国立国語研究所学術情報リポジトリ
Survey of standardisation in Tsuruoka, Japan : Comparison of results from three surveys
conducted at 20-year intervals
言語: eng 出版者:
公開日: 2019-03-25 キーワード (Ja):
キーワード (En):
作成者: 米田, 正人, YONEDA, Masato メールアドレス:
所属:
メタデータ
https://doi.org/10.15084/00001976
URL
ノ4メ)ごz,z6sθLinguistics 2 (October,1997) 24−39 (Article]
Survey of standardisation in Tsuruoka, Japan:
Comparison of results from three surveys conducted
at 2e−year intervals
Masato YONEDA
The National Language Research lnstitute
Keywords
the National Language Research lnstitute, the Tsuruoka dialect,
standardisation, segmental phonemes, accent Abstract
This paper presents some preliminary results on the changes in the Tsuruoka dialect in the post−war years. The National Language Research lnstitute conducted three surveys on linguistic changes in Tsuruoka in 195e, 1971 and 1991. This paper makes the foilowing points. (1)The third survey, conducted in 1ee1 after a 20−year interval, made it possible to investigate linguistic change over a period of forty years. (2)Standardisation ef segmental phonemes is progressing rapidly and word accent more gradually in Tsuruoka.
(3)The results of the three surveys support a model of language change in society proposed by Nomoto and Egawa in 1974.
Intreduction
This paper will present some preliminary results from a research project entitled The linguistic life of a local society 一 changes in the Tsuruoka dialect in the post−war years 一 which, by conducting large−scaie surveys at regrtlar intervals, attempts te trace the spread of standard Japanese in one part of northern Japan.
The National Language Research lnstitute has conducted three surveys on
linguistic change in Tsuruoka(鶴醐), a city on the west coast of northem Japan, in 1950,
in 1971, and again in 1991, The purpose of the surveys was to investigate the status of the local dialect and the infiuence of various social factors on the spread of the standard language. This was of pardcular interest in the rapidly changing society of post−war Japan when the project began. This paper, using results from the third survey, makes the following points.
(1) The third survey, conducted in 1991 after a 20−year interval, made it possible to investigate linguistic change over a period of forty years.
(2) Standardisation of segmental phonemes is progressing rapidly and word accent more gradually in Tsuruoka.
24
(3) The results of the three surveys support a model of langrtage change in society . proposed by Nomoto and Egawa in 1974.
Project erganisatien
The project is funded by the Japanese Ministry of Education and based at the
National Language Research lnstitute in Tokyo. The following persons and
organisations co−operated in conducting the research related to the 1991 phase of the project: Kiyoshi Egawa, Masato Yoneda, Seiju Sugito, Yasuo Kumagai, Masao Aizawa,
Masamitsu lto, Kikuo Maekawa, Yoshimitsu Ozaki, Shoichi Yokoyama, Masaru lnoue,
Takuichiro Onishi, Rieko lkeda, Hiroe Shirasawa, Tokie Tsujino, Michiyo Tsukada,
Yoshiko lsobe, Junko Yoneda (the Natienal Language Research lnstitute), Ryoichi Sato
(Ferris University), Motohisa lmaishi (Hiroshima Women s University), Fumio lnoue (Tokyo University of Foreign LaRgu ages), Makoto Takad a (Tsukuba University), Shinji Sanada
(Osaka University), Toshiaki Suzuki (Toyama University), Yasuo Yoshioka (Kumamoto Junior College), Motoei Sawaki (Shinshu University), Kazuo Kato (Kanazawa University),
Kazuyuki Sato (Hirosaki University), Hiroyuki Kanazawa (Okayama University),
Yoshimichi Mizuno (Kyoto lnstitute of Technology), Tetsuo Nitta (Hiroshirna Bunkyo Women s College), Katsumi Shibuya (Osaka University), Koichi Shinozaki (Tokyo Metropolitan University), Shingo Hayano (Sophia University, Graduate School), Shiro Hori
(Tsuruoka City Library).
Background
Tokyo, the capital of Japan, is the centre of economy, culture and language.
Throughout Japan, local dialects are changing under the infiuence of Standard
Japanese. The change involves the adoption of the pronunciation, vocabulary, etc., of the language of the Tokyo area, which is, roughly defined, Standard Japanese.Figure 1 shows the location of Tsuruoka, a small city on the Japan Sea coast of Yamagata Prefecture, in northem Jap an. Figure 2 shows the main dialect groupings of Japan. The dialect of Tsuruoka is part of the Tohoku dialect group 一 Tohoku is the
northeastern part of mainland Japan. Tsuruoka itself covers about 235 square
kilometres and has a population of about a hundred thousand. The central old city area was the target ef the three surveys. The target area is six kilometres wide and four kilometres long, and has a population of about sixty thousand.Figure 3 shows the numbers of informants in each survey. A survey consists of two sep arate p arts: an area survey and a p anel survey. ln the area survey, we take a random sample of people to get a synchronic picture of the spoken language in the
JAPAN
sbf
/塾
,,,SX
Hekkeido dialect
!.
Tsuruoka
e
., Kyeto, 6tl
ヌα・、粟Σへ ノ▽
「「
秩G,・
]。 U・召\
;.SNE t A r v,
r 鑑
o
v一
2eokmcl ty x o/]
昏一rて≦〜1・1
覧 ・( 二♪く.。「㍉
;乙バ・
. Oseka
つ!
! ノya吊agarδ一Pfef
。 、,
F、
●
Tokyo
Hokufiku d{elect 嫡。
u叩、k. di、1,、竃タ ・ !ジ
殿磐蛾・・節鳴鐘諮1二轡n
篤蝉濃lia
cu dialecr
dialect
dialect
Figure 1: Map of Japan Figure 2: Dialect map oi Japan
area. ln the panel survey, we trace the same people over the years to investigate how particular individuals have changed their way of speaking.
In 1950, we conducted an area survey only, using 577 inforrnants randomly selected from the Register of Residents kept at the Municipal Office. ln 1971, we conducted an area survey with 457 new infomiants, again randomly selected. ln addition, we conducted a panel survey with a hundred and seven of the 577 informants from the original 1950 area survey. Finally, in 1991, we conducted an area survey with 405 new randomly selected informants, and a panel survey with 314 informants. Of these 314, 261 were informants from the 1971 area survey and 53 were informants from both the 1950 area survey and the 1971 panel survey.
We emp}oyed the same methodology in 1950, in 1971 and in 1991, to allow close comparison of the results from the three surveys. The method was similar for both the area survey and the panel survey. Each informant was interviewed individually using a questionnaire. The interview lasted about 40 minutes. The area sur vey also included an
additional questionnaire which was mailed to the informants ahead of time and
collected at the time of the interviews. Below is shown the distribution of theinformants who participated in the area surveys by sex, age and educational background.
26
Table 1
Tetal
: Distribution of informants by sex, age and 195e o/0 577 (100)
educational background
1971 0/0
457 (100) 405
1991
0/o
(leo)
Sex Male
Female
243 (42.1)
334 (57.9)
204 (44.6)
253 (55.4)
181 (44.7)
224 (55.3)
Age
15−1920−24 25−34 35−44 45−54 55−69
98 64 130 118 84 83
(17.e)
(11.1)
(22.5)
(20.5)
(14.6)
(14.4)
60 (13.1)
50 (IO.9)
88 (19.3)
IOI (22.1)
75 (16.4)
83 (18.2)
47 (11.6)
21 (5.2)
65 (16.0)
10e (24.7)
61 (15.1)
lll (27.4)
Educatien Low 383 (66.4) 189 (41.3) le5 (25.9)
Middle 169 (29.3) 217 (47.4) 199 (49.1)
High 25 (4.3) 51 (11.1) 101 (24.9)
The informants ages ranged from 15 to 69. With each survey, the number of informants in the lower age group decreased, and the number in the highest group increased. Educational background is described as low, middle and higk: up to rniddle school, up to high school, and beyond junior college, respectively. In Tsuruoka, as in the rest of Japan, the number of people with a higher educational background is mcreaslng.
Results of the Survey
A wide range of items were included in the survey, including: pronunciation, accent,
grammar, and vocabulary; items of personal history including age, education, occupa−
tion, place of birth etc.; and items related to ianguage use. Although we are still working on the analysis of the data from the 1991 survey, we have some results on pronun−
ciation, including word accent. ln each of the three surveys, we investigated a total of 31 items,
standard }anguage, including differences Rasalisation, and vowel quality. ln the phonemes we were investigating are emb
lgso
Su・ve−
1 577
1971 1991
A[ea N f Area
um. ! Survey i m.... i Survey
paneM・urveNX,,,,.6,,.3,.,1,: i
Panei survey
Figure 3: Numbers of informants in each survey, Tsuruoka
representing nine types of difference between the Tsuruoka dialect and the
in labialisation, palatalisation, voicing,
following list of words we surveyed, the oldened .
TyPe
Labialisation I
Labialisation ll
Palatalisation
Voicing
Nasalisation
Centralisation 1
Centralisation ll
Vowel shift I
Vowel shift E
Word accent
乱
㎎ 頭
S
日
署曜瓜 務中 三三三子突 火西蛇百戯曲背甲松柿靴帯解開口猫開窓鈴野烏平地島町雑芥聖杯糸息 中 扇 猫旗背烏団
短 卿 吻
舘
コの ユ
由圃一田 圃瓢比垣
﹇ ﹇ ﹇ ︹︒.−飯磁劇−n ㎏
コむ コ
ユ︹ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ﹇ ︹ ︹ ﹇ ﹇ ﹇ ﹇
コ ﹈﹈a 野晒齢 韻鳳叫圃糾
﹇ ﹇ ﹇ ﹇ ﹇The pitch marks are those standardly used fail in pitch and r a rise in pitch.
28
Tsuruoha
[kwajo:bi]
[suiikwa]
[Φ¢b詞
[ejaktu]
[垂ipe]
[d3e:mulfo]
[fenaga]
[aSe]
[m。d。虚]
[kagi]
[ktUdztuコ
[hado]
[hada]
[had3i]
[k曲d3封
[negoコ
[6b封
[mado]
[s磁dztα]
[kidzine]
[karas&]
[s&md
[tfizi]
[s&ma]
繊dz垂wa]
[tSizi]
[karas婁]
[entots樹
[egi]
[&do]
〔をgi〕
[nergo]
[hadaコ
[fernalga]
[karralsi]
[tsdzi rwa]
in
gloss
Tuesday
water melon snake one hundred beard
tax office
back sweat pine
Japanese persimmon shoes
plgeon
fiagbee mouth
cat belt
window
bell
fox crow Chinese ink
map
island paper fan
governor mustard chimney
stat正on
thread breath
cat fiag
back crow
paper fan
works on Japanese: 1 represents a
5とθndanζノrXj
︹UOnUQJ
窪pt−mN==pt===:tNsc $
. 囚\
t s
SN 磁、一_哨_
一 8ss h − k 、田 SN Ns SN SN 、田
Age Gnoup 15 20 25 35
1 1 1 i 19 24 34 44
Figure 4 : Non−labialisation f
Standand rZj 0 10
9。
W。
≠U︒艶葡3︒舶抱
亡﹂一44仁︾ =﹂一QJ虞JeU、 ㍉ムー_
「一△\
\、
皆一一一一へ へ
一轍、 \、
!va 、\、 \
ヘ へ
/ 賦 △ げ \、
、、、
、、、
国\、
、、、、、
、磁
一 1991
一一一狽嶋鼈鼈黶@1971
一一一一?一一一一 G950
Age Gnoup 15 2e 25
1 1 1 19 24 34
Figure 6: Devoicing
麗﹂一4り﹂4 5ーバ﹁
45
ロ﹂一Q↓質Jeロsとθndendr2ワ
∩UO OQU
1
︑
̲
\ 風
橘
〜7 \
叙、
, ∈ヨー一一一田、、
巴 、、、
『改 N N N N N x N NNs
s s s s s N s.M
Age Gnoup
15 20 25 35 45 55 1 1 1 1 1 1 19 24 34 44 54 69
F 9σrθ5:〜on−pa ata〃sation
5しと∂ndandrXj lGO
\ 、戦、
\
go 甥\、
\
80
\ \x
70 \
\、
地
60 凹\
、田一一一一紋 、、
へ
50 \
、
40 ・
、、
301 \、
、
OnU 21
一一鼈黶@1991
一一一一etU一一一一一19ア1 一一一一一b甲一一一一1950
へ
巴、
、、、
、、、
\巴
AgθGnoup
15 20 25 351 1 1 I IS 24 34 44
Figure 7 : Non−labiaiisation ZT
トー一一←
︻﹂﹂鱗 4に﹂ 賃﹂一9︻h冒2∪The level of standardisation turns out to vary accordiRg to the type of phoneme:
consonants are more standardised than vowels, which are more standardised than
word accent (pitch contour).
Consona凱毛s
The graph in figure 4 shows the rate of standardisation for non−labialisation tyPe L This is the item which had become most standardised by 1991. With non−labialisation
tyPe L standardisation has reached its final stage, with all the age groups showing over 900/o 一most people produce the standard phones most of the time. ln the graph the !991 result is plotted with filled circles. In non−labialisation tyPe 1, the velar [k] of the standard language sometimes corresponds to a labio−velar [kW] ln Tsuruoka dialect.
Examples are the initial[k]in standard∫apanese火曜臼 [kajo:bi]℃uesday and the media1[k]呈n西瓜 [surtika] water−melon , which are pronounced in the dialect as[k:wa jo:bi] and [suiikwa]. The vertical axis of the graph, represents the rate of use of the standard language pronunciation. It can be seen that almost everybody uses the standard−Japanese pronunciation. Standardisation is proceeding smoothly, because the local dialect itself already has the [k] phoneme.
Figure 5 shows the rate of standardisation for non−palatalisation. The
standardisation has also reached its final stage by 1991. Standardisation is proceeding smoothly, because the dialect has the consonants [s] and [z].Figure 6 shows another example of this type. lntervocalic voiceless consonants
[k], [t], and [ts] in the standard language correspond to their voiced counterparts in the dialect. The dialect already has the consonants [k,t,s] , and thus standardisation is proceeding smoothly as in the previous case.
Figure 7, for non−la bialisation tyPe ll, shows a slightly different pattern. The oldest age group lags behind the other groups in standardisation. The same is true for figure 8, showing non−nasalisation. ln the dialect, inter−vocalic [b], [d], and [g] nasalise a preceding vowe}, as in [6bi], which is pronounced as [obi] in standard Japanese.
Except for the oldest age group, most of the informants pronounce the word without nasalisation. We assume this pattern represents a stage of standardisation immediately preceding the previous one.
Vowel position
Finally, figure 9 shows a pattem where only the youngest groups have arrived at the ceiling. The age groups from 35 up in 1991 retain approximately the level of standardisation of 20 years ago. The same is true for three other types of vowel change,
30
shown in figure le 一 figure 12. lt is interesting tkat these items all concem vowei quality. This suggests that vowels are harder to change than consonaRts.
We combined the data for all the segrnental phonemes, in the same way as in previous surveys, to give a total score for standardisation. We gave one point for a
staRdard pronunciation and no points for a Tsuruoka dialect pronunciation. We
assigned points for all 31 items iRvestigated and calculated the total scores for each informant. 1 will refer to these total scores as the pltonetic scores. Figure 13 shows the distribution of cumulative phonetic scores. Tlte graph s X−axis shows the phonetic scores and its Y−axis the percentage of informants. Over the period of the three surveys, the curve becomes progressively steeper. By 199!, over 900/o of the informants scored between 27 and 31 points, and the lowest score was 9 points.The graph in figure 14 shows the distribution of phonetic scores by age. lt shows that, overall, standardisation is progressing smoothly with time. ln 1950, standard−
isation was most prominent in the 25 to 34 age group. ln 1971, standardisation was highest in the youngest age group (15 to 19 years) and phonetic score decreased progressively with age. By 1991, standardisation was all but complete, except for the oldest age groups.
Standa nd rZ)
∩UnU nUQJ
︷一N s
巴一一一雛ノ
× sIL一一
s.s 鳳承 s
/臥
t N 一 N
x N N N
一 1991
一一一l一一 1971
一一一一?一一一一 195g
× s
x
NNN
N s
N
s ×
s × s
x
s
xNEヨー一一一一一一鼎鰺
湾ge Gnoup
StandardrX?
00 1
15 20 25 35 1 1 1 ]
19 24 34 44
Figure 8 : Non−nasafisation
に﹂ーバ触
45
田!
〆2ヨ ,陽、
i N h ..
一一一 臥 N N x N N N N N N
一一一一垂蝿黶@1 991 一一一一ムー一艸一一197G
−m一一一浮宙鼈鼈鼈黷f95e
×
N×
sts柏__
軸『鴫一凹
Age Gnoup
5i∩﹂=﹂ρQ
15 20 25 35
1 ] 1 ll9 24 34 44
Figure 9 : Non−centratisation f
45 トート 頁J4 56
亡﹂一Q︾5Lt∂nda戸d r/ノ 0
︷0 へ N
N
×
s
s
×
.× s 戦、
× 凍 N
\tS〆/田陶一鴨一一eq、
N N
一一
黶@1991
一一一一一 1971
一一一一壕鼈鼈鼈黶@1950
x
s
x
s
×
NN唐m Ss>!x
也、
s li ss 、、、巴
Age Gnoup i5 20
t l
19 24
ご﹂42つ﹂
Figure 10: Vowel shift ll
StandardrZ?
00nUO OOJΩU7
1
∩︶nUnUOOO ︻bF34∩﹂2イー
へ N
︐田
x
k×
N N×
田一一一一固、
l s
s l
t
=J14つ﹂4 ごJ14
45
×s.,.../ Al × s
にり一∩﹂=UCU
傘
\ \
無、、 \
ヘ へ
・ \ 、
、、、
、
、田一一一一一一口凹
1991
一一一@th 一一一 1971 一一一一c一一一一一1950
Age Gnoup 15 20 25 35
1 1 1 1
19 24 34 4A
Figure 12 : Non−centralisation ZT
に﹂一4バ﹃口﹂ ﹁OiQり虞﹂︹0
5日目∂ndsρd r2ツ
160
瞠rへ 、
9G ・
戦、
\、
80 M\、
\、
70m 八一 ̲
60 \ \
・・
@ \\\
40 油\ 五
、、、、、
も
30 、E1
20
10
15 2e
I I
19 24
︻﹂42う﹂
Figure fl: Vovvel shift f
Cu〃uノεt!レθr7ノ
00 0︹コ
ーAge GnouF?
35 45 55
1 1 144 54 69
◎ 伯91 一,一▲一一一一一19ア1 一一一一c一一一一一1950
♂
afZfiiZ]
戸ノ. ♂
}
︐︐ノ弁 一
〆〆 ︐奔
戸ずF
︐轟/ 〆
ノgコ.一 ノロ臼!孟r﹃− 〆
32
t
O 5 16 15 20 25 30
Phonett c Score rFu l l Score 一一 3ff
Figure 13 : Cumulative phonetic scores
Phone t lc Scone 30
25
20
15
10
5
!田一一一一囚\、
ノ ヘヘ ロ難ノ@ \臥
、 、 、 、 、 、 、 、
×
s
199G
×
N
杓、
s N s s一@s
x
.M
傘
一一A一一 197a
一一一聯yヨρ一一一騨195〔}
Age Gnoup
﹃︾lGJ震﹂ρ0
ん7ノ
︑
、
︑ ︑ ︑
、
\一虹 ︑竃︑
︑︑−︑覧︑\
︑
一︑歪
t5 2e
I I
IS 24
口J4
2つ﹂ =﹂一﹂﹃つ﹂4 ︻﹂一445
本 1991 △一一一19ア1
一一一帥yヨ俸需一一一1950
{.一一一一m.一一Pm.ntLL:::i?:::::i#Lm
\∴
\⁝ \狛
Figure 74 : Phonetic scores by age
0 1 2 3 4 5
AcOθnと SCO戸θ (Fu i l Score = 5]
rigure 15:Distribution oi accent scores
Word Accent
Japanese has phonemic pitch accent. Both the inventory of pitch contours and the pitch contours of individual words vary between dialects. To investigate accent in Tsuruoka, the same five words were tested in each of the three surveys:
II1tzz@rd 一Zligtk2{akyo 一ZEtgun2keuruoha ,ECnglSIIi h
猫 [nelko] 〔nergO] cat
旗 [ha「ta] [hada] fiag 背中 [se「naka] [∫e「nalga] back 烏 [kalrasu1] [ka「ralsi] crow 団扇 [ur「t∫ilwa】 繊dz碁wa] paper fan
The pitch c◎ntours used in the Tsuruoka dialect for two of these words,旗and団 扇,are not found in standard Japanese at a11. The contours of the other words are used in standard Japanese, but not in those words. For analysis of the survey results, we follow the same method as for the phonetic scores: we score one for a standard Japanese accent and zero for a local accent. Figure 15 shows the distribution of accent score, and figure 16 shows the distribution of accent score by age. We see here that
standardisation has not progressed very much over the years. Notably, even the youngest group has achieved only 600/o standardisation, and the oldest group barely 350/o.
By comparing this graph to the graph in figtare 14, it can be seen that the total accent score obtained by the youngest age group in 1991, about 600/o, is almost the same as their phonetic score in 1950. These results support the oft−repeated claim that it is
difficult to change one s accent. Compared with standardisation of segmental
phonemes, accent standardisation is proceeding slowiy.The Nomoto−Egawa Model
It should be possible to explain these results, and account for the pattems with a simple model of language change. Figure 17 summarizes one such model, a model proposed by Kikuo Nemoto and Kiyoshi Egawa in 1974 in their report on our 1971 survey.
Cemparifig the pattems of standardisation for different age groups in 1950 and 1971, Nomoto and Egawa suggested the following progression. lnitially, they assumed,
there was a stage, stage O, where there was no standardisation for any age groups. ln the next stage, stage 1, standardisation had not progressed much. They noted a distribution with a peak in the 25 to 34 age group. ln the next stage, stage II,
standardisation was more advanced in the younger age groups than in the older age groups. Lastly, in stage III, all the age groups have reached saturation level.
Thus standardisation of the language proceeded in three stages. ln the first stage,
people around the age of 30 were the most infiuenced by the standard language. During the shift from the first to the second stage, standardisation in the lower age groups progressed rapidly, relative to that in the higher age groups. ln the third stage, the standard language is used throughout the society, With the older age groups catching up with the younger ones.
Some justification can be given for this model based on social changes between the two surveys. Before 1950, transport was less developed and individuals contacts with eutside areas were resuicted. Given these circumstances, speakers around the age of thirty were the most likely to have a wide range of interaction with others, and so were more open to standardisation. This was the biggest impetus for the first stage ef standardisation in 1950. By 1971, transport had improved greatly, increasing individual mobility. Furthermore, with the development of the mass media during the 1960 s, the exposure of all age groups to the standard language increased. The Royal Wedding in 1959 and the Tokyo Olympic Games in 1964, helped to spread television throughout Japan. As a result, we find rapid standardisation in the lewer age groups, which are
34
Accen t Scorre 5
4
3
2
t
sとθ〃(Vardr■ノ
ioe
車 1991
一一一 dl 一一一 1 971
一一一一mコー一一一一 歪950
90
8S 78
60
M
鋤!
.A.
×
N域、
一x 、旗_
へ鴨r転 辱\
鞠、
輻「s
.va…魯隔胸一一一田\
t .一?s?g一一 一一一一一ee
Age Group
貫﹂一QJπJρQ
se
40 30 2e
w
15 20
1 1
19 24
=﹂42つ﹂
綴J4 34
﹇﹂一44=﹂ept一一NX一一Hs一.一s.
Sヒage 玉1董
af
T
/
t
、!一へ、
、ノ「 \ N
/ x
r
Stage 1{
×
xMX.
S七age I .
Stage O 巴…一一田…一網隔一輔一一va一 一一一一me..一...
10 2fi 30
40 50 60十
Age Gnoup
Figure f6:Accent seores by age Figure f7 : Nomoto−ffgawa rnodel
the most sensitive to language. This was the main impetus for the second stage.
Of course, when Nomoto and Egawa proposed this model, they were working from rather sparse data. We now have the means to check their ideas. Retuming to the survey results we have seen so far, we see that phonetic standardisation (figure 14) was at Stage 1 or the beginning of Stage II in 1950, at Stage II in 1971, and at Stage III in 1991. Accent standardisation (figure 16), which has progressed more slowly, was at Stage 1 in 1991.
1 should like to look now at the change in particular groups of individuals over time. So far, the informants ages have always been plotted on the horizontal axis. lf the same person is plotted always in the same place, we can see just when people learn the standard langtiage. The graph in figure i8 shows the phonetic scores in relation to the informant s date of birth, so unlike the previous graphs, the age of informants is lower and further right on the horizontal axis. This type of graph makes it easier to see the change for a given age group. Comparing the results for 1950 afid 1971, standard−
isation for informants born between 1916 and 1935, that is, those who were 15 to 34 years old in 1950, had progressed. However, comparing the results for 1971 and 1991,
there was little change in the level of the standardisation for informants born between
Phonetic Scone 3C
25 /tr 一/
・/・
./ /〔ヨ\、
20十 一2N J. .一 r 一hs一.m
!田 x!
ノ!
1
15十 ・!
ノ ,[ゴ 〆一 凹 〆〆
Ge
5
.,,s5
廓 agsa
一一一?ー一一 哩971
一一一一b一一一一一i950
Yean oF BIrth
1886 1896 1906 191S 1926 」936 1946
: I I { I l l
1895 1995 1915 1925 IS35 194S 1955
ゆθ丁8 : Phonet o soo「θby yea「oデb th (Fu Dsoσ θ二3か
Accen t Score 5
4
3
2
..A / ...tr !一 一門配一
1 ,.一一!一k一
,田一一一一一一一囚、、、
巴一一一一一一一田〆一一/ \\田 Ye∂〜へ
1956
:
1965
0ヂθ〆rth f966
1
i975
1886 1896 a906 1916 1926
1 1 1 1 1
a895 1905 19t5 1925 1935
F 9しf「e 19:Accent sσσ8 θby yθar of bir th (Fα〃score=5)
36
1936
:
1945
1946
E
1955
1956
1
1965
1966
1
1S75
1926 and 1955 一 people who were 15 to 44 years eld in 1971. By 1971 then, this age group had stopped changing: in other words, this is the third stage of the Nomoto−Egawa model. ln the case of accent though, it can be seen that there is a progressive improvement for all age groups, plotted in figure 19. Presumably the improvement will continue into the future. This is in rnarked contrast to figure 18.
Conclusion
In these three surveys, the oldest informant was bom in 1881, and was 69 years old in 1950. The youngest informant was bom in 1975, and was 15 years old in 1991.
Thus we were able to obtain data ofi informants of an age range of about 100 years by conducting three longitudinal surveys of linguistic change at twenty−year intervals.
These data enable us to make a number of substantial observations on language
change in real time: (1) standardisation of pronunciation of segmental phonemes isprogressing rapidly, and ef pitch accent more gradually in Tsuruoka; (2)
standardisation in Tsuruoka validates the Nomoto−Egawa model empirically.
This paper is our first report on the 1991 survey. We are working on a more detailed quantitative analysis of the phonetic data. This wi11 include correlations with language consciousness, and other social factors, such as education, residential history, and occupation. We are doing the same with the lexical and grammatical items, which have not been discussed in this paper:
In addition to the 1991 survey, we conducted another survey in 1992 to investigate the effect of situation on language use in Tsuruoka. This survey was motivated by our belief that in the past surveys, we were looking at the most standardised aspects of language life, narRely the language used in the most formal situations. By introducing another variable, i.e. situations, into our study, we hope to look at the informants language ability in a more compreheAsive manner. This, in turn, will help us to evaluate the results of our p ast surveys with respect to eur initial purpose, namely an investigation of language standardisation in the everyday lives of dialect speakers.
References
The National Language Research lnstitute (1953) Kokuntsu Kofeugo Kenkywfo Hofeoku 5 Chiiki Shakai no Gengo Seileatsu: Tsuruoha ni ofeeru Jittai Chousa [Language Life in a Regional Society: An Actual lnvestigation of Tsuruoka City]. Tokyo: Shuei Shuppan.
(1974) Kokun tsu Koleugo Kenfeywfo Hokoleu 52 Chiilei Shakai no Gengo Seikatsu:
Tsecruoka ni okeru 20 Nen Mae to no Hihaleu [Language Life Style in a Regional Society: A Comparison of the Past and the Present on Tsuruoka City with a Twenty−Year lnterval].
Tokyo: Shuei Shuppan.
(1988) An lntroduction to the National Language Research lnstitute: A sketch of its achievements (third edition).
NOMOTO, Kikuo. (1975) How much has been Standardized over the Past Twenty Years?
Fre(i C. C・Pen9・(ed.)Language in/mPanese Society. Tokyo:Univers量ty of Tokyo Press.
CHAMBERS, F. K. (1se5) Sociolinguistic Theory: Linguistic Variation and its Social Signtficance. Oxford UK & Camridge USA: Blackwell.
Acknewledgeinents
I should llke to express speclal appreciation to John PhMips, Takako Ayusawa,
Prem Motowani, Polly Szatrowski, Margaret Maeda and Kenjiro Matsuda for their help in preparing this paper. 1 should aiso like to thank Jack Chambers and Der−Hwa Ran for valuable comments on the interpretation of the data in relation to the Nomoto−Egawa model .
Using the Nornoto−Egawa Model, it is necessary to examine the factors which influence the speed of standaradisation for each age group. A more detailed stat1stical and sociological study of the cause−effect relations of the patterns of change at each stage will be the subject of a future paper.
This paper is a revised version of an abstract of a presentatien given at Method NEII : IBternatlona}CGnference on Method in Dialectology he}d at the Univers三ty of Vict面a,
British Cplumbia, CANADA in August 1993.
(Received i February 1997>
Masato YONEDA(米田正人)
The National Language Resear¢h 1nstitute 3−9−14, Nishigaoka, Kita−ku, Tokyo 115 Japan myoneda@kokken. go. jp
38
『鑓本:語秘学』2(1997年10月)24−39 〔研究論文〕
鶴岡市における共通語化の調査
一約20年間隔で行われた3回の調査を比較して一 米田 正人
(国立国語研究所)
キーワード
国立国語研究所,鶴岡方言,共通語化,音声,アクセント
概 要
国立国語研究所では昭和25年度と昭和46年の2度にわたって文部省科学硬究費の交付を受け,山 形県鶴岡市において地域社会に於ける書語生活の実態調査を実施した。それにより,戦後四半世紀 の急激な社会変化の中で方言が共通語化していく過程について,その実態や社:会的な要因を明らか にした。本研究は,これらの成果を受け継ぎ,鶴岡市において約20年間隔の第3次調査を実施する とともに,言語変化を将来に向けて経年的に調査記述していくための基礎構築をN的として行われ
た。
また,本報告は平成3年度および4年度の文部省科学研究費補助金(総合研究(A)),研究課題名
「地域社会の言語生活一鶴岡市における戦後の変化一(課題番号03301060)(研究代表者 江川清)の交 付を受けて行った調査研究のうち,音声,アクセントの共通藷化について一部をまとめたものであ
り,平成5年8月,カナダのビクトリア大学で行われたMethods蟹(方言研究の方法論に関する国際 会議)でm頭発表した内容に加筆訂正したものである。