
| MHJ |
Mana-bon Hōjōki |
| MIM |
Mana Ise monogatari |
| MK |
Midō kanpakuki |
| MOH |
Man’yōgana Ogura hyakushu |
| MTG |
Mana Tsurezuregusa |
| MYHI |
Man’yō Hyakunin isshu |
| MYS |
Man’yōshū |
| NR |
Nihon ryōiki |
| RM |
Ruiju myōgishō |
| SE |
Sanbō ekotoba |
| SMK |
Shōmonki |
| SMYS |
Shinsen Man’yōshū |
| SNSM |
Shoku Nihongi senmyō |
| TA |
Tegami anmon |
| TF |
Tōdaiji fujumon-kō |
| TKM |
Tamakatsuma |
| WGR |
Wago ryakuin |
| WR |
Wakan rōeishū |
5.2 Chinese
| MZ |
Mèngzǐ |
| SJ |
Shījīng |
The phoneme /ñ/ is not transcribed as /ɴ/, since uppercase ‹ɴ› and small caps ‹ɴ› (needed to indicate morphographic spelling) are visually almost indistinguishable.