Tibetan Dialects: A Case Study in Areal Phonology
Tibetan Dialects: A Case Study in Areal Phonology
Tibetan Dialects: A Case Study in Areal Phonology
Themchen Tibetan
Balti (Kharkoo)
Labrang Tibetan
Purki
Nyinpa Cone
Tod Tibetan
Spiti Tibetan
Nangchenpa Tibetan
Ladakhi (Leh)
Baima
Zhongu Tibetan
Khalong Tibetan
gSerpa
Kham Tibetan
Bragg.yab Tibetan
Humla Bhotia
Drokpa Tibetan (Zhongba)
Lhasa Tibetan
Shigatse Tibetan
gLo Tibetan
Kyirong
Tibetan
South Mustang Tibetan
(Jharkhot)
Dingri Tibetan
Yolmo
Southern Cuona
Walungge
Solu
SoluSherpa
Sherpa(Lukla)
(Hile)
Dzongkha
Denjongka
Sogpho Tibetan
Northern Cuona
Dongwang Tibetan
Kami Tibetan
Rgyalthang Tibetan
Up to 3 elements
st
We
ern
le
dia
cts
Southern dialects
Nishi 1986:
Central (U-Tsang)
Western Archaic
Western Innovative
Southern
Khams
Amdo
Bradley
1997:
Tournadre
2014:
Western
Central
Khams
Amdo
North-Western
Western
Central
South-Western
Southern
South-Eastern
Eastern
North-Eastern
A promising candidate
phonological inventories
k/
g/
x/
A bad idea
Table 1: Binary matrix for three small inventories
/p, t, k/
/b, d, g/
/f, , x/
p
1
0
0
t
1
0
0
k
1
0
0
b
0
1
0
d
0
1
0
g
0
1
0
f
0
0
1
0
0
1
x
0
0
1
9
an inventory consisting of /p, t, k/ will
be analysed as having absolutely
An alternative
Instead of whole phonemes use their
component features (IPA, Hayess, etc.) or
subsets of features found in the phonemes
of the inventories.
More fine-grained analyses are possible
(size of feature subsets can be controlled,
weights can be used).
10
1
0
0
1
0
0
1
0
0
0
1
0
0
1
0
0
1
0
0
0
1
0
0
1
0
0
1
plosive
1
1
0
fricative
0
0
1
voiceless
1
0
1
voiced
0
1
0
bilabial
1
1
1
denti-alveolar
1
1
1
velar
1
1
0
plosive
3
3
0
fricative
0
0
3
voiceless
3
0
3
voiced
0
3
0
bilabial
1
1
1
denti-alveolar
1
1
1
velar
1
1
1
plosive, voiceles
3
0
0
plosive, voiced
0
3
0
12
plosive, bilabial
1
1
0
fricative, voiceless
0
0
3
2-tuples
of
features
with
weights
plosive fricative voiceless voiced bilabial denti-alveolar
3
0 (first 3several0 columns)
1
1
3
0
0
3
0
3
3
0
1
1
1
1
velar
1
1
1
plosive, voiceles
3
0
0
plosive, voiced
0
3
0
13
plosive, bilabial
1
1
0
fricative, voiceless
0
0
3
A continuum
Individual feature without weights
(inventory as a feature set).
14
The procedure
1. Construct a matrix for consonant
inventories based on the IPA features.
2. Calculate pairwise Manhattan distances.
3. Clusterise the languages using the
Partitioning Around Medoids methods.
4. Investigate the geographical layout of
the resulting clusters.
16
17
18
19
20
21
23
25
Discussion
28
So
uth
De
njo
iga
tse
Sh
ibe
tan
tan
Tib
e
ibe
tan
gri
T
Din
aT
gb
a)
Pu
rki
(Zh
on
Lh
as
tan
tan
eh
)
Dz
on
gk
ha
Wa
lun
gg
e
kh
i (L
La
da
kla
(Lu
e)
Hil
iti T
ibe
Sp
erp
a
on
a
Cu
tan
ibe
a(
Sh
erp
Sh
ibe
So
lu
lu
So
ern
No
r th
ark
oo
)
ark
ho
t)
(Jh
To
dT
eta
n
eta
Tib
Kh
lti (
Ba
gL
o
Yo
lm
ng
So
ka
uth
ern
Cu
on
a
Tib
aT
ok
p
Dr
ibe
tan
bT
Ba
im
ho
tia
on
gT
ibe
tan
Ky
ir
Bra
g
g.y
a
ng
sta
Mu
ibe
tan
aB
Hu
ml
eta
n
Tib
tan
eta
Tib
nT
he
Th
em
c
am
Kh
an
g
on
e
ibe
ng
T
La
bra
erp
gS
inp
aC
Ny
Rg
ya
lth
Na
gp
ho
Tib
ng
eta
ch
n
en
pa
Tib
eta
Zh
n
on
gu
Tib
eta
n
Ka
mi
Tib
eta
Kh
n
alo
ng
T
ibe
Do
tan
ng
wa
ng
Tib
eta
n
So
20
40
60
80
100
43
35
31
44
46
uth
44
P
la
te
au (<30)
9
-w
41
2
35
35
es
ter
n
rim
40
36
(~
34
27
3344
32
37
32
35
)
29
10
40
27
27
37
49
44
+)
33
485
(4
0
61
st
So
33
5 9 40
Ea
35
56
30
Lhasa Tibetan
31
Shigatse Tibetan
32
Ladakhi
33
34
Kami Tibetan
35
Zhongu Tibetan
36
Sogpho Tibetan
37
38
39
Lhasa
Shigatse
Ladakhi
Kami Tibetan
Zhongu Tibetan
Sogpho Tibetan
A wider picture: western varieties pattern with TGMT and Tshangla; eastern ones
with Shixing, rGyalrongic, Lizu, and Ersu but not Qiang, Puxi, and Pumi
rGyalrongic
Qiangic
Shixing
TGMT
Tshangla
45
What to do next